SLURM_JOB_ID = 9198410 SLURM_JOB_NAME = nvr_elm_llm:dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume RUN_NAME = tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume OUTPUT_DIR = runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume NNODES = 8 SLURM_JOB_ID = 9198410 SLURM_JOB_NAME = nvr_elm_llm:dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume RUN_NAME = tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume OUTPUT_DIR = runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume NNODES = 8 SLURM_JOB_ID = 9198410 SLURM_JOB_NAME = nvr_elm_llm:dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume RUN_NAME = tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume OUTPUT_DIR = runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume NNODES = 8 SLURM_JOB_ID = 9198410 SLURM_JOB_NAME = nvr_elm_llm:dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume RUN_NAME = tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume OUTPUT_DIR = runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume NNODES = 8 SLURM_JOB_ID = 9198410 SLURM_JOB_NAME = nvr_elm_llm:dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume RUN_NAME = tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume OUTPUT_DIR = runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume NNODES = 8 SLURM_JOB_ID = 9198410 SLURM_JOB_NAME = nvr_elm_llm:dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume RUN_NAME = tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume OUTPUT_DIR = runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume NNODES = 8 NODES = batch-block1-2023 batch-block1-3060 batch-block1-3564 batch-block1-2095 batch-block1-2096 batch-block1-2081 batch-block1-3462 batch-block1-3261 NODE_RANK = 3 GPUS_PER_NODE = 8 NODES = batch-block1-2023 batch-block1-3060 batch-block1-3564 batch-block1-2095 batch-block1-2096 batch-block1-2081 batch-block1-3462 batch-block1-3261 NODE_RANK = 5 GPUS_PER_NODE = 8 NODES = batch-block1-2023 batch-block1-3060 batch-block1-3564 batch-block1-2095 batch-block1-2096 batch-block1-2081 batch-block1-3462 batch-block1-3261 NODE_RANK = 4 GPUS_PER_NODE = 8 NODES = batch-block1-2023 batch-block1-3060 batch-block1-3564 batch-block1-2095 batch-block1-2096 batch-block1-2081 batch-block1-3462 batch-block1-3261 NODE_RANK = 6 GPUS_PER_NODE = 8 NODES = batch-block1-2023 batch-block1-3060 batch-block1-3564 batch-block1-2095 batch-block1-2096 batch-block1-2081 batch-block1-3462 batch-block1-3261 NODE_RANK = 7 GPUS_PER_NODE = 8 MASTER_ADDR = batch-block1-2023 MASTER_PORT = 25001 GLOBAL_TRAIN_BATCH_SIZE = GRADIENT_ACCUMULATION_STEPS = PER_DEVICE_TRAIN_BATCH_SIZE = NODES = batch-block1-2023 batch-block1-3060 batch-block1-3564 batch-block1-2095 batch-block1-2096 batch-block1-2081 batch-block1-3462 batch-block1-3261 NODE_RANK = 2 GPUS_PER_NODE = 8 MASTER_ADDR = batch-block1-2023 MASTER_PORT = 25001 GLOBAL_TRAIN_BATCH_SIZE = GRADIENT_ACCUMULATION_STEPS = PER_DEVICE_TRAIN_BATCH_SIZE = MASTER_ADDR = batch-block1-2023 MASTER_PORT = 25001 GLOBAL_TRAIN_BATCH_SIZE = GRADIENT_ACCUMULATION_STEPS = PER_DEVICE_TRAIN_BATCH_SIZE = Starting training from base model: Qwen/Qwen3-8B MASTER_ADDR = batch-block1-2023 MASTER_PORT = 25001 GLOBAL_TRAIN_BATCH_SIZE = GRADIENT_ACCUMULATION_STEPS = PER_DEVICE_TRAIN_BATCH_SIZE = Starting training from base model: Qwen/Qwen3-8B MASTER_ADDR = batch-block1-2023 MASTER_PORT = 25001 GLOBAL_TRAIN_BATCH_SIZE = GRADIENT_ACCUMULATION_STEPS = PER_DEVICE_TRAIN_BATCH_SIZE = Starting training from base model: Qwen/Qwen3-8B Starting training from base model: Qwen/Qwen3-8B MASTER_ADDR = batch-block1-2023 MASTER_PORT = 25001 GLOBAL_TRAIN_BATCH_SIZE = GRADIENT_ACCUMULATION_STEPS = PER_DEVICE_TRAIN_BATCH_SIZE = Starting training from base model: Qwen/Qwen3-8B Starting training from base model: Qwen/Qwen3-8B SLURM_JOB_ID = 9198410 SLURM_JOB_NAME = nvr_elm_llm:dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume RUN_NAME = tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume OUTPUT_DIR = runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume NNODES = 8 NODES = batch-block1-2023 batch-block1-3060 batch-block1-3564 batch-block1-2095 batch-block1-2096 batch-block1-2081 batch-block1-3462 batch-block1-3261 NODE_RANK = 0 GPUS_PER_NODE = 8 SLURM_JOB_ID = 9198410 SLURM_JOB_NAME = nvr_elm_llm:dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume RUN_NAME = tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume OUTPUT_DIR = runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume NNODES = 8 MASTER_ADDR = batch-block1-2023 MASTER_PORT = 25001 GLOBAL_TRAIN_BATCH_SIZE = GRADIENT_ACCUMULATION_STEPS = PER_DEVICE_TRAIN_BATCH_SIZE = NODES = batch-block1-2023 batch-block1-3060 batch-block1-3564 batch-block1-2095 batch-block1-2096 batch-block1-2081 batch-block1-3462 batch-block1-3261 NODE_RANK = 1 GPUS_PER_NODE = 8 Starting training from base model: Qwen/Qwen3-8B MASTER_ADDR = batch-block1-2023 MASTER_PORT = 25001 GLOBAL_TRAIN_BATCH_SIZE = GRADIENT_ACCUMULATION_STEPS = PER_DEVICE_TRAIN_BATCH_SIZE = Starting training from base model: Qwen/Qwen3-8B Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 Imported prefix tree collator v1 [2026-04-15 09:07:26,399] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,640] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,643] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,676] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,706] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,706] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,706] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,708] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,713] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,713] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,719] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,754] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,818] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,818] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,830] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,831] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,838] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,875] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,878] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,917] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,930] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,940] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,946] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,949] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,961] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:26,983] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,024] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,026] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,027] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,036] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,036] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,058] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,075] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,097] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,099] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,104] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,123] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,129] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,131] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,132] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,158] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,158] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,159] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,178] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,189] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,196] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,196] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,197] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,202] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,218] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,245] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,266] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,292] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,320] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,344] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,371] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,380] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,389] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,390] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,394] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,394] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,398] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,398] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:27,401] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2026-04-15 09:07:35,049] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,311] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,347] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,364] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,419] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,419] [INFO] [comm.py:706:init_distributed] Initializing TorchBackend in DeepSpeed with backend nccl [2026-04-15 09:07:35,474] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,490] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,515] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,516] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,518] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,522] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,525] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,539] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,574] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,635] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,685] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,714] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,721] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,725] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,809] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,838] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,846] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,884] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,888] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,892] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,912] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,934] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,935] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,944] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,961] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,964] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,975] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:35,986] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,014] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,021] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,037] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,037] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,043] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,051] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,079] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,127] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,131] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,133] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,142] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,161] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,187] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,202] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,206] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,222] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,303] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,324] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,401] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,414] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,426] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,502] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,503] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,507] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,589] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,680] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,752] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,814] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,816] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,830] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:36,862] [INFO] [comm.py:675:init_distributed] cdb=None [2026-04-15 09:07:37,188] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,188] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,194] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,197] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,208] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,208] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,234] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,264] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,264] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,276] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,280] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,288] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,285] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,330] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,392] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,394] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,416] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,420] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,555] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,578] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,582] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,667] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,675] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:37,686] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,600] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,601] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,602] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,607] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,612] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,612] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,645] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,653] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,662] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,670] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,678] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,686] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,692] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,694] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,697] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,698] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,699] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,710] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,713] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,714] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,713] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,715] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,731] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,731] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,731] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:42,731] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:48,713] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:48,716] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:48,747] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:48,823] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:48,825] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:48,864] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:48,878] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:48,895] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:48,928] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:48,945] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:52,684] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:52,727] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:52,759] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:52,787] [INFO] [config.py:744:__init__] Config mesh_device None world_size = 64 [2026-04-15 09:07:56,853] [INFO] [partition_parameters.py:348:__exit__] finished initializing model - num_params = 399, num_elems = 8.19B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Qwen templates for Qwen/Qwen3-8B Using Prefix Tree collator args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2096 Using Prefix Tree collator Using Prefix Tree collator args.report_to: ['wandb'] args.report_to: ['wandb'] Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2095 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3060 Using Prefix Tree collator Using Prefix Tree collator args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2023 Using Prefix Tree collator args.report_to: ['wandb'] args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2023 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3261 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2023 Using Prefix Tree collator Using Prefix Tree collator Using Prefix Tree collator args.report_to: ['wandb'] args.report_to: ['wandb'] Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2023 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2096 Using Prefix Tree collator args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3060 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3060 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3564 Using Prefix Tree collator Using Prefix Tree collator Using Prefix Tree collator args.report_to: ['wandb'] args.report_to: ['wandb'] Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2096 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3060 Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3261 args.report_to: ['wandb'] Using Prefix Tree collator Using Prefix Tree collator Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3564 Using Prefix Tree collator args.report_to: ['wandb'] Using Prefix Tree collator Using Prefix Tree collator args.report_to: ['wandb'] Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3462 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3462 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3261 args.report_to: ['wandb'] args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3261 Using Prefix Tree collator args.report_to: ['wandb'] args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2023 args.report_to: ['wandb'] Using Prefix Tree collator Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3060 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2095 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2081 Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3261 args.report_to: ['wandb'] args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2096 Using Prefix Tree collator Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3462 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3462 Using Prefix Tree collator Using Prefix Tree collator Using Prefix Tree collator args.report_to: ['wandb'] Using Prefix Tree collator args.report_to: ['wandb'] Using Prefix Tree collator Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2096 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2096 Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3564 args.report_to: ['wandb'] args.report_to: ['wandb'] args.report_to: ['wandb'] Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3462 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2081 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2081 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2095 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3462 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2023 Using Prefix Tree collator Using Prefix Tree collatorargs.report_to: ['wandb'] Using Prefix Tree collator Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3060 args.report_to: ['wandb'] Using Prefix Tree collator args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3060 Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3564 args.report_to: ['wandb'] args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2081 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3462 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2081 Using Prefix Tree collator Using Prefix Tree collator args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2023 Using Prefix Tree collator args.report_to: ['wandb'] args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3261 Using Prefix Tree collator args.report_to: ['wandb'] Using Prefix Tree collator Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2095 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2096 Using Prefix Tree collatorUsing Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3261 args.report_to: ['wandb'] args.report_to: ['wandb'] Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2081 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3462 args.report_to: ['wandb'] args.report_to: ['wandb']args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3564 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-3564 args.report_to: ['wandb'] Using Prefix Tree collator Using Prefix Tree collatorUsing Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3060 Using Prefix Tree collator args.report_to: ['wandb'] Using Prefix Tree collator args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2095 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2081 Using Prefix Tree collator args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2095 args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2095 args.report_to: ['wandb'] args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2023 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3564 Using Prefix Tree collator Using Prefix Tree collator args.report_to: ['wandb'] args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-35_batch-block1-2095 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2096 Using Prefix Tree collator Using Prefix Tree collator args.report_to: ['wandb'] Using Prefix Tree collator args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-2081 args.report_to: ['wandb'] args.report_to: ['wandb'] args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3564 args.logging_dir: runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume/runs/Apr15_09-07-36_batch-block1-3261 Parameter Offload: Total persistent parameters: 308224 in 145 params Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 6362 Attention mask shape: torch.Size([1, 1, 6362, 6362]) Position ids shape: torch.Size([1, 6362]) Input IDs shape: torch.Size([1, 6362]) Labels shape: torch.Size([1, 6362]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 10523 Attention mask shape: torch.Size([1, 1, 10523, 10523]) Position ids shape: torch.Size([1, 10523]) Input IDs shape: torch.Size([1, 10523]) Labels shape: torch.Size([1, 10523]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 18410 Attention mask shape: torch.Size([1, 1, 18410, 18410]) Position ids shape: torch.Size([1, 18410]) Input IDs shape: torch.Size([1, 18410]) Labels shape: torch.Size([1, 18410]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 12385 Attention mask shape: torch.Size([1, 1, 12385, 12385]) Position ids shape: torch.Size([1, 12385]) Input IDs shape: torch.Size([1, 12385]) Labels shape: torch.Size([1, 12385]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 18098 Attention mask shape: torch.Size([1, 1, 18098, 18098]) Position ids shape: torch.Size([1, 18098]) Input IDs shape: torch.Size([1, 18098]) Labels shape: torch.Size([1, 18098]) Final batch size: 1, sequence length: 18060 Attention mask shape: torch.Size([1, 1, 18060, 18060]) Position ids shape: torch.Size([1, 18060]) Input IDs shape: torch.Size([1, 18060]) Labels shape: torch.Size([1, 18060]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 17299 Attention mask shape: torch.Size([1, 1, 17299, 17299]) Position ids shape: torch.Size([1, 17299]) Input IDs shape: torch.Size([1, 17299]) Labels shape: torch.Size([1, 17299]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 12656 Attention mask shape: torch.Size([1, 1, 12656, 12656]) Position ids shape: torch.Size([1, 12656]) Input IDs shape: torch.Size([1, 12656]) Labels shape: torch.Size([1, 12656]) Final batch size: 1, sequence length: 18915 Attention mask shape: torch.Size([1, 1, 18915, 18915]) Position ids shape: torch.Size([1, 18915]) Input IDs shape: torch.Size([1, 18915]) Labels shape: torch.Size([1, 18915]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 9091 Attention mask shape: torch.Size([1, 1, 9091, 9091]) Position ids shape: torch.Size([1, 9091]) Input IDs shape: torch.Size([1, 9091]) Labels shape: torch.Size([1, 9091]) Final batch size: 1, sequence length: 24365 Attention mask shape: torch.Size([1, 1, 24365, 24365]) Position ids shape: torch.Size([1, 24365]) Input IDs shape: torch.Size([1, 24365]) Labels shape: torch.Size([1, 24365]) Final batch size: 1, sequence length: 28841 Attention mask shape: torch.Size([1, 1, 28841, 28841]) Position ids shape: torch.Size([1, 28841]) Input IDs shape: torch.Size([1, 28841]) Labels shape: torch.Size([1, 28841]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 29113 Attention mask shape: torch.Size([1, 1, 29113, 29113]) Position ids shape: torch.Size([1, 29113]) Input IDs shape: torch.Size([1, 29113]) Labels shape: torch.Size([1, 29113]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 26976 Attention mask shape: torch.Size([1, 1, 26976, 26976]) Position ids shape: torch.Size([1, 26976]) Input IDs shape: torch.Size([1, 26976]) Labels shape: torch.Size([1, 26976]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 11819 Attention mask shape: torch.Size([1, 1, 11819, 11819]) Position ids shape: torch.Size([1, 11819]) Input IDs shape: torch.Size([1, 11819]) Labels shape: torch.Size([1, 11819]) Final batch size: 1, sequence length: 27541 Attention mask shape: torch.Size([1, 1, 27541, 27541]) Position ids shape: torch.Size([1, 27541]) Input IDs shape: torch.Size([1, 27541]) Labels shape: torch.Size([1, 27541]) Final batch size: 1, sequence length: 25172 Attention mask shape: torch.Size([1, 1, 25172, 25172]) Position ids shape: torch.Size([1, 25172]) Input IDs shape: torch.Size([1, 25172]) Labels shape: torch.Size([1, 25172]) Final batch size: 1, sequence length: 21450 Attention mask shape: torch.Size([1, 1, 21450, 21450]) Position ids shape: torch.Size([1, 21450]) Input IDs shape: torch.Size([1, 21450]) Labels shape: torch.Size([1, 21450]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 7722 Attention mask shape: torch.Size([1, 1, 7722, 7722]) Position ids shape: torch.Size([1, 7722]) Input IDs shape: torch.Size([1, 7722]) Labels shape: torch.Size([1, 7722]) Final batch size: 1, sequence length: 20827 Attention mask shape: torch.Size([1, 1, 20827, 20827]) Position ids shape: torch.Size([1, 20827]) Input IDs shape: torch.Size([1, 20827]) Labels shape: torch.Size([1, 20827]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 24880 Attention mask shape: torch.Size([1, 1, 24880, 24880]) Position ids shape: torch.Size([1, 24880]) Input IDs shape: torch.Size([1, 24880]) Labels shape: torch.Size([1, 24880]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 21596 Attention mask shape: torch.Size([1, 1, 21596, 21596]) Position ids shape: torch.Size([1, 21596]) Input IDs shape: torch.Size([1, 21596]) Labels shape: torch.Size([1, 21596]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 6948 Attention mask shape: torch.Size([1, 1, 6948, 6948]) Position ids shape: torch.Size([1, 6948]) Input IDs shape: torch.Size([1, 6948]) Labels shape: torch.Size([1, 6948]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 29875 Attention mask shape: torch.Size([1, 1, 29875, 29875]) Position ids shape: torch.Size([1, 29875]) Input IDs shape: torch.Size([1, 29875]) Labels shape: torch.Size([1, 29875]) Final batch size: 1, sequence length: 21491 Attention mask shape: torch.Size([1, 1, 21491, 21491]) Position ids shape: torch.Size([1, 21491]) Input IDs shape: torch.Size([1, 21491]) Labels shape: torch.Size([1, 21491]) Final batch size: 1, sequence length: 30623 Attention mask shape: torch.Size([1, 1, 30623, 30623]) Position ids shape: torch.Size([1, 30623]) Input IDs shape: torch.Size([1, 30623]) Labels shape: torch.Size([1, 30623]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 14009 Attention mask shape: torch.Size([1, 1, 14009, 14009]) Position ids shape: torch.Size([1, 14009]) Input IDs shape: torch.Size([1, 14009]) Labels shape: torch.Size([1, 14009]) Final batch size: 1, sequence length: 19705 Attention mask shape: torch.Size([1, 1, 19705, 19705]) Position ids shape: torch.Size([1, 19705]) Input IDs shape: torch.Size([1, 19705]) Labels shape: torch.Size([1, 19705]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 13215 Attention mask shape: torch.Size([1, 1, 13215, 13215]) Position ids shape: torch.Size([1, 13215]) Input IDs shape: torch.Size([1, 13215]) Labels shape: torch.Size([1, 13215]) Final batch size: 1, sequence length: 25014 Attention mask shape: torch.Size([1, 1, 25014, 25014]) Position ids shape: torch.Size([1, 25014]) Input IDs shape: torch.Size([1, 25014]) Labels shape: torch.Size([1, 25014]) Final batch size: 1, sequence length: 33367 Attention mask shape: torch.Size([1, 1, 33367, 33367]) Position ids shape: torch.Size([1, 33367]) Input IDs shape: torch.Size([1, 33367]) Labels shape: torch.Size([1, 33367]) Final batch size: 1, sequence length: 22098 Attention mask shape: torch.Size([1, 1, 22098, 22098]) Position ids shape: torch.Size([1, 22098]) Input IDs shape: torch.Size([1, 22098]) Labels shape: torch.Size([1, 22098]) Final batch size: 1, sequence length: 38249 Attention mask shape: torch.Size([1, 1, 38249, 38249]) Position ids shape: torch.Size([1, 38249]) Input IDs shape: torch.Size([1, 38249]) Labels shape: torch.Size([1, 38249]) Final batch size: 1, sequence length: 20363 Attention mask shape: torch.Size([1, 1, 20363, 20363]) Position ids shape: torch.Size([1, 20363]) Input IDs shape: torch.Size([1, 20363]) Labels shape: torch.Size([1, 20363]) Final batch size: 1, sequence length: 19586 Attention mask shape: torch.Size([1, 1, 19586, 19586]) Position ids shape: torch.Size([1, 19586]) Input IDs shape: torch.Size([1, 19586]) Labels shape: torch.Size([1, 19586]) Final batch size: 1, sequence length: 27265 Attention mask shape: torch.Size([1, 1, 27265, 27265]) Position ids shape: torch.Size([1, 27265]) Input IDs shape: torch.Size([1, 27265]) Labels shape: torch.Size([1, 27265]) Final batch size: 1, sequence length: 17400 Attention mask shape: torch.Size([1, 1, 17400, 17400]) Position ids shape: torch.Size([1, 17400]) Input IDs shape: torch.Size([1, 17400]) Labels shape: torch.Size([1, 17400]) Final batch size: 1, sequence length: 28641 Attention mask shape: torch.Size([1, 1, 28641, 28641]) Position ids shape: torch.Size([1, 28641]) Input IDs shape: torch.Size([1, 28641]) Labels shape: torch.Size([1, 28641]) Final batch size: 1, sequence length: 15656 Attention mask shape: torch.Size([1, 1, 15656, 15656]) Position ids shape: torch.Size([1, 15656]) Input IDs shape: torch.Size([1, 15656]) Labels shape: torch.Size([1, 15656]) Final batch size: 1, sequence length: 24001 Attention mask shape: torch.Size([1, 1, 24001, 24001]) Position ids shape: torch.Size([1, 24001]) Input IDs shape: torch.Size([1, 24001]) Labels shape: torch.Size([1, 24001]) Final batch size: 1, sequence length: 30101 Attention mask shape: torch.Size([1, 1, 30101, 30101]) Position ids shape: torch.Size([1, 30101]) Input IDs shape: torch.Size([1, 30101]) Labels shape: torch.Size([1, 30101]) Final batch size: 1, sequence length: 39142 Attention mask shape: torch.Size([1, 1, 39142, 39142]) Position ids shape: torch.Size([1, 39142]) Input IDs shape: torch.Size([1, 39142]) Labels shape: torch.Size([1, 39142]) Final batch size: 1, sequence length: 26306 Attention mask shape: torch.Size([1, 1, 26306, 26306]) Position ids shape: torch.Size([1, 26306]) Input IDs shape: torch.Size([1, 26306]) Labels shape: torch.Size([1, 26306]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 29632 Attention mask shape: torch.Size([1, 1, 29632, 29632]) Position ids shape: torch.Size([1, 29632]) Input IDs shape: torch.Size([1, 29632]) Labels shape: torch.Size([1, 29632]) Final batch size: 1, sequence length: 18565 Attention mask shape: torch.Size([1, 1, 18565, 18565]) Position ids shape: torch.Size([1, 18565]) Input IDs shape: torch.Size([1, 18565]) Labels shape: torch.Size([1, 18565]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 18606 Attention mask shape: torch.Size([1, 1, 18606, 18606]) Position ids shape: torch.Size([1, 18606]) Input IDs shape: torch.Size([1, 18606]) Labels shape: torch.Size([1, 18606]) Final batch size: 1, sequence length: 27243 Attention mask shape: torch.Size([1, 1, 27243, 27243]) Position ids shape: torch.Size([1, 27243]) Input IDs shape: torch.Size([1, 27243]) Labels shape: torch.Size([1, 27243]) Final batch size: 1, sequence length: 15217 Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 15217, 15217]) Position ids shape: torch.Size([1, 15217]) Input IDs shape: torch.Size([1, 15217]) Labels shape: torch.Size([1, 15217]) Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30066 Attention mask shape: torch.Size([1, 1, 30066, 30066]) Position ids shape: torch.Size([1, 30066]) Input IDs shape: torch.Size([1, 30066]) Labels shape: torch.Size([1, 30066]) Final batch size: 1, sequence length: 26939 Attention mask shape: torch.Size([1, 1, 26939, 26939]) Position ids shape: torch.Size([1, 26939]) Input IDs shape: torch.Size([1, 26939]) Labels shape: torch.Size([1, 26939]) Final batch size: 1, sequence length: 21348 Attention mask shape: torch.Size([1, 1, 21348, 21348]) Position ids shape: torch.Size([1, 21348]) Input IDs shape: torch.Size([1, 21348]) Labels shape: torch.Size([1, 21348]) Final batch size: 1, sequence length: 22625 Attention mask shape: torch.Size([1, 1, 22625, 22625]) Position ids shape: torch.Size([1, 22625]) Input IDs shape: torch.Size([1, 22625]) Labels shape: torch.Size([1, 22625]) Final batch size: 1, sequence length: 32609 Attention mask shape: torch.Size([1, 1, 32609, 32609]) Position ids shape: torch.Size([1, 32609]) Input IDs shape: torch.Size([1, 32609]) Labels shape: torch.Size([1, 32609]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 37945 Attention mask shape: torch.Size([1, 1, 37945, 37945]) Position ids shape: torch.Size([1, 37945]) Input IDs shape: torch.Size([1, 37945]) Labels shape: torch.Size([1, 37945]) Final batch size: 1, sequence length: 27441 Attention mask shape: torch.Size([1, 1, 27441, 27441]) Position ids shape: torch.Size([1, 27441]) Input IDs shape: torch.Size([1, 27441]) Labels shape: torch.Size([1, 27441]) Final batch size: 1, sequence length: 36511 Attention mask shape: torch.Size([1, 1, 36511, 36511]) Position ids shape: torch.Size([1, 36511]) Input IDs shape: torch.Size([1, 36511]) Labels shape: torch.Size([1, 36511]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21567 Attention mask shape: torch.Size([1, 1, 21567, 21567]) Position ids shape: torch.Size([1, 21567]) Input IDs shape: torch.Size([1, 21567]) Labels shape: torch.Size([1, 21567]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 13509 Attention mask shape: torch.Size([1, 1, 13509, 13509]) Position ids shape: torch.Size([1, 13509]) Input IDs shape: torch.Size([1, 13509]) Labels shape: torch.Size([1, 13509]) Final batch size: 1, sequence length: 30802 Attention mask shape: torch.Size([1, 1, 30802, 30802]) Position ids shape: torch.Size([1, 30802]) Input IDs shape: torch.Size([1, 30802]) Labels shape: torch.Size([1, 30802]) Final batch size: 1, sequence length: 24622 Attention mask shape: torch.Size([1, 1, 24622, 24622]) Position ids shape: torch.Size([1, 24622]) Input IDs shape: torch.Size([1, 24622]) Labels shape: torch.Size([1, 24622]) Final batch size: 1, sequence length: 32352 Attention mask shape: torch.Size([1, 1, 32352, 32352]) Position ids shape: torch.Size([1, 32352]) Input IDs shape: torch.Size([1, 32352]) Labels shape: torch.Size([1, 32352]) Final batch size: 1, sequence length: 16587 Attention mask shape: torch.Size([1, 1, 16587, 16587]) Position ids shape: torch.Size([1, 16587]) Input IDs shape: torch.Size([1, 16587]) Labels shape: torch.Size([1, 16587]) Final batch size: 1, sequence length: 16677 Attention mask shape: torch.Size([1, 1, 16677, 16677]) Position ids shape: torch.Size([1, 16677]) Input IDs shape: torch.Size([1, 16677]) Labels shape: torch.Size([1, 16677]) Final batch size: 1, sequence length: 17778 Attention mask shape: torch.Size([1, 1, 17778, 17778]) Position ids shape: torch.Size([1, 17778]) Input IDs shape: torch.Size([1, 17778]) Labels shape: torch.Size([1, 17778]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 15508 Attention mask shape: torch.Size([1, 1, 15508, 15508]) Position ids shape: torch.Size([1, 15508]) Input IDs shape: torch.Size([1, 15508]) Labels shape: torch.Size([1, 15508]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 20611 Attention mask shape: torch.Size([1, 1, 20611, 20611]) Position ids shape: torch.Size([1, 20611]) Input IDs shape: torch.Size([1, 20611]) Labels shape: torch.Size([1, 20611]) Final batch size: 1, sequence length: 25447 Attention mask shape: torch.Size([1, 1, 25447, 25447]) Position ids shape: torch.Size([1, 25447]) Input IDs shape: torch.Size([1, 25447]) Labels shape: torch.Size([1, 25447]) Final batch size: 1, sequence length: 35153 Attention mask shape: torch.Size([1, 1, 35153, 35153]) Position ids shape: torch.Size([1, 35153]) Input IDs shape: torch.Size([1, 35153]) Labels shape: torch.Size([1, 35153]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 33839 Attention mask shape: torch.Size([1, 1, 33839, 33839]) Position ids shape: torch.Size([1, 33839]) Input IDs shape: torch.Size([1, 33839]) Labels shape: torch.Size([1, 33839]) Final batch size: 1, sequence length: 20422 Attention mask shape: torch.Size([1, 1, 20422, 20422]) Position ids shape: torch.Size([1, 20422]) Input IDs shape: torch.Size([1, 20422]) Labels shape: torch.Size([1, 20422]) Final batch size: 1, sequence length: 23258 Attention mask shape: torch.Size([1, 1, 23258, 23258]) Position ids shape: torch.Size([1, 23258]) Input IDs shape: torch.Size([1, 23258]) Labels shape: torch.Size([1, 23258]) Final batch size: 1, sequence length: 30109 Attention mask shape: torch.Size([1, 1, 30109, 30109]) Position ids shape: torch.Size([1, 30109]) Input IDs shape: torch.Size([1, 30109]) Labels shape: torch.Size([1, 30109]) Final batch size: 1, sequence length: 39836 Attention mask shape: torch.Size([1, 1, 39836, 39836]) Position ids shape: torch.Size([1, 39836]) Input IDs shape: torch.Size([1, 39836]) Labels shape: torch.Size([1, 39836]) Final batch size: 1, sequence length: 40745 Attention mask shape: torch.Size([1, 1, 40745, 40745]) Position ids shape: torch.Size([1, 40745]) Input IDs shape: torch.Size([1, 40745]) Labels shape: torch.Size([1, 40745]) Final batch size: 1, sequence length: 19036 Attention mask shape: torch.Size([1, 1, 19036, 19036]) Position ids shape: torch.Size([1, 19036]) Input IDs shape: torch.Size([1, 19036]) Labels shape: torch.Size([1, 19036]) Final batch size: 1, sequence length: 36271 Attention mask shape: torch.Size([1, 1, 36271, 36271]) Position ids shape: torch.Size([1, 36271]) Input IDs shape: torch.Size([1, 36271]) Labels shape: torch.Size([1, 36271]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40496 Attention mask shape: torch.Size([1, 1, 40496, 40496]) Position ids shape: torch.Size([1, 40496]) Input IDs shape: torch.Size([1, 40496]) Labels shape: torch.Size([1, 40496]) Final batch size: 1, sequence length: 13095 Attention mask shape: torch.Size([1, 1, 13095, 13095]) Position ids shape: torch.Size([1, 13095]) Input IDs shape: torch.Size([1, 13095]) Labels shape: torch.Size([1, 13095]) Final batch size: 1, sequence length: 21758 Attention mask shape: torch.Size([1, 1, 21758, 21758]) Position ids shape: torch.Size([1, 21758]) Input IDs shape: torch.Size([1, 21758]) Labels shape: torch.Size([1, 21758]) Final batch size: 1, sequence length: 34142 Attention mask shape: torch.Size([1, 1, 34142, 34142]) Position ids shape: torch.Size([1, 34142]) Input IDs shape: torch.Size([1, 34142]) Labels shape: torch.Size([1, 34142]) Final batch size: 1, sequence length: 13297 Attention mask shape: torch.Size([1, 1, 13297, 13297]) Position ids shape: torch.Size([1, 13297]) Input IDs shape: torch.Size([1, 13297]) Labels shape: torch.Size([1, 13297]) Final batch size: 1, sequence length: 12653 Attention mask shape: torch.Size([1, 1, 12653, 12653]) Position ids shape: torch.Size([1, 12653]) Input IDs shape: torch.Size([1, 12653]) Labels shape: torch.Size([1, 12653]) Final batch size: 1, sequence length: 34937 Attention mask shape: torch.Size([1, 1, 34937, 34937]) Position ids shape: torch.Size([1, 34937]) Input IDs shape: torch.Size([1, 34937]) Labels shape: torch.Size([1, 34937]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 9947 Attention mask shape: torch.Size([1, 1, 9947, 9947]) Position ids shape: torch.Size([1, 9947]) Input IDs shape: torch.Size([1, 9947]) Labels shape: torch.Size([1, 9947]) Final batch size: 1, sequence length: 24298 Attention mask shape: torch.Size([1, 1, 24298, 24298]) Position ids shape: torch.Size([1, 24298]) Input IDs shape: torch.Size([1, 24298]) Labels shape: torch.Size([1, 24298]) Final batch size: 1, sequence length: 12224 Attention mask shape: torch.Size([1, 1, 12224, 12224]) Position ids shape: torch.Size([1, 12224]) Input IDs shape: torch.Size([1, 12224]) Labels shape: torch.Size([1, 12224]) Final batch size: 1, sequence length: 22786 Attention mask shape: torch.Size([1, 1, 22786, 22786]) Position ids shape: torch.Size([1, 22786]) Input IDs shape: torch.Size([1, 22786]) Labels shape: torch.Size([1, 22786]) Final batch size: 1, sequence length: 32564 Attention mask shape: torch.Size([1, 1, 32564, 32564]) Position ids shape: torch.Size([1, 32564]) Input IDs shape: torch.Size([1, 32564]) Labels shape: torch.Size([1, 32564]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21683 Attention mask shape: torch.Size([1, 1, 21683, 21683]) Position ids shape: torch.Size([1, 21683]) Input IDs shape: torch.Size([1, 21683]) Labels shape: torch.Size([1, 21683]) Final batch size: 1, sequence length: 22623 Attention mask shape: torch.Size([1, 1, 22623, 22623]) Position ids shape: torch.Size([1, 22623]) Input IDs shape: torch.Size([1, 22623]) Labels shape: torch.Size([1, 22623]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17373 Attention mask shape: torch.Size([1, 1, 17373, 17373]) Position ids shape: torch.Size([1, 17373]) Input IDs shape: torch.Size([1, 17373]) Labels shape: torch.Size([1, 17373]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 14995 Attention mask shape: torch.Size([1, 1, 14995, 14995]) Position ids shape: torch.Size([1, 14995]) Input IDs shape: torch.Size([1, 14995]) Labels shape: torch.Size([1, 14995]) Final batch size: 1, sequence length: 32472 Attention mask shape: torch.Size([1, 1, 32472, 32472]) Position ids shape: torch.Size([1, 32472]) Input IDs shape: torch.Size([1, 32472]) Labels shape: torch.Size([1, 32472]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 35864 Attention mask shape: torch.Size([1, 1, 35864, 35864]) Position ids shape: torch.Size([1, 35864]) Input IDs shape: torch.Size([1, 35864]) Labels shape: torch.Size([1, 35864]) Final batch size: 1, sequence length: 32344 Attention mask shape: torch.Size([1, 1, 32344, 32344]) Position ids shape: torch.Size([1, 32344]) Input IDs shape: torch.Size([1, 32344]) Labels shape: torch.Size([1, 32344]) Final batch size: 1, sequence length: 24424 Attention mask shape: torch.Size([1, 1, 24424, 24424]) Position ids shape: torch.Size([1, 24424]) Input IDs shape: torch.Size([1, 24424]) Labels shape: torch.Size([1, 24424]) Final batch size: 1, sequence length: 28634 Attention mask shape: torch.Size([1, 1, 28634, 28634]) Position ids shape: torch.Size([1, 28634]) Input IDs shape: torch.Size([1, 28634]) Labels shape: torch.Size([1, 28634]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32753 Attention mask shape: torch.Size([1, 1, 32753, 32753]) Position ids shape: torch.Size([1, 32753]) Input IDs shape: torch.Size([1, 32753]) Labels shape: torch.Size([1, 32753]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26006 Attention mask shape: torch.Size([1, 1, 26006, 26006]) Position ids shape: torch.Size([1, 26006]) Input IDs shape: torch.Size([1, 26006]) Labels shape: torch.Size([1, 26006]) Final batch size: 1, sequence length: 26537 Attention mask shape: torch.Size([1, 1, 26537, 26537]) Position ids shape: torch.Size([1, 26537]) Input IDs shape: torch.Size([1, 26537]) Labels shape: torch.Size([1, 26537]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32919 Attention mask shape: torch.Size([1, 1, 32919, 32919]) Position ids shape: torch.Size([1, 32919]) Input IDs shape: torch.Size([1, 32919]) Labels shape: torch.Size([1, 32919]) Final batch size: 1, sequence length: 32514 Attention mask shape: torch.Size([1, 1, 32514, 32514]) Position ids shape: torch.Size([1, 32514]) Input IDs shape: torch.Size([1, 32514]) Labels shape: torch.Size([1, 32514]) Final batch size: 1, sequence length: 23362 Attention mask shape: torch.Size([1, 1, 23362, 23362]) Position ids shape: torch.Size([1, 23362]) Input IDs shape: torch.Size([1, 23362]) Labels shape: torch.Size([1, 23362]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 37168 Attention mask shape: torch.Size([1, 1, 37168, 37168]) Position ids shape: torch.Size([1, 37168]) Input IDs shape: torch.Size([1, 37168]) Labels shape: torch.Size([1, 37168]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28861 Attention mask shape: torch.Size([1, 1, 28861, 28861]) Position ids shape: torch.Size([1, 28861]) Input IDs shape: torch.Size([1, 28861]) Labels shape: torch.Size([1, 28861]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.4291, 'grad_norm': 2.128983729982784, 'learning_rate': 0.0, 'num_tokens': -inf, 'epoch': 0.12} Final batch size: 1, sequence length: 7360 Attention mask shape: torch.Size([1, 1, 7360, 7360]) Position ids shape: torch.Size([1, 7360]) Input IDs shape: torch.Size([1, 7360]) Labels shape: torch.Size([1, 7360]) Final batch size: 1, sequence length: 4858 Attention mask shape: torch.Size([1, 1, 4858, 4858]) Position ids shape: torch.Size([1, 4858]) Input IDs shape: torch.Size([1, 4858]) Labels shape: torch.Size([1, 4858]) Final batch size: 1, sequence length: 10301 Attention mask shape: torch.Size([1, 1, 10301, 10301]) Position ids shape: torch.Size([1, 10301]) Input IDs shape: torch.Size([1, 10301]) Labels shape: torch.Size([1, 10301]) Final batch size: 1, sequence length: 11728 Attention mask shape: torch.Size([1, 1, 11728, 11728]) Position ids shape: torch.Size([1, 11728]) Input IDs shape: torch.Size([1, 11728]) Labels shape: torch.Size([1, 11728]) Final batch size: 1, sequence length: 6316 Attention mask shape: torch.Size([1, 1, 6316, 6316]) Position ids shape: torch.Size([1, 6316]) Input IDs shape: torch.Size([1, 6316]) Labels shape: torch.Size([1, 6316]) Final batch size: 1, sequence length: 11548 Attention mask shape: torch.Size([1, 1, 11548, 11548]) Position ids shape: torch.Size([1, 11548]) Input IDs shape: torch.Size([1, 11548]) Labels shape: torch.Size([1, 11548]) Final batch size: 1, sequence length: 12293 Attention mask shape: torch.Size([1, 1, 12293, 12293]) Position ids shape: torch.Size([1, 12293]) Input IDs shape: torch.Size([1, 12293]) Labels shape: torch.Size([1, 12293]) Final batch size: 1, sequence length: 13355 Attention mask shape: torch.Size([1, 1, 13355, 13355]) Position ids shape: torch.Size([1, 13355]) Input IDs shape: torch.Size([1, 13355]) Labels shape: torch.Size([1, 13355]) Final batch size: 1, sequence length: 14363 Attention mask shape: torch.Size([1, 1, 14363, 14363]) Position ids shape: torch.Size([1, 14363]) Input IDs shape: torch.Size([1, 14363]) Labels shape: torch.Size([1, 14363]) Final batch size: 1, sequence length: 16827 Attention mask shape: torch.Size([1, 1, 16827, 16827]) Position ids shape: torch.Size([1, 16827]) Input IDs shape: torch.Size([1, 16827]) Labels shape: torch.Size([1, 16827]) Final batch size: 1, sequence length: 17108 Attention mask shape: torch.Size([1, 1, 17108, 17108]) Position ids shape: torch.Size([1, 17108]) Input IDs shape: torch.Size([1, 17108]) Labels shape: torch.Size([1, 17108]) Final batch size: 1, sequence length: 14704 Attention mask shape: torch.Size([1, 1, 14704, 14704]) Position ids shape: torch.Size([1, 14704]) Input IDs shape: torch.Size([1, 14704]) Labels shape: torch.Size([1, 14704]) Final batch size: 1, sequence length: 18496 Attention mask shape: torch.Size([1, 1, 18496, 18496]) Position ids shape: torch.Size([1, 18496]) Input IDs shape: torch.Size([1, 18496]) Labels shape: torch.Size([1, 18496]) Final batch size: 1, sequence length: 14587 Attention mask shape: torch.Size([1, 1, 14587, 14587]) Position ids shape: torch.Size([1, 14587]) Input IDs shape: torch.Size([1, 14587]) Labels shape: torch.Size([1, 14587]) Final batch size: 1, sequence length: 16536 Attention mask shape: torch.Size([1, 1, 16536, 16536]) Position ids shape: torch.Size([1, 16536]) Input IDs shape: torch.Size([1, 16536]) Labels shape: torch.Size([1, 16536]) Final batch size: 1, sequence length: 7221 Attention mask shape: torch.Size([1, 1, 7221, 7221]) Position ids shape: torch.Size([1, 7221]) Input IDs shape: torch.Size([1, 7221]) Labels shape: torch.Size([1, 7221]) Final batch size: 1, sequence length: 18741 Attention mask shape: torch.Size([1, 1, 18741, 18741]) Position ids shape: torch.Size([1, 18741]) Input IDs shape: torch.Size([1, 18741]) Labels shape: torch.Size([1, 18741]) Final batch size: 1, sequence length: 17733 Attention mask shape: torch.Size([1, 1, 17733, 17733]) Position ids shape: torch.Size([1, 17733]) Input IDs shape: torch.Size([1, 17733]) Labels shape: torch.Size([1, 17733]) Final batch size: 1, sequence length: 18927 Attention mask shape: torch.Size([1, 1, 18927, 18927]) Position ids shape: torch.Size([1, 18927]) Input IDs shape: torch.Size([1, 18927]) Labels shape: torch.Size([1, 18927]) Final batch size: 1, sequence length: 18393 Attention mask shape: torch.Size([1, 1, 18393, 18393]) Position ids shape: torch.Size([1, 18393]) Input IDs shape: torch.Size([1, 18393]) Labels shape: torch.Size([1, 18393]) Final batch size: 1, sequence length: 18653 Attention mask shape: torch.Size([1, 1, 18653, 18653]) Position ids shape: torch.Size([1, 18653]) Input IDs shape: torch.Size([1, 18653]) Labels shape: torch.Size([1, 18653]) Final batch size: 1, sequence length: 13638 Attention mask shape: torch.Size([1, 1, 13638, 13638]) Position ids shape: torch.Size([1, 13638]) Input IDs shape: torch.Size([1, 13638]) Labels shape: torch.Size([1, 13638]) Final batch size: 1, sequence length: 17415 Attention mask shape: torch.Size([1, 1, 17415, 17415]) Position ids shape: torch.Size([1, 17415]) Input IDs shape: torch.Size([1, 17415]) Labels shape: torch.Size([1, 17415]) Final batch size: 1, sequence length: 20933 Attention mask shape: torch.Size([1, 1, 20933, 20933]) Position ids shape: torch.Size([1, 20933]) Input IDs shape: torch.Size([1, 20933]) Labels shape: torch.Size([1, 20933]) Final batch size: 1, sequence length: 22004 Attention mask shape: torch.Size([1, 1, 22004, 22004]) Position ids shape: torch.Size([1, 22004]) Input IDs shape: torch.Size([1, 22004]) Labels shape: torch.Size([1, 22004]) Final batch size: 1, sequence length: 17220 Attention mask shape: torch.Size([1, 1, 17220, 17220]) Position ids shape: torch.Size([1, 17220]) Input IDs shape: torch.Size([1, 17220]) Labels shape: torch.Size([1, 17220]) Final batch size: 1, sequence length: 19414 Attention mask shape: torch.Size([1, 1, 19414, 19414]) Position ids shape: torch.Size([1, 19414]) Input IDs shape: torch.Size([1, 19414]) Labels shape: torch.Size([1, 19414]) Final batch size: 1, sequence length: 16750 Attention mask shape: torch.Size([1, 1, 16750, 16750]) Position ids shape: torch.Size([1, 16750]) Input IDs shape: torch.Size([1, 16750]) Labels shape: torch.Size([1, 16750]) Final batch size: 1, sequence length: 21420 Attention mask shape: torch.Size([1, 1, 21420, 21420]) Position ids shape: torch.Size([1, 21420]) Input IDs shape: torch.Size([1, 21420]) Labels shape: torch.Size([1, 21420]) Final batch size: 1, sequence length: 22391 Attention mask shape: torch.Size([1, 1, 22391, 22391]) Position ids shape: torch.Size([1, 22391]) Input IDs shape: torch.Size([1, 22391]) Labels shape: torch.Size([1, 22391]) Final batch size: 1, sequence length: 20612 Attention mask shape: torch.Size([1, 1, 20612, 20612]) Position ids shape: torch.Size([1, 20612]) Input IDs shape: torch.Size([1, 20612]) Labels shape: torch.Size([1, 20612]) Final batch size: 1, sequence length: 22887 Attention mask shape: torch.Size([1, 1, 22887, 22887]) Position ids shape: torch.Size([1, 22887]) Input IDs shape: torch.Size([1, 22887]) Labels shape: torch.Size([1, 22887]) Final batch size: 1, sequence length: 11067 Attention mask shape: torch.Size([1, 1, 11067, 11067]) Position ids shape: torch.Size([1, 11067]) Input IDs shape: torch.Size([1, 11067]) Labels shape: torch.Size([1, 11067]) Final batch size: 1, sequence length: 20579 Attention mask shape: torch.Size([1, 1, 20579, 20579]) Position ids shape: torch.Size([1, 20579]) Input IDs shape: torch.Size([1, 20579]) Labels shape: torch.Size([1, 20579]) Final batch size: 1, sequence length: 25747 Attention mask shape: torch.Size([1, 1, 25747, 25747]) Position ids shape: torch.Size([1, 25747]) Input IDs shape: torch.Size([1, 25747]) Labels shape: torch.Size([1, 25747]) Final batch size: 1, sequence length: 24988 Attention mask shape: torch.Size([1, 1, 24988, 24988]) Position ids shape: torch.Size([1, 24988]) Input IDs shape: torch.Size([1, 24988]) Labels shape: torch.Size([1, 24988]) Final batch size: 1, sequence length: 25477 Attention mask shape: torch.Size([1, 1, 25477, 25477]) Position ids shape: torch.Size([1, 25477]) Input IDs shape: torch.Size([1, 25477]) Labels shape: torch.Size([1, 25477]) Final batch size: 1, sequence length: 26663 Attention mask shape: torch.Size([1, 1, 26663, 26663]) Position ids shape: torch.Size([1, 26663]) Input IDs shape: torch.Size([1, 26663]) Labels shape: torch.Size([1, 26663]) Final batch size: 1, sequence length: 15317 Attention mask shape: torch.Size([1, 1, 15317, 15317]) Position ids shape: torch.Size([1, 15317]) Input IDs shape: torch.Size([1, 15317]) Labels shape: torch.Size([1, 15317]) Final batch size: 1, sequence length: 25651 Attention mask shape: torch.Size([1, 1, 25651, 25651]) Position ids shape: torch.Size([1, 25651]) Input IDs shape: torch.Size([1, 25651]) Labels shape: torch.Size([1, 25651]) Final batch size: 1, sequence length: 27447 Attention mask shape: torch.Size([1, 1, 27447, 27447]) Position ids shape: torch.Size([1, 27447]) Input IDs shape: torch.Size([1, 27447]) Labels shape: torch.Size([1, 27447]) Final batch size: 1, sequence length: 11184 Attention mask shape: torch.Size([1, 1, 11184, 11184]) Position ids shape: torch.Size([1, 11184]) Input IDs shape: torch.Size([1, 11184]) Labels shape: torch.Size([1, 11184]) Final batch size: 1, sequence length: 19552 Attention mask shape: torch.Size([1, 1, 19552, 19552]) Position ids shape: torch.Size([1, 19552]) Input IDs shape: torch.Size([1, 19552]) Labels shape: torch.Size([1, 19552]) Final batch size: 1, sequence length: 19869 Attention mask shape: torch.Size([1, 1, 19869, 19869]) Position ids shape: torch.Size([1, 19869]) Input IDs shape: torch.Size([1, 19869]) Labels shape: torch.Size([1, 19869]) Final batch size: 1, sequence length: 10719 Attention mask shape: torch.Size([1, 1, 10719, 10719]) Position ids shape: torch.Size([1, 10719]) Input IDs shape: torch.Size([1, 10719]) Labels shape: torch.Size([1, 10719]) Final batch size: 1, sequence length: 16915 Attention mask shape: torch.Size([1, 1, 16915, 16915]) Position ids shape: torch.Size([1, 16915]) Input IDs shape: torch.Size([1, 16915]) Labels shape: torch.Size([1, 16915]) Final batch size: 1, sequence length: 30031 Attention mask shape: torch.Size([1, 1, 30031, 30031]) Position ids shape: torch.Size([1, 30031]) Input IDs shape: torch.Size([1, 30031]) Labels shape: torch.Size([1, 30031]) Final batch size: 1, sequence length: 17911 Attention mask shape: torch.Size([1, 1, 17911, 17911]) Position ids shape: torch.Size([1, 17911]) Input IDs shape: torch.Size([1, 17911]) Labels shape: torch.Size([1, 17911]) Final batch size: 1, sequence length: 30981 Attention mask shape: torch.Size([1, 1, 30981, 30981]) Position ids shape: torch.Size([1, 30981]) Input IDs shape: torch.Size([1, 30981]) Labels shape: torch.Size([1, 30981]) Final batch size: 1, sequence length: 17395 Attention mask shape: torch.Size([1, 1, 17395, 17395]) Position ids shape: torch.Size([1, 17395]) Input IDs shape: torch.Size([1, 17395]) Labels shape: torch.Size([1, 17395]) Final batch size: 1, sequence length: 28777 Attention mask shape: torch.Size([1, 1, 28777, 28777]) Position ids shape: torch.Size([1, 28777]) Input IDs shape: torch.Size([1, 28777]) Labels shape: torch.Size([1, 28777]) Final batch size: 1, sequence length: 30601 Attention mask shape: torch.Size([1, 1, 30601, 30601]) Position ids shape: torch.Size([1, 30601]) Input IDs shape: torch.Size([1, 30601]) Labels shape: torch.Size([1, 30601]) Final batch size: 1, sequence length: 16953 Attention mask shape: torch.Size([1, 1, 16953, 16953]) Position ids shape: torch.Size([1, 16953]) Input IDs shape: torch.Size([1, 16953]) Labels shape: torch.Size([1, 16953]) Final batch size: 1, sequence length: 24782 Attention mask shape: torch.Size([1, 1, 24782, 24782]) Position ids shape: torch.Size([1, 24782]) Input IDs shape: torch.Size([1, 24782]) Labels shape: torch.Size([1, 24782]) Final batch size: 1, sequence length: 18376 Attention mask shape: torch.Size([1, 1, 18376, 18376]) Position ids shape: torch.Size([1, 18376]) Input IDs shape: torch.Size([1, 18376]) Labels shape: torch.Size([1, 18376]) Final batch size: 1, sequence length: 28060 Attention mask shape: torch.Size([1, 1, 28060, 28060]) Position ids shape: torch.Size([1, 28060]) Input IDs shape: torch.Size([1, 28060]) Labels shape: torch.Size([1, 28060]) Final batch size: 1, sequence length: 32466 Attention mask shape: torch.Size([1, 1, 32466, 32466]) Position ids shape: torch.Size([1, 32466]) Input IDs shape: torch.Size([1, 32466]) Labels shape: torch.Size([1, 32466]) Final batch size: 1, sequence length: 27334 Attention mask shape: torch.Size([1, 1, 27334, 27334]) Position ids shape: torch.Size([1, 27334]) Input IDs shape: torch.Size([1, 27334]) Labels shape: torch.Size([1, 27334]) Final batch size: 1, sequence length: 26033 Attention mask shape: torch.Size([1, 1, 26033, 26033]) Position ids shape: torch.Size([1, 26033]) Input IDs shape: torch.Size([1, 26033]) Labels shape: torch.Size([1, 26033]) Final batch size: 1, sequence length: 15924 Attention mask shape: torch.Size([1, 1, 15924, 15924]) Position ids shape: torch.Size([1, 15924]) Input IDs shape: torch.Size([1, 15924]) Labels shape: torch.Size([1, 15924]) Final batch size: 1, sequence length: 27480 Attention mask shape: torch.Size([1, 1, 27480, 27480]) Position ids shape: torch.Size([1, 27480]) Input IDs shape: torch.Size([1, 27480]) Labels shape: torch.Size([1, 27480]) Final batch size: 1, sequence length: 12245 Attention mask shape: torch.Size([1, 1, 12245, 12245]) Position ids shape: torch.Size([1, 12245]) Input IDs shape: torch.Size([1, 12245]) Labels shape: torch.Size([1, 12245]) Final batch size: 1, sequence length: 21988 Attention mask shape: torch.Size([1, 1, 21988, 21988]) Position ids shape: torch.Size([1, 21988]) Input IDs shape: torch.Size([1, 21988]) Labels shape: torch.Size([1, 21988]) Final batch size: 1, sequence length: 33725 Attention mask shape: torch.Size([1, 1, 33725, 33725]) Position ids shape: torch.Size([1, 33725]) Input IDs shape: torch.Size([1, 33725]) Labels shape: torch.Size([1, 33725]) Final batch size: 1, sequence length: 37025 Attention mask shape: torch.Size([1, 1, 37025, 37025]) Position ids shape: torch.Size([1, 37025]) Input IDs shape: torch.Size([1, 37025]) Labels shape: torch.Size([1, 37025]) Final batch size: 1, sequence length: 27385 Attention mask shape: torch.Size([1, 1, 27385, 27385]) Position ids shape: torch.Size([1, 27385]) Input IDs shape: torch.Size([1, 27385]) Labels shape: torch.Size([1, 27385]) Final batch size: 1, sequence length: 36175 Attention mask shape: torch.Size([1, 1, 36175, 36175]) Position ids shape: torch.Size([1, 36175]) Input IDs shape: torch.Size([1, 36175]) Labels shape: torch.Size([1, 36175]) Final batch size: 1, sequence length: 14873 Attention mask shape: torch.Size([1, 1, 14873, 14873]) Position ids shape: torch.Size([1, 14873]) Input IDs shape: torch.Size([1, 14873]) Labels shape: torch.Size([1, 14873]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 37511 Attention mask shape: torch.Size([1, 1, 37511, 37511]) Position ids shape: torch.Size([1, 37511]) Input IDs shape: torch.Size([1, 37511]) Labels shape: torch.Size([1, 37511]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26479 Attention mask shape: torch.Size([1, 1, 26479, 26479]) Position ids shape: torch.Size([1, 26479]) Input IDs shape: torch.Size([1, 26479]) Labels shape: torch.Size([1, 26479]) Final batch size: 1, sequence length: 37738 Attention mask shape: torch.Size([1, 1, 37738, 37738]) Position ids shape: torch.Size([1, 37738]) Input IDs shape: torch.Size([1, 37738]) Labels shape: torch.Size([1, 37738]) Final batch size: 1, sequence length: 37158 Attention mask shape: torch.Size([1, 1, 37158, 37158]) Position ids shape: torch.Size([1, 37158]) Input IDs shape: torch.Size([1, 37158]) Labels shape: torch.Size([1, 37158]) Final batch size: 1, sequence length: 32835 Attention mask shape: torch.Size([1, 1, 32835, 32835]) Position ids shape: torch.Size([1, 32835]) Input IDs shape: torch.Size([1, 32835]) Labels shape: torch.Size([1, 32835]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 29586 Attention mask shape: torch.Size([1, 1, 29586, 29586]) Position ids shape: torch.Size([1, 29586]) Input IDs shape: torch.Size([1, 29586]) Labels shape: torch.Size([1, 29586]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31879 Attention mask shape: torch.Size([1, 1, 31879, 31879]) Position ids shape: torch.Size([1, 31879]) Input IDs shape: torch.Size([1, 31879]) Labels shape: torch.Size([1, 31879]) Final batch size: 1, sequence length: 36130 Attention mask shape: torch.Size([1, 1, 36130, 36130]) Position ids shape: torch.Size([1, 36130]) Input IDs shape: torch.Size([1, 36130]) Labels shape: torch.Size([1, 36130]) Final batch size: 1, sequence length: 38986 Attention mask shape: torch.Size([1, 1, 38986, 38986]) Position ids shape: torch.Size([1, 38986]) Input IDs shape: torch.Size([1, 38986]) Labels shape: torch.Size([1, 38986]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40237 Attention mask shape: torch.Size([1, 1, 40237, 40237]) Position ids shape: torch.Size([1, 40237]) Input IDs shape: torch.Size([1, 40237]) Labels shape: torch.Size([1, 40237]) Final batch size: 1, sequence length: 40269 Attention mask shape: torch.Size([1, 1, 40269, 40269]) Position ids shape: torch.Size([1, 40269]) Input IDs shape: torch.Size([1, 40269]) Labels shape: torch.Size([1, 40269]) Final batch size: 1, sequence length: 31348 Attention mask shape: torch.Size([1, 1, 31348, 31348]) Position ids shape: torch.Size([1, 31348]) Input IDs shape: torch.Size([1, 31348]) Labels shape: torch.Size([1, 31348]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 27291 Attention mask shape: torch.Size([1, 1, 27291, 27291]) Position ids shape: torch.Size([1, 27291]) Input IDs shape: torch.Size([1, 27291]) Labels shape: torch.Size([1, 27291]) Final batch size: 1, sequence length: 40397 Attention mask shape: torch.Size([1, 1, 40397, 40397]) Position ids shape: torch.Size([1, 40397]) Input IDs shape: torch.Size([1, 40397]) Labels shape: torch.Size([1, 40397]) Final batch size: 1, sequence length: 32318 Attention mask shape: torch.Size([1, 1, 32318, 32318]) Position ids shape: torch.Size([1, 32318]) Input IDs shape: torch.Size([1, 32318]) Labels shape: torch.Size([1, 32318]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28631 Attention mask shape: torch.Size([1, 1, 28631, 28631]) Position ids shape: torch.Size([1, 28631]) Input IDs shape: torch.Size([1, 28631]) Labels shape: torch.Size([1, 28631]) Final batch size: 1, sequence length: 17711 Attention mask shape: torch.Size([1, 1, 17711, 17711]) Position ids shape: torch.Size([1, 17711]) Input IDs shape: torch.Size([1, 17711]) Labels shape: torch.Size([1, 17711]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36947 Attention mask shape: torch.Size([1, 1, 36947, 36947]) Position ids shape: torch.Size([1, 36947]) Input IDs shape: torch.Size([1, 36947]) Labels shape: torch.Size([1, 36947]) Final batch size: 1, sequence length: 20509 Attention mask shape: torch.Size([1, 1, 20509, 20509]) Position ids shape: torch.Size([1, 20509]) Input IDs shape: torch.Size([1, 20509]) Labels shape: torch.Size([1, 20509]) Final batch size: 1, sequence length: 23740 Attention mask shape: torch.Size([1, 1, 23740, 23740]) Position ids shape: torch.Size([1, 23740]) Input IDs shape: torch.Size([1, 23740]) Labels shape: torch.Size([1, 23740]) Final batch size: 1, sequence length: 32247 Attention mask shape: torch.Size([1, 1, 32247, 32247]) Position ids shape: torch.Size([1, 32247]) Input IDs shape: torch.Size([1, 32247]) Labels shape: torch.Size([1, 32247]) Final batch size: 1, sequence length: 21225 Attention mask shape: torch.Size([1, 1, 21225, 21225]) Position ids shape: torch.Size([1, 21225]) Input IDs shape: torch.Size([1, 21225]) Labels shape: torch.Size([1, 21225]) Final batch size: 1, sequence length: 22896 Attention mask shape: torch.Size([1, 1, 22896, 22896]) Position ids shape: torch.Size([1, 22896]) Input IDs shape: torch.Size([1, 22896]) Labels shape: torch.Size([1, 22896]) Final batch size: 1, sequence length: 33125 Attention mask shape: torch.Size([1, 1, 33125, 33125]) Position ids shape: torch.Size([1, 33125]) Input IDs shape: torch.Size([1, 33125]) Labels shape: torch.Size([1, 33125]) Final batch size: 1, sequence length: 24939 Attention mask shape: torch.Size([1, 1, 24939, 24939]) Position ids shape: torch.Size([1, 24939]) Input IDs shape: torch.Size([1, 24939]) Labels shape: torch.Size([1, 24939]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22205 Attention mask shape: torch.Size([1, 1, 22205, 22205]) Position ids shape: torch.Size([1, 22205]) Input IDs shape: torch.Size([1, 22205]) Labels shape: torch.Size([1, 22205]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 10198 Attention mask shape: torch.Size([1, 1, 10198, 10198]) Position ids shape: torch.Size([1, 10198]) Input IDs shape: torch.Size([1, 10198]) Labels shape: torch.Size([1, 10198]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28149 Attention mask shape: torch.Size([1, 1, 28149, 28149]) Position ids shape: torch.Size([1, 28149]) Input IDs shape: torch.Size([1, 28149]) Labels shape: torch.Size([1, 28149]) Final batch size: 1, sequence length: 28046 Attention mask shape: torch.Size([1, 1, 28046, 28046]) Position ids shape: torch.Size([1, 28046]) Input IDs shape: torch.Size([1, 28046]) Labels shape: torch.Size([1, 28046]) Final batch size: 1, sequence length: 38360 Attention mask shape: torch.Size([1, 1, 38360, 38360]) Position ids shape: torch.Size([1, 38360]) Input IDs shape: torch.Size([1, 38360]) Labels shape: torch.Size([1, 38360]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32767 Attention mask shape: torch.Size([1, 1, 32767, 32767]) Position ids shape: torch.Size([1, 32767]) Input IDs shape: torch.Size([1, 32767]) Labels shape: torch.Size([1, 32767]) Final batch size: 1, sequence length: 36567 Attention mask shape: torch.Size([1, 1, 36567, 36567]) Position ids shape: torch.Size([1, 36567]) Input IDs shape: torch.Size([1, 36567]) Labels shape: torch.Size([1, 36567]) Final batch size: 1, sequence length: 34268 Attention mask shape: torch.Size([1, 1, 34268, 34268]) Position ids shape: torch.Size([1, 34268]) Input IDs shape: torch.Size([1, 34268]) Labels shape: torch.Size([1, 34268]) Final batch size: 1, sequence length: 32313 Attention mask shape: torch.Size([1, 1, 32313, 32313]) Position ids shape: torch.Size([1, 32313]) Input IDs shape: torch.Size([1, 32313]) Labels shape: torch.Size([1, 32313]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.4601, 'grad_norm': 2.379637236085563, 'learning_rate': 2.5e-06, 'num_tokens': -inf, 'epoch': 0.25} Final batch size: 1, sequence length: 5377 Attention mask shape: torch.Size([1, 1, 5377, 5377]) Position ids shape: torch.Size([1, 5377]) Input IDs shape: torch.Size([1, 5377]) Labels shape: torch.Size([1, 5377]) Final batch size: 1, sequence length: 7998 Attention mask shape: torch.Size([1, 1, 7998, 7998]) Position ids shape: torch.Size([1, 7998]) Input IDs shape: torch.Size([1, 7998]) Labels shape: torch.Size([1, 7998]) Final batch size: 1, sequence length: 8436 Attention mask shape: torch.Size([1, 1, 8436, 8436]) Position ids shape: torch.Size([1, 8436]) Input IDs shape: torch.Size([1, 8436]) Labels shape: torch.Size([1, 8436]) Final batch size: 1, sequence length: 7402 Attention mask shape: torch.Size([1, 1, 7402, 7402]) Position ids shape: torch.Size([1, 7402]) Input IDs shape: torch.Size([1, 7402]) Labels shape: torch.Size([1, 7402]) Final batch size: 1, sequence length: 10576 Attention mask shape: torch.Size([1, 1, 10576, 10576]) Position ids shape: torch.Size([1, 10576]) Input IDs shape: torch.Size([1, 10576]) Labels shape: torch.Size([1, 10576]) Final batch size: 1, sequence length: 11709 Attention mask shape: torch.Size([1, 1, 11709, 11709]) Position ids shape: torch.Size([1, 11709]) Input IDs shape: torch.Size([1, 11709]) Labels shape: torch.Size([1, 11709]) Final batch size: 1, sequence length: 11365 Attention mask shape: torch.Size([1, 1, 11365, 11365]) Position ids shape: torch.Size([1, 11365]) Input IDs shape: torch.Size([1, 11365]) Labels shape: torch.Size([1, 11365]) Final batch size: 1, sequence length: 14127 Attention mask shape: torch.Size([1, 1, 14127, 14127]) Position ids shape: torch.Size([1, 14127]) Input IDs shape: torch.Size([1, 14127]) Labels shape: torch.Size([1, 14127]) Final batch size: 1, sequence length: 8655 Attention mask shape: torch.Size([1, 1, 8655, 8655]) Position ids shape: torch.Size([1, 8655]) Input IDs shape: torch.Size([1, 8655]) Labels shape: torch.Size([1, 8655]) Final batch size: 1, sequence length: 7344 Attention mask shape: torch.Size([1, 1, 7344, 7344]) Position ids shape: torch.Size([1, 7344]) Input IDs shape: torch.Size([1, 7344]) Labels shape: torch.Size([1, 7344]) Final batch size: 1, sequence length: 15079 Attention mask shape: torch.Size([1, 1, 15079, 15079]) Position ids shape: torch.Size([1, 15079]) Input IDs shape: torch.Size([1, 15079]) Labels shape: torch.Size([1, 15079]) Final batch size: 1, sequence length: 11623 Attention mask shape: torch.Size([1, 1, 11623, 11623]) Position ids shape: torch.Size([1, 11623]) Input IDs shape: torch.Size([1, 11623]) Labels shape: torch.Size([1, 11623]) Final batch size: 1, sequence length: 15308 Attention mask shape: torch.Size([1, 1, 15308, 15308]) Position ids shape: torch.Size([1, 15308]) Input IDs shape: torch.Size([1, 15308]) Labels shape: torch.Size([1, 15308]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 11678 Attention mask shape: torch.Size([1, 1, 11678, 11678]) Position ids shape: torch.Size([1, 11678]) Input IDs shape: torch.Size([1, 11678]) Labels shape: torch.Size([1, 11678]) Final batch size: 1, sequence length: 15478 Attention mask shape: torch.Size([1, 1, 15478, 15478]) Position ids shape: torch.Size([1, 15478]) Input IDs shape: torch.Size([1, 15478]) Labels shape: torch.Size([1, 15478]) Final batch size: 1, sequence length: 16535 Attention mask shape: torch.Size([1, 1, 16535, 16535]) Position ids shape: torch.Size([1, 16535]) Input IDs shape: torch.Size([1, 16535]) Labels shape: torch.Size([1, 16535]) Final batch size: 1, sequence length: 14648 Attention mask shape: torch.Size([1, 1, 14648, 14648]) Position ids shape: torch.Size([1, 14648]) Input IDs shape: torch.Size([1, 14648]) Labels shape: torch.Size([1, 14648]) Final batch size: 1, sequence length: 13639 Attention mask shape: torch.Size([1, 1, 13639, 13639]) Position ids shape: torch.Size([1, 13639]) Input IDs shape: torch.Size([1, 13639]) Labels shape: torch.Size([1, 13639]) Final batch size: 1, sequence length: 18953 Attention mask shape: torch.Size([1, 1, 18953, 18953]) Position ids shape: torch.Size([1, 18953]) Input IDs shape: torch.Size([1, 18953]) Labels shape: torch.Size([1, 18953]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 19170 Attention mask shape: torch.Size([1, 1, 19170, 19170]) Position ids shape: torch.Size([1, 19170]) Input IDs shape: torch.Size([1, 19170]) Labels shape: torch.Size([1, 19170]) Final batch size: 1, sequence length: 19138 Attention mask shape: torch.Size([1, 1, 19138, 19138]) Position ids shape: torch.Size([1, 19138]) Input IDs shape: torch.Size([1, 19138]) Labels shape: torch.Size([1, 19138]) Final batch size: 1, sequence length: 19338 Attention mask shape: torch.Size([1, 1, 19338, 19338]) Position ids shape: torch.Size([1, 19338]) Input IDs shape: torch.Size([1, 19338]) Labels shape: torch.Size([1, 19338]) Final batch size: 1, sequence length: 18307 Attention mask shape: torch.Size([1, 1, 18307, 18307]) Position ids shape: torch.Size([1, 18307]) Input IDs shape: torch.Size([1, 18307]) Labels shape: torch.Size([1, 18307]) Final batch size: 1, sequence length: 12006 Attention mask shape: torch.Size([1, 1, 12006, 12006]) Position ids shape: torch.Size([1, 12006]) Input IDs shape: torch.Size([1, 12006]) Labels shape: torch.Size([1, 12006]) Final batch size: 1, sequence length: 22561 Attention mask shape: torch.Size([1, 1, 22561, 22561]) Position ids shape: torch.Size([1, 22561]) Input IDs shape: torch.Size([1, 22561]) Labels shape: torch.Size([1, 22561]) Final batch size: 1, sequence length: 18836 Attention mask shape: torch.Size([1, 1, 18836, 18836]) Position ids shape: torch.Size([1, 18836]) Input IDs shape: torch.Size([1, 18836]) Labels shape: torch.Size([1, 18836]) Final batch size: 1, sequence length: 20854 Attention mask shape: torch.Size([1, 1, 20854, 20854]) Position ids shape: torch.Size([1, 20854]) Input IDs shape: torch.Size([1, 20854]) Labels shape: torch.Size([1, 20854]) Final batch size: 1, sequence length: 19330 Attention mask shape: torch.Size([1, 1, 19330, 19330]) Position ids shape: torch.Size([1, 19330]) Input IDs shape: torch.Size([1, 19330]) Labels shape: torch.Size([1, 19330]) Final batch size: 1, sequence length: 18395 Attention mask shape: torch.Size([1, 1, 18395, 18395]) Position ids shape: torch.Size([1, 18395]) Input IDs shape: torch.Size([1, 18395]) Labels shape: torch.Size([1, 18395]) Final batch size: 1, sequence length: 17058 Attention mask shape: torch.Size([1, 1, 17058, 17058]) Position ids shape: torch.Size([1, 17058]) Input IDs shape: torch.Size([1, 17058]) Labels shape: torch.Size([1, 17058]) Final batch size: 1, sequence length: 20770 Attention mask shape: torch.Size([1, 1, 20770, 20770]) Position ids shape: torch.Size([1, 20770]) Input IDs shape: torch.Size([1, 20770]) Labels shape: torch.Size([1, 20770]) Final batch size: 1, sequence length: 21982 Attention mask shape: torch.Size([1, 1, 21982, 21982]) Position ids shape: torch.Size([1, 21982]) Input IDs shape: torch.Size([1, 21982]) Labels shape: torch.Size([1, 21982]) Final batch size: 1, sequence length: 16060 Attention mask shape: torch.Size([1, 1, 16060, 16060]) Position ids shape: torch.Size([1, 16060]) Input IDs shape: torch.Size([1, 16060]) Labels shape: torch.Size([1, 16060]) Final batch size: 1, sequence length: 20714 Attention mask shape: torch.Size([1, 1, 20714, 20714]) Position ids shape: torch.Size([1, 20714]) Input IDs shape: torch.Size([1, 20714]) Labels shape: torch.Size([1, 20714]) Final batch size: 1, sequence length: 21858 Attention mask shape: torch.Size([1, 1, 21858, 21858]) Position ids shape: torch.Size([1, 21858]) Input IDs shape: torch.Size([1, 21858]) Labels shape: torch.Size([1, 21858]) Final batch size: 1, sequence length: 24248 Attention mask shape: torch.Size([1, 1, 24248, 24248]) Position ids shape: torch.Size([1, 24248]) Input IDs shape: torch.Size([1, 24248]) Labels shape: torch.Size([1, 24248]) Final batch size: 1, sequence length: 18527 Attention mask shape: torch.Size([1, 1, 18527, 18527]) Position ids shape: torch.Size([1, 18527]) Input IDs shape: torch.Size([1, 18527]) Labels shape: torch.Size([1, 18527]) Final batch size: 1, sequence length: 8839 Attention mask shape: torch.Size([1, 1, 8839, 8839]) Position ids shape: torch.Size([1, 8839]) Input IDs shape: torch.Size([1, 8839]) Labels shape: torch.Size([1, 8839]) Final batch size: 1, sequence length: 18823 Attention mask shape: torch.Size([1, 1, 18823, 18823]) Position ids shape: torch.Size([1, 18823]) Input IDs shape: torch.Size([1, 18823]) Labels shape: torch.Size([1, 18823]) Final batch size: 1, sequence length: 15913 Attention mask shape: torch.Size([1, 1, 15913, 15913]) Position ids shape: torch.Size([1, 15913]) Input IDs shape: torch.Size([1, 15913]) Labels shape: torch.Size([1, 15913]) Final batch size: 1, sequence length: 18325 Attention mask shape: torch.Size([1, 1, 18325, 18325]) Position ids shape: torch.Size([1, 18325]) Input IDs shape: torch.Size([1, 18325]) Labels shape: torch.Size([1, 18325]) Final batch size: 1, sequence length: 25451 Attention mask shape: torch.Size([1, 1, 25451, 25451]) Position ids shape: torch.Size([1, 25451]) Input IDs shape: torch.Size([1, 25451]) Labels shape: torch.Size([1, 25451]) Final batch size: 1, sequence length: 23694 Attention mask shape: torch.Size([1, 1, 23694, 23694]) Position ids shape: torch.Size([1, 23694]) Input IDs shape: torch.Size([1, 23694]) Labels shape: torch.Size([1, 23694]) Final batch size: 1, sequence length: 23560 Attention mask shape: torch.Size([1, 1, 23560, 23560]) Position ids shape: torch.Size([1, 23560]) Input IDs shape: torch.Size([1, 23560]) Labels shape: torch.Size([1, 23560]) Final batch size: 1, sequence length: 24433 Attention mask shape: torch.Size([1, 1, 24433, 24433]) Position ids shape: torch.Size([1, 24433]) Input IDs shape: torch.Size([1, 24433]) Labels shape: torch.Size([1, 24433]) Final batch size: 1, sequence length: 26356 Attention mask shape: torch.Size([1, 1, 26356, 26356]) Position ids shape: torch.Size([1, 26356]) Input IDs shape: torch.Size([1, 26356]) Labels shape: torch.Size([1, 26356]) Final batch size: 1, sequence length: 21405 Attention mask shape: torch.Size([1, 1, 21405, 21405]) Position ids shape: torch.Size([1, 21405]) Input IDs shape: torch.Size([1, 21405]) Labels shape: torch.Size([1, 21405]) Final batch size: 1, sequence length: 5801 Attention mask shape: torch.Size([1, 1, 5801, 5801]) Position ids shape: torch.Size([1, 5801]) Input IDs shape: torch.Size([1, 5801]) Labels shape: torch.Size([1, 5801]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 29098 Attention mask shape: torch.Size([1, 1, 29098, 29098]) Position ids shape: torch.Size([1, 29098]) Input IDs shape: torch.Size([1, 29098]) Labels shape: torch.Size([1, 29098]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 9380 Attention mask shape: torch.Size([1, 1, 9380, 9380]) Position ids shape: torch.Size([1, 9380]) Input IDs shape: torch.Size([1, 9380]) Labels shape: torch.Size([1, 9380]) Final batch size: 1, sequence length: 30687 Attention mask shape: torch.Size([1, 1, 30687, 30687]) Position ids shape: torch.Size([1, 30687]) Input IDs shape: torch.Size([1, 30687]) Labels shape: torch.Size([1, 30687]) Final batch size: 1, sequence length: 22735 Attention mask shape: torch.Size([1, 1, 22735, 22735]) Position ids shape: torch.Size([1, 22735]) Input IDs shape: torch.Size([1, 22735]) Labels shape: torch.Size([1, 22735]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 28684 Attention mask shape: torch.Size([1, 1, 28684, 28684]) Position ids shape: torch.Size([1, 28684]) Input IDs shape: torch.Size([1, 28684]) Labels shape: torch.Size([1, 28684]) Final batch size: 1, sequence length: 19045 Attention mask shape: torch.Size([1, 1, 19045, 19045]) Position ids shape: torch.Size([1, 19045]) Input IDs shape: torch.Size([1, 19045]) Labels shape: torch.Size([1, 19045]) Final batch size: 1, sequence length: 15875 Attention mask shape: torch.Size([1, 1, 15875, 15875]) Position ids shape: torch.Size([1, 15875]) Input IDs shape: torch.Size([1, 15875]) Labels shape: torch.Size([1, 15875]) Final batch size: 1, sequence length: 30346 Attention mask shape: torch.Size([1, 1, 30346, 30346]) Position ids shape: torch.Size([1, 30346]) Input IDs shape: torch.Size([1, 30346]) Labels shape: torch.Size([1, 30346]) Final batch size: 1, sequence length: 19239 Attention mask shape: torch.Size([1, 1, 19239, 19239]) Position ids shape: torch.Size([1, 19239]) Input IDs shape: torch.Size([1, 19239]) Labels shape: torch.Size([1, 19239]) Final batch size: 1, sequence length: 31414 Attention mask shape: torch.Size([1, 1, 31414, 31414]) Position ids shape: torch.Size([1, 31414]) Input IDs shape: torch.Size([1, 31414]) Labels shape: torch.Size([1, 31414]) Final batch size: 1, sequence length: 26072 Attention mask shape: torch.Size([1, 1, 26072, 26072]) Position ids shape: torch.Size([1, 26072]) Input IDs shape: torch.Size([1, 26072]) Labels shape: torch.Size([1, 26072]) Final batch size: 1, sequence length: 32328 Attention mask shape: torch.Size([1, 1, 32328, 32328]) Position ids shape: torch.Size([1, 32328]) Input IDs shape: torch.Size([1, 32328]) Labels shape: torch.Size([1, 32328]) Final batch size: 1, sequence length: 32885 Attention mask shape: torch.Size([1, 1, 32885, 32885]) Position ids shape: torch.Size([1, 32885]) Input IDs shape: torch.Size([1, 32885]) Labels shape: torch.Size([1, 32885]) Final batch size: 1, sequence length: 17465 Attention mask shape: torch.Size([1, 1, 17465, 17465]) Position ids shape: torch.Size([1, 17465]) Input IDs shape: torch.Size([1, 17465]) Labels shape: torch.Size([1, 17465]) Final batch size: 1, sequence length: 23881 Attention mask shape: torch.Size([1, 1, 23881, 23881]) Position ids shape: torch.Size([1, 23881]) Input IDs shape: torch.Size([1, 23881]) Labels shape: torch.Size([1, 23881]) Final batch size: 1, sequence length: 31712 Attention mask shape: torch.Size([1, 1, 31712, 31712]) Position ids shape: torch.Size([1, 31712]) Input IDs shape: torch.Size([1, 31712]) Labels shape: torch.Size([1, 31712]) Final batch size: 1, sequence length: 33871 Attention mask shape: torch.Size([1, 1, 33871, 33871]) Position ids shape: torch.Size([1, 33871]) Input IDs shape: torch.Size([1, 33871]) Labels shape: torch.Size([1, 33871]) Final batch size: 1, sequence length: 32529 Attention mask shape: torch.Size([1, 1, 32529, 32529]) Position ids shape: torch.Size([1, 32529]) Input IDs shape: torch.Size([1, 32529]) Labels shape: torch.Size([1, 32529]) Final batch size: 1, sequence length: 29355 Attention mask shape: torch.Size([1, 1, 29355, 29355]) Position ids shape: torch.Size([1, 29355]) Input IDs shape: torch.Size([1, 29355]) Labels shape: torch.Size([1, 29355]) Final batch size: 1, sequence length: 21615 Attention mask shape: torch.Size([1, 1, 21615, 21615]) Position ids shape: torch.Size([1, 21615]) Input IDs shape: torch.Size([1, 21615]) Labels shape: torch.Size([1, 21615]) Final batch size: 1, sequence length: 32636 Attention mask shape: torch.Size([1, 1, 32636, 32636]) Position ids shape: torch.Size([1, 32636]) Input IDs shape: torch.Size([1, 32636]) Labels shape: torch.Size([1, 32636]) Final batch size: 1, sequence length: 27972 Attention mask shape: torch.Size([1, 1, 27972, 27972]) Position ids shape: torch.Size([1, 27972]) Input IDs shape: torch.Size([1, 27972]) Labels shape: torch.Size([1, 27972]) Final batch size: 1, sequence length: 26689 Attention mask shape: torch.Size([1, 1, 26689, 26689]) Position ids shape: torch.Size([1, 26689]) Input IDs shape: torch.Size([1, 26689]) Labels shape: torch.Size([1, 26689]) Final batch size: 1, sequence length: 30689 Attention mask shape: torch.Size([1, 1, 30689, 30689]) Position ids shape: torch.Size([1, 30689]) Input IDs shape: torch.Size([1, 30689]) Labels shape: torch.Size([1, 30689]) Final batch size: 1, sequence length: 35397 Final batch size: 1, sequence length: 15077 Attention mask shape: torch.Size([1, 1, 35397, 35397]) Position ids shape: torch.Size([1, 35397]) Attention mask shape: torch.Size([1, 1, 15077, 15077]) Position ids shape: torch.Size([1, 15077]) Input IDs shape: torch.Size([1, 15077]) Input IDs shape: torch.Size([1, 35397]) Labels shape: torch.Size([1, 35397]) Labels shape: torch.Size([1, 15077]) Final batch size: 1, sequence length: 29150 Attention mask shape: torch.Size([1, 1, 29150, 29150]) Position ids shape: torch.Size([1, 29150]) Input IDs shape: torch.Size([1, 29150]) Labels shape: torch.Size([1, 29150]) Final batch size: 1, sequence length: 31860 Attention mask shape: torch.Size([1, 1, 31860, 31860]) Position ids shape: torch.Size([1, 31860]) Input IDs shape: torch.Size([1, 31860]) Labels shape: torch.Size([1, 31860]) Final batch size: 1, sequence length: 34581 Attention mask shape: torch.Size([1, 1, 34581, 34581]) Position ids shape: torch.Size([1, 34581]) Input IDs shape: torch.Size([1, 34581]) Labels shape: torch.Size([1, 34581]) Final batch size: 1, sequence length: 34921 Attention mask shape: torch.Size([1, 1, 34921, 34921]) Position ids shape: torch.Size([1, 34921]) Input IDs shape: torch.Size([1, 34921]) Labels shape: torch.Size([1, 34921]) Final batch size: 1, sequence length: 22264 Attention mask shape: torch.Size([1, 1, 22264, 22264]) Position ids shape: torch.Size([1, 22264]) Input IDs shape: torch.Size([1, 22264]) Labels shape: torch.Size([1, 22264]) Final batch size: 1, sequence length: 16050 Attention mask shape: torch.Size([1, 1, 16050, 16050]) Position ids shape: torch.Size([1, 16050]) Input IDs shape: torch.Size([1, 16050]) Labels shape: torch.Size([1, 16050]) Final batch size: 1, sequence length: 18963 Attention mask shape: torch.Size([1, 1, 18963, 18963]) Position ids shape: torch.Size([1, 18963]) Input IDs shape: torch.Size([1, 18963]) Labels shape: torch.Size([1, 18963]) Final batch size: 1, sequence length: 24781 Attention mask shape: torch.Size([1, 1, 24781, 24781]) Position ids shape: torch.Size([1, 24781]) Input IDs shape: torch.Size([1, 24781]) Labels shape: torch.Size([1, 24781]) Final batch size: 1, sequence length: 18177 Attention mask shape: torch.Size([1, 1, 18177, 18177]) Position ids shape: torch.Size([1, 18177]) Input IDs shape: torch.Size([1, 18177]) Labels shape: torch.Size([1, 18177]) Final batch size: 1, sequence length: 24002 Attention mask shape: torch.Size([1, 1, 24002, 24002]) Position ids shape: torch.Size([1, 24002]) Input IDs shape: torch.Size([1, 24002]) Labels shape: torch.Size([1, 24002]) Final batch size: 1, sequence length: 16025 Attention mask shape: torch.Size([1, 1, 16025, 16025]) Position ids shape: torch.Size([1, 16025]) Input IDs shape: torch.Size([1, 16025]) Labels shape: torch.Size([1, 16025]) Final batch size: 1, sequence length: 25741 Attention mask shape: torch.Size([1, 1, 25741, 25741]) Position ids shape: torch.Size([1, 25741]) Input IDs shape: torch.Size([1, 25741]) Labels shape: torch.Size([1, 25741]) Final batch size: 1, sequence length: 14976 Attention mask shape: torch.Size([1, 1, 14976, 14976]) Position ids shape: torch.Size([1, 14976]) Input IDs shape: torch.Size([1, 14976]) Labels shape: torch.Size([1, 14976]) Final batch size: 1, sequence length: 17243 Attention mask shape: torch.Size([1, 1, 17243, 17243]) Position ids shape: torch.Size([1, 17243]) Input IDs shape: torch.Size([1, 17243]) Labels shape: torch.Size([1, 17243]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31022 Attention mask shape: torch.Size([1, 1, 31022, 31022]) Position ids shape: torch.Size([1, 31022]) Input IDs shape: torch.Size([1, 31022]) Labels shape: torch.Size([1, 31022]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25674 Attention mask shape: torch.Size([1, 1, 25674, 25674]) Position ids shape: torch.Size([1, 25674]) Input IDs shape: torch.Size([1, 25674]) Labels shape: torch.Size([1, 25674]) Final batch size: 1, sequence length: 28397 Attention mask shape: torch.Size([1, 1, 28397, 28397]) Position ids shape: torch.Size([1, 28397]) Input IDs shape: torch.Size([1, 28397]) Labels shape: torch.Size([1, 28397]) Final batch size: 1, sequence length: 28018 Attention mask shape: torch.Size([1, 1, 28018, 28018]) Position ids shape: torch.Size([1, 28018]) Input IDs shape: torch.Size([1, 28018]) Labels shape: torch.Size([1, 28018]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40488 Attention mask shape: torch.Size([1, 1, 40488, 40488]) Position ids shape: torch.Size([1, 40488]) Input IDs shape: torch.Size([1, 40488]) Labels shape: torch.Size([1, 40488]) Final batch size: 1, sequence length: 26922 Attention mask shape: torch.Size([1, 1, 26922, 26922]) Position ids shape: torch.Size([1, 26922]) Input IDs shape: torch.Size([1, 26922]) Labels shape: torch.Size([1, 26922]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36596 Attention mask shape: torch.Size([1, 1, 36596, 36596]) Position ids shape: torch.Size([1, 36596]) Input IDs shape: torch.Size([1, 36596]) Labels shape: torch.Size([1, 36596]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 11611 Attention mask shape: torch.Size([1, 1, 11611, 11611]) Position ids shape: torch.Size([1, 11611]) Input IDs shape: torch.Size([1, 11611]) Labels shape: torch.Size([1, 11611]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24043 Attention mask shape: torch.Size([1, 1, 24043, 24043]) Position ids shape: torch.Size([1, 24043]) Input IDs shape: torch.Size([1, 24043]) Labels shape: torch.Size([1, 24043]) Final batch size: 1, sequence length: 34853 Attention mask shape: torch.Size([1, 1, 34853, 34853]) Position ids shape: torch.Size([1, 34853]) Input IDs shape: torch.Size([1, 34853]) Labels shape: torch.Size([1, 34853]) Final batch size: 1, sequence length: 28086 Attention mask shape: torch.Size([1, 1, 28086, 28086]) Position ids shape: torch.Size([1, 28086]) Input IDs shape: torch.Size([1, 28086]) Labels shape: torch.Size([1, 28086]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16337 Attention mask shape: torch.Size([1, 1, 16337, 16337]) Position ids shape: torch.Size([1, 16337]) Input IDs shape: torch.Size([1, 16337]) Labels shape: torch.Size([1, 16337]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 12218 Attention mask shape: torch.Size([1, 1, 12218, 12218]) Position ids shape: torch.Size([1, 12218]) Input IDs shape: torch.Size([1, 12218]) Labels shape: torch.Size([1, 12218]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36334 Attention mask shape: torch.Size([1, 1, 36334, 36334]) Position ids shape: torch.Size([1, 36334]) Input IDs shape: torch.Size([1, 36334]) Labels shape: torch.Size([1, 36334]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32917 Attention mask shape: torch.Size([1, 1, 32917, 32917]) Position ids shape: torch.Size([1, 32917]) Input IDs shape: torch.Size([1, 32917]) Labels shape: torch.Size([1, 32917]) {'loss': 0.4335, 'grad_norm': 1.9508074001444418, 'learning_rate': 5e-06, 'num_tokens': -inf, 'epoch': 0.38} Final batch size: 1, sequence length: 5818 Attention mask shape: torch.Size([1, 1, 5818, 5818]) Position ids shape: torch.Size([1, 5818]) Input IDs shape: torch.Size([1, 5818]) Labels shape: torch.Size([1, 5818]) Final batch size: 1, sequence length: 6215 Attention mask shape: torch.Size([1, 1, 6215, 6215]) Position ids shape: torch.Size([1, 6215]) Input IDs shape: torch.Size([1, 6215]) Labels shape: torch.Size([1, 6215]) Final batch size: 1, sequence length: 6871 Attention mask shape: torch.Size([1, 1, 6871, 6871]) Position ids shape: torch.Size([1, 6871]) Input IDs shape: torch.Size([1, 6871]) Labels shape: torch.Size([1, 6871]) Final batch size: 1, sequence length: 8623 Attention mask shape: torch.Size([1, 1, 8623, 8623]) Position ids shape: torch.Size([1, 8623]) Input IDs shape: torch.Size([1, 8623]) Labels shape: torch.Size([1, 8623]) Final batch size: 1, sequence length: 5917 Attention mask shape: torch.Size([1, 1, 5917, 5917]) Position ids shape: torch.Size([1, 5917]) Input IDs shape: torch.Size([1, 5917]) Labels shape: torch.Size([1, 5917]) Final batch size: 1, sequence length: 6034 Attention mask shape: torch.Size([1, 1, 6034, 6034]) Position ids shape: torch.Size([1, 6034]) Input IDs shape: torch.Size([1, 6034]) Labels shape: torch.Size([1, 6034]) Final batch size: 1, sequence length: 11616 Attention mask shape: torch.Size([1, 1, 11616, 11616]) Position ids shape: torch.Size([1, 11616]) Input IDs shape: torch.Size([1, 11616]) Labels shape: torch.Size([1, 11616]) Final batch size: 1, sequence length: 12275 Attention mask shape: torch.Size([1, 1, 12275, 12275]) Position ids shape: torch.Size([1, 12275]) Input IDs shape: torch.Size([1, 12275]) Labels shape: torch.Size([1, 12275]) Final batch size: 1, sequence length: 12317 Attention mask shape: torch.Size([1, 1, 12317, 12317]) Position ids shape: torch.Size([1, 12317]) Input IDs shape: torch.Size([1, 12317]) Labels shape: torch.Size([1, 12317]) Final batch size: 1, sequence length: 13202 Attention mask shape: torch.Size([1, 1, 13202, 13202]) Position ids shape: torch.Size([1, 13202]) Input IDs shape: torch.Size([1, 13202]) Labels shape: torch.Size([1, 13202]) Final batch size: 1, sequence length: 10107 Attention mask shape: torch.Size([1, 1, 10107, 10107]) Position ids shape: torch.Size([1, 10107]) Input IDs shape: torch.Size([1, 10107]) Labels shape: torch.Size([1, 10107]) Final batch size: 1, sequence length: 9517 Attention mask shape: torch.Size([1, 1, 9517, 9517]) Position ids shape: torch.Size([1, 9517]) Input IDs shape: torch.Size([1, 9517]) Labels shape: torch.Size([1, 9517]) Final batch size: 1, sequence length: 11648 Attention mask shape: torch.Size([1, 1, 11648, 11648]) Position ids shape: torch.Size([1, 11648]) Input IDs shape: torch.Size([1, 11648]) Labels shape: torch.Size([1, 11648]) Final batch size: 1, sequence length: 8117 Attention mask shape: torch.Size([1, 1, 8117, 8117]) Position ids shape: torch.Size([1, 8117]) Input IDs shape: torch.Size([1, 8117]) Labels shape: torch.Size([1, 8117]) Final batch size: 1, sequence length: 12813 Attention mask shape: torch.Size([1, 1, 12813, 12813]) Position ids shape: torch.Size([1, 12813]) Input IDs shape: torch.Size([1, 12813]) Labels shape: torch.Size([1, 12813]) Final batch size: 1, sequence length: 16331 Attention mask shape: torch.Size([1, 1, 16331, 16331]) Position ids shape: torch.Size([1, 16331]) Input IDs shape: torch.Size([1, 16331]) Labels shape: torch.Size([1, 16331]) Final batch size: 1, sequence length: 13428 Attention mask shape: torch.Size([1, 1, 13428, 13428]) Position ids shape: torch.Size([1, 13428]) Input IDs shape: torch.Size([1, 13428]) Labels shape: torch.Size([1, 13428]) Final batch size: 1, sequence length: 15066 Attention mask shape: torch.Size([1, 1, 15066, 15066]) Position ids shape: torch.Size([1, 15066]) Input IDs shape: torch.Size([1, 15066]) Labels shape: torch.Size([1, 15066]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 10136 Attention mask shape: torch.Size([1, 1, 10136, 10136]) Position ids shape: torch.Size([1, 10136]) Input IDs shape: torch.Size([1, 10136]) Labels shape: torch.Size([1, 10136]) Final batch size: 1, sequence length: 12553 Attention mask shape: torch.Size([1, 1, 12553, 12553]) Position ids shape: torch.Size([1, 12553]) Input IDs shape: torch.Size([1, 12553]) Labels shape: torch.Size([1, 12553]) Final batch size: 1, sequence length: 19847 Attention mask shape: torch.Size([1, 1, 19847, 19847]) Position ids shape: torch.Size([1, 19847]) Input IDs shape: torch.Size([1, 19847]) Labels shape: torch.Size([1, 19847]) Final batch size: 1, sequence length: 15499 Attention mask shape: torch.Size([1, 1, 15499, 15499]) Position ids shape: torch.Size([1, 15499]) Input IDs shape: torch.Size([1, 15499]) Labels shape: torch.Size([1, 15499]) Final batch size: 1, sequence length: 19221 Attention mask shape: torch.Size([1, 1, 19221, 19221]) Position ids shape: torch.Size([1, 19221]) Input IDs shape: torch.Size([1, 19221]) Labels shape: torch.Size([1, 19221]) Final batch size: 1, sequence length: 18414 Attention mask shape: torch.Size([1, 1, 18414, 18414]) Position ids shape: torch.Size([1, 18414]) Input IDs shape: torch.Size([1, 18414]) Labels shape: torch.Size([1, 18414]) Final batch size: 1, sequence length: 18469 Attention mask shape: torch.Size([1, 1, 18469, 18469]) Position ids shape: torch.Size([1, 18469]) Input IDs shape: torch.Size([1, 18469]) Labels shape: torch.Size([1, 18469]) Final batch size: 1, sequence length: 15226 Attention mask shape: torch.Size([1, 1, 15226, 15226]) Position ids shape: torch.Size([1, 15226]) Input IDs shape: torch.Size([1, 15226]) Labels shape: torch.Size([1, 15226]) Final batch size: 1, sequence length: 15029 Attention mask shape: torch.Size([1, 1, 15029, 15029]) Position ids shape: torch.Size([1, 15029]) Input IDs shape: torch.Size([1, 15029]) Labels shape: torch.Size([1, 15029]) Final batch size: 1, sequence length: 21556 Attention mask shape: torch.Size([1, 1, 21556, 21556]) Position ids shape: torch.Size([1, 21556]) Input IDs shape: torch.Size([1, 21556]) Labels shape: torch.Size([1, 21556]) Final batch size: 1, sequence length: 21028 Attention mask shape: torch.Size([1, 1, 21028, 21028]) Position ids shape: torch.Size([1, 21028]) Input IDs shape: torch.Size([1, 21028]) Labels shape: torch.Size([1, 21028]) Final batch size: 1, sequence length: 18438 Attention mask shape: torch.Size([1, 1, 18438, 18438]) Position ids shape: torch.Size([1, 18438]) Input IDs shape: torch.Size([1, 18438]) Labels shape: torch.Size([1, 18438]) Final batch size: 1, sequence length: 16299 Attention mask shape: torch.Size([1, 1, 16299, 16299]) Position ids shape: torch.Size([1, 16299]) Input IDs shape: torch.Size([1, 16299]) Labels shape: torch.Size([1, 16299]) Final batch size: 1, sequence length: 17587 Attention mask shape: torch.Size([1, 1, 17587, 17587]) Position ids shape: torch.Size([1, 17587]) Input IDs shape: torch.Size([1, 17587]) Labels shape: torch.Size([1, 17587]) Final batch size: 1, sequence length: 19512 Attention mask shape: torch.Size([1, 1, 19512, 19512]) Position ids shape: torch.Size([1, 19512]) Input IDs shape: torch.Size([1, 19512]) Labels shape: torch.Size([1, 19512]) Final batch size: 1, sequence length: 22079 Attention mask shape: torch.Size([1, 1, 22079, 22079]) Position ids shape: torch.Size([1, 22079]) Input IDs shape: torch.Size([1, 22079]) Labels shape: torch.Size([1, 22079]) Final batch size: 1, sequence length: 22618 Attention mask shape: torch.Size([1, 1, 22618, 22618]) Position ids shape: torch.Size([1, 22618]) Input IDs shape: torch.Size([1, 22618]) Labels shape: torch.Size([1, 22618]) Final batch size: 1, sequence length: 22718 Attention mask shape: torch.Size([1, 1, 22718, 22718]) Position ids shape: torch.Size([1, 22718]) Input IDs shape: torch.Size([1, 22718]) Labels shape: torch.Size([1, 22718]) Final batch size: 1, sequence length: 22138 Attention mask shape: torch.Size([1, 1, 22138, 22138]) Position ids shape: torch.Size([1, 22138]) Input IDs shape: torch.Size([1, 22138]) Labels shape: torch.Size([1, 22138]) Final batch size: 1, sequence length: 23973 Attention mask shape: torch.Size([1, 1, 23973, 23973]) Position ids shape: torch.Size([1, 23973]) Input IDs shape: torch.Size([1, 23973]) Labels shape: torch.Size([1, 23973]) Final batch size: 1, sequence length: 25622 Attention mask shape: torch.Size([1, 1, 25622, 25622]) Position ids shape: torch.Size([1, 25622]) Input IDs shape: torch.Size([1, 25622]) Labels shape: torch.Size([1, 25622]) Final batch size: 1, sequence length: 7584 Attention mask shape: torch.Size([1, 1, 7584, 7584]) Position ids shape: torch.Size([1, 7584]) Input IDs shape: torch.Size([1, 7584]) Labels shape: torch.Size([1, 7584]) Final batch size: 1, sequence length: 24694 Attention mask shape: torch.Size([1, 1, 24694, 24694]) Position ids shape: torch.Size([1, 24694]) Input IDs shape: torch.Size([1, 24694]) Labels shape: torch.Size([1, 24694]) Final batch size: 1, sequence length: 14784 Attention mask shape: torch.Size([1, 1, 14784, 14784]) Position ids shape: torch.Size([1, 14784]) Input IDs shape: torch.Size([1, 14784]) Labels shape: torch.Size([1, 14784]) Final batch size: 1, sequence length: 19428 Attention mask shape: torch.Size([1, 1, 19428, 19428]) Position ids shape: torch.Size([1, 19428]) Input IDs shape: torch.Size([1, 19428]) Labels shape: torch.Size([1, 19428]) Final batch size: 1, sequence length: 22763 Attention mask shape: torch.Size([1, 1, 22763, 22763]) Position ids shape: torch.Size([1, 22763]) Input IDs shape: torch.Size([1, 22763]) Labels shape: torch.Size([1, 22763]) Final batch size: 1, sequence length: 25999 Attention mask shape: torch.Size([1, 1, 25999, 25999]) Position ids shape: torch.Size([1, 25999]) Input IDs shape: torch.Size([1, 25999]) Labels shape: torch.Size([1, 25999]) Final batch size: 1, sequence length: 24909 Attention mask shape: torch.Size([1, 1, 24909, 24909]) Position ids shape: torch.Size([1, 24909]) Input IDs shape: torch.Size([1, 24909]) Labels shape: torch.Size([1, 24909]) Final batch size: 1, sequence length: 27566 Attention mask shape: torch.Size([1, 1, 27566, 27566]) Position ids shape: torch.Size([1, 27566]) Input IDs shape: torch.Size([1, 27566]) Labels shape: torch.Size([1, 27566]) Final batch size: 1, sequence length: 28497 Attention mask shape: torch.Size([1, 1, 28497, 28497]) Position ids shape: torch.Size([1, 28497]) Input IDs shape: torch.Size([1, 28497]) Labels shape: torch.Size([1, 28497]) Final batch size: 1, sequence length: 27327 Attention mask shape: torch.Size([1, 1, 27327, 27327]) Position ids shape: torch.Size([1, 27327]) Input IDs shape: torch.Size([1, 27327]) Labels shape: torch.Size([1, 27327]) Final batch size: 1, sequence length: 22235 Attention mask shape: torch.Size([1, 1, 22235, 22235]) Position ids shape: torch.Size([1, 22235]) Input IDs shape: torch.Size([1, 22235]) Labels shape: torch.Size([1, 22235]) Final batch size: 1, sequence length: 23766 Attention mask shape: torch.Size([1, 1, 23766, 23766]) Position ids shape: torch.Size([1, 23766]) Input IDs shape: torch.Size([1, 23766]) Labels shape: torch.Size([1, 23766]) Final batch size: 1, sequence length: 10269 Attention mask shape: torch.Size([1, 1, 10269, 10269]) Position ids shape: torch.Size([1, 10269]) Input IDs shape: torch.Size([1, 10269]) Labels shape: torch.Size([1, 10269]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 15184 Attention mask shape: torch.Size([1, 1, 15184, 15184]) Position ids shape: torch.Size([1, 15184]) Input IDs shape: torch.Size([1, 15184]) Labels shape: torch.Size([1, 15184]) Final batch size: 1, sequence length: 28749 Attention mask shape: torch.Size([1, 1, 28749, 28749]) Position ids shape: torch.Size([1, 28749]) Input IDs shape: torch.Size([1, 28749]) Labels shape: torch.Size([1, 28749]) Final batch size: 1, sequence length: 15588 Attention mask shape: torch.Size([1, 1, 15588, 15588]) Position ids shape: torch.Size([1, 15588]) Input IDs shape: torch.Size([1, 15588]) Labels shape: torch.Size([1, 15588]) Final batch size: 1, sequence length: 15221 Attention mask shape: torch.Size([1, 1, 15221, 15221]) Position ids shape: torch.Size([1, 15221]) Input IDs shape: torch.Size([1, 15221]) Labels shape: torch.Size([1, 15221]) Final batch size: 1, sequence length: 24499 Attention mask shape: torch.Size([1, 1, 24499, 24499]) Position ids shape: torch.Size([1, 24499]) Input IDs shape: torch.Size([1, 24499]) Labels shape: torch.Size([1, 24499]) Final batch size: 1, sequence length: 21826 Attention mask shape: torch.Size([1, 1, 21826, 21826]) Position ids shape: torch.Size([1, 21826]) Input IDs shape: torch.Size([1, 21826]) Labels shape: torch.Size([1, 21826]) Final batch size: 1, sequence length: 23851 Attention mask shape: torch.Size([1, 1, 23851, 23851]) Position ids shape: torch.Size([1, 23851]) Input IDs shape: torch.Size([1, 23851]) Labels shape: torch.Size([1, 23851]) Final batch size: 1, sequence length: 30356 Attention mask shape: torch.Size([1, 1, 30356, 30356]) Position ids shape: torch.Size([1, 30356]) Input IDs shape: torch.Size([1, 30356]) Labels shape: torch.Size([1, 30356]) Final batch size: 1, sequence length: 11795 Attention mask shape: torch.Size([1, 1, 11795, 11795]) Position ids shape: torch.Size([1, 11795]) Input IDs shape: torch.Size([1, 11795]) Labels shape: torch.Size([1, 11795]) Final batch size: 1, sequence length: 21034 Attention mask shape: torch.Size([1, 1, 21034, 21034]) Position ids shape: torch.Size([1, 21034]) Input IDs shape: torch.Size([1, 21034]) Labels shape: torch.Size([1, 21034]) Final batch size: 1, sequence length: 34673 Attention mask shape: torch.Size([1, 1, 34673, 34673]) Position ids shape: torch.Size([1, 34673]) Input IDs shape: torch.Size([1, 34673]) Labels shape: torch.Size([1, 34673]) Final batch size: 1, sequence length: 36777 Attention mask shape: torch.Size([1, 1, 36777, 36777]) Position ids shape: torch.Size([1, 36777]) Input IDs shape: torch.Size([1, 36777]) Labels shape: torch.Size([1, 36777]) Final batch size: 1, sequence length: 20755 Attention mask shape: torch.Size([1, 1, 20755, 20755]) Position ids shape: torch.Size([1, 20755]) Input IDs shape: torch.Size([1, 20755]) Labels shape: torch.Size([1, 20755]) Final batch size: 1, sequence length: 27489 Attention mask shape: torch.Size([1, 1, 27489, 27489]) Position ids shape: torch.Size([1, 27489]) Input IDs shape: torch.Size([1, 27489]) Labels shape: torch.Size([1, 27489]) Final batch size: 1, sequence length: 36723 Attention mask shape: torch.Size([1, 1, 36723, 36723]) Position ids shape: torch.Size([1, 36723]) Input IDs shape: torch.Size([1, 36723]) Labels shape: torch.Size([1, 36723]) Final batch size: 1, sequence length: 40325 Attention mask shape: torch.Size([1, 1, 40325, 40325]) Position ids shape: torch.Size([1, 40325]) Input IDs shape: torch.Size([1, 40325]) Labels shape: torch.Size([1, 40325]) Final batch size: 1, sequence length: 38848 Attention mask shape: torch.Size([1, 1, 38848, 38848]) Position ids shape: torch.Size([1, 38848]) Input IDs shape: torch.Size([1, 38848]) Labels shape: torch.Size([1, 38848]) Final batch size: 1, sequence length: 28176 Attention mask shape: torch.Size([1, 1, 28176, 28176]) Position ids shape: torch.Size([1, 28176]) Input IDs shape: torch.Size([1, 28176]) Labels shape: torch.Size([1, 28176]) Final batch size: 1, sequence length: 39253 Attention mask shape: torch.Size([1, 1, 39253, 39253]) Position ids shape: torch.Size([1, 39253]) Input IDs shape: torch.Size([1, 39253]) Labels shape: torch.Size([1, 39253]) Final batch size: 1, sequence length: 25946 Attention mask shape: torch.Size([1, 1, 25946, 25946]) Position ids shape: torch.Size([1, 25946]) Input IDs shape: torch.Size([1, 25946]) Labels shape: torch.Size([1, 25946]) Final batch size: 1, sequence length: 6882 Attention mask shape: torch.Size([1, 1, 6882, 6882]) Position ids shape: torch.Size([1, 6882]) Input IDs shape: torch.Size([1, 6882]) Labels shape: torch.Size([1, 6882]) Final batch size: 1, sequence length: 38104 Attention mask shape: torch.Size([1, 1, 38104, 38104]) Position ids shape: torch.Size([1, 38104]) Input IDs shape: torch.Size([1, 38104]) Labels shape: torch.Size([1, 38104]) Final batch size: 1, sequence length: 35904 Attention mask shape: torch.Size([1, 1, 35904, 35904]) Position ids shape: torch.Size([1, 35904]) Input IDs shape: torch.Size([1, 35904]) Labels shape: torch.Size([1, 35904]) Final batch size: 1, sequence length: 31903 Attention mask shape: torch.Size([1, 1, 31903, 31903]) Position ids shape: torch.Size([1, 31903]) Input IDs shape: torch.Size([1, 31903]) Labels shape: torch.Size([1, 31903]) Final batch size: 1, sequence length: 22366 Attention mask shape: torch.Size([1, 1, 22366, 22366]) Position ids shape: torch.Size([1, 22366]) Input IDs shape: torch.Size([1, 22366]) Labels shape: torch.Size([1, 22366]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34289 Attention mask shape: torch.Size([1, 1, 34289, 34289]) Position ids shape: torch.Size([1, 34289]) Input IDs shape: torch.Size([1, 34289]) Labels shape: torch.Size([1, 34289]) Final batch size: 1, sequence length: 10469 Attention mask shape: torch.Size([1, 1, 10469, 10469]) Position ids shape: torch.Size([1, 10469]) Input IDs shape: torch.Size([1, 10469]) Labels shape: torch.Size([1, 10469]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38049 Attention mask shape: torch.Size([1, 1, 38049, 38049]) Position ids shape: torch.Size([1, 38049]) Input IDs shape: torch.Size([1, 38049]) Labels shape: torch.Size([1, 38049]) Final batch size: 1, sequence length: 36786 Attention mask shape: torch.Size([1, 1, 36786, 36786]) Position ids shape: torch.Size([1, 36786]) Input IDs shape: torch.Size([1, 36786]) Labels shape: torch.Size([1, 36786]) Final batch size: 1, sequence length: 40317 Attention mask shape: torch.Size([1, 1, 40317, 40317]) Position ids shape: torch.Size([1, 40317]) Input IDs shape: torch.Size([1, 40317]) Labels shape: torch.Size([1, 40317]) Final batch size: 1, sequence length: 37285 Attention mask shape: torch.Size([1, 1, 37285, 37285]) Position ids shape: torch.Size([1, 37285]) Input IDs shape: torch.Size([1, 37285]) Labels shape: torch.Size([1, 37285]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24801 Attention mask shape: torch.Size([1, 1, 24801, 24801]) Position ids shape: torch.Size([1, 24801]) Input IDs shape: torch.Size([1, 24801]) Labels shape: torch.Size([1, 24801]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31714 Attention mask shape: torch.Size([1, 1, 31714, 31714]) Position ids shape: torch.Size([1, 31714]) Input IDs shape: torch.Size([1, 31714]) Labels shape: torch.Size([1, 31714]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16649 Attention mask shape: torch.Size([1, 1, 16649, 16649]) Position ids shape: torch.Size([1, 16649]) Input IDs shape: torch.Size([1, 16649]) Labels shape: torch.Size([1, 16649]) Final batch size: 1, sequence length: 19437 Attention mask shape: torch.Size([1, 1, 19437, 19437]) Position ids shape: torch.Size([1, 19437]) Input IDs shape: torch.Size([1, 19437]) Labels shape: torch.Size([1, 19437]) Final batch size: 1, sequence length: 40937 Attention mask shape: torch.Size([1, 1, 40937, 40937]) Position ids shape: torch.Size([1, 40937]) Input IDs shape: torch.Size([1, 40937]) Labels shape: torch.Size([1, 40937]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36128 Attention mask shape: torch.Size([1, 1, 36128, 36128]) Position ids shape: torch.Size([1, 36128]) Input IDs shape: torch.Size([1, 36128]) Labels shape: torch.Size([1, 36128]) Final batch size: 1, sequence length: 40657 Attention mask shape: torch.Size([1, 1, 40657, 40657]) Position ids shape: torch.Size([1, 40657]) Input IDs shape: torch.Size([1, 40657]) Labels shape: torch.Size([1, 40657]) Final batch size: 1, sequence length: 21547 Attention mask shape: torch.Size([1, 1, 21547, 21547]) Position ids shape: torch.Size([1, 21547]) Input IDs shape: torch.Size([1, 21547]) Labels shape: torch.Size([1, 21547]) Final batch size: 1, sequence length: 38686 Attention mask shape: torch.Size([1, 1, 38686, 38686]) Position ids shape: torch.Size([1, 38686]) Input IDs shape: torch.Size([1, 38686]) Labels shape: torch.Size([1, 38686]) Final batch size: 1, sequence length: 19639 Attention mask shape: torch.Size([1, 1, 19639, 19639]) Position ids shape: torch.Size([1, 19639]) Input IDs shape: torch.Size([1, 19639]) Labels shape: torch.Size([1, 19639]) Final batch size: 1, sequence length: 25963 Attention mask shape: torch.Size([1, 1, 25963, 25963]) Position ids shape: torch.Size([1, 25963]) Input IDs shape: torch.Size([1, 25963]) Labels shape: torch.Size([1, 25963]) Final batch size: 1, sequence length: 24551 Attention mask shape: torch.Size([1, 1, 24551, 24551]) Position ids shape: torch.Size([1, 24551]) Input IDs shape: torch.Size([1, 24551]) Labels shape: torch.Size([1, 24551]) Final batch size: 1, sequence length: 34540 Attention mask shape: torch.Size([1, 1, 34540, 34540]) Position ids shape: torch.Size([1, 34540]) Input IDs shape: torch.Size([1, 34540]) Labels shape: torch.Size([1, 34540]) Final batch size: 1, sequence length: 17882 Attention mask shape: torch.Size([1, 1, 17882, 17882]) Position ids shape: torch.Size([1, 17882]) Input IDs shape: torch.Size([1, 17882]) Labels shape: torch.Size([1, 17882]) Final batch size: 1, sequence length: 17379 Attention mask shape: torch.Size([1, 1, 17379, 17379]) Position ids shape: torch.Size([1, 17379]) Input IDs shape: torch.Size([1, 17379]) Labels shape: torch.Size([1, 17379]) Final batch size: 1, sequence length: 27947 Attention mask shape: torch.Size([1, 1, 27947, 27947]) Position ids shape: torch.Size([1, 27947]) Input IDs shape: torch.Size([1, 27947]) Labels shape: torch.Size([1, 27947]) Final batch size: 1, sequence length: 26660 Attention mask shape: torch.Size([1, 1, 26660, 26660]) Position ids shape: torch.Size([1, 26660]) Input IDs shape: torch.Size([1, 26660]) Labels shape: torch.Size([1, 26660]) Final batch size: 1, sequence length: 21496 Attention mask shape: torch.Size([1, 1, 21496, 21496]) Position ids shape: torch.Size([1, 21496]) Input IDs shape: torch.Size([1, 21496]) Labels shape: torch.Size([1, 21496]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20843 Attention mask shape: torch.Size([1, 1, 20843, 20843]) Position ids shape: torch.Size([1, 20843]) Input IDs shape: torch.Size([1, 20843]) Labels shape: torch.Size([1, 20843]) Final batch size: 1, sequence length: 37946 Attention mask shape: torch.Size([1, 1, 37946, 37946]) Position ids shape: torch.Size([1, 37946]) Input IDs shape: torch.Size([1, 37946]) Labels shape: torch.Size([1, 37946]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36317 Attention mask shape: torch.Size([1, 1, 36317, 36317]) Position ids shape: torch.Size([1, 36317]) Input IDs shape: torch.Size([1, 36317]) Labels shape: torch.Size([1, 36317]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7681 Attention mask shape: torch.Size([1, 1, 7681, 7681]) Position ids shape: torch.Size([1, 7681]) Input IDs shape: torch.Size([1, 7681]) Labels shape: torch.Size([1, 7681]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.4325, 'grad_norm': 1.7246965281405975, 'learning_rate': 7.500000000000001e-06, 'num_tokens': -inf, 'epoch': 0.5} Final batch size: 1, sequence length: 5328 Attention mask shape: torch.Size([1, 1, 5328, 5328]) Position ids shape: torch.Size([1, 5328]) Input IDs shape: torch.Size([1, 5328]) Labels shape: torch.Size([1, 5328]) Final batch size: 1, sequence length: 5525 Attention mask shape: torch.Size([1, 1, 5525, 5525]) Position ids shape: torch.Size([1, 5525]) Input IDs shape: torch.Size([1, 5525]) Labels shape: torch.Size([1, 5525]) Final batch size: 1, sequence length: 11464 Attention mask shape: torch.Size([1, 1, 11464, 11464]) Position ids shape: torch.Size([1, 11464]) Input IDs shape: torch.Size([1, 11464]) Labels shape: torch.Size([1, 11464]) Final batch size: 1, sequence length: 10273 Attention mask shape: torch.Size([1, 1, 10273, 10273]) Position ids shape: torch.Size([1, 10273]) Input IDs shape: torch.Size([1, 10273]) Labels shape: torch.Size([1, 10273]) Final batch size: 1, sequence length: 12419 Attention mask shape: torch.Size([1, 1, 12419, 12419]) Position ids shape: torch.Size([1, 12419]) Input IDs shape: torch.Size([1, 12419]) Labels shape: torch.Size([1, 12419]) Final batch size: 1, sequence length: 10434 Attention mask shape: torch.Size([1, 1, 10434, 10434]) Position ids shape: torch.Size([1, 10434]) Input IDs shape: torch.Size([1, 10434]) Labels shape: torch.Size([1, 10434]) Final batch size: 1, sequence length: 13607 Attention mask shape: torch.Size([1, 1, 13607, 13607]) Position ids shape: torch.Size([1, 13607]) Input IDs shape: torch.Size([1, 13607]) Labels shape: torch.Size([1, 13607]) Final batch size: 1, sequence length: 13657 Attention mask shape: torch.Size([1, 1, 13657, 13657]) Position ids shape: torch.Size([1, 13657]) Input IDs shape: torch.Size([1, 13657]) Labels shape: torch.Size([1, 13657]) Final batch size: 1, sequence length: 14226 Attention mask shape: torch.Size([1, 1, 14226, 14226]) Position ids shape: torch.Size([1, 14226]) Input IDs shape: torch.Size([1, 14226]) Labels shape: torch.Size([1, 14226]) Final batch size: 1, sequence length: 9648 Attention mask shape: torch.Size([1, 1, 9648, 9648]) Position ids shape: torch.Size([1, 9648]) Input IDs shape: torch.Size([1, 9648]) Labels shape: torch.Size([1, 9648]) Final batch size: 1, sequence length: 11225 Attention mask shape: torch.Size([1, 1, 11225, 11225]) Position ids shape: torch.Size([1, 11225]) Input IDs shape: torch.Size([1, 11225]) Labels shape: torch.Size([1, 11225]) Final batch size: 1, sequence length: 16398 Attention mask shape: torch.Size([1, 1, 16398, 16398]) Position ids shape: torch.Size([1, 16398]) Input IDs shape: torch.Size([1, 16398]) Labels shape: torch.Size([1, 16398]) Final batch size: 1, sequence length: 16756 Attention mask shape: torch.Size([1, 1, 16756, 16756]) Position ids shape: torch.Size([1, 16756]) Input IDs shape: torch.Size([1, 16756]) Labels shape: torch.Size([1, 16756]) Final batch size: 1, sequence length: 14360 Attention mask shape: torch.Size([1, 1, 14360, 14360]) Position ids shape: torch.Size([1, 14360]) Input IDs shape: torch.Size([1, 14360]) Labels shape: torch.Size([1, 14360]) Final batch size: 1, sequence length: 16223 Attention mask shape: torch.Size([1, 1, 16223, 16223]) Position ids shape: torch.Size([1, 16223]) Input IDs shape: torch.Size([1, 16223]) Labels shape: torch.Size([1, 16223]) Final batch size: 1, sequence length: 18408 Attention mask shape: torch.Size([1, 1, 18408, 18408]) Position ids shape: torch.Size([1, 18408]) Input IDs shape: torch.Size([1, 18408]) Labels shape: torch.Size([1, 18408]) Final batch size: 1, sequence length: 16053 Attention mask shape: torch.Size([1, 1, 16053, 16053]) Position ids shape: torch.Size([1, 16053]) Input IDs shape: torch.Size([1, 16053]) Labels shape: torch.Size([1, 16053]) Final batch size: 1, sequence length: 15243 Attention mask shape: torch.Size([1, 1, 15243, 15243]) Position ids shape: torch.Size([1, 15243]) Input IDs shape: torch.Size([1, 15243]) Labels shape: torch.Size([1, 15243]) Final batch size: 1, sequence length: 17294 Attention mask shape: torch.Size([1, 1, 17294, 17294]) Position ids shape: torch.Size([1, 17294]) Input IDs shape: torch.Size([1, 17294]) Labels shape: torch.Size([1, 17294]) Final batch size: 1, sequence length: 20433 Attention mask shape: torch.Size([1, 1, 20433, 20433]) Position ids shape: torch.Size([1, 20433]) Input IDs shape: torch.Size([1, 20433]) Labels shape: torch.Size([1, 20433]) Final batch size: 1, sequence length: 21455 Attention mask shape: torch.Size([1, 1, 21455, 21455]) Position ids shape: torch.Size([1, 21455]) Input IDs shape: torch.Size([1, 21455]) Labels shape: torch.Size([1, 21455]) Final batch size: 1, sequence length: 17117 Attention mask shape: torch.Size([1, 1, 17117, 17117]) Position ids shape: torch.Size([1, 17117]) Input IDs shape: torch.Size([1, 17117]) Labels shape: torch.Size([1, 17117]) Final batch size: 1, sequence length: 19278 Attention mask shape: torch.Size([1, 1, 19278, 19278]) Position ids shape: torch.Size([1, 19278]) Input IDs shape: torch.Size([1, 19278]) Labels shape: torch.Size([1, 19278]) Final batch size: 1, sequence length: 10017 Attention mask shape: torch.Size([1, 1, 10017, 10017]) Position ids shape: torch.Size([1, 10017]) Input IDs shape: torch.Size([1, 10017]) Labels shape: torch.Size([1, 10017]) Final batch size: 1, sequence length: 19767 Attention mask shape: torch.Size([1, 1, 19767, 19767]) Position ids shape: torch.Size([1, 19767]) Input IDs shape: torch.Size([1, 19767]) Labels shape: torch.Size([1, 19767]) Final batch size: 1, sequence length: 19892 Attention mask shape: torch.Size([1, 1, 19892, 19892]) Position ids shape: torch.Size([1, 19892]) Input IDs shape: torch.Size([1, 19892]) Labels shape: torch.Size([1, 19892]) Final batch size: 1, sequence length: 17843 Attention mask shape: torch.Size([1, 1, 17843, 17843]) Position ids shape: torch.Size([1, 17843]) Input IDs shape: torch.Size([1, 17843]) Labels shape: torch.Size([1, 17843]) Final batch size: 1, sequence length: 13623 Attention mask shape: torch.Size([1, 1, 13623, 13623]) Position ids shape: torch.Size([1, 13623]) Input IDs shape: torch.Size([1, 13623]) Labels shape: torch.Size([1, 13623]) Final batch size: 1, sequence length: 14437 Attention mask shape: torch.Size([1, 1, 14437, 14437]) Position ids shape: torch.Size([1, 14437]) Input IDs shape: torch.Size([1, 14437]) Labels shape: torch.Size([1, 14437]) Final batch size: 1, sequence length: 20695 Attention mask shape: torch.Size([1, 1, 20695, 20695]) Position ids shape: torch.Size([1, 20695]) Input IDs shape: torch.Size([1, 20695]) Labels shape: torch.Size([1, 20695]) Final batch size: 1, sequence length: 19259 Attention mask shape: torch.Size([1, 1, 19259, 19259]) Position ids shape: torch.Size([1, 19259]) Input IDs shape: torch.Size([1, 19259]) Labels shape: torch.Size([1, 19259]) Final batch size: 1, sequence length: 6378 Attention mask shape: torch.Size([1, 1, 6378, 6378]) Position ids shape: torch.Size([1, 6378]) Input IDs shape: torch.Size([1, 6378]) Labels shape: torch.Size([1, 6378]) Final batch size: 1, sequence length: 23334 Attention mask shape: torch.Size([1, 1, 23334, 23334]) Position ids shape: torch.Size([1, 23334]) Input IDs shape: torch.Size([1, 23334]) Labels shape: torch.Size([1, 23334]) Final batch size: 1, sequence length: 5734 Attention mask shape: torch.Size([1, 1, 5734, 5734]) Position ids shape: torch.Size([1, 5734]) Input IDs shape: torch.Size([1, 5734]) Labels shape: torch.Size([1, 5734]) Final batch size: 1, sequence length: 14429 Attention mask shape: torch.Size([1, 1, 14429, 14429]) Position ids shape: torch.Size([1, 14429]) Input IDs shape: torch.Size([1, 14429]) Labels shape: torch.Size([1, 14429]) Final batch size: 1, sequence length: 14025 Attention mask shape: torch.Size([1, 1, 14025, 14025]) Position ids shape: torch.Size([1, 14025]) Input IDs shape: torch.Size([1, 14025]) Labels shape: torch.Size([1, 14025]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 19492 Attention mask shape: torch.Size([1, 1, 19492, 19492]) Position ids shape: torch.Size([1, 19492]) Input IDs shape: torch.Size([1, 19492]) Labels shape: torch.Size([1, 19492]) Final batch size: 1, sequence length: 25405 Attention mask shape: torch.Size([1, 1, 25405, 25405]) Position ids shape: torch.Size([1, 25405]) Input IDs shape: torch.Size([1, 25405]) Labels shape: torch.Size([1, 25405]) Final batch size: 1, sequence length: 17985 Attention mask shape: torch.Size([1, 1, 17985, 17985]) Position ids shape: torch.Size([1, 17985]) Input IDs shape: torch.Size([1, 17985]) Labels shape: torch.Size([1, 17985]) Final batch size: 1, sequence length: 26063 Attention mask shape: torch.Size([1, 1, 26063, 26063]) Position ids shape: torch.Size([1, 26063]) Input IDs shape: torch.Size([1, 26063]) Labels shape: torch.Size([1, 26063]) Final batch size: 1, sequence length: 22932 Attention mask shape: torch.Size([1, 1, 22932, 22932]) Position ids shape: torch.Size([1, 22932]) Input IDs shape: torch.Size([1, 22932]) Labels shape: torch.Size([1, 22932]) Final batch size: 1, sequence length: 28166 Attention mask shape: torch.Size([1, 1, 28166, 28166]) Position ids shape: torch.Size([1, 28166]) Input IDs shape: torch.Size([1, 28166]) Labels shape: torch.Size([1, 28166]) Final batch size: 1, sequence length: 26639 Attention mask shape: torch.Size([1, 1, 26639, 26639]) Position ids shape: torch.Size([1, 26639]) Input IDs shape: torch.Size([1, 26639]) Labels shape: torch.Size([1, 26639]) Final batch size: 1, sequence length: 17951 Attention mask shape: torch.Size([1, 1, 17951, 17951]) Position ids shape: torch.Size([1, 17951]) Input IDs shape: torch.Size([1, 17951]) Labels shape: torch.Size([1, 17951]) Final batch size: 1, sequence length: 25381 Attention mask shape: torch.Size([1, 1, 25381, 25381]) Position ids shape: torch.Size([1, 25381]) Input IDs shape: torch.Size([1, 25381]) Labels shape: torch.Size([1, 25381]) Final batch size: 1, sequence length: 24885 Attention mask shape: torch.Size([1, 1, 24885, 24885]) Position ids shape: torch.Size([1, 24885]) Input IDs shape: torch.Size([1, 24885]) Labels shape: torch.Size([1, 24885]) Final batch size: 1, sequence length: 25548 Attention mask shape: torch.Size([1, 1, 25548, 25548]) Position ids shape: torch.Size([1, 25548]) Input IDs shape: torch.Size([1, 25548]) Labels shape: torch.Size([1, 25548]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 28623 Attention mask shape: torch.Size([1, 1, 28623, 28623]) Position ids shape: torch.Size([1, 28623]) Input IDs shape: torch.Size([1, 28623]) Labels shape: torch.Size([1, 28623]) Final batch size: 1, sequence length: 28910 Attention mask shape: torch.Size([1, 1, 28910, 28910]) Position ids shape: torch.Size([1, 28910]) Input IDs shape: torch.Size([1, 28910]) Labels shape: torch.Size([1, 28910]) Final batch size: 1, sequence length: 30236 Attention mask shape: torch.Size([1, 1, 30236, 30236]) Position ids shape: torch.Size([1, 30236]) Input IDs shape: torch.Size([1, 30236]) Labels shape: torch.Size([1, 30236]) Final batch size: 1, sequence length: 29824 Attention mask shape: torch.Size([1, 1, 29824, 29824]) Position ids shape: torch.Size([1, 29824]) Input IDs shape: torch.Size([1, 29824]) Labels shape: torch.Size([1, 29824]) Final batch size: 1, sequence length: 16716 Attention mask shape: torch.Size([1, 1, 16716, 16716]) Position ids shape: torch.Size([1, 16716]) Input IDs shape: torch.Size([1, 16716]) Labels shape: torch.Size([1, 16716]) Final batch size: 1, sequence length: 25540 Attention mask shape: torch.Size([1, 1, 25540, 25540]) Position ids shape: torch.Size([1, 25540]) Input IDs shape: torch.Size([1, 25540]) Labels shape: torch.Size([1, 25540]) Final batch size: 1, sequence length: 28823 Attention mask shape: torch.Size([1, 1, 28823, 28823]) Position ids shape: torch.Size([1, 28823]) Input IDs shape: torch.Size([1, 28823]) Labels shape: torch.Size([1, 28823]) Final batch size: 1, sequence length: 32660 Attention mask shape: torch.Size([1, 1, 32660, 32660]) Position ids shape: torch.Size([1, 32660]) Input IDs shape: torch.Size([1, 32660]) Labels shape: torch.Size([1, 32660]) Final batch size: 1, sequence length: 17634 Attention mask shape: torch.Size([1, 1, 17634, 17634]) Position ids shape: torch.Size([1, 17634]) Input IDs shape: torch.Size([1, 17634]) Labels shape: torch.Size([1, 17634]) Final batch size: 1, sequence length: 28206 Attention mask shape: torch.Size([1, 1, 28206, 28206]) Position ids shape: torch.Size([1, 28206]) Input IDs shape: torch.Size([1, 28206]) Labels shape: torch.Size([1, 28206]) Final batch size: 1, sequence length: 32515 Attention mask shape: torch.Size([1, 1, 32515, 32515]) Position ids shape: torch.Size([1, 32515]) Input IDs shape: torch.Size([1, 32515]) Labels shape: torch.Size([1, 32515]) Final batch size: 1, sequence length: 31464 Attention mask shape: torch.Size([1, 1, 31464, 31464]) Position ids shape: torch.Size([1, 31464]) Input IDs shape: torch.Size([1, 31464]) Labels shape: torch.Size([1, 31464]) Final batch size: 1, sequence length: 33601 Attention mask shape: torch.Size([1, 1, 33601, 33601]) Position ids shape: torch.Size([1, 33601]) Input IDs shape: torch.Size([1, 33601]) Labels shape: torch.Size([1, 33601]) Final batch size: 1, sequence length: 9029 Attention mask shape: torch.Size([1, 1, 9029, 9029]) Position ids shape: torch.Size([1, 9029]) Input IDs shape: torch.Size([1, 9029]) Labels shape: torch.Size([1, 9029]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 32786 Attention mask shape: torch.Size([1, 1, 32786, 32786]) Position ids shape: torch.Size([1, 32786]) Input IDs shape: torch.Size([1, 32786]) Labels shape: torch.Size([1, 32786]) Final batch size: 1, sequence length: 27659 Attention mask shape: torch.Size([1, 1, 27659, 27659]) Position ids shape: torch.Size([1, 27659]) Input IDs shape: torch.Size([1, 27659]) Labels shape: torch.Size([1, 27659]) Final batch size: 1, sequence length: 27484 Attention mask shape: torch.Size([1, 1, 27484, 27484]) Position ids shape: torch.Size([1, 27484]) Input IDs shape: torch.Size([1, 27484]) Labels shape: torch.Size([1, 27484]) Final batch size: 1, sequence length: 10132 Attention mask shape: torch.Size([1, 1, 10132, 10132]) Position ids shape: torch.Size([1, 10132]) Input IDs shape: torch.Size([1, 10132]) Labels shape: torch.Size([1, 10132]) Final batch size: 1, sequence length: 36970 Attention mask shape: torch.Size([1, 1, 36970, 36970]) Position ids shape: torch.Size([1, 36970]) Input IDs shape: torch.Size([1, 36970]) Labels shape: torch.Size([1, 36970]) Final batch size: 1, sequence length: 35411 Attention mask shape: torch.Size([1, 1, 35411, 35411]) Position ids shape: torch.Size([1, 35411]) Input IDs shape: torch.Size([1, 35411]) Labels shape: torch.Size([1, 35411]) Final batch size: 1, sequence length: 37555 Attention mask shape: torch.Size([1, 1, 37555, 37555]) Position ids shape: torch.Size([1, 37555]) Input IDs shape: torch.Size([1, 37555]) Labels shape: torch.Size([1, 37555]) Final batch size: 1, sequence length: 29561 Attention mask shape: torch.Size([1, 1, 29561, 29561]) Position ids shape: torch.Size([1, 29561]) Input IDs shape: torch.Size([1, 29561]) Labels shape: torch.Size([1, 29561]) Final batch size: 1, sequence length: 20951 Attention mask shape: torch.Size([1, 1, 20951, 20951]) Position ids shape: torch.Size([1, 20951]) Input IDs shape: torch.Size([1, 20951]) Labels shape: torch.Size([1, 20951]) Final batch size: 1, sequence length: 37016 Attention mask shape: torch.Size([1, 1, 37016, 37016]) Position ids shape: torch.Size([1, 37016]) Input IDs shape: torch.Size([1, 37016]) Labels shape: torch.Size([1, 37016]) Final batch size: 1, sequence length: 15513 Attention mask shape: torch.Size([1, 1, 15513, 15513]) Position ids shape: torch.Size([1, 15513]) Input IDs shape: torch.Size([1, 15513]) Labels shape: torch.Size([1, 15513]) Final batch size: 1, sequence length: 36131 Attention mask shape: torch.Size([1, 1, 36131, 36131]) Position ids shape: torch.Size([1, 36131]) Input IDs shape: torch.Size([1, 36131]) Labels shape: torch.Size([1, 36131]) Final batch size: 1, sequence length: 35775 Attention mask shape: torch.Size([1, 1, 35775, 35775]) Position ids shape: torch.Size([1, 35775]) Input IDs shape: torch.Size([1, 35775]) Labels shape: torch.Size([1, 35775]) Final batch size: 1, sequence length: 20533 Attention mask shape: torch.Size([1, 1, 20533, 20533]) Position ids shape: torch.Size([1, 20533]) Input IDs shape: torch.Size([1, 20533]) Labels shape: torch.Size([1, 20533]) Final batch size: 1, sequence length: 38712 Attention mask shape: torch.Size([1, 1, 38712, 38712]) Position ids shape: torch.Size([1, 38712]) Input IDs shape: torch.Size([1, 38712]) Labels shape: torch.Size([1, 38712]) Final batch size: 1, sequence length: 36292 Attention mask shape: torch.Size([1, 1, 36292, 36292]) Position ids shape: torch.Size([1, 36292]) Input IDs shape: torch.Size([1, 36292]) Labels shape: torch.Size([1, 36292]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38071 Attention mask shape: torch.Size([1, 1, 38071, 38071]) Position ids shape: torch.Size([1, 38071]) Input IDs shape: torch.Size([1, 38071]) Labels shape: torch.Size([1, 38071]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40753 Attention mask shape: torch.Size([1, 1, 40753, 40753]) Position ids shape: torch.Size([1, 40753]) Input IDs shape: torch.Size([1, 40753]) Labels shape: torch.Size([1, 40753]) Final batch size: 1, sequence length: 26781 Attention mask shape: torch.Size([1, 1, 26781, 26781]) Position ids shape: torch.Size([1, 26781]) Input IDs shape: torch.Size([1, 26781]) Labels shape: torch.Size([1, 26781]) Final batch size: 1, sequence length: 36124 Attention mask shape: torch.Size([1, 1, 36124, 36124]) Position ids shape: torch.Size([1, 36124]) Input IDs shape: torch.Size([1, 36124]) Labels shape: torch.Size([1, 36124]) Final batch size: 1, sequence length: 35525 Attention mask shape: torch.Size([1, 1, 35525, 35525]) Position ids shape: torch.Size([1, 35525]) Input IDs shape: torch.Size([1, 35525]) Labels shape: torch.Size([1, 35525]) Final batch size: 1, sequence length: 31084 Attention mask shape: torch.Size([1, 1, 31084, 31084]) Position ids shape: torch.Size([1, 31084]) Input IDs shape: torch.Size([1, 31084]) Labels shape: torch.Size([1, 31084]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 28879 Attention mask shape: torch.Size([1, 1, 28879, 28879]) Position ids shape: torch.Size([1, 28879]) Input IDs shape: torch.Size([1, 28879]) Labels shape: torch.Size([1, 28879]) Final batch size: 1, sequence length: 36185 Attention mask shape: torch.Size([1, 1, 36185, 36185]) Position ids shape: torch.Size([1, 36185]) Input IDs shape: torch.Size([1, 36185]) Labels shape: torch.Size([1, 36185]) Final batch size: 1, sequence length: 14372 Attention mask shape: torch.Size([1, 1, 14372, 14372]) Position ids shape: torch.Size([1, 14372]) Input IDs shape: torch.Size([1, 14372]) Labels shape: torch.Size([1, 14372]) Final batch size: 1, sequence length: 17914 Attention mask shape: torch.Size([1, 1, 17914, 17914]) Position ids shape: torch.Size([1, 17914]) Input IDs shape: torch.Size([1, 17914]) Labels shape: torch.Size([1, 17914]) Final batch size: 1, sequence length: 38210 Attention mask shape: torch.Size([1, 1, 38210, 38210]) Position ids shape: torch.Size([1, 38210]) Input IDs shape: torch.Size([1, 38210]) Labels shape: torch.Size([1, 38210]) Final batch size: 1, sequence length: 30009 Attention mask shape: torch.Size([1, 1, 30009, 30009]) Position ids shape: torch.Size([1, 30009]) Input IDs shape: torch.Size([1, 30009]) Labels shape: torch.Size([1, 30009]) Final batch size: 1, sequence length: 33422 Attention mask shape: torch.Size([1, 1, 33422, 33422]) Position ids shape: torch.Size([1, 33422]) Input IDs shape: torch.Size([1, 33422]) Labels shape: torch.Size([1, 33422]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25529 Attention mask shape: torch.Size([1, 1, 25529, 25529]) Position ids shape: torch.Size([1, 25529]) Input IDs shape: torch.Size([1, 25529]) Labels shape: torch.Size([1, 25529]) Final batch size: 1, sequence length: 26562 Attention mask shape: torch.Size([1, 1, 26562, 26562]) Position ids shape: torch.Size([1, 26562]) Input IDs shape: torch.Size([1, 26562]) Labels shape: torch.Size([1, 26562]) Final batch size: 1, sequence length: 18884 Attention mask shape: torch.Size([1, 1, 18884, 18884]) Position ids shape: torch.Size([1, 18884]) Input IDs shape: torch.Size([1, 18884]) Labels shape: torch.Size([1, 18884]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16409 Attention mask shape: torch.Size([1, 1, 16409, 16409]) Position ids shape: torch.Size([1, 16409]) Input IDs shape: torch.Size([1, 16409]) Labels shape: torch.Size([1, 16409]) Final batch size: 1, sequence length: 30653 Attention mask shape: torch.Size([1, 1, 30653, 30653]) Position ids shape: torch.Size([1, 30653]) Input IDs shape: torch.Size([1, 30653]) Labels shape: torch.Size([1, 30653]) Final batch size: 1, sequence length: 15031 Attention mask shape: torch.Size([1, 1, 15031, 15031]) Position ids shape: torch.Size([1, 15031]) Input IDs shape: torch.Size([1, 15031]) Labels shape: torch.Size([1, 15031]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25850 Attention mask shape: torch.Size([1, 1, 25850, 25850]) Position ids shape: torch.Size([1, 25850]) Input IDs shape: torch.Size([1, 25850]) Labels shape: torch.Size([1, 25850]) Final batch size: 1, sequence length: 37159 Attention mask shape: torch.Size([1, 1, 37159, 37159]) Position ids shape: torch.Size([1, 37159]) Input IDs shape: torch.Size([1, 37159]) Labels shape: torch.Size([1, 37159]) Final batch size: 1, sequence length: 20307 Attention mask shape: torch.Size([1, 1, 20307, 20307]) Position ids shape: torch.Size([1, 20307]) Input IDs shape: torch.Size([1, 20307]) Labels shape: torch.Size([1, 20307]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36534 Attention mask shape: torch.Size([1, 1, 36534, 36534]) Position ids shape: torch.Size([1, 36534]) Input IDs shape: torch.Size([1, 36534]) Labels shape: torch.Size([1, 36534]) Final batch size: 1, sequence length: 31448 Attention mask shape: torch.Size([1, 1, 31448, 31448]) Position ids shape: torch.Size([1, 31448]) Input IDs shape: torch.Size([1, 31448]) Labels shape: torch.Size([1, 31448]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31042 Attention mask shape: torch.Size([1, 1, 31042, 31042]) Position ids shape: torch.Size([1, 31042]) Input IDs shape: torch.Size([1, 31042]) Labels shape: torch.Size([1, 31042]) Final batch size: 1, sequence length: 7448 Attention mask shape: torch.Size([1, 1, 7448, 7448]) Position ids shape: torch.Size([1, 7448]) Input IDs shape: torch.Size([1, 7448]) Labels shape: torch.Size([1, 7448]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40605 Attention mask shape: torch.Size([1, 1, 40605, 40605]) Position ids shape: torch.Size([1, 40605]) Input IDs shape: torch.Size([1, 40605]) Labels shape: torch.Size([1, 40605]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 5405 Attention mask shape: torch.Size([1, 1, 5405, 5405]) Position ids shape: torch.Size([1, 5405]) Input IDs shape: torch.Size([1, 5405]) Labels shape: torch.Size([1, 5405]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.3902, 'grad_norm': 1.0776006555304134, 'learning_rate': 1e-05, 'num_tokens': -inf, 'epoch': 0.62} Final batch size: 1, sequence length: 4641 Attention mask shape: torch.Size([1, 1, 4641, 4641]) Position ids shape: torch.Size([1, 4641]) Input IDs shape: torch.Size([1, 4641]) Labels shape: torch.Size([1, 4641]) Final batch size: 1, sequence length: 4223 Attention mask shape: torch.Size([1, 1, 4223, 4223]) Position ids shape: torch.Size([1, 4223]) Input IDs shape: torch.Size([1, 4223]) Labels shape: torch.Size([1, 4223]) Final batch size: 1, sequence length: 5754 Attention mask shape: torch.Size([1, 1, 5754, 5754]) Position ids shape: torch.Size([1, 5754]) Input IDs shape: torch.Size([1, 5754]) Labels shape: torch.Size([1, 5754]) Final batch size: 1, sequence length: 8177 Attention mask shape: torch.Size([1, 1, 8177, 8177]) Position ids shape: torch.Size([1, 8177]) Input IDs shape: torch.Size([1, 8177]) Labels shape: torch.Size([1, 8177]) Final batch size: 1, sequence length: 7795 Attention mask shape: torch.Size([1, 1, 7795, 7795]) Position ids shape: torch.Size([1, 7795]) Input IDs shape: torch.Size([1, 7795]) Labels shape: torch.Size([1, 7795]) Final batch size: 1, sequence length: 9021 Attention mask shape: torch.Size([1, 1, 9021, 9021]) Position ids shape: torch.Size([1, 9021]) Input IDs shape: torch.Size([1, 9021]) Labels shape: torch.Size([1, 9021]) Final batch size: 1, sequence length: 9027 Attention mask shape: torch.Size([1, 1, 9027, 9027]) Position ids shape: torch.Size([1, 9027]) Input IDs shape: torch.Size([1, 9027]) Labels shape: torch.Size([1, 9027]) Final batch size: 1, sequence length: 12501 Attention mask shape: torch.Size([1, 1, 12501, 12501]) Position ids shape: torch.Size([1, 12501]) Input IDs shape: torch.Size([1, 12501]) Labels shape: torch.Size([1, 12501]) Final batch size: 1, sequence length: 13186 Attention mask shape: torch.Size([1, 1, 13186, 13186]) Position ids shape: torch.Size([1, 13186]) Input IDs shape: torch.Size([1, 13186]) Labels shape: torch.Size([1, 13186]) Final batch size: 1, sequence length: 11974 Attention mask shape: torch.Size([1, 1, 11974, 11974]) Position ids shape: torch.Size([1, 11974]) Input IDs shape: torch.Size([1, 11974]) Labels shape: torch.Size([1, 11974]) Final batch size: 1, sequence length: 15283 Attention mask shape: torch.Size([1, 1, 15283, 15283]) Position ids shape: torch.Size([1, 15283]) Input IDs shape: torch.Size([1, 15283]) Labels shape: torch.Size([1, 15283]) Final batch size: 1, sequence length: 16330 Attention mask shape: torch.Size([1, 1, 16330, 16330]) Position ids shape: torch.Size([1, 16330]) Input IDs shape: torch.Size([1, 16330]) Labels shape: torch.Size([1, 16330]) Final batch size: 1, sequence length: 16104 Attention mask shape: torch.Size([1, 1, 16104, 16104]) Position ids shape: torch.Size([1, 16104]) Input IDs shape: torch.Size([1, 16104]) Labels shape: torch.Size([1, 16104]) Final batch size: 1, sequence length: 16496 Attention mask shape: torch.Size([1, 1, 16496, 16496]) Position ids shape: torch.Size([1, 16496]) Input IDs shape: torch.Size([1, 16496]) Labels shape: torch.Size([1, 16496]) Final batch size: 1, sequence length: 13957 Attention mask shape: torch.Size([1, 1, 13957, 13957]) Position ids shape: torch.Size([1, 13957]) Input IDs shape: torch.Size([1, 13957]) Labels shape: torch.Size([1, 13957]) Final batch size: 1, sequence length: 18958 Attention mask shape: torch.Size([1, 1, 18958, 18958]) Position ids shape: torch.Size([1, 18958]) Input IDs shape: torch.Size([1, 18958]) Labels shape: torch.Size([1, 18958]) Final batch size: 1, sequence length: 16088 Attention mask shape: torch.Size([1, 1, 16088, 16088]) Position ids shape: torch.Size([1, 16088]) Input IDs shape: torch.Size([1, 16088]) Labels shape: torch.Size([1, 16088]) Final batch size: 1, sequence length: 16014 Attention mask shape: torch.Size([1, 1, 16014, 16014]) Position ids shape: torch.Size([1, 16014]) Input IDs shape: torch.Size([1, 16014]) Labels shape: torch.Size([1, 16014]) Final batch size: 1, sequence length: 19603 Attention mask shape: torch.Size([1, 1, 19603, 19603]) Position ids shape: torch.Size([1, 19603]) Input IDs shape: torch.Size([1, 19603]) Labels shape: torch.Size([1, 19603]) Final batch size: 1, sequence length: 17092 Attention mask shape: torch.Size([1, 1, 17092, 17092]) Position ids shape: torch.Size([1, 17092]) Input IDs shape: torch.Size([1, 17092]) Labels shape: torch.Size([1, 17092]) Final batch size: 1, sequence length: 16366 Attention mask shape: torch.Size([1, 1, 16366, 16366]) Position ids shape: torch.Size([1, 16366]) Input IDs shape: torch.Size([1, 16366]) Labels shape: torch.Size([1, 16366]) Final batch size: 1, sequence length: 18938 Attention mask shape: torch.Size([1, 1, 18938, 18938]) Position ids shape: torch.Size([1, 18938]) Input IDs shape: torch.Size([1, 18938]) Labels shape: torch.Size([1, 18938]) Final batch size: 1, sequence length: 14492 Attention mask shape: torch.Size([1, 1, 14492, 14492]) Position ids shape: torch.Size([1, 14492]) Input IDs shape: torch.Size([1, 14492]) Labels shape: torch.Size([1, 14492]) Final batch size: 1, sequence length: 18415 Attention mask shape: torch.Size([1, 1, 18415, 18415]) Position ids shape: torch.Size([1, 18415]) Input IDs shape: torch.Size([1, 18415]) Labels shape: torch.Size([1, 18415]) Final batch size: 1, sequence length: 20726 Attention mask shape: torch.Size([1, 1, 20726, 20726]) Position ids shape: torch.Size([1, 20726]) Input IDs shape: torch.Size([1, 20726]) Labels shape: torch.Size([1, 20726]) Final batch size: 1, sequence length: 18034 Attention mask shape: torch.Size([1, 1, 18034, 18034]) Position ids shape: torch.Size([1, 18034]) Input IDs shape: torch.Size([1, 18034]) Labels shape: torch.Size([1, 18034]) Final batch size: 1, sequence length: 13468 Attention mask shape: torch.Size([1, 1, 13468, 13468]) Position ids shape: torch.Size([1, 13468]) Input IDs shape: torch.Size([1, 13468]) Labels shape: torch.Size([1, 13468]) Final batch size: 1, sequence length: 23132 Attention mask shape: torch.Size([1, 1, 23132, 23132]) Position ids shape: torch.Size([1, 23132]) Input IDs shape: torch.Size([1, 23132]) Labels shape: torch.Size([1, 23132]) Final batch size: 1, sequence length: 21404 Attention mask shape: torch.Size([1, 1, 21404, 21404]) Position ids shape: torch.Size([1, 21404]) Input IDs shape: torch.Size([1, 21404]) Labels shape: torch.Size([1, 21404]) Final batch size: 1, sequence length: 22771 Attention mask shape: torch.Size([1, 1, 22771, 22771]) Position ids shape: torch.Size([1, 22771]) Input IDs shape: torch.Size([1, 22771]) Labels shape: torch.Size([1, 22771]) Final batch size: 1, sequence length: 23102 Attention mask shape: torch.Size([1, 1, 23102, 23102]) Position ids shape: torch.Size([1, 23102]) Input IDs shape: torch.Size([1, 23102]) Labels shape: torch.Size([1, 23102]) Final batch size: 1, sequence length: 16590 Attention mask shape: torch.Size([1, 1, 16590, 16590]) Position ids shape: torch.Size([1, 16590]) Input IDs shape: torch.Size([1, 16590]) Labels shape: torch.Size([1, 16590]) Final batch size: 1, sequence length: 12756 Attention mask shape: torch.Size([1, 1, 12756, 12756]) Position ids shape: torch.Size([1, 12756]) Input IDs shape: torch.Size([1, 12756]) Labels shape: torch.Size([1, 12756]) Final batch size: 1, sequence length: 24432 Attention mask shape: torch.Size([1, 1, 24432, 24432]) Position ids shape: torch.Size([1, 24432]) Input IDs shape: torch.Size([1, 24432]) Labels shape: torch.Size([1, 24432]) Final batch size: 1, sequence length: 21132 Attention mask shape: torch.Size([1, 1, 21132, 21132]) Position ids shape: torch.Size([1, 21132]) Input IDs shape: torch.Size([1, 21132]) Labels shape: torch.Size([1, 21132]) Final batch size: 1, sequence length: 23971 Attention mask shape: torch.Size([1, 1, 23971, 23971]) Position ids shape: torch.Size([1, 23971]) Input IDs shape: torch.Size([1, 23971]) Labels shape: torch.Size([1, 23971]) Final batch size: 1, sequence length: 25351 Attention mask shape: torch.Size([1, 1, 25351, 25351]) Position ids shape: torch.Size([1, 25351]) Input IDs shape: torch.Size([1, 25351]) Labels shape: torch.Size([1, 25351]) Final batch size: 1, sequence length: 19034 Attention mask shape: torch.Size([1, 1, 19034, 19034]) Position ids shape: torch.Size([1, 19034]) Input IDs shape: torch.Size([1, 19034]) Labels shape: torch.Size([1, 19034]) Final batch size: 1, sequence length: 17514 Attention mask shape: torch.Size([1, 1, 17514, 17514]) Position ids shape: torch.Size([1, 17514]) Input IDs shape: torch.Size([1, 17514]) Labels shape: torch.Size([1, 17514]) Final batch size: 1, sequence length: 26596 Attention mask shape: torch.Size([1, 1, 26596, 26596]) Position ids shape: torch.Size([1, 26596]) Input IDs shape: torch.Size([1, 26596]) Labels shape: torch.Size([1, 26596]) Final batch size: 1, sequence length: 21705 Attention mask shape: torch.Size([1, 1, 21705, 21705]) Position ids shape: torch.Size([1, 21705]) Input IDs shape: torch.Size([1, 21705]) Labels shape: torch.Size([1, 21705]) Final batch size: 1, sequence length: 16291 Attention mask shape: torch.Size([1, 1, 16291, 16291]) Position ids shape: torch.Size([1, 16291]) Input IDs shape: torch.Size([1, 16291]) Labels shape: torch.Size([1, 16291]) Final batch size: 1, sequence length: 20817 Attention mask shape: torch.Size([1, 1, 20817, 20817]) Position ids shape: torch.Size([1, 20817]) Input IDs shape: torch.Size([1, 20817]) Labels shape: torch.Size([1, 20817]) Final batch size: 1, sequence length: 26375 Attention mask shape: torch.Size([1, 1, 26375, 26375]) Position ids shape: torch.Size([1, 26375]) Input IDs shape: torch.Size([1, 26375]) Labels shape: torch.Size([1, 26375]) Final batch size: 1, sequence length: 27780 Attention mask shape: torch.Size([1, 1, 27780, 27780]) Position ids shape: torch.Size([1, 27780]) Input IDs shape: torch.Size([1, 27780]) Labels shape: torch.Size([1, 27780]) Final batch size: 1, sequence length: 23894 Attention mask shape: torch.Size([1, 1, 23894, 23894]) Position ids shape: torch.Size([1, 23894]) Input IDs shape: torch.Size([1, 23894]) Labels shape: torch.Size([1, 23894]) Final batch size: 1, sequence length: 26766 Attention mask shape: torch.Size([1, 1, 26766, 26766]) Position ids shape: torch.Size([1, 26766]) Input IDs shape: torch.Size([1, 26766]) Labels shape: torch.Size([1, 26766]) Final batch size: 1, sequence length: 29132 Attention mask shape: torch.Size([1, 1, 29132, 29132]) Position ids shape: torch.Size([1, 29132]) Input IDs shape: torch.Size([1, 29132]) Labels shape: torch.Size([1, 29132]) Final batch size: 1, sequence length: 28412 Attention mask shape: torch.Size([1, 1, 28412, 28412]) Position ids shape: torch.Size([1, 28412]) Input IDs shape: torch.Size([1, 28412]) Labels shape: torch.Size([1, 28412]) Final batch size: 1, sequence length: 28574 Attention mask shape: torch.Size([1, 1, 28574, 28574]) Position ids shape: torch.Size([1, 28574]) Input IDs shape: torch.Size([1, 28574]) Labels shape: torch.Size([1, 28574]) Final batch size: 1, sequence length: 30165 Attention mask shape: torch.Size([1, 1, 30165, 30165]) Position ids shape: torch.Size([1, 30165]) Input IDs shape: torch.Size([1, 30165]) Labels shape: torch.Size([1, 30165]) Final batch size: 1, sequence length: 32022 Attention mask shape: torch.Size([1, 1, 32022, 32022]) Position ids shape: torch.Size([1, 32022]) Input IDs shape: torch.Size([1, 32022]) Labels shape: torch.Size([1, 32022]) Final batch size: 1, sequence length: 18869 Attention mask shape: torch.Size([1, 1, 18869, 18869]) Position ids shape: torch.Size([1, 18869]) Input IDs shape: torch.Size([1, 18869]) Labels shape: torch.Size([1, 18869]) Final batch size: 1, sequence length: 22043 Attention mask shape: torch.Size([1, 1, 22043, 22043]) Position ids shape: torch.Size([1, 22043]) Input IDs shape: torch.Size([1, 22043]) Labels shape: torch.Size([1, 22043]) Final batch size: 1, sequence length: 25388 Attention mask shape: torch.Size([1, 1, 25388, 25388]) Position ids shape: torch.Size([1, 25388]) Input IDs shape: torch.Size([1, 25388]) Labels shape: torch.Size([1, 25388]) Final batch size: 1, sequence length: 29168 Attention mask shape: torch.Size([1, 1, 29168, 29168]) Position ids shape: torch.Size([1, 29168]) Input IDs shape: torch.Size([1, 29168]) Labels shape: torch.Size([1, 29168]) Final batch size: 1, sequence length: 34457 Attention mask shape: torch.Size([1, 1, 34457, 34457]) Position ids shape: torch.Size([1, 34457]) Input IDs shape: torch.Size([1, 34457]) Labels shape: torch.Size([1, 34457]) Final batch size: 1, sequence length: 27785 Attention mask shape: torch.Size([1, 1, 27785, 27785]) Position ids shape: torch.Size([1, 27785]) Input IDs shape: torch.Size([1, 27785]) Labels shape: torch.Size([1, 27785]) Final batch size: 1, sequence length: 29830 Attention mask shape: torch.Size([1, 1, 29830, 29830]) Position ids shape: torch.Size([1, 29830]) Input IDs shape: torch.Size([1, 29830]) Labels shape: torch.Size([1, 29830]) Final batch size: 1, sequence length: 22328 Attention mask shape: torch.Size([1, 1, 22328, 22328]) Position ids shape: torch.Size([1, 22328]) Input IDs shape: torch.Size([1, 22328]) Labels shape: torch.Size([1, 22328]) Final batch size: 1, sequence length: 34326 Attention mask shape: torch.Size([1, 1, 34326, 34326]) Position ids shape: torch.Size([1, 34326]) Input IDs shape: torch.Size([1, 34326]) Labels shape: torch.Size([1, 34326]) Final batch size: 1, sequence length: 23810 Attention mask shape: torch.Size([1, 1, 23810, 23810]) Position ids shape: torch.Size([1, 23810]) Input IDs shape: torch.Size([1, 23810]) Labels shape: torch.Size([1, 23810]) Final batch size: 1, sequence length: 13790 Attention mask shape: torch.Size([1, 1, 13790, 13790]) Position ids shape: torch.Size([1, 13790]) Input IDs shape: torch.Size([1, 13790]) Labels shape: torch.Size([1, 13790]) Final batch size: 1, sequence length: 28845 Attention mask shape: torch.Size([1, 1, 28845, 28845]) Position ids shape: torch.Size([1, 28845]) Input IDs shape: torch.Size([1, 28845]) Labels shape: torch.Size([1, 28845]) Final batch size: 1, sequence length: 24653 Attention mask shape: torch.Size([1, 1, 24653, 24653]) Position ids shape: torch.Size([1, 24653]) Input IDs shape: torch.Size([1, 24653]) Labels shape: torch.Size([1, 24653]) Final batch size: 1, sequence length: 35868 Attention mask shape: torch.Size([1, 1, 35868, 35868]) Position ids shape: torch.Size([1, 35868]) Input IDs shape: torch.Size([1, 35868]) Labels shape: torch.Size([1, 35868]) Final batch size: 1, sequence length: 37039 Attention mask shape: torch.Size([1, 1, 37039, 37039]) Position ids shape: torch.Size([1, 37039]) Input IDs shape: torch.Size([1, 37039]) Labels shape: torch.Size([1, 37039]) Final batch size: 1, sequence length: 26465 Attention mask shape: torch.Size([1, 1, 26465, 26465]) Position ids shape: torch.Size([1, 26465]) Input IDs shape: torch.Size([1, 26465]) Labels shape: torch.Size([1, 26465]) Final batch size: 1, sequence length: 35790 Attention mask shape: torch.Size([1, 1, 35790, 35790]) Position ids shape: torch.Size([1, 35790]) Input IDs shape: torch.Size([1, 35790]) Labels shape: torch.Size([1, 35790]) Final batch size: 1, sequence length: 34874 Attention mask shape: torch.Size([1, 1, 34874, 34874]) Position ids shape: torch.Size([1, 34874]) Input IDs shape: torch.Size([1, 34874]) Labels shape: torch.Size([1, 34874]) Final batch size: 1, sequence length: 38919 Attention mask shape: torch.Size([1, 1, 38919, 38919]) Position ids shape: torch.Size([1, 38919]) Input IDs shape: torch.Size([1, 38919]) Labels shape: torch.Size([1, 38919]) Final batch size: 1, sequence length: 38617 Attention mask shape: torch.Size([1, 1, 38617, 38617]) Position ids shape: torch.Size([1, 38617]) Input IDs shape: torch.Size([1, 38617]) Labels shape: torch.Size([1, 38617]) Final batch size: 1, sequence length: 19744 Attention mask shape: torch.Size([1, 1, 19744, 19744]) Position ids shape: torch.Size([1, 19744]) Input IDs shape: torch.Size([1, 19744]) Labels shape: torch.Size([1, 19744]) Final batch size: 1, sequence length: 13986 Attention mask shape: torch.Size([1, 1, 13986, 13986]) Position ids shape: torch.Size([1, 13986]) Input IDs shape: torch.Size([1, 13986]) Labels shape: torch.Size([1, 13986]) Final batch size: 1, sequence length: 40717 Attention mask shape: torch.Size([1, 1, 40717, 40717]) Position ids shape: torch.Size([1, 40717]) Input IDs shape: torch.Size([1, 40717]) Labels shape: torch.Size([1, 40717]) Final batch size: 1, sequence length: 39150 Attention mask shape: torch.Size([1, 1, 39150, 39150]) Position ids shape: torch.Size([1, 39150]) Input IDs shape: torch.Size([1, 39150]) Labels shape: torch.Size([1, 39150]) Final batch size: 1, sequence length: 35803 Attention mask shape: torch.Size([1, 1, 35803, 35803]) Position ids shape: torch.Size([1, 35803]) Input IDs shape: torch.Size([1, 35803]) Labels shape: torch.Size([1, 35803]) Final batch size: 1, sequence length: 19204 Attention mask shape: torch.Size([1, 1, 19204, 19204]) Position ids shape: torch.Size([1, 19204]) Input IDs shape: torch.Size([1, 19204]) Labels shape: torch.Size([1, 19204]) Final batch size: 1, sequence length: 23245 Attention mask shape: torch.Size([1, 1, 23245, 23245]) Position ids shape: torch.Size([1, 23245]) Input IDs shape: torch.Size([1, 23245]) Labels shape: torch.Size([1, 23245]) Final batch size: 1, sequence length: 22282 Attention mask shape: torch.Size([1, 1, 22282, 22282]) Position ids shape: torch.Size([1, 22282]) Input IDs shape: torch.Size([1, 22282]) Labels shape: torch.Size([1, 22282]) Final batch size: 1, sequence length: 35999 Attention mask shape: torch.Size([1, 1, 35999, 35999]) Position ids shape: torch.Size([1, 35999]) Input IDs shape: torch.Size([1, 35999]) Labels shape: torch.Size([1, 35999]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36794 Attention mask shape: torch.Size([1, 1, 36794, 36794]) Position ids shape: torch.Size([1, 36794]) Input IDs shape: torch.Size([1, 36794]) Labels shape: torch.Size([1, 36794]) Final batch size: 1, sequence length: 22014 Attention mask shape: torch.Size([1, 1, 22014, 22014]) Position ids shape: torch.Size([1, 22014]) Input IDs shape: torch.Size([1, 22014]) Labels shape: torch.Size([1, 22014]) Final batch size: 1, sequence length: 31340 Attention mask shape: torch.Size([1, 1, 31340, 31340]) Position ids shape: torch.Size([1, 31340]) Input IDs shape: torch.Size([1, 31340]) Labels shape: torch.Size([1, 31340]) Final batch size: 1, sequence length: 18379 Attention mask shape: torch.Size([1, 1, 18379, 18379]) Position ids shape: torch.Size([1, 18379]) Input IDs shape: torch.Size([1, 18379]) Labels shape: torch.Size([1, 18379]) Final batch size: 1, sequence length: 31506 Attention mask shape: torch.Size([1, 1, 31506, 31506]) Position ids shape: torch.Size([1, 31506]) Input IDs shape: torch.Size([1, 31506]) Labels shape: torch.Size([1, 31506]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 11644 Attention mask shape: torch.Size([1, 1, 11644, 11644]) Position ids shape: torch.Size([1, 11644]) Input IDs shape: torch.Size([1, 11644]) Labels shape: torch.Size([1, 11644]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32974 Attention mask shape: torch.Size([1, 1, 32974, 32974]) Position ids shape: torch.Size([1, 32974]) Input IDs shape: torch.Size([1, 32974]) Labels shape: torch.Size([1, 32974]) Final batch size: 1, sequence length: 25560 Attention mask shape: torch.Size([1, 1, 25560, 25560]) Position ids shape: torch.Size([1, 25560]) Input IDs shape: torch.Size([1, 25560]) Labels shape: torch.Size([1, 25560]) Final batch size: 1, sequence length: 12418 Attention mask shape: torch.Size([1, 1, 12418, 12418]) Position ids shape: torch.Size([1, 12418]) Input IDs shape: torch.Size([1, 12418]) Labels shape: torch.Size([1, 12418]) Final batch size: 1, sequence length: 19846 Attention mask shape: torch.Size([1, 1, 19846, 19846]) Position ids shape: torch.Size([1, 19846]) Input IDs shape: torch.Size([1, 19846]) Labels shape: torch.Size([1, 19846]) Final batch size: 1, sequence length: 28002 Attention mask shape: torch.Size([1, 1, 28002, 28002]) Position ids shape: torch.Size([1, 28002]) Input IDs shape: torch.Size([1, 28002]) Labels shape: torch.Size([1, 28002]) Final batch size: 1, sequence length: 40874 Attention mask shape: torch.Size([1, 1, 40874, 40874]) Position ids shape: torch.Size([1, 40874]) Input IDs shape: torch.Size([1, 40874]) Labels shape: torch.Size([1, 40874]) Final batch size: 1, sequence length: 40065 Attention mask shape: torch.Size([1, 1, 40065, 40065]) Position ids shape: torch.Size([1, 40065]) Input IDs shape: torch.Size([1, 40065]) Labels shape: torch.Size([1, 40065]) Final batch size: 1, sequence length: 31359 Attention mask shape: torch.Size([1, 1, 31359, 31359]) Position ids shape: torch.Size([1, 31359]) Input IDs shape: torch.Size([1, 31359]) Labels shape: torch.Size([1, 31359]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28781 Attention mask shape: torch.Size([1, 1, 28781, 28781]) Position ids shape: torch.Size([1, 28781]) Input IDs shape: torch.Size([1, 28781]) Labels shape: torch.Size([1, 28781]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7364 Attention mask shape: torch.Size([1, 1, 7364, 7364]) Position ids shape: torch.Size([1, 7364]) Input IDs shape: torch.Size([1, 7364]) Labels shape: torch.Size([1, 7364]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 25923 Attention mask shape: torch.Size([1, 1, 25923, 25923]) Position ids shape: torch.Size([1, 25923]) Input IDs shape: torch.Size([1, 25923]) Labels shape: torch.Size([1, 25923]) Final batch size: 1, sequence length: 9309 Attention mask shape: torch.Size([1, 1, 9309, 9309]) Position ids shape: torch.Size([1, 9309]) Input IDs shape: torch.Size([1, 9309]) Labels shape: torch.Size([1, 9309]) Final batch size: 1, sequence length: 39919 Attention mask shape: torch.Size([1, 1, 39919, 39919]) Position ids shape: torch.Size([1, 39919]) Input IDs shape: torch.Size([1, 39919]) Labels shape: torch.Size([1, 39919]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26221 Attention mask shape: torch.Size([1, 1, 26221, 26221]) Position ids shape: torch.Size([1, 26221]) Input IDs shape: torch.Size([1, 26221]) Labels shape: torch.Size([1, 26221]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17224 Attention mask shape: torch.Size([1, 1, 17224, 17224]) Position ids shape: torch.Size([1, 17224]) Input IDs shape: torch.Size([1, 17224]) Labels shape: torch.Size([1, 17224]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 18654 Attention mask shape: torch.Size([1, 1, 18654, 18654]) Position ids shape: torch.Size([1, 18654]) Input IDs shape: torch.Size([1, 18654]) Labels shape: torch.Size([1, 18654]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 29526 Attention mask shape: torch.Size([1, 1, 29526, 29526]) Position ids shape: torch.Size([1, 29526]) Input IDs shape: torch.Size([1, 29526]) Labels shape: torch.Size([1, 29526]) Final batch size: 1, sequence length: 25820 Attention mask shape: torch.Size([1, 1, 25820, 25820]) Position ids shape: torch.Size([1, 25820]) Input IDs shape: torch.Size([1, 25820]) Labels shape: torch.Size([1, 25820]) Final batch size: 1, sequence length: 30709 Attention mask shape: torch.Size([1, 1, 30709, 30709]) Position ids shape: torch.Size([1, 30709]) Input IDs shape: torch.Size([1, 30709]) Labels shape: torch.Size([1, 30709]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.3813, 'grad_norm': 1.0720989736243394, 'learning_rate': 9.993147673772869e-06, 'num_tokens': -inf, 'epoch': 0.75} Final batch size: 1, sequence length: 7461 Attention mask shape: torch.Size([1, 1, 7461, 7461]) Position ids shape: torch.Size([1, 7461]) Input IDs shape: torch.Size([1, 7461]) Labels shape: torch.Size([1, 7461]) Final batch size: 1, sequence length: 12279 Attention mask shape: torch.Size([1, 1, 12279, 12279]) Position ids shape: torch.Size([1, 12279]) Input IDs shape: torch.Size([1, 12279]) Labels shape: torch.Size([1, 12279]) Final batch size: 1, sequence length: 9858 Attention mask shape: torch.Size([1, 1, 9858, 9858]) Position ids shape: torch.Size([1, 9858]) Input IDs shape: torch.Size([1, 9858]) Labels shape: torch.Size([1, 9858]) Final batch size: 1, sequence length: 10826 Attention mask shape: torch.Size([1, 1, 10826, 10826]) Position ids shape: torch.Size([1, 10826]) Input IDs shape: torch.Size([1, 10826]) Labels shape: torch.Size([1, 10826]) Final batch size: 1, sequence length: 12096 Attention mask shape: torch.Size([1, 1, 12096, 12096]) Position ids shape: torch.Size([1, 12096]) Input IDs shape: torch.Size([1, 12096]) Labels shape: torch.Size([1, 12096]) Final batch size: 1, sequence length: 13427 Attention mask shape: torch.Size([1, 1, 13427, 13427]) Position ids shape: torch.Size([1, 13427]) Input IDs shape: torch.Size([1, 13427]) Labels shape: torch.Size([1, 13427]) Final batch size: 1, sequence length: 12622 Attention mask shape: torch.Size([1, 1, 12622, 12622]) Position ids shape: torch.Size([1, 12622]) Input IDs shape: torch.Size([1, 12622]) Labels shape: torch.Size([1, 12622]) Final batch size: 1, sequence length: 15933 Attention mask shape: torch.Size([1, 1, 15933, 15933]) Position ids shape: torch.Size([1, 15933]) Input IDs shape: torch.Size([1, 15933]) Labels shape: torch.Size([1, 15933]) Final batch size: 1, sequence length: 13550 Attention mask shape: torch.Size([1, 1, 13550, 13550]) Position ids shape: torch.Size([1, 13550]) Input IDs shape: torch.Size([1, 13550]) Labels shape: torch.Size([1, 13550]) Final batch size: 1, sequence length: 16450 Attention mask shape: torch.Size([1, 1, 16450, 16450]) Position ids shape: torch.Size([1, 16450]) Input IDs shape: torch.Size([1, 16450]) Labels shape: torch.Size([1, 16450]) Final batch size: 1, sequence length: 12960 Attention mask shape: torch.Size([1, 1, 12960, 12960]) Position ids shape: torch.Size([1, 12960]) Input IDs shape: torch.Size([1, 12960]) Labels shape: torch.Size([1, 12960]) Final batch size: 1, sequence length: 16927 Attention mask shape: torch.Size([1, 1, 16927, 16927]) Position ids shape: torch.Size([1, 16927]) Input IDs shape: torch.Size([1, 16927]) Labels shape: torch.Size([1, 16927]) Final batch size: 1, sequence length: 17861 Attention mask shape: torch.Size([1, 1, 17861, 17861]) Position ids shape: torch.Size([1, 17861]) Input IDs shape: torch.Size([1, 17861]) Labels shape: torch.Size([1, 17861]) Final batch size: 1, sequence length: 17934 Attention mask shape: torch.Size([1, 1, 17934, 17934]) Position ids shape: torch.Size([1, 17934]) Input IDs shape: torch.Size([1, 17934]) Labels shape: torch.Size([1, 17934]) Final batch size: 1, sequence length: 17245 Attention mask shape: torch.Size([1, 1, 17245, 17245]) Position ids shape: torch.Size([1, 17245]) Input IDs shape: torch.Size([1, 17245]) Labels shape: torch.Size([1, 17245]) Final batch size: 1, sequence length: 15673 Attention mask shape: torch.Size([1, 1, 15673, 15673]) Position ids shape: torch.Size([1, 15673]) Input IDs shape: torch.Size([1, 15673]) Labels shape: torch.Size([1, 15673]) Final batch size: 1, sequence length: 18214 Attention mask shape: torch.Size([1, 1, 18214, 18214]) Position ids shape: torch.Size([1, 18214]) Input IDs shape: torch.Size([1, 18214]) Labels shape: torch.Size([1, 18214]) Final batch size: 1, sequence length: 13789 Attention mask shape: torch.Size([1, 1, 13789, 13789]) Position ids shape: torch.Size([1, 13789]) Input IDs shape: torch.Size([1, 13789]) Labels shape: torch.Size([1, 13789]) Final batch size: 1, sequence length: 21675 Attention mask shape: torch.Size([1, 1, 21675, 21675]) Position ids shape: torch.Size([1, 21675]) Input IDs shape: torch.Size([1, 21675]) Labels shape: torch.Size([1, 21675]) Final batch size: 1, sequence length: 12728 Attention mask shape: torch.Size([1, 1, 12728, 12728]) Position ids shape: torch.Size([1, 12728]) Input IDs shape: torch.Size([1, 12728]) Labels shape: torch.Size([1, 12728]) Final batch size: 1, sequence length: 19899 Attention mask shape: torch.Size([1, 1, 19899, 19899]) Position ids shape: torch.Size([1, 19899]) Input IDs shape: torch.Size([1, 19899]) Labels shape: torch.Size([1, 19899]) Final batch size: 1, sequence length: 20065 Attention mask shape: torch.Size([1, 1, 20065, 20065]) Position ids shape: torch.Size([1, 20065]) Input IDs shape: torch.Size([1, 20065]) Labels shape: torch.Size([1, 20065]) Final batch size: 1, sequence length: 21091 Attention mask shape: torch.Size([1, 1, 21091, 21091]) Position ids shape: torch.Size([1, 21091]) Input IDs shape: torch.Size([1, 21091]) Labels shape: torch.Size([1, 21091]) Final batch size: 1, sequence length: 16036 Attention mask shape: torch.Size([1, 1, 16036, 16036]) Position ids shape: torch.Size([1, 16036]) Input IDs shape: torch.Size([1, 16036]) Labels shape: torch.Size([1, 16036]) Final batch size: 1, sequence length: 20292 Attention mask shape: torch.Size([1, 1, 20292, 20292]) Position ids shape: torch.Size([1, 20292]) Input IDs shape: torch.Size([1, 20292]) Labels shape: torch.Size([1, 20292]) Final batch size: 1, sequence length: 19187 Attention mask shape: torch.Size([1, 1, 19187, 19187]) Position ids shape: torch.Size([1, 19187]) Input IDs shape: torch.Size([1, 19187]) Labels shape: torch.Size([1, 19187]) Final batch size: 1, sequence length: 23004 Attention mask shape: torch.Size([1, 1, 23004, 23004]) Position ids shape: torch.Size([1, 23004]) Input IDs shape: torch.Size([1, 23004]) Labels shape: torch.Size([1, 23004]) Final batch size: 1, sequence length: 25176 Attention mask shape: torch.Size([1, 1, 25176, 25176]) Position ids shape: torch.Size([1, 25176]) Input IDs shape: torch.Size([1, 25176]) Labels shape: torch.Size([1, 25176]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 24633 Attention mask shape: torch.Size([1, 1, 24633, 24633]) Position ids shape: torch.Size([1, 24633]) Input IDs shape: torch.Size([1, 24633]) Labels shape: torch.Size([1, 24633]) Final batch size: 1, sequence length: 23896 Attention mask shape: torch.Size([1, 1, 23896, 23896]) Position ids shape: torch.Size([1, 23896]) Input IDs shape: torch.Size([1, 23896]) Labels shape: torch.Size([1, 23896]) Final batch size: 1, sequence length: 25435 Attention mask shape: torch.Size([1, 1, 25435, 25435]) Position ids shape: torch.Size([1, 25435]) Input IDs shape: torch.Size([1, 25435]) Labels shape: torch.Size([1, 25435]) Final batch size: 1, sequence length: 24965 Attention mask shape: torch.Size([1, 1, 24965, 24965]) Position ids shape: torch.Size([1, 24965]) Input IDs shape: torch.Size([1, 24965]) Labels shape: torch.Size([1, 24965]) Final batch size: 1, sequence length: 21586 Attention mask shape: torch.Size([1, 1, 21586, 21586]) Position ids shape: torch.Size([1, 21586]) Input IDs shape: torch.Size([1, 21586]) Labels shape: torch.Size([1, 21586]) Final batch size: 1, sequence length: 22775 Attention mask shape: torch.Size([1, 1, 22775, 22775]) Position ids shape: torch.Size([1, 22775]) Input IDs shape: torch.Size([1, 22775]) Labels shape: torch.Size([1, 22775]) Final batch size: 1, sequence length: 14212 Attention mask shape: torch.Size([1, 1, 14212, 14212]) Position ids shape: torch.Size([1, 14212]) Input IDs shape: torch.Size([1, 14212]) Labels shape: torch.Size([1, 14212]) Final batch size: 1, sequence length: 22778 Attention mask shape: torch.Size([1, 1, 22778, 22778]) Position ids shape: torch.Size([1, 22778]) Input IDs shape: torch.Size([1, 22778]) Labels shape: torch.Size([1, 22778]) Final batch size: 1, sequence length: 17763 Attention mask shape: torch.Size([1, 1, 17763, 17763]) Position ids shape: torch.Size([1, 17763]) Input IDs shape: torch.Size([1, 17763]) Labels shape: torch.Size([1, 17763]) Final batch size: 1, sequence length: 25600 Attention mask shape: torch.Size([1, 1, 25600, 25600]) Position ids shape: torch.Size([1, 25600]) Input IDs shape: torch.Size([1, 25600]) Labels shape: torch.Size([1, 25600]) Final batch size: 1, sequence length: 25035 Attention mask shape: torch.Size([1, 1, 25035, 25035]) Position ids shape: torch.Size([1, 25035]) Input IDs shape: torch.Size([1, 25035]) Labels shape: torch.Size([1, 25035]) Final batch size: 1, sequence length: 27222 Attention mask shape: torch.Size([1, 1, 27222, 27222]) Position ids shape: torch.Size([1, 27222]) Input IDs shape: torch.Size([1, 27222]) Labels shape: torch.Size([1, 27222]) Final batch size: 1, sequence length: 24956 Attention mask shape: torch.Size([1, 1, 24956, 24956]) Position ids shape: torch.Size([1, 24956]) Input IDs shape: torch.Size([1, 24956]) Labels shape: torch.Size([1, 24956]) Final batch size: 1, sequence length: 26247 Attention mask shape: torch.Size([1, 1, 26247, 26247]) Position ids shape: torch.Size([1, 26247]) Input IDs shape: torch.Size([1, 26247]) Labels shape: torch.Size([1, 26247]) Final batch size: 1, sequence length: 26316 Attention mask shape: torch.Size([1, 1, 26316, 26316]) Position ids shape: torch.Size([1, 26316]) Input IDs shape: torch.Size([1, 26316]) Labels shape: torch.Size([1, 26316]) Final batch size: 1, sequence length: 20582 Attention mask shape: torch.Size([1, 1, 20582, 20582]) Position ids shape: torch.Size([1, 20582]) Input IDs shape: torch.Size([1, 20582]) Labels shape: torch.Size([1, 20582]) Final batch size: 1, sequence length: 19953 Attention mask shape: torch.Size([1, 1, 19953, 19953]) Position ids shape: torch.Size([1, 19953]) Input IDs shape: torch.Size([1, 19953]) Labels shape: torch.Size([1, 19953]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 3010 Attention mask shape: torch.Size([1, 1, 3010, 3010]) Position ids shape: torch.Size([1, 3010]) Input IDs shape: torch.Size([1, 3010]) Labels shape: torch.Size([1, 3010]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 28926 Attention mask shape: torch.Size([1, 1, 28926, 28926]) Position ids shape: torch.Size([1, 28926]) Input IDs shape: torch.Size([1, 28926]) Labels shape: torch.Size([1, 28926]) Final batch size: 1, sequence length: 25325 Attention mask shape: torch.Size([1, 1, 25325, 25325]) Position ids shape: torch.Size([1, 25325]) Input IDs shape: torch.Size([1, 25325]) Labels shape: torch.Size([1, 25325]) Final batch size: 1, sequence length: 24927 Attention mask shape: torch.Size([1, 1, 24927, 24927]) Position ids shape: torch.Size([1, 24927]) Input IDs shape: torch.Size([1, 24927]) Labels shape: torch.Size([1, 24927]) Final batch size: 1, sequence length: 13946 Attention mask shape: torch.Size([1, 1, 13946, 13946]) Position ids shape: torch.Size([1, 13946]) Input IDs shape: torch.Size([1, 13946]) Labels shape: torch.Size([1, 13946]) Final batch size: 1, sequence length: 13453 Attention mask shape: torch.Size([1, 1, 13453, 13453]) Position ids shape: torch.Size([1, 13453]) Input IDs shape: torch.Size([1, 13453]) Labels shape: torch.Size([1, 13453]) Final batch size: 1, sequence length: 18832 Attention mask shape: torch.Size([1, 1, 18832, 18832]) Position ids shape: torch.Size([1, 18832]) Input IDs shape: torch.Size([1, 18832]) Labels shape: torch.Size([1, 18832]) Final batch size: 1, sequence length: 26500 Attention mask shape: torch.Size([1, 1, 26500, 26500]) Position ids shape: torch.Size([1, 26500]) Input IDs shape: torch.Size([1, 26500]) Labels shape: torch.Size([1, 26500]) Final batch size: 1, sequence length: 28552 Attention mask shape: torch.Size([1, 1, 28552, 28552]) Position ids shape: torch.Size([1, 28552]) Input IDs shape: torch.Size([1, 28552]) Labels shape: torch.Size([1, 28552]) Final batch size: 1, sequence length: 6978 Attention mask shape: torch.Size([1, 1, 6978, 6978]) Position ids shape: torch.Size([1, 6978]) Input IDs shape: torch.Size([1, 6978]) Labels shape: torch.Size([1, 6978]) Final batch size: 1, sequence length: 31208 Attention mask shape: torch.Size([1, 1, 31208, 31208]) Position ids shape: torch.Size([1, 31208]) Input IDs shape: torch.Size([1, 31208]) Labels shape: torch.Size([1, 31208]) Final batch size: 1, sequence length: 24808 Attention mask shape: torch.Size([1, 1, 24808, 24808]) Position ids shape: torch.Size([1, 24808]) Input IDs shape: torch.Size([1, 24808]) Labels shape: torch.Size([1, 24808]) Final batch size: 1, sequence length: 17049 Attention mask shape: torch.Size([1, 1, 17049, 17049]) Position ids shape: torch.Size([1, 17049]) Input IDs shape: torch.Size([1, 17049]) Labels shape: torch.Size([1, 17049]) Final batch size: 1, sequence length: 25979 Attention mask shape: torch.Size([1, 1, 25979, 25979]) Position ids shape: torch.Size([1, 25979]) Input IDs shape: torch.Size([1, 25979]) Labels shape: torch.Size([1, 25979]) Final batch size: 1, sequence length: 28644 Attention mask shape: torch.Size([1, 1, 28644, 28644]) Position ids shape: torch.Size([1, 28644]) Input IDs shape: torch.Size([1, 28644]) Labels shape: torch.Size([1, 28644]) Final batch size: 1, sequence length: 16571 Attention mask shape: torch.Size([1, 1, 16571, 16571]) Position ids shape: torch.Size([1, 16571]) Input IDs shape: torch.Size([1, 16571]) Labels shape: torch.Size([1, 16571]) Final batch size: 1, sequence length: 32513 Attention mask shape: torch.Size([1, 1, 32513, 32513]) Position ids shape: torch.Size([1, 32513]) Input IDs shape: torch.Size([1, 32513]) Labels shape: torch.Size([1, 32513]) Final batch size: 1, sequence length: 32101 Attention mask shape: torch.Size([1, 1, 32101, 32101]) Position ids shape: torch.Size([1, 32101]) Input IDs shape: torch.Size([1, 32101]) Labels shape: torch.Size([1, 32101]) Final batch size: 1, sequence length: 32100 Attention mask shape: torch.Size([1, 1, 32100, 32100]) Position ids shape: torch.Size([1, 32100]) Input IDs shape: torch.Size([1, 32100]) Labels shape: torch.Size([1, 32100]) Final batch size: 1, sequence length: 21290 Attention mask shape: torch.Size([1, 1, 21290, 21290]) Position ids shape: torch.Size([1, 21290]) Input IDs shape: torch.Size([1, 21290]) Labels shape: torch.Size([1, 21290]) Final batch size: 1, sequence length: 31662 Attention mask shape: torch.Size([1, 1, 31662, 31662]) Position ids shape: torch.Size([1, 31662]) Input IDs shape: torch.Size([1, 31662]) Labels shape: torch.Size([1, 31662]) Final batch size: 1, sequence length: 37381 Attention mask shape: torch.Size([1, 1, 37381, 37381]) Position ids shape: torch.Size([1, 37381]) Input IDs shape: torch.Size([1, 37381]) Labels shape: torch.Size([1, 37381]) Final batch size: 1, sequence length: 35014 Attention mask shape: torch.Size([1, 1, 35014, 35014]) Position ids shape: torch.Size([1, 35014]) Input IDs shape: torch.Size([1, 35014]) Labels shape: torch.Size([1, 35014]) Final batch size: 1, sequence length: 25147 Attention mask shape: torch.Size([1, 1, 25147, 25147]) Position ids shape: torch.Size([1, 25147]) Input IDs shape: torch.Size([1, 25147]) Labels shape: torch.Size([1, 25147]) Final batch size: 1, sequence length: 28377 Attention mask shape: torch.Size([1, 1, 28377, 28377]) Position ids shape: torch.Size([1, 28377]) Input IDs shape: torch.Size([1, 28377]) Labels shape: torch.Size([1, 28377]) Final batch size: 1, sequence length: 16714 Attention mask shape: torch.Size([1, 1, 16714, 16714]) Position ids shape: torch.Size([1, 16714]) Input IDs shape: torch.Size([1, 16714]) Labels shape: torch.Size([1, 16714]) Final batch size: 1, sequence length: 24676 Attention mask shape: torch.Size([1, 1, 24676, 24676]) Position ids shape: torch.Size([1, 24676]) Input IDs shape: torch.Size([1, 24676]) Labels shape: torch.Size([1, 24676]) Final batch size: 1, sequence length: 37391 Attention mask shape: torch.Size([1, 1, 37391, 37391]) Position ids shape: torch.Size([1, 37391]) Input IDs shape: torch.Size([1, 37391]) Labels shape: torch.Size([1, 37391]) Final batch size: 1, sequence length: 39474 Attention mask shape: torch.Size([1, 1, 39474, 39474]) Position ids shape: torch.Size([1, 39474]) Input IDs shape: torch.Size([1, 39474]) Labels shape: torch.Size([1, 39474]) Final batch size: 1, sequence length: 12426 Attention mask shape: torch.Size([1, 1, 12426, 12426]) Position ids shape: torch.Size([1, 12426]) Input IDs shape: torch.Size([1, 12426]) Labels shape: torch.Size([1, 12426]) Final batch size: 1, sequence length: 26789 Attention mask shape: torch.Size([1, 1, 26789, 26789]) Position ids shape: torch.Size([1, 26789]) Input IDs shape: torch.Size([1, 26789]) Labels shape: torch.Size([1, 26789]) Final batch size: 1, sequence length: 33621 Attention mask shape: torch.Size([1, 1, 33621, 33621]) Position ids shape: torch.Size([1, 33621]) Input IDs shape: torch.Size([1, 33621]) Labels shape: torch.Size([1, 33621]) Final batch size: 1, sequence length: 37720 Attention mask shape: torch.Size([1, 1, 37720, 37720]) Position ids shape: torch.Size([1, 37720]) Input IDs shape: torch.Size([1, 37720]) Labels shape: torch.Size([1, 37720]) Final batch size: 1, sequence length: 38391 Attention mask shape: torch.Size([1, 1, 38391, 38391]) Position ids shape: torch.Size([1, 38391]) Input IDs shape: torch.Size([1, 38391]) Labels shape: torch.Size([1, 38391]) Final batch size: 1, sequence length: 25774 Attention mask shape: torch.Size([1, 1, 25774, 25774]) Position ids shape: torch.Size([1, 25774]) Input IDs shape: torch.Size([1, 25774]) Labels shape: torch.Size([1, 25774]) Final batch size: 1, sequence length: 26454 Attention mask shape: torch.Size([1, 1, 26454, 26454]) Position ids shape: torch.Size([1, 26454]) Input IDs shape: torch.Size([1, 26454]) Labels shape: torch.Size([1, 26454]) Final batch size: 1, sequence length: 25219 Attention mask shape: torch.Size([1, 1, 25219, 25219]) Position ids shape: torch.Size([1, 25219]) Input IDs shape: torch.Size([1, 25219]) Labels shape: torch.Size([1, 25219]) Final batch size: 1, sequence length: 36047 Attention mask shape: torch.Size([1, 1, 36047, 36047]) Position ids shape: torch.Size([1, 36047]) Input IDs shape: torch.Size([1, 36047]) Labels shape: torch.Size([1, 36047]) Final batch size: 1, sequence length: 16816 Attention mask shape: torch.Size([1, 1, 16816, 16816]) Position ids shape: torch.Size([1, 16816]) Input IDs shape: torch.Size([1, 16816]) Labels shape: torch.Size([1, 16816]) Final batch size: 1, sequence length: 35058 Attention mask shape: torch.Size([1, 1, 35058, 35058]) Position ids shape: torch.Size([1, 35058]) Input IDs shape: torch.Size([1, 35058]) Labels shape: torch.Size([1, 35058]) Final batch size: 1, sequence length: 39955 Attention mask shape: torch.Size([1, 1, 39955, 39955]) Position ids shape: torch.Size([1, 39955]) Input IDs shape: torch.Size([1, 39955]) Labels shape: torch.Size([1, 39955]) Final batch size: 1, sequence length: 30635 Attention mask shape: torch.Size([1, 1, 30635, 30635]) Position ids shape: torch.Size([1, 30635]) Input IDs shape: torch.Size([1, 30635]) Labels shape: torch.Size([1, 30635]) Final batch size: 1, sequence length: 31447 Attention mask shape: torch.Size([1, 1, 31447, 31447]) Position ids shape: torch.Size([1, 31447]) Input IDs shape: torch.Size([1, 31447]) Labels shape: torch.Size([1, 31447]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 35261 Attention mask shape: torch.Size([1, 1, 35261, 35261]) Position ids shape: torch.Size([1, 35261]) Input IDs shape: torch.Size([1, 35261]) Labels shape: torch.Size([1, 35261]) Final batch size: 1, sequence length: 22710 Attention mask shape: torch.Size([1, 1, 22710, 22710]) Position ids shape: torch.Size([1, 22710]) Input IDs shape: torch.Size([1, 22710]) Labels shape: torch.Size([1, 22710]) Final batch size: 1, sequence length: 38577 Attention mask shape: torch.Size([1, 1, 38577, 38577]) Position ids shape: torch.Size([1, 38577]) Input IDs shape: torch.Size([1, 38577]) Labels shape: torch.Size([1, 38577]) Final batch size: 1, sequence length: 21321 Attention mask shape: torch.Size([1, 1, 21321, 21321]) Position ids shape: torch.Size([1, 21321]) Input IDs shape: torch.Size([1, 21321]) Labels shape: torch.Size([1, 21321]) Final batch size: 1, sequence length: 26387 Attention mask shape: torch.Size([1, 1, 26387, 26387]) Position ids shape: torch.Size([1, 26387]) Input IDs shape: torch.Size([1, 26387]) Labels shape: torch.Size([1, 26387]) Final batch size: 1, sequence length: 19441 Attention mask shape: torch.Size([1, 1, 19441, 19441]) Position ids shape: torch.Size([1, 19441]) Input IDs shape: torch.Size([1, 19441]) Labels shape: torch.Size([1, 19441]) Final batch size: 1, sequence length: 30259 Attention mask shape: torch.Size([1, 1, 30259, 30259]) Position ids shape: torch.Size([1, 30259]) Input IDs shape: torch.Size([1, 30259]) Labels shape: torch.Size([1, 30259]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40590 Attention mask shape: torch.Size([1, 1, 40590, 40590]) Position ids shape: torch.Size([1, 40590]) Input IDs shape: torch.Size([1, 40590]) Labels shape: torch.Size([1, 40590]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32079 Attention mask shape: torch.Size([1, 1, 32079, 32079]) Position ids shape: torch.Size([1, 32079]) Input IDs shape: torch.Size([1, 32079]) Labels shape: torch.Size([1, 32079]) Final batch size: 1, sequence length: 32129 Attention mask shape: torch.Size([1, 1, 32129, 32129]) Position ids shape: torch.Size([1, 32129]) Input IDs shape: torch.Size([1, 32129]) Labels shape: torch.Size([1, 32129]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28448 Attention mask shape: torch.Size([1, 1, 28448, 28448]) Position ids shape: torch.Size([1, 28448]) Input IDs shape: torch.Size([1, 28448]) Labels shape: torch.Size([1, 28448]) Final batch size: 1, sequence length: 27021 Attention mask shape: torch.Size([1, 1, 27021, 27021]) Position ids shape: torch.Size([1, 27021]) Input IDs shape: torch.Size([1, 27021]) Labels shape: torch.Size([1, 27021]) Final batch size: 1, sequence length: 27371 Attention mask shape: torch.Size([1, 1, 27371, 27371]) Position ids shape: torch.Size([1, 27371]) Input IDs shape: torch.Size([1, 27371]) Labels shape: torch.Size([1, 27371]) Final batch size: 1, sequence length: 24623 Attention mask shape: torch.Size([1, 1, 24623, 24623]) Position ids shape: torch.Size([1, 24623]) Input IDs shape: torch.Size([1, 24623]) Labels shape: torch.Size([1, 24623]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33014 Attention mask shape: torch.Size([1, 1, 33014, 33014]) Position ids shape: torch.Size([1, 33014]) Input IDs shape: torch.Size([1, 33014]) Labels shape: torch.Size([1, 33014]) Final batch size: 1, sequence length: 14136 Attention mask shape: torch.Size([1, 1, 14136, 14136]) Position ids shape: torch.Size([1, 14136]) Input IDs shape: torch.Size([1, 14136]) Labels shape: torch.Size([1, 14136]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 23208 Attention mask shape: torch.Size([1, 1, 23208, 23208]) Position ids shape: torch.Size([1, 23208]) Input IDs shape: torch.Size([1, 23208]) Labels shape: torch.Size([1, 23208]) Final batch size: 1, sequence length: 36356 Attention mask shape: torch.Size([1, 1, 36356, 36356]) Position ids shape: torch.Size([1, 36356]) Input IDs shape: torch.Size([1, 36356]) Labels shape: torch.Size([1, 36356]) Final batch size: 1, sequence length: 30712 Attention mask shape: torch.Size([1, 1, 30712, 30712]) Position ids shape: torch.Size([1, 30712]) Input IDs shape: torch.Size([1, 30712]) Labels shape: torch.Size([1, 30712]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19027 Attention mask shape: torch.Size([1, 1, 19027, 19027]) Position ids shape: torch.Size([1, 19027]) Input IDs shape: torch.Size([1, 19027]) Labels shape: torch.Size([1, 19027]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.3745, 'grad_norm': 1.375671812827236, 'learning_rate': 9.972609476841368e-06, 'num_tokens': -inf, 'epoch': 0.88} Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 10507 Attention mask shape: torch.Size([1, 1, 10507, 10507]) Position ids shape: torch.Size([1, 10507]) Input IDs shape: torch.Size([1, 10507]) Labels shape: torch.Size([1, 10507]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 26271 Attention mask shape: torch.Size([1, 1, 26271, 26271]) Position ids shape: torch.Size([1, 26271]) Input IDs shape: torch.Size([1, 26271]) Labels shape: torch.Size([1, 26271]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 31381 Attention mask shape: torch.Size([1, 1, 31381, 31381]) Position ids shape: torch.Size([1, 31381]) Input IDs shape: torch.Size([1, 31381]) Labels shape: torch.Size([1, 31381]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.3714, 'grad_norm': 1.3029617874957764, 'learning_rate': 9.938441702975689e-06, 'num_tokens': -inf, 'epoch': 1.0} Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 6362 Attention mask shape: torch.Size([1, 1, 6362, 6362]) Position ids shape: torch.Size([1, 6362]) Input IDs shape: torch.Size([1, 6362]) Labels shape: torch.Size([1, 6362]) Final batch size: 1, sequence length: 10523 Attention mask shape: torch.Size([1, 1, 10523, 10523]) Position ids shape: torch.Size([1, 10523]) Input IDs shape: torch.Size([1, 10523]) Labels shape: torch.Size([1, 10523]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 12385 Attention mask shape: torch.Size([1, 1, 12385, 12385]) Position ids shape: torch.Size([1, 12385]) Input IDs shape: torch.Size([1, 12385]) Labels shape: torch.Size([1, 12385]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 18410 Attention mask shape: torch.Size([1, 1, 18410, 18410]) Position ids shape: torch.Size([1, 18410]) Input IDs shape: torch.Size([1, 18410]) Labels shape: torch.Size([1, 18410]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 18098 Attention mask shape: torch.Size([1, 1, 18098, 18098]) Position ids shape: torch.Size([1, 18098]) Input IDs shape: torch.Size([1, 18098]) Labels shape: torch.Size([1, 18098]) Final batch size: 1, sequence length: 12656 Attention mask shape: torch.Size([1, 1, 12656, 12656]) Position ids shape: torch.Size([1, 12656]) Input IDs shape: torch.Size([1, 12656]) Labels shape: torch.Size([1, 12656]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 17299 Attention mask shape: torch.Size([1, 1, 17299, 17299]) Position ids shape: torch.Size([1, 17299]) Input IDs shape: torch.Size([1, 17299]) Labels shape: torch.Size([1, 17299]) Final batch size: 1, sequence length: 9091 Attention mask shape: torch.Size([1, 1, 9091, 9091]) Position ids shape: torch.Size([1, 9091]) Input IDs shape: torch.Size([1, 9091]) Labels shape: torch.Size([1, 9091]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 24365 Attention mask shape: torch.Size([1, 1, 24365, 24365]) Position ids shape: torch.Size([1, 24365]) Input IDs shape: torch.Size([1, 24365]) Labels shape: torch.Size([1, 24365]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 18060 Attention mask shape: torch.Size([1, 1, 18060, 18060]) Position ids shape: torch.Size([1, 18060]) Input IDs shape: torch.Size([1, 18060]) Labels shape: torch.Size([1, 18060]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 28841 Attention mask shape: torch.Size([1, 1, 28841, 28841]) Position ids shape: torch.Size([1, 28841]) Input IDs shape: torch.Size([1, 28841]) Labels shape: torch.Size([1, 28841]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 29113 Attention mask shape: torch.Size([1, 1, 29113, 29113]) Position ids shape: torch.Size([1, 29113]) Input IDs shape: torch.Size([1, 29113]) Labels shape: torch.Size([1, 29113]) Final batch size: 1, sequence length: 26976 Attention mask shape: torch.Size([1, 1, 26976, 26976]) Position ids shape: torch.Size([1, 26976]) Input IDs shape: torch.Size([1, 26976]) Labels shape: torch.Size([1, 26976]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 18915 Attention mask shape: torch.Size([1, 1, 18915, 18915]) Position ids shape: torch.Size([1, 18915]) Input IDs shape: torch.Size([1, 18915]) Labels shape: torch.Size([1, 18915]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 11819 Attention mask shape: torch.Size([1, 1, 11819, 11819]) Position ids shape: torch.Size([1, 11819]) Input IDs shape: torch.Size([1, 11819]) Labels shape: torch.Size([1, 11819]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 24880 Attention mask shape: torch.Size([1, 1, 24880, 24880]) Position ids shape: torch.Size([1, 24880]) Input IDs shape: torch.Size([1, 24880]) Labels shape: torch.Size([1, 24880]) Final batch size: 1, sequence length: 21596 Attention mask shape: torch.Size([1, 1, 21596, 21596]) Position ids shape: torch.Size([1, 21596]) Input IDs shape: torch.Size([1, 21596]) Labels shape: torch.Size([1, 21596]) Final batch size: 1, sequence length: 25172 Attention mask shape: torch.Size([1, 1, 25172, 25172]) Position ids shape: torch.Size([1, 25172]) Input IDs shape: torch.Size([1, 25172]) Labels shape: torch.Size([1, 25172]) Final batch size: 1, sequence length: 27541 Attention mask shape: torch.Size([1, 1, 27541, 27541]) Position ids shape: torch.Size([1, 27541]) Input IDs shape: torch.Size([1, 27541]) Labels shape: torch.Size([1, 27541]) Final batch size: 1, sequence length: 6948 Attention mask shape: torch.Size([1, 1, 6948, 6948]) Position ids shape: torch.Size([1, 6948]) Input IDs shape: torch.Size([1, 6948]) Labels shape: torch.Size([1, 6948]) Final batch size: 1, sequence length: 7722 Attention mask shape: torch.Size([1, 1, 7722, 7722]) Position ids shape: torch.Size([1, 7722]) Input IDs shape: torch.Size([1, 7722]) Labels shape: torch.Size([1, 7722]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 21450 Attention mask shape: torch.Size([1, 1, 21450, 21450]) Position ids shape: torch.Size([1, 21450]) Input IDs shape: torch.Size([1, 21450]) Labels shape: torch.Size([1, 21450]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 21491 Attention mask shape: torch.Size([1, 1, 21491, 21491]) Position ids shape: torch.Size([1, 21491]) Input IDs shape: torch.Size([1, 21491]) Labels shape: torch.Size([1, 21491]) Final batch size: 1, sequence length: 14009 Attention mask shape: torch.Size([1, 1, 14009, 14009]) Position ids shape: torch.Size([1, 14009]) Input IDs shape: torch.Size([1, 14009]) Labels shape: torch.Size([1, 14009]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 29875 Attention mask shape: torch.Size([1, 1, 29875, 29875]) Position ids shape: torch.Size([1, 29875]) Input IDs shape: torch.Size([1, 29875]) Labels shape: torch.Size([1, 29875]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 20827 Attention mask shape: torch.Size([1, 1, 20827, 20827]) Position ids shape: torch.Size([1, 20827]) Input IDs shape: torch.Size([1, 20827]) Labels shape: torch.Size([1, 20827]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 19705 Attention mask shape: torch.Size([1, 1, 19705, 19705]) Position ids shape: torch.Size([1, 19705]) Input IDs shape: torch.Size([1, 19705]) Labels shape: torch.Size([1, 19705]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 20363 Attention mask shape: torch.Size([1, 1, 20363, 20363]) Position ids shape: torch.Size([1, 20363]) Input IDs shape: torch.Size([1, 20363]) Labels shape: torch.Size([1, 20363]) Final batch size: 1, sequence length: 25014 Attention mask shape: torch.Size([1, 1, 25014, 25014]) Position ids shape: torch.Size([1, 25014]) Input IDs shape: torch.Size([1, 25014]) Labels shape: torch.Size([1, 25014]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 13215 Attention mask shape: torch.Size([1, 1, 13215, 13215]) Position ids shape: torch.Size([1, 13215]) Input IDs shape: torch.Size([1, 13215]) Labels shape: torch.Size([1, 13215]) Final batch size: 1, sequence length: 30623 Attention mask shape: torch.Size([1, 1, 30623, 30623]) Position ids shape: torch.Size([1, 30623]) Input IDs shape: torch.Size([1, 30623]) Labels shape: torch.Size([1, 30623]) Final batch size: 1, sequence length: 17400 Attention mask shape: torch.Size([1, 1, 17400, 17400]) Position ids shape: torch.Size([1, 17400]) Input IDs shape: torch.Size([1, 17400]) Labels shape: torch.Size([1, 17400]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 33367 Attention mask shape: torch.Size([1, 1, 33367, 33367]) Position ids shape: torch.Size([1, 33367]) Input IDs shape: torch.Size([1, 33367]) Labels shape: torch.Size([1, 33367]) Final batch size: 1, sequence length: 28641 Attention mask shape: torch.Size([1, 1, 28641, 28641]) Position ids shape: torch.Size([1, 28641]) Input IDs shape: torch.Size([1, 28641]) Labels shape: torch.Size([1, 28641]) Final batch size: 1, sequence length: 15217 Attention mask shape: torch.Size([1, 1, 15217, 15217]) Position ids shape: torch.Size([1, 15217]) Input IDs shape: torch.Size([1, 15217]) Labels shape: torch.Size([1, 15217]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 27265 Attention mask shape: torch.Size([1, 1, 27265, 27265]) Position ids shape: torch.Size([1, 27265]) Input IDs shape: torch.Size([1, 27265]) Labels shape: torch.Size([1, 27265]) Final batch size: 1, sequence length: 24001 Attention mask shape: torch.Size([1, 1, 24001, 24001]) Position ids shape: torch.Size([1, 24001]) Input IDs shape: torch.Size([1, 24001]) Labels shape: torch.Size([1, 24001]) Final batch size: 1, sequence length: 22098 Attention mask shape: torch.Size([1, 1, 22098, 22098]) Position ids shape: torch.Size([1, 22098]) Input IDs shape: torch.Size([1, 22098]) Labels shape: torch.Size([1, 22098]) Final batch size: 1, sequence length: 15656 Attention mask shape: torch.Size([1, 1, 15656, 15656]) Position ids shape: torch.Size([1, 15656]) Input IDs shape: torch.Size([1, 15656]) Labels shape: torch.Size([1, 15656]) Final batch size: 1, sequence length: 27243 Attention mask shape: torch.Size([1, 1, 27243, 27243]) Position ids shape: torch.Size([1, 27243]) Input IDs shape: torch.Size([1, 27243]) Labels shape: torch.Size([1, 27243]) Final batch size: 1, sequence length: 38249 Attention mask shape: torch.Size([1, 1, 38249, 38249]) Position ids shape: torch.Size([1, 38249]) Input IDs shape: torch.Size([1, 38249]) Labels shape: torch.Size([1, 38249]) Final batch size: 1, sequence length: 21348 Attention mask shape: torch.Size([1, 1, 21348, 21348]) Position ids shape: torch.Size([1, 21348]) Input IDs shape: torch.Size([1, 21348]) Labels shape: torch.Size([1, 21348]) Final batch size: 1, sequence length: 29632 Attention mask shape: torch.Size([1, 1, 29632, 29632]) Position ids shape: torch.Size([1, 29632]) Input IDs shape: torch.Size([1, 29632]) Labels shape: torch.Size([1, 29632]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 30066 Attention mask shape: torch.Size([1, 1, 30066, 30066]) Position ids shape: torch.Size([1, 30066]) Input IDs shape: torch.Size([1, 30066]) Labels shape: torch.Size([1, 30066]) Final batch size: 1, sequence length: 32609 Attention mask shape: torch.Size([1, 1, 32609, 32609]) Position ids shape: torch.Size([1, 32609]) Input IDs shape: torch.Size([1, 32609]) Labels shape: torch.Size([1, 32609]) Final batch size: 1, sequence length: 18565 Final batch size: 1, sequence length: 18606 Attention mask shape: torch.Size([1, 1, 18565, 18565]) Position ids shape: torch.Size([1, 18565]) Input IDs shape: torch.Size([1, 18565]) Labels shape: torch.Size([1, 18565]) Attention mask shape: torch.Size([1, 1, 18606, 18606]) Position ids shape: torch.Size([1, 18606]) Input IDs shape: torch.Size([1, 18606]) Labels shape: torch.Size([1, 18606]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 39142 Attention mask shape: torch.Size([1, 1, 39142, 39142]) Position ids shape: torch.Size([1, 39142]) Input IDs shape: torch.Size([1, 39142]) Labels shape: torch.Size([1, 39142]) Final batch size: 1, sequence length: 22625 Attention mask shape: torch.Size([1, 1, 22625, 22625]) Position ids shape: torch.Size([1, 22625]) Input IDs shape: torch.Size([1, 22625]) Labels shape: torch.Size([1, 22625]) Final batch size: 1, sequence length: 26939 Attention mask shape: torch.Size([1, 1, 26939, 26939]) Position ids shape: torch.Size([1, 26939]) Input IDs shape: torch.Size([1, 26939]) Labels shape: torch.Size([1, 26939]) Final batch size: 1, sequence length: 13509 Attention mask shape: torch.Size([1, 1, 13509, 13509]) Position ids shape: torch.Size([1, 13509]) Input IDs shape: torch.Size([1, 13509]) Labels shape: torch.Size([1, 13509]) Final batch size: 1, sequence length: 17778 Attention mask shape: torch.Size([1, 1, 17778, 17778]) Position ids shape: torch.Size([1, 17778]) Input IDs shape: torch.Size([1, 17778]) Labels shape: torch.Size([1, 17778]) Final batch size: 1, sequence length: 21567 Attention mask shape: torch.Size([1, 1, 21567, 21567]) Position ids shape: torch.Size([1, 21567]) Input IDs shape: torch.Size([1, 21567]) Labels shape: torch.Size([1, 21567]) Final batch size: 1, sequence length: 27441 Attention mask shape: torch.Size([1, 1, 27441, 27441]) Position ids shape: torch.Size([1, 27441]) Input IDs shape: torch.Size([1, 27441]) Labels shape: torch.Size([1, 27441]) Final batch size: 1, sequence length: 30802 Attention mask shape: torch.Size([1, 1, 30802, 30802]) Position ids shape: torch.Size([1, 30802]) Input IDs shape: torch.Size([1, 30802]) Labels shape: torch.Size([1, 30802]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26306 Attention mask shape: torch.Size([1, 1, 26306, 26306]) Position ids shape: torch.Size([1, 26306]) Input IDs shape: torch.Size([1, 26306]) Labels shape: torch.Size([1, 26306]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 25447 Attention mask shape: torch.Size([1, 1, 25447, 25447]) Position ids shape: torch.Size([1, 25447]) Input IDs shape: torch.Size([1, 25447]) Labels shape: torch.Size([1, 25447]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19586 Attention mask shape: torch.Size([1, 1, 19586, 19586]) Position ids shape: torch.Size([1, 19586]) Input IDs shape: torch.Size([1, 19586]) Labels shape: torch.Size([1, 19586]) Final batch size: 1, sequence length: 33839 Attention mask shape: torch.Size([1, 1, 33839, 33839]) Position ids shape: torch.Size([1, 33839]) Input IDs shape: torch.Size([1, 33839]) Labels shape: torch.Size([1, 33839]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40745 Attention mask shape: torch.Size([1, 1, 40745, 40745]) Position ids shape: torch.Size([1, 40745]) Input IDs shape: torch.Size([1, 40745]) Labels shape: torch.Size([1, 40745]) Final batch size: 1, sequence length: 30101 Attention mask shape: torch.Size([1, 1, 30101, 30101]) Position ids shape: torch.Size([1, 30101]) Input IDs shape: torch.Size([1, 30101]) Labels shape: torch.Size([1, 30101]) Final batch size: 1, sequence length: 40496 Attention mask shape: torch.Size([1, 1, 40496, 40496]) Position ids shape: torch.Size([1, 40496]) Input IDs shape: torch.Size([1, 40496]) Labels shape: torch.Size([1, 40496]) Final batch size: 1, sequence length: 35153 Attention mask shape: torch.Size([1, 1, 35153, 35153]) Position ids shape: torch.Size([1, 35153]) Input IDs shape: torch.Size([1, 35153]) Labels shape: torch.Size([1, 35153]) Final batch size: 1, sequence length: 15508 Attention mask shape: torch.Size([1, 1, 15508, 15508]) Position ids shape: torch.Size([1, 15508]) Input IDs shape: torch.Size([1, 15508]) Labels shape: torch.Size([1, 15508]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32352 Attention mask shape: torch.Size([1, 1, 32352, 32352]) Position ids shape: torch.Size([1, 32352]) Input IDs shape: torch.Size([1, 32352]) Labels shape: torch.Size([1, 32352]) Final batch size: 1, sequence length: 36511 Attention mask shape: torch.Size([1, 1, 36511, 36511]) Position ids shape: torch.Size([1, 36511]) Input IDs shape: torch.Size([1, 36511]) Labels shape: torch.Size([1, 36511]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 30109 Attention mask shape: torch.Size([1, 1, 30109, 30109]) Position ids shape: torch.Size([1, 30109]) Input IDs shape: torch.Size([1, 30109]) Labels shape: torch.Size([1, 30109]) Final batch size: 1, sequence length: 23258 Attention mask shape: torch.Size([1, 1, 23258, 23258]) Position ids shape: torch.Size([1, 23258]) Input IDs shape: torch.Size([1, 23258]) Labels shape: torch.Size([1, 23258]) Final batch size: 1, sequence length: 16677 Attention mask shape: torch.Size([1, 1, 16677, 16677]) Position ids shape: torch.Size([1, 16677]) Input IDs shape: torch.Size([1, 16677]) Labels shape: torch.Size([1, 16677]) Final batch size: 1, sequence length: 9947 Attention mask shape: torch.Size([1, 1, 9947, 9947]) Position ids shape: torch.Size([1, 9947]) Input IDs shape: torch.Size([1, 9947]) Labels shape: torch.Size([1, 9947]) Final batch size: 1, sequence length: 19036 Attention mask shape: torch.Size([1, 1, 19036, 19036]) Position ids shape: torch.Size([1, 19036]) Input IDs shape: torch.Size([1, 19036]) Labels shape: torch.Size([1, 19036]) Final batch size: 1, sequence length: 24622 Attention mask shape: torch.Size([1, 1, 24622, 24622]) Position ids shape: torch.Size([1, 24622]) Input IDs shape: torch.Size([1, 24622]) Labels shape: torch.Size([1, 24622]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20422 Attention mask shape: torch.Size([1, 1, 20422, 20422]) Position ids shape: torch.Size([1, 20422]) Input IDs shape: torch.Size([1, 20422]) Labels shape: torch.Size([1, 20422]) Final batch size: 1, sequence length: 36271 Attention mask shape: torch.Size([1, 1, 36271, 36271]) Position ids shape: torch.Size([1, 36271]) Input IDs shape: torch.Size([1, 36271]) Labels shape: torch.Size([1, 36271]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 13095 Attention mask shape: torch.Size([1, 1, 13095, 13095]) Position ids shape: torch.Size([1, 13095]) Input IDs shape: torch.Size([1, 13095]) Labels shape: torch.Size([1, 13095]) Final batch size: 1, sequence length: 39836 Attention mask shape: torch.Size([1, 1, 39836, 39836]) Position ids shape: torch.Size([1, 39836]) Input IDs shape: torch.Size([1, 39836]) Labels shape: torch.Size([1, 39836]) Final batch size: 1, sequence length: 16587 Attention mask shape: torch.Size([1, 1, 16587, 16587]) Position ids shape: torch.Size([1, 16587]) Input IDs shape: torch.Size([1, 16587]) Labels shape: torch.Size([1, 16587]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20611 Attention mask shape: torch.Size([1, 1, 20611, 20611]) Position ids shape: torch.Size([1, 20611]) Input IDs shape: torch.Size([1, 20611]) Labels shape: torch.Size([1, 20611]) Final batch size: 1, sequence length: 37945 Attention mask shape: torch.Size([1, 1, 37945, 37945]) Position ids shape: torch.Size([1, 37945]) Input IDs shape: torch.Size([1, 37945]) Labels shape: torch.Size([1, 37945]) Final batch size: 1, sequence length: 13297 Attention mask shape: torch.Size([1, 1, 13297, 13297]) Position ids shape: torch.Size([1, 13297]) Input IDs shape: torch.Size([1, 13297]) Labels shape: torch.Size([1, 13297]) Final batch size: 1, sequence length: 12653 Attention mask shape: torch.Size([1, 1, 12653, 12653]) Position ids shape: torch.Size([1, 12653]) Input IDs shape: torch.Size([1, 12653]) Labels shape: torch.Size([1, 12653]) Final batch size: 1, sequence length: 34937 Attention mask shape: torch.Size([1, 1, 34937, 34937]) Position ids shape: torch.Size([1, 34937]) Input IDs shape: torch.Size([1, 34937]) Labels shape: torch.Size([1, 34937]) Final batch size: 1, sequence length: 24298 Attention mask shape: torch.Size([1, 1, 24298, 24298]) Position ids shape: torch.Size([1, 24298]) Input IDs shape: torch.Size([1, 24298]) Labels shape: torch.Size([1, 24298]) Final batch size: 1, sequence length: 22786 Attention mask shape: torch.Size([1, 1, 22786, 22786]) Position ids shape: torch.Size([1, 22786]) Input IDs shape: torch.Size([1, 22786]) Labels shape: torch.Size([1, 22786]) Final batch size: 1, sequence length: 34142 Attention mask shape: torch.Size([1, 1, 34142, 34142]) Position ids shape: torch.Size([1, 34142]) Input IDs shape: torch.Size([1, 34142]) Labels shape: torch.Size([1, 34142]) Final batch size: 1, sequence length: 21758 Attention mask shape: torch.Size([1, 1, 21758, 21758]) Position ids shape: torch.Size([1, 21758]) Input IDs shape: torch.Size([1, 21758]) Labels shape: torch.Size([1, 21758]) Final batch size: 1, sequence length: 12224 Attention mask shape: torch.Size([1, 1, 12224, 12224]) Position ids shape: torch.Size([1, 12224]) Input IDs shape: torch.Size([1, 12224]) Labels shape: torch.Size([1, 12224]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32472 Attention mask shape: torch.Size([1, 1, 32472, 32472]) Position ids shape: torch.Size([1, 32472]) Input IDs shape: torch.Size([1, 32472]) Labels shape: torch.Size([1, 32472]) Final batch size: 1, sequence length: 28634 Attention mask shape: torch.Size([1, 1, 28634, 28634]) Position ids shape: torch.Size([1, 28634]) Input IDs shape: torch.Size([1, 28634]) Labels shape: torch.Size([1, 28634]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21683 Attention mask shape: torch.Size([1, 1, 21683, 21683]) Position ids shape: torch.Size([1, 21683]) Input IDs shape: torch.Size([1, 21683]) Labels shape: torch.Size([1, 21683]) Final batch size: 1, sequence length: 14995 Attention mask shape: torch.Size([1, 1, 14995, 14995]) Position ids shape: torch.Size([1, 14995]) Input IDs shape: torch.Size([1, 14995]) Labels shape: torch.Size([1, 14995]) Final batch size: 1, sequence length: 22623 Attention mask shape: torch.Size([1, 1, 22623, 22623]) Position ids shape: torch.Size([1, 22623]) Input IDs shape: torch.Size([1, 22623]) Labels shape: torch.Size([1, 22623]) Final batch size: 1, sequence length: 32753 Attention mask shape: torch.Size([1, 1, 32753, 32753]) Position ids shape: torch.Size([1, 32753]) Input IDs shape: torch.Size([1, 32753]) Labels shape: torch.Size([1, 32753]) Final batch size: 1, sequence length: 17373 Attention mask shape: torch.Size([1, 1, 17373, 17373]) Position ids shape: torch.Size([1, 17373]) Input IDs shape: torch.Size([1, 17373]) Labels shape: torch.Size([1, 17373]) Final batch size: 1, sequence length: 24424 Attention mask shape: torch.Size([1, 1, 24424, 24424]) Position ids shape: torch.Size([1, 24424]) Input IDs shape: torch.Size([1, 24424]) Labels shape: torch.Size([1, 24424]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 35864 Attention mask shape: torch.Size([1, 1, 35864, 35864]) Position ids shape: torch.Size([1, 35864]) Input IDs shape: torch.Size([1, 35864]) Labels shape: torch.Size([1, 35864]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32564 Attention mask shape: torch.Size([1, 1, 32564, 32564]) Position ids shape: torch.Size([1, 32564]) Input IDs shape: torch.Size([1, 32564]) Labels shape: torch.Size([1, 32564]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32344 Attention mask shape: torch.Size([1, 1, 32344, 32344]) Position ids shape: torch.Size([1, 32344]) Input IDs shape: torch.Size([1, 32344]) Labels shape: torch.Size([1, 32344]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26537 Attention mask shape: torch.Size([1, 1, 26537, 26537]) Position ids shape: torch.Size([1, 26537]) Input IDs shape: torch.Size([1, 26537]) Labels shape: torch.Size([1, 26537]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32514 Attention mask shape: torch.Size([1, 1, 32514, 32514]) Position ids shape: torch.Size([1, 32514]) Input IDs shape: torch.Size([1, 32514]) Labels shape: torch.Size([1, 32514]) Final batch size: 1, sequence length: 32919 Attention mask shape: torch.Size([1, 1, 32919, 32919]) Position ids shape: torch.Size([1, 32919]) Input IDs shape: torch.Size([1, 32919]) Labels shape: torch.Size([1, 32919]) Final batch size: 1, sequence length: 23362 Attention mask shape: torch.Size([1, 1, 23362, 23362]) Position ids shape: torch.Size([1, 23362]) Input IDs shape: torch.Size([1, 23362]) Labels shape: torch.Size([1, 23362]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 37168 Attention mask shape: torch.Size([1, 1, 37168, 37168]) Position ids shape: torch.Size([1, 37168]) Input IDs shape: torch.Size([1, 37168]) Labels shape: torch.Size([1, 37168]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28861 Attention mask shape: torch.Size([1, 1, 28861, 28861]) Position ids shape: torch.Size([1, 28861]) Input IDs shape: torch.Size([1, 28861]) Labels shape: torch.Size([1, 28861]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26006 Attention mask shape: torch.Size([1, 1, 26006, 26006]) Position ids shape: torch.Size([1, 26006]) Input IDs shape: torch.Size([1, 26006]) Labels shape: torch.Size([1, 26006]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.3534, 'grad_norm': 0.9634744200772533, 'learning_rate': 9.890738003669029e-06, 'num_tokens': -inf, 'epoch': 1.12} Final batch size: 1, sequence length: 6316 Attention mask shape: torch.Size([1, 1, 6316, 6316]) Position ids shape: torch.Size([1, 6316]) Input IDs shape: torch.Size([1, 6316]) Labels shape: torch.Size([1, 6316]) Final batch size: 1, sequence length: 7360 Attention mask shape: torch.Size([1, 1, 7360, 7360]) Position ids shape: torch.Size([1, 7360]) Input IDs shape: torch.Size([1, 7360]) Labels shape: torch.Size([1, 7360]) Final batch size: 1, sequence length: 4858 Attention mask shape: torch.Size([1, 1, 4858, 4858]) Position ids shape: torch.Size([1, 4858]) Input IDs shape: torch.Size([1, 4858]) Labels shape: torch.Size([1, 4858]) Final batch size: 1, sequence length: 10301 Attention mask shape: torch.Size([1, 1, 10301, 10301]) Position ids shape: torch.Size([1, 10301]) Input IDs shape: torch.Size([1, 10301]) Labels shape: torch.Size([1, 10301]) Final batch size: 1, sequence length: 13355 Attention mask shape: torch.Size([1, 1, 13355, 13355]) Position ids shape: torch.Size([1, 13355]) Input IDs shape: torch.Size([1, 13355]) Labels shape: torch.Size([1, 13355]) Final batch size: 1, sequence length: 12293 Attention mask shape: torch.Size([1, 1, 12293, 12293]) Position ids shape: torch.Size([1, 12293]) Input IDs shape: torch.Size([1, 12293]) Labels shape: torch.Size([1, 12293]) Final batch size: 1, sequence length: 14363 Attention mask shape: torch.Size([1, 1, 14363, 14363]) Position ids shape: torch.Size([1, 14363]) Input IDs shape: torch.Size([1, 14363]) Labels shape: torch.Size([1, 14363]) Final batch size: 1, sequence length: 11548 Attention mask shape: torch.Size([1, 1, 11548, 11548]) Position ids shape: torch.Size([1, 11548]) Input IDs shape: torch.Size([1, 11548]) Labels shape: torch.Size([1, 11548]) Final batch size: 1, sequence length: 11728 Attention mask shape: torch.Size([1, 1, 11728, 11728]) Position ids shape: torch.Size([1, 11728]) Input IDs shape: torch.Size([1, 11728]) Labels shape: torch.Size([1, 11728]) Final batch size: 1, sequence length: 16827 Attention mask shape: torch.Size([1, 1, 16827, 16827]) Position ids shape: torch.Size([1, 16827]) Input IDs shape: torch.Size([1, 16827]) Labels shape: torch.Size([1, 16827]) Final batch size: 1, sequence length: 17108 Attention mask shape: torch.Size([1, 1, 17108, 17108]) Position ids shape: torch.Size([1, 17108]) Input IDs shape: torch.Size([1, 17108]) Labels shape: torch.Size([1, 17108]) Final batch size: 1, sequence length: 14704 Attention mask shape: torch.Size([1, 1, 14704, 14704]) Position ids shape: torch.Size([1, 14704]) Input IDs shape: torch.Size([1, 14704]) Labels shape: torch.Size([1, 14704]) Final batch size: 1, sequence length: 16536 Attention mask shape: torch.Size([1, 1, 16536, 16536]) Position ids shape: torch.Size([1, 16536]) Input IDs shape: torch.Size([1, 16536]) Labels shape: torch.Size([1, 16536]) Final batch size: 1, sequence length: 14587 Attention mask shape: torch.Size([1, 1, 14587, 14587]) Position ids shape: torch.Size([1, 14587]) Input IDs shape: torch.Size([1, 14587]) Labels shape: torch.Size([1, 14587]) Final batch size: 1, sequence length: 17415 Attention mask shape: torch.Size([1, 1, 17415, 17415]) Position ids shape: torch.Size([1, 17415]) Input IDs shape: torch.Size([1, 17415]) Labels shape: torch.Size([1, 17415]) Final batch size: 1, sequence length: 17220 Attention mask shape: torch.Size([1, 1, 17220, 17220]) Position ids shape: torch.Size([1, 17220]) Input IDs shape: torch.Size([1, 17220]) Labels shape: torch.Size([1, 17220]) Final batch size: 1, sequence length: 7221 Attention mask shape: torch.Size([1, 1, 7221, 7221]) Position ids shape: torch.Size([1, 7221]) Input IDs shape: torch.Size([1, 7221]) Labels shape: torch.Size([1, 7221]) Final batch size: 1, sequence length: 17733 Attention mask shape: torch.Size([1, 1, 17733, 17733]) Position ids shape: torch.Size([1, 17733]) Input IDs shape: torch.Size([1, 17733]) Labels shape: torch.Size([1, 17733]) Final batch size: 1, sequence length: 18741 Attention mask shape: torch.Size([1, 1, 18741, 18741]) Position ids shape: torch.Size([1, 18741]) Input IDs shape: torch.Size([1, 18741]) Labels shape: torch.Size([1, 18741]) Final batch size: 1, sequence length: 18927 Attention mask shape: torch.Size([1, 1, 18927, 18927]) Position ids shape: torch.Size([1, 18927]) Input IDs shape: torch.Size([1, 18927]) Labels shape: torch.Size([1, 18927]) Final batch size: 1, sequence length: 18496 Attention mask shape: torch.Size([1, 1, 18496, 18496]) Position ids shape: torch.Size([1, 18496]) Input IDs shape: torch.Size([1, 18496]) Labels shape: torch.Size([1, 18496]) Final batch size: 1, sequence length: 18653 Attention mask shape: torch.Size([1, 1, 18653, 18653]) Position ids shape: torch.Size([1, 18653]) Input IDs shape: torch.Size([1, 18653]) Labels shape: torch.Size([1, 18653]) Final batch size: 1, sequence length: 20933 Attention mask shape: torch.Size([1, 1, 20933, 20933]) Position ids shape: torch.Size([1, 20933]) Input IDs shape: torch.Size([1, 20933]) Labels shape: torch.Size([1, 20933]) Final batch size: 1, sequence length: 16750 Attention mask shape: torch.Size([1, 1, 16750, 16750]) Position ids shape: torch.Size([1, 16750]) Input IDs shape: torch.Size([1, 16750]) Labels shape: torch.Size([1, 16750]) Final batch size: 1, sequence length: 18393 Attention mask shape: torch.Size([1, 1, 18393, 18393]) Position ids shape: torch.Size([1, 18393]) Input IDs shape: torch.Size([1, 18393]) Labels shape: torch.Size([1, 18393]) Final batch size: 1, sequence length: 13638 Attention mask shape: torch.Size([1, 1, 13638, 13638]) Position ids shape: torch.Size([1, 13638]) Input IDs shape: torch.Size([1, 13638]) Labels shape: torch.Size([1, 13638]) Final batch size: 1, sequence length: 22004 Attention mask shape: torch.Size([1, 1, 22004, 22004]) Position ids shape: torch.Size([1, 22004]) Input IDs shape: torch.Size([1, 22004]) Labels shape: torch.Size([1, 22004]) Final batch size: 1, sequence length: 19414 Attention mask shape: torch.Size([1, 1, 19414, 19414]) Position ids shape: torch.Size([1, 19414]) Input IDs shape: torch.Size([1, 19414]) Labels shape: torch.Size([1, 19414]) Final batch size: 1, sequence length: 21420 Attention mask shape: torch.Size([1, 1, 21420, 21420]) Position ids shape: torch.Size([1, 21420]) Input IDs shape: torch.Size([1, 21420]) Labels shape: torch.Size([1, 21420]) Final batch size: 1, sequence length: 20612 Attention mask shape: torch.Size([1, 1, 20612, 20612]) Position ids shape: torch.Size([1, 20612]) Input IDs shape: torch.Size([1, 20612]) Labels shape: torch.Size([1, 20612]) Final batch size: 1, sequence length: 22391 Attention mask shape: torch.Size([1, 1, 22391, 22391]) Position ids shape: torch.Size([1, 22391]) Input IDs shape: torch.Size([1, 22391]) Labels shape: torch.Size([1, 22391]) Final batch size: 1, sequence length: 11067 Attention mask shape: torch.Size([1, 1, 11067, 11067]) Position ids shape: torch.Size([1, 11067]) Input IDs shape: torch.Size([1, 11067]) Labels shape: torch.Size([1, 11067]) Final batch size: 1, sequence length: 10719 Attention mask shape: torch.Size([1, 1, 10719, 10719]) Position ids shape: torch.Size([1, 10719]) Input IDs shape: torch.Size([1, 10719]) Labels shape: torch.Size([1, 10719]) Final batch size: 1, sequence length: 24988 Attention mask shape: torch.Size([1, 1, 24988, 24988]) Position ids shape: torch.Size([1, 24988]) Input IDs shape: torch.Size([1, 24988]) Labels shape: torch.Size([1, 24988]) Final batch size: 1, sequence length: 25477 Attention mask shape: torch.Size([1, 1, 25477, 25477]) Position ids shape: torch.Size([1, 25477]) Input IDs shape: torch.Size([1, 25477]) Labels shape: torch.Size([1, 25477]) Final batch size: 1, sequence length: 20579 Attention mask shape: torch.Size([1, 1, 20579, 20579]) Position ids shape: torch.Size([1, 20579]) Input IDs shape: torch.Size([1, 20579]) Labels shape: torch.Size([1, 20579]) Final batch size: 1, sequence length: 11184 Attention mask shape: torch.Size([1, 1, 11184, 11184]) Position ids shape: torch.Size([1, 11184]) Input IDs shape: torch.Size([1, 11184]) Labels shape: torch.Size([1, 11184]) Final batch size: 1, sequence length: 25747 Attention mask shape: torch.Size([1, 1, 25747, 25747]) Position ids shape: torch.Size([1, 25747]) Input IDs shape: torch.Size([1, 25747]) Labels shape: torch.Size([1, 25747]) Final batch size: 1, sequence length: 26663 Attention mask shape: torch.Size([1, 1, 26663, 26663]) Position ids shape: torch.Size([1, 26663]) Input IDs shape: torch.Size([1, 26663]) Labels shape: torch.Size([1, 26663]) Final batch size: 1, sequence length: 25651 Attention mask shape: torch.Size([1, 1, 25651, 25651]) Position ids shape: torch.Size([1, 25651]) Input IDs shape: torch.Size([1, 25651]) Labels shape: torch.Size([1, 25651]) Final batch size: 1, sequence length: 27447 Attention mask shape: torch.Size([1, 1, 27447, 27447]) Position ids shape: torch.Size([1, 27447]) Input IDs shape: torch.Size([1, 27447]) Labels shape: torch.Size([1, 27447]) Final batch size: 1, sequence length: 22887 Attention mask shape: torch.Size([1, 1, 22887, 22887]) Position ids shape: torch.Size([1, 22887]) Input IDs shape: torch.Size([1, 22887]) Labels shape: torch.Size([1, 22887]) Final batch size: 1, sequence length: 15317 Attention mask shape: torch.Size([1, 1, 15317, 15317]) Position ids shape: torch.Size([1, 15317]) Input IDs shape: torch.Size([1, 15317]) Labels shape: torch.Size([1, 15317]) Final batch size: 1, sequence length: 19552 Attention mask shape: torch.Size([1, 1, 19552, 19552]) Position ids shape: torch.Size([1, 19552]) Input IDs shape: torch.Size([1, 19552]) Labels shape: torch.Size([1, 19552]) Final batch size: 1, sequence length: 17395 Attention mask shape: torch.Size([1, 1, 17395, 17395]) Position ids shape: torch.Size([1, 17395]) Input IDs shape: torch.Size([1, 17395]) Labels shape: torch.Size([1, 17395]) Final batch size: 1, sequence length: 27480 Attention mask shape: torch.Size([1, 1, 27480, 27480]) Position ids shape: torch.Size([1, 27480]) Input IDs shape: torch.Size([1, 27480]) Labels shape: torch.Size([1, 27480]) Final batch size: 1, sequence length: 17911 Attention mask shape: torch.Size([1, 1, 17911, 17911]) Position ids shape: torch.Size([1, 17911]) Input IDs shape: torch.Size([1, 17911]) Labels shape: torch.Size([1, 17911]) Final batch size: 1, sequence length: 30981 Attention mask shape: torch.Size([1, 1, 30981, 30981]) Position ids shape: torch.Size([1, 30981]) Input IDs shape: torch.Size([1, 30981]) Labels shape: torch.Size([1, 30981]) Final batch size: 1, sequence length: 19869 Attention mask shape: torch.Size([1, 1, 19869, 19869]) Position ids shape: torch.Size([1, 19869]) Input IDs shape: torch.Size([1, 19869]) Labels shape: torch.Size([1, 19869]) Final batch size: 1, sequence length: 16915 Attention mask shape: torch.Size([1, 1, 16915, 16915]) Position ids shape: torch.Size([1, 16915]) Input IDs shape: torch.Size([1, 16915]) Labels shape: torch.Size([1, 16915]) Final batch size: 1, sequence length: 30031 Attention mask shape: torch.Size([1, 1, 30031, 30031]) Position ids shape: torch.Size([1, 30031]) Input IDs shape: torch.Size([1, 30031]) Labels shape: torch.Size([1, 30031]) Final batch size: 1, sequence length: 28777 Attention mask shape: torch.Size([1, 1, 28777, 28777]) Position ids shape: torch.Size([1, 28777]) Input IDs shape: torch.Size([1, 28777]) Labels shape: torch.Size([1, 28777]) Final batch size: 1, sequence length: 16953 Attention mask shape: torch.Size([1, 1, 16953, 16953]) Position ids shape: torch.Size([1, 16953]) Input IDs shape: torch.Size([1, 16953]) Labels shape: torch.Size([1, 16953]) Final batch size: 1, sequence length: 30601 Attention mask shape: torch.Size([1, 1, 30601, 30601]) Position ids shape: torch.Size([1, 30601]) Input IDs shape: torch.Size([1, 30601]) Labels shape: torch.Size([1, 30601]) Final batch size: 1, sequence length: 24782 Attention mask shape: torch.Size([1, 1, 24782, 24782]) Position ids shape: torch.Size([1, 24782]) Input IDs shape: torch.Size([1, 24782]) Labels shape: torch.Size([1, 24782]) Final batch size: 1, sequence length: 32466 Attention mask shape: torch.Size([1, 1, 32466, 32466]) Position ids shape: torch.Size([1, 32466]) Input IDs shape: torch.Size([1, 32466]) Labels shape: torch.Size([1, 32466]) Final batch size: 1, sequence length: 28060 Attention mask shape: torch.Size([1, 1, 28060, 28060]) Position ids shape: torch.Size([1, 28060]) Input IDs shape: torch.Size([1, 28060]) Labels shape: torch.Size([1, 28060]) Final batch size: 1, sequence length: 18376 Attention mask shape: torch.Size([1, 1, 18376, 18376]) Position ids shape: torch.Size([1, 18376]) Input IDs shape: torch.Size([1, 18376]) Labels shape: torch.Size([1, 18376]) Final batch size: 1, sequence length: 15924 Attention mask shape: torch.Size([1, 1, 15924, 15924]) Position ids shape: torch.Size([1, 15924]) Input IDs shape: torch.Size([1, 15924]) Labels shape: torch.Size([1, 15924]) Final batch size: 1, sequence length: 27385 Attention mask shape: torch.Size([1, 1, 27385, 27385]) Position ids shape: torch.Size([1, 27385]) Input IDs shape: torch.Size([1, 27385]) Labels shape: torch.Size([1, 27385]) Final batch size: 1, sequence length: 12245 Attention mask shape: torch.Size([1, 1, 12245, 12245]) Position ids shape: torch.Size([1, 12245]) Input IDs shape: torch.Size([1, 12245]) Labels shape: torch.Size([1, 12245]) Final batch size: 1, sequence length: 33725 Attention mask shape: torch.Size([1, 1, 33725, 33725]) Position ids shape: torch.Size([1, 33725]) Input IDs shape: torch.Size([1, 33725]) Labels shape: torch.Size([1, 33725]) Final batch size: 1, sequence length: 26033 Attention mask shape: torch.Size([1, 1, 26033, 26033]) Position ids shape: torch.Size([1, 26033]) Input IDs shape: torch.Size([1, 26033]) Labels shape: torch.Size([1, 26033]) Final batch size: 1, sequence length: 14873 Attention mask shape: torch.Size([1, 1, 14873, 14873]) Position ids shape: torch.Size([1, 14873]) Input IDs shape: torch.Size([1, 14873]) Labels shape: torch.Size([1, 14873]) Final batch size: 1, sequence length: 27334 Attention mask shape: torch.Size([1, 1, 27334, 27334]) Position ids shape: torch.Size([1, 27334]) Input IDs shape: torch.Size([1, 27334]) Labels shape: torch.Size([1, 27334]) Final batch size: 1, sequence length: 26479 Attention mask shape: torch.Size([1, 1, 26479, 26479]) Position ids shape: torch.Size([1, 26479]) Input IDs shape: torch.Size([1, 26479]) Labels shape: torch.Size([1, 26479]) Final batch size: 1, sequence length: 37025 Attention mask shape: torch.Size([1, 1, 37025, 37025]) Position ids shape: torch.Size([1, 37025]) Input IDs shape: torch.Size([1, 37025]) Labels shape: torch.Size([1, 37025]) Final batch size: 1, sequence length: 21988 Attention mask shape: torch.Size([1, 1, 21988, 21988]) Position ids shape: torch.Size([1, 21988]) Input IDs shape: torch.Size([1, 21988]) Labels shape: torch.Size([1, 21988]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 37738 Attention mask shape: torch.Size([1, 1, 37738, 37738]) Position ids shape: torch.Size([1, 37738]) Input IDs shape: torch.Size([1, 37738]) Labels shape: torch.Size([1, 37738]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32835 Attention mask shape: torch.Size([1, 1, 32835, 32835]) Position ids shape: torch.Size([1, 32835]) Input IDs shape: torch.Size([1, 32835]) Labels shape: torch.Size([1, 32835]) Final batch size: 1, sequence length: 36175 Attention mask shape: torch.Size([1, 1, 36175, 36175]) Position ids shape: torch.Size([1, 36175]) Input IDs shape: torch.Size([1, 36175]) Labels shape: torch.Size([1, 36175]) Final batch size: 1, sequence length: 37511 Attention mask shape: torch.Size([1, 1, 37511, 37511]) Position ids shape: torch.Size([1, 37511]) Input IDs shape: torch.Size([1, 37511]) Labels shape: torch.Size([1, 37511]) Final batch size: 1, sequence length: 37158 Attention mask shape: torch.Size([1, 1, 37158, 37158]) Position ids shape: torch.Size([1, 37158]) Input IDs shape: torch.Size([1, 37158]) Labels shape: torch.Size([1, 37158]) Final batch size: 1, sequence length: 38986 Attention mask shape: torch.Size([1, 1, 38986, 38986]) Position ids shape: torch.Size([1, 38986]) Input IDs shape: torch.Size([1, 38986]) Labels shape: torch.Size([1, 38986]) Final batch size: 1, sequence length: 40397 Attention mask shape: torch.Size([1, 1, 40397, 40397]) Position ids shape: torch.Size([1, 40397]) Input IDs shape: torch.Size([1, 40397]) Labels shape: torch.Size([1, 40397]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 29586 Attention mask shape: torch.Size([1, 1, 29586, 29586]) Position ids shape: torch.Size([1, 29586]) Input IDs shape: torch.Size([1, 29586]) Labels shape: torch.Size([1, 29586]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40269 Attention mask shape: torch.Size([1, 1, 40269, 40269]) Position ids shape: torch.Size([1, 40269]) Input IDs shape: torch.Size([1, 40269]) Labels shape: torch.Size([1, 40269]) Final batch size: 1, sequence length: 36130 Attention mask shape: torch.Size([1, 1, 36130, 36130]) Position ids shape: torch.Size([1, 36130]) Input IDs shape: torch.Size([1, 36130]) Labels shape: torch.Size([1, 36130]) Final batch size: 1, sequence length: 31879 Attention mask shape: torch.Size([1, 1, 31879, 31879]) Position ids shape: torch.Size([1, 31879]) Input IDs shape: torch.Size([1, 31879]) Labels shape: torch.Size([1, 31879]) Final batch size: 1, sequence length: 40237 Attention mask shape: torch.Size([1, 1, 40237, 40237]) Position ids shape: torch.Size([1, 40237]) Input IDs shape: torch.Size([1, 40237]) Labels shape: torch.Size([1, 40237]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31348 Attention mask shape: torch.Size([1, 1, 31348, 31348]) Position ids shape: torch.Size([1, 31348]) Input IDs shape: torch.Size([1, 31348]) Labels shape: torch.Size([1, 31348]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 27291 Attention mask shape: torch.Size([1, 1, 27291, 27291]) Position ids shape: torch.Size([1, 27291]) Input IDs shape: torch.Size([1, 27291]) Labels shape: torch.Size([1, 27291]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32318 Attention mask shape: torch.Size([1, 1, 32318, 32318]) Position ids shape: torch.Size([1, 32318]) Input IDs shape: torch.Size([1, 32318]) Labels shape: torch.Size([1, 32318]) Final batch size: 1, sequence length: 20509 Attention mask shape: torch.Size([1, 1, 20509, 20509]) Position ids shape: torch.Size([1, 20509]) Input IDs shape: torch.Size([1, 20509]) Labels shape: torch.Size([1, 20509]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17711 Attention mask shape: torch.Size([1, 1, 17711, 17711]) Position ids shape: torch.Size([1, 17711]) Input IDs shape: torch.Size([1, 17711]) Labels shape: torch.Size([1, 17711]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36947 Attention mask shape: torch.Size([1, 1, 36947, 36947]) Position ids shape: torch.Size([1, 36947]) Input IDs shape: torch.Size([1, 36947]) Labels shape: torch.Size([1, 36947]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28631 Attention mask shape: torch.Size([1, 1, 28631, 28631]) Position ids shape: torch.Size([1, 28631]) Input IDs shape: torch.Size([1, 28631]) Labels shape: torch.Size([1, 28631]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32247 Attention mask shape: torch.Size([1, 1, 32247, 32247]) Position ids shape: torch.Size([1, 32247]) Input IDs shape: torch.Size([1, 32247]) Labels shape: torch.Size([1, 32247]) Final batch size: 1, sequence length: 24939 Attention mask shape: torch.Size([1, 1, 24939, 24939]) Position ids shape: torch.Size([1, 24939]) Input IDs shape: torch.Size([1, 24939]) Labels shape: torch.Size([1, 24939]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 23740 Attention mask shape: torch.Size([1, 1, 23740, 23740]) Position ids shape: torch.Size([1, 23740]) Input IDs shape: torch.Size([1, 23740]) Labels shape: torch.Size([1, 23740]) Final batch size: 1, sequence length: 21225 Attention mask shape: torch.Size([1, 1, 21225, 21225]) Position ids shape: torch.Size([1, 21225]) Input IDs shape: torch.Size([1, 21225]) Labels shape: torch.Size([1, 21225]) Final batch size: 1, sequence length: 33125 Attention mask shape: torch.Size([1, 1, 33125, 33125]) Position ids shape: torch.Size([1, 33125]) Input IDs shape: torch.Size([1, 33125]) Labels shape: torch.Size([1, 33125]) Final batch size: 1, sequence length: 10198 Attention mask shape: torch.Size([1, 1, 10198, 10198]) Position ids shape: torch.Size([1, 10198]) Input IDs shape: torch.Size([1, 10198]) Labels shape: torch.Size([1, 10198]) Final batch size: 1, sequence length: 22896 Attention mask shape: torch.Size([1, 1, 22896, 22896]) Position ids shape: torch.Size([1, 22896]) Input IDs shape: torch.Size([1, 22896]) Labels shape: torch.Size([1, 22896]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22205 Attention mask shape: torch.Size([1, 1, 22205, 22205]) Position ids shape: torch.Size([1, 22205]) Input IDs shape: torch.Size([1, 22205]) Labels shape: torch.Size([1, 22205]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28046 Attention mask shape: torch.Size([1, 1, 28046, 28046]) Position ids shape: torch.Size([1, 28046]) Input IDs shape: torch.Size([1, 28046]) Labels shape: torch.Size([1, 28046]) Final batch size: 1, sequence length: 28149 Attention mask shape: torch.Size([1, 1, 28149, 28149]) Position ids shape: torch.Size([1, 28149]) Input IDs shape: torch.Size([1, 28149]) Labels shape: torch.Size([1, 28149]) Final batch size: 1, sequence length: 38360 Attention mask shape: torch.Size([1, 1, 38360, 38360]) Position ids shape: torch.Size([1, 38360]) Input IDs shape: torch.Size([1, 38360]) Labels shape: torch.Size([1, 38360]) Final batch size: 1, sequence length: 32767 Attention mask shape: torch.Size([1, 1, 32767, 32767]) Position ids shape: torch.Size([1, 32767]) Input IDs shape: torch.Size([1, 32767]) Labels shape: torch.Size([1, 32767]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36567 Attention mask shape: torch.Size([1, 1, 36567, 36567]) Position ids shape: torch.Size([1, 36567]) Input IDs shape: torch.Size([1, 36567]) Labels shape: torch.Size([1, 36567]) Final batch size: 1, sequence length: 34268 Attention mask shape: torch.Size([1, 1, 34268, 34268]) Position ids shape: torch.Size([1, 34268]) Input IDs shape: torch.Size([1, 34268]) Labels shape: torch.Size([1, 34268]) Final batch size: 1, sequence length: 32313 Attention mask shape: torch.Size([1, 1, 32313, 32313]) Position ids shape: torch.Size([1, 32313]) Input IDs shape: torch.Size([1, 32313]) Labels shape: torch.Size([1, 32313]) {'loss': 0.3633, 'grad_norm': 0.8625059366617047, 'learning_rate': 9.829629131445342e-06, 'num_tokens': -inf, 'epoch': 1.25} Final batch size: 1, sequence length: 5377 Attention mask shape: torch.Size([1, 1, 5377, 5377]) Position ids shape: torch.Size([1, 5377]) Input IDs shape: torch.Size([1, 5377]) Labels shape: torch.Size([1, 5377]) Final batch size: 1, sequence length: 7998 Attention mask shape: torch.Size([1, 1, 7998, 7998]) Position ids shape: torch.Size([1, 7998]) Input IDs shape: torch.Size([1, 7998]) Labels shape: torch.Size([1, 7998]) Final batch size: 1, sequence length: 7402 Attention mask shape: torch.Size([1, 1, 7402, 7402]) Position ids shape: torch.Size([1, 7402]) Input IDs shape: torch.Size([1, 7402]) Labels shape: torch.Size([1, 7402]) Final batch size: 1, sequence length: 10576 Attention mask shape: torch.Size([1, 1, 10576, 10576]) Position ids shape: torch.Size([1, 10576]) Input IDs shape: torch.Size([1, 10576]) Labels shape: torch.Size([1, 10576]) Final batch size: 1, sequence length: 11709 Attention mask shape: torch.Size([1, 1, 11709, 11709]) Position ids shape: torch.Size([1, 11709]) Input IDs shape: torch.Size([1, 11709]) Labels shape: torch.Size([1, 11709]) Final batch size: 1, sequence length: 11365 Attention mask shape: torch.Size([1, 1, 11365, 11365]) Position ids shape: torch.Size([1, 11365]) Input IDs shape: torch.Size([1, 11365]) Labels shape: torch.Size([1, 11365]) Final batch size: 1, sequence length: 7344 Attention mask shape: torch.Size([1, 1, 7344, 7344]) Position ids shape: torch.Size([1, 7344]) Input IDs shape: torch.Size([1, 7344]) Labels shape: torch.Size([1, 7344]) Final batch size: 1, sequence length: 14127 Attention mask shape: torch.Size([1, 1, 14127, 14127]) Position ids shape: torch.Size([1, 14127]) Input IDs shape: torch.Size([1, 14127]) Labels shape: torch.Size([1, 14127]) Final batch size: 1, sequence length: 8436 Attention mask shape: torch.Size([1, 1, 8436, 8436]) Position ids shape: torch.Size([1, 8436]) Input IDs shape: torch.Size([1, 8436]) Labels shape: torch.Size([1, 8436]) Final batch size: 1, sequence length: 11678 Attention mask shape: torch.Size([1, 1, 11678, 11678]) Position ids shape: torch.Size([1, 11678]) Input IDs shape: torch.Size([1, 11678]) Labels shape: torch.Size([1, 11678]) Final batch size: 1, sequence length: 8655 Attention mask shape: torch.Size([1, 1, 8655, 8655]) Position ids shape: torch.Size([1, 8655]) Input IDs shape: torch.Size([1, 8655]) Labels shape: torch.Size([1, 8655]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 15079 Attention mask shape: torch.Size([1, 1, 15079, 15079]) Position ids shape: torch.Size([1, 15079]) Input IDs shape: torch.Size([1, 15079]) Labels shape: torch.Size([1, 15079]) Final batch size: 1, sequence length: 15308 Attention mask shape: torch.Size([1, 1, 15308, 15308]) Position ids shape: torch.Size([1, 15308]) Input IDs shape: torch.Size([1, 15308]) Labels shape: torch.Size([1, 15308]) Final batch size: 1, sequence length: 15478 Attention mask shape: torch.Size([1, 1, 15478, 15478]) Position ids shape: torch.Size([1, 15478]) Input IDs shape: torch.Size([1, 15478]) Labels shape: torch.Size([1, 15478]) Final batch size: 1, sequence length: 11623 Attention mask shape: torch.Size([1, 1, 11623, 11623]) Position ids shape: torch.Size([1, 11623]) Input IDs shape: torch.Size([1, 11623]) Labels shape: torch.Size([1, 11623]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 16535 Attention mask shape: torch.Size([1, 1, 16535, 16535]) Position ids shape: torch.Size([1, 16535]) Input IDs shape: torch.Size([1, 16535]) Labels shape: torch.Size([1, 16535]) Final batch size: 1, sequence length: 14648 Attention mask shape: torch.Size([1, 1, 14648, 14648]) Position ids shape: torch.Size([1, 14648]) Input IDs shape: torch.Size([1, 14648]) Labels shape: torch.Size([1, 14648]) Final batch size: 1, sequence length: 18307 Attention mask shape: torch.Size([1, 1, 18307, 18307]) Position ids shape: torch.Size([1, 18307]) Input IDs shape: torch.Size([1, 18307]) Labels shape: torch.Size([1, 18307]) Final batch size: 1, sequence length: 13639 Attention mask shape: torch.Size([1, 1, 13639, 13639]) Position ids shape: torch.Size([1, 13639]) Input IDs shape: torch.Size([1, 13639]) Labels shape: torch.Size([1, 13639]) Final batch size: 1, sequence length: 19170 Attention mask shape: torch.Size([1, 1, 19170, 19170]) Position ids shape: torch.Size([1, 19170]) Input IDs shape: torch.Size([1, 19170]) Labels shape: torch.Size([1, 19170]) Final batch size: 1, sequence length: 18836 Attention mask shape: torch.Size([1, 1, 18836, 18836]) Position ids shape: torch.Size([1, 18836]) Input IDs shape: torch.Size([1, 18836]) Labels shape: torch.Size([1, 18836]) Final batch size: 1, sequence length: 19138 Attention mask shape: torch.Size([1, 1, 19138, 19138]) Position ids shape: torch.Size([1, 19138]) Input IDs shape: torch.Size([1, 19138]) Labels shape: torch.Size([1, 19138]) Final batch size: 1, sequence length: 19338 Attention mask shape: torch.Size([1, 1, 19338, 19338]) Position ids shape: torch.Size([1, 19338]) Input IDs shape: torch.Size([1, 19338]) Labels shape: torch.Size([1, 19338]) Final batch size: 1, sequence length: 12006 Attention mask shape: torch.Size([1, 1, 12006, 12006]) Position ids shape: torch.Size([1, 12006]) Input IDs shape: torch.Size([1, 12006]) Labels shape: torch.Size([1, 12006]) Final batch size: 1, sequence length: 19330 Attention mask shape: torch.Size([1, 1, 19330, 19330]) Position ids shape: torch.Size([1, 19330]) Input IDs shape: torch.Size([1, 19330]) Labels shape: torch.Size([1, 19330]) Final batch size: 1, sequence length: 8839 Attention mask shape: torch.Size([1, 1, 8839, 8839]) Position ids shape: torch.Size([1, 8839]) Input IDs shape: torch.Size([1, 8839]) Labels shape: torch.Size([1, 8839]) Final batch size: 1, sequence length: 18953 Attention mask shape: torch.Size([1, 1, 18953, 18953]) Position ids shape: torch.Size([1, 18953]) Input IDs shape: torch.Size([1, 18953]) Labels shape: torch.Size([1, 18953]) Final batch size: 1, sequence length: 17058 Attention mask shape: torch.Size([1, 1, 17058, 17058]) Position ids shape: torch.Size([1, 17058]) Input IDs shape: torch.Size([1, 17058]) Labels shape: torch.Size([1, 17058]) Final batch size: 1, sequence length: 20770 Attention mask shape: torch.Size([1, 1, 20770, 20770]) Position ids shape: torch.Size([1, 20770]) Input IDs shape: torch.Size([1, 20770]) Labels shape: torch.Size([1, 20770]) Final batch size: 1, sequence length: 22561 Attention mask shape: torch.Size([1, 1, 22561, 22561]) Position ids shape: torch.Size([1, 22561]) Input IDs shape: torch.Size([1, 22561]) Labels shape: torch.Size([1, 22561]) Final batch size: 1, sequence length: 20714 Attention mask shape: torch.Size([1, 1, 20714, 20714]) Position ids shape: torch.Size([1, 20714]) Input IDs shape: torch.Size([1, 20714]) Labels shape: torch.Size([1, 20714]) Final batch size: 1, sequence length: 18395 Attention mask shape: torch.Size([1, 1, 18395, 18395]) Position ids shape: torch.Size([1, 18395]) Input IDs shape: torch.Size([1, 18395]) Labels shape: torch.Size([1, 18395]) Final batch size: 1, sequence length: 21982 Attention mask shape: torch.Size([1, 1, 21982, 21982]) Position ids shape: torch.Size([1, 21982]) Input IDs shape: torch.Size([1, 21982]) Labels shape: torch.Size([1, 21982]) Final batch size: 1, sequence length: 21858 Attention mask shape: torch.Size([1, 1, 21858, 21858]) Position ids shape: torch.Size([1, 21858]) Input IDs shape: torch.Size([1, 21858]) Labels shape: torch.Size([1, 21858]) Final batch size: 1, sequence length: 20854 Attention mask shape: torch.Size([1, 1, 20854, 20854]) Position ids shape: torch.Size([1, 20854]) Input IDs shape: torch.Size([1, 20854]) Labels shape: torch.Size([1, 20854]) Final batch size: 1, sequence length: 18823 Attention mask shape: torch.Size([1, 1, 18823, 18823]) Position ids shape: torch.Size([1, 18823]) Input IDs shape: torch.Size([1, 18823]) Labels shape: torch.Size([1, 18823]) Final batch size: 1, sequence length: 18325 Attention mask shape: torch.Size([1, 1, 18325, 18325]) Position ids shape: torch.Size([1, 18325]) Input IDs shape: torch.Size([1, 18325]) Labels shape: torch.Size([1, 18325]) Final batch size: 1, sequence length: 18527 Attention mask shape: torch.Size([1, 1, 18527, 18527]) Position ids shape: torch.Size([1, 18527]) Input IDs shape: torch.Size([1, 18527]) Labels shape: torch.Size([1, 18527]) Final batch size: 1, sequence length: 24248 Attention mask shape: torch.Size([1, 1, 24248, 24248]) Position ids shape: torch.Size([1, 24248]) Input IDs shape: torch.Size([1, 24248]) Labels shape: torch.Size([1, 24248]) Final batch size: 1, sequence length: 21405 Attention mask shape: torch.Size([1, 1, 21405, 21405]) Position ids shape: torch.Size([1, 21405]) Input IDs shape: torch.Size([1, 21405]) Labels shape: torch.Size([1, 21405]) Final batch size: 1, sequence length: 23560 Attention mask shape: torch.Size([1, 1, 23560, 23560]) Position ids shape: torch.Size([1, 23560]) Input IDs shape: torch.Size([1, 23560]) Labels shape: torch.Size([1, 23560]) Final batch size: 1, sequence length: 15913 Attention mask shape: torch.Size([1, 1, 15913, 15913]) Position ids shape: torch.Size([1, 15913]) Input IDs shape: torch.Size([1, 15913]) Labels shape: torch.Size([1, 15913]) Final batch size: 1, sequence length: 26356 Attention mask shape: torch.Size([1, 1, 26356, 26356]) Position ids shape: torch.Size([1, 26356]) Input IDs shape: torch.Size([1, 26356]) Labels shape: torch.Size([1, 26356]) Final batch size: 1, sequence length: 23694 Attention mask shape: torch.Size([1, 1, 23694, 23694]) Position ids shape: torch.Size([1, 23694]) Input IDs shape: torch.Size([1, 23694]) Labels shape: torch.Size([1, 23694]) Final batch size: 1, sequence length: 25451 Attention mask shape: torch.Size([1, 1, 25451, 25451]) Position ids shape: torch.Size([1, 25451]) Input IDs shape: torch.Size([1, 25451]) Labels shape: torch.Size([1, 25451]) Final batch size: 1, sequence length: 24433 Attention mask shape: torch.Size([1, 1, 24433, 24433]) Position ids shape: torch.Size([1, 24433]) Input IDs shape: torch.Size([1, 24433]) Labels shape: torch.Size([1, 24433]) Final batch size: 1, sequence length: 29098 Attention mask shape: torch.Size([1, 1, 29098, 29098]) Position ids shape: torch.Size([1, 29098]) Input IDs shape: torch.Size([1, 29098]) Labels shape: torch.Size([1, 29098]) Final batch size: 1, sequence length: 16060 Attention mask shape: torch.Size([1, 1, 16060, 16060]) Position ids shape: torch.Size([1, 16060]) Input IDs shape: torch.Size([1, 16060]) Labels shape: torch.Size([1, 16060]) Final batch size: 1, sequence length: 22735 Attention mask shape: torch.Size([1, 1, 22735, 22735]) Position ids shape: torch.Size([1, 22735]) Input IDs shape: torch.Size([1, 22735]) Labels shape: torch.Size([1, 22735]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 5801 Attention mask shape: torch.Size([1, 1, 5801, 5801]) Position ids shape: torch.Size([1, 5801]) Input IDs shape: torch.Size([1, 5801]) Labels shape: torch.Size([1, 5801]) Final batch size: 1, sequence length: 9380 Attention mask shape: torch.Size([1, 1, 9380, 9380]) Position ids shape: torch.Size([1, 9380]) Input IDs shape: torch.Size([1, 9380]) Labels shape: torch.Size([1, 9380]) Final batch size: 1, sequence length: 30687 Attention mask shape: torch.Size([1, 1, 30687, 30687]) Position ids shape: torch.Size([1, 30687]) Input IDs shape: torch.Size([1, 30687]) Labels shape: torch.Size([1, 30687]) Final batch size: 1, sequence length: 21615 Attention mask shape: torch.Size([1, 1, 21615, 21615]) Position ids shape: torch.Size([1, 21615]) Input IDs shape: torch.Size([1, 21615]) Labels shape: torch.Size([1, 21615]) Final batch size: 1, sequence length: 26072 Attention mask shape: torch.Size([1, 1, 26072, 26072]) Position ids shape: torch.Size([1, 26072]) Input IDs shape: torch.Size([1, 26072]) Labels shape: torch.Size([1, 26072]) Final batch size: 1, sequence length: 19239 Attention mask shape: torch.Size([1, 1, 19239, 19239]) Position ids shape: torch.Size([1, 19239]) Input IDs shape: torch.Size([1, 19239]) Labels shape: torch.Size([1, 19239]) Final batch size: 1, sequence length: 15875 Attention mask shape: torch.Size([1, 1, 15875, 15875]) Position ids shape: torch.Size([1, 15875]) Input IDs shape: torch.Size([1, 15875]) Labels shape: torch.Size([1, 15875]) Final batch size: 1, sequence length: 32885 Attention mask shape: torch.Size([1, 1, 32885, 32885]) Position ids shape: torch.Size([1, 32885]) Input IDs shape: torch.Size([1, 32885]) Labels shape: torch.Size([1, 32885]) Final batch size: 1, sequence length: 28684 Attention mask shape: torch.Size([1, 1, 28684, 28684]) Position ids shape: torch.Size([1, 28684]) Input IDs shape: torch.Size([1, 28684]) Labels shape: torch.Size([1, 28684]) Final batch size: 1, sequence length: 19045 Attention mask shape: torch.Size([1, 1, 19045, 19045]) Position ids shape: torch.Size([1, 19045]) Input IDs shape: torch.Size([1, 19045]) Labels shape: torch.Size([1, 19045]) Final batch size: 1, sequence length: 32328 Attention mask shape: torch.Size([1, 1, 32328, 32328]) Position ids shape: torch.Size([1, 32328]) Input IDs shape: torch.Size([1, 32328]) Labels shape: torch.Size([1, 32328]) Final batch size: 1, sequence length: 17465 Attention mask shape: torch.Size([1, 1, 17465, 17465]) Position ids shape: torch.Size([1, 17465]) Input IDs shape: torch.Size([1, 17465]) Labels shape: torch.Size([1, 17465]) Final batch size: 1, sequence length: 29355 Final batch size: 1, sequence length: 23881 Attention mask shape: torch.Size([1, 1, 29355, 29355]) Position ids shape: torch.Size([1, 29355]) Input IDs shape: torch.Size([1, 29355]) Labels shape: torch.Size([1, 29355]) Attention mask shape: torch.Size([1, 1, 23881, 23881]) Position ids shape: torch.Size([1, 23881]) Input IDs shape: torch.Size([1, 23881]) Labels shape: torch.Size([1, 23881]) Final batch size: 1, sequence length: 31414 Attention mask shape: torch.Size([1, 1, 31414, 31414]) Position ids shape: torch.Size([1, 31414]) Input IDs shape: torch.Size([1, 31414]) Labels shape: torch.Size([1, 31414]) Final batch size: 1, sequence length: 16050 Attention mask shape: torch.Size([1, 1, 16050, 16050]) Position ids shape: torch.Size([1, 16050]) Input IDs shape: torch.Size([1, 16050]) Labels shape: torch.Size([1, 16050]) Final batch size: 1, sequence length: 33871 Attention mask shape: torch.Size([1, 1, 33871, 33871]) Position ids shape: torch.Size([1, 33871]) Input IDs shape: torch.Size([1, 33871]) Labels shape: torch.Size([1, 33871]) Final batch size: 1, sequence length: 35397 Attention mask shape: torch.Size([1, 1, 35397, 35397]) Position ids shape: torch.Size([1, 35397]) Input IDs shape: torch.Size([1, 35397]) Labels shape: torch.Size([1, 35397]) Final batch size: 1, sequence length: 30689 Attention mask shape: torch.Size([1, 1, 30689, 30689]) Position ids shape: torch.Size([1, 30689]) Input IDs shape: torch.Size([1, 30689]) Labels shape: torch.Size([1, 30689]) Final batch size: 1, sequence length: 31860 Attention mask shape: torch.Size([1, 1, 31860, 31860]) Position ids shape: torch.Size([1, 31860]) Input IDs shape: torch.Size([1, 31860]) Labels shape: torch.Size([1, 31860]) Final batch size: 1, sequence length: 34581 Attention mask shape: torch.Size([1, 1, 34581, 34581]) Position ids shape: torch.Size([1, 34581]) Input IDs shape: torch.Size([1, 34581]) Labels shape: torch.Size([1, 34581]) Final batch size: 1, sequence length: 27972 Attention mask shape: torch.Size([1, 1, 27972, 27972]) Position ids shape: torch.Size([1, 27972]) Input IDs shape: torch.Size([1, 27972]) Labels shape: torch.Size([1, 27972]) Final batch size: 1, sequence length: 26689 Attention mask shape: torch.Size([1, 1, 26689, 26689]) Position ids shape: torch.Size([1, 26689]) Input IDs shape: torch.Size([1, 26689]) Labels shape: torch.Size([1, 26689]) Final batch size: 1, sequence length: 32529 Attention mask shape: torch.Size([1, 1, 32529, 32529]) Position ids shape: torch.Size([1, 32529]) Input IDs shape: torch.Size([1, 32529]) Labels shape: torch.Size([1, 32529]) Final batch size: 1, sequence length: 29150 Attention mask shape: torch.Size([1, 1, 29150, 29150]) Position ids shape: torch.Size([1, 29150]) Input IDs shape: torch.Size([1, 29150]) Labels shape: torch.Size([1, 29150]) Final batch size: 1, sequence length: 30346 Final batch size: 1, sequence length: 32636 Attention mask shape: torch.Size([1, 1, 32636, 32636]) Position ids shape: torch.Size([1, 32636]) Input IDs shape: torch.Size([1, 32636]) Labels shape: torch.Size([1, 32636]) Final batch size: 1, sequence length: 18177 Attention mask shape: torch.Size([1, 1, 18177, 18177]) Position ids shape: torch.Size([1, 18177]) Input IDs shape: torch.Size([1, 18177]) Labels shape: torch.Size([1, 18177]) Final batch size: 1, sequence length: 34921 Attention mask shape: torch.Size([1, 1, 34921, 34921]) Position ids shape: torch.Size([1, 34921]) Input IDs shape: torch.Size([1, 34921]) Labels shape: torch.Size([1, 34921]) Attention mask shape: torch.Size([1, 1, 30346, 30346]) Position ids shape: torch.Size([1, 30346]) Input IDs shape: torch.Size([1, 30346]) Labels shape: torch.Size([1, 30346]) Final batch size: 1, sequence length: 31712 Attention mask shape: torch.Size([1, 1, 31712, 31712]) Position ids shape: torch.Size([1, 31712]) Input IDs shape: torch.Size([1, 31712]) Labels shape: torch.Size([1, 31712]) Final batch size: 1, sequence length: 22264 Attention mask shape: torch.Size([1, 1, 22264, 22264]) Position ids shape: torch.Size([1, 22264]) Input IDs shape: torch.Size([1, 22264]) Labels shape: torch.Size([1, 22264]) Final batch size: 1, sequence length: 15077 Attention mask shape: torch.Size([1, 1, 15077, 15077]) Position ids shape: torch.Size([1, 15077]) Input IDs shape: torch.Size([1, 15077]) Labels shape: torch.Size([1, 15077]) Final batch size: 1, sequence length: 18963 Attention mask shape: torch.Size([1, 1, 18963, 18963]) Position ids shape: torch.Size([1, 18963]) Input IDs shape: torch.Size([1, 18963]) Labels shape: torch.Size([1, 18963]) Final batch size: 1, sequence length: 24781 Attention mask shape: torch.Size([1, 1, 24781, 24781]) Position ids shape: torch.Size([1, 24781]) Input IDs shape: torch.Size([1, 24781]) Labels shape: torch.Size([1, 24781]) Final batch size: 1, sequence length: 16025 Attention mask shape: torch.Size([1, 1, 16025, 16025]) Position ids shape: torch.Size([1, 16025]) Input IDs shape: torch.Size([1, 16025]) Labels shape: torch.Size([1, 16025]) Final batch size: 1, sequence length: 14976 Attention mask shape: torch.Size([1, 1, 14976, 14976]) Position ids shape: torch.Size([1, 14976]) Input IDs shape: torch.Size([1, 14976]) Labels shape: torch.Size([1, 14976]) Final batch size: 1, sequence length: 25741 Attention mask shape: torch.Size([1, 1, 25741, 25741]) Position ids shape: torch.Size([1, 25741]) Input IDs shape: torch.Size([1, 25741]) Labels shape: torch.Size([1, 25741]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17243 Attention mask shape: torch.Size([1, 1, 17243, 17243]) Position ids shape: torch.Size([1, 17243]) Input IDs shape: torch.Size([1, 17243]) Labels shape: torch.Size([1, 17243]) Final batch size: 1, sequence length: 31022 Attention mask shape: torch.Size([1, 1, 31022, 31022]) Position ids shape: torch.Size([1, 31022]) Input IDs shape: torch.Size([1, 31022]) Labels shape: torch.Size([1, 31022]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28018 Attention mask shape: torch.Size([1, 1, 28018, 28018]) Position ids shape: torch.Size([1, 28018]) Input IDs shape: torch.Size([1, 28018]) Labels shape: torch.Size([1, 28018]) Final batch size: 1, sequence length: 26922 Attention mask shape: torch.Size([1, 1, 26922, 26922]) Position ids shape: torch.Size([1, 26922]) Input IDs shape: torch.Size([1, 26922]) Labels shape: torch.Size([1, 26922]) Final batch size: 1, sequence length: 24002 Attention mask shape: torch.Size([1, 1, 24002, 24002]) Position ids shape: torch.Size([1, 24002]) Input IDs shape: torch.Size([1, 24002]) Labels shape: torch.Size([1, 24002]) Final batch size: 1, sequence length: 40488 Attention mask shape: torch.Size([1, 1, 40488, 40488]) Position ids shape: torch.Size([1, 40488]) Input IDs shape: torch.Size([1, 40488]) Labels shape: torch.Size([1, 40488]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36596 Attention mask shape: torch.Size([1, 1, 36596, 36596]) Position ids shape: torch.Size([1, 36596]) Input IDs shape: torch.Size([1, 36596]) Labels shape: torch.Size([1, 36596]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25674 Attention mask shape: torch.Size([1, 1, 25674, 25674]) Position ids shape: torch.Size([1, 25674]) Input IDs shape: torch.Size([1, 25674]) Labels shape: torch.Size([1, 25674]) Final batch size: 1, sequence length: 28397 Attention mask shape: torch.Size([1, 1, 28397, 28397]) Position ids shape: torch.Size([1, 28397]) Input IDs shape: torch.Size([1, 28397]) Labels shape: torch.Size([1, 28397]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 11611 Attention mask shape: torch.Size([1, 1, 11611, 11611]) Position ids shape: torch.Size([1, 11611]) Input IDs shape: torch.Size([1, 11611]) Labels shape: torch.Size([1, 11611]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34853 Attention mask shape: torch.Size([1, 1, 34853, 34853]) Position ids shape: torch.Size([1, 34853]) Input IDs shape: torch.Size([1, 34853]) Labels shape: torch.Size([1, 34853]) Final batch size: 1, sequence length: 24043 Attention mask shape: torch.Size([1, 1, 24043, 24043]) Position ids shape: torch.Size([1, 24043]) Input IDs shape: torch.Size([1, 24043]) Labels shape: torch.Size([1, 24043]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16337 Attention mask shape: torch.Size([1, 1, 16337, 16337]) Position ids shape: torch.Size([1, 16337]) Input IDs shape: torch.Size([1, 16337]) Labels shape: torch.Size([1, 16337]) Final batch size: 1, sequence length: 28086 Attention mask shape: torch.Size([1, 1, 28086, 28086]) Position ids shape: torch.Size([1, 28086]) Input IDs shape: torch.Size([1, 28086]) Labels shape: torch.Size([1, 28086]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 12218 Attention mask shape: torch.Size([1, 1, 12218, 12218]) Position ids shape: torch.Size([1, 12218]) Input IDs shape: torch.Size([1, 12218]) Labels shape: torch.Size([1, 12218]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36334 Attention mask shape: torch.Size([1, 1, 36334, 36334]) Position ids shape: torch.Size([1, 36334]) Input IDs shape: torch.Size([1, 36334]) Labels shape: torch.Size([1, 36334]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32917 Attention mask shape: torch.Size([1, 1, 32917, 32917]) Position ids shape: torch.Size([1, 32917]) Input IDs shape: torch.Size([1, 32917]) Labels shape: torch.Size([1, 32917]) {'loss': 0.3433, 'grad_norm': 0.7266844746570459, 'learning_rate': 9.755282581475769e-06, 'num_tokens': -inf, 'epoch': 1.38} Final batch size: 1, sequence length: 5818 Attention mask shape: torch.Size([1, 1, 5818, 5818]) Position ids shape: torch.Size([1, 5818]) Input IDs shape: torch.Size([1, 5818]) Labels shape: torch.Size([1, 5818]) Final batch size: 1, sequence length: 6215 Attention mask shape: torch.Size([1, 1, 6215, 6215]) Position ids shape: torch.Size([1, 6215]) Input IDs shape: torch.Size([1, 6215]) Labels shape: torch.Size([1, 6215]) Final batch size: 1, sequence length: 6871 Attention mask shape: torch.Size([1, 1, 6871, 6871]) Position ids shape: torch.Size([1, 6871]) Input IDs shape: torch.Size([1, 6871]) Labels shape: torch.Size([1, 6871]) Final batch size: 1, sequence length: 8623 Attention mask shape: torch.Size([1, 1, 8623, 8623]) Position ids shape: torch.Size([1, 8623]) Input IDs shape: torch.Size([1, 8623]) Labels shape: torch.Size([1, 8623]) Final batch size: 1, sequence length: 5917 Attention mask shape: torch.Size([1, 1, 5917, 5917]) Position ids shape: torch.Size([1, 5917]) Input IDs shape: torch.Size([1, 5917]) Labels shape: torch.Size([1, 5917]) Final batch size: 1, sequence length: 6034 Attention mask shape: torch.Size([1, 1, 6034, 6034]) Position ids shape: torch.Size([1, 6034]) Input IDs shape: torch.Size([1, 6034]) Labels shape: torch.Size([1, 6034]) Final batch size: 1, sequence length: 11616 Attention mask shape: torch.Size([1, 1, 11616, 11616]) Position ids shape: torch.Size([1, 11616]) Input IDs shape: torch.Size([1, 11616]) Labels shape: torch.Size([1, 11616]) Final batch size: 1, sequence length: 12275 Attention mask shape: torch.Size([1, 1, 12275, 12275]) Position ids shape: torch.Size([1, 12275]) Input IDs shape: torch.Size([1, 12275]) Labels shape: torch.Size([1, 12275]) Final batch size: 1, sequence length: 12317 Attention mask shape: torch.Size([1, 1, 12317, 12317]) Position ids shape: torch.Size([1, 12317]) Input IDs shape: torch.Size([1, 12317]) Labels shape: torch.Size([1, 12317]) Final batch size: 1, sequence length: 9517 Attention mask shape: torch.Size([1, 1, 9517, 9517]) Position ids shape: torch.Size([1, 9517]) Input IDs shape: torch.Size([1, 9517]) Labels shape: torch.Size([1, 9517]) Final batch size: 1, sequence length: 10107 Attention mask shape: torch.Size([1, 1, 10107, 10107]) Position ids shape: torch.Size([1, 10107]) Input IDs shape: torch.Size([1, 10107]) Labels shape: torch.Size([1, 10107]) Final batch size: 1, sequence length: 11648 Attention mask shape: torch.Size([1, 1, 11648, 11648]) Position ids shape: torch.Size([1, 11648]) Input IDs shape: torch.Size([1, 11648]) Labels shape: torch.Size([1, 11648]) Final batch size: 1, sequence length: 8117 Attention mask shape: torch.Size([1, 1, 8117, 8117]) Position ids shape: torch.Size([1, 8117]) Input IDs shape: torch.Size([1, 8117]) Labels shape: torch.Size([1, 8117]) Final batch size: 1, sequence length: 15029 Attention mask shape: torch.Size([1, 1, 15029, 15029]) Position ids shape: torch.Size([1, 15029]) Input IDs shape: torch.Size([1, 15029]) Labels shape: torch.Size([1, 15029]) Final batch size: 1, sequence length: 10136 Attention mask shape: torch.Size([1, 1, 10136, 10136]) Position ids shape: torch.Size([1, 10136]) Input IDs shape: torch.Size([1, 10136]) Labels shape: torch.Size([1, 10136]) Final batch size: 1, sequence length: 13202 Attention mask shape: torch.Size([1, 1, 13202, 13202]) Position ids shape: torch.Size([1, 13202]) Input IDs shape: torch.Size([1, 13202]) Labels shape: torch.Size([1, 13202]) Final batch size: 1, sequence length: 16331 Attention mask shape: torch.Size([1, 1, 16331, 16331]) Position ids shape: torch.Size([1, 16331]) Input IDs shape: torch.Size([1, 16331]) Labels shape: torch.Size([1, 16331]) Final batch size: 1, sequence length: 15066 Attention mask shape: torch.Size([1, 1, 15066, 15066]) Position ids shape: torch.Size([1, 15066]) Input IDs shape: torch.Size([1, 15066]) Labels shape: torch.Size([1, 15066]) Final batch size: 1, sequence length: 12813 Attention mask shape: torch.Size([1, 1, 12813, 12813]) Position ids shape: torch.Size([1, 12813]) Input IDs shape: torch.Size([1, 12813]) Labels shape: torch.Size([1, 12813]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 13428 Attention mask shape: torch.Size([1, 1, 13428, 13428]) Position ids shape: torch.Size([1, 13428]) Input IDs shape: torch.Size([1, 13428]) Labels shape: torch.Size([1, 13428]) Final batch size: 1, sequence length: 12553 Attention mask shape: torch.Size([1, 1, 12553, 12553]) Position ids shape: torch.Size([1, 12553]) Input IDs shape: torch.Size([1, 12553]) Labels shape: torch.Size([1, 12553]) Final batch size: 1, sequence length: 17587 Attention mask shape: torch.Size([1, 1, 17587, 17587]) Position ids shape: torch.Size([1, 17587]) Input IDs shape: torch.Size([1, 17587]) Labels shape: torch.Size([1, 17587]) Final batch size: 1, sequence length: 15499 Attention mask shape: torch.Size([1, 1, 15499, 15499]) Position ids shape: torch.Size([1, 15499]) Input IDs shape: torch.Size([1, 15499]) Labels shape: torch.Size([1, 15499]) Final batch size: 1, sequence length: 18414 Attention mask shape: torch.Size([1, 1, 18414, 18414]) Position ids shape: torch.Size([1, 18414]) Input IDs shape: torch.Size([1, 18414]) Labels shape: torch.Size([1, 18414]) Final batch size: 1, sequence length: 19847 Attention mask shape: torch.Size([1, 1, 19847, 19847]) Position ids shape: torch.Size([1, 19847]) Input IDs shape: torch.Size([1, 19847]) Labels shape: torch.Size([1, 19847]) Final batch size: 1, sequence length: 18469 Attention mask shape: torch.Size([1, 1, 18469, 18469]) Position ids shape: torch.Size([1, 18469]) Input IDs shape: torch.Size([1, 18469]) Labels shape: torch.Size([1, 18469]) Final batch size: 1, sequence length: 18438 Attention mask shape: torch.Size([1, 1, 18438, 18438]) Position ids shape: torch.Size([1, 18438]) Input IDs shape: torch.Size([1, 18438]) Labels shape: torch.Size([1, 18438]) Final batch size: 1, sequence length: 19512 Attention mask shape: torch.Size([1, 1, 19512, 19512]) Position ids shape: torch.Size([1, 19512]) Input IDs shape: torch.Size([1, 19512]) Labels shape: torch.Size([1, 19512]) Final batch size: 1, sequence length: 21556 Attention mask shape: torch.Size([1, 1, 21556, 21556]) Position ids shape: torch.Size([1, 21556]) Input IDs shape: torch.Size([1, 21556]) Labels shape: torch.Size([1, 21556]) Final batch size: 1, sequence length: 21028 Attention mask shape: torch.Size([1, 1, 21028, 21028]) Position ids shape: torch.Size([1, 21028]) Input IDs shape: torch.Size([1, 21028]) Labels shape: torch.Size([1, 21028]) Final batch size: 1, sequence length: 22079 Attention mask shape: torch.Size([1, 1, 22079, 22079]) Position ids shape: torch.Size([1, 22079]) Input IDs shape: torch.Size([1, 22079]) Labels shape: torch.Size([1, 22079]) Final batch size: 1, sequence length: 22618 Attention mask shape: torch.Size([1, 1, 22618, 22618]) Position ids shape: torch.Size([1, 22618]) Input IDs shape: torch.Size([1, 22618]) Labels shape: torch.Size([1, 22618]) Final batch size: 1, sequence length: 19221 Attention mask shape: torch.Size([1, 1, 19221, 19221]) Position ids shape: torch.Size([1, 19221]) Input IDs shape: torch.Size([1, 19221]) Labels shape: torch.Size([1, 19221]) Final batch size: 1, sequence length: 15226 Attention mask shape: torch.Size([1, 1, 15226, 15226]) Position ids shape: torch.Size([1, 15226]) Input IDs shape: torch.Size([1, 15226]) Labels shape: torch.Size([1, 15226]) Final batch size: 1, sequence length: 22138 Attention mask shape: torch.Size([1, 1, 22138, 22138]) Position ids shape: torch.Size([1, 22138]) Input IDs shape: torch.Size([1, 22138]) Labels shape: torch.Size([1, 22138]) Final batch size: 1, sequence length: 23973 Attention mask shape: torch.Size([1, 1, 23973, 23973]) Position ids shape: torch.Size([1, 23973]) Input IDs shape: torch.Size([1, 23973]) Labels shape: torch.Size([1, 23973]) Final batch size: 1, sequence length: 22718 Attention mask shape: torch.Size([1, 1, 22718, 22718]) Position ids shape: torch.Size([1, 22718]) Input IDs shape: torch.Size([1, 22718]) Labels shape: torch.Size([1, 22718]) Final batch size: 1, sequence length: 7584 Attention mask shape: torch.Size([1, 1, 7584, 7584]) Position ids shape: torch.Size([1, 7584]) Input IDs shape: torch.Size([1, 7584]) Labels shape: torch.Size([1, 7584]) Final batch size: 1, sequence length: 25622 Attention mask shape: torch.Size([1, 1, 25622, 25622]) Position ids shape: torch.Size([1, 25622]) Input IDs shape: torch.Size([1, 25622]) Labels shape: torch.Size([1, 25622]) Final batch size: 1, sequence length: 25999 Attention mask shape: torch.Size([1, 1, 25999, 25999]) Position ids shape: torch.Size([1, 25999]) Input IDs shape: torch.Size([1, 25999]) Labels shape: torch.Size([1, 25999]) Final batch size: 1, sequence length: 24694 Attention mask shape: torch.Size([1, 1, 24694, 24694]) Position ids shape: torch.Size([1, 24694]) Input IDs shape: torch.Size([1, 24694]) Labels shape: torch.Size([1, 24694]) Final batch size: 1, sequence length: 14784 Attention mask shape: torch.Size([1, 1, 14784, 14784]) Position ids shape: torch.Size([1, 14784]) Input IDs shape: torch.Size([1, 14784]) Labels shape: torch.Size([1, 14784]) Final batch size: 1, sequence length: 16299 Attention mask shape: torch.Size([1, 1, 16299, 16299]) Position ids shape: torch.Size([1, 16299]) Input IDs shape: torch.Size([1, 16299]) Labels shape: torch.Size([1, 16299]) Final batch size: 1, sequence length: 23766 Attention mask shape: torch.Size([1, 1, 23766, 23766]) Position ids shape: torch.Size([1, 23766]) Input IDs shape: torch.Size([1, 23766]) Labels shape: torch.Size([1, 23766]) Final batch size: 1, sequence length: 19428 Attention mask shape: torch.Size([1, 1, 19428, 19428]) Position ids shape: torch.Size([1, 19428]) Input IDs shape: torch.Size([1, 19428]) Labels shape: torch.Size([1, 19428]) Final batch size: 1, sequence length: 24499 Attention mask shape: torch.Size([1, 1, 24499, 24499]) Position ids shape: torch.Size([1, 24499]) Input IDs shape: torch.Size([1, 24499]) Labels shape: torch.Size([1, 24499]) Final batch size: 1, sequence length: 22763 Attention mask shape: torch.Size([1, 1, 22763, 22763]) Position ids shape: torch.Size([1, 22763]) Input IDs shape: torch.Size([1, 22763]) Labels shape: torch.Size([1, 22763]) Final batch size: 1, sequence length: 15221 Attention mask shape: torch.Size([1, 1, 15221, 15221]) Position ids shape: torch.Size([1, 15221]) Input IDs shape: torch.Size([1, 15221]) Labels shape: torch.Size([1, 15221]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 28497 Attention mask shape: torch.Size([1, 1, 28497, 28497]) Position ids shape: torch.Size([1, 28497]) Input IDs shape: torch.Size([1, 28497]) Labels shape: torch.Size([1, 28497]) Final batch size: 1, sequence length: 27566 Attention mask shape: torch.Size([1, 1, 27566, 27566]) Position ids shape: torch.Size([1, 27566]) Input IDs shape: torch.Size([1, 27566]) Labels shape: torch.Size([1, 27566]) Final batch size: 1, sequence length: 22235 Attention mask shape: torch.Size([1, 1, 22235, 22235]) Position ids shape: torch.Size([1, 22235]) Input IDs shape: torch.Size([1, 22235]) Labels shape: torch.Size([1, 22235]) Final batch size: 1, sequence length: 27327 Attention mask shape: torch.Size([1, 1, 27327, 27327]) Position ids shape: torch.Size([1, 27327]) Input IDs shape: torch.Size([1, 27327]) Labels shape: torch.Size([1, 27327]) Final batch size: 1, sequence length: 24909 Attention mask shape: torch.Size([1, 1, 24909, 24909]) Position ids shape: torch.Size([1, 24909]) Input IDs shape: torch.Size([1, 24909]) Labels shape: torch.Size([1, 24909]) Final batch size: 1, sequence length: 15184 Attention mask shape: torch.Size([1, 1, 15184, 15184]) Position ids shape: torch.Size([1, 15184]) Input IDs shape: torch.Size([1, 15184]) Labels shape: torch.Size([1, 15184]) Final batch size: 1, sequence length: 28749 Attention mask shape: torch.Size([1, 1, 28749, 28749]) Position ids shape: torch.Size([1, 28749]) Input IDs shape: torch.Size([1, 28749]) Labels shape: torch.Size([1, 28749]) Final batch size: 1, sequence length: 30356 Attention mask shape: torch.Size([1, 1, 30356, 30356]) Position ids shape: torch.Size([1, 30356]) Input IDs shape: torch.Size([1, 30356]) Labels shape: torch.Size([1, 30356]) Final batch size: 1, sequence length: 15588 Attention mask shape: torch.Size([1, 1, 15588, 15588]) Position ids shape: torch.Size([1, 15588]) Input IDs shape: torch.Size([1, 15588]) Labels shape: torch.Size([1, 15588]) Final batch size: 1, sequence length: 21826 Attention mask shape: torch.Size([1, 1, 21826, 21826]) Position ids shape: torch.Size([1, 21826]) Input IDs shape: torch.Size([1, 21826]) Labels shape: torch.Size([1, 21826]) Final batch size: 1, sequence length: 11795 Attention mask shape: torch.Size([1, 1, 11795, 11795]) Position ids shape: torch.Size([1, 11795]) Input IDs shape: torch.Size([1, 11795]) Labels shape: torch.Size([1, 11795]) Final batch size: 1, sequence length: 23851 Attention mask shape: torch.Size([1, 1, 23851, 23851]) Position ids shape: torch.Size([1, 23851]) Input IDs shape: torch.Size([1, 23851]) Labels shape: torch.Size([1, 23851]) Final batch size: 1, sequence length: 10269 Attention mask shape: torch.Size([1, 1, 10269, 10269]) Position ids shape: torch.Size([1, 10269]) Input IDs shape: torch.Size([1, 10269]) Labels shape: torch.Size([1, 10269]) Final batch size: 1, sequence length: 36777 Attention mask shape: torch.Size([1, 1, 36777, 36777]) Position ids shape: torch.Size([1, 36777]) Input IDs shape: torch.Size([1, 36777]) Labels shape: torch.Size([1, 36777]) Final batch size: 1, sequence length: 20755 Attention mask shape: torch.Size([1, 1, 20755, 20755]) Position ids shape: torch.Size([1, 20755]) Input IDs shape: torch.Size([1, 20755]) Labels shape: torch.Size([1, 20755]) Final batch size: 1, sequence length: 34673 Attention mask shape: torch.Size([1, 1, 34673, 34673]) Position ids shape: torch.Size([1, 34673]) Input IDs shape: torch.Size([1, 34673]) Labels shape: torch.Size([1, 34673]) Final batch size: 1, sequence length: 21034 Attention mask shape: torch.Size([1, 1, 21034, 21034]) Position ids shape: torch.Size([1, 21034]) Input IDs shape: torch.Size([1, 21034]) Labels shape: torch.Size([1, 21034]) Final batch size: 1, sequence length: 27489 Attention mask shape: torch.Size([1, 1, 27489, 27489]) Position ids shape: torch.Size([1, 27489]) Input IDs shape: torch.Size([1, 27489]) Labels shape: torch.Size([1, 27489]) Final batch size: 1, sequence length: 22366 Attention mask shape: torch.Size([1, 1, 22366, 22366]) Position ids shape: torch.Size([1, 22366]) Input IDs shape: torch.Size([1, 22366]) Labels shape: torch.Size([1, 22366]) Final batch size: 1, sequence length: 36723 Attention mask shape: torch.Size([1, 1, 36723, 36723]) Position ids shape: torch.Size([1, 36723]) Input IDs shape: torch.Size([1, 36723]) Labels shape: torch.Size([1, 36723]) Final batch size: 1, sequence length: 38049 Attention mask shape: torch.Size([1, 1, 38049, 38049]) Position ids shape: torch.Size([1, 38049]) Input IDs shape: torch.Size([1, 38049]) Labels shape: torch.Size([1, 38049]) Final batch size: 1, sequence length: 38848 Attention mask shape: torch.Size([1, 1, 38848, 38848]) Position ids shape: torch.Size([1, 38848]) Input IDs shape: torch.Size([1, 38848]) Labels shape: torch.Size([1, 38848]) Final batch size: 1, sequence length: 28176 Attention mask shape: torch.Size([1, 1, 28176, 28176]) Position ids shape: torch.Size([1, 28176]) Input IDs shape: torch.Size([1, 28176]) Labels shape: torch.Size([1, 28176]) Final batch size: 1, sequence length: 37285 Attention mask shape: torch.Size([1, 1, 37285, 37285]) Position ids shape: torch.Size([1, 37285]) Input IDs shape: torch.Size([1, 37285]) Labels shape: torch.Size([1, 37285]) Final batch size: 1, sequence length: 31903 Attention mask shape: torch.Size([1, 1, 31903, 31903]) Position ids shape: torch.Size([1, 31903]) Input IDs shape: torch.Size([1, 31903]) Labels shape: torch.Size([1, 31903]) Final batch size: 1, sequence length: 40325 Attention mask shape: torch.Size([1, 1, 40325, 40325]) Position ids shape: torch.Size([1, 40325]) Input IDs shape: torch.Size([1, 40325]) Labels shape: torch.Size([1, 40325]) Final batch size: 1, sequence length: 35904 Attention mask shape: torch.Size([1, 1, 35904, 35904]) Position ids shape: torch.Size([1, 35904]) Input IDs shape: torch.Size([1, 35904]) Labels shape: torch.Size([1, 35904]) Final batch size: 1, sequence length: 25946 Attention mask shape: torch.Size([1, 1, 25946, 25946]) Position ids shape: torch.Size([1, 25946]) Input IDs shape: torch.Size([1, 25946]) Labels shape: torch.Size([1, 25946]) Final batch size: 1, sequence length: 6882 Attention mask shape: torch.Size([1, 1, 6882, 6882]) Position ids shape: torch.Size([1, 6882]) Input IDs shape: torch.Size([1, 6882]) Labels shape: torch.Size([1, 6882]) Final batch size: 1, sequence length: 39253 Attention mask shape: torch.Size([1, 1, 39253, 39253]) Position ids shape: torch.Size([1, 39253]) Input IDs shape: torch.Size([1, 39253]) Labels shape: torch.Size([1, 39253]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38104 Attention mask shape: torch.Size([1, 1, 38104, 38104]) Position ids shape: torch.Size([1, 38104]) Input IDs shape: torch.Size([1, 38104]) Labels shape: torch.Size([1, 38104]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24801 Attention mask shape: torch.Size([1, 1, 24801, 24801]) Position ids shape: torch.Size([1, 24801]) Input IDs shape: torch.Size([1, 24801]) Labels shape: torch.Size([1, 24801]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34289 Attention mask shape: torch.Size([1, 1, 34289, 34289]) Position ids shape: torch.Size([1, 34289]) Input IDs shape: torch.Size([1, 34289]) Labels shape: torch.Size([1, 34289]) Final batch size: 1, sequence length: 36786 Attention mask shape: torch.Size([1, 1, 36786, 36786]) Position ids shape: torch.Size([1, 36786]) Input IDs shape: torch.Size([1, 36786]) Labels shape: torch.Size([1, 36786]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 10469 Attention mask shape: torch.Size([1, 1, 10469, 10469]) Position ids shape: torch.Size([1, 10469]) Input IDs shape: torch.Size([1, 10469]) Labels shape: torch.Size([1, 10469]) Final batch size: 1, sequence length: 40937 Attention mask shape: torch.Size([1, 1, 40937, 40937]) Position ids shape: torch.Size([1, 40937]) Input IDs shape: torch.Size([1, 40937]) Labels shape: torch.Size([1, 40937]) Final batch size: 1, sequence length: 40317 Attention mask shape: torch.Size([1, 1, 40317, 40317]) Position ids shape: torch.Size([1, 40317]) Input IDs shape: torch.Size([1, 40317]) Labels shape: torch.Size([1, 40317]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16649 Attention mask shape: torch.Size([1, 1, 16649, 16649]) Position ids shape: torch.Size([1, 16649]) Input IDs shape: torch.Size([1, 16649]) Labels shape: torch.Size([1, 16649]) Final batch size: 1, sequence length: 31714 Attention mask shape: torch.Size([1, 1, 31714, 31714]) Position ids shape: torch.Size([1, 31714]) Input IDs shape: torch.Size([1, 31714]) Labels shape: torch.Size([1, 31714]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19437 Attention mask shape: torch.Size([1, 1, 19437, 19437]) Position ids shape: torch.Size([1, 19437]) Input IDs shape: torch.Size([1, 19437]) Labels shape: torch.Size([1, 19437]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19639 Attention mask shape: torch.Size([1, 1, 19639, 19639]) Position ids shape: torch.Size([1, 19639]) Input IDs shape: torch.Size([1, 19639]) Labels shape: torch.Size([1, 19639]) Final batch size: 1, sequence length: 25963 Attention mask shape: torch.Size([1, 1, 25963, 25963]) Position ids shape: torch.Size([1, 25963]) Input IDs shape: torch.Size([1, 25963]) Labels shape: torch.Size([1, 25963]) Final batch size: 1, sequence length: 40657 Attention mask shape: torch.Size([1, 1, 40657, 40657]) Position ids shape: torch.Size([1, 40657]) Input IDs shape: torch.Size([1, 40657]) Labels shape: torch.Size([1, 40657]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38686 Attention mask shape: torch.Size([1, 1, 38686, 38686]) Position ids shape: torch.Size([1, 38686]) Input IDs shape: torch.Size([1, 38686]) Labels shape: torch.Size([1, 38686]) Final batch size: 1, sequence length: 36128 Attention mask shape: torch.Size([1, 1, 36128, 36128]) Position ids shape: torch.Size([1, 36128]) Input IDs shape: torch.Size([1, 36128]) Labels shape: torch.Size([1, 36128]) Final batch size: 1, sequence length: 24551 Attention mask shape: torch.Size([1, 1, 24551, 24551]) Position ids shape: torch.Size([1, 24551]) Input IDs shape: torch.Size([1, 24551]) Labels shape: torch.Size([1, 24551]) Final batch size: 1, sequence length: 21547 Attention mask shape: torch.Size([1, 1, 21547, 21547]) Position ids shape: torch.Size([1, 21547]) Input IDs shape: torch.Size([1, 21547]) Labels shape: torch.Size([1, 21547]) Final batch size: 1, sequence length: 34540 Attention mask shape: torch.Size([1, 1, 34540, 34540]) Position ids shape: torch.Size([1, 34540]) Input IDs shape: torch.Size([1, 34540]) Labels shape: torch.Size([1, 34540]) Final batch size: 1, sequence length: 17882 Attention mask shape: torch.Size([1, 1, 17882, 17882]) Position ids shape: torch.Size([1, 17882]) Input IDs shape: torch.Size([1, 17882]) Labels shape: torch.Size([1, 17882]) Final batch size: 1, sequence length: 27947 Attention mask shape: torch.Size([1, 1, 27947, 27947]) Position ids shape: torch.Size([1, 27947]) Input IDs shape: torch.Size([1, 27947]) Labels shape: torch.Size([1, 27947]) Final batch size: 1, sequence length: 26660 Attention mask shape: torch.Size([1, 1, 26660, 26660]) Position ids shape: torch.Size([1, 26660]) Input IDs shape: torch.Size([1, 26660]) Labels shape: torch.Size([1, 26660]) Final batch size: 1, sequence length: 17379 Attention mask shape: torch.Size([1, 1, 17379, 17379]) Position ids shape: torch.Size([1, 17379]) Input IDs shape: torch.Size([1, 17379]) Labels shape: torch.Size([1, 17379]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21496 Attention mask shape: torch.Size([1, 1, 21496, 21496]) Position ids shape: torch.Size([1, 21496]) Input IDs shape: torch.Size([1, 21496]) Labels shape: torch.Size([1, 21496]) Final batch size: 1, sequence length: 20843 Attention mask shape: torch.Size([1, 1, 20843, 20843]) Position ids shape: torch.Size([1, 20843]) Input IDs shape: torch.Size([1, 20843]) Labels shape: torch.Size([1, 20843]) Final batch size: 1, sequence length: 37946 Attention mask shape: torch.Size([1, 1, 37946, 37946]) Position ids shape: torch.Size([1, 37946]) Input IDs shape: torch.Size([1, 37946]) Labels shape: torch.Size([1, 37946]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7681 Attention mask shape: torch.Size([1, 1, 7681, 7681]) Position ids shape: torch.Size([1, 7681]) Input IDs shape: torch.Size([1, 7681]) Labels shape: torch.Size([1, 7681]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36317 Attention mask shape: torch.Size([1, 1, 36317, 36317]) Position ids shape: torch.Size([1, 36317]) Input IDs shape: torch.Size([1, 36317]) Labels shape: torch.Size([1, 36317]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.3432, 'grad_norm': 0.6601561997352322, 'learning_rate': 9.667902132486009e-06, 'num_tokens': -inf, 'epoch': 1.5} Final batch size: 1, sequence length: 5525 Attention mask shape: torch.Size([1, 1, 5525, 5525]) Position ids shape: torch.Size([1, 5525]) Input IDs shape: torch.Size([1, 5525]) Labels shape: torch.Size([1, 5525]) Final batch size: 1, sequence length: 5328 Attention mask shape: torch.Size([1, 1, 5328, 5328]) Position ids shape: torch.Size([1, 5328]) Input IDs shape: torch.Size([1, 5328]) Labels shape: torch.Size([1, 5328]) Final batch size: 1, sequence length: 10273 Attention mask shape: torch.Size([1, 1, 10273, 10273]) Position ids shape: torch.Size([1, 10273]) Input IDs shape: torch.Size([1, 10273]) Labels shape: torch.Size([1, 10273]) Final batch size: 1, sequence length: 11464 Attention mask shape: torch.Size([1, 1, 11464, 11464]) Position ids shape: torch.Size([1, 11464]) Input IDs shape: torch.Size([1, 11464]) Labels shape: torch.Size([1, 11464]) Final batch size: 1, sequence length: 12419 Attention mask shape: torch.Size([1, 1, 12419, 12419]) Position ids shape: torch.Size([1, 12419]) Input IDs shape: torch.Size([1, 12419]) Labels shape: torch.Size([1, 12419]) Final batch size: 1, sequence length: 10434 Attention mask shape: torch.Size([1, 1, 10434, 10434]) Position ids shape: torch.Size([1, 10434]) Input IDs shape: torch.Size([1, 10434]) Labels shape: torch.Size([1, 10434]) Final batch size: 1, sequence length: 13607 Attention mask shape: torch.Size([1, 1, 13607, 13607]) Position ids shape: torch.Size([1, 13607]) Input IDs shape: torch.Size([1, 13607]) Labels shape: torch.Size([1, 13607]) Final batch size: 1, sequence length: 13657 Attention mask shape: torch.Size([1, 1, 13657, 13657]) Position ids shape: torch.Size([1, 13657]) Input IDs shape: torch.Size([1, 13657]) Labels shape: torch.Size([1, 13657]) Final batch size: 1, sequence length: 9648 Final batch size: 1, sequence length: 14226 Attention mask shape: torch.Size([1, 1, 9648, 9648]) Position ids shape: torch.Size([1, 9648]) Input IDs shape: torch.Size([1, 9648]) Labels shape: torch.Size([1, 9648]) Attention mask shape: torch.Size([1, 1, 14226, 14226]) Position ids shape: torch.Size([1, 14226]) Input IDs shape: torch.Size([1, 14226]) Labels shape: torch.Size([1, 14226]) Final batch size: 1, sequence length: 14360 Attention mask shape: torch.Size([1, 1, 14360, 14360]) Position ids shape: torch.Size([1, 14360]) Input IDs shape: torch.Size([1, 14360]) Labels shape: torch.Size([1, 14360]) Final batch size: 1, sequence length: 16398 Attention mask shape: torch.Size([1, 1, 16398, 16398]) Position ids shape: torch.Size([1, 16398]) Input IDs shape: torch.Size([1, 16398]) Labels shape: torch.Size([1, 16398]) Final batch size: 1, sequence length: 16053 Attention mask shape: torch.Size([1, 1, 16053, 16053]) Position ids shape: torch.Size([1, 16053]) Input IDs shape: torch.Size([1, 16053]) Labels shape: torch.Size([1, 16053]) Final batch size: 1, sequence length: 16756 Attention mask shape: torch.Size([1, 1, 16756, 16756]) Position ids shape: torch.Size([1, 16756]) Input IDs shape: torch.Size([1, 16756]) Labels shape: torch.Size([1, 16756]) Final batch size: 1, sequence length: 11225 Attention mask shape: torch.Size([1, 1, 11225, 11225]) Position ids shape: torch.Size([1, 11225]) Input IDs shape: torch.Size([1, 11225]) Labels shape: torch.Size([1, 11225]) Final batch size: 1, sequence length: 16223 Attention mask shape: torch.Size([1, 1, 16223, 16223]) Position ids shape: torch.Size([1, 16223]) Input IDs shape: torch.Size([1, 16223]) Labels shape: torch.Size([1, 16223]) Final batch size: 1, sequence length: 18408 Attention mask shape: torch.Size([1, 1, 18408, 18408]) Position ids shape: torch.Size([1, 18408]) Input IDs shape: torch.Size([1, 18408]) Labels shape: torch.Size([1, 18408]) Final batch size: 1, sequence length: 19892 Attention mask shape: torch.Size([1, 1, 19892, 19892]) Position ids shape: torch.Size([1, 19892]) Input IDs shape: torch.Size([1, 19892]) Labels shape: torch.Size([1, 19892]) Final batch size: 1, sequence length: 17117 Attention mask shape: torch.Size([1, 1, 17117, 17117]) Position ids shape: torch.Size([1, 17117]) Input IDs shape: torch.Size([1, 17117]) Labels shape: torch.Size([1, 17117]) Final batch size: 1, sequence length: 15243 Attention mask shape: torch.Size([1, 1, 15243, 15243]) Position ids shape: torch.Size([1, 15243]) Input IDs shape: torch.Size([1, 15243]) Labels shape: torch.Size([1, 15243]) Final batch size: 1, sequence length: 17294 Attention mask shape: torch.Size([1, 1, 17294, 17294]) Position ids shape: torch.Size([1, 17294]) Input IDs shape: torch.Size([1, 17294]) Labels shape: torch.Size([1, 17294]) Final batch size: 1, sequence length: 17843 Attention mask shape: torch.Size([1, 1, 17843, 17843]) Position ids shape: torch.Size([1, 17843]) Input IDs shape: torch.Size([1, 17843]) Labels shape: torch.Size([1, 17843]) Final batch size: 1, sequence length: 10017 Attention mask shape: torch.Size([1, 1, 10017, 10017]) Position ids shape: torch.Size([1, 10017]) Input IDs shape: torch.Size([1, 10017]) Labels shape: torch.Size([1, 10017]) Final batch size: 1, sequence length: 14437 Attention mask shape: torch.Size([1, 1, 14437, 14437]) Position ids shape: torch.Size([1, 14437]) Input IDs shape: torch.Size([1, 14437]) Labels shape: torch.Size([1, 14437]) Final batch size: 1, sequence length: 14025 Attention mask shape: torch.Size([1, 1, 14025, 14025]) Position ids shape: torch.Size([1, 14025]) Input IDs shape: torch.Size([1, 14025]) Labels shape: torch.Size([1, 14025]) Final batch size: 1, sequence length: 19767 Attention mask shape: torch.Size([1, 1, 19767, 19767]) Position ids shape: torch.Size([1, 19767]) Input IDs shape: torch.Size([1, 19767]) Labels shape: torch.Size([1, 19767]) Final batch size: 1, sequence length: 20695 Attention mask shape: torch.Size([1, 1, 20695, 20695]) Position ids shape: torch.Size([1, 20695]) Input IDs shape: torch.Size([1, 20695]) Labels shape: torch.Size([1, 20695]) Final batch size: 1, sequence length: 19259 Attention mask shape: torch.Size([1, 1, 19259, 19259]) Position ids shape: torch.Size([1, 19259]) Input IDs shape: torch.Size([1, 19259]) Labels shape: torch.Size([1, 19259]) Final batch size: 1, sequence length: 21455 Attention mask shape: torch.Size([1, 1, 21455, 21455]) Position ids shape: torch.Size([1, 21455]) Input IDs shape: torch.Size([1, 21455]) Labels shape: torch.Size([1, 21455]) Final batch size: 1, sequence length: 19278 Attention mask shape: torch.Size([1, 1, 19278, 19278]) Position ids shape: torch.Size([1, 19278]) Input IDs shape: torch.Size([1, 19278]) Labels shape: torch.Size([1, 19278]) Final batch size: 1, sequence length: 20433 Attention mask shape: torch.Size([1, 1, 20433, 20433]) Position ids shape: torch.Size([1, 20433]) Input IDs shape: torch.Size([1, 20433]) Labels shape: torch.Size([1, 20433]) Final batch size: 1, sequence length: 5734 Attention mask shape: torch.Size([1, 1, 5734, 5734]) Position ids shape: torch.Size([1, 5734]) Input IDs shape: torch.Size([1, 5734]) Labels shape: torch.Size([1, 5734]) Final batch size: 1, sequence length: 13623 Attention mask shape: torch.Size([1, 1, 13623, 13623]) Position ids shape: torch.Size([1, 13623]) Input IDs shape: torch.Size([1, 13623]) Labels shape: torch.Size([1, 13623]) Final batch size: 1, sequence length: 22932 Attention mask shape: torch.Size([1, 1, 22932, 22932]) Position ids shape: torch.Size([1, 22932]) Input IDs shape: torch.Size([1, 22932]) Labels shape: torch.Size([1, 22932]) Final batch size: 1, sequence length: 25405 Attention mask shape: torch.Size([1, 1, 25405, 25405]) Position ids shape: torch.Size([1, 25405]) Input IDs shape: torch.Size([1, 25405]) Labels shape: torch.Size([1, 25405]) Final batch size: 1, sequence length: 6378 Attention mask shape: torch.Size([1, 1, 6378, 6378]) Position ids shape: torch.Size([1, 6378]) Input IDs shape: torch.Size([1, 6378]) Labels shape: torch.Size([1, 6378]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 17985 Attention mask shape: torch.Size([1, 1, 17985, 17985]) Position ids shape: torch.Size([1, 17985]) Input IDs shape: torch.Size([1, 17985]) Labels shape: torch.Size([1, 17985]) Final batch size: 1, sequence length: 19492 Attention mask shape: torch.Size([1, 1, 19492, 19492]) Position ids shape: torch.Size([1, 19492]) Input IDs shape: torch.Size([1, 19492]) Labels shape: torch.Size([1, 19492]) Final batch size: 1, sequence length: 14429 Attention mask shape: torch.Size([1, 1, 14429, 14429]) Position ids shape: torch.Size([1, 14429]) Input IDs shape: torch.Size([1, 14429]) Labels shape: torch.Size([1, 14429]) Final batch size: 1, sequence length: 23334 Attention mask shape: torch.Size([1, 1, 23334, 23334]) Position ids shape: torch.Size([1, 23334]) Input IDs shape: torch.Size([1, 23334]) Labels shape: torch.Size([1, 23334]) Final batch size: 1, sequence length: 25548 Attention mask shape: torch.Size([1, 1, 25548, 25548]) Position ids shape: torch.Size([1, 25548]) Input IDs shape: torch.Size([1, 25548]) Labels shape: torch.Size([1, 25548]) Final batch size: 1, sequence length: 24885 Attention mask shape: torch.Size([1, 1, 24885, 24885]) Position ids shape: torch.Size([1, 24885]) Input IDs shape: torch.Size([1, 24885]) Labels shape: torch.Size([1, 24885]) Final batch size: 1, sequence length: 17951 Attention mask shape: torch.Size([1, 1, 17951, 17951]) Position ids shape: torch.Size([1, 17951]) Input IDs shape: torch.Size([1, 17951]) Labels shape: torch.Size([1, 17951]) Final batch size: 1, sequence length: 26639 Attention mask shape: torch.Size([1, 1, 26639, 26639]) Position ids shape: torch.Size([1, 26639]) Input IDs shape: torch.Size([1, 26639]) Labels shape: torch.Size([1, 26639]) Final batch size: 1, sequence length: 28166 Attention mask shape: torch.Size([1, 1, 28166, 28166]) Position ids shape: torch.Size([1, 28166]) Input IDs shape: torch.Size([1, 28166]) Labels shape: torch.Size([1, 28166]) Final batch size: 1, sequence length: 25381 Attention mask shape: torch.Size([1, 1, 25381, 25381]) Position ids shape: torch.Size([1, 25381]) Input IDs shape: torch.Size([1, 25381]) Labels shape: torch.Size([1, 25381]) Final batch size: 1, sequence length: 26063 Attention mask shape: torch.Size([1, 1, 26063, 26063]) Position ids shape: torch.Size([1, 26063]) Input IDs shape: torch.Size([1, 26063]) Labels shape: torch.Size([1, 26063]) Final batch size: 1, sequence length: 27484 Attention mask shape: torch.Size([1, 1, 27484, 27484]) Position ids shape: torch.Size([1, 27484]) Input IDs shape: torch.Size([1, 27484]) Labels shape: torch.Size([1, 27484]) Final batch size: 1, sequence length: 28623 Attention mask shape: torch.Size([1, 1, 28623, 28623]) Position ids shape: torch.Size([1, 28623]) Input IDs shape: torch.Size([1, 28623]) Labels shape: torch.Size([1, 28623]) Final batch size: 1, sequence length: 28910 Attention mask shape: torch.Size([1, 1, 28910, 28910]) Position ids shape: torch.Size([1, 28910]) Input IDs shape: torch.Size([1, 28910]) Labels shape: torch.Size([1, 28910]) Final batch size: 1, sequence length: 30236 Attention mask shape: torch.Size([1, 1, 30236, 30236]) Position ids shape: torch.Size([1, 30236]) Input IDs shape: torch.Size([1, 30236]) Labels shape: torch.Size([1, 30236]) Final batch size: 1, sequence length: 16716 Attention mask shape: torch.Size([1, 1, 16716, 16716]) Position ids shape: torch.Size([1, 16716]) Input IDs shape: torch.Size([1, 16716]) Labels shape: torch.Size([1, 16716]) Final batch size: 1, sequence length: 28823 Attention mask shape: torch.Size([1, 1, 28823, 28823]) Position ids shape: torch.Size([1, 28823]) Input IDs shape: torch.Size([1, 28823]) Labels shape: torch.Size([1, 28823]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 25540 Attention mask shape: torch.Size([1, 1, 25540, 25540]) Position ids shape: torch.Size([1, 25540]) Input IDs shape: torch.Size([1, 25540]) Labels shape: torch.Size([1, 25540]) Final batch size: 1, sequence length: 32660 Attention mask shape: torch.Size([1, 1, 32660, 32660]) Position ids shape: torch.Size([1, 32660]) Input IDs shape: torch.Size([1, 32660]) Labels shape: torch.Size([1, 32660]) Final batch size: 1, sequence length: 29824 Attention mask shape: torch.Size([1, 1, 29824, 29824]) Position ids shape: torch.Size([1, 29824]) Input IDs shape: torch.Size([1, 29824]) Labels shape: torch.Size([1, 29824]) Final batch size: 1, sequence length: 28206 Attention mask shape: torch.Size([1, 1, 28206, 28206]) Position ids shape: torch.Size([1, 28206]) Input IDs shape: torch.Size([1, 28206]) Labels shape: torch.Size([1, 28206]) Final batch size: 1, sequence length: 31464 Attention mask shape: torch.Size([1, 1, 31464, 31464]) Position ids shape: torch.Size([1, 31464]) Input IDs shape: torch.Size([1, 31464]) Labels shape: torch.Size([1, 31464]) Final batch size: 1, sequence length: 32515 Attention mask shape: torch.Size([1, 1, 32515, 32515]) Position ids shape: torch.Size([1, 32515]) Input IDs shape: torch.Size([1, 32515]) Labels shape: torch.Size([1, 32515]) Final batch size: 1, sequence length: 33601 Attention mask shape: torch.Size([1, 1, 33601, 33601]) Position ids shape: torch.Size([1, 33601]) Input IDs shape: torch.Size([1, 33601]) Labels shape: torch.Size([1, 33601]) Final batch size: 1, sequence length: 17634 Attention mask shape: torch.Size([1, 1, 17634, 17634]) Position ids shape: torch.Size([1, 17634]) Input IDs shape: torch.Size([1, 17634]) Labels shape: torch.Size([1, 17634]) Final batch size: 1, sequence length: 27659 Attention mask shape: torch.Size([1, 1, 27659, 27659]) Position ids shape: torch.Size([1, 27659]) Input IDs shape: torch.Size([1, 27659]) Labels shape: torch.Size([1, 27659]) Final batch size: 1, sequence length: 32786 Attention mask shape: torch.Size([1, 1, 32786, 32786]) Position ids shape: torch.Size([1, 32786]) Input IDs shape: torch.Size([1, 32786]) Labels shape: torch.Size([1, 32786]) Final batch size: 1, sequence length: 29561 Attention mask shape: torch.Size([1, 1, 29561, 29561]) Position ids shape: torch.Size([1, 29561]) Input IDs shape: torch.Size([1, 29561]) Labels shape: torch.Size([1, 29561]) Final batch size: 1, sequence length: 9029 Attention mask shape: torch.Size([1, 1, 9029, 9029]) Position ids shape: torch.Size([1, 9029]) Input IDs shape: torch.Size([1, 9029]) Labels shape: torch.Size([1, 9029]) Final batch size: 1, sequence length: 10132 Attention mask shape: torch.Size([1, 1, 10132, 10132]) Position ids shape: torch.Size([1, 10132]) Input IDs shape: torch.Size([1, 10132]) Labels shape: torch.Size([1, 10132]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 20951 Attention mask shape: torch.Size([1, 1, 20951, 20951]) Position ids shape: torch.Size([1, 20951]) Input IDs shape: torch.Size([1, 20951]) Labels shape: torch.Size([1, 20951]) Final batch size: 1, sequence length: 37555 Attention mask shape: torch.Size([1, 1, 37555, 37555]) Position ids shape: torch.Size([1, 37555]) Input IDs shape: torch.Size([1, 37555]) Labels shape: torch.Size([1, 37555]) Final batch size: 1, sequence length: 35411 Attention mask shape: torch.Size([1, 1, 35411, 35411]) Position ids shape: torch.Size([1, 35411]) Input IDs shape: torch.Size([1, 35411]) Labels shape: torch.Size([1, 35411]) Final batch size: 1, sequence length: 15513 Attention mask shape: torch.Size([1, 1, 15513, 15513]) Position ids shape: torch.Size([1, 15513]) Input IDs shape: torch.Size([1, 15513]) Labels shape: torch.Size([1, 15513]) Final batch size: 1, sequence length: 36131 Attention mask shape: torch.Size([1, 1, 36131, 36131]) Position ids shape: torch.Size([1, 36131]) Input IDs shape: torch.Size([1, 36131]) Labels shape: torch.Size([1, 36131]) Final batch size: 1, sequence length: 36970 Attention mask shape: torch.Size([1, 1, 36970, 36970]) Position ids shape: torch.Size([1, 36970]) Input IDs shape: torch.Size([1, 36970]) Labels shape: torch.Size([1, 36970]) Final batch size: 1, sequence length: 37016 Attention mask shape: torch.Size([1, 1, 37016, 37016]) Position ids shape: torch.Size([1, 37016]) Input IDs shape: torch.Size([1, 37016]) Labels shape: torch.Size([1, 37016]) Final batch size: 1, sequence length: 20533 Attention mask shape: torch.Size([1, 1, 20533, 20533]) Position ids shape: torch.Size([1, 20533]) Input IDs shape: torch.Size([1, 20533]) Labels shape: torch.Size([1, 20533]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26562 Attention mask shape: torch.Size([1, 1, 26562, 26562]) Position ids shape: torch.Size([1, 26562]) Input IDs shape: torch.Size([1, 26562]) Labels shape: torch.Size([1, 26562]) Final batch size: 1, sequence length: 14372 Attention mask shape: torch.Size([1, 1, 14372, 14372]) Position ids shape: torch.Size([1, 14372]) Input IDs shape: torch.Size([1, 14372]) Labels shape: torch.Size([1, 14372]) Final batch size: 1, sequence length: 38712 Attention mask shape: torch.Size([1, 1, 38712, 38712]) Position ids shape: torch.Size([1, 38712]) Input IDs shape: torch.Size([1, 38712]) Labels shape: torch.Size([1, 38712]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36292 Attention mask shape: torch.Size([1, 1, 36292, 36292]) Position ids shape: torch.Size([1, 36292]) Input IDs shape: torch.Size([1, 36292]) Labels shape: torch.Size([1, 36292]) Final batch size: 1, sequence length: 28879 Attention mask shape: torch.Size([1, 1, 28879, 28879]) Position ids shape: torch.Size([1, 28879]) Input IDs shape: torch.Size([1, 28879]) Labels shape: torch.Size([1, 28879]) Final batch size: 1, sequence length: 38071 Attention mask shape: torch.Size([1, 1, 38071, 38071]) Position ids shape: torch.Size([1, 38071]) Input IDs shape: torch.Size([1, 38071]) Labels shape: torch.Size([1, 38071]) Final batch size: 1, sequence length: 26781 Attention mask shape: torch.Size([1, 1, 26781, 26781]) Position ids shape: torch.Size([1, 26781]) Input IDs shape: torch.Size([1, 26781]) Labels shape: torch.Size([1, 26781]) Final batch size: 1, sequence length: 31084 Attention mask shape: torch.Size([1, 1, 31084, 31084]) Position ids shape: torch.Size([1, 31084]) Input IDs shape: torch.Size([1, 31084]) Labels shape: torch.Size([1, 31084]) Final batch size: 1, sequence length: 17914 Attention mask shape: torch.Size([1, 1, 17914, 17914]) Position ids shape: torch.Size([1, 17914]) Input IDs shape: torch.Size([1, 17914]) Labels shape: torch.Size([1, 17914]) Final batch size: 1, sequence length: 33422 Attention mask shape: torch.Size([1, 1, 33422, 33422]) Position ids shape: torch.Size([1, 33422]) Input IDs shape: torch.Size([1, 33422]) Labels shape: torch.Size([1, 33422]) Final batch size: 1, sequence length: 35525 Attention mask shape: torch.Size([1, 1, 35525, 35525]) Position ids shape: torch.Size([1, 35525]) Input IDs shape: torch.Size([1, 35525]) Labels shape: torch.Size([1, 35525]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 36124 Attention mask shape: torch.Size([1, 1, 36124, 36124]) Position ids shape: torch.Size([1, 36124]) Input IDs shape: torch.Size([1, 36124]) Labels shape: torch.Size([1, 36124]) Final batch size: 1, sequence length: 36185 Attention mask shape: torch.Size([1, 1, 36185, 36185]) Position ids shape: torch.Size([1, 36185]) Input IDs shape: torch.Size([1, 36185]) Labels shape: torch.Size([1, 36185]) Final batch size: 1, sequence length: 30009 Attention mask shape: torch.Size([1, 1, 30009, 30009]) Position ids shape: torch.Size([1, 30009]) Input IDs shape: torch.Size([1, 30009]) Labels shape: torch.Size([1, 30009]) Final batch size: 1, sequence length: 25529 Attention mask shape: torch.Size([1, 1, 25529, 25529]) Position ids shape: torch.Size([1, 25529]) Input IDs shape: torch.Size([1, 25529]) Labels shape: torch.Size([1, 25529]) Final batch size: 1, sequence length: 35775 Attention mask shape: torch.Size([1, 1, 35775, 35775]) Position ids shape: torch.Size([1, 35775]) Input IDs shape: torch.Size([1, 35775]) Labels shape: torch.Size([1, 35775]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38210 Attention mask shape: torch.Size([1, 1, 38210, 38210]) Position ids shape: torch.Size([1, 38210]) Input IDs shape: torch.Size([1, 38210]) Labels shape: torch.Size([1, 38210]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16409 Attention mask shape: torch.Size([1, 1, 16409, 16409]) Position ids shape: torch.Size([1, 16409]) Input IDs shape: torch.Size([1, 16409]) Labels shape: torch.Size([1, 16409]) Final batch size: 1, sequence length: 40753 Attention mask shape: torch.Size([1, 1, 40753, 40753]) Position ids shape: torch.Size([1, 40753]) Input IDs shape: torch.Size([1, 40753]) Labels shape: torch.Size([1, 40753]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 15031 Attention mask shape: torch.Size([1, 1, 15031, 15031]) Position ids shape: torch.Size([1, 15031]) Input IDs shape: torch.Size([1, 15031]) Labels shape: torch.Size([1, 15031]) Final batch size: 1, sequence length: 36534 Attention mask shape: torch.Size([1, 1, 36534, 36534]) Position ids shape: torch.Size([1, 36534]) Input IDs shape: torch.Size([1, 36534]) Labels shape: torch.Size([1, 36534]) Final batch size: 1, sequence length: 31042 Attention mask shape: torch.Size([1, 1, 31042, 31042]) Position ids shape: torch.Size([1, 31042]) Input IDs shape: torch.Size([1, 31042]) Labels shape: torch.Size([1, 31042]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30653 Attention mask shape: torch.Size([1, 1, 30653, 30653]) Position ids shape: torch.Size([1, 30653]) Input IDs shape: torch.Size([1, 30653]) Labels shape: torch.Size([1, 30653]) Final batch size: 1, sequence length: 37159 Attention mask shape: torch.Size([1, 1, 37159, 37159]) Position ids shape: torch.Size([1, 37159]) Input IDs shape: torch.Size([1, 37159]) Labels shape: torch.Size([1, 37159]) Final batch size: 1, sequence length: 18884 Attention mask shape: torch.Size([1, 1, 18884, 18884]) Position ids shape: torch.Size([1, 18884]) Input IDs shape: torch.Size([1, 18884]) Labels shape: torch.Size([1, 18884]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31448 Attention mask shape: torch.Size([1, 1, 31448, 31448]) Position ids shape: torch.Size([1, 31448]) Input IDs shape: torch.Size([1, 31448]) Labels shape: torch.Size([1, 31448]) Final batch size: 1, sequence length: 25850 Attention mask shape: torch.Size([1, 1, 25850, 25850]) Position ids shape: torch.Size([1, 25850]) Input IDs shape: torch.Size([1, 25850]) Labels shape: torch.Size([1, 25850]) Final batch size: 1, sequence length: 20307 Attention mask shape: torch.Size([1, 1, 20307, 20307]) Position ids shape: torch.Size([1, 20307]) Input IDs shape: torch.Size([1, 20307]) Labels shape: torch.Size([1, 20307]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7448 Attention mask shape: torch.Size([1, 1, 7448, 7448]) Position ids shape: torch.Size([1, 7448]) Input IDs shape: torch.Size([1, 7448]) Labels shape: torch.Size([1, 7448]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40605 Attention mask shape: torch.Size([1, 1, 40605, 40605]) Position ids shape: torch.Size([1, 40605]) Input IDs shape: torch.Size([1, 40605]) Labels shape: torch.Size([1, 40605]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 5405 Attention mask shape: torch.Size([1, 1, 5405, 5405]) Position ids shape: torch.Size([1, 5405]) Input IDs shape: torch.Size([1, 5405]) Labels shape: torch.Size([1, 5405]) {'loss': 0.3205, 'grad_norm': 0.5224832888014513, 'learning_rate': 9.567727288213005e-06, 'num_tokens': -inf, 'epoch': 1.62} Final batch size: 1, sequence length: 4641 Attention mask shape: torch.Size([1, 1, 4641, 4641]) Position ids shape: torch.Size([1, 4641]) Input IDs shape: torch.Size([1, 4641]) Labels shape: torch.Size([1, 4641]) Final batch size: 1, sequence length: 4223 Attention mask shape: torch.Size([1, 1, 4223, 4223]) Position ids shape: torch.Size([1, 4223]) Input IDs shape: torch.Size([1, 4223]) Labels shape: torch.Size([1, 4223]) Final batch size: 1, sequence length: 5754 Attention mask shape: torch.Size([1, 1, 5754, 5754]) Position ids shape: torch.Size([1, 5754]) Input IDs shape: torch.Size([1, 5754]) Labels shape: torch.Size([1, 5754]) Final batch size: 1, sequence length: 7795 Attention mask shape: torch.Size([1, 1, 7795, 7795]) Position ids shape: torch.Size([1, 7795]) Input IDs shape: torch.Size([1, 7795]) Labels shape: torch.Size([1, 7795]) Final batch size: 1, sequence length: 8177 Attention mask shape: torch.Size([1, 1, 8177, 8177]) Position ids shape: torch.Size([1, 8177]) Input IDs shape: torch.Size([1, 8177]) Labels shape: torch.Size([1, 8177]) Final batch size: 1, sequence length: 9021 Attention mask shape: torch.Size([1, 1, 9021, 9021]) Position ids shape: torch.Size([1, 9021]) Input IDs shape: torch.Size([1, 9021]) Labels shape: torch.Size([1, 9021]) Final batch size: 1, sequence length: 12501 Attention mask shape: torch.Size([1, 1, 12501, 12501]) Position ids shape: torch.Size([1, 12501]) Input IDs shape: torch.Size([1, 12501]) Labels shape: torch.Size([1, 12501]) Final batch size: 1, sequence length: 13186 Attention mask shape: torch.Size([1, 1, 13186, 13186]) Position ids shape: torch.Size([1, 13186]) Input IDs shape: torch.Size([1, 13186]) Labels shape: torch.Size([1, 13186]) Final batch size: 1, sequence length: 11974 Attention mask shape: torch.Size([1, 1, 11974, 11974]) Position ids shape: torch.Size([1, 11974]) Input IDs shape: torch.Size([1, 11974]) Labels shape: torch.Size([1, 11974]) Final batch size: 1, sequence length: 9027 Attention mask shape: torch.Size([1, 1, 9027, 9027]) Position ids shape: torch.Size([1, 9027]) Input IDs shape: torch.Size([1, 9027]) Labels shape: torch.Size([1, 9027]) Final batch size: 1, sequence length: 16104 Attention mask shape: torch.Size([1, 1, 16104, 16104]) Position ids shape: torch.Size([1, 16104]) Input IDs shape: torch.Size([1, 16104]) Labels shape: torch.Size([1, 16104]) Final batch size: 1, sequence length: 16330 Attention mask shape: torch.Size([1, 1, 16330, 16330]) Position ids shape: torch.Size([1, 16330]) Input IDs shape: torch.Size([1, 16330]) Labels shape: torch.Size([1, 16330]) Final batch size: 1, sequence length: 13957 Attention mask shape: torch.Size([1, 1, 13957, 13957]) Position ids shape: torch.Size([1, 13957]) Input IDs shape: torch.Size([1, 13957]) Labels shape: torch.Size([1, 13957]) Final batch size: 1, sequence length: 16496 Attention mask shape: torch.Size([1, 1, 16496, 16496]) Position ids shape: torch.Size([1, 16496]) Input IDs shape: torch.Size([1, 16496]) Labels shape: torch.Size([1, 16496]) Final batch size: 1, sequence length: 17092 Attention mask shape: torch.Size([1, 1, 17092, 17092]) Position ids shape: torch.Size([1, 17092]) Input IDs shape: torch.Size([1, 17092]) Labels shape: torch.Size([1, 17092]) Final batch size: 1, sequence length: 16366 Attention mask shape: torch.Size([1, 1, 16366, 16366]) Position ids shape: torch.Size([1, 16366]) Input IDs shape: torch.Size([1, 16366]) Labels shape: torch.Size([1, 16366]) Final batch size: 1, sequence length: 18958 Attention mask shape: torch.Size([1, 1, 18958, 18958]) Position ids shape: torch.Size([1, 18958]) Input IDs shape: torch.Size([1, 18958]) Labels shape: torch.Size([1, 18958]) Final batch size: 1, sequence length: 15283 Attention mask shape: torch.Size([1, 1, 15283, 15283]) Position ids shape: torch.Size([1, 15283]) Input IDs shape: torch.Size([1, 15283]) Labels shape: torch.Size([1, 15283]) Final batch size: 1, sequence length: 18415 Attention mask shape: torch.Size([1, 1, 18415, 18415]) Position ids shape: torch.Size([1, 18415]) Input IDs shape: torch.Size([1, 18415]) Labels shape: torch.Size([1, 18415]) Final batch size: 1, sequence length: 16014 Attention mask shape: torch.Size([1, 1, 16014, 16014]) Position ids shape: torch.Size([1, 16014]) Input IDs shape: torch.Size([1, 16014]) Labels shape: torch.Size([1, 16014]) Final batch size: 1, sequence length: 14492 Attention mask shape: torch.Size([1, 1, 14492, 14492]) Position ids shape: torch.Size([1, 14492]) Input IDs shape: torch.Size([1, 14492]) Labels shape: torch.Size([1, 14492]) Final batch size: 1, sequence length: 18034 Attention mask shape: torch.Size([1, 1, 18034, 18034]) Position ids shape: torch.Size([1, 18034]) Input IDs shape: torch.Size([1, 18034]) Labels shape: torch.Size([1, 18034]) Final batch size: 1, sequence length: 18938 Attention mask shape: torch.Size([1, 1, 18938, 18938]) Position ids shape: torch.Size([1, 18938]) Input IDs shape: torch.Size([1, 18938]) Labels shape: torch.Size([1, 18938]) Final batch size: 1, sequence length: 16088 Attention mask shape: torch.Size([1, 1, 16088, 16088]) Position ids shape: torch.Size([1, 16088]) Input IDs shape: torch.Size([1, 16088]) Labels shape: torch.Size([1, 16088]) Final batch size: 1, sequence length: 19603 Attention mask shape: torch.Size([1, 1, 19603, 19603]) Position ids shape: torch.Size([1, 19603]) Input IDs shape: torch.Size([1, 19603]) Labels shape: torch.Size([1, 19603]) Final batch size: 1, sequence length: 21705 Attention mask shape: torch.Size([1, 1, 21705, 21705]) Position ids shape: torch.Size([1, 21705]) Input IDs shape: torch.Size([1, 21705]) Labels shape: torch.Size([1, 21705]) Final batch size: 1, sequence length: 19034 Attention mask shape: torch.Size([1, 1, 19034, 19034]) Position ids shape: torch.Size([1, 19034]) Input IDs shape: torch.Size([1, 19034]) Labels shape: torch.Size([1, 19034]) Final batch size: 1, sequence length: 20726 Attention mask shape: torch.Size([1, 1, 20726, 20726]) Position ids shape: torch.Size([1, 20726]) Input IDs shape: torch.Size([1, 20726]) Labels shape: torch.Size([1, 20726]) Final batch size: 1, sequence length: 23132 Attention mask shape: torch.Size([1, 1, 23132, 23132]) Position ids shape: torch.Size([1, 23132]) Input IDs shape: torch.Size([1, 23132]) Labels shape: torch.Size([1, 23132]) Final batch size: 1, sequence length: 23102 Attention mask shape: torch.Size([1, 1, 23102, 23102]) Position ids shape: torch.Size([1, 23102]) Input IDs shape: torch.Size([1, 23102]) Labels shape: torch.Size([1, 23102]) Final batch size: 1, sequence length: 24432 Attention mask shape: torch.Size([1, 1, 24432, 24432]) Position ids shape: torch.Size([1, 24432]) Input IDs shape: torch.Size([1, 24432]) Labels shape: torch.Size([1, 24432]) Final batch size: 1, sequence length: 22771 Attention mask shape: torch.Size([1, 1, 22771, 22771]) Position ids shape: torch.Size([1, 22771]) Input IDs shape: torch.Size([1, 22771]) Labels shape: torch.Size([1, 22771]) Final batch size: 1, sequence length: 21404 Attention mask shape: torch.Size([1, 1, 21404, 21404]) Position ids shape: torch.Size([1, 21404]) Input IDs shape: torch.Size([1, 21404]) Labels shape: torch.Size([1, 21404]) Final batch size: 1, sequence length: 13468 Attention mask shape: torch.Size([1, 1, 13468, 13468]) Position ids shape: torch.Size([1, 13468]) Input IDs shape: torch.Size([1, 13468]) Labels shape: torch.Size([1, 13468]) Final batch size: 1, sequence length: 23971 Attention mask shape: torch.Size([1, 1, 23971, 23971]) Position ids shape: torch.Size([1, 23971]) Input IDs shape: torch.Size([1, 23971]) Labels shape: torch.Size([1, 23971]) Final batch size: 1, sequence length: 21132 Attention mask shape: torch.Size([1, 1, 21132, 21132]) Position ids shape: torch.Size([1, 21132]) Input IDs shape: torch.Size([1, 21132]) Labels shape: torch.Size([1, 21132]) Final batch size: 1, sequence length: 16590 Attention mask shape: torch.Size([1, 1, 16590, 16590]) Position ids shape: torch.Size([1, 16590]) Input IDs shape: torch.Size([1, 16590]) Labels shape: torch.Size([1, 16590]) Final batch size: 1, sequence length: 23894 Attention mask shape: torch.Size([1, 1, 23894, 23894]) Position ids shape: torch.Size([1, 23894]) Input IDs shape: torch.Size([1, 23894]) Labels shape: torch.Size([1, 23894]) Final batch size: 1, sequence length: 17514 Attention mask shape: torch.Size([1, 1, 17514, 17514]) Position ids shape: torch.Size([1, 17514]) Input IDs shape: torch.Size([1, 17514]) Labels shape: torch.Size([1, 17514]) Final batch size: 1, sequence length: 22328 Attention mask shape: torch.Size([1, 1, 22328, 22328]) Position ids shape: torch.Size([1, 22328]) Input IDs shape: torch.Size([1, 22328]) Labels shape: torch.Size([1, 22328]) Final batch size: 1, sequence length: 12756 Attention mask shape: torch.Size([1, 1, 12756, 12756]) Position ids shape: torch.Size([1, 12756]) Input IDs shape: torch.Size([1, 12756]) Labels shape: torch.Size([1, 12756]) Final batch size: 1, sequence length: 16291 Attention mask shape: torch.Size([1, 1, 16291, 16291]) Position ids shape: torch.Size([1, 16291]) Input IDs shape: torch.Size([1, 16291]) Labels shape: torch.Size([1, 16291]) Final batch size: 1, sequence length: 20817 Attention mask shape: torch.Size([1, 1, 20817, 20817]) Position ids shape: torch.Size([1, 20817]) Input IDs shape: torch.Size([1, 20817]) Labels shape: torch.Size([1, 20817]) Final batch size: 1, sequence length: 13790 Attention mask shape: torch.Size([1, 1, 13790, 13790]) Position ids shape: torch.Size([1, 13790]) Input IDs shape: torch.Size([1, 13790]) Labels shape: torch.Size([1, 13790]) Final batch size: 1, sequence length: 26375 Attention mask shape: torch.Size([1, 1, 26375, 26375]) Position ids shape: torch.Size([1, 26375]) Input IDs shape: torch.Size([1, 26375]) Labels shape: torch.Size([1, 26375]) Final batch size: 1, sequence length: 27780 Attention mask shape: torch.Size([1, 1, 27780, 27780]) Position ids shape: torch.Size([1, 27780]) Input IDs shape: torch.Size([1, 27780]) Labels shape: torch.Size([1, 27780]) Final batch size: 1, sequence length: 26766 Attention mask shape: torch.Size([1, 1, 26766, 26766]) Position ids shape: torch.Size([1, 26766]) Input IDs shape: torch.Size([1, 26766]) Labels shape: torch.Size([1, 26766]) Final batch size: 1, sequence length: 29132 Attention mask shape: torch.Size([1, 1, 29132, 29132]) Position ids shape: torch.Size([1, 29132]) Input IDs shape: torch.Size([1, 29132]) Labels shape: torch.Size([1, 29132]) Final batch size: 1, sequence length: 25351 Attention mask shape: torch.Size([1, 1, 25351, 25351]) Position ids shape: torch.Size([1, 25351]) Input IDs shape: torch.Size([1, 25351]) Labels shape: torch.Size([1, 25351]) Final batch size: 1, sequence length: 26596 Attention mask shape: torch.Size([1, 1, 26596, 26596]) Position ids shape: torch.Size([1, 26596]) Input IDs shape: torch.Size([1, 26596]) Labels shape: torch.Size([1, 26596]) Final batch size: 1, sequence length: 30165 Attention mask shape: torch.Size([1, 1, 30165, 30165]) Position ids shape: torch.Size([1, 30165]) Input IDs shape: torch.Size([1, 30165]) Labels shape: torch.Size([1, 30165]) Final batch size: 1, sequence length: 18869 Attention mask shape: torch.Size([1, 1, 18869, 18869]) Position ids shape: torch.Size([1, 18869]) Input IDs shape: torch.Size([1, 18869]) Labels shape: torch.Size([1, 18869]) Final batch size: 1, sequence length: 27785 Attention mask shape: torch.Size([1, 1, 27785, 27785]) Position ids shape: torch.Size([1, 27785]) Input IDs shape: torch.Size([1, 27785]) Labels shape: torch.Size([1, 27785]) Final batch size: 1, sequence length: 32022 Attention mask shape: torch.Size([1, 1, 32022, 32022]) Position ids shape: torch.Size([1, 32022]) Input IDs shape: torch.Size([1, 32022]) Labels shape: torch.Size([1, 32022]) Final batch size: 1, sequence length: 25388 Attention mask shape: torch.Size([1, 1, 25388, 25388]) Position ids shape: torch.Size([1, 25388]) Input IDs shape: torch.Size([1, 25388]) Labels shape: torch.Size([1, 25388]) Final batch size: 1, sequence length: 28574 Attention mask shape: torch.Size([1, 1, 28574, 28574]) Position ids shape: torch.Size([1, 28574]) Input IDs shape: torch.Size([1, 28574]) Labels shape: torch.Size([1, 28574]) Final batch size: 1, sequence length: 29168 Attention mask shape: torch.Size([1, 1, 29168, 29168]) Position ids shape: torch.Size([1, 29168]) Input IDs shape: torch.Size([1, 29168]) Labels shape: torch.Size([1, 29168]) Final batch size: 1, sequence length: 34457 Attention mask shape: torch.Size([1, 1, 34457, 34457]) Position ids shape: torch.Size([1, 34457]) Input IDs shape: torch.Size([1, 34457]) Labels shape: torch.Size([1, 34457]) Final batch size: 1, sequence length: 29830 Attention mask shape: torch.Size([1, 1, 29830, 29830]) Position ids shape: torch.Size([1, 29830]) Input IDs shape: torch.Size([1, 29830]) Labels shape: torch.Size([1, 29830]) Final batch size: 1, sequence length: 28412 Attention mask shape: torch.Size([1, 1, 28412, 28412]) Position ids shape: torch.Size([1, 28412]) Input IDs shape: torch.Size([1, 28412]) Labels shape: torch.Size([1, 28412]) Final batch size: 1, sequence length: 28845 Attention mask shape: torch.Size([1, 1, 28845, 28845]) Position ids shape: torch.Size([1, 28845]) Input IDs shape: torch.Size([1, 28845]) Labels shape: torch.Size([1, 28845]) Final batch size: 1, sequence length: 23810 Attention mask shape: torch.Size([1, 1, 23810, 23810]) Position ids shape: torch.Size([1, 23810]) Input IDs shape: torch.Size([1, 23810]) Labels shape: torch.Size([1, 23810]) Final batch size: 1, sequence length: 34326 Attention mask shape: torch.Size([1, 1, 34326, 34326]) Position ids shape: torch.Size([1, 34326]) Input IDs shape: torch.Size([1, 34326]) Labels shape: torch.Size([1, 34326]) Final batch size: 1, sequence length: 22043 Attention mask shape: torch.Size([1, 1, 22043, 22043]) Position ids shape: torch.Size([1, 22043]) Input IDs shape: torch.Size([1, 22043]) Labels shape: torch.Size([1, 22043]) Final batch size: 1, sequence length: 24653 Attention mask shape: torch.Size([1, 1, 24653, 24653]) Position ids shape: torch.Size([1, 24653]) Input IDs shape: torch.Size([1, 24653]) Labels shape: torch.Size([1, 24653]) Final batch size: 1, sequence length: 35868 Attention mask shape: torch.Size([1, 1, 35868, 35868]) Position ids shape: torch.Size([1, 35868]) Input IDs shape: torch.Size([1, 35868]) Labels shape: torch.Size([1, 35868]) Final batch size: 1, sequence length: 35790 Final batch size: 1, sequence length: 26465 Attention mask shape: torch.Size([1, 1, 35790, 35790]) Position ids shape: torch.Size([1, 35790]) Input IDs shape: torch.Size([1, 35790]) Labels shape: torch.Size([1, 35790]) Attention mask shape: torch.Size([1, 1, 26465, 26465]) Position ids shape: torch.Size([1, 26465]) Input IDs shape: torch.Size([1, 26465]) Labels shape: torch.Size([1, 26465]) Final batch size: 1, sequence length: 34874 Attention mask shape: torch.Size([1, 1, 34874, 34874]) Position ids shape: torch.Size([1, 34874]) Input IDs shape: torch.Size([1, 34874]) Labels shape: torch.Size([1, 34874]) Final batch size: 1, sequence length: 38919 Attention mask shape: torch.Size([1, 1, 38919, 38919]) Position ids shape: torch.Size([1, 38919]) Input IDs shape: torch.Size([1, 38919]) Labels shape: torch.Size([1, 38919]) Final batch size: 1, sequence length: 25923 Attention mask shape: torch.Size([1, 1, 25923, 25923]) Position ids shape: torch.Size([1, 25923]) Input IDs shape: torch.Size([1, 25923]) Labels shape: torch.Size([1, 25923]) Final batch size: 1, sequence length: 19744 Attention mask shape: torch.Size([1, 1, 19744, 19744]) Position ids shape: torch.Size([1, 19744]) Input IDs shape: torch.Size([1, 19744]) Labels shape: torch.Size([1, 19744]) Final batch size: 1, sequence length: 38617 Attention mask shape: torch.Size([1, 1, 38617, 38617]) Position ids shape: torch.Size([1, 38617]) Input IDs shape: torch.Size([1, 38617]) Labels shape: torch.Size([1, 38617]) Final batch size: 1, sequence length: 37039 Attention mask shape: torch.Size([1, 1, 37039, 37039]) Position ids shape: torch.Size([1, 37039]) Input IDs shape: torch.Size([1, 37039]) Labels shape: torch.Size([1, 37039]) Final batch size: 1, sequence length: 40717 Attention mask shape: torch.Size([1, 1, 40717, 40717]) Position ids shape: torch.Size([1, 40717]) Input IDs shape: torch.Size([1, 40717]) Labels shape: torch.Size([1, 40717]) Final batch size: 1, sequence length: 36794 Attention mask shape: torch.Size([1, 1, 36794, 36794]) Position ids shape: torch.Size([1, 36794]) Input IDs shape: torch.Size([1, 36794]) Labels shape: torch.Size([1, 36794]) Final batch size: 1, sequence length: 13986 Attention mask shape: torch.Size([1, 1, 13986, 13986]) Position ids shape: torch.Size([1, 13986]) Input IDs shape: torch.Size([1, 13986]) Labels shape: torch.Size([1, 13986]) Final batch size: 1, sequence length: 35803 Attention mask shape: torch.Size([1, 1, 35803, 35803]) Position ids shape: torch.Size([1, 35803]) Input IDs shape: torch.Size([1, 35803]) Labels shape: torch.Size([1, 35803]) Final batch size: 1, sequence length: 19204 Attention mask shape: torch.Size([1, 1, 19204, 19204]) Position ids shape: torch.Size([1, 19204]) Input IDs shape: torch.Size([1, 19204]) Labels shape: torch.Size([1, 19204]) Final batch size: 1, sequence length: 22014 Attention mask shape: torch.Size([1, 1, 22014, 22014]) Position ids shape: torch.Size([1, 22014]) Input IDs shape: torch.Size([1, 22014]) Labels shape: torch.Size([1, 22014]) Final batch size: 1, sequence length: 35999 Attention mask shape: torch.Size([1, 1, 35999, 35999]) Position ids shape: torch.Size([1, 35999]) Input IDs shape: torch.Size([1, 35999]) Labels shape: torch.Size([1, 35999]) Final batch size: 1, sequence length: 22282 Attention mask shape: torch.Size([1, 1, 22282, 22282]) Position ids shape: torch.Size([1, 22282]) Input IDs shape: torch.Size([1, 22282]) Labels shape: torch.Size([1, 22282]) Final batch size: 1, sequence length: 39150 Attention mask shape: torch.Size([1, 1, 39150, 39150]) Position ids shape: torch.Size([1, 39150]) Input IDs shape: torch.Size([1, 39150]) Labels shape: torch.Size([1, 39150]) Final batch size: 1, sequence length: 28002 Attention mask shape: torch.Size([1, 1, 28002, 28002]) Position ids shape: torch.Size([1, 28002]) Input IDs shape: torch.Size([1, 28002]) Labels shape: torch.Size([1, 28002]) Final batch size: 1, sequence length: 31506 Attention mask shape: torch.Size([1, 1, 31506, 31506]) Position ids shape: torch.Size([1, 31506]) Input IDs shape: torch.Size([1, 31506]) Labels shape: torch.Size([1, 31506]) Final batch size: 1, sequence length: 23245 Attention mask shape: torch.Size([1, 1, 23245, 23245]) Position ids shape: torch.Size([1, 23245]) Input IDs shape: torch.Size([1, 23245]) Labels shape: torch.Size([1, 23245]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28781 Attention mask shape: torch.Size([1, 1, 28781, 28781]) Position ids shape: torch.Size([1, 28781]) Input IDs shape: torch.Size([1, 28781]) Labels shape: torch.Size([1, 28781]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25560 Attention mask shape: torch.Size([1, 1, 25560, 25560]) Position ids shape: torch.Size([1, 25560]) Input IDs shape: torch.Size([1, 25560]) Labels shape: torch.Size([1, 25560]) Final batch size: 1, sequence length: 40065 Attention mask shape: torch.Size([1, 1, 40065, 40065]) Position ids shape: torch.Size([1, 40065]) Input IDs shape: torch.Size([1, 40065]) Labels shape: torch.Size([1, 40065]) Final batch size: 1, sequence length: 31340 Attention mask shape: torch.Size([1, 1, 31340, 31340]) Position ids shape: torch.Size([1, 31340]) Input IDs shape: torch.Size([1, 31340]) Labels shape: torch.Size([1, 31340]) Final batch size: 1, sequence length: 19846 Attention mask shape: torch.Size([1, 1, 19846, 19846]) Position ids shape: torch.Size([1, 19846]) Input IDs shape: torch.Size([1, 19846]) Labels shape: torch.Size([1, 19846]) Final batch size: 1, sequence length: 18379 Attention mask shape: torch.Size([1, 1, 18379, 18379]) Position ids shape: torch.Size([1, 18379]) Input IDs shape: torch.Size([1, 18379]) Labels shape: torch.Size([1, 18379]) Final batch size: 1, sequence length: 11644 Attention mask shape: torch.Size([1, 1, 11644, 11644]) Position ids shape: torch.Size([1, 11644]) Input IDs shape: torch.Size([1, 11644]) Labels shape: torch.Size([1, 11644]) Final batch size: 1, sequence length: 32974 Attention mask shape: torch.Size([1, 1, 32974, 32974]) Position ids shape: torch.Size([1, 32974]) Input IDs shape: torch.Size([1, 32974]) Labels shape: torch.Size([1, 32974]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31359 Attention mask shape: torch.Size([1, 1, 31359, 31359]) Position ids shape: torch.Size([1, 31359]) Input IDs shape: torch.Size([1, 31359]) Labels shape: torch.Size([1, 31359]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7364 Attention mask shape: torch.Size([1, 1, 7364, 7364]) Position ids shape: torch.Size([1, 7364]) Input IDs shape: torch.Size([1, 7364]) Labels shape: torch.Size([1, 7364]) Final batch size: 1, sequence length: 40874 Attention mask shape: torch.Size([1, 1, 40874, 40874]) Position ids shape: torch.Size([1, 40874]) Input IDs shape: torch.Size([1, 40874]) Labels shape: torch.Size([1, 40874]) Final batch size: 1, sequence length: 12418 Attention mask shape: torch.Size([1, 1, 12418, 12418]) Position ids shape: torch.Size([1, 12418]) Input IDs shape: torch.Size([1, 12418]) Labels shape: torch.Size([1, 12418]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 9309 Attention mask shape: torch.Size([1, 1, 9309, 9309]) Position ids shape: torch.Size([1, 9309]) Input IDs shape: torch.Size([1, 9309]) Labels shape: torch.Size([1, 9309]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 39919 Attention mask shape: torch.Size([1, 1, 39919, 39919]) Position ids shape: torch.Size([1, 39919]) Input IDs shape: torch.Size([1, 39919]) Labels shape: torch.Size([1, 39919]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26221 Attention mask shape: torch.Size([1, 1, 26221, 26221]) Position ids shape: torch.Size([1, 26221]) Input IDs shape: torch.Size([1, 26221]) Labels shape: torch.Size([1, 26221]) Final batch size: 1, sequence length: 17224 Attention mask shape: torch.Size([1, 1, 17224, 17224]) Position ids shape: torch.Size([1, 17224]) Input IDs shape: torch.Size([1, 17224]) Labels shape: torch.Size([1, 17224]) Final batch size: 1, sequence length: 18654 Attention mask shape: torch.Size([1, 1, 18654, 18654]) Position ids shape: torch.Size([1, 18654]) Input IDs shape: torch.Size([1, 18654]) Labels shape: torch.Size([1, 18654]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 29526 Attention mask shape: torch.Size([1, 1, 29526, 29526]) Position ids shape: torch.Size([1, 29526]) Input IDs shape: torch.Size([1, 29526]) Labels shape: torch.Size([1, 29526]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25820 Attention mask shape: torch.Size([1, 1, 25820, 25820]) Position ids shape: torch.Size([1, 25820]) Input IDs shape: torch.Size([1, 25820]) Labels shape: torch.Size([1, 25820]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30709 Attention mask shape: torch.Size([1, 1, 30709, 30709]) Position ids shape: torch.Size([1, 30709]) Input IDs shape: torch.Size([1, 30709]) Labels shape: torch.Size([1, 30709]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.3274, 'grad_norm': 0.5799930855931837, 'learning_rate': 9.45503262094184e-06, 'num_tokens': -inf, 'epoch': 1.75} Final batch size: 1, sequence length: 9858 Attention mask shape: torch.Size([1, 1, 9858, 9858]) Position ids shape: torch.Size([1, 9858]) Input IDs shape: torch.Size([1, 9858]) Labels shape: torch.Size([1, 9858]) Final batch size: 1, sequence length: 10826 Attention mask shape: torch.Size([1, 1, 10826, 10826]) Position ids shape: torch.Size([1, 10826]) Input IDs shape: torch.Size([1, 10826]) Labels shape: torch.Size([1, 10826]) Final batch size: 1, sequence length: 13427 Attention mask shape: torch.Size([1, 1, 13427, 13427]) Position ids shape: torch.Size([1, 13427]) Input IDs shape: torch.Size([1, 13427]) Labels shape: torch.Size([1, 13427]) Final batch size: 1, sequence length: 12279 Attention mask shape: torch.Size([1, 1, 12279, 12279]) Position ids shape: torch.Size([1, 12279]) Input IDs shape: torch.Size([1, 12279]) Labels shape: torch.Size([1, 12279]) Final batch size: 1, sequence length: 12096 Attention mask shape: torch.Size([1, 1, 12096, 12096]) Position ids shape: torch.Size([1, 12096]) Input IDs shape: torch.Size([1, 12096]) Labels shape: torch.Size([1, 12096]) Final batch size: 1, sequence length: 7461 Attention mask shape: torch.Size([1, 1, 7461, 7461]) Position ids shape: torch.Size([1, 7461]) Input IDs shape: torch.Size([1, 7461]) Labels shape: torch.Size([1, 7461]) Final batch size: 1, sequence length: 12622 Attention mask shape: torch.Size([1, 1, 12622, 12622]) Position ids shape: torch.Size([1, 12622]) Input IDs shape: torch.Size([1, 12622]) Labels shape: torch.Size([1, 12622]) Final batch size: 1, sequence length: 12960 Attention mask shape: torch.Size([1, 1, 12960, 12960]) Position ids shape: torch.Size([1, 12960]) Input IDs shape: torch.Size([1, 12960]) Labels shape: torch.Size([1, 12960]) Final batch size: 1, sequence length: 15933 Attention mask shape: torch.Size([1, 1, 15933, 15933]) Position ids shape: torch.Size([1, 15933]) Input IDs shape: torch.Size([1, 15933]) Labels shape: torch.Size([1, 15933]) Final batch size: 1, sequence length: 13550 Attention mask shape: torch.Size([1, 1, 13550, 13550]) Position ids shape: torch.Size([1, 13550]) Input IDs shape: torch.Size([1, 13550]) Labels shape: torch.Size([1, 13550]) Final batch size: 1, sequence length: 17934 Attention mask shape: torch.Size([1, 1, 17934, 17934]) Position ids shape: torch.Size([1, 17934]) Input IDs shape: torch.Size([1, 17934]) Labels shape: torch.Size([1, 17934]) Final batch size: 1, sequence length: 15673 Attention mask shape: torch.Size([1, 1, 15673, 15673]) Position ids shape: torch.Size([1, 15673]) Input IDs shape: torch.Size([1, 15673]) Labels shape: torch.Size([1, 15673]) Final batch size: 1, sequence length: 13789 Attention mask shape: torch.Size([1, 1, 13789, 13789]) Position ids shape: torch.Size([1, 13789]) Input IDs shape: torch.Size([1, 13789]) Labels shape: torch.Size([1, 13789]) Final batch size: 1, sequence length: 17245 Attention mask shape: torch.Size([1, 1, 17245, 17245]) Position ids shape: torch.Size([1, 17245]) Input IDs shape: torch.Size([1, 17245]) Labels shape: torch.Size([1, 17245]) Final batch size: 1, sequence length: 18214 Attention mask shape: torch.Size([1, 1, 18214, 18214]) Position ids shape: torch.Size([1, 18214]) Input IDs shape: torch.Size([1, 18214]) Labels shape: torch.Size([1, 18214]) Final batch size: 1, sequence length: 16450 Attention mask shape: torch.Size([1, 1, 16450, 16450]) Position ids shape: torch.Size([1, 16450]) Input IDs shape: torch.Size([1, 16450]) Labels shape: torch.Size([1, 16450]) Final batch size: 1, sequence length: 17861 Attention mask shape: torch.Size([1, 1, 17861, 17861]) Position ids shape: torch.Size([1, 17861]) Input IDs shape: torch.Size([1, 17861]) Labels shape: torch.Size([1, 17861]) Final batch size: 1, sequence length: 16927 Attention mask shape: torch.Size([1, 1, 16927, 16927]) Position ids shape: torch.Size([1, 16927]) Input IDs shape: torch.Size([1, 16927]) Labels shape: torch.Size([1, 16927]) Final batch size: 1, sequence length: 19899 Attention mask shape: torch.Size([1, 1, 19899, 19899]) Position ids shape: torch.Size([1, 19899]) Input IDs shape: torch.Size([1, 19899]) Labels shape: torch.Size([1, 19899]) Final batch size: 1, sequence length: 12728 Attention mask shape: torch.Size([1, 1, 12728, 12728]) Position ids shape: torch.Size([1, 12728]) Input IDs shape: torch.Size([1, 12728]) Labels shape: torch.Size([1, 12728]) Final batch size: 1, sequence length: 20292 Attention mask shape: torch.Size([1, 1, 20292, 20292]) Position ids shape: torch.Size([1, 20292]) Input IDs shape: torch.Size([1, 20292]) Labels shape: torch.Size([1, 20292]) Final batch size: 1, sequence length: 21091 Attention mask shape: torch.Size([1, 1, 21091, 21091]) Position ids shape: torch.Size([1, 21091]) Input IDs shape: torch.Size([1, 21091]) Labels shape: torch.Size([1, 21091]) Final batch size: 1, sequence length: 21675 Attention mask shape: torch.Size([1, 1, 21675, 21675]) Position ids shape: torch.Size([1, 21675]) Input IDs shape: torch.Size([1, 21675]) Labels shape: torch.Size([1, 21675]) Final batch size: 1, sequence length: 23004 Attention mask shape: torch.Size([1, 1, 23004, 23004]) Position ids shape: torch.Size([1, 23004]) Input IDs shape: torch.Size([1, 23004]) Labels shape: torch.Size([1, 23004]) Final batch size: 1, sequence length: 16036 Attention mask shape: torch.Size([1, 1, 16036, 16036]) Position ids shape: torch.Size([1, 16036]) Input IDs shape: torch.Size([1, 16036]) Labels shape: torch.Size([1, 16036]) Final batch size: 1, sequence length: 20065 Attention mask shape: torch.Size([1, 1, 20065, 20065]) Position ids shape: torch.Size([1, 20065]) Input IDs shape: torch.Size([1, 20065]) Labels shape: torch.Size([1, 20065]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 25176 Attention mask shape: torch.Size([1, 1, 25176, 25176]) Position ids shape: torch.Size([1, 25176]) Input IDs shape: torch.Size([1, 25176]) Labels shape: torch.Size([1, 25176]) Final batch size: 1, sequence length: 23896 Attention mask shape: torch.Size([1, 1, 23896, 23896]) Position ids shape: torch.Size([1, 23896]) Input IDs shape: torch.Size([1, 23896]) Labels shape: torch.Size([1, 23896]) Final batch size: 1, sequence length: 21586 Attention mask shape: torch.Size([1, 1, 21586, 21586]) Position ids shape: torch.Size([1, 21586]) Input IDs shape: torch.Size([1, 21586]) Labels shape: torch.Size([1, 21586]) Final batch size: 1, sequence length: 19953 Attention mask shape: torch.Size([1, 1, 19953, 19953]) Position ids shape: torch.Size([1, 19953]) Input IDs shape: torch.Size([1, 19953]) Labels shape: torch.Size([1, 19953]) Final batch size: 1, sequence length: 24633 Attention mask shape: torch.Size([1, 1, 24633, 24633]) Position ids shape: torch.Size([1, 24633]) Input IDs shape: torch.Size([1, 24633]) Labels shape: torch.Size([1, 24633]) Final batch size: 1, sequence length: 25035 Attention mask shape: torch.Size([1, 1, 25035, 25035]) Position ids shape: torch.Size([1, 25035]) Input IDs shape: torch.Size([1, 25035]) Labels shape: torch.Size([1, 25035]) Final batch size: 1, sequence length: 20582 Attention mask shape: torch.Size([1, 1, 20582, 20582]) Position ids shape: torch.Size([1, 20582]) Input IDs shape: torch.Size([1, 20582]) Labels shape: torch.Size([1, 20582]) Final batch size: 1, sequence length: 26316 Attention mask shape: torch.Size([1, 1, 26316, 26316]) Position ids shape: torch.Size([1, 26316]) Input IDs shape: torch.Size([1, 26316]) Labels shape: torch.Size([1, 26316]) Final batch size: 1, sequence length: 19187 Attention mask shape: torch.Size([1, 1, 19187, 19187]) Position ids shape: torch.Size([1, 19187]) Input IDs shape: torch.Size([1, 19187]) Labels shape: torch.Size([1, 19187]) Final batch size: 1, sequence length: 24808 Attention mask shape: torch.Size([1, 1, 24808, 24808]) Position ids shape: torch.Size([1, 24808]) Input IDs shape: torch.Size([1, 24808]) Labels shape: torch.Size([1, 24808]) Final batch size: 1, sequence length: 17763 Attention mask shape: torch.Size([1, 1, 17763, 17763]) Position ids shape: torch.Size([1, 17763]) Input IDs shape: torch.Size([1, 17763]) Labels shape: torch.Size([1, 17763]) Final batch size: 1, sequence length: 25600 Attention mask shape: torch.Size([1, 1, 25600, 25600]) Position ids shape: torch.Size([1, 25600]) Input IDs shape: torch.Size([1, 25600]) Labels shape: torch.Size([1, 25600]) Final batch size: 1, sequence length: 22778 Attention mask shape: torch.Size([1, 1, 22778, 22778]) Position ids shape: torch.Size([1, 22778]) Input IDs shape: torch.Size([1, 22778]) Labels shape: torch.Size([1, 22778]) Final batch size: 1, sequence length: 25435 Attention mask shape: torch.Size([1, 1, 25435, 25435]) Position ids shape: torch.Size([1, 25435]) Input IDs shape: torch.Size([1, 25435]) Labels shape: torch.Size([1, 25435]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 24956 Attention mask shape: torch.Size([1, 1, 24956, 24956]) Position ids shape: torch.Size([1, 24956]) Input IDs shape: torch.Size([1, 24956]) Labels shape: torch.Size([1, 24956]) Final batch size: 1, sequence length: 25325 Attention mask shape: torch.Size([1, 1, 25325, 25325]) Position ids shape: torch.Size([1, 25325]) Input IDs shape: torch.Size([1, 25325]) Labels shape: torch.Size([1, 25325]) Final batch size: 1, sequence length: 26247 Attention mask shape: torch.Size([1, 1, 26247, 26247]) Position ids shape: torch.Size([1, 26247]) Input IDs shape: torch.Size([1, 26247]) Labels shape: torch.Size([1, 26247]) Final batch size: 1, sequence length: 24965 Attention mask shape: torch.Size([1, 1, 24965, 24965]) Position ids shape: torch.Size([1, 24965]) Input IDs shape: torch.Size([1, 24965]) Labels shape: torch.Size([1, 24965]) Final batch size: 1, sequence length: 22775 Attention mask shape: torch.Size([1, 1, 22775, 22775]) Position ids shape: torch.Size([1, 22775]) Input IDs shape: torch.Size([1, 22775]) Labels shape: torch.Size([1, 22775]) Final batch size: 1, sequence length: 14212 Attention mask shape: torch.Size([1, 1, 14212, 14212]) Position ids shape: torch.Size([1, 14212]) Input IDs shape: torch.Size([1, 14212]) Labels shape: torch.Size([1, 14212]) Final batch size: 1, sequence length: 13946 Attention mask shape: torch.Size([1, 1, 13946, 13946]) Position ids shape: torch.Size([1, 13946]) Input IDs shape: torch.Size([1, 13946]) Labels shape: torch.Size([1, 13946]) Final batch size: 1, sequence length: 28926 Attention mask shape: torch.Size([1, 1, 28926, 28926]) Position ids shape: torch.Size([1, 28926]) Input IDs shape: torch.Size([1, 28926]) Labels shape: torch.Size([1, 28926]) Final batch size: 1, sequence length: 3010 Attention mask shape: torch.Size([1, 1, 3010, 3010]) Position ids shape: torch.Size([1, 3010]) Input IDs shape: torch.Size([1, 3010]) Labels shape: torch.Size([1, 3010]) Final batch size: 1, sequence length: 27222 Attention mask shape: torch.Size([1, 1, 27222, 27222]) Position ids shape: torch.Size([1, 27222]) Input IDs shape: torch.Size([1, 27222]) Labels shape: torch.Size([1, 27222]) Final batch size: 1, sequence length: 24927 Attention mask shape: torch.Size([1, 1, 24927, 24927]) Position ids shape: torch.Size([1, 24927]) Input IDs shape: torch.Size([1, 24927]) Labels shape: torch.Size([1, 24927]) Final batch size: 1, sequence length: 18832 Attention mask shape: torch.Size([1, 1, 18832, 18832]) Position ids shape: torch.Size([1, 18832]) Input IDs shape: torch.Size([1, 18832]) Labels shape: torch.Size([1, 18832]) Final batch size: 1, sequence length: 26500 Attention mask shape: torch.Size([1, 1, 26500, 26500]) Position ids shape: torch.Size([1, 26500]) Input IDs shape: torch.Size([1, 26500]) Labels shape: torch.Size([1, 26500]) Final batch size: 1, sequence length: 13453 Attention mask shape: torch.Size([1, 1, 13453, 13453]) Position ids shape: torch.Size([1, 13453]) Input IDs shape: torch.Size([1, 13453]) Labels shape: torch.Size([1, 13453]) Final batch size: 1, sequence length: 6978 Attention mask shape: torch.Size([1, 1, 6978, 6978]) Position ids shape: torch.Size([1, 6978]) Input IDs shape: torch.Size([1, 6978]) Labels shape: torch.Size([1, 6978]) Final batch size: 1, sequence length: 26454 Attention mask shape: torch.Size([1, 1, 26454, 26454]) Position ids shape: torch.Size([1, 26454]) Input IDs shape: torch.Size([1, 26454]) Labels shape: torch.Size([1, 26454]) Final batch size: 1, sequence length: 28552 Attention mask shape: torch.Size([1, 1, 28552, 28552]) Position ids shape: torch.Size([1, 28552]) Input IDs shape: torch.Size([1, 28552]) Labels shape: torch.Size([1, 28552]) Final batch size: 1, sequence length: 12426 Attention mask shape: torch.Size([1, 1, 12426, 12426]) Position ids shape: torch.Size([1, 12426]) Input IDs shape: torch.Size([1, 12426]) Labels shape: torch.Size([1, 12426]) Final batch size: 1, sequence length: 31208 Attention mask shape: torch.Size([1, 1, 31208, 31208]) Position ids shape: torch.Size([1, 31208]) Input IDs shape: torch.Size([1, 31208]) Labels shape: torch.Size([1, 31208]) Final batch size: 1, sequence length: 28644 Attention mask shape: torch.Size([1, 1, 28644, 28644]) Position ids shape: torch.Size([1, 28644]) Input IDs shape: torch.Size([1, 28644]) Labels shape: torch.Size([1, 28644]) Final batch size: 1, sequence length: 32101 Attention mask shape: torch.Size([1, 1, 32101, 32101]) Position ids shape: torch.Size([1, 32101]) Input IDs shape: torch.Size([1, 32101]) Labels shape: torch.Size([1, 32101]) Final batch size: 1, sequence length: 17049 Attention mask shape: torch.Size([1, 1, 17049, 17049]) Position ids shape: torch.Size([1, 17049]) Input IDs shape: torch.Size([1, 17049]) Labels shape: torch.Size([1, 17049]) Final batch size: 1, sequence length: 25979 Attention mask shape: torch.Size([1, 1, 25979, 25979]) Position ids shape: torch.Size([1, 25979]) Input IDs shape: torch.Size([1, 25979]) Labels shape: torch.Size([1, 25979]) Final batch size: 1, sequence length: 32100 Attention mask shape: torch.Size([1, 1, 32100, 32100]) Position ids shape: torch.Size([1, 32100]) Input IDs shape: torch.Size([1, 32100]) Labels shape: torch.Size([1, 32100]) Final batch size: 1, sequence length: 16714 Attention mask shape: torch.Size([1, 1, 16714, 16714]) Position ids shape: torch.Size([1, 16714]) Input IDs shape: torch.Size([1, 16714]) Labels shape: torch.Size([1, 16714]) Final batch size: 1, sequence length: 32513 Attention mask shape: torch.Size([1, 1, 32513, 32513]) Position ids shape: torch.Size([1, 32513]) Input IDs shape: torch.Size([1, 32513]) Labels shape: torch.Size([1, 32513]) Final batch size: 1, sequence length: 21290 Attention mask shape: torch.Size([1, 1, 21290, 21290]) Position ids shape: torch.Size([1, 21290]) Input IDs shape: torch.Size([1, 21290]) Labels shape: torch.Size([1, 21290]) Final batch size: 1, sequence length: 16571 Attention mask shape: torch.Size([1, 1, 16571, 16571]) Position ids shape: torch.Size([1, 16571]) Input IDs shape: torch.Size([1, 16571]) Labels shape: torch.Size([1, 16571]) Final batch size: 1, sequence length: 37381 Attention mask shape: torch.Size([1, 1, 37381, 37381]) Position ids shape: torch.Size([1, 37381]) Input IDs shape: torch.Size([1, 37381]) Labels shape: torch.Size([1, 37381]) Final batch size: 1, sequence length: 31662 Attention mask shape: torch.Size([1, 1, 31662, 31662]) Position ids shape: torch.Size([1, 31662]) Input IDs shape: torch.Size([1, 31662]) Labels shape: torch.Size([1, 31662]) Final batch size: 1, sequence length: 28377 Attention mask shape: torch.Size([1, 1, 28377, 28377]) Position ids shape: torch.Size([1, 28377]) Input IDs shape: torch.Size([1, 28377]) Labels shape: torch.Size([1, 28377]) Final batch size: 1, sequence length: 35014 Attention mask shape: torch.Size([1, 1, 35014, 35014]) Position ids shape: torch.Size([1, 35014]) Input IDs shape: torch.Size([1, 35014]) Labels shape: torch.Size([1, 35014]) Final batch size: 1, sequence length: 24676 Attention mask shape: torch.Size([1, 1, 24676, 24676]) Position ids shape: torch.Size([1, 24676]) Input IDs shape: torch.Size([1, 24676]) Labels shape: torch.Size([1, 24676]) Final batch size: 1, sequence length: 37391 Attention mask shape: torch.Size([1, 1, 37391, 37391]) Position ids shape: torch.Size([1, 37391]) Input IDs shape: torch.Size([1, 37391]) Labels shape: torch.Size([1, 37391]) Final batch size: 1, sequence length: 33621 Attention mask shape: torch.Size([1, 1, 33621, 33621]) Position ids shape: torch.Size([1, 33621]) Input IDs shape: torch.Size([1, 33621]) Labels shape: torch.Size([1, 33621]) Final batch size: 1, sequence length: 30635 Attention mask shape: torch.Size([1, 1, 30635, 30635]) Position ids shape: torch.Size([1, 30635]) Input IDs shape: torch.Size([1, 30635]) Labels shape: torch.Size([1, 30635]) Final batch size: 1, sequence length: 25147 Attention mask shape: torch.Size([1, 1, 25147, 25147]) Position ids shape: torch.Size([1, 25147]) Input IDs shape: torch.Size([1, 25147]) Labels shape: torch.Size([1, 25147]) Final batch size: 1, sequence length: 35058 Attention mask shape: torch.Size([1, 1, 35058, 35058]) Position ids shape: torch.Size([1, 35058]) Input IDs shape: torch.Size([1, 35058]) Labels shape: torch.Size([1, 35058]) Final batch size: 1, sequence length: 27021 Attention mask shape: torch.Size([1, 1, 27021, 27021]) Position ids shape: torch.Size([1, 27021]) Input IDs shape: torch.Size([1, 27021]) Labels shape: torch.Size([1, 27021]) Final batch size: 1, sequence length: 39474 Attention mask shape: torch.Size([1, 1, 39474, 39474]) Position ids shape: torch.Size([1, 39474]) Input IDs shape: torch.Size([1, 39474]) Labels shape: torch.Size([1, 39474]) Final batch size: 1, sequence length: 38391 Attention mask shape: torch.Size([1, 1, 38391, 38391]) Position ids shape: torch.Size([1, 38391]) Input IDs shape: torch.Size([1, 38391]) Labels shape: torch.Size([1, 38391]) Final batch size: 1, sequence length: 25219 Attention mask shape: torch.Size([1, 1, 25219, 25219]) Position ids shape: torch.Size([1, 25219]) Input IDs shape: torch.Size([1, 25219]) Labels shape: torch.Size([1, 25219]) Final batch size: 1, sequence length: 26789 Attention mask shape: torch.Size([1, 1, 26789, 26789]) Position ids shape: torch.Size([1, 26789]) Input IDs shape: torch.Size([1, 26789]) Labels shape: torch.Size([1, 26789]) Final batch size: 1, sequence length: 37720 Attention mask shape: torch.Size([1, 1, 37720, 37720]) Position ids shape: torch.Size([1, 37720]) Input IDs shape: torch.Size([1, 37720]) Labels shape: torch.Size([1, 37720]) Final batch size: 1, sequence length: 16816 Attention mask shape: torch.Size([1, 1, 16816, 16816]) Position ids shape: torch.Size([1, 16816]) Input IDs shape: torch.Size([1, 16816]) Labels shape: torch.Size([1, 16816]) Final batch size: 1, sequence length: 26387 Attention mask shape: torch.Size([1, 1, 26387, 26387]) Position ids shape: torch.Size([1, 26387]) Input IDs shape: torch.Size([1, 26387]) Labels shape: torch.Size([1, 26387]) Final batch size: 1, sequence length: 39955 Attention mask shape: torch.Size([1, 1, 39955, 39955]) Position ids shape: torch.Size([1, 39955]) Input IDs shape: torch.Size([1, 39955]) Labels shape: torch.Size([1, 39955]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 30259 Attention mask shape: torch.Size([1, 1, 30259, 30259]) Position ids shape: torch.Size([1, 30259]) Input IDs shape: torch.Size([1, 30259]) Labels shape: torch.Size([1, 30259]) Final batch size: 1, sequence length: 31447 Attention mask shape: torch.Size([1, 1, 31447, 31447]) Position ids shape: torch.Size([1, 31447]) Input IDs shape: torch.Size([1, 31447]) Labels shape: torch.Size([1, 31447]) Final batch size: 1, sequence length: 25774 Attention mask shape: torch.Size([1, 1, 25774, 25774]) Position ids shape: torch.Size([1, 25774]) Input IDs shape: torch.Size([1, 25774]) Labels shape: torch.Size([1, 25774]) Final batch size: 1, sequence length: 35261 Attention mask shape: torch.Size([1, 1, 35261, 35261]) Position ids shape: torch.Size([1, 35261]) Input IDs shape: torch.Size([1, 35261]) Labels shape: torch.Size([1, 35261]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22710 Attention mask shape: torch.Size([1, 1, 22710, 22710]) Position ids shape: torch.Size([1, 22710]) Input IDs shape: torch.Size([1, 22710]) Labels shape: torch.Size([1, 22710]) Final batch size: 1, sequence length: 36047 Attention mask shape: torch.Size([1, 1, 36047, 36047]) Position ids shape: torch.Size([1, 36047]) Input IDs shape: torch.Size([1, 36047]) Labels shape: torch.Size([1, 36047]) Final batch size: 1, sequence length: 21321 Attention mask shape: torch.Size([1, 1, 21321, 21321]) Position ids shape: torch.Size([1, 21321]) Input IDs shape: torch.Size([1, 21321]) Labels shape: torch.Size([1, 21321]) Final batch size: 1, sequence length: 19441 Attention mask shape: torch.Size([1, 1, 19441, 19441]) Position ids shape: torch.Size([1, 19441]) Input IDs shape: torch.Size([1, 19441]) Labels shape: torch.Size([1, 19441]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40590 Attention mask shape: torch.Size([1, 1, 40590, 40590]) Position ids shape: torch.Size([1, 40590]) Input IDs shape: torch.Size([1, 40590]) Labels shape: torch.Size([1, 40590]) Final batch size: 1, sequence length: 38577 Attention mask shape: torch.Size([1, 1, 38577, 38577]) Position ids shape: torch.Size([1, 38577]) Input IDs shape: torch.Size([1, 38577]) Labels shape: torch.Size([1, 38577]) Final batch size: 1, sequence length: 32079 Attention mask shape: torch.Size([1, 1, 32079, 32079]) Position ids shape: torch.Size([1, 32079]) Input IDs shape: torch.Size([1, 32079]) Labels shape: torch.Size([1, 32079]) Final batch size: 1, sequence length: 32129 Attention mask shape: torch.Size([1, 1, 32129, 32129]) Position ids shape: torch.Size([1, 32129]) Input IDs shape: torch.Size([1, 32129]) Labels shape: torch.Size([1, 32129]) Final batch size: 1, sequence length: 28448 Attention mask shape: torch.Size([1, 1, 28448, 28448]) Position ids shape: torch.Size([1, 28448]) Input IDs shape: torch.Size([1, 28448]) Labels shape: torch.Size([1, 28448]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 14136 Attention mask shape: torch.Size([1, 1, 14136, 14136]) Position ids shape: torch.Size([1, 14136]) Input IDs shape: torch.Size([1, 14136]) Labels shape: torch.Size([1, 14136]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33014 Attention mask shape: torch.Size([1, 1, 33014, 33014]) Position ids shape: torch.Size([1, 33014]) Input IDs shape: torch.Size([1, 33014]) Labels shape: torch.Size([1, 33014]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24623 Attention mask shape: torch.Size([1, 1, 24623, 24623]) Position ids shape: torch.Size([1, 24623]) Input IDs shape: torch.Size([1, 24623]) Labels shape: torch.Size([1, 24623]) Final batch size: 1, sequence length: 27371 Attention mask shape: torch.Size([1, 1, 27371, 27371]) Position ids shape: torch.Size([1, 27371]) Input IDs shape: torch.Size([1, 27371]) Labels shape: torch.Size([1, 27371]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36356 Attention mask shape: torch.Size([1, 1, 36356, 36356]) Position ids shape: torch.Size([1, 36356]) Input IDs shape: torch.Size([1, 36356]) Labels shape: torch.Size([1, 36356]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30712 Attention mask shape: torch.Size([1, 1, 30712, 30712]) Position ids shape: torch.Size([1, 30712]) Input IDs shape: torch.Size([1, 30712]) Labels shape: torch.Size([1, 30712]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19027 Attention mask shape: torch.Size([1, 1, 19027, 19027]) Position ids shape: torch.Size([1, 19027]) Input IDs shape: torch.Size([1, 19027]) Labels shape: torch.Size([1, 19027]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 23208 Attention mask shape: torch.Size([1, 1, 23208, 23208]) Position ids shape: torch.Size([1, 23208]) Input IDs shape: torch.Size([1, 23208]) Labels shape: torch.Size([1, 23208]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.3191, 'grad_norm': 0.5277645687433418, 'learning_rate': 9.330127018922195e-06, 'num_tokens': -inf, 'epoch': 1.88} Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 10507 Attention mask shape: torch.Size([1, 1, 10507, 10507]) Position ids shape: torch.Size([1, 10507]) Input IDs shape: torch.Size([1, 10507]) Labels shape: torch.Size([1, 10507]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 21841 Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 26271 Attention mask shape: torch.Size([1, 1, 26271, 26271]) Position ids shape: torch.Size([1, 26271]) Input IDs shape: torch.Size([1, 26271]) Labels shape: torch.Size([1, 26271]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 31381 Attention mask shape: torch.Size([1, 1, 31381, 31381]) Position ids shape: torch.Size([1, 31381]) Input IDs shape: torch.Size([1, 31381]) Labels shape: torch.Size([1, 31381]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.3134, 'grad_norm': 0.4407525210858162, 'learning_rate': 9.193352839727122e-06, 'num_tokens': -inf, 'epoch': 2.0} Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 6362 Attention mask shape: torch.Size([1, 1, 6362, 6362]) Position ids shape: torch.Size([1, 6362]) Input IDs shape: torch.Size([1, 6362]) Labels shape: torch.Size([1, 6362]) Final batch size: 1, sequence length: 10523 Attention mask shape: torch.Size([1, 1, 10523, 10523]) Position ids shape: torch.Size([1, 10523]) Input IDs shape: torch.Size([1, 10523]) Labels shape: torch.Size([1, 10523]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 12385 Attention mask shape: torch.Size([1, 1, 12385, 12385]) Position ids shape: torch.Size([1, 12385]) Input IDs shape: torch.Size([1, 12385]) Labels shape: torch.Size([1, 12385]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 18410 Attention mask shape: torch.Size([1, 1, 18410, 18410]) Position ids shape: torch.Size([1, 18410]) Input IDs shape: torch.Size([1, 18410]) Labels shape: torch.Size([1, 18410]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 18098 Attention mask shape: torch.Size([1, 1, 18098, 18098]) Position ids shape: torch.Size([1, 18098]) Input IDs shape: torch.Size([1, 18098]) Labels shape: torch.Size([1, 18098]) Final batch size: 1, sequence length: 12656 Attention mask shape: torch.Size([1, 1, 12656, 12656]) Position ids shape: torch.Size([1, 12656]) Input IDs shape: torch.Size([1, 12656]) Labels shape: torch.Size([1, 12656]) Final batch size: 1, sequence length: 17299 Attention mask shape: torch.Size([1, 1, 17299, 17299]) Position ids shape: torch.Size([1, 17299]) Input IDs shape: torch.Size([1, 17299]) Labels shape: torch.Size([1, 17299]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 24365 Attention mask shape: torch.Size([1, 1, 24365, 24365]) Position ids shape: torch.Size([1, 24365]) Input IDs shape: torch.Size([1, 24365]) Labels shape: torch.Size([1, 24365]) Final batch size: 1, sequence length: 18060 Attention mask shape: torch.Size([1, 1, 18060, 18060]) Position ids shape: torch.Size([1, 18060]) Input IDs shape: torch.Size([1, 18060]) Labels shape: torch.Size([1, 18060]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 9091 Attention mask shape: torch.Size([1, 1, 9091, 9091]) Position ids shape: torch.Size([1, 9091]) Input IDs shape: torch.Size([1, 9091]) Labels shape: torch.Size([1, 9091]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 18915 Attention mask shape: torch.Size([1, 1, 18915, 18915]) Position ids shape: torch.Size([1, 18915]) Input IDs shape: torch.Size([1, 18915]) Labels shape: torch.Size([1, 18915]) Final batch size: 1, sequence length: 11819 Attention mask shape: torch.Size([1, 1, 11819, 11819]) Position ids shape: torch.Size([1, 11819]) Input IDs shape: torch.Size([1, 11819]) Labels shape: torch.Size([1, 11819]) Final batch size: 1, sequence length: 28841 Attention mask shape: torch.Size([1, 1, 28841, 28841]) Position ids shape: torch.Size([1, 28841]) Input IDs shape: torch.Size([1, 28841]) Labels shape: torch.Size([1, 28841]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 26976 Attention mask shape: torch.Size([1, 1, 26976, 26976]) Position ids shape: torch.Size([1, 26976]) Input IDs shape: torch.Size([1, 26976]) Labels shape: torch.Size([1, 26976]) Final batch size: 1, sequence length: 29113 Attention mask shape: torch.Size([1, 1, 29113, 29113]) Position ids shape: torch.Size([1, 29113]) Input IDs shape: torch.Size([1, 29113]) Labels shape: torch.Size([1, 29113]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 27541 Attention mask shape: torch.Size([1, 1, 27541, 27541]) Position ids shape: torch.Size([1, 27541]) Input IDs shape: torch.Size([1, 27541]) Labels shape: torch.Size([1, 27541]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 24880 Attention mask shape: torch.Size([1, 1, 24880, 24880]) Position ids shape: torch.Size([1, 24880]) Input IDs shape: torch.Size([1, 24880]) Labels shape: torch.Size([1, 24880]) Final batch size: 1, sequence length: 7722 Attention mask shape: torch.Size([1, 1, 7722, 7722]) Position ids shape: torch.Size([1, 7722]) Input IDs shape: torch.Size([1, 7722]) Labels shape: torch.Size([1, 7722]) Final batch size: 1, sequence length: 25172 Attention mask shape: torch.Size([1, 1, 25172, 25172]) Position ids shape: torch.Size([1, 25172]) Input IDs shape: torch.Size([1, 25172]) Labels shape: torch.Size([1, 25172]) Final batch size: 1, sequence length: 21596 Attention mask shape: torch.Size([1, 1, 21596, 21596]) Position ids shape: torch.Size([1, 21596]) Input IDs shape: torch.Size([1, 21596]) Labels shape: torch.Size([1, 21596]) Final batch size: 1, sequence length: 21450 Attention mask shape: torch.Size([1, 1, 21450, 21450]) Position ids shape: torch.Size([1, 21450]) Input IDs shape: torch.Size([1, 21450]) Labels shape: torch.Size([1, 21450]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 19705 Attention mask shape: torch.Size([1, 1, 19705, 19705]) Position ids shape: torch.Size([1, 19705]) Input IDs shape: torch.Size([1, 19705]) Labels shape: torch.Size([1, 19705]) Final batch size: 1, sequence length: 6948 Attention mask shape: torch.Size([1, 1, 6948, 6948]) Position ids shape: torch.Size([1, 6948]) Input IDs shape: torch.Size([1, 6948]) Labels shape: torch.Size([1, 6948]) Final batch size: 1, sequence length: 21491 Attention mask shape: torch.Size([1, 1, 21491, 21491]) Position ids shape: torch.Size([1, 21491]) Input IDs shape: torch.Size([1, 21491]) Labels shape: torch.Size([1, 21491]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 14009 Attention mask shape: torch.Size([1, 1, 14009, 14009]) Position ids shape: torch.Size([1, 14009]) Input IDs shape: torch.Size([1, 14009]) Labels shape: torch.Size([1, 14009]) Final batch size: 1, sequence length: 29875 Attention mask shape: torch.Size([1, 1, 29875, 29875]) Position ids shape: torch.Size([1, 29875]) Input IDs shape: torch.Size([1, 29875]) Labels shape: torch.Size([1, 29875]) Final batch size: 1, sequence length: 30623 Attention mask shape: torch.Size([1, 1, 30623, 30623]) Position ids shape: torch.Size([1, 30623]) Input IDs shape: torch.Size([1, 30623]) Labels shape: torch.Size([1, 30623]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 20827 Attention mask shape: torch.Size([1, 1, 20827, 20827]) Position ids shape: torch.Size([1, 20827]) Input IDs shape: torch.Size([1, 20827]) Labels shape: torch.Size([1, 20827]) Final batch size: 1, sequence length: 25014 Attention mask shape: torch.Size([1, 1, 25014, 25014]) Position ids shape: torch.Size([1, 25014]) Input IDs shape: torch.Size([1, 25014]) Labels shape: torch.Size([1, 25014]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 20363 Attention mask shape: torch.Size([1, 1, 20363, 20363]) Position ids shape: torch.Size([1, 20363]) Input IDs shape: torch.Size([1, 20363]) Labels shape: torch.Size([1, 20363]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 17400 Attention mask shape: torch.Size([1, 1, 17400, 17400]) Position ids shape: torch.Size([1, 17400]) Input IDs shape: torch.Size([1, 17400]) Labels shape: torch.Size([1, 17400]) Final batch size: 1, sequence length: 13215 Attention mask shape: torch.Size([1, 1, 13215, 13215]) Position ids shape: torch.Size([1, 13215]) Input IDs shape: torch.Size([1, 13215]) Labels shape: torch.Size([1, 13215]) Final batch size: 1, sequence length: 28641 Attention mask shape: torch.Size([1, 1, 28641, 28641]) Position ids shape: torch.Size([1, 28641]) Input IDs shape: torch.Size([1, 28641]) Labels shape: torch.Size([1, 28641]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 33367 Attention mask shape: torch.Size([1, 1, 33367, 33367]) Position ids shape: torch.Size([1, 33367]) Input IDs shape: torch.Size([1, 33367]) Labels shape: torch.Size([1, 33367]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 27265 Attention mask shape: torch.Size([1, 1, 27265, 27265]) Position ids shape: torch.Size([1, 27265]) Input IDs shape: torch.Size([1, 27265]) Labels shape: torch.Size([1, 27265]) Final batch size: 1, sequence length: 22098 Attention mask shape: torch.Size([1, 1, 22098, 22098]) Position ids shape: torch.Size([1, 22098]) Input IDs shape: torch.Size([1, 22098]) Labels shape: torch.Size([1, 22098]) Final batch size: 1, sequence length: 24001 Attention mask shape: torch.Size([1, 1, 24001, 24001]) Position ids shape: torch.Size([1, 24001]) Input IDs shape: torch.Size([1, 24001]) Labels shape: torch.Size([1, 24001]) Final batch size: 1, sequence length: 15656 Attention mask shape: torch.Size([1, 1, 15656, 15656]) Position ids shape: torch.Size([1, 15656]) Input IDs shape: torch.Size([1, 15656]) Labels shape: torch.Size([1, 15656]) Final batch size: 1, sequence length: 27243 Attention mask shape: torch.Size([1, 1, 27243, 27243]) Position ids shape: torch.Size([1, 27243]) Input IDs shape: torch.Size([1, 27243]) Labels shape: torch.Size([1, 27243]) Final batch size: 1, sequence length: 18606 Attention mask shape: torch.Size([1, 1, 18606, 18606]) Position ids shape: torch.Size([1, 18606]) Input IDs shape: torch.Size([1, 18606]) Labels shape: torch.Size([1, 18606]) Final batch size: 1, sequence length: 38249 Attention mask shape: torch.Size([1, 1, 38249, 38249]) Position ids shape: torch.Size([1, 38249]) Input IDs shape: torch.Size([1, 38249]) Labels shape: torch.Size([1, 38249]) Final batch size: 1, sequence length: 21348 Attention mask shape: torch.Size([1, 1, 21348, 21348]) Position ids shape: torch.Size([1, 21348]) Input IDs shape: torch.Size([1, 21348]) Labels shape: torch.Size([1, 21348]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 19586 Attention mask shape: torch.Size([1, 1, 19586, 19586]) Position ids shape: torch.Size([1, 19586]) Input IDs shape: torch.Size([1, 19586]) Labels shape: torch.Size([1, 19586]) Final batch size: 1, sequence length: 29632 Attention mask shape: torch.Size([1, 1, 29632, 29632]) Position ids shape: torch.Size([1, 29632]) Input IDs shape: torch.Size([1, 29632]) Labels shape: torch.Size([1, 29632]) Final batch size: 1, sequence length: 30066 Attention mask shape: torch.Size([1, 1, 30066, 30066]) Position ids shape: torch.Size([1, 30066]) Input IDs shape: torch.Size([1, 30066]) Labels shape: torch.Size([1, 30066]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 22625 Attention mask shape: torch.Size([1, 1, 22625, 22625]) Position ids shape: torch.Size([1, 22625]) Input IDs shape: torch.Size([1, 22625]) Labels shape: torch.Size([1, 22625]) Final batch size: 1, sequence length: 32609 Attention mask shape: torch.Size([1, 1, 32609, 32609]) Position ids shape: torch.Size([1, 32609]) Input IDs shape: torch.Size([1, 32609]) Labels shape: torch.Size([1, 32609]) Final batch size: 1, sequence length: 27441 Attention mask shape: torch.Size([1, 1, 27441, 27441]) Position ids shape: torch.Size([1, 27441]) Input IDs shape: torch.Size([1, 27441]) Labels shape: torch.Size([1, 27441]) Final batch size: 1, sequence length: 39142 Attention mask shape: torch.Size([1, 1, 39142, 39142]) Position ids shape: torch.Size([1, 39142]) Input IDs shape: torch.Size([1, 39142]) Labels shape: torch.Size([1, 39142]) Final batch size: 1, sequence length: 17778 Attention mask shape: torch.Size([1, 1, 17778, 17778]) Position ids shape: torch.Size([1, 17778]) Input IDs shape: torch.Size([1, 17778]) Labels shape: torch.Size([1, 17778]) Final batch size: 1, sequence length: 26939 Attention mask shape: torch.Size([1, 1, 26939, 26939]) Position ids shape: torch.Size([1, 26939]) Input IDs shape: torch.Size([1, 26939]) Labels shape: torch.Size([1, 26939]) Final batch size: 1, sequence length: 13509 Attention mask shape: torch.Size([1, 1, 13509, 13509]) Position ids shape: torch.Size([1, 13509]) Input IDs shape: torch.Size([1, 13509]) Labels shape: torch.Size([1, 13509]) Final batch size: 1, sequence length: 15217 Attention mask shape: torch.Size([1, 1, 15217, 15217]) Position ids shape: torch.Size([1, 15217]) Input IDs shape: torch.Size([1, 15217]) Labels shape: torch.Size([1, 15217]) Final batch size: 1, sequence length: 18565 Attention mask shape: torch.Size([1, 1, 18565, 18565]) Position ids shape: torch.Size([1, 18565]) Input IDs shape: torch.Size([1, 18565]) Labels shape: torch.Size([1, 18565]) Final batch size: 1, sequence length: 21567 Attention mask shape: torch.Size([1, 1, 21567, 21567]) Position ids shape: torch.Size([1, 21567]) Input IDs shape: torch.Size([1, 21567]) Labels shape: torch.Size([1, 21567]) Final batch size: 1, sequence length: 26306 Attention mask shape: torch.Size([1, 1, 26306, 26306]) Position ids shape: torch.Size([1, 26306]) Input IDs shape: torch.Size([1, 26306]) Labels shape: torch.Size([1, 26306]) Final batch size: 1, sequence length: 30802 Attention mask shape: torch.Size([1, 1, 30802, 30802]) Position ids shape: torch.Size([1, 30802]) Input IDs shape: torch.Size([1, 30802]) Labels shape: torch.Size([1, 30802]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 35153 Attention mask shape: torch.Size([1, 1, 35153, 35153]) Position ids shape: torch.Size([1, 35153]) Input IDs shape: torch.Size([1, 35153]) Labels shape: torch.Size([1, 35153]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 40496 Attention mask shape: torch.Size([1, 1, 40496, 40496]) Position ids shape: torch.Size([1, 40496]) Input IDs shape: torch.Size([1, 40496]) Labels shape: torch.Size([1, 40496]) Final batch size: 1, sequence length: 33839 Attention mask shape: torch.Size([1, 1, 33839, 33839]) Position ids shape: torch.Size([1, 33839]) Input IDs shape: torch.Size([1, 33839]) Labels shape: torch.Size([1, 33839]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40745 Attention mask shape: torch.Size([1, 1, 40745, 40745]) Position ids shape: torch.Size([1, 40745]) Input IDs shape: torch.Size([1, 40745]) Labels shape: torch.Size([1, 40745]) Final batch size: 1, sequence length: 25447 Attention mask shape: torch.Size([1, 1, 25447, 25447]) Position ids shape: torch.Size([1, 25447]) Input IDs shape: torch.Size([1, 25447]) Labels shape: torch.Size([1, 25447]) Final batch size: 1, sequence length: 30101 Attention mask shape: torch.Size([1, 1, 30101, 30101]) Position ids shape: torch.Size([1, 30101]) Input IDs shape: torch.Size([1, 30101]) Labels shape: torch.Size([1, 30101]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32352 Attention mask shape: torch.Size([1, 1, 32352, 32352]) Position ids shape: torch.Size([1, 32352]) Input IDs shape: torch.Size([1, 32352]) Labels shape: torch.Size([1, 32352]) Final batch size: 1, sequence length: 36511 Attention mask shape: torch.Size([1, 1, 36511, 36511]) Position ids shape: torch.Size([1, 36511]) Input IDs shape: torch.Size([1, 36511]) Labels shape: torch.Size([1, 36511]) Final batch size: 1, sequence length: 15508 Attention mask shape: torch.Size([1, 1, 15508, 15508]) Position ids shape: torch.Size([1, 15508]) Input IDs shape: torch.Size([1, 15508]) Labels shape: torch.Size([1, 15508]) Final batch size: 1, sequence length: 30109 Attention mask shape: torch.Size([1, 1, 30109, 30109]) Position ids shape: torch.Size([1, 30109]) Input IDs shape: torch.Size([1, 30109]) Labels shape: torch.Size([1, 30109]) Final batch size: 1, sequence length: 24622 Attention mask shape: torch.Size([1, 1, 24622, 24622]) Position ids shape: torch.Size([1, 24622]) Input IDs shape: torch.Size([1, 24622]) Labels shape: torch.Size([1, 24622]) Final batch size: 1, sequence length: 19036 Attention mask shape: torch.Size([1, 1, 19036, 19036]) Position ids shape: torch.Size([1, 19036]) Input IDs shape: torch.Size([1, 19036]) Labels shape: torch.Size([1, 19036]) Final batch size: 1, sequence length: 9947 Attention mask shape: torch.Size([1, 1, 9947, 9947]) Position ids shape: torch.Size([1, 9947]) Input IDs shape: torch.Size([1, 9947]) Labels shape: torch.Size([1, 9947]) Final batch size: 1, sequence length: 16677 Attention mask shape: torch.Size([1, 1, 16677, 16677]) Position ids shape: torch.Size([1, 16677]) Input IDs shape: torch.Size([1, 16677]) Labels shape: torch.Size([1, 16677]) Final batch size: 1, sequence length: 23258 Attention mask shape: torch.Size([1, 1, 23258, 23258]) Position ids shape: torch.Size([1, 23258]) Input IDs shape: torch.Size([1, 23258]) Labels shape: torch.Size([1, 23258]) Final batch size: 1, sequence length: 36271 Attention mask shape: torch.Size([1, 1, 36271, 36271]) Position ids shape: torch.Size([1, 36271]) Input IDs shape: torch.Size([1, 36271]) Labels shape: torch.Size([1, 36271]) Final batch size: 1, sequence length: 20422 Attention mask shape: torch.Size([1, 1, 20422, 20422]) Position ids shape: torch.Size([1, 20422]) Input IDs shape: torch.Size([1, 20422]) Labels shape: torch.Size([1, 20422]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 13095 Attention mask shape: torch.Size([1, 1, 13095, 13095]) Position ids shape: torch.Size([1, 13095]) Input IDs shape: torch.Size([1, 13095]) Labels shape: torch.Size([1, 13095]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 16587 Attention mask shape: torch.Size([1, 1, 16587, 16587]) Position ids shape: torch.Size([1, 16587]) Input IDs shape: torch.Size([1, 16587]) Labels shape: torch.Size([1, 16587]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 39836 Attention mask shape: torch.Size([1, 1, 39836, 39836]) Position ids shape: torch.Size([1, 39836]) Input IDs shape: torch.Size([1, 39836]) Labels shape: torch.Size([1, 39836]) Final batch size: 1, sequence length: 20611 Attention mask shape: torch.Size([1, 1, 20611, 20611]) Position ids shape: torch.Size([1, 20611]) Input IDs shape: torch.Size([1, 20611]) Labels shape: torch.Size([1, 20611]) Final batch size: 1, sequence length: 37945 Attention mask shape: torch.Size([1, 1, 37945, 37945]) Position ids shape: torch.Size([1, 37945]) Input IDs shape: torch.Size([1, 37945]) Labels shape: torch.Size([1, 37945]) Final batch size: 1, sequence length: 34937 Attention mask shape: torch.Size([1, 1, 34937, 34937]) Position ids shape: torch.Size([1, 34937]) Input IDs shape: torch.Size([1, 34937]) Labels shape: torch.Size([1, 34937]) Final batch size: 1, sequence length: 13297 Attention mask shape: torch.Size([1, 1, 13297, 13297]) Position ids shape: torch.Size([1, 13297]) Input IDs shape: torch.Size([1, 13297]) Labels shape: torch.Size([1, 13297]) Final batch size: 1, sequence length: 24298 Attention mask shape: torch.Size([1, 1, 24298, 24298]) Position ids shape: torch.Size([1, 24298]) Input IDs shape: torch.Size([1, 24298]) Labels shape: torch.Size([1, 24298]) Final batch size: 1, sequence length: 34142 Attention mask shape: torch.Size([1, 1, 34142, 34142]) Position ids shape: torch.Size([1, 34142]) Input IDs shape: torch.Size([1, 34142]) Labels shape: torch.Size([1, 34142]) Final batch size: 1, sequence length: 22786 Attention mask shape: torch.Size([1, 1, 22786, 22786]) Position ids shape: torch.Size([1, 22786]) Input IDs shape: torch.Size([1, 22786]) Labels shape: torch.Size([1, 22786]) Final batch size: 1, sequence length: 21758 Attention mask shape: torch.Size([1, 1, 21758, 21758]) Position ids shape: torch.Size([1, 21758]) Input IDs shape: torch.Size([1, 21758]) Labels shape: torch.Size([1, 21758]) Final batch size: 1, sequence length: 12224 Attention mask shape: torch.Size([1, 1, 12224, 12224]) Position ids shape: torch.Size([1, 12224]) Input IDs shape: torch.Size([1, 12224]) Labels shape: torch.Size([1, 12224]) Final batch size: 1, sequence length: 12653 Attention mask shape: torch.Size([1, 1, 12653, 12653]) Position ids shape: torch.Size([1, 12653]) Input IDs shape: torch.Size([1, 12653]) Labels shape: torch.Size([1, 12653]) Final batch size: 1, sequence length: 28634 Attention mask shape: torch.Size([1, 1, 28634, 28634]) Position ids shape: torch.Size([1, 28634]) Input IDs shape: torch.Size([1, 28634]) Labels shape: torch.Size([1, 28634]) Final batch size: 1, sequence length: 32472 Attention mask shape: torch.Size([1, 1, 32472, 32472]) Position ids shape: torch.Size([1, 32472]) Input IDs shape: torch.Size([1, 32472]) Labels shape: torch.Size([1, 32472]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 14995 Attention mask shape: torch.Size([1, 1, 14995, 14995]) Position ids shape: torch.Size([1, 14995]) Input IDs shape: torch.Size([1, 14995]) Labels shape: torch.Size([1, 14995]) Final batch size: 1, sequence length: 21683 Attention mask shape: torch.Size([1, 1, 21683, 21683]) Position ids shape: torch.Size([1, 21683]) Input IDs shape: torch.Size([1, 21683]) Labels shape: torch.Size([1, 21683]) Final batch size: 1, sequence length: 17373 Attention mask shape: torch.Size([1, 1, 17373, 17373]) Position ids shape: torch.Size([1, 17373]) Input IDs shape: torch.Size([1, 17373]) Labels shape: torch.Size([1, 17373]) Final batch size: 1, sequence length: 22623 Attention mask shape: torch.Size([1, 1, 22623, 22623]) Position ids shape: torch.Size([1, 22623]) Input IDs shape: torch.Size([1, 22623]) Labels shape: torch.Size([1, 22623]) Final batch size: 1, sequence length: 32753 Attention mask shape: torch.Size([1, 1, 32753, 32753]) Position ids shape: torch.Size([1, 32753]) Input IDs shape: torch.Size([1, 32753]) Labels shape: torch.Size([1, 32753]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24424 Attention mask shape: torch.Size([1, 1, 24424, 24424]) Position ids shape: torch.Size([1, 24424]) Input IDs shape: torch.Size([1, 24424]) Labels shape: torch.Size([1, 24424]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32564 Attention mask shape: torch.Size([1, 1, 32564, 32564]) Position ids shape: torch.Size([1, 32564]) Input IDs shape: torch.Size([1, 32564]) Labels shape: torch.Size([1, 32564]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 35864 Attention mask shape: torch.Size([1, 1, 35864, 35864]) Position ids shape: torch.Size([1, 35864]) Input IDs shape: torch.Size([1, 35864]) Labels shape: torch.Size([1, 35864]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32344 Attention mask shape: torch.Size([1, 1, 32344, 32344]) Position ids shape: torch.Size([1, 32344]) Input IDs shape: torch.Size([1, 32344]) Labels shape: torch.Size([1, 32344]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26537 Attention mask shape: torch.Size([1, 1, 26537, 26537]) Position ids shape: torch.Size([1, 26537]) Input IDs shape: torch.Size([1, 26537]) Labels shape: torch.Size([1, 26537]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32919 Attention mask shape: torch.Size([1, 1, 32919, 32919]) Position ids shape: torch.Size([1, 32919]) Input IDs shape: torch.Size([1, 32919]) Labels shape: torch.Size([1, 32919]) Final batch size: 1, sequence length: 37168 Attention mask shape: torch.Size([1, 1, 37168, 37168]) Position ids shape: torch.Size([1, 37168]) Input IDs shape: torch.Size([1, 37168]) Labels shape: torch.Size([1, 37168]) Final batch size: 1, sequence length: 32514 Attention mask shape: torch.Size([1, 1, 32514, 32514]) Position ids shape: torch.Size([1, 32514]) Input IDs shape: torch.Size([1, 32514]) Labels shape: torch.Size([1, 32514]) Final batch size: 1, sequence length: 23362 Attention mask shape: torch.Size([1, 1, 23362, 23362]) Position ids shape: torch.Size([1, 23362]) Input IDs shape: torch.Size([1, 23362]) Labels shape: torch.Size([1, 23362]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28861 Attention mask shape: torch.Size([1, 1, 28861, 28861]) Position ids shape: torch.Size([1, 28861]) Input IDs shape: torch.Size([1, 28861]) Labels shape: torch.Size([1, 28861]) Final batch size: 1, sequence length: 26006 Attention mask shape: torch.Size([1, 1, 26006, 26006]) Position ids shape: torch.Size([1, 26006]) Input IDs shape: torch.Size([1, 26006]) Labels shape: torch.Size([1, 26006]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.3054, 'grad_norm': 0.3866579396690133, 'learning_rate': 9.045084971874738e-06, 'num_tokens': -inf, 'epoch': 2.12} Final batch size: 1, sequence length: 7360 Attention mask shape: torch.Size([1, 1, 7360, 7360]) Position ids shape: torch.Size([1, 7360]) Input IDs shape: torch.Size([1, 7360]) Labels shape: torch.Size([1, 7360]) Final batch size: 1, sequence length: 4858 Attention mask shape: torch.Size([1, 1, 4858, 4858]) Position ids shape: torch.Size([1, 4858]) Input IDs shape: torch.Size([1, 4858]) Labels shape: torch.Size([1, 4858]) Final batch size: 1, sequence length: 6316 Attention mask shape: torch.Size([1, 1, 6316, 6316]) Position ids shape: torch.Size([1, 6316]) Input IDs shape: torch.Size([1, 6316]) Labels shape: torch.Size([1, 6316]) Final batch size: 1, sequence length: 11728 Attention mask shape: torch.Size([1, 1, 11728, 11728]) Position ids shape: torch.Size([1, 11728]) Input IDs shape: torch.Size([1, 11728]) Labels shape: torch.Size([1, 11728]) Final batch size: 1, sequence length: 11548 Attention mask shape: torch.Size([1, 1, 11548, 11548]) Position ids shape: torch.Size([1, 11548]) Input IDs shape: torch.Size([1, 11548]) Labels shape: torch.Size([1, 11548]) Final batch size: 1, sequence length: 12293 Attention mask shape: torch.Size([1, 1, 12293, 12293]) Position ids shape: torch.Size([1, 12293]) Input IDs shape: torch.Size([1, 12293]) Labels shape: torch.Size([1, 12293]) Final batch size: 1, sequence length: 10301 Attention mask shape: torch.Size([1, 1, 10301, 10301]) Position ids shape: torch.Size([1, 10301]) Input IDs shape: torch.Size([1, 10301]) Labels shape: torch.Size([1, 10301]) Final batch size: 1, sequence length: 16827 Attention mask shape: torch.Size([1, 1, 16827, 16827]) Position ids shape: torch.Size([1, 16827]) Input IDs shape: torch.Size([1, 16827]) Labels shape: torch.Size([1, 16827]) Final batch size: 1, sequence length: 13355 Attention mask shape: torch.Size([1, 1, 13355, 13355]) Position ids shape: torch.Size([1, 13355]) Input IDs shape: torch.Size([1, 13355]) Labels shape: torch.Size([1, 13355]) Final batch size: 1, sequence length: 17108 Attention mask shape: torch.Size([1, 1, 17108, 17108]) Position ids shape: torch.Size([1, 17108]) Input IDs shape: torch.Size([1, 17108]) Labels shape: torch.Size([1, 17108]) Final batch size: 1, sequence length: 14363 Attention mask shape: torch.Size([1, 1, 14363, 14363]) Position ids shape: torch.Size([1, 14363]) Input IDs shape: torch.Size([1, 14363]) Labels shape: torch.Size([1, 14363]) Final batch size: 1, sequence length: 14587 Attention mask shape: torch.Size([1, 1, 14587, 14587]) Position ids shape: torch.Size([1, 14587]) Input IDs shape: torch.Size([1, 14587]) Labels shape: torch.Size([1, 14587]) Final batch size: 1, sequence length: 14704 Attention mask shape: torch.Size([1, 1, 14704, 14704]) Position ids shape: torch.Size([1, 14704]) Input IDs shape: torch.Size([1, 14704]) Labels shape: torch.Size([1, 14704]) Final batch size: 1, sequence length: 16536 Attention mask shape: torch.Size([1, 1, 16536, 16536]) Position ids shape: torch.Size([1, 16536]) Input IDs shape: torch.Size([1, 16536]) Labels shape: torch.Size([1, 16536]) Final batch size: 1, sequence length: 13638 Attention mask shape: torch.Size([1, 1, 13638, 13638]) Position ids shape: torch.Size([1, 13638]) Input IDs shape: torch.Size([1, 13638]) Labels shape: torch.Size([1, 13638]) Final batch size: 1, sequence length: 17415 Attention mask shape: torch.Size([1, 1, 17415, 17415]) Position ids shape: torch.Size([1, 17415]) Input IDs shape: torch.Size([1, 17415]) Labels shape: torch.Size([1, 17415]) Final batch size: 1, sequence length: 17220 Attention mask shape: torch.Size([1, 1, 17220, 17220]) Position ids shape: torch.Size([1, 17220]) Input IDs shape: torch.Size([1, 17220]) Labels shape: torch.Size([1, 17220]) Final batch size: 1, sequence length: 7221 Attention mask shape: torch.Size([1, 1, 7221, 7221]) Position ids shape: torch.Size([1, 7221]) Input IDs shape: torch.Size([1, 7221]) Labels shape: torch.Size([1, 7221]) Final batch size: 1, sequence length: 19414 Attention mask shape: torch.Size([1, 1, 19414, 19414]) Position ids shape: torch.Size([1, 19414]) Input IDs shape: torch.Size([1, 19414]) Labels shape: torch.Size([1, 19414]) Final batch size: 1, sequence length: 17733 Attention mask shape: torch.Size([1, 1, 17733, 17733]) Position ids shape: torch.Size([1, 17733]) Input IDs shape: torch.Size([1, 17733]) Labels shape: torch.Size([1, 17733]) Final batch size: 1, sequence length: 18741 Attention mask shape: torch.Size([1, 1, 18741, 18741]) Position ids shape: torch.Size([1, 18741]) Input IDs shape: torch.Size([1, 18741]) Labels shape: torch.Size([1, 18741]) Final batch size: 1, sequence length: 18653 Attention mask shape: torch.Size([1, 1, 18653, 18653]) Position ids shape: torch.Size([1, 18653]) Input IDs shape: torch.Size([1, 18653]) Labels shape: torch.Size([1, 18653]) Final batch size: 1, sequence length: 18496 Attention mask shape: torch.Size([1, 1, 18496, 18496]) Position ids shape: torch.Size([1, 18496]) Input IDs shape: torch.Size([1, 18496]) Labels shape: torch.Size([1, 18496]) Final batch size: 1, sequence length: 16750 Attention mask shape: torch.Size([1, 1, 16750, 16750]) Position ids shape: torch.Size([1, 16750]) Input IDs shape: torch.Size([1, 16750]) Labels shape: torch.Size([1, 16750]) Final batch size: 1, sequence length: 20933 Attention mask shape: torch.Size([1, 1, 20933, 20933]) Position ids shape: torch.Size([1, 20933]) Input IDs shape: torch.Size([1, 20933]) Labels shape: torch.Size([1, 20933]) Final batch size: 1, sequence length: 18393 Attention mask shape: torch.Size([1, 1, 18393, 18393]) Position ids shape: torch.Size([1, 18393]) Input IDs shape: torch.Size([1, 18393]) Labels shape: torch.Size([1, 18393]) Final batch size: 1, sequence length: 18927 Attention mask shape: torch.Size([1, 1, 18927, 18927]) Position ids shape: torch.Size([1, 18927]) Input IDs shape: torch.Size([1, 18927]) Labels shape: torch.Size([1, 18927]) Final batch size: 1, sequence length: 22391 Attention mask shape: torch.Size([1, 1, 22391, 22391]) Position ids shape: torch.Size([1, 22391]) Input IDs shape: torch.Size([1, 22391]) Labels shape: torch.Size([1, 22391]) Final batch size: 1, sequence length: 11067 Attention mask shape: torch.Size([1, 1, 11067, 11067]) Position ids shape: torch.Size([1, 11067]) Input IDs shape: torch.Size([1, 11067]) Labels shape: torch.Size([1, 11067]) Final batch size: 1, sequence length: 21420 Attention mask shape: torch.Size([1, 1, 21420, 21420]) Position ids shape: torch.Size([1, 21420]) Input IDs shape: torch.Size([1, 21420]) Labels shape: torch.Size([1, 21420]) Final batch size: 1, sequence length: 22004 Attention mask shape: torch.Size([1, 1, 22004, 22004]) Position ids shape: torch.Size([1, 22004]) Input IDs shape: torch.Size([1, 22004]) Labels shape: torch.Size([1, 22004]) Final batch size: 1, sequence length: 20612 Attention mask shape: torch.Size([1, 1, 20612, 20612]) Position ids shape: torch.Size([1, 20612]) Input IDs shape: torch.Size([1, 20612]) Labels shape: torch.Size([1, 20612]) Final batch size: 1, sequence length: 22887 Attention mask shape: torch.Size([1, 1, 22887, 22887]) Position ids shape: torch.Size([1, 22887]) Input IDs shape: torch.Size([1, 22887]) Labels shape: torch.Size([1, 22887]) Final batch size: 1, sequence length: 25747 Attention mask shape: torch.Size([1, 1, 25747, 25747]) Position ids shape: torch.Size([1, 25747]) Input IDs shape: torch.Size([1, 25747]) Labels shape: torch.Size([1, 25747]) Final batch size: 1, sequence length: 10719 Attention mask shape: torch.Size([1, 1, 10719, 10719]) Position ids shape: torch.Size([1, 10719]) Input IDs shape: torch.Size([1, 10719]) Labels shape: torch.Size([1, 10719]) Final batch size: 1, sequence length: 24988 Attention mask shape: torch.Size([1, 1, 24988, 24988]) Position ids shape: torch.Size([1, 24988]) Input IDs shape: torch.Size([1, 24988]) Labels shape: torch.Size([1, 24988]) Final batch size: 1, sequence length: 25477 Attention mask shape: torch.Size([1, 1, 25477, 25477]) Position ids shape: torch.Size([1, 25477]) Input IDs shape: torch.Size([1, 25477]) Labels shape: torch.Size([1, 25477]) Final batch size: 1, sequence length: 20579 Attention mask shape: torch.Size([1, 1, 20579, 20579]) Position ids shape: torch.Size([1, 20579]) Input IDs shape: torch.Size([1, 20579]) Labels shape: torch.Size([1, 20579]) Final batch size: 1, sequence length: 15317 Attention mask shape: torch.Size([1, 1, 15317, 15317]) Position ids shape: torch.Size([1, 15317]) Input IDs shape: torch.Size([1, 15317]) Labels shape: torch.Size([1, 15317]) Final batch size: 1, sequence length: 11184 Attention mask shape: torch.Size([1, 1, 11184, 11184]) Position ids shape: torch.Size([1, 11184]) Input IDs shape: torch.Size([1, 11184]) Labels shape: torch.Size([1, 11184]) Final batch size: 1, sequence length: 26663 Attention mask shape: torch.Size([1, 1, 26663, 26663]) Position ids shape: torch.Size([1, 26663]) Input IDs shape: torch.Size([1, 26663]) Labels shape: torch.Size([1, 26663]) Final batch size: 1, sequence length: 19552 Attention mask shape: torch.Size([1, 1, 19552, 19552]) Position ids shape: torch.Size([1, 19552]) Input IDs shape: torch.Size([1, 19552]) Labels shape: torch.Size([1, 19552]) Final batch size: 1, sequence length: 27447 Attention mask shape: torch.Size([1, 1, 27447, 27447]) Position ids shape: torch.Size([1, 27447]) Input IDs shape: torch.Size([1, 27447]) Labels shape: torch.Size([1, 27447]) Final batch size: 1, sequence length: 16915 Attention mask shape: torch.Size([1, 1, 16915, 16915]) Position ids shape: torch.Size([1, 16915]) Input IDs shape: torch.Size([1, 16915]) Labels shape: torch.Size([1, 16915]) Final batch size: 1, sequence length: 25651 Attention mask shape: torch.Size([1, 1, 25651, 25651]) Position ids shape: torch.Size([1, 25651]) Input IDs shape: torch.Size([1, 25651]) Labels shape: torch.Size([1, 25651]) Final batch size: 1, sequence length: 30031 Attention mask shape: torch.Size([1, 1, 30031, 30031]) Position ids shape: torch.Size([1, 30031]) Input IDs shape: torch.Size([1, 30031]) Labels shape: torch.Size([1, 30031]) Final batch size: 1, sequence length: 27334 Attention mask shape: torch.Size([1, 1, 27334, 27334]) Position ids shape: torch.Size([1, 27334]) Input IDs shape: torch.Size([1, 27334]) Labels shape: torch.Size([1, 27334]) Final batch size: 1, sequence length: 17395 Attention mask shape: torch.Size([1, 1, 17395, 17395]) Position ids shape: torch.Size([1, 17395]) Input IDs shape: torch.Size([1, 17395]) Labels shape: torch.Size([1, 17395]) Final batch size: 1, sequence length: 28777 Attention mask shape: torch.Size([1, 1, 28777, 28777]) Position ids shape: torch.Size([1, 28777]) Input IDs shape: torch.Size([1, 28777]) Labels shape: torch.Size([1, 28777]) Final batch size: 1, sequence length: 19869 Attention mask shape: torch.Size([1, 1, 19869, 19869]) Position ids shape: torch.Size([1, 19869]) Input IDs shape: torch.Size([1, 19869]) Labels shape: torch.Size([1, 19869]) Final batch size: 1, sequence length: 26033 Attention mask shape: torch.Size([1, 1, 26033, 26033]) Position ids shape: torch.Size([1, 26033]) Input IDs shape: torch.Size([1, 26033]) Labels shape: torch.Size([1, 26033]) Final batch size: 1, sequence length: 30981 Attention mask shape: torch.Size([1, 1, 30981, 30981]) Position ids shape: torch.Size([1, 30981]) Input IDs shape: torch.Size([1, 30981]) Labels shape: torch.Size([1, 30981]) Final batch size: 1, sequence length: 17911 Attention mask shape: torch.Size([1, 1, 17911, 17911]) Position ids shape: torch.Size([1, 17911]) Input IDs shape: torch.Size([1, 17911]) Labels shape: torch.Size([1, 17911]) Final batch size: 1, sequence length: 16953 Attention mask shape: torch.Size([1, 1, 16953, 16953]) Position ids shape: torch.Size([1, 16953]) Input IDs shape: torch.Size([1, 16953]) Labels shape: torch.Size([1, 16953]) Final batch size: 1, sequence length: 32466 Attention mask shape: torch.Size([1, 1, 32466, 32466]) Position ids shape: torch.Size([1, 32466]) Input IDs shape: torch.Size([1, 32466]) Labels shape: torch.Size([1, 32466]) Final batch size: 1, sequence length: 30601 Attention mask shape: torch.Size([1, 1, 30601, 30601]) Position ids shape: torch.Size([1, 30601]) Input IDs shape: torch.Size([1, 30601]) Labels shape: torch.Size([1, 30601]) Final batch size: 1, sequence length: 28060 Attention mask shape: torch.Size([1, 1, 28060, 28060]) Position ids shape: torch.Size([1, 28060]) Input IDs shape: torch.Size([1, 28060]) Labels shape: torch.Size([1, 28060]) Final batch size: 1, sequence length: 18376 Attention mask shape: torch.Size([1, 1, 18376, 18376]) Position ids shape: torch.Size([1, 18376]) Input IDs shape: torch.Size([1, 18376]) Labels shape: torch.Size([1, 18376]) Final batch size: 1, sequence length: 27480 Attention mask shape: torch.Size([1, 1, 27480, 27480]) Position ids shape: torch.Size([1, 27480]) Input IDs shape: torch.Size([1, 27480]) Labels shape: torch.Size([1, 27480]) Final batch size: 1, sequence length: 24782 Attention mask shape: torch.Size([1, 1, 24782, 24782]) Position ids shape: torch.Size([1, 24782]) Input IDs shape: torch.Size([1, 24782]) Labels shape: torch.Size([1, 24782]) Final batch size: 1, sequence length: 21988 Attention mask shape: torch.Size([1, 1, 21988, 21988]) Position ids shape: torch.Size([1, 21988]) Input IDs shape: torch.Size([1, 21988]) Labels shape: torch.Size([1, 21988]) Final batch size: 1, sequence length: 15924 Attention mask shape: torch.Size([1, 1, 15924, 15924]) Position ids shape: torch.Size([1, 15924]) Input IDs shape: torch.Size([1, 15924]) Labels shape: torch.Size([1, 15924]) Final batch size: 1, sequence length: 12245 Attention mask shape: torch.Size([1, 1, 12245, 12245]) Position ids shape: torch.Size([1, 12245]) Input IDs shape: torch.Size([1, 12245]) Labels shape: torch.Size([1, 12245]) Final batch size: 1, sequence length: 27385 Attention mask shape: torch.Size([1, 1, 27385, 27385]) Position ids shape: torch.Size([1, 27385]) Input IDs shape: torch.Size([1, 27385]) Labels shape: torch.Size([1, 27385]) Final batch size: 1, sequence length: 14873 Attention mask shape: torch.Size([1, 1, 14873, 14873]) Position ids shape: torch.Size([1, 14873]) Input IDs shape: torch.Size([1, 14873]) Labels shape: torch.Size([1, 14873]) Final batch size: 1, sequence length: 33725 Attention mask shape: torch.Size([1, 1, 33725, 33725]) Position ids shape: torch.Size([1, 33725]) Input IDs shape: torch.Size([1, 33725]) Labels shape: torch.Size([1, 33725]) Final batch size: 1, sequence length: 37025 Attention mask shape: torch.Size([1, 1, 37025, 37025]) Position ids shape: torch.Size([1, 37025]) Input IDs shape: torch.Size([1, 37025]) Labels shape: torch.Size([1, 37025]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 36175 Attention mask shape: torch.Size([1, 1, 36175, 36175]) Position ids shape: torch.Size([1, 36175]) Input IDs shape: torch.Size([1, 36175]) Labels shape: torch.Size([1, 36175]) Final batch size: 1, sequence length: 32835 Attention mask shape: torch.Size([1, 1, 32835, 32835]) Position ids shape: torch.Size([1, 32835]) Input IDs shape: torch.Size([1, 32835]) Labels shape: torch.Size([1, 32835]) Final batch size: 1, sequence length: 37738 Attention mask shape: torch.Size([1, 1, 37738, 37738]) Position ids shape: torch.Size([1, 37738]) Input IDs shape: torch.Size([1, 37738]) Labels shape: torch.Size([1, 37738]) Final batch size: 1, sequence length: 37158 Attention mask shape: torch.Size([1, 1, 37158, 37158]) Position ids shape: torch.Size([1, 37158]) Input IDs shape: torch.Size([1, 37158]) Labels shape: torch.Size([1, 37158]) Final batch size: 1, sequence length: 37511 Attention mask shape: torch.Size([1, 1, 37511, 37511]) Position ids shape: torch.Size([1, 37511]) Input IDs shape: torch.Size([1, 37511]) Labels shape: torch.Size([1, 37511]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38986 Attention mask shape: torch.Size([1, 1, 38986, 38986]) Position ids shape: torch.Size([1, 38986]) Input IDs shape: torch.Size([1, 38986]) Labels shape: torch.Size([1, 38986]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 29586 Attention mask shape: torch.Size([1, 1, 29586, 29586]) Position ids shape: torch.Size([1, 29586]) Input IDs shape: torch.Size([1, 29586]) Labels shape: torch.Size([1, 29586]) Final batch size: 1, sequence length: 40397 Attention mask shape: torch.Size([1, 1, 40397, 40397]) Position ids shape: torch.Size([1, 40397]) Input IDs shape: torch.Size([1, 40397]) Labels shape: torch.Size([1, 40397]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40269 Attention mask shape: torch.Size([1, 1, 40269, 40269]) Position ids shape: torch.Size([1, 40269]) Input IDs shape: torch.Size([1, 40269]) Labels shape: torch.Size([1, 40269]) Final batch size: 1, sequence length: 36130 Attention mask shape: torch.Size([1, 1, 36130, 36130]) Position ids shape: torch.Size([1, 36130]) Input IDs shape: torch.Size([1, 36130]) Labels shape: torch.Size([1, 36130]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31879 Attention mask shape: torch.Size([1, 1, 31879, 31879]) Position ids shape: torch.Size([1, 31879]) Input IDs shape: torch.Size([1, 31879]) Labels shape: torch.Size([1, 31879]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40237 Attention mask shape: torch.Size([1, 1, 40237, 40237]) Position ids shape: torch.Size([1, 40237]) Input IDs shape: torch.Size([1, 40237]) Labels shape: torch.Size([1, 40237]) Final batch size: 1, sequence length: 26479 Attention mask shape: torch.Size([1, 1, 26479, 26479]) Position ids shape: torch.Size([1, 26479]) Input IDs shape: torch.Size([1, 26479]) Labels shape: torch.Size([1, 26479]) Final batch size: 1, sequence length: 31348 Attention mask shape: torch.Size([1, 1, 31348, 31348]) Position ids shape: torch.Size([1, 31348]) Input IDs shape: torch.Size([1, 31348]) Labels shape: torch.Size([1, 31348]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 27291 Attention mask shape: torch.Size([1, 1, 27291, 27291]) Position ids shape: torch.Size([1, 27291]) Input IDs shape: torch.Size([1, 27291]) Labels shape: torch.Size([1, 27291]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32318 Attention mask shape: torch.Size([1, 1, 32318, 32318]) Position ids shape: torch.Size([1, 32318]) Input IDs shape: torch.Size([1, 32318]) Labels shape: torch.Size([1, 32318]) Final batch size: 1, sequence length: 20509 Attention mask shape: torch.Size([1, 1, 20509, 20509]) Position ids shape: torch.Size([1, 20509]) Input IDs shape: torch.Size([1, 20509]) Labels shape: torch.Size([1, 20509]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17711 Attention mask shape: torch.Size([1, 1, 17711, 17711]) Position ids shape: torch.Size([1, 17711]) Input IDs shape: torch.Size([1, 17711]) Labels shape: torch.Size([1, 17711]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32247 Attention mask shape: torch.Size([1, 1, 32247, 32247]) Position ids shape: torch.Size([1, 32247]) Input IDs shape: torch.Size([1, 32247]) Labels shape: torch.Size([1, 32247]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28631 Attention mask shape: torch.Size([1, 1, 28631, 28631]) Position ids shape: torch.Size([1, 28631]) Input IDs shape: torch.Size([1, 28631]) Labels shape: torch.Size([1, 28631]) Final batch size: 1, sequence length: 24939 Attention mask shape: torch.Size([1, 1, 24939, 24939]) Position ids shape: torch.Size([1, 24939]) Input IDs shape: torch.Size([1, 24939]) Labels shape: torch.Size([1, 24939]) Final batch size: 1, sequence length: 36947 Attention mask shape: torch.Size([1, 1, 36947, 36947]) Position ids shape: torch.Size([1, 36947]) Input IDs shape: torch.Size([1, 36947]) Labels shape: torch.Size([1, 36947]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21225 Attention mask shape: torch.Size([1, 1, 21225, 21225]) Position ids shape: torch.Size([1, 21225]) Input IDs shape: torch.Size([1, 21225]) Labels shape: torch.Size([1, 21225]) Final batch size: 1, sequence length: 23740 Attention mask shape: torch.Size([1, 1, 23740, 23740]) Position ids shape: torch.Size([1, 23740]) Input IDs shape: torch.Size([1, 23740]) Labels shape: torch.Size([1, 23740]) Final batch size: 1, sequence length: 22896 Attention mask shape: torch.Size([1, 1, 22896, 22896]) Position ids shape: torch.Size([1, 22896]) Input IDs shape: torch.Size([1, 22896]) Labels shape: torch.Size([1, 22896]) Final batch size: 1, sequence length: 10198 Attention mask shape: torch.Size([1, 1, 10198, 10198]) Position ids shape: torch.Size([1, 10198]) Input IDs shape: torch.Size([1, 10198]) Labels shape: torch.Size([1, 10198]) Final batch size: 1, sequence length: 33125 Attention mask shape: torch.Size([1, 1, 33125, 33125]) Position ids shape: torch.Size([1, 33125]) Input IDs shape: torch.Size([1, 33125]) Labels shape: torch.Size([1, 33125]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22205 Attention mask shape: torch.Size([1, 1, 22205, 22205]) Position ids shape: torch.Size([1, 22205]) Input IDs shape: torch.Size([1, 22205]) Labels shape: torch.Size([1, 22205]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28046 Attention mask shape: torch.Size([1, 1, 28046, 28046]) Position ids shape: torch.Size([1, 28046]) Input IDs shape: torch.Size([1, 28046]) Labels shape: torch.Size([1, 28046]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28149 Attention mask shape: torch.Size([1, 1, 28149, 28149]) Position ids shape: torch.Size([1, 28149]) Input IDs shape: torch.Size([1, 28149]) Labels shape: torch.Size([1, 28149]) Final batch size: 1, sequence length: 38360 Attention mask shape: torch.Size([1, 1, 38360, 38360]) Position ids shape: torch.Size([1, 38360]) Input IDs shape: torch.Size([1, 38360]) Labels shape: torch.Size([1, 38360]) Final batch size: 1, sequence length: 32767 Attention mask shape: torch.Size([1, 1, 32767, 32767]) Position ids shape: torch.Size([1, 32767]) Input IDs shape: torch.Size([1, 32767]) Labels shape: torch.Size([1, 32767]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36567 Attention mask shape: torch.Size([1, 1, 36567, 36567]) Position ids shape: torch.Size([1, 36567]) Input IDs shape: torch.Size([1, 36567]) Labels shape: torch.Size([1, 36567]) Final batch size: 1, sequence length: 34268 Attention mask shape: torch.Size([1, 1, 34268, 34268]) Position ids shape: torch.Size([1, 34268]) Input IDs shape: torch.Size([1, 34268]) Labels shape: torch.Size([1, 34268]) Final batch size: 1, sequence length: 32313 Attention mask shape: torch.Size([1, 1, 32313, 32313]) Position ids shape: torch.Size([1, 32313]) Input IDs shape: torch.Size([1, 32313]) Labels shape: torch.Size([1, 32313]) {'loss': 0.3268, 'grad_norm': 0.4106039684301258, 'learning_rate': 8.885729807284855e-06, 'num_tokens': -inf, 'epoch': 2.25} Final batch size: 1, sequence length: 5377 Attention mask shape: torch.Size([1, 1, 5377, 5377]) Position ids shape: torch.Size([1, 5377]) Input IDs shape: torch.Size([1, 5377]) Labels shape: torch.Size([1, 5377]) Final batch size: 1, sequence length: 7998 Attention mask shape: torch.Size([1, 1, 7998, 7998]) Position ids shape: torch.Size([1, 7998]) Input IDs shape: torch.Size([1, 7998]) Labels shape: torch.Size([1, 7998]) Final batch size: 1, sequence length: 7402 Attention mask shape: torch.Size([1, 1, 7402, 7402]) Position ids shape: torch.Size([1, 7402]) Input IDs shape: torch.Size([1, 7402]) Labels shape: torch.Size([1, 7402]) Final batch size: 1, sequence length: 10576 Attention mask shape: torch.Size([1, 1, 10576, 10576]) Position ids shape: torch.Size([1, 10576]) Input IDs shape: torch.Size([1, 10576]) Labels shape: torch.Size([1, 10576]) Final batch size: 1, sequence length: 8655 Attention mask shape: torch.Size([1, 1, 8655, 8655]) Position ids shape: torch.Size([1, 8655]) Input IDs shape: torch.Size([1, 8655]) Labels shape: torch.Size([1, 8655]) Final batch size: 1, sequence length: 11709 Attention mask shape: torch.Size([1, 1, 11709, 11709]) Position ids shape: torch.Size([1, 11709]) Input IDs shape: torch.Size([1, 11709]) Labels shape: torch.Size([1, 11709]) Final batch size: 1, sequence length: 11365 Attention mask shape: torch.Size([1, 1, 11365, 11365]) Position ids shape: torch.Size([1, 11365]) Input IDs shape: torch.Size([1, 11365]) Labels shape: torch.Size([1, 11365]) Final batch size: 1, sequence length: 7344 Attention mask shape: torch.Size([1, 1, 7344, 7344]) Position ids shape: torch.Size([1, 7344]) Input IDs shape: torch.Size([1, 7344]) Labels shape: torch.Size([1, 7344]) Final batch size: 1, sequence length: 8436 Attention mask shape: torch.Size([1, 1, 8436, 8436]) Position ids shape: torch.Size([1, 8436]) Input IDs shape: torch.Size([1, 8436]) Labels shape: torch.Size([1, 8436]) Final batch size: 1, sequence length: 14127 Attention mask shape: torch.Size([1, 1, 14127, 14127]) Position ids shape: torch.Size([1, 14127]) Input IDs shape: torch.Size([1, 14127]) Labels shape: torch.Size([1, 14127]) Final batch size: 1, sequence length: 11678 Attention mask shape: torch.Size([1, 1, 11678, 11678]) Position ids shape: torch.Size([1, 11678]) Input IDs shape: torch.Size([1, 11678]) Labels shape: torch.Size([1, 11678]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 11623 Attention mask shape: torch.Size([1, 1, 11623, 11623]) Position ids shape: torch.Size([1, 11623]) Input IDs shape: torch.Size([1, 11623]) Labels shape: torch.Size([1, 11623]) Final batch size: 1, sequence length: 15079 Attention mask shape: torch.Size([1, 1, 15079, 15079]) Position ids shape: torch.Size([1, 15079]) Input IDs shape: torch.Size([1, 15079]) Labels shape: torch.Size([1, 15079]) Final batch size: 1, sequence length: 15308 Attention mask shape: torch.Size([1, 1, 15308, 15308]) Position ids shape: torch.Size([1, 15308]) Input IDs shape: torch.Size([1, 15308]) Labels shape: torch.Size([1, 15308]) Final batch size: 1, sequence length: 15478 Attention mask shape: torch.Size([1, 1, 15478, 15478]) Position ids shape: torch.Size([1, 15478]) Input IDs shape: torch.Size([1, 15478]) Labels shape: torch.Size([1, 15478]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 16535 Attention mask shape: torch.Size([1, 1, 16535, 16535]) Position ids shape: torch.Size([1, 16535]) Input IDs shape: torch.Size([1, 16535]) Labels shape: torch.Size([1, 16535]) Final batch size: 1, sequence length: 14648 Attention mask shape: torch.Size([1, 1, 14648, 14648]) Position ids shape: torch.Size([1, 14648]) Input IDs shape: torch.Size([1, 14648]) Labels shape: torch.Size([1, 14648]) Final batch size: 1, sequence length: 13639 Attention mask shape: torch.Size([1, 1, 13639, 13639]) Position ids shape: torch.Size([1, 13639]) Input IDs shape: torch.Size([1, 13639]) Labels shape: torch.Size([1, 13639]) Final batch size: 1, sequence length: 18836 Attention mask shape: torch.Size([1, 1, 18836, 18836]) Position ids shape: torch.Size([1, 18836]) Input IDs shape: torch.Size([1, 18836]) Labels shape: torch.Size([1, 18836]) Final batch size: 1, sequence length: 19338 Attention mask shape: torch.Size([1, 1, 19338, 19338]) Position ids shape: torch.Size([1, 19338]) Input IDs shape: torch.Size([1, 19338]) Labels shape: torch.Size([1, 19338]) Final batch size: 1, sequence length: 19138 Attention mask shape: torch.Size([1, 1, 19138, 19138]) Position ids shape: torch.Size([1, 19138]) Input IDs shape: torch.Size([1, 19138]) Labels shape: torch.Size([1, 19138]) Final batch size: 1, sequence length: 18307 Attention mask shape: torch.Size([1, 1, 18307, 18307]) Position ids shape: torch.Size([1, 18307]) Input IDs shape: torch.Size([1, 18307]) Labels shape: torch.Size([1, 18307]) Final batch size: 1, sequence length: 12006 Attention mask shape: torch.Size([1, 1, 12006, 12006]) Position ids shape: torch.Size([1, 12006]) Input IDs shape: torch.Size([1, 12006]) Labels shape: torch.Size([1, 12006]) Final batch size: 1, sequence length: 18953 Attention mask shape: torch.Size([1, 1, 18953, 18953]) Position ids shape: torch.Size([1, 18953]) Input IDs shape: torch.Size([1, 18953]) Labels shape: torch.Size([1, 18953]) Final batch size: 1, sequence length: 19330 Attention mask shape: torch.Size([1, 1, 19330, 19330]) Position ids shape: torch.Size([1, 19330]) Input IDs shape: torch.Size([1, 19330]) Labels shape: torch.Size([1, 19330]) Final batch size: 1, sequence length: 8839 Attention mask shape: torch.Size([1, 1, 8839, 8839]) Position ids shape: torch.Size([1, 8839]) Input IDs shape: torch.Size([1, 8839]) Labels shape: torch.Size([1, 8839]) Final batch size: 1, sequence length: 20714 Attention mask shape: torch.Size([1, 1, 20714, 20714]) Position ids shape: torch.Size([1, 20714]) Input IDs shape: torch.Size([1, 20714]) Labels shape: torch.Size([1, 20714]) Final batch size: 1, sequence length: 19170 Attention mask shape: torch.Size([1, 1, 19170, 19170]) Position ids shape: torch.Size([1, 19170]) Input IDs shape: torch.Size([1, 19170]) Labels shape: torch.Size([1, 19170]) Final batch size: 1, sequence length: 18325 Attention mask shape: torch.Size([1, 1, 18325, 18325]) Position ids shape: torch.Size([1, 18325]) Input IDs shape: torch.Size([1, 18325]) Labels shape: torch.Size([1, 18325]) Final batch size: 1, sequence length: 18395 Attention mask shape: torch.Size([1, 1, 18395, 18395]) Position ids shape: torch.Size([1, 18395]) Input IDs shape: torch.Size([1, 18395]) Labels shape: torch.Size([1, 18395]) Final batch size: 1, sequence length: 20770 Attention mask shape: torch.Size([1, 1, 20770, 20770]) Position ids shape: torch.Size([1, 20770]) Input IDs shape: torch.Size([1, 20770]) Labels shape: torch.Size([1, 20770]) Final batch size: 1, sequence length: 18527 Attention mask shape: torch.Size([1, 1, 18527, 18527]) Position ids shape: torch.Size([1, 18527]) Input IDs shape: torch.Size([1, 18527]) Labels shape: torch.Size([1, 18527]) Final batch size: 1, sequence length: 17058 Attention mask shape: torch.Size([1, 1, 17058, 17058]) Position ids shape: torch.Size([1, 17058]) Input IDs shape: torch.Size([1, 17058]) Labels shape: torch.Size([1, 17058]) Final batch size: 1, sequence length: 21982 Attention mask shape: torch.Size([1, 1, 21982, 21982]) Position ids shape: torch.Size([1, 21982]) Input IDs shape: torch.Size([1, 21982]) Labels shape: torch.Size([1, 21982]) Final batch size: 1, sequence length: 22561 Attention mask shape: torch.Size([1, 1, 22561, 22561]) Position ids shape: torch.Size([1, 22561]) Input IDs shape: torch.Size([1, 22561]) Labels shape: torch.Size([1, 22561]) Final batch size: 1, sequence length: 20854 Attention mask shape: torch.Size([1, 1, 20854, 20854]) Position ids shape: torch.Size([1, 20854]) Input IDs shape: torch.Size([1, 20854]) Labels shape: torch.Size([1, 20854]) Final batch size: 1, sequence length: 18823 Attention mask shape: torch.Size([1, 1, 18823, 18823]) Position ids shape: torch.Size([1, 18823]) Input IDs shape: torch.Size([1, 18823]) Labels shape: torch.Size([1, 18823]) Final batch size: 1, sequence length: 21405 Attention mask shape: torch.Size([1, 1, 21405, 21405]) Position ids shape: torch.Size([1, 21405]) Input IDs shape: torch.Size([1, 21405]) Labels shape: torch.Size([1, 21405]) Final batch size: 1, sequence length: 21858 Attention mask shape: torch.Size([1, 1, 21858, 21858]) Position ids shape: torch.Size([1, 21858]) Input IDs shape: torch.Size([1, 21858]) Labels shape: torch.Size([1, 21858]) Final batch size: 1, sequence length: 25451 Attention mask shape: torch.Size([1, 1, 25451, 25451]) Position ids shape: torch.Size([1, 25451]) Input IDs shape: torch.Size([1, 25451]) Labels shape: torch.Size([1, 25451]) Final batch size: 1, sequence length: 23560 Final batch size: 1, sequence length: 26356 Attention mask shape: torch.Size([1, 1, 23560, 23560]) Attention mask shape: torch.Size([1, 1, 26356, 26356]) Position ids shape: torch.Size([1, 26356]) Input IDs shape: torch.Size([1, 26356]) Labels shape: torch.Size([1, 26356]) Position ids shape: torch.Size([1, 23560]) Input IDs shape: torch.Size([1, 23560]) Labels shape: torch.Size([1, 23560]) Final batch size: 1, sequence length: 15913 Attention mask shape: torch.Size([1, 1, 15913, 15913]) Position ids shape: torch.Size([1, 15913]) Input IDs shape: torch.Size([1, 15913]) Labels shape: torch.Size([1, 15913]) Final batch size: 1, sequence length: 24248 Attention mask shape: torch.Size([1, 1, 24248, 24248]) Position ids shape: torch.Size([1, 24248]) Input IDs shape: torch.Size([1, 24248]) Labels shape: torch.Size([1, 24248]) Final batch size: 1, sequence length: 23694 Attention mask shape: torch.Size([1, 1, 23694, 23694]) Position ids shape: torch.Size([1, 23694]) Input IDs shape: torch.Size([1, 23694]) Labels shape: torch.Size([1, 23694]) Final batch size: 1, sequence length: 24433 Attention mask shape: torch.Size([1, 1, 24433, 24433]) Position ids shape: torch.Size([1, 24433]) Input IDs shape: torch.Size([1, 24433]) Labels shape: torch.Size([1, 24433]) Final batch size: 1, sequence length: 29098 Attention mask shape: torch.Size([1, 1, 29098, 29098]) Position ids shape: torch.Size([1, 29098]) Input IDs shape: torch.Size([1, 29098]) Labels shape: torch.Size([1, 29098]) Final batch size: 1, sequence length: 22735 Attention mask shape: torch.Size([1, 1, 22735, 22735]) Position ids shape: torch.Size([1, 22735]) Input IDs shape: torch.Size([1, 22735]) Labels shape: torch.Size([1, 22735]) Final batch size: 1, sequence length: 9380 Attention mask shape: torch.Size([1, 1, 9380, 9380]) Position ids shape: torch.Size([1, 9380]) Input IDs shape: torch.Size([1, 9380]) Labels shape: torch.Size([1, 9380]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 5801 Attention mask shape: torch.Size([1, 1, 5801, 5801]) Position ids shape: torch.Size([1, 5801]) Input IDs shape: torch.Size([1, 5801]) Labels shape: torch.Size([1, 5801]) Final batch size: 1, sequence length: 28684 Attention mask shape: torch.Size([1, 1, 28684, 28684]) Position ids shape: torch.Size([1, 28684]) Input IDs shape: torch.Size([1, 28684]) Labels shape: torch.Size([1, 28684]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 19045 Attention mask shape: torch.Size([1, 1, 19045, 19045]) Position ids shape: torch.Size([1, 19045]) Input IDs shape: torch.Size([1, 19045]) Labels shape: torch.Size([1, 19045]) Final batch size: 1, sequence length: 16060 Attention mask shape: torch.Size([1, 1, 16060, 16060]) Position ids shape: torch.Size([1, 16060]) Input IDs shape: torch.Size([1, 16060]) Labels shape: torch.Size([1, 16060]) Final batch size: 1, sequence length: 26072 Attention mask shape: torch.Size([1, 1, 26072, 26072]) Position ids shape: torch.Size([1, 26072]) Input IDs shape: torch.Size([1, 26072]) Labels shape: torch.Size([1, 26072]) Final batch size: 1, sequence length: 19239 Attention mask shape: torch.Size([1, 1, 19239, 19239]) Position ids shape: torch.Size([1, 19239]) Input IDs shape: torch.Size([1, 19239]) Labels shape: torch.Size([1, 19239]) Final batch size: 1, sequence length: 32328 Attention mask shape: torch.Size([1, 1, 32328, 32328]) Position ids shape: torch.Size([1, 32328]) Input IDs shape: torch.Size([1, 32328]) Labels shape: torch.Size([1, 32328]) Final batch size: 1, sequence length: 21615 Attention mask shape: torch.Size([1, 1, 21615, 21615]) Position ids shape: torch.Size([1, 21615]) Input IDs shape: torch.Size([1, 21615]) Labels shape: torch.Size([1, 21615]) Final batch size: 1, sequence length: 17465 Attention mask shape: torch.Size([1, 1, 17465, 17465]) Position ids shape: torch.Size([1, 17465]) Input IDs shape: torch.Size([1, 17465]) Labels shape: torch.Size([1, 17465]) Final batch size: 1, sequence length: 15875 Attention mask shape: torch.Size([1, 1, 15875, 15875]) Position ids shape: torch.Size([1, 15875]) Input IDs shape: torch.Size([1, 15875]) Labels shape: torch.Size([1, 15875]) Final batch size: 1, sequence length: 32885 Attention mask shape: torch.Size([1, 1, 32885, 32885]) Position ids shape: torch.Size([1, 32885]) Input IDs shape: torch.Size([1, 32885]) Labels shape: torch.Size([1, 32885]) Final batch size: 1, sequence length: 16050 Attention mask shape: torch.Size([1, 1, 16050, 16050]) Position ids shape: torch.Size([1, 16050]) Input IDs shape: torch.Size([1, 16050]) Labels shape: torch.Size([1, 16050]) Final batch size: 1, sequence length: 31414 Attention mask shape: torch.Size([1, 1, 31414, 31414]) Position ids shape: torch.Size([1, 31414]) Input IDs shape: torch.Size([1, 31414]) Labels shape: torch.Size([1, 31414]) Final batch size: 1, sequence length: 26689 Attention mask shape: torch.Size([1, 1, 26689, 26689]) Position ids shape: torch.Size([1, 26689]) Input IDs shape: torch.Size([1, 26689]) Labels shape: torch.Size([1, 26689]) Final batch size: 1, sequence length: 23881 Attention mask shape: torch.Size([1, 1, 23881, 23881]) Position ids shape: torch.Size([1, 23881]) Input IDs shape: torch.Size([1, 23881]) Labels shape: torch.Size([1, 23881]) Final batch size: 1, sequence length: 30687 Attention mask shape: torch.Size([1, 1, 30687, 30687]) Position ids shape: torch.Size([1, 30687]) Input IDs shape: torch.Size([1, 30687]) Labels shape: torch.Size([1, 30687]) Final batch size: 1, sequence length: 35397 Attention mask shape: torch.Size([1, 1, 35397, 35397]) Position ids shape: torch.Size([1, 35397]) Input IDs shape: torch.Size([1, 35397]) Labels shape: torch.Size([1, 35397]) Final batch size: 1, sequence length: 29355 Attention mask shape: torch.Size([1, 1, 29355, 29355]) Position ids shape: torch.Size([1, 29355]) Input IDs shape: torch.Size([1, 29355]) Labels shape: torch.Size([1, 29355]) Final batch size: 1, sequence length: 33871 Attention mask shape: torch.Size([1, 1, 33871, 33871]) Position ids shape: torch.Size([1, 33871]) Input IDs shape: torch.Size([1, 33871]) Labels shape: torch.Size([1, 33871]) Final batch size: 1, sequence length: 30689 Attention mask shape: torch.Size([1, 1, 30689, 30689]) Position ids shape: torch.Size([1, 30689]) Input IDs shape: torch.Size([1, 30689]) Labels shape: torch.Size([1, 30689]) Final batch size: 1, sequence length: 27972 Attention mask shape: torch.Size([1, 1, 27972, 27972]) Position ids shape: torch.Size([1, 27972]) Input IDs shape: torch.Size([1, 27972]) Labels shape: torch.Size([1, 27972]) Final batch size: 1, sequence length: 30346 Attention mask shape: torch.Size([1, 1, 30346, 30346]) Position ids shape: torch.Size([1, 30346]) Input IDs shape: torch.Size([1, 30346]) Labels shape: torch.Size([1, 30346]) Final batch size: 1, sequence length: 32529 Attention mask shape: torch.Size([1, 1, 32529, 32529]) Position ids shape: torch.Size([1, 32529]) Input IDs shape: torch.Size([1, 32529]) Labels shape: torch.Size([1, 32529]) Final batch size: 1, sequence length: 32636 Attention mask shape: torch.Size([1, 1, 32636, 32636]) Position ids shape: torch.Size([1, 32636]) Input IDs shape: torch.Size([1, 32636]) Labels shape: torch.Size([1, 32636]) Final batch size: 1, sequence length: 29150 Attention mask shape: torch.Size([1, 1, 29150, 29150]) Position ids shape: torch.Size([1, 29150]) Input IDs shape: torch.Size([1, 29150]) Labels shape: torch.Size([1, 29150]) Final batch size: 1, sequence length: 31860 Attention mask shape: torch.Size([1, 1, 31860, 31860]) Position ids shape: torch.Size([1, 31860]) Input IDs shape: torch.Size([1, 31860]) Labels shape: torch.Size([1, 31860]) Final batch size: 1, sequence length: 34921 Attention mask shape: torch.Size([1, 1, 34921, 34921]) Position ids shape: torch.Size([1, 34921]) Input IDs shape: torch.Size([1, 34921]) Labels shape: torch.Size([1, 34921]) Final batch size: 1, sequence length: 34581 Attention mask shape: torch.Size([1, 1, 34581, 34581]) Position ids shape: torch.Size([1, 34581]) Input IDs shape: torch.Size([1, 34581]) Labels shape: torch.Size([1, 34581]) Final batch size: 1, sequence length: 22264 Attention mask shape: torch.Size([1, 1, 22264, 22264]) Position ids shape: torch.Size([1, 22264]) Input IDs shape: torch.Size([1, 22264]) Labels shape: torch.Size([1, 22264]) Final batch size: 1, sequence length: 31712 Attention mask shape: torch.Size([1, 1, 31712, 31712]) Position ids shape: torch.Size([1, 31712]) Input IDs shape: torch.Size([1, 31712]) Labels shape: torch.Size([1, 31712]) Final batch size: 1, sequence length: 15077 Attention mask shape: torch.Size([1, 1, 15077, 15077]) Position ids shape: torch.Size([1, 15077]) Input IDs shape: torch.Size([1, 15077]) Labels shape: torch.Size([1, 15077]) Final batch size: 1, sequence length: 24781 Attention mask shape: torch.Size([1, 1, 24781, 24781]) Position ids shape: torch.Size([1, 24781]) Input IDs shape: torch.Size([1, 24781]) Labels shape: torch.Size([1, 24781]) Final batch size: 1, sequence length: 18963 Attention mask shape: torch.Size([1, 1, 18963, 18963]) Position ids shape: torch.Size([1, 18963]) Input IDs shape: torch.Size([1, 18963]) Labels shape: torch.Size([1, 18963]) Final batch size: 1, sequence length: 18177 Attention mask shape: torch.Size([1, 1, 18177, 18177]) Position ids shape: torch.Size([1, 18177]) Input IDs shape: torch.Size([1, 18177]) Labels shape: torch.Size([1, 18177]) Final batch size: 1, sequence length: 16025 Attention mask shape: torch.Size([1, 1, 16025, 16025]) Position ids shape: torch.Size([1, 16025]) Input IDs shape: torch.Size([1, 16025]) Labels shape: torch.Size([1, 16025]) Final batch size: 1, sequence length: 14976 Attention mask shape: torch.Size([1, 1, 14976, 14976]) Position ids shape: torch.Size([1, 14976]) Input IDs shape: torch.Size([1, 14976]) Labels shape: torch.Size([1, 14976]) Final batch size: 1, sequence length: 25741 Attention mask shape: torch.Size([1, 1, 25741, 25741]) Position ids shape: torch.Size([1, 25741]) Input IDs shape: torch.Size([1, 25741]) Labels shape: torch.Size([1, 25741]) Final batch size: 1, sequence length: 17243 Attention mask shape: torch.Size([1, 1, 17243, 17243]) Position ids shape: torch.Size([1, 17243]) Input IDs shape: torch.Size([1, 17243]) Labels shape: torch.Size([1, 17243]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31022 Attention mask shape: torch.Size([1, 1, 31022, 31022]) Position ids shape: torch.Size([1, 31022]) Input IDs shape: torch.Size([1, 31022]) Labels shape: torch.Size([1, 31022]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24002 Attention mask shape: torch.Size([1, 1, 24002, 24002]) Position ids shape: torch.Size([1, 24002]) Input IDs shape: torch.Size([1, 24002]) Labels shape: torch.Size([1, 24002]) Final batch size: 1, sequence length: 28018 Attention mask shape: torch.Size([1, 1, 28018, 28018]) Position ids shape: torch.Size([1, 28018]) Input IDs shape: torch.Size([1, 28018]) Labels shape: torch.Size([1, 28018]) Final batch size: 1, sequence length: 26922 Attention mask shape: torch.Size([1, 1, 26922, 26922]) Position ids shape: torch.Size([1, 26922]) Input IDs shape: torch.Size([1, 26922]) Labels shape: torch.Size([1, 26922]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40488 Attention mask shape: torch.Size([1, 1, 40488, 40488]) Position ids shape: torch.Size([1, 40488]) Input IDs shape: torch.Size([1, 40488]) Labels shape: torch.Size([1, 40488]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36596 Attention mask shape: torch.Size([1, 1, 36596, 36596]) Position ids shape: torch.Size([1, 36596]) Input IDs shape: torch.Size([1, 36596]) Labels shape: torch.Size([1, 36596]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25674 Attention mask shape: torch.Size([1, 1, 25674, 25674]) Position ids shape: torch.Size([1, 25674]) Input IDs shape: torch.Size([1, 25674]) Labels shape: torch.Size([1, 25674]) Final batch size: 1, sequence length: 28397 Attention mask shape: torch.Size([1, 1, 28397, 28397]) Position ids shape: torch.Size([1, 28397]) Input IDs shape: torch.Size([1, 28397]) Labels shape: torch.Size([1, 28397]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 11611 Attention mask shape: torch.Size([1, 1, 11611, 11611]) Position ids shape: torch.Size([1, 11611]) Input IDs shape: torch.Size([1, 11611]) Labels shape: torch.Size([1, 11611]) Final batch size: 1, sequence length: 24043 Attention mask shape: torch.Size([1, 1, 24043, 24043]) Position ids shape: torch.Size([1, 24043]) Input IDs shape: torch.Size([1, 24043]) Labels shape: torch.Size([1, 24043]) Final batch size: 1, sequence length: 34853 Attention mask shape: torch.Size([1, 1, 34853, 34853]) Position ids shape: torch.Size([1, 34853]) Input IDs shape: torch.Size([1, 34853]) Labels shape: torch.Size([1, 34853]) Final batch size: 1, sequence length: 28086 Attention mask shape: torch.Size([1, 1, 28086, 28086]) Position ids shape: torch.Size([1, 28086]) Input IDs shape: torch.Size([1, 28086]) Labels shape: torch.Size([1, 28086]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16337 Attention mask shape: torch.Size([1, 1, 16337, 16337]) Position ids shape: torch.Size([1, 16337]) Input IDs shape: torch.Size([1, 16337]) Labels shape: torch.Size([1, 16337]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 12218 Attention mask shape: torch.Size([1, 1, 12218, 12218]) Position ids shape: torch.Size([1, 12218]) Input IDs shape: torch.Size([1, 12218]) Labels shape: torch.Size([1, 12218]) Final batch size: 1, sequence length: 36334 Attention mask shape: torch.Size([1, 1, 36334, 36334]) Position ids shape: torch.Size([1, 36334]) Input IDs shape: torch.Size([1, 36334]) Labels shape: torch.Size([1, 36334]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32917 Attention mask shape: torch.Size([1, 1, 32917, 32917]) Position ids shape: torch.Size([1, 32917]) Input IDs shape: torch.Size([1, 32917]) Labels shape: torch.Size([1, 32917]) {'loss': 0.3083, 'grad_norm': 0.41707709085786676, 'learning_rate': 8.715724127386971e-06, 'num_tokens': -inf, 'epoch': 2.38} Final batch size: 1, sequence length: 5818 Attention mask shape: torch.Size([1, 1, 5818, 5818]) Position ids shape: torch.Size([1, 5818]) Input IDs shape: torch.Size([1, 5818]) Labels shape: torch.Size([1, 5818]) Final batch size: 1, sequence length: 6215 Attention mask shape: torch.Size([1, 1, 6215, 6215]) Position ids shape: torch.Size([1, 6215]) Input IDs shape: torch.Size([1, 6215]) Labels shape: torch.Size([1, 6215]) Final batch size: 1, sequence length: 6871 Attention mask shape: torch.Size([1, 1, 6871, 6871]) Position ids shape: torch.Size([1, 6871]) Input IDs shape: torch.Size([1, 6871]) Labels shape: torch.Size([1, 6871]) Final batch size: 1, sequence length: 8623 Attention mask shape: torch.Size([1, 1, 8623, 8623]) Position ids shape: torch.Size([1, 8623]) Input IDs shape: torch.Size([1, 8623]) Labels shape: torch.Size([1, 8623]) Final batch size: 1, sequence length: 5917 Attention mask shape: torch.Size([1, 1, 5917, 5917]) Position ids shape: torch.Size([1, 5917]) Input IDs shape: torch.Size([1, 5917]) Labels shape: torch.Size([1, 5917]) Final batch size: 1, sequence length: 6034 Attention mask shape: torch.Size([1, 1, 6034, 6034]) Position ids shape: torch.Size([1, 6034]) Input IDs shape: torch.Size([1, 6034]) Labels shape: torch.Size([1, 6034]) Final batch size: 1, sequence length: 11616 Attention mask shape: torch.Size([1, 1, 11616, 11616]) Position ids shape: torch.Size([1, 11616]) Input IDs shape: torch.Size([1, 11616]) Labels shape: torch.Size([1, 11616]) Final batch size: 1, sequence length: 12275 Attention mask shape: torch.Size([1, 1, 12275, 12275]) Position ids shape: torch.Size([1, 12275]) Input IDs shape: torch.Size([1, 12275]) Labels shape: torch.Size([1, 12275]) Final batch size: 1, sequence length: 12317 Attention mask shape: torch.Size([1, 1, 12317, 12317]) Position ids shape: torch.Size([1, 12317]) Input IDs shape: torch.Size([1, 12317]) Labels shape: torch.Size([1, 12317]) Final batch size: 1, sequence length: 13202 Attention mask shape: torch.Size([1, 1, 13202, 13202]) Position ids shape: torch.Size([1, 13202]) Input IDs shape: torch.Size([1, 13202]) Labels shape: torch.Size([1, 13202]) Final batch size: 1, sequence length: 9517 Attention mask shape: torch.Size([1, 1, 9517, 9517]) Position ids shape: torch.Size([1, 9517]) Input IDs shape: torch.Size([1, 9517]) Labels shape: torch.Size([1, 9517]) Final batch size: 1, sequence length: 10107 Attention mask shape: torch.Size([1, 1, 10107, 10107]) Position ids shape: torch.Size([1, 10107]) Input IDs shape: torch.Size([1, 10107]) Labels shape: torch.Size([1, 10107]) Final batch size: 1, sequence length: 11648 Attention mask shape: torch.Size([1, 1, 11648, 11648]) Position ids shape: torch.Size([1, 11648]) Input IDs shape: torch.Size([1, 11648]) Labels shape: torch.Size([1, 11648]) Final batch size: 1, sequence length: 15029 Attention mask shape: torch.Size([1, 1, 15029, 15029]) Position ids shape: torch.Size([1, 15029]) Input IDs shape: torch.Size([1, 15029]) Labels shape: torch.Size([1, 15029]) Final batch size: 1, sequence length: 8117 Attention mask shape: torch.Size([1, 1, 8117, 8117]) Position ids shape: torch.Size([1, 8117]) Input IDs shape: torch.Size([1, 8117]) Labels shape: torch.Size([1, 8117]) Final batch size: 1, sequence length: 10136 Attention mask shape: torch.Size([1, 1, 10136, 10136]) Position ids shape: torch.Size([1, 10136]) Input IDs shape: torch.Size([1, 10136]) Labels shape: torch.Size([1, 10136]) Final batch size: 1, sequence length: 13428 Attention mask shape: torch.Size([1, 1, 13428, 13428]) Position ids shape: torch.Size([1, 13428]) Input IDs shape: torch.Size([1, 13428]) Labels shape: torch.Size([1, 13428]) Final batch size: 1, sequence length: 16331 Attention mask shape: torch.Size([1, 1, 16331, 16331]) Position ids shape: torch.Size([1, 16331]) Input IDs shape: torch.Size([1, 16331]) Labels shape: torch.Size([1, 16331]) Final batch size: 1, sequence length: 12813 Attention mask shape: torch.Size([1, 1, 12813, 12813]) Position ids shape: torch.Size([1, 12813]) Input IDs shape: torch.Size([1, 12813]) Labels shape: torch.Size([1, 12813]) Final batch size: 1, sequence length: 15499 Attention mask shape: torch.Size([1, 1, 15499, 15499]) Position ids shape: torch.Size([1, 15499]) Input IDs shape: torch.Size([1, 15499]) Labels shape: torch.Size([1, 15499]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 17587 Attention mask shape: torch.Size([1, 1, 17587, 17587]) Position ids shape: torch.Size([1, 17587]) Input IDs shape: torch.Size([1, 17587]) Labels shape: torch.Size([1, 17587]) Final batch size: 1, sequence length: 12553 Attention mask shape: torch.Size([1, 1, 12553, 12553]) Position ids shape: torch.Size([1, 12553]) Input IDs shape: torch.Size([1, 12553]) Labels shape: torch.Size([1, 12553]) Final batch size: 1, sequence length: 15066 Attention mask shape: torch.Size([1, 1, 15066, 15066]) Position ids shape: torch.Size([1, 15066]) Input IDs shape: torch.Size([1, 15066]) Labels shape: torch.Size([1, 15066]) Final batch size: 1, sequence length: 18414 Attention mask shape: torch.Size([1, 1, 18414, 18414]) Position ids shape: torch.Size([1, 18414]) Input IDs shape: torch.Size([1, 18414]) Labels shape: torch.Size([1, 18414]) Final batch size: 1, sequence length: 18469 Attention mask shape: torch.Size([1, 1, 18469, 18469]) Position ids shape: torch.Size([1, 18469]) Input IDs shape: torch.Size([1, 18469]) Labels shape: torch.Size([1, 18469]) Final batch size: 1, sequence length: 19221 Attention mask shape: torch.Size([1, 1, 19221, 19221]) Position ids shape: torch.Size([1, 19221]) Input IDs shape: torch.Size([1, 19221]) Labels shape: torch.Size([1, 19221]) Final batch size: 1, sequence length: 21028 Attention mask shape: torch.Size([1, 1, 21028, 21028]) Position ids shape: torch.Size([1, 21028]) Input IDs shape: torch.Size([1, 21028]) Labels shape: torch.Size([1, 21028]) Final batch size: 1, sequence length: 15226 Attention mask shape: torch.Size([1, 1, 15226, 15226]) Position ids shape: torch.Size([1, 15226]) Input IDs shape: torch.Size([1, 15226]) Labels shape: torch.Size([1, 15226]) Final batch size: 1, sequence length: 19847 Attention mask shape: torch.Size([1, 1, 19847, 19847]) Position ids shape: torch.Size([1, 19847]) Input IDs shape: torch.Size([1, 19847]) Labels shape: torch.Size([1, 19847]) Final batch size: 1, sequence length: 21556 Attention mask shape: torch.Size([1, 1, 21556, 21556]) Position ids shape: torch.Size([1, 21556]) Input IDs shape: torch.Size([1, 21556]) Labels shape: torch.Size([1, 21556]) Final batch size: 1, sequence length: 19512 Attention mask shape: torch.Size([1, 1, 19512, 19512]) Position ids shape: torch.Size([1, 19512]) Input IDs shape: torch.Size([1, 19512]) Labels shape: torch.Size([1, 19512]) Final batch size: 1, sequence length: 18438 Attention mask shape: torch.Size([1, 1, 18438, 18438]) Position ids shape: torch.Size([1, 18438]) Input IDs shape: torch.Size([1, 18438]) Labels shape: torch.Size([1, 18438]) Final batch size: 1, sequence length: 22079 Attention mask shape: torch.Size([1, 1, 22079, 22079]) Position ids shape: torch.Size([1, 22079]) Input IDs shape: torch.Size([1, 22079]) Labels shape: torch.Size([1, 22079]) Final batch size: 1, sequence length: 22618 Attention mask shape: torch.Size([1, 1, 22618, 22618]) Position ids shape: torch.Size([1, 22618]) Input IDs shape: torch.Size([1, 22618]) Labels shape: torch.Size([1, 22618]) Final batch size: 1, sequence length: 22138 Attention mask shape: torch.Size([1, 1, 22138, 22138]) Position ids shape: torch.Size([1, 22138]) Input IDs shape: torch.Size([1, 22138]) Labels shape: torch.Size([1, 22138]) Final batch size: 1, sequence length: 22718 Attention mask shape: torch.Size([1, 1, 22718, 22718]) Position ids shape: torch.Size([1, 22718]) Input IDs shape: torch.Size([1, 22718]) Labels shape: torch.Size([1, 22718]) Final batch size: 1, sequence length: 7584 Attention mask shape: torch.Size([1, 1, 7584, 7584]) Position ids shape: torch.Size([1, 7584]) Input IDs shape: torch.Size([1, 7584]) Labels shape: torch.Size([1, 7584]) Final batch size: 1, sequence length: 19428 Attention mask shape: torch.Size([1, 1, 19428, 19428]) Position ids shape: torch.Size([1, 19428]) Input IDs shape: torch.Size([1, 19428]) Labels shape: torch.Size([1, 19428]) Final batch size: 1, sequence length: 25999 Attention mask shape: torch.Size([1, 1, 25999, 25999]) Position ids shape: torch.Size([1, 25999]) Input IDs shape: torch.Size([1, 25999]) Labels shape: torch.Size([1, 25999]) Final batch size: 1, sequence length: 14784 Attention mask shape: torch.Size([1, 1, 14784, 14784]) Position ids shape: torch.Size([1, 14784]) Input IDs shape: torch.Size([1, 14784]) Labels shape: torch.Size([1, 14784]) Final batch size: 1, sequence length: 22763 Attention mask shape: torch.Size([1, 1, 22763, 22763]) Position ids shape: torch.Size([1, 22763]) Input IDs shape: torch.Size([1, 22763]) Labels shape: torch.Size([1, 22763]) Final batch size: 1, sequence length: 25622 Attention mask shape: torch.Size([1, 1, 25622, 25622]) Position ids shape: torch.Size([1, 25622]) Input IDs shape: torch.Size([1, 25622]) Labels shape: torch.Size([1, 25622]) Final batch size: 1, sequence length: 23973 Attention mask shape: torch.Size([1, 1, 23973, 23973]) Position ids shape: torch.Size([1, 23973]) Input IDs shape: torch.Size([1, 23973]) Labels shape: torch.Size([1, 23973]) Final batch size: 1, sequence length: 23766 Attention mask shape: torch.Size([1, 1, 23766, 23766]) Position ids shape: torch.Size([1, 23766]) Input IDs shape: torch.Size([1, 23766]) Labels shape: torch.Size([1, 23766]) Final batch size: 1, sequence length: 16299 Attention mask shape: torch.Size([1, 1, 16299, 16299]) Position ids shape: torch.Size([1, 16299]) Input IDs shape: torch.Size([1, 16299]) Labels shape: torch.Size([1, 16299]) Final batch size: 1, sequence length: 24694 Attention mask shape: torch.Size([1, 1, 24694, 24694]) Position ids shape: torch.Size([1, 24694]) Input IDs shape: torch.Size([1, 24694]) Labels shape: torch.Size([1, 24694]) Final batch size: 1, sequence length: 24499 Attention mask shape: torch.Size([1, 1, 24499, 24499]) Position ids shape: torch.Size([1, 24499]) Input IDs shape: torch.Size([1, 24499]) Labels shape: torch.Size([1, 24499]) Final batch size: 1, sequence length: 24909 Attention mask shape: torch.Size([1, 1, 24909, 24909]) Position ids shape: torch.Size([1, 24909]) Input IDs shape: torch.Size([1, 24909]) Labels shape: torch.Size([1, 24909]) Final batch size: 1, sequence length: 15221 Attention mask shape: torch.Size([1, 1, 15221, 15221]) Position ids shape: torch.Size([1, 15221]) Input IDs shape: torch.Size([1, 15221]) Labels shape: torch.Size([1, 15221]) Final batch size: 1, sequence length: 22235 Attention mask shape: torch.Size([1, 1, 22235, 22235]) Position ids shape: torch.Size([1, 22235]) Input IDs shape: torch.Size([1, 22235]) Labels shape: torch.Size([1, 22235]) Final batch size: 1, sequence length: 15184 Attention mask shape: torch.Size([1, 1, 15184, 15184]) Position ids shape: torch.Size([1, 15184]) Input IDs shape: torch.Size([1, 15184]) Labels shape: torch.Size([1, 15184]) Final batch size: 1, sequence length: 27566 Attention mask shape: torch.Size([1, 1, 27566, 27566]) Position ids shape: torch.Size([1, 27566]) Input IDs shape: torch.Size([1, 27566]) Labels shape: torch.Size([1, 27566]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 28497 Attention mask shape: torch.Size([1, 1, 28497, 28497]) Position ids shape: torch.Size([1, 28497]) Input IDs shape: torch.Size([1, 28497]) Labels shape: torch.Size([1, 28497]) Final batch size: 1, sequence length: 21826 Attention mask shape: torch.Size([1, 1, 21826, 21826]) Position ids shape: torch.Size([1, 21826]) Input IDs shape: torch.Size([1, 21826]) Labels shape: torch.Size([1, 21826]) Final batch size: 1, sequence length: 27327 Attention mask shape: torch.Size([1, 1, 27327, 27327]) Position ids shape: torch.Size([1, 27327]) Input IDs shape: torch.Size([1, 27327]) Labels shape: torch.Size([1, 27327]) Final batch size: 1, sequence length: 28749 Attention mask shape: torch.Size([1, 1, 28749, 28749]) Position ids shape: torch.Size([1, 28749]) Input IDs shape: torch.Size([1, 28749]) Labels shape: torch.Size([1, 28749]) Final batch size: 1, sequence length: 15588 Attention mask shape: torch.Size([1, 1, 15588, 15588]) Position ids shape: torch.Size([1, 15588]) Input IDs shape: torch.Size([1, 15588]) Labels shape: torch.Size([1, 15588]) Final batch size: 1, sequence length: 30356 Attention mask shape: torch.Size([1, 1, 30356, 30356]) Position ids shape: torch.Size([1, 30356]) Input IDs shape: torch.Size([1, 30356]) Labels shape: torch.Size([1, 30356]) Final batch size: 1, sequence length: 23851 Attention mask shape: torch.Size([1, 1, 23851, 23851]) Position ids shape: torch.Size([1, 23851]) Input IDs shape: torch.Size([1, 23851]) Labels shape: torch.Size([1, 23851]) Final batch size: 1, sequence length: 11795 Attention mask shape: torch.Size([1, 1, 11795, 11795]) Position ids shape: torch.Size([1, 11795]) Input IDs shape: torch.Size([1, 11795]) Labels shape: torch.Size([1, 11795]) Final batch size: 1, sequence length: 10269 Attention mask shape: torch.Size([1, 1, 10269, 10269]) Position ids shape: torch.Size([1, 10269]) Input IDs shape: torch.Size([1, 10269]) Labels shape: torch.Size([1, 10269]) Final batch size: 1, sequence length: 36777 Attention mask shape: torch.Size([1, 1, 36777, 36777]) Position ids shape: torch.Size([1, 36777]) Input IDs shape: torch.Size([1, 36777]) Labels shape: torch.Size([1, 36777]) Final batch size: 1, sequence length: 20755 Attention mask shape: torch.Size([1, 1, 20755, 20755]) Position ids shape: torch.Size([1, 20755]) Input IDs shape: torch.Size([1, 20755]) Labels shape: torch.Size([1, 20755]) Final batch size: 1, sequence length: 34673 Attention mask shape: torch.Size([1, 1, 34673, 34673]) Position ids shape: torch.Size([1, 34673]) Input IDs shape: torch.Size([1, 34673]) Labels shape: torch.Size([1, 34673]) Final batch size: 1, sequence length: 21034 Attention mask shape: torch.Size([1, 1, 21034, 21034]) Position ids shape: torch.Size([1, 21034]) Input IDs shape: torch.Size([1, 21034]) Labels shape: torch.Size([1, 21034]) Final batch size: 1, sequence length: 27489 Attention mask shape: torch.Size([1, 1, 27489, 27489]) Position ids shape: torch.Size([1, 27489]) Input IDs shape: torch.Size([1, 27489]) Labels shape: torch.Size([1, 27489]) Final batch size: 1, sequence length: 22366 Attention mask shape: torch.Size([1, 1, 22366, 22366]) Position ids shape: torch.Size([1, 22366]) Input IDs shape: torch.Size([1, 22366]) Labels shape: torch.Size([1, 22366]) Final batch size: 1, sequence length: 38049 Attention mask shape: torch.Size([1, 1, 38049, 38049]) Position ids shape: torch.Size([1, 38049]) Input IDs shape: torch.Size([1, 38049]) Labels shape: torch.Size([1, 38049]) Final batch size: 1, sequence length: 38848 Attention mask shape: torch.Size([1, 1, 38848, 38848]) Position ids shape: torch.Size([1, 38848]) Input IDs shape: torch.Size([1, 38848]) Labels shape: torch.Size([1, 38848]) Final batch size: 1, sequence length: 36723 Attention mask shape: torch.Size([1, 1, 36723, 36723]) Position ids shape: torch.Size([1, 36723]) Input IDs shape: torch.Size([1, 36723]) Labels shape: torch.Size([1, 36723]) Final batch size: 1, sequence length: 28176 Attention mask shape: torch.Size([1, 1, 28176, 28176]) Position ids shape: torch.Size([1, 28176]) Input IDs shape: torch.Size([1, 28176]) Labels shape: torch.Size([1, 28176]) Final batch size: 1, sequence length: 37285 Attention mask shape: torch.Size([1, 1, 37285, 37285]) Position ids shape: torch.Size([1, 37285]) Input IDs shape: torch.Size([1, 37285]) Labels shape: torch.Size([1, 37285]) Final batch size: 1, sequence length: 31903 Attention mask shape: torch.Size([1, 1, 31903, 31903]) Position ids shape: torch.Size([1, 31903]) Input IDs shape: torch.Size([1, 31903]) Labels shape: torch.Size([1, 31903]) Final batch size: 1, sequence length: 40325 Attention mask shape: torch.Size([1, 1, 40325, 40325]) Position ids shape: torch.Size([1, 40325]) Input IDs shape: torch.Size([1, 40325]) Labels shape: torch.Size([1, 40325]) Final batch size: 1, sequence length: 35904 Attention mask shape: torch.Size([1, 1, 35904, 35904]) Position ids shape: torch.Size([1, 35904]) Input IDs shape: torch.Size([1, 35904]) Labels shape: torch.Size([1, 35904]) Final batch size: 1, sequence length: 25946 Attention mask shape: torch.Size([1, 1, 25946, 25946]) Position ids shape: torch.Size([1, 25946]) Input IDs shape: torch.Size([1, 25946]) Labels shape: torch.Size([1, 25946]) Final batch size: 1, sequence length: 6882 Attention mask shape: torch.Size([1, 1, 6882, 6882]) Position ids shape: torch.Size([1, 6882]) Input IDs shape: torch.Size([1, 6882]) Labels shape: torch.Size([1, 6882]) Final batch size: 1, sequence length: 39253 Attention mask shape: torch.Size([1, 1, 39253, 39253]) Position ids shape: torch.Size([1, 39253]) Input IDs shape: torch.Size([1, 39253]) Labels shape: torch.Size([1, 39253]) Final batch size: 1, sequence length: 38104 Attention mask shape: torch.Size([1, 1, 38104, 38104]) Position ids shape: torch.Size([1, 38104]) Input IDs shape: torch.Size([1, 38104]) Labels shape: torch.Size([1, 38104]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40317 Attention mask shape: torch.Size([1, 1, 40317, 40317]) Position ids shape: torch.Size([1, 40317]) Input IDs shape: torch.Size([1, 40317]) Labels shape: torch.Size([1, 40317]) Final batch size: 1, sequence length: 24801 Attention mask shape: torch.Size([1, 1, 24801, 24801]) Position ids shape: torch.Size([1, 24801]) Input IDs shape: torch.Size([1, 24801]) Labels shape: torch.Size([1, 24801]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34289 Attention mask shape: torch.Size([1, 1, 34289, 34289]) Position ids shape: torch.Size([1, 34289]) Input IDs shape: torch.Size([1, 34289]) Labels shape: torch.Size([1, 34289]) Final batch size: 1, sequence length: 36786 Attention mask shape: torch.Size([1, 1, 36786, 36786]) Position ids shape: torch.Size([1, 36786]) Input IDs shape: torch.Size([1, 36786]) Labels shape: torch.Size([1, 36786]) Final batch size: 1, sequence length: 40937 Attention mask shape: torch.Size([1, 1, 40937, 40937]) Position ids shape: torch.Size([1, 40937]) Input IDs shape: torch.Size([1, 40937]) Labels shape: torch.Size([1, 40937]) Final batch size: 1, sequence length: 10469 Attention mask shape: torch.Size([1, 1, 10469, 10469]) Position ids shape: torch.Size([1, 10469]) Input IDs shape: torch.Size([1, 10469]) Labels shape: torch.Size([1, 10469]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19639 Attention mask shape: torch.Size([1, 1, 19639, 19639]) Position ids shape: torch.Size([1, 19639]) Input IDs shape: torch.Size([1, 19639]) Labels shape: torch.Size([1, 19639]) Final batch size: 1, sequence length: 40657 Attention mask shape: torch.Size([1, 1, 40657, 40657]) Position ids shape: torch.Size([1, 40657]) Input IDs shape: torch.Size([1, 40657]) Labels shape: torch.Size([1, 40657]) Final batch size: 1, sequence length: 19437 Attention mask shape: torch.Size([1, 1, 19437, 19437]) Position ids shape: torch.Size([1, 19437]) Input IDs shape: torch.Size([1, 19437]) Labels shape: torch.Size([1, 19437]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16649 Attention mask shape: torch.Size([1, 1, 16649, 16649]) Position ids shape: torch.Size([1, 16649]) Input IDs shape: torch.Size([1, 16649]) Labels shape: torch.Size([1, 16649]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25963 Attention mask shape: torch.Size([1, 1, 25963, 25963]) Position ids shape: torch.Size([1, 25963]) Input IDs shape: torch.Size([1, 25963]) Labels shape: torch.Size([1, 25963]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38686 Attention mask shape: torch.Size([1, 1, 38686, 38686]) Position ids shape: torch.Size([1, 38686]) Input IDs shape: torch.Size([1, 38686]) Labels shape: torch.Size([1, 38686]) Final batch size: 1, sequence length: 36128 Attention mask shape: torch.Size([1, 1, 36128, 36128]) Position ids shape: torch.Size([1, 36128]) Input IDs shape: torch.Size([1, 36128]) Labels shape: torch.Size([1, 36128]) Final batch size: 1, sequence length: 21547 Attention mask shape: torch.Size([1, 1, 21547, 21547]) Position ids shape: torch.Size([1, 21547]) Input IDs shape: torch.Size([1, 21547]) Labels shape: torch.Size([1, 21547]) Final batch size: 1, sequence length: 34540 Attention mask shape: torch.Size([1, 1, 34540, 34540]) Position ids shape: torch.Size([1, 34540]) Input IDs shape: torch.Size([1, 34540]) Labels shape: torch.Size([1, 34540]) Final batch size: 1, sequence length: 17882 Attention mask shape: torch.Size([1, 1, 17882, 17882]) Position ids shape: torch.Size([1, 17882]) Input IDs shape: torch.Size([1, 17882]) Labels shape: torch.Size([1, 17882]) Final batch size: 1, sequence length: 31714 Attention mask shape: torch.Size([1, 1, 31714, 31714]) Position ids shape: torch.Size([1, 31714]) Input IDs shape: torch.Size([1, 31714]) Labels shape: torch.Size([1, 31714]) Final batch size: 1, sequence length: 17379 Attention mask shape: torch.Size([1, 1, 17379, 17379]) Position ids shape: torch.Size([1, 17379]) Input IDs shape: torch.Size([1, 17379]) Labels shape: torch.Size([1, 17379]) Final batch size: 1, sequence length: 27947 Attention mask shape: torch.Size([1, 1, 27947, 27947]) Position ids shape: torch.Size([1, 27947]) Input IDs shape: torch.Size([1, 27947]) Labels shape: torch.Size([1, 27947]) Final batch size: 1, sequence length: 26660 Attention mask shape: torch.Size([1, 1, 26660, 26660]) Position ids shape: torch.Size([1, 26660]) Input IDs shape: torch.Size([1, 26660]) Labels shape: torch.Size([1, 26660]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20843 Attention mask shape: torch.Size([1, 1, 20843, 20843]) Position ids shape: torch.Size([1, 20843]) Input IDs shape: torch.Size([1, 20843]) Labels shape: torch.Size([1, 20843]) Final batch size: 1, sequence length: 37946 Attention mask shape: torch.Size([1, 1, 37946, 37946]) Position ids shape: torch.Size([1, 37946]) Input IDs shape: torch.Size([1, 37946]) Labels shape: torch.Size([1, 37946]) Final batch size: 1, sequence length: 21496 Attention mask shape: torch.Size([1, 1, 21496, 21496]) Position ids shape: torch.Size([1, 21496]) Input IDs shape: torch.Size([1, 21496]) Labels shape: torch.Size([1, 21496]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24551 Attention mask shape: torch.Size([1, 1, 24551, 24551]) Position ids shape: torch.Size([1, 24551]) Input IDs shape: torch.Size([1, 24551]) Labels shape: torch.Size([1, 24551]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7681 Attention mask shape: torch.Size([1, 1, 7681, 7681]) Position ids shape: torch.Size([1, 7681]) Input IDs shape: torch.Size([1, 7681]) Labels shape: torch.Size([1, 7681]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36317 Attention mask shape: torch.Size([1, 1, 36317, 36317]) Position ids shape: torch.Size([1, 36317]) Input IDs shape: torch.Size([1, 36317]) Labels shape: torch.Size([1, 36317]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.3096, 'grad_norm': 0.41839991237313046, 'learning_rate': 8.535533905932739e-06, 'num_tokens': -inf, 'epoch': 2.5} Final batch size: 1, sequence length: 5328 Attention mask shape: torch.Size([1, 1, 5328, 5328]) Position ids shape: torch.Size([1, 5328]) Input IDs shape: torch.Size([1, 5328]) Labels shape: torch.Size([1, 5328]) Final batch size: 1, sequence length: 5525 Attention mask shape: torch.Size([1, 1, 5525, 5525]) Position ids shape: torch.Size([1, 5525]) Input IDs shape: torch.Size([1, 5525]) Labels shape: torch.Size([1, 5525]) Final batch size: 1, sequence length: 10273 Attention mask shape: torch.Size([1, 1, 10273, 10273]) Position ids shape: torch.Size([1, 10273]) Input IDs shape: torch.Size([1, 10273]) Labels shape: torch.Size([1, 10273]) Final batch size: 1, sequence length: 11464 Attention mask shape: torch.Size([1, 1, 11464, 11464]) Position ids shape: torch.Size([1, 11464]) Input IDs shape: torch.Size([1, 11464]) Labels shape: torch.Size([1, 11464]) Final batch size: 1, sequence length: 12419 Attention mask shape: torch.Size([1, 1, 12419, 12419]) Position ids shape: torch.Size([1, 12419]) Input IDs shape: torch.Size([1, 12419]) Labels shape: torch.Size([1, 12419]) Final batch size: 1, sequence length: 13607 Attention mask shape: torch.Size([1, 1, 13607, 13607]) Position ids shape: torch.Size([1, 13607]) Input IDs shape: torch.Size([1, 13607]) Labels shape: torch.Size([1, 13607]) Final batch size: 1, sequence length: 14226 Attention mask shape: torch.Size([1, 1, 14226, 14226]) Position ids shape: torch.Size([1, 14226]) Input IDs shape: torch.Size([1, 14226]) Labels shape: torch.Size([1, 14226]) Final batch size: 1, sequence length: 9648 Attention mask shape: torch.Size([1, 1, 9648, 9648]) Position ids shape: torch.Size([1, 9648]) Input IDs shape: torch.Size([1, 9648]) Labels shape: torch.Size([1, 9648]) Final batch size: 1, sequence length: 13657 Attention mask shape: torch.Size([1, 1, 13657, 13657]) Position ids shape: torch.Size([1, 13657]) Input IDs shape: torch.Size([1, 13657]) Labels shape: torch.Size([1, 13657]) Final batch size: 1, sequence length: 14360 Attention mask shape: torch.Size([1, 1, 14360, 14360]) Position ids shape: torch.Size([1, 14360]) Input IDs shape: torch.Size([1, 14360]) Labels shape: torch.Size([1, 14360]) Final batch size: 1, sequence length: 10434 Attention mask shape: torch.Size([1, 1, 10434, 10434]) Position ids shape: torch.Size([1, 10434]) Input IDs shape: torch.Size([1, 10434]) Labels shape: torch.Size([1, 10434]) Final batch size: 1, sequence length: 16398 Attention mask shape: torch.Size([1, 1, 16398, 16398]) Position ids shape: torch.Size([1, 16398]) Input IDs shape: torch.Size([1, 16398]) Labels shape: torch.Size([1, 16398]) Final batch size: 1, sequence length: 16756 Attention mask shape: torch.Size([1, 1, 16756, 16756]) Position ids shape: torch.Size([1, 16756]) Input IDs shape: torch.Size([1, 16756]) Labels shape: torch.Size([1, 16756]) Final batch size: 1, sequence length: 16053 Attention mask shape: torch.Size([1, 1, 16053, 16053]) Position ids shape: torch.Size([1, 16053]) Input IDs shape: torch.Size([1, 16053]) Labels shape: torch.Size([1, 16053]) Final batch size: 1, sequence length: 17294 Attention mask shape: torch.Size([1, 1, 17294, 17294]) Position ids shape: torch.Size([1, 17294]) Input IDs shape: torch.Size([1, 17294]) Labels shape: torch.Size([1, 17294]) Final batch size: 1, sequence length: 18408 Attention mask shape: torch.Size([1, 1, 18408, 18408]) Position ids shape: torch.Size([1, 18408]) Input IDs shape: torch.Size([1, 18408]) Labels shape: torch.Size([1, 18408]) Final batch size: 1, sequence length: 16223 Attention mask shape: torch.Size([1, 1, 16223, 16223]) Position ids shape: torch.Size([1, 16223]) Input IDs shape: torch.Size([1, 16223]) Labels shape: torch.Size([1, 16223]) Final batch size: 1, sequence length: 10017 Attention mask shape: torch.Size([1, 1, 10017, 10017]) Position ids shape: torch.Size([1, 10017]) Input IDs shape: torch.Size([1, 10017]) Labels shape: torch.Size([1, 10017]) Final batch size: 1, sequence length: 15243 Attention mask shape: torch.Size([1, 1, 15243, 15243]) Position ids shape: torch.Size([1, 15243]) Input IDs shape: torch.Size([1, 15243]) Labels shape: torch.Size([1, 15243]) Final batch size: 1, sequence length: 21455 Attention mask shape: torch.Size([1, 1, 21455, 21455]) Position ids shape: torch.Size([1, 21455]) Input IDs shape: torch.Size([1, 21455]) Labels shape: torch.Size([1, 21455]) Final batch size: 1, sequence length: 19278 Attention mask shape: torch.Size([1, 1, 19278, 19278]) Position ids shape: torch.Size([1, 19278]) Input IDs shape: torch.Size([1, 19278]) Labels shape: torch.Size([1, 19278]) Final batch size: 1, sequence length: 20433 Attention mask shape: torch.Size([1, 1, 20433, 20433]) Position ids shape: torch.Size([1, 20433]) Input IDs shape: torch.Size([1, 20433]) Labels shape: torch.Size([1, 20433]) Final batch size: 1, sequence length: 14437 Attention mask shape: torch.Size([1, 1, 14437, 14437]) Position ids shape: torch.Size([1, 14437]) Input IDs shape: torch.Size([1, 14437]) Labels shape: torch.Size([1, 14437]) Final batch size: 1, sequence length: 17117 Attention mask shape: torch.Size([1, 1, 17117, 17117]) Position ids shape: torch.Size([1, 17117]) Input IDs shape: torch.Size([1, 17117]) Labels shape: torch.Size([1, 17117]) Final batch size: 1, sequence length: 20695 Attention mask shape: torch.Size([1, 1, 20695, 20695]) Position ids shape: torch.Size([1, 20695]) Input IDs shape: torch.Size([1, 20695]) Labels shape: torch.Size([1, 20695]) Final batch size: 1, sequence length: 17843 Attention mask shape: torch.Size([1, 1, 17843, 17843]) Position ids shape: torch.Size([1, 17843]) Input IDs shape: torch.Size([1, 17843]) Labels shape: torch.Size([1, 17843]) Final batch size: 1, sequence length: 19767 Attention mask shape: torch.Size([1, 1, 19767, 19767]) Position ids shape: torch.Size([1, 19767]) Input IDs shape: torch.Size([1, 19767]) Labels shape: torch.Size([1, 19767]) Final batch size: 1, sequence length: 19259 Attention mask shape: torch.Size([1, 1, 19259, 19259]) Position ids shape: torch.Size([1, 19259]) Input IDs shape: torch.Size([1, 19259]) Labels shape: torch.Size([1, 19259]) Final batch size: 1, sequence length: 19892 Attention mask shape: torch.Size([1, 1, 19892, 19892]) Position ids shape: torch.Size([1, 19892]) Input IDs shape: torch.Size([1, 19892]) Labels shape: torch.Size([1, 19892]) Final batch size: 1, sequence length: 6378 Attention mask shape: torch.Size([1, 1, 6378, 6378]) Position ids shape: torch.Size([1, 6378]) Input IDs shape: torch.Size([1, 6378]) Labels shape: torch.Size([1, 6378]) Final batch size: 1, sequence length: 11225 Attention mask shape: torch.Size([1, 1, 11225, 11225]) Position ids shape: torch.Size([1, 11225]) Input IDs shape: torch.Size([1, 11225]) Labels shape: torch.Size([1, 11225]) Final batch size: 1, sequence length: 14025 Attention mask shape: torch.Size([1, 1, 14025, 14025]) Position ids shape: torch.Size([1, 14025]) Input IDs shape: torch.Size([1, 14025]) Labels shape: torch.Size([1, 14025]) Final batch size: 1, sequence length: 14429 Attention mask shape: torch.Size([1, 1, 14429, 14429]) Position ids shape: torch.Size([1, 14429]) Input IDs shape: torch.Size([1, 14429]) Labels shape: torch.Size([1, 14429]) Final batch size: 1, sequence length: 25405 Attention mask shape: torch.Size([1, 1, 25405, 25405]) Position ids shape: torch.Size([1, 25405]) Input IDs shape: torch.Size([1, 25405]) Labels shape: torch.Size([1, 25405]) Final batch size: 1, sequence length: 19492 Attention mask shape: torch.Size([1, 1, 19492, 19492]) Position ids shape: torch.Size([1, 19492]) Input IDs shape: torch.Size([1, 19492]) Labels shape: torch.Size([1, 19492]) Final batch size: 1, sequence length: 5734 Attention mask shape: torch.Size([1, 1, 5734, 5734]) Position ids shape: torch.Size([1, 5734]) Input IDs shape: torch.Size([1, 5734]) Labels shape: torch.Size([1, 5734]) Final batch size: 1, sequence length: 23334 Attention mask shape: torch.Size([1, 1, 23334, 23334]) Position ids shape: torch.Size([1, 23334]) Input IDs shape: torch.Size([1, 23334]) Labels shape: torch.Size([1, 23334]) Final batch size: 1, sequence length: 13623 Attention mask shape: torch.Size([1, 1, 13623, 13623]) Position ids shape: torch.Size([1, 13623]) Input IDs shape: torch.Size([1, 13623]) Labels shape: torch.Size([1, 13623]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 17985 Attention mask shape: torch.Size([1, 1, 17985, 17985]) Position ids shape: torch.Size([1, 17985]) Input IDs shape: torch.Size([1, 17985]) Labels shape: torch.Size([1, 17985]) Final batch size: 1, sequence length: 22932 Attention mask shape: torch.Size([1, 1, 22932, 22932]) Position ids shape: torch.Size([1, 22932]) Input IDs shape: torch.Size([1, 22932]) Labels shape: torch.Size([1, 22932]) Final batch size: 1, sequence length: 26063 Attention mask shape: torch.Size([1, 1, 26063, 26063]) Position ids shape: torch.Size([1, 26063]) Input IDs shape: torch.Size([1, 26063]) Labels shape: torch.Size([1, 26063]) Final batch size: 1, sequence length: 17951 Attention mask shape: torch.Size([1, 1, 17951, 17951]) Position ids shape: torch.Size([1, 17951]) Input IDs shape: torch.Size([1, 17951]) Labels shape: torch.Size([1, 17951]) Final batch size: 1, sequence length: 24885 Attention mask shape: torch.Size([1, 1, 24885, 24885]) Position ids shape: torch.Size([1, 24885]) Input IDs shape: torch.Size([1, 24885]) Labels shape: torch.Size([1, 24885]) Final batch size: 1, sequence length: 25381 Attention mask shape: torch.Size([1, 1, 25381, 25381]) Position ids shape: torch.Size([1, 25381]) Input IDs shape: torch.Size([1, 25381]) Labels shape: torch.Size([1, 25381]) Final batch size: 1, sequence length: 27484 Attention mask shape: torch.Size([1, 1, 27484, 27484]) Position ids shape: torch.Size([1, 27484]) Input IDs shape: torch.Size([1, 27484]) Labels shape: torch.Size([1, 27484]) Final batch size: 1, sequence length: 28623 Attention mask shape: torch.Size([1, 1, 28623, 28623]) Position ids shape: torch.Size([1, 28623]) Input IDs shape: torch.Size([1, 28623]) Labels shape: torch.Size([1, 28623]) Final batch size: 1, sequence length: 16716 Attention mask shape: torch.Size([1, 1, 16716, 16716]) Position ids shape: torch.Size([1, 16716]) Input IDs shape: torch.Size([1, 16716]) Labels shape: torch.Size([1, 16716]) Final batch size: 1, sequence length: 29561 Attention mask shape: torch.Size([1, 1, 29561, 29561]) Position ids shape: torch.Size([1, 29561]) Input IDs shape: torch.Size([1, 29561]) Labels shape: torch.Size([1, 29561]) Final batch size: 1, sequence length: 25548 Attention mask shape: torch.Size([1, 1, 25548, 25548]) Position ids shape: torch.Size([1, 25548]) Input IDs shape: torch.Size([1, 25548]) Labels shape: torch.Size([1, 25548]) Final batch size: 1, sequence length: 26639 Attention mask shape: torch.Size([1, 1, 26639, 26639]) Position ids shape: torch.Size([1, 26639]) Input IDs shape: torch.Size([1, 26639]) Labels shape: torch.Size([1, 26639]) Final batch size: 1, sequence length: 28910 Attention mask shape: torch.Size([1, 1, 28910, 28910]) Position ids shape: torch.Size([1, 28910]) Input IDs shape: torch.Size([1, 28910]) Labels shape: torch.Size([1, 28910]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 28166 Attention mask shape: torch.Size([1, 1, 28166, 28166]) Position ids shape: torch.Size([1, 28166]) Input IDs shape: torch.Size([1, 28166]) Labels shape: torch.Size([1, 28166]) Final batch size: 1, sequence length: 27659 Attention mask shape: torch.Size([1, 1, 27659, 27659]) Position ids shape: torch.Size([1, 27659]) Input IDs shape: torch.Size([1, 27659]) Labels shape: torch.Size([1, 27659]) Final batch size: 1, sequence length: 29824 Attention mask shape: torch.Size([1, 1, 29824, 29824]) Position ids shape: torch.Size([1, 29824]) Input IDs shape: torch.Size([1, 29824]) Labels shape: torch.Size([1, 29824]) Final batch size: 1, sequence length: 25540 Attention mask shape: torch.Size([1, 1, 25540, 25540]) Position ids shape: torch.Size([1, 25540]) Input IDs shape: torch.Size([1, 25540]) Labels shape: torch.Size([1, 25540]) Final batch size: 1, sequence length: 30236 Attention mask shape: torch.Size([1, 1, 30236, 30236]) Position ids shape: torch.Size([1, 30236]) Input IDs shape: torch.Size([1, 30236]) Labels shape: torch.Size([1, 30236]) Final batch size: 1, sequence length: 28206 Attention mask shape: torch.Size([1, 1, 28206, 28206]) Position ids shape: torch.Size([1, 28206]) Input IDs shape: torch.Size([1, 28206]) Labels shape: torch.Size([1, 28206]) Final batch size: 1, sequence length: 17634 Attention mask shape: torch.Size([1, 1, 17634, 17634]) Position ids shape: torch.Size([1, 17634]) Input IDs shape: torch.Size([1, 17634]) Labels shape: torch.Size([1, 17634]) Final batch size: 1, sequence length: 28823 Attention mask shape: torch.Size([1, 1, 28823, 28823]) Position ids shape: torch.Size([1, 28823]) Input IDs shape: torch.Size([1, 28823]) Labels shape: torch.Size([1, 28823]) Final batch size: 1, sequence length: 32515 Attention mask shape: torch.Size([1, 1, 32515, 32515]) Position ids shape: torch.Size([1, 32515]) Input IDs shape: torch.Size([1, 32515]) Labels shape: torch.Size([1, 32515]) Final batch size: 1, sequence length: 31464 Attention mask shape: torch.Size([1, 1, 31464, 31464]) Position ids shape: torch.Size([1, 31464]) Input IDs shape: torch.Size([1, 31464]) Labels shape: torch.Size([1, 31464]) Final batch size: 1, sequence length: 32786 Attention mask shape: torch.Size([1, 1, 32786, 32786]) Position ids shape: torch.Size([1, 32786]) Input IDs shape: torch.Size([1, 32786]) Labels shape: torch.Size([1, 32786]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 33601 Attention mask shape: torch.Size([1, 1, 33601, 33601]) Position ids shape: torch.Size([1, 33601]) Input IDs shape: torch.Size([1, 33601]) Labels shape: torch.Size([1, 33601]) Final batch size: 1, sequence length: 9029 Attention mask shape: torch.Size([1, 1, 9029, 9029]) Position ids shape: torch.Size([1, 9029]) Input IDs shape: torch.Size([1, 9029]) Labels shape: torch.Size([1, 9029]) Final batch size: 1, sequence length: 32660 Attention mask shape: torch.Size([1, 1, 32660, 32660]) Position ids shape: torch.Size([1, 32660]) Input IDs shape: torch.Size([1, 32660]) Labels shape: torch.Size([1, 32660]) Final batch size: 1, sequence length: 17914 Attention mask shape: torch.Size([1, 1, 17914, 17914]) Position ids shape: torch.Size([1, 17914]) Input IDs shape: torch.Size([1, 17914]) Labels shape: torch.Size([1, 17914]) Final batch size: 1, sequence length: 20951 Attention mask shape: torch.Size([1, 1, 20951, 20951]) Position ids shape: torch.Size([1, 20951]) Input IDs shape: torch.Size([1, 20951]) Labels shape: torch.Size([1, 20951]) Final batch size: 1, sequence length: 36131 Attention mask shape: torch.Size([1, 1, 36131, 36131]) Position ids shape: torch.Size([1, 36131]) Input IDs shape: torch.Size([1, 36131]) Labels shape: torch.Size([1, 36131]) Final batch size: 1, sequence length: 37555 Attention mask shape: torch.Size([1, 1, 37555, 37555]) Position ids shape: torch.Size([1, 37555]) Input IDs shape: torch.Size([1, 37555]) Labels shape: torch.Size([1, 37555]) Final batch size: 1, sequence length: 10132 Attention mask shape: torch.Size([1, 1, 10132, 10132]) Position ids shape: torch.Size([1, 10132]) Input IDs shape: torch.Size([1, 10132]) Labels shape: torch.Size([1, 10132]) Final batch size: 1, sequence length: 15513 Attention mask shape: torch.Size([1, 1, 15513, 15513]) Position ids shape: torch.Size([1, 15513]) Input IDs shape: torch.Size([1, 15513]) Labels shape: torch.Size([1, 15513]) Final batch size: 1, sequence length: 37016 Attention mask shape: torch.Size([1, 1, 37016, 37016]) Position ids shape: torch.Size([1, 37016]) Input IDs shape: torch.Size([1, 37016]) Labels shape: torch.Size([1, 37016]) Final batch size: 1, sequence length: 36970 Attention mask shape: torch.Size([1, 1, 36970, 36970]) Position ids shape: torch.Size([1, 36970]) Input IDs shape: torch.Size([1, 36970]) Labels shape: torch.Size([1, 36970]) Final batch size: 1, sequence length: 35411 Attention mask shape: torch.Size([1, 1, 35411, 35411]) Position ids shape: torch.Size([1, 35411]) Input IDs shape: torch.Size([1, 35411]) Labels shape: torch.Size([1, 35411]) Final batch size: 1, sequence length: 36124 Attention mask shape: torch.Size([1, 1, 36124, 36124]) Position ids shape: torch.Size([1, 36124]) Input IDs shape: torch.Size([1, 36124]) Labels shape: torch.Size([1, 36124]) Final batch size: 1, sequence length: 26562 Attention mask shape: torch.Size([1, 1, 26562, 26562]) Position ids shape: torch.Size([1, 26562]) Input IDs shape: torch.Size([1, 26562]) Labels shape: torch.Size([1, 26562]) Final batch size: 1, sequence length: 20533 Attention mask shape: torch.Size([1, 1, 20533, 20533]) Position ids shape: torch.Size([1, 20533]) Input IDs shape: torch.Size([1, 20533]) Labels shape: torch.Size([1, 20533]) Final batch size: 1, sequence length: 35525 Attention mask shape: torch.Size([1, 1, 35525, 35525]) Position ids shape: torch.Size([1, 35525]) Input IDs shape: torch.Size([1, 35525]) Labels shape: torch.Size([1, 35525]) Final batch size: 1, sequence length: 38712 Attention mask shape: torch.Size([1, 1, 38712, 38712]) Position ids shape: torch.Size([1, 38712]) Input IDs shape: torch.Size([1, 38712]) Labels shape: torch.Size([1, 38712]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25529 Attention mask shape: torch.Size([1, 1, 25529, 25529]) Position ids shape: torch.Size([1, 25529]) Input IDs shape: torch.Size([1, 25529]) Labels shape: torch.Size([1, 25529]) Final batch size: 1, sequence length: 14372 Attention mask shape: torch.Size([1, 1, 14372, 14372]) Position ids shape: torch.Size([1, 14372]) Input IDs shape: torch.Size([1, 14372]) Labels shape: torch.Size([1, 14372]) Final batch size: 1, sequence length: 36292 Attention mask shape: torch.Size([1, 1, 36292, 36292]) Position ids shape: torch.Size([1, 36292]) Input IDs shape: torch.Size([1, 36292]) Labels shape: torch.Size([1, 36292]) Final batch size: 1, sequence length: 26781 Attention mask shape: torch.Size([1, 1, 26781, 26781]) Position ids shape: torch.Size([1, 26781]) Input IDs shape: torch.Size([1, 26781]) Labels shape: torch.Size([1, 26781]) Final batch size: 1, sequence length: 28879 Attention mask shape: torch.Size([1, 1, 28879, 28879]) Position ids shape: torch.Size([1, 28879]) Input IDs shape: torch.Size([1, 28879]) Labels shape: torch.Size([1, 28879]) Final batch size: 1, sequence length: 31084 Attention mask shape: torch.Size([1, 1, 31084, 31084]) Position ids shape: torch.Size([1, 31084]) Input IDs shape: torch.Size([1, 31084]) Labels shape: torch.Size([1, 31084]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 38071 Attention mask shape: torch.Size([1, 1, 38071, 38071]) Position ids shape: torch.Size([1, 38071]) Input IDs shape: torch.Size([1, 38071]) Labels shape: torch.Size([1, 38071]) Final batch size: 1, sequence length: 33422 Attention mask shape: torch.Size([1, 1, 33422, 33422]) Position ids shape: torch.Size([1, 33422]) Input IDs shape: torch.Size([1, 33422]) Labels shape: torch.Size([1, 33422]) Final batch size: 1, sequence length: 30009 Attention mask shape: torch.Size([1, 1, 30009, 30009]) Position ids shape: torch.Size([1, 30009]) Input IDs shape: torch.Size([1, 30009]) Labels shape: torch.Size([1, 30009]) Final batch size: 1, sequence length: 35775 Attention mask shape: torch.Size([1, 1, 35775, 35775]) Position ids shape: torch.Size([1, 35775]) Input IDs shape: torch.Size([1, 35775]) Labels shape: torch.Size([1, 35775]) Final batch size: 1, sequence length: 36185 Attention mask shape: torch.Size([1, 1, 36185, 36185]) Position ids shape: torch.Size([1, 36185]) Input IDs shape: torch.Size([1, 36185]) Labels shape: torch.Size([1, 36185]) Final batch size: 1, sequence length: 38210 Attention mask shape: torch.Size([1, 1, 38210, 38210]) Position ids shape: torch.Size([1, 38210]) Input IDs shape: torch.Size([1, 38210]) Labels shape: torch.Size([1, 38210]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16409 Attention mask shape: torch.Size([1, 1, 16409, 16409]) Position ids shape: torch.Size([1, 16409]) Input IDs shape: torch.Size([1, 16409]) Labels shape: torch.Size([1, 16409]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40753 Attention mask shape: torch.Size([1, 1, 40753, 40753]) Position ids shape: torch.Size([1, 40753]) Input IDs shape: torch.Size([1, 40753]) Labels shape: torch.Size([1, 40753]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31042 Attention mask shape: torch.Size([1, 1, 31042, 31042]) Position ids shape: torch.Size([1, 31042]) Input IDs shape: torch.Size([1, 31042]) Labels shape: torch.Size([1, 31042]) Final batch size: 1, sequence length: 36534 Attention mask shape: torch.Size([1, 1, 36534, 36534]) Position ids shape: torch.Size([1, 36534]) Input IDs shape: torch.Size([1, 36534]) Labels shape: torch.Size([1, 36534]) Final batch size: 1, sequence length: 15031 Attention mask shape: torch.Size([1, 1, 15031, 15031]) Position ids shape: torch.Size([1, 15031]) Input IDs shape: torch.Size([1, 15031]) Labels shape: torch.Size([1, 15031]) Final batch size: 1, sequence length: 37159 Attention mask shape: torch.Size([1, 1, 37159, 37159]) Position ids shape: torch.Size([1, 37159]) Input IDs shape: torch.Size([1, 37159]) Labels shape: torch.Size([1, 37159]) Final batch size: 1, sequence length: 30653 Attention mask shape: torch.Size([1, 1, 30653, 30653]) Position ids shape: torch.Size([1, 30653]) Input IDs shape: torch.Size([1, 30653]) Labels shape: torch.Size([1, 30653]) Final batch size: 1, sequence length: 18884 Attention mask shape: torch.Size([1, 1, 18884, 18884]) Position ids shape: torch.Size([1, 18884]) Input IDs shape: torch.Size([1, 18884]) Labels shape: torch.Size([1, 18884]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20307 Attention mask shape: torch.Size([1, 1, 20307, 20307]) Position ids shape: torch.Size([1, 20307]) Input IDs shape: torch.Size([1, 20307]) Labels shape: torch.Size([1, 20307]) Final batch size: 1, sequence length: 25850 Attention mask shape: torch.Size([1, 1, 25850, 25850]) Position ids shape: torch.Size([1, 25850]) Input IDs shape: torch.Size([1, 25850]) Labels shape: torch.Size([1, 25850]) Final batch size: 1, sequence length: 31448 Attention mask shape: torch.Size([1, 1, 31448, 31448]) Position ids shape: torch.Size([1, 31448]) Input IDs shape: torch.Size([1, 31448]) Labels shape: torch.Size([1, 31448]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7448 Attention mask shape: torch.Size([1, 1, 7448, 7448]) Position ids shape: torch.Size([1, 7448]) Input IDs shape: torch.Size([1, 7448]) Labels shape: torch.Size([1, 7448]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40605 Attention mask shape: torch.Size([1, 1, 40605, 40605]) Position ids shape: torch.Size([1, 40605]) Input IDs shape: torch.Size([1, 40605]) Labels shape: torch.Size([1, 40605]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 5405 Attention mask shape: torch.Size([1, 1, 5405, 5405]) Position ids shape: torch.Size([1, 5405]) Input IDs shape: torch.Size([1, 5405]) Labels shape: torch.Size([1, 5405]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2912, 'grad_norm': 0.35285215146954907, 'learning_rate': 8.345653031794292e-06, 'num_tokens': -inf, 'epoch': 2.62} Final batch size: 1, sequence length: 4641 Attention mask shape: torch.Size([1, 1, 4641, 4641]) Position ids shape: torch.Size([1, 4641]) Input IDs shape: torch.Size([1, 4641]) Labels shape: torch.Size([1, 4641]) Final batch size: 1, sequence length: 4223 Attention mask shape: torch.Size([1, 1, 4223, 4223]) Position ids shape: torch.Size([1, 4223]) Input IDs shape: torch.Size([1, 4223]) Labels shape: torch.Size([1, 4223]) Final batch size: 1, sequence length: 7795 Final batch size: 1, sequence length: 5754 Attention mask shape: torch.Size([1, 1, 7795, 7795]) Position ids shape: torch.Size([1, 7795]) Input IDs shape: torch.Size([1, 7795]) Labels shape: torch.Size([1, 7795]) Attention mask shape: torch.Size([1, 1, 5754, 5754]) Position ids shape: torch.Size([1, 5754]) Input IDs shape: torch.Size([1, 5754]) Labels shape: torch.Size([1, 5754]) Final batch size: 1, sequence length: 8177 Attention mask shape: torch.Size([1, 1, 8177, 8177]) Position ids shape: torch.Size([1, 8177]) Input IDs shape: torch.Size([1, 8177]) Labels shape: torch.Size([1, 8177]) Final batch size: 1, sequence length: 9021 Attention mask shape: torch.Size([1, 1, 9021, 9021]) Position ids shape: torch.Size([1, 9021]) Input IDs shape: torch.Size([1, 9021]) Labels shape: torch.Size([1, 9021]) Final batch size: 1, sequence length: 12501 Attention mask shape: torch.Size([1, 1, 12501, 12501]) Position ids shape: torch.Size([1, 12501]) Input IDs shape: torch.Size([1, 12501]) Labels shape: torch.Size([1, 12501]) Final batch size: 1, sequence length: 9027 Attention mask shape: torch.Size([1, 1, 9027, 9027]) Position ids shape: torch.Size([1, 9027]) Input IDs shape: torch.Size([1, 9027]) Labels shape: torch.Size([1, 9027]) Final batch size: 1, sequence length: 13186 Attention mask shape: torch.Size([1, 1, 13186, 13186]) Position ids shape: torch.Size([1, 13186]) Input IDs shape: torch.Size([1, 13186]) Labels shape: torch.Size([1, 13186]) Final batch size: 1, sequence length: 15283 Attention mask shape: torch.Size([1, 1, 15283, 15283]) Position ids shape: torch.Size([1, 15283]) Input IDs shape: torch.Size([1, 15283]) Labels shape: torch.Size([1, 15283]) Final batch size: 1, sequence length: 16330 Attention mask shape: torch.Size([1, 1, 16330, 16330]) Position ids shape: torch.Size([1, 16330]) Input IDs shape: torch.Size([1, 16330]) Labels shape: torch.Size([1, 16330]) Final batch size: 1, sequence length: 11974 Attention mask shape: torch.Size([1, 1, 11974, 11974]) Position ids shape: torch.Size([1, 11974]) Input IDs shape: torch.Size([1, 11974]) Labels shape: torch.Size([1, 11974]) Final batch size: 1, sequence length: 13957 Attention mask shape: torch.Size([1, 1, 13957, 13957]) Position ids shape: torch.Size([1, 13957]) Input IDs shape: torch.Size([1, 13957]) Labels shape: torch.Size([1, 13957]) Final batch size: 1, sequence length: 16496 Attention mask shape: torch.Size([1, 1, 16496, 16496]) Position ids shape: torch.Size([1, 16496]) Input IDs shape: torch.Size([1, 16496]) Labels shape: torch.Size([1, 16496]) Final batch size: 1, sequence length: 18958 Attention mask shape: torch.Size([1, 1, 18958, 18958]) Position ids shape: torch.Size([1, 18958]) Input IDs shape: torch.Size([1, 18958]) Labels shape: torch.Size([1, 18958]) Final batch size: 1, sequence length: 16366 Attention mask shape: torch.Size([1, 1, 16366, 16366]) Position ids shape: torch.Size([1, 16366]) Input IDs shape: torch.Size([1, 16366]) Labels shape: torch.Size([1, 16366]) Final batch size: 1, sequence length: 16104 Attention mask shape: torch.Size([1, 1, 16104, 16104]) Position ids shape: torch.Size([1, 16104]) Input IDs shape: torch.Size([1, 16104]) Labels shape: torch.Size([1, 16104]) Final batch size: 1, sequence length: 18415 Attention mask shape: torch.Size([1, 1, 18415, 18415]) Position ids shape: torch.Size([1, 18415]) Input IDs shape: torch.Size([1, 18415]) Labels shape: torch.Size([1, 18415]) Final batch size: 1, sequence length: 16088 Attention mask shape: torch.Size([1, 1, 16088, 16088]) Position ids shape: torch.Size([1, 16088]) Input IDs shape: torch.Size([1, 16088]) Labels shape: torch.Size([1, 16088]) Final batch size: 1, sequence length: 14492 Attention mask shape: torch.Size([1, 1, 14492, 14492]) Position ids shape: torch.Size([1, 14492]) Input IDs shape: torch.Size([1, 14492]) Labels shape: torch.Size([1, 14492]) Final batch size: 1, sequence length: 18938 Attention mask shape: torch.Size([1, 1, 18938, 18938]) Position ids shape: torch.Size([1, 18938]) Input IDs shape: torch.Size([1, 18938]) Labels shape: torch.Size([1, 18938]) Final batch size: 1, sequence length: 16014 Attention mask shape: torch.Size([1, 1, 16014, 16014]) Position ids shape: torch.Size([1, 16014]) Input IDs shape: torch.Size([1, 16014]) Labels shape: torch.Size([1, 16014]) Final batch size: 1, sequence length: 17092 Attention mask shape: torch.Size([1, 1, 17092, 17092]) Position ids shape: torch.Size([1, 17092]) Input IDs shape: torch.Size([1, 17092]) Labels shape: torch.Size([1, 17092]) Final batch size: 1, sequence length: 19603 Attention mask shape: torch.Size([1, 1, 19603, 19603]) Position ids shape: torch.Size([1, 19603]) Input IDs shape: torch.Size([1, 19603]) Labels shape: torch.Size([1, 19603]) Final batch size: 1, sequence length: 21132 Attention mask shape: torch.Size([1, 1, 21132, 21132]) Position ids shape: torch.Size([1, 21132]) Input IDs shape: torch.Size([1, 21132]) Labels shape: torch.Size([1, 21132]) Final batch size: 1, sequence length: 19034 Attention mask shape: torch.Size([1, 1, 19034, 19034]) Position ids shape: torch.Size([1, 19034]) Input IDs shape: torch.Size([1, 19034]) Labels shape: torch.Size([1, 19034]) Final batch size: 1, sequence length: 13468 Attention mask shape: torch.Size([1, 1, 13468, 13468]) Position ids shape: torch.Size([1, 13468]) Input IDs shape: torch.Size([1, 13468]) Labels shape: torch.Size([1, 13468]) Final batch size: 1, sequence length: 18034 Attention mask shape: torch.Size([1, 1, 18034, 18034]) Position ids shape: torch.Size([1, 18034]) Input IDs shape: torch.Size([1, 18034]) Labels shape: torch.Size([1, 18034]) Final batch size: 1, sequence length: 23132 Attention mask shape: torch.Size([1, 1, 23132, 23132]) Position ids shape: torch.Size([1, 23132]) Input IDs shape: torch.Size([1, 23132]) Labels shape: torch.Size([1, 23132]) Final batch size: 1, sequence length: 20726 Attention mask shape: torch.Size([1, 1, 20726, 20726]) Position ids shape: torch.Size([1, 20726]) Input IDs shape: torch.Size([1, 20726]) Labels shape: torch.Size([1, 20726]) Final batch size: 1, sequence length: 21705 Attention mask shape: torch.Size([1, 1, 21705, 21705]) Position ids shape: torch.Size([1, 21705]) Input IDs shape: torch.Size([1, 21705]) Labels shape: torch.Size([1, 21705]) Final batch size: 1, sequence length: 21404 Attention mask shape: torch.Size([1, 1, 21404, 21404]) Position ids shape: torch.Size([1, 21404]) Input IDs shape: torch.Size([1, 21404]) Labels shape: torch.Size([1, 21404]) Final batch size: 1, sequence length: 23102 Attention mask shape: torch.Size([1, 1, 23102, 23102]) Position ids shape: torch.Size([1, 23102]) Input IDs shape: torch.Size([1, 23102]) Labels shape: torch.Size([1, 23102]) Final batch size: 1, sequence length: 23971 Attention mask shape: torch.Size([1, 1, 23971, 23971]) Position ids shape: torch.Size([1, 23971]) Input IDs shape: torch.Size([1, 23971]) Labels shape: torch.Size([1, 23971]) Final batch size: 1, sequence length: 24432 Attention mask shape: torch.Size([1, 1, 24432, 24432]) Position ids shape: torch.Size([1, 24432]) Input IDs shape: torch.Size([1, 24432]) Labels shape: torch.Size([1, 24432]) Final batch size: 1, sequence length: 12756 Attention mask shape: torch.Size([1, 1, 12756, 12756]) Position ids shape: torch.Size([1, 12756]) Input IDs shape: torch.Size([1, 12756]) Labels shape: torch.Size([1, 12756]) Final batch size: 1, sequence length: 16291 Attention mask shape: torch.Size([1, 1, 16291, 16291]) Position ids shape: torch.Size([1, 16291]) Input IDs shape: torch.Size([1, 16291]) Labels shape: torch.Size([1, 16291]) Final batch size: 1, sequence length: 22771 Attention mask shape: torch.Size([1, 1, 22771, 22771]) Position ids shape: torch.Size([1, 22771]) Input IDs shape: torch.Size([1, 22771]) Labels shape: torch.Size([1, 22771]) Final batch size: 1, sequence length: 17514 Attention mask shape: torch.Size([1, 1, 17514, 17514]) Position ids shape: torch.Size([1, 17514]) Input IDs shape: torch.Size([1, 17514]) Labels shape: torch.Size([1, 17514]) Final batch size: 1, sequence length: 23894 Attention mask shape: torch.Size([1, 1, 23894, 23894]) Position ids shape: torch.Size([1, 23894]) Input IDs shape: torch.Size([1, 23894]) Labels shape: torch.Size([1, 23894]) Final batch size: 1, sequence length: 16590 Attention mask shape: torch.Size([1, 1, 16590, 16590]) Position ids shape: torch.Size([1, 16590]) Input IDs shape: torch.Size([1, 16590]) Labels shape: torch.Size([1, 16590]) Final batch size: 1, sequence length: 25351 Attention mask shape: torch.Size([1, 1, 25351, 25351]) Position ids shape: torch.Size([1, 25351]) Input IDs shape: torch.Size([1, 25351]) Labels shape: torch.Size([1, 25351]) Final batch size: 1, sequence length: 26375 Attention mask shape: torch.Size([1, 1, 26375, 26375]) Position ids shape: torch.Size([1, 26375]) Input IDs shape: torch.Size([1, 26375]) Labels shape: torch.Size([1, 26375]) Final batch size: 1, sequence length: 27780 Attention mask shape: torch.Size([1, 1, 27780, 27780]) Position ids shape: torch.Size([1, 27780]) Input IDs shape: torch.Size([1, 27780]) Labels shape: torch.Size([1, 27780]) Final batch size: 1, sequence length: 27785 Attention mask shape: torch.Size([1, 1, 27785, 27785]) Position ids shape: torch.Size([1, 27785]) Input IDs shape: torch.Size([1, 27785]) Labels shape: torch.Size([1, 27785]) Final batch size: 1, sequence length: 13790 Attention mask shape: torch.Size([1, 1, 13790, 13790]) Position ids shape: torch.Size([1, 13790]) Input IDs shape: torch.Size([1, 13790]) Labels shape: torch.Size([1, 13790]) Final batch size: 1, sequence length: 26596 Attention mask shape: torch.Size([1, 1, 26596, 26596]) Position ids shape: torch.Size([1, 26596]) Input IDs shape: torch.Size([1, 26596]) Labels shape: torch.Size([1, 26596]) Final batch size: 1, sequence length: 20817 Attention mask shape: torch.Size([1, 1, 20817, 20817]) Position ids shape: torch.Size([1, 20817]) Input IDs shape: torch.Size([1, 20817]) Labels shape: torch.Size([1, 20817]) Final batch size: 1, sequence length: 26766 Attention mask shape: torch.Size([1, 1, 26766, 26766]) Position ids shape: torch.Size([1, 26766]) Input IDs shape: torch.Size([1, 26766]) Labels shape: torch.Size([1, 26766]) Final batch size: 1, sequence length: 28412 Attention mask shape: torch.Size([1, 1, 28412, 28412]) Position ids shape: torch.Size([1, 28412]) Input IDs shape: torch.Size([1, 28412]) Labels shape: torch.Size([1, 28412]) Final batch size: 1, sequence length: 28574 Attention mask shape: torch.Size([1, 1, 28574, 28574]) Position ids shape: torch.Size([1, 28574]) Input IDs shape: torch.Size([1, 28574]) Labels shape: torch.Size([1, 28574]) Final batch size: 1, sequence length: 32022 Attention mask shape: torch.Size([1, 1, 32022, 32022]) Position ids shape: torch.Size([1, 32022]) Input IDs shape: torch.Size([1, 32022]) Labels shape: torch.Size([1, 32022]) Final batch size: 1, sequence length: 18869 Attention mask shape: torch.Size([1, 1, 18869, 18869]) Position ids shape: torch.Size([1, 18869]) Input IDs shape: torch.Size([1, 18869]) Labels shape: torch.Size([1, 18869]) Final batch size: 1, sequence length: 25388 Attention mask shape: torch.Size([1, 1, 25388, 25388]) Position ids shape: torch.Size([1, 25388]) Input IDs shape: torch.Size([1, 25388]) Labels shape: torch.Size([1, 25388]) Final batch size: 1, sequence length: 30165 Attention mask shape: torch.Size([1, 1, 30165, 30165]) Position ids shape: torch.Size([1, 30165]) Input IDs shape: torch.Size([1, 30165]) Labels shape: torch.Size([1, 30165]) Final batch size: 1, sequence length: 29132 Attention mask shape: torch.Size([1, 1, 29132, 29132]) Position ids shape: torch.Size([1, 29132]) Input IDs shape: torch.Size([1, 29132]) Labels shape: torch.Size([1, 29132]) Final batch size: 1, sequence length: 29168 Attention mask shape: torch.Size([1, 1, 29168, 29168]) Position ids shape: torch.Size([1, 29168]) Input IDs shape: torch.Size([1, 29168]) Labels shape: torch.Size([1, 29168]) Final batch size: 1, sequence length: 34326 Attention mask shape: torch.Size([1, 1, 34326, 34326]) Position ids shape: torch.Size([1, 34326]) Input IDs shape: torch.Size([1, 34326]) Labels shape: torch.Size([1, 34326]) Final batch size: 1, sequence length: 23810 Attention mask shape: torch.Size([1, 1, 23810, 23810]) Position ids shape: torch.Size([1, 23810]) Input IDs shape: torch.Size([1, 23810]) Labels shape: torch.Size([1, 23810]) Final batch size: 1, sequence length: 29830 Attention mask shape: torch.Size([1, 1, 29830, 29830]) Position ids shape: torch.Size([1, 29830]) Input IDs shape: torch.Size([1, 29830]) Labels shape: torch.Size([1, 29830]) Final batch size: 1, sequence length: 34457 Attention mask shape: torch.Size([1, 1, 34457, 34457]) Position ids shape: torch.Size([1, 34457]) Input IDs shape: torch.Size([1, 34457]) Labels shape: torch.Size([1, 34457]) Final batch size: 1, sequence length: 28845 Attention mask shape: torch.Size([1, 1, 28845, 28845]) Position ids shape: torch.Size([1, 28845]) Input IDs shape: torch.Size([1, 28845]) Labels shape: torch.Size([1, 28845]) Final batch size: 1, sequence length: 22043 Attention mask shape: torch.Size([1, 1, 22043, 22043]) Position ids shape: torch.Size([1, 22043]) Input IDs shape: torch.Size([1, 22043]) Labels shape: torch.Size([1, 22043]) Final batch size: 1, sequence length: 24653 Attention mask shape: torch.Size([1, 1, 24653, 24653]) Position ids shape: torch.Size([1, 24653]) Input IDs shape: torch.Size([1, 24653]) Labels shape: torch.Size([1, 24653]) Final batch size: 1, sequence length: 22328 Attention mask shape: torch.Size([1, 1, 22328, 22328]) Position ids shape: torch.Size([1, 22328]) Input IDs shape: torch.Size([1, 22328]) Labels shape: torch.Size([1, 22328]) Final batch size: 1, sequence length: 35790 Attention mask shape: torch.Size([1, 1, 35790, 35790]) Position ids shape: torch.Size([1, 35790]) Input IDs shape: torch.Size([1, 35790]) Labels shape: torch.Size([1, 35790]) Final batch size: 1, sequence length: 26465 Attention mask shape: torch.Size([1, 1, 26465, 26465]) Position ids shape: torch.Size([1, 26465]) Input IDs shape: torch.Size([1, 26465]) Labels shape: torch.Size([1, 26465]) Final batch size: 1, sequence length: 34874 Attention mask shape: torch.Size([1, 1, 34874, 34874]) Position ids shape: torch.Size([1, 34874]) Input IDs shape: torch.Size([1, 34874]) Labels shape: torch.Size([1, 34874]) Final batch size: 1, sequence length: 38919 Attention mask shape: torch.Size([1, 1, 38919, 38919]) Position ids shape: torch.Size([1, 38919]) Input IDs shape: torch.Size([1, 38919]) Labels shape: torch.Size([1, 38919]) Final batch size: 1, sequence length: 13986 Attention mask shape: torch.Size([1, 1, 13986, 13986]) Position ids shape: torch.Size([1, 13986]) Input IDs shape: torch.Size([1, 13986]) Labels shape: torch.Size([1, 13986]) Final batch size: 1, sequence length: 38617 Attention mask shape: torch.Size([1, 1, 38617, 38617]) Position ids shape: torch.Size([1, 38617]) Input IDs shape: torch.Size([1, 38617]) Labels shape: torch.Size([1, 38617]) Final batch size: 1, sequence length: 22014 Attention mask shape: torch.Size([1, 1, 22014, 22014]) Position ids shape: torch.Size([1, 22014]) Input IDs shape: torch.Size([1, 22014]) Labels shape: torch.Size([1, 22014]) Final batch size: 1, sequence length: 37039 Attention mask shape: torch.Size([1, 1, 37039, 37039]) Position ids shape: torch.Size([1, 37039]) Input IDs shape: torch.Size([1, 37039]) Labels shape: torch.Size([1, 37039]) Final batch size: 1, sequence length: 19204 Attention mask shape: torch.Size([1, 1, 19204, 19204]) Position ids shape: torch.Size([1, 19204]) Input IDs shape: torch.Size([1, 19204]) Labels shape: torch.Size([1, 19204]) Final batch size: 1, sequence length: 19744 Attention mask shape: torch.Size([1, 1, 19744, 19744]) Position ids shape: torch.Size([1, 19744]) Input IDs shape: torch.Size([1, 19744]) Labels shape: torch.Size([1, 19744]) Final batch size: 1, sequence length: 35868 Attention mask shape: torch.Size([1, 1, 35868, 35868]) Position ids shape: torch.Size([1, 35868]) Input IDs shape: torch.Size([1, 35868]) Labels shape: torch.Size([1, 35868]) Final batch size: 1, sequence length: 39150 Attention mask shape: torch.Size([1, 1, 39150, 39150]) Position ids shape: torch.Size([1, 39150]) Input IDs shape: torch.Size([1, 39150]) Labels shape: torch.Size([1, 39150]) Final batch size: 1, sequence length: 36794 Attention mask shape: torch.Size([1, 1, 36794, 36794]) Position ids shape: torch.Size([1, 36794]) Input IDs shape: torch.Size([1, 36794]) Labels shape: torch.Size([1, 36794]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40717 Attention mask shape: torch.Size([1, 1, 40717, 40717]) Position ids shape: torch.Size([1, 40717]) Input IDs shape: torch.Size([1, 40717]) Labels shape: torch.Size([1, 40717]) Final batch size: 1, sequence length: 35999 Attention mask shape: torch.Size([1, 1, 35999, 35999]) Position ids shape: torch.Size([1, 35999]) Input IDs shape: torch.Size([1, 35999]) Labels shape: torch.Size([1, 35999]) Final batch size: 1, sequence length: 35803 Attention mask shape: torch.Size([1, 1, 35803, 35803]) Position ids shape: torch.Size([1, 35803]) Input IDs shape: torch.Size([1, 35803]) Labels shape: torch.Size([1, 35803]) Final batch size: 1, sequence length: 22282 Attention mask shape: torch.Size([1, 1, 22282, 22282]) Position ids shape: torch.Size([1, 22282]) Input IDs shape: torch.Size([1, 22282]) Labels shape: torch.Size([1, 22282]) Final batch size: 1, sequence length: 23245 Attention mask shape: torch.Size([1, 1, 23245, 23245]) Position ids shape: torch.Size([1, 23245]) Input IDs shape: torch.Size([1, 23245]) Labels shape: torch.Size([1, 23245]) Final batch size: 1, sequence length: 31506 Attention mask shape: torch.Size([1, 1, 31506, 31506]) Position ids shape: torch.Size([1, 31506]) Input IDs shape: torch.Size([1, 31506]) Labels shape: torch.Size([1, 31506]) Final batch size: 1, sequence length: 32974 Attention mask shape: torch.Size([1, 1, 32974, 32974]) Position ids shape: torch.Size([1, 32974]) Input IDs shape: torch.Size([1, 32974]) Labels shape: torch.Size([1, 32974]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28002 Attention mask shape: torch.Size([1, 1, 28002, 28002]) Position ids shape: torch.Size([1, 28002]) Input IDs shape: torch.Size([1, 28002]) Labels shape: torch.Size([1, 28002]) Final batch size: 1, sequence length: 39919 Attention mask shape: torch.Size([1, 1, 39919, 39919]) Position ids shape: torch.Size([1, 39919]) Input IDs shape: torch.Size([1, 39919]) Labels shape: torch.Size([1, 39919]) Final batch size: 1, sequence length: 28781 Attention mask shape: torch.Size([1, 1, 28781, 28781]) Position ids shape: torch.Size([1, 28781]) Input IDs shape: torch.Size([1, 28781]) Labels shape: torch.Size([1, 28781]) Final batch size: 1, sequence length: 12418 Attention mask shape: torch.Size([1, 1, 12418, 12418]) Position ids shape: torch.Size([1, 12418]) Input IDs shape: torch.Size([1, 12418]) Labels shape: torch.Size([1, 12418]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31340 Attention mask shape: torch.Size([1, 1, 31340, 31340]) Position ids shape: torch.Size([1, 31340]) Input IDs shape: torch.Size([1, 31340]) Labels shape: torch.Size([1, 31340]) Final batch size: 1, sequence length: 11644 Attention mask shape: torch.Size([1, 1, 11644, 11644]) Position ids shape: torch.Size([1, 11644]) Input IDs shape: torch.Size([1, 11644]) Labels shape: torch.Size([1, 11644]) Final batch size: 1, sequence length: 40065 Attention mask shape: torch.Size([1, 1, 40065, 40065]) Position ids shape: torch.Size([1, 40065]) Input IDs shape: torch.Size([1, 40065]) Labels shape: torch.Size([1, 40065]) Final batch size: 1, sequence length: 19846 Attention mask shape: torch.Size([1, 1, 19846, 19846]) Position ids shape: torch.Size([1, 19846]) Input IDs shape: torch.Size([1, 19846]) Labels shape: torch.Size([1, 19846]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31359 Attention mask shape: torch.Size([1, 1, 31359, 31359]) Position ids shape: torch.Size([1, 31359]) Input IDs shape: torch.Size([1, 31359]) Labels shape: torch.Size([1, 31359]) Final batch size: 1, sequence length: 18379 Attention mask shape: torch.Size([1, 1, 18379, 18379]) Position ids shape: torch.Size([1, 18379]) Input IDs shape: torch.Size([1, 18379]) Labels shape: torch.Size([1, 18379]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7364 Attention mask shape: torch.Size([1, 1, 7364, 7364]) Position ids shape: torch.Size([1, 7364]) Input IDs shape: torch.Size([1, 7364]) Labels shape: torch.Size([1, 7364]) Final batch size: 1, sequence length: 40874 Attention mask shape: torch.Size([1, 1, 40874, 40874]) Position ids shape: torch.Size([1, 40874]) Input IDs shape: torch.Size([1, 40874]) Labels shape: torch.Size([1, 40874]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 9309 Attention mask shape: torch.Size([1, 1, 9309, 9309]) Position ids shape: torch.Size([1, 9309]) Input IDs shape: torch.Size([1, 9309]) Labels shape: torch.Size([1, 9309]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 25560 Attention mask shape: torch.Size([1, 1, 25560, 25560]) Position ids shape: torch.Size([1, 25560]) Input IDs shape: torch.Size([1, 25560]) Labels shape: torch.Size([1, 25560]) Final batch size: 1, sequence length: 26221 Attention mask shape: torch.Size([1, 1, 26221, 26221]) Position ids shape: torch.Size([1, 26221]) Input IDs shape: torch.Size([1, 26221]) Labels shape: torch.Size([1, 26221]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25923 Attention mask shape: torch.Size([1, 1, 25923, 25923]) Position ids shape: torch.Size([1, 25923]) Input IDs shape: torch.Size([1, 25923]) Labels shape: torch.Size([1, 25923]) Final batch size: 1, sequence length: 17224 Attention mask shape: torch.Size([1, 1, 17224, 17224]) Position ids shape: torch.Size([1, 17224]) Input IDs shape: torch.Size([1, 17224]) Labels shape: torch.Size([1, 17224]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 18654 Attention mask shape: torch.Size([1, 1, 18654, 18654]) Position ids shape: torch.Size([1, 18654]) Input IDs shape: torch.Size([1, 18654]) Labels shape: torch.Size([1, 18654]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25820 Attention mask shape: torch.Size([1, 1, 25820, 25820]) Position ids shape: torch.Size([1, 25820]) Input IDs shape: torch.Size([1, 25820]) Labels shape: torch.Size([1, 25820]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30709 Attention mask shape: torch.Size([1, 1, 30709, 30709]) Position ids shape: torch.Size([1, 30709]) Input IDs shape: torch.Size([1, 30709]) Labels shape: torch.Size([1, 30709]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 29526 Attention mask shape: torch.Size([1, 1, 29526, 29526]) Position ids shape: torch.Size([1, 29526]) Input IDs shape: torch.Size([1, 29526]) Labels shape: torch.Size([1, 29526]) {'loss': 0.2996, 'grad_norm': 0.3130928289727393, 'learning_rate': 8.146601955249187e-06, 'num_tokens': -inf, 'epoch': 2.75} Final batch size: 1, sequence length: 7461 Attention mask shape: torch.Size([1, 1, 7461, 7461]) Position ids shape: torch.Size([1, 7461]) Input IDs shape: torch.Size([1, 7461]) Labels shape: torch.Size([1, 7461]) Final batch size: 1, sequence length: 12279 Attention mask shape: torch.Size([1, 1, 12279, 12279]) Position ids shape: torch.Size([1, 12279]) Input IDs shape: torch.Size([1, 12279]) Labels shape: torch.Size([1, 12279]) Final batch size: 1, sequence length: 10826 Attention mask shape: torch.Size([1, 1, 10826, 10826]) Position ids shape: torch.Size([1, 10826]) Input IDs shape: torch.Size([1, 10826]) Labels shape: torch.Size([1, 10826]) Final batch size: 1, sequence length: 13427 Attention mask shape: torch.Size([1, 1, 13427, 13427]) Position ids shape: torch.Size([1, 13427]) Input IDs shape: torch.Size([1, 13427]) Labels shape: torch.Size([1, 13427]) Final batch size: 1, sequence length: 9858 Attention mask shape: torch.Size([1, 1, 9858, 9858]) Position ids shape: torch.Size([1, 9858]) Input IDs shape: torch.Size([1, 9858]) Labels shape: torch.Size([1, 9858]) Final batch size: 1, sequence length: 12096 Attention mask shape: torch.Size([1, 1, 12096, 12096]) Position ids shape: torch.Size([1, 12096]) Input IDs shape: torch.Size([1, 12096]) Labels shape: torch.Size([1, 12096]) Final batch size: 1, sequence length: 12960 Attention mask shape: torch.Size([1, 1, 12960, 12960]) Position ids shape: torch.Size([1, 12960]) Input IDs shape: torch.Size([1, 12960]) Labels shape: torch.Size([1, 12960]) Final batch size: 1, sequence length: 12622 Attention mask shape: torch.Size([1, 1, 12622, 12622]) Position ids shape: torch.Size([1, 12622]) Input IDs shape: torch.Size([1, 12622]) Labels shape: torch.Size([1, 12622]) Final batch size: 1, sequence length: 13550 Attention mask shape: torch.Size([1, 1, 13550, 13550]) Position ids shape: torch.Size([1, 13550]) Input IDs shape: torch.Size([1, 13550]) Labels shape: torch.Size([1, 13550]) Final batch size: 1, sequence length: 15933 Attention mask shape: torch.Size([1, 1, 15933, 15933]) Position ids shape: torch.Size([1, 15933]) Input IDs shape: torch.Size([1, 15933]) Labels shape: torch.Size([1, 15933]) Final batch size: 1, sequence length: 15673 Attention mask shape: torch.Size([1, 1, 15673, 15673]) Position ids shape: torch.Size([1, 15673]) Input IDs shape: torch.Size([1, 15673]) Labels shape: torch.Size([1, 15673]) Final batch size: 1, sequence length: 17861 Attention mask shape: torch.Size([1, 1, 17861, 17861]) Position ids shape: torch.Size([1, 17861]) Input IDs shape: torch.Size([1, 17861]) Labels shape: torch.Size([1, 17861]) Final batch size: 1, sequence length: 16927 Attention mask shape: torch.Size([1, 1, 16927, 16927]) Position ids shape: torch.Size([1, 16927]) Input IDs shape: torch.Size([1, 16927]) Labels shape: torch.Size([1, 16927]) Final batch size: 1, sequence length: 17934 Attention mask shape: torch.Size([1, 1, 17934, 17934]) Position ids shape: torch.Size([1, 17934]) Input IDs shape: torch.Size([1, 17934]) Labels shape: torch.Size([1, 17934]) Final batch size: 1, sequence length: 13789 Attention mask shape: torch.Size([1, 1, 13789, 13789]) Position ids shape: torch.Size([1, 13789]) Input IDs shape: torch.Size([1, 13789]) Labels shape: torch.Size([1, 13789]) Final batch size: 1, sequence length: 16450 Attention mask shape: torch.Size([1, 1, 16450, 16450]) Position ids shape: torch.Size([1, 16450]) Input IDs shape: torch.Size([1, 16450]) Labels shape: torch.Size([1, 16450]) Final batch size: 1, sequence length: 18214 Attention mask shape: torch.Size([1, 1, 18214, 18214]) Position ids shape: torch.Size([1, 18214]) Input IDs shape: torch.Size([1, 18214]) Labels shape: torch.Size([1, 18214]) Final batch size: 1, sequence length: 17245 Attention mask shape: torch.Size([1, 1, 17245, 17245]) Position ids shape: torch.Size([1, 17245]) Input IDs shape: torch.Size([1, 17245]) Labels shape: torch.Size([1, 17245]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 19899 Attention mask shape: torch.Size([1, 1, 19899, 19899]) Position ids shape: torch.Size([1, 19899]) Input IDs shape: torch.Size([1, 19899]) Labels shape: torch.Size([1, 19899]) Final batch size: 1, sequence length: 12728 Attention mask shape: torch.Size([1, 1, 12728, 12728]) Position ids shape: torch.Size([1, 12728]) Input IDs shape: torch.Size([1, 12728]) Labels shape: torch.Size([1, 12728]) Final batch size: 1, sequence length: 20065 Attention mask shape: torch.Size([1, 1, 20065, 20065]) Position ids shape: torch.Size([1, 20065]) Input IDs shape: torch.Size([1, 20065]) Labels shape: torch.Size([1, 20065]) Final batch size: 1, sequence length: 21091 Attention mask shape: torch.Size([1, 1, 21091, 21091]) Position ids shape: torch.Size([1, 21091]) Input IDs shape: torch.Size([1, 21091]) Labels shape: torch.Size([1, 21091]) Final batch size: 1, sequence length: 21675 Attention mask shape: torch.Size([1, 1, 21675, 21675]) Position ids shape: torch.Size([1, 21675]) Input IDs shape: torch.Size([1, 21675]) Labels shape: torch.Size([1, 21675]) Final batch size: 1, sequence length: 20292 Attention mask shape: torch.Size([1, 1, 20292, 20292]) Position ids shape: torch.Size([1, 20292]) Input IDs shape: torch.Size([1, 20292]) Labels shape: torch.Size([1, 20292]) Final batch size: 1, sequence length: 23004 Attention mask shape: torch.Size([1, 1, 23004, 23004]) Position ids shape: torch.Size([1, 23004]) Input IDs shape: torch.Size([1, 23004]) Labels shape: torch.Size([1, 23004]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 16036 Attention mask shape: torch.Size([1, 1, 16036, 16036]) Position ids shape: torch.Size([1, 16036]) Input IDs shape: torch.Size([1, 16036]) Labels shape: torch.Size([1, 16036]) Final batch size: 1, sequence length: 25600 Attention mask shape: torch.Size([1, 1, 25600, 25600]) Position ids shape: torch.Size([1, 25600]) Input IDs shape: torch.Size([1, 25600]) Labels shape: torch.Size([1, 25600]) Final batch size: 1, sequence length: 23896 Attention mask shape: torch.Size([1, 1, 23896, 23896]) Position ids shape: torch.Size([1, 23896]) Input IDs shape: torch.Size([1, 23896]) Labels shape: torch.Size([1, 23896]) Final batch size: 1, sequence length: 24633 Attention mask shape: torch.Size([1, 1, 24633, 24633]) Position ids shape: torch.Size([1, 24633]) Input IDs shape: torch.Size([1, 24633]) Labels shape: torch.Size([1, 24633]) Final batch size: 1, sequence length: 24808 Attention mask shape: torch.Size([1, 1, 24808, 24808]) Position ids shape: torch.Size([1, 24808]) Input IDs shape: torch.Size([1, 24808]) Labels shape: torch.Size([1, 24808]) Final batch size: 1, sequence length: 21586 Attention mask shape: torch.Size([1, 1, 21586, 21586]) Position ids shape: torch.Size([1, 21586]) Input IDs shape: torch.Size([1, 21586]) Labels shape: torch.Size([1, 21586]) Final batch size: 1, sequence length: 22775 Attention mask shape: torch.Size([1, 1, 22775, 22775]) Position ids shape: torch.Size([1, 22775]) Input IDs shape: torch.Size([1, 22775]) Labels shape: torch.Size([1, 22775]) Final batch size: 1, sequence length: 24956 Attention mask shape: torch.Size([1, 1, 24956, 24956]) Position ids shape: torch.Size([1, 24956]) Input IDs shape: torch.Size([1, 24956]) Labels shape: torch.Size([1, 24956]) Final batch size: 1, sequence length: 25435 Attention mask shape: torch.Size([1, 1, 25435, 25435]) Position ids shape: torch.Size([1, 25435]) Input IDs shape: torch.Size([1, 25435]) Labels shape: torch.Size([1, 25435]) Final batch size: 1, sequence length: 26316 Attention mask shape: torch.Size([1, 1, 26316, 26316]) Position ids shape: torch.Size([1, 26316]) Input IDs shape: torch.Size([1, 26316]) Labels shape: torch.Size([1, 26316]) Final batch size: 1, sequence length: 14212 Attention mask shape: torch.Size([1, 1, 14212, 14212]) Position ids shape: torch.Size([1, 14212]) Input IDs shape: torch.Size([1, 14212]) Labels shape: torch.Size([1, 14212]) Final batch size: 1, sequence length: 19187 Attention mask shape: torch.Size([1, 1, 19187, 19187]) Position ids shape: torch.Size([1, 19187]) Input IDs shape: torch.Size([1, 19187]) Labels shape: torch.Size([1, 19187]) Final batch size: 1, sequence length: 25035 Attention mask shape: torch.Size([1, 1, 25035, 25035]) Position ids shape: torch.Size([1, 25035]) Input IDs shape: torch.Size([1, 25035]) Labels shape: torch.Size([1, 25035]) Final batch size: 1, sequence length: 17763 Attention mask shape: torch.Size([1, 1, 17763, 17763]) Position ids shape: torch.Size([1, 17763]) Input IDs shape: torch.Size([1, 17763]) Labels shape: torch.Size([1, 17763]) Final batch size: 1, sequence length: 26520 Final batch size: 1, sequence length: 3010 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Attention mask shape: torch.Size([1, 1, 3010, 3010]) Position ids shape: torch.Size([1, 3010]) Input IDs shape: torch.Size([1, 3010]) Labels shape: torch.Size([1, 3010]) Final batch size: 1, sequence length: 24927 Attention mask shape: torch.Size([1, 1, 24927, 24927]) Position ids shape: torch.Size([1, 24927]) Input IDs shape: torch.Size([1, 24927]) Labels shape: torch.Size([1, 24927]) Final batch size: 1, sequence length: 24965 Attention mask shape: torch.Size([1, 1, 24965, 24965]) Position ids shape: torch.Size([1, 24965]) Input IDs shape: torch.Size([1, 24965]) Labels shape: torch.Size([1, 24965]) Final batch size: 1, sequence length: 25176 Attention mask shape: torch.Size([1, 1, 25176, 25176]) Position ids shape: torch.Size([1, 25176]) Input IDs shape: torch.Size([1, 25176]) Labels shape: torch.Size([1, 25176]) Final batch size: 1, sequence length: 22778 Attention mask shape: torch.Size([1, 1, 22778, 22778]) Position ids shape: torch.Size([1, 22778]) Input IDs shape: torch.Size([1, 22778]) Labels shape: torch.Size([1, 22778]) Final batch size: 1, sequence length: 26247 Attention mask shape: torch.Size([1, 1, 26247, 26247]) Position ids shape: torch.Size([1, 26247]) Input IDs shape: torch.Size([1, 26247]) Labels shape: torch.Size([1, 26247]) Final batch size: 1, sequence length: 28926 Attention mask shape: torch.Size([1, 1, 28926, 28926]) Position ids shape: torch.Size([1, 28926]) Input IDs shape: torch.Size([1, 28926]) Labels shape: torch.Size([1, 28926]) Final batch size: 1, sequence length: 25325 Attention mask shape: torch.Size([1, 1, 25325, 25325]) Position ids shape: torch.Size([1, 25325]) Input IDs shape: torch.Size([1, 25325]) Labels shape: torch.Size([1, 25325]) Final batch size: 1, sequence length: 20582 Attention mask shape: torch.Size([1, 1, 20582, 20582]) Position ids shape: torch.Size([1, 20582]) Input IDs shape: torch.Size([1, 20582]) Labels shape: torch.Size([1, 20582]) Final batch size: 1, sequence length: 26500 Attention mask shape: torch.Size([1, 1, 26500, 26500]) Position ids shape: torch.Size([1, 26500]) Input IDs shape: torch.Size([1, 26500]) Labels shape: torch.Size([1, 26500]) Final batch size: 1, sequence length: 27222 Attention mask shape: torch.Size([1, 1, 27222, 27222]) Position ids shape: torch.Size([1, 27222]) Input IDs shape: torch.Size([1, 27222]) Labels shape: torch.Size([1, 27222]) Final batch size: 1, sequence length: 13946 Attention mask shape: torch.Size([1, 1, 13946, 13946]) Position ids shape: torch.Size([1, 13946]) Input IDs shape: torch.Size([1, 13946]) Labels shape: torch.Size([1, 13946]) Final batch size: 1, sequence length: 16714 Attention mask shape: torch.Size([1, 1, 16714, 16714]) Position ids shape: torch.Size([1, 16714]) Input IDs shape: torch.Size([1, 16714]) Labels shape: torch.Size([1, 16714]) Final batch size: 1, sequence length: 28552 Attention mask shape: torch.Size([1, 1, 28552, 28552]) Position ids shape: torch.Size([1, 28552]) Input IDs shape: torch.Size([1, 28552]) Labels shape: torch.Size([1, 28552]) Final batch size: 1, sequence length: 28644 Attention mask shape: torch.Size([1, 1, 28644, 28644]) Position ids shape: torch.Size([1, 28644]) Input IDs shape: torch.Size([1, 28644]) Labels shape: torch.Size([1, 28644]) Final batch size: 1, sequence length: 18832 Attention mask shape: torch.Size([1, 1, 18832, 18832]) Position ids shape: torch.Size([1, 18832]) Input IDs shape: torch.Size([1, 18832]) Labels shape: torch.Size([1, 18832]) Final batch size: 1, sequence length: 6978 Attention mask shape: torch.Size([1, 1, 6978, 6978]) Position ids shape: torch.Size([1, 6978]) Input IDs shape: torch.Size([1, 6978]) Labels shape: torch.Size([1, 6978]) Final batch size: 1, sequence length: 12426 Attention mask shape: torch.Size([1, 1, 12426, 12426]) Position ids shape: torch.Size([1, 12426]) Input IDs shape: torch.Size([1, 12426]) Labels shape: torch.Size([1, 12426]) Final batch size: 1, sequence length: 31208 Attention mask shape: torch.Size([1, 1, 31208, 31208]) Position ids shape: torch.Size([1, 31208]) Input IDs shape: torch.Size([1, 31208]) Labels shape: torch.Size([1, 31208]) Final batch size: 1, sequence length: 13453 Attention mask shape: torch.Size([1, 1, 13453, 13453]) Position ids shape: torch.Size([1, 13453]) Input IDs shape: torch.Size([1, 13453]) Labels shape: torch.Size([1, 13453]) Final batch size: 1, sequence length: 32513 Attention mask shape: torch.Size([1, 1, 32513, 32513]) Position ids shape: torch.Size([1, 32513]) Input IDs shape: torch.Size([1, 32513]) Labels shape: torch.Size([1, 32513]) Final batch size: 1, sequence length: 25979 Attention mask shape: torch.Size([1, 1, 25979, 25979]) Position ids shape: torch.Size([1, 25979]) Input IDs shape: torch.Size([1, 25979]) Labels shape: torch.Size([1, 25979]) Final batch size: 1, sequence length: 16571 Attention mask shape: torch.Size([1, 1, 16571, 16571]) Position ids shape: torch.Size([1, 16571]) Input IDs shape: torch.Size([1, 16571]) Labels shape: torch.Size([1, 16571]) Final batch size: 1, sequence length: 17049 Attention mask shape: torch.Size([1, 1, 17049, 17049]) Position ids shape: torch.Size([1, 17049]) Input IDs shape: torch.Size([1, 17049]) Labels shape: torch.Size([1, 17049]) Final batch size: 1, sequence length: 21290 Attention mask shape: torch.Size([1, 1, 21290, 21290]) Position ids shape: torch.Size([1, 21290]) Input IDs shape: torch.Size([1, 21290]) Labels shape: torch.Size([1, 21290]) Final batch size: 1, sequence length: 24676 Attention mask shape: torch.Size([1, 1, 24676, 24676]) Position ids shape: torch.Size([1, 24676]) Input IDs shape: torch.Size([1, 24676]) Labels shape: torch.Size([1, 24676]) Final batch size: 1, sequence length: 19953 Attention mask shape: torch.Size([1, 1, 19953, 19953]) Position ids shape: torch.Size([1, 19953]) Input IDs shape: torch.Size([1, 19953]) Labels shape: torch.Size([1, 19953]) Final batch size: 1, sequence length: 31662 Final batch size: 1, sequence length: 30635 Attention mask shape: torch.Size([1, 1, 30635, 30635]) Position ids shape: torch.Size([1, 30635]) Input IDs shape: torch.Size([1, 30635]) Labels shape: torch.Size([1, 30635]) Final batch size: 1, sequence length: 32101 Attention mask shape: torch.Size([1, 1, 32101, 32101]) Position ids shape: torch.Size([1, 32101]) Input IDs shape: torch.Size([1, 32101]) Labels shape: torch.Size([1, 32101]) Attention mask shape: torch.Size([1, 1, 31662, 31662]) Position ids shape: torch.Size([1, 31662]) Input IDs shape: torch.Size([1, 31662]) Labels shape: torch.Size([1, 31662]) Final batch size: 1, sequence length: 37720 Attention mask shape: torch.Size([1, 1, 37720, 37720]) Position ids shape: torch.Size([1, 37720]) Input IDs shape: torch.Size([1, 37720]) Labels shape: torch.Size([1, 37720]) Final batch size: 1, sequence length: 37381 Attention mask shape: torch.Size([1, 1, 37381, 37381]) Position ids shape: torch.Size([1, 37381]) Input IDs shape: torch.Size([1, 37381]) Labels shape: torch.Size([1, 37381]) Final batch size: 1, sequence length: 32100 Attention mask shape: torch.Size([1, 1, 32100, 32100]) Position ids shape: torch.Size([1, 32100]) Input IDs shape: torch.Size([1, 32100]) Labels shape: torch.Size([1, 32100]) Final batch size: 1, sequence length: 35058 Attention mask shape: torch.Size([1, 1, 35058, 35058]) Position ids shape: torch.Size([1, 35058]) Input IDs shape: torch.Size([1, 35058]) Labels shape: torch.Size([1, 35058]) Final batch size: 1, sequence length: 37391 Attention mask shape: torch.Size([1, 1, 37391, 37391]) Position ids shape: torch.Size([1, 37391]) Input IDs shape: torch.Size([1, 37391]) Labels shape: torch.Size([1, 37391]) Final batch size: 1, sequence length: 25774 Attention mask shape: torch.Size([1, 1, 25774, 25774]) Position ids shape: torch.Size([1, 25774]) Input IDs shape: torch.Size([1, 25774]) Labels shape: torch.Size([1, 25774]) Final batch size: 1, sequence length: 35014 Attention mask shape: torch.Size([1, 1, 35014, 35014]) Position ids shape: torch.Size([1, 35014]) Input IDs shape: torch.Size([1, 35014]) Labels shape: torch.Size([1, 35014]) Final batch size: 1, sequence length: 28377 Attention mask shape: torch.Size([1, 1, 28377, 28377]) Position ids shape: torch.Size([1, 28377]) Input IDs shape: torch.Size([1, 28377]) Labels shape: torch.Size([1, 28377]) Final batch size: 1, sequence length: 39955 Attention mask shape: torch.Size([1, 1, 39955, 39955]) Position ids shape: torch.Size([1, 39955]) Input IDs shape: torch.Size([1, 39955]) Labels shape: torch.Size([1, 39955]) Final batch size: 1, sequence length: 39474 Attention mask shape: torch.Size([1, 1, 39474, 39474]) Position ids shape: torch.Size([1, 39474]) Input IDs shape: torch.Size([1, 39474]) Labels shape: torch.Size([1, 39474]) Final batch size: 1, sequence length: 25147 Attention mask shape: torch.Size([1, 1, 25147, 25147]) Position ids shape: torch.Size([1, 25147]) Input IDs shape: torch.Size([1, 25147]) Labels shape: torch.Size([1, 25147]) Final batch size: 1, sequence length: 33621 Attention mask shape: torch.Size([1, 1, 33621, 33621]) Position ids shape: torch.Size([1, 33621]) Input IDs shape: torch.Size([1, 33621]) Labels shape: torch.Size([1, 33621]) Final batch size: 1, sequence length: 36047 Attention mask shape: torch.Size([1, 1, 36047, 36047]) Position ids shape: torch.Size([1, 36047]) Input IDs shape: torch.Size([1, 36047]) Labels shape: torch.Size([1, 36047]) Final batch size: 1, sequence length: 38391 Attention mask shape: torch.Size([1, 1, 38391, 38391]) Position ids shape: torch.Size([1, 38391]) Input IDs shape: torch.Size([1, 38391]) Labels shape: torch.Size([1, 38391]) Final batch size: 1, sequence length: 25219 Attention mask shape: torch.Size([1, 1, 25219, 25219]) Position ids shape: torch.Size([1, 25219]) Input IDs shape: torch.Size([1, 25219]) Labels shape: torch.Size([1, 25219]) Final batch size: 1, sequence length: 16816 Attention mask shape: torch.Size([1, 1, 16816, 16816]) Position ids shape: torch.Size([1, 16816]) Input IDs shape: torch.Size([1, 16816]) Labels shape: torch.Size([1, 16816]) Final batch size: 1, sequence length: 26789 Attention mask shape: torch.Size([1, 1, 26789, 26789]) Position ids shape: torch.Size([1, 26789]) Input IDs shape: torch.Size([1, 26789]) Labels shape: torch.Size([1, 26789]) Final batch size: 1, sequence length: 30259 Attention mask shape: torch.Size([1, 1, 30259, 30259]) Position ids shape: torch.Size([1, 30259]) Input IDs shape: torch.Size([1, 30259]) Labels shape: torch.Size([1, 30259]) Final batch size: 1, sequence length: 26454 Attention mask shape: torch.Size([1, 1, 26454, 26454]) Position ids shape: torch.Size([1, 26454]) Input IDs shape: torch.Size([1, 26454]) Labels shape: torch.Size([1, 26454]) Final batch size: 1, sequence length: 26387 Attention mask shape: torch.Size([1, 1, 26387, 26387]) Position ids shape: torch.Size([1, 26387]) Input IDs shape: torch.Size([1, 26387]) Labels shape: torch.Size([1, 26387]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22710 Attention mask shape: torch.Size([1, 1, 22710, 22710]) Position ids shape: torch.Size([1, 22710]) Input IDs shape: torch.Size([1, 22710]) Labels shape: torch.Size([1, 22710]) Final batch size: 1, sequence length: 38577 Attention mask shape: torch.Size([1, 1, 38577, 38577]) Position ids shape: torch.Size([1, 38577]) Input IDs shape: torch.Size([1, 38577]) Labels shape: torch.Size([1, 38577]) Final batch size: 1, sequence length: 21321 Attention mask shape: torch.Size([1, 1, 21321, 21321]) Position ids shape: torch.Size([1, 21321]) Input IDs shape: torch.Size([1, 21321]) Labels shape: torch.Size([1, 21321]) Final batch size: 1, sequence length: 19441 Attention mask shape: torch.Size([1, 1, 19441, 19441]) Position ids shape: torch.Size([1, 19441]) Input IDs shape: torch.Size([1, 19441]) Labels shape: torch.Size([1, 19441]) Final batch size: 1, sequence length: 31447 Attention mask shape: torch.Size([1, 1, 31447, 31447]) Position ids shape: torch.Size([1, 31447]) Input IDs shape: torch.Size([1, 31447]) Labels shape: torch.Size([1, 31447]) Final batch size: 1, sequence length: 27021 Attention mask shape: torch.Size([1, 1, 27021, 27021]) Position ids shape: torch.Size([1, 27021]) Input IDs shape: torch.Size([1, 27021]) Labels shape: torch.Size([1, 27021]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 35261 Attention mask shape: torch.Size([1, 1, 35261, 35261]) Position ids shape: torch.Size([1, 35261]) Input IDs shape: torch.Size([1, 35261]) Labels shape: torch.Size([1, 35261]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40590 Attention mask shape: torch.Size([1, 1, 40590, 40590]) Position ids shape: torch.Size([1, 40590]) Input IDs shape: torch.Size([1, 40590]) Labels shape: torch.Size([1, 40590]) Final batch size: 1, sequence length: 28448 Attention mask shape: torch.Size([1, 1, 28448, 28448]) Position ids shape: torch.Size([1, 28448]) Input IDs shape: torch.Size([1, 28448]) Labels shape: torch.Size([1, 28448]) Final batch size: 1, sequence length: 32079 Attention mask shape: torch.Size([1, 1, 32079, 32079]) Position ids shape: torch.Size([1, 32079]) Input IDs shape: torch.Size([1, 32079]) Labels shape: torch.Size([1, 32079]) Final batch size: 1, sequence length: 27371 Attention mask shape: torch.Size([1, 1, 27371, 27371]) Position ids shape: torch.Size([1, 27371]) Input IDs shape: torch.Size([1, 27371]) Labels shape: torch.Size([1, 27371]) Final batch size: 1, sequence length: 32129 Attention mask shape: torch.Size([1, 1, 32129, 32129]) Position ids shape: torch.Size([1, 32129]) Input IDs shape: torch.Size([1, 32129]) Labels shape: torch.Size([1, 32129]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 14136 Attention mask shape: torch.Size([1, 1, 14136, 14136]) Position ids shape: torch.Size([1, 14136]) Input IDs shape: torch.Size([1, 14136]) Labels shape: torch.Size([1, 14136]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33014 Attention mask shape: torch.Size([1, 1, 33014, 33014]) Position ids shape: torch.Size([1, 33014]) Input IDs shape: torch.Size([1, 33014]) Labels shape: torch.Size([1, 33014]) Final batch size: 1, sequence length: 24623 Attention mask shape: torch.Size([1, 1, 24623, 24623]) Position ids shape: torch.Size([1, 24623]) Input IDs shape: torch.Size([1, 24623]) Labels shape: torch.Size([1, 24623]) Final batch size: 1, sequence length: 36356 Attention mask shape: torch.Size([1, 1, 36356, 36356]) Position ids shape: torch.Size([1, 36356]) Input IDs shape: torch.Size([1, 36356]) Labels shape: torch.Size([1, 36356]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19027 Attention mask shape: torch.Size([1, 1, 19027, 19027]) Position ids shape: torch.Size([1, 19027]) Input IDs shape: torch.Size([1, 19027]) Labels shape: torch.Size([1, 19027]) Final batch size: 1, sequence length: 30712 Attention mask shape: torch.Size([1, 1, 30712, 30712]) Position ids shape: torch.Size([1, 30712]) Input IDs shape: torch.Size([1, 30712]) Labels shape: torch.Size([1, 30712]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 23208 Attention mask shape: torch.Size([1, 1, 23208, 23208]) Position ids shape: torch.Size([1, 23208]) Input IDs shape: torch.Size([1, 23208]) Labels shape: torch.Size([1, 23208]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2938, 'grad_norm': 0.30469065469568796, 'learning_rate': 7.938926261462366e-06, 'num_tokens': -inf, 'epoch': 2.88} Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 10507 Attention mask shape: torch.Size([1, 1, 10507, 10507]) Position ids shape: torch.Size([1, 10507]) Input IDs shape: torch.Size([1, 10507]) Labels shape: torch.Size([1, 10507]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 26271 Attention mask shape: torch.Size([1, 1, 26271, 26271]) Position ids shape: torch.Size([1, 26271]) Input IDs shape: torch.Size([1, 26271]) Labels shape: torch.Size([1, 26271]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 31381 Attention mask shape: torch.Size([1, 1, 31381, 31381]) Position ids shape: torch.Size([1, 31381]) Input IDs shape: torch.Size([1, 31381]) Labels shape: torch.Size([1, 31381]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2878, 'grad_norm': 0.324777119499745, 'learning_rate': 7.723195175075136e-06, 'num_tokens': -inf, 'epoch': 3.0} Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 6362 Attention mask shape: torch.Size([1, 1, 6362, 6362]) Position ids shape: torch.Size([1, 6362]) Input IDs shape: torch.Size([1, 6362]) Labels shape: torch.Size([1, 6362]) Final batch size: 1, sequence length: 10523 Attention mask shape: torch.Size([1, 1, 10523, 10523]) Position ids shape: torch.Size([1, 10523]) Input IDs shape: torch.Size([1, 10523]) Labels shape: torch.Size([1, 10523]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 18410 Attention mask shape: torch.Size([1, 1, 18410, 18410]) Position ids shape: torch.Size([1, 18410]) Input IDs shape: torch.Size([1, 18410]) Labels shape: torch.Size([1, 18410]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 12385 Attention mask shape: torch.Size([1, 1, 12385, 12385]) Position ids shape: torch.Size([1, 12385]) Input IDs shape: torch.Size([1, 12385]) Labels shape: torch.Size([1, 12385]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 19427 Labels shape: torch.Size([1, 16187]) Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 18098 Attention mask shape: torch.Size([1, 1, 18098, 18098]) Position ids shape: torch.Size([1, 18098]) Input IDs shape: torch.Size([1, 18098]) Labels shape: torch.Size([1, 18098]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 12656 Attention mask shape: torch.Size([1, 1, 12656, 12656]) Position ids shape: torch.Size([1, 12656]) Input IDs shape: torch.Size([1, 12656]) Labels shape: torch.Size([1, 12656]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 17299 Attention mask shape: torch.Size([1, 1, 17299, 17299]) Position ids shape: torch.Size([1, 17299]) Input IDs shape: torch.Size([1, 17299]) Labels shape: torch.Size([1, 17299]) Final batch size: 1, sequence length: 11819 Attention mask shape: torch.Size([1, 1, 11819, 11819]) Position ids shape: torch.Size([1, 11819]) Input IDs shape: torch.Size([1, 11819]) Labels shape: torch.Size([1, 11819]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 24365 Attention mask shape: torch.Size([1, 1, 24365, 24365]) Position ids shape: torch.Size([1, 24365]) Input IDs shape: torch.Size([1, 24365]) Labels shape: torch.Size([1, 24365]) Final batch size: 1, sequence length: 18060 Attention mask shape: torch.Size([1, 1, 18060, 18060]) Position ids shape: torch.Size([1, 18060]) Input IDs shape: torch.Size([1, 18060]) Labels shape: torch.Size([1, 18060]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 9091 Attention mask shape: torch.Size([1, 1, 9091, 9091]) Position ids shape: torch.Size([1, 9091]) Input IDs shape: torch.Size([1, 9091]) Labels shape: torch.Size([1, 9091]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 6948 Attention mask shape: torch.Size([1, 1, 6948, 6948]) Position ids shape: torch.Size([1, 6948]) Input IDs shape: torch.Size([1, 6948]) Labels shape: torch.Size([1, 6948]) Final batch size: 1, sequence length: 28841 Attention mask shape: torch.Size([1, 1, 28841, 28841]) Position ids shape: torch.Size([1, 28841]) Input IDs shape: torch.Size([1, 28841]) Labels shape: torch.Size([1, 28841]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 29113 Attention mask shape: torch.Size([1, 1, 29113, 29113]) Position ids shape: torch.Size([1, 29113]) Input IDs shape: torch.Size([1, 29113]) Labels shape: torch.Size([1, 29113]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 18915 Attention mask shape: torch.Size([1, 1, 18915, 18915]) Position ids shape: torch.Size([1, 18915]) Input IDs shape: torch.Size([1, 18915]) Labels shape: torch.Size([1, 18915]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 26976 Attention mask shape: torch.Size([1, 1, 26976, 26976]) Position ids shape: torch.Size([1, 26976]) Input IDs shape: torch.Size([1, 26976]) Labels shape: torch.Size([1, 26976]) Final batch size: 1, sequence length: 21450 Attention mask shape: torch.Size([1, 1, 21450, 21450]) Position ids shape: torch.Size([1, 21450]) Input IDs shape: torch.Size([1, 21450]) Labels shape: torch.Size([1, 21450]) Final batch size: 1, sequence length: 13215 Attention mask shape: torch.Size([1, 1, 13215, 13215]) Position ids shape: torch.Size([1, 13215]) Input IDs shape: torch.Size([1, 13215]) Labels shape: torch.Size([1, 13215]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 25172 Attention mask shape: torch.Size([1, 1, 25172, 25172]) Position ids shape: torch.Size([1, 25172]) Input IDs shape: torch.Size([1, 25172]) Labels shape: torch.Size([1, 25172]) Final batch size: 1, sequence length: 24880 Attention mask shape: torch.Size([1, 1, 24880, 24880]) Position ids shape: torch.Size([1, 24880]) Input IDs shape: torch.Size([1, 24880]) Labels shape: torch.Size([1, 24880]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 27541 Attention mask shape: torch.Size([1, 1, 27541, 27541]) Position ids shape: torch.Size([1, 27541]) Input IDs shape: torch.Size([1, 27541]) Labels shape: torch.Size([1, 27541]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 7722 Attention mask shape: torch.Size([1, 1, 7722, 7722]) Position ids shape: torch.Size([1, 7722]) Input IDs shape: torch.Size([1, 7722]) Labels shape: torch.Size([1, 7722]) Final batch size: 1, sequence length: 21596 Attention mask shape: torch.Size([1, 1, 21596, 21596]) Position ids shape: torch.Size([1, 21596]) Input IDs shape: torch.Size([1, 21596]) Labels shape: torch.Size([1, 21596]) Final batch size: 1, sequence length: 21491 Attention mask shape: torch.Size([1, 1, 21491, 21491]) Position ids shape: torch.Size([1, 21491]) Input IDs shape: torch.Size([1, 21491]) Labels shape: torch.Size([1, 21491]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 29875 Attention mask shape: torch.Size([1, 1, 29875, 29875]) Position ids shape: torch.Size([1, 29875]) Input IDs shape: torch.Size([1, 29875]) Labels shape: torch.Size([1, 29875]) Final batch size: 1, sequence length: 20827 Attention mask shape: torch.Size([1, 1, 20827, 20827]) Position ids shape: torch.Size([1, 20827]) Input IDs shape: torch.Size([1, 20827]) Labels shape: torch.Size([1, 20827]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 20363 Attention mask shape: torch.Size([1, 1, 20363, 20363]) Position ids shape: torch.Size([1, 20363]) Input IDs shape: torch.Size([1, 20363]) Labels shape: torch.Size([1, 20363]) Final batch size: 1, sequence length: 25014 Attention mask shape: torch.Size([1, 1, 25014, 25014]) Position ids shape: torch.Size([1, 25014]) Input IDs shape: torch.Size([1, 25014]) Labels shape: torch.Size([1, 25014]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 14009 Attention mask shape: torch.Size([1, 1, 14009, 14009]) Position ids shape: torch.Size([1, 14009]) Input IDs shape: torch.Size([1, 14009]) Labels shape: torch.Size([1, 14009]) Final batch size: 1, sequence length: 30623 Attention mask shape: torch.Size([1, 1, 30623, 30623]) Position ids shape: torch.Size([1, 30623]) Input IDs shape: torch.Size([1, 30623]) Labels shape: torch.Size([1, 30623]) Final batch size: 1, sequence length: 17400 Attention mask shape: torch.Size([1, 1, 17400, 17400]) Position ids shape: torch.Size([1, 17400]) Input IDs shape: torch.Size([1, 17400]) Labels shape: torch.Size([1, 17400]) Final batch size: 1, sequence length: 33367 Attention mask shape: torch.Size([1, 1, 33367, 33367]) Position ids shape: torch.Size([1, 33367]) Input IDs shape: torch.Size([1, 33367]) Labels shape: torch.Size([1, 33367]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 28641 Attention mask shape: torch.Size([1, 1, 28641, 28641]) Position ids shape: torch.Size([1, 28641]) Input IDs shape: torch.Size([1, 28641]) Labels shape: torch.Size([1, 28641]) Final batch size: 1, sequence length: 19705 Attention mask shape: torch.Size([1, 1, 19705, 19705]) Position ids shape: torch.Size([1, 19705]) Input IDs shape: torch.Size([1, 19705]) Labels shape: torch.Size([1, 19705]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 27265 Attention mask shape: torch.Size([1, 1, 27265, 27265]) Position ids shape: torch.Size([1, 27265]) Input IDs shape: torch.Size([1, 27265]) Labels shape: torch.Size([1, 27265]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 24001 Attention mask shape: torch.Size([1, 1, 24001, 24001]) Position ids shape: torch.Size([1, 24001]) Input IDs shape: torch.Size([1, 24001]) Labels shape: torch.Size([1, 24001]) Final batch size: 1, sequence length: 22098 Attention mask shape: torch.Size([1, 1, 22098, 22098]) Position ids shape: torch.Size([1, 22098]) Input IDs shape: torch.Size([1, 22098]) Labels shape: torch.Size([1, 22098]) Final batch size: 1, sequence length: 15656 Attention mask shape: torch.Size([1, 1, 15656, 15656]) Position ids shape: torch.Size([1, 15656]) Input IDs shape: torch.Size([1, 15656]) Labels shape: torch.Size([1, 15656]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 26306 Attention mask shape: torch.Size([1, 1, 26306, 26306]) Position ids shape: torch.Size([1, 26306]) Input IDs shape: torch.Size([1, 26306]) Labels shape: torch.Size([1, 26306]) Final batch size: 1, sequence length: 27243 Attention mask shape: torch.Size([1, 1, 27243, 27243]) Position ids shape: torch.Size([1, 27243]) Input IDs shape: torch.Size([1, 27243]) Labels shape: torch.Size([1, 27243]) Final batch size: 1, sequence length: 21348 Attention mask shape: torch.Size([1, 1, 21348, 21348]) Position ids shape: torch.Size([1, 21348]) Input IDs shape: torch.Size([1, 21348]) Labels shape: torch.Size([1, 21348]) Final batch size: 1, sequence length: 18606 Attention mask shape: torch.Size([1, 1, 18606, 18606]) Position ids shape: torch.Size([1, 18606]) Input IDs shape: torch.Size([1, 18606]) Labels shape: torch.Size([1, 18606]) Final batch size: 1, sequence length: 38249 Attention mask shape: torch.Size([1, 1, 38249, 38249]) Position ids shape: torch.Size([1, 38249]) Input IDs shape: torch.Size([1, 38249]) Labels shape: torch.Size([1, 38249]) Final batch size: 1, sequence length: 19586 Attention mask shape: torch.Size([1, 1, 19586, 19586]) Position ids shape: torch.Size([1, 19586]) Input IDs shape: torch.Size([1, 19586]) Labels shape: torch.Size([1, 19586]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 39142 Attention mask shape: torch.Size([1, 1, 39142, 39142]) Position ids shape: torch.Size([1, 39142]) Input IDs shape: torch.Size([1, 39142]) Labels shape: torch.Size([1, 39142]) Final batch size: 1, sequence length: 29632 Attention mask shape: torch.Size([1, 1, 29632, 29632]) Position ids shape: torch.Size([1, 29632]) Input IDs shape: torch.Size([1, 29632]) Labels shape: torch.Size([1, 29632]) Final batch size: 1, sequence length: 30066 Attention mask shape: torch.Size([1, 1, 30066, 30066]) Position ids shape: torch.Size([1, 30066]) Input IDs shape: torch.Size([1, 30066]) Labels shape: torch.Size([1, 30066]) Final batch size: 1, sequence length: 40745 Attention mask shape: torch.Size([1, 1, 40745, 40745]) Position ids shape: torch.Size([1, 40745]) Input IDs shape: torch.Size([1, 40745]) Labels shape: torch.Size([1, 40745]) Final batch size: 1, sequence length: 24622 Attention mask shape: torch.Size([1, 1, 24622, 24622]) Position ids shape: torch.Size([1, 24622]) Input IDs shape: torch.Size([1, 24622]) Labels shape: torch.Size([1, 24622]) Final batch size: 1, sequence length: 18565 Attention mask shape: torch.Size([1, 1, 18565, 18565]) Position ids shape: torch.Size([1, 18565]) Input IDs shape: torch.Size([1, 18565]) Labels shape: torch.Size([1, 18565]) Final batch size: 1, sequence length: 27441 Attention mask shape: torch.Size([1, 1, 27441, 27441]) Position ids shape: torch.Size([1, 27441]) Input IDs shape: torch.Size([1, 27441]) Labels shape: torch.Size([1, 27441]) Final batch size: 1, sequence length: 36271 Attention mask shape: torch.Size([1, 1, 36271, 36271]) Position ids shape: torch.Size([1, 36271]) Input IDs shape: torch.Size([1, 36271]) Labels shape: torch.Size([1, 36271]) Final batch size: 1, sequence length: 17778 Attention mask shape: torch.Size([1, 1, 17778, 17778]) Position ids shape: torch.Size([1, 17778]) Input IDs shape: torch.Size([1, 17778]) Labels shape: torch.Size([1, 17778]) Final batch size: 1, sequence length: 22625 Attention mask shape: torch.Size([1, 1, 22625, 22625]) Position ids shape: torch.Size([1, 22625]) Input IDs shape: torch.Size([1, 22625]) Labels shape: torch.Size([1, 22625]) Final batch size: 1, sequence length: 26939 Attention mask shape: torch.Size([1, 1, 26939, 26939]) Position ids shape: torch.Size([1, 26939]) Input IDs shape: torch.Size([1, 26939]) Labels shape: torch.Size([1, 26939]) Final batch size: 1, sequence length: 21567 Attention mask shape: torch.Size([1, 1, 21567, 21567]) Position ids shape: torch.Size([1, 21567]) Input IDs shape: torch.Size([1, 21567]) Labels shape: torch.Size([1, 21567]) Final batch size: 1, sequence length: 13509 Attention mask shape: torch.Size([1, 1, 13509, 13509]) Position ids shape: torch.Size([1, 13509]) Input IDs shape: torch.Size([1, 13509]) Labels shape: torch.Size([1, 13509]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30802 Attention mask shape: torch.Size([1, 1, 30802, 30802]) Position ids shape: torch.Size([1, 30802]) Input IDs shape: torch.Size([1, 30802]) Labels shape: torch.Size([1, 30802]) Final batch size: 1, sequence length: 40496 Attention mask shape: torch.Size([1, 1, 40496, 40496]) Position ids shape: torch.Size([1, 40496]) Input IDs shape: torch.Size([1, 40496]) Labels shape: torch.Size([1, 40496]) Final batch size: 1, sequence length: 15217 Attention mask shape: torch.Size([1, 1, 15217, 15217]) Position ids shape: torch.Size([1, 15217]) Input IDs shape: torch.Size([1, 15217]) Labels shape: torch.Size([1, 15217]) Final batch size: 1, sequence length: 35153 Attention mask shape: torch.Size([1, 1, 35153, 35153]) Position ids shape: torch.Size([1, 35153]) Input IDs shape: torch.Size([1, 35153]) Labels shape: torch.Size([1, 35153]) Final batch size: 1, sequence length: 32609 Attention mask shape: torch.Size([1, 1, 32609, 32609]) Position ids shape: torch.Size([1, 32609]) Input IDs shape: torch.Size([1, 32609]) Labels shape: torch.Size([1, 32609]) Final batch size: 1, sequence length: 33839 Attention mask shape: torch.Size([1, 1, 33839, 33839]) Position ids shape: torch.Size([1, 33839]) Input IDs shape: torch.Size([1, 33839]) Labels shape: torch.Size([1, 33839]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 25447 Attention mask shape: torch.Size([1, 1, 25447, 25447]) Position ids shape: torch.Size([1, 25447]) Input IDs shape: torch.Size([1, 25447]) Labels shape: torch.Size([1, 25447]) Final batch size: 1, sequence length: 30101 Attention mask shape: torch.Size([1, 1, 30101, 30101]) Position ids shape: torch.Size([1, 30101]) Input IDs shape: torch.Size([1, 30101]) Labels shape: torch.Size([1, 30101]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 15508 Attention mask shape: torch.Size([1, 1, 15508, 15508]) Position ids shape: torch.Size([1, 15508]) Input IDs shape: torch.Size([1, 15508]) Labels shape: torch.Size([1, 15508]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36511 Attention mask shape: torch.Size([1, 1, 36511, 36511]) Position ids shape: torch.Size([1, 36511]) Input IDs shape: torch.Size([1, 36511]) Labels shape: torch.Size([1, 36511]) Final batch size: 1, sequence length: 32352 Attention mask shape: torch.Size([1, 1, 32352, 32352]) Position ids shape: torch.Size([1, 32352]) Input IDs shape: torch.Size([1, 32352]) Labels shape: torch.Size([1, 32352]) Final batch size: 1, sequence length: 23258 Attention mask shape: torch.Size([1, 1, 23258, 23258]) Position ids shape: torch.Size([1, 23258]) Input IDs shape: torch.Size([1, 23258]) Labels shape: torch.Size([1, 23258]) Final batch size: 1, sequence length: 39836 Attention mask shape: torch.Size([1, 1, 39836, 39836]) Position ids shape: torch.Size([1, 39836]) Input IDs shape: torch.Size([1, 39836]) Labels shape: torch.Size([1, 39836]) Final batch size: 1, sequence length: 30109 Attention mask shape: torch.Size([1, 1, 30109, 30109]) Position ids shape: torch.Size([1, 30109]) Input IDs shape: torch.Size([1, 30109]) Labels shape: torch.Size([1, 30109]) Final batch size: 1, sequence length: 9947 Attention mask shape: torch.Size([1, 1, 9947, 9947]) Position ids shape: torch.Size([1, 9947]) Input IDs shape: torch.Size([1, 9947]) Labels shape: torch.Size([1, 9947]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19036 Attention mask shape: torch.Size([1, 1, 19036, 19036]) Position ids shape: torch.Size([1, 19036]) Input IDs shape: torch.Size([1, 19036]) Labels shape: torch.Size([1, 19036]) Final batch size: 1, sequence length: 16677 Attention mask shape: torch.Size([1, 1, 16677, 16677]) Position ids shape: torch.Size([1, 16677]) Input IDs shape: torch.Size([1, 16677]) Labels shape: torch.Size([1, 16677]) Final batch size: 1, sequence length: 21758 Attention mask shape: torch.Size([1, 1, 21758, 21758]) Position ids shape: torch.Size([1, 21758]) Input IDs shape: torch.Size([1, 21758]) Labels shape: torch.Size([1, 21758]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20422 Attention mask shape: torch.Size([1, 1, 20422, 20422]) Position ids shape: torch.Size([1, 20422]) Input IDs shape: torch.Size([1, 20422]) Labels shape: torch.Size([1, 20422]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 16587 Attention mask shape: torch.Size([1, 1, 16587, 16587]) Position ids shape: torch.Size([1, 16587]) Input IDs shape: torch.Size([1, 16587]) Labels shape: torch.Size([1, 16587]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 13095 Attention mask shape: torch.Size([1, 1, 13095, 13095]) Position ids shape: torch.Size([1, 13095]) Input IDs shape: torch.Size([1, 13095]) Labels shape: torch.Size([1, 13095]) Final batch size: 1, sequence length: 37945 Attention mask shape: torch.Size([1, 1, 37945, 37945]) Position ids shape: torch.Size([1, 37945]) Input IDs shape: torch.Size([1, 37945]) Labels shape: torch.Size([1, 37945]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20611 Attention mask shape: torch.Size([1, 1, 20611, 20611]) Position ids shape: torch.Size([1, 20611]) Input IDs shape: torch.Size([1, 20611]) Labels shape: torch.Size([1, 20611]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 13297 Attention mask shape: torch.Size([1, 1, 13297, 13297]) Position ids shape: torch.Size([1, 13297]) Input IDs shape: torch.Size([1, 13297]) Labels shape: torch.Size([1, 13297]) Final batch size: 1, sequence length: 24298 Attention mask shape: torch.Size([1, 1, 24298, 24298]) Position ids shape: torch.Size([1, 24298]) Input IDs shape: torch.Size([1, 24298]) Labels shape: torch.Size([1, 24298]) Final batch size: 1, sequence length: 34937 Attention mask shape: torch.Size([1, 1, 34937, 34937]) Position ids shape: torch.Size([1, 34937]) Input IDs shape: torch.Size([1, 34937]) Labels shape: torch.Size([1, 34937]) Final batch size: 1, sequence length: 34142 Attention mask shape: torch.Size([1, 1, 34142, 34142]) Position ids shape: torch.Size([1, 34142]) Input IDs shape: torch.Size([1, 34142]) Labels shape: torch.Size([1, 34142]) Final batch size: 1, sequence length: 22786 Attention mask shape: torch.Size([1, 1, 22786, 22786]) Position ids shape: torch.Size([1, 22786]) Input IDs shape: torch.Size([1, 22786]) Labels shape: torch.Size([1, 22786]) Final batch size: 1, sequence length: 12224 Attention mask shape: torch.Size([1, 1, 12224, 12224]) Position ids shape: torch.Size([1, 12224]) Input IDs shape: torch.Size([1, 12224]) Labels shape: torch.Size([1, 12224]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28634 Attention mask shape: torch.Size([1, 1, 28634, 28634]) Position ids shape: torch.Size([1, 28634]) Input IDs shape: torch.Size([1, 28634]) Labels shape: torch.Size([1, 28634]) Final batch size: 1, sequence length: 32753 Attention mask shape: torch.Size([1, 1, 32753, 32753]) Position ids shape: torch.Size([1, 32753]) Input IDs shape: torch.Size([1, 32753]) Labels shape: torch.Size([1, 32753]) Final batch size: 1, sequence length: 32472 Attention mask shape: torch.Size([1, 1, 32472, 32472]) Position ids shape: torch.Size([1, 32472]) Input IDs shape: torch.Size([1, 32472]) Labels shape: torch.Size([1, 32472]) Final batch size: 1, sequence length: 21683 Attention mask shape: torch.Size([1, 1, 21683, 21683]) Position ids shape: torch.Size([1, 21683]) Input IDs shape: torch.Size([1, 21683]) Labels shape: torch.Size([1, 21683]) Final batch size: 1, sequence length: 17373 Attention mask shape: torch.Size([1, 1, 17373, 17373]) Position ids shape: torch.Size([1, 17373]) Input IDs shape: torch.Size([1, 17373]) Labels shape: torch.Size([1, 17373]) Final batch size: 1, sequence length: 12653 Attention mask shape: torch.Size([1, 1, 12653, 12653]) Position ids shape: torch.Size([1, 12653]) Input IDs shape: torch.Size([1, 12653]) Labels shape: torch.Size([1, 12653]) Final batch size: 1, sequence length: 22623 Attention mask shape: torch.Size([1, 1, 22623, 22623]) Position ids shape: torch.Size([1, 22623]) Input IDs shape: torch.Size([1, 22623]) Labels shape: torch.Size([1, 22623]) Final batch size: 1, sequence length: 14995 Attention mask shape: torch.Size([1, 1, 14995, 14995]) Position ids shape: torch.Size([1, 14995]) Input IDs shape: torch.Size([1, 14995]) Labels shape: torch.Size([1, 14995]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24424 Attention mask shape: torch.Size([1, 1, 24424, 24424]) Position ids shape: torch.Size([1, 24424]) Input IDs shape: torch.Size([1, 24424]) Labels shape: torch.Size([1, 24424]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32564 Attention mask shape: torch.Size([1, 1, 32564, 32564]) Position ids shape: torch.Size([1, 32564]) Input IDs shape: torch.Size([1, 32564]) Labels shape: torch.Size([1, 32564]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 35864 Attention mask shape: torch.Size([1, 1, 35864, 35864]) Position ids shape: torch.Size([1, 35864]) Input IDs shape: torch.Size([1, 35864]) Labels shape: torch.Size([1, 35864]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26006 Attention mask shape: torch.Size([1, 1, 26006, 26006]) Position ids shape: torch.Size([1, 26006]) Input IDs shape: torch.Size([1, 26006]) Labels shape: torch.Size([1, 26006]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32344 Attention mask shape: torch.Size([1, 1, 32344, 32344]) Position ids shape: torch.Size([1, 32344]) Input IDs shape: torch.Size([1, 32344]) Labels shape: torch.Size([1, 32344]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26537 Attention mask shape: torch.Size([1, 1, 26537, 26537]) Position ids shape: torch.Size([1, 26537]) Input IDs shape: torch.Size([1, 26537]) Labels shape: torch.Size([1, 26537]) Final batch size: 1, sequence length: 37168 Attention mask shape: torch.Size([1, 1, 37168, 37168]) Position ids shape: torch.Size([1, 37168]) Input IDs shape: torch.Size([1, 37168]) Labels shape: torch.Size([1, 37168]) Final batch size: 1, sequence length: 32919 Attention mask shape: torch.Size([1, 1, 32919, 32919]) Position ids shape: torch.Size([1, 32919]) Input IDs shape: torch.Size([1, 32919]) Labels shape: torch.Size([1, 32919]) Final batch size: 1, sequence length: 32514 Attention mask shape: torch.Size([1, 1, 32514, 32514]) Position ids shape: torch.Size([1, 32514]) Input IDs shape: torch.Size([1, 32514]) Labels shape: torch.Size([1, 32514]) Final batch size: 1, sequence length: 23362 Attention mask shape: torch.Size([1, 1, 23362, 23362]) Position ids shape: torch.Size([1, 23362]) Input IDs shape: torch.Size([1, 23362]) Labels shape: torch.Size([1, 23362]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28861 Attention mask shape: torch.Size([1, 1, 28861, 28861]) Position ids shape: torch.Size([1, 28861]) Input IDs shape: torch.Size([1, 28861]) Labels shape: torch.Size([1, 28861]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2809, 'grad_norm': 0.34051833183011415, 'learning_rate': 7.500000000000001e-06, 'num_tokens': -inf, 'epoch': 3.12} Final batch size: 1, sequence length: 6316 Attention mask shape: torch.Size([1, 1, 6316, 6316]) Position ids shape: torch.Size([1, 6316]) Input IDs shape: torch.Size([1, 6316]) Labels shape: torch.Size([1, 6316]) Final batch size: 1, sequence length: 4858 Attention mask shape: torch.Size([1, 1, 4858, 4858]) Position ids shape: torch.Size([1, 4858]) Input IDs shape: torch.Size([1, 4858]) Labels shape: torch.Size([1, 4858]) Final batch size: 1, sequence length: 7360 Attention mask shape: torch.Size([1, 1, 7360, 7360]) Position ids shape: torch.Size([1, 7360]) Input IDs shape: torch.Size([1, 7360]) Labels shape: torch.Size([1, 7360]) Final batch size: 1, sequence length: 10301 Attention mask shape: torch.Size([1, 1, 10301, 10301]) Position ids shape: torch.Size([1, 10301]) Input IDs shape: torch.Size([1, 10301]) Labels shape: torch.Size([1, 10301]) Final batch size: 1, sequence length: 12293 Attention mask shape: torch.Size([1, 1, 12293, 12293]) Position ids shape: torch.Size([1, 12293]) Input IDs shape: torch.Size([1, 12293]) Labels shape: torch.Size([1, 12293]) Final batch size: 1, sequence length: 13355 Attention mask shape: torch.Size([1, 1, 13355, 13355]) Position ids shape: torch.Size([1, 13355]) Input IDs shape: torch.Size([1, 13355]) Labels shape: torch.Size([1, 13355]) Final batch size: 1, sequence length: 14363 Attention mask shape: torch.Size([1, 1, 14363, 14363]) Position ids shape: torch.Size([1, 14363]) Input IDs shape: torch.Size([1, 14363]) Labels shape: torch.Size([1, 14363]) Final batch size: 1, sequence length: 11548 Attention mask shape: torch.Size([1, 1, 11548, 11548]) Position ids shape: torch.Size([1, 11548]) Input IDs shape: torch.Size([1, 11548]) Labels shape: torch.Size([1, 11548]) Final batch size: 1, sequence length: 11728 Attention mask shape: torch.Size([1, 1, 11728, 11728]) Position ids shape: torch.Size([1, 11728]) Input IDs shape: torch.Size([1, 11728]) Labels shape: torch.Size([1, 11728]) Final batch size: 1, sequence length: 16827 Attention mask shape: torch.Size([1, 1, 16827, 16827]) Position ids shape: torch.Size([1, 16827]) Input IDs shape: torch.Size([1, 16827]) Labels shape: torch.Size([1, 16827]) Final batch size: 1, sequence length: 14704 Attention mask shape: torch.Size([1, 1, 14704, 14704]) Position ids shape: torch.Size([1, 14704]) Input IDs shape: torch.Size([1, 14704]) Labels shape: torch.Size([1, 14704]) Final batch size: 1, sequence length: 16536 Attention mask shape: torch.Size([1, 1, 16536, 16536]) Position ids shape: torch.Size([1, 16536]) Input IDs shape: torch.Size([1, 16536]) Labels shape: torch.Size([1, 16536]) Final batch size: 1, sequence length: 17415 Attention mask shape: torch.Size([1, 1, 17415, 17415]) Position ids shape: torch.Size([1, 17415]) Input IDs shape: torch.Size([1, 17415]) Labels shape: torch.Size([1, 17415]) Final batch size: 1, sequence length: 17108 Attention mask shape: torch.Size([1, 1, 17108, 17108]) Position ids shape: torch.Size([1, 17108]) Input IDs shape: torch.Size([1, 17108]) Labels shape: torch.Size([1, 17108]) Final batch size: 1, sequence length: 14587 Attention mask shape: torch.Size([1, 1, 14587, 14587]) Position ids shape: torch.Size([1, 14587]) Input IDs shape: torch.Size([1, 14587]) Labels shape: torch.Size([1, 14587]) Final batch size: 1, sequence length: 18653 Attention mask shape: torch.Size([1, 1, 18653, 18653]) Position ids shape: torch.Size([1, 18653]) Input IDs shape: torch.Size([1, 18653]) Labels shape: torch.Size([1, 18653]) Final batch size: 1, sequence length: 19414 Attention mask shape: torch.Size([1, 1, 19414, 19414]) Position ids shape: torch.Size([1, 19414]) Input IDs shape: torch.Size([1, 19414]) Labels shape: torch.Size([1, 19414]) Final batch size: 1, sequence length: 18741 Attention mask shape: torch.Size([1, 1, 18741, 18741]) Position ids shape: torch.Size([1, 18741]) Input IDs shape: torch.Size([1, 18741]) Labels shape: torch.Size([1, 18741]) Final batch size: 1, sequence length: 18496 Attention mask shape: torch.Size([1, 1, 18496, 18496]) Position ids shape: torch.Size([1, 18496]) Input IDs shape: torch.Size([1, 18496]) Labels shape: torch.Size([1, 18496]) Final batch size: 1, sequence length: 18927 Attention mask shape: torch.Size([1, 1, 18927, 18927]) Position ids shape: torch.Size([1, 18927]) Input IDs shape: torch.Size([1, 18927]) Labels shape: torch.Size([1, 18927]) Final batch size: 1, sequence length: 16750 Attention mask shape: torch.Size([1, 1, 16750, 16750]) Position ids shape: torch.Size([1, 16750]) Input IDs shape: torch.Size([1, 16750]) Labels shape: torch.Size([1, 16750]) Final batch size: 1, sequence length: 17733 Attention mask shape: torch.Size([1, 1, 17733, 17733]) Position ids shape: torch.Size([1, 17733]) Input IDs shape: torch.Size([1, 17733]) Labels shape: torch.Size([1, 17733]) Final batch size: 1, sequence length: 13638 Attention mask shape: torch.Size([1, 1, 13638, 13638]) Position ids shape: torch.Size([1, 13638]) Input IDs shape: torch.Size([1, 13638]) Labels shape: torch.Size([1, 13638]) Final batch size: 1, sequence length: 20612 Attention mask shape: torch.Size([1, 1, 20612, 20612]) Position ids shape: torch.Size([1, 20612]) Input IDs shape: torch.Size([1, 20612]) Labels shape: torch.Size([1, 20612]) Final batch size: 1, sequence length: 7221 Attention mask shape: torch.Size([1, 1, 7221, 7221]) Position ids shape: torch.Size([1, 7221]) Input IDs shape: torch.Size([1, 7221]) Labels shape: torch.Size([1, 7221]) Final batch size: 1, sequence length: 22391 Attention mask shape: torch.Size([1, 1, 22391, 22391]) Position ids shape: torch.Size([1, 22391]) Input IDs shape: torch.Size([1, 22391]) Labels shape: torch.Size([1, 22391]) Final batch size: 1, sequence length: 21420 Attention mask shape: torch.Size([1, 1, 21420, 21420]) Position ids shape: torch.Size([1, 21420]) Input IDs shape: torch.Size([1, 21420]) Labels shape: torch.Size([1, 21420]) Final batch size: 1, sequence length: 17220 Attention mask shape: torch.Size([1, 1, 17220, 17220]) Position ids shape: torch.Size([1, 17220]) Input IDs shape: torch.Size([1, 17220]) Labels shape: torch.Size([1, 17220]) Final batch size: 1, sequence length: 20933 Attention mask shape: torch.Size([1, 1, 20933, 20933]) Position ids shape: torch.Size([1, 20933]) Input IDs shape: torch.Size([1, 20933]) Labels shape: torch.Size([1, 20933]) Final batch size: 1, sequence length: 18393 Attention mask shape: torch.Size([1, 1, 18393, 18393]) Position ids shape: torch.Size([1, 18393]) Input IDs shape: torch.Size([1, 18393]) Labels shape: torch.Size([1, 18393]) Final batch size: 1, sequence length: 22004 Attention mask shape: torch.Size([1, 1, 22004, 22004]) Position ids shape: torch.Size([1, 22004]) Input IDs shape: torch.Size([1, 22004]) Labels shape: torch.Size([1, 22004]) Final batch size: 1, sequence length: 20579 Attention mask shape: torch.Size([1, 1, 20579, 20579]) Position ids shape: torch.Size([1, 20579]) Input IDs shape: torch.Size([1, 20579]) Labels shape: torch.Size([1, 20579]) Final batch size: 1, sequence length: 11067 Attention mask shape: torch.Size([1, 1, 11067, 11067]) Position ids shape: torch.Size([1, 11067]) Input IDs shape: torch.Size([1, 11067]) Labels shape: torch.Size([1, 11067]) Final batch size: 1, sequence length: 22887 Attention mask shape: torch.Size([1, 1, 22887, 22887]) Position ids shape: torch.Size([1, 22887]) Input IDs shape: torch.Size([1, 22887]) Labels shape: torch.Size([1, 22887]) Final batch size: 1, sequence length: 25747 Attention mask shape: torch.Size([1, 1, 25747, 25747]) Position ids shape: torch.Size([1, 25747]) Input IDs shape: torch.Size([1, 25747]) Labels shape: torch.Size([1, 25747]) Final batch size: 1, sequence length: 24988 Attention mask shape: torch.Size([1, 1, 24988, 24988]) Position ids shape: torch.Size([1, 24988]) Input IDs shape: torch.Size([1, 24988]) Labels shape: torch.Size([1, 24988]) Final batch size: 1, sequence length: 10719 Attention mask shape: torch.Size([1, 1, 10719, 10719]) Position ids shape: torch.Size([1, 10719]) Input IDs shape: torch.Size([1, 10719]) Labels shape: torch.Size([1, 10719]) Final batch size: 1, sequence length: 27447 Attention mask shape: torch.Size([1, 1, 27447, 27447]) Position ids shape: torch.Size([1, 27447]) Input IDs shape: torch.Size([1, 27447]) Labels shape: torch.Size([1, 27447]) Final batch size: 1, sequence length: 11184 Attention mask shape: torch.Size([1, 1, 11184, 11184]) Position ids shape: torch.Size([1, 11184]) Input IDs shape: torch.Size([1, 11184]) Labels shape: torch.Size([1, 11184]) Final batch size: 1, sequence length: 25477 Attention mask shape: torch.Size([1, 1, 25477, 25477]) Position ids shape: torch.Size([1, 25477]) Input IDs shape: torch.Size([1, 25477]) Labels shape: torch.Size([1, 25477]) Final batch size: 1, sequence length: 15317 Attention mask shape: torch.Size([1, 1, 15317, 15317]) Position ids shape: torch.Size([1, 15317]) Input IDs shape: torch.Size([1, 15317]) Labels shape: torch.Size([1, 15317]) Final batch size: 1, sequence length: 16915 Attention mask shape: torch.Size([1, 1, 16915, 16915]) Position ids shape: torch.Size([1, 16915]) Input IDs shape: torch.Size([1, 16915]) Labels shape: torch.Size([1, 16915]) Final batch size: 1, sequence length: 25651 Attention mask shape: torch.Size([1, 1, 25651, 25651]) Position ids shape: torch.Size([1, 25651]) Input IDs shape: torch.Size([1, 25651]) Labels shape: torch.Size([1, 25651]) Final batch size: 1, sequence length: 26479 Attention mask shape: torch.Size([1, 1, 26479, 26479]) Position ids shape: torch.Size([1, 26479]) Input IDs shape: torch.Size([1, 26479]) Labels shape: torch.Size([1, 26479]) Final batch size: 1, sequence length: 19552 Attention mask shape: torch.Size([1, 1, 19552, 19552]) Position ids shape: torch.Size([1, 19552]) Input IDs shape: torch.Size([1, 19552]) Labels shape: torch.Size([1, 19552]) Final batch size: 1, sequence length: 26663 Attention mask shape: torch.Size([1, 1, 26663, 26663]) Position ids shape: torch.Size([1, 26663]) Input IDs shape: torch.Size([1, 26663]) Labels shape: torch.Size([1, 26663]) Final batch size: 1, sequence length: 27480 Attention mask shape: torch.Size([1, 1, 27480, 27480]) Position ids shape: torch.Size([1, 27480]) Input IDs shape: torch.Size([1, 27480]) Labels shape: torch.Size([1, 27480]) Final batch size: 1, sequence length: 28777 Attention mask shape: torch.Size([1, 1, 28777, 28777]) Position ids shape: torch.Size([1, 28777]) Input IDs shape: torch.Size([1, 28777]) Labels shape: torch.Size([1, 28777]) Final batch size: 1, sequence length: 17395 Attention mask shape: torch.Size([1, 1, 17395, 17395]) Position ids shape: torch.Size([1, 17395]) Input IDs shape: torch.Size([1, 17395]) Labels shape: torch.Size([1, 17395]) Final batch size: 1, sequence length: 27334 Attention mask shape: torch.Size([1, 1, 27334, 27334]) Position ids shape: torch.Size([1, 27334]) Input IDs shape: torch.Size([1, 27334]) Labels shape: torch.Size([1, 27334]) Final batch size: 1, sequence length: 19869 Attention mask shape: torch.Size([1, 1, 19869, 19869]) Position ids shape: torch.Size([1, 19869]) Input IDs shape: torch.Size([1, 19869]) Labels shape: torch.Size([1, 19869]) Final batch size: 1, sequence length: 30031 Attention mask shape: torch.Size([1, 1, 30031, 30031]) Position ids shape: torch.Size([1, 30031]) Input IDs shape: torch.Size([1, 30031]) Labels shape: torch.Size([1, 30031]) Final batch size: 1, sequence length: 30981 Attention mask shape: torch.Size([1, 1, 30981, 30981]) Position ids shape: torch.Size([1, 30981]) Input IDs shape: torch.Size([1, 30981]) Labels shape: torch.Size([1, 30981]) Final batch size: 1, sequence length: 17911 Attention mask shape: torch.Size([1, 1, 17911, 17911]) Position ids shape: torch.Size([1, 17911]) Input IDs shape: torch.Size([1, 17911]) Labels shape: torch.Size([1, 17911]) Final batch size: 1, sequence length: 16953 Attention mask shape: torch.Size([1, 1, 16953, 16953]) Position ids shape: torch.Size([1, 16953]) Input IDs shape: torch.Size([1, 16953]) Labels shape: torch.Size([1, 16953]) Final batch size: 1, sequence length: 18376 Attention mask shape: torch.Size([1, 1, 18376, 18376]) Position ids shape: torch.Size([1, 18376]) Input IDs shape: torch.Size([1, 18376]) Labels shape: torch.Size([1, 18376]) Final batch size: 1, sequence length: 26033 Attention mask shape: torch.Size([1, 1, 26033, 26033]) Position ids shape: torch.Size([1, 26033]) Input IDs shape: torch.Size([1, 26033]) Labels shape: torch.Size([1, 26033]) Final batch size: 1, sequence length: 15924 Attention mask shape: torch.Size([1, 1, 15924, 15924]) Position ids shape: torch.Size([1, 15924]) Input IDs shape: torch.Size([1, 15924]) Labels shape: torch.Size([1, 15924]) Final batch size: 1, sequence length: 24782 Attention mask shape: torch.Size([1, 1, 24782, 24782]) Position ids shape: torch.Size([1, 24782]) Input IDs shape: torch.Size([1, 24782]) Labels shape: torch.Size([1, 24782]) Final batch size: 1, sequence length: 32466 Attention mask shape: torch.Size([1, 1, 32466, 32466]) Position ids shape: torch.Size([1, 32466]) Input IDs shape: torch.Size([1, 32466]) Labels shape: torch.Size([1, 32466]) Final batch size: 1, sequence length: 21988 Attention mask shape: torch.Size([1, 1, 21988, 21988]) Position ids shape: torch.Size([1, 21988]) Input IDs shape: torch.Size([1, 21988]) Labels shape: torch.Size([1, 21988]) Final batch size: 1, sequence length: 30601 Attention mask shape: torch.Size([1, 1, 30601, 30601]) Position ids shape: torch.Size([1, 30601]) Input IDs shape: torch.Size([1, 30601]) Labels shape: torch.Size([1, 30601]) Final batch size: 1, sequence length: 14873 Attention mask shape: torch.Size([1, 1, 14873, 14873]) Position ids shape: torch.Size([1, 14873]) Input IDs shape: torch.Size([1, 14873]) Labels shape: torch.Size([1, 14873]) Final batch size: 1, sequence length: 27385 Attention mask shape: torch.Size([1, 1, 27385, 27385]) Position ids shape: torch.Size([1, 27385]) Input IDs shape: torch.Size([1, 27385]) Labels shape: torch.Size([1, 27385]) Final batch size: 1, sequence length: 28060 Attention mask shape: torch.Size([1, 1, 28060, 28060]) Position ids shape: torch.Size([1, 28060]) Input IDs shape: torch.Size([1, 28060]) Labels shape: torch.Size([1, 28060]) Final batch size: 1, sequence length: 12245 Attention mask shape: torch.Size([1, 1, 12245, 12245]) Position ids shape: torch.Size([1, 12245]) Input IDs shape: torch.Size([1, 12245]) Labels shape: torch.Size([1, 12245]) Final batch size: 1, sequence length: 37025 Attention mask shape: torch.Size([1, 1, 37025, 37025]) Position ids shape: torch.Size([1, 37025]) Input IDs shape: torch.Size([1, 37025]) Labels shape: torch.Size([1, 37025]) Final batch size: 1, sequence length: 33725 Attention mask shape: torch.Size([1, 1, 33725, 33725]) Position ids shape: torch.Size([1, 33725]) Input IDs shape: torch.Size([1, 33725]) Labels shape: torch.Size([1, 33725]) Final batch size: 1, sequence length: 36175 Attention mask shape: torch.Size([1, 1, 36175, 36175]) Position ids shape: torch.Size([1, 36175]) Input IDs shape: torch.Size([1, 36175]) Labels shape: torch.Size([1, 36175]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 37738 Attention mask shape: torch.Size([1, 1, 37738, 37738]) Position ids shape: torch.Size([1, 37738]) Input IDs shape: torch.Size([1, 37738]) Labels shape: torch.Size([1, 37738]) Final batch size: 1, sequence length: 32835 Attention mask shape: torch.Size([1, 1, 32835, 32835]) Position ids shape: torch.Size([1, 32835]) Input IDs shape: torch.Size([1, 32835]) Labels shape: torch.Size([1, 32835]) Final batch size: 1, sequence length: 37511 Attention mask shape: torch.Size([1, 1, 37511, 37511]) Position ids shape: torch.Size([1, 37511]) Input IDs shape: torch.Size([1, 37511]) Labels shape: torch.Size([1, 37511]) Final batch size: 1, sequence length: 40397 Attention mask shape: torch.Size([1, 1, 40397, 40397]) Position ids shape: torch.Size([1, 40397]) Input IDs shape: torch.Size([1, 40397]) Labels shape: torch.Size([1, 40397]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 37158 Attention mask shape: torch.Size([1, 1, 37158, 37158]) Position ids shape: torch.Size([1, 37158]) Input IDs shape: torch.Size([1, 37158]) Labels shape: torch.Size([1, 37158]) Final batch size: 1, sequence length: 38986 Attention mask shape: torch.Size([1, 1, 38986, 38986]) Position ids shape: torch.Size([1, 38986]) Input IDs shape: torch.Size([1, 38986]) Labels shape: torch.Size([1, 38986]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 29586 Attention mask shape: torch.Size([1, 1, 29586, 29586]) Position ids shape: torch.Size([1, 29586]) Input IDs shape: torch.Size([1, 29586]) Labels shape: torch.Size([1, 29586]) Final batch size: 1, sequence length: 36130 Attention mask shape: torch.Size([1, 1, 36130, 36130]) Position ids shape: torch.Size([1, 36130]) Input IDs shape: torch.Size([1, 36130]) Labels shape: torch.Size([1, 36130]) Final batch size: 1, sequence length: 31879 Attention mask shape: torch.Size([1, 1, 31879, 31879]) Position ids shape: torch.Size([1, 31879]) Input IDs shape: torch.Size([1, 31879]) Labels shape: torch.Size([1, 31879]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40269 Attention mask shape: torch.Size([1, 1, 40269, 40269]) Position ids shape: torch.Size([1, 40269]) Input IDs shape: torch.Size([1, 40269]) Labels shape: torch.Size([1, 40269]) Final batch size: 1, sequence length: 40237 Attention mask shape: torch.Size([1, 1, 40237, 40237]) Position ids shape: torch.Size([1, 40237]) Input IDs shape: torch.Size([1, 40237]) Labels shape: torch.Size([1, 40237]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31348 Attention mask shape: torch.Size([1, 1, 31348, 31348]) Position ids shape: torch.Size([1, 31348]) Input IDs shape: torch.Size([1, 31348]) Labels shape: torch.Size([1, 31348]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32318 Attention mask shape: torch.Size([1, 1, 32318, 32318]) Position ids shape: torch.Size([1, 32318]) Input IDs shape: torch.Size([1, 32318]) Labels shape: torch.Size([1, 32318]) Final batch size: 1, sequence length: 27291 Attention mask shape: torch.Size([1, 1, 27291, 27291]) Position ids shape: torch.Size([1, 27291]) Input IDs shape: torch.Size([1, 27291]) Labels shape: torch.Size([1, 27291]) Final batch size: 1, sequence length: 20509 Attention mask shape: torch.Size([1, 1, 20509, 20509]) Position ids shape: torch.Size([1, 20509]) Input IDs shape: torch.Size([1, 20509]) Labels shape: torch.Size([1, 20509]) Final batch size: 1, sequence length: 17711 Attention mask shape: torch.Size([1, 1, 17711, 17711]) Position ids shape: torch.Size([1, 17711]) Input IDs shape: torch.Size([1, 17711]) Labels shape: torch.Size([1, 17711]) Final batch size: 1, sequence length: 36947 Attention mask shape: torch.Size([1, 1, 36947, 36947]) Position ids shape: torch.Size([1, 36947]) Input IDs shape: torch.Size([1, 36947]) Labels shape: torch.Size([1, 36947]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28631 Attention mask shape: torch.Size([1, 1, 28631, 28631]) Position ids shape: torch.Size([1, 28631]) Input IDs shape: torch.Size([1, 28631]) Labels shape: torch.Size([1, 28631]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24939 Attention mask shape: torch.Size([1, 1, 24939, 24939]) Position ids shape: torch.Size([1, 24939]) Input IDs shape: torch.Size([1, 24939]) Labels shape: torch.Size([1, 24939]) Final batch size: 1, sequence length: 32247 Attention mask shape: torch.Size([1, 1, 32247, 32247]) Position ids shape: torch.Size([1, 32247]) Input IDs shape: torch.Size([1, 32247]) Labels shape: torch.Size([1, 32247]) Final batch size: 1, sequence length: 21225 Attention mask shape: torch.Size([1, 1, 21225, 21225]) Position ids shape: torch.Size([1, 21225]) Input IDs shape: torch.Size([1, 21225]) Labels shape: torch.Size([1, 21225]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28046 Attention mask shape: torch.Size([1, 1, 28046, 28046]) Position ids shape: torch.Size([1, 28046]) Input IDs shape: torch.Size([1, 28046]) Labels shape: torch.Size([1, 28046]) Final batch size: 1, sequence length: 23740 Attention mask shape: torch.Size([1, 1, 23740, 23740]) Position ids shape: torch.Size([1, 23740]) Input IDs shape: torch.Size([1, 23740]) Labels shape: torch.Size([1, 23740]) Final batch size: 1, sequence length: 22896 Attention mask shape: torch.Size([1, 1, 22896, 22896]) Position ids shape: torch.Size([1, 22896]) Input IDs shape: torch.Size([1, 22896]) Labels shape: torch.Size([1, 22896]) Final batch size: 1, sequence length: 33125 Attention mask shape: torch.Size([1, 1, 33125, 33125]) Position ids shape: torch.Size([1, 33125]) Input IDs shape: torch.Size([1, 33125]) Labels shape: torch.Size([1, 33125]) Final batch size: 1, sequence length: 10198 Attention mask shape: torch.Size([1, 1, 10198, 10198]) Position ids shape: torch.Size([1, 10198]) Input IDs shape: torch.Size([1, 10198]) Labels shape: torch.Size([1, 10198]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22205 Attention mask shape: torch.Size([1, 1, 22205, 22205]) Position ids shape: torch.Size([1, 22205]) Input IDs shape: torch.Size([1, 22205]) Labels shape: torch.Size([1, 22205]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28149 Attention mask shape: torch.Size([1, 1, 28149, 28149]) Position ids shape: torch.Size([1, 28149]) Input IDs shape: torch.Size([1, 28149]) Labels shape: torch.Size([1, 28149]) Final batch size: 1, sequence length: 38360 Attention mask shape: torch.Size([1, 1, 38360, 38360]) Position ids shape: torch.Size([1, 38360]) Input IDs shape: torch.Size([1, 38360]) Labels shape: torch.Size([1, 38360]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32767 Attention mask shape: torch.Size([1, 1, 32767, 32767]) Position ids shape: torch.Size([1, 32767]) Input IDs shape: torch.Size([1, 32767]) Labels shape: torch.Size([1, 32767]) Final batch size: 1, sequence length: 36567 Attention mask shape: torch.Size([1, 1, 36567, 36567]) Position ids shape: torch.Size([1, 36567]) Input IDs shape: torch.Size([1, 36567]) Labels shape: torch.Size([1, 36567]) Final batch size: 1, sequence length: 34268 Attention mask shape: torch.Size([1, 1, 34268, 34268]) Position ids shape: torch.Size([1, 34268]) Input IDs shape: torch.Size([1, 34268]) Labels shape: torch.Size([1, 34268]) Final batch size: 1, sequence length: 32313 Attention mask shape: torch.Size([1, 1, 32313, 32313]) Position ids shape: torch.Size([1, 32313]) Input IDs shape: torch.Size([1, 32313]) Labels shape: torch.Size([1, 32313]) {'loss': 0.3045, 'grad_norm': 0.3799426826809637, 'learning_rate': 7.269952498697734e-06, 'num_tokens': -inf, 'epoch': 3.25} Final batch size: 1, sequence length: 5377 Attention mask shape: torch.Size([1, 1, 5377, 5377]) Position ids shape: torch.Size([1, 5377]) Input IDs shape: torch.Size([1, 5377]) Labels shape: torch.Size([1, 5377]) Final batch size: 1, sequence length: 7998 Attention mask shape: torch.Size([1, 1, 7998, 7998]) Position ids shape: torch.Size([1, 7998]) Input IDs shape: torch.Size([1, 7998]) Labels shape: torch.Size([1, 7998]) Final batch size: 1, sequence length: 7402 Attention mask shape: torch.Size([1, 1, 7402, 7402]) Position ids shape: torch.Size([1, 7402]) Input IDs shape: torch.Size([1, 7402]) Labels shape: torch.Size([1, 7402]) Final batch size: 1, sequence length: 8436 Attention mask shape: torch.Size([1, 1, 8436, 8436]) Position ids shape: torch.Size([1, 8436]) Input IDs shape: torch.Size([1, 8436]) Labels shape: torch.Size([1, 8436]) Final batch size: 1, sequence length: 10576 Attention mask shape: torch.Size([1, 1, 10576, 10576]) Position ids shape: torch.Size([1, 10576]) Input IDs shape: torch.Size([1, 10576]) Labels shape: torch.Size([1, 10576]) Final batch size: 1, sequence length: 8655 Attention mask shape: torch.Size([1, 1, 8655, 8655]) Position ids shape: torch.Size([1, 8655]) Input IDs shape: torch.Size([1, 8655]) Labels shape: torch.Size([1, 8655]) Final batch size: 1, sequence length: 11709 Attention mask shape: torch.Size([1, 1, 11709, 11709]) Position ids shape: torch.Size([1, 11709]) Input IDs shape: torch.Size([1, 11709]) Labels shape: torch.Size([1, 11709]) Final batch size: 1, sequence length: 7344 Attention mask shape: torch.Size([1, 1, 7344, 7344]) Position ids shape: torch.Size([1, 7344]) Input IDs shape: torch.Size([1, 7344]) Labels shape: torch.Size([1, 7344]) Final batch size: 1, sequence length: 11365 Attention mask shape: torch.Size([1, 1, 11365, 11365]) Position ids shape: torch.Size([1, 11365]) Input IDs shape: torch.Size([1, 11365]) Labels shape: torch.Size([1, 11365]) Final batch size: 1, sequence length: 14127 Attention mask shape: torch.Size([1, 1, 14127, 14127]) Position ids shape: torch.Size([1, 14127]) Input IDs shape: torch.Size([1, 14127]) Labels shape: torch.Size([1, 14127]) Final batch size: 1, sequence length: 11678 Attention mask shape: torch.Size([1, 1, 11678, 11678]) Position ids shape: torch.Size([1, 11678]) Input IDs shape: torch.Size([1, 11678]) Labels shape: torch.Size([1, 11678]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 11623 Attention mask shape: torch.Size([1, 1, 11623, 11623]) Position ids shape: torch.Size([1, 11623]) Input IDs shape: torch.Size([1, 11623]) Labels shape: torch.Size([1, 11623]) Final batch size: 1, sequence length: 15079 Attention mask shape: torch.Size([1, 1, 15079, 15079]) Position ids shape: torch.Size([1, 15079]) Input IDs shape: torch.Size([1, 15079]) Labels shape: torch.Size([1, 15079]) Final batch size: 1, sequence length: 15308 Attention mask shape: torch.Size([1, 1, 15308, 15308]) Position ids shape: torch.Size([1, 15308]) Input IDs shape: torch.Size([1, 15308]) Labels shape: torch.Size([1, 15308]) Final batch size: 1, sequence length: 15478 Attention mask shape: torch.Size([1, 1, 15478, 15478]) Position ids shape: torch.Size([1, 15478]) Input IDs shape: torch.Size([1, 15478]) Labels shape: torch.Size([1, 15478]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 16535 Attention mask shape: torch.Size([1, 1, 16535, 16535]) Position ids shape: torch.Size([1, 16535]) Input IDs shape: torch.Size([1, 16535]) Labels shape: torch.Size([1, 16535]) Final batch size: 1, sequence length: 18307 Attention mask shape: torch.Size([1, 1, 18307, 18307]) Position ids shape: torch.Size([1, 18307]) Input IDs shape: torch.Size([1, 18307]) Labels shape: torch.Size([1, 18307]) Final batch size: 1, sequence length: 16060 Attention mask shape: torch.Size([1, 1, 16060, 16060]) Position ids shape: torch.Size([1, 16060]) Input IDs shape: torch.Size([1, 16060]) Labels shape: torch.Size([1, 16060]) Final batch size: 1, sequence length: 18953 Attention mask shape: torch.Size([1, 1, 18953, 18953]) Position ids shape: torch.Size([1, 18953]) Input IDs shape: torch.Size([1, 18953]) Labels shape: torch.Size([1, 18953]) Final batch size: 1, sequence length: 19170 Attention mask shape: torch.Size([1, 1, 19170, 19170]) Position ids shape: torch.Size([1, 19170]) Input IDs shape: torch.Size([1, 19170]) Labels shape: torch.Size([1, 19170]) Final batch size: 1, sequence length: 13639 Attention mask shape: torch.Size([1, 1, 13639, 13639]) Position ids shape: torch.Size([1, 13639]) Input IDs shape: torch.Size([1, 13639]) Labels shape: torch.Size([1, 13639]) Final batch size: 1, sequence length: 17058 Attention mask shape: torch.Size([1, 1, 17058, 17058]) Position ids shape: torch.Size([1, 17058]) Input IDs shape: torch.Size([1, 17058]) Labels shape: torch.Size([1, 17058]) Final batch size: 1, sequence length: 14648 Attention mask shape: torch.Size([1, 1, 14648, 14648]) Position ids shape: torch.Size([1, 14648]) Input IDs shape: torch.Size([1, 14648]) Labels shape: torch.Size([1, 14648]) Final batch size: 1, sequence length: 19338 Attention mask shape: torch.Size([1, 1, 19338, 19338]) Position ids shape: torch.Size([1, 19338]) Input IDs shape: torch.Size([1, 19338]) Labels shape: torch.Size([1, 19338]) Final batch size: 1, sequence length: 18836 Attention mask shape: torch.Size([1, 1, 18836, 18836]) Position ids shape: torch.Size([1, 18836]) Input IDs shape: torch.Size([1, 18836]) Labels shape: torch.Size([1, 18836]) Final batch size: 1, sequence length: 8839 Attention mask shape: torch.Size([1, 1, 8839, 8839]) Position ids shape: torch.Size([1, 8839]) Input IDs shape: torch.Size([1, 8839]) Labels shape: torch.Size([1, 8839]) Final batch size: 1, sequence length: 18527 Attention mask shape: torch.Size([1, 1, 18527, 18527]) Position ids shape: torch.Size([1, 18527]) Input IDs shape: torch.Size([1, 18527]) Labels shape: torch.Size([1, 18527]) Final batch size: 1, sequence length: 20770 Attention mask shape: torch.Size([1, 1, 20770, 20770]) Position ids shape: torch.Size([1, 20770]) Input IDs shape: torch.Size([1, 20770]) Labels shape: torch.Size([1, 20770]) Final batch size: 1, sequence length: 22561 Attention mask shape: torch.Size([1, 1, 22561, 22561]) Position ids shape: torch.Size([1, 22561]) Input IDs shape: torch.Size([1, 22561]) Labels shape: torch.Size([1, 22561]) Final batch size: 1, sequence length: 18325 Attention mask shape: torch.Size([1, 1, 18325, 18325]) Position ids shape: torch.Size([1, 18325]) Input IDs shape: torch.Size([1, 18325]) Labels shape: torch.Size([1, 18325]) Final batch size: 1, sequence length: 20854 Attention mask shape: torch.Size([1, 1, 20854, 20854]) Position ids shape: torch.Size([1, 20854]) Input IDs shape: torch.Size([1, 20854]) Labels shape: torch.Size([1, 20854]) Final batch size: 1, sequence length: 12006 Attention mask shape: torch.Size([1, 1, 12006, 12006]) Position ids shape: torch.Size([1, 12006]) Input IDs shape: torch.Size([1, 12006]) Labels shape: torch.Size([1, 12006]) Final batch size: 1, sequence length: 19330 Attention mask shape: torch.Size([1, 1, 19330, 19330]) Position ids shape: torch.Size([1, 19330]) Input IDs shape: torch.Size([1, 19330]) Labels shape: torch.Size([1, 19330]) Final batch size: 1, sequence length: 20714 Attention mask shape: torch.Size([1, 1, 20714, 20714]) Position ids shape: torch.Size([1, 20714]) Input IDs shape: torch.Size([1, 20714]) Labels shape: torch.Size([1, 20714]) Final batch size: 1, sequence length: 19138 Attention mask shape: torch.Size([1, 1, 19138, 19138]) Position ids shape: torch.Size([1, 19138]) Input IDs shape: torch.Size([1, 19138]) Labels shape: torch.Size([1, 19138]) Final batch size: 1, sequence length: 21982 Attention mask shape: torch.Size([1, 1, 21982, 21982]) Position ids shape: torch.Size([1, 21982]) Input IDs shape: torch.Size([1, 21982]) Labels shape: torch.Size([1, 21982]) Final batch size: 1, sequence length: 18823 Attention mask shape: torch.Size([1, 1, 18823, 18823]) Position ids shape: torch.Size([1, 18823]) Input IDs shape: torch.Size([1, 18823]) Labels shape: torch.Size([1, 18823]) Final batch size: 1, sequence length: 23560 Attention mask shape: torch.Size([1, 1, 23560, 23560]) Position ids shape: torch.Size([1, 23560]) Input IDs shape: torch.Size([1, 23560]) Labels shape: torch.Size([1, 23560]) Final batch size: 1, sequence length: 21858 Attention mask shape: torch.Size([1, 1, 21858, 21858]) Position ids shape: torch.Size([1, 21858]) Input IDs shape: torch.Size([1, 21858]) Labels shape: torch.Size([1, 21858]) Final batch size: 1, sequence length: 18395 Attention mask shape: torch.Size([1, 1, 18395, 18395]) Position ids shape: torch.Size([1, 18395]) Input IDs shape: torch.Size([1, 18395]) Labels shape: torch.Size([1, 18395]) Final batch size: 1, sequence length: 25451 Attention mask shape: torch.Size([1, 1, 25451, 25451]) Position ids shape: torch.Size([1, 25451]) Input IDs shape: torch.Size([1, 25451]) Labels shape: torch.Size([1, 25451]) Final batch size: 1, sequence length: 26356 Attention mask shape: torch.Size([1, 1, 26356, 26356]) Position ids shape: torch.Size([1, 26356]) Input IDs shape: torch.Size([1, 26356]) Labels shape: torch.Size([1, 26356]) Final batch size: 1, sequence length: 24248 Attention mask shape: torch.Size([1, 1, 24248, 24248]) Position ids shape: torch.Size([1, 24248]) Input IDs shape: torch.Size([1, 24248]) Labels shape: torch.Size([1, 24248]) Final batch size: 1, sequence length: 15913 Attention mask shape: torch.Size([1, 1, 15913, 15913]) Position ids shape: torch.Size([1, 15913]) Input IDs shape: torch.Size([1, 15913]) Labels shape: torch.Size([1, 15913]) Final batch size: 1, sequence length: 21405 Attention mask shape: torch.Size([1, 1, 21405, 21405]) Position ids shape: torch.Size([1, 21405]) Input IDs shape: torch.Size([1, 21405]) Labels shape: torch.Size([1, 21405]) Final batch size: 1, sequence length: 24433 Attention mask shape: torch.Size([1, 1, 24433, 24433]) Position ids shape: torch.Size([1, 24433]) Input IDs shape: torch.Size([1, 24433]) Labels shape: torch.Size([1, 24433]) Final batch size: 1, sequence length: 23694 Attention mask shape: torch.Size([1, 1, 23694, 23694]) Position ids shape: torch.Size([1, 23694]) Input IDs shape: torch.Size([1, 23694]) Labels shape: torch.Size([1, 23694]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 29098 Attention mask shape: torch.Size([1, 1, 29098, 29098]) Position ids shape: torch.Size([1, 29098]) Input IDs shape: torch.Size([1, 29098]) Labels shape: torch.Size([1, 29098]) Final batch size: 1, sequence length: 9380 Attention mask shape: torch.Size([1, 1, 9380, 9380]) Position ids shape: torch.Size([1, 9380]) Input IDs shape: torch.Size([1, 9380]) Labels shape: torch.Size([1, 9380]) Final batch size: 1, sequence length: 22735 Attention mask shape: torch.Size([1, 1, 22735, 22735]) Position ids shape: torch.Size([1, 22735]) Input IDs shape: torch.Size([1, 22735]) Labels shape: torch.Size([1, 22735]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 19239 Attention mask shape: torch.Size([1, 1, 19239, 19239]) Position ids shape: torch.Size([1, 19239]) Input IDs shape: torch.Size([1, 19239]) Labels shape: torch.Size([1, 19239]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 5801 Attention mask shape: torch.Size([1, 1, 5801, 5801]) Position ids shape: torch.Size([1, 5801]) Input IDs shape: torch.Size([1, 5801]) Labels shape: torch.Size([1, 5801]) Final batch size: 1, sequence length: 31414 Attention mask shape: torch.Size([1, 1, 31414, 31414]) Position ids shape: torch.Size([1, 31414]) Input IDs shape: torch.Size([1, 31414]) Labels shape: torch.Size([1, 31414]) Final batch size: 1, sequence length: 32885 Attention mask shape: torch.Size([1, 1, 32885, 32885]) Position ids shape: torch.Size([1, 32885]) Input IDs shape: torch.Size([1, 32885]) Labels shape: torch.Size([1, 32885]) Final batch size: 1, sequence length: 21615 Attention mask shape: torch.Size([1, 1, 21615, 21615]) Position ids shape: torch.Size([1, 21615]) Input IDs shape: torch.Size([1, 21615]) Labels shape: torch.Size([1, 21615]) Final batch size: 1, sequence length: 28684 Attention mask shape: torch.Size([1, 1, 28684, 28684]) Position ids shape: torch.Size([1, 28684]) Input IDs shape: torch.Size([1, 28684]) Labels shape: torch.Size([1, 28684]) Final batch size: 1, sequence length: 19045 Attention mask shape: torch.Size([1, 1, 19045, 19045]) Position ids shape: torch.Size([1, 19045]) Input IDs shape: torch.Size([1, 19045]) Labels shape: torch.Size([1, 19045]) Final batch size: 1, sequence length: 32328 Attention mask shape: torch.Size([1, 1, 32328, 32328]) Position ids shape: torch.Size([1, 32328]) Input IDs shape: torch.Size([1, 32328]) Labels shape: torch.Size([1, 32328]) Final batch size: 1, sequence length: 30687 Attention mask shape: torch.Size([1, 1, 30687, 30687]) Position ids shape: torch.Size([1, 30687]) Input IDs shape: torch.Size([1, 30687]) Labels shape: torch.Size([1, 30687]) Final batch size: 1, sequence length: 17465 Attention mask shape: torch.Size([1, 1, 17465, 17465]) Position ids shape: torch.Size([1, 17465]) Input IDs shape: torch.Size([1, 17465]) Labels shape: torch.Size([1, 17465]) Final batch size: 1, sequence length: 23881 Attention mask shape: torch.Size([1, 1, 23881, 23881]) Position ids shape: torch.Size([1, 23881]) Input IDs shape: torch.Size([1, 23881]) Labels shape: torch.Size([1, 23881]) Final batch size: 1, sequence length: 15875 Attention mask shape: torch.Size([1, 1, 15875, 15875]) Position ids shape: torch.Size([1, 15875]) Input IDs shape: torch.Size([1, 15875]) Labels shape: torch.Size([1, 15875]) Final batch size: 1, sequence length: 26072 Attention mask shape: torch.Size([1, 1, 26072, 26072]) Position ids shape: torch.Size([1, 26072]) Input IDs shape: torch.Size([1, 26072]) Labels shape: torch.Size([1, 26072]) Final batch size: 1, sequence length: 16050 Attention mask shape: torch.Size([1, 1, 16050, 16050]) Position ids shape: torch.Size([1, 16050]) Input IDs shape: torch.Size([1, 16050]) Labels shape: torch.Size([1, 16050]) Final batch size: 1, sequence length: 33871 Attention mask shape: torch.Size([1, 1, 33871, 33871]) Position ids shape: torch.Size([1, 33871]) Input IDs shape: torch.Size([1, 33871]) Labels shape: torch.Size([1, 33871]) Final batch size: 1, sequence length: 25674 Attention mask shape: torch.Size([1, 1, 25674, 25674]) Position ids shape: torch.Size([1, 25674]) Input IDs shape: torch.Size([1, 25674]) Labels shape: torch.Size([1, 25674]) Final batch size: 1, sequence length: 31860 Attention mask shape: torch.Size([1, 1, 31860, 31860]) Position ids shape: torch.Size([1, 31860]) Input IDs shape: torch.Size([1, 31860]) Labels shape: torch.Size([1, 31860]) Final batch size: 1, sequence length: 29355 Attention mask shape: torch.Size([1, 1, 29355, 29355]) Position ids shape: torch.Size([1, 29355]) Input IDs shape: torch.Size([1, 29355]) Labels shape: torch.Size([1, 29355]) Final batch size: 1, sequence length: 29150 Attention mask shape: torch.Size([1, 1, 29150, 29150]) Position ids shape: torch.Size([1, 29150]) Input IDs shape: torch.Size([1, 29150]) Labels shape: torch.Size([1, 29150]) Final batch size: 1, sequence length: 27972 Attention mask shape: torch.Size([1, 1, 27972, 27972]) Position ids shape: torch.Size([1, 27972]) Input IDs shape: torch.Size([1, 27972]) Labels shape: torch.Size([1, 27972]) Final batch size: 1, sequence length: 32529 Attention mask shape: torch.Size([1, 1, 32529, 32529]) Position ids shape: torch.Size([1, 32529]) Input IDs shape: torch.Size([1, 32529]) Labels shape: torch.Size([1, 32529]) Final batch size: 1, sequence length: 35397 Attention mask shape: torch.Size([1, 1, 35397, 35397]) Position ids shape: torch.Size([1, 35397]) Input IDs shape: torch.Size([1, 35397]) Labels shape: torch.Size([1, 35397]) Final batch size: 1, sequence length: 30689 Attention mask shape: torch.Size([1, 1, 30689, 30689]) Position ids shape: torch.Size([1, 30689]) Input IDs shape: torch.Size([1, 30689]) Labels shape: torch.Size([1, 30689]) Final batch size: 1, sequence length: 32636 Attention mask shape: torch.Size([1, 1, 32636, 32636]) Position ids shape: torch.Size([1, 32636]) Input IDs shape: torch.Size([1, 32636]) Labels shape: torch.Size([1, 32636]) Final batch size: 1, sequence length: 26689 Attention mask shape: torch.Size([1, 1, 26689, 26689]) Position ids shape: torch.Size([1, 26689]) Input IDs shape: torch.Size([1, 26689]) Labels shape: torch.Size([1, 26689]) Final batch size: 1, sequence length: 31712 Attention mask shape: torch.Size([1, 1, 31712, 31712]) Position ids shape: torch.Size([1, 31712]) Input IDs shape: torch.Size([1, 31712]) Labels shape: torch.Size([1, 31712]) Final batch size: 1, sequence length: 34921 Attention mask shape: torch.Size([1, 1, 34921, 34921]) Position ids shape: torch.Size([1, 34921]) Input IDs shape: torch.Size([1, 34921]) Labels shape: torch.Size([1, 34921]) Final batch size: 1, sequence length: 30346 Attention mask shape: torch.Size([1, 1, 30346, 30346]) Position ids shape: torch.Size([1, 30346]) Input IDs shape: torch.Size([1, 30346]) Labels shape: torch.Size([1, 30346]) Final batch size: 1, sequence length: 34581 Attention mask shape: torch.Size([1, 1, 34581, 34581]) Position ids shape: torch.Size([1, 34581]) Input IDs shape: torch.Size([1, 34581]) Labels shape: torch.Size([1, 34581]) Final batch size: 1, sequence length: 15077 Attention mask shape: torch.Size([1, 1, 15077, 15077]) Position ids shape: torch.Size([1, 15077]) Input IDs shape: torch.Size([1, 15077]) Labels shape: torch.Size([1, 15077]) Final batch size: 1, sequence length: 24781 Attention mask shape: torch.Size([1, 1, 24781, 24781]) Position ids shape: torch.Size([1, 24781]) Input IDs shape: torch.Size([1, 24781]) Labels shape: torch.Size([1, 24781]) Final batch size: 1, sequence length: 22264 Attention mask shape: torch.Size([1, 1, 22264, 22264]) Position ids shape: torch.Size([1, 22264]) Input IDs shape: torch.Size([1, 22264]) Labels shape: torch.Size([1, 22264]) Final batch size: 1, sequence length: 18963 Attention mask shape: torch.Size([1, 1, 18963, 18963]) Position ids shape: torch.Size([1, 18963]) Input IDs shape: torch.Size([1, 18963]) Labels shape: torch.Size([1, 18963]) Final batch size: 1, sequence length: 18177 Attention mask shape: torch.Size([1, 1, 18177, 18177]) Position ids shape: torch.Size([1, 18177]) Input IDs shape: torch.Size([1, 18177]) Labels shape: torch.Size([1, 18177]) Final batch size: 1, sequence length: 25741 Attention mask shape: torch.Size([1, 1, 25741, 25741]) Position ids shape: torch.Size([1, 25741]) Input IDs shape: torch.Size([1, 25741]) Labels shape: torch.Size([1, 25741]) Final batch size: 1, sequence length: 16025 Attention mask shape: torch.Size([1, 1, 16025, 16025]) Position ids shape: torch.Size([1, 16025]) Input IDs shape: torch.Size([1, 16025]) Labels shape: torch.Size([1, 16025]) Final batch size: 1, sequence length: 17243 Attention mask shape: torch.Size([1, 1, 17243, 17243]) Position ids shape: torch.Size([1, 17243]) Input IDs shape: torch.Size([1, 17243]) Labels shape: torch.Size([1, 17243]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 14976 Attention mask shape: torch.Size([1, 1, 14976, 14976]) Position ids shape: torch.Size([1, 14976]) Input IDs shape: torch.Size([1, 14976]) Labels shape: torch.Size([1, 14976]) Final batch size: 1, sequence length: 28018 Attention mask shape: torch.Size([1, 1, 28018, 28018]) Position ids shape: torch.Size([1, 28018]) Input IDs shape: torch.Size([1, 28018]) Labels shape: torch.Size([1, 28018]) Final batch size: 1, sequence length: 31022 Attention mask shape: torch.Size([1, 1, 31022, 31022]) Position ids shape: torch.Size([1, 31022]) Input IDs shape: torch.Size([1, 31022]) Labels shape: torch.Size([1, 31022]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40488 Attention mask shape: torch.Size([1, 1, 40488, 40488]) Position ids shape: torch.Size([1, 40488]) Input IDs shape: torch.Size([1, 40488]) Labels shape: torch.Size([1, 40488]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26922 Attention mask shape: torch.Size([1, 1, 26922, 26922]) Position ids shape: torch.Size([1, 26922]) Input IDs shape: torch.Size([1, 26922]) Labels shape: torch.Size([1, 26922]) Final batch size: 1, sequence length: 24002 Attention mask shape: torch.Size([1, 1, 24002, 24002]) Position ids shape: torch.Size([1, 24002]) Input IDs shape: torch.Size([1, 24002]) Labels shape: torch.Size([1, 24002]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36596 Attention mask shape: torch.Size([1, 1, 36596, 36596]) Position ids shape: torch.Size([1, 36596]) Input IDs shape: torch.Size([1, 36596]) Labels shape: torch.Size([1, 36596]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 11611 Attention mask shape: torch.Size([1, 1, 11611, 11611]) Position ids shape: torch.Size([1, 11611]) Input IDs shape: torch.Size([1, 11611]) Labels shape: torch.Size([1, 11611]) Final batch size: 1, sequence length: 28397 Attention mask shape: torch.Size([1, 1, 28397, 28397]) Position ids shape: torch.Size([1, 28397]) Input IDs shape: torch.Size([1, 28397]) Labels shape: torch.Size([1, 28397]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24043 Attention mask shape: torch.Size([1, 1, 24043, 24043]) Position ids shape: torch.Size([1, 24043]) Input IDs shape: torch.Size([1, 24043]) Labels shape: torch.Size([1, 24043]) Final batch size: 1, sequence length: 28086 Attention mask shape: torch.Size([1, 1, 28086, 28086]) Position ids shape: torch.Size([1, 28086]) Input IDs shape: torch.Size([1, 28086]) Labels shape: torch.Size([1, 28086]) Final batch size: 1, sequence length: 34853 Attention mask shape: torch.Size([1, 1, 34853, 34853]) Position ids shape: torch.Size([1, 34853]) Input IDs shape: torch.Size([1, 34853]) Labels shape: torch.Size([1, 34853]) Final batch size: 1, sequence length: 16337 Attention mask shape: torch.Size([1, 1, 16337, 16337]) Position ids shape: torch.Size([1, 16337]) Input IDs shape: torch.Size([1, 16337]) Labels shape: torch.Size([1, 16337]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 12218 Attention mask shape: torch.Size([1, 1, 12218, 12218]) Position ids shape: torch.Size([1, 12218]) Input IDs shape: torch.Size([1, 12218]) Labels shape: torch.Size([1, 12218]) Final batch size: 1, sequence length: 36334 Attention mask shape: torch.Size([1, 1, 36334, 36334]) Position ids shape: torch.Size([1, 36334]) Input IDs shape: torch.Size([1, 36334]) Labels shape: torch.Size([1, 36334]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32917 Attention mask shape: torch.Size([1, 1, 32917, 32917]) Position ids shape: torch.Size([1, 32917]) Input IDs shape: torch.Size([1, 32917]) Labels shape: torch.Size([1, 32917]) {'loss': 0.2861, 'grad_norm': 0.3241177604695199, 'learning_rate': 7.033683215379002e-06, 'num_tokens': -inf, 'epoch': 3.38} Final batch size: 1, sequence length: 5818 Attention mask shape: torch.Size([1, 1, 5818, 5818]) Position ids shape: torch.Size([1, 5818]) Input IDs shape: torch.Size([1, 5818]) Labels shape: torch.Size([1, 5818]) Final batch size: 1, sequence length: 6215 Attention mask shape: torch.Size([1, 1, 6215, 6215]) Position ids shape: torch.Size([1, 6215]) Input IDs shape: torch.Size([1, 6215]) Labels shape: torch.Size([1, 6215]) Final batch size: 1, sequence length: 6871 Attention mask shape: torch.Size([1, 1, 6871, 6871]) Position ids shape: torch.Size([1, 6871]) Input IDs shape: torch.Size([1, 6871]) Labels shape: torch.Size([1, 6871]) Final batch size: 1, sequence length: 10107 Attention mask shape: torch.Size([1, 1, 10107, 10107]) Position ids shape: torch.Size([1, 10107]) Input IDs shape: torch.Size([1, 10107]) Labels shape: torch.Size([1, 10107]) Final batch size: 1, sequence length: 5917 Attention mask shape: torch.Size([1, 1, 5917, 5917]) Position ids shape: torch.Size([1, 5917]) Input IDs shape: torch.Size([1, 5917]) Labels shape: torch.Size([1, 5917]) Final batch size: 1, sequence length: 6034 Attention mask shape: torch.Size([1, 1, 6034, 6034]) Position ids shape: torch.Size([1, 6034]) Input IDs shape: torch.Size([1, 6034]) Labels shape: torch.Size([1, 6034]) Final batch size: 1, sequence length: 8623 Attention mask shape: torch.Size([1, 1, 8623, 8623]) Position ids shape: torch.Size([1, 8623]) Input IDs shape: torch.Size([1, 8623]) Labels shape: torch.Size([1, 8623]) Final batch size: 1, sequence length: 12317 Attention mask shape: torch.Size([1, 1, 12317, 12317]) Position ids shape: torch.Size([1, 12317]) Input IDs shape: torch.Size([1, 12317]) Labels shape: torch.Size([1, 12317]) Final batch size: 1, sequence length: 13202 Attention mask shape: torch.Size([1, 1, 13202, 13202]) Position ids shape: torch.Size([1, 13202]) Input IDs shape: torch.Size([1, 13202]) Labels shape: torch.Size([1, 13202]) Final batch size: 1, sequence length: 9517 Attention mask shape: torch.Size([1, 1, 9517, 9517]) Position ids shape: torch.Size([1, 9517]) Input IDs shape: torch.Size([1, 9517]) Labels shape: torch.Size([1, 9517]) Final batch size: 1, sequence length: 11616 Attention mask shape: torch.Size([1, 1, 11616, 11616]) Position ids shape: torch.Size([1, 11616]) Input IDs shape: torch.Size([1, 11616]) Labels shape: torch.Size([1, 11616]) Final batch size: 1, sequence length: 15029 Attention mask shape: torch.Size([1, 1, 15029, 15029]) Position ids shape: torch.Size([1, 15029]) Input IDs shape: torch.Size([1, 15029]) Labels shape: torch.Size([1, 15029]) Final batch size: 1, sequence length: 12813 Attention mask shape: torch.Size([1, 1, 12813, 12813]) Position ids shape: torch.Size([1, 12813]) Input IDs shape: torch.Size([1, 12813]) Labels shape: torch.Size([1, 12813]) Final batch size: 1, sequence length: 12275 Attention mask shape: torch.Size([1, 1, 12275, 12275]) Position ids shape: torch.Size([1, 12275]) Input IDs shape: torch.Size([1, 12275]) Labels shape: torch.Size([1, 12275]) Final batch size: 1, sequence length: 11648 Attention mask shape: torch.Size([1, 1, 11648, 11648]) Position ids shape: torch.Size([1, 11648]) Input IDs shape: torch.Size([1, 11648]) Labels shape: torch.Size([1, 11648]) Final batch size: 1, sequence length: 8117 Attention mask shape: torch.Size([1, 1, 8117, 8117]) Position ids shape: torch.Size([1, 8117]) Input IDs shape: torch.Size([1, 8117]) Labels shape: torch.Size([1, 8117]) Final batch size: 1, sequence length: 13428 Attention mask shape: torch.Size([1, 1, 13428, 13428]) Position ids shape: torch.Size([1, 13428]) Input IDs shape: torch.Size([1, 13428]) Labels shape: torch.Size([1, 13428]) Final batch size: 1, sequence length: 15066 Attention mask shape: torch.Size([1, 1, 15066, 15066]) Position ids shape: torch.Size([1, 15066]) Input IDs shape: torch.Size([1, 15066]) Labels shape: torch.Size([1, 15066]) Final batch size: 1, sequence length: 10136 Attention mask shape: torch.Size([1, 1, 10136, 10136]) Position ids shape: torch.Size([1, 10136]) Input IDs shape: torch.Size([1, 10136]) Labels shape: torch.Size([1, 10136]) Final batch size: 1, sequence length: 16299 Attention mask shape: torch.Size([1, 1, 16299, 16299]) Position ids shape: torch.Size([1, 16299]) Input IDs shape: torch.Size([1, 16299]) Labels shape: torch.Size([1, 16299]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 16331 Attention mask shape: torch.Size([1, 1, 16331, 16331]) Position ids shape: torch.Size([1, 16331]) Input IDs shape: torch.Size([1, 16331]) Labels shape: torch.Size([1, 16331]) Final batch size: 1, sequence length: 12553 Attention mask shape: torch.Size([1, 1, 12553, 12553]) Position ids shape: torch.Size([1, 12553]) Input IDs shape: torch.Size([1, 12553]) Labels shape: torch.Size([1, 12553]) Final batch size: 1, sequence length: 15499 Attention mask shape: torch.Size([1, 1, 15499, 15499]) Position ids shape: torch.Size([1, 15499]) Input IDs shape: torch.Size([1, 15499]) Labels shape: torch.Size([1, 15499]) Final batch size: 1, sequence length: 18414 Attention mask shape: torch.Size([1, 1, 18414, 18414]) Position ids shape: torch.Size([1, 18414]) Input IDs shape: torch.Size([1, 18414]) Labels shape: torch.Size([1, 18414]) Final batch size: 1, sequence length: 19847 Attention mask shape: torch.Size([1, 1, 19847, 19847]) Position ids shape: torch.Size([1, 19847]) Input IDs shape: torch.Size([1, 19847]) Labels shape: torch.Size([1, 19847]) Final batch size: 1, sequence length: 19221 Attention mask shape: torch.Size([1, 1, 19221, 19221]) Position ids shape: torch.Size([1, 19221]) Input IDs shape: torch.Size([1, 19221]) Labels shape: torch.Size([1, 19221]) Final batch size: 1, sequence length: 15226 Attention mask shape: torch.Size([1, 1, 15226, 15226]) Position ids shape: torch.Size([1, 15226]) Input IDs shape: torch.Size([1, 15226]) Labels shape: torch.Size([1, 15226]) Final batch size: 1, sequence length: 17587 Attention mask shape: torch.Size([1, 1, 17587, 17587]) Position ids shape: torch.Size([1, 17587]) Input IDs shape: torch.Size([1, 17587]) Labels shape: torch.Size([1, 17587]) Final batch size: 1, sequence length: 19512 Attention mask shape: torch.Size([1, 1, 19512, 19512]) Position ids shape: torch.Size([1, 19512]) Input IDs shape: torch.Size([1, 19512]) Labels shape: torch.Size([1, 19512]) Final batch size: 1, sequence length: 22079 Attention mask shape: torch.Size([1, 1, 22079, 22079]) Position ids shape: torch.Size([1, 22079]) Input IDs shape: torch.Size([1, 22079]) Labels shape: torch.Size([1, 22079]) Final batch size: 1, sequence length: 21556 Attention mask shape: torch.Size([1, 1, 21556, 21556]) Position ids shape: torch.Size([1, 21556]) Input IDs shape: torch.Size([1, 21556]) Labels shape: torch.Size([1, 21556]) Final batch size: 1, sequence length: 18438 Attention mask shape: torch.Size([1, 1, 18438, 18438]) Position ids shape: torch.Size([1, 18438]) Input IDs shape: torch.Size([1, 18438]) Labels shape: torch.Size([1, 18438]) Final batch size: 1, sequence length: 21028 Attention mask shape: torch.Size([1, 1, 21028, 21028]) Position ids shape: torch.Size([1, 21028]) Input IDs shape: torch.Size([1, 21028]) Labels shape: torch.Size([1, 21028]) Final batch size: 1, sequence length: 18469 Attention mask shape: torch.Size([1, 1, 18469, 18469]) Position ids shape: torch.Size([1, 18469]) Input IDs shape: torch.Size([1, 18469]) Labels shape: torch.Size([1, 18469]) Final batch size: 1, sequence length: 22618 Attention mask shape: torch.Size([1, 1, 22618, 22618]) Position ids shape: torch.Size([1, 22618]) Input IDs shape: torch.Size([1, 22618]) Labels shape: torch.Size([1, 22618]) Final batch size: 1, sequence length: 22138 Attention mask shape: torch.Size([1, 1, 22138, 22138]) Position ids shape: torch.Size([1, 22138]) Input IDs shape: torch.Size([1, 22138]) Labels shape: torch.Size([1, 22138]) Final batch size: 1, sequence length: 15221 Attention mask shape: torch.Size([1, 1, 15221, 15221]) Position ids shape: torch.Size([1, 15221]) Input IDs shape: torch.Size([1, 15221]) Labels shape: torch.Size([1, 15221]) Final batch size: 1, sequence length: 23766 Attention mask shape: torch.Size([1, 1, 23766, 23766]) Position ids shape: torch.Size([1, 23766]) Input IDs shape: torch.Size([1, 23766]) Labels shape: torch.Size([1, 23766]) Final batch size: 1, sequence length: 10269 Attention mask shape: torch.Size([1, 1, 10269, 10269]) Position ids shape: torch.Size([1, 10269]) Input IDs shape: torch.Size([1, 10269]) Labels shape: torch.Size([1, 10269]) Final batch size: 1, sequence length: 24499 Attention mask shape: torch.Size([1, 1, 24499, 24499]) Position ids shape: torch.Size([1, 24499]) Input IDs shape: torch.Size([1, 24499]) Labels shape: torch.Size([1, 24499]) Final batch size: 1, sequence length: 23973 Attention mask shape: torch.Size([1, 1, 23973, 23973]) Position ids shape: torch.Size([1, 23973]) Input IDs shape: torch.Size([1, 23973]) Labels shape: torch.Size([1, 23973]) Final batch size: 1, sequence length: 22718 Attention mask shape: torch.Size([1, 1, 22718, 22718]) Position ids shape: torch.Size([1, 22718]) Input IDs shape: torch.Size([1, 22718]) Labels shape: torch.Size([1, 22718]) Final batch size: 1, sequence length: 25622 Attention mask shape: torch.Size([1, 1, 25622, 25622]) Position ids shape: torch.Size([1, 25622]) Input IDs shape: torch.Size([1, 25622]) Labels shape: torch.Size([1, 25622]) Final batch size: 1, sequence length: 19428 Attention mask shape: torch.Size([1, 1, 19428, 19428]) Position ids shape: torch.Size([1, 19428]) Input IDs shape: torch.Size([1, 19428]) Labels shape: torch.Size([1, 19428]) Final batch size: 1, sequence length: 24694 Attention mask shape: torch.Size([1, 1, 24694, 24694]) Position ids shape: torch.Size([1, 24694]) Input IDs shape: torch.Size([1, 24694]) Labels shape: torch.Size([1, 24694]) Final batch size: 1, sequence length: 7584 Attention mask shape: torch.Size([1, 1, 7584, 7584]) Position ids shape: torch.Size([1, 7584]) Input IDs shape: torch.Size([1, 7584]) Labels shape: torch.Size([1, 7584]) Final batch size: 1, sequence length: 25999 Attention mask shape: torch.Size([1, 1, 25999, 25999]) Position ids shape: torch.Size([1, 25999]) Input IDs shape: torch.Size([1, 25999]) Labels shape: torch.Size([1, 25999]) Final batch size: 1, sequence length: 22235 Attention mask shape: torch.Size([1, 1, 22235, 22235]) Position ids shape: torch.Size([1, 22235]) Input IDs shape: torch.Size([1, 22235]) Labels shape: torch.Size([1, 22235]) Final batch size: 1, sequence length: 14784 Attention mask shape: torch.Size([1, 1, 14784, 14784]) Position ids shape: torch.Size([1, 14784]) Input IDs shape: torch.Size([1, 14784]) Labels shape: torch.Size([1, 14784]) Final batch size: 1, sequence length: 27566 Attention mask shape: torch.Size([1, 1, 27566, 27566]) Position ids shape: torch.Size([1, 27566]) Input IDs shape: torch.Size([1, 27566]) Labels shape: torch.Size([1, 27566]) Final batch size: 1, sequence length: 28497 Attention mask shape: torch.Size([1, 1, 28497, 28497]) Position ids shape: torch.Size([1, 28497]) Input IDs shape: torch.Size([1, 28497]) Labels shape: torch.Size([1, 28497]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 15184 Attention mask shape: torch.Size([1, 1, 15184, 15184]) Position ids shape: torch.Size([1, 15184]) Input IDs shape: torch.Size([1, 15184]) Labels shape: torch.Size([1, 15184]) Final batch size: 1, sequence length: 22763 Attention mask shape: torch.Size([1, 1, 22763, 22763]) Position ids shape: torch.Size([1, 22763]) Input IDs shape: torch.Size([1, 22763]) Labels shape: torch.Size([1, 22763]) Final batch size: 1, sequence length: 27327 Attention mask shape: torch.Size([1, 1, 27327, 27327]) Position ids shape: torch.Size([1, 27327]) Input IDs shape: torch.Size([1, 27327]) Labels shape: torch.Size([1, 27327]) Final batch size: 1, sequence length: 24909 Attention mask shape: torch.Size([1, 1, 24909, 24909]) Position ids shape: torch.Size([1, 24909]) Input IDs shape: torch.Size([1, 24909]) Labels shape: torch.Size([1, 24909]) Final batch size: 1, sequence length: 28749 Attention mask shape: torch.Size([1, 1, 28749, 28749]) Position ids shape: torch.Size([1, 28749]) Input IDs shape: torch.Size([1, 28749]) Labels shape: torch.Size([1, 28749]) Final batch size: 1, sequence length: 15588 Attention mask shape: torch.Size([1, 1, 15588, 15588]) Position ids shape: torch.Size([1, 15588]) Input IDs shape: torch.Size([1, 15588]) Labels shape: torch.Size([1, 15588]) Final batch size: 1, sequence length: 21826 Attention mask shape: torch.Size([1, 1, 21826, 21826]) Position ids shape: torch.Size([1, 21826]) Input IDs shape: torch.Size([1, 21826]) Labels shape: torch.Size([1, 21826]) Final batch size: 1, sequence length: 23851 Attention mask shape: torch.Size([1, 1, 23851, 23851]) Position ids shape: torch.Size([1, 23851]) Input IDs shape: torch.Size([1, 23851]) Labels shape: torch.Size([1, 23851]) Final batch size: 1, sequence length: 30356 Attention mask shape: torch.Size([1, 1, 30356, 30356]) Position ids shape: torch.Size([1, 30356]) Input IDs shape: torch.Size([1, 30356]) Labels shape: torch.Size([1, 30356]) Final batch size: 1, sequence length: 22366 Attention mask shape: torch.Size([1, 1, 22366, 22366]) Position ids shape: torch.Size([1, 22366]) Input IDs shape: torch.Size([1, 22366]) Labels shape: torch.Size([1, 22366]) Final batch size: 1, sequence length: 21034 Attention mask shape: torch.Size([1, 1, 21034, 21034]) Position ids shape: torch.Size([1, 21034]) Input IDs shape: torch.Size([1, 21034]) Labels shape: torch.Size([1, 21034]) Final batch size: 1, sequence length: 11795 Attention mask shape: torch.Size([1, 1, 11795, 11795]) Position ids shape: torch.Size([1, 11795]) Input IDs shape: torch.Size([1, 11795]) Labels shape: torch.Size([1, 11795]) Final batch size: 1, sequence length: 20755 Attention mask shape: torch.Size([1, 1, 20755, 20755]) Position ids shape: torch.Size([1, 20755]) Input IDs shape: torch.Size([1, 20755]) Labels shape: torch.Size([1, 20755]) Final batch size: 1, sequence length: 36777 Attention mask shape: torch.Size([1, 1, 36777, 36777]) Position ids shape: torch.Size([1, 36777]) Input IDs shape: torch.Size([1, 36777]) Labels shape: torch.Size([1, 36777]) Final batch size: 1, sequence length: 34673 Attention mask shape: torch.Size([1, 1, 34673, 34673]) Position ids shape: torch.Size([1, 34673]) Input IDs shape: torch.Size([1, 34673]) Labels shape: torch.Size([1, 34673]) Final batch size: 1, sequence length: 27489 Attention mask shape: torch.Size([1, 1, 27489, 27489]) Position ids shape: torch.Size([1, 27489]) Input IDs shape: torch.Size([1, 27489]) Labels shape: torch.Size([1, 27489]) Final batch size: 1, sequence length: 38049 Attention mask shape: torch.Size([1, 1, 38049, 38049]) Position ids shape: torch.Size([1, 38049]) Input IDs shape: torch.Size([1, 38049]) Labels shape: torch.Size([1, 38049]) Final batch size: 1, sequence length: 36723 Attention mask shape: torch.Size([1, 1, 36723, 36723]) Position ids shape: torch.Size([1, 36723]) Input IDs shape: torch.Size([1, 36723]) Labels shape: torch.Size([1, 36723]) Final batch size: 1, sequence length: 38848 Attention mask shape: torch.Size([1, 1, 38848, 38848]) Position ids shape: torch.Size([1, 38848]) Input IDs shape: torch.Size([1, 38848]) Labels shape: torch.Size([1, 38848]) Final batch size: 1, sequence length: 28176 Attention mask shape: torch.Size([1, 1, 28176, 28176]) Position ids shape: torch.Size([1, 28176]) Input IDs shape: torch.Size([1, 28176]) Labels shape: torch.Size([1, 28176]) Final batch size: 1, sequence length: 37285 Attention mask shape: torch.Size([1, 1, 37285, 37285]) Position ids shape: torch.Size([1, 37285]) Input IDs shape: torch.Size([1, 37285]) Labels shape: torch.Size([1, 37285]) Final batch size: 1, sequence length: 31903 Attention mask shape: torch.Size([1, 1, 31903, 31903]) Position ids shape: torch.Size([1, 31903]) Input IDs shape: torch.Size([1, 31903]) Labels shape: torch.Size([1, 31903]) Final batch size: 1, sequence length: 40325 Attention mask shape: torch.Size([1, 1, 40325, 40325]) Position ids shape: torch.Size([1, 40325]) Input IDs shape: torch.Size([1, 40325]) Labels shape: torch.Size([1, 40325]) Final batch size: 1, sequence length: 24801 Attention mask shape: torch.Size([1, 1, 24801, 24801]) Position ids shape: torch.Size([1, 24801]) Input IDs shape: torch.Size([1, 24801]) Labels shape: torch.Size([1, 24801]) Final batch size: 1, sequence length: 39253 Attention mask shape: torch.Size([1, 1, 39253, 39253]) Position ids shape: torch.Size([1, 39253]) Input IDs shape: torch.Size([1, 39253]) Labels shape: torch.Size([1, 39253]) Final batch size: 1, sequence length: 25946 Attention mask shape: torch.Size([1, 1, 25946, 25946]) Position ids shape: torch.Size([1, 25946]) Input IDs shape: torch.Size([1, 25946]) Labels shape: torch.Size([1, 25946]) Final batch size: 1, sequence length: 38104 Attention mask shape: torch.Size([1, 1, 38104, 38104]) Position ids shape: torch.Size([1, 38104]) Input IDs shape: torch.Size([1, 38104]) Labels shape: torch.Size([1, 38104]) Final batch size: 1, sequence length: 6882 Attention mask shape: torch.Size([1, 1, 6882, 6882]) Position ids shape: torch.Size([1, 6882]) Input IDs shape: torch.Size([1, 6882]) Labels shape: torch.Size([1, 6882]) Final batch size: 1, sequence length: 35904 Attention mask shape: torch.Size([1, 1, 35904, 35904]) Position ids shape: torch.Size([1, 35904]) Input IDs shape: torch.Size([1, 35904]) Labels shape: torch.Size([1, 35904]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34289 Attention mask shape: torch.Size([1, 1, 34289, 34289]) Position ids shape: torch.Size([1, 34289]) Input IDs shape: torch.Size([1, 34289]) Labels shape: torch.Size([1, 34289]) Final batch size: 1, sequence length: 40317 Attention mask shape: torch.Size([1, 1, 40317, 40317]) Position ids shape: torch.Size([1, 40317]) Input IDs shape: torch.Size([1, 40317]) Labels shape: torch.Size([1, 40317]) Final batch size: 1, sequence length: 36786 Attention mask shape: torch.Size([1, 1, 36786, 36786]) Position ids shape: torch.Size([1, 36786]) Input IDs shape: torch.Size([1, 36786]) Labels shape: torch.Size([1, 36786]) Final batch size: 1, sequence length: 40937 Attention mask shape: torch.Size([1, 1, 40937, 40937]) Position ids shape: torch.Size([1, 40937]) Input IDs shape: torch.Size([1, 40937]) Labels shape: torch.Size([1, 40937]) Final batch size: 1, sequence length: 10469 Attention mask shape: torch.Size([1, 1, 10469, 10469]) Position ids shape: torch.Size([1, 10469]) Input IDs shape: torch.Size([1, 10469]) Labels shape: torch.Size([1, 10469]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16649 Attention mask shape: torch.Size([1, 1, 16649, 16649]) Position ids shape: torch.Size([1, 16649]) Input IDs shape: torch.Size([1, 16649]) Labels shape: torch.Size([1, 16649]) Final batch size: 1, sequence length: 19437 Attention mask shape: torch.Size([1, 1, 19437, 19437]) Position ids shape: torch.Size([1, 19437]) Input IDs shape: torch.Size([1, 19437]) Labels shape: torch.Size([1, 19437]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25963 Attention mask shape: torch.Size([1, 1, 25963, 25963]) Position ids shape: torch.Size([1, 25963]) Input IDs shape: torch.Size([1, 25963]) Labels shape: torch.Size([1, 25963]) Final batch size: 1, sequence length: 19639 Attention mask shape: torch.Size([1, 1, 19639, 19639]) Position ids shape: torch.Size([1, 19639]) Input IDs shape: torch.Size([1, 19639]) Labels shape: torch.Size([1, 19639]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40657 Attention mask shape: torch.Size([1, 1, 40657, 40657]) Position ids shape: torch.Size([1, 40657]) Input IDs shape: torch.Size([1, 40657]) Labels shape: torch.Size([1, 40657]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38686 Attention mask shape: torch.Size([1, 1, 38686, 38686]) Position ids shape: torch.Size([1, 38686]) Input IDs shape: torch.Size([1, 38686]) Labels shape: torch.Size([1, 38686]) Final batch size: 1, sequence length: 36128 Attention mask shape: torch.Size([1, 1, 36128, 36128]) Position ids shape: torch.Size([1, 36128]) Input IDs shape: torch.Size([1, 36128]) Labels shape: torch.Size([1, 36128]) Final batch size: 1, sequence length: 21547 Attention mask shape: torch.Size([1, 1, 21547, 21547]) Position ids shape: torch.Size([1, 21547]) Input IDs shape: torch.Size([1, 21547]) Labels shape: torch.Size([1, 21547]) Final batch size: 1, sequence length: 17882 Attention mask shape: torch.Size([1, 1, 17882, 17882]) Position ids shape: torch.Size([1, 17882]) Input IDs shape: torch.Size([1, 17882]) Labels shape: torch.Size([1, 17882]) Final batch size: 1, sequence length: 34540 Attention mask shape: torch.Size([1, 1, 34540, 34540]) Position ids shape: torch.Size([1, 34540]) Input IDs shape: torch.Size([1, 34540]) Labels shape: torch.Size([1, 34540]) Final batch size: 1, sequence length: 31714 Attention mask shape: torch.Size([1, 1, 31714, 31714]) Position ids shape: torch.Size([1, 31714]) Input IDs shape: torch.Size([1, 31714]) Labels shape: torch.Size([1, 31714]) Final batch size: 1, sequence length: 17379 Attention mask shape: torch.Size([1, 1, 17379, 17379]) Position ids shape: torch.Size([1, 17379]) Input IDs shape: torch.Size([1, 17379]) Labels shape: torch.Size([1, 17379]) Final batch size: 1, sequence length: 26660 Attention mask shape: torch.Size([1, 1, 26660, 26660]) Position ids shape: torch.Size([1, 26660]) Input IDs shape: torch.Size([1, 26660]) Labels shape: torch.Size([1, 26660]) Final batch size: 1, sequence length: 27947 Attention mask shape: torch.Size([1, 1, 27947, 27947]) Position ids shape: torch.Size([1, 27947]) Input IDs shape: torch.Size([1, 27947]) Labels shape: torch.Size([1, 27947]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 37946 Attention mask shape: torch.Size([1, 1, 37946, 37946]) Position ids shape: torch.Size([1, 37946]) Input IDs shape: torch.Size([1, 37946]) Labels shape: torch.Size([1, 37946]) Final batch size: 1, sequence length: 20843 Attention mask shape: torch.Size([1, 1, 20843, 20843]) Position ids shape: torch.Size([1, 20843]) Input IDs shape: torch.Size([1, 20843]) Labels shape: torch.Size([1, 20843]) Final batch size: 1, sequence length: 21496 Attention mask shape: torch.Size([1, 1, 21496, 21496]) Position ids shape: torch.Size([1, 21496]) Input IDs shape: torch.Size([1, 21496]) Labels shape: torch.Size([1, 21496]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24551 Attention mask shape: torch.Size([1, 1, 24551, 24551]) Position ids shape: torch.Size([1, 24551]) Input IDs shape: torch.Size([1, 24551]) Labels shape: torch.Size([1, 24551]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7681 Attention mask shape: torch.Size([1, 1, 7681, 7681]) Position ids shape: torch.Size([1, 7681]) Input IDs shape: torch.Size([1, 7681]) Labels shape: torch.Size([1, 7681]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36317 Attention mask shape: torch.Size([1, 1, 36317, 36317]) Position ids shape: torch.Size([1, 36317]) Input IDs shape: torch.Size([1, 36317]) Labels shape: torch.Size([1, 36317]) {'loss': 0.2876, 'grad_norm': 0.2759266268011296, 'learning_rate': 6.7918397477265e-06, 'num_tokens': -inf, 'epoch': 3.5} Final batch size: 1, sequence length: 5328 Attention mask shape: torch.Size([1, 1, 5328, 5328]) Position ids shape: torch.Size([1, 5328]) Input IDs shape: torch.Size([1, 5328]) Labels shape: torch.Size([1, 5328]) Final batch size: 1, sequence length: 5525 Attention mask shape: torch.Size([1, 1, 5525, 5525]) Position ids shape: torch.Size([1, 5525]) Input IDs shape: torch.Size([1, 5525]) Labels shape: torch.Size([1, 5525]) Final batch size: 1, sequence length: 10273 Attention mask shape: torch.Size([1, 1, 10273, 10273]) Position ids shape: torch.Size([1, 10273]) Input IDs shape: torch.Size([1, 10273]) Labels shape: torch.Size([1, 10273]) Final batch size: 1, sequence length: 11225 Attention mask shape: torch.Size([1, 1, 11225, 11225]) Position ids shape: torch.Size([1, 11225]) Input IDs shape: torch.Size([1, 11225]) Labels shape: torch.Size([1, 11225]) Final batch size: 1, sequence length: 12419 Attention mask shape: torch.Size([1, 1, 12419, 12419]) Position ids shape: torch.Size([1, 12419]) Input IDs shape: torch.Size([1, 12419]) Labels shape: torch.Size([1, 12419]) Final batch size: 1, sequence length: 10434 Attention mask shape: torch.Size([1, 1, 10434, 10434]) Position ids shape: torch.Size([1, 10434]) Input IDs shape: torch.Size([1, 10434]) Labels shape: torch.Size([1, 10434]) Final batch size: 1, sequence length: 13607 Attention mask shape: torch.Size([1, 1, 13607, 13607]) Position ids shape: torch.Size([1, 13607]) Input IDs shape: torch.Size([1, 13607]) Labels shape: torch.Size([1, 13607]) Final batch size: 1, sequence length: 13657 Attention mask shape: torch.Size([1, 1, 13657, 13657]) Position ids shape: torch.Size([1, 13657]) Input IDs shape: torch.Size([1, 13657]) Labels shape: torch.Size([1, 13657]) Final batch size: 1, sequence length: 14360 Attention mask shape: torch.Size([1, 1, 14360, 14360]) Position ids shape: torch.Size([1, 14360]) Input IDs shape: torch.Size([1, 14360]) Labels shape: torch.Size([1, 14360]) Final batch size: 1, sequence length: 11464 Attention mask shape: torch.Size([1, 1, 11464, 11464]) Position ids shape: torch.Size([1, 11464]) Input IDs shape: torch.Size([1, 11464]) Labels shape: torch.Size([1, 11464]) Final batch size: 1, sequence length: 16053 Attention mask shape: torch.Size([1, 1, 16053, 16053]) Position ids shape: torch.Size([1, 16053]) Input IDs shape: torch.Size([1, 16053]) Labels shape: torch.Size([1, 16053]) Final batch size: 1, sequence length: 14226 Attention mask shape: torch.Size([1, 1, 14226, 14226]) Position ids shape: torch.Size([1, 14226]) Input IDs shape: torch.Size([1, 14226]) Labels shape: torch.Size([1, 14226]) Final batch size: 1, sequence length: 9648 Attention mask shape: torch.Size([1, 1, 9648, 9648]) Position ids shape: torch.Size([1, 9648]) Input IDs shape: torch.Size([1, 9648]) Labels shape: torch.Size([1, 9648]) Final batch size: 1, sequence length: 16756 Attention mask shape: torch.Size([1, 1, 16756, 16756]) Position ids shape: torch.Size([1, 16756]) Input IDs shape: torch.Size([1, 16756]) Labels shape: torch.Size([1, 16756]) Final batch size: 1, sequence length: 17294 Attention mask shape: torch.Size([1, 1, 17294, 17294]) Position ids shape: torch.Size([1, 17294]) Input IDs shape: torch.Size([1, 17294]) Labels shape: torch.Size([1, 17294]) Final batch size: 1, sequence length: 18408 Attention mask shape: torch.Size([1, 1, 18408, 18408]) Position ids shape: torch.Size([1, 18408]) Input IDs shape: torch.Size([1, 18408]) Labels shape: torch.Size([1, 18408]) Final batch size: 1, sequence length: 16223 Attention mask shape: torch.Size([1, 1, 16223, 16223]) Position ids shape: torch.Size([1, 16223]) Input IDs shape: torch.Size([1, 16223]) Labels shape: torch.Size([1, 16223]) Final batch size: 1, sequence length: 16398 Attention mask shape: torch.Size([1, 1, 16398, 16398]) Position ids shape: torch.Size([1, 16398]) Input IDs shape: torch.Size([1, 16398]) Labels shape: torch.Size([1, 16398]) Final batch size: 1, sequence length: 19278 Attention mask shape: torch.Size([1, 1, 19278, 19278]) Position ids shape: torch.Size([1, 19278]) Input IDs shape: torch.Size([1, 19278]) Labels shape: torch.Size([1, 19278]) Final batch size: 1, sequence length: 15243 Attention mask shape: torch.Size([1, 1, 15243, 15243]) Position ids shape: torch.Size([1, 15243]) Input IDs shape: torch.Size([1, 15243]) Labels shape: torch.Size([1, 15243]) Final batch size: 1, sequence length: 17843 Attention mask shape: torch.Size([1, 1, 17843, 17843]) Position ids shape: torch.Size([1, 17843]) Input IDs shape: torch.Size([1, 17843]) Labels shape: torch.Size([1, 17843]) Final batch size: 1, sequence length: 17117 Attention mask shape: torch.Size([1, 1, 17117, 17117]) Position ids shape: torch.Size([1, 17117]) Input IDs shape: torch.Size([1, 17117]) Labels shape: torch.Size([1, 17117]) Final batch size: 1, sequence length: 10017 Attention mask shape: torch.Size([1, 1, 10017, 10017]) Position ids shape: torch.Size([1, 10017]) Input IDs shape: torch.Size([1, 10017]) Labels shape: torch.Size([1, 10017]) Final batch size: 1, sequence length: 21455 Attention mask shape: torch.Size([1, 1, 21455, 21455]) Position ids shape: torch.Size([1, 21455]) Input IDs shape: torch.Size([1, 21455]) Labels shape: torch.Size([1, 21455]) Final batch size: 1, sequence length: 20433 Attention mask shape: torch.Size([1, 1, 20433, 20433]) Position ids shape: torch.Size([1, 20433]) Input IDs shape: torch.Size([1, 20433]) Labels shape: torch.Size([1, 20433]) Final batch size: 1, sequence length: 19892 Attention mask shape: torch.Size([1, 1, 19892, 19892]) Position ids shape: torch.Size([1, 19892]) Input IDs shape: torch.Size([1, 19892]) Labels shape: torch.Size([1, 19892]) Final batch size: 1, sequence length: 14025 Attention mask shape: torch.Size([1, 1, 14025, 14025]) Position ids shape: torch.Size([1, 14025]) Input IDs shape: torch.Size([1, 14025]) Labels shape: torch.Size([1, 14025]) Final batch size: 1, sequence length: 20695 Attention mask shape: torch.Size([1, 1, 20695, 20695]) Position ids shape: torch.Size([1, 20695]) Input IDs shape: torch.Size([1, 20695]) Labels shape: torch.Size([1, 20695]) Final batch size: 1, sequence length: 19259 Attention mask shape: torch.Size([1, 1, 19259, 19259]) Position ids shape: torch.Size([1, 19259]) Input IDs shape: torch.Size([1, 19259]) Labels shape: torch.Size([1, 19259]) Final batch size: 1, sequence length: 5734 Attention mask shape: torch.Size([1, 1, 5734, 5734]) Position ids shape: torch.Size([1, 5734]) Input IDs shape: torch.Size([1, 5734]) Labels shape: torch.Size([1, 5734]) Final batch size: 1, sequence length: 19767 Attention mask shape: torch.Size([1, 1, 19767, 19767]) Position ids shape: torch.Size([1, 19767]) Input IDs shape: torch.Size([1, 19767]) Labels shape: torch.Size([1, 19767]) Final batch size: 1, sequence length: 14437 Attention mask shape: torch.Size([1, 1, 14437, 14437]) Position ids shape: torch.Size([1, 14437]) Input IDs shape: torch.Size([1, 14437]) Labels shape: torch.Size([1, 14437]) Final batch size: 1, sequence length: 6378 Attention mask shape: torch.Size([1, 1, 6378, 6378]) Position ids shape: torch.Size([1, 6378]) Input IDs shape: torch.Size([1, 6378]) Labels shape: torch.Size([1, 6378]) Final batch size: 1, sequence length: 14429 Attention mask shape: torch.Size([1, 1, 14429, 14429]) Position ids shape: torch.Size([1, 14429]) Input IDs shape: torch.Size([1, 14429]) Labels shape: torch.Size([1, 14429]) Final batch size: 1, sequence length: 13623 Attention mask shape: torch.Size([1, 1, 13623, 13623]) Position ids shape: torch.Size([1, 13623]) Input IDs shape: torch.Size([1, 13623]) Labels shape: torch.Size([1, 13623]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 23334 Attention mask shape: torch.Size([1, 1, 23334, 23334]) Position ids shape: torch.Size([1, 23334]) Input IDs shape: torch.Size([1, 23334]) Labels shape: torch.Size([1, 23334]) Final batch size: 1, sequence length: 17985 Attention mask shape: torch.Size([1, 1, 17985, 17985]) Position ids shape: torch.Size([1, 17985]) Input IDs shape: torch.Size([1, 17985]) Labels shape: torch.Size([1, 17985]) Final batch size: 1, sequence length: 25405 Attention mask shape: torch.Size([1, 1, 25405, 25405]) Position ids shape: torch.Size([1, 25405]) Input IDs shape: torch.Size([1, 25405]) Labels shape: torch.Size([1, 25405]) Final batch size: 1, sequence length: 19492 Attention mask shape: torch.Size([1, 1, 19492, 19492]) Position ids shape: torch.Size([1, 19492]) Input IDs shape: torch.Size([1, 19492]) Labels shape: torch.Size([1, 19492]) Final batch size: 1, sequence length: 17951 Attention mask shape: torch.Size([1, 1, 17951, 17951]) Position ids shape: torch.Size([1, 17951]) Input IDs shape: torch.Size([1, 17951]) Labels shape: torch.Size([1, 17951]) Final batch size: 1, sequence length: 26063 Attention mask shape: torch.Size([1, 1, 26063, 26063]) Position ids shape: torch.Size([1, 26063]) Input IDs shape: torch.Size([1, 26063]) Labels shape: torch.Size([1, 26063]) Final batch size: 1, sequence length: 25548 Attention mask shape: torch.Size([1, 1, 25548, 25548]) Position ids shape: torch.Size([1, 25548]) Input IDs shape: torch.Size([1, 25548]) Labels shape: torch.Size([1, 25548]) Final batch size: 1, sequence length: 24885 Attention mask shape: torch.Size([1, 1, 24885, 24885]) Position ids shape: torch.Size([1, 24885]) Input IDs shape: torch.Size([1, 24885]) Labels shape: torch.Size([1, 24885]) Final batch size: 1, sequence length: 27659 Attention mask shape: torch.Size([1, 1, 27659, 27659]) Position ids shape: torch.Size([1, 27659]) Input IDs shape: torch.Size([1, 27659]) Labels shape: torch.Size([1, 27659]) Final batch size: 1, sequence length: 22932 Attention mask shape: torch.Size([1, 1, 22932, 22932]) Position ids shape: torch.Size([1, 22932]) Input IDs shape: torch.Size([1, 22932]) Labels shape: torch.Size([1, 22932]) Final batch size: 1, sequence length: 25381 Attention mask shape: torch.Size([1, 1, 25381, 25381]) Position ids shape: torch.Size([1, 25381]) Input IDs shape: torch.Size([1, 25381]) Labels shape: torch.Size([1, 25381]) Final batch size: 1, sequence length: 27484 Attention mask shape: torch.Size([1, 1, 27484, 27484]) Position ids shape: torch.Size([1, 27484]) Input IDs shape: torch.Size([1, 27484]) Labels shape: torch.Size([1, 27484]) Final batch size: 1, sequence length: 29561 Attention mask shape: torch.Size([1, 1, 29561, 29561]) Position ids shape: torch.Size([1, 29561]) Input IDs shape: torch.Size([1, 29561]) Labels shape: torch.Size([1, 29561]) Final batch size: 1, sequence length: 16716 Attention mask shape: torch.Size([1, 1, 16716, 16716]) Position ids shape: torch.Size([1, 16716]) Input IDs shape: torch.Size([1, 16716]) Labels shape: torch.Size([1, 16716]) Final batch size: 1, sequence length: 28623 Attention mask shape: torch.Size([1, 1, 28623, 28623]) Position ids shape: torch.Size([1, 28623]) Input IDs shape: torch.Size([1, 28623]) Labels shape: torch.Size([1, 28623]) Final batch size: 1, sequence length: 26639 Attention mask shape: torch.Size([1, 1, 26639, 26639]) Position ids shape: torch.Size([1, 26639]) Input IDs shape: torch.Size([1, 26639]) Labels shape: torch.Size([1, 26639]) Final batch size: 1, sequence length: 30236 Attention mask shape: torch.Size([1, 1, 30236, 30236]) Position ids shape: torch.Size([1, 30236]) Input IDs shape: torch.Size([1, 30236]) Labels shape: torch.Size([1, 30236]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 28910 Attention mask shape: torch.Size([1, 1, 28910, 28910]) Position ids shape: torch.Size([1, 28910]) Input IDs shape: torch.Size([1, 28910]) Labels shape: torch.Size([1, 28910]) Final batch size: 1, sequence length: 28166 Attention mask shape: torch.Size([1, 1, 28166, 28166]) Position ids shape: torch.Size([1, 28166]) Input IDs shape: torch.Size([1, 28166]) Labels shape: torch.Size([1, 28166]) Final batch size: 1, sequence length: 28206 Attention mask shape: torch.Size([1, 1, 28206, 28206]) Position ids shape: torch.Size([1, 28206]) Input IDs shape: torch.Size([1, 28206]) Labels shape: torch.Size([1, 28206]) Final batch size: 1, sequence length: 29824 Attention mask shape: torch.Size([1, 1, 29824, 29824]) Position ids shape: torch.Size([1, 29824]) Input IDs shape: torch.Size([1, 29824]) Labels shape: torch.Size([1, 29824]) Final batch size: 1, sequence length: 25540 Attention mask shape: torch.Size([1, 1, 25540, 25540]) Position ids shape: torch.Size([1, 25540]) Input IDs shape: torch.Size([1, 25540]) Labels shape: torch.Size([1, 25540]) Final batch size: 1, sequence length: 31464 Final batch size: 1, sequence length: 32515 Attention mask shape: torch.Size([1, 1, 31464, 31464]) Position ids shape: torch.Size([1, 31464]) Input IDs shape: torch.Size([1, 31464]) Labels shape: torch.Size([1, 31464]) Attention mask shape: torch.Size([1, 1, 32515, 32515]) Position ids shape: torch.Size([1, 32515]) Input IDs shape: torch.Size([1, 32515]) Labels shape: torch.Size([1, 32515]) Final batch size: 1, sequence length: 17634 Attention mask shape: torch.Size([1, 1, 17634, 17634]) Position ids shape: torch.Size([1, 17634]) Input IDs shape: torch.Size([1, 17634]) Labels shape: torch.Size([1, 17634]) Final batch size: 1, sequence length: 28823 Attention mask shape: torch.Size([1, 1, 28823, 28823]) Position ids shape: torch.Size([1, 28823]) Input IDs shape: torch.Size([1, 28823]) Labels shape: torch.Size([1, 28823]) Final batch size: 1, sequence length: 32786 Attention mask shape: torch.Size([1, 1, 32786, 32786]) Position ids shape: torch.Size([1, 32786]) Input IDs shape: torch.Size([1, 32786]) Labels shape: torch.Size([1, 32786]) Final batch size: 1, sequence length: 33601 Attention mask shape: torch.Size([1, 1, 33601, 33601]) Position ids shape: torch.Size([1, 33601]) Input IDs shape: torch.Size([1, 33601]) Labels shape: torch.Size([1, 33601]) Final batch size: 1, sequence length: 32660 Attention mask shape: torch.Size([1, 1, 32660, 32660]) Position ids shape: torch.Size([1, 32660]) Input IDs shape: torch.Size([1, 32660]) Labels shape: torch.Size([1, 32660]) Final batch size: 1, sequence length: 9029 Attention mask shape: torch.Size([1, 1, 9029, 9029]) Position ids shape: torch.Size([1, 9029]) Input IDs shape: torch.Size([1, 9029]) Labels shape: torch.Size([1, 9029]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 17914 Attention mask shape: torch.Size([1, 1, 17914, 17914]) Position ids shape: torch.Size([1, 17914]) Input IDs shape: torch.Size([1, 17914]) Labels shape: torch.Size([1, 17914]) Final batch size: 1, sequence length: 20951 Attention mask shape: torch.Size([1, 1, 20951, 20951]) Position ids shape: torch.Size([1, 20951]) Input IDs shape: torch.Size([1, 20951]) Labels shape: torch.Size([1, 20951]) Final batch size: 1, sequence length: 36131 Attention mask shape: torch.Size([1, 1, 36131, 36131]) Position ids shape: torch.Size([1, 36131]) Input IDs shape: torch.Size([1, 36131]) Labels shape: torch.Size([1, 36131]) Final batch size: 1, sequence length: 10132 Attention mask shape: torch.Size([1, 1, 10132, 10132]) Position ids shape: torch.Size([1, 10132]) Input IDs shape: torch.Size([1, 10132]) Labels shape: torch.Size([1, 10132]) Final batch size: 1, sequence length: 37555 Attention mask shape: torch.Size([1, 1, 37555, 37555]) Position ids shape: torch.Size([1, 37555]) Input IDs shape: torch.Size([1, 37555]) Labels shape: torch.Size([1, 37555]) Final batch size: 1, sequence length: 35411 Attention mask shape: torch.Size([1, 1, 35411, 35411]) Position ids shape: torch.Size([1, 35411]) Input IDs shape: torch.Size([1, 35411]) Labels shape: torch.Size([1, 35411]) Final batch size: 1, sequence length: 36970 Attention mask shape: torch.Size([1, 1, 36970, 36970]) Position ids shape: torch.Size([1, 36970]) Input IDs shape: torch.Size([1, 36970]) Labels shape: torch.Size([1, 36970]) Final batch size: 1, sequence length: 15513 Attention mask shape: torch.Size([1, 1, 15513, 15513]) Position ids shape: torch.Size([1, 15513]) Input IDs shape: torch.Size([1, 15513]) Labels shape: torch.Size([1, 15513]) Final batch size: 1, sequence length: 37016 Attention mask shape: torch.Size([1, 1, 37016, 37016]) Position ids shape: torch.Size([1, 37016]) Input IDs shape: torch.Size([1, 37016]) Labels shape: torch.Size([1, 37016]) Final batch size: 1, sequence length: 36124 Attention mask shape: torch.Size([1, 1, 36124, 36124]) Position ids shape: torch.Size([1, 36124]) Input IDs shape: torch.Size([1, 36124]) Labels shape: torch.Size([1, 36124]) Final batch size: 1, sequence length: 25529 Attention mask shape: torch.Size([1, 1, 25529, 25529]) Position ids shape: torch.Size([1, 25529]) Input IDs shape: torch.Size([1, 25529]) Labels shape: torch.Size([1, 25529]) Final batch size: 1, sequence length: 20533 Attention mask shape: torch.Size([1, 1, 20533, 20533]) Position ids shape: torch.Size([1, 20533]) Input IDs shape: torch.Size([1, 20533]) Labels shape: torch.Size([1, 20533]) Final batch size: 1, sequence length: 35525 Attention mask shape: torch.Size([1, 1, 35525, 35525]) Position ids shape: torch.Size([1, 35525]) Input IDs shape: torch.Size([1, 35525]) Labels shape: torch.Size([1, 35525]) Final batch size: 1, sequence length: 26562 Attention mask shape: torch.Size([1, 1, 26562, 26562]) Position ids shape: torch.Size([1, 26562]) Input IDs shape: torch.Size([1, 26562]) Labels shape: torch.Size([1, 26562]) Final batch size: 1, sequence length: 38712 Attention mask shape: torch.Size([1, 1, 38712, 38712]) Position ids shape: torch.Size([1, 38712]) Input IDs shape: torch.Size([1, 38712]) Labels shape: torch.Size([1, 38712]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38071 Attention mask shape: torch.Size([1, 1, 38071, 38071]) Position ids shape: torch.Size([1, 38071]) Input IDs shape: torch.Size([1, 38071]) Labels shape: torch.Size([1, 38071]) Final batch size: 1, sequence length: 14372 Attention mask shape: torch.Size([1, 1, 14372, 14372]) Position ids shape: torch.Size([1, 14372]) Input IDs shape: torch.Size([1, 14372]) Labels shape: torch.Size([1, 14372]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26781 Attention mask shape: torch.Size([1, 1, 26781, 26781]) Position ids shape: torch.Size([1, 26781]) Input IDs shape: torch.Size([1, 26781]) Labels shape: torch.Size([1, 26781]) Final batch size: 1, sequence length: 36292 Attention mask shape: torch.Size([1, 1, 36292, 36292]) Position ids shape: torch.Size([1, 36292]) Input IDs shape: torch.Size([1, 36292]) Labels shape: torch.Size([1, 36292]) Final batch size: 1, sequence length: 31084 Attention mask shape: torch.Size([1, 1, 31084, 31084]) Position ids shape: torch.Size([1, 31084]) Input IDs shape: torch.Size([1, 31084]) Labels shape: torch.Size([1, 31084]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 38210 Attention mask shape: torch.Size([1, 1, 38210, 38210]) Position ids shape: torch.Size([1, 38210]) Input IDs shape: torch.Size([1, 38210]) Labels shape: torch.Size([1, 38210]) Final batch size: 1, sequence length: 33422 Attention mask shape: torch.Size([1, 1, 33422, 33422]) Position ids shape: torch.Size([1, 33422]) Input IDs shape: torch.Size([1, 33422]) Labels shape: torch.Size([1, 33422]) Final batch size: 1, sequence length: 28879 Attention mask shape: torch.Size([1, 1, 28879, 28879]) Position ids shape: torch.Size([1, 28879]) Input IDs shape: torch.Size([1, 28879]) Labels shape: torch.Size([1, 28879]) Final batch size: 1, sequence length: 30009 Attention mask shape: torch.Size([1, 1, 30009, 30009]) Position ids shape: torch.Size([1, 30009]) Input IDs shape: torch.Size([1, 30009]) Labels shape: torch.Size([1, 30009]) Final batch size: 1, sequence length: 36185 Attention mask shape: torch.Size([1, 1, 36185, 36185]) Position ids shape: torch.Size([1, 36185]) Input IDs shape: torch.Size([1, 36185]) Labels shape: torch.Size([1, 36185]) Final batch size: 1, sequence length: 35775 Attention mask shape: torch.Size([1, 1, 35775, 35775]) Position ids shape: torch.Size([1, 35775]) Input IDs shape: torch.Size([1, 35775]) Labels shape: torch.Size([1, 35775]) Final batch size: 1, sequence length: 16409 Attention mask shape: torch.Size([1, 1, 16409, 16409]) Position ids shape: torch.Size([1, 16409]) Input IDs shape: torch.Size([1, 16409]) Labels shape: torch.Size([1, 16409]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40753 Attention mask shape: torch.Size([1, 1, 40753, 40753]) Position ids shape: torch.Size([1, 40753]) Input IDs shape: torch.Size([1, 40753]) Labels shape: torch.Size([1, 40753]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36534 Attention mask shape: torch.Size([1, 1, 36534, 36534]) Position ids shape: torch.Size([1, 36534]) Input IDs shape: torch.Size([1, 36534]) Labels shape: torch.Size([1, 36534]) Final batch size: 1, sequence length: 31042 Attention mask shape: torch.Size([1, 1, 31042, 31042]) Position ids shape: torch.Size([1, 31042]) Input IDs shape: torch.Size([1, 31042]) Labels shape: torch.Size([1, 31042]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 15031 Attention mask shape: torch.Size([1, 1, 15031, 15031]) Position ids shape: torch.Size([1, 15031]) Input IDs shape: torch.Size([1, 15031]) Labels shape: torch.Size([1, 15031]) Final batch size: 1, sequence length: 37159 Attention mask shape: torch.Size([1, 1, 37159, 37159]) Position ids shape: torch.Size([1, 37159]) Input IDs shape: torch.Size([1, 37159]) Labels shape: torch.Size([1, 37159]) Final batch size: 1, sequence length: 30653 Attention mask shape: torch.Size([1, 1, 30653, 30653]) Position ids shape: torch.Size([1, 30653]) Input IDs shape: torch.Size([1, 30653]) Labels shape: torch.Size([1, 30653]) Final batch size: 1, sequence length: 18884 Attention mask shape: torch.Size([1, 1, 18884, 18884]) Position ids shape: torch.Size([1, 18884]) Input IDs shape: torch.Size([1, 18884]) Labels shape: torch.Size([1, 18884]) Final batch size: 1, sequence length: 7448 Attention mask shape: torch.Size([1, 1, 7448, 7448]) Position ids shape: torch.Size([1, 7448]) Input IDs shape: torch.Size([1, 7448]) Labels shape: torch.Size([1, 7448]) Final batch size: 1, sequence length: 20307 Attention mask shape: torch.Size([1, 1, 20307, 20307]) Position ids shape: torch.Size([1, 20307]) Input IDs shape: torch.Size([1, 20307]) Labels shape: torch.Size([1, 20307]) Final batch size: 1, sequence length: 31448 Attention mask shape: torch.Size([1, 1, 31448, 31448]) Position ids shape: torch.Size([1, 31448]) Input IDs shape: torch.Size([1, 31448]) Labels shape: torch.Size([1, 31448]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25850 Attention mask shape: torch.Size([1, 1, 25850, 25850]) Position ids shape: torch.Size([1, 25850]) Input IDs shape: torch.Size([1, 25850]) Labels shape: torch.Size([1, 25850]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40605 Attention mask shape: torch.Size([1, 1, 40605, 40605]) Position ids shape: torch.Size([1, 40605]) Input IDs shape: torch.Size([1, 40605]) Labels shape: torch.Size([1, 40605]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 5405 Attention mask shape: torch.Size([1, 1, 5405, 5405]) Position ids shape: torch.Size([1, 5405]) Input IDs shape: torch.Size([1, 5405]) Labels shape: torch.Size([1, 5405]) {'loss': 0.2721, 'grad_norm': 0.2913038528875607, 'learning_rate': 6.545084971874738e-06, 'num_tokens': -inf, 'epoch': 3.62} Final batch size: 1, sequence length: 4223 Attention mask shape: torch.Size([1, 1, 4223, 4223]) Position ids shape: torch.Size([1, 4223]) Input IDs shape: torch.Size([1, 4223]) Labels shape: torch.Size([1, 4223]) Final batch size: 1, sequence length: 7795 Attention mask shape: torch.Size([1, 1, 7795, 7795]) Position ids shape: torch.Size([1, 7795]) Input IDs shape: torch.Size([1, 7795]) Labels shape: torch.Size([1, 7795]) Final batch size: 1, sequence length: 8177 Attention mask shape: torch.Size([1, 1, 8177, 8177]) Position ids shape: torch.Size([1, 8177]) Input IDs shape: torch.Size([1, 8177]) Labels shape: torch.Size([1, 8177]) Final batch size: 1, sequence length: 4641 Attention mask shape: torch.Size([1, 1, 4641, 4641]) Position ids shape: torch.Size([1, 4641]) Input IDs shape: torch.Size([1, 4641]) Labels shape: torch.Size([1, 4641]) Final batch size: 1, sequence length: 9021 Attention mask shape: torch.Size([1, 1, 9021, 9021]) Position ids shape: torch.Size([1, 9021]) Input IDs shape: torch.Size([1, 9021]) Labels shape: torch.Size([1, 9021]) Final batch size: 1, sequence length: 12501 Attention mask shape: torch.Size([1, 1, 12501, 12501]) Position ids shape: torch.Size([1, 12501]) Input IDs shape: torch.Size([1, 12501]) Labels shape: torch.Size([1, 12501]) Final batch size: 1, sequence length: 5754 Attention mask shape: torch.Size([1, 1, 5754, 5754]) Position ids shape: torch.Size([1, 5754]) Input IDs shape: torch.Size([1, 5754]) Labels shape: torch.Size([1, 5754]) Final batch size: 1, sequence length: 9027 Attention mask shape: torch.Size([1, 1, 9027, 9027]) Position ids shape: torch.Size([1, 9027]) Input IDs shape: torch.Size([1, 9027]) Labels shape: torch.Size([1, 9027]) Final batch size: 1, sequence length: 16330 Attention mask shape: torch.Size([1, 1, 16330, 16330]) Position ids shape: torch.Size([1, 16330]) Input IDs shape: torch.Size([1, 16330]) Labels shape: torch.Size([1, 16330]) Final batch size: 1, sequence length: 13186 Attention mask shape: torch.Size([1, 1, 13186, 13186]) Position ids shape: torch.Size([1, 13186]) Input IDs shape: torch.Size([1, 13186]) Labels shape: torch.Size([1, 13186]) Final batch size: 1, sequence length: 11974 Attention mask shape: torch.Size([1, 1, 11974, 11974]) Position ids shape: torch.Size([1, 11974]) Input IDs shape: torch.Size([1, 11974]) Labels shape: torch.Size([1, 11974]) Final batch size: 1, sequence length: 16104 Attention mask shape: torch.Size([1, 1, 16104, 16104]) Position ids shape: torch.Size([1, 16104]) Input IDs shape: torch.Size([1, 16104]) Labels shape: torch.Size([1, 16104]) Final batch size: 1, sequence length: 16496 Attention mask shape: torch.Size([1, 1, 16496, 16496]) Position ids shape: torch.Size([1, 16496]) Input IDs shape: torch.Size([1, 16496]) Labels shape: torch.Size([1, 16496]) Final batch size: 1, sequence length: 13957 Attention mask shape: torch.Size([1, 1, 13957, 13957]) Position ids shape: torch.Size([1, 13957]) Input IDs shape: torch.Size([1, 13957]) Labels shape: torch.Size([1, 13957]) Final batch size: 1, sequence length: 17092 Attention mask shape: torch.Size([1, 1, 17092, 17092]) Position ids shape: torch.Size([1, 17092]) Input IDs shape: torch.Size([1, 17092]) Labels shape: torch.Size([1, 17092]) Final batch size: 1, sequence length: 16366 Attention mask shape: torch.Size([1, 1, 16366, 16366]) Position ids shape: torch.Size([1, 16366]) Input IDs shape: torch.Size([1, 16366]) Labels shape: torch.Size([1, 16366]) Final batch size: 1, sequence length: 18958 Attention mask shape: torch.Size([1, 1, 18958, 18958]) Position ids shape: torch.Size([1, 18958]) Input IDs shape: torch.Size([1, 18958]) Labels shape: torch.Size([1, 18958]) Final batch size: 1, sequence length: 15283 Attention mask shape: torch.Size([1, 1, 15283, 15283]) Position ids shape: torch.Size([1, 15283]) Input IDs shape: torch.Size([1, 15283]) Labels shape: torch.Size([1, 15283]) Final batch size: 1, sequence length: 16088 Attention mask shape: torch.Size([1, 1, 16088, 16088]) Position ids shape: torch.Size([1, 16088]) Input IDs shape: torch.Size([1, 16088]) Labels shape: torch.Size([1, 16088]) Final batch size: 1, sequence length: 18938 Attention mask shape: torch.Size([1, 1, 18938, 18938]) Position ids shape: torch.Size([1, 18938]) Input IDs shape: torch.Size([1, 18938]) Labels shape: torch.Size([1, 18938]) Final batch size: 1, sequence length: 19034 Attention mask shape: torch.Size([1, 1, 19034, 19034]) Position ids shape: torch.Size([1, 19034]) Input IDs shape: torch.Size([1, 19034]) Labels shape: torch.Size([1, 19034]) Final batch size: 1, sequence length: 19603 Attention mask shape: torch.Size([1, 1, 19603, 19603]) Position ids shape: torch.Size([1, 19603]) Input IDs shape: torch.Size([1, 19603]) Labels shape: torch.Size([1, 19603]) Final batch size: 1, sequence length: 18034 Attention mask shape: torch.Size([1, 1, 18034, 18034]) Position ids shape: torch.Size([1, 18034]) Input IDs shape: torch.Size([1, 18034]) Labels shape: torch.Size([1, 18034]) Final batch size: 1, sequence length: 18415 Attention mask shape: torch.Size([1, 1, 18415, 18415]) Position ids shape: torch.Size([1, 18415]) Input IDs shape: torch.Size([1, 18415]) Labels shape: torch.Size([1, 18415]) Final batch size: 1, sequence length: 21132 Attention mask shape: torch.Size([1, 1, 21132, 21132]) Position ids shape: torch.Size([1, 21132]) Input IDs shape: torch.Size([1, 21132]) Labels shape: torch.Size([1, 21132]) Final batch size: 1, sequence length: 16014 Attention mask shape: torch.Size([1, 1, 16014, 16014]) Position ids shape: torch.Size([1, 16014]) Input IDs shape: torch.Size([1, 16014]) Labels shape: torch.Size([1, 16014]) Final batch size: 1, sequence length: 14492 Attention mask shape: torch.Size([1, 1, 14492, 14492]) Position ids shape: torch.Size([1, 14492]) Input IDs shape: torch.Size([1, 14492]) Labels shape: torch.Size([1, 14492]) Final batch size: 1, sequence length: 13468 Attention mask shape: torch.Size([1, 1, 13468, 13468]) Position ids shape: torch.Size([1, 13468]) Input IDs shape: torch.Size([1, 13468]) Labels shape: torch.Size([1, 13468]) Final batch size: 1, sequence length: 21705 Attention mask shape: torch.Size([1, 1, 21705, 21705]) Position ids shape: torch.Size([1, 21705]) Input IDs shape: torch.Size([1, 21705]) Labels shape: torch.Size([1, 21705]) Final batch size: 1, sequence length: 23132 Attention mask shape: torch.Size([1, 1, 23132, 23132]) Position ids shape: torch.Size([1, 23132]) Input IDs shape: torch.Size([1, 23132]) Labels shape: torch.Size([1, 23132]) Final batch size: 1, sequence length: 23102 Attention mask shape: torch.Size([1, 1, 23102, 23102]) Position ids shape: torch.Size([1, 23102]) Input IDs shape: torch.Size([1, 23102]) Labels shape: torch.Size([1, 23102]) Final batch size: 1, sequence length: 22328 Attention mask shape: torch.Size([1, 1, 22328, 22328]) Position ids shape: torch.Size([1, 22328]) Input IDs shape: torch.Size([1, 22328]) Labels shape: torch.Size([1, 22328]) Final batch size: 1, sequence length: 21404 Attention mask shape: torch.Size([1, 1, 21404, 21404]) Position ids shape: torch.Size([1, 21404]) Input IDs shape: torch.Size([1, 21404]) Labels shape: torch.Size([1, 21404]) Final batch size: 1, sequence length: 20726 Attention mask shape: torch.Size([1, 1, 20726, 20726]) Position ids shape: torch.Size([1, 20726]) Input IDs shape: torch.Size([1, 20726]) Labels shape: torch.Size([1, 20726]) Final batch size: 1, sequence length: 24432 Attention mask shape: torch.Size([1, 1, 24432, 24432]) Position ids shape: torch.Size([1, 24432]) Input IDs shape: torch.Size([1, 24432]) Labels shape: torch.Size([1, 24432]) Final batch size: 1, sequence length: 12756 Attention mask shape: torch.Size([1, 1, 12756, 12756]) Position ids shape: torch.Size([1, 12756]) Input IDs shape: torch.Size([1, 12756]) Labels shape: torch.Size([1, 12756]) Final batch size: 1, sequence length: 25351 Attention mask shape: torch.Size([1, 1, 25351, 25351]) Position ids shape: torch.Size([1, 25351]) Input IDs shape: torch.Size([1, 25351]) Labels shape: torch.Size([1, 25351]) Final batch size: 1, sequence length: 22771 Attention mask shape: torch.Size([1, 1, 22771, 22771]) Position ids shape: torch.Size([1, 22771]) Input IDs shape: torch.Size([1, 22771]) Labels shape: torch.Size([1, 22771]) Final batch size: 1, sequence length: 16590 Attention mask shape: torch.Size([1, 1, 16590, 16590]) Position ids shape: torch.Size([1, 16590]) Input IDs shape: torch.Size([1, 16590]) Labels shape: torch.Size([1, 16590]) Final batch size: 1, sequence length: 23971 Attention mask shape: torch.Size([1, 1, 23971, 23971]) Position ids shape: torch.Size([1, 23971]) Input IDs shape: torch.Size([1, 23971]) Labels shape: torch.Size([1, 23971]) Final batch size: 1, sequence length: 16291 Attention mask shape: torch.Size([1, 1, 16291, 16291]) Position ids shape: torch.Size([1, 16291]) Input IDs shape: torch.Size([1, 16291]) Labels shape: torch.Size([1, 16291]) Final batch size: 1, sequence length: 17514 Attention mask shape: torch.Size([1, 1, 17514, 17514]) Position ids shape: torch.Size([1, 17514]) Input IDs shape: torch.Size([1, 17514]) Labels shape: torch.Size([1, 17514]) Final batch size: 1, sequence length: 27785 Attention mask shape: torch.Size([1, 1, 27785, 27785]) Position ids shape: torch.Size([1, 27785]) Input IDs shape: torch.Size([1, 27785]) Labels shape: torch.Size([1, 27785]) Final batch size: 1, sequence length: 13790 Attention mask shape: torch.Size([1, 1, 13790, 13790]) Position ids shape: torch.Size([1, 13790]) Input IDs shape: torch.Size([1, 13790]) Labels shape: torch.Size([1, 13790]) Final batch size: 1, sequence length: 23894 Attention mask shape: torch.Size([1, 1, 23894, 23894]) Position ids shape: torch.Size([1, 23894]) Input IDs shape: torch.Size([1, 23894]) Labels shape: torch.Size([1, 23894]) Final batch size: 1, sequence length: 26375 Attention mask shape: torch.Size([1, 1, 26375, 26375]) Position ids shape: torch.Size([1, 26375]) Input IDs shape: torch.Size([1, 26375]) Labels shape: torch.Size([1, 26375]) Final batch size: 1, sequence length: 27780 Attention mask shape: torch.Size([1, 1, 27780, 27780]) Position ids shape: torch.Size([1, 27780]) Input IDs shape: torch.Size([1, 27780]) Labels shape: torch.Size([1, 27780]) Final batch size: 1, sequence length: 26596 Attention mask shape: torch.Size([1, 1, 26596, 26596]) Position ids shape: torch.Size([1, 26596]) Input IDs shape: torch.Size([1, 26596]) Labels shape: torch.Size([1, 26596]) Final batch size: 1, sequence length: 23810 Attention mask shape: torch.Size([1, 1, 23810, 23810]) Position ids shape: torch.Size([1, 23810]) Input IDs shape: torch.Size([1, 23810]) Labels shape: torch.Size([1, 23810]) Final batch size: 1, sequence length: 18869 Attention mask shape: torch.Size([1, 1, 18869, 18869]) Position ids shape: torch.Size([1, 18869]) Input IDs shape: torch.Size([1, 18869]) Labels shape: torch.Size([1, 18869]) Final batch size: 1, sequence length: 20817 Attention mask shape: torch.Size([1, 1, 20817, 20817]) Position ids shape: torch.Size([1, 20817]) Input IDs shape: torch.Size([1, 20817]) Labels shape: torch.Size([1, 20817]) Final batch size: 1, sequence length: 28574 Attention mask shape: torch.Size([1, 1, 28574, 28574]) Position ids shape: torch.Size([1, 28574]) Input IDs shape: torch.Size([1, 28574]) Labels shape: torch.Size([1, 28574]) Final batch size: 1, sequence length: 32022 Attention mask shape: torch.Size([1, 1, 32022, 32022]) Position ids shape: torch.Size([1, 32022]) Input IDs shape: torch.Size([1, 32022]) Labels shape: torch.Size([1, 32022]) Final batch size: 1, sequence length: 30165 Attention mask shape: torch.Size([1, 1, 30165, 30165]) Position ids shape: torch.Size([1, 30165]) Input IDs shape: torch.Size([1, 30165]) Labels shape: torch.Size([1, 30165]) Final batch size: 1, sequence length: 28412 Attention mask shape: torch.Size([1, 1, 28412, 28412]) Position ids shape: torch.Size([1, 28412]) Input IDs shape: torch.Size([1, 28412]) Labels shape: torch.Size([1, 28412]) Final batch size: 1, sequence length: 25388 Attention mask shape: torch.Size([1, 1, 25388, 25388]) Position ids shape: torch.Size([1, 25388]) Input IDs shape: torch.Size([1, 25388]) Labels shape: torch.Size([1, 25388]) Final batch size: 1, sequence length: 26766 Attention mask shape: torch.Size([1, 1, 26766, 26766]) Position ids shape: torch.Size([1, 26766]) Input IDs shape: torch.Size([1, 26766]) Labels shape: torch.Size([1, 26766]) Final batch size: 1, sequence length: 29132 Attention mask shape: torch.Size([1, 1, 29132, 29132]) Position ids shape: torch.Size([1, 29132]) Input IDs shape: torch.Size([1, 29132]) Labels shape: torch.Size([1, 29132]) Final batch size: 1, sequence length: 28845 Attention mask shape: torch.Size([1, 1, 28845, 28845]) Position ids shape: torch.Size([1, 28845]) Input IDs shape: torch.Size([1, 28845]) Labels shape: torch.Size([1, 28845]) Final batch size: 1, sequence length: 29168 Attention mask shape: torch.Size([1, 1, 29168, 29168]) Position ids shape: torch.Size([1, 29168]) Input IDs shape: torch.Size([1, 29168]) Labels shape: torch.Size([1, 29168]) Final batch size: 1, sequence length: 34457 Attention mask shape: torch.Size([1, 1, 34457, 34457]) Position ids shape: torch.Size([1, 34457]) Input IDs shape: torch.Size([1, 34457]) Labels shape: torch.Size([1, 34457]) Final batch size: 1, sequence length: 29830 Attention mask shape: torch.Size([1, 1, 29830, 29830]) Position ids shape: torch.Size([1, 29830]) Input IDs shape: torch.Size([1, 29830]) Labels shape: torch.Size([1, 29830]) Final batch size: 1, sequence length: 34326 Attention mask shape: torch.Size([1, 1, 34326, 34326]) Position ids shape: torch.Size([1, 34326]) Input IDs shape: torch.Size([1, 34326]) Labels shape: torch.Size([1, 34326]) Final batch size: 1, sequence length: 22043 Attention mask shape: torch.Size([1, 1, 22043, 22043]) Position ids shape: torch.Size([1, 22043]) Input IDs shape: torch.Size([1, 22043]) Labels shape: torch.Size([1, 22043]) Final batch size: 1, sequence length: 35790 Attention mask shape: torch.Size([1, 1, 35790, 35790]) Position ids shape: torch.Size([1, 35790]) Input IDs shape: torch.Size([1, 35790]) Labels shape: torch.Size([1, 35790]) Final batch size: 1, sequence length: 24653 Attention mask shape: torch.Size([1, 1, 24653, 24653]) Position ids shape: torch.Size([1, 24653]) Input IDs shape: torch.Size([1, 24653]) Labels shape: torch.Size([1, 24653]) Final batch size: 1, sequence length: 22014 Attention mask shape: torch.Size([1, 1, 22014, 22014]) Position ids shape: torch.Size([1, 22014]) Input IDs shape: torch.Size([1, 22014]) Labels shape: torch.Size([1, 22014]) Final batch size: 1, sequence length: 26465 Attention mask shape: torch.Size([1, 1, 26465, 26465]) Position ids shape: torch.Size([1, 26465]) Input IDs shape: torch.Size([1, 26465]) Labels shape: torch.Size([1, 26465]) Final batch size: 1, sequence length: 34874 Attention mask shape: torch.Size([1, 1, 34874, 34874]) Position ids shape: torch.Size([1, 34874]) Input IDs shape: torch.Size([1, 34874]) Labels shape: torch.Size([1, 34874]) Final batch size: 1, sequence length: 38617 Attention mask shape: torch.Size([1, 1, 38617, 38617]) Position ids shape: torch.Size([1, 38617]) Input IDs shape: torch.Size([1, 38617]) Labels shape: torch.Size([1, 38617]) Final batch size: 1, sequence length: 13986 Attention mask shape: torch.Size([1, 1, 13986, 13986]) Position ids shape: torch.Size([1, 13986]) Input IDs shape: torch.Size([1, 13986]) Labels shape: torch.Size([1, 13986]) Final batch size: 1, sequence length: 38919 Attention mask shape: torch.Size([1, 1, 38919, 38919]) Position ids shape: torch.Size([1, 38919]) Input IDs shape: torch.Size([1, 38919]) Labels shape: torch.Size([1, 38919]) Final batch size: 1, sequence length: 19744 Attention mask shape: torch.Size([1, 1, 19744, 19744]) Position ids shape: torch.Size([1, 19744]) Input IDs shape: torch.Size([1, 19744]) Labels shape: torch.Size([1, 19744]) Final batch size: 1, sequence length: 35868 Attention mask shape: torch.Size([1, 1, 35868, 35868]) Position ids shape: torch.Size([1, 35868]) Input IDs shape: torch.Size([1, 35868]) Labels shape: torch.Size([1, 35868]) Final batch size: 1, sequence length: 35803 Attention mask shape: torch.Size([1, 1, 35803, 35803]) Position ids shape: torch.Size([1, 35803]) Input IDs shape: torch.Size([1, 35803]) Labels shape: torch.Size([1, 35803]) Final batch size: 1, sequence length: 39150 Attention mask shape: torch.Size([1, 1, 39150, 39150]) Position ids shape: torch.Size([1, 39150]) Input IDs shape: torch.Size([1, 39150]) Labels shape: torch.Size([1, 39150]) Final batch size: 1, sequence length: 37039 Attention mask shape: torch.Size([1, 1, 37039, 37039]) Position ids shape: torch.Size([1, 37039]) Input IDs shape: torch.Size([1, 37039]) Labels shape: torch.Size([1, 37039]) Final batch size: 1, sequence length: 36794 Attention mask shape: torch.Size([1, 1, 36794, 36794]) Position ids shape: torch.Size([1, 36794]) Input IDs shape: torch.Size([1, 36794]) Labels shape: torch.Size([1, 36794]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19204 Attention mask shape: torch.Size([1, 1, 19204, 19204]) Position ids shape: torch.Size([1, 19204]) Input IDs shape: torch.Size([1, 19204]) Labels shape: torch.Size([1, 19204]) Final batch size: 1, sequence length: 40717 Attention mask shape: torch.Size([1, 1, 40717, 40717]) Position ids shape: torch.Size([1, 40717]) Input IDs shape: torch.Size([1, 40717]) Labels shape: torch.Size([1, 40717]) Final batch size: 1, sequence length: 23245 Attention mask shape: torch.Size([1, 1, 23245, 23245]) Position ids shape: torch.Size([1, 23245]) Input IDs shape: torch.Size([1, 23245]) Labels shape: torch.Size([1, 23245]) Final batch size: 1, sequence length: 25923 Attention mask shape: torch.Size([1, 1, 25923, 25923]) Position ids shape: torch.Size([1, 25923]) Input IDs shape: torch.Size([1, 25923]) Labels shape: torch.Size([1, 25923]) Final batch size: 1, sequence length: 22282 Attention mask shape: torch.Size([1, 1, 22282, 22282]) Position ids shape: torch.Size([1, 22282]) Input IDs shape: torch.Size([1, 22282]) Labels shape: torch.Size([1, 22282]) Final batch size: 1, sequence length: 35999 Attention mask shape: torch.Size([1, 1, 35999, 35999]) Position ids shape: torch.Size([1, 35999]) Input IDs shape: torch.Size([1, 35999]) Labels shape: torch.Size([1, 35999]) Final batch size: 1, sequence length: 28002 Attention mask shape: torch.Size([1, 1, 28002, 28002]) Position ids shape: torch.Size([1, 28002]) Input IDs shape: torch.Size([1, 28002]) Labels shape: torch.Size([1, 28002]) Final batch size: 1, sequence length: 31506 Attention mask shape: torch.Size([1, 1, 31506, 31506]) Position ids shape: torch.Size([1, 31506]) Input IDs shape: torch.Size([1, 31506]) Labels shape: torch.Size([1, 31506]) Final batch size: 1, sequence length: 32974 Attention mask shape: torch.Size([1, 1, 32974, 32974]) Position ids shape: torch.Size([1, 32974]) Input IDs shape: torch.Size([1, 32974]) Labels shape: torch.Size([1, 32974]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40065 Attention mask shape: torch.Size([1, 1, 40065, 40065]) Position ids shape: torch.Size([1, 40065]) Input IDs shape: torch.Size([1, 40065]) Labels shape: torch.Size([1, 40065]) Final batch size: 1, sequence length: 39919 Attention mask shape: torch.Size([1, 1, 39919, 39919]) Position ids shape: torch.Size([1, 39919]) Input IDs shape: torch.Size([1, 39919]) Labels shape: torch.Size([1, 39919]) Final batch size: 1, sequence length: 28781 Attention mask shape: torch.Size([1, 1, 28781, 28781]) Position ids shape: torch.Size([1, 28781]) Input IDs shape: torch.Size([1, 28781]) Labels shape: torch.Size([1, 28781]) Final batch size: 1, sequence length: 19846 Attention mask shape: torch.Size([1, 1, 19846, 19846]) Position ids shape: torch.Size([1, 19846]) Input IDs shape: torch.Size([1, 19846]) Labels shape: torch.Size([1, 19846]) Final batch size: 1, sequence length: 31340 Attention mask shape: torch.Size([1, 1, 31340, 31340]) Position ids shape: torch.Size([1, 31340]) Input IDs shape: torch.Size([1, 31340]) Labels shape: torch.Size([1, 31340]) Final batch size: 1, sequence length: 11644 Attention mask shape: torch.Size([1, 1, 11644, 11644]) Position ids shape: torch.Size([1, 11644]) Input IDs shape: torch.Size([1, 11644]) Labels shape: torch.Size([1, 11644]) Final batch size: 1, sequence length: 12418 Attention mask shape: torch.Size([1, 1, 12418, 12418]) Position ids shape: torch.Size([1, 12418]) Input IDs shape: torch.Size([1, 12418]) Labels shape: torch.Size([1, 12418]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31359 Attention mask shape: torch.Size([1, 1, 31359, 31359]) Position ids shape: torch.Size([1, 31359]) Input IDs shape: torch.Size([1, 31359]) Labels shape: torch.Size([1, 31359]) Final batch size: 1, sequence length: 18379 Attention mask shape: torch.Size([1, 1, 18379, 18379]) Position ids shape: torch.Size([1, 18379]) Input IDs shape: torch.Size([1, 18379]) Labels shape: torch.Size([1, 18379]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7364 Attention mask shape: torch.Size([1, 1, 7364, 7364]) Position ids shape: torch.Size([1, 7364]) Input IDs shape: torch.Size([1, 7364]) Labels shape: torch.Size([1, 7364]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40874 Attention mask shape: torch.Size([1, 1, 40874, 40874]) Position ids shape: torch.Size([1, 40874]) Input IDs shape: torch.Size([1, 40874]) Labels shape: torch.Size([1, 40874]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25560 Attention mask shape: torch.Size([1, 1, 25560, 25560]) Position ids shape: torch.Size([1, 25560]) Input IDs shape: torch.Size([1, 25560]) Labels shape: torch.Size([1, 25560]) Final batch size: 1, sequence length: 9309 Attention mask shape: torch.Size([1, 1, 9309, 9309]) Position ids shape: torch.Size([1, 9309]) Input IDs shape: torch.Size([1, 9309]) Labels shape: torch.Size([1, 9309]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26221 Attention mask shape: torch.Size([1, 1, 26221, 26221]) Position ids shape: torch.Size([1, 26221]) Input IDs shape: torch.Size([1, 26221]) Labels shape: torch.Size([1, 26221]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17224 Attention mask shape: torch.Size([1, 1, 17224, 17224]) Position ids shape: torch.Size([1, 17224]) Input IDs shape: torch.Size([1, 17224]) Labels shape: torch.Size([1, 17224]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 18654 Attention mask shape: torch.Size([1, 1, 18654, 18654]) Position ids shape: torch.Size([1, 18654]) Input IDs shape: torch.Size([1, 18654]) Labels shape: torch.Size([1, 18654]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25820 Attention mask shape: torch.Size([1, 1, 25820, 25820]) Position ids shape: torch.Size([1, 25820]) Input IDs shape: torch.Size([1, 25820]) Labels shape: torch.Size([1, 25820]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30709 Attention mask shape: torch.Size([1, 1, 30709, 30709]) Position ids shape: torch.Size([1, 30709]) Input IDs shape: torch.Size([1, 30709]) Labels shape: torch.Size([1, 30709]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 29526 Attention mask shape: torch.Size([1, 1, 29526, 29526]) Position ids shape: torch.Size([1, 29526]) Input IDs shape: torch.Size([1, 29526]) Labels shape: torch.Size([1, 29526]) {'loss': 0.2817, 'grad_norm': 0.3136304605317824, 'learning_rate': 6.294095225512604e-06, 'num_tokens': -inf, 'epoch': 3.75} Final batch size: 1, sequence length: 7461 Attention mask shape: torch.Size([1, 1, 7461, 7461]) Position ids shape: torch.Size([1, 7461]) Input IDs shape: torch.Size([1, 7461]) Labels shape: torch.Size([1, 7461]) Final batch size: 1, sequence length: 12279 Attention mask shape: torch.Size([1, 1, 12279, 12279]) Position ids shape: torch.Size([1, 12279]) Input IDs shape: torch.Size([1, 12279]) Labels shape: torch.Size([1, 12279]) Final batch size: 1, sequence length: 9858 Attention mask shape: torch.Size([1, 1, 9858, 9858]) Position ids shape: torch.Size([1, 9858]) Input IDs shape: torch.Size([1, 9858]) Labels shape: torch.Size([1, 9858]) Final batch size: 1, sequence length: 10826 Attention mask shape: torch.Size([1, 1, 10826, 10826]) Position ids shape: torch.Size([1, 10826]) Input IDs shape: torch.Size([1, 10826]) Labels shape: torch.Size([1, 10826]) Final batch size: 1, sequence length: 12096 Attention mask shape: torch.Size([1, 1, 12096, 12096]) Position ids shape: torch.Size([1, 12096]) Input IDs shape: torch.Size([1, 12096]) Labels shape: torch.Size([1, 12096]) Final batch size: 1, sequence length: 13427 Attention mask shape: torch.Size([1, 1, 13427, 13427]) Position ids shape: torch.Size([1, 13427]) Input IDs shape: torch.Size([1, 13427]) Labels shape: torch.Size([1, 13427]) Final batch size: 1, sequence length: 15933 Attention mask shape: torch.Size([1, 1, 15933, 15933]) Position ids shape: torch.Size([1, 15933]) Input IDs shape: torch.Size([1, 15933]) Labels shape: torch.Size([1, 15933]) Final batch size: 1, sequence length: 12622 Attention mask shape: torch.Size([1, 1, 12622, 12622]) Position ids shape: torch.Size([1, 12622]) Input IDs shape: torch.Size([1, 12622]) Labels shape: torch.Size([1, 12622]) Final batch size: 1, sequence length: 13550 Attention mask shape: torch.Size([1, 1, 13550, 13550]) Position ids shape: torch.Size([1, 13550]) Input IDs shape: torch.Size([1, 13550]) Labels shape: torch.Size([1, 13550]) Final batch size: 1, sequence length: 12960 Attention mask shape: torch.Size([1, 1, 12960, 12960]) Position ids shape: torch.Size([1, 12960]) Input IDs shape: torch.Size([1, 12960]) Labels shape: torch.Size([1, 12960]) Final batch size: 1, sequence length: 16450 Attention mask shape: torch.Size([1, 1, 16450, 16450]) Position ids shape: torch.Size([1, 16450]) Input IDs shape: torch.Size([1, 16450]) Labels shape: torch.Size([1, 16450]) Final batch size: 1, sequence length: 15673 Attention mask shape: torch.Size([1, 1, 15673, 15673]) Position ids shape: torch.Size([1, 15673]) Input IDs shape: torch.Size([1, 15673]) Labels shape: torch.Size([1, 15673]) Final batch size: 1, sequence length: 16927 Attention mask shape: torch.Size([1, 1, 16927, 16927]) Position ids shape: torch.Size([1, 16927]) Input IDs shape: torch.Size([1, 16927]) Labels shape: torch.Size([1, 16927]) Final batch size: 1, sequence length: 18214 Attention mask shape: torch.Size([1, 1, 18214, 18214]) Position ids shape: torch.Size([1, 18214]) Input IDs shape: torch.Size([1, 18214]) Labels shape: torch.Size([1, 18214]) Final batch size: 1, sequence length: 17934 Attention mask shape: torch.Size([1, 1, 17934, 17934]) Position ids shape: torch.Size([1, 17934]) Input IDs shape: torch.Size([1, 17934]) Labels shape: torch.Size([1, 17934]) Final batch size: 1, sequence length: 13789 Attention mask shape: torch.Size([1, 1, 13789, 13789]) Position ids shape: torch.Size([1, 13789]) Input IDs shape: torch.Size([1, 13789]) Labels shape: torch.Size([1, 13789]) Final batch size: 1, sequence length: 20292 Attention mask shape: torch.Size([1, 1, 20292, 20292]) Position ids shape: torch.Size([1, 20292]) Input IDs shape: torch.Size([1, 20292]) Labels shape: torch.Size([1, 20292]) Final batch size: 1, sequence length: 17245 Attention mask shape: torch.Size([1, 1, 17245, 17245]) Position ids shape: torch.Size([1, 17245]) Input IDs shape: torch.Size([1, 17245]) Labels shape: torch.Size([1, 17245]) Final batch size: 1, sequence length: 17861 Attention mask shape: torch.Size([1, 1, 17861, 17861]) Position ids shape: torch.Size([1, 17861]) Input IDs shape: torch.Size([1, 17861]) Labels shape: torch.Size([1, 17861]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 19899 Attention mask shape: torch.Size([1, 1, 19899, 19899]) Position ids shape: torch.Size([1, 19899]) Input IDs shape: torch.Size([1, 19899]) Labels shape: torch.Size([1, 19899]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 12728 Attention mask shape: torch.Size([1, 1, 12728, 12728]) Position ids shape: torch.Size([1, 12728]) Input IDs shape: torch.Size([1, 12728]) Labels shape: torch.Size([1, 12728]) Final batch size: 1, sequence length: 21091 Attention mask shape: torch.Size([1, 1, 21091, 21091]) Position ids shape: torch.Size([1, 21091]) Input IDs shape: torch.Size([1, 21091]) Labels shape: torch.Size([1, 21091]) Final batch size: 1, sequence length: 16036 Attention mask shape: torch.Size([1, 1, 16036, 16036]) Position ids shape: torch.Size([1, 16036]) Input IDs shape: torch.Size([1, 16036]) Labels shape: torch.Size([1, 16036]) Final batch size: 1, sequence length: 23004 Attention mask shape: torch.Size([1, 1, 23004, 23004]) Position ids shape: torch.Size([1, 23004]) Input IDs shape: torch.Size([1, 23004]) Labels shape: torch.Size([1, 23004]) Final batch size: 1, sequence length: 20065 Attention mask shape: torch.Size([1, 1, 20065, 20065]) Position ids shape: torch.Size([1, 20065]) Input IDs shape: torch.Size([1, 20065]) Labels shape: torch.Size([1, 20065]) Final batch size: 1, sequence length: 19187 Attention mask shape: torch.Size([1, 1, 19187, 19187]) Position ids shape: torch.Size([1, 19187]) Input IDs shape: torch.Size([1, 19187]) Labels shape: torch.Size([1, 19187]) Final batch size: 1, sequence length: 25435 Attention mask shape: torch.Size([1, 1, 25435, 25435]) Position ids shape: torch.Size([1, 25435]) Input IDs shape: torch.Size([1, 25435]) Labels shape: torch.Size([1, 25435]) Final batch size: 1, sequence length: 21675 Attention mask shape: torch.Size([1, 1, 21675, 21675]) Position ids shape: torch.Size([1, 21675]) Input IDs shape: torch.Size([1, 21675]) Labels shape: torch.Size([1, 21675]) Final batch size: 1, sequence length: 25600 Attention mask shape: torch.Size([1, 1, 25600, 25600]) Position ids shape: torch.Size([1, 25600]) Input IDs shape: torch.Size([1, 25600]) Labels shape: torch.Size([1, 25600]) Final batch size: 1, sequence length: 24633 Attention mask shape: torch.Size([1, 1, 24633, 24633]) Position ids shape: torch.Size([1, 24633]) Input IDs shape: torch.Size([1, 24633]) Labels shape: torch.Size([1, 24633]) Final batch size: 1, sequence length: 13946 Attention mask shape: torch.Size([1, 1, 13946, 13946]) Position ids shape: torch.Size([1, 13946]) Input IDs shape: torch.Size([1, 13946]) Labels shape: torch.Size([1, 13946]) Final batch size: 1, sequence length: 25176 Attention mask shape: torch.Size([1, 1, 25176, 25176]) Position ids shape: torch.Size([1, 25176]) Input IDs shape: torch.Size([1, 25176]) Labels shape: torch.Size([1, 25176]) Final batch size: 1, sequence length: 24965 Attention mask shape: torch.Size([1, 1, 24965, 24965]) Position ids shape: torch.Size([1, 24965]) Input IDs shape: torch.Size([1, 24965]) Labels shape: torch.Size([1, 24965]) Final batch size: 1, sequence length: 21586 Attention mask shape: torch.Size([1, 1, 21586, 21586]) Position ids shape: torch.Size([1, 21586]) Input IDs shape: torch.Size([1, 21586]) Labels shape: torch.Size([1, 21586]) Final batch size: 1, sequence length: 26316 Attention mask shape: torch.Size([1, 1, 26316, 26316]) Position ids shape: torch.Size([1, 26316]) Input IDs shape: torch.Size([1, 26316]) Labels shape: torch.Size([1, 26316]) Final batch size: 1, sequence length: 24956 Attention mask shape: torch.Size([1, 1, 24956, 24956]) Position ids shape: torch.Size([1, 24956]) Input IDs shape: torch.Size([1, 24956]) Labels shape: torch.Size([1, 24956]) Final batch size: 1, sequence length: 17763 Attention mask shape: torch.Size([1, 1, 17763, 17763]) Position ids shape: torch.Size([1, 17763]) Input IDs shape: torch.Size([1, 17763]) Labels shape: torch.Size([1, 17763]) Final batch size: 1, sequence length: 3010 Attention mask shape: torch.Size([1, 1, 3010, 3010]) Position ids shape: torch.Size([1, 3010]) Input IDs shape: torch.Size([1, 3010]) Labels shape: torch.Size([1, 3010]) Final batch size: 1, sequence length: 25325 Attention mask shape: torch.Size([1, 1, 25325, 25325]) Position ids shape: torch.Size([1, 25325]) Input IDs shape: torch.Size([1, 25325]) Labels shape: torch.Size([1, 25325]) Final batch size: 1, sequence length: 22778 Attention mask shape: torch.Size([1, 1, 22778, 22778]) Position ids shape: torch.Size([1, 22778]) Input IDs shape: torch.Size([1, 22778]) Labels shape: torch.Size([1, 22778]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 24808 Attention mask shape: torch.Size([1, 1, 24808, 24808]) Position ids shape: torch.Size([1, 24808]) Input IDs shape: torch.Size([1, 24808]) Labels shape: torch.Size([1, 24808]) Final batch size: 1, sequence length: 23896 Attention mask shape: torch.Size([1, 1, 23896, 23896]) Position ids shape: torch.Size([1, 23896]) Input IDs shape: torch.Size([1, 23896]) Labels shape: torch.Size([1, 23896]) Final batch size: 1, sequence length: 26247 Attention mask shape: torch.Size([1, 1, 26247, 26247]) Position ids shape: torch.Size([1, 26247]) Input IDs shape: torch.Size([1, 26247]) Labels shape: torch.Size([1, 26247]) Final batch size: 1, sequence length: 25035 Attention mask shape: torch.Size([1, 1, 25035, 25035]) Position ids shape: torch.Size([1, 25035]) Input IDs shape: torch.Size([1, 25035]) Labels shape: torch.Size([1, 25035]) Final batch size: 1, sequence length: 18832 Attention mask shape: torch.Size([1, 1, 18832, 18832]) Position ids shape: torch.Size([1, 18832]) Input IDs shape: torch.Size([1, 18832]) Labels shape: torch.Size([1, 18832]) Final batch size: 1, sequence length: 20582 Attention mask shape: torch.Size([1, 1, 20582, 20582]) Position ids shape: torch.Size([1, 20582]) Input IDs shape: torch.Size([1, 20582]) Labels shape: torch.Size([1, 20582]) Final batch size: 1, sequence length: 26454 Attention mask shape: torch.Size([1, 1, 26454, 26454]) Position ids shape: torch.Size([1, 26454]) Input IDs shape: torch.Size([1, 26454]) Labels shape: torch.Size([1, 26454]) Final batch size: 1, sequence length: 26500 Attention mask shape: torch.Size([1, 1, 26500, 26500]) Position ids shape: torch.Size([1, 26500]) Input IDs shape: torch.Size([1, 26500]) Labels shape: torch.Size([1, 26500]) Final batch size: 1, sequence length: 22775 Attention mask shape: torch.Size([1, 1, 22775, 22775]) Position ids shape: torch.Size([1, 22775]) Input IDs shape: torch.Size([1, 22775]) Labels shape: torch.Size([1, 22775]) Final batch size: 1, sequence length: 14212 Attention mask shape: torch.Size([1, 1, 14212, 14212]) Position ids shape: torch.Size([1, 14212]) Input IDs shape: torch.Size([1, 14212]) Labels shape: torch.Size([1, 14212]) Final batch size: 1, sequence length: 24927 Attention mask shape: torch.Size([1, 1, 24927, 24927]) Position ids shape: torch.Size([1, 24927]) Input IDs shape: torch.Size([1, 24927]) Labels shape: torch.Size([1, 24927]) Final batch size: 1, sequence length: 16714 Attention mask shape: torch.Size([1, 1, 16714, 16714]) Position ids shape: torch.Size([1, 16714]) Input IDs shape: torch.Size([1, 16714]) Labels shape: torch.Size([1, 16714]) Final batch size: 1, sequence length: 27222 Attention mask shape: torch.Size([1, 1, 27222, 27222]) Position ids shape: torch.Size([1, 27222]) Input IDs shape: torch.Size([1, 27222]) Labels shape: torch.Size([1, 27222]) Final batch size: 1, sequence length: 28926 Attention mask shape: torch.Size([1, 1, 28926, 28926]) Position ids shape: torch.Size([1, 28926]) Input IDs shape: torch.Size([1, 28926]) Labels shape: torch.Size([1, 28926]) Final batch size: 1, sequence length: 6978 Attention mask shape: torch.Size([1, 1, 6978, 6978]) Position ids shape: torch.Size([1, 6978]) Input IDs shape: torch.Size([1, 6978]) Labels shape: torch.Size([1, 6978]) Final batch size: 1, sequence length: 28552 Attention mask shape: torch.Size([1, 1, 28552, 28552]) Position ids shape: torch.Size([1, 28552]) Input IDs shape: torch.Size([1, 28552]) Labels shape: torch.Size([1, 28552]) Final batch size: 1, sequence length: 13453 Attention mask shape: torch.Size([1, 1, 13453, 13453]) Position ids shape: torch.Size([1, 13453]) Input IDs shape: torch.Size([1, 13453]) Labels shape: torch.Size([1, 13453]) Final batch size: 1, sequence length: 12426 Attention mask shape: torch.Size([1, 1, 12426, 12426]) Position ids shape: torch.Size([1, 12426]) Input IDs shape: torch.Size([1, 12426]) Labels shape: torch.Size([1, 12426]) Final batch size: 1, sequence length: 28644 Attention mask shape: torch.Size([1, 1, 28644, 28644]) Position ids shape: torch.Size([1, 28644]) Input IDs shape: torch.Size([1, 28644]) Labels shape: torch.Size([1, 28644]) Final batch size: 1, sequence length: 17049 Attention mask shape: torch.Size([1, 1, 17049, 17049]) Position ids shape: torch.Size([1, 17049]) Input IDs shape: torch.Size([1, 17049]) Labels shape: torch.Size([1, 17049]) Final batch size: 1, sequence length: 19953 Attention mask shape: torch.Size([1, 1, 19953, 19953]) Position ids shape: torch.Size([1, 19953]) Input IDs shape: torch.Size([1, 19953]) Labels shape: torch.Size([1, 19953]) Final batch size: 1, sequence length: 32101 Attention mask shape: torch.Size([1, 1, 32101, 32101]) Position ids shape: torch.Size([1, 32101]) Input IDs shape: torch.Size([1, 32101]) Labels shape: torch.Size([1, 32101]) Final batch size: 1, sequence length: 31208 Attention mask shape: torch.Size([1, 1, 31208, 31208]) Position ids shape: torch.Size([1, 31208]) Input IDs shape: torch.Size([1, 31208]) Labels shape: torch.Size([1, 31208]) Final batch size: 1, sequence length: 32513 Attention mask shape: torch.Size([1, 1, 32513, 32513]) Position ids shape: torch.Size([1, 32513]) Input IDs shape: torch.Size([1, 32513]) Labels shape: torch.Size([1, 32513]) Final batch size: 1, sequence length: 25979 Attention mask shape: torch.Size([1, 1, 25979, 25979]) Position ids shape: torch.Size([1, 25979]) Input IDs shape: torch.Size([1, 25979]) Labels shape: torch.Size([1, 25979]) Final batch size: 1, sequence length: 32100 Attention mask shape: torch.Size([1, 1, 32100, 32100]) Position ids shape: torch.Size([1, 32100]) Input IDs shape: torch.Size([1, 32100]) Labels shape: torch.Size([1, 32100]) Final batch size: 1, sequence length: 16571 Attention mask shape: torch.Size([1, 1, 16571, 16571]) Position ids shape: torch.Size([1, 16571]) Input IDs shape: torch.Size([1, 16571]) Labels shape: torch.Size([1, 16571]) Final batch size: 1, sequence length: 30635 Attention mask shape: torch.Size([1, 1, 30635, 30635]) Position ids shape: torch.Size([1, 30635]) Input IDs shape: torch.Size([1, 30635]) Labels shape: torch.Size([1, 30635]) Final batch size: 1, sequence length: 24676 Attention mask shape: torch.Size([1, 1, 24676, 24676]) Position ids shape: torch.Size([1, 24676]) Input IDs shape: torch.Size([1, 24676]) Labels shape: torch.Size([1, 24676]) Final batch size: 1, sequence length: 21290 Final batch size: 1, sequence length: 35058 Attention mask shape: torch.Size([1, 1, 21290, 21290]) Position ids shape: torch.Size([1, 21290]) Input IDs shape: torch.Size([1, 21290]) Labels shape: torch.Size([1, 21290]) Attention mask shape: torch.Size([1, 1, 35058, 35058]) Position ids shape: torch.Size([1, 35058]) Input IDs shape: torch.Size([1, 35058]) Labels shape: torch.Size([1, 35058]) Final batch size: 1, sequence length: 31662 Attention mask shape: torch.Size([1, 1, 31662, 31662]) Position ids shape: torch.Size([1, 31662]) Input IDs shape: torch.Size([1, 31662]) Labels shape: torch.Size([1, 31662]) Final batch size: 1, sequence length: 25774 Attention mask shape: torch.Size([1, 1, 25774, 25774]) Position ids shape: torch.Size([1, 25774]) Input IDs shape: torch.Size([1, 25774]) Labels shape: torch.Size([1, 25774]) Final batch size: 1, sequence length: 37381 Attention mask shape: torch.Size([1, 1, 37381, 37381]) Position ids shape: torch.Size([1, 37381]) Input IDs shape: torch.Size([1, 37381]) Labels shape: torch.Size([1, 37381]) Final batch size: 1, sequence length: 37720 Attention mask shape: torch.Size([1, 1, 37720, 37720]) Position ids shape: torch.Size([1, 37720]) Input IDs shape: torch.Size([1, 37720]) Labels shape: torch.Size([1, 37720]) Final batch size: 1, sequence length: 35014 Attention mask shape: torch.Size([1, 1, 35014, 35014]) Position ids shape: torch.Size([1, 35014]) Input IDs shape: torch.Size([1, 35014]) Labels shape: torch.Size([1, 35014]) Final batch size: 1, sequence length: 37391 Attention mask shape: torch.Size([1, 1, 37391, 37391]) Position ids shape: torch.Size([1, 37391]) Input IDs shape: torch.Size([1, 37391]) Labels shape: torch.Size([1, 37391]) Final batch size: 1, sequence length: 25147 Attention mask shape: torch.Size([1, 1, 25147, 25147]) Position ids shape: torch.Size([1, 25147]) Input IDs shape: torch.Size([1, 25147]) Labels shape: torch.Size([1, 25147]) Final batch size: 1, sequence length: 33621 Attention mask shape: torch.Size([1, 1, 33621, 33621]) Position ids shape: torch.Size([1, 33621]) Input IDs shape: torch.Size([1, 33621]) Labels shape: torch.Size([1, 33621]) Final batch size: 1, sequence length: 28377 Attention mask shape: torch.Size([1, 1, 28377, 28377]) Position ids shape: torch.Size([1, 28377]) Input IDs shape: torch.Size([1, 28377]) Labels shape: torch.Size([1, 28377]) Final batch size: 1, sequence length: 39955 Attention mask shape: torch.Size([1, 1, 39955, 39955]) Position ids shape: torch.Size([1, 39955]) Input IDs shape: torch.Size([1, 39955]) Labels shape: torch.Size([1, 39955]) Final batch size: 1, sequence length: 36047 Attention mask shape: torch.Size([1, 1, 36047, 36047]) Position ids shape: torch.Size([1, 36047]) Input IDs shape: torch.Size([1, 36047]) Labels shape: torch.Size([1, 36047]) Final batch size: 1, sequence length: 39474 Attention mask shape: torch.Size([1, 1, 39474, 39474]) Position ids shape: torch.Size([1, 39474]) Input IDs shape: torch.Size([1, 39474]) Labels shape: torch.Size([1, 39474]) Final batch size: 1, sequence length: 38391 Attention mask shape: torch.Size([1, 1, 38391, 38391]) Position ids shape: torch.Size([1, 38391]) Input IDs shape: torch.Size([1, 38391]) Labels shape: torch.Size([1, 38391]) Final batch size: 1, sequence length: 25219 Attention mask shape: torch.Size([1, 1, 25219, 25219]) Position ids shape: torch.Size([1, 25219]) Input IDs shape: torch.Size([1, 25219]) Labels shape: torch.Size([1, 25219]) Final batch size: 1, sequence length: 26789 Attention mask shape: torch.Size([1, 1, 26789, 26789]) Position ids shape: torch.Size([1, 26789]) Input IDs shape: torch.Size([1, 26789]) Labels shape: torch.Size([1, 26789]) Final batch size: 1, sequence length: 16816 Attention mask shape: torch.Size([1, 1, 16816, 16816]) Position ids shape: torch.Size([1, 16816]) Input IDs shape: torch.Size([1, 16816]) Labels shape: torch.Size([1, 16816]) Final batch size: 1, sequence length: 30259 Attention mask shape: torch.Size([1, 1, 30259, 30259]) Position ids shape: torch.Size([1, 30259]) Input IDs shape: torch.Size([1, 30259]) Labels shape: torch.Size([1, 30259]) Final batch size: 1, sequence length: 26387 Attention mask shape: torch.Size([1, 1, 26387, 26387]) Position ids shape: torch.Size([1, 26387]) Input IDs shape: torch.Size([1, 26387]) Labels shape: torch.Size([1, 26387]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 22710 Attention mask shape: torch.Size([1, 1, 22710, 22710]) Position ids shape: torch.Size([1, 22710]) Input IDs shape: torch.Size([1, 22710]) Labels shape: torch.Size([1, 22710]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38577 Attention mask shape: torch.Size([1, 1, 38577, 38577]) Position ids shape: torch.Size([1, 38577]) Input IDs shape: torch.Size([1, 38577]) Labels shape: torch.Size([1, 38577]) Final batch size: 1, sequence length: 21321 Attention mask shape: torch.Size([1, 1, 21321, 21321]) Position ids shape: torch.Size([1, 21321]) Input IDs shape: torch.Size([1, 21321]) Labels shape: torch.Size([1, 21321]) Final batch size: 1, sequence length: 19441 Attention mask shape: torch.Size([1, 1, 19441, 19441]) Position ids shape: torch.Size([1, 19441]) Input IDs shape: torch.Size([1, 19441]) Labels shape: torch.Size([1, 19441]) Final batch size: 1, sequence length: 31447 Attention mask shape: torch.Size([1, 1, 31447, 31447]) Position ids shape: torch.Size([1, 31447]) Input IDs shape: torch.Size([1, 31447]) Labels shape: torch.Size([1, 31447]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40590 Attention mask shape: torch.Size([1, 1, 40590, 40590]) Position ids shape: torch.Size([1, 40590]) Input IDs shape: torch.Size([1, 40590]) Labels shape: torch.Size([1, 40590]) Final batch size: 1, sequence length: 35261 Attention mask shape: torch.Size([1, 1, 35261, 35261]) Position ids shape: torch.Size([1, 35261]) Input IDs shape: torch.Size([1, 35261]) Labels shape: torch.Size([1, 35261]) Final batch size: 1, sequence length: 28448 Attention mask shape: torch.Size([1, 1, 28448, 28448]) Position ids shape: torch.Size([1, 28448]) Input IDs shape: torch.Size([1, 28448]) Labels shape: torch.Size([1, 28448]) Final batch size: 1, sequence length: 32079 Attention mask shape: torch.Size([1, 1, 32079, 32079]) Position ids shape: torch.Size([1, 32079]) Input IDs shape: torch.Size([1, 32079]) Labels shape: torch.Size([1, 32079]) Final batch size: 1, sequence length: 27371 Attention mask shape: torch.Size([1, 1, 27371, 27371]) Position ids shape: torch.Size([1, 27371]) Input IDs shape: torch.Size([1, 27371]) Labels shape: torch.Size([1, 27371]) Final batch size: 1, sequence length: 32129 Attention mask shape: torch.Size([1, 1, 32129, 32129]) Position ids shape: torch.Size([1, 32129]) Input IDs shape: torch.Size([1, 32129]) Labels shape: torch.Size([1, 32129]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36356 Attention mask shape: torch.Size([1, 1, 36356, 36356]) Position ids shape: torch.Size([1, 36356]) Input IDs shape: torch.Size([1, 36356]) Labels shape: torch.Size([1, 36356]) Final batch size: 1, sequence length: 14136 Attention mask shape: torch.Size([1, 1, 14136, 14136]) Position ids shape: torch.Size([1, 14136]) Input IDs shape: torch.Size([1, 14136]) Labels shape: torch.Size([1, 14136]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24623 Attention mask shape: torch.Size([1, 1, 24623, 24623]) Position ids shape: torch.Size([1, 24623]) Input IDs shape: torch.Size([1, 24623]) Labels shape: torch.Size([1, 24623]) Final batch size: 1, sequence length: 27021 Attention mask shape: torch.Size([1, 1, 27021, 27021]) Position ids shape: torch.Size([1, 27021]) Input IDs shape: torch.Size([1, 27021]) Labels shape: torch.Size([1, 27021]) Final batch size: 1, sequence length: 33014 Attention mask shape: torch.Size([1, 1, 33014, 33014]) Position ids shape: torch.Size([1, 33014]) Input IDs shape: torch.Size([1, 33014]) Labels shape: torch.Size([1, 33014]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30712 Attention mask shape: torch.Size([1, 1, 30712, 30712]) Position ids shape: torch.Size([1, 30712]) Input IDs shape: torch.Size([1, 30712]) Labels shape: torch.Size([1, 30712]) Final batch size: 1, sequence length: 19027 Attention mask shape: torch.Size([1, 1, 19027, 19027]) Position ids shape: torch.Size([1, 19027]) Input IDs shape: torch.Size([1, 19027]) Labels shape: torch.Size([1, 19027]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 23208 Attention mask shape: torch.Size([1, 1, 23208, 23208]) Position ids shape: torch.Size([1, 23208]) Input IDs shape: torch.Size([1, 23208]) Labels shape: torch.Size([1, 23208]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2765, 'grad_norm': 0.31482526514749304, 'learning_rate': 6.039558454088796e-06, 'num_tokens': -inf, 'epoch': 3.88} Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 10507 Attention mask shape: torch.Size([1, 1, 10507, 10507]) Position ids shape: torch.Size([1, 10507]) Input IDs shape: torch.Size([1, 10507]) Labels shape: torch.Size([1, 10507]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 26271 Attention mask shape: torch.Size([1, 1, 26271, 26271]) Position ids shape: torch.Size([1, 26271]) Input IDs shape: torch.Size([1, 26271]) Labels shape: torch.Size([1, 26271]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 31381 Attention mask shape: torch.Size([1, 1, 31381, 31381]) Position ids shape: torch.Size([1, 31381]) Input IDs shape: torch.Size([1, 31381]) Labels shape: torch.Size([1, 31381]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2693, 'grad_norm': 0.287428613099651, 'learning_rate': 5.782172325201155e-06, 'num_tokens': -inf, 'epoch': 4.0} Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 6362 Attention mask shape: torch.Size([1, 1, 6362, 6362]) Position ids shape: torch.Size([1, 6362]) Input IDs shape: torch.Size([1, 6362]) Labels shape: torch.Size([1, 6362]) Final batch size: 1, sequence length: 10523 Attention mask shape: torch.Size([1, 1, 10523, 10523]) Position ids shape: torch.Size([1, 10523]) Input IDs shape: torch.Size([1, 10523]) Labels shape: torch.Size([1, 10523]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 12385 Attention mask shape: torch.Size([1, 1, 12385, 12385]) Position ids shape: torch.Size([1, 12385]) Input IDs shape: torch.Size([1, 12385]) Labels shape: torch.Size([1, 12385]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 18410 Attention mask shape: torch.Size([1, 1, 18410, 18410]) Position ids shape: torch.Size([1, 18410]) Input IDs shape: torch.Size([1, 18410]) Labels shape: torch.Size([1, 18410]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 18098 Attention mask shape: torch.Size([1, 1, 18098, 18098]) Position ids shape: torch.Size([1, 18098]) Input IDs shape: torch.Size([1, 18098]) Labels shape: torch.Size([1, 18098]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 17299 Attention mask shape: torch.Size([1, 1, 17299, 17299]) Position ids shape: torch.Size([1, 17299]) Input IDs shape: torch.Size([1, 17299]) Labels shape: torch.Size([1, 17299]) Final batch size: 1, sequence length: 9091 Attention mask shape: torch.Size([1, 1, 9091, 9091]) Position ids shape: torch.Size([1, 9091]) Input IDs shape: torch.Size([1, 9091]) Labels shape: torch.Size([1, 9091]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 12656 Attention mask shape: torch.Size([1, 1, 12656, 12656]) Position ids shape: torch.Size([1, 12656]) Input IDs shape: torch.Size([1, 12656]) Labels shape: torch.Size([1, 12656]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 18915 Attention mask shape: torch.Size([1, 1, 18915, 18915]) Position ids shape: torch.Size([1, 18915]) Input IDs shape: torch.Size([1, 18915]) Labels shape: torch.Size([1, 18915]) Final batch size: 1, sequence length: 18060 Attention mask shape: torch.Size([1, 1, 18060, 18060]) Position ids shape: torch.Size([1, 18060]) Input IDs shape: torch.Size([1, 18060]) Labels shape: torch.Size([1, 18060]) Final batch size: 1, sequence length: 24365 Attention mask shape: torch.Size([1, 1, 24365, 24365]) Position ids shape: torch.Size([1, 24365]) Input IDs shape: torch.Size([1, 24365]) Labels shape: torch.Size([1, 24365]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 6948 Attention mask shape: torch.Size([1, 1, 6948, 6948]) Position ids shape: torch.Size([1, 6948]) Input IDs shape: torch.Size([1, 6948]) Labels shape: torch.Size([1, 6948]) Final batch size: 1, sequence length: 11819 Attention mask shape: torch.Size([1, 1, 11819, 11819]) Position ids shape: torch.Size([1, 11819]) Input IDs shape: torch.Size([1, 11819]) Labels shape: torch.Size([1, 11819]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 29113 Attention mask shape: torch.Size([1, 1, 29113, 29113]) Position ids shape: torch.Size([1, 29113]) Input IDs shape: torch.Size([1, 29113]) Labels shape: torch.Size([1, 29113]) Final batch size: 1, sequence length: 26976 Attention mask shape: torch.Size([1, 1, 26976, 26976]) Position ids shape: torch.Size([1, 26976]) Input IDs shape: torch.Size([1, 26976]) Labels shape: torch.Size([1, 26976]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 28841 Attention mask shape: torch.Size([1, 1, 28841, 28841]) Position ids shape: torch.Size([1, 28841]) Input IDs shape: torch.Size([1, 28841]) Labels shape: torch.Size([1, 28841]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 21450 Attention mask shape: torch.Size([1, 1, 21450, 21450]) Position ids shape: torch.Size([1, 21450]) Input IDs shape: torch.Size([1, 21450]) Labels shape: torch.Size([1, 21450]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 24880 Attention mask shape: torch.Size([1, 1, 24880, 24880]) Position ids shape: torch.Size([1, 24880]) Input IDs shape: torch.Size([1, 24880]) Labels shape: torch.Size([1, 24880]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 13215 Attention mask shape: torch.Size([1, 1, 13215, 13215]) Position ids shape: torch.Size([1, 13215]) Input IDs shape: torch.Size([1, 13215]) Labels shape: torch.Size([1, 13215]) Final batch size: 1, sequence length: 25172 Attention mask shape: torch.Size([1, 1, 25172, 25172]) Position ids shape: torch.Size([1, 25172]) Input IDs shape: torch.Size([1, 25172]) Labels shape: torch.Size([1, 25172]) Final batch size: 1, sequence length: 27541 Attention mask shape: torch.Size([1, 1, 27541, 27541]) Position ids shape: torch.Size([1, 27541]) Input IDs shape: torch.Size([1, 27541]) Labels shape: torch.Size([1, 27541]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 21491 Attention mask shape: torch.Size([1, 1, 21491, 21491]) Position ids shape: torch.Size([1, 21491]) Input IDs shape: torch.Size([1, 21491]) Labels shape: torch.Size([1, 21491]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 7722 Attention mask shape: torch.Size([1, 1, 7722, 7722]) Position ids shape: torch.Size([1, 7722]) Input IDs shape: torch.Size([1, 7722]) Labels shape: torch.Size([1, 7722]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 21596 Attention mask shape: torch.Size([1, 1, 21596, 21596]) Position ids shape: torch.Size([1, 21596]) Input IDs shape: torch.Size([1, 21596]) Labels shape: torch.Size([1, 21596]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 29875 Attention mask shape: torch.Size([1, 1, 29875, 29875]) Position ids shape: torch.Size([1, 29875]) Input IDs shape: torch.Size([1, 29875]) Labels shape: torch.Size([1, 29875]) Final batch size: 1, sequence length: 20827 Attention mask shape: torch.Size([1, 1, 20827, 20827]) Position ids shape: torch.Size([1, 20827]) Input IDs shape: torch.Size([1, 20827]) Labels shape: torch.Size([1, 20827]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 20363 Attention mask shape: torch.Size([1, 1, 20363, 20363]) Position ids shape: torch.Size([1, 20363]) Input IDs shape: torch.Size([1, 20363]) Labels shape: torch.Size([1, 20363]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 25014 Attention mask shape: torch.Size([1, 1, 25014, 25014]) Position ids shape: torch.Size([1, 25014]) Input IDs shape: torch.Size([1, 25014]) Labels shape: torch.Size([1, 25014]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 30623 Attention mask shape: torch.Size([1, 1, 30623, 30623]) Position ids shape: torch.Size([1, 30623]) Input IDs shape: torch.Size([1, 30623]) Labels shape: torch.Size([1, 30623]) Final batch size: 1, sequence length: 17400 Attention mask shape: torch.Size([1, 1, 17400, 17400]) Position ids shape: torch.Size([1, 17400]) Input IDs shape: torch.Size([1, 17400]) Labels shape: torch.Size([1, 17400]) Final batch size: 1, sequence length: 14009 Attention mask shape: torch.Size([1, 1, 14009, 14009]) Position ids shape: torch.Size([1, 14009]) Input IDs shape: torch.Size([1, 14009]) Labels shape: torch.Size([1, 14009]) Final batch size: 1, sequence length: 19705 Attention mask shape: torch.Size([1, 1, 19705, 19705]) Position ids shape: torch.Size([1, 19705]) Input IDs shape: torch.Size([1, 19705]) Labels shape: torch.Size([1, 19705]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 28641 Attention mask shape: torch.Size([1, 1, 28641, 28641]) Position ids shape: torch.Size([1, 28641]) Input IDs shape: torch.Size([1, 28641]) Labels shape: torch.Size([1, 28641]) Final batch size: 1, sequence length: 27265 Attention mask shape: torch.Size([1, 1, 27265, 27265]) Position ids shape: torch.Size([1, 27265]) Input IDs shape: torch.Size([1, 27265]) Labels shape: torch.Size([1, 27265]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 33367 Attention mask shape: torch.Size([1, 1, 33367, 33367]) Position ids shape: torch.Size([1, 33367]) Input IDs shape: torch.Size([1, 33367]) Labels shape: torch.Size([1, 33367]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 15656 Attention mask shape: torch.Size([1, 1, 15656, 15656]) Position ids shape: torch.Size([1, 15656]) Input IDs shape: torch.Size([1, 15656]) Labels shape: torch.Size([1, 15656]) Final batch size: 1, sequence length: 22098 Attention mask shape: torch.Size([1, 1, 22098, 22098]) Position ids shape: torch.Size([1, 22098]) Input IDs shape: torch.Size([1, 22098]) Labels shape: torch.Size([1, 22098]) Final batch size: 1, sequence length: 24001 Attention mask shape: torch.Size([1, 1, 24001, 24001]) Position ids shape: torch.Size([1, 24001]) Input IDs shape: torch.Size([1, 24001]) Labels shape: torch.Size([1, 24001]) Final batch size: 1, sequence length: 27243 Attention mask shape: torch.Size([1, 1, 27243, 27243]) Position ids shape: torch.Size([1, 27243]) Input IDs shape: torch.Size([1, 27243]) Labels shape: torch.Size([1, 27243]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 26306 Attention mask shape: torch.Size([1, 1, 26306, 26306]) Position ids shape: torch.Size([1, 26306]) Input IDs shape: torch.Size([1, 26306]) Labels shape: torch.Size([1, 26306]) Final batch size: 1, sequence length: 18606 Attention mask shape: torch.Size([1, 1, 18606, 18606]) Position ids shape: torch.Size([1, 18606]) Input IDs shape: torch.Size([1, 18606]) Labels shape: torch.Size([1, 18606]) Final batch size: 1, sequence length: 38249 Attention mask shape: torch.Size([1, 1, 38249, 38249]) Position ids shape: torch.Size([1, 38249]) Input IDs shape: torch.Size([1, 38249]) Labels shape: torch.Size([1, 38249]) Final batch size: 1, sequence length: 29632 Attention mask shape: torch.Size([1, 1, 29632, 29632]) Position ids shape: torch.Size([1, 29632]) Input IDs shape: torch.Size([1, 29632]) Labels shape: torch.Size([1, 29632]) Final batch size: 1, sequence length: 30066 Attention mask shape: torch.Size([1, 1, 30066, 30066]) Position ids shape: torch.Size([1, 30066]) Input IDs shape: torch.Size([1, 30066]) Labels shape: torch.Size([1, 30066]) Final batch size: 1, sequence length: 30101 Attention mask shape: torch.Size([1, 1, 30101, 30101]) Position ids shape: torch.Size([1, 30101]) Input IDs shape: torch.Size([1, 30101]) Labels shape: torch.Size([1, 30101]) Final batch size: 1, sequence length: 24622 Attention mask shape: torch.Size([1, 1, 24622, 24622]) Position ids shape: torch.Size([1, 24622]) Input IDs shape: torch.Size([1, 24622]) Labels shape: torch.Size([1, 24622]) Final batch size: 1, sequence length: 39142 Attention mask shape: torch.Size([1, 1, 39142, 39142]) Position ids shape: torch.Size([1, 39142]) Input IDs shape: torch.Size([1, 39142]) Labels shape: torch.Size([1, 39142]) Final batch size: 1, sequence length: 21348 Attention mask shape: torch.Size([1, 1, 21348, 21348]) Position ids shape: torch.Size([1, 21348]) Input IDs shape: torch.Size([1, 21348]) Labels shape: torch.Size([1, 21348]) Final batch size: 1, sequence length: 40745 Final batch size: 1, sequence length: 17778 Attention mask shape: torch.Size([1, 1, 40745, 40745]) Position ids shape: torch.Size([1, 40745]) Input IDs shape: torch.Size([1, 40745]) Labels shape: torch.Size([1, 40745]) Attention mask shape: torch.Size([1, 1, 17778, 17778]) Position ids shape: torch.Size([1, 17778]) Input IDs shape: torch.Size([1, 17778]) Labels shape: torch.Size([1, 17778]) Final batch size: 1, sequence length: 22625 Attention mask shape: torch.Size([1, 1, 22625, 22625]) Position ids shape: torch.Size([1, 22625]) Input IDs shape: torch.Size([1, 22625]) Labels shape: torch.Size([1, 22625]) Final batch size: 1, sequence length: 15217 Attention mask shape: torch.Size([1, 1, 15217, 15217]) Position ids shape: torch.Size([1, 15217]) Input IDs shape: torch.Size([1, 15217]) Labels shape: torch.Size([1, 15217]) Final batch size: 1, sequence length: 27441 Attention mask shape: torch.Size([1, 1, 27441, 27441]) Position ids shape: torch.Size([1, 27441]) Input IDs shape: torch.Size([1, 27441]) Labels shape: torch.Size([1, 27441]) Final batch size: 1, sequence length: 18565 Attention mask shape: torch.Size([1, 1, 18565, 18565]) Position ids shape: torch.Size([1, 18565]) Input IDs shape: torch.Size([1, 18565]) Labels shape: torch.Size([1, 18565]) Final batch size: 1, sequence length: 13509 Attention mask shape: torch.Size([1, 1, 13509, 13509]) Position ids shape: torch.Size([1, 13509]) Input IDs shape: torch.Size([1, 13509]) Labels shape: torch.Size([1, 13509]) Final batch size: 1, sequence length: 21567 Attention mask shape: torch.Size([1, 1, 21567, 21567]) Position ids shape: torch.Size([1, 21567]) Input IDs shape: torch.Size([1, 21567]) Labels shape: torch.Size([1, 21567]) Final batch size: 1, sequence length: 26939 Attention mask shape: torch.Size([1, 1, 26939, 26939]) Position ids shape: torch.Size([1, 26939]) Input IDs shape: torch.Size([1, 26939]) Labels shape: torch.Size([1, 26939]) Final batch size: 1, sequence length: 30802 Attention mask shape: torch.Size([1, 1, 30802, 30802]) Position ids shape: torch.Size([1, 30802]) Input IDs shape: torch.Size([1, 30802]) Labels shape: torch.Size([1, 30802]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36271 Attention mask shape: torch.Size([1, 1, 36271, 36271]) Position ids shape: torch.Size([1, 36271]) Input IDs shape: torch.Size([1, 36271]) Labels shape: torch.Size([1, 36271]) Final batch size: 1, sequence length: 40496 Attention mask shape: torch.Size([1, 1, 40496, 40496]) Position ids shape: torch.Size([1, 40496]) Input IDs shape: torch.Size([1, 40496]) Labels shape: torch.Size([1, 40496]) Final batch size: 1, sequence length: 37945 Attention mask shape: torch.Size([1, 1, 37945, 37945]) Position ids shape: torch.Size([1, 37945]) Input IDs shape: torch.Size([1, 37945]) Labels shape: torch.Size([1, 37945]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32609 Attention mask shape: torch.Size([1, 1, 32609, 32609]) Position ids shape: torch.Size([1, 32609]) Input IDs shape: torch.Size([1, 32609]) Labels shape: torch.Size([1, 32609]) Final batch size: 1, sequence length: 35153 Attention mask shape: torch.Size([1, 1, 35153, 35153]) Position ids shape: torch.Size([1, 35153]) Input IDs shape: torch.Size([1, 35153]) Labels shape: torch.Size([1, 35153]) Final batch size: 1, sequence length: 33839 Attention mask shape: torch.Size([1, 1, 33839, 33839]) Position ids shape: torch.Size([1, 33839]) Input IDs shape: torch.Size([1, 33839]) Labels shape: torch.Size([1, 33839]) Final batch size: 1, sequence length: 25447 Attention mask shape: torch.Size([1, 1, 25447, 25447]) Position ids shape: torch.Size([1, 25447]) Input IDs shape: torch.Size([1, 25447]) Labels shape: torch.Size([1, 25447]) Final batch size: 1, sequence length: 36511 Attention mask shape: torch.Size([1, 1, 36511, 36511]) Position ids shape: torch.Size([1, 36511]) Input IDs shape: torch.Size([1, 36511]) Labels shape: torch.Size([1, 36511]) Final batch size: 1, sequence length: 39836 Attention mask shape: torch.Size([1, 1, 39836, 39836]) Position ids shape: torch.Size([1, 39836]) Input IDs shape: torch.Size([1, 39836]) Labels shape: torch.Size([1, 39836]) Final batch size: 1, sequence length: 23258 Attention mask shape: torch.Size([1, 1, 23258, 23258]) Position ids shape: torch.Size([1, 23258]) Input IDs shape: torch.Size([1, 23258]) Labels shape: torch.Size([1, 23258]) Final batch size: 1, sequence length: 15508 Attention mask shape: torch.Size([1, 1, 15508, 15508]) Position ids shape: torch.Size([1, 15508]) Input IDs shape: torch.Size([1, 15508]) Labels shape: torch.Size([1, 15508]) Final batch size: 1, sequence length: 16587 Attention mask shape: torch.Size([1, 1, 16587, 16587]) Position ids shape: torch.Size([1, 16587]) Input IDs shape: torch.Size([1, 16587]) Labels shape: torch.Size([1, 16587]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30109 Attention mask shape: torch.Size([1, 1, 30109, 30109]) Position ids shape: torch.Size([1, 30109]) Input IDs shape: torch.Size([1, 30109]) Labels shape: torch.Size([1, 30109]) Final batch size: 1, sequence length: 32352 Attention mask shape: torch.Size([1, 1, 32352, 32352]) Position ids shape: torch.Size([1, 32352]) Input IDs shape: torch.Size([1, 32352]) Labels shape: torch.Size([1, 32352]) Final batch size: 1, sequence length: 20611 Attention mask shape: torch.Size([1, 1, 20611, 20611]) Position ids shape: torch.Size([1, 20611]) Input IDs shape: torch.Size([1, 20611]) Labels shape: torch.Size([1, 20611]) Final batch size: 1, sequence length: 19036 Attention mask shape: torch.Size([1, 1, 19036, 19036]) Position ids shape: torch.Size([1, 19036]) Input IDs shape: torch.Size([1, 19036]) Labels shape: torch.Size([1, 19036]) Final batch size: 1, sequence length: 9947 Attention mask shape: torch.Size([1, 1, 9947, 9947]) Position ids shape: torch.Size([1, 9947]) Input IDs shape: torch.Size([1, 9947]) Labels shape: torch.Size([1, 9947]) Final batch size: 1, sequence length: 16677 Attention mask shape: torch.Size([1, 1, 16677, 16677]) Position ids shape: torch.Size([1, 16677]) Input IDs shape: torch.Size([1, 16677]) Labels shape: torch.Size([1, 16677]) Final batch size: 1, sequence length: 21758 Attention mask shape: torch.Size([1, 1, 21758, 21758]) Position ids shape: torch.Size([1, 21758]) Input IDs shape: torch.Size([1, 21758]) Labels shape: torch.Size([1, 21758]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20422 Attention mask shape: torch.Size([1, 1, 20422, 20422]) Position ids shape: torch.Size([1, 20422]) Input IDs shape: torch.Size([1, 20422]) Labels shape: torch.Size([1, 20422]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 13095 Attention mask shape: torch.Size([1, 1, 13095, 13095]) Position ids shape: torch.Size([1, 13095]) Input IDs shape: torch.Size([1, 13095]) Labels shape: torch.Size([1, 13095]) Final batch size: 1, sequence length: 13297 Attention mask shape: torch.Size([1, 1, 13297, 13297]) Position ids shape: torch.Size([1, 13297]) Input IDs shape: torch.Size([1, 13297]) Labels shape: torch.Size([1, 13297]) Final batch size: 1, sequence length: 34937 Attention mask shape: torch.Size([1, 1, 34937, 34937]) Position ids shape: torch.Size([1, 34937]) Input IDs shape: torch.Size([1, 34937]) Labels shape: torch.Size([1, 34937]) Final batch size: 1, sequence length: 24298 Attention mask shape: torch.Size([1, 1, 24298, 24298]) Position ids shape: torch.Size([1, 24298]) Input IDs shape: torch.Size([1, 24298]) Labels shape: torch.Size([1, 24298]) Final batch size: 1, sequence length: 22786 Attention mask shape: torch.Size([1, 1, 22786, 22786]) Position ids shape: torch.Size([1, 22786]) Input IDs shape: torch.Size([1, 22786]) Labels shape: torch.Size([1, 22786]) Final batch size: 1, sequence length: 12653 Attention mask shape: torch.Size([1, 1, 12653, 12653]) Position ids shape: torch.Size([1, 12653]) Input IDs shape: torch.Size([1, 12653]) Labels shape: torch.Size([1, 12653]) Final batch size: 1, sequence length: 34142 Attention mask shape: torch.Size([1, 1, 34142, 34142]) Position ids shape: torch.Size([1, 34142]) Input IDs shape: torch.Size([1, 34142]) Labels shape: torch.Size([1, 34142]) Final batch size: 1, sequence length: 12224 Attention mask shape: torch.Size([1, 1, 12224, 12224]) Position ids shape: torch.Size([1, 12224]) Input IDs shape: torch.Size([1, 12224]) Labels shape: torch.Size([1, 12224]) Final batch size: 1, sequence length: 32472 Attention mask shape: torch.Size([1, 1, 32472, 32472]) Position ids shape: torch.Size([1, 32472]) Input IDs shape: torch.Size([1, 32472]) Labels shape: torch.Size([1, 32472]) Final batch size: 1, sequence length: 28634 Attention mask shape: torch.Size([1, 1, 28634, 28634]) Position ids shape: torch.Size([1, 28634]) Input IDs shape: torch.Size([1, 28634]) Labels shape: torch.Size([1, 28634]) Final batch size: 1, sequence length: 32753 Attention mask shape: torch.Size([1, 1, 32753, 32753]) Position ids shape: torch.Size([1, 32753]) Input IDs shape: torch.Size([1, 32753]) Labels shape: torch.Size([1, 32753]) Final batch size: 1, sequence length: 21683 Attention mask shape: torch.Size([1, 1, 21683, 21683]) Position ids shape: torch.Size([1, 21683]) Input IDs shape: torch.Size([1, 21683]) Labels shape: torch.Size([1, 21683]) Final batch size: 1, sequence length: 14995 Attention mask shape: torch.Size([1, 1, 14995, 14995]) Position ids shape: torch.Size([1, 14995]) Input IDs shape: torch.Size([1, 14995]) Labels shape: torch.Size([1, 14995]) Final batch size: 1, sequence length: 17373 Attention mask shape: torch.Size([1, 1, 17373, 17373]) Position ids shape: torch.Size([1, 17373]) Input IDs shape: torch.Size([1, 17373]) Labels shape: torch.Size([1, 17373]) Final batch size: 1, sequence length: 32564 Attention mask shape: torch.Size([1, 1, 32564, 32564]) Position ids shape: torch.Size([1, 32564]) Input IDs shape: torch.Size([1, 32564]) Labels shape: torch.Size([1, 32564]) Final batch size: 1, sequence length: 22623 Attention mask shape: torch.Size([1, 1, 22623, 22623]) Position ids shape: torch.Size([1, 22623]) Input IDs shape: torch.Size([1, 22623]) Labels shape: torch.Size([1, 22623]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24424 Attention mask shape: torch.Size([1, 1, 24424, 24424]) Position ids shape: torch.Size([1, 24424]) Input IDs shape: torch.Size([1, 24424]) Labels shape: torch.Size([1, 24424]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19586 Attention mask shape: torch.Size([1, 1, 19586, 19586]) Position ids shape: torch.Size([1, 19586]) Input IDs shape: torch.Size([1, 19586]) Labels shape: torch.Size([1, 19586]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 35864 Attention mask shape: torch.Size([1, 1, 35864, 35864]) Position ids shape: torch.Size([1, 35864]) Input IDs shape: torch.Size([1, 35864]) Labels shape: torch.Size([1, 35864]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32344 Attention mask shape: torch.Size([1, 1, 32344, 32344]) Position ids shape: torch.Size([1, 32344]) Input IDs shape: torch.Size([1, 32344]) Labels shape: torch.Size([1, 32344]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26537 Attention mask shape: torch.Size([1, 1, 26537, 26537]) Position ids shape: torch.Size([1, 26537]) Input IDs shape: torch.Size([1, 26537]) Labels shape: torch.Size([1, 26537]) Final batch size: 1, sequence length: 32919 Attention mask shape: torch.Size([1, 1, 32919, 32919]) Position ids shape: torch.Size([1, 32919]) Input IDs shape: torch.Size([1, 32919]) Labels shape: torch.Size([1, 32919]) Final batch size: 1, sequence length: 32514 Attention mask shape: torch.Size([1, 1, 32514, 32514]) Position ids shape: torch.Size([1, 32514]) Input IDs shape: torch.Size([1, 32514]) Labels shape: torch.Size([1, 32514]) Final batch size: 1, sequence length: 37168 Attention mask shape: torch.Size([1, 1, 37168, 37168]) Position ids shape: torch.Size([1, 37168]) Input IDs shape: torch.Size([1, 37168]) Labels shape: torch.Size([1, 37168]) Final batch size: 1, sequence length: 23362 Attention mask shape: torch.Size([1, 1, 23362, 23362]) Position ids shape: torch.Size([1, 23362]) Input IDs shape: torch.Size([1, 23362]) Labels shape: torch.Size([1, 23362]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28861 Attention mask shape: torch.Size([1, 1, 28861, 28861]) Position ids shape: torch.Size([1, 28861]) Input IDs shape: torch.Size([1, 28861]) Labels shape: torch.Size([1, 28861]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26006 Attention mask shape: torch.Size([1, 1, 26006, 26006]) Position ids shape: torch.Size([1, 26006]) Input IDs shape: torch.Size([1, 26006]) Labels shape: torch.Size([1, 26006]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2627, 'grad_norm': 0.27396124133167743, 'learning_rate': 5.522642316338268e-06, 'num_tokens': -inf, 'epoch': 4.12} Final batch size: 1, sequence length: 6316 Attention mask shape: torch.Size([1, 1, 6316, 6316]) Position ids shape: torch.Size([1, 6316]) Input IDs shape: torch.Size([1, 6316]) Labels shape: torch.Size([1, 6316]) Final batch size: 1, sequence length: 7360 Attention mask shape: torch.Size([1, 1, 7360, 7360]) Position ids shape: torch.Size([1, 7360]) Input IDs shape: torch.Size([1, 7360]) Labels shape: torch.Size([1, 7360]) Final batch size: 1, sequence length: 4858 Attention mask shape: torch.Size([1, 1, 4858, 4858]) Position ids shape: torch.Size([1, 4858]) Input IDs shape: torch.Size([1, 4858]) Labels shape: torch.Size([1, 4858]) Final batch size: 1, sequence length: 10301 Attention mask shape: torch.Size([1, 1, 10301, 10301]) Position ids shape: torch.Size([1, 10301]) Input IDs shape: torch.Size([1, 10301]) Labels shape: torch.Size([1, 10301]) Final batch size: 1, sequence length: 11548 Attention mask shape: torch.Size([1, 1, 11548, 11548]) Position ids shape: torch.Size([1, 11548]) Input IDs shape: torch.Size([1, 11548]) Labels shape: torch.Size([1, 11548]) Final batch size: 1, sequence length: 11728 Attention mask shape: torch.Size([1, 1, 11728, 11728]) Position ids shape: torch.Size([1, 11728]) Input IDs shape: torch.Size([1, 11728]) Labels shape: torch.Size([1, 11728]) Final batch size: 1, sequence length: 12293 Attention mask shape: torch.Size([1, 1, 12293, 12293]) Position ids shape: torch.Size([1, 12293]) Input IDs shape: torch.Size([1, 12293]) Labels shape: torch.Size([1, 12293]) Final batch size: 1, sequence length: 14363 Attention mask shape: torch.Size([1, 1, 14363, 14363]) Position ids shape: torch.Size([1, 14363]) Input IDs shape: torch.Size([1, 14363]) Labels shape: torch.Size([1, 14363]) Final batch size: 1, sequence length: 16827 Attention mask shape: torch.Size([1, 1, 16827, 16827]) Position ids shape: torch.Size([1, 16827]) Input IDs shape: torch.Size([1, 16827]) Labels shape: torch.Size([1, 16827]) Final batch size: 1, sequence length: 13355 Attention mask shape: torch.Size([1, 1, 13355, 13355]) Position ids shape: torch.Size([1, 13355]) Input IDs shape: torch.Size([1, 13355]) Labels shape: torch.Size([1, 13355]) Final batch size: 1, sequence length: 14587 Attention mask shape: torch.Size([1, 1, 14587, 14587]) Position ids shape: torch.Size([1, 14587]) Input IDs shape: torch.Size([1, 14587]) Labels shape: torch.Size([1, 14587]) Final batch size: 1, sequence length: 14704 Attention mask shape: torch.Size([1, 1, 14704, 14704]) Position ids shape: torch.Size([1, 14704]) Input IDs shape: torch.Size([1, 14704]) Labels shape: torch.Size([1, 14704]) Final batch size: 1, sequence length: 17108 Attention mask shape: torch.Size([1, 1, 17108, 17108]) Position ids shape: torch.Size([1, 17108]) Input IDs shape: torch.Size([1, 17108]) Labels shape: torch.Size([1, 17108]) Final batch size: 1, sequence length: 16536 Attention mask shape: torch.Size([1, 1, 16536, 16536]) Position ids shape: torch.Size([1, 16536]) Input IDs shape: torch.Size([1, 16536]) Labels shape: torch.Size([1, 16536]) Final batch size: 1, sequence length: 13638 Attention mask shape: torch.Size([1, 1, 13638, 13638]) Position ids shape: torch.Size([1, 13638]) Input IDs shape: torch.Size([1, 13638]) Labels shape: torch.Size([1, 13638]) Final batch size: 1, sequence length: 17415 Attention mask shape: torch.Size([1, 1, 17415, 17415]) Position ids shape: torch.Size([1, 17415]) Input IDs shape: torch.Size([1, 17415]) Labels shape: torch.Size([1, 17415]) Final batch size: 1, sequence length: 18653 Attention mask shape: torch.Size([1, 1, 18653, 18653]) Position ids shape: torch.Size([1, 18653]) Input IDs shape: torch.Size([1, 18653]) Labels shape: torch.Size([1, 18653]) Final batch size: 1, sequence length: 18927 Final batch size: 1, sequence length: 18741 Attention mask shape: torch.Size([1, 1, 18927, 18927]) Position ids shape: torch.Size([1, 18927]) Input IDs shape: torch.Size([1, 18927]) Labels shape: torch.Size([1, 18927]) Attention mask shape: torch.Size([1, 1, 18741, 18741]) Position ids shape: torch.Size([1, 18741]) Input IDs shape: torch.Size([1, 18741]) Labels shape: torch.Size([1, 18741]) Final batch size: 1, sequence length: 16750 Attention mask shape: torch.Size([1, 1, 16750, 16750]) Position ids shape: torch.Size([1, 16750]) Input IDs shape: torch.Size([1, 16750]) Labels shape: torch.Size([1, 16750]) Final batch size: 1, sequence length: 18496 Attention mask shape: torch.Size([1, 1, 18496, 18496]) Position ids shape: torch.Size([1, 18496]) Input IDs shape: torch.Size([1, 18496]) Labels shape: torch.Size([1, 18496]) Final batch size: 1, sequence length: 7221 Attention mask shape: torch.Size([1, 1, 7221, 7221]) Position ids shape: torch.Size([1, 7221]) Input IDs shape: torch.Size([1, 7221]) Labels shape: torch.Size([1, 7221]) Final batch size: 1, sequence length: 17220 Attention mask shape: torch.Size([1, 1, 17220, 17220]) Position ids shape: torch.Size([1, 17220]) Input IDs shape: torch.Size([1, 17220]) Labels shape: torch.Size([1, 17220]) Final batch size: 1, sequence length: 18393 Attention mask shape: torch.Size([1, 1, 18393, 18393]) Position ids shape: torch.Size([1, 18393]) Input IDs shape: torch.Size([1, 18393]) Labels shape: torch.Size([1, 18393]) Final batch size: 1, sequence length: 20933 Attention mask shape: torch.Size([1, 1, 20933, 20933]) Position ids shape: torch.Size([1, 20933]) Input IDs shape: torch.Size([1, 20933]) Labels shape: torch.Size([1, 20933]) Final batch size: 1, sequence length: 17733 Attention mask shape: torch.Size([1, 1, 17733, 17733]) Position ids shape: torch.Size([1, 17733]) Input IDs shape: torch.Size([1, 17733]) Labels shape: torch.Size([1, 17733]) Final batch size: 1, sequence length: 19414 Attention mask shape: torch.Size([1, 1, 19414, 19414]) Position ids shape: torch.Size([1, 19414]) Input IDs shape: torch.Size([1, 19414]) Labels shape: torch.Size([1, 19414]) Final batch size: 1, sequence length: 20612 Attention mask shape: torch.Size([1, 1, 20612, 20612]) Position ids shape: torch.Size([1, 20612]) Input IDs shape: torch.Size([1, 20612]) Labels shape: torch.Size([1, 20612]) Final batch size: 1, sequence length: 21420 Attention mask shape: torch.Size([1, 1, 21420, 21420]) Position ids shape: torch.Size([1, 21420]) Input IDs shape: torch.Size([1, 21420]) Labels shape: torch.Size([1, 21420]) Final batch size: 1, sequence length: 22887 Attention mask shape: torch.Size([1, 1, 22887, 22887]) Position ids shape: torch.Size([1, 22887]) Input IDs shape: torch.Size([1, 22887]) Labels shape: torch.Size([1, 22887]) Final batch size: 1, sequence length: 22004 Attention mask shape: torch.Size([1, 1, 22004, 22004]) Position ids shape: torch.Size([1, 22004]) Input IDs shape: torch.Size([1, 22004]) Labels shape: torch.Size([1, 22004]) Final batch size: 1, sequence length: 11067 Attention mask shape: torch.Size([1, 1, 11067, 11067]) Position ids shape: torch.Size([1, 11067]) Input IDs shape: torch.Size([1, 11067]) Labels shape: torch.Size([1, 11067]) Final batch size: 1, sequence length: 22391 Attention mask shape: torch.Size([1, 1, 22391, 22391]) Position ids shape: torch.Size([1, 22391]) Input IDs shape: torch.Size([1, 22391]) Labels shape: torch.Size([1, 22391]) Final batch size: 1, sequence length: 10719 Attention mask shape: torch.Size([1, 1, 10719, 10719]) Position ids shape: torch.Size([1, 10719]) Input IDs shape: torch.Size([1, 10719]) Labels shape: torch.Size([1, 10719]) Final batch size: 1, sequence length: 25747 Attention mask shape: torch.Size([1, 1, 25747, 25747]) Position ids shape: torch.Size([1, 25747]) Input IDs shape: torch.Size([1, 25747]) Labels shape: torch.Size([1, 25747]) Final batch size: 1, sequence length: 24988 Attention mask shape: torch.Size([1, 1, 24988, 24988]) Position ids shape: torch.Size([1, 24988]) Input IDs shape: torch.Size([1, 24988]) Labels shape: torch.Size([1, 24988]) Final batch size: 1, sequence length: 11184 Attention mask shape: torch.Size([1, 1, 11184, 11184]) Position ids shape: torch.Size([1, 11184]) Input IDs shape: torch.Size([1, 11184]) Labels shape: torch.Size([1, 11184]) Final batch size: 1, sequence length: 25477 Attention mask shape: torch.Size([1, 1, 25477, 25477]) Position ids shape: torch.Size([1, 25477]) Input IDs shape: torch.Size([1, 25477]) Labels shape: torch.Size([1, 25477]) Final batch size: 1, sequence length: 20579 Attention mask shape: torch.Size([1, 1, 20579, 20579]) Position ids shape: torch.Size([1, 20579]) Input IDs shape: torch.Size([1, 20579]) Labels shape: torch.Size([1, 20579]) Final batch size: 1, sequence length: 15317 Attention mask shape: torch.Size([1, 1, 15317, 15317]) Position ids shape: torch.Size([1, 15317]) Input IDs shape: torch.Size([1, 15317]) Labels shape: torch.Size([1, 15317]) Final batch size: 1, sequence length: 16915 Attention mask shape: torch.Size([1, 1, 16915, 16915]) Position ids shape: torch.Size([1, 16915]) Input IDs shape: torch.Size([1, 16915]) Labels shape: torch.Size([1, 16915]) Final batch size: 1, sequence length: 17395 Attention mask shape: torch.Size([1, 1, 17395, 17395]) Position ids shape: torch.Size([1, 17395]) Input IDs shape: torch.Size([1, 17395]) Labels shape: torch.Size([1, 17395]) Final batch size: 1, sequence length: 26663 Attention mask shape: torch.Size([1, 1, 26663, 26663]) Position ids shape: torch.Size([1, 26663]) Input IDs shape: torch.Size([1, 26663]) Labels shape: torch.Size([1, 26663]) Final batch size: 1, sequence length: 27447 Attention mask shape: torch.Size([1, 1, 27447, 27447]) Position ids shape: torch.Size([1, 27447]) Input IDs shape: torch.Size([1, 27447]) Labels shape: torch.Size([1, 27447]) Final batch size: 1, sequence length: 25651 Attention mask shape: torch.Size([1, 1, 25651, 25651]) Position ids shape: torch.Size([1, 25651]) Input IDs shape: torch.Size([1, 25651]) Labels shape: torch.Size([1, 25651]) Final batch size: 1, sequence length: 19552 Attention mask shape: torch.Size([1, 1, 19552, 19552]) Position ids shape: torch.Size([1, 19552]) Input IDs shape: torch.Size([1, 19552]) Labels shape: torch.Size([1, 19552]) Final batch size: 1, sequence length: 30031 Attention mask shape: torch.Size([1, 1, 30031, 30031]) Position ids shape: torch.Size([1, 30031]) Input IDs shape: torch.Size([1, 30031]) Labels shape: torch.Size([1, 30031]) Final batch size: 1, sequence length: 27480 Attention mask shape: torch.Size([1, 1, 27480, 27480]) Position ids shape: torch.Size([1, 27480]) Input IDs shape: torch.Size([1, 27480]) Labels shape: torch.Size([1, 27480]) Final batch size: 1, sequence length: 19869 Attention mask shape: torch.Size([1, 1, 19869, 19869]) Position ids shape: torch.Size([1, 19869]) Input IDs shape: torch.Size([1, 19869]) Labels shape: torch.Size([1, 19869]) Final batch size: 1, sequence length: 28777 Attention mask shape: torch.Size([1, 1, 28777, 28777]) Position ids shape: torch.Size([1, 28777]) Input IDs shape: torch.Size([1, 28777]) Labels shape: torch.Size([1, 28777]) Final batch size: 1, sequence length: 27334 Final batch size: 1, sequence length: 30981 Attention mask shape: torch.Size([1, 1, 30981, 30981]) Position ids shape: torch.Size([1, 30981]) Input IDs shape: torch.Size([1, 30981]) Labels shape: torch.Size([1, 30981]) Attention mask shape: torch.Size([1, 1, 27334, 27334]) Position ids shape: torch.Size([1, 27334]) Input IDs shape: torch.Size([1, 27334]) Labels shape: torch.Size([1, 27334]) Final batch size: 1, sequence length: 26033 Attention mask shape: torch.Size([1, 1, 26033, 26033]) Position ids shape: torch.Size([1, 26033]) Input IDs shape: torch.Size([1, 26033]) Labels shape: torch.Size([1, 26033]) Final batch size: 1, sequence length: 16953 Attention mask shape: torch.Size([1, 1, 16953, 16953]) Position ids shape: torch.Size([1, 16953]) Input IDs shape: torch.Size([1, 16953]) Labels shape: torch.Size([1, 16953]) Final batch size: 1, sequence length: 24782 Attention mask shape: torch.Size([1, 1, 24782, 24782]) Position ids shape: torch.Size([1, 24782]) Input IDs shape: torch.Size([1, 24782]) Labels shape: torch.Size([1, 24782]) Final batch size: 1, sequence length: 18376 Attention mask shape: torch.Size([1, 1, 18376, 18376]) Position ids shape: torch.Size([1, 18376]) Input IDs shape: torch.Size([1, 18376]) Labels shape: torch.Size([1, 18376]) Final batch size: 1, sequence length: 17911 Attention mask shape: torch.Size([1, 1, 17911, 17911]) Position ids shape: torch.Size([1, 17911]) Input IDs shape: torch.Size([1, 17911]) Labels shape: torch.Size([1, 17911]) Final batch size: 1, sequence length: 21988 Attention mask shape: torch.Size([1, 1, 21988, 21988]) Position ids shape: torch.Size([1, 21988]) Input IDs shape: torch.Size([1, 21988]) Labels shape: torch.Size([1, 21988]) Final batch size: 1, sequence length: 30601 Attention mask shape: torch.Size([1, 1, 30601, 30601]) Position ids shape: torch.Size([1, 30601]) Input IDs shape: torch.Size([1, 30601]) Labels shape: torch.Size([1, 30601]) Final batch size: 1, sequence length: 15924 Attention mask shape: torch.Size([1, 1, 15924, 15924]) Position ids shape: torch.Size([1, 15924]) Input IDs shape: torch.Size([1, 15924]) Labels shape: torch.Size([1, 15924]) Final batch size: 1, sequence length: 32466 Attention mask shape: torch.Size([1, 1, 32466, 32466]) Position ids shape: torch.Size([1, 32466]) Input IDs shape: torch.Size([1, 32466]) Labels shape: torch.Size([1, 32466]) Final batch size: 1, sequence length: 28060 Attention mask shape: torch.Size([1, 1, 28060, 28060]) Position ids shape: torch.Size([1, 28060]) Input IDs shape: torch.Size([1, 28060]) Labels shape: torch.Size([1, 28060]) Final batch size: 1, sequence length: 27385 Attention mask shape: torch.Size([1, 1, 27385, 27385]) Position ids shape: torch.Size([1, 27385]) Input IDs shape: torch.Size([1, 27385]) Labels shape: torch.Size([1, 27385]) Final batch size: 1, sequence length: 12245 Attention mask shape: torch.Size([1, 1, 12245, 12245]) Position ids shape: torch.Size([1, 12245]) Input IDs shape: torch.Size([1, 12245]) Labels shape: torch.Size([1, 12245]) Final batch size: 1, sequence length: 14873 Attention mask shape: torch.Size([1, 1, 14873, 14873]) Position ids shape: torch.Size([1, 14873]) Input IDs shape: torch.Size([1, 14873]) Labels shape: torch.Size([1, 14873]) Final batch size: 1, sequence length: 37158 Attention mask shape: torch.Size([1, 1, 37158, 37158]) Position ids shape: torch.Size([1, 37158]) Input IDs shape: torch.Size([1, 37158]) Labels shape: torch.Size([1, 37158]) Final batch size: 1, sequence length: 33725 Attention mask shape: torch.Size([1, 1, 33725, 33725]) Position ids shape: torch.Size([1, 33725]) Input IDs shape: torch.Size([1, 33725]) Labels shape: torch.Size([1, 33725]) Final batch size: 1, sequence length: 37025 Attention mask shape: torch.Size([1, 1, 37025, 37025]) Position ids shape: torch.Size([1, 37025]) Input IDs shape: torch.Size([1, 37025]) Labels shape: torch.Size([1, 37025]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 40397 Attention mask shape: torch.Size([1, 1, 40397, 40397]) Position ids shape: torch.Size([1, 40397]) Input IDs shape: torch.Size([1, 40397]) Labels shape: torch.Size([1, 40397]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 37738 Attention mask shape: torch.Size([1, 1, 37738, 37738]) Position ids shape: torch.Size([1, 37738]) Input IDs shape: torch.Size([1, 37738]) Labels shape: torch.Size([1, 37738]) Final batch size: 1, sequence length: 32835 Attention mask shape: torch.Size([1, 1, 32835, 32835]) Position ids shape: torch.Size([1, 32835]) Input IDs shape: torch.Size([1, 32835]) Labels shape: torch.Size([1, 32835]) Final batch size: 1, sequence length: 37511 Attention mask shape: torch.Size([1, 1, 37511, 37511]) Position ids shape: torch.Size([1, 37511]) Input IDs shape: torch.Size([1, 37511]) Labels shape: torch.Size([1, 37511]) Final batch size: 1, sequence length: 36175 Attention mask shape: torch.Size([1, 1, 36175, 36175]) Position ids shape: torch.Size([1, 36175]) Input IDs shape: torch.Size([1, 36175]) Labels shape: torch.Size([1, 36175]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38986 Attention mask shape: torch.Size([1, 1, 38986, 38986]) Position ids shape: torch.Size([1, 38986]) Input IDs shape: torch.Size([1, 38986]) Labels shape: torch.Size([1, 38986]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31348 Attention mask shape: torch.Size([1, 1, 31348, 31348]) Position ids shape: torch.Size([1, 31348]) Input IDs shape: torch.Size([1, 31348]) Labels shape: torch.Size([1, 31348]) Final batch size: 1, sequence length: 29586 Attention mask shape: torch.Size([1, 1, 29586, 29586]) Position ids shape: torch.Size([1, 29586]) Input IDs shape: torch.Size([1, 29586]) Labels shape: torch.Size([1, 29586]) Final batch size: 1, sequence length: 26479 Attention mask shape: torch.Size([1, 1, 26479, 26479]) Position ids shape: torch.Size([1, 26479]) Input IDs shape: torch.Size([1, 26479]) Labels shape: torch.Size([1, 26479]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40269 Attention mask shape: torch.Size([1, 1, 40269, 40269]) Position ids shape: torch.Size([1, 40269]) Input IDs shape: torch.Size([1, 40269]) Labels shape: torch.Size([1, 40269]) Final batch size: 1, sequence length: 36130 Attention mask shape: torch.Size([1, 1, 36130, 36130]) Position ids shape: torch.Size([1, 36130]) Input IDs shape: torch.Size([1, 36130]) Labels shape: torch.Size([1, 36130]) Final batch size: 1, sequence length: 31879 Attention mask shape: torch.Size([1, 1, 31879, 31879]) Position ids shape: torch.Size([1, 31879]) Input IDs shape: torch.Size([1, 31879]) Labels shape: torch.Size([1, 31879]) Final batch size: 1, sequence length: 40237 Attention mask shape: torch.Size([1, 1, 40237, 40237]) Position ids shape: torch.Size([1, 40237]) Input IDs shape: torch.Size([1, 40237]) Labels shape: torch.Size([1, 40237]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32318 Attention mask shape: torch.Size([1, 1, 32318, 32318]) Position ids shape: torch.Size([1, 32318]) Input IDs shape: torch.Size([1, 32318]) Labels shape: torch.Size([1, 32318]) Final batch size: 1, sequence length: 27291 Attention mask shape: torch.Size([1, 1, 27291, 27291]) Position ids shape: torch.Size([1, 27291]) Input IDs shape: torch.Size([1, 27291]) Labels shape: torch.Size([1, 27291]) Final batch size: 1, sequence length: 20509 Attention mask shape: torch.Size([1, 1, 20509, 20509]) Position ids shape: torch.Size([1, 20509]) Input IDs shape: torch.Size([1, 20509]) Labels shape: torch.Size([1, 20509]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22205 Attention mask shape: torch.Size([1, 1, 22205, 22205]) Position ids shape: torch.Size([1, 22205]) Input IDs shape: torch.Size([1, 22205]) Labels shape: torch.Size([1, 22205]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33125 Attention mask shape: torch.Size([1, 1, 33125, 33125]) Position ids shape: torch.Size([1, 33125]) Input IDs shape: torch.Size([1, 33125]) Labels shape: torch.Size([1, 33125]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36947 Attention mask shape: torch.Size([1, 1, 36947, 36947]) Position ids shape: torch.Size([1, 36947]) Input IDs shape: torch.Size([1, 36947]) Labels shape: torch.Size([1, 36947]) Final batch size: 1, sequence length: 28631 Attention mask shape: torch.Size([1, 1, 28631, 28631]) Position ids shape: torch.Size([1, 28631]) Input IDs shape: torch.Size([1, 28631]) Labels shape: torch.Size([1, 28631]) Final batch size: 1, sequence length: 10198 Attention mask shape: torch.Size([1, 1, 10198, 10198]) Position ids shape: torch.Size([1, 10198]) Input IDs shape: torch.Size([1, 10198]) Labels shape: torch.Size([1, 10198]) Final batch size: 1, sequence length: 24939 Attention mask shape: torch.Size([1, 1, 24939, 24939]) Position ids shape: torch.Size([1, 24939]) Input IDs shape: torch.Size([1, 24939]) Labels shape: torch.Size([1, 24939]) Final batch size: 1, sequence length: 32247 Attention mask shape: torch.Size([1, 1, 32247, 32247]) Position ids shape: torch.Size([1, 32247]) Input IDs shape: torch.Size([1, 32247]) Labels shape: torch.Size([1, 32247]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21225 Attention mask shape: torch.Size([1, 1, 21225, 21225]) Position ids shape: torch.Size([1, 21225]) Input IDs shape: torch.Size([1, 21225]) Labels shape: torch.Size([1, 21225]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17711 Attention mask shape: torch.Size([1, 1, 17711, 17711]) Position ids shape: torch.Size([1, 17711]) Input IDs shape: torch.Size([1, 17711]) Labels shape: torch.Size([1, 17711]) Final batch size: 1, sequence length: 23740 Attention mask shape: torch.Size([1, 1, 23740, 23740]) Position ids shape: torch.Size([1, 23740]) Input IDs shape: torch.Size([1, 23740]) Labels shape: torch.Size([1, 23740]) Final batch size: 1, sequence length: 28046 Attention mask shape: torch.Size([1, 1, 28046, 28046]) Position ids shape: torch.Size([1, 28046]) Input IDs shape: torch.Size([1, 28046]) Labels shape: torch.Size([1, 28046]) Final batch size: 1, sequence length: 22896 Attention mask shape: torch.Size([1, 1, 22896, 22896]) Position ids shape: torch.Size([1, 22896]) Input IDs shape: torch.Size([1, 22896]) Labels shape: torch.Size([1, 22896]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28149 Attention mask shape: torch.Size([1, 1, 28149, 28149]) Position ids shape: torch.Size([1, 28149]) Input IDs shape: torch.Size([1, 28149]) Labels shape: torch.Size([1, 28149]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38360 Attention mask shape: torch.Size([1, 1, 38360, 38360]) Position ids shape: torch.Size([1, 38360]) Input IDs shape: torch.Size([1, 38360]) Labels shape: torch.Size([1, 38360]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32767 Attention mask shape: torch.Size([1, 1, 32767, 32767]) Position ids shape: torch.Size([1, 32767]) Input IDs shape: torch.Size([1, 32767]) Labels shape: torch.Size([1, 32767]) Final batch size: 1, sequence length: 36567 Attention mask shape: torch.Size([1, 1, 36567, 36567]) Position ids shape: torch.Size([1, 36567]) Input IDs shape: torch.Size([1, 36567]) Labels shape: torch.Size([1, 36567]) Final batch size: 1, sequence length: 34268 Attention mask shape: torch.Size([1, 1, 34268, 34268]) Position ids shape: torch.Size([1, 34268]) Input IDs shape: torch.Size([1, 34268]) Labels shape: torch.Size([1, 34268]) Final batch size: 1, sequence length: 32313 Attention mask shape: torch.Size([1, 1, 32313, 32313]) Position ids shape: torch.Size([1, 32313]) Input IDs shape: torch.Size([1, 32313]) Labels shape: torch.Size([1, 32313]) {'loss': 0.289, 'grad_norm': 0.3255988588661155, 'learning_rate': 5.2616797812147205e-06, 'num_tokens': -inf, 'epoch': 4.25} Final batch size: 1, sequence length: 7998 Attention mask shape: torch.Size([1, 1, 7998, 7998]) Position ids shape: torch.Size([1, 7998]) Input IDs shape: torch.Size([1, 7998]) Labels shape: torch.Size([1, 7998]) Final batch size: 1, sequence length: 7402 Attention mask shape: torch.Size([1, 1, 7402, 7402]) Position ids shape: torch.Size([1, 7402]) Input IDs shape: torch.Size([1, 7402]) Labels shape: torch.Size([1, 7402]) Final batch size: 1, sequence length: 8436 Attention mask shape: torch.Size([1, 1, 8436, 8436]) Position ids shape: torch.Size([1, 8436]) Input IDs shape: torch.Size([1, 8436]) Labels shape: torch.Size([1, 8436]) Final batch size: 1, sequence length: 10576 Attention mask shape: torch.Size([1, 1, 10576, 10576]) Position ids shape: torch.Size([1, 10576]) Input IDs shape: torch.Size([1, 10576]) Labels shape: torch.Size([1, 10576]) Final batch size: 1, sequence length: 8655 Attention mask shape: torch.Size([1, 1, 8655, 8655]) Position ids shape: torch.Size([1, 8655]) Input IDs shape: torch.Size([1, 8655]) Labels shape: torch.Size([1, 8655]) Final batch size: 1, sequence length: 5377 Attention mask shape: torch.Size([1, 1, 5377, 5377]) Position ids shape: torch.Size([1, 5377]) Input IDs shape: torch.Size([1, 5377]) Labels shape: torch.Size([1, 5377]) Final batch size: 1, sequence length: 11709 Attention mask shape: torch.Size([1, 1, 11709, 11709]) Position ids shape: torch.Size([1, 11709]) Input IDs shape: torch.Size([1, 11709]) Labels shape: torch.Size([1, 11709]) Final batch size: 1, sequence length: 7344 Attention mask shape: torch.Size([1, 1, 7344, 7344]) Position ids shape: torch.Size([1, 7344]) Input IDs shape: torch.Size([1, 7344]) Labels shape: torch.Size([1, 7344]) Final batch size: 1, sequence length: 11365 Attention mask shape: torch.Size([1, 1, 11365, 11365]) Position ids shape: torch.Size([1, 11365]) Input IDs shape: torch.Size([1, 11365]) Labels shape: torch.Size([1, 11365]) Final batch size: 1, sequence length: 14127 Attention mask shape: torch.Size([1, 1, 14127, 14127]) Position ids shape: torch.Size([1, 14127]) Input IDs shape: torch.Size([1, 14127]) Labels shape: torch.Size([1, 14127]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 15079 Attention mask shape: torch.Size([1, 1, 15079, 15079]) Position ids shape: torch.Size([1, 15079]) Input IDs shape: torch.Size([1, 15079]) Labels shape: torch.Size([1, 15079]) Final batch size: 1, sequence length: 15308 Attention mask shape: torch.Size([1, 1, 15308, 15308]) Position ids shape: torch.Size([1, 15308]) Input IDs shape: torch.Size([1, 15308]) Labels shape: torch.Size([1, 15308]) Final batch size: 1, sequence length: 15478 Attention mask shape: torch.Size([1, 1, 15478, 15478]) Position ids shape: torch.Size([1, 15478]) Input IDs shape: torch.Size([1, 15478]) Labels shape: torch.Size([1, 15478]) Final batch size: 1, sequence length: 11623 Attention mask shape: torch.Size([1, 1, 11623, 11623]) Position ids shape: torch.Size([1, 11623]) Input IDs shape: torch.Size([1, 11623]) Labels shape: torch.Size([1, 11623]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 11678 Attention mask shape: torch.Size([1, 1, 11678, 11678]) Position ids shape: torch.Size([1, 11678]) Input IDs shape: torch.Size([1, 11678]) Labels shape: torch.Size([1, 11678]) Final batch size: 1, sequence length: 16535 Attention mask shape: torch.Size([1, 1, 16535, 16535]) Position ids shape: torch.Size([1, 16535]) Input IDs shape: torch.Size([1, 16535]) Labels shape: torch.Size([1, 16535]) Final batch size: 1, sequence length: 14648 Attention mask shape: torch.Size([1, 1, 14648, 14648]) Position ids shape: torch.Size([1, 14648]) Input IDs shape: torch.Size([1, 14648]) Labels shape: torch.Size([1, 14648]) Final batch size: 1, sequence length: 18953 Attention mask shape: torch.Size([1, 1, 18953, 18953]) Position ids shape: torch.Size([1, 18953]) Input IDs shape: torch.Size([1, 18953]) Labels shape: torch.Size([1, 18953]) Final batch size: 1, sequence length: 13639 Attention mask shape: torch.Size([1, 1, 13639, 13639]) Position ids shape: torch.Size([1, 13639]) Input IDs shape: torch.Size([1, 13639]) Labels shape: torch.Size([1, 13639]) Final batch size: 1, sequence length: 17058 Attention mask shape: torch.Size([1, 1, 17058, 17058]) Position ids shape: torch.Size([1, 17058]) Input IDs shape: torch.Size([1, 17058]) Labels shape: torch.Size([1, 17058]) Final batch size: 1, sequence length: 18836 Final batch size: 1, sequence length: 18307 Attention mask shape: torch.Size([1, 1, 18836, 18836]) Position ids shape: torch.Size([1, 18836]) Input IDs shape: torch.Size([1, 18836]) Labels shape: torch.Size([1, 18836]) Attention mask shape: torch.Size([1, 1, 18307, 18307]) Position ids shape: torch.Size([1, 18307]) Input IDs shape: torch.Size([1, 18307]) Labels shape: torch.Size([1, 18307]) Final batch size: 1, sequence length: 19330 Attention mask shape: torch.Size([1, 1, 19330, 19330]) Position ids shape: torch.Size([1, 19330]) Input IDs shape: torch.Size([1, 19330]) Labels shape: torch.Size([1, 19330]) Final batch size: 1, sequence length: 19138 Attention mask shape: torch.Size([1, 1, 19138, 19138]) Position ids shape: torch.Size([1, 19138]) Input IDs shape: torch.Size([1, 19138]) Labels shape: torch.Size([1, 19138]) Final batch size: 1, sequence length: 8839 Attention mask shape: torch.Size([1, 1, 8839, 8839]) Position ids shape: torch.Size([1, 8839]) Input IDs shape: torch.Size([1, 8839]) Labels shape: torch.Size([1, 8839]) Final batch size: 1, sequence length: 20714 Attention mask shape: torch.Size([1, 1, 20714, 20714]) Position ids shape: torch.Size([1, 20714]) Input IDs shape: torch.Size([1, 20714]) Labels shape: torch.Size([1, 20714]) Final batch size: 1, sequence length: 19170 Attention mask shape: torch.Size([1, 1, 19170, 19170]) Position ids shape: torch.Size([1, 19170]) Input IDs shape: torch.Size([1, 19170]) Labels shape: torch.Size([1, 19170]) Final batch size: 1, sequence length: 22561 Attention mask shape: torch.Size([1, 1, 22561, 22561]) Position ids shape: torch.Size([1, 22561]) Input IDs shape: torch.Size([1, 22561]) Labels shape: torch.Size([1, 22561]) Final batch size: 1, sequence length: 12006 Attention mask shape: torch.Size([1, 1, 12006, 12006]) Position ids shape: torch.Size([1, 12006]) Input IDs shape: torch.Size([1, 12006]) Labels shape: torch.Size([1, 12006]) Final batch size: 1, sequence length: 19338 Attention mask shape: torch.Size([1, 1, 19338, 19338]) Position ids shape: torch.Size([1, 19338]) Input IDs shape: torch.Size([1, 19338]) Labels shape: torch.Size([1, 19338]) Final batch size: 1, sequence length: 18395 Attention mask shape: torch.Size([1, 1, 18395, 18395]) Position ids shape: torch.Size([1, 18395]) Input IDs shape: torch.Size([1, 18395]) Labels shape: torch.Size([1, 18395]) Final batch size: 1, sequence length: 20770 Attention mask shape: torch.Size([1, 1, 20770, 20770]) Position ids shape: torch.Size([1, 20770]) Input IDs shape: torch.Size([1, 20770]) Labels shape: torch.Size([1, 20770]) Final batch size: 1, sequence length: 21982 Attention mask shape: torch.Size([1, 1, 21982, 21982]) Position ids shape: torch.Size([1, 21982]) Input IDs shape: torch.Size([1, 21982]) Labels shape: torch.Size([1, 21982]) Final batch size: 1, sequence length: 18527 Attention mask shape: torch.Size([1, 1, 18527, 18527]) Position ids shape: torch.Size([1, 18527]) Input IDs shape: torch.Size([1, 18527]) Labels shape: torch.Size([1, 18527]) Final batch size: 1, sequence length: 18325 Attention mask shape: torch.Size([1, 1, 18325, 18325]) Position ids shape: torch.Size([1, 18325]) Input IDs shape: torch.Size([1, 18325]) Labels shape: torch.Size([1, 18325]) Final batch size: 1, sequence length: 20854 Attention mask shape: torch.Size([1, 1, 20854, 20854]) Position ids shape: torch.Size([1, 20854]) Input IDs shape: torch.Size([1, 20854]) Labels shape: torch.Size([1, 20854]) Final batch size: 1, sequence length: 23560 Attention mask shape: torch.Size([1, 1, 23560, 23560]) Position ids shape: torch.Size([1, 23560]) Input IDs shape: torch.Size([1, 23560]) Labels shape: torch.Size([1, 23560]) Final batch size: 1, sequence length: 18823 Attention mask shape: torch.Size([1, 1, 18823, 18823]) Position ids shape: torch.Size([1, 18823]) Input IDs shape: torch.Size([1, 18823]) Labels shape: torch.Size([1, 18823]) Final batch size: 1, sequence length: 21405 Attention mask shape: torch.Size([1, 1, 21405, 21405]) Position ids shape: torch.Size([1, 21405]) Input IDs shape: torch.Size([1, 21405]) Labels shape: torch.Size([1, 21405]) Final batch size: 1, sequence length: 21858 Attention mask shape: torch.Size([1, 1, 21858, 21858]) Position ids shape: torch.Size([1, 21858]) Input IDs shape: torch.Size([1, 21858]) Labels shape: torch.Size([1, 21858]) Final batch size: 1, sequence length: 25451 Attention mask shape: torch.Size([1, 1, 25451, 25451]) Position ids shape: torch.Size([1, 25451]) Input IDs shape: torch.Size([1, 25451]) Labels shape: torch.Size([1, 25451]) Final batch size: 1, sequence length: 15913 Attention mask shape: torch.Size([1, 1, 15913, 15913]) Position ids shape: torch.Size([1, 15913]) Input IDs shape: torch.Size([1, 15913]) Labels shape: torch.Size([1, 15913]) Final batch size: 1, sequence length: 24248 Attention mask shape: torch.Size([1, 1, 24248, 24248]) Position ids shape: torch.Size([1, 24248]) Input IDs shape: torch.Size([1, 24248]) Labels shape: torch.Size([1, 24248]) Final batch size: 1, sequence length: 26356 Attention mask shape: torch.Size([1, 1, 26356, 26356]) Position ids shape: torch.Size([1, 26356]) Input IDs shape: torch.Size([1, 26356]) Labels shape: torch.Size([1, 26356]) Final batch size: 1, sequence length: 23694 Attention mask shape: torch.Size([1, 1, 23694, 23694]) Position ids shape: torch.Size([1, 23694]) Input IDs shape: torch.Size([1, 23694]) Labels shape: torch.Size([1, 23694]) Final batch size: 1, sequence length: 24433 Attention mask shape: torch.Size([1, 1, 24433, 24433]) Position ids shape: torch.Size([1, 24433]) Input IDs shape: torch.Size([1, 24433]) Labels shape: torch.Size([1, 24433]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 16060 Attention mask shape: torch.Size([1, 1, 16060, 16060]) Position ids shape: torch.Size([1, 16060]) Input IDs shape: torch.Size([1, 16060]) Labels shape: torch.Size([1, 16060]) Final batch size: 1, sequence length: 22735 Attention mask shape: torch.Size([1, 1, 22735, 22735]) Position ids shape: torch.Size([1, 22735]) Input IDs shape: torch.Size([1, 22735]) Labels shape: torch.Size([1, 22735]) Final batch size: 1, sequence length: 5801 Attention mask shape: torch.Size([1, 1, 5801, 5801]) Position ids shape: torch.Size([1, 5801]) Input IDs shape: torch.Size([1, 5801]) Labels shape: torch.Size([1, 5801]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 29098 Attention mask shape: torch.Size([1, 1, 29098, 29098]) Position ids shape: torch.Size([1, 29098]) Input IDs shape: torch.Size([1, 29098]) Labels shape: torch.Size([1, 29098]) Final batch size: 1, sequence length: 9380 Attention mask shape: torch.Size([1, 1, 9380, 9380]) Position ids shape: torch.Size([1, 9380]) Input IDs shape: torch.Size([1, 9380]) Labels shape: torch.Size([1, 9380]) Final batch size: 1, sequence length: 28684 Attention mask shape: torch.Size([1, 1, 28684, 28684]) Position ids shape: torch.Size([1, 28684]) Input IDs shape: torch.Size([1, 28684]) Labels shape: torch.Size([1, 28684]) Final batch size: 1, sequence length: 21615 Attention mask shape: torch.Size([1, 1, 21615, 21615]) Position ids shape: torch.Size([1, 21615]) Input IDs shape: torch.Size([1, 21615]) Labels shape: torch.Size([1, 21615]) Final batch size: 1, sequence length: 19045 Attention mask shape: torch.Size([1, 1, 19045, 19045]) Position ids shape: torch.Size([1, 19045]) Input IDs shape: torch.Size([1, 19045]) Labels shape: torch.Size([1, 19045]) Final batch size: 1, sequence length: 31414 Attention mask shape: torch.Size([1, 1, 31414, 31414]) Position ids shape: torch.Size([1, 31414]) Input IDs shape: torch.Size([1, 31414]) Labels shape: torch.Size([1, 31414]) Final batch size: 1, sequence length: 26072 Attention mask shape: torch.Size([1, 1, 26072, 26072]) Position ids shape: torch.Size([1, 26072]) Input IDs shape: torch.Size([1, 26072]) Labels shape: torch.Size([1, 26072]) Final batch size: 1, sequence length: 17465 Attention mask shape: torch.Size([1, 1, 17465, 17465]) Position ids shape: torch.Size([1, 17465]) Input IDs shape: torch.Size([1, 17465]) Labels shape: torch.Size([1, 17465]) Final batch size: 1, sequence length: 19239 Attention mask shape: torch.Size([1, 1, 19239, 19239]) Position ids shape: torch.Size([1, 19239]) Input IDs shape: torch.Size([1, 19239]) Labels shape: torch.Size([1, 19239]) Final batch size: 1, sequence length: 32885 Attention mask shape: torch.Size([1, 1, 32885, 32885]) Position ids shape: torch.Size([1, 32885]) Input IDs shape: torch.Size([1, 32885]) Labels shape: torch.Size([1, 32885]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 32328 Attention mask shape: torch.Size([1, 1, 32328, 32328]) Position ids shape: torch.Size([1, 32328]) Input IDs shape: torch.Size([1, 32328]) Labels shape: torch.Size([1, 32328]) Final batch size: 1, sequence length: 23881 Attention mask shape: torch.Size([1, 1, 23881, 23881]) Position ids shape: torch.Size([1, 23881]) Input IDs shape: torch.Size([1, 23881]) Labels shape: torch.Size([1, 23881]) Final batch size: 1, sequence length: 15875 Attention mask shape: torch.Size([1, 1, 15875, 15875]) Position ids shape: torch.Size([1, 15875]) Input IDs shape: torch.Size([1, 15875]) Labels shape: torch.Size([1, 15875]) Final batch size: 1, sequence length: 34921 Attention mask shape: torch.Size([1, 1, 34921, 34921]) Position ids shape: torch.Size([1, 34921]) Input IDs shape: torch.Size([1, 34921]) Labels shape: torch.Size([1, 34921]) Final batch size: 1, sequence length: 33871 Attention mask shape: torch.Size([1, 1, 33871, 33871]) Position ids shape: torch.Size([1, 33871]) Input IDs shape: torch.Size([1, 33871]) Labels shape: torch.Size([1, 33871]) Final batch size: 1, sequence length: 30687 Attention mask shape: torch.Size([1, 1, 30687, 30687]) Position ids shape: torch.Size([1, 30687]) Input IDs shape: torch.Size([1, 30687]) Labels shape: torch.Size([1, 30687]) Final batch size: 1, sequence length: 29355 Attention mask shape: torch.Size([1, 1, 29355, 29355]) Position ids shape: torch.Size([1, 29355]) Input IDs shape: torch.Size([1, 29355]) Labels shape: torch.Size([1, 29355]) Final batch size: 1, sequence length: 27972 Attention mask shape: torch.Size([1, 1, 27972, 27972]) Position ids shape: torch.Size([1, 27972]) Input IDs shape: torch.Size([1, 27972]) Labels shape: torch.Size([1, 27972]) Final batch size: 1, sequence length: 16050 Attention mask shape: torch.Size([1, 1, 16050, 16050]) Position ids shape: torch.Size([1, 16050]) Input IDs shape: torch.Size([1, 16050]) Labels shape: torch.Size([1, 16050]) Final batch size: 1, sequence length: 32529 Attention mask shape: torch.Size([1, 1, 32529, 32529]) Position ids shape: torch.Size([1, 32529]) Input IDs shape: torch.Size([1, 32529]) Labels shape: torch.Size([1, 32529]) Final batch size: 1, sequence length: 35397 Attention mask shape: torch.Size([1, 1, 35397, 35397]) Position ids shape: torch.Size([1, 35397]) Input IDs shape: torch.Size([1, 35397]) Labels shape: torch.Size([1, 35397]) Final batch size: 1, sequence length: 30689 Attention mask shape: torch.Size([1, 1, 30689, 30689]) Position ids shape: torch.Size([1, 30689]) Input IDs shape: torch.Size([1, 30689]) Labels shape: torch.Size([1, 30689]) Final batch size: 1, sequence length: 32636 Attention mask shape: torch.Size([1, 1, 32636, 32636]) Position ids shape: torch.Size([1, 32636]) Input IDs shape: torch.Size([1, 32636]) Labels shape: torch.Size([1, 32636]) Final batch size: 1, sequence length: 31860 Attention mask shape: torch.Size([1, 1, 31860, 31860]) Position ids shape: torch.Size([1, 31860]) Input IDs shape: torch.Size([1, 31860]) Labels shape: torch.Size([1, 31860]) Final batch size: 1, sequence length: 29150 Attention mask shape: torch.Size([1, 1, 29150, 29150]) Position ids shape: torch.Size([1, 29150]) Input IDs shape: torch.Size([1, 29150]) Labels shape: torch.Size([1, 29150]) Final batch size: 1, sequence length: 31712 Attention mask shape: torch.Size([1, 1, 31712, 31712]) Position ids shape: torch.Size([1, 31712]) Input IDs shape: torch.Size([1, 31712]) Labels shape: torch.Size([1, 31712]) Final batch size: 1, sequence length: 30346 Attention mask shape: torch.Size([1, 1, 30346, 30346]) Position ids shape: torch.Size([1, 30346]) Input IDs shape: torch.Size([1, 30346]) Labels shape: torch.Size([1, 30346]) Final batch size: 1, sequence length: 26689 Attention mask shape: torch.Size([1, 1, 26689, 26689]) Position ids shape: torch.Size([1, 26689]) Input IDs shape: torch.Size([1, 26689]) Labels shape: torch.Size([1, 26689]) Final batch size: 1, sequence length: 34581 Attention mask shape: torch.Size([1, 1, 34581, 34581]) Position ids shape: torch.Size([1, 34581]) Input IDs shape: torch.Size([1, 34581]) Labels shape: torch.Size([1, 34581]) Final batch size: 1, sequence length: 24781 Attention mask shape: torch.Size([1, 1, 24781, 24781]) Position ids shape: torch.Size([1, 24781]) Input IDs shape: torch.Size([1, 24781]) Labels shape: torch.Size([1, 24781]) Final batch size: 1, sequence length: 15077 Attention mask shape: torch.Size([1, 1, 15077, 15077]) Position ids shape: torch.Size([1, 15077]) Input IDs shape: torch.Size([1, 15077]) Labels shape: torch.Size([1, 15077]) Final batch size: 1, sequence length: 22264 Attention mask shape: torch.Size([1, 1, 22264, 22264]) Position ids shape: torch.Size([1, 22264]) Input IDs shape: torch.Size([1, 22264]) Labels shape: torch.Size([1, 22264]) Final batch size: 1, sequence length: 18963 Attention mask shape: torch.Size([1, 1, 18963, 18963]) Position ids shape: torch.Size([1, 18963]) Input IDs shape: torch.Size([1, 18963]) Labels shape: torch.Size([1, 18963]) Final batch size: 1, sequence length: 17243 Attention mask shape: torch.Size([1, 1, 17243, 17243]) Position ids shape: torch.Size([1, 17243]) Input IDs shape: torch.Size([1, 17243]) Labels shape: torch.Size([1, 17243]) Final batch size: 1, sequence length: 16025 Attention mask shape: torch.Size([1, 1, 16025, 16025]) Position ids shape: torch.Size([1, 16025]) Input IDs shape: torch.Size([1, 16025]) Labels shape: torch.Size([1, 16025]) Final batch size: 1, sequence length: 18177 Attention mask shape: torch.Size([1, 1, 18177, 18177]) Position ids shape: torch.Size([1, 18177]) Input IDs shape: torch.Size([1, 18177]) Labels shape: torch.Size([1, 18177]) Final batch size: 1, sequence length: 25741 Attention mask shape: torch.Size([1, 1, 25741, 25741]) Position ids shape: torch.Size([1, 25741]) Input IDs shape: torch.Size([1, 25741]) Labels shape: torch.Size([1, 25741]) Final batch size: 1, sequence length: 14976 Attention mask shape: torch.Size([1, 1, 14976, 14976]) Position ids shape: torch.Size([1, 14976]) Input IDs shape: torch.Size([1, 14976]) Labels shape: torch.Size([1, 14976]) Final batch size: 1, sequence length: 34853 Attention mask shape: torch.Size([1, 1, 34853, 34853]) Position ids shape: torch.Size([1, 34853]) Input IDs shape: torch.Size([1, 34853]) Labels shape: torch.Size([1, 34853]) Final batch size: 1, sequence length: 31022 Attention mask shape: torch.Size([1, 1, 31022, 31022]) Position ids shape: torch.Size([1, 31022]) Input IDs shape: torch.Size([1, 31022]) Labels shape: torch.Size([1, 31022]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28018 Attention mask shape: torch.Size([1, 1, 28018, 28018]) Position ids shape: torch.Size([1, 28018]) Input IDs shape: torch.Size([1, 28018]) Labels shape: torch.Size([1, 28018]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40488 Attention mask shape: torch.Size([1, 1, 40488, 40488]) Position ids shape: torch.Size([1, 40488]) Input IDs shape: torch.Size([1, 40488]) Labels shape: torch.Size([1, 40488]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26922 Attention mask shape: torch.Size([1, 1, 26922, 26922]) Position ids shape: torch.Size([1, 26922]) Input IDs shape: torch.Size([1, 26922]) Labels shape: torch.Size([1, 26922]) Final batch size: 1, sequence length: 24002 Attention mask shape: torch.Size([1, 1, 24002, 24002]) Position ids shape: torch.Size([1, 24002]) Input IDs shape: torch.Size([1, 24002]) Labels shape: torch.Size([1, 24002]) Final batch size: 1, sequence length: 12218 Attention mask shape: torch.Size([1, 1, 12218, 12218]) Position ids shape: torch.Size([1, 12218]) Input IDs shape: torch.Size([1, 12218]) Labels shape: torch.Size([1, 12218]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25674 Attention mask shape: torch.Size([1, 1, 25674, 25674]) Position ids shape: torch.Size([1, 25674]) Input IDs shape: torch.Size([1, 25674]) Labels shape: torch.Size([1, 25674]) Final batch size: 1, sequence length: 36596 Attention mask shape: torch.Size([1, 1, 36596, 36596]) Position ids shape: torch.Size([1, 36596]) Input IDs shape: torch.Size([1, 36596]) Labels shape: torch.Size([1, 36596]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28397 Attention mask shape: torch.Size([1, 1, 28397, 28397]) Position ids shape: torch.Size([1, 28397]) Input IDs shape: torch.Size([1, 28397]) Labels shape: torch.Size([1, 28397]) Final batch size: 1, sequence length: 11611 Attention mask shape: torch.Size([1, 1, 11611, 11611]) Position ids shape: torch.Size([1, 11611]) Input IDs shape: torch.Size([1, 11611]) Labels shape: torch.Size([1, 11611]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24043 Attention mask shape: torch.Size([1, 1, 24043, 24043]) Position ids shape: torch.Size([1, 24043]) Input IDs shape: torch.Size([1, 24043]) Labels shape: torch.Size([1, 24043]) Final batch size: 1, sequence length: 16337 Attention mask shape: torch.Size([1, 1, 16337, 16337]) Position ids shape: torch.Size([1, 16337]) Input IDs shape: torch.Size([1, 16337]) Labels shape: torch.Size([1, 16337]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28086 Attention mask shape: torch.Size([1, 1, 28086, 28086]) Position ids shape: torch.Size([1, 28086]) Input IDs shape: torch.Size([1, 28086]) Labels shape: torch.Size([1, 28086]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36334 Attention mask shape: torch.Size([1, 1, 36334, 36334]) Position ids shape: torch.Size([1, 36334]) Input IDs shape: torch.Size([1, 36334]) Labels shape: torch.Size([1, 36334]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32917 Attention mask shape: torch.Size([1, 1, 32917, 32917]) Position ids shape: torch.Size([1, 32917]) Input IDs shape: torch.Size([1, 32917]) Labels shape: torch.Size([1, 32917]) {'loss': 0.272, 'grad_norm': 0.33806122411888334, 'learning_rate': 5e-06, 'num_tokens': -inf, 'epoch': 4.38} Final batch size: 1, sequence length: 5818 Attention mask shape: torch.Size([1, 1, 5818, 5818]) Position ids shape: torch.Size([1, 5818]) Input IDs shape: torch.Size([1, 5818]) Labels shape: torch.Size([1, 5818]) Final batch size: 1, sequence length: 6215 Attention mask shape: torch.Size([1, 1, 6215, 6215]) Position ids shape: torch.Size([1, 6215]) Input IDs shape: torch.Size([1, 6215]) Labels shape: torch.Size([1, 6215]) Final batch size: 1, sequence length: 6871 Attention mask shape: torch.Size([1, 1, 6871, 6871]) Position ids shape: torch.Size([1, 6871]) Input IDs shape: torch.Size([1, 6871]) Labels shape: torch.Size([1, 6871]) Final batch size: 1, sequence length: 5917 Attention mask shape: torch.Size([1, 1, 5917, 5917]) Position ids shape: torch.Size([1, 5917]) Input IDs shape: torch.Size([1, 5917]) Labels shape: torch.Size([1, 5917]) Final batch size: 1, sequence length: 6034 Attention mask shape: torch.Size([1, 1, 6034, 6034]) Position ids shape: torch.Size([1, 6034]) Input IDs shape: torch.Size([1, 6034]) Labels shape: torch.Size([1, 6034]) Final batch size: 1, sequence length: 8623 Attention mask shape: torch.Size([1, 1, 8623, 8623]) Position ids shape: torch.Size([1, 8623]) Input IDs shape: torch.Size([1, 8623]) Labels shape: torch.Size([1, 8623]) Final batch size: 1, sequence length: 11616 Attention mask shape: torch.Size([1, 1, 11616, 11616]) Position ids shape: torch.Size([1, 11616]) Input IDs shape: torch.Size([1, 11616]) Labels shape: torch.Size([1, 11616]) Final batch size: 1, sequence length: 12275 Attention mask shape: torch.Size([1, 1, 12275, 12275]) Position ids shape: torch.Size([1, 12275]) Input IDs shape: torch.Size([1, 12275]) Labels shape: torch.Size([1, 12275]) Final batch size: 1, sequence length: 12317 Attention mask shape: torch.Size([1, 1, 12317, 12317]) Position ids shape: torch.Size([1, 12317]) Input IDs shape: torch.Size([1, 12317]) Labels shape: torch.Size([1, 12317]) Final batch size: 1, sequence length: 10107 Attention mask shape: torch.Size([1, 1, 10107, 10107]) Position ids shape: torch.Size([1, 10107]) Input IDs shape: torch.Size([1, 10107]) Labels shape: torch.Size([1, 10107]) Final batch size: 1, sequence length: 13202 Attention mask shape: torch.Size([1, 1, 13202, 13202]) Position ids shape: torch.Size([1, 13202]) Input IDs shape: torch.Size([1, 13202]) Labels shape: torch.Size([1, 13202]) Final batch size: 1, sequence length: 9517 Attention mask shape: torch.Size([1, 1, 9517, 9517]) Position ids shape: torch.Size([1, 9517]) Input IDs shape: torch.Size([1, 9517]) Labels shape: torch.Size([1, 9517]) Final batch size: 1, sequence length: 15029 Attention mask shape: torch.Size([1, 1, 15029, 15029]) Position ids shape: torch.Size([1, 15029]) Input IDs shape: torch.Size([1, 15029]) Labels shape: torch.Size([1, 15029]) Final batch size: 1, sequence length: 12813 Attention mask shape: torch.Size([1, 1, 12813, 12813]) Position ids shape: torch.Size([1, 12813]) Input IDs shape: torch.Size([1, 12813]) Labels shape: torch.Size([1, 12813]) Final batch size: 1, sequence length: 8117 Attention mask shape: torch.Size([1, 1, 8117, 8117]) Position ids shape: torch.Size([1, 8117]) Input IDs shape: torch.Size([1, 8117]) Labels shape: torch.Size([1, 8117]) Final batch size: 1, sequence length: 10136 Attention mask shape: torch.Size([1, 1, 10136, 10136]) Position ids shape: torch.Size([1, 10136]) Input IDs shape: torch.Size([1, 10136]) Labels shape: torch.Size([1, 10136]) Final batch size: 1, sequence length: 13428 Attention mask shape: torch.Size([1, 1, 13428, 13428]) Position ids shape: torch.Size([1, 13428]) Input IDs shape: torch.Size([1, 13428]) Labels shape: torch.Size([1, 13428]) Final batch size: 1, sequence length: 16331 Attention mask shape: torch.Size([1, 1, 16331, 16331]) Position ids shape: torch.Size([1, 16331]) Input IDs shape: torch.Size([1, 16331]) Labels shape: torch.Size([1, 16331]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 11648 Attention mask shape: torch.Size([1, 1, 11648, 11648]) Position ids shape: torch.Size([1, 11648]) Input IDs shape: torch.Size([1, 11648]) Labels shape: torch.Size([1, 11648]) Final batch size: 1, sequence length: 12553 Attention mask shape: torch.Size([1, 1, 12553, 12553]) Position ids shape: torch.Size([1, 12553]) Input IDs shape: torch.Size([1, 12553]) Labels shape: torch.Size([1, 12553]) Final batch size: 1, sequence length: 15499 Attention mask shape: torch.Size([1, 1, 15499, 15499]) Position ids shape: torch.Size([1, 15499]) Input IDs shape: torch.Size([1, 15499]) Labels shape: torch.Size([1, 15499]) Final batch size: 1, sequence length: 15066 Attention mask shape: torch.Size([1, 1, 15066, 15066]) Position ids shape: torch.Size([1, 15066]) Input IDs shape: torch.Size([1, 15066]) Labels shape: torch.Size([1, 15066]) Final batch size: 1, sequence length: 17587 Attention mask shape: torch.Size([1, 1, 17587, 17587]) Position ids shape: torch.Size([1, 17587]) Input IDs shape: torch.Size([1, 17587]) Labels shape: torch.Size([1, 17587]) Final batch size: 1, sequence length: 19847 Attention mask shape: torch.Size([1, 1, 19847, 19847]) Position ids shape: torch.Size([1, 19847]) Input IDs shape: torch.Size([1, 19847]) Labels shape: torch.Size([1, 19847]) Final batch size: 1, sequence length: 19512 Attention mask shape: torch.Size([1, 1, 19512, 19512]) Position ids shape: torch.Size([1, 19512]) Input IDs shape: torch.Size([1, 19512]) Labels shape: torch.Size([1, 19512]) Final batch size: 1, sequence length: 19221 Attention mask shape: torch.Size([1, 1, 19221, 19221]) Position ids shape: torch.Size([1, 19221]) Input IDs shape: torch.Size([1, 19221]) Labels shape: torch.Size([1, 19221]) Final batch size: 1, sequence length: 21028 Attention mask shape: torch.Size([1, 1, 21028, 21028]) Position ids shape: torch.Size([1, 21028]) Input IDs shape: torch.Size([1, 21028]) Labels shape: torch.Size([1, 21028]) Final batch size: 1, sequence length: 21556 Attention mask shape: torch.Size([1, 1, 21556, 21556]) Position ids shape: torch.Size([1, 21556]) Input IDs shape: torch.Size([1, 21556]) Labels shape: torch.Size([1, 21556]) Final batch size: 1, sequence length: 18438 Attention mask shape: torch.Size([1, 1, 18438, 18438]) Position ids shape: torch.Size([1, 18438]) Input IDs shape: torch.Size([1, 18438]) Labels shape: torch.Size([1, 18438]) Final batch size: 1, sequence length: 22079 Attention mask shape: torch.Size([1, 1, 22079, 22079]) Position ids shape: torch.Size([1, 22079]) Input IDs shape: torch.Size([1, 22079]) Labels shape: torch.Size([1, 22079]) Final batch size: 1, sequence length: 18414 Attention mask shape: torch.Size([1, 1, 18414, 18414]) Position ids shape: torch.Size([1, 18414]) Input IDs shape: torch.Size([1, 18414]) Labels shape: torch.Size([1, 18414]) Final batch size: 1, sequence length: 18469 Attention mask shape: torch.Size([1, 1, 18469, 18469]) Position ids shape: torch.Size([1, 18469]) Input IDs shape: torch.Size([1, 18469]) Labels shape: torch.Size([1, 18469]) Final batch size: 1, sequence length: 22138 Attention mask shape: torch.Size([1, 1, 22138, 22138]) Position ids shape: torch.Size([1, 22138]) Input IDs shape: torch.Size([1, 22138]) Labels shape: torch.Size([1, 22138]) Final batch size: 1, sequence length: 15221 Attention mask shape: torch.Size([1, 1, 15221, 15221]) Position ids shape: torch.Size([1, 15221]) Input IDs shape: torch.Size([1, 15221]) Labels shape: torch.Size([1, 15221]) Final batch size: 1, sequence length: 15226 Attention mask shape: torch.Size([1, 1, 15226, 15226]) Position ids shape: torch.Size([1, 15226]) Input IDs shape: torch.Size([1, 15226]) Labels shape: torch.Size([1, 15226]) Final batch size: 1, sequence length: 22718 Attention mask shape: torch.Size([1, 1, 22718, 22718]) Position ids shape: torch.Size([1, 22718]) Input IDs shape: torch.Size([1, 22718]) Labels shape: torch.Size([1, 22718]) Final batch size: 1, sequence length: 24499 Attention mask shape: torch.Size([1, 1, 24499, 24499]) Position ids shape: torch.Size([1, 24499]) Input IDs shape: torch.Size([1, 24499]) Labels shape: torch.Size([1, 24499]) Final batch size: 1, sequence length: 23973 Attention mask shape: torch.Size([1, 1, 23973, 23973]) Position ids shape: torch.Size([1, 23973]) Input IDs shape: torch.Size([1, 23973]) Labels shape: torch.Size([1, 23973]) Final batch size: 1, sequence length: 22618 Attention mask shape: torch.Size([1, 1, 22618, 22618]) Position ids shape: torch.Size([1, 22618]) Input IDs shape: torch.Size([1, 22618]) Labels shape: torch.Size([1, 22618]) Final batch size: 1, sequence length: 25622 Attention mask shape: torch.Size([1, 1, 25622, 25622]) Position ids shape: torch.Size([1, 25622]) Input IDs shape: torch.Size([1, 25622]) Labels shape: torch.Size([1, 25622]) Final batch size: 1, sequence length: 23766 Attention mask shape: torch.Size([1, 1, 23766, 23766]) Position ids shape: torch.Size([1, 23766]) Input IDs shape: torch.Size([1, 23766]) Labels shape: torch.Size([1, 23766]) Final batch size: 1, sequence length: 19428 Attention mask shape: torch.Size([1, 1, 19428, 19428]) Position ids shape: torch.Size([1, 19428]) Input IDs shape: torch.Size([1, 19428]) Labels shape: torch.Size([1, 19428]) Final batch size: 1, sequence length: 25999 Attention mask shape: torch.Size([1, 1, 25999, 25999]) Position ids shape: torch.Size([1, 25999]) Input IDs shape: torch.Size([1, 25999]) Labels shape: torch.Size([1, 25999]) Final batch size: 1, sequence length: 16299 Attention mask shape: torch.Size([1, 1, 16299, 16299]) Position ids shape: torch.Size([1, 16299]) Input IDs shape: torch.Size([1, 16299]) Labels shape: torch.Size([1, 16299]) Final batch size: 1, sequence length: 24694 Attention mask shape: torch.Size([1, 1, 24694, 24694]) Position ids shape: torch.Size([1, 24694]) Input IDs shape: torch.Size([1, 24694]) Labels shape: torch.Size([1, 24694]) Final batch size: 1, sequence length: 22235 Attention mask shape: torch.Size([1, 1, 22235, 22235]) Position ids shape: torch.Size([1, 22235]) Input IDs shape: torch.Size([1, 22235]) Labels shape: torch.Size([1, 22235]) Final batch size: 1, sequence length: 22763 Attention mask shape: torch.Size([1, 1, 22763, 22763]) Position ids shape: torch.Size([1, 22763]) Input IDs shape: torch.Size([1, 22763]) Labels shape: torch.Size([1, 22763]) Final batch size: 1, sequence length: 14784 Attention mask shape: torch.Size([1, 1, 14784, 14784]) Position ids shape: torch.Size([1, 14784]) Input IDs shape: torch.Size([1, 14784]) Labels shape: torch.Size([1, 14784]) Final batch size: 1, sequence length: 24909 Attention mask shape: torch.Size([1, 1, 24909, 24909]) Position ids shape: torch.Size([1, 24909]) Input IDs shape: torch.Size([1, 24909]) Labels shape: torch.Size([1, 24909]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 7584 Attention mask shape: torch.Size([1, 1, 7584, 7584]) Position ids shape: torch.Size([1, 7584]) Input IDs shape: torch.Size([1, 7584]) Labels shape: torch.Size([1, 7584]) Final batch size: 1, sequence length: 27327 Attention mask shape: torch.Size([1, 1, 27327, 27327]) Position ids shape: torch.Size([1, 27327]) Input IDs shape: torch.Size([1, 27327]) Labels shape: torch.Size([1, 27327]) Final batch size: 1, sequence length: 27566 Attention mask shape: torch.Size([1, 1, 27566, 27566]) Position ids shape: torch.Size([1, 27566]) Input IDs shape: torch.Size([1, 27566]) Labels shape: torch.Size([1, 27566]) Final batch size: 1, sequence length: 28497 Attention mask shape: torch.Size([1, 1, 28497, 28497]) Position ids shape: torch.Size([1, 28497]) Input IDs shape: torch.Size([1, 28497]) Labels shape: torch.Size([1, 28497]) Final batch size: 1, sequence length: 28749 Attention mask shape: torch.Size([1, 1, 28749, 28749]) Position ids shape: torch.Size([1, 28749]) Input IDs shape: torch.Size([1, 28749]) Labels shape: torch.Size([1, 28749]) Final batch size: 1, sequence length: 21826 Attention mask shape: torch.Size([1, 1, 21826, 21826]) Position ids shape: torch.Size([1, 21826]) Input IDs shape: torch.Size([1, 21826]) Labels shape: torch.Size([1, 21826]) Final batch size: 1, sequence length: 15184 Attention mask shape: torch.Size([1, 1, 15184, 15184]) Position ids shape: torch.Size([1, 15184]) Input IDs shape: torch.Size([1, 15184]) Labels shape: torch.Size([1, 15184]) Final batch size: 1, sequence length: 15588 Attention mask shape: torch.Size([1, 1, 15588, 15588]) Position ids shape: torch.Size([1, 15588]) Input IDs shape: torch.Size([1, 15588]) Labels shape: torch.Size([1, 15588]) Final batch size: 1, sequence length: 10269 Attention mask shape: torch.Size([1, 1, 10269, 10269]) Position ids shape: torch.Size([1, 10269]) Input IDs shape: torch.Size([1, 10269]) Labels shape: torch.Size([1, 10269]) Final batch size: 1, sequence length: 11795 Attention mask shape: torch.Size([1, 1, 11795, 11795]) Position ids shape: torch.Size([1, 11795]) Input IDs shape: torch.Size([1, 11795]) Labels shape: torch.Size([1, 11795]) Final batch size: 1, sequence length: 30356 Attention mask shape: torch.Size([1, 1, 30356, 30356]) Position ids shape: torch.Size([1, 30356]) Input IDs shape: torch.Size([1, 30356]) Labels shape: torch.Size([1, 30356]) Final batch size: 1, sequence length: 22366 Attention mask shape: torch.Size([1, 1, 22366, 22366]) Position ids shape: torch.Size([1, 22366]) Input IDs shape: torch.Size([1, 22366]) Labels shape: torch.Size([1, 22366]) Final batch size: 1, sequence length: 31714 Attention mask shape: torch.Size([1, 1, 31714, 31714]) Position ids shape: torch.Size([1, 31714]) Input IDs shape: torch.Size([1, 31714]) Labels shape: torch.Size([1, 31714]) Final batch size: 1, sequence length: 23851 Attention mask shape: torch.Size([1, 1, 23851, 23851]) Position ids shape: torch.Size([1, 23851]) Input IDs shape: torch.Size([1, 23851]) Labels shape: torch.Size([1, 23851]) Final batch size: 1, sequence length: 36777 Attention mask shape: torch.Size([1, 1, 36777, 36777]) Position ids shape: torch.Size([1, 36777]) Input IDs shape: torch.Size([1, 36777]) Labels shape: torch.Size([1, 36777]) Final batch size: 1, sequence length: 20755 Attention mask shape: torch.Size([1, 1, 20755, 20755]) Position ids shape: torch.Size([1, 20755]) Input IDs shape: torch.Size([1, 20755]) Labels shape: torch.Size([1, 20755]) Final batch size: 1, sequence length: 21034 Attention mask shape: torch.Size([1, 1, 21034, 21034]) Position ids shape: torch.Size([1, 21034]) Input IDs shape: torch.Size([1, 21034]) Labels shape: torch.Size([1, 21034]) Final batch size: 1, sequence length: 27489 Attention mask shape: torch.Size([1, 1, 27489, 27489]) Position ids shape: torch.Size([1, 27489]) Input IDs shape: torch.Size([1, 27489]) Labels shape: torch.Size([1, 27489]) Final batch size: 1, sequence length: 34673 Attention mask shape: torch.Size([1, 1, 34673, 34673]) Position ids shape: torch.Size([1, 34673]) Input IDs shape: torch.Size([1, 34673]) Labels shape: torch.Size([1, 34673]) Final batch size: 1, sequence length: 31903 Attention mask shape: torch.Size([1, 1, 31903, 31903]) Position ids shape: torch.Size([1, 31903]) Input IDs shape: torch.Size([1, 31903]) Labels shape: torch.Size([1, 31903]) Final batch size: 1, sequence length: 38049 Attention mask shape: torch.Size([1, 1, 38049, 38049]) Position ids shape: torch.Size([1, 38049]) Input IDs shape: torch.Size([1, 38049]) Labels shape: torch.Size([1, 38049]) Final batch size: 1, sequence length: 36723 Attention mask shape: torch.Size([1, 1, 36723, 36723]) Position ids shape: torch.Size([1, 36723]) Input IDs shape: torch.Size([1, 36723]) Labels shape: torch.Size([1, 36723]) Final batch size: 1, sequence length: 38848 Attention mask shape: torch.Size([1, 1, 38848, 38848]) Position ids shape: torch.Size([1, 38848]) Input IDs shape: torch.Size([1, 38848]) Labels shape: torch.Size([1, 38848]) Final batch size: 1, sequence length: 28176 Attention mask shape: torch.Size([1, 1, 28176, 28176]) Position ids shape: torch.Size([1, 28176]) Input IDs shape: torch.Size([1, 28176]) Labels shape: torch.Size([1, 28176]) Final batch size: 1, sequence length: 40325 Attention mask shape: torch.Size([1, 1, 40325, 40325]) Position ids shape: torch.Size([1, 40325]) Input IDs shape: torch.Size([1, 40325]) Labels shape: torch.Size([1, 40325]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 37285 Attention mask shape: torch.Size([1, 1, 37285, 37285]) Position ids shape: torch.Size([1, 37285]) Input IDs shape: torch.Size([1, 37285]) Labels shape: torch.Size([1, 37285]) Final batch size: 1, sequence length: 39253 Attention mask shape: torch.Size([1, 1, 39253, 39253]) Position ids shape: torch.Size([1, 39253]) Input IDs shape: torch.Size([1, 39253]) Labels shape: torch.Size([1, 39253]) Final batch size: 1, sequence length: 24801 Attention mask shape: torch.Size([1, 1, 24801, 24801]) Position ids shape: torch.Size([1, 24801]) Input IDs shape: torch.Size([1, 24801]) Labels shape: torch.Size([1, 24801]) Final batch size: 1, sequence length: 25946 Attention mask shape: torch.Size([1, 1, 25946, 25946]) Position ids shape: torch.Size([1, 25946]) Input IDs shape: torch.Size([1, 25946]) Labels shape: torch.Size([1, 25946]) Final batch size: 1, sequence length: 6882 Attention mask shape: torch.Size([1, 1, 6882, 6882]) Position ids shape: torch.Size([1, 6882]) Input IDs shape: torch.Size([1, 6882]) Labels shape: torch.Size([1, 6882]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38104 Attention mask shape: torch.Size([1, 1, 38104, 38104]) Position ids shape: torch.Size([1, 38104]) Input IDs shape: torch.Size([1, 38104]) Labels shape: torch.Size([1, 38104]) Final batch size: 1, sequence length: 34289 Attention mask shape: torch.Size([1, 1, 34289, 34289]) Position ids shape: torch.Size([1, 34289]) Input IDs shape: torch.Size([1, 34289]) Labels shape: torch.Size([1, 34289]) Final batch size: 1, sequence length: 40317 Attention mask shape: torch.Size([1, 1, 40317, 40317]) Position ids shape: torch.Size([1, 40317]) Input IDs shape: torch.Size([1, 40317]) Labels shape: torch.Size([1, 40317]) Final batch size: 1, sequence length: 40937 Attention mask shape: torch.Size([1, 1, 40937, 40937]) Position ids shape: torch.Size([1, 40937]) Input IDs shape: torch.Size([1, 40937]) Labels shape: torch.Size([1, 40937]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 10469 Attention mask shape: torch.Size([1, 1, 10469, 10469]) Position ids shape: torch.Size([1, 10469]) Input IDs shape: torch.Size([1, 10469]) Labels shape: torch.Size([1, 10469]) Final batch size: 1, sequence length: 35904 Attention mask shape: torch.Size([1, 1, 35904, 35904]) Position ids shape: torch.Size([1, 35904]) Input IDs shape: torch.Size([1, 35904]) Labels shape: torch.Size([1, 35904]) Final batch size: 1, sequence length: 36786 Attention mask shape: torch.Size([1, 1, 36786, 36786]) Position ids shape: torch.Size([1, 36786]) Input IDs shape: torch.Size([1, 36786]) Labels shape: torch.Size([1, 36786]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16649 Attention mask shape: torch.Size([1, 1, 16649, 16649]) Position ids shape: torch.Size([1, 16649]) Input IDs shape: torch.Size([1, 16649]) Labels shape: torch.Size([1, 16649]) Final batch size: 1, sequence length: 19437 Attention mask shape: torch.Size([1, 1, 19437, 19437]) Position ids shape: torch.Size([1, 19437]) Input IDs shape: torch.Size([1, 19437]) Labels shape: torch.Size([1, 19437]) Final batch size: 1, sequence length: 19639 Attention mask shape: torch.Size([1, 1, 19639, 19639]) Position ids shape: torch.Size([1, 19639]) Input IDs shape: torch.Size([1, 19639]) Labels shape: torch.Size([1, 19639]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40657 Attention mask shape: torch.Size([1, 1, 40657, 40657]) Position ids shape: torch.Size([1, 40657]) Input IDs shape: torch.Size([1, 40657]) Labels shape: torch.Size([1, 40657]) Final batch size: 1, sequence length: 25963 Attention mask shape: torch.Size([1, 1, 25963, 25963]) Position ids shape: torch.Size([1, 25963]) Input IDs shape: torch.Size([1, 25963]) Labels shape: torch.Size([1, 25963]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26660 Attention mask shape: torch.Size([1, 1, 26660, 26660]) Position ids shape: torch.Size([1, 26660]) Input IDs shape: torch.Size([1, 26660]) Labels shape: torch.Size([1, 26660]) Final batch size: 1, sequence length: 38686 Attention mask shape: torch.Size([1, 1, 38686, 38686]) Position ids shape: torch.Size([1, 38686]) Input IDs shape: torch.Size([1, 38686]) Labels shape: torch.Size([1, 38686]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36128 Attention mask shape: torch.Size([1, 1, 36128, 36128]) Position ids shape: torch.Size([1, 36128]) Input IDs shape: torch.Size([1, 36128]) Labels shape: torch.Size([1, 36128]) Final batch size: 1, sequence length: 21547 Attention mask shape: torch.Size([1, 1, 21547, 21547]) Position ids shape: torch.Size([1, 21547]) Input IDs shape: torch.Size([1, 21547]) Labels shape: torch.Size([1, 21547]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34540 Attention mask shape: torch.Size([1, 1, 34540, 34540]) Position ids shape: torch.Size([1, 34540]) Input IDs shape: torch.Size([1, 34540]) Labels shape: torch.Size([1, 34540]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17882 Attention mask shape: torch.Size([1, 1, 17882, 17882]) Position ids shape: torch.Size([1, 17882]) Input IDs shape: torch.Size([1, 17882]) Labels shape: torch.Size([1, 17882]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 27947 Attention mask shape: torch.Size([1, 1, 27947, 27947]) Position ids shape: torch.Size([1, 27947]) Input IDs shape: torch.Size([1, 27947]) Labels shape: torch.Size([1, 27947]) Final batch size: 1, sequence length: 17379 Attention mask shape: torch.Size([1, 1, 17379, 17379]) Position ids shape: torch.Size([1, 17379]) Input IDs shape: torch.Size([1, 17379]) Labels shape: torch.Size([1, 17379]) Final batch size: 1, sequence length: 36317 Attention mask shape: torch.Size([1, 1, 36317, 36317]) Position ids shape: torch.Size([1, 36317]) Input IDs shape: torch.Size([1, 36317]) Labels shape: torch.Size([1, 36317]) Final batch size: 1, sequence length: 37946 Attention mask shape: torch.Size([1, 1, 37946, 37946]) Position ids shape: torch.Size([1, 37946]) Input IDs shape: torch.Size([1, 37946]) Labels shape: torch.Size([1, 37946]) Final batch size: 1, sequence length: 20843 Attention mask shape: torch.Size([1, 1, 20843, 20843]) Position ids shape: torch.Size([1, 20843]) Input IDs shape: torch.Size([1, 20843]) Labels shape: torch.Size([1, 20843]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21496 Attention mask shape: torch.Size([1, 1, 21496, 21496]) Position ids shape: torch.Size([1, 21496]) Input IDs shape: torch.Size([1, 21496]) Labels shape: torch.Size([1, 21496]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24551 Attention mask shape: torch.Size([1, 1, 24551, 24551]) Position ids shape: torch.Size([1, 24551]) Input IDs shape: torch.Size([1, 24551]) Labels shape: torch.Size([1, 24551]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7681 Attention mask shape: torch.Size([1, 1, 7681, 7681]) Position ids shape: torch.Size([1, 7681]) Input IDs shape: torch.Size([1, 7681]) Labels shape: torch.Size([1, 7681]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2742, 'grad_norm': 0.27956884322049164, 'learning_rate': 4.738320218785281e-06, 'num_tokens': -inf, 'epoch': 4.5} Final batch size: 1, sequence length: 5328 Attention mask shape: torch.Size([1, 1, 5328, 5328]) Position ids shape: torch.Size([1, 5328]) Input IDs shape: torch.Size([1, 5328]) Labels shape: torch.Size([1, 5328]) Final batch size: 1, sequence length: 5525 Attention mask shape: torch.Size([1, 1, 5525, 5525]) Position ids shape: torch.Size([1, 5525]) Input IDs shape: torch.Size([1, 5525]) Labels shape: torch.Size([1, 5525]) Final batch size: 1, sequence length: 10273 Attention mask shape: torch.Size([1, 1, 10273, 10273]) Position ids shape: torch.Size([1, 10273]) Input IDs shape: torch.Size([1, 10273]) Labels shape: torch.Size([1, 10273]) Final batch size: 1, sequence length: 11464 Attention mask shape: torch.Size([1, 1, 11464, 11464]) Position ids shape: torch.Size([1, 11464]) Input IDs shape: torch.Size([1, 11464]) Labels shape: torch.Size([1, 11464]) Final batch size: 1, sequence length: 12419 Attention mask shape: torch.Size([1, 1, 12419, 12419]) Position ids shape: torch.Size([1, 12419]) Input IDs shape: torch.Size([1, 12419]) Labels shape: torch.Size([1, 12419]) Final batch size: 1, sequence length: 9648 Attention mask shape: torch.Size([1, 1, 9648, 9648]) Position ids shape: torch.Size([1, 9648]) Input IDs shape: torch.Size([1, 9648]) Labels shape: torch.Size([1, 9648]) Final batch size: 1, sequence length: 13657 Attention mask shape: torch.Size([1, 1, 13657, 13657]) Position ids shape: torch.Size([1, 13657]) Input IDs shape: torch.Size([1, 13657]) Labels shape: torch.Size([1, 13657]) Final batch size: 1, sequence length: 14226 Attention mask shape: torch.Size([1, 1, 14226, 14226]) Position ids shape: torch.Size([1, 14226]) Input IDs shape: torch.Size([1, 14226]) Labels shape: torch.Size([1, 14226]) Final batch size: 1, sequence length: 14360 Attention mask shape: torch.Size([1, 1, 14360, 14360]) Position ids shape: torch.Size([1, 14360]) Input IDs shape: torch.Size([1, 14360]) Labels shape: torch.Size([1, 14360]) Final batch size: 1, sequence length: 10434 Attention mask shape: torch.Size([1, 1, 10434, 10434]) Position ids shape: torch.Size([1, 10434]) Input IDs shape: torch.Size([1, 10434]) Labels shape: torch.Size([1, 10434]) Final batch size: 1, sequence length: 16053 Attention mask shape: torch.Size([1, 1, 16053, 16053]) Position ids shape: torch.Size([1, 16053]) Input IDs shape: torch.Size([1, 16053]) Labels shape: torch.Size([1, 16053]) Final batch size: 1, sequence length: 13607 Attention mask shape: torch.Size([1, 1, 13607, 13607]) Position ids shape: torch.Size([1, 13607]) Input IDs shape: torch.Size([1, 13607]) Labels shape: torch.Size([1, 13607]) Final batch size: 1, sequence length: 16756 Attention mask shape: torch.Size([1, 1, 16756, 16756]) Position ids shape: torch.Size([1, 16756]) Input IDs shape: torch.Size([1, 16756]) Labels shape: torch.Size([1, 16756]) Final batch size: 1, sequence length: 11225 Attention mask shape: torch.Size([1, 1, 11225, 11225]) Position ids shape: torch.Size([1, 11225]) Input IDs shape: torch.Size([1, 11225]) Labels shape: torch.Size([1, 11225]) Final batch size: 1, sequence length: 16223 Attention mask shape: torch.Size([1, 1, 16223, 16223]) Position ids shape: torch.Size([1, 16223]) Input IDs shape: torch.Size([1, 16223]) Labels shape: torch.Size([1, 16223]) Final batch size: 1, sequence length: 16398 Attention mask shape: torch.Size([1, 1, 16398, 16398]) Position ids shape: torch.Size([1, 16398]) Input IDs shape: torch.Size([1, 16398]) Labels shape: torch.Size([1, 16398]) Final batch size: 1, sequence length: 10017 Attention mask shape: torch.Size([1, 1, 10017, 10017]) Position ids shape: torch.Size([1, 10017]) Input IDs shape: torch.Size([1, 10017]) Labels shape: torch.Size([1, 10017]) Final batch size: 1, sequence length: 17294 Attention mask shape: torch.Size([1, 1, 17294, 17294]) Position ids shape: torch.Size([1, 17294]) Input IDs shape: torch.Size([1, 17294]) Labels shape: torch.Size([1, 17294]) Final batch size: 1, sequence length: 17843 Attention mask shape: torch.Size([1, 1, 17843, 17843]) Position ids shape: torch.Size([1, 17843]) Input IDs shape: torch.Size([1, 17843]) Labels shape: torch.Size([1, 17843]) Final batch size: 1, sequence length: 21455 Attention mask shape: torch.Size([1, 1, 21455, 21455]) Position ids shape: torch.Size([1, 21455]) Input IDs shape: torch.Size([1, 21455]) Labels shape: torch.Size([1, 21455]) Final batch size: 1, sequence length: 19892 Attention mask shape: torch.Size([1, 1, 19892, 19892]) Position ids shape: torch.Size([1, 19892]) Input IDs shape: torch.Size([1, 19892]) Labels shape: torch.Size([1, 19892]) Final batch size: 1, sequence length: 18408 Attention mask shape: torch.Size([1, 1, 18408, 18408]) Position ids shape: torch.Size([1, 18408]) Input IDs shape: torch.Size([1, 18408]) Labels shape: torch.Size([1, 18408]) Final batch size: 1, sequence length: 20433 Attention mask shape: torch.Size([1, 1, 20433, 20433]) Position ids shape: torch.Size([1, 20433]) Input IDs shape: torch.Size([1, 20433]) Labels shape: torch.Size([1, 20433]) Final batch size: 1, sequence length: 17117 Attention mask shape: torch.Size([1, 1, 17117, 17117]) Position ids shape: torch.Size([1, 17117]) Input IDs shape: torch.Size([1, 17117]) Labels shape: torch.Size([1, 17117]) Final batch size: 1, sequence length: 19278 Attention mask shape: torch.Size([1, 1, 19278, 19278]) Position ids shape: torch.Size([1, 19278]) Input IDs shape: torch.Size([1, 19278]) Labels shape: torch.Size([1, 19278]) Final batch size: 1, sequence length: 14437 Attention mask shape: torch.Size([1, 1, 14437, 14437]) Position ids shape: torch.Size([1, 14437]) Input IDs shape: torch.Size([1, 14437]) Labels shape: torch.Size([1, 14437]) Final batch size: 1, sequence length: 15243 Attention mask shape: torch.Size([1, 1, 15243, 15243]) Position ids shape: torch.Size([1, 15243]) Input IDs shape: torch.Size([1, 15243]) Labels shape: torch.Size([1, 15243]) Final batch size: 1, sequence length: 20695 Attention mask shape: torch.Size([1, 1, 20695, 20695]) Position ids shape: torch.Size([1, 20695]) Input IDs shape: torch.Size([1, 20695]) Labels shape: torch.Size([1, 20695]) Final batch size: 1, sequence length: 19767 Attention mask shape: torch.Size([1, 1, 19767, 19767]) Position ids shape: torch.Size([1, 19767]) Input IDs shape: torch.Size([1, 19767]) Labels shape: torch.Size([1, 19767]) Final batch size: 1, sequence length: 19259 Attention mask shape: torch.Size([1, 1, 19259, 19259]) Position ids shape: torch.Size([1, 19259]) Input IDs shape: torch.Size([1, 19259]) Labels shape: torch.Size([1, 19259]) Final batch size: 1, sequence length: 5734 Attention mask shape: torch.Size([1, 1, 5734, 5734]) Position ids shape: torch.Size([1, 5734]) Input IDs shape: torch.Size([1, 5734]) Labels shape: torch.Size([1, 5734]) Final batch size: 1, sequence length: 14025 Attention mask shape: torch.Size([1, 1, 14025, 14025]) Position ids shape: torch.Size([1, 14025]) Input IDs shape: torch.Size([1, 14025]) Labels shape: torch.Size([1, 14025]) Final batch size: 1, sequence length: 13623 Attention mask shape: torch.Size([1, 1, 13623, 13623]) Position ids shape: torch.Size([1, 13623]) Input IDs shape: torch.Size([1, 13623]) Labels shape: torch.Size([1, 13623]) Final batch size: 1, sequence length: 6378 Attention mask shape: torch.Size([1, 1, 6378, 6378]) Position ids shape: torch.Size([1, 6378]) Input IDs shape: torch.Size([1, 6378]) Labels shape: torch.Size([1, 6378]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 22932 Attention mask shape: torch.Size([1, 1, 22932, 22932]) Position ids shape: torch.Size([1, 22932]) Input IDs shape: torch.Size([1, 22932]) Labels shape: torch.Size([1, 22932]) Final batch size: 1, sequence length: 23334 Attention mask shape: torch.Size([1, 1, 23334, 23334]) Position ids shape: torch.Size([1, 23334]) Input IDs shape: torch.Size([1, 23334]) Labels shape: torch.Size([1, 23334]) Final batch size: 1, sequence length: 25405 Attention mask shape: torch.Size([1, 1, 25405, 25405]) Position ids shape: torch.Size([1, 25405]) Input IDs shape: torch.Size([1, 25405]) Labels shape: torch.Size([1, 25405]) Final batch size: 1, sequence length: 19492 Attention mask shape: torch.Size([1, 1, 19492, 19492]) Position ids shape: torch.Size([1, 19492]) Input IDs shape: torch.Size([1, 19492]) Labels shape: torch.Size([1, 19492]) Final batch size: 1, sequence length: 14429 Attention mask shape: torch.Size([1, 1, 14429, 14429]) Position ids shape: torch.Size([1, 14429]) Input IDs shape: torch.Size([1, 14429]) Labels shape: torch.Size([1, 14429]) Final batch size: 1, sequence length: 26063 Attention mask shape: torch.Size([1, 1, 26063, 26063]) Position ids shape: torch.Size([1, 26063]) Input IDs shape: torch.Size([1, 26063]) Labels shape: torch.Size([1, 26063]) Final batch size: 1, sequence length: 26639 Attention mask shape: torch.Size([1, 1, 26639, 26639]) Position ids shape: torch.Size([1, 26639]) Input IDs shape: torch.Size([1, 26639]) Labels shape: torch.Size([1, 26639]) Final batch size: 1, sequence length: 25548 Attention mask shape: torch.Size([1, 1, 25548, 25548]) Position ids shape: torch.Size([1, 25548]) Input IDs shape: torch.Size([1, 25548]) Labels shape: torch.Size([1, 25548]) Final batch size: 1, sequence length: 25381 Attention mask shape: torch.Size([1, 1, 25381, 25381]) Position ids shape: torch.Size([1, 25381]) Input IDs shape: torch.Size([1, 25381]) Labels shape: torch.Size([1, 25381]) Final batch size: 1, sequence length: 27659 Attention mask shape: torch.Size([1, 1, 27659, 27659]) Position ids shape: torch.Size([1, 27659]) Input IDs shape: torch.Size([1, 27659]) Labels shape: torch.Size([1, 27659]) Final batch size: 1, sequence length: 29561 Attention mask shape: torch.Size([1, 1, 29561, 29561]) Position ids shape: torch.Size([1, 29561]) Input IDs shape: torch.Size([1, 29561]) Labels shape: torch.Size([1, 29561]) Final batch size: 1, sequence length: 24885 Attention mask shape: torch.Size([1, 1, 24885, 24885]) Position ids shape: torch.Size([1, 24885]) Input IDs shape: torch.Size([1, 24885]) Labels shape: torch.Size([1, 24885]) Final batch size: 1, sequence length: 27484 Attention mask shape: torch.Size([1, 1, 27484, 27484]) Position ids shape: torch.Size([1, 27484]) Input IDs shape: torch.Size([1, 27484]) Labels shape: torch.Size([1, 27484]) Final batch size: 1, sequence length: 16716 Attention mask shape: torch.Size([1, 1, 16716, 16716]) Position ids shape: torch.Size([1, 16716]) Input IDs shape: torch.Size([1, 16716]) Labels shape: torch.Size([1, 16716]) Final batch size: 1, sequence length: 28623 Attention mask shape: torch.Size([1, 1, 28623, 28623]) Position ids shape: torch.Size([1, 28623]) Input IDs shape: torch.Size([1, 28623]) Labels shape: torch.Size([1, 28623]) Final batch size: 1, sequence length: 17985 Attention mask shape: torch.Size([1, 1, 17985, 17985]) Position ids shape: torch.Size([1, 17985]) Input IDs shape: torch.Size([1, 17985]) Labels shape: torch.Size([1, 17985]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 28166 Attention mask shape: torch.Size([1, 1, 28166, 28166]) Position ids shape: torch.Size([1, 28166]) Input IDs shape: torch.Size([1, 28166]) Labels shape: torch.Size([1, 28166]) Final batch size: 1, sequence length: 17951 Attention mask shape: torch.Size([1, 1, 17951, 17951]) Position ids shape: torch.Size([1, 17951]) Input IDs shape: torch.Size([1, 17951]) Labels shape: torch.Size([1, 17951]) Final batch size: 1, sequence length: 30236 Attention mask shape: torch.Size([1, 1, 30236, 30236]) Position ids shape: torch.Size([1, 30236]) Input IDs shape: torch.Size([1, 30236]) Labels shape: torch.Size([1, 30236]) Final batch size: 1, sequence length: 28823 Attention mask shape: torch.Size([1, 1, 28823, 28823]) Position ids shape: torch.Size([1, 28823]) Input IDs shape: torch.Size([1, 28823]) Labels shape: torch.Size([1, 28823]) Final batch size: 1, sequence length: 29824 Attention mask shape: torch.Size([1, 1, 29824, 29824]) Position ids shape: torch.Size([1, 29824]) Input IDs shape: torch.Size([1, 29824]) Labels shape: torch.Size([1, 29824]) Final batch size: 1, sequence length: 28910 Attention mask shape: torch.Size([1, 1, 28910, 28910]) Position ids shape: torch.Size([1, 28910]) Input IDs shape: torch.Size([1, 28910]) Labels shape: torch.Size([1, 28910]) Final batch size: 1, sequence length: 25540 Attention mask shape: torch.Size([1, 1, 25540, 25540]) Position ids shape: torch.Size([1, 25540]) Input IDs shape: torch.Size([1, 25540]) Labels shape: torch.Size([1, 25540]) Final batch size: 1, sequence length: 28206 Attention mask shape: torch.Size([1, 1, 28206, 28206]) Position ids shape: torch.Size([1, 28206]) Input IDs shape: torch.Size([1, 28206]) Labels shape: torch.Size([1, 28206]) Final batch size: 1, sequence length: 17634 Attention mask shape: torch.Size([1, 1, 17634, 17634]) Position ids shape: torch.Size([1, 17634]) Input IDs shape: torch.Size([1, 17634]) Labels shape: torch.Size([1, 17634]) Final batch size: 1, sequence length: 31464 Attention mask shape: torch.Size([1, 1, 31464, 31464]) Position ids shape: torch.Size([1, 31464]) Input IDs shape: torch.Size([1, 31464]) Labels shape: torch.Size([1, 31464]) Final batch size: 1, sequence length: 32515 Attention mask shape: torch.Size([1, 1, 32515, 32515]) Position ids shape: torch.Size([1, 32515]) Input IDs shape: torch.Size([1, 32515]) Labels shape: torch.Size([1, 32515]) Final batch size: 1, sequence length: 33601 Attention mask shape: torch.Size([1, 1, 33601, 33601]) Position ids shape: torch.Size([1, 33601]) Input IDs shape: torch.Size([1, 33601]) Labels shape: torch.Size([1, 33601]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 32660 Attention mask shape: torch.Size([1, 1, 32660, 32660]) Position ids shape: torch.Size([1, 32660]) Input IDs shape: torch.Size([1, 32660]) Labels shape: torch.Size([1, 32660]) Final batch size: 1, sequence length: 32786 Attention mask shape: torch.Size([1, 1, 32786, 32786]) Position ids shape: torch.Size([1, 32786]) Input IDs shape: torch.Size([1, 32786]) Labels shape: torch.Size([1, 32786]) Final batch size: 1, sequence length: 9029 Attention mask shape: torch.Size([1, 1, 9029, 9029]) Position ids shape: torch.Size([1, 9029]) Input IDs shape: torch.Size([1, 9029]) Labels shape: torch.Size([1, 9029]) Final batch size: 1, sequence length: 35411 Attention mask shape: torch.Size([1, 1, 35411, 35411]) Position ids shape: torch.Size([1, 35411]) Input IDs shape: torch.Size([1, 35411]) Labels shape: torch.Size([1, 35411]) Final batch size: 1, sequence length: 17914 Attention mask shape: torch.Size([1, 1, 17914, 17914]) Position ids shape: torch.Size([1, 17914]) Input IDs shape: torch.Size([1, 17914]) Labels shape: torch.Size([1, 17914]) Final batch size: 1, sequence length: 37555 Attention mask shape: torch.Size([1, 1, 37555, 37555]) Position ids shape: torch.Size([1, 37555]) Input IDs shape: torch.Size([1, 37555]) Labels shape: torch.Size([1, 37555]) Final batch size: 1, sequence length: 36131 Attention mask shape: torch.Size([1, 1, 36131, 36131]) Position ids shape: torch.Size([1, 36131]) Input IDs shape: torch.Size([1, 36131]) Labels shape: torch.Size([1, 36131]) Final batch size: 1, sequence length: 10132 Attention mask shape: torch.Size([1, 1, 10132, 10132]) Position ids shape: torch.Size([1, 10132]) Input IDs shape: torch.Size([1, 10132]) Labels shape: torch.Size([1, 10132]) Final batch size: 1, sequence length: 20951 Attention mask shape: torch.Size([1, 1, 20951, 20951]) Position ids shape: torch.Size([1, 20951]) Input IDs shape: torch.Size([1, 20951]) Labels shape: torch.Size([1, 20951]) Final batch size: 1, sequence length: 15513 Attention mask shape: torch.Size([1, 1, 15513, 15513]) Position ids shape: torch.Size([1, 15513]) Input IDs shape: torch.Size([1, 15513]) Labels shape: torch.Size([1, 15513]) Final batch size: 1, sequence length: 36970 Attention mask shape: torch.Size([1, 1, 36970, 36970]) Position ids shape: torch.Size([1, 36970]) Input IDs shape: torch.Size([1, 36970]) Labels shape: torch.Size([1, 36970]) Final batch size: 1, sequence length: 37016 Attention mask shape: torch.Size([1, 1, 37016, 37016]) Position ids shape: torch.Size([1, 37016]) Input IDs shape: torch.Size([1, 37016]) Labels shape: torch.Size([1, 37016]) Final batch size: 1, sequence length: 36124 Attention mask shape: torch.Size([1, 1, 36124, 36124]) Position ids shape: torch.Size([1, 36124]) Input IDs shape: torch.Size([1, 36124]) Labels shape: torch.Size([1, 36124]) Final batch size: 1, sequence length: 25529 Attention mask shape: torch.Size([1, 1, 25529, 25529]) Position ids shape: torch.Size([1, 25529]) Input IDs shape: torch.Size([1, 25529]) Labels shape: torch.Size([1, 25529]) Final batch size: 1, sequence length: 20533 Attention mask shape: torch.Size([1, 1, 20533, 20533]) Position ids shape: torch.Size([1, 20533]) Input IDs shape: torch.Size([1, 20533]) Labels shape: torch.Size([1, 20533]) Final batch size: 1, sequence length: 38712 Attention mask shape: torch.Size([1, 1, 38712, 38712]) Position ids shape: torch.Size([1, 38712]) Input IDs shape: torch.Size([1, 38712]) Labels shape: torch.Size([1, 38712]) Final batch size: 1, sequence length: 35525 Attention mask shape: torch.Size([1, 1, 35525, 35525]) Position ids shape: torch.Size([1, 35525]) Input IDs shape: torch.Size([1, 35525]) Labels shape: torch.Size([1, 35525]) Final batch size: 1, sequence length: 26562 Attention mask shape: torch.Size([1, 1, 26562, 26562]) Position ids shape: torch.Size([1, 26562]) Input IDs shape: torch.Size([1, 26562]) Labels shape: torch.Size([1, 26562]) Final batch size: 1, sequence length: 14372 Attention mask shape: torch.Size([1, 1, 14372, 14372]) Position ids shape: torch.Size([1, 14372]) Input IDs shape: torch.Size([1, 14372]) Labels shape: torch.Size([1, 14372]) Final batch size: 1, sequence length: 38071 Attention mask shape: torch.Size([1, 1, 38071, 38071]) Position ids shape: torch.Size([1, 38071]) Input IDs shape: torch.Size([1, 38071]) Labels shape: torch.Size([1, 38071]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36292 Attention mask shape: torch.Size([1, 1, 36292, 36292]) Position ids shape: torch.Size([1, 36292]) Input IDs shape: torch.Size([1, 36292]) Labels shape: torch.Size([1, 36292]) Final batch size: 1, sequence length: 26781 Attention mask shape: torch.Size([1, 1, 26781, 26781]) Position ids shape: torch.Size([1, 26781]) Input IDs shape: torch.Size([1, 26781]) Labels shape: torch.Size([1, 26781]) Final batch size: 1, sequence length: 31084 Attention mask shape: torch.Size([1, 1, 31084, 31084]) Position ids shape: torch.Size([1, 31084]) Input IDs shape: torch.Size([1, 31084]) Labels shape: torch.Size([1, 31084]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 28879 Attention mask shape: torch.Size([1, 1, 28879, 28879]) Position ids shape: torch.Size([1, 28879]) Input IDs shape: torch.Size([1, 28879]) Labels shape: torch.Size([1, 28879]) Final batch size: 1, sequence length: 33422 Attention mask shape: torch.Size([1, 1, 33422, 33422]) Position ids shape: torch.Size([1, 33422]) Input IDs shape: torch.Size([1, 33422]) Labels shape: torch.Size([1, 33422]) Final batch size: 1, sequence length: 38210 Attention mask shape: torch.Size([1, 1, 38210, 38210]) Position ids shape: torch.Size([1, 38210]) Input IDs shape: torch.Size([1, 38210]) Labels shape: torch.Size([1, 38210]) Final batch size: 1, sequence length: 35775 Attention mask shape: torch.Size([1, 1, 35775, 35775]) Position ids shape: torch.Size([1, 35775]) Input IDs shape: torch.Size([1, 35775]) Labels shape: torch.Size([1, 35775]) Final batch size: 1, sequence length: 36185 Attention mask shape: torch.Size([1, 1, 36185, 36185]) Position ids shape: torch.Size([1, 36185]) Input IDs shape: torch.Size([1, 36185]) Labels shape: torch.Size([1, 36185]) Final batch size: 1, sequence length: 16409 Attention mask shape: torch.Size([1, 1, 16409, 16409]) Position ids shape: torch.Size([1, 16409]) Input IDs shape: torch.Size([1, 16409]) Labels shape: torch.Size([1, 16409]) Final batch size: 1, sequence length: 30009 Attention mask shape: torch.Size([1, 1, 30009, 30009]) Position ids shape: torch.Size([1, 30009]) Input IDs shape: torch.Size([1, 30009]) Labels shape: torch.Size([1, 30009]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40753 Attention mask shape: torch.Size([1, 1, 40753, 40753]) Position ids shape: torch.Size([1, 40753]) Input IDs shape: torch.Size([1, 40753]) Labels shape: torch.Size([1, 40753]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30653 Attention mask shape: torch.Size([1, 1, 30653, 30653]) Position ids shape: torch.Size([1, 30653]) Input IDs shape: torch.Size([1, 30653]) Labels shape: torch.Size([1, 30653]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31042 Attention mask shape: torch.Size([1, 1, 31042, 31042]) Position ids shape: torch.Size([1, 31042]) Input IDs shape: torch.Size([1, 31042]) Labels shape: torch.Size([1, 31042]) Final batch size: 1, sequence length: 36534 Attention mask shape: torch.Size([1, 1, 36534, 36534]) Position ids shape: torch.Size([1, 36534]) Input IDs shape: torch.Size([1, 36534]) Labels shape: torch.Size([1, 36534]) Final batch size: 1, sequence length: 37159 Attention mask shape: torch.Size([1, 1, 37159, 37159]) Position ids shape: torch.Size([1, 37159]) Input IDs shape: torch.Size([1, 37159]) Labels shape: torch.Size([1, 37159]) Final batch size: 1, sequence length: 15031 Attention mask shape: torch.Size([1, 1, 15031, 15031]) Position ids shape: torch.Size([1, 15031]) Input IDs shape: torch.Size([1, 15031]) Labels shape: torch.Size([1, 15031]) Final batch size: 1, sequence length: 18884 Attention mask shape: torch.Size([1, 1, 18884, 18884]) Position ids shape: torch.Size([1, 18884]) Input IDs shape: torch.Size([1, 18884]) Labels shape: torch.Size([1, 18884]) Final batch size: 1, sequence length: 7448 Attention mask shape: torch.Size([1, 1, 7448, 7448]) Position ids shape: torch.Size([1, 7448]) Input IDs shape: torch.Size([1, 7448]) Labels shape: torch.Size([1, 7448]) Final batch size: 1, sequence length: 20307 Attention mask shape: torch.Size([1, 1, 20307, 20307]) Position ids shape: torch.Size([1, 20307]) Input IDs shape: torch.Size([1, 20307]) Labels shape: torch.Size([1, 20307]) Final batch size: 1, sequence length: 25850 Attention mask shape: torch.Size([1, 1, 25850, 25850]) Position ids shape: torch.Size([1, 25850]) Input IDs shape: torch.Size([1, 25850]) Labels shape: torch.Size([1, 25850]) Final batch size: 1, sequence length: 31448 Attention mask shape: torch.Size([1, 1, 31448, 31448]) Position ids shape: torch.Size([1, 31448]) Input IDs shape: torch.Size([1, 31448]) Labels shape: torch.Size([1, 31448]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40605 Attention mask shape: torch.Size([1, 1, 40605, 40605]) Position ids shape: torch.Size([1, 40605]) Input IDs shape: torch.Size([1, 40605]) Labels shape: torch.Size([1, 40605]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 5405 Attention mask shape: torch.Size([1, 1, 5405, 5405]) Position ids shape: torch.Size([1, 5405]) Input IDs shape: torch.Size([1, 5405]) Labels shape: torch.Size([1, 5405]) {'loss': 0.2598, 'grad_norm': 0.2609480757949517, 'learning_rate': 4.477357683661734e-06, 'num_tokens': -inf, 'epoch': 4.62} Final batch size: 1, sequence length: 4641 Attention mask shape: torch.Size([1, 1, 4641, 4641]) Position ids shape: torch.Size([1, 4641]) Input IDs shape: torch.Size([1, 4641]) Labels shape: torch.Size([1, 4641]) Final batch size: 1, sequence length: 4223 Attention mask shape: torch.Size([1, 1, 4223, 4223]) Position ids shape: torch.Size([1, 4223]) Input IDs shape: torch.Size([1, 4223]) Labels shape: torch.Size([1, 4223]) Final batch size: 1, sequence length: 5754 Attention mask shape: torch.Size([1, 1, 5754, 5754]) Position ids shape: torch.Size([1, 5754]) Input IDs shape: torch.Size([1, 5754]) Labels shape: torch.Size([1, 5754]) Final batch size: 1, sequence length: 7795 Attention mask shape: torch.Size([1, 1, 7795, 7795]) Position ids shape: torch.Size([1, 7795]) Input IDs shape: torch.Size([1, 7795]) Labels shape: torch.Size([1, 7795]) Final batch size: 1, sequence length: 9021 Attention mask shape: torch.Size([1, 1, 9021, 9021]) Position ids shape: torch.Size([1, 9021]) Input IDs shape: torch.Size([1, 9021]) Labels shape: torch.Size([1, 9021]) Final batch size: 1, sequence length: 8177 Attention mask shape: torch.Size([1, 1, 8177, 8177]) Position ids shape: torch.Size([1, 8177]) Input IDs shape: torch.Size([1, 8177]) Labels shape: torch.Size([1, 8177]) Final batch size: 1, sequence length: 12501 Attention mask shape: torch.Size([1, 1, 12501, 12501]) Position ids shape: torch.Size([1, 12501]) Input IDs shape: torch.Size([1, 12501]) Labels shape: torch.Size([1, 12501]) Final batch size: 1, sequence length: 9027 Attention mask shape: torch.Size([1, 1, 9027, 9027]) Position ids shape: torch.Size([1, 9027]) Input IDs shape: torch.Size([1, 9027]) Labels shape: torch.Size([1, 9027]) Final batch size: 1, sequence length: 11974 Attention mask shape: torch.Size([1, 1, 11974, 11974]) Position ids shape: torch.Size([1, 11974]) Input IDs shape: torch.Size([1, 11974]) Labels shape: torch.Size([1, 11974]) Final batch size: 1, sequence length: 16330 Attention mask shape: torch.Size([1, 1, 16330, 16330]) Position ids shape: torch.Size([1, 16330]) Input IDs shape: torch.Size([1, 16330]) Labels shape: torch.Size([1, 16330]) Final batch size: 1, sequence length: 13186 Attention mask shape: torch.Size([1, 1, 13186, 13186]) Position ids shape: torch.Size([1, 13186]) Input IDs shape: torch.Size([1, 13186]) Labels shape: torch.Size([1, 13186]) Final batch size: 1, sequence length: 15283 Attention mask shape: torch.Size([1, 1, 15283, 15283]) Position ids shape: torch.Size([1, 15283]) Input IDs shape: torch.Size([1, 15283]) Labels shape: torch.Size([1, 15283]) Final batch size: 1, sequence length: 16104 Attention mask shape: torch.Size([1, 1, 16104, 16104]) Position ids shape: torch.Size([1, 16104]) Input IDs shape: torch.Size([1, 16104]) Labels shape: torch.Size([1, 16104]) Final batch size: 1, sequence length: 16496 Attention mask shape: torch.Size([1, 1, 16496, 16496]) Position ids shape: torch.Size([1, 16496]) Input IDs shape: torch.Size([1, 16496]) Labels shape: torch.Size([1, 16496]) Final batch size: 1, sequence length: 13957 Attention mask shape: torch.Size([1, 1, 13957, 13957]) Position ids shape: torch.Size([1, 13957]) Input IDs shape: torch.Size([1, 13957]) Labels shape: torch.Size([1, 13957]) Final batch size: 1, sequence length: 17092 Attention mask shape: torch.Size([1, 1, 17092, 17092]) Position ids shape: torch.Size([1, 17092]) Input IDs shape: torch.Size([1, 17092]) Labels shape: torch.Size([1, 17092]) Final batch size: 1, sequence length: 16366 Attention mask shape: torch.Size([1, 1, 16366, 16366]) Position ids shape: torch.Size([1, 16366]) Input IDs shape: torch.Size([1, 16366]) Labels shape: torch.Size([1, 16366]) Final batch size: 1, sequence length: 16088 Attention mask shape: torch.Size([1, 1, 16088, 16088]) Position ids shape: torch.Size([1, 16088]) Input IDs shape: torch.Size([1, 16088]) Labels shape: torch.Size([1, 16088]) Final batch size: 1, sequence length: 18958 Attention mask shape: torch.Size([1, 1, 18958, 18958]) Position ids shape: torch.Size([1, 18958]) Input IDs shape: torch.Size([1, 18958]) Labels shape: torch.Size([1, 18958]) Final batch size: 1, sequence length: 18415 Attention mask shape: torch.Size([1, 1, 18415, 18415]) Position ids shape: torch.Size([1, 18415]) Input IDs shape: torch.Size([1, 18415]) Labels shape: torch.Size([1, 18415]) Final batch size: 1, sequence length: 18938 Attention mask shape: torch.Size([1, 1, 18938, 18938]) Position ids shape: torch.Size([1, 18938]) Input IDs shape: torch.Size([1, 18938]) Labels shape: torch.Size([1, 18938]) Final batch size: 1, sequence length: 16014 Attention mask shape: torch.Size([1, 1, 16014, 16014]) Position ids shape: torch.Size([1, 16014]) Input IDs shape: torch.Size([1, 16014]) Labels shape: torch.Size([1, 16014]) Final batch size: 1, sequence length: 14492 Attention mask shape: torch.Size([1, 1, 14492, 14492]) Position ids shape: torch.Size([1, 14492]) Input IDs shape: torch.Size([1, 14492]) Labels shape: torch.Size([1, 14492]) Final batch size: 1, sequence length: 19034 Attention mask shape: torch.Size([1, 1, 19034, 19034]) Position ids shape: torch.Size([1, 19034]) Input IDs shape: torch.Size([1, 19034]) Labels shape: torch.Size([1, 19034]) Final batch size: 1, sequence length: 19603 Attention mask shape: torch.Size([1, 1, 19603, 19603]) Position ids shape: torch.Size([1, 19603]) Input IDs shape: torch.Size([1, 19603]) Labels shape: torch.Size([1, 19603]) Final batch size: 1, sequence length: 21132 Attention mask shape: torch.Size([1, 1, 21132, 21132]) Position ids shape: torch.Size([1, 21132]) Input IDs shape: torch.Size([1, 21132]) Labels shape: torch.Size([1, 21132]) Final batch size: 1, sequence length: 18034 Attention mask shape: torch.Size([1, 1, 18034, 18034]) Position ids shape: torch.Size([1, 18034]) Input IDs shape: torch.Size([1, 18034]) Labels shape: torch.Size([1, 18034]) Final batch size: 1, sequence length: 21705 Attention mask shape: torch.Size([1, 1, 21705, 21705]) Position ids shape: torch.Size([1, 21705]) Input IDs shape: torch.Size([1, 21705]) Labels shape: torch.Size([1, 21705]) Final batch size: 1, sequence length: 13468 Attention mask shape: torch.Size([1, 1, 13468, 13468]) Position ids shape: torch.Size([1, 13468]) Input IDs shape: torch.Size([1, 13468]) Labels shape: torch.Size([1, 13468]) Final batch size: 1, sequence length: 23132 Attention mask shape: torch.Size([1, 1, 23132, 23132]) Position ids shape: torch.Size([1, 23132]) Input IDs shape: torch.Size([1, 23132]) Labels shape: torch.Size([1, 23132]) Final batch size: 1, sequence length: 20726 Attention mask shape: torch.Size([1, 1, 20726, 20726]) Position ids shape: torch.Size([1, 20726]) Input IDs shape: torch.Size([1, 20726]) Labels shape: torch.Size([1, 20726]) Final batch size: 1, sequence length: 24432 Attention mask shape: torch.Size([1, 1, 24432, 24432]) Position ids shape: torch.Size([1, 24432]) Input IDs shape: torch.Size([1, 24432]) Labels shape: torch.Size([1, 24432]) Final batch size: 1, sequence length: 22771 Attention mask shape: torch.Size([1, 1, 22771, 22771]) Position ids shape: torch.Size([1, 22771]) Input IDs shape: torch.Size([1, 22771]) Labels shape: torch.Size([1, 22771]) Final batch size: 1, sequence length: 23102 Attention mask shape: torch.Size([1, 1, 23102, 23102]) Position ids shape: torch.Size([1, 23102]) Input IDs shape: torch.Size([1, 23102]) Labels shape: torch.Size([1, 23102]) Final batch size: 1, sequence length: 12756 Attention mask shape: torch.Size([1, 1, 12756, 12756]) Position ids shape: torch.Size([1, 12756]) Input IDs shape: torch.Size([1, 12756]) Labels shape: torch.Size([1, 12756]) Final batch size: 1, sequence length: 21404 Attention mask shape: torch.Size([1, 1, 21404, 21404]) Position ids shape: torch.Size([1, 21404]) Input IDs shape: torch.Size([1, 21404]) Labels shape: torch.Size([1, 21404]) Final batch size: 1, sequence length: 16291 Attention mask shape: torch.Size([1, 1, 16291, 16291]) Position ids shape: torch.Size([1, 16291]) Input IDs shape: torch.Size([1, 16291]) Labels shape: torch.Size([1, 16291]) Final batch size: 1, sequence length: 23971 Attention mask shape: torch.Size([1, 1, 23971, 23971]) Position ids shape: torch.Size([1, 23971]) Input IDs shape: torch.Size([1, 23971]) Labels shape: torch.Size([1, 23971]) Final batch size: 1, sequence length: 23894 Attention mask shape: torch.Size([1, 1, 23894, 23894]) Position ids shape: torch.Size([1, 23894]) Input IDs shape: torch.Size([1, 23894]) Labels shape: torch.Size([1, 23894]) Final batch size: 1, sequence length: 25351 Attention mask shape: torch.Size([1, 1, 25351, 25351]) Position ids shape: torch.Size([1, 25351]) Input IDs shape: torch.Size([1, 25351]) Labels shape: torch.Size([1, 25351]) Final batch size: 1, sequence length: 17514 Attention mask shape: torch.Size([1, 1, 17514, 17514]) Position ids shape: torch.Size([1, 17514]) Input IDs shape: torch.Size([1, 17514]) Labels shape: torch.Size([1, 17514]) Final batch size: 1, sequence length: 27785 Attention mask shape: torch.Size([1, 1, 27785, 27785]) Position ids shape: torch.Size([1, 27785]) Input IDs shape: torch.Size([1, 27785]) Labels shape: torch.Size([1, 27785]) Final batch size: 1, sequence length: 13790 Attention mask shape: torch.Size([1, 1, 13790, 13790]) Position ids shape: torch.Size([1, 13790]) Input IDs shape: torch.Size([1, 13790]) Labels shape: torch.Size([1, 13790]) Final batch size: 1, sequence length: 16590 Attention mask shape: torch.Size([1, 1, 16590, 16590]) Position ids shape: torch.Size([1, 16590]) Input IDs shape: torch.Size([1, 16590]) Labels shape: torch.Size([1, 16590]) Final batch size: 1, sequence length: 23810 Attention mask shape: torch.Size([1, 1, 23810, 23810]) Position ids shape: torch.Size([1, 23810]) Input IDs shape: torch.Size([1, 23810]) Labels shape: torch.Size([1, 23810]) Final batch size: 1, sequence length: 20817 Attention mask shape: torch.Size([1, 1, 20817, 20817]) Position ids shape: torch.Size([1, 20817]) Input IDs shape: torch.Size([1, 20817]) Labels shape: torch.Size([1, 20817]) Final batch size: 1, sequence length: 26375 Attention mask shape: torch.Size([1, 1, 26375, 26375]) Position ids shape: torch.Size([1, 26375]) Input IDs shape: torch.Size([1, 26375]) Labels shape: torch.Size([1, 26375]) Final batch size: 1, sequence length: 27780 Attention mask shape: torch.Size([1, 1, 27780, 27780]) Position ids shape: torch.Size([1, 27780]) Input IDs shape: torch.Size([1, 27780]) Labels shape: torch.Size([1, 27780]) Final batch size: 1, sequence length: 28412 Attention mask shape: torch.Size([1, 1, 28412, 28412]) Position ids shape: torch.Size([1, 28412]) Input IDs shape: torch.Size([1, 28412]) Labels shape: torch.Size([1, 28412]) Final batch size: 1, sequence length: 18869 Attention mask shape: torch.Size([1, 1, 18869, 18869]) Position ids shape: torch.Size([1, 18869]) Input IDs shape: torch.Size([1, 18869]) Labels shape: torch.Size([1, 18869]) Final batch size: 1, sequence length: 28574 Attention mask shape: torch.Size([1, 1, 28574, 28574]) Position ids shape: torch.Size([1, 28574]) Input IDs shape: torch.Size([1, 28574]) Labels shape: torch.Size([1, 28574]) Final batch size: 1, sequence length: 32022 Attention mask shape: torch.Size([1, 1, 32022, 32022]) Position ids shape: torch.Size([1, 32022]) Input IDs shape: torch.Size([1, 32022]) Labels shape: torch.Size([1, 32022]) Final batch size: 1, sequence length: 26596 Attention mask shape: torch.Size([1, 1, 26596, 26596]) Position ids shape: torch.Size([1, 26596]) Input IDs shape: torch.Size([1, 26596]) Labels shape: torch.Size([1, 26596]) Final batch size: 1, sequence length: 30165 Attention mask shape: torch.Size([1, 1, 30165, 30165]) Position ids shape: torch.Size([1, 30165]) Input IDs shape: torch.Size([1, 30165]) Labels shape: torch.Size([1, 30165]) Final batch size: 1, sequence length: 26766 Attention mask shape: torch.Size([1, 1, 26766, 26766]) Position ids shape: torch.Size([1, 26766]) Input IDs shape: torch.Size([1, 26766]) Labels shape: torch.Size([1, 26766]) Final batch size: 1, sequence length: 25388 Attention mask shape: torch.Size([1, 1, 25388, 25388]) Position ids shape: torch.Size([1, 25388]) Input IDs shape: torch.Size([1, 25388]) Labels shape: torch.Size([1, 25388]) Final batch size: 1, sequence length: 28845 Attention mask shape: torch.Size([1, 1, 28845, 28845]) Position ids shape: torch.Size([1, 28845]) Input IDs shape: torch.Size([1, 28845]) Labels shape: torch.Size([1, 28845]) Final batch size: 1, sequence length: 29168 Attention mask shape: torch.Size([1, 1, 29168, 29168]) Position ids shape: torch.Size([1, 29168]) Input IDs shape: torch.Size([1, 29168]) Labels shape: torch.Size([1, 29168]) Final batch size: 1, sequence length: 29132 Attention mask shape: torch.Size([1, 1, 29132, 29132]) Position ids shape: torch.Size([1, 29132]) Input IDs shape: torch.Size([1, 29132]) Labels shape: torch.Size([1, 29132]) Final batch size: 1, sequence length: 34457 Attention mask shape: torch.Size([1, 1, 34457, 34457]) Position ids shape: torch.Size([1, 34457]) Input IDs shape: torch.Size([1, 34457]) Labels shape: torch.Size([1, 34457]) Final batch size: 1, sequence length: 34326 Attention mask shape: torch.Size([1, 1, 34326, 34326]) Position ids shape: torch.Size([1, 34326]) Input IDs shape: torch.Size([1, 34326]) Labels shape: torch.Size([1, 34326]) Final batch size: 1, sequence length: 22043 Attention mask shape: torch.Size([1, 1, 22043, 22043]) Position ids shape: torch.Size([1, 22043]) Input IDs shape: torch.Size([1, 22043]) Labels shape: torch.Size([1, 22043]) Final batch size: 1, sequence length: 29830 Attention mask shape: torch.Size([1, 1, 29830, 29830]) Position ids shape: torch.Size([1, 29830]) Input IDs shape: torch.Size([1, 29830]) Labels shape: torch.Size([1, 29830]) Final batch size: 1, sequence length: 24653 Attention mask shape: torch.Size([1, 1, 24653, 24653]) Position ids shape: torch.Size([1, 24653]) Input IDs shape: torch.Size([1, 24653]) Labels shape: torch.Size([1, 24653]) Final batch size: 1, sequence length: 35790 Attention mask shape: torch.Size([1, 1, 35790, 35790]) Position ids shape: torch.Size([1, 35790]) Input IDs shape: torch.Size([1, 35790]) Labels shape: torch.Size([1, 35790]) Final batch size: 1, sequence length: 22328 Attention mask shape: torch.Size([1, 1, 22328, 22328]) Position ids shape: torch.Size([1, 22328]) Input IDs shape: torch.Size([1, 22328]) Labels shape: torch.Size([1, 22328]) Final batch size: 1, sequence length: 26465 Attention mask shape: torch.Size([1, 1, 26465, 26465]) Position ids shape: torch.Size([1, 26465]) Input IDs shape: torch.Size([1, 26465]) Labels shape: torch.Size([1, 26465]) Final batch size: 1, sequence length: 13986 Attention mask shape: torch.Size([1, 1, 13986, 13986]) Position ids shape: torch.Size([1, 13986]) Input IDs shape: torch.Size([1, 13986]) Labels shape: torch.Size([1, 13986]) Final batch size: 1, sequence length: 34874 Attention mask shape: torch.Size([1, 1, 34874, 34874]) Position ids shape: torch.Size([1, 34874]) Input IDs shape: torch.Size([1, 34874]) Labels shape: torch.Size([1, 34874]) Final batch size: 1, sequence length: 22014 Attention mask shape: torch.Size([1, 1, 22014, 22014]) Position ids shape: torch.Size([1, 22014]) Input IDs shape: torch.Size([1, 22014]) Labels shape: torch.Size([1, 22014]) Final batch size: 1, sequence length: 38617 Attention mask shape: torch.Size([1, 1, 38617, 38617]) Position ids shape: torch.Size([1, 38617]) Input IDs shape: torch.Size([1, 38617]) Labels shape: torch.Size([1, 38617]) Final batch size: 1, sequence length: 40717 Attention mask shape: torch.Size([1, 1, 40717, 40717]) Position ids shape: torch.Size([1, 40717]) Input IDs shape: torch.Size([1, 40717]) Labels shape: torch.Size([1, 40717]) Final batch size: 1, sequence length: 19204 Attention mask shape: torch.Size([1, 1, 19204, 19204]) Position ids shape: torch.Size([1, 19204]) Input IDs shape: torch.Size([1, 19204]) Labels shape: torch.Size([1, 19204]) Final batch size: 1, sequence length: 35803 Attention mask shape: torch.Size([1, 1, 35803, 35803]) Position ids shape: torch.Size([1, 35803]) Input IDs shape: torch.Size([1, 35803]) Labels shape: torch.Size([1, 35803]) Final batch size: 1, sequence length: 39150 Attention mask shape: torch.Size([1, 1, 39150, 39150]) Position ids shape: torch.Size([1, 39150]) Input IDs shape: torch.Size([1, 39150]) Labels shape: torch.Size([1, 39150]) Final batch size: 1, sequence length: 36794 Attention mask shape: torch.Size([1, 1, 36794, 36794]) Position ids shape: torch.Size([1, 36794]) Input IDs shape: torch.Size([1, 36794]) Labels shape: torch.Size([1, 36794]) Final batch size: 1, sequence length: 37039 Final batch size: 1, sequence length: 23245 Attention mask shape: torch.Size([1, 1, 37039, 37039]) Position ids shape: torch.Size([1, 37039]) Input IDs shape: torch.Size([1, 37039]) Labels shape: torch.Size([1, 37039]) Attention mask shape: torch.Size([1, 1, 23245, 23245]) Position ids shape: torch.Size([1, 23245]) Input IDs shape: torch.Size([1, 23245]) Labels shape: torch.Size([1, 23245]) Final batch size: 1, sequence length: 35868 Attention mask shape: torch.Size([1, 1, 35868, 35868]) Position ids shape: torch.Size([1, 35868]) Input IDs shape: torch.Size([1, 35868]) Labels shape: torch.Size([1, 35868]) Final batch size: 1, sequence length: 35999 Attention mask shape: torch.Size([1, 1, 35999, 35999]) Position ids shape: torch.Size([1, 35999]) Input IDs shape: torch.Size([1, 35999]) Labels shape: torch.Size([1, 35999]) Final batch size: 1, sequence length: 19744 Attention mask shape: torch.Size([1, 1, 19744, 19744]) Position ids shape: torch.Size([1, 19744]) Input IDs shape: torch.Size([1, 19744]) Labels shape: torch.Size([1, 19744]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31506 Attention mask shape: torch.Size([1, 1, 31506, 31506]) Position ids shape: torch.Size([1, 31506]) Input IDs shape: torch.Size([1, 31506]) Labels shape: torch.Size([1, 31506]) Final batch size: 1, sequence length: 28002 Attention mask shape: torch.Size([1, 1, 28002, 28002]) Position ids shape: torch.Size([1, 28002]) Input IDs shape: torch.Size([1, 28002]) Labels shape: torch.Size([1, 28002]) Final batch size: 1, sequence length: 32974 Attention mask shape: torch.Size([1, 1, 32974, 32974]) Position ids shape: torch.Size([1, 32974]) Input IDs shape: torch.Size([1, 32974]) Labels shape: torch.Size([1, 32974]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40065 Attention mask shape: torch.Size([1, 1, 40065, 40065]) Position ids shape: torch.Size([1, 40065]) Input IDs shape: torch.Size([1, 40065]) Labels shape: torch.Size([1, 40065]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38919 Attention mask shape: torch.Size([1, 1, 38919, 38919]) Position ids shape: torch.Size([1, 38919]) Input IDs shape: torch.Size([1, 38919]) Labels shape: torch.Size([1, 38919]) Final batch size: 1, sequence length: 39919 Attention mask shape: torch.Size([1, 1, 39919, 39919]) Position ids shape: torch.Size([1, 39919]) Input IDs shape: torch.Size([1, 39919]) Labels shape: torch.Size([1, 39919]) Final batch size: 1, sequence length: 28781 Attention mask shape: torch.Size([1, 1, 28781, 28781]) Position ids shape: torch.Size([1, 28781]) Input IDs shape: torch.Size([1, 28781]) Labels shape: torch.Size([1, 28781]) Final batch size: 1, sequence length: 22282 Attention mask shape: torch.Size([1, 1, 22282, 22282]) Position ids shape: torch.Size([1, 22282]) Input IDs shape: torch.Size([1, 22282]) Labels shape: torch.Size([1, 22282]) Final batch size: 1, sequence length: 11644 Attention mask shape: torch.Size([1, 1, 11644, 11644]) Position ids shape: torch.Size([1, 11644]) Input IDs shape: torch.Size([1, 11644]) Labels shape: torch.Size([1, 11644]) Final batch size: 1, sequence length: 19846 Attention mask shape: torch.Size([1, 1, 19846, 19846]) Position ids shape: torch.Size([1, 19846]) Input IDs shape: torch.Size([1, 19846]) Labels shape: torch.Size([1, 19846]) Final batch size: 1, sequence length: 31340 Attention mask shape: torch.Size([1, 1, 31340, 31340]) Position ids shape: torch.Size([1, 31340]) Input IDs shape: torch.Size([1, 31340]) Labels shape: torch.Size([1, 31340]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31359 Attention mask shape: torch.Size([1, 1, 31359, 31359]) Position ids shape: torch.Size([1, 31359]) Input IDs shape: torch.Size([1, 31359]) Labels shape: torch.Size([1, 31359]) Final batch size: 1, sequence length: 18379 Attention mask shape: torch.Size([1, 1, 18379, 18379]) Position ids shape: torch.Size([1, 18379]) Input IDs shape: torch.Size([1, 18379]) Labels shape: torch.Size([1, 18379]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40874 Attention mask shape: torch.Size([1, 1, 40874, 40874]) Position ids shape: torch.Size([1, 40874]) Input IDs shape: torch.Size([1, 40874]) Labels shape: torch.Size([1, 40874]) Final batch size: 1, sequence length: 7364 Attention mask shape: torch.Size([1, 1, 7364, 7364]) Position ids shape: torch.Size([1, 7364]) Input IDs shape: torch.Size([1, 7364]) Labels shape: torch.Size([1, 7364]) Final batch size: 1, sequence length: 12418 Attention mask shape: torch.Size([1, 1, 12418, 12418]) Position ids shape: torch.Size([1, 12418]) Input IDs shape: torch.Size([1, 12418]) Labels shape: torch.Size([1, 12418]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 9309 Attention mask shape: torch.Size([1, 1, 9309, 9309]) Position ids shape: torch.Size([1, 9309]) Input IDs shape: torch.Size([1, 9309]) Labels shape: torch.Size([1, 9309]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 25560 Attention mask shape: torch.Size([1, 1, 25560, 25560]) Position ids shape: torch.Size([1, 25560]) Input IDs shape: torch.Size([1, 25560]) Labels shape: torch.Size([1, 25560]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26221 Attention mask shape: torch.Size([1, 1, 26221, 26221]) Position ids shape: torch.Size([1, 26221]) Input IDs shape: torch.Size([1, 26221]) Labels shape: torch.Size([1, 26221]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25923 Attention mask shape: torch.Size([1, 1, 25923, 25923]) Position ids shape: torch.Size([1, 25923]) Input IDs shape: torch.Size([1, 25923]) Labels shape: torch.Size([1, 25923]) Final batch size: 1, sequence length: 18654 Attention mask shape: torch.Size([1, 1, 18654, 18654]) Position ids shape: torch.Size([1, 18654]) Input IDs shape: torch.Size([1, 18654]) Labels shape: torch.Size([1, 18654]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17224 Attention mask shape: torch.Size([1, 1, 17224, 17224]) Position ids shape: torch.Size([1, 17224]) Input IDs shape: torch.Size([1, 17224]) Labels shape: torch.Size([1, 17224]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25820 Attention mask shape: torch.Size([1, 1, 25820, 25820]) Position ids shape: torch.Size([1, 25820]) Input IDs shape: torch.Size([1, 25820]) Labels shape: torch.Size([1, 25820]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30709 Attention mask shape: torch.Size([1, 1, 30709, 30709]) Position ids shape: torch.Size([1, 30709]) Input IDs shape: torch.Size([1, 30709]) Labels shape: torch.Size([1, 30709]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 29526 Attention mask shape: torch.Size([1, 1, 29526, 29526]) Position ids shape: torch.Size([1, 29526]) Input IDs shape: torch.Size([1, 29526]) Labels shape: torch.Size([1, 29526]) {'loss': 0.2698, 'grad_norm': 0.30723149472614836, 'learning_rate': 4.217827674798845e-06, 'num_tokens': -inf, 'epoch': 4.75} Final batch size: 1, sequence length: 7461 Attention mask shape: torch.Size([1, 1, 7461, 7461]) Position ids shape: torch.Size([1, 7461]) Input IDs shape: torch.Size([1, 7461]) Labels shape: torch.Size([1, 7461]) Final batch size: 1, sequence length: 12279 Attention mask shape: torch.Size([1, 1, 12279, 12279]) Position ids shape: torch.Size([1, 12279]) Input IDs shape: torch.Size([1, 12279]) Labels shape: torch.Size([1, 12279]) Final batch size: 1, sequence length: 10826 Attention mask shape: torch.Size([1, 1, 10826, 10826]) Position ids shape: torch.Size([1, 10826]) Input IDs shape: torch.Size([1, 10826]) Labels shape: torch.Size([1, 10826]) Final batch size: 1, sequence length: 9858 Attention mask shape: torch.Size([1, 1, 9858, 9858]) Position ids shape: torch.Size([1, 9858]) Input IDs shape: torch.Size([1, 9858]) Labels shape: torch.Size([1, 9858]) Final batch size: 1, sequence length: 12096 Attention mask shape: torch.Size([1, 1, 12096, 12096]) Position ids shape: torch.Size([1, 12096]) Input IDs shape: torch.Size([1, 12096]) Labels shape: torch.Size([1, 12096]) Final batch size: 1, sequence length: 12960 Attention mask shape: torch.Size([1, 1, 12960, 12960]) Position ids shape: torch.Size([1, 12960]) Input IDs shape: torch.Size([1, 12960]) Labels shape: torch.Size([1, 12960]) Final batch size: 1, sequence length: 13427 Attention mask shape: torch.Size([1, 1, 13427, 13427]) Position ids shape: torch.Size([1, 13427]) Input IDs shape: torch.Size([1, 13427]) Labels shape: torch.Size([1, 13427]) Final batch size: 1, sequence length: 12622 Attention mask shape: torch.Size([1, 1, 12622, 12622]) Position ids shape: torch.Size([1, 12622]) Input IDs shape: torch.Size([1, 12622]) Labels shape: torch.Size([1, 12622]) Final batch size: 1, sequence length: 15933 Attention mask shape: torch.Size([1, 1, 15933, 15933]) Position ids shape: torch.Size([1, 15933]) Input IDs shape: torch.Size([1, 15933]) Labels shape: torch.Size([1, 15933]) Final batch size: 1, sequence length: 13550 Attention mask shape: torch.Size([1, 1, 13550, 13550]) Position ids shape: torch.Size([1, 13550]) Input IDs shape: torch.Size([1, 13550]) Labels shape: torch.Size([1, 13550]) Final batch size: 1, sequence length: 15673 Attention mask shape: torch.Size([1, 1, 15673, 15673]) Position ids shape: torch.Size([1, 15673]) Input IDs shape: torch.Size([1, 15673]) Labels shape: torch.Size([1, 15673]) Final batch size: 1, sequence length: 17861 Attention mask shape: torch.Size([1, 1, 17861, 17861]) Position ids shape: torch.Size([1, 17861]) Input IDs shape: torch.Size([1, 17861]) Labels shape: torch.Size([1, 17861]) Final batch size: 1, sequence length: 13789 Attention mask shape: torch.Size([1, 1, 13789, 13789]) Position ids shape: torch.Size([1, 13789]) Input IDs shape: torch.Size([1, 13789]) Labels shape: torch.Size([1, 13789]) Final batch size: 1, sequence length: 18214 Attention mask shape: torch.Size([1, 1, 18214, 18214]) Position ids shape: torch.Size([1, 18214]) Input IDs shape: torch.Size([1, 18214]) Labels shape: torch.Size([1, 18214]) Final batch size: 1, sequence length: 17245 Attention mask shape: torch.Size([1, 1, 17245, 17245]) Position ids shape: torch.Size([1, 17245]) Input IDs shape: torch.Size([1, 17245]) Labels shape: torch.Size([1, 17245]) Final batch size: 1, sequence length: 16450 Attention mask shape: torch.Size([1, 1, 16450, 16450]) Position ids shape: torch.Size([1, 16450]) Input IDs shape: torch.Size([1, 16450]) Labels shape: torch.Size([1, 16450]) Final batch size: 1, sequence length: 20292 Attention mask shape: torch.Size([1, 1, 20292, 20292]) Position ids shape: torch.Size([1, 20292]) Input IDs shape: torch.Size([1, 20292]) Labels shape: torch.Size([1, 20292]) Final batch size: 1, sequence length: 17934 Attention mask shape: torch.Size([1, 1, 17934, 17934]) Position ids shape: torch.Size([1, 17934]) Input IDs shape: torch.Size([1, 17934]) Labels shape: torch.Size([1, 17934]) Final batch size: 1, sequence length: 19899 Attention mask shape: torch.Size([1, 1, 19899, 19899]) Position ids shape: torch.Size([1, 19899]) Input IDs shape: torch.Size([1, 19899]) Labels shape: torch.Size([1, 19899]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 12728 Attention mask shape: torch.Size([1, 1, 12728, 12728]) Position ids shape: torch.Size([1, 12728]) Input IDs shape: torch.Size([1, 12728]) Labels shape: torch.Size([1, 12728]) Final batch size: 1, sequence length: 16927 Attention mask shape: torch.Size([1, 1, 16927, 16927]) Position ids shape: torch.Size([1, 16927]) Input IDs shape: torch.Size([1, 16927]) Labels shape: torch.Size([1, 16927]) Final batch size: 1, sequence length: 20065 Attention mask shape: torch.Size([1, 1, 20065, 20065]) Position ids shape: torch.Size([1, 20065]) Input IDs shape: torch.Size([1, 20065]) Labels shape: torch.Size([1, 20065]) Final batch size: 1, sequence length: 23004 Attention mask shape: torch.Size([1, 1, 23004, 23004]) Position ids shape: torch.Size([1, 23004]) Input IDs shape: torch.Size([1, 23004]) Labels shape: torch.Size([1, 23004]) Final batch size: 1, sequence length: 16036 Attention mask shape: torch.Size([1, 1, 16036, 16036]) Position ids shape: torch.Size([1, 16036]) Input IDs shape: torch.Size([1, 16036]) Labels shape: torch.Size([1, 16036]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 19187 Attention mask shape: torch.Size([1, 1, 19187, 19187]) Position ids shape: torch.Size([1, 19187]) Input IDs shape: torch.Size([1, 19187]) Labels shape: torch.Size([1, 19187]) Final batch size: 1, sequence length: 21091 Attention mask shape: torch.Size([1, 1, 21091, 21091]) Position ids shape: torch.Size([1, 21091]) Input IDs shape: torch.Size([1, 21091]) Labels shape: torch.Size([1, 21091]) Final batch size: 1, sequence length: 21675 Attention mask shape: torch.Size([1, 1, 21675, 21675]) Position ids shape: torch.Size([1, 21675]) Input IDs shape: torch.Size([1, 21675]) Labels shape: torch.Size([1, 21675]) Final batch size: 1, sequence length: 25435 Attention mask shape: torch.Size([1, 1, 25435, 25435]) Position ids shape: torch.Size([1, 25435]) Input IDs shape: torch.Size([1, 25435]) Labels shape: torch.Size([1, 25435]) Final batch size: 1, sequence length: 22775 Attention mask shape: torch.Size([1, 1, 22775, 22775]) Position ids shape: torch.Size([1, 22775]) Input IDs shape: torch.Size([1, 22775]) Labels shape: torch.Size([1, 22775]) Final batch size: 1, sequence length: 24965 Attention mask shape: torch.Size([1, 1, 24965, 24965]) Position ids shape: torch.Size([1, 24965]) Input IDs shape: torch.Size([1, 24965]) Labels shape: torch.Size([1, 24965]) Final batch size: 1, sequence length: 25176 Attention mask shape: torch.Size([1, 1, 25176, 25176]) Position ids shape: torch.Size([1, 25176]) Input IDs shape: torch.Size([1, 25176]) Labels shape: torch.Size([1, 25176]) Final batch size: 1, sequence length: 21586 Attention mask shape: torch.Size([1, 1, 21586, 21586]) Position ids shape: torch.Size([1, 21586]) Input IDs shape: torch.Size([1, 21586]) Labels shape: torch.Size([1, 21586]) Final batch size: 1, sequence length: 26316 Attention mask shape: torch.Size([1, 1, 26316, 26316]) Position ids shape: torch.Size([1, 26316]) Input IDs shape: torch.Size([1, 26316]) Labels shape: torch.Size([1, 26316]) Final batch size: 1, sequence length: 13946 Attention mask shape: torch.Size([1, 1, 13946, 13946]) Position ids shape: torch.Size([1, 13946]) Input IDs shape: torch.Size([1, 13946]) Labels shape: torch.Size([1, 13946]) Final batch size: 1, sequence length: 25600 Attention mask shape: torch.Size([1, 1, 25600, 25600]) Position ids shape: torch.Size([1, 25600]) Input IDs shape: torch.Size([1, 25600]) Labels shape: torch.Size([1, 25600]) Final batch size: 1, sequence length: 23896 Attention mask shape: torch.Size([1, 1, 23896, 23896]) Position ids shape: torch.Size([1, 23896]) Input IDs shape: torch.Size([1, 23896]) Labels shape: torch.Size([1, 23896]) Final batch size: 1, sequence length: 22778 Attention mask shape: torch.Size([1, 1, 22778, 22778]) Position ids shape: torch.Size([1, 22778]) Input IDs shape: torch.Size([1, 22778]) Labels shape: torch.Size([1, 22778]) Final batch size: 1, sequence length: 24808 Attention mask shape: torch.Size([1, 1, 24808, 24808]) Position ids shape: torch.Size([1, 24808]) Input IDs shape: torch.Size([1, 24808]) Labels shape: torch.Size([1, 24808]) Final batch size: 1, sequence length: 24956 Attention mask shape: torch.Size([1, 1, 24956, 24956]) Position ids shape: torch.Size([1, 24956]) Input IDs shape: torch.Size([1, 24956]) Labels shape: torch.Size([1, 24956]) Final batch size: 1, sequence length: 25325 Attention mask shape: torch.Size([1, 1, 25325, 25325]) Position ids shape: torch.Size([1, 25325]) Input IDs shape: torch.Size([1, 25325]) Labels shape: torch.Size([1, 25325]) Final batch size: 1, sequence length: 26247 Attention mask shape: torch.Size([1, 1, 26247, 26247]) Position ids shape: torch.Size([1, 26247]) Input IDs shape: torch.Size([1, 26247]) Labels shape: torch.Size([1, 26247]) Final batch size: 1, sequence length: 24633 Attention mask shape: torch.Size([1, 1, 24633, 24633]) Position ids shape: torch.Size([1, 24633]) Input IDs shape: torch.Size([1, 24633]) Labels shape: torch.Size([1, 24633]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 28926 Attention mask shape: torch.Size([1, 1, 28926, 28926]) Position ids shape: torch.Size([1, 28926]) Input IDs shape: torch.Size([1, 28926]) Labels shape: torch.Size([1, 28926]) Final batch size: 1, sequence length: 18832 Attention mask shape: torch.Size([1, 1, 18832, 18832]) Position ids shape: torch.Size([1, 18832]) Input IDs shape: torch.Size([1, 18832]) Labels shape: torch.Size([1, 18832]) Final batch size: 1, sequence length: 3010 Attention mask shape: torch.Size([1, 1, 3010, 3010]) Position ids shape: torch.Size([1, 3010]) Input IDs shape: torch.Size([1, 3010]) Labels shape: torch.Size([1, 3010]) Final batch size: 1, sequence length: 24927 Attention mask shape: torch.Size([1, 1, 24927, 24927]) Position ids shape: torch.Size([1, 24927]) Input IDs shape: torch.Size([1, 24927]) Labels shape: torch.Size([1, 24927]) Final batch size: 1, sequence length: 25035 Attention mask shape: torch.Size([1, 1, 25035, 25035]) Position ids shape: torch.Size([1, 25035]) Input IDs shape: torch.Size([1, 25035]) Labels shape: torch.Size([1, 25035]) Final batch size: 1, sequence length: 14212 Attention mask shape: torch.Size([1, 1, 14212, 14212]) Position ids shape: torch.Size([1, 14212]) Input IDs shape: torch.Size([1, 14212]) Labels shape: torch.Size([1, 14212]) Final batch size: 1, sequence length: 20582 Attention mask shape: torch.Size([1, 1, 20582, 20582]) Position ids shape: torch.Size([1, 20582]) Input IDs shape: torch.Size([1, 20582]) Labels shape: torch.Size([1, 20582]) Final batch size: 1, sequence length: 26500 Attention mask shape: torch.Size([1, 1, 26500, 26500]) Position ids shape: torch.Size([1, 26500]) Input IDs shape: torch.Size([1, 26500]) Labels shape: torch.Size([1, 26500]) Final batch size: 1, sequence length: 6978 Attention mask shape: torch.Size([1, 1, 6978, 6978]) Position ids shape: torch.Size([1, 6978]) Input IDs shape: torch.Size([1, 6978]) Labels shape: torch.Size([1, 6978]) Final batch size: 1, sequence length: 28552 Attention mask shape: torch.Size([1, 1, 28552, 28552]) Position ids shape: torch.Size([1, 28552]) Input IDs shape: torch.Size([1, 28552]) Labels shape: torch.Size([1, 28552]) Final batch size: 1, sequence length: 27222 Attention mask shape: torch.Size([1, 1, 27222, 27222]) Position ids shape: torch.Size([1, 27222]) Input IDs shape: torch.Size([1, 27222]) Labels shape: torch.Size([1, 27222]) Final batch size: 1, sequence length: 17763 Attention mask shape: torch.Size([1, 1, 17763, 17763]) Position ids shape: torch.Size([1, 17763]) Input IDs shape: torch.Size([1, 17763]) Labels shape: torch.Size([1, 17763]) Final batch size: 1, sequence length: 16714 Attention mask shape: torch.Size([1, 1, 16714, 16714]) Position ids shape: torch.Size([1, 16714]) Input IDs shape: torch.Size([1, 16714]) Labels shape: torch.Size([1, 16714]) Final batch size: 1, sequence length: 13453 Attention mask shape: torch.Size([1, 1, 13453, 13453]) Position ids shape: torch.Size([1, 13453]) Input IDs shape: torch.Size([1, 13453]) Labels shape: torch.Size([1, 13453]) Final batch size: 1, sequence length: 31208 Attention mask shape: torch.Size([1, 1, 31208, 31208]) Position ids shape: torch.Size([1, 31208]) Input IDs shape: torch.Size([1, 31208]) Labels shape: torch.Size([1, 31208]) Final batch size: 1, sequence length: 12426 Attention mask shape: torch.Size([1, 1, 12426, 12426]) Position ids shape: torch.Size([1, 12426]) Input IDs shape: torch.Size([1, 12426]) Labels shape: torch.Size([1, 12426]) Final batch size: 1, sequence length: 28644 Attention mask shape: torch.Size([1, 1, 28644, 28644]) Position ids shape: torch.Size([1, 28644]) Input IDs shape: torch.Size([1, 28644]) Labels shape: torch.Size([1, 28644]) Final batch size: 1, sequence length: 25979 Attention mask shape: torch.Size([1, 1, 25979, 25979]) Position ids shape: torch.Size([1, 25979]) Input IDs shape: torch.Size([1, 25979]) Labels shape: torch.Size([1, 25979]) Final batch size: 1, sequence length: 16571 Attention mask shape: torch.Size([1, 1, 16571, 16571]) Position ids shape: torch.Size([1, 16571]) Input IDs shape: torch.Size([1, 16571]) Labels shape: torch.Size([1, 16571]) Final batch size: 1, sequence length: 17049 Attention mask shape: torch.Size([1, 1, 17049, 17049]) Position ids shape: torch.Size([1, 17049]) Input IDs shape: torch.Size([1, 17049]) Labels shape: torch.Size([1, 17049]) Final batch size: 1, sequence length: 32513 Attention mask shape: torch.Size([1, 1, 32513, 32513]) Position ids shape: torch.Size([1, 32513]) Input IDs shape: torch.Size([1, 32513]) Labels shape: torch.Size([1, 32513]) Final batch size: 1, sequence length: 32100 Attention mask shape: torch.Size([1, 1, 32100, 32100]) Position ids shape: torch.Size([1, 32100]) Input IDs shape: torch.Size([1, 32100]) Labels shape: torch.Size([1, 32100]) Final batch size: 1, sequence length: 35058 Attention mask shape: torch.Size([1, 1, 35058, 35058]) Position ids shape: torch.Size([1, 35058]) Input IDs shape: torch.Size([1, 35058]) Labels shape: torch.Size([1, 35058]) Final batch size: 1, sequence length: 19953 Attention mask shape: torch.Size([1, 1, 19953, 19953]) Position ids shape: torch.Size([1, 19953]) Input IDs shape: torch.Size([1, 19953]) Labels shape: torch.Size([1, 19953]) Final batch size: 1, sequence length: 30635 Attention mask shape: torch.Size([1, 1, 30635, 30635]) Position ids shape: torch.Size([1, 30635]) Input IDs shape: torch.Size([1, 30635]) Labels shape: torch.Size([1, 30635]) Final batch size: 1, sequence length: 32101 Attention mask shape: torch.Size([1, 1, 32101, 32101]) Position ids shape: torch.Size([1, 32101]) Input IDs shape: torch.Size([1, 32101]) Labels shape: torch.Size([1, 32101]) Final batch size: 1, sequence length: 31662 Attention mask shape: torch.Size([1, 1, 31662, 31662]) Position ids shape: torch.Size([1, 31662]) Input IDs shape: torch.Size([1, 31662]) Labels shape: torch.Size([1, 31662]) Final batch size: 1, sequence length: 37381 Attention mask shape: torch.Size([1, 1, 37381, 37381]) Position ids shape: torch.Size([1, 37381]) Input IDs shape: torch.Size([1, 37381]) Labels shape: torch.Size([1, 37381]) Final batch size: 1, sequence length: 24676 Attention mask shape: torch.Size([1, 1, 24676, 24676]) Position ids shape: torch.Size([1, 24676]) Input IDs shape: torch.Size([1, 24676]) Labels shape: torch.Size([1, 24676]) Final batch size: 1, sequence length: 37720 Attention mask shape: torch.Size([1, 1, 37720, 37720]) Position ids shape: torch.Size([1, 37720]) Input IDs shape: torch.Size([1, 37720]) Labels shape: torch.Size([1, 37720]) Final batch size: 1, sequence length: 35014 Attention mask shape: torch.Size([1, 1, 35014, 35014]) Position ids shape: torch.Size([1, 35014]) Input IDs shape: torch.Size([1, 35014]) Labels shape: torch.Size([1, 35014]) Final batch size: 1, sequence length: 21290 Attention mask shape: torch.Size([1, 1, 21290, 21290]) Position ids shape: torch.Size([1, 21290]) Input IDs shape: torch.Size([1, 21290]) Labels shape: torch.Size([1, 21290]) Final batch size: 1, sequence length: 37391 Attention mask shape: torch.Size([1, 1, 37391, 37391]) Position ids shape: torch.Size([1, 37391]) Input IDs shape: torch.Size([1, 37391]) Labels shape: torch.Size([1, 37391]) Final batch size: 1, sequence length: 25147 Attention mask shape: torch.Size([1, 1, 25147, 25147]) Position ids shape: torch.Size([1, 25147]) Input IDs shape: torch.Size([1, 25147]) Labels shape: torch.Size([1, 25147]) Final batch size: 1, sequence length: 33621 Attention mask shape: torch.Size([1, 1, 33621, 33621]) Position ids shape: torch.Size([1, 33621]) Input IDs shape: torch.Size([1, 33621]) Labels shape: torch.Size([1, 33621]) Final batch size: 1, sequence length: 28377 Attention mask shape: torch.Size([1, 1, 28377, 28377]) Position ids shape: torch.Size([1, 28377]) Input IDs shape: torch.Size([1, 28377]) Labels shape: torch.Size([1, 28377]) Final batch size: 1, sequence length: 39955 Attention mask shape: torch.Size([1, 1, 39955, 39955]) Position ids shape: torch.Size([1, 39955]) Input IDs shape: torch.Size([1, 39955]) Labels shape: torch.Size([1, 39955]) Final batch size: 1, sequence length: 39474 Attention mask shape: torch.Size([1, 1, 39474, 39474]) Position ids shape: torch.Size([1, 39474]) Input IDs shape: torch.Size([1, 39474]) Labels shape: torch.Size([1, 39474]) Final batch size: 1, sequence length: 25774 Attention mask shape: torch.Size([1, 1, 25774, 25774]) Position ids shape: torch.Size([1, 25774]) Input IDs shape: torch.Size([1, 25774]) Labels shape: torch.Size([1, 25774]) Final batch size: 1, sequence length: 38391 Attention mask shape: torch.Size([1, 1, 38391, 38391]) Position ids shape: torch.Size([1, 38391]) Input IDs shape: torch.Size([1, 38391]) Labels shape: torch.Size([1, 38391]) Final batch size: 1, sequence length: 25219 Attention mask shape: torch.Size([1, 1, 25219, 25219]) Position ids shape: torch.Size([1, 25219]) Input IDs shape: torch.Size([1, 25219]) Labels shape: torch.Size([1, 25219]) Final batch size: 1, sequence length: 36047 Attention mask shape: torch.Size([1, 1, 36047, 36047]) Position ids shape: torch.Size([1, 36047]) Input IDs shape: torch.Size([1, 36047]) Labels shape: torch.Size([1, 36047]) Final batch size: 1, sequence length: 26789 Attention mask shape: torch.Size([1, 1, 26789, 26789]) Position ids shape: torch.Size([1, 26789]) Input IDs shape: torch.Size([1, 26789]) Labels shape: torch.Size([1, 26789]) Final batch size: 1, sequence length: 16816 Attention mask shape: torch.Size([1, 1, 16816, 16816]) Position ids shape: torch.Size([1, 16816]) Input IDs shape: torch.Size([1, 16816]) Labels shape: torch.Size([1, 16816]) Final batch size: 1, sequence length: 30259 Attention mask shape: torch.Size([1, 1, 30259, 30259]) Position ids shape: torch.Size([1, 30259]) Input IDs shape: torch.Size([1, 30259]) Labels shape: torch.Size([1, 30259]) Final batch size: 1, sequence length: 26387 Attention mask shape: torch.Size([1, 1, 26387, 26387]) Position ids shape: torch.Size([1, 26387]) Input IDs shape: torch.Size([1, 26387]) Labels shape: torch.Size([1, 26387]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 26454 Attention mask shape: torch.Size([1, 1, 26454, 26454]) Position ids shape: torch.Size([1, 26454]) Input IDs shape: torch.Size([1, 26454]) Labels shape: torch.Size([1, 26454]) Final batch size: 1, sequence length: 22710 Attention mask shape: torch.Size([1, 1, 22710, 22710]) Position ids shape: torch.Size([1, 22710]) Input IDs shape: torch.Size([1, 22710]) Labels shape: torch.Size([1, 22710]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38577 Attention mask shape: torch.Size([1, 1, 38577, 38577]) Position ids shape: torch.Size([1, 38577]) Input IDs shape: torch.Size([1, 38577]) Labels shape: torch.Size([1, 38577]) Final batch size: 1, sequence length: 21321 Attention mask shape: torch.Size([1, 1, 21321, 21321]) Position ids shape: torch.Size([1, 21321]) Input IDs shape: torch.Size([1, 21321]) Labels shape: torch.Size([1, 21321]) Final batch size: 1, sequence length: 19441 Attention mask shape: torch.Size([1, 1, 19441, 19441]) Position ids shape: torch.Size([1, 19441]) Input IDs shape: torch.Size([1, 19441]) Labels shape: torch.Size([1, 19441]) Final batch size: 1, sequence length: 31447 Attention mask shape: torch.Size([1, 1, 31447, 31447]) Position ids shape: torch.Size([1, 31447]) Input IDs shape: torch.Size([1, 31447]) Labels shape: torch.Size([1, 31447]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40590 Attention mask shape: torch.Size([1, 1, 40590, 40590]) Position ids shape: torch.Size([1, 40590]) Input IDs shape: torch.Size([1, 40590]) Labels shape: torch.Size([1, 40590]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 35261 Attention mask shape: torch.Size([1, 1, 35261, 35261]) Position ids shape: torch.Size([1, 35261]) Input IDs shape: torch.Size([1, 35261]) Labels shape: torch.Size([1, 35261]) Final batch size: 1, sequence length: 28448 Attention mask shape: torch.Size([1, 1, 28448, 28448]) Position ids shape: torch.Size([1, 28448]) Input IDs shape: torch.Size([1, 28448]) Labels shape: torch.Size([1, 28448]) Final batch size: 1, sequence length: 32079 Attention mask shape: torch.Size([1, 1, 32079, 32079]) Position ids shape: torch.Size([1, 32079]) Input IDs shape: torch.Size([1, 32079]) Labels shape: torch.Size([1, 32079]) Final batch size: 1, sequence length: 27371 Attention mask shape: torch.Size([1, 1, 27371, 27371]) Position ids shape: torch.Size([1, 27371]) Input IDs shape: torch.Size([1, 27371]) Labels shape: torch.Size([1, 27371]) Final batch size: 1, sequence length: 32129 Attention mask shape: torch.Size([1, 1, 32129, 32129]) Position ids shape: torch.Size([1, 32129]) Input IDs shape: torch.Size([1, 32129]) Labels shape: torch.Size([1, 32129]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36356 Attention mask shape: torch.Size([1, 1, 36356, 36356]) Position ids shape: torch.Size([1, 36356]) Input IDs shape: torch.Size([1, 36356]) Labels shape: torch.Size([1, 36356]) Final batch size: 1, sequence length: 14136 Attention mask shape: torch.Size([1, 1, 14136, 14136]) Position ids shape: torch.Size([1, 14136]) Input IDs shape: torch.Size([1, 14136]) Labels shape: torch.Size([1, 14136]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24623 Attention mask shape: torch.Size([1, 1, 24623, 24623]) Position ids shape: torch.Size([1, 24623]) Input IDs shape: torch.Size([1, 24623]) Labels shape: torch.Size([1, 24623]) Final batch size: 1, sequence length: 33014 Attention mask shape: torch.Size([1, 1, 33014, 33014]) Position ids shape: torch.Size([1, 33014]) Input IDs shape: torch.Size([1, 33014]) Labels shape: torch.Size([1, 33014]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 27021 Attention mask shape: torch.Size([1, 1, 27021, 27021]) Position ids shape: torch.Size([1, 27021]) Input IDs shape: torch.Size([1, 27021]) Labels shape: torch.Size([1, 27021]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30712 Attention mask shape: torch.Size([1, 1, 30712, 30712]) Position ids shape: torch.Size([1, 30712]) Input IDs shape: torch.Size([1, 30712]) Labels shape: torch.Size([1, 30712]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19027 Attention mask shape: torch.Size([1, 1, 19027, 19027]) Position ids shape: torch.Size([1, 19027]) Input IDs shape: torch.Size([1, 19027]) Labels shape: torch.Size([1, 19027]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 23208 Attention mask shape: torch.Size([1, 1, 23208, 23208]) Position ids shape: torch.Size([1, 23208]) Input IDs shape: torch.Size([1, 23208]) Labels shape: torch.Size([1, 23208]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2655, 'grad_norm': 0.2950905115742646, 'learning_rate': 3.960441545911205e-06, 'num_tokens': -inf, 'epoch': 4.88} Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 10507 Attention mask shape: torch.Size([1, 1, 10507, 10507]) Position ids shape: torch.Size([1, 10507]) Input IDs shape: torch.Size([1, 10507]) Labels shape: torch.Size([1, 10507]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 26271 Attention mask shape: torch.Size([1, 1, 26271, 26271]) Position ids shape: torch.Size([1, 26271]) Input IDs shape: torch.Size([1, 26271]) Labels shape: torch.Size([1, 26271]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 31381 Attention mask shape: torch.Size([1, 1, 31381, 31381]) Position ids shape: torch.Size([1, 31381]) Input IDs shape: torch.Size([1, 31381]) Labels shape: torch.Size([1, 31381]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2575, 'grad_norm': 0.29575384671731464, 'learning_rate': 3.705904774487396e-06, 'num_tokens': -inf, 'epoch': 5.0} Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 6362 Attention mask shape: torch.Size([1, 1, 6362, 6362]) Position ids shape: torch.Size([1, 6362]) Input IDs shape: torch.Size([1, 6362]) Labels shape: torch.Size([1, 6362]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 10523 Attention mask shape: torch.Size([1, 1, 10523, 10523]) Position ids shape: torch.Size([1, 10523]) Input IDs shape: torch.Size([1, 10523]) Labels shape: torch.Size([1, 10523]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 12385 Attention mask shape: torch.Size([1, 1, 12385, 12385]) Position ids shape: torch.Size([1, 12385]) Input IDs shape: torch.Size([1, 12385]) Labels shape: torch.Size([1, 12385]) Final batch size: 1, sequence length: 18410 Attention mask shape: torch.Size([1, 1, 18410, 18410]) Position ids shape: torch.Size([1, 18410]) Input IDs shape: torch.Size([1, 18410]) Labels shape: torch.Size([1, 18410]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 18098 Attention mask shape: torch.Size([1, 1, 18098, 18098]) Position ids shape: torch.Size([1, 18098]) Input IDs shape: torch.Size([1, 18098]) Labels shape: torch.Size([1, 18098]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 24365 Attention mask shape: torch.Size([1, 1, 24365, 24365]) Position ids shape: torch.Size([1, 24365]) Input IDs shape: torch.Size([1, 24365]) Labels shape: torch.Size([1, 24365]) Final batch size: 1, sequence length: 17299 Attention mask shape: torch.Size([1, 1, 17299, 17299]) Position ids shape: torch.Size([1, 17299]) Input IDs shape: torch.Size([1, 17299]) Labels shape: torch.Size([1, 17299]) Final batch size: 1, sequence length: 9091 Attention mask shape: torch.Size([1, 1, 9091, 9091]) Position ids shape: torch.Size([1, 9091]) Input IDs shape: torch.Size([1, 9091]) Labels shape: torch.Size([1, 9091]) Final batch size: 1, sequence length: 18060 Attention mask shape: torch.Size([1, 1, 18060, 18060]) Position ids shape: torch.Size([1, 18060]) Input IDs shape: torch.Size([1, 18060]) Labels shape: torch.Size([1, 18060]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 12656 Attention mask shape: torch.Size([1, 1, 12656, 12656]) Position ids shape: torch.Size([1, 12656]) Input IDs shape: torch.Size([1, 12656]) Labels shape: torch.Size([1, 12656]) Final batch size: 1, sequence length: 18915 Attention mask shape: torch.Size([1, 1, 18915, 18915]) Position ids shape: torch.Size([1, 18915]) Input IDs shape: torch.Size([1, 18915]) Labels shape: torch.Size([1, 18915]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 11819 Attention mask shape: torch.Size([1, 1, 11819, 11819]) Position ids shape: torch.Size([1, 11819]) Input IDs shape: torch.Size([1, 11819]) Labels shape: torch.Size([1, 11819]) Final batch size: 1, sequence length: 26976 Attention mask shape: torch.Size([1, 1, 26976, 26976]) Position ids shape: torch.Size([1, 26976]) Input IDs shape: torch.Size([1, 26976]) Labels shape: torch.Size([1, 26976]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 6948 Attention mask shape: torch.Size([1, 1, 6948, 6948]) Position ids shape: torch.Size([1, 6948]) Input IDs shape: torch.Size([1, 6948]) Labels shape: torch.Size([1, 6948]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 28841 Attention mask shape: torch.Size([1, 1, 28841, 28841]) Position ids shape: torch.Size([1, 28841]) Input IDs shape: torch.Size([1, 28841]) Labels shape: torch.Size([1, 28841]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 29113 Attention mask shape: torch.Size([1, 1, 29113, 29113]) Position ids shape: torch.Size([1, 29113]) Input IDs shape: torch.Size([1, 29113]) Labels shape: torch.Size([1, 29113]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 27541 Attention mask shape: torch.Size([1, 1, 27541, 27541]) Position ids shape: torch.Size([1, 27541]) Input IDs shape: torch.Size([1, 27541]) Labels shape: torch.Size([1, 27541]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 24880 Attention mask shape: torch.Size([1, 1, 24880, 24880]) Position ids shape: torch.Size([1, 24880]) Input IDs shape: torch.Size([1, 24880]) Labels shape: torch.Size([1, 24880]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 21450 Attention mask shape: torch.Size([1, 1, 21450, 21450]) Position ids shape: torch.Size([1, 21450]) Input IDs shape: torch.Size([1, 21450]) Labels shape: torch.Size([1, 21450]) Final batch size: 1, sequence length: 13215 Attention mask shape: torch.Size([1, 1, 13215, 13215]) Position ids shape: torch.Size([1, 13215]) Input IDs shape: torch.Size([1, 13215]) Labels shape: torch.Size([1, 13215]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 25172 Attention mask shape: torch.Size([1, 1, 25172, 25172]) Position ids shape: torch.Size([1, 25172]) Input IDs shape: torch.Size([1, 25172]) Labels shape: torch.Size([1, 25172]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 21596 Attention mask shape: torch.Size([1, 1, 21596, 21596]) Position ids shape: torch.Size([1, 21596]) Input IDs shape: torch.Size([1, 21596]) Labels shape: torch.Size([1, 21596]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 7722 Attention mask shape: torch.Size([1, 1, 7722, 7722]) Position ids shape: torch.Size([1, 7722]) Input IDs shape: torch.Size([1, 7722]) Labels shape: torch.Size([1, 7722]) Final batch size: 1, sequence length: 21491 Attention mask shape: torch.Size([1, 1, 21491, 21491]) Position ids shape: torch.Size([1, 21491]) Input IDs shape: torch.Size([1, 21491]) Labels shape: torch.Size([1, 21491]) Final batch size: 1, sequence length: 20827 Attention mask shape: torch.Size([1, 1, 20827, 20827]) Position ids shape: torch.Size([1, 20827]) Input IDs shape: torch.Size([1, 20827]) Labels shape: torch.Size([1, 20827]) Final batch size: 1, sequence length: 29875 Attention mask shape: torch.Size([1, 1, 29875, 29875]) Position ids shape: torch.Size([1, 29875]) Input IDs shape: torch.Size([1, 29875]) Labels shape: torch.Size([1, 29875]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 20363 Attention mask shape: torch.Size([1, 1, 20363, 20363]) Position ids shape: torch.Size([1, 20363]) Input IDs shape: torch.Size([1, 20363]) Labels shape: torch.Size([1, 20363]) Final batch size: 1, sequence length: 25014 Attention mask shape: torch.Size([1, 1, 25014, 25014]) Position ids shape: torch.Size([1, 25014]) Input IDs shape: torch.Size([1, 25014]) Labels shape: torch.Size([1, 25014]) Final batch size: 1, sequence length: 30623 Attention mask shape: torch.Size([1, 1, 30623, 30623]) Position ids shape: torch.Size([1, 30623]) Input IDs shape: torch.Size([1, 30623]) Labels shape: torch.Size([1, 30623]) Final batch size: 1, sequence length: 14009 Attention mask shape: torch.Size([1, 1, 14009, 14009]) Position ids shape: torch.Size([1, 14009]) Input IDs shape: torch.Size([1, 14009]) Labels shape: torch.Size([1, 14009]) Final batch size: 1, sequence length: 17400 Attention mask shape: torch.Size([1, 1, 17400, 17400]) Position ids shape: torch.Size([1, 17400]) Input IDs shape: torch.Size([1, 17400]) Labels shape: torch.Size([1, 17400]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 19705 Attention mask shape: torch.Size([1, 1, 19705, 19705]) Position ids shape: torch.Size([1, 19705]) Input IDs shape: torch.Size([1, 19705]) Labels shape: torch.Size([1, 19705]) Final batch size: 1, sequence length: 28641 Attention mask shape: torch.Size([1, 1, 28641, 28641]) Position ids shape: torch.Size([1, 28641]) Input IDs shape: torch.Size([1, 28641]) Labels shape: torch.Size([1, 28641]) Final batch size: 1, sequence length: 27265 Attention mask shape: torch.Size([1, 1, 27265, 27265]) Position ids shape: torch.Size([1, 27265]) Input IDs shape: torch.Size([1, 27265]) Labels shape: torch.Size([1, 27265]) Final batch size: 1, sequence length: 33367 Attention mask shape: torch.Size([1, 1, 33367, 33367]) Position ids shape: torch.Size([1, 33367]) Input IDs shape: torch.Size([1, 33367]) Labels shape: torch.Size([1, 33367]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 22625 Attention mask shape: torch.Size([1, 1, 22625, 22625]) Position ids shape: torch.Size([1, 22625]) Input IDs shape: torch.Size([1, 22625]) Labels shape: torch.Size([1, 22625]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 24001 Attention mask shape: torch.Size([1, 1, 24001, 24001]) Position ids shape: torch.Size([1, 24001]) Input IDs shape: torch.Size([1, 24001]) Labels shape: torch.Size([1, 24001]) Final batch size: 1, sequence length: 22098 Attention mask shape: torch.Size([1, 1, 22098, 22098]) Position ids shape: torch.Size([1, 22098]) Input IDs shape: torch.Size([1, 22098]) Labels shape: torch.Size([1, 22098]) Final batch size: 1, sequence length: 27243 Attention mask shape: torch.Size([1, 1, 27243, 27243]) Position ids shape: torch.Size([1, 27243]) Input IDs shape: torch.Size([1, 27243]) Labels shape: torch.Size([1, 27243]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 15656 Attention mask shape: torch.Size([1, 1, 15656, 15656]) Position ids shape: torch.Size([1, 15656]) Input IDs shape: torch.Size([1, 15656]) Labels shape: torch.Size([1, 15656]) Final batch size: 1, sequence length: 26306 Attention mask shape: torch.Size([1, 1, 26306, 26306]) Position ids shape: torch.Size([1, 26306]) Input IDs shape: torch.Size([1, 26306]) Labels shape: torch.Size([1, 26306]) Final batch size: 1, sequence length: 18606 Attention mask shape: torch.Size([1, 1, 18606, 18606]) Position ids shape: torch.Size([1, 18606]) Input IDs shape: torch.Size([1, 18606]) Labels shape: torch.Size([1, 18606]) Final batch size: 1, sequence length: 21348 Attention mask shape: torch.Size([1, 1, 21348, 21348]) Position ids shape: torch.Size([1, 21348]) Input IDs shape: torch.Size([1, 21348]) Labels shape: torch.Size([1, 21348]) Final batch size: 1, sequence length: 19586 Attention mask shape: torch.Size([1, 1, 19586, 19586]) Position ids shape: torch.Size([1, 19586]) Input IDs shape: torch.Size([1, 19586]) Labels shape: torch.Size([1, 19586]) Final batch size: 1, sequence length: 30066 Attention mask shape: torch.Size([1, 1, 30066, 30066]) Position ids shape: torch.Size([1, 30066]) Input IDs shape: torch.Size([1, 30066]) Labels shape: torch.Size([1, 30066]) Final batch size: 1, sequence length: 30101 Attention mask shape: torch.Size([1, 1, 30101, 30101]) Position ids shape: torch.Size([1, 30101]) Input IDs shape: torch.Size([1, 30101]) Labels shape: torch.Size([1, 30101]) Final batch size: 1, sequence length: 17778 Attention mask shape: torch.Size([1, 1, 17778, 17778]) Position ids shape: torch.Size([1, 17778]) Input IDs shape: torch.Size([1, 17778]) Labels shape: torch.Size([1, 17778]) Final batch size: 1, sequence length: 27441 Attention mask shape: torch.Size([1, 1, 27441, 27441]) Position ids shape: torch.Size([1, 27441]) Input IDs shape: torch.Size([1, 27441]) Labels shape: torch.Size([1, 27441]) Final batch size: 1, sequence length: 24622 Attention mask shape: torch.Size([1, 1, 24622, 24622]) Position ids shape: torch.Size([1, 24622]) Input IDs shape: torch.Size([1, 24622]) Labels shape: torch.Size([1, 24622]) Final batch size: 1, sequence length: 40745 Attention mask shape: torch.Size([1, 1, 40745, 40745]) Position ids shape: torch.Size([1, 40745]) Input IDs shape: torch.Size([1, 40745]) Labels shape: torch.Size([1, 40745]) Final batch size: 1, sequence length: 39142 Attention mask shape: torch.Size([1, 1, 39142, 39142]) Position ids shape: torch.Size([1, 39142]) Input IDs shape: torch.Size([1, 39142]) Labels shape: torch.Size([1, 39142]) Final batch size: 1, sequence length: 18565 Attention mask shape: torch.Size([1, 1, 18565, 18565]) Position ids shape: torch.Size([1, 18565]) Input IDs shape: torch.Size([1, 18565]) Labels shape: torch.Size([1, 18565]) Final batch size: 1, sequence length: 26939 Attention mask shape: torch.Size([1, 1, 26939, 26939]) Position ids shape: torch.Size([1, 26939]) Input IDs shape: torch.Size([1, 26939]) Labels shape: torch.Size([1, 26939]) Final batch size: 1, sequence length: 13509 Attention mask shape: torch.Size([1, 1, 13509, 13509]) Position ids shape: torch.Size([1, 13509]) Input IDs shape: torch.Size([1, 13509]) Labels shape: torch.Size([1, 13509]) Final batch size: 1, sequence length: 30802 Attention mask shape: torch.Size([1, 1, 30802, 30802]) Position ids shape: torch.Size([1, 30802]) Input IDs shape: torch.Size([1, 30802]) Labels shape: torch.Size([1, 30802]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 37945 Attention mask shape: torch.Size([1, 1, 37945, 37945]) Position ids shape: torch.Size([1, 37945]) Input IDs shape: torch.Size([1, 37945]) Labels shape: torch.Size([1, 37945]) Final batch size: 1, sequence length: 38249 Attention mask shape: torch.Size([1, 1, 38249, 38249]) Position ids shape: torch.Size([1, 38249]) Input IDs shape: torch.Size([1, 38249]) Labels shape: torch.Size([1, 38249]) Final batch size: 1, sequence length: 21567 Attention mask shape: torch.Size([1, 1, 21567, 21567]) Position ids shape: torch.Size([1, 21567]) Input IDs shape: torch.Size([1, 21567]) Labels shape: torch.Size([1, 21567]) Final batch size: 1, sequence length: 36271 Attention mask shape: torch.Size([1, 1, 36271, 36271]) Position ids shape: torch.Size([1, 36271]) Input IDs shape: torch.Size([1, 36271]) Labels shape: torch.Size([1, 36271]) Final batch size: 1, sequence length: 32609 Attention mask shape: torch.Size([1, 1, 32609, 32609]) Position ids shape: torch.Size([1, 32609]) Input IDs shape: torch.Size([1, 32609]) Labels shape: torch.Size([1, 32609]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19036 Attention mask shape: torch.Size([1, 1, 19036, 19036]) Position ids shape: torch.Size([1, 19036]) Input IDs shape: torch.Size([1, 19036]) Labels shape: torch.Size([1, 19036]) Final batch size: 1, sequence length: 35153 Attention mask shape: torch.Size([1, 1, 35153, 35153]) Position ids shape: torch.Size([1, 35153]) Input IDs shape: torch.Size([1, 35153]) Labels shape: torch.Size([1, 35153]) Final batch size: 1, sequence length: 25447 Attention mask shape: torch.Size([1, 1, 25447, 25447]) Position ids shape: torch.Size([1, 25447]) Final batch size: 1, sequence length: 15217Input IDs shape: torch.Size([1, 25447]) Labels shape: torch.Size([1, 25447]) Attention mask shape: torch.Size([1, 1, 15217, 15217]) Position ids shape: torch.Size([1, 15217]) Input IDs shape: torch.Size([1, 15217]) Labels shape: torch.Size([1, 15217]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 33839 Attention mask shape: torch.Size([1, 1, 33839, 33839]) Position ids shape: torch.Size([1, 33839]) Input IDs shape: torch.Size([1, 33839]) Labels shape: torch.Size([1, 33839]) Final batch size: 1, sequence length: 40496 Attention mask shape: torch.Size([1, 1, 40496, 40496]) Position ids shape: torch.Size([1, 40496]) Input IDs shape: torch.Size([1, 40496]) Labels shape: torch.Size([1, 40496]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36511 Attention mask shape: torch.Size([1, 1, 36511, 36511]) Position ids shape: torch.Size([1, 36511]) Input IDs shape: torch.Size([1, 36511]) Labels shape: torch.Size([1, 36511]) Final batch size: 1, sequence length: 15508 Attention mask shape: torch.Size([1, 1, 15508, 15508]) Position ids shape: torch.Size([1, 15508]) Input IDs shape: torch.Size([1, 15508]) Labels shape: torch.Size([1, 15508]) Final batch size: 1, sequence length: 39836 Attention mask shape: torch.Size([1, 1, 39836, 39836]) Position ids shape: torch.Size([1, 39836]) Input IDs shape: torch.Size([1, 39836]) Labels shape: torch.Size([1, 39836]) Final batch size: 1, sequence length: 32352 Attention mask shape: torch.Size([1, 1, 32352, 32352]) Position ids shape: torch.Size([1, 32352]) Input IDs shape: torch.Size([1, 32352]) Labels shape: torch.Size([1, 32352]) Final batch size: 1, sequence length: 16587 Attention mask shape: torch.Size([1, 1, 16587, 16587]) Position ids shape: torch.Size([1, 16587]) Input IDs shape: torch.Size([1, 16587]) Labels shape: torch.Size([1, 16587]) Final batch size: 1, sequence length: 23258 Attention mask shape: torch.Size([1, 1, 23258, 23258]) Position ids shape: torch.Size([1, 23258]) Input IDs shape: torch.Size([1, 23258]) Labels shape: torch.Size([1, 23258]) Final batch size: 1, sequence length: 30109 Attention mask shape: torch.Size([1, 1, 30109, 30109]) Position ids shape: torch.Size([1, 30109]) Input IDs shape: torch.Size([1, 30109]) Labels shape: torch.Size([1, 30109]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20611 Attention mask shape: torch.Size([1, 1, 20611, 20611]) Position ids shape: torch.Size([1, 20611]) Input IDs shape: torch.Size([1, 20611]) Labels shape: torch.Size([1, 20611]) Final batch size: 1, sequence length: 9947 Attention mask shape: torch.Size([1, 1, 9947, 9947]) Position ids shape: torch.Size([1, 9947]) Input IDs shape: torch.Size([1, 9947]) Labels shape: torch.Size([1, 9947]) Final batch size: 1, sequence length: 16677 Attention mask shape: torch.Size([1, 1, 16677, 16677]) Position ids shape: torch.Size([1, 16677]) Input IDs shape: torch.Size([1, 16677]) Labels shape: torch.Size([1, 16677]) Final batch size: 1, sequence length: 21758 Attention mask shape: torch.Size([1, 1, 21758, 21758]) Position ids shape: torch.Size([1, 21758]) Input IDs shape: torch.Size([1, 21758]) Labels shape: torch.Size([1, 21758]) Final batch size: 1, sequence length: 29632 Attention mask shape: torch.Size([1, 1, 29632, 29632]) Position ids shape: torch.Size([1, 29632]) Input IDs shape: torch.Size([1, 29632]) Labels shape: torch.Size([1, 29632]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 13095 Attention mask shape: torch.Size([1, 1, 13095, 13095]) Position ids shape: torch.Size([1, 13095]) Input IDs shape: torch.Size([1, 13095]) Labels shape: torch.Size([1, 13095]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 12224 Attention mask shape: torch.Size([1, 1, 12224, 12224]) Position ids shape: torch.Size([1, 12224]) Input IDs shape: torch.Size([1, 12224]) Labels shape: torch.Size([1, 12224]) Final batch size: 1, sequence length: 20422 Attention mask shape: torch.Size([1, 1, 20422, 20422]) Position ids shape: torch.Size([1, 20422]) Input IDs shape: torch.Size([1, 20422]) Labels shape: torch.Size([1, 20422]) Final batch size: 1, sequence length: 34937 Attention mask shape: torch.Size([1, 1, 34937, 34937]) Position ids shape: torch.Size([1, 34937]) Input IDs shape: torch.Size([1, 34937]) Labels shape: torch.Size([1, 34937]) Final batch size: 1, sequence length: 24298 Attention mask shape: torch.Size([1, 1, 24298, 24298]) Position ids shape: torch.Size([1, 24298]) Input IDs shape: torch.Size([1, 24298]) Labels shape: torch.Size([1, 24298]) Final batch size: 1, sequence length: 22786 Attention mask shape: torch.Size([1, 1, 22786, 22786]) Position ids shape: torch.Size([1, 22786]) Input IDs shape: torch.Size([1, 22786]) Labels shape: torch.Size([1, 22786]) Final batch size: 1, sequence length: 34142 Attention mask shape: torch.Size([1, 1, 34142, 34142]) Position ids shape: torch.Size([1, 34142]) Input IDs shape: torch.Size([1, 34142]) Labels shape: torch.Size([1, 34142]) Final batch size: 1, sequence length: 12653 Attention mask shape: torch.Size([1, 1, 12653, 12653]) Position ids shape: torch.Size([1, 12653]) Input IDs shape: torch.Size([1, 12653]) Labels shape: torch.Size([1, 12653]) Final batch size: 1, sequence length: 13297 Attention mask shape: torch.Size([1, 1, 13297, 13297]) Position ids shape: torch.Size([1, 13297]) Input IDs shape: torch.Size([1, 13297]) Labels shape: torch.Size([1, 13297]) Final batch size: 1, sequence length: 28634 Attention mask shape: torch.Size([1, 1, 28634, 28634]) Position ids shape: torch.Size([1, 28634]) Input IDs shape: torch.Size([1, 28634]) Labels shape: torch.Size([1, 28634]) Final batch size: 1, sequence length: 17373 Attention mask shape: torch.Size([1, 1, 17373, 17373]) Position ids shape: torch.Size([1, 17373]) Input IDs shape: torch.Size([1, 17373]) Labels shape: torch.Size([1, 17373]) Final batch size: 1, sequence length: 32472 Attention mask shape: torch.Size([1, 1, 32472, 32472]) Position ids shape: torch.Size([1, 32472]) Input IDs shape: torch.Size([1, 32472]) Labels shape: torch.Size([1, 32472]) Final batch size: 1, sequence length: 32564 Attention mask shape: torch.Size([1, 1, 32564, 32564]) Position ids shape: torch.Size([1, 32564]) Input IDs shape: torch.Size([1, 32564]) Labels shape: torch.Size([1, 32564]) Final batch size: 1, sequence length: 32753 Attention mask shape: torch.Size([1, 1, 32753, 32753]) Position ids shape: torch.Size([1, 32753]) Input IDs shape: torch.Size([1, 32753]) Labels shape: torch.Size([1, 32753]) Final batch size: 1, sequence length: 14995 Attention mask shape: torch.Size([1, 1, 14995, 14995]) Position ids shape: torch.Size([1, 14995]) Input IDs shape: torch.Size([1, 14995]) Labels shape: torch.Size([1, 14995]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22623 Attention mask shape: torch.Size([1, 1, 22623, 22623]) Position ids shape: torch.Size([1, 22623]) Input IDs shape: torch.Size([1, 22623]) Labels shape: torch.Size([1, 22623]) Final batch size: 1, sequence length: 21683 Attention mask shape: torch.Size([1, 1, 21683, 21683]) Position ids shape: torch.Size([1, 21683]) Input IDs shape: torch.Size([1, 21683]) Labels shape: torch.Size([1, 21683]) Final batch size: 1, sequence length: 24424 Attention mask shape: torch.Size([1, 1, 24424, 24424]) Position ids shape: torch.Size([1, 24424]) Input IDs shape: torch.Size([1, 24424]) Labels shape: torch.Size([1, 24424]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 35864 Attention mask shape: torch.Size([1, 1, 35864, 35864]) Position ids shape: torch.Size([1, 35864]) Input IDs shape: torch.Size([1, 35864]) Labels shape: torch.Size([1, 35864]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32344 Attention mask shape: torch.Size([1, 1, 32344, 32344]) Position ids shape: torch.Size([1, 32344]) Input IDs shape: torch.Size([1, 32344]) Labels shape: torch.Size([1, 32344]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26537 Attention mask shape: torch.Size([1, 1, 26537, 26537]) Position ids shape: torch.Size([1, 26537]) Input IDs shape: torch.Size([1, 26537]) Labels shape: torch.Size([1, 26537]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32919 Attention mask shape: torch.Size([1, 1, 32919, 32919]) Position ids shape: torch.Size([1, 32919]) Input IDs shape: torch.Size([1, 32919]) Labels shape: torch.Size([1, 32919]) Final batch size: 1, sequence length: 32514 Attention mask shape: torch.Size([1, 1, 32514, 32514]) Position ids shape: torch.Size([1, 32514]) Input IDs shape: torch.Size([1, 32514]) Labels shape: torch.Size([1, 32514]) Final batch size: 1, sequence length: 37168 Attention mask shape: torch.Size([1, 1, 37168, 37168]) Position ids shape: torch.Size([1, 37168]) Input IDs shape: torch.Size([1, 37168]) Labels shape: torch.Size([1, 37168]) Final batch size: 1, sequence length: 23362 Attention mask shape: torch.Size([1, 1, 23362, 23362]) Position ids shape: torch.Size([1, 23362]) Input IDs shape: torch.Size([1, 23362]) Labels shape: torch.Size([1, 23362]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28861 Attention mask shape: torch.Size([1, 1, 28861, 28861]) Position ids shape: torch.Size([1, 28861]) Input IDs shape: torch.Size([1, 28861]) Labels shape: torch.Size([1, 28861]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26006 Attention mask shape: torch.Size([1, 1, 26006, 26006]) Position ids shape: torch.Size([1, 26006]) Input IDs shape: torch.Size([1, 26006]) Labels shape: torch.Size([1, 26006]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2515, 'grad_norm': 0.28022842799402736, 'learning_rate': 3.4549150281252635e-06, 'num_tokens': -inf, 'epoch': 5.12} Final batch size: 1, sequence length: 6316 Attention mask shape: torch.Size([1, 1, 6316, 6316]) Position ids shape: torch.Size([1, 6316]) Input IDs shape: torch.Size([1, 6316]) Labels shape: torch.Size([1, 6316]) Final batch size: 1, sequence length: 7360 Attention mask shape: torch.Size([1, 1, 7360, 7360]) Position ids shape: torch.Size([1, 7360]) Input IDs shape: torch.Size([1, 7360]) Labels shape: torch.Size([1, 7360]) Final batch size: 1, sequence length: 4858 Attention mask shape: torch.Size([1, 1, 4858, 4858]) Position ids shape: torch.Size([1, 4858]) Input IDs shape: torch.Size([1, 4858]) Labels shape: torch.Size([1, 4858]) Final batch size: 1, sequence length: 10301 Attention mask shape: torch.Size([1, 1, 10301, 10301]) Position ids shape: torch.Size([1, 10301]) Input IDs shape: torch.Size([1, 10301]) Labels shape: torch.Size([1, 10301]) Final batch size: 1, sequence length: 11548 Attention mask shape: torch.Size([1, 1, 11548, 11548]) Position ids shape: torch.Size([1, 11548]) Input IDs shape: torch.Size([1, 11548]) Labels shape: torch.Size([1, 11548]) Final batch size: 1, sequence length: 11728 Attention mask shape: torch.Size([1, 1, 11728, 11728]) Position ids shape: torch.Size([1, 11728]) Input IDs shape: torch.Size([1, 11728]) Labels shape: torch.Size([1, 11728]) Final batch size: 1, sequence length: 12293 Attention mask shape: torch.Size([1, 1, 12293, 12293]) Position ids shape: torch.Size([1, 12293]) Input IDs shape: torch.Size([1, 12293]) Labels shape: torch.Size([1, 12293]) Final batch size: 1, sequence length: 13355 Attention mask shape: torch.Size([1, 1, 13355, 13355]) Position ids shape: torch.Size([1, 13355]) Input IDs shape: torch.Size([1, 13355]) Labels shape: torch.Size([1, 13355]) Final batch size: 1, sequence length: 14363 Attention mask shape: torch.Size([1, 1, 14363, 14363]) Position ids shape: torch.Size([1, 14363]) Input IDs shape: torch.Size([1, 14363]) Labels shape: torch.Size([1, 14363]) Final batch size: 1, sequence length: 16827 Attention mask shape: torch.Size([1, 1, 16827, 16827]) Position ids shape: torch.Size([1, 16827]) Input IDs shape: torch.Size([1, 16827]) Labels shape: torch.Size([1, 16827]) Final batch size: 1, sequence length: 14587 Attention mask shape: torch.Size([1, 1, 14587, 14587]) Position ids shape: torch.Size([1, 14587]) Input IDs shape: torch.Size([1, 14587]) Labels shape: torch.Size([1, 14587]) Final batch size: 1, sequence length: 14704 Attention mask shape: torch.Size([1, 1, 14704, 14704]) Position ids shape: torch.Size([1, 14704]) Input IDs shape: torch.Size([1, 14704]) Labels shape: torch.Size([1, 14704]) Final batch size: 1, sequence length: 17108 Attention mask shape: torch.Size([1, 1, 17108, 17108]) Position ids shape: torch.Size([1, 17108]) Input IDs shape: torch.Size([1, 17108]) Labels shape: torch.Size([1, 17108]) Final batch size: 1, sequence length: 13638 Attention mask shape: torch.Size([1, 1, 13638, 13638]) Position ids shape: torch.Size([1, 13638]) Input IDs shape: torch.Size([1, 13638]) Labels shape: torch.Size([1, 13638]) Final batch size: 1, sequence length: 16536 Attention mask shape: torch.Size([1, 1, 16536, 16536]) Position ids shape: torch.Size([1, 16536]) Input IDs shape: torch.Size([1, 16536]) Labels shape: torch.Size([1, 16536]) Final batch size: 1, sequence length: 17415 Attention mask shape: torch.Size([1, 1, 17415, 17415]) Position ids shape: torch.Size([1, 17415]) Input IDs shape: torch.Size([1, 17415]) Labels shape: torch.Size([1, 17415]) Final batch size: 1, sequence length: 19414 Attention mask shape: torch.Size([1, 1, 19414, 19414]) Position ids shape: torch.Size([1, 19414]) Input IDs shape: torch.Size([1, 19414]) Labels shape: torch.Size([1, 19414]) Final batch size: 1, sequence length: 18927 Attention mask shape: torch.Size([1, 1, 18927, 18927]) Position ids shape: torch.Size([1, 18927]) Input IDs shape: torch.Size([1, 18927]) Labels shape: torch.Size([1, 18927]) Final batch size: 1, sequence length: 18741 Attention mask shape: torch.Size([1, 1, 18741, 18741]) Position ids shape: torch.Size([1, 18741]) Input IDs shape: torch.Size([1, 18741]) Labels shape: torch.Size([1, 18741]) Final batch size: 1, sequence length: 18496 Attention mask shape: torch.Size([1, 1, 18496, 18496]) Position ids shape: torch.Size([1, 18496]) Input IDs shape: torch.Size([1, 18496]) Labels shape: torch.Size([1, 18496]) Final batch size: 1, sequence length: 17220 Attention mask shape: torch.Size([1, 1, 17220, 17220]) Position ids shape: torch.Size([1, 17220]) Input IDs shape: torch.Size([1, 17220]) Labels shape: torch.Size([1, 17220]) Final batch size: 1, sequence length: 7221 Attention mask shape: torch.Size([1, 1, 7221, 7221]) Position ids shape: torch.Size([1, 7221]) Input IDs shape: torch.Size([1, 7221]) Labels shape: torch.Size([1, 7221]) Final batch size: 1, sequence length: 21420 Attention mask shape: torch.Size([1, 1, 21420, 21420]) Position ids shape: torch.Size([1, 21420]) Input IDs shape: torch.Size([1, 21420]) Labels shape: torch.Size([1, 21420]) Final batch size: 1, sequence length: 18393 Attention mask shape: torch.Size([1, 1, 18393, 18393]) Position ids shape: torch.Size([1, 18393]) Input IDs shape: torch.Size([1, 18393]) Labels shape: torch.Size([1, 18393]) Final batch size: 1, sequence length: 20933 Attention mask shape: torch.Size([1, 1, 20933, 20933]) Position ids shape: torch.Size([1, 20933]) Input IDs shape: torch.Size([1, 20933]) Labels shape: torch.Size([1, 20933]) Final batch size: 1, sequence length: 17733 Attention mask shape: torch.Size([1, 1, 17733, 17733]) Position ids shape: torch.Size([1, 17733]) Input IDs shape: torch.Size([1, 17733]) Labels shape: torch.Size([1, 17733]) Final batch size: 1, sequence length: 18653 Attention mask shape: torch.Size([1, 1, 18653, 18653]) Position ids shape: torch.Size([1, 18653]) Input IDs shape: torch.Size([1, 18653]) Labels shape: torch.Size([1, 18653]) Final batch size: 1, sequence length: 16750 Attention mask shape: torch.Size([1, 1, 16750, 16750]) Position ids shape: torch.Size([1, 16750]) Input IDs shape: torch.Size([1, 16750]) Labels shape: torch.Size([1, 16750]) Final batch size: 1, sequence length: 20612 Attention mask shape: torch.Size([1, 1, 20612, 20612]) Position ids shape: torch.Size([1, 20612]) Input IDs shape: torch.Size([1, 20612]) Labels shape: torch.Size([1, 20612]) Final batch size: 1, sequence length: 22391 Attention mask shape: torch.Size([1, 1, 22391, 22391]) Position ids shape: torch.Size([1, 22391]) Input IDs shape: torch.Size([1, 22391]) Labels shape: torch.Size([1, 22391]) Final batch size: 1, sequence length: 20579 Attention mask shape: torch.Size([1, 1, 20579, 20579]) Position ids shape: torch.Size([1, 20579]) Input IDs shape: torch.Size([1, 20579]) Labels shape: torch.Size([1, 20579]) Final batch size: 1, sequence length: 22887 Attention mask shape: torch.Size([1, 1, 22887, 22887]) Position ids shape: torch.Size([1, 22887]) Input IDs shape: torch.Size([1, 22887]) Labels shape: torch.Size([1, 22887]) Final batch size: 1, sequence length: 22004 Attention mask shape: torch.Size([1, 1, 22004, 22004]) Position ids shape: torch.Size([1, 22004]) Input IDs shape: torch.Size([1, 22004]) Labels shape: torch.Size([1, 22004]) Final batch size: 1, sequence length: 11067 Attention mask shape: torch.Size([1, 1, 11067, 11067]) Position ids shape: torch.Size([1, 11067]) Input IDs shape: torch.Size([1, 11067]) Labels shape: torch.Size([1, 11067]) Final batch size: 1, sequence length: 10719 Attention mask shape: torch.Size([1, 1, 10719, 10719]) Position ids shape: torch.Size([1, 10719]) Input IDs shape: torch.Size([1, 10719]) Labels shape: torch.Size([1, 10719]) Final batch size: 1, sequence length: 24988 Attention mask shape: torch.Size([1, 1, 24988, 24988]) Position ids shape: torch.Size([1, 24988]) Input IDs shape: torch.Size([1, 24988]) Labels shape: torch.Size([1, 24988]) Final batch size: 1, sequence length: 11184 Attention mask shape: torch.Size([1, 1, 11184, 11184]) Position ids shape: torch.Size([1, 11184]) Input IDs shape: torch.Size([1, 11184]) Labels shape: torch.Size([1, 11184]) Final batch size: 1, sequence length: 25477 Attention mask shape: torch.Size([1, 1, 25477, 25477]) Position ids shape: torch.Size([1, 25477]) Input IDs shape: torch.Size([1, 25477]) Labels shape: torch.Size([1, 25477]) Final batch size: 1, sequence length: 25651 Attention mask shape: torch.Size([1, 1, 25651, 25651]) Position ids shape: torch.Size([1, 25651]) Input IDs shape: torch.Size([1, 25651]) Labels shape: torch.Size([1, 25651]) Final batch size: 1, sequence length: 25747 Attention mask shape: torch.Size([1, 1, 25747, 25747]) Position ids shape: torch.Size([1, 25747]) Input IDs shape: torch.Size([1, 25747]) Labels shape: torch.Size([1, 25747]) Final batch size: 1, sequence length: 15317 Attention mask shape: torch.Size([1, 1, 15317, 15317]) Position ids shape: torch.Size([1, 15317]) Input IDs shape: torch.Size([1, 15317]) Labels shape: torch.Size([1, 15317]) Final batch size: 1, sequence length: 26663 Attention mask shape: torch.Size([1, 1, 26663, 26663]) Position ids shape: torch.Size([1, 26663]) Input IDs shape: torch.Size([1, 26663]) Labels shape: torch.Size([1, 26663]) Final batch size: 1, sequence length: 16915 Attention mask shape: torch.Size([1, 1, 16915, 16915]) Position ids shape: torch.Size([1, 16915]) Input IDs shape: torch.Size([1, 16915]) Labels shape: torch.Size([1, 16915]) Final batch size: 1, sequence length: 27447 Attention mask shape: torch.Size([1, 1, 27447, 27447]) Position ids shape: torch.Size([1, 27447]) Input IDs shape: torch.Size([1, 27447]) Labels shape: torch.Size([1, 27447]) Final batch size: 1, sequence length: 17395 Attention mask shape: torch.Size([1, 1, 17395, 17395]) Position ids shape: torch.Size([1, 17395]) Input IDs shape: torch.Size([1, 17395]) Labels shape: torch.Size([1, 17395]) Final batch size: 1, sequence length: 19552 Attention mask shape: torch.Size([1, 1, 19552, 19552]) Position ids shape: torch.Size([1, 19552]) Input IDs shape: torch.Size([1, 19552]) Labels shape: torch.Size([1, 19552]) Final batch size: 1, sequence length: 28777 Attention mask shape: torch.Size([1, 1, 28777, 28777]) Position ids shape: torch.Size([1, 28777]) Input IDs shape: torch.Size([1, 28777]) Labels shape: torch.Size([1, 28777]) Final batch size: 1, sequence length: 30031 Attention mask shape: torch.Size([1, 1, 30031, 30031]) Position ids shape: torch.Size([1, 30031]) Input IDs shape: torch.Size([1, 30031]) Labels shape: torch.Size([1, 30031]) Final batch size: 1, sequence length: 30981 Attention mask shape: torch.Size([1, 1, 30981, 30981]) Position ids shape: torch.Size([1, 30981]) Input IDs shape: torch.Size([1, 30981]) Labels shape: torch.Size([1, 30981]) Final batch size: 1, sequence length: 19869 Attention mask shape: torch.Size([1, 1, 19869, 19869]) Position ids shape: torch.Size([1, 19869]) Input IDs shape: torch.Size([1, 19869]) Labels shape: torch.Size([1, 19869]) Final batch size: 1, sequence length: 17911 Attention mask shape: torch.Size([1, 1, 17911, 17911]) Position ids shape: torch.Size([1, 17911]) Input IDs shape: torch.Size([1, 17911]) Labels shape: torch.Size([1, 17911]) Final batch size: 1, sequence length: 16953 Attention mask shape: torch.Size([1, 1, 16953, 16953]) Position ids shape: torch.Size([1, 16953]) Input IDs shape: torch.Size([1, 16953]) Labels shape: torch.Size([1, 16953]) Final batch size: 1, sequence length: 27334 Attention mask shape: torch.Size([1, 1, 27334, 27334]) Position ids shape: torch.Size([1, 27334]) Input IDs shape: torch.Size([1, 27334]) Labels shape: torch.Size([1, 27334]) Final batch size: 1, sequence length: 27480 Final batch size: 1, sequence length: 32466 Attention mask shape: torch.Size([1, 1, 27480, 27480]) Position ids shape: torch.Size([1, 27480]) Input IDs shape: torch.Size([1, 27480]) Labels shape: torch.Size([1, 27480]) Attention mask shape: torch.Size([1, 1, 32466, 32466]) Position ids shape: torch.Size([1, 32466]) Input IDs shape: torch.Size([1, 32466]) Labels shape: torch.Size([1, 32466]) Final batch size: 1, sequence length: 26033 Attention mask shape: torch.Size([1, 1, 26033, 26033]) Position ids shape: torch.Size([1, 26033]) Input IDs shape: torch.Size([1, 26033]) Labels shape: torch.Size([1, 26033]) Final batch size: 1, sequence length: 18376 Attention mask shape: torch.Size([1, 1, 18376, 18376]) Position ids shape: torch.Size([1, 18376]) Input IDs shape: torch.Size([1, 18376]) Labels shape: torch.Size([1, 18376]) Final batch size: 1, sequence length: 21988 Attention mask shape: torch.Size([1, 1, 21988, 21988]) Position ids shape: torch.Size([1, 21988]) Input IDs shape: torch.Size([1, 21988]) Labels shape: torch.Size([1, 21988]) Final batch size: 1, sequence length: 30601 Attention mask shape: torch.Size([1, 1, 30601, 30601]) Position ids shape: torch.Size([1, 30601]) Input IDs shape: torch.Size([1, 30601]) Labels shape: torch.Size([1, 30601]) Final batch size: 1, sequence length: 15924 Attention mask shape: torch.Size([1, 1, 15924, 15924]) Position ids shape: torch.Size([1, 15924]) Input IDs shape: torch.Size([1, 15924]) Labels shape: torch.Size([1, 15924]) Final batch size: 1, sequence length: 28060 Attention mask shape: torch.Size([1, 1, 28060, 28060]) Position ids shape: torch.Size([1, 28060]) Input IDs shape: torch.Size([1, 28060]) Labels shape: torch.Size([1, 28060]) Final batch size: 1, sequence length: 27385 Attention mask shape: torch.Size([1, 1, 27385, 27385]) Position ids shape: torch.Size([1, 27385]) Input IDs shape: torch.Size([1, 27385]) Labels shape: torch.Size([1, 27385]) Final batch size: 1, sequence length: 24782 Attention mask shape: torch.Size([1, 1, 24782, 24782]) Position ids shape: torch.Size([1, 24782]) Input IDs shape: torch.Size([1, 24782]) Labels shape: torch.Size([1, 24782]) Final batch size: 1, sequence length: 12245 Attention mask shape: torch.Size([1, 1, 12245, 12245]) Position ids shape: torch.Size([1, 12245]) Input IDs shape: torch.Size([1, 12245]) Labels shape: torch.Size([1, 12245]) Final batch size: 1, sequence length: 14873 Attention mask shape: torch.Size([1, 1, 14873, 14873]) Position ids shape: torch.Size([1, 14873]) Input IDs shape: torch.Size([1, 14873]) Labels shape: torch.Size([1, 14873]) Final batch size: 1, sequence length: 33725 Attention mask shape: torch.Size([1, 1, 33725, 33725]) Position ids shape: torch.Size([1, 33725]) Input IDs shape: torch.Size([1, 33725]) Labels shape: torch.Size([1, 33725]) Final batch size: 1, sequence length: 37158 Attention mask shape: torch.Size([1, 1, 37158, 37158]) Position ids shape: torch.Size([1, 37158]) Input IDs shape: torch.Size([1, 37158]) Labels shape: torch.Size([1, 37158]) Final batch size: 1, sequence length: 37025 Attention mask shape: torch.Size([1, 1, 37025, 37025]) Position ids shape: torch.Size([1, 37025]) Input IDs shape: torch.Size([1, 37025]) Labels shape: torch.Size([1, 37025]) Final batch size: 1, sequence length: 26479 Attention mask shape: torch.Size([1, 1, 26479, 26479]) Position ids shape: torch.Size([1, 26479]) Input IDs shape: torch.Size([1, 26479]) Labels shape: torch.Size([1, 26479]) Final batch size: 1, sequence length: 37738 Attention mask shape: torch.Size([1, 1, 37738, 37738]) Position ids shape: torch.Size([1, 37738]) Input IDs shape: torch.Size([1, 37738]) Labels shape: torch.Size([1, 37738]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 37511 Attention mask shape: torch.Size([1, 1, 37511, 37511]) Position ids shape: torch.Size([1, 37511]) Input IDs shape: torch.Size([1, 37511]) Labels shape: torch.Size([1, 37511]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40397 Attention mask shape: torch.Size([1, 1, 40397, 40397]) Position ids shape: torch.Size([1, 40397]) Input IDs shape: torch.Size([1, 40397]) Labels shape: torch.Size([1, 40397]) Final batch size: 1, sequence length: 32835 Attention mask shape: torch.Size([1, 1, 32835, 32835]) Position ids shape: torch.Size([1, 32835]) Input IDs shape: torch.Size([1, 32835]) Labels shape: torch.Size([1, 32835]) Final batch size: 1, sequence length: 36175 Attention mask shape: torch.Size([1, 1, 36175, 36175]) Position ids shape: torch.Size([1, 36175]) Input IDs shape: torch.Size([1, 36175]) Labels shape: torch.Size([1, 36175]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 38986 Attention mask shape: torch.Size([1, 1, 38986, 38986]) Position ids shape: torch.Size([1, 38986]) Input IDs shape: torch.Size([1, 38986]) Labels shape: torch.Size([1, 38986]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31879 Attention mask shape: torch.Size([1, 1, 31879, 31879]) Position ids shape: torch.Size([1, 31879]) Input IDs shape: torch.Size([1, 31879]) Labels shape: torch.Size([1, 31879]) Final batch size: 1, sequence length: 36130 Attention mask shape: torch.Size([1, 1, 36130, 36130]) Position ids shape: torch.Size([1, 36130]) Input IDs shape: torch.Size([1, 36130]) Labels shape: torch.Size([1, 36130]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 29586 Attention mask shape: torch.Size([1, 1, 29586, 29586]) Position ids shape: torch.Size([1, 29586]) Input IDs shape: torch.Size([1, 29586]) Labels shape: torch.Size([1, 29586]) Final batch size: 1, sequence length: 40269 Attention mask shape: torch.Size([1, 1, 40269, 40269]) Position ids shape: torch.Size([1, 40269]) Input IDs shape: torch.Size([1, 40269]) Labels shape: torch.Size([1, 40269]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40237 Attention mask shape: torch.Size([1, 1, 40237, 40237]) Position ids shape: torch.Size([1, 40237]) Input IDs shape: torch.Size([1, 40237]) Labels shape: torch.Size([1, 40237]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31348 Attention mask shape: torch.Size([1, 1, 31348, 31348]) Position ids shape: torch.Size([1, 31348]) Input IDs shape: torch.Size([1, 31348]) Labels shape: torch.Size([1, 31348]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20509 Attention mask shape: torch.Size([1, 1, 20509, 20509]) Position ids shape: torch.Size([1, 20509]) Input IDs shape: torch.Size([1, 20509]) Labels shape: torch.Size([1, 20509]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32318 Attention mask shape: torch.Size([1, 1, 32318, 32318]) Position ids shape: torch.Size([1, 32318]) Input IDs shape: torch.Size([1, 32318]) Labels shape: torch.Size([1, 32318]) Final batch size: 1, sequence length: 27291 Attention mask shape: torch.Size([1, 1, 27291, 27291]) Position ids shape: torch.Size([1, 27291]) Input IDs shape: torch.Size([1, 27291]) Labels shape: torch.Size([1, 27291]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33125 Attention mask shape: torch.Size([1, 1, 33125, 33125]) Position ids shape: torch.Size([1, 33125]) Input IDs shape: torch.Size([1, 33125]) Labels shape: torch.Size([1, 33125]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28631 Attention mask shape: torch.Size([1, 1, 28631, 28631]) Position ids shape: torch.Size([1, 28631]) Input IDs shape: torch.Size([1, 28631]) Labels shape: torch.Size([1, 28631]) Final batch size: 1, sequence length: 36947 Attention mask shape: torch.Size([1, 1, 36947, 36947]) Position ids shape: torch.Size([1, 36947]) Input IDs shape: torch.Size([1, 36947]) Labels shape: torch.Size([1, 36947]) Final batch size: 1, sequence length: 10198 Attention mask shape: torch.Size([1, 1, 10198, 10198]) Position ids shape: torch.Size([1, 10198]) Input IDs shape: torch.Size([1, 10198]) Labels shape: torch.Size([1, 10198]) Final batch size: 1, sequence length: 32247 Attention mask shape: torch.Size([1, 1, 32247, 32247]) Position ids shape: torch.Size([1, 32247]) Input IDs shape: torch.Size([1, 32247]) Labels shape: torch.Size([1, 32247]) Final batch size: 1, sequence length: 24939 Attention mask shape: torch.Size([1, 1, 24939, 24939]) Position ids shape: torch.Size([1, 24939]) Input IDs shape: torch.Size([1, 24939]) Labels shape: torch.Size([1, 24939]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21225 Attention mask shape: torch.Size([1, 1, 21225, 21225]) Position ids shape: torch.Size([1, 21225]) Input IDs shape: torch.Size([1, 21225]) Labels shape: torch.Size([1, 21225]) Final batch size: 1, sequence length: 17711 Attention mask shape: torch.Size([1, 1, 17711, 17711]) Position ids shape: torch.Size([1, 17711]) Input IDs shape: torch.Size([1, 17711]) Labels shape: torch.Size([1, 17711]) Final batch size: 1, sequence length: 23740 Attention mask shape: torch.Size([1, 1, 23740, 23740]) Position ids shape: torch.Size([1, 23740]) Input IDs shape: torch.Size([1, 23740]) Labels shape: torch.Size([1, 23740]) Final batch size: 1, sequence length: 28046 Attention mask shape: torch.Size([1, 1, 28046, 28046]) Position ids shape: torch.Size([1, 28046]) Input IDs shape: torch.Size([1, 28046]) Labels shape: torch.Size([1, 28046]) Final batch size: 1, sequence length: 22896 Attention mask shape: torch.Size([1, 1, 22896, 22896]) Position ids shape: torch.Size([1, 22896]) Input IDs shape: torch.Size([1, 22896]) Labels shape: torch.Size([1, 22896]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22205 Attention mask shape: torch.Size([1, 1, 22205, 22205]) Position ids shape: torch.Size([1, 22205]) Input IDs shape: torch.Size([1, 22205]) Labels shape: torch.Size([1, 22205]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28149 Attention mask shape: torch.Size([1, 1, 28149, 28149]) Position ids shape: torch.Size([1, 28149]) Input IDs shape: torch.Size([1, 28149]) Labels shape: torch.Size([1, 28149]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38360 Attention mask shape: torch.Size([1, 1, 38360, 38360]) Position ids shape: torch.Size([1, 38360]) Input IDs shape: torch.Size([1, 38360]) Labels shape: torch.Size([1, 38360]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32767 Attention mask shape: torch.Size([1, 1, 32767, 32767]) Position ids shape: torch.Size([1, 32767]) Input IDs shape: torch.Size([1, 32767]) Labels shape: torch.Size([1, 32767]) Final batch size: 1, sequence length: 36567 Attention mask shape: torch.Size([1, 1, 36567, 36567]) Position ids shape: torch.Size([1, 36567]) Input IDs shape: torch.Size([1, 36567]) Labels shape: torch.Size([1, 36567]) Final batch size: 1, sequence length: 34268 Attention mask shape: torch.Size([1, 1, 34268, 34268]) Position ids shape: torch.Size([1, 34268]) Input IDs shape: torch.Size([1, 34268]) Labels shape: torch.Size([1, 34268]) Final batch size: 1, sequence length: 32313 Attention mask shape: torch.Size([1, 1, 32313, 32313]) Position ids shape: torch.Size([1, 32313]) Input IDs shape: torch.Size([1, 32313]) Labels shape: torch.Size([1, 32313]) {'loss': 0.2794, 'grad_norm': 0.2624260462163811, 'learning_rate': 3.2081602522734987e-06, 'num_tokens': -inf, 'epoch': 5.25} Final batch size: 1, sequence length: 5377 Attention mask shape: torch.Size([1, 1, 5377, 5377]) Position ids shape: torch.Size([1, 5377]) Input IDs shape: torch.Size([1, 5377]) Labels shape: torch.Size([1, 5377]) Final batch size: 1, sequence length: 7998 Attention mask shape: torch.Size([1, 1, 7998, 7998]) Position ids shape: torch.Size([1, 7998]) Input IDs shape: torch.Size([1, 7998]) Labels shape: torch.Size([1, 7998]) Final batch size: 1, sequence length: 7402 Attention mask shape: torch.Size([1, 1, 7402, 7402]) Position ids shape: torch.Size([1, 7402]) Input IDs shape: torch.Size([1, 7402]) Labels shape: torch.Size([1, 7402]) Final batch size: 1, sequence length: 8436 Attention mask shape: torch.Size([1, 1, 8436, 8436]) Position ids shape: torch.Size([1, 8436]) Input IDs shape: torch.Size([1, 8436]) Labels shape: torch.Size([1, 8436]) Final batch size: 1, sequence length: 10576 Attention mask shape: torch.Size([1, 1, 10576, 10576]) Position ids shape: torch.Size([1, 10576]) Input IDs shape: torch.Size([1, 10576]) Labels shape: torch.Size([1, 10576]) Final batch size: 1, sequence length: 11709 Attention mask shape: torch.Size([1, 1, 11709, 11709]) Position ids shape: torch.Size([1, 11709]) Input IDs shape: torch.Size([1, 11709]) Labels shape: torch.Size([1, 11709]) Final batch size: 1, sequence length: 7344 Attention mask shape: torch.Size([1, 1, 7344, 7344]) Position ids shape: torch.Size([1, 7344]) Input IDs shape: torch.Size([1, 7344]) Labels shape: torch.Size([1, 7344]) Final batch size: 1, sequence length: 11365 Attention mask shape: torch.Size([1, 1, 11365, 11365]) Position ids shape: torch.Size([1, 11365]) Input IDs shape: torch.Size([1, 11365]) Labels shape: torch.Size([1, 11365]) Final batch size: 1, sequence length: 11678 Attention mask shape: torch.Size([1, 1, 11678, 11678]) Position ids shape: torch.Size([1, 11678]) Input IDs shape: torch.Size([1, 11678]) Labels shape: torch.Size([1, 11678]) Final batch size: 1, sequence length: 8655 Attention mask shape: torch.Size([1, 1, 8655, 8655]) Position ids shape: torch.Size([1, 8655]) Input IDs shape: torch.Size([1, 8655]) Labels shape: torch.Size([1, 8655]) Final batch size: 1, sequence length: 11623 Attention mask shape: torch.Size([1, 1, 11623, 11623]) Position ids shape: torch.Size([1, 11623]) Input IDs shape: torch.Size([1, 11623]) Labels shape: torch.Size([1, 11623]) Final batch size: 1, sequence length: 15079 Attention mask shape: torch.Size([1, 1, 15079, 15079]) Position ids shape: torch.Size([1, 15079]) Input IDs shape: torch.Size([1, 15079]) Labels shape: torch.Size([1, 15079]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 14127 Attention mask shape: torch.Size([1, 1, 14127, 14127]) Position ids shape: torch.Size([1, 14127]) Input IDs shape: torch.Size([1, 14127]) Labels shape: torch.Size([1, 14127]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 15478 Attention mask shape: torch.Size([1, 1, 15478, 15478]) Position ids shape: torch.Size([1, 15478]) Input IDs shape: torch.Size([1, 15478]) Labels shape: torch.Size([1, 15478]) Final batch size: 1, sequence length: 16535 Attention mask shape: torch.Size([1, 1, 16535, 16535]) Position ids shape: torch.Size([1, 16535]) Input IDs shape: torch.Size([1, 16535]) Labels shape: torch.Size([1, 16535]) Final batch size: 1, sequence length: 13639 Attention mask shape: torch.Size([1, 1, 13639, 13639]) Position ids shape: torch.Size([1, 13639]) Input IDs shape: torch.Size([1, 13639]) Labels shape: torch.Size([1, 13639]) Final batch size: 1, sequence length: 14648 Attention mask shape: torch.Size([1, 1, 14648, 14648]) Position ids shape: torch.Size([1, 14648]) Input IDs shape: torch.Size([1, 14648]) Labels shape: torch.Size([1, 14648]) Final batch size: 1, sequence length: 15308 Attention mask shape: torch.Size([1, 1, 15308, 15308]) Position ids shape: torch.Size([1, 15308]) Input IDs shape: torch.Size([1, 15308]) Labels shape: torch.Size([1, 15308]) Final batch size: 1, sequence length: 19170 Attention mask shape: torch.Size([1, 1, 19170, 19170]) Position ids shape: torch.Size([1, 19170]) Input IDs shape: torch.Size([1, 19170]) Labels shape: torch.Size([1, 19170]) Final batch size: 1, sequence length: 19338 Attention mask shape: torch.Size([1, 1, 19338, 19338]) Position ids shape: torch.Size([1, 19338]) Input IDs shape: torch.Size([1, 19338]) Labels shape: torch.Size([1, 19338]) Final batch size: 1, sequence length: 18953 Attention mask shape: torch.Size([1, 1, 18953, 18953]) Position ids shape: torch.Size([1, 18953]) Input IDs shape: torch.Size([1, 18953]) Labels shape: torch.Size([1, 18953]) Final batch size: 1, sequence length: 17058 Attention mask shape: torch.Size([1, 1, 17058, 17058]) Position ids shape: torch.Size([1, 17058]) Input IDs shape: torch.Size([1, 17058]) Labels shape: torch.Size([1, 17058]) Final batch size: 1, sequence length: 18307 Attention mask shape: torch.Size([1, 1, 18307, 18307]) Position ids shape: torch.Size([1, 18307]) Input IDs shape: torch.Size([1, 18307]) Labels shape: torch.Size([1, 18307]) Final batch size: 1, sequence length: 19138 Attention mask shape: torch.Size([1, 1, 19138, 19138]) Position ids shape: torch.Size([1, 19138]) Input IDs shape: torch.Size([1, 19138]) Labels shape: torch.Size([1, 19138]) Final batch size: 1, sequence length: 19330 Attention mask shape: torch.Size([1, 1, 19330, 19330]) Position ids shape: torch.Size([1, 19330]) Input IDs shape: torch.Size([1, 19330]) Labels shape: torch.Size([1, 19330]) Final batch size: 1, sequence length: 8839 Attention mask shape: torch.Size([1, 1, 8839, 8839]) Position ids shape: torch.Size([1, 8839]) Input IDs shape: torch.Size([1, 8839]) Labels shape: torch.Size([1, 8839]) Final batch size: 1, sequence length: 20714 Attention mask shape: torch.Size([1, 1, 20714, 20714]) Position ids shape: torch.Size([1, 20714]) Input IDs shape: torch.Size([1, 20714]) Labels shape: torch.Size([1, 20714]) Final batch size: 1, sequence length: 18527 Attention mask shape: torch.Size([1, 1, 18527, 18527]) Position ids shape: torch.Size([1, 18527]) Input IDs shape: torch.Size([1, 18527]) Labels shape: torch.Size([1, 18527]) Final batch size: 1, sequence length: 20770 Attention mask shape: torch.Size([1, 1, 20770, 20770]) Position ids shape: torch.Size([1, 20770]) Input IDs shape: torch.Size([1, 20770]) Labels shape: torch.Size([1, 20770]) Final batch size: 1, sequence length: 22561 Attention mask shape: torch.Size([1, 1, 22561, 22561]) Position ids shape: torch.Size([1, 22561]) Input IDs shape: torch.Size([1, 22561]) Labels shape: torch.Size([1, 22561]) Final batch size: 1, sequence length: 12006 Attention mask shape: torch.Size([1, 1, 12006, 12006]) Position ids shape: torch.Size([1, 12006]) Input IDs shape: torch.Size([1, 12006]) Labels shape: torch.Size([1, 12006]) Final batch size: 1, sequence length: 18836 Attention mask shape: torch.Size([1, 1, 18836, 18836]) Position ids shape: torch.Size([1, 18836]) Input IDs shape: torch.Size([1, 18836]) Labels shape: torch.Size([1, 18836]) Final batch size: 1, sequence length: 18395 Attention mask shape: torch.Size([1, 1, 18395, 18395]) Position ids shape: torch.Size([1, 18395]) Input IDs shape: torch.Size([1, 18395]) Labels shape: torch.Size([1, 18395]) Final batch size: 1, sequence length: 20854 Attention mask shape: torch.Size([1, 1, 20854, 20854]) Position ids shape: torch.Size([1, 20854]) Input IDs shape: torch.Size([1, 20854]) Labels shape: torch.Size([1, 20854]) Final batch size: 1, sequence length: 18823 Attention mask shape: torch.Size([1, 1, 18823, 18823]) Position ids shape: torch.Size([1, 18823]) Input IDs shape: torch.Size([1, 18823]) Labels shape: torch.Size([1, 18823]) Final batch size: 1, sequence length: 21982 Attention mask shape: torch.Size([1, 1, 21982, 21982]) Position ids shape: torch.Size([1, 21982]) Input IDs shape: torch.Size([1, 21982]) Labels shape: torch.Size([1, 21982]) Final batch size: 1, sequence length: 18325 Attention mask shape: torch.Size([1, 1, 18325, 18325]) Position ids shape: torch.Size([1, 18325]) Input IDs shape: torch.Size([1, 18325]) Labels shape: torch.Size([1, 18325]) Final batch size: 1, sequence length: 21858 Attention mask shape: torch.Size([1, 1, 21858, 21858]) Position ids shape: torch.Size([1, 21858]) Input IDs shape: torch.Size([1, 21858]) Labels shape: torch.Size([1, 21858]) Final batch size: 1, sequence length: 15913 Attention mask shape: torch.Size([1, 1, 15913, 15913]) Position ids shape: torch.Size([1, 15913]) Input IDs shape: torch.Size([1, 15913]) Labels shape: torch.Size([1, 15913]) Final batch size: 1, sequence length: 21405 Attention mask shape: torch.Size([1, 1, 21405, 21405]) Position ids shape: torch.Size([1, 21405]) Input IDs shape: torch.Size([1, 21405]) Labels shape: torch.Size([1, 21405]) Final batch size: 1, sequence length: 23560 Attention mask shape: torch.Size([1, 1, 23560, 23560]) Position ids shape: torch.Size([1, 23560]) Input IDs shape: torch.Size([1, 23560]) Labels shape: torch.Size([1, 23560]) Final batch size: 1, sequence length: 24248 Attention mask shape: torch.Size([1, 1, 24248, 24248]) Position ids shape: torch.Size([1, 24248]) Input IDs shape: torch.Size([1, 24248]) Labels shape: torch.Size([1, 24248]) Final batch size: 1, sequence length: 25451 Attention mask shape: torch.Size([1, 1, 25451, 25451]) Position ids shape: torch.Size([1, 25451]) Input IDs shape: torch.Size([1, 25451]) Labels shape: torch.Size([1, 25451]) Final batch size: 1, sequence length: 24433 Attention mask shape: torch.Size([1, 1, 24433, 24433]) Position ids shape: torch.Size([1, 24433]) Input IDs shape: torch.Size([1, 24433]) Labels shape: torch.Size([1, 24433]) Final batch size: 1, sequence length: 23694 Attention mask shape: torch.Size([1, 1, 23694, 23694]) Position ids shape: torch.Size([1, 23694]) Input IDs shape: torch.Size([1, 23694]) Labels shape: torch.Size([1, 23694]) Final batch size: 1, sequence length: 26356 Attention mask shape: torch.Size([1, 1, 26356, 26356]) Position ids shape: torch.Size([1, 26356]) Input IDs shape: torch.Size([1, 26356]) Labels shape: torch.Size([1, 26356]) Final batch size: 1, sequence length: 22735 Attention mask shape: torch.Size([1, 1, 22735, 22735]) Position ids shape: torch.Size([1, 22735]) Input IDs shape: torch.Size([1, 22735]) Labels shape: torch.Size([1, 22735]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 16060 Attention mask shape: torch.Size([1, 1, 16060, 16060]) Position ids shape: torch.Size([1, 16060]) Input IDs shape: torch.Size([1, 16060]) Labels shape: torch.Size([1, 16060]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 29098 Attention mask shape: torch.Size([1, 1, 29098, 29098]) Position ids shape: torch.Size([1, 29098]) Input IDs shape: torch.Size([1, 29098]) Labels shape: torch.Size([1, 29098]) Final batch size: 1, sequence length: 5801 Attention mask shape: torch.Size([1, 1, 5801, 5801]) Position ids shape: torch.Size([1, 5801]) Input IDs shape: torch.Size([1, 5801]) Labels shape: torch.Size([1, 5801]) Final batch size: 1, sequence length: 9380 Attention mask shape: torch.Size([1, 1, 9380, 9380]) Position ids shape: torch.Size([1, 9380]) Input IDs shape: torch.Size([1, 9380]) Labels shape: torch.Size([1, 9380]) Final batch size: 1, sequence length: 28684 Attention mask shape: torch.Size([1, 1, 28684, 28684]) Position ids shape: torch.Size([1, 28684]) Input IDs shape: torch.Size([1, 28684]) Labels shape: torch.Size([1, 28684]) Final batch size: 1, sequence length: 21615 Attention mask shape: torch.Size([1, 1, 21615, 21615]) Position ids shape: torch.Size([1, 21615]) Input IDs shape: torch.Size([1, 21615]) Labels shape: torch.Size([1, 21615]) Final batch size: 1, sequence length: 31414 Attention mask shape: torch.Size([1, 1, 31414, 31414]) Position ids shape: torch.Size([1, 31414]) Input IDs shape: torch.Size([1, 31414]) Labels shape: torch.Size([1, 31414]) Final batch size: 1, sequence length: 26072 Attention mask shape: torch.Size([1, 1, 26072, 26072]) Position ids shape: torch.Size([1, 26072]) Input IDs shape: torch.Size([1, 26072]) Labels shape: torch.Size([1, 26072]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 32328 Attention mask shape: torch.Size([1, 1, 32328, 32328]) Position ids shape: torch.Size([1, 32328]) Input IDs shape: torch.Size([1, 32328]) Labels shape: torch.Size([1, 32328]) Final batch size: 1, sequence length: 19239 Attention mask shape: torch.Size([1, 1, 19239, 19239]) Position ids shape: torch.Size([1, 19239]) Input IDs shape: torch.Size([1, 19239]) Labels shape: torch.Size([1, 19239]) Final batch size: 1, sequence length: 19045 Attention mask shape: torch.Size([1, 1, 19045, 19045]) Position ids shape: torch.Size([1, 19045]) Input IDs shape: torch.Size([1, 19045]) Labels shape: torch.Size([1, 19045]) Final batch size: 1, sequence length: 32885 Attention mask shape: torch.Size([1, 1, 32885, 32885]) Position ids shape: torch.Size([1, 32885]) Input IDs shape: torch.Size([1, 32885]) Labels shape: torch.Size([1, 32885]) Final batch size: 1, sequence length: 23881 Attention mask shape: torch.Size([1, 1, 23881, 23881]) Position ids shape: torch.Size([1, 23881]) Input IDs shape: torch.Size([1, 23881]) Labels shape: torch.Size([1, 23881]) Final batch size: 1, sequence length: 29355 Attention mask shape: torch.Size([1, 1, 29355, 29355]) Position ids shape: torch.Size([1, 29355]) Input IDs shape: torch.Size([1, 29355]) Labels shape: torch.Size([1, 29355]) Final batch size: 1, sequence length: 17465 Attention mask shape: torch.Size([1, 1, 17465, 17465]) Position ids shape: torch.Size([1, 17465]) Input IDs shape: torch.Size([1, 17465]) Labels shape: torch.Size([1, 17465]) Final batch size: 1, sequence length: 30687 Attention mask shape: torch.Size([1, 1, 30687, 30687]) Position ids shape: torch.Size([1, 30687]) Input IDs shape: torch.Size([1, 30687]) Labels shape: torch.Size([1, 30687]) Final batch size: 1, sequence length: 15875 Attention mask shape: torch.Size([1, 1, 15875, 15875]) Position ids shape: torch.Size([1, 15875]) Input IDs shape: torch.Size([1, 15875]) Labels shape: torch.Size([1, 15875]) Final batch size: 1, sequence length: 30689 Attention mask shape: torch.Size([1, 1, 30689, 30689]) Position ids shape: torch.Size([1, 30689]) Input IDs shape: torch.Size([1, 30689]) Labels shape: torch.Size([1, 30689]) Final batch size: 1, sequence length: 33871 Attention mask shape: torch.Size([1, 1, 33871, 33871]) Position ids shape: torch.Size([1, 33871]) Input IDs shape: torch.Size([1, 33871]) Labels shape: torch.Size([1, 33871]) Final batch size: 1, sequence length: 27972 Attention mask shape: torch.Size([1, 1, 27972, 27972]) Position ids shape: torch.Size([1, 27972]) Input IDs shape: torch.Size([1, 27972]) Labels shape: torch.Size([1, 27972]) Final batch size: 1, sequence length: 34921 Attention mask shape: torch.Size([1, 1, 34921, 34921]) Position ids shape: torch.Size([1, 34921]) Input IDs shape: torch.Size([1, 34921]) Labels shape: torch.Size([1, 34921]) Final batch size: 1, sequence length: 16050 Attention mask shape: torch.Size([1, 1, 16050, 16050]) Position ids shape: torch.Size([1, 16050]) Input IDs shape: torch.Size([1, 16050]) Labels shape: torch.Size([1, 16050]) Final batch size: 1, sequence length: 32529 Attention mask shape: torch.Size([1, 1, 32529, 32529]) Position ids shape: torch.Size([1, 32529]) Input IDs shape: torch.Size([1, 32529]) Labels shape: torch.Size([1, 32529]) Final batch size: 1, sequence length: 35397 Attention mask shape: torch.Size([1, 1, 35397, 35397]) Position ids shape: torch.Size([1, 35397]) Input IDs shape: torch.Size([1, 35397]) Labels shape: torch.Size([1, 35397]) Final batch size: 1, sequence length: 30346 Attention mask shape: torch.Size([1, 1, 30346, 30346]) Position ids shape: torch.Size([1, 30346]) Input IDs shape: torch.Size([1, 30346]) Labels shape: torch.Size([1, 30346]) Final batch size: 1, sequence length: 29150 Attention mask shape: torch.Size([1, 1, 29150, 29150]) Position ids shape: torch.Size([1, 29150]) Input IDs shape: torch.Size([1, 29150]) Labels shape: torch.Size([1, 29150]) Final batch size: 1, sequence length: 32636 Attention mask shape: torch.Size([1, 1, 32636, 32636]) Position ids shape: torch.Size([1, 32636]) Input IDs shape: torch.Size([1, 32636]) Labels shape: torch.Size([1, 32636]) Final batch size: 1, sequence length: 31860 Attention mask shape: torch.Size([1, 1, 31860, 31860]) Position ids shape: torch.Size([1, 31860]) Input IDs shape: torch.Size([1, 31860]) Labels shape: torch.Size([1, 31860]) Final batch size: 1, sequence length: 26689 Attention mask shape: torch.Size([1, 1, 26689, 26689]) Position ids shape: torch.Size([1, 26689]) Input IDs shape: torch.Size([1, 26689]) Labels shape: torch.Size([1, 26689]) Final batch size: 1, sequence length: 34581 Attention mask shape: torch.Size([1, 1, 34581, 34581]) Position ids shape: torch.Size([1, 34581]) Input IDs shape: torch.Size([1, 34581]) Labels shape: torch.Size([1, 34581]) Final batch size: 1, sequence length: 31712 Attention mask shape: torch.Size([1, 1, 31712, 31712]) Position ids shape: torch.Size([1, 31712]) Input IDs shape: torch.Size([1, 31712]) Labels shape: torch.Size([1, 31712]) Final batch size: 1, sequence length: 24781 Attention mask shape: torch.Size([1, 1, 24781, 24781]) Position ids shape: torch.Size([1, 24781]) Input IDs shape: torch.Size([1, 24781]) Labels shape: torch.Size([1, 24781]) Final batch size: 1, sequence length: 15077 Attention mask shape: torch.Size([1, 1, 15077, 15077]) Position ids shape: torch.Size([1, 15077]) Input IDs shape: torch.Size([1, 15077]) Labels shape: torch.Size([1, 15077]) Final batch size: 1, sequence length: 22264 Attention mask shape: torch.Size([1, 1, 22264, 22264]) Position ids shape: torch.Size([1, 22264]) Input IDs shape: torch.Size([1, 22264]) Labels shape: torch.Size([1, 22264]) Final batch size: 1, sequence length: 18963 Attention mask shape: torch.Size([1, 1, 18963, 18963]) Position ids shape: torch.Size([1, 18963]) Input IDs shape: torch.Size([1, 18963]) Labels shape: torch.Size([1, 18963]) Final batch size: 1, sequence length: 14976 Attention mask shape: torch.Size([1, 1, 14976, 14976]) Position ids shape: torch.Size([1, 14976]) Input IDs shape: torch.Size([1, 14976]) Labels shape: torch.Size([1, 14976]) Final batch size: 1, sequence length: 17243 Attention mask shape: torch.Size([1, 1, 17243, 17243]) Position ids shape: torch.Size([1, 17243]) Input IDs shape: torch.Size([1, 17243]) Labels shape: torch.Size([1, 17243]) Final batch size: 1, sequence length: 18177 Attention mask shape: torch.Size([1, 1, 18177, 18177]) Position ids shape: torch.Size([1, 18177]) Input IDs shape: torch.Size([1, 18177]) Labels shape: torch.Size([1, 18177]) Final batch size: 1, sequence length: 16025 Attention mask shape: torch.Size([1, 1, 16025, 16025]) Position ids shape: torch.Size([1, 16025]) Input IDs shape: torch.Size([1, 16025]) Labels shape: torch.Size([1, 16025]) Final batch size: 1, sequence length: 25741 Attention mask shape: torch.Size([1, 1, 25741, 25741]) Position ids shape: torch.Size([1, 25741]) Input IDs shape: torch.Size([1, 25741]) Labels shape: torch.Size([1, 25741]) Final batch size: 1, sequence length: 28018 Attention mask shape: torch.Size([1, 1, 28018, 28018]) Position ids shape: torch.Size([1, 28018]) Input IDs shape: torch.Size([1, 28018]) Labels shape: torch.Size([1, 28018]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31022 Attention mask shape: torch.Size([1, 1, 31022, 31022]) Position ids shape: torch.Size([1, 31022]) Input IDs shape: torch.Size([1, 31022]) Labels shape: torch.Size([1, 31022]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24002 Attention mask shape: torch.Size([1, 1, 24002, 24002]) Position ids shape: torch.Size([1, 24002]) Input IDs shape: torch.Size([1, 24002]) Labels shape: torch.Size([1, 24002]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40488 Attention mask shape: torch.Size([1, 1, 40488, 40488]) Position ids shape: torch.Size([1, 40488]) Input IDs shape: torch.Size([1, 40488]) Labels shape: torch.Size([1, 40488]) Final batch size: 1, sequence length: 26922 Attention mask shape: torch.Size([1, 1, 26922, 26922]) Position ids shape: torch.Size([1, 26922]) Input IDs shape: torch.Size([1, 26922]) Labels shape: torch.Size([1, 26922]) Final batch size: 1, sequence length: 25674 Attention mask shape: torch.Size([1, 1, 25674, 25674]) Position ids shape: torch.Size([1, 25674]) Input IDs shape: torch.Size([1, 25674]) Labels shape: torch.Size([1, 25674]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36596 Attention mask shape: torch.Size([1, 1, 36596, 36596]) Position ids shape: torch.Size([1, 36596]) Input IDs shape: torch.Size([1, 36596]) Labels shape: torch.Size([1, 36596]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 11611 Attention mask shape: torch.Size([1, 1, 11611, 11611]) Position ids shape: torch.Size([1, 11611]) Input IDs shape: torch.Size([1, 11611]) Labels shape: torch.Size([1, 11611]) Final batch size: 1, sequence length: 28397 Attention mask shape: torch.Size([1, 1, 28397, 28397]) Position ids shape: torch.Size([1, 28397]) Input IDs shape: torch.Size([1, 28397]) Labels shape: torch.Size([1, 28397]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24043 Attention mask shape: torch.Size([1, 1, 24043, 24043]) Position ids shape: torch.Size([1, 24043]) Input IDs shape: torch.Size([1, 24043]) Labels shape: torch.Size([1, 24043]) Final batch size: 1, sequence length: 34853 Attention mask shape: torch.Size([1, 1, 34853, 34853]) Position ids shape: torch.Size([1, 34853]) Input IDs shape: torch.Size([1, 34853]) Labels shape: torch.Size([1, 34853]) Final batch size: 1, sequence length: 16337 Attention mask shape: torch.Size([1, 1, 16337, 16337]) Position ids shape: torch.Size([1, 16337]) Input IDs shape: torch.Size([1, 16337]) Labels shape: torch.Size([1, 16337]) Final batch size: 1, sequence length: 28086 Attention mask shape: torch.Size([1, 1, 28086, 28086]) Position ids shape: torch.Size([1, 28086]) Input IDs shape: torch.Size([1, 28086]) Labels shape: torch.Size([1, 28086]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 12218 Attention mask shape: torch.Size([1, 1, 12218, 12218]) Position ids shape: torch.Size([1, 12218]) Input IDs shape: torch.Size([1, 12218]) Labels shape: torch.Size([1, 12218]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36334 Attention mask shape: torch.Size([1, 1, 36334, 36334]) Position ids shape: torch.Size([1, 36334]) Input IDs shape: torch.Size([1, 36334]) Labels shape: torch.Size([1, 36334]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32917 Attention mask shape: torch.Size([1, 1, 32917, 32917]) Position ids shape: torch.Size([1, 32917]) Input IDs shape: torch.Size([1, 32917]) Labels shape: torch.Size([1, 32917]) {'loss': 0.2628, 'grad_norm': 0.29879438601593994, 'learning_rate': 2.966316784621e-06, 'num_tokens': -inf, 'epoch': 5.38} Final batch size: 1, sequence length: 5818 Attention mask shape: torch.Size([1, 1, 5818, 5818]) Position ids shape: torch.Size([1, 5818]) Input IDs shape: torch.Size([1, 5818]) Labels shape: torch.Size([1, 5818]) Final batch size: 1, sequence length: 8623 Attention mask shape: torch.Size([1, 1, 8623, 8623]) Position ids shape: torch.Size([1, 8623]) Input IDs shape: torch.Size([1, 8623]) Labels shape: torch.Size([1, 8623]) Final batch size: 1, sequence length: 6871 Attention mask shape: torch.Size([1, 1, 6871, 6871]) Position ids shape: torch.Size([1, 6871]) Input IDs shape: torch.Size([1, 6871]) Labels shape: torch.Size([1, 6871]) Final batch size: 1, sequence length: 6034 Attention mask shape: torch.Size([1, 1, 6034, 6034]) Position ids shape: torch.Size([1, 6034]) Input IDs shape: torch.Size([1, 6034]) Labels shape: torch.Size([1, 6034]) Final batch size: 1, sequence length: 11616 Attention mask shape: torch.Size([1, 1, 11616, 11616]) Position ids shape: torch.Size([1, 11616]) Input IDs shape: torch.Size([1, 11616]) Labels shape: torch.Size([1, 11616]) Final batch size: 1, sequence length: 12275 Attention mask shape: torch.Size([1, 1, 12275, 12275]) Position ids shape: torch.Size([1, 12275]) Input IDs shape: torch.Size([1, 12275]) Labels shape: torch.Size([1, 12275]) Final batch size: 1, sequence length: 6215 Attention mask shape: torch.Size([1, 1, 6215, 6215]) Position ids shape: torch.Size([1, 6215]) Input IDs shape: torch.Size([1, 6215]) Labels shape: torch.Size([1, 6215]) Final batch size: 1, sequence length: 10107 Attention mask shape: torch.Size([1, 1, 10107, 10107]) Position ids shape: torch.Size([1, 10107]) Input IDs shape: torch.Size([1, 10107]) Labels shape: torch.Size([1, 10107]) Final batch size: 1, sequence length: 13202 Attention mask shape: torch.Size([1, 1, 13202, 13202]) Position ids shape: torch.Size([1, 13202]) Input IDs shape: torch.Size([1, 13202]) Labels shape: torch.Size([1, 13202]) Final batch size: 1, sequence length: 9517 Attention mask shape: torch.Size([1, 1, 9517, 9517]) Position ids shape: torch.Size([1, 9517]) Input IDs shape: torch.Size([1, 9517]) Labels shape: torch.Size([1, 9517]) Final batch size: 1, sequence length: 5917 Attention mask shape: torch.Size([1, 1, 5917, 5917]) Position ids shape: torch.Size([1, 5917]) Input IDs shape: torch.Size([1, 5917]) Labels shape: torch.Size([1, 5917]) Final batch size: 1, sequence length: 12813 Attention mask shape: torch.Size([1, 1, 12813, 12813]) Position ids shape: torch.Size([1, 12813]) Input IDs shape: torch.Size([1, 12813]) Labels shape: torch.Size([1, 12813]) Final batch size: 1, sequence length: 11648 Attention mask shape: torch.Size([1, 1, 11648, 11648]) Position ids shape: torch.Size([1, 11648]) Input IDs shape: torch.Size([1, 11648]) Labels shape: torch.Size([1, 11648]) Final batch size: 1, sequence length: 8117 Attention mask shape: torch.Size([1, 1, 8117, 8117]) Position ids shape: torch.Size([1, 8117]) Input IDs shape: torch.Size([1, 8117]) Labels shape: torch.Size([1, 8117]) Final batch size: 1, sequence length: 15029 Attention mask shape: torch.Size([1, 1, 15029, 15029]) Position ids shape: torch.Size([1, 15029]) Input IDs shape: torch.Size([1, 15029]) Labels shape: torch.Size([1, 15029]) Final batch size: 1, sequence length: 12317 Attention mask shape: torch.Size([1, 1, 12317, 12317]) Position ids shape: torch.Size([1, 12317]) Input IDs shape: torch.Size([1, 12317]) Labels shape: torch.Size([1, 12317]) Final batch size: 1, sequence length: 13428 Attention mask shape: torch.Size([1, 1, 13428, 13428]) Position ids shape: torch.Size([1, 13428]) Input IDs shape: torch.Size([1, 13428]) Labels shape: torch.Size([1, 13428]) Final batch size: 1, sequence length: 15066 Attention mask shape: torch.Size([1, 1, 15066, 15066]) Position ids shape: torch.Size([1, 15066]) Input IDs shape: torch.Size([1, 15066]) Labels shape: torch.Size([1, 15066]) Final batch size: 1, sequence length: 15499 Attention mask shape: torch.Size([1, 1, 15499, 15499]) Position ids shape: torch.Size([1, 15499]) Input IDs shape: torch.Size([1, 15499]) Labels shape: torch.Size([1, 15499]) Final batch size: 1, sequence length: 10136 Attention mask shape: torch.Size([1, 1, 10136, 10136]) Position ids shape: torch.Size([1, 10136]) Input IDs shape: torch.Size([1, 10136]) Labels shape: torch.Size([1, 10136]) Final batch size: 1, sequence length: 12553 Attention mask shape: torch.Size([1, 1, 12553, 12553]) Position ids shape: torch.Size([1, 12553]) Input IDs shape: torch.Size([1, 12553]) Labels shape: torch.Size([1, 12553]) Final batch size: 1, sequence length: 16331 Attention mask shape: torch.Size([1, 1, 16331, 16331]) Position ids shape: torch.Size([1, 16331]) Input IDs shape: torch.Size([1, 16331]) Labels shape: torch.Size([1, 16331]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 18414 Attention mask shape: torch.Size([1, 1, 18414, 18414]) Position ids shape: torch.Size([1, 18414]) Input IDs shape: torch.Size([1, 18414]) Labels shape: torch.Size([1, 18414]) Final batch size: 1, sequence length: 19847 Attention mask shape: torch.Size([1, 1, 19847, 19847]) Position ids shape: torch.Size([1, 19847]) Input IDs shape: torch.Size([1, 19847]) Labels shape: torch.Size([1, 19847]) Final batch size: 1, sequence length: 19512 Final batch size: 1, sequence length: 21028 Attention mask shape: torch.Size([1, 1, 19512, 19512]) Position ids shape: torch.Size([1, 19512]) Input IDs shape: torch.Size([1, 19512]) Labels shape: torch.Size([1, 19512]) Attention mask shape: torch.Size([1, 1, 21028, 21028]) Position ids shape: torch.Size([1, 21028]) Input IDs shape: torch.Size([1, 21028]) Labels shape: torch.Size([1, 21028]) Final batch size: 1, sequence length: 21556 Attention mask shape: torch.Size([1, 1, 21556, 21556]) Position ids shape: torch.Size([1, 21556]) Input IDs shape: torch.Size([1, 21556]) Labels shape: torch.Size([1, 21556]) Final batch size: 1, sequence length: 18438 Attention mask shape: torch.Size([1, 1, 18438, 18438]) Position ids shape: torch.Size([1, 18438]) Input IDs shape: torch.Size([1, 18438]) Labels shape: torch.Size([1, 18438]) Final batch size: 1, sequence length: 17587 Attention mask shape: torch.Size([1, 1, 17587, 17587]) Position ids shape: torch.Size([1, 17587]) Input IDs shape: torch.Size([1, 17587]) Labels shape: torch.Size([1, 17587]) Final batch size: 1, sequence length: 22079 Attention mask shape: torch.Size([1, 1, 22079, 22079]) Position ids shape: torch.Size([1, 22079]) Input IDs shape: torch.Size([1, 22079]) Labels shape: torch.Size([1, 22079]) Final batch size: 1, sequence length: 18469 Attention mask shape: torch.Size([1, 1, 18469, 18469]) Position ids shape: torch.Size([1, 18469]) Input IDs shape: torch.Size([1, 18469]) Labels shape: torch.Size([1, 18469]) Final batch size: 1, sequence length: 19221 Attention mask shape: torch.Size([1, 1, 19221, 19221]) Position ids shape: torch.Size([1, 19221]) Input IDs shape: torch.Size([1, 19221]) Labels shape: torch.Size([1, 19221]) Final batch size: 1, sequence length: 22138 Attention mask shape: torch.Size([1, 1, 22138, 22138]) Position ids shape: torch.Size([1, 22138]) Input IDs shape: torch.Size([1, 22138]) Labels shape: torch.Size([1, 22138]) Final batch size: 1, sequence length: 15226 Attention mask shape: torch.Size([1, 1, 15226, 15226]) Position ids shape: torch.Size([1, 15226]) Input IDs shape: torch.Size([1, 15226]) Labels shape: torch.Size([1, 15226]) Final batch size: 1, sequence length: 15221 Attention mask shape: torch.Size([1, 1, 15221, 15221]) Position ids shape: torch.Size([1, 15221]) Input IDs shape: torch.Size([1, 15221]) Labels shape: torch.Size([1, 15221]) Final batch size: 1, sequence length: 22618 Attention mask shape: torch.Size([1, 1, 22618, 22618]) Position ids shape: torch.Size([1, 22618]) Input IDs shape: torch.Size([1, 22618]) Labels shape: torch.Size([1, 22618]) Final batch size: 1, sequence length: 23766 Attention mask shape: torch.Size([1, 1, 23766, 23766]) Position ids shape: torch.Size([1, 23766]) Input IDs shape: torch.Size([1, 23766]) Labels shape: torch.Size([1, 23766]) Final batch size: 1, sequence length: 23973 Attention mask shape: torch.Size([1, 1, 23973, 23973]) Position ids shape: torch.Size([1, 23973]) Input IDs shape: torch.Size([1, 23973]) Labels shape: torch.Size([1, 23973]) Final batch size: 1, sequence length: 24499 Attention mask shape: torch.Size([1, 1, 24499, 24499]) Position ids shape: torch.Size([1, 24499]) Input IDs shape: torch.Size([1, 24499]) Labels shape: torch.Size([1, 24499]) Final batch size: 1, sequence length: 25622 Attention mask shape: torch.Size([1, 1, 25622, 25622]) Position ids shape: torch.Size([1, 25622]) Input IDs shape: torch.Size([1, 25622]) Labels shape: torch.Size([1, 25622]) Final batch size: 1, sequence length: 22718 Attention mask shape: torch.Size([1, 1, 22718, 22718]) Position ids shape: torch.Size([1, 22718]) Input IDs shape: torch.Size([1, 22718]) Labels shape: torch.Size([1, 22718]) Final batch size: 1, sequence length: 19428 Attention mask shape: torch.Size([1, 1, 19428, 19428]) Position ids shape: torch.Size([1, 19428]) Input IDs shape: torch.Size([1, 19428]) Labels shape: torch.Size([1, 19428]) Final batch size: 1, sequence length: 24694 Attention mask shape: torch.Size([1, 1, 24694, 24694]) Position ids shape: torch.Size([1, 24694]) Input IDs shape: torch.Size([1, 24694]) Labels shape: torch.Size([1, 24694]) Final batch size: 1, sequence length: 25999 Attention mask shape: torch.Size([1, 1, 25999, 25999]) Position ids shape: torch.Size([1, 25999]) Input IDs shape: torch.Size([1, 25999]) Labels shape: torch.Size([1, 25999]) Final batch size: 1, sequence length: 16299 Attention mask shape: torch.Size([1, 1, 16299, 16299]) Position ids shape: torch.Size([1, 16299]) Input IDs shape: torch.Size([1, 16299]) Labels shape: torch.Size([1, 16299]) Final batch size: 1, sequence length: 7584 Attention mask shape: torch.Size([1, 1, 7584, 7584]) Position ids shape: torch.Size([1, 7584]) Input IDs shape: torch.Size([1, 7584]) Labels shape: torch.Size([1, 7584]) Final batch size: 1, sequence length: 27566 Attention mask shape: torch.Size([1, 1, 27566, 27566]) Position ids shape: torch.Size([1, 27566]) Input IDs shape: torch.Size([1, 27566]) Labels shape: torch.Size([1, 27566]) Final batch size: 1, sequence length: 14784 Attention mask shape: torch.Size([1, 1, 14784, 14784]) Position ids shape: torch.Size([1, 14784]) Input IDs shape: torch.Size([1, 14784]) Labels shape: torch.Size([1, 14784]) Final batch size: 1, sequence length: 24909 Attention mask shape: torch.Size([1, 1, 24909, 24909]) Position ids shape: torch.Size([1, 24909]) Input IDs shape: torch.Size([1, 24909]) Labels shape: torch.Size([1, 24909]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 22235 Attention mask shape: torch.Size([1, 1, 22235, 22235]) Position ids shape: torch.Size([1, 22235]) Input IDs shape: torch.Size([1, 22235]) Labels shape: torch.Size([1, 22235]) Final batch size: 1, sequence length: 28497 Attention mask shape: torch.Size([1, 1, 28497, 28497]) Position ids shape: torch.Size([1, 28497]) Input IDs shape: torch.Size([1, 28497]) Labels shape: torch.Size([1, 28497]) Final batch size: 1, sequence length: 22763 Attention mask shape: torch.Size([1, 1, 22763, 22763]) Position ids shape: torch.Size([1, 22763]) Input IDs shape: torch.Size([1, 22763]) Labels shape: torch.Size([1, 22763]) Final batch size: 1, sequence length: 28749 Attention mask shape: torch.Size([1, 1, 28749, 28749]) Position ids shape: torch.Size([1, 28749]) Input IDs shape: torch.Size([1, 28749]) Labels shape: torch.Size([1, 28749]) Final batch size: 1, sequence length: 15184 Attention mask shape: torch.Size([1, 1, 15184, 15184]) Position ids shape: torch.Size([1, 15184]) Input IDs shape: torch.Size([1, 15184]) Labels shape: torch.Size([1, 15184]) Final batch size: 1, sequence length: 27327 Attention mask shape: torch.Size([1, 1, 27327, 27327]) Position ids shape: torch.Size([1, 27327]) Input IDs shape: torch.Size([1, 27327]) Labels shape: torch.Size([1, 27327]) Final batch size: 1, sequence length: 21826 Attention mask shape: torch.Size([1, 1, 21826, 21826]) Position ids shape: torch.Size([1, 21826]) Input IDs shape: torch.Size([1, 21826]) Labels shape: torch.Size([1, 21826]) Final batch size: 1, sequence length: 15588 Attention mask shape: torch.Size([1, 1, 15588, 15588]) Position ids shape: torch.Size([1, 15588]) Input IDs shape: torch.Size([1, 15588]) Labels shape: torch.Size([1, 15588]) Final batch size: 1, sequence length: 10269 Attention mask shape: torch.Size([1, 1, 10269, 10269]) Position ids shape: torch.Size([1, 10269]) Input IDs shape: torch.Size([1, 10269]) Labels shape: torch.Size([1, 10269]) Final batch size: 1, sequence length: 30356 Attention mask shape: torch.Size([1, 1, 30356, 30356]) Position ids shape: torch.Size([1, 30356]) Input IDs shape: torch.Size([1, 30356]) Labels shape: torch.Size([1, 30356]) Final batch size: 1, sequence length: 22366 Attention mask shape: torch.Size([1, 1, 22366, 22366]) Position ids shape: torch.Size([1, 22366]) Input IDs shape: torch.Size([1, 22366]) Labels shape: torch.Size([1, 22366]) Final batch size: 1, sequence length: 23851 Attention mask shape: torch.Size([1, 1, 23851, 23851]) Position ids shape: torch.Size([1, 23851]) Input IDs shape: torch.Size([1, 23851]) Labels shape: torch.Size([1, 23851]) Final batch size: 1, sequence length: 11795 Attention mask shape: torch.Size([1, 1, 11795, 11795]) Position ids shape: torch.Size([1, 11795]) Input IDs shape: torch.Size([1, 11795]) Labels shape: torch.Size([1, 11795]) Final batch size: 1, sequence length: 21034 Attention mask shape: torch.Size([1, 1, 21034, 21034]) Position ids shape: torch.Size([1, 21034]) Input IDs shape: torch.Size([1, 21034]) Labels shape: torch.Size([1, 21034]) Final batch size: 1, sequence length: 34673 Attention mask shape: torch.Size([1, 1, 34673, 34673]) Position ids shape: torch.Size([1, 34673]) Input IDs shape: torch.Size([1, 34673]) Labels shape: torch.Size([1, 34673]) Final batch size: 1, sequence length: 36777 Attention mask shape: torch.Size([1, 1, 36777, 36777]) Position ids shape: torch.Size([1, 36777]) Input IDs shape: torch.Size([1, 36777]) Labels shape: torch.Size([1, 36777]) Final batch size: 1, sequence length: 20755 Attention mask shape: torch.Size([1, 1, 20755, 20755]) Position ids shape: torch.Size([1, 20755]) Input IDs shape: torch.Size([1, 20755]) Labels shape: torch.Size([1, 20755]) Final batch size: 1, sequence length: 27489 Attention mask shape: torch.Size([1, 1, 27489, 27489]) Position ids shape: torch.Size([1, 27489]) Input IDs shape: torch.Size([1, 27489]) Labels shape: torch.Size([1, 27489]) Final batch size: 1, sequence length: 31903 Attention mask shape: torch.Size([1, 1, 31903, 31903]) Position ids shape: torch.Size([1, 31903]) Input IDs shape: torch.Size([1, 31903]) Labels shape: torch.Size([1, 31903]) Final batch size: 1, sequence length: 36723 Attention mask shape: torch.Size([1, 1, 36723, 36723]) Position ids shape: torch.Size([1, 36723]) Input IDs shape: torch.Size([1, 36723]) Labels shape: torch.Size([1, 36723]) Final batch size: 1, sequence length: 37285 Attention mask shape: torch.Size([1, 1, 37285, 37285]) Position ids shape: torch.Size([1, 37285]) Input IDs shape: torch.Size([1, 37285]) Labels shape: torch.Size([1, 37285]) Final batch size: 1, sequence length: 38848 Attention mask shape: torch.Size([1, 1, 38848, 38848]) Position ids shape: torch.Size([1, 38848]) Input IDs shape: torch.Size([1, 38848]) Labels shape: torch.Size([1, 38848]) Final batch size: 1, sequence length: 28176 Attention mask shape: torch.Size([1, 1, 28176, 28176]) Position ids shape: torch.Size([1, 28176]) Input IDs shape: torch.Size([1, 28176]) Labels shape: torch.Size([1, 28176]) Final batch size: 1, sequence length: 40325 Attention mask shape: torch.Size([1, 1, 40325, 40325]) Position ids shape: torch.Size([1, 40325]) Input IDs shape: torch.Size([1, 40325]) Labels shape: torch.Size([1, 40325]) Final batch size: 1, sequence length: 39253 Attention mask shape: torch.Size([1, 1, 39253, 39253]) Position ids shape: torch.Size([1, 39253]) Input IDs shape: torch.Size([1, 39253]) Labels shape: torch.Size([1, 39253]) Final batch size: 1, sequence length: 38049 Attention mask shape: torch.Size([1, 1, 38049, 38049]) Position ids shape: torch.Size([1, 38049]) Input IDs shape: torch.Size([1, 38049]) Labels shape: torch.Size([1, 38049]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25946 Attention mask shape: torch.Size([1, 1, 25946, 25946]) Position ids shape: torch.Size([1, 25946]) Input IDs shape: torch.Size([1, 25946]) Labels shape: torch.Size([1, 25946]) Final batch size: 1, sequence length: 6882 Attention mask shape: torch.Size([1, 1, 6882, 6882]) Position ids shape: torch.Size([1, 6882]) Input IDs shape: torch.Size([1, 6882]) Labels shape: torch.Size([1, 6882]) Final batch size: 1, sequence length: 24801 Attention mask shape: torch.Size([1, 1, 24801, 24801]) Position ids shape: torch.Size([1, 24801]) Input IDs shape: torch.Size([1, 24801]) Labels shape: torch.Size([1, 24801]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38104 Attention mask shape: torch.Size([1, 1, 38104, 38104]) Position ids shape: torch.Size([1, 38104]) Input IDs shape: torch.Size([1, 38104]) Labels shape: torch.Size([1, 38104]) Final batch size: 1, sequence length: 31714 Attention mask shape: torch.Size([1, 1, 31714, 31714]) Position ids shape: torch.Size([1, 31714]) Input IDs shape: torch.Size([1, 31714]) Labels shape: torch.Size([1, 31714]) Final batch size: 1, sequence length: 34289 Attention mask shape: torch.Size([1, 1, 34289, 34289]) Position ids shape: torch.Size([1, 34289]) Input IDs shape: torch.Size([1, 34289]) Labels shape: torch.Size([1, 34289]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40317 Attention mask shape: torch.Size([1, 1, 40317, 40317]) Position ids shape: torch.Size([1, 40317]) Input IDs shape: torch.Size([1, 40317]) Labels shape: torch.Size([1, 40317]) Final batch size: 1, sequence length: 40937 Attention mask shape: torch.Size([1, 1, 40937, 40937]) Position ids shape: torch.Size([1, 40937]) Input IDs shape: torch.Size([1, 40937]) Labels shape: torch.Size([1, 40937]) Final batch size: 1, sequence length: 36786 Attention mask shape: torch.Size([1, 1, 36786, 36786]) Position ids shape: torch.Size([1, 36786]) Input IDs shape: torch.Size([1, 36786]) Labels shape: torch.Size([1, 36786]) Final batch size: 1, sequence length: 35904 Attention mask shape: torch.Size([1, 1, 35904, 35904]) Position ids shape: torch.Size([1, 35904]) Input IDs shape: torch.Size([1, 35904]) Labels shape: torch.Size([1, 35904]) Final batch size: 1, sequence length: 40657 Attention mask shape: torch.Size([1, 1, 40657, 40657]) Position ids shape: torch.Size([1, 40657]) Input IDs shape: torch.Size([1, 40657]) Labels shape: torch.Size([1, 40657]) Final batch size: 1, sequence length: 10469 Attention mask shape: torch.Size([1, 1, 10469, 10469]) Position ids shape: torch.Size([1, 10469]) Input IDs shape: torch.Size([1, 10469]) Labels shape: torch.Size([1, 10469]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19639 Attention mask shape: torch.Size([1, 1, 19639, 19639]) Position ids shape: torch.Size([1, 19639]) Input IDs shape: torch.Size([1, 19639]) Labels shape: torch.Size([1, 19639]) Final batch size: 1, sequence length: 16649 Attention mask shape: torch.Size([1, 1, 16649, 16649]) Position ids shape: torch.Size([1, 16649]) Input IDs shape: torch.Size([1, 16649]) Labels shape: torch.Size([1, 16649]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19437 Attention mask shape: torch.Size([1, 1, 19437, 19437]) Position ids shape: torch.Size([1, 19437]) Input IDs shape: torch.Size([1, 19437]) Labels shape: torch.Size([1, 19437]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26660 Attention mask shape: torch.Size([1, 1, 26660, 26660]) Position ids shape: torch.Size([1, 26660]) Input IDs shape: torch.Size([1, 26660]) Labels shape: torch.Size([1, 26660]) Final batch size: 1, sequence length: 38686 Attention mask shape: torch.Size([1, 1, 38686, 38686]) Position ids shape: torch.Size([1, 38686]) Input IDs shape: torch.Size([1, 38686]) Labels shape: torch.Size([1, 38686]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36128 Attention mask shape: torch.Size([1, 1, 36128, 36128]) Position ids shape: torch.Size([1, 36128]) Input IDs shape: torch.Size([1, 36128]) Labels shape: torch.Size([1, 36128]) Final batch size: 1, sequence length: 21547 Attention mask shape: torch.Size([1, 1, 21547, 21547]) Position ids shape: torch.Size([1, 21547]) Input IDs shape: torch.Size([1, 21547]) Labels shape: torch.Size([1, 21547]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34540 Attention mask shape: torch.Size([1, 1, 34540, 34540]) Position ids shape: torch.Size([1, 34540]) Input IDs shape: torch.Size([1, 34540]) Labels shape: torch.Size([1, 34540]) Final batch size: 1, sequence length: 17882 Attention mask shape: torch.Size([1, 1, 17882, 17882]) Position ids shape: torch.Size([1, 17882]) Input IDs shape: torch.Size([1, 17882]) Labels shape: torch.Size([1, 17882]) Final batch size: 1, sequence length: 25963 Attention mask shape: torch.Size([1, 1, 25963, 25963]) Position ids shape: torch.Size([1, 25963]) Input IDs shape: torch.Size([1, 25963]) Labels shape: torch.Size([1, 25963]) Final batch size: 1, sequence length: 27947 Attention mask shape: torch.Size([1, 1, 27947, 27947]) Position ids shape: torch.Size([1, 27947]) Input IDs shape: torch.Size([1, 27947]) Labels shape: torch.Size([1, 27947]) Final batch size: 1, sequence length: 17379 Attention mask shape: torch.Size([1, 1, 17379, 17379]) Position ids shape: torch.Size([1, 17379]) Input IDs shape: torch.Size([1, 17379]) Labels shape: torch.Size([1, 17379]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 37946 Attention mask shape: torch.Size([1, 1, 37946, 37946]) Position ids shape: torch.Size([1, 37946]) Input IDs shape: torch.Size([1, 37946]) Labels shape: torch.Size([1, 37946]) Final batch size: 1, sequence length: 20843 Attention mask shape: torch.Size([1, 1, 20843, 20843]) Position ids shape: torch.Size([1, 20843]) Input IDs shape: torch.Size([1, 20843]) Labels shape: torch.Size([1, 20843]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21496 Attention mask shape: torch.Size([1, 1, 21496, 21496]) Position ids shape: torch.Size([1, 21496]) Input IDs shape: torch.Size([1, 21496]) Labels shape: torch.Size([1, 21496]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36317 Attention mask shape: torch.Size([1, 1, 36317, 36317]) Position ids shape: torch.Size([1, 36317]) Input IDs shape: torch.Size([1, 36317]) Labels shape: torch.Size([1, 36317]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24551 Attention mask shape: torch.Size([1, 1, 24551, 24551]) Position ids shape: torch.Size([1, 24551]) Input IDs shape: torch.Size([1, 24551]) Labels shape: torch.Size([1, 24551]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7681 Attention mask shape: torch.Size([1, 1, 7681, 7681]) Position ids shape: torch.Size([1, 7681]) Input IDs shape: torch.Size([1, 7681]) Labels shape: torch.Size([1, 7681]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2658, 'grad_norm': 0.2726003475845015, 'learning_rate': 2.7300475013022666e-06, 'num_tokens': -inf, 'epoch': 5.5} Final batch size: 1, sequence length: 5328 Attention mask shape: torch.Size([1, 1, 5328, 5328]) Position ids shape: torch.Size([1, 5328]) Input IDs shape: torch.Size([1, 5328]) Labels shape: torch.Size([1, 5328]) Final batch size: 1, sequence length: 5525 Attention mask shape: torch.Size([1, 1, 5525, 5525]) Position ids shape: torch.Size([1, 5525]) Input IDs shape: torch.Size([1, 5525]) Labels shape: torch.Size([1, 5525]) Final batch size: 1, sequence length: 10273 Attention mask shape: torch.Size([1, 1, 10273, 10273]) Position ids shape: torch.Size([1, 10273]) Input IDs shape: torch.Size([1, 10273]) Labels shape: torch.Size([1, 10273]) Final batch size: 1, sequence length: 11464 Attention mask shape: torch.Size([1, 1, 11464, 11464]) Position ids shape: torch.Size([1, 11464]) Input IDs shape: torch.Size([1, 11464]) Labels shape: torch.Size([1, 11464]) Final batch size: 1, sequence length: 12419 Attention mask shape: torch.Size([1, 1, 12419, 12419]) Position ids shape: torch.Size([1, 12419]) Input IDs shape: torch.Size([1, 12419]) Labels shape: torch.Size([1, 12419]) Final batch size: 1, sequence length: 10434 Attention mask shape: torch.Size([1, 1, 10434, 10434]) Position ids shape: torch.Size([1, 10434]) Input IDs shape: torch.Size([1, 10434]) Labels shape: torch.Size([1, 10434]) Final batch size: 1, sequence length: 13607 Attention mask shape: torch.Size([1, 1, 13607, 13607]) Position ids shape: torch.Size([1, 13607]) Input IDs shape: torch.Size([1, 13607]) Labels shape: torch.Size([1, 13607]) Final batch size: 1, sequence length: 14226 Attention mask shape: torch.Size([1, 1, 14226, 14226]) Position ids shape: torch.Size([1, 14226]) Input IDs shape: torch.Size([1, 14226]) Labels shape: torch.Size([1, 14226]) Final batch size: 1, sequence length: 9648 Attention mask shape: torch.Size([1, 1, 9648, 9648]) Position ids shape: torch.Size([1, 9648]) Input IDs shape: torch.Size([1, 9648]) Labels shape: torch.Size([1, 9648]) Final batch size: 1, sequence length: 13657 Attention mask shape: torch.Size([1, 1, 13657, 13657]) Position ids shape: torch.Size([1, 13657]) Input IDs shape: torch.Size([1, 13657]) Labels shape: torch.Size([1, 13657]) Final batch size: 1, sequence length: 14360 Attention mask shape: torch.Size([1, 1, 14360, 14360]) Position ids shape: torch.Size([1, 14360]) Input IDs shape: torch.Size([1, 14360]) Labels shape: torch.Size([1, 14360]) Final batch size: 1, sequence length: 16398 Attention mask shape: torch.Size([1, 1, 16398, 16398]) Position ids shape: torch.Size([1, 16398]) Input IDs shape: torch.Size([1, 16398]) Labels shape: torch.Size([1, 16398]) Final batch size: 1, sequence length: 16053 Attention mask shape: torch.Size([1, 1, 16053, 16053]) Position ids shape: torch.Size([1, 16053]) Input IDs shape: torch.Size([1, 16053]) Labels shape: torch.Size([1, 16053]) Final batch size: 1, sequence length: 16756 Attention mask shape: torch.Size([1, 1, 16756, 16756]) Position ids shape: torch.Size([1, 16756]) Input IDs shape: torch.Size([1, 16756]) Labels shape: torch.Size([1, 16756]) Final batch size: 1, sequence length: 16223 Attention mask shape: torch.Size([1, 1, 16223, 16223]) Position ids shape: torch.Size([1, 16223]) Input IDs shape: torch.Size([1, 16223]) Labels shape: torch.Size([1, 16223]) Final batch size: 1, sequence length: 19278 Attention mask shape: torch.Size([1, 1, 19278, 19278]) Position ids shape: torch.Size([1, 19278]) Input IDs shape: torch.Size([1, 19278]) Labels shape: torch.Size([1, 19278]) Final batch size: 1, sequence length: 17294 Attention mask shape: torch.Size([1, 1, 17294, 17294]) Position ids shape: torch.Size([1, 17294]) Input IDs shape: torch.Size([1, 17294]) Labels shape: torch.Size([1, 17294]) Final batch size: 1, sequence length: 17117 Attention mask shape: torch.Size([1, 1, 17117, 17117]) Position ids shape: torch.Size([1, 17117]) Input IDs shape: torch.Size([1, 17117]) Labels shape: torch.Size([1, 17117]) Final batch size: 1, sequence length: 14025 Attention mask shape: torch.Size([1, 1, 14025, 14025]) Position ids shape: torch.Size([1, 14025]) Input IDs shape: torch.Size([1, 14025]) Labels shape: torch.Size([1, 14025]) Final batch size: 1, sequence length: 10017 Attention mask shape: torch.Size([1, 1, 10017, 10017]) Position ids shape: torch.Size([1, 10017]) Input IDs shape: torch.Size([1, 10017]) Labels shape: torch.Size([1, 10017]) Final batch size: 1, sequence length: 14437 Attention mask shape: torch.Size([1, 1, 14437, 14437]) Position ids shape: torch.Size([1, 14437]) Input IDs shape: torch.Size([1, 14437]) Labels shape: torch.Size([1, 14437]) Final batch size: 1, sequence length: 18408 Attention mask shape: torch.Size([1, 1, 18408, 18408]) Position ids shape: torch.Size([1, 18408]) Input IDs shape: torch.Size([1, 18408]) Labels shape: torch.Size([1, 18408]) Final batch size: 1, sequence length: 19892 Attention mask shape: torch.Size([1, 1, 19892, 19892]) Position ids shape: torch.Size([1, 19892]) Input IDs shape: torch.Size([1, 19892]) Labels shape: torch.Size([1, 19892]) Final batch size: 1, sequence length: 15243 Attention mask shape: torch.Size([1, 1, 15243, 15243]) Position ids shape: torch.Size([1, 15243]) Input IDs shape: torch.Size([1, 15243]) Labels shape: torch.Size([1, 15243]) Final batch size: 1, sequence length: 11225 Attention mask shape: torch.Size([1, 1, 11225, 11225]) Position ids shape: torch.Size([1, 11225]) Input IDs shape: torch.Size([1, 11225]) Labels shape: torch.Size([1, 11225]) Final batch size: 1, sequence length: 20695 Attention mask shape: torch.Size([1, 1, 20695, 20695]) Position ids shape: torch.Size([1, 20695]) Input IDs shape: torch.Size([1, 20695]) Labels shape: torch.Size([1, 20695]) Final batch size: 1, sequence length: 19259 Attention mask shape: torch.Size([1, 1, 19259, 19259]) Position ids shape: torch.Size([1, 19259]) Input IDs shape: torch.Size([1, 19259]) Labels shape: torch.Size([1, 19259]) Final batch size: 1, sequence length: 17843 Attention mask shape: torch.Size([1, 1, 17843, 17843]) Position ids shape: torch.Size([1, 17843]) Input IDs shape: torch.Size([1, 17843]) Labels shape: torch.Size([1, 17843]) Final batch size: 1, sequence length: 21455 Attention mask shape: torch.Size([1, 1, 21455, 21455]) Position ids shape: torch.Size([1, 21455]) Input IDs shape: torch.Size([1, 21455]) Labels shape: torch.Size([1, 21455]) Final batch size: 1, sequence length: 19767 Attention mask shape: torch.Size([1, 1, 19767, 19767]) Position ids shape: torch.Size([1, 19767]) Input IDs shape: torch.Size([1, 19767]) Labels shape: torch.Size([1, 19767]) Final batch size: 1, sequence length: 20433 Attention mask shape: torch.Size([1, 1, 20433, 20433]) Position ids shape: torch.Size([1, 20433]) Input IDs shape: torch.Size([1, 20433]) Labels shape: torch.Size([1, 20433]) Final batch size: 1, sequence length: 23334 Attention mask shape: torch.Size([1, 1, 23334, 23334]) Position ids shape: torch.Size([1, 23334]) Input IDs shape: torch.Size([1, 23334]) Labels shape: torch.Size([1, 23334]) Final batch size: 1, sequence length: 22932 Attention mask shape: torch.Size([1, 1, 22932, 22932]) Position ids shape: torch.Size([1, 22932]) Input IDs shape: torch.Size([1, 22932]) Labels shape: torch.Size([1, 22932]) Final batch size: 1, sequence length: 25405 Attention mask shape: torch.Size([1, 1, 25405, 25405]) Position ids shape: torch.Size([1, 25405]) Input IDs shape: torch.Size([1, 25405]) Labels shape: torch.Size([1, 25405]) Final batch size: 1, sequence length: 5734 Attention mask shape: torch.Size([1, 1, 5734, 5734]) Position ids shape: torch.Size([1, 5734]) Input IDs shape: torch.Size([1, 5734]) Labels shape: torch.Size([1, 5734]) Final batch size: 1, sequence length: 19492 Attention mask shape: torch.Size([1, 1, 19492, 19492]) Position ids shape: torch.Size([1, 19492]) Input IDs shape: torch.Size([1, 19492]) Labels shape: torch.Size([1, 19492]) Final batch size: 1, sequence length: 6378 Attention mask shape: torch.Size([1, 1, 6378, 6378]) Position ids shape: torch.Size([1, 6378]) Input IDs shape: torch.Size([1, 6378]) Labels shape: torch.Size([1, 6378]) Final batch size: 1, sequence length: 25548 Attention mask shape: torch.Size([1, 1, 25548, 25548]) Position ids shape: torch.Size([1, 25548]) Input IDs shape: torch.Size([1, 25548]) Labels shape: torch.Size([1, 25548]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 14429 Attention mask shape: torch.Size([1, 1, 14429, 14429]) Position ids shape: torch.Size([1, 14429]) Input IDs shape: torch.Size([1, 14429]) Labels shape: torch.Size([1, 14429]) Final batch size: 1, sequence length: 13623 Attention mask shape: torch.Size([1, 1, 13623, 13623]) Position ids shape: torch.Size([1, 13623]) Input IDs shape: torch.Size([1, 13623]) Labels shape: torch.Size([1, 13623]) Final batch size: 1, sequence length: 26063 Attention mask shape: torch.Size([1, 1, 26063, 26063]) Position ids shape: torch.Size([1, 26063]) Input IDs shape: torch.Size([1, 26063]) Labels shape: torch.Size([1, 26063]) Final batch size: 1, sequence length: 25381 Attention mask shape: torch.Size([1, 1, 25381, 25381]) Position ids shape: torch.Size([1, 25381]) Input IDs shape: torch.Size([1, 25381]) Labels shape: torch.Size([1, 25381]) Final batch size: 1, sequence length: 16716 Attention mask shape: torch.Size([1, 1, 16716, 16716]) Position ids shape: torch.Size([1, 16716]) Input IDs shape: torch.Size([1, 16716]) Labels shape: torch.Size([1, 16716]) Final batch size: 1, sequence length: 28166 Attention mask shape: torch.Size([1, 1, 28166, 28166]) Position ids shape: torch.Size([1, 28166]) Input IDs shape: torch.Size([1, 28166]) Labels shape: torch.Size([1, 28166]) Final batch size: 1, sequence length: 27659 Attention mask shape: torch.Size([1, 1, 27659, 27659]) Position ids shape: torch.Size([1, 27659]) Input IDs shape: torch.Size([1, 27659]) Labels shape: torch.Size([1, 27659]) Final batch size: 1, sequence length: 29561 Attention mask shape: torch.Size([1, 1, 29561, 29561]) Position ids shape: torch.Size([1, 29561]) Input IDs shape: torch.Size([1, 29561]) Labels shape: torch.Size([1, 29561]) Final batch size: 1, sequence length: 24885 Attention mask shape: torch.Size([1, 1, 24885, 24885]) Position ids shape: torch.Size([1, 24885]) Input IDs shape: torch.Size([1, 24885]) Labels shape: torch.Size([1, 24885]) Final batch size: 1, sequence length: 17985 Attention mask shape: torch.Size([1, 1, 17985, 17985]) Position ids shape: torch.Size([1, 17985]) Input IDs shape: torch.Size([1, 17985]) Labels shape: torch.Size([1, 17985]) Final batch size: 1, sequence length: 28623 Attention mask shape: torch.Size([1, 1, 28623, 28623]) Position ids shape: torch.Size([1, 28623]) Input IDs shape: torch.Size([1, 28623]) Labels shape: torch.Size([1, 28623]) Final batch size: 1, sequence length: 28910 Attention mask shape: torch.Size([1, 1, 28910, 28910]) Position ids shape: torch.Size([1, 28910]) Input IDs shape: torch.Size([1, 28910]) Labels shape: torch.Size([1, 28910]) Final batch size: 1, sequence length: 17951 Attention mask shape: torch.Size([1, 1, 17951, 17951]) Position ids shape: torch.Size([1, 17951]) Input IDs shape: torch.Size([1, 17951]) Labels shape: torch.Size([1, 17951]) Final batch size: 1, sequence length: 28206 Attention mask shape: torch.Size([1, 1, 28206, 28206]) Position ids shape: torch.Size([1, 28206]) Input IDs shape: torch.Size([1, 28206]) Labels shape: torch.Size([1, 28206]) Final batch size: 1, sequence length: 30236 Attention mask shape: torch.Size([1, 1, 30236, 30236]) Position ids shape: torch.Size([1, 30236]) Input IDs shape: torch.Size([1, 30236]) Labels shape: torch.Size([1, 30236]) Final batch size: 1, sequence length: 28823 Attention mask shape: torch.Size([1, 1, 28823, 28823]) Position ids shape: torch.Size([1, 28823]) Input IDs shape: torch.Size([1, 28823]) Labels shape: torch.Size([1, 28823]) Final batch size: 1, sequence length: 26639 Attention mask shape: torch.Size([1, 1, 26639, 26639]) Position ids shape: torch.Size([1, 26639]) Input IDs shape: torch.Size([1, 26639]) Labels shape: torch.Size([1, 26639]) Final batch size: 1, sequence length: 29824 Attention mask shape: torch.Size([1, 1, 29824, 29824]) Position ids shape: torch.Size([1, 29824]) Input IDs shape: torch.Size([1, 29824]) Labels shape: torch.Size([1, 29824]) Final batch size: 1, sequence length: 17634 Attention mask shape: torch.Size([1, 1, 17634, 17634]) Position ids shape: torch.Size([1, 17634]) Input IDs shape: torch.Size([1, 17634]) Labels shape: torch.Size([1, 17634]) Final batch size: 1, sequence length: 27484 Attention mask shape: torch.Size([1, 1, 27484, 27484]) Position ids shape: torch.Size([1, 27484]) Input IDs shape: torch.Size([1, 27484]) Labels shape: torch.Size([1, 27484]) Final batch size: 1, sequence length: 25540 Attention mask shape: torch.Size([1, 1, 25540, 25540]) Position ids shape: torch.Size([1, 25540]) Input IDs shape: torch.Size([1, 25540]) Labels shape: torch.Size([1, 25540]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 31464 Attention mask shape: torch.Size([1, 1, 31464, 31464]) Position ids shape: torch.Size([1, 31464]) Input IDs shape: torch.Size([1, 31464]) Labels shape: torch.Size([1, 31464]) Final batch size: 1, sequence length: 32515 Attention mask shape: torch.Size([1, 1, 32515, 32515]) Position ids shape: torch.Size([1, 32515]) Input IDs shape: torch.Size([1, 32515]) Labels shape: torch.Size([1, 32515]) Final batch size: 1, sequence length: 33601 Attention mask shape: torch.Size([1, 1, 33601, 33601]) Position ids shape: torch.Size([1, 33601]) Input IDs shape: torch.Size([1, 33601]) Labels shape: torch.Size([1, 33601]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 32660 Attention mask shape: torch.Size([1, 1, 32660, 32660]) Position ids shape: torch.Size([1, 32660]) Input IDs shape: torch.Size([1, 32660]) Labels shape: torch.Size([1, 32660]) Final batch size: 1, sequence length: 32786 Attention mask shape: torch.Size([1, 1, 32786, 32786]) Position ids shape: torch.Size([1, 32786]) Input IDs shape: torch.Size([1, 32786]) Labels shape: torch.Size([1, 32786]) Final batch size: 1, sequence length: 9029 Attention mask shape: torch.Size([1, 1, 9029, 9029]) Position ids shape: torch.Size([1, 9029]) Input IDs shape: torch.Size([1, 9029]) Labels shape: torch.Size([1, 9029]) Final batch size: 1, sequence length: 17914 Attention mask shape: torch.Size([1, 1, 17914, 17914]) Position ids shape: torch.Size([1, 17914]) Input IDs shape: torch.Size([1, 17914]) Labels shape: torch.Size([1, 17914]) Final batch size: 1, sequence length: 36131 Attention mask shape: torch.Size([1, 1, 36131, 36131]) Position ids shape: torch.Size([1, 36131]) Input IDs shape: torch.Size([1, 36131]) Labels shape: torch.Size([1, 36131]) Final batch size: 1, sequence length: 35411 Attention mask shape: torch.Size([1, 1, 35411, 35411]) Position ids shape: torch.Size([1, 35411]) Input IDs shape: torch.Size([1, 35411]) Labels shape: torch.Size([1, 35411]) Final batch size: 1, sequence length: 10132 Attention mask shape: torch.Size([1, 1, 10132, 10132]) Position ids shape: torch.Size([1, 10132]) Input IDs shape: torch.Size([1, 10132]) Labels shape: torch.Size([1, 10132]) Final batch size: 1, sequence length: 37555 Attention mask shape: torch.Size([1, 1, 37555, 37555]) Position ids shape: torch.Size([1, 37555]) Input IDs shape: torch.Size([1, 37555]) Labels shape: torch.Size([1, 37555]) Final batch size: 1, sequence length: 20951 Attention mask shape: torch.Size([1, 1, 20951, 20951]) Position ids shape: torch.Size([1, 20951]) Input IDs shape: torch.Size([1, 20951]) Labels shape: torch.Size([1, 20951]) Final batch size: 1, sequence length: 37016 Attention mask shape: torch.Size([1, 1, 37016, 37016]) Position ids shape: torch.Size([1, 37016]) Input IDs shape: torch.Size([1, 37016]) Labels shape: torch.Size([1, 37016]) Final batch size: 1, sequence length: 15513 Attention mask shape: torch.Size([1, 1, 15513, 15513]) Position ids shape: torch.Size([1, 15513]) Input IDs shape: torch.Size([1, 15513]) Labels shape: torch.Size([1, 15513]) Final batch size: 1, sequence length: 36970 Attention mask shape: torch.Size([1, 1, 36970, 36970]) Position ids shape: torch.Size([1, 36970]) Input IDs shape: torch.Size([1, 36970]) Labels shape: torch.Size([1, 36970]) Final batch size: 1, sequence length: 36124 Attention mask shape: torch.Size([1, 1, 36124, 36124]) Position ids shape: torch.Size([1, 36124]) Input IDs shape: torch.Size([1, 36124]) Labels shape: torch.Size([1, 36124]) Final batch size: 1, sequence length: 25529 Attention mask shape: torch.Size([1, 1, 25529, 25529]) Position ids shape: torch.Size([1, 25529]) Input IDs shape: torch.Size([1, 25529]) Labels shape: torch.Size([1, 25529]) Final batch size: 1, sequence length: 35525 Attention mask shape: torch.Size([1, 1, 35525, 35525]) Position ids shape: torch.Size([1, 35525]) Input IDs shape: torch.Size([1, 35525]) Labels shape: torch.Size([1, 35525]) Final batch size: 1, sequence length: 20533 Attention mask shape: torch.Size([1, 1, 20533, 20533]) Position ids shape: torch.Size([1, 20533]) Input IDs shape: torch.Size([1, 20533]) Labels shape: torch.Size([1, 20533]) Final batch size: 1, sequence length: 38712 Attention mask shape: torch.Size([1, 1, 38712, 38712]) Position ids shape: torch.Size([1, 38712]) Input IDs shape: torch.Size([1, 38712]) Labels shape: torch.Size([1, 38712]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 14372 Attention mask shape: torch.Size([1, 1, 14372, 14372]) Position ids shape: torch.Size([1, 14372]) Input IDs shape: torch.Size([1, 14372]) Labels shape: torch.Size([1, 14372]) Final batch size: 1, sequence length: 33422 Attention mask shape: torch.Size([1, 1, 33422, 33422]) Position ids shape: torch.Size([1, 33422]) Input IDs shape: torch.Size([1, 33422]) Labels shape: torch.Size([1, 33422]) Final batch size: 1, sequence length: 38071 Attention mask shape: torch.Size([1, 1, 38071, 38071]) Position ids shape: torch.Size([1, 38071]) Input IDs shape: torch.Size([1, 38071]) Labels shape: torch.Size([1, 38071]) Final batch size: 1, sequence length: 36292 Attention mask shape: torch.Size([1, 1, 36292, 36292]) Position ids shape: torch.Size([1, 36292]) Input IDs shape: torch.Size([1, 36292]) Labels shape: torch.Size([1, 36292]) Final batch size: 1, sequence length: 26781 Attention mask shape: torch.Size([1, 1, 26781, 26781]) Position ids shape: torch.Size([1, 26781]) Input IDs shape: torch.Size([1, 26781]) Labels shape: torch.Size([1, 26781]) Final batch size: 1, sequence length: 28879 Attention mask shape: torch.Size([1, 1, 28879, 28879]) Position ids shape: torch.Size([1, 28879]) Input IDs shape: torch.Size([1, 28879]) Labels shape: torch.Size([1, 28879]) Final batch size: 1, sequence length: 31084 Attention mask shape: torch.Size([1, 1, 31084, 31084]) Position ids shape: torch.Size([1, 31084]) Input IDs shape: torch.Size([1, 31084]) Labels shape: torch.Size([1, 31084]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 26562 Attention mask shape: torch.Size([1, 1, 26562, 26562]) Position ids shape: torch.Size([1, 26562]) Input IDs shape: torch.Size([1, 26562]) Labels shape: torch.Size([1, 26562]) Final batch size: 1, sequence length: 38210 Attention mask shape: torch.Size([1, 1, 38210, 38210]) Position ids shape: torch.Size([1, 38210]) Input IDs shape: torch.Size([1, 38210]) Labels shape: torch.Size([1, 38210]) Final batch size: 1, sequence length: 35775 Attention mask shape: torch.Size([1, 1, 35775, 35775]) Position ids shape: torch.Size([1, 35775]) Input IDs shape: torch.Size([1, 35775]) Labels shape: torch.Size([1, 35775]) Final batch size: 1, sequence length: 30009 Attention mask shape: torch.Size([1, 1, 30009, 30009]) Position ids shape: torch.Size([1, 30009]) Input IDs shape: torch.Size([1, 30009]) Labels shape: torch.Size([1, 30009]) Final batch size: 1, sequence length: 40753 Attention mask shape: torch.Size([1, 1, 40753, 40753]) Position ids shape: torch.Size([1, 40753]) Input IDs shape: torch.Size([1, 40753]) Labels shape: torch.Size([1, 40753]) Final batch size: 1, sequence length: 16409 Attention mask shape: torch.Size([1, 1, 16409, 16409]) Position ids shape: torch.Size([1, 16409]) Input IDs shape: torch.Size([1, 16409]) Labels shape: torch.Size([1, 16409]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36185 Attention mask shape: torch.Size([1, 1, 36185, 36185]) Position ids shape: torch.Size([1, 36185]) Input IDs shape: torch.Size([1, 36185]) Labels shape: torch.Size([1, 36185]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31042 Attention mask shape: torch.Size([1, 1, 31042, 31042]) Position ids shape: torch.Size([1, 31042]) Input IDs shape: torch.Size([1, 31042]) Labels shape: torch.Size([1, 31042]) Final batch size: 1, sequence length: 36534 Attention mask shape: torch.Size([1, 1, 36534, 36534]) Position ids shape: torch.Size([1, 36534]) Input IDs shape: torch.Size([1, 36534]) Labels shape: torch.Size([1, 36534]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30653 Attention mask shape: torch.Size([1, 1, 30653, 30653]) Position ids shape: torch.Size([1, 30653]) Input IDs shape: torch.Size([1, 30653]) Labels shape: torch.Size([1, 30653]) Final batch size: 1, sequence length: 37159 Attention mask shape: torch.Size([1, 1, 37159, 37159]) Position ids shape: torch.Size([1, 37159]) Input IDs shape: torch.Size([1, 37159]) Labels shape: torch.Size([1, 37159]) Final batch size: 1, sequence length: 15031 Attention mask shape: torch.Size([1, 1, 15031, 15031]) Position ids shape: torch.Size([1, 15031]) Input IDs shape: torch.Size([1, 15031]) Labels shape: torch.Size([1, 15031]) Final batch size: 1, sequence length: 18884 Attention mask shape: torch.Size([1, 1, 18884, 18884]) Position ids shape: torch.Size([1, 18884]) Input IDs shape: torch.Size([1, 18884]) Labels shape: torch.Size([1, 18884]) Final batch size: 1, sequence length: 7448 Attention mask shape: torch.Size([1, 1, 7448, 7448]) Position ids shape: torch.Size([1, 7448]) Input IDs shape: torch.Size([1, 7448]) Labels shape: torch.Size([1, 7448]) Final batch size: 1, sequence length: 20307 Attention mask shape: torch.Size([1, 1, 20307, 20307]) Position ids shape: torch.Size([1, 20307]) Input IDs shape: torch.Size([1, 20307]) Labels shape: torch.Size([1, 20307]) Final batch size: 1, sequence length: 31448 Attention mask shape: torch.Size([1, 1, 31448, 31448]) Position ids shape: torch.Size([1, 31448]) Input IDs shape: torch.Size([1, 31448]) Labels shape: torch.Size([1, 31448]) Final batch size: 1, sequence length: 25850 Attention mask shape: torch.Size([1, 1, 25850, 25850]) Position ids shape: torch.Size([1, 25850]) Input IDs shape: torch.Size([1, 25850]) Labels shape: torch.Size([1, 25850]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40605 Attention mask shape: torch.Size([1, 1, 40605, 40605]) Position ids shape: torch.Size([1, 40605]) Input IDs shape: torch.Size([1, 40605]) Labels shape: torch.Size([1, 40605]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 5405 Attention mask shape: torch.Size([1, 1, 5405, 5405]) Position ids shape: torch.Size([1, 5405]) Input IDs shape: torch.Size([1, 5405]) Labels shape: torch.Size([1, 5405]) {'loss': 0.2525, 'grad_norm': 0.24147867136554246, 'learning_rate': 2.5000000000000015e-06, 'num_tokens': -inf, 'epoch': 5.62} Final batch size: 1, sequence length: 4641 Attention mask shape: torch.Size([1, 1, 4641, 4641]) Position ids shape: torch.Size([1, 4641]) Input IDs shape: torch.Size([1, 4641]) Labels shape: torch.Size([1, 4641]) Final batch size: 1, sequence length: 4223 Attention mask shape: torch.Size([1, 1, 4223, 4223]) Position ids shape: torch.Size([1, 4223]) Input IDs shape: torch.Size([1, 4223]) Labels shape: torch.Size([1, 4223]) Final batch size: 1, sequence length: 5754 Attention mask shape: torch.Size([1, 1, 5754, 5754]) Position ids shape: torch.Size([1, 5754]) Input IDs shape: torch.Size([1, 5754]) Labels shape: torch.Size([1, 5754]) Final batch size: 1, sequence length: 7795 Attention mask shape: torch.Size([1, 1, 7795, 7795]) Position ids shape: torch.Size([1, 7795]) Input IDs shape: torch.Size([1, 7795]) Labels shape: torch.Size([1, 7795]) Final batch size: 1, sequence length: 8177 Attention mask shape: torch.Size([1, 1, 8177, 8177]) Position ids shape: torch.Size([1, 8177]) Input IDs shape: torch.Size([1, 8177]) Labels shape: torch.Size([1, 8177]) Final batch size: 1, sequence length: 9021 Attention mask shape: torch.Size([1, 1, 9021, 9021]) Position ids shape: torch.Size([1, 9021]) Input IDs shape: torch.Size([1, 9021]) Labels shape: torch.Size([1, 9021]) Final batch size: 1, sequence length: 12501 Attention mask shape: torch.Size([1, 1, 12501, 12501]) Position ids shape: torch.Size([1, 12501]) Input IDs shape: torch.Size([1, 12501]) Labels shape: torch.Size([1, 12501]) Final batch size: 1, sequence length: 9027 Attention mask shape: torch.Size([1, 1, 9027, 9027]) Position ids shape: torch.Size([1, 9027]) Input IDs shape: torch.Size([1, 9027]) Labels shape: torch.Size([1, 9027]) Final batch size: 1, sequence length: 11974 Attention mask shape: torch.Size([1, 1, 11974, 11974]) Position ids shape: torch.Size([1, 11974]) Input IDs shape: torch.Size([1, 11974]) Labels shape: torch.Size([1, 11974]) Final batch size: 1, sequence length: 13186 Attention mask shape: torch.Size([1, 1, 13186, 13186]) Position ids shape: torch.Size([1, 13186]) Input IDs shape: torch.Size([1, 13186]) Labels shape: torch.Size([1, 13186]) Final batch size: 1, sequence length: 16330 Attention mask shape: torch.Size([1, 1, 16330, 16330]) Position ids shape: torch.Size([1, 16330]) Input IDs shape: torch.Size([1, 16330]) Labels shape: torch.Size([1, 16330]) Final batch size: 1, sequence length: 16104 Attention mask shape: torch.Size([1, 1, 16104, 16104]) Position ids shape: torch.Size([1, 16104]) Input IDs shape: torch.Size([1, 16104]) Labels shape: torch.Size([1, 16104]) Final batch size: 1, sequence length: 16496 Attention mask shape: torch.Size([1, 1, 16496, 16496]) Position ids shape: torch.Size([1, 16496]) Input IDs shape: torch.Size([1, 16496]) Labels shape: torch.Size([1, 16496]) Final batch size: 1, sequence length: 13957 Attention mask shape: torch.Size([1, 1, 13957, 13957]) Position ids shape: torch.Size([1, 13957]) Input IDs shape: torch.Size([1, 13957]) Labels shape: torch.Size([1, 13957]) Final batch size: 1, sequence length: 16366 Attention mask shape: torch.Size([1, 1, 16366, 16366]) Position ids shape: torch.Size([1, 16366]) Input IDs shape: torch.Size([1, 16366]) Labels shape: torch.Size([1, 16366]) Final batch size: 1, sequence length: 18958 Attention mask shape: torch.Size([1, 1, 18958, 18958]) Position ids shape: torch.Size([1, 18958]) Input IDs shape: torch.Size([1, 18958]) Labels shape: torch.Size([1, 18958]) Final batch size: 1, sequence length: 16088 Attention mask shape: torch.Size([1, 1, 16088, 16088]) Position ids shape: torch.Size([1, 16088]) Input IDs shape: torch.Size([1, 16088]) Labels shape: torch.Size([1, 16088]) Final batch size: 1, sequence length: 18415 Attention mask shape: torch.Size([1, 1, 18415, 18415]) Position ids shape: torch.Size([1, 18415]) Input IDs shape: torch.Size([1, 18415]) Labels shape: torch.Size([1, 18415]) Final batch size: 1, sequence length: 15283 Attention mask shape: torch.Size([1, 1, 15283, 15283]) Position ids shape: torch.Size([1, 15283]) Input IDs shape: torch.Size([1, 15283]) Labels shape: torch.Size([1, 15283]) Final batch size: 1, sequence length: 16014 Attention mask shape: torch.Size([1, 1, 16014, 16014]) Position ids shape: torch.Size([1, 16014]) Input IDs shape: torch.Size([1, 16014]) Labels shape: torch.Size([1, 16014]) Final batch size: 1, sequence length: 17092 Attention mask shape: torch.Size([1, 1, 17092, 17092]) Position ids shape: torch.Size([1, 17092]) Input IDs shape: torch.Size([1, 17092]) Labels shape: torch.Size([1, 17092]) Final batch size: 1, sequence length: 18034 Attention mask shape: torch.Size([1, 1, 18034, 18034]) Position ids shape: torch.Size([1, 18034]) Input IDs shape: torch.Size([1, 18034]) Labels shape: torch.Size([1, 18034]) Final batch size: 1, sequence length: 19034 Attention mask shape: torch.Size([1, 1, 19034, 19034]) Position ids shape: torch.Size([1, 19034]) Input IDs shape: torch.Size([1, 19034]) Labels shape: torch.Size([1, 19034]) Final batch size: 1, sequence length: 19603 Attention mask shape: torch.Size([1, 1, 19603, 19603]) Position ids shape: torch.Size([1, 19603]) Input IDs shape: torch.Size([1, 19603]) Labels shape: torch.Size([1, 19603]) Final batch size: 1, sequence length: 18938 Attention mask shape: torch.Size([1, 1, 18938, 18938]) Position ids shape: torch.Size([1, 18938]) Input IDs shape: torch.Size([1, 18938]) Labels shape: torch.Size([1, 18938]) Final batch size: 1, sequence length: 21132 Attention mask shape: torch.Size([1, 1, 21132, 21132]) Position ids shape: torch.Size([1, 21132]) Input IDs shape: torch.Size([1, 21132]) Labels shape: torch.Size([1, 21132]) Final batch size: 1, sequence length: 14492 Attention mask shape: torch.Size([1, 1, 14492, 14492]) Position ids shape: torch.Size([1, 14492]) Input IDs shape: torch.Size([1, 14492]) Labels shape: torch.Size([1, 14492]) Final batch size: 1, sequence length: 13468 Attention mask shape: torch.Size([1, 1, 13468, 13468]) Position ids shape: torch.Size([1, 13468]) Input IDs shape: torch.Size([1, 13468]) Labels shape: torch.Size([1, 13468]) Final batch size: 1, sequence length: 23132 Attention mask shape: torch.Size([1, 1, 23132, 23132]) Position ids shape: torch.Size([1, 23132]) Input IDs shape: torch.Size([1, 23132]) Labels shape: torch.Size([1, 23132]) Final batch size: 1, sequence length: 23102 Attention mask shape: torch.Size([1, 1, 23102, 23102]) Position ids shape: torch.Size([1, 23102]) Input IDs shape: torch.Size([1, 23102]) Labels shape: torch.Size([1, 23102]) Final batch size: 1, sequence length: 20726 Attention mask shape: torch.Size([1, 1, 20726, 20726]) Position ids shape: torch.Size([1, 20726]) Input IDs shape: torch.Size([1, 20726]) Labels shape: torch.Size([1, 20726]) Final batch size: 1, sequence length: 22328 Attention mask shape: torch.Size([1, 1, 22328, 22328]) Position ids shape: torch.Size([1, 22328]) Input IDs shape: torch.Size([1, 22328]) Labels shape: torch.Size([1, 22328]) Final batch size: 1, sequence length: 24432 Attention mask shape: torch.Size([1, 1, 24432, 24432]) Position ids shape: torch.Size([1, 24432]) Input IDs shape: torch.Size([1, 24432]) Labels shape: torch.Size([1, 24432]) Final batch size: 1, sequence length: 21705 Attention mask shape: torch.Size([1, 1, 21705, 21705]) Position ids shape: torch.Size([1, 21705]) Input IDs shape: torch.Size([1, 21705]) Labels shape: torch.Size([1, 21705]) Final batch size: 1, sequence length: 22771 Attention mask shape: torch.Size([1, 1, 22771, 22771]) Position ids shape: torch.Size([1, 22771]) Input IDs shape: torch.Size([1, 22771]) Labels shape: torch.Size([1, 22771]) Final batch size: 1, sequence length: 21404 Attention mask shape: torch.Size([1, 1, 21404, 21404]) Position ids shape: torch.Size([1, 21404]) Input IDs shape: torch.Size([1, 21404]) Labels shape: torch.Size([1, 21404]) Final batch size: 1, sequence length: 12756 Attention mask shape: torch.Size([1, 1, 12756, 12756]) Position ids shape: torch.Size([1, 12756]) Input IDs shape: torch.Size([1, 12756]) Labels shape: torch.Size([1, 12756]) Final batch size: 1, sequence length: 23971 Attention mask shape: torch.Size([1, 1, 23971, 23971]) Position ids shape: torch.Size([1, 23971]) Input IDs shape: torch.Size([1, 23971]) Labels shape: torch.Size([1, 23971]) Final batch size: 1, sequence length: 16590 Attention mask shape: torch.Size([1, 1, 16590, 16590]) Position ids shape: torch.Size([1, 16590]) Input IDs shape: torch.Size([1, 16590]) Labels shape: torch.Size([1, 16590]) Final batch size: 1, sequence length: 17514 Attention mask shape: torch.Size([1, 1, 17514, 17514]) Position ids shape: torch.Size([1, 17514]) Input IDs shape: torch.Size([1, 17514]) Labels shape: torch.Size([1, 17514]) Final batch size: 1, sequence length: 25351 Attention mask shape: torch.Size([1, 1, 25351, 25351]) Position ids shape: torch.Size([1, 25351]) Input IDs shape: torch.Size([1, 25351]) Labels shape: torch.Size([1, 25351]) Final batch size: 1, sequence length: 16291 Attention mask shape: torch.Size([1, 1, 16291, 16291]) Position ids shape: torch.Size([1, 16291]) Input IDs shape: torch.Size([1, 16291]) Labels shape: torch.Size([1, 16291]) Final batch size: 1, sequence length: 27785 Attention mask shape: torch.Size([1, 1, 27785, 27785]) Position ids shape: torch.Size([1, 27785]) Input IDs shape: torch.Size([1, 27785]) Labels shape: torch.Size([1, 27785]) Final batch size: 1, sequence length: 23894 Attention mask shape: torch.Size([1, 1, 23894, 23894]) Position ids shape: torch.Size([1, 23894]) Input IDs shape: torch.Size([1, 23894]) Labels shape: torch.Size([1, 23894]) Final batch size: 1, sequence length: 28412 Attention mask shape: torch.Size([1, 1, 28412, 28412]) Position ids shape: torch.Size([1, 28412]) Input IDs shape: torch.Size([1, 28412]) Labels shape: torch.Size([1, 28412]) Final batch size: 1, sequence length: 26375 Attention mask shape: torch.Size([1, 1, 26375, 26375]) Position ids shape: torch.Size([1, 26375]) Input IDs shape: torch.Size([1, 26375]) Labels shape: torch.Size([1, 26375]) Final batch size: 1, sequence length: 27780 Attention mask shape: torch.Size([1, 1, 27780, 27780]) Position ids shape: torch.Size([1, 27780]) Input IDs shape: torch.Size([1, 27780]) Labels shape: torch.Size([1, 27780]) Final batch size: 1, sequence length: 20817 Attention mask shape: torch.Size([1, 1, 20817, 20817]) Position ids shape: torch.Size([1, 20817]) Input IDs shape: torch.Size([1, 20817]) Labels shape: torch.Size([1, 20817]) Final batch size: 1, sequence length: 23810 Attention mask shape: torch.Size([1, 1, 23810, 23810]) Position ids shape: torch.Size([1, 23810]) Input IDs shape: torch.Size([1, 23810]) Labels shape: torch.Size([1, 23810]) Final batch size: 1, sequence length: 26596 Attention mask shape: torch.Size([1, 1, 26596, 26596]) Position ids shape: torch.Size([1, 26596]) Input IDs shape: torch.Size([1, 26596]) Labels shape: torch.Size([1, 26596]) Final batch size: 1, sequence length: 26766 Attention mask shape: torch.Size([1, 1, 26766, 26766]) Position ids shape: torch.Size([1, 26766]) Input IDs shape: torch.Size([1, 26766]) Labels shape: torch.Size([1, 26766]) Final batch size: 1, sequence length: 30165 Attention mask shape: torch.Size([1, 1, 30165, 30165]) Position ids shape: torch.Size([1, 30165]) Input IDs shape: torch.Size([1, 30165]) Labels shape: torch.Size([1, 30165]) Final batch size: 1, sequence length: 29168 Attention mask shape: torch.Size([1, 1, 29168, 29168]) Position ids shape: torch.Size([1, 29168]) Input IDs shape: torch.Size([1, 29168]) Labels shape: torch.Size([1, 29168]) Final batch size: 1, sequence length: 28574 Attention mask shape: torch.Size([1, 1, 28574, 28574]) Position ids shape: torch.Size([1, 28574]) Input IDs shape: torch.Size([1, 28574]) Labels shape: torch.Size([1, 28574]) Final batch size: 1, sequence length: 32022 Attention mask shape: torch.Size([1, 1, 32022, 32022]) Position ids shape: torch.Size([1, 32022]) Input IDs shape: torch.Size([1, 32022]) Labels shape: torch.Size([1, 32022]) Final batch size: 1, sequence length: 18869 Attention mask shape: torch.Size([1, 1, 18869, 18869]) Position ids shape: torch.Size([1, 18869]) Input IDs shape: torch.Size([1, 18869]) Labels shape: torch.Size([1, 18869]) Final batch size: 1, sequence length: 13790 Attention mask shape: torch.Size([1, 1, 13790, 13790]) Position ids shape: torch.Size([1, 13790]) Input IDs shape: torch.Size([1, 13790]) Labels shape: torch.Size([1, 13790]) Final batch size: 1, sequence length: 25388 Attention mask shape: torch.Size([1, 1, 25388, 25388]) Position ids shape: torch.Size([1, 25388]) Input IDs shape: torch.Size([1, 25388]) Labels shape: torch.Size([1, 25388]) Final batch size: 1, sequence length: 28845 Attention mask shape: torch.Size([1, 1, 28845, 28845]) Position ids shape: torch.Size([1, 28845]) Input IDs shape: torch.Size([1, 28845]) Labels shape: torch.Size([1, 28845]) Final batch size: 1, sequence length: 29132 Attention mask shape: torch.Size([1, 1, 29132, 29132]) Position ids shape: torch.Size([1, 29132]) Input IDs shape: torch.Size([1, 29132]) Labels shape: torch.Size([1, 29132]) Final batch size: 1, sequence length: 34457 Attention mask shape: torch.Size([1, 1, 34457, 34457]) Position ids shape: torch.Size([1, 34457]) Input IDs shape: torch.Size([1, 34457]) Labels shape: torch.Size([1, 34457]) Final batch size: 1, sequence length: 34326 Attention mask shape: torch.Size([1, 1, 34326, 34326]) Position ids shape: torch.Size([1, 34326]) Input IDs shape: torch.Size([1, 34326]) Labels shape: torch.Size([1, 34326]) Final batch size: 1, sequence length: 22043 Attention mask shape: torch.Size([1, 1, 22043, 22043]) Position ids shape: torch.Size([1, 22043]) Input IDs shape: torch.Size([1, 22043]) Labels shape: torch.Size([1, 22043]) Final batch size: 1, sequence length: 24653 Attention mask shape: torch.Size([1, 1, 24653, 24653]) Position ids shape: torch.Size([1, 24653]) Input IDs shape: torch.Size([1, 24653]) Labels shape: torch.Size([1, 24653]) Final batch size: 1, sequence length: 35790 Attention mask shape: torch.Size([1, 1, 35790, 35790]) Position ids shape: torch.Size([1, 35790]) Input IDs shape: torch.Size([1, 35790]) Labels shape: torch.Size([1, 35790]) Final batch size: 1, sequence length: 29830 Attention mask shape: torch.Size([1, 1, 29830, 29830]) Position ids shape: torch.Size([1, 29830]) Input IDs shape: torch.Size([1, 29830]) Labels shape: torch.Size([1, 29830]) Final batch size: 1, sequence length: 26465 Attention mask shape: torch.Size([1, 1, 26465, 26465]) Position ids shape: torch.Size([1, 26465]) Input IDs shape: torch.Size([1, 26465]) Labels shape: torch.Size([1, 26465]) Final batch size: 1, sequence length: 13986 Attention mask shape: torch.Size([1, 1, 13986, 13986]) Position ids shape: torch.Size([1, 13986]) Input IDs shape: torch.Size([1, 13986]) Labels shape: torch.Size([1, 13986]) Final batch size: 1, sequence length: 34874 Attention mask shape: torch.Size([1, 1, 34874, 34874]) Position ids shape: torch.Size([1, 34874]) Input IDs shape: torch.Size([1, 34874]) Labels shape: torch.Size([1, 34874]) Final batch size: 1, sequence length: 22014 Attention mask shape: torch.Size([1, 1, 22014, 22014]) Position ids shape: torch.Size([1, 22014]) Input IDs shape: torch.Size([1, 22014]) Labels shape: torch.Size([1, 22014]) Final batch size: 1, sequence length: 23245 Attention mask shape: torch.Size([1, 1, 23245, 23245]) Position ids shape: torch.Size([1, 23245]) Input IDs shape: torch.Size([1, 23245]) Labels shape: torch.Size([1, 23245]) Final batch size: 1, sequence length: 38617 Attention mask shape: torch.Size([1, 1, 38617, 38617]) Position ids shape: torch.Size([1, 38617]) Input IDs shape: torch.Size([1, 38617]) Labels shape: torch.Size([1, 38617]) Final batch size: 1, sequence length: 35868 Attention mask shape: torch.Size([1, 1, 35868, 35868]) Position ids shape: torch.Size([1, 35868]) Input IDs shape: torch.Size([1, 35868]) Labels shape: torch.Size([1, 35868]) Final batch size: 1, sequence length: 37039 Attention mask shape: torch.Size([1, 1, 37039, 37039]) Position ids shape: torch.Size([1, 37039]) Input IDs shape: torch.Size([1, 37039]) Labels shape: torch.Size([1, 37039]) Final batch size: 1, sequence length: 36794 Attention mask shape: torch.Size([1, 1, 36794, 36794]) Position ids shape: torch.Size([1, 36794]) Input IDs shape: torch.Size([1, 36794]) Labels shape: torch.Size([1, 36794]) Final batch size: 1, sequence length: 39150 Attention mask shape: torch.Size([1, 1, 39150, 39150]) Position ids shape: torch.Size([1, 39150]) Input IDs shape: torch.Size([1, 39150]) Labels shape: torch.Size([1, 39150]) Final batch size: 1, sequence length: 19204 Attention mask shape: torch.Size([1, 1, 19204, 19204]) Position ids shape: torch.Size([1, 19204]) Input IDs shape: torch.Size([1, 19204]) Labels shape: torch.Size([1, 19204]) Final batch size: 1, sequence length: 35803 Attention mask shape: torch.Size([1, 1, 35803, 35803]) Position ids shape: torch.Size([1, 35803]) Input IDs shape: torch.Size([1, 35803]) Labels shape: torch.Size([1, 35803]) Final batch size: 1, sequence length: 25923 Attention mask shape: torch.Size([1, 1, 25923, 25923]) Position ids shape: torch.Size([1, 25923]) Input IDs shape: torch.Size([1, 25923]) Labels shape: torch.Size([1, 25923]) Final batch size: 1, sequence length: 35999 Attention mask shape: torch.Size([1, 1, 35999, 35999]) Position ids shape: torch.Size([1, 35999]) Input IDs shape: torch.Size([1, 35999]) Labels shape: torch.Size([1, 35999]) Final batch size: 1, sequence length: 19744 Attention mask shape: torch.Size([1, 1, 19744, 19744]) Position ids shape: torch.Size([1, 19744]) Input IDs shape: torch.Size([1, 19744]) Labels shape: torch.Size([1, 19744]) Final batch size: 1, sequence length: 28002 Attention mask shape: torch.Size([1, 1, 28002, 28002]) Position ids shape: torch.Size([1, 28002]) Input IDs shape: torch.Size([1, 28002]) Labels shape: torch.Size([1, 28002]) Final batch size: 1, sequence length: 31506 Attention mask shape: torch.Size([1, 1, 31506, 31506]) Position ids shape: torch.Size([1, 31506]) Input IDs shape: torch.Size([1, 31506]) Labels shape: torch.Size([1, 31506]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32974 Attention mask shape: torch.Size([1, 1, 32974, 32974]) Position ids shape: torch.Size([1, 32974]) Input IDs shape: torch.Size([1, 32974]) Labels shape: torch.Size([1, 32974]) Final batch size: 1, sequence length: 28781 Attention mask shape: torch.Size([1, 1, 28781, 28781]) Position ids shape: torch.Size([1, 28781]) Input IDs shape: torch.Size([1, 28781]) Labels shape: torch.Size([1, 28781]) Final batch size: 1, sequence length: 38919 Attention mask shape: torch.Size([1, 1, 38919, 38919]) Position ids shape: torch.Size([1, 38919]) Input IDs shape: torch.Size([1, 38919]) Labels shape: torch.Size([1, 38919]) Final batch size: 1, sequence length: 39919 Attention mask shape: torch.Size([1, 1, 39919, 39919]) Position ids shape: torch.Size([1, 39919]) Input IDs shape: torch.Size([1, 39919]) Labels shape: torch.Size([1, 39919]) Final batch size: 1, sequence length: 40717 Attention mask shape: torch.Size([1, 1, 40717, 40717]) Position ids shape: torch.Size([1, 40717]) Input IDs shape: torch.Size([1, 40717]) Labels shape: torch.Size([1, 40717]) Final batch size: 1, sequence length: 22282 Attention mask shape: torch.Size([1, 1, 22282, 22282]) Position ids shape: torch.Size([1, 22282]) Input IDs shape: torch.Size([1, 22282]) Labels shape: torch.Size([1, 22282]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31340 Attention mask shape: torch.Size([1, 1, 31340, 31340]) Position ids shape: torch.Size([1, 31340]) Input IDs shape: torch.Size([1, 31340]) Labels shape: torch.Size([1, 31340]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40065 Attention mask shape: torch.Size([1, 1, 40065, 40065]) Position ids shape: torch.Size([1, 40065]) Input IDs shape: torch.Size([1, 40065]) Labels shape: torch.Size([1, 40065]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19846 Attention mask shape: torch.Size([1, 1, 19846, 19846]) Position ids shape: torch.Size([1, 19846]) Input IDs shape: torch.Size([1, 19846]) Labels shape: torch.Size([1, 19846]) Final batch size: 1, sequence length: 11644 Attention mask shape: torch.Size([1, 1, 11644, 11644]) Position ids shape: torch.Size([1, 11644]) Input IDs shape: torch.Size([1, 11644]) Labels shape: torch.Size([1, 11644]) Final batch size: 1, sequence length: 31359 Attention mask shape: torch.Size([1, 1, 31359, 31359]) Position ids shape: torch.Size([1, 31359]) Input IDs shape: torch.Size([1, 31359]) Labels shape: torch.Size([1, 31359]) Final batch size: 1, sequence length: 18379 Attention mask shape: torch.Size([1, 1, 18379, 18379]) Position ids shape: torch.Size([1, 18379]) Input IDs shape: torch.Size([1, 18379]) Labels shape: torch.Size([1, 18379]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 12418 Attention mask shape: torch.Size([1, 1, 12418, 12418]) Position ids shape: torch.Size([1, 12418]) Input IDs shape: torch.Size([1, 12418]) Labels shape: torch.Size([1, 12418]) Final batch size: 1, sequence length: 7364 Attention mask shape: torch.Size([1, 1, 7364, 7364]) Position ids shape: torch.Size([1, 7364]) Input IDs shape: torch.Size([1, 7364]) Labels shape: torch.Size([1, 7364]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 9309 Attention mask shape: torch.Size([1, 1, 9309, 9309]) Position ids shape: torch.Size([1, 9309]) Input IDs shape: torch.Size([1, 9309]) Labels shape: torch.Size([1, 9309]) Final batch size: 1, sequence length: 40874 Attention mask shape: torch.Size([1, 1, 40874, 40874]) Position ids shape: torch.Size([1, 40874]) Input IDs shape: torch.Size([1, 40874]) Labels shape: torch.Size([1, 40874]) Final batch size: 1, sequence length: 25560 Attention mask shape: torch.Size([1, 1, 25560, 25560]) Position ids shape: torch.Size([1, 25560]) Input IDs shape: torch.Size([1, 25560]) Labels shape: torch.Size([1, 25560]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 26221 Attention mask shape: torch.Size([1, 1, 26221, 26221]) Position ids shape: torch.Size([1, 26221]) Input IDs shape: torch.Size([1, 26221]) Labels shape: torch.Size([1, 26221]) Final batch size: 1, sequence length: 17224 Attention mask shape: torch.Size([1, 1, 17224, 17224]) Position ids shape: torch.Size([1, 17224]) Input IDs shape: torch.Size([1, 17224]) Labels shape: torch.Size([1, 17224]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 18654 Attention mask shape: torch.Size([1, 1, 18654, 18654]) Position ids shape: torch.Size([1, 18654]) Input IDs shape: torch.Size([1, 18654]) Labels shape: torch.Size([1, 18654]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25820 Attention mask shape: torch.Size([1, 1, 25820, 25820]) Position ids shape: torch.Size([1, 25820]) Input IDs shape: torch.Size([1, 25820]) Labels shape: torch.Size([1, 25820]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30709 Attention mask shape: torch.Size([1, 1, 30709, 30709]) Position ids shape: torch.Size([1, 30709]) Input IDs shape: torch.Size([1, 30709]) Labels shape: torch.Size([1, 30709]) Final batch size: 1, sequence length: 29526 Attention mask shape: torch.Size([1, 1, 29526, 29526]) Position ids shape: torch.Size([1, 29526]) Input IDs shape: torch.Size([1, 29526]) Labels shape: torch.Size([1, 29526]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2625, 'grad_norm': 0.23342245542561751, 'learning_rate': 2.2768048249248648e-06, 'num_tokens': -inf, 'epoch': 5.75} Final batch size: 1, sequence length: 7461 Attention mask shape: torch.Size([1, 1, 7461, 7461]) Position ids shape: torch.Size([1, 7461]) Input IDs shape: torch.Size([1, 7461]) Labels shape: torch.Size([1, 7461]) Final batch size: 1, sequence length: 10826 Attention mask shape: torch.Size([1, 1, 10826, 10826]) Position ids shape: torch.Size([1, 10826]) Input IDs shape: torch.Size([1, 10826]) Labels shape: torch.Size([1, 10826]) Final batch size: 1, sequence length: 13427 Attention mask shape: torch.Size([1, 1, 13427, 13427]) Position ids shape: torch.Size([1, 13427]) Input IDs shape: torch.Size([1, 13427]) Labels shape: torch.Size([1, 13427]) Final batch size: 1, sequence length: 9858 Attention mask shape: torch.Size([1, 1, 9858, 9858]) Position ids shape: torch.Size([1, 9858]) Input IDs shape: torch.Size([1, 9858]) Labels shape: torch.Size([1, 9858]) Final batch size: 1, sequence length: 12279 Attention mask shape: torch.Size([1, 1, 12279, 12279]) Position ids shape: torch.Size([1, 12279]) Input IDs shape: torch.Size([1, 12279]) Labels shape: torch.Size([1, 12279]) Final batch size: 1, sequence length: 12096 Attention mask shape: torch.Size([1, 1, 12096, 12096]) Position ids shape: torch.Size([1, 12096]) Input IDs shape: torch.Size([1, 12096]) Labels shape: torch.Size([1, 12096]) Final batch size: 1, sequence length: 12622 Attention mask shape: torch.Size([1, 1, 12622, 12622]) Position ids shape: torch.Size([1, 12622]) Input IDs shape: torch.Size([1, 12622]) Labels shape: torch.Size([1, 12622]) Final batch size: 1, sequence length: 13550 Attention mask shape: torch.Size([1, 1, 13550, 13550]) Position ids shape: torch.Size([1, 13550]) Input IDs shape: torch.Size([1, 13550]) Labels shape: torch.Size([1, 13550]) Final batch size: 1, sequence length: 15933 Attention mask shape: torch.Size([1, 1, 15933, 15933]) Position ids shape: torch.Size([1, 15933]) Input IDs shape: torch.Size([1, 15933]) Labels shape: torch.Size([1, 15933]) Final batch size: 1, sequence length: 16927 Attention mask shape: torch.Size([1, 1, 16927, 16927]) Position ids shape: torch.Size([1, 16927]) Input IDs shape: torch.Size([1, 16927]) Labels shape: torch.Size([1, 16927]) Final batch size: 1, sequence length: 15673 Attention mask shape: torch.Size([1, 1, 15673, 15673]) Position ids shape: torch.Size([1, 15673]) Input IDs shape: torch.Size([1, 15673]) Labels shape: torch.Size([1, 15673]) Final batch size: 1, sequence length: 13789 Attention mask shape: torch.Size([1, 1, 13789, 13789]) Position ids shape: torch.Size([1, 13789]) Input IDs shape: torch.Size([1, 13789]) Labels shape: torch.Size([1, 13789]) Final batch size: 1, sequence length: 12960 Attention mask shape: torch.Size([1, 1, 12960, 12960]) Position ids shape: torch.Size([1, 12960]) Input IDs shape: torch.Size([1, 12960]) Labels shape: torch.Size([1, 12960]) Final batch size: 1, sequence length: 17245 Attention mask shape: torch.Size([1, 1, 17245, 17245]) Position ids shape: torch.Size([1, 17245]) Input IDs shape: torch.Size([1, 17245]) Labels shape: torch.Size([1, 17245]) Final batch size: 1, sequence length: 19899 Attention mask shape: torch.Size([1, 1, 19899, 19899]) Position ids shape: torch.Size([1, 19899]) Input IDs shape: torch.Size([1, 19899]) Labels shape: torch.Size([1, 19899]) Final batch size: 1, sequence length: 18214 Attention mask shape: torch.Size([1, 1, 18214, 18214]) Position ids shape: torch.Size([1, 18214]) Input IDs shape: torch.Size([1, 18214]) Labels shape: torch.Size([1, 18214]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 16450 Attention mask shape: torch.Size([1, 1, 16450, 16450]) Position ids shape: torch.Size([1, 16450]) Input IDs shape: torch.Size([1, 16450]) Labels shape: torch.Size([1, 16450]) Final batch size: 1, sequence length: 17934 Attention mask shape: torch.Size([1, 1, 17934, 17934]) Position ids shape: torch.Size([1, 17934]) Input IDs shape: torch.Size([1, 17934]) Labels shape: torch.Size([1, 17934]) Final batch size: 1, sequence length: 17861 Attention mask shape: torch.Size([1, 1, 17861, 17861]) Position ids shape: torch.Size([1, 17861]) Input IDs shape: torch.Size([1, 17861]) Labels shape: torch.Size([1, 17861]) Final batch size: 1, sequence length: 20065 Attention mask shape: torch.Size([1, 1, 20065, 20065]) Position ids shape: torch.Size([1, 20065]) Input IDs shape: torch.Size([1, 20065]) Labels shape: torch.Size([1, 20065]) Final batch size: 1, sequence length: 20292 Attention mask shape: torch.Size([1, 1, 20292, 20292]) Position ids shape: torch.Size([1, 20292]) Input IDs shape: torch.Size([1, 20292]) Labels shape: torch.Size([1, 20292]) Final batch size: 1, sequence length: 12728 Attention mask shape: torch.Size([1, 1, 12728, 12728]) Position ids shape: torch.Size([1, 12728]) Input IDs shape: torch.Size([1, 12728]) Labels shape: torch.Size([1, 12728]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 21675 Attention mask shape: torch.Size([1, 1, 21675, 21675]) Position ids shape: torch.Size([1, 21675]) Input IDs shape: torch.Size([1, 21675]) Labels shape: torch.Size([1, 21675]) Final batch size: 1, sequence length: 16036 Attention mask shape: torch.Size([1, 1, 16036, 16036]) Position ids shape: torch.Size([1, 16036]) Input IDs shape: torch.Size([1, 16036]) Labels shape: torch.Size([1, 16036]) Final batch size: 1, sequence length: 23896 Attention mask shape: torch.Size([1, 1, 23896, 23896]) Position ids shape: torch.Size([1, 23896]) Input IDs shape: torch.Size([1, 23896]) Labels shape: torch.Size([1, 23896]) Final batch size: 1, sequence length: 24956 Attention mask shape: torch.Size([1, 1, 24956, 24956]) Position ids shape: torch.Size([1, 24956]) Input IDs shape: torch.Size([1, 24956]) Labels shape: torch.Size([1, 24956]) Final batch size: 1, sequence length: 21091 Attention mask shape: torch.Size([1, 1, 21091, 21091]) Position ids shape: torch.Size([1, 21091]) Input IDs shape: torch.Size([1, 21091]) Labels shape: torch.Size([1, 21091]) Final batch size: 1, sequence length: 23004 Attention mask shape: torch.Size([1, 1, 23004, 23004]) Position ids shape: torch.Size([1, 23004]) Input IDs shape: torch.Size([1, 23004]) Labels shape: torch.Size([1, 23004]) Final batch size: 1, sequence length: 14212 Attention mask shape: torch.Size([1, 1, 14212, 14212]) Position ids shape: torch.Size([1, 14212]) Input IDs shape: torch.Size([1, 14212]) Labels shape: torch.Size([1, 14212]) Final batch size: 1, sequence length: 25435 Attention mask shape: torch.Size([1, 1, 25435, 25435]) Position ids shape: torch.Size([1, 25435]) Input IDs shape: torch.Size([1, 25435]) Labels shape: torch.Size([1, 25435]) Final batch size: 1, sequence length: 25176 Attention mask shape: torch.Size([1, 1, 25176, 25176]) Position ids shape: torch.Size([1, 25176]) Input IDs shape: torch.Size([1, 25176]) Labels shape: torch.Size([1, 25176]) Final batch size: 1, sequence length: 21586 Attention mask shape: torch.Size([1, 1, 21586, 21586]) Position ids shape: torch.Size([1, 21586]) Input IDs shape: torch.Size([1, 21586]) Labels shape: torch.Size([1, 21586]) Final batch size: 1, sequence length: 26316 Attention mask shape: torch.Size([1, 1, 26316, 26316]) Position ids shape: torch.Size([1, 26316]) Input IDs shape: torch.Size([1, 26316]) Labels shape: torch.Size([1, 26316]) Final batch size: 1, sequence length: 24808 Attention mask shape: torch.Size([1, 1, 24808, 24808]) Position ids shape: torch.Size([1, 24808]) Input IDs shape: torch.Size([1, 24808]) Labels shape: torch.Size([1, 24808]) Final batch size: 1, sequence length: 25035 Attention mask shape: torch.Size([1, 1, 25035, 25035]) Position ids shape: torch.Size([1, 25035]) Input IDs shape: torch.Size([1, 25035]) Labels shape: torch.Size([1, 25035]) Final batch size: 1, sequence length: 25600 Attention mask shape: torch.Size([1, 1, 25600, 25600]) Position ids shape: torch.Size([1, 25600]) Input IDs shape: torch.Size([1, 25600]) Labels shape: torch.Size([1, 25600]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 19187 Attention mask shape: torch.Size([1, 1, 19187, 19187]) Position ids shape: torch.Size([1, 19187]) Input IDs shape: torch.Size([1, 19187]) Labels shape: torch.Size([1, 19187]) Final batch size: 1, sequence length: 24927 Attention mask shape: torch.Size([1, 1, 24927, 24927]) Position ids shape: torch.Size([1, 24927]) Input IDs shape: torch.Size([1, 24927]) Labels shape: torch.Size([1, 24927]) Final batch size: 1, sequence length: 22778 Attention mask shape: torch.Size([1, 1, 22778, 22778]) Position ids shape: torch.Size([1, 22778]) Input IDs shape: torch.Size([1, 22778]) Labels shape: torch.Size([1, 22778]) Final batch size: 1, sequence length: 26247 Attention mask shape: torch.Size([1, 1, 26247, 26247]) Position ids shape: torch.Size([1, 26247]) Input IDs shape: torch.Size([1, 26247]) Labels shape: torch.Size([1, 26247]) Final batch size: 1, sequence length: 20582 Attention mask shape: torch.Size([1, 1, 20582, 20582]) Position ids shape: torch.Size([1, 20582]) Input IDs shape: torch.Size([1, 20582]) Labels shape: torch.Size([1, 20582]) Final batch size: 1, sequence length: 26500 Attention mask shape: torch.Size([1, 1, 26500, 26500]) Position ids shape: torch.Size([1, 26500]) Input IDs shape: torch.Size([1, 26500]) Labels shape: torch.Size([1, 26500]) Final batch size: 1, sequence length: 24633 Attention mask shape: torch.Size([1, 1, 24633, 24633]) Position ids shape: torch.Size([1, 24633]) Input IDs shape: torch.Size([1, 24633]) Labels shape: torch.Size([1, 24633]) Final batch size: 1, sequence length: 28926 Attention mask shape: torch.Size([1, 1, 28926, 28926]) Position ids shape: torch.Size([1, 28926]) Input IDs shape: torch.Size([1, 28926]) Labels shape: torch.Size([1, 28926]) Final batch size: 1, sequence length: 22775 Attention mask shape: torch.Size([1, 1, 22775, 22775]) Position ids shape: torch.Size([1, 22775]) Input IDs shape: torch.Size([1, 22775]) Labels shape: torch.Size([1, 22775]) Final batch size: 1, sequence length: 25325 Attention mask shape: torch.Size([1, 1, 25325, 25325]) Position ids shape: torch.Size([1, 25325]) Input IDs shape: torch.Size([1, 25325]) Labels shape: torch.Size([1, 25325]) Final batch size: 1, sequence length: 13946 Attention mask shape: torch.Size([1, 1, 13946, 13946]) Position ids shape: torch.Size([1, 13946]) Input IDs shape: torch.Size([1, 13946]) Labels shape: torch.Size([1, 13946]) Final batch size: 1, sequence length: 3010 Attention mask shape: torch.Size([1, 1, 3010, 3010]) Position ids shape: torch.Size([1, 3010]) Input IDs shape: torch.Size([1, 3010]) Labels shape: torch.Size([1, 3010]) Final batch size: 1, sequence length: 27222 Attention mask shape: torch.Size([1, 1, 27222, 27222]) Position ids shape: torch.Size([1, 27222]) Input IDs shape: torch.Size([1, 27222]) Labels shape: torch.Size([1, 27222]) Final batch size: 1, sequence length: 17763 Attention mask shape: torch.Size([1, 1, 17763, 17763]) Position ids shape: torch.Size([1, 17763]) Input IDs shape: torch.Size([1, 17763]) Labels shape: torch.Size([1, 17763]) Final batch size: 1, sequence length: 19953 Attention mask shape: torch.Size([1, 1, 19953, 19953]) Position ids shape: torch.Size([1, 19953]) Input IDs shape: torch.Size([1, 19953]) Labels shape: torch.Size([1, 19953]) Final batch size: 1, sequence length: 24965 Attention mask shape: torch.Size([1, 1, 24965, 24965]) Position ids shape: torch.Size([1, 24965]) Input IDs shape: torch.Size([1, 24965]) Labels shape: torch.Size([1, 24965]) Final batch size: 1, sequence length: 16714 Attention mask shape: torch.Size([1, 1, 16714, 16714]) Position ids shape: torch.Size([1, 16714]) Input IDs shape: torch.Size([1, 16714]) Labels shape: torch.Size([1, 16714]) Final batch size: 1, sequence length: 13453 Attention mask shape: torch.Size([1, 1, 13453, 13453]) Position ids shape: torch.Size([1, 13453]) Input IDs shape: torch.Size([1, 13453]) Labels shape: torch.Size([1, 13453]) Final batch size: 1, sequence length: 28552 Attention mask shape: torch.Size([1, 1, 28552, 28552]) Position ids shape: torch.Size([1, 28552]) Input IDs shape: torch.Size([1, 28552]) Labels shape: torch.Size([1, 28552]) Final batch size: 1, sequence length: 28644 Attention mask shape: torch.Size([1, 1, 28644, 28644]) Position ids shape: torch.Size([1, 28644]) Input IDs shape: torch.Size([1, 28644]) Labels shape: torch.Size([1, 28644]) Final batch size: 1, sequence length: 6978 Attention mask shape: torch.Size([1, 1, 6978, 6978]) Position ids shape: torch.Size([1, 6978]) Input IDs shape: torch.Size([1, 6978]) Labels shape: torch.Size([1, 6978]) Final batch size: 1, sequence length: 32513 Attention mask shape: torch.Size([1, 1, 32513, 32513]) Position ids shape: torch.Size([1, 32513]) Input IDs shape: torch.Size([1, 32513]) Labels shape: torch.Size([1, 32513]) Final batch size: 1, sequence length: 32101 Attention mask shape: torch.Size([1, 1, 32101, 32101]) Position ids shape: torch.Size([1, 32101]) Input IDs shape: torch.Size([1, 32101]) Labels shape: torch.Size([1, 32101]) Final batch size: 1, sequence length: 12426 Attention mask shape: torch.Size([1, 1, 12426, 12426]) Position ids shape: torch.Size([1, 12426]) Input IDs shape: torch.Size([1, 12426]) Labels shape: torch.Size([1, 12426]) Final batch size: 1, sequence length: 32100 Attention mask shape: torch.Size([1, 1, 32100, 32100]) Position ids shape: torch.Size([1, 32100]) Input IDs shape: torch.Size([1, 32100]) Labels shape: torch.Size([1, 32100]) Final batch size: 1, sequence length: 18832 Attention mask shape: torch.Size([1, 1, 18832, 18832]) Position ids shape: torch.Size([1, 18832]) Input IDs shape: torch.Size([1, 18832]) Labels shape: torch.Size([1, 18832]) Final batch size: 1, sequence length: 35058 Attention mask shape: torch.Size([1, 1, 35058, 35058]) Position ids shape: torch.Size([1, 35058]) Input IDs shape: torch.Size([1, 35058]) Labels shape: torch.Size([1, 35058]) Final batch size: 1, sequence length: 25979 Attention mask shape: torch.Size([1, 1, 25979, 25979]) Position ids shape: torch.Size([1, 25979]) Input IDs shape: torch.Size([1, 25979]) Labels shape: torch.Size([1, 25979]) Final batch size: 1, sequence length: 16571 Attention mask shape: torch.Size([1, 1, 16571, 16571]) Position ids shape: torch.Size([1, 16571]) Input IDs shape: torch.Size([1, 16571]) Labels shape: torch.Size([1, 16571]) Final batch size: 1, sequence length: 30635 Attention mask shape: torch.Size([1, 1, 30635, 30635]) Position ids shape: torch.Size([1, 30635]) Input IDs shape: torch.Size([1, 30635]) Labels shape: torch.Size([1, 30635]) Final batch size: 1, sequence length: 31662 Attention mask shape: torch.Size([1, 1, 31662, 31662]) Position ids shape: torch.Size([1, 31662]) Input IDs shape: torch.Size([1, 31662]) Labels shape: torch.Size([1, 31662]) Final batch size: 1, sequence length: 24676 Attention mask shape: torch.Size([1, 1, 24676, 24676]) Position ids shape: torch.Size([1, 24676]) Input IDs shape: torch.Size([1, 24676]) Labels shape: torch.Size([1, 24676]) Final batch size: 1, sequence length: 31208 Attention mask shape: torch.Size([1, 1, 31208, 31208]) Position ids shape: torch.Size([1, 31208]) Input IDs shape: torch.Size([1, 31208]) Labels shape: torch.Size([1, 31208]) Final batch size: 1, sequence length: 37381 Attention mask shape: torch.Size([1, 1, 37381, 37381]) Position ids shape: torch.Size([1, 37381]) Input IDs shape: torch.Size([1, 37381]) Labels shape: torch.Size([1, 37381]) Final batch size: 1, sequence length: 17049 Attention mask shape: torch.Size([1, 1, 17049, 17049]) Position ids shape: torch.Size([1, 17049]) Input IDs shape: torch.Size([1, 17049]) Labels shape: torch.Size([1, 17049]) Final batch size: 1, sequence length: 21290 Attention mask shape: torch.Size([1, 1, 21290, 21290]) Position ids shape: torch.Size([1, 21290]) Input IDs shape: torch.Size([1, 21290]) Labels shape: torch.Size([1, 21290]) Final batch size: 1, sequence length: 26789 Attention mask shape: torch.Size([1, 1, 26789, 26789]) Position ids shape: torch.Size([1, 26789]) Input IDs shape: torch.Size([1, 26789]) Labels shape: torch.Size([1, 26789]) Final batch size: 1, sequence length: 37391 Attention mask shape: torch.Size([1, 1, 37391, 37391]) Position ids shape: torch.Size([1, 37391]) Input IDs shape: torch.Size([1, 37391]) Labels shape: torch.Size([1, 37391]) Final batch size: 1, sequence length: 35014 Attention mask shape: torch.Size([1, 1, 35014, 35014]) Position ids shape: torch.Size([1, 35014]) Input IDs shape: torch.Size([1, 35014]) Labels shape: torch.Size([1, 35014]) Final batch size: 1, sequence length: 28377 Attention mask shape: torch.Size([1, 1, 28377, 28377]) Position ids shape: torch.Size([1, 28377]) Input IDs shape: torch.Size([1, 28377]) Labels shape: torch.Size([1, 28377]) Final batch size: 1, sequence length: 25147 Attention mask shape: torch.Size([1, 1, 25147, 25147]) Position ids shape: torch.Size([1, 25147]) Input IDs shape: torch.Size([1, 25147]) Labels shape: torch.Size([1, 25147]) Final batch size: 1, sequence length: 25774 Attention mask shape: torch.Size([1, 1, 25774, 25774]) Position ids shape: torch.Size([1, 25774]) Input IDs shape: torch.Size([1, 25774]) Labels shape: torch.Size([1, 25774]) Final batch size: 1, sequence length: 25219 Attention mask shape: torch.Size([1, 1, 25219, 25219]) Position ids shape: torch.Size([1, 25219]) Input IDs shape: torch.Size([1, 25219]) Labels shape: torch.Size([1, 25219]) Final batch size: 1, sequence length: 36047 Attention mask shape: torch.Size([1, 1, 36047, 36047]) Position ids shape: torch.Size([1, 36047]) Input IDs shape: torch.Size([1, 36047]) Labels shape: torch.Size([1, 36047]) Final batch size: 1, sequence length: 16816 Attention mask shape: torch.Size([1, 1, 16816, 16816]) Position ids shape: torch.Size([1, 16816]) Input IDs shape: torch.Size([1, 16816]) Labels shape: torch.Size([1, 16816]) Final batch size: 1, sequence length: 39474 Attention mask shape: torch.Size([1, 1, 39474, 39474]) Position ids shape: torch.Size([1, 39474]) Input IDs shape: torch.Size([1, 39474]) Labels shape: torch.Size([1, 39474]) Final batch size: 1, sequence length: 30259 Attention mask shape: torch.Size([1, 1, 30259, 30259]) Position ids shape: torch.Size([1, 30259]) Input IDs shape: torch.Size([1, 30259]) Labels shape: torch.Size([1, 30259]) Final batch size: 1, sequence length: 38391 Attention mask shape: torch.Size([1, 1, 38391, 38391]) Position ids shape: torch.Size([1, 38391]) Input IDs shape: torch.Size([1, 38391]) Labels shape: torch.Size([1, 38391]) Final batch size: 1, sequence length: 22710 Attention mask shape: torch.Size([1, 1, 22710, 22710]) Position ids shape: torch.Size([1, 22710]) Input IDs shape: torch.Size([1, 22710]) Labels shape: torch.Size([1, 22710]) Final batch size: 1, sequence length: 27021 Attention mask shape: torch.Size([1, 1, 27021, 27021]) Position ids shape: torch.Size([1, 27021]) Input IDs shape: torch.Size([1, 27021]) Labels shape: torch.Size([1, 27021]) Final batch size: 1, sequence length: 26387 Attention mask shape: torch.Size([1, 1, 26387, 26387]) Position ids shape: torch.Size([1, 26387]) Input IDs shape: torch.Size([1, 26387]) Labels shape: torch.Size([1, 26387]) Final batch size: 1, sequence length: 26454 Attention mask shape: torch.Size([1, 1, 26454, 26454]) Position ids shape: torch.Size([1, 26454]) Input IDs shape: torch.Size([1, 26454]) Labels shape: torch.Size([1, 26454]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38577 Attention mask shape: torch.Size([1, 1, 38577, 38577]) Position ids shape: torch.Size([1, 38577]) Input IDs shape: torch.Size([1, 38577]) Labels shape: torch.Size([1, 38577]) Final batch size: 1, sequence length: 21321 Attention mask shape: torch.Size([1, 1, 21321, 21321]) Position ids shape: torch.Size([1, 21321]) Input IDs shape: torch.Size([1, 21321]) Labels shape: torch.Size([1, 21321]) Final batch size: 1, sequence length: 33621 Attention mask shape: torch.Size([1, 1, 33621, 33621]) Position ids shape: torch.Size([1, 33621]) Input IDs shape: torch.Size([1, 33621]) Labels shape: torch.Size([1, 33621]) Final batch size: 1, sequence length: 37720 Attention mask shape: torch.Size([1, 1, 37720, 37720]) Position ids shape: torch.Size([1, 37720]) Input IDs shape: torch.Size([1, 37720]) Labels shape: torch.Size([1, 37720]) Final batch size: 1, sequence length: 31447 Attention mask shape: torch.Size([1, 1, 31447, 31447]) Position ids shape: torch.Size([1, 31447]) Input IDs shape: torch.Size([1, 31447]) Labels shape: torch.Size([1, 31447]) Final batch size: 1, sequence length: 19441 Attention mask shape: torch.Size([1, 1, 19441, 19441]) Position ids shape: torch.Size([1, 19441]) Input IDs shape: torch.Size([1, 19441]) Labels shape: torch.Size([1, 19441]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 35261 Attention mask shape: torch.Size([1, 1, 35261, 35261]) Position ids shape: torch.Size([1, 35261]) Input IDs shape: torch.Size([1, 35261]) Labels shape: torch.Size([1, 35261]) Final batch size: 1, sequence length: 40590 Attention mask shape: torch.Size([1, 1, 40590, 40590]) Position ids shape: torch.Size([1, 40590]) Input IDs shape: torch.Size([1, 40590]) Labels shape: torch.Size([1, 40590]) Final batch size: 1, sequence length: 39955 Attention mask shape: torch.Size([1, 1, 39955, 39955]) Position ids shape: torch.Size([1, 39955]) Input IDs shape: torch.Size([1, 39955]) Labels shape: torch.Size([1, 39955]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28448 Attention mask shape: torch.Size([1, 1, 28448, 28448]) Position ids shape: torch.Size([1, 28448]) Input IDs shape: torch.Size([1, 28448]) Labels shape: torch.Size([1, 28448]) Final batch size: 1, sequence length: 32079 Attention mask shape: torch.Size([1, 1, 32079, 32079]) Position ids shape: torch.Size([1, 32079]) Input IDs shape: torch.Size([1, 32079]) Labels shape: torch.Size([1, 32079]) Final batch size: 1, sequence length: 32129 Attention mask shape: torch.Size([1, 1, 32129, 32129]) Position ids shape: torch.Size([1, 32129]) Input IDs shape: torch.Size([1, 32129]) Labels shape: torch.Size([1, 32129]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36356 Attention mask shape: torch.Size([1, 1, 36356, 36356]) Position ids shape: torch.Size([1, 36356]) Input IDs shape: torch.Size([1, 36356]) Labels shape: torch.Size([1, 36356]) Final batch size: 1, sequence length: 14136 Attention mask shape: torch.Size([1, 1, 14136, 14136]) Position ids shape: torch.Size([1, 14136]) Input IDs shape: torch.Size([1, 14136]) Labels shape: torch.Size([1, 14136]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24623 Attention mask shape: torch.Size([1, 1, 24623, 24623]) Position ids shape: torch.Size([1, 24623]) Input IDs shape: torch.Size([1, 24623]) Labels shape: torch.Size([1, 24623]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33014 Attention mask shape: torch.Size([1, 1, 33014, 33014]) Position ids shape: torch.Size([1, 33014]) Input IDs shape: torch.Size([1, 33014]) Labels shape: torch.Size([1, 33014]) Final batch size: 1, sequence length: 27371 Attention mask shape: torch.Size([1, 1, 27371, 27371]) Position ids shape: torch.Size([1, 27371]) Input IDs shape: torch.Size([1, 27371]) Labels shape: torch.Size([1, 27371]) Final batch size: 1, sequence length: 30712 Attention mask shape: torch.Size([1, 1, 30712, 30712]) Position ids shape: torch.Size([1, 30712]) Input IDs shape: torch.Size([1, 30712]) Labels shape: torch.Size([1, 30712]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19027 Attention mask shape: torch.Size([1, 1, 19027, 19027]) Position ids shape: torch.Size([1, 19027]) Input IDs shape: torch.Size([1, 19027]) Labels shape: torch.Size([1, 19027]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 23208 Attention mask shape: torch.Size([1, 1, 23208, 23208]) Position ids shape: torch.Size([1, 23208]) Input IDs shape: torch.Size([1, 23208]) Labels shape: torch.Size([1, 23208]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2587, 'grad_norm': 0.22048673162645635, 'learning_rate': 2.061073738537635e-06, 'num_tokens': -inf, 'epoch': 5.88} Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 10507 Attention mask shape: torch.Size([1, 1, 10507, 10507]) Position ids shape: torch.Size([1, 10507]) Input IDs shape: torch.Size([1, 10507]) Labels shape: torch.Size([1, 10507]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 26271 Attention mask shape: torch.Size([1, 1, 26271, 26271]) Position ids shape: torch.Size([1, 26271]) Input IDs shape: torch.Size([1, 26271]) Labels shape: torch.Size([1, 26271]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 31381 Attention mask shape: torch.Size([1, 1, 31381, 31381]) Position ids shape: torch.Size([1, 31381]) Input IDs shape: torch.Size([1, 31381]) Labels shape: torch.Size([1, 31381]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2502, 'grad_norm': 0.2509582936056328, 'learning_rate': 1.8533980447508138e-06, 'num_tokens': -inf, 'epoch': 6.0} Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 6362 Attention mask shape: torch.Size([1, 1, 6362, 6362]) Position ids shape: torch.Size([1, 6362]) Input IDs shape: torch.Size([1, 6362]) Labels shape: torch.Size([1, 6362]) Final batch size: 1, sequence length: 10523 Attention mask shape: torch.Size([1, 1, 10523, 10523]) Position ids shape: torch.Size([1, 10523]) Input IDs shape: torch.Size([1, 10523]) Labels shape: torch.Size([1, 10523]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 12385 Attention mask shape: torch.Size([1, 1, 12385, 12385]) Position ids shape: torch.Size([1, 12385]) Input IDs shape: torch.Size([1, 12385]) Labels shape: torch.Size([1, 12385]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 18410 Attention mask shape: torch.Size([1, 1, 18410, 18410]) Position ids shape: torch.Size([1, 18410]) Input IDs shape: torch.Size([1, 18410]) Labels shape: torch.Size([1, 18410]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 18098 Attention mask shape: torch.Size([1, 1, 18098, 18098]) Position ids shape: torch.Size([1, 18098]) Input IDs shape: torch.Size([1, 18098]) Labels shape: torch.Size([1, 18098]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 12656 Attention mask shape: torch.Size([1, 1, 12656, 12656]) Position ids shape: torch.Size([1, 12656]) Input IDs shape: torch.Size([1, 12656]) Labels shape: torch.Size([1, 12656]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 24365 Attention mask shape: torch.Size([1, 1, 24365, 24365]) Position ids shape: torch.Size([1, 24365]) Input IDs shape: torch.Size([1, 24365]) Labels shape: torch.Size([1, 24365]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 17299 Attention mask shape: torch.Size([1, 1, 17299, 17299]) Position ids shape: torch.Size([1, 17299]) Input IDs shape: torch.Size([1, 17299]) Labels shape: torch.Size([1, 17299]) Final batch size: 1, sequence length: 18060 Attention mask shape: torch.Size([1, 1, 18060, 18060]) Position ids shape: torch.Size([1, 18060]) Input IDs shape: torch.Size([1, 18060]) Labels shape: torch.Size([1, 18060]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 11819 Attention mask shape: torch.Size([1, 1, 11819, 11819]) Position ids shape: torch.Size([1, 11819]) Input IDs shape: torch.Size([1, 11819]) Labels shape: torch.Size([1, 11819]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 9091 Attention mask shape: torch.Size([1, 1, 9091, 9091]) Position ids shape: torch.Size([1, 9091]) Input IDs shape: torch.Size([1, 9091]) Labels shape: torch.Size([1, 9091]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 18915 Attention mask shape: torch.Size([1, 1, 18915, 18915]) Position ids shape: torch.Size([1, 18915]) Input IDs shape: torch.Size([1, 18915]) Labels shape: torch.Size([1, 18915]) Final batch size: 1, sequence length: 26976 Attention mask shape: torch.Size([1, 1, 26976, 26976]) Position ids shape: torch.Size([1, 26976]) Input IDs shape: torch.Size([1, 26976]) Labels shape: torch.Size([1, 26976]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 6948 Attention mask shape: torch.Size([1, 1, 6948, 6948]) Position ids shape: torch.Size([1, 6948]) Input IDs shape: torch.Size([1, 6948]) Labels shape: torch.Size([1, 6948]) Final batch size: 1, sequence length: 28841 Attention mask shape: torch.Size([1, 1, 28841, 28841]) Position ids shape: torch.Size([1, 28841]) Input IDs shape: torch.Size([1, 28841]) Labels shape: torch.Size([1, 28841]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 27541 Attention mask shape: torch.Size([1, 1, 27541, 27541]) Position ids shape: torch.Size([1, 27541]) Input IDs shape: torch.Size([1, 27541]) Labels shape: torch.Size([1, 27541]) Final batch size: 1, sequence length: 13215 Attention mask shape: torch.Size([1, 1, 13215, 13215]) Position ids shape: torch.Size([1, 13215]) Input IDs shape: torch.Size([1, 13215]) Labels shape: torch.Size([1, 13215]) Final batch size: 1, sequence length: 24880 Attention mask shape: torch.Size([1, 1, 24880, 24880]) Position ids shape: torch.Size([1, 24880]) Input IDs shape: torch.Size([1, 24880]) Labels shape: torch.Size([1, 24880]) Final batch size: 1, sequence length: 7722 Attention mask shape: torch.Size([1, 1, 7722, 7722]) Position ids shape: torch.Size([1, 7722]) Input IDs shape: torch.Size([1, 7722]) Labels shape: torch.Size([1, 7722]) Final batch size: 1, sequence length: 25172 Attention mask shape: torch.Size([1, 1, 25172, 25172]) Position ids shape: torch.Size([1, 25172]) Input IDs shape: torch.Size([1, 25172]) Labels shape: torch.Size([1, 25172]) Final batch size: 1, sequence length: 29113 Attention mask shape: torch.Size([1, 1, 29113, 29113]) Position ids shape: torch.Size([1, 29113]) Input IDs shape: torch.Size([1, 29113]) Labels shape: torch.Size([1, 29113]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 21450 Attention mask shape: torch.Size([1, 1, 21450, 21450]) Position ids shape: torch.Size([1, 21450]) Input IDs shape: torch.Size([1, 21450]) Labels shape: torch.Size([1, 21450]) Final batch size: 1, sequence length: 21596 Attention mask shape: torch.Size([1, 1, 21596, 21596]) Position ids shape: torch.Size([1, 21596]) Input IDs shape: torch.Size([1, 21596]) Labels shape: torch.Size([1, 21596]) Final batch size: 1, sequence length: 21491 Attention mask shape: torch.Size([1, 1, 21491, 21491]) Position ids shape: torch.Size([1, 21491]) Input IDs shape: torch.Size([1, 21491]) Labels shape: torch.Size([1, 21491]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 29875 Attention mask shape: torch.Size([1, 1, 29875, 29875]) Position ids shape: torch.Size([1, 29875]) Input IDs shape: torch.Size([1, 29875]) Labels shape: torch.Size([1, 29875]) Final batch size: 1, sequence length: 14009 Attention mask shape: torch.Size([1, 1, 14009, 14009]) Position ids shape: torch.Size([1, 14009]) Input IDs shape: torch.Size([1, 14009]) Labels shape: torch.Size([1, 14009]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 20363 Attention mask shape: torch.Size([1, 1, 20363, 20363]) Position ids shape: torch.Size([1, 20363]) Input IDs shape: torch.Size([1, 20363]) Labels shape: torch.Size([1, 20363]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 20827 Attention mask shape: torch.Size([1, 1, 20827, 20827]) Position ids shape: torch.Size([1, 20827]) Input IDs shape: torch.Size([1, 20827]) Labels shape: torch.Size([1, 20827]) Final batch size: 1, sequence length: 30623 Attention mask shape: torch.Size([1, 1, 30623, 30623]) Position ids shape: torch.Size([1, 30623]) Input IDs shape: torch.Size([1, 30623]) Labels shape: torch.Size([1, 30623]) Final batch size: 1, sequence length: 17400 Attention mask shape: torch.Size([1, 1, 17400, 17400]) Position ids shape: torch.Size([1, 17400]) Input IDs shape: torch.Size([1, 17400]) Labels shape: torch.Size([1, 17400]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 19705 Attention mask shape: torch.Size([1, 1, 19705, 19705]) Position ids shape: torch.Size([1, 19705]) Input IDs shape: torch.Size([1, 19705]) Labels shape: torch.Size([1, 19705]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 28641 Attention mask shape: torch.Size([1, 1, 28641, 28641]) Position ids shape: torch.Size([1, 28641]) Input IDs shape: torch.Size([1, 28641]) Labels shape: torch.Size([1, 28641]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 22098 Attention mask shape: torch.Size([1, 1, 22098, 22098]) Position ids shape: torch.Size([1, 22098]) Input IDs shape: torch.Size([1, 22098]) Labels shape: torch.Size([1, 22098]) Final batch size: 1, sequence length: 30101 Attention mask shape: torch.Size([1, 1, 30101, 30101]) Position ids shape: torch.Size([1, 30101]) Input IDs shape: torch.Size([1, 30101]) Labels shape: torch.Size([1, 30101]) Final batch size: 1, sequence length: 22625 Attention mask shape: torch.Size([1, 1, 22625, 22625]) Position ids shape: torch.Size([1, 22625]) Input IDs shape: torch.Size([1, 22625]) Labels shape: torch.Size([1, 22625]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 27243 Attention mask shape: torch.Size([1, 1, 27243, 27243]) Position ids shape: torch.Size([1, 27243]) Input IDs shape: torch.Size([1, 27243]) Labels shape: torch.Size([1, 27243]) Final batch size: 1, sequence length: 27265 Attention mask shape: torch.Size([1, 1, 27265, 27265]) Position ids shape: torch.Size([1, 27265]) Input IDs shape: torch.Size([1, 27265]) Labels shape: torch.Size([1, 27265]) Final batch size: 1, sequence length: 24001 Attention mask shape: torch.Size([1, 1, 24001, 24001]) Position ids shape: torch.Size([1, 24001]) Input IDs shape: torch.Size([1, 24001]) Labels shape: torch.Size([1, 24001]) Final batch size: 1, sequence length: 21348 Attention mask shape: torch.Size([1, 1, 21348, 21348]) Position ids shape: torch.Size([1, 21348]) Input IDs shape: torch.Size([1, 21348]) Labels shape: torch.Size([1, 21348]) Final batch size: 1, sequence length: 30066 Attention mask shape: torch.Size([1, 1, 30066, 30066]) Position ids shape: torch.Size([1, 30066]) Input IDs shape: torch.Size([1, 30066]) Labels shape: torch.Size([1, 30066]) Final batch size: 1, sequence length: 39142 Attention mask shape: torch.Size([1, 1, 39142, 39142]) Position ids shape: torch.Size([1, 39142]) Input IDs shape: torch.Size([1, 39142]) Labels shape: torch.Size([1, 39142]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 24622 Attention mask shape: torch.Size([1, 1, 24622, 24622]) Position ids shape: torch.Size([1, 24622]) Input IDs shape: torch.Size([1, 24622]) Labels shape: torch.Size([1, 24622]) Final batch size: 1, sequence length: 17778 Attention mask shape: torch.Size([1, 1, 17778, 17778]) Position ids shape: torch.Size([1, 17778]) Input IDs shape: torch.Size([1, 17778]) Labels shape: torch.Size([1, 17778]) Final batch size: 1, sequence length: 19586 Attention mask shape: torch.Size([1, 1, 19586, 19586]) Position ids shape: torch.Size([1, 19586]) Input IDs shape: torch.Size([1, 19586]) Labels shape: torch.Size([1, 19586]) Final batch size: 1, sequence length: 26939 Attention mask shape: torch.Size([1, 1, 26939, 26939]) Position ids shape: torch.Size([1, 26939]) Input IDs shape: torch.Size([1, 26939]) Labels shape: torch.Size([1, 26939]) Final batch size: 1, sequence length: 18565 Attention mask shape: torch.Size([1, 1, 18565, 18565]) Position ids shape: torch.Size([1, 18565]) Input IDs shape: torch.Size([1, 18565]) Labels shape: torch.Size([1, 18565]) Final batch size: 1, sequence length: 21567 Attention mask shape: torch.Size([1, 1, 21567, 21567]) Position ids shape: torch.Size([1, 21567]) Input IDs shape: torch.Size([1, 21567]) Labels shape: torch.Size([1, 21567]) Final batch size: 1, sequence length: 33367 Attention mask shape: torch.Size([1, 1, 33367, 33367]) Position ids shape: torch.Size([1, 33367]) Input IDs shape: torch.Size([1, 33367]) Labels shape: torch.Size([1, 33367]) Final batch size: 1, sequence length: 27441 Attention mask shape: torch.Size([1, 1, 27441, 27441]) Position ids shape: torch.Size([1, 27441]) Input IDs shape: torch.Size([1, 27441]) Labels shape: torch.Size([1, 27441]) Final batch size: 1, sequence length: 38249 Attention mask shape: torch.Size([1, 1, 38249, 38249]) Position ids shape: torch.Size([1, 38249]) Input IDs shape: torch.Size([1, 38249]) Labels shape: torch.Size([1, 38249]) Final batch size: 1, sequence length: 36271 Attention mask shape: torch.Size([1, 1, 36271, 36271]) Position ids shape: torch.Size([1, 36271]) Input IDs shape: torch.Size([1, 36271]) Labels shape: torch.Size([1, 36271]) Final batch size: 1, sequence length: 30802 Attention mask shape: torch.Size([1, 1, 30802, 30802]) Position ids shape: torch.Size([1, 30802]) Input IDs shape: torch.Size([1, 30802]) Labels shape: torch.Size([1, 30802]) Final batch size: 1, sequence length: 25014 Attention mask shape: torch.Size([1, 1, 25014, 25014]) Position ids shape: torch.Size([1, 25014]) Input IDs shape: torch.Size([1, 25014]) Labels shape: torch.Size([1, 25014]) Final batch size: 1, sequence length: 15217 Attention mask shape: torch.Size([1, 1, 15217, 15217]) Position ids shape: torch.Size([1, 15217]) Input IDs shape: torch.Size([1, 15217]) Labels shape: torch.Size([1, 15217]) Final batch size: 1, sequence length: 29632 Attention mask shape: torch.Size([1, 1, 29632, 29632]) Position ids shape: torch.Size([1, 29632]) Input IDs shape: torch.Size([1, 29632]) Labels shape: torch.Size([1, 29632]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16587 Attention mask shape: torch.Size([1, 1, 16587, 16587]) Position ids shape: torch.Size([1, 16587]) Input IDs shape: torch.Size([1, 16587]) Labels shape: torch.Size([1, 16587]) Final batch size: 1, sequence length: 15656 Attention mask shape: torch.Size([1, 1, 15656, 15656]) Position ids shape: torch.Size([1, 15656]) Input IDs shape: torch.Size([1, 15656]) Labels shape: torch.Size([1, 15656]) Final batch size: 1, sequence length: 35153 Attention mask shape: torch.Size([1, 1, 35153, 35153]) Position ids shape: torch.Size([1, 35153]) Input IDs shape: torch.Size([1, 35153]) Labels shape: torch.Size([1, 35153]) Final batch size: 1, sequence length: 19036 Attention mask shape: torch.Size([1, 1, 19036, 19036]) Position ids shape: torch.Size([1, 19036]) Input IDs shape: torch.Size([1, 19036]) Labels shape: torch.Size([1, 19036]) Final batch size: 1, sequence length: 25447 Attention mask shape: torch.Size([1, 1, 25447, 25447]) Position ids shape: torch.Size([1, 25447]) Input IDs shape: torch.Size([1, 25447]) Labels shape: torch.Size([1, 25447]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32609 Attention mask shape: torch.Size([1, 1, 32609, 32609]) Position ids shape: torch.Size([1, 32609]) Input IDs shape: torch.Size([1, 32609]) Labels shape: torch.Size([1, 32609]) Final batch size: 1, sequence length: 33839 Attention mask shape: torch.Size([1, 1, 33839, 33839]) Position ids shape: torch.Size([1, 33839]) Input IDs shape: torch.Size([1, 33839]) Labels shape: torch.Size([1, 33839]) Final batch size: 1, sequence length: 39836 Attention mask shape: torch.Size([1, 1, 39836, 39836]) Position ids shape: torch.Size([1, 39836]) Input IDs shape: torch.Size([1, 39836]) Labels shape: torch.Size([1, 39836]) Final batch size: 1, sequence length: 15508 Final batch size: 1, sequence length: 32352 Attention mask shape: torch.Size([1, 1, 15508, 15508]) Position ids shape: torch.Size([1, 15508]) Input IDs shape: torch.Size([1, 15508]) Labels shape: torch.Size([1, 15508]) Attention mask shape: torch.Size([1, 1, 32352, 32352]) Position ids shape: torch.Size([1, 32352]) Input IDs shape: torch.Size([1, 32352]) Labels shape: torch.Size([1, 32352]) Final batch size: 1, sequence length: 26306 Attention mask shape: torch.Size([1, 1, 26306, 26306]) Position ids shape: torch.Size([1, 26306]) Input IDs shape: torch.Size([1, 26306]) Labels shape: torch.Size([1, 26306]) Final batch size: 1, sequence length: 16677 Attention mask shape: torch.Size([1, 1, 16677, 16677]) Position ids shape: torch.Size([1, 16677]) Input IDs shape: torch.Size([1, 16677]) Labels shape: torch.Size([1, 16677]) Final batch size: 1, sequence length: 40496 Attention mask shape: torch.Size([1, 1, 40496, 40496]) Position ids shape: torch.Size([1, 40496]) Input IDs shape: torch.Size([1, 40496]) Labels shape: torch.Size([1, 40496]) Final batch size: 1, sequence length: 13509 Attention mask shape: torch.Size([1, 1, 13509, 13509]) Position ids shape: torch.Size([1, 13509]) Input IDs shape: torch.Size([1, 13509]) Labels shape: torch.Size([1, 13509]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 9947 Attention mask shape: torch.Size([1, 1, 9947, 9947]) Position ids shape: torch.Size([1, 9947]) Input IDs shape: torch.Size([1, 9947]) Labels shape: torch.Size([1, 9947]) Final batch size: 1, sequence length: 21758 Attention mask shape: torch.Size([1, 1, 21758, 21758]) Position ids shape: torch.Size([1, 21758]) Input IDs shape: torch.Size([1, 21758]) Labels shape: torch.Size([1, 21758]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40745 Attention mask shape: torch.Size([1, 1, 40745, 40745]) Position ids shape: torch.Size([1, 40745]) Input IDs shape: torch.Size([1, 40745]) Labels shape: torch.Size([1, 40745]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36511 Attention mask shape: torch.Size([1, 1, 36511, 36511]) Position ids shape: torch.Size([1, 36511]) Input IDs shape: torch.Size([1, 36511]) Labels shape: torch.Size([1, 36511]) Final batch size: 1, sequence length: 37945 Attention mask shape: torch.Size([1, 1, 37945, 37945]) Position ids shape: torch.Size([1, 37945]) Input IDs shape: torch.Size([1, 37945]) Labels shape: torch.Size([1, 37945]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 18606 Attention mask shape: torch.Size([1, 1, 18606, 18606]) Position ids shape: torch.Size([1, 18606]) Input IDs shape: torch.Size([1, 18606]) Labels shape: torch.Size([1, 18606]) Final batch size: 1, sequence length: 23258 Attention mask shape: torch.Size([1, 1, 23258, 23258]) Position ids shape: torch.Size([1, 23258]) Input IDs shape: torch.Size([1, 23258]) Labels shape: torch.Size([1, 23258]) Final batch size: 1, sequence length: 20611 Attention mask shape: torch.Size([1, 1, 20611, 20611]) Position ids shape: torch.Size([1, 20611]) Input IDs shape: torch.Size([1, 20611]) Labels shape: torch.Size([1, 20611]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30109 Attention mask shape: torch.Size([1, 1, 30109, 30109]) Position ids shape: torch.Size([1, 30109]) Input IDs shape: torch.Size([1, 30109]) Labels shape: torch.Size([1, 30109]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 13297 Attention mask shape: torch.Size([1, 1, 13297, 13297]) Position ids shape: torch.Size([1, 13297]) Input IDs shape: torch.Size([1, 13297]) Labels shape: torch.Size([1, 13297]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20422 Attention mask shape: torch.Size([1, 1, 20422, 20422]) Position ids shape: torch.Size([1, 20422]) Input IDs shape: torch.Size([1, 20422]) Labels shape: torch.Size([1, 20422]) Final batch size: 1, sequence length: 12224 Attention mask shape: torch.Size([1, 1, 12224, 12224]) Position ids shape: torch.Size([1, 12224]) Input IDs shape: torch.Size([1, 12224]) Labels shape: torch.Size([1, 12224]) Final batch size: 1, sequence length: 13095 Attention mask shape: torch.Size([1, 1, 13095, 13095]) Position ids shape: torch.Size([1, 13095]) Input IDs shape: torch.Size([1, 13095]) Labels shape: torch.Size([1, 13095]) Final batch size: 1, sequence length: 34142 Attention mask shape: torch.Size([1, 1, 34142, 34142]) Position ids shape: torch.Size([1, 34142]) Input IDs shape: torch.Size([1, 34142]) Labels shape: torch.Size([1, 34142]) Final batch size: 1, sequence length: 28634 Attention mask shape: torch.Size([1, 1, 28634, 28634]) Position ids shape: torch.Size([1, 28634]) Input IDs shape: torch.Size([1, 28634]) Labels shape: torch.Size([1, 28634]) Final batch size: 1, sequence length: 32472 Attention mask shape: torch.Size([1, 1, 32472, 32472]) Position ids shape: torch.Size([1, 32472]) Input IDs shape: torch.Size([1, 32472]) Labels shape: torch.Size([1, 32472]) Final batch size: 1, sequence length: 17373 Attention mask shape: torch.Size([1, 1, 17373, 17373]) Position ids shape: torch.Size([1, 17373]) Input IDs shape: torch.Size([1, 17373]) Labels shape: torch.Size([1, 17373]) Final batch size: 1, sequence length: 24298 Attention mask shape: torch.Size([1, 1, 24298, 24298]) Position ids shape: torch.Size([1, 24298]) Input IDs shape: torch.Size([1, 24298]) Labels shape: torch.Size([1, 24298]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34937 Attention mask shape: torch.Size([1, 1, 34937, 34937]) Position ids shape: torch.Size([1, 34937]) Input IDs shape: torch.Size([1, 34937]) Labels shape: torch.Size([1, 34937]) Final batch size: 1, sequence length: 12653 Attention mask shape: torch.Size([1, 1, 12653, 12653]) Position ids shape: torch.Size([1, 12653]) Input IDs shape: torch.Size([1, 12653]) Labels shape: torch.Size([1, 12653]) Final batch size: 1, sequence length: 21683 Attention mask shape: torch.Size([1, 1, 21683, 21683]) Position ids shape: torch.Size([1, 21683]) Input IDs shape: torch.Size([1, 21683]) Labels shape: torch.Size([1, 21683]) Final batch size: 1, sequence length: 22786 Attention mask shape: torch.Size([1, 1, 22786, 22786]) Position ids shape: torch.Size([1, 22786]) Input IDs shape: torch.Size([1, 22786]) Labels shape: torch.Size([1, 22786]) Final batch size: 1, sequence length: 32753 Attention mask shape: torch.Size([1, 1, 32753, 32753]) Position ids shape: torch.Size([1, 32753]) Input IDs shape: torch.Size([1, 32753]) Labels shape: torch.Size([1, 32753]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22623 Attention mask shape: torch.Size([1, 1, 22623, 22623]) Position ids shape: torch.Size([1, 22623]) Input IDs shape: torch.Size([1, 22623]) Labels shape: torch.Size([1, 22623]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 14995 Attention mask shape: torch.Size([1, 1, 14995, 14995]) Position ids shape: torch.Size([1, 14995]) Input IDs shape: torch.Size([1, 14995]) Labels shape: torch.Size([1, 14995]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32564 Attention mask shape: torch.Size([1, 1, 32564, 32564]) Position ids shape: torch.Size([1, 32564]) Input IDs shape: torch.Size([1, 32564]) Labels shape: torch.Size([1, 32564]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24424 Attention mask shape: torch.Size([1, 1, 24424, 24424]) Position ids shape: torch.Size([1, 24424]) Input IDs shape: torch.Size([1, 24424]) Labels shape: torch.Size([1, 24424]) Final batch size: 1, sequence length: 35864 Attention mask shape: torch.Size([1, 1, 35864, 35864]) Position ids shape: torch.Size([1, 35864]) Input IDs shape: torch.Size([1, 35864]) Labels shape: torch.Size([1, 35864]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32344 Attention mask shape: torch.Size([1, 1, 32344, 32344]) Position ids shape: torch.Size([1, 32344]) Input IDs shape: torch.Size([1, 32344]) Labels shape: torch.Size([1, 32344]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26537 Attention mask shape: torch.Size([1, 1, 26537, 26537]) Position ids shape: torch.Size([1, 26537]) Input IDs shape: torch.Size([1, 26537]) Labels shape: torch.Size([1, 26537]) Final batch size: 1, sequence length: 26006 Attention mask shape: torch.Size([1, 1, 26006, 26006]) Position ids shape: torch.Size([1, 26006]) Input IDs shape: torch.Size([1, 26006]) Labels shape: torch.Size([1, 26006]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32919 Attention mask shape: torch.Size([1, 1, 32919, 32919]) Position ids shape: torch.Size([1, 32919]) Input IDs shape: torch.Size([1, 32919]) Labels shape: torch.Size([1, 32919]) Final batch size: 1, sequence length: 37168 Attention mask shape: torch.Size([1, 1, 37168, 37168]) Position ids shape: torch.Size([1, 37168]) Input IDs shape: torch.Size([1, 37168]) Labels shape: torch.Size([1, 37168]) Final batch size: 1, sequence length: 23362 Attention mask shape: torch.Size([1, 1, 23362, 23362]) Position ids shape: torch.Size([1, 23362]) Input IDs shape: torch.Size([1, 23362]) Labels shape: torch.Size([1, 23362]) Final batch size: 1, sequence length: 32514 Attention mask shape: torch.Size([1, 1, 32514, 32514]) Position ids shape: torch.Size([1, 32514]) Input IDs shape: torch.Size([1, 32514]) Labels shape: torch.Size([1, 32514]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28861 Attention mask shape: torch.Size([1, 1, 28861, 28861]) Position ids shape: torch.Size([1, 28861]) Input IDs shape: torch.Size([1, 28861]) Labels shape: torch.Size([1, 28861]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2447, 'grad_norm': 0.23790091087712154, 'learning_rate': 1.6543469682057105e-06, 'num_tokens': -inf, 'epoch': 6.12} Final batch size: 1, sequence length: 4858 Attention mask shape: torch.Size([1, 1, 4858, 4858]) Position ids shape: torch.Size([1, 4858]) Input IDs shape: torch.Size([1, 4858]) Labels shape: torch.Size([1, 4858]) Final batch size: 1, sequence length: 6316 Attention mask shape: torch.Size([1, 1, 6316, 6316]) Position ids shape: torch.Size([1, 6316]) Input IDs shape: torch.Size([1, 6316]) Labels shape: torch.Size([1, 6316]) Final batch size: 1, sequence length: 7360 Attention mask shape: torch.Size([1, 1, 7360, 7360]) Position ids shape: torch.Size([1, 7360]) Input IDs shape: torch.Size([1, 7360]) Labels shape: torch.Size([1, 7360]) Final batch size: 1, sequence length: 10301 Attention mask shape: torch.Size([1, 1, 10301, 10301]) Position ids shape: torch.Size([1, 10301]) Input IDs shape: torch.Size([1, 10301]) Labels shape: torch.Size([1, 10301]) Final batch size: 1, sequence length: 11548 Attention mask shape: torch.Size([1, 1, 11548, 11548]) Position ids shape: torch.Size([1, 11548]) Input IDs shape: torch.Size([1, 11548]) Labels shape: torch.Size([1, 11548]) Final batch size: 1, sequence length: 11728 Attention mask shape: torch.Size([1, 1, 11728, 11728]) Position ids shape: torch.Size([1, 11728]) Input IDs shape: torch.Size([1, 11728]) Labels shape: torch.Size([1, 11728]) Final batch size: 1, sequence length: 12293 Attention mask shape: torch.Size([1, 1, 12293, 12293]) Position ids shape: torch.Size([1, 12293]) Input IDs shape: torch.Size([1, 12293]) Labels shape: torch.Size([1, 12293]) Final batch size: 1, sequence length: 13355 Attention mask shape: torch.Size([1, 1, 13355, 13355]) Position ids shape: torch.Size([1, 13355]) Input IDs shape: torch.Size([1, 13355]) Labels shape: torch.Size([1, 13355]) Final batch size: 1, sequence length: 14363 Attention mask shape: torch.Size([1, 1, 14363, 14363]) Position ids shape: torch.Size([1, 14363]) Input IDs shape: torch.Size([1, 14363]) Labels shape: torch.Size([1, 14363]) Final batch size: 1, sequence length: 17108 Attention mask shape: torch.Size([1, 1, 17108, 17108]) Position ids shape: torch.Size([1, 17108]) Input IDs shape: torch.Size([1, 17108]) Labels shape: torch.Size([1, 17108]) Final batch size: 1, sequence length: 14704 Attention mask shape: torch.Size([1, 1, 14704, 14704]) Position ids shape: torch.Size([1, 14704]) Input IDs shape: torch.Size([1, 14704]) Labels shape: torch.Size([1, 14704]) Final batch size: 1, sequence length: 17415 Attention mask shape: torch.Size([1, 1, 17415, 17415]) Position ids shape: torch.Size([1, 17415]) Input IDs shape: torch.Size([1, 17415]) Labels shape: torch.Size([1, 17415]) Final batch size: 1, sequence length: 14587 Attention mask shape: torch.Size([1, 1, 14587, 14587]) Position ids shape: torch.Size([1, 14587]) Input IDs shape: torch.Size([1, 14587]) Labels shape: torch.Size([1, 14587]) Final batch size: 1, sequence length: 18653 Attention mask shape: torch.Size([1, 1, 18653, 18653]) Position ids shape: torch.Size([1, 18653]) Input IDs shape: torch.Size([1, 18653]) Labels shape: torch.Size([1, 18653]) Final batch size: 1, sequence length: 17220 Attention mask shape: torch.Size([1, 1, 17220, 17220]) Position ids shape: torch.Size([1, 17220]) Input IDs shape: torch.Size([1, 17220]) Labels shape: torch.Size([1, 17220]) Final batch size: 1, sequence length: 18741 Attention mask shape: torch.Size([1, 1, 18741, 18741]) Position ids shape: torch.Size([1, 18741]) Input IDs shape: torch.Size([1, 18741]) Labels shape: torch.Size([1, 18741]) Final batch size: 1, sequence length: 17733 Attention mask shape: torch.Size([1, 1, 17733, 17733]) Position ids shape: torch.Size([1, 17733]) Input IDs shape: torch.Size([1, 17733]) Labels shape: torch.Size([1, 17733]) Final batch size: 1, sequence length: 16750 Attention mask shape: torch.Size([1, 1, 16750, 16750]) Position ids shape: torch.Size([1, 16750]) Input IDs shape: torch.Size([1, 16750]) Labels shape: torch.Size([1, 16750]) Final batch size: 1, sequence length: 7221 Attention mask shape: torch.Size([1, 1, 7221, 7221]) Position ids shape: torch.Size([1, 7221]) Input IDs shape: torch.Size([1, 7221]) Labels shape: torch.Size([1, 7221]) Final batch size: 1, sequence length: 13638 Attention mask shape: torch.Size([1, 1, 13638, 13638]) Position ids shape: torch.Size([1, 13638]) Input IDs shape: torch.Size([1, 13638]) Labels shape: torch.Size([1, 13638]) Final batch size: 1, sequence length: 21420 Attention mask shape: torch.Size([1, 1, 21420, 21420]) Position ids shape: torch.Size([1, 21420]) Input IDs shape: torch.Size([1, 21420]) Labels shape: torch.Size([1, 21420]) Final batch size: 1, sequence length: 16827 Attention mask shape: torch.Size([1, 1, 16827, 16827]) Position ids shape: torch.Size([1, 16827]) Input IDs shape: torch.Size([1, 16827]) Labels shape: torch.Size([1, 16827]) Final batch size: 1, sequence length: 18496 Attention mask shape: torch.Size([1, 1, 18496, 18496]) Position ids shape: torch.Size([1, 18496]) Input IDs shape: torch.Size([1, 18496]) Labels shape: torch.Size([1, 18496]) Final batch size: 1, sequence length: 16536 Attention mask shape: torch.Size([1, 1, 16536, 16536]) Position ids shape: torch.Size([1, 16536]) Input IDs shape: torch.Size([1, 16536]) Labels shape: torch.Size([1, 16536]) Final batch size: 1, sequence length: 18927 Attention mask shape: torch.Size([1, 1, 18927, 18927]) Position ids shape: torch.Size([1, 18927]) Input IDs shape: torch.Size([1, 18927]) Labels shape: torch.Size([1, 18927]) Final batch size: 1, sequence length: 19414 Attention mask shape: torch.Size([1, 1, 19414, 19414]) Position ids shape: torch.Size([1, 19414]) Input IDs shape: torch.Size([1, 19414]) Labels shape: torch.Size([1, 19414]) Final batch size: 1, sequence length: 20612 Attention mask shape: torch.Size([1, 1, 20612, 20612]) Position ids shape: torch.Size([1, 20612]) Input IDs shape: torch.Size([1, 20612]) Labels shape: torch.Size([1, 20612]) Final batch size: 1, sequence length: 18393 Attention mask shape: torch.Size([1, 1, 18393, 18393]) Position ids shape: torch.Size([1, 18393]) Input IDs shape: torch.Size([1, 18393]) Labels shape: torch.Size([1, 18393]) Final batch size: 1, sequence length: 20933 Attention mask shape: torch.Size([1, 1, 20933, 20933]) Position ids shape: torch.Size([1, 20933]) Input IDs shape: torch.Size([1, 20933]) Labels shape: torch.Size([1, 20933]) Final batch size: 1, sequence length: 20579 Attention mask shape: torch.Size([1, 1, 20579, 20579]) Position ids shape: torch.Size([1, 20579]) Input IDs shape: torch.Size([1, 20579]) Labels shape: torch.Size([1, 20579]) Final batch size: 1, sequence length: 22887 Attention mask shape: torch.Size([1, 1, 22887, 22887]) Position ids shape: torch.Size([1, 22887]) Input IDs shape: torch.Size([1, 22887]) Labels shape: torch.Size([1, 22887]) Final batch size: 1, sequence length: 22391 Attention mask shape: torch.Size([1, 1, 22391, 22391]) Position ids shape: torch.Size([1, 22391]) Input IDs shape: torch.Size([1, 22391]) Labels shape: torch.Size([1, 22391]) Final batch size: 1, sequence length: 11067 Attention mask shape: torch.Size([1, 1, 11067, 11067]) Position ids shape: torch.Size([1, 11067]) Input IDs shape: torch.Size([1, 11067]) Labels shape: torch.Size([1, 11067]) Final batch size: 1, sequence length: 24988 Attention mask shape: torch.Size([1, 1, 24988, 24988]) Position ids shape: torch.Size([1, 24988]) Input IDs shape: torch.Size([1, 24988]) Labels shape: torch.Size([1, 24988]) Final batch size: 1, sequence length: 25747 Attention mask shape: torch.Size([1, 1, 25747, 25747]) Position ids shape: torch.Size([1, 25747]) Input IDs shape: torch.Size([1, 25747]) Labels shape: torch.Size([1, 25747]) Final batch size: 1, sequence length: 22004 Attention mask shape: torch.Size([1, 1, 22004, 22004]) Position ids shape: torch.Size([1, 22004]) Input IDs shape: torch.Size([1, 22004]) Labels shape: torch.Size([1, 22004]) Final batch size: 1, sequence length: 25651 Attention mask shape: torch.Size([1, 1, 25651, 25651]) Position ids shape: torch.Size([1, 25651]) Input IDs shape: torch.Size([1, 25651]) Labels shape: torch.Size([1, 25651]) Final batch size: 1, sequence length: 15317 Attention mask shape: torch.Size([1, 1, 15317, 15317]) Position ids shape: torch.Size([1, 15317]) Input IDs shape: torch.Size([1, 15317]) Labels shape: torch.Size([1, 15317]) Final batch size: 1, sequence length: 26663 Attention mask shape: torch.Size([1, 1, 26663, 26663]) Position ids shape: torch.Size([1, 26663]) Input IDs shape: torch.Size([1, 26663]) Labels shape: torch.Size([1, 26663]) Final batch size: 1, sequence length: 11184 Attention mask shape: torch.Size([1, 1, 11184, 11184]) Position ids shape: torch.Size([1, 11184]) Input IDs shape: torch.Size([1, 11184]) Labels shape: torch.Size([1, 11184]) Final batch size: 1, sequence length: 16915 Attention mask shape: torch.Size([1, 1, 16915, 16915]) Position ids shape: torch.Size([1, 16915]) Input IDs shape: torch.Size([1, 16915]) Labels shape: torch.Size([1, 16915]) Final batch size: 1, sequence length: 25477 Attention mask shape: torch.Size([1, 1, 25477, 25477]) Position ids shape: torch.Size([1, 25477]) Input IDs shape: torch.Size([1, 25477]) Labels shape: torch.Size([1, 25477]) Final batch size: 1, sequence length: 19552 Attention mask shape: torch.Size([1, 1, 19552, 19552]) Position ids shape: torch.Size([1, 19552]) Input IDs shape: torch.Size([1, 19552]) Labels shape: torch.Size([1, 19552]) Final batch size: 1, sequence length: 10719 Attention mask shape: torch.Size([1, 1, 10719, 10719]) Position ids shape: torch.Size([1, 10719]) Input IDs shape: torch.Size([1, 10719]) Labels shape: torch.Size([1, 10719]) Final batch size: 1, sequence length: 26479 Attention mask shape: torch.Size([1, 1, 26479, 26479]) Position ids shape: torch.Size([1, 26479]) Input IDs shape: torch.Size([1, 26479]) Labels shape: torch.Size([1, 26479]) Final batch size: 1, sequence length: 28777 Attention mask shape: torch.Size([1, 1, 28777, 28777]) Position ids shape: torch.Size([1, 28777]) Input IDs shape: torch.Size([1, 28777]) Labels shape: torch.Size([1, 28777]) Final batch size: 1, sequence length: 30031 Attention mask shape: torch.Size([1, 1, 30031, 30031]) Position ids shape: torch.Size([1, 30031]) Input IDs shape: torch.Size([1, 30031]) Labels shape: torch.Size([1, 30031]) Final batch size: 1, sequence length: 27447 Attention mask shape: torch.Size([1, 1, 27447, 27447]) Position ids shape: torch.Size([1, 27447]) Input IDs shape: torch.Size([1, 27447]) Labels shape: torch.Size([1, 27447]) Final batch size: 1, sequence length: 17395 Attention mask shape: torch.Size([1, 1, 17395, 17395]) Position ids shape: torch.Size([1, 17395]) Input IDs shape: torch.Size([1, 17395]) Labels shape: torch.Size([1, 17395]) Final batch size: 1, sequence length: 30981 Attention mask shape: torch.Size([1, 1, 30981, 30981]) Position ids shape: torch.Size([1, 30981]) Input IDs shape: torch.Size([1, 30981]) Labels shape: torch.Size([1, 30981]) Final batch size: 1, sequence length: 27334 Attention mask shape: torch.Size([1, 1, 27334, 27334]) Position ids shape: torch.Size([1, 27334]) Input IDs shape: torch.Size([1, 27334]) Labels shape: torch.Size([1, 27334]) Final batch size: 1, sequence length: 19869 Attention mask shape: torch.Size([1, 1, 19869, 19869]) Position ids shape: torch.Size([1, 19869]) Input IDs shape: torch.Size([1, 19869]) Labels shape: torch.Size([1, 19869]) Final batch size: 1, sequence length: 27480 Attention mask shape: torch.Size([1, 1, 27480, 27480]) Position ids shape: torch.Size([1, 27480]) Input IDs shape: torch.Size([1, 27480]) Labels shape: torch.Size([1, 27480]) Final batch size: 1, sequence length: 16953 Attention mask shape: torch.Size([1, 1, 16953, 16953]) Position ids shape: torch.Size([1, 16953]) Input IDs shape: torch.Size([1, 16953]) Labels shape: torch.Size([1, 16953]) Final batch size: 1, sequence length: 17911 Attention mask shape: torch.Size([1, 1, 17911, 17911]) Position ids shape: torch.Size([1, 17911]) Input IDs shape: torch.Size([1, 17911]) Labels shape: torch.Size([1, 17911]) Final batch size: 1, sequence length: 18376 Attention mask shape: torch.Size([1, 1, 18376, 18376]) Position ids shape: torch.Size([1, 18376]) Input IDs shape: torch.Size([1, 18376]) Labels shape: torch.Size([1, 18376]) Final batch size: 1, sequence length: 26033 Attention mask shape: torch.Size([1, 1, 26033, 26033]) Position ids shape: torch.Size([1, 26033]) Input IDs shape: torch.Size([1, 26033]) Labels shape: torch.Size([1, 26033]) Final batch size: 1, sequence length: 32466 Attention mask shape: torch.Size([1, 1, 32466, 32466]) Position ids shape: torch.Size([1, 32466]) Input IDs shape: torch.Size([1, 32466]) Labels shape: torch.Size([1, 32466]) Final batch size: 1, sequence length: 21988 Attention mask shape: torch.Size([1, 1, 21988, 21988]) Position ids shape: torch.Size([1, 21988]) Input IDs shape: torch.Size([1, 21988]) Labels shape: torch.Size([1, 21988]) Final batch size: 1, sequence length: 24782 Attention mask shape: torch.Size([1, 1, 24782, 24782]) Position ids shape: torch.Size([1, 24782]) Input IDs shape: torch.Size([1, 24782]) Labels shape: torch.Size([1, 24782]) Final batch size: 1, sequence length: 14873 Attention mask shape: torch.Size([1, 1, 14873, 14873]) Position ids shape: torch.Size([1, 14873]) Input IDs shape: torch.Size([1, 14873]) Labels shape: torch.Size([1, 14873]) Final batch size: 1, sequence length: 30601 Attention mask shape: torch.Size([1, 1, 30601, 30601]) Position ids shape: torch.Size([1, 30601]) Input IDs shape: torch.Size([1, 30601]) Labels shape: torch.Size([1, 30601]) Final batch size: 1, sequence length: 28060 Attention mask shape: torch.Size([1, 1, 28060, 28060]) Position ids shape: torch.Size([1, 28060]) Input IDs shape: torch.Size([1, 28060]) Labels shape: torch.Size([1, 28060]) Final batch size: 1, sequence length: 12245 Attention mask shape: torch.Size([1, 1, 12245, 12245]) Position ids shape: torch.Size([1, 12245]) Input IDs shape: torch.Size([1, 12245]) Labels shape: torch.Size([1, 12245]) Final batch size: 1, sequence length: 33725 Attention mask shape: torch.Size([1, 1, 33725, 33725]) Position ids shape: torch.Size([1, 33725]) Input IDs shape: torch.Size([1, 33725]) Labels shape: torch.Size([1, 33725]) Final batch size: 1, sequence length: 15924 Attention mask shape: torch.Size([1, 1, 15924, 15924]) Position ids shape: torch.Size([1, 15924]) Input IDs shape: torch.Size([1, 15924]) Labels shape: torch.Size([1, 15924]) Final batch size: 1, sequence length: 37738 Attention mask shape: torch.Size([1, 1, 37738, 37738]) Position ids shape: torch.Size([1, 37738]) Input IDs shape: torch.Size([1, 37738]) Labels shape: torch.Size([1, 37738]) Final batch size: 1, sequence length: 37511 Attention mask shape: torch.Size([1, 1, 37511, 37511]) Position ids shape: torch.Size([1, 37511]) Input IDs shape: torch.Size([1, 37511]) Labels shape: torch.Size([1, 37511]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32835 Attention mask shape: torch.Size([1, 1, 32835, 32835]) Position ids shape: torch.Size([1, 32835]) Input IDs shape: torch.Size([1, 32835]) Labels shape: torch.Size([1, 32835]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 27385 Attention mask shape: torch.Size([1, 1, 27385, 27385]) Position ids shape: torch.Size([1, 27385]) Input IDs shape: torch.Size([1, 27385]) Labels shape: torch.Size([1, 27385]) Final batch size: 1, sequence length: 40397 Attention mask shape: torch.Size([1, 1, 40397, 40397]) Position ids shape: torch.Size([1, 40397]) Input IDs shape: torch.Size([1, 40397]) Labels shape: torch.Size([1, 40397]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 37158 Attention mask shape: torch.Size([1, 1, 37158, 37158]) Position ids shape: torch.Size([1, 37158]) Input IDs shape: torch.Size([1, 37158]) Labels shape: torch.Size([1, 37158]) Final batch size: 1, sequence length: 36175 Attention mask shape: torch.Size([1, 1, 36175, 36175]) Position ids shape: torch.Size([1, 36175]) Input IDs shape: torch.Size([1, 36175]) Labels shape: torch.Size([1, 36175]) Final batch size: 1, sequence length: 31879 Attention mask shape: torch.Size([1, 1, 31879, 31879]) Position ids shape: torch.Size([1, 31879]) Input IDs shape: torch.Size([1, 31879]) Labels shape: torch.Size([1, 31879]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 37025 Attention mask shape: torch.Size([1, 1, 37025, 37025]) Position ids shape: torch.Size([1, 37025]) Input IDs shape: torch.Size([1, 37025]) Labels shape: torch.Size([1, 37025]) Final batch size: 1, sequence length: 29586 Attention mask shape: torch.Size([1, 1, 29586, 29586]) Position ids shape: torch.Size([1, 29586]) Input IDs shape: torch.Size([1, 29586]) Labels shape: torch.Size([1, 29586]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40269 Attention mask shape: torch.Size([1, 1, 40269, 40269]) Position ids shape: torch.Size([1, 40269]) Input IDs shape: torch.Size([1, 40269]) Labels shape: torch.Size([1, 40269]) Final batch size: 1, sequence length: 36130 Attention mask shape: torch.Size([1, 1, 36130, 36130]) Position ids shape: torch.Size([1, 36130]) Input IDs shape: torch.Size([1, 36130]) Labels shape: torch.Size([1, 36130]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38986 Attention mask shape: torch.Size([1, 1, 38986, 38986]) Position ids shape: torch.Size([1, 38986]) Input IDs shape: torch.Size([1, 38986]) Labels shape: torch.Size([1, 38986]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31348 Attention mask shape: torch.Size([1, 1, 31348, 31348]) Position ids shape: torch.Size([1, 31348]) Input IDs shape: torch.Size([1, 31348]) Labels shape: torch.Size([1, 31348]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32318 Attention mask shape: torch.Size([1, 1, 32318, 32318]) Position ids shape: torch.Size([1, 32318]) Input IDs shape: torch.Size([1, 32318]) Labels shape: torch.Size([1, 32318]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20509 Attention mask shape: torch.Size([1, 1, 20509, 20509]) Position ids shape: torch.Size([1, 20509]) Input IDs shape: torch.Size([1, 20509]) Labels shape: torch.Size([1, 20509]) Final batch size: 1, sequence length: 40237 Attention mask shape: torch.Size([1, 1, 40237, 40237]) Position ids shape: torch.Size([1, 40237]) Input IDs shape: torch.Size([1, 40237]) Labels shape: torch.Size([1, 40237]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 10198 Attention mask shape: torch.Size([1, 1, 10198, 10198]) Position ids shape: torch.Size([1, 10198]) Input IDs shape: torch.Size([1, 10198]) Labels shape: torch.Size([1, 10198]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24939 Attention mask shape: torch.Size([1, 1, 24939, 24939]) Position ids shape: torch.Size([1, 24939]) Input IDs shape: torch.Size([1, 24939]) Labels shape: torch.Size([1, 24939]) Final batch size: 1, sequence length: 32247 Attention mask shape: torch.Size([1, 1, 32247, 32247]) Position ids shape: torch.Size([1, 32247]) Input IDs shape: torch.Size([1, 32247]) Labels shape: torch.Size([1, 32247]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17711 Attention mask shape: torch.Size([1, 1, 17711, 17711]) Position ids shape: torch.Size([1, 17711]) Input IDs shape: torch.Size([1, 17711]) Labels shape: torch.Size([1, 17711]) Final batch size: 1, sequence length: 36947 Attention mask shape: torch.Size([1, 1, 36947, 36947]) Position ids shape: torch.Size([1, 36947]) Input IDs shape: torch.Size([1, 36947]) Labels shape: torch.Size([1, 36947]) Final batch size: 1, sequence length: 28046 Attention mask shape: torch.Size([1, 1, 28046, 28046]) Position ids shape: torch.Size([1, 28046]) Input IDs shape: torch.Size([1, 28046]) Labels shape: torch.Size([1, 28046]) Final batch size: 1, sequence length: 23740 Attention mask shape: torch.Size([1, 1, 23740, 23740]) Position ids shape: torch.Size([1, 23740]) Input IDs shape: torch.Size([1, 23740]) Labels shape: torch.Size([1, 23740]) Final batch size: 1, sequence length: 21225 Attention mask shape: torch.Size([1, 1, 21225, 21225]) Position ids shape: torch.Size([1, 21225]) Input IDs shape: torch.Size([1, 21225]) Labels shape: torch.Size([1, 21225]) Final batch size: 1, sequence length: 22896 Attention mask shape: torch.Size([1, 1, 22896, 22896]) Position ids shape: torch.Size([1, 22896]) Input IDs shape: torch.Size([1, 22896]) Labels shape: torch.Size([1, 22896]) Final batch size: 1, sequence length: 28631 Attention mask shape: torch.Size([1, 1, 28631, 28631]) Position ids shape: torch.Size([1, 28631]) Input IDs shape: torch.Size([1, 28631]) Labels shape: torch.Size([1, 28631]) Final batch size: 1, sequence length: 27291 Attention mask shape: torch.Size([1, 1, 27291, 27291]) Position ids shape: torch.Size([1, 27291]) Input IDs shape: torch.Size([1, 27291]) Labels shape: torch.Size([1, 27291]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33125 Attention mask shape: torch.Size([1, 1, 33125, 33125]) Position ids shape: torch.Size([1, 33125]) Input IDs shape: torch.Size([1, 33125]) Labels shape: torch.Size([1, 33125]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22205 Attention mask shape: torch.Size([1, 1, 22205, 22205]) Position ids shape: torch.Size([1, 22205]) Input IDs shape: torch.Size([1, 22205]) Labels shape: torch.Size([1, 22205]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28149 Attention mask shape: torch.Size([1, 1, 28149, 28149]) Position ids shape: torch.Size([1, 28149]) Input IDs shape: torch.Size([1, 28149]) Labels shape: torch.Size([1, 28149]) Final batch size: 1, sequence length: 38360 Attention mask shape: torch.Size([1, 1, 38360, 38360]) Position ids shape: torch.Size([1, 38360]) Input IDs shape: torch.Size([1, 38360]) Labels shape: torch.Size([1, 38360]) Final batch size: 1, sequence length: 32767 Attention mask shape: torch.Size([1, 1, 32767, 32767]) Position ids shape: torch.Size([1, 32767]) Input IDs shape: torch.Size([1, 32767]) Labels shape: torch.Size([1, 32767]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36567 Attention mask shape: torch.Size([1, 1, 36567, 36567]) Position ids shape: torch.Size([1, 36567]) Input IDs shape: torch.Size([1, 36567]) Labels shape: torch.Size([1, 36567]) Final batch size: 1, sequence length: 34268 Attention mask shape: torch.Size([1, 1, 34268, 34268]) Position ids shape: torch.Size([1, 34268]) Input IDs shape: torch.Size([1, 34268]) Labels shape: torch.Size([1, 34268]) Final batch size: 1, sequence length: 32313 Attention mask shape: torch.Size([1, 1, 32313, 32313]) Position ids shape: torch.Size([1, 32313]) Input IDs shape: torch.Size([1, 32313]) Labels shape: torch.Size([1, 32313]) {'loss': 0.2738, 'grad_norm': 0.18869471130399465, 'learning_rate': 1.4644660940672628e-06, 'num_tokens': -inf, 'epoch': 6.25} Final batch size: 1, sequence length: 5377 Attention mask shape: torch.Size([1, 1, 5377, 5377]) Position ids shape: torch.Size([1, 5377]) Input IDs shape: torch.Size([1, 5377]) Labels shape: torch.Size([1, 5377]) Final batch size: 1, sequence length: 7998 Attention mask shape: torch.Size([1, 1, 7998, 7998]) Position ids shape: torch.Size([1, 7998]) Input IDs shape: torch.Size([1, 7998]) Labels shape: torch.Size([1, 7998]) Final batch size: 1, sequence length: 7402 Attention mask shape: torch.Size([1, 1, 7402, 7402]) Position ids shape: torch.Size([1, 7402]) Input IDs shape: torch.Size([1, 7402]) Labels shape: torch.Size([1, 7402]) Final batch size: 1, sequence length: 8436 Attention mask shape: torch.Size([1, 1, 8436, 8436]) Position ids shape: torch.Size([1, 8436]) Input IDs shape: torch.Size([1, 8436]) Labels shape: torch.Size([1, 8436]) Final batch size: 1, sequence length: 8655 Attention mask shape: torch.Size([1, 1, 8655, 8655]) Position ids shape: torch.Size([1, 8655]) Input IDs shape: torch.Size([1, 8655]) Labels shape: torch.Size([1, 8655]) Final batch size: 1, sequence length: 10576 Attention mask shape: torch.Size([1, 1, 10576, 10576]) Position ids shape: torch.Size([1, 10576]) Input IDs shape: torch.Size([1, 10576]) Labels shape: torch.Size([1, 10576]) Final batch size: 1, sequence length: 11709 Attention mask shape: torch.Size([1, 1, 11709, 11709]) Position ids shape: torch.Size([1, 11709]) Input IDs shape: torch.Size([1, 11709]) Labels shape: torch.Size([1, 11709]) Final batch size: 1, sequence length: 7344 Attention mask shape: torch.Size([1, 1, 7344, 7344]) Position ids shape: torch.Size([1, 7344]) Input IDs shape: torch.Size([1, 7344]) Labels shape: torch.Size([1, 7344]) Final batch size: 1, sequence length: 11678 Attention mask shape: torch.Size([1, 1, 11678, 11678]) Position ids shape: torch.Size([1, 11678]) Input IDs shape: torch.Size([1, 11678]) Labels shape: torch.Size([1, 11678]) Final batch size: 1, sequence length: 11365 Attention mask shape: torch.Size([1, 1, 11365, 11365]) Position ids shape: torch.Size([1, 11365]) Input IDs shape: torch.Size([1, 11365]) Labels shape: torch.Size([1, 11365]) Final batch size: 1, sequence length: 14127 Attention mask shape: torch.Size([1, 1, 14127, 14127]) Position ids shape: torch.Size([1, 14127]) Input IDs shape: torch.Size([1, 14127]) Labels shape: torch.Size([1, 14127]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 15079 Attention mask shape: torch.Size([1, 1, 15079, 15079]) Position ids shape: torch.Size([1, 15079]) Input IDs shape: torch.Size([1, 15079]) Labels shape: torch.Size([1, 15079]) Final batch size: 1, sequence length: 15308 Attention mask shape: torch.Size([1, 1, 15308, 15308]) Position ids shape: torch.Size([1, 15308]) Input IDs shape: torch.Size([1, 15308]) Labels shape: torch.Size([1, 15308]) Final batch size: 1, sequence length: 13639 Attention mask shape: torch.Size([1, 1, 13639, 13639]) Position ids shape: torch.Size([1, 13639]) Input IDs shape: torch.Size([1, 13639]) Labels shape: torch.Size([1, 13639]) Final batch size: 1, sequence length: 16535 Attention mask shape: torch.Size([1, 1, 16535, 16535]) Position ids shape: torch.Size([1, 16535]) Input IDs shape: torch.Size([1, 16535]) Labels shape: torch.Size([1, 16535]) Final batch size: 1, sequence length: 18307 Attention mask shape: torch.Size([1, 1, 18307, 18307]) Position ids shape: torch.Size([1, 18307]) Input IDs shape: torch.Size([1, 18307]) Labels shape: torch.Size([1, 18307]) Final batch size: 1, sequence length: 11623 Attention mask shape: torch.Size([1, 1, 11623, 11623]) Position ids shape: torch.Size([1, 11623]) Input IDs shape: torch.Size([1, 11623]) Labels shape: torch.Size([1, 11623]) Final batch size: 1, sequence length: 19170 Attention mask shape: torch.Size([1, 1, 19170, 19170]) Position ids shape: torch.Size([1, 19170]) Input IDs shape: torch.Size([1, 19170]) Labels shape: torch.Size([1, 19170]) Final batch size: 1, sequence length: 16060 Attention mask shape: torch.Size([1, 1, 16060, 16060]) Position ids shape: torch.Size([1, 16060]) Input IDs shape: torch.Size([1, 16060]) Labels shape: torch.Size([1, 16060]) Final batch size: 1, sequence length: 15478 Attention mask shape: torch.Size([1, 1, 15478, 15478]) Position ids shape: torch.Size([1, 15478]) Input IDs shape: torch.Size([1, 15478]) Labels shape: torch.Size([1, 15478]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 18953 Attention mask shape: torch.Size([1, 1, 18953, 18953]) Position ids shape: torch.Size([1, 18953]) Input IDs shape: torch.Size([1, 18953]) Labels shape: torch.Size([1, 18953]) Final batch size: 1, sequence length: 19138 Attention mask shape: torch.Size([1, 1, 19138, 19138]) Position ids shape: torch.Size([1, 19138]) Input IDs shape: torch.Size([1, 19138]) Labels shape: torch.Size([1, 19138]) Final batch size: 1, sequence length: 19338 Attention mask shape: torch.Size([1, 1, 19338, 19338]) Position ids shape: torch.Size([1, 19338]) Input IDs shape: torch.Size([1, 19338]) Labels shape: torch.Size([1, 19338]) Final batch size: 1, sequence length: 14648 Attention mask shape: torch.Size([1, 1, 14648, 14648]) Position ids shape: torch.Size([1, 14648]) Input IDs shape: torch.Size([1, 14648]) Labels shape: torch.Size([1, 14648]) Final batch size: 1, sequence length: 12006 Attention mask shape: torch.Size([1, 1, 12006, 12006]) Position ids shape: torch.Size([1, 12006]) Input IDs shape: torch.Size([1, 12006]) Labels shape: torch.Size([1, 12006]) Final batch size: 1, sequence length: 18836 Attention mask shape: torch.Size([1, 1, 18836, 18836]) Position ids shape: torch.Size([1, 18836]) Input IDs shape: torch.Size([1, 18836]) Labels shape: torch.Size([1, 18836]) Final batch size: 1, sequence length: 18527 Attention mask shape: torch.Size([1, 1, 18527, 18527]) Position ids shape: torch.Size([1, 18527]) Input IDs shape: torch.Size([1, 18527]) Labels shape: torch.Size([1, 18527]) Final batch size: 1, sequence length: 22561 Attention mask shape: torch.Size([1, 1, 22561, 22561]) Position ids shape: torch.Size([1, 22561]) Input IDs shape: torch.Size([1, 22561]) Labels shape: torch.Size([1, 22561]) Final batch size: 1, sequence length: 8839 Attention mask shape: torch.Size([1, 1, 8839, 8839]) Position ids shape: torch.Size([1, 8839]) Input IDs shape: torch.Size([1, 8839]) Labels shape: torch.Size([1, 8839]) Final batch size: 1, sequence length: 17058 Attention mask shape: torch.Size([1, 1, 17058, 17058]) Position ids shape: torch.Size([1, 17058]) Input IDs shape: torch.Size([1, 17058]) Labels shape: torch.Size([1, 17058]) Final batch size: 1, sequence length: 18325 Attention mask shape: torch.Size([1, 1, 18325, 18325]) Position ids shape: torch.Size([1, 18325]) Input IDs shape: torch.Size([1, 18325]) Labels shape: torch.Size([1, 18325]) Final batch size: 1, sequence length: 19330 Attention mask shape: torch.Size([1, 1, 19330, 19330]) Position ids shape: torch.Size([1, 19330]) Input IDs shape: torch.Size([1, 19330]) Labels shape: torch.Size([1, 19330]) Final batch size: 1, sequence length: 20770 Attention mask shape: torch.Size([1, 1, 20770, 20770]) Position ids shape: torch.Size([1, 20770]) Input IDs shape: torch.Size([1, 20770]) Labels shape: torch.Size([1, 20770]) Final batch size: 1, sequence length: 21982 Attention mask shape: torch.Size([1, 1, 21982, 21982]) Position ids shape: torch.Size([1, 21982]) Input IDs shape: torch.Size([1, 21982]) Labels shape: torch.Size([1, 21982]) Final batch size: 1, sequence length: 18395 Attention mask shape: torch.Size([1, 1, 18395, 18395]) Position ids shape: torch.Size([1, 18395]) Input IDs shape: torch.Size([1, 18395]) Labels shape: torch.Size([1, 18395]) Final batch size: 1, sequence length: 20854 Attention mask shape: torch.Size([1, 1, 20854, 20854]) Position ids shape: torch.Size([1, 20854]) Input IDs shape: torch.Size([1, 20854]) Labels shape: torch.Size([1, 20854]) Final batch size: 1, sequence length: 18823 Attention mask shape: torch.Size([1, 1, 18823, 18823]) Position ids shape: torch.Size([1, 18823]) Input IDs shape: torch.Size([1, 18823]) Labels shape: torch.Size([1, 18823]) Final batch size: 1, sequence length: 20714 Attention mask shape: torch.Size([1, 1, 20714, 20714]) Position ids shape: torch.Size([1, 20714]) Input IDs shape: torch.Size([1, 20714]) Labels shape: torch.Size([1, 20714]) Final batch size: 1, sequence length: 23560 Attention mask shape: torch.Size([1, 1, 23560, 23560]) Position ids shape: torch.Size([1, 23560]) Input IDs shape: torch.Size([1, 23560]) Labels shape: torch.Size([1, 23560]) Final batch size: 1, sequence length: 24248 Attention mask shape: torch.Size([1, 1, 24248, 24248]) Position ids shape: torch.Size([1, 24248]) Input IDs shape: torch.Size([1, 24248]) Labels shape: torch.Size([1, 24248]) Final batch size: 1, sequence length: 25451 Attention mask shape: torch.Size([1, 1, 25451, 25451]) Position ids shape: torch.Size([1, 25451]) Input IDs shape: torch.Size([1, 25451]) Labels shape: torch.Size([1, 25451]) Final batch size: 1, sequence length: 15913 Attention mask shape: torch.Size([1, 1, 15913, 15913]) Position ids shape: torch.Size([1, 15913]) Input IDs shape: torch.Size([1, 15913]) Labels shape: torch.Size([1, 15913]) Final batch size: 1, sequence length: 23694 Attention mask shape: torch.Size([1, 1, 23694, 23694]) Position ids shape: torch.Size([1, 23694]) Input IDs shape: torch.Size([1, 23694]) Labels shape: torch.Size([1, 23694]) Final batch size: 1, sequence length: 24433 Attention mask shape: torch.Size([1, 1, 24433, 24433]) Position ids shape: torch.Size([1, 24433]) Input IDs shape: torch.Size([1, 24433]) Labels shape: torch.Size([1, 24433]) Final batch size: 1, sequence length: 29098 Attention mask shape: torch.Size([1, 1, 29098, 29098]) Position ids shape: torch.Size([1, 29098]) Input IDs shape: torch.Size([1, 29098]) Labels shape: torch.Size([1, 29098]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 21405 Attention mask shape: torch.Size([1, 1, 21405, 21405]) Position ids shape: torch.Size([1, 21405]) Input IDs shape: torch.Size([1, 21405]) Labels shape: torch.Size([1, 21405]) Final batch size: 1, sequence length: 28684 Attention mask shape: torch.Size([1, 1, 28684, 28684]) Position ids shape: torch.Size([1, 28684]) Input IDs shape: torch.Size([1, 28684]) Labels shape: torch.Size([1, 28684]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 26356 Attention mask shape: torch.Size([1, 1, 26356, 26356]) Position ids shape: torch.Size([1, 26356]) Input IDs shape: torch.Size([1, 26356]) Labels shape: torch.Size([1, 26356]) Final batch size: 1, sequence length: 31414 Attention mask shape: torch.Size([1, 1, 31414, 31414]) Position ids shape: torch.Size([1, 31414]) Input IDs shape: torch.Size([1, 31414]) Labels shape: torch.Size([1, 31414]) Final batch size: 1, sequence length: 9380 Attention mask shape: torch.Size([1, 1, 9380, 9380]) Position ids shape: torch.Size([1, 9380]) Input IDs shape: torch.Size([1, 9380]) Labels shape: torch.Size([1, 9380]) Final batch size: 1, sequence length: 5801 Attention mask shape: torch.Size([1, 1, 5801, 5801]) Position ids shape: torch.Size([1, 5801]) Input IDs shape: torch.Size([1, 5801]) Labels shape: torch.Size([1, 5801]) Final batch size: 1, sequence length: 21858 Attention mask shape: torch.Size([1, 1, 21858, 21858]) Position ids shape: torch.Size([1, 21858]) Input IDs shape: torch.Size([1, 21858]) Labels shape: torch.Size([1, 21858]) Final batch size: 1, sequence length: 15875 Attention mask shape: torch.Size([1, 1, 15875, 15875]) Position ids shape: torch.Size([1, 15875]) Input IDs shape: torch.Size([1, 15875]) Labels shape: torch.Size([1, 15875]) Final batch size: 1, sequence length: 22735 Attention mask shape: torch.Size([1, 1, 22735, 22735]) Position ids shape: torch.Size([1, 22735]) Input IDs shape: torch.Size([1, 22735]) Labels shape: torch.Size([1, 22735]) Final batch size: 1, sequence length: 17465 Attention mask shape: torch.Size([1, 1, 17465, 17465]) Position ids shape: torch.Size([1, 17465]) Input IDs shape: torch.Size([1, 17465]) Labels shape: torch.Size([1, 17465]) Final batch size: 1, sequence length: 32885 Attention mask shape: torch.Size([1, 1, 32885, 32885]) Position ids shape: torch.Size([1, 32885]) Input IDs shape: torch.Size([1, 32885]) Labels shape: torch.Size([1, 32885]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 19045 Attention mask shape: torch.Size([1, 1, 19045, 19045]) Position ids shape: torch.Size([1, 19045]) Input IDs shape: torch.Size([1, 19045]) Labels shape: torch.Size([1, 19045]) Final batch size: 1, sequence length: 32328 Attention mask shape: torch.Size([1, 1, 32328, 32328]) Position ids shape: torch.Size([1, 32328]) Input IDs shape: torch.Size([1, 32328]) Labels shape: torch.Size([1, 32328]) Final batch size: 1, sequence length: 19239 Attention mask shape: torch.Size([1, 1, 19239, 19239]) Position ids shape: torch.Size([1, 19239]) Input IDs shape: torch.Size([1, 19239]) Labels shape: torch.Size([1, 19239]) Final batch size: 1, sequence length: 23881 Attention mask shape: torch.Size([1, 1, 23881, 23881]) Position ids shape: torch.Size([1, 23881]) Input IDs shape: torch.Size([1, 23881]) Labels shape: torch.Size([1, 23881]) Final batch size: 1, sequence length: 25674 Attention mask shape: torch.Size([1, 1, 25674, 25674]) Position ids shape: torch.Size([1, 25674]) Input IDs shape: torch.Size([1, 25674]) Labels shape: torch.Size([1, 25674]) Final batch size: 1, sequence length: 30346 Attention mask shape: torch.Size([1, 1, 30346, 30346]) Position ids shape: torch.Size([1, 30346]) Input IDs shape: torch.Size([1, 30346]) Labels shape: torch.Size([1, 30346]) Final batch size: 1, sequence length: 29355 Attention mask shape: torch.Size([1, 1, 29355, 29355]) Position ids shape: torch.Size([1, 29355]) Input IDs shape: torch.Size([1, 29355]) Labels shape: torch.Size([1, 29355]) Final batch size: 1, sequence length: 21615 Attention mask shape: torch.Size([1, 1, 21615, 21615]) Position ids shape: torch.Size([1, 21615]) Input IDs shape: torch.Size([1, 21615]) Labels shape: torch.Size([1, 21615]) Final batch size: 1, sequence length: 27972 Attention mask shape: torch.Size([1, 1, 27972, 27972]) Position ids shape: torch.Size([1, 27972]) Input IDs shape: torch.Size([1, 27972]) Labels shape: torch.Size([1, 27972]) Final batch size: 1, sequence length: 30687 Attention mask shape: torch.Size([1, 1, 30687, 30687]) Position ids shape: torch.Size([1, 30687]) Input IDs shape: torch.Size([1, 30687]) Labels shape: torch.Size([1, 30687]) Final batch size: 1, sequence length: 32529 Attention mask shape: torch.Size([1, 1, 32529, 32529]) Position ids shape: torch.Size([1, 32529]) Input IDs shape: torch.Size([1, 32529]) Labels shape: torch.Size([1, 32529]) Final batch size: 1, sequence length: 35397 Attention mask shape: torch.Size([1, 1, 35397, 35397]) Position ids shape: torch.Size([1, 35397]) Input IDs shape: torch.Size([1, 35397]) Labels shape: torch.Size([1, 35397]) Final batch size: 1, sequence length: 32636 Attention mask shape: torch.Size([1, 1, 32636, 32636]) Position ids shape: torch.Size([1, 32636]) Input IDs shape: torch.Size([1, 32636]) Labels shape: torch.Size([1, 32636]) Final batch size: 1, sequence length: 26072 Attention mask shape: torch.Size([1, 1, 26072, 26072]) Position ids shape: torch.Size([1, 26072]) Input IDs shape: torch.Size([1, 26072]) Labels shape: torch.Size([1, 26072]) Final batch size: 1, sequence length: 29150 Attention mask shape: torch.Size([1, 1, 29150, 29150]) Position ids shape: torch.Size([1, 29150]) Input IDs shape: torch.Size([1, 29150]) Labels shape: torch.Size([1, 29150]) Final batch size: 1, sequence length: 31860 Attention mask shape: torch.Size([1, 1, 31860, 31860]) Position ids shape: torch.Size([1, 31860]) Input IDs shape: torch.Size([1, 31860]) Labels shape: torch.Size([1, 31860]) Final batch size: 1, sequence length: 16050 Attention mask shape: torch.Size([1, 1, 16050, 16050]) Position ids shape: torch.Size([1, 16050]) Input IDs shape: torch.Size([1, 16050]) Labels shape: torch.Size([1, 16050]) Final batch size: 1, sequence length: 26689 Attention mask shape: torch.Size([1, 1, 26689, 26689]) Position ids shape: torch.Size([1, 26689]) Input IDs shape: torch.Size([1, 26689]) Labels shape: torch.Size([1, 26689]) Final batch size: 1, sequence length: 33871 Attention mask shape: torch.Size([1, 1, 33871, 33871]) Position ids shape: torch.Size([1, 33871]) Input IDs shape: torch.Size([1, 33871]) Labels shape: torch.Size([1, 33871]) Final batch size: 1, sequence length: 34581 Attention mask shape: torch.Size([1, 1, 34581, 34581]) Position ids shape: torch.Size([1, 34581]) Input IDs shape: torch.Size([1, 34581]) Labels shape: torch.Size([1, 34581]) Final batch size: 1, sequence length: 31712 Attention mask shape: torch.Size([1, 1, 31712, 31712]) Position ids shape: torch.Size([1, 31712]) Input IDs shape: torch.Size([1, 31712]) Labels shape: torch.Size([1, 31712]) Final batch size: 1, sequence length: 22264 Attention mask shape: torch.Size([1, 1, 22264, 22264]) Position ids shape: torch.Size([1, 22264]) Input IDs shape: torch.Size([1, 22264]) Labels shape: torch.Size([1, 22264]) Final batch size: 1, sequence length: 30689 Attention mask shape: torch.Size([1, 1, 30689, 30689]) Position ids shape: torch.Size([1, 30689]) Input IDs shape: torch.Size([1, 30689]) Labels shape: torch.Size([1, 30689]) Final batch size: 1, sequence length: 15077 Attention mask shape: torch.Size([1, 1, 15077, 15077]) Position ids shape: torch.Size([1, 15077]) Input IDs shape: torch.Size([1, 15077]) Labels shape: torch.Size([1, 15077]) Final batch size: 1, sequence length: 24781 Attention mask shape: torch.Size([1, 1, 24781, 24781]) Position ids shape: torch.Size([1, 24781]) Input IDs shape: torch.Size([1, 24781]) Labels shape: torch.Size([1, 24781]) Final batch size: 1, sequence length: 34921 Attention mask shape: torch.Size([1, 1, 34921, 34921]) Position ids shape: torch.Size([1, 34921]) Input IDs shape: torch.Size([1, 34921]) Labels shape: torch.Size([1, 34921]) Final batch size: 1, sequence length: 18963 Attention mask shape: torch.Size([1, 1, 18963, 18963]) Position ids shape: torch.Size([1, 18963]) Input IDs shape: torch.Size([1, 18963]) Labels shape: torch.Size([1, 18963]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16025 Attention mask shape: torch.Size([1, 1, 16025, 16025]) Position ids shape: torch.Size([1, 16025]) Input IDs shape: torch.Size([1, 16025]) Labels shape: torch.Size([1, 16025]) Final batch size: 1, sequence length: 25741 Attention mask shape: torch.Size([1, 1, 25741, 25741]) Position ids shape: torch.Size([1, 25741]) Input IDs shape: torch.Size([1, 25741]) Labels shape: torch.Size([1, 25741]) Final batch size: 1, sequence length: 28018 Attention mask shape: torch.Size([1, 1, 28018, 28018]) Position ids shape: torch.Size([1, 28018]) Input IDs shape: torch.Size([1, 28018]) Labels shape: torch.Size([1, 28018]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 18177 Attention mask shape: torch.Size([1, 1, 18177, 18177]) Position ids shape: torch.Size([1, 18177]) Input IDs shape: torch.Size([1, 18177]) Labels shape: torch.Size([1, 18177]) Final batch size: 1, sequence length: 31022 Attention mask shape: torch.Size([1, 1, 31022, 31022]) Position ids shape: torch.Size([1, 31022]) Input IDs shape: torch.Size([1, 31022]) Labels shape: torch.Size([1, 31022]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17243 Attention mask shape: torch.Size([1, 1, 17243, 17243]) Position ids shape: torch.Size([1, 17243]) Input IDs shape: torch.Size([1, 17243]) Labels shape: torch.Size([1, 17243]) Final batch size: 1, sequence length: 14976 Attention mask shape: torch.Size([1, 1, 14976, 14976]) Position ids shape: torch.Size([1, 14976]) Input IDs shape: torch.Size([1, 14976]) Labels shape: torch.Size([1, 14976]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24002 Attention mask shape: torch.Size([1, 1, 24002, 24002]) Position ids shape: torch.Size([1, 24002]) Input IDs shape: torch.Size([1, 24002]) Labels shape: torch.Size([1, 24002]) Final batch size: 1, sequence length: 26922 Attention mask shape: torch.Size([1, 1, 26922, 26922]) Position ids shape: torch.Size([1, 26922]) Input IDs shape: torch.Size([1, 26922]) Labels shape: torch.Size([1, 26922]) Final batch size: 1, sequence length: 40488 Attention mask shape: torch.Size([1, 1, 40488, 40488]) Position ids shape: torch.Size([1, 40488]) Input IDs shape: torch.Size([1, 40488]) Labels shape: torch.Size([1, 40488]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 11611 Attention mask shape: torch.Size([1, 1, 11611, 11611]) Position ids shape: torch.Size([1, 11611]) Input IDs shape: torch.Size([1, 11611]) Labels shape: torch.Size([1, 11611]) Final batch size: 1, sequence length: 28397 Attention mask shape: torch.Size([1, 1, 28397, 28397]) Position ids shape: torch.Size([1, 28397]) Input IDs shape: torch.Size([1, 28397]) Labels shape: torch.Size([1, 28397]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36596 Attention mask shape: torch.Size([1, 1, 36596, 36596]) Position ids shape: torch.Size([1, 36596]) Input IDs shape: torch.Size([1, 36596]) Labels shape: torch.Size([1, 36596]) Final batch size: 1, sequence length: 24043 Attention mask shape: torch.Size([1, 1, 24043, 24043]) Position ids shape: torch.Size([1, 24043]) Input IDs shape: torch.Size([1, 24043]) Labels shape: torch.Size([1, 24043]) Final batch size: 1, sequence length: 28086 Attention mask shape: torch.Size([1, 1, 28086, 28086]) Position ids shape: torch.Size([1, 28086]) Input IDs shape: torch.Size([1, 28086]) Labels shape: torch.Size([1, 28086]) Final batch size: 1, sequence length: 16337 Attention mask shape: torch.Size([1, 1, 16337, 16337]) Position ids shape: torch.Size([1, 16337]) Input IDs shape: torch.Size([1, 16337]) Labels shape: torch.Size([1, 16337]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34853 Attention mask shape: torch.Size([1, 1, 34853, 34853]) Position ids shape: torch.Size([1, 34853]) Input IDs shape: torch.Size([1, 34853]) Labels shape: torch.Size([1, 34853]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36334 Attention mask shape: torch.Size([1, 1, 36334, 36334]) Position ids shape: torch.Size([1, 36334]) Input IDs shape: torch.Size([1, 36334]) Labels shape: torch.Size([1, 36334]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 12218 Attention mask shape: torch.Size([1, 1, 12218, 12218]) Position ids shape: torch.Size([1, 12218]) Input IDs shape: torch.Size([1, 12218]) Labels shape: torch.Size([1, 12218]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32917 Attention mask shape: torch.Size([1, 1, 32917, 32917]) Position ids shape: torch.Size([1, 32917]) Input IDs shape: torch.Size([1, 32917]) Labels shape: torch.Size([1, 32917]) {'loss': 0.2573, 'grad_norm': 0.1661350362703919, 'learning_rate': 1.2842758726130283e-06, 'num_tokens': -inf, 'epoch': 6.38} Final batch size: 1, sequence length: 5818 Attention mask shape: torch.Size([1, 1, 5818, 5818]) Position ids shape: torch.Size([1, 5818]) Input IDs shape: torch.Size([1, 5818]) Labels shape: torch.Size([1, 5818]) Final batch size: 1, sequence length: 6215 Attention mask shape: torch.Size([1, 1, 6215, 6215]) Position ids shape: torch.Size([1, 6215]) Input IDs shape: torch.Size([1, 6215]) Labels shape: torch.Size([1, 6215]) Final batch size: 1, sequence length: 8623 Attention mask shape: torch.Size([1, 1, 8623, 8623]) Position ids shape: torch.Size([1, 8623]) Input IDs shape: torch.Size([1, 8623]) Labels shape: torch.Size([1, 8623]) Final batch size: 1, sequence length: 6871 Attention mask shape: torch.Size([1, 1, 6871, 6871]) Position ids shape: torch.Size([1, 6871]) Input IDs shape: torch.Size([1, 6871]) Labels shape: torch.Size([1, 6871]) Final batch size: 1, sequence length: 5917 Attention mask shape: torch.Size([1, 1, 5917, 5917]) Position ids shape: torch.Size([1, 5917]) Input IDs shape: torch.Size([1, 5917]) Labels shape: torch.Size([1, 5917]) Final batch size: 1, sequence length: 11616 Attention mask shape: torch.Size([1, 1, 11616, 11616]) Position ids shape: torch.Size([1, 11616]) Input IDs shape: torch.Size([1, 11616]) Labels shape: torch.Size([1, 11616]) Final batch size: 1, sequence length: 12275 Attention mask shape: torch.Size([1, 1, 12275, 12275]) Position ids shape: torch.Size([1, 12275]) Input IDs shape: torch.Size([1, 12275]) Labels shape: torch.Size([1, 12275]) Final batch size: 1, sequence length: 10107 Attention mask shape: torch.Size([1, 1, 10107, 10107]) Position ids shape: torch.Size([1, 10107]) Input IDs shape: torch.Size([1, 10107]) Labels shape: torch.Size([1, 10107]) Final batch size: 1, sequence length: 12317 Attention mask shape: torch.Size([1, 1, 12317, 12317]) Position ids shape: torch.Size([1, 12317]) Input IDs shape: torch.Size([1, 12317]) Labels shape: torch.Size([1, 12317]) Final batch size: 1, sequence length: 13202 Attention mask shape: torch.Size([1, 1, 13202, 13202]) Position ids shape: torch.Size([1, 13202]) Input IDs shape: torch.Size([1, 13202]) Labels shape: torch.Size([1, 13202]) Final batch size: 1, sequence length: 9517 Attention mask shape: torch.Size([1, 1, 9517, 9517]) Position ids shape: torch.Size([1, 9517]) Input IDs shape: torch.Size([1, 9517]) Labels shape: torch.Size([1, 9517]) Final batch size: 1, sequence length: 11648 Attention mask shape: torch.Size([1, 1, 11648, 11648]) Position ids shape: torch.Size([1, 11648]) Input IDs shape: torch.Size([1, 11648]) Labels shape: torch.Size([1, 11648]) Final batch size: 1, sequence length: 6034 Attention mask shape: torch.Size([1, 1, 6034, 6034]) Position ids shape: torch.Size([1, 6034]) Input IDs shape: torch.Size([1, 6034]) Labels shape: torch.Size([1, 6034]) Final batch size: 1, sequence length: 15499 Attention mask shape: torch.Size([1, 1, 15499, 15499]) Position ids shape: torch.Size([1, 15499]) Input IDs shape: torch.Size([1, 15499]) Labels shape: torch.Size([1, 15499]) Final batch size: 1, sequence length: 8117 Final batch size: 1, sequence length: 15029 Attention mask shape: torch.Size([1, 1, 8117, 8117]) Position ids shape: torch.Size([1, 8117]) Input IDs shape: torch.Size([1, 8117]) Labels shape: torch.Size([1, 8117]) Attention mask shape: torch.Size([1, 1, 15029, 15029]) Position ids shape: torch.Size([1, 15029]) Input IDs shape: torch.Size([1, 15029]) Labels shape: torch.Size([1, 15029]) Final batch size: 1, sequence length: 10136 Attention mask shape: torch.Size([1, 1, 10136, 10136]) Position ids shape: torch.Size([1, 10136]) Input IDs shape: torch.Size([1, 10136]) Labels shape: torch.Size([1, 10136]) Final batch size: 1, sequence length: 13428 Attention mask shape: torch.Size([1, 1, 13428, 13428]) Position ids shape: torch.Size([1, 13428]) Input IDs shape: torch.Size([1, 13428]) Labels shape: torch.Size([1, 13428]) Final batch size: 1, sequence length: 12813 Attention mask shape: torch.Size([1, 1, 12813, 12813]) Position ids shape: torch.Size([1, 12813]) Input IDs shape: torch.Size([1, 12813]) Labels shape: torch.Size([1, 12813]) Final batch size: 1, sequence length: 15066 Attention mask shape: torch.Size([1, 1, 15066, 15066]) Position ids shape: torch.Size([1, 15066]) Input IDs shape: torch.Size([1, 15066]) Labels shape: torch.Size([1, 15066]) Final batch size: 1, sequence length: 16299 Attention mask shape: torch.Size([1, 1, 16299, 16299]) Position ids shape: torch.Size([1, 16299]) Input IDs shape: torch.Size([1, 16299]) Labels shape: torch.Size([1, 16299]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 17587 Attention mask shape: torch.Size([1, 1, 17587, 17587]) Position ids shape: torch.Size([1, 17587]) Input IDs shape: torch.Size([1, 17587]) Labels shape: torch.Size([1, 17587]) Final batch size: 1, sequence length: 18414 Attention mask shape: torch.Size([1, 1, 18414, 18414]) Position ids shape: torch.Size([1, 18414]) Input IDs shape: torch.Size([1, 18414]) Labels shape: torch.Size([1, 18414]) Final batch size: 1, sequence length: 12553 Attention mask shape: torch.Size([1, 1, 12553, 12553]) Position ids shape: torch.Size([1, 12553]) Input IDs shape: torch.Size([1, 12553]) Labels shape: torch.Size([1, 12553]) Final batch size: 1, sequence length: 16331 Attention mask shape: torch.Size([1, 1, 16331, 16331]) Position ids shape: torch.Size([1, 16331]) Input IDs shape: torch.Size([1, 16331]) Labels shape: torch.Size([1, 16331]) Final batch size: 1, sequence length: 18438 Attention mask shape: torch.Size([1, 1, 18438, 18438]) Position ids shape: torch.Size([1, 18438]) Input IDs shape: torch.Size([1, 18438]) Labels shape: torch.Size([1, 18438]) Final batch size: 1, sequence length: 19847 Attention mask shape: torch.Size([1, 1, 19847, 19847]) Position ids shape: torch.Size([1, 19847]) Input IDs shape: torch.Size([1, 19847]) Labels shape: torch.Size([1, 19847]) Final batch size: 1, sequence length: 19221 Attention mask shape: torch.Size([1, 1, 19221, 19221]) Position ids shape: torch.Size([1, 19221]) Input IDs shape: torch.Size([1, 19221]) Labels shape: torch.Size([1, 19221]) Final batch size: 1, sequence length: 21556 Attention mask shape: torch.Size([1, 1, 21556, 21556]) Position ids shape: torch.Size([1, 21556]) Input IDs shape: torch.Size([1, 21556]) Labels shape: torch.Size([1, 21556]) Final batch size: 1, sequence length: 15226 Attention mask shape: torch.Size([1, 1, 15226, 15226]) Position ids shape: torch.Size([1, 15226]) Input IDs shape: torch.Size([1, 15226]) Labels shape: torch.Size([1, 15226]) Final batch size: 1, sequence length: 22079 Attention mask shape: torch.Size([1, 1, 22079, 22079]) Position ids shape: torch.Size([1, 22079]) Input IDs shape: torch.Size([1, 22079]) Labels shape: torch.Size([1, 22079]) Final batch size: 1, sequence length: 21028 Attention mask shape: torch.Size([1, 1, 21028, 21028]) Position ids shape: torch.Size([1, 21028]) Input IDs shape: torch.Size([1, 21028]) Labels shape: torch.Size([1, 21028]) Final batch size: 1, sequence length: 22618 Attention mask shape: torch.Size([1, 1, 22618, 22618]) Position ids shape: torch.Size([1, 22618]) Input IDs shape: torch.Size([1, 22618]) Labels shape: torch.Size([1, 22618]) Final batch size: 1, sequence length: 15221 Attention mask shape: torch.Size([1, 1, 15221, 15221]) Position ids shape: torch.Size([1, 15221]) Input IDs shape: torch.Size([1, 15221]) Labels shape: torch.Size([1, 15221]) Final batch size: 1, sequence length: 10269 Attention mask shape: torch.Size([1, 1, 10269, 10269]) Position ids shape: torch.Size([1, 10269]) Input IDs shape: torch.Size([1, 10269]) Labels shape: torch.Size([1, 10269]) Final batch size: 1, sequence length: 23766 Attention mask shape: torch.Size([1, 1, 23766, 23766]) Position ids shape: torch.Size([1, 23766]) Input IDs shape: torch.Size([1, 23766]) Labels shape: torch.Size([1, 23766]) Final batch size: 1, sequence length: 22718 Attention mask shape: torch.Size([1, 1, 22718, 22718]) Position ids shape: torch.Size([1, 22718]) Input IDs shape: torch.Size([1, 22718]) Labels shape: torch.Size([1, 22718]) Final batch size: 1, sequence length: 22138 Attention mask shape: torch.Size([1, 1, 22138, 22138]) Position ids shape: torch.Size([1, 22138]) Input IDs shape: torch.Size([1, 22138]) Labels shape: torch.Size([1, 22138]) Final batch size: 1, sequence length: 18469 Attention mask shape: torch.Size([1, 1, 18469, 18469]) Position ids shape: torch.Size([1, 18469]) Input IDs shape: torch.Size([1, 18469]) Labels shape: torch.Size([1, 18469]) Final batch size: 1, sequence length: 19512 Attention mask shape: torch.Size([1, 1, 19512, 19512]) Position ids shape: torch.Size([1, 19512]) Input IDs shape: torch.Size([1, 19512]) Labels shape: torch.Size([1, 19512]) Final batch size: 1, sequence length: 7584 Attention mask shape: torch.Size([1, 1, 7584, 7584]) Position ids shape: torch.Size([1, 7584]) Input IDs shape: torch.Size([1, 7584]) Labels shape: torch.Size([1, 7584]) Final batch size: 1, sequence length: 19428 Attention mask shape: torch.Size([1, 1, 19428, 19428]) Position ids shape: torch.Size([1, 19428]) Input IDs shape: torch.Size([1, 19428]) Labels shape: torch.Size([1, 19428]) Final batch size: 1, sequence length: 23973 Attention mask shape: torch.Size([1, 1, 23973, 23973]) Position ids shape: torch.Size([1, 23973]) Input IDs shape: torch.Size([1, 23973]) Labels shape: torch.Size([1, 23973]) Final batch size: 1, sequence length: 24499 Attention mask shape: torch.Size([1, 1, 24499, 24499]) Position ids shape: torch.Size([1, 24499]) Input IDs shape: torch.Size([1, 24499]) Labels shape: torch.Size([1, 24499]) Final batch size: 1, sequence length: 22763 Attention mask shape: torch.Size([1, 1, 22763, 22763]) Position ids shape: torch.Size([1, 22763]) Input IDs shape: torch.Size([1, 22763]) Labels shape: torch.Size([1, 22763]) Final batch size: 1, sequence length: 25999 Attention mask shape: torch.Size([1, 1, 25999, 25999]) Position ids shape: torch.Size([1, 25999]) Input IDs shape: torch.Size([1, 25999]) Labels shape: torch.Size([1, 25999]) Final batch size: 1, sequence length: 24694 Attention mask shape: torch.Size([1, 1, 24694, 24694]) Position ids shape: torch.Size([1, 24694]) Input IDs shape: torch.Size([1, 24694]) Labels shape: torch.Size([1, 24694]) Final batch size: 1, sequence length: 24909 Attention mask shape: torch.Size([1, 1, 24909, 24909]) Position ids shape: torch.Size([1, 24909]) Input IDs shape: torch.Size([1, 24909]) Labels shape: torch.Size([1, 24909]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 22235 Attention mask shape: torch.Size([1, 1, 22235, 22235]) Position ids shape: torch.Size([1, 22235]) Input IDs shape: torch.Size([1, 22235]) Labels shape: torch.Size([1, 22235]) Final batch size: 1, sequence length: 28497 Attention mask shape: torch.Size([1, 1, 28497, 28497]) Position ids shape: torch.Size([1, 28497]) Input IDs shape: torch.Size([1, 28497]) Labels shape: torch.Size([1, 28497]) Final batch size: 1, sequence length: 28749 Attention mask shape: torch.Size([1, 1, 28749, 28749]) Position ids shape: torch.Size([1, 28749]) Input IDs shape: torch.Size([1, 28749]) Labels shape: torch.Size([1, 28749]) Final batch size: 1, sequence length: 14784 Attention mask shape: torch.Size([1, 1, 14784, 14784]) Position ids shape: torch.Size([1, 14784]) Input IDs shape: torch.Size([1, 14784]) Labels shape: torch.Size([1, 14784]) Final batch size: 1, sequence length: 15184 Attention mask shape: torch.Size([1, 1, 15184, 15184]) Position ids shape: torch.Size([1, 15184]) Input IDs shape: torch.Size([1, 15184]) Labels shape: torch.Size([1, 15184]) Final batch size: 1, sequence length: 25622 Attention mask shape: torch.Size([1, 1, 25622, 25622]) Position ids shape: torch.Size([1, 25622]) Input IDs shape: torch.Size([1, 25622]) Labels shape: torch.Size([1, 25622]) Final batch size: 1, sequence length: 27566 Attention mask shape: torch.Size([1, 1, 27566, 27566]) Position ids shape: torch.Size([1, 27566]) Input IDs shape: torch.Size([1, 27566]) Labels shape: torch.Size([1, 27566]) Final batch size: 1, sequence length: 27327 Attention mask shape: torch.Size([1, 1, 27327, 27327]) Position ids shape: torch.Size([1, 27327]) Input IDs shape: torch.Size([1, 27327]) Labels shape: torch.Size([1, 27327]) Final batch size: 1, sequence length: 21826 Attention mask shape: torch.Size([1, 1, 21826, 21826]) Position ids shape: torch.Size([1, 21826]) Input IDs shape: torch.Size([1, 21826]) Labels shape: torch.Size([1, 21826]) Final batch size: 1, sequence length: 30356 Attention mask shape: torch.Size([1, 1, 30356, 30356]) Position ids shape: torch.Size([1, 30356]) Input IDs shape: torch.Size([1, 30356]) Labels shape: torch.Size([1, 30356]) Final batch size: 1, sequence length: 23851 Attention mask shape: torch.Size([1, 1, 23851, 23851]) Position ids shape: torch.Size([1, 23851]) Input IDs shape: torch.Size([1, 23851]) Labels shape: torch.Size([1, 23851]) Final batch size: 1, sequence length: 22366 Attention mask shape: torch.Size([1, 1, 22366, 22366]) Position ids shape: torch.Size([1, 22366]) Input IDs shape: torch.Size([1, 22366]) Labels shape: torch.Size([1, 22366]) Final batch size: 1, sequence length: 21034 Attention mask shape: torch.Size([1, 1, 21034, 21034]) Position ids shape: torch.Size([1, 21034]) Input IDs shape: torch.Size([1, 21034]) Labels shape: torch.Size([1, 21034]) Final batch size: 1, sequence length: 11795 Attention mask shape: torch.Size([1, 1, 11795, 11795]) Position ids shape: torch.Size([1, 11795]) Input IDs shape: torch.Size([1, 11795]) Labels shape: torch.Size([1, 11795]) Final batch size: 1, sequence length: 31903 Attention mask shape: torch.Size([1, 1, 31903, 31903]) Position ids shape: torch.Size([1, 31903]) Input IDs shape: torch.Size([1, 31903]) Labels shape: torch.Size([1, 31903]) Final batch size: 1, sequence length: 27489 Attention mask shape: torch.Size([1, 1, 27489, 27489]) Position ids shape: torch.Size([1, 27489]) Input IDs shape: torch.Size([1, 27489]) Labels shape: torch.Size([1, 27489]) Final batch size: 1, sequence length: 34673 Attention mask shape: torch.Size([1, 1, 34673, 34673]) Position ids shape: torch.Size([1, 34673]) Input IDs shape: torch.Size([1, 34673]) Labels shape: torch.Size([1, 34673]) Final batch size: 1, sequence length: 36777 Attention mask shape: torch.Size([1, 1, 36777, 36777]) Position ids shape: torch.Size([1, 36777]) Input IDs shape: torch.Size([1, 36777]) Labels shape: torch.Size([1, 36777]) Final batch size: 1, sequence length: 15588 Attention mask shape: torch.Size([1, 1, 15588, 15588]) Position ids shape: torch.Size([1, 15588]) Input IDs shape: torch.Size([1, 15588]) Labels shape: torch.Size([1, 15588]) Final batch size: 1, sequence length: 37285 Attention mask shape: torch.Size([1, 1, 37285, 37285]) Position ids shape: torch.Size([1, 37285]) Input IDs shape: torch.Size([1, 37285]) Labels shape: torch.Size([1, 37285]) Final batch size: 1, sequence length: 36723 Attention mask shape: torch.Size([1, 1, 36723, 36723]) Position ids shape: torch.Size([1, 36723]) Input IDs shape: torch.Size([1, 36723]) Labels shape: torch.Size([1, 36723]) Final batch size: 1, sequence length: 38848 Attention mask shape: torch.Size([1, 1, 38848, 38848]) Position ids shape: torch.Size([1, 38848]) Input IDs shape: torch.Size([1, 38848]) Labels shape: torch.Size([1, 38848]) Final batch size: 1, sequence length: 28176 Attention mask shape: torch.Size([1, 1, 28176, 28176]) Position ids shape: torch.Size([1, 28176]) Input IDs shape: torch.Size([1, 28176]) Labels shape: torch.Size([1, 28176]) Final batch size: 1, sequence length: 38049 Attention mask shape: torch.Size([1, 1, 38049, 38049]) Position ids shape: torch.Size([1, 38049]) Input IDs shape: torch.Size([1, 38049]) Labels shape: torch.Size([1, 38049]) Final batch size: 1, sequence length: 40325 Attention mask shape: torch.Size([1, 1, 40325, 40325]) Position ids shape: torch.Size([1, 40325]) Input IDs shape: torch.Size([1, 40325]) Labels shape: torch.Size([1, 40325]) Final batch size: 1, sequence length: 39253 Attention mask shape: torch.Size([1, 1, 39253, 39253]) Position ids shape: torch.Size([1, 39253]) Input IDs shape: torch.Size([1, 39253]) Labels shape: torch.Size([1, 39253]) Final batch size: 1, sequence length: 24801 Attention mask shape: torch.Size([1, 1, 24801, 24801]) Position ids shape: torch.Size([1, 24801]) Input IDs shape: torch.Size([1, 24801]) Labels shape: torch.Size([1, 24801]) Final batch size: 1, sequence length: 38104 Attention mask shape: torch.Size([1, 1, 38104, 38104]) Position ids shape: torch.Size([1, 38104]) Input IDs shape: torch.Size([1, 38104]) Labels shape: torch.Size([1, 38104]) Final batch size: 1, sequence length: 25946 Attention mask shape: torch.Size([1, 1, 25946, 25946]) Position ids shape: torch.Size([1, 25946]) Input IDs shape: torch.Size([1, 25946]) Labels shape: torch.Size([1, 25946]) Final batch size: 1, sequence length: 6882 Attention mask shape: torch.Size([1, 1, 6882, 6882]) Position ids shape: torch.Size([1, 6882]) Input IDs shape: torch.Size([1, 6882]) Labels shape: torch.Size([1, 6882]) Final batch size: 1, sequence length: 20755 Attention mask shape: torch.Size([1, 1, 20755, 20755]) Position ids shape: torch.Size([1, 20755]) Input IDs shape: torch.Size([1, 20755]) Labels shape: torch.Size([1, 20755]) Final batch size: 1, sequence length: 34289 Attention mask shape: torch.Size([1, 1, 34289, 34289]) Position ids shape: torch.Size([1, 34289]) Input IDs shape: torch.Size([1, 34289]) Labels shape: torch.Size([1, 34289]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40317 Attention mask shape: torch.Size([1, 1, 40317, 40317]) Position ids shape: torch.Size([1, 40317]) Input IDs shape: torch.Size([1, 40317]) Labels shape: torch.Size([1, 40317]) Final batch size: 1, sequence length: 35904 Attention mask shape: torch.Size([1, 1, 35904, 35904]) Position ids shape: torch.Size([1, 35904]) Input IDs shape: torch.Size([1, 35904]) Labels shape: torch.Size([1, 35904]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36786 Attention mask shape: torch.Size([1, 1, 36786, 36786]) Position ids shape: torch.Size([1, 36786]) Input IDs shape: torch.Size([1, 36786]) Labels shape: torch.Size([1, 36786]) Final batch size: 1, sequence length: 40937 Attention mask shape: torch.Size([1, 1, 40937, 40937]) Position ids shape: torch.Size([1, 40937]) Input IDs shape: torch.Size([1, 40937]) Labels shape: torch.Size([1, 40937]) Final batch size: 1, sequence length: 40657 Attention mask shape: torch.Size([1, 1, 40657, 40657]) Position ids shape: torch.Size([1, 40657]) Input IDs shape: torch.Size([1, 40657]) Labels shape: torch.Size([1, 40657]) Final batch size: 1, sequence length: 10469 Attention mask shape: torch.Size([1, 1, 10469, 10469]) Position ids shape: torch.Size([1, 10469]) Input IDs shape: torch.Size([1, 10469]) Labels shape: torch.Size([1, 10469]) Final batch size: 1, sequence length: 19639 Attention mask shape: torch.Size([1, 1, 19639, 19639]) Position ids shape: torch.Size([1, 19639]) Input IDs shape: torch.Size([1, 19639]) Labels shape: torch.Size([1, 19639]) Final batch size: 1, sequence length: 16649 Attention mask shape: torch.Size([1, 1, 16649, 16649]) Position ids shape: torch.Size([1, 16649]) Input IDs shape: torch.Size([1, 16649]) Labels shape: torch.Size([1, 16649]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31714 Attention mask shape: torch.Size([1, 1, 31714, 31714]) Position ids shape: torch.Size([1, 31714]) Input IDs shape: torch.Size([1, 31714]) Labels shape: torch.Size([1, 31714]) Final batch size: 1, sequence length: 19437 Attention mask shape: torch.Size([1, 1, 19437, 19437]) Position ids shape: torch.Size([1, 19437]) Input IDs shape: torch.Size([1, 19437]) Labels shape: torch.Size([1, 19437]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38686 Attention mask shape: torch.Size([1, 1, 38686, 38686]) Position ids shape: torch.Size([1, 38686]) Input IDs shape: torch.Size([1, 38686]) Labels shape: torch.Size([1, 38686]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25963 Attention mask shape: torch.Size([1, 1, 25963, 25963]) Position ids shape: torch.Size([1, 25963]) Input IDs shape: torch.Size([1, 25963]) Labels shape: torch.Size([1, 25963]) Final batch size: 1, sequence length: 21547 Attention mask shape: torch.Size([1, 1, 21547, 21547]) Position ids shape: torch.Size([1, 21547]) Input IDs shape: torch.Size([1, 21547]) Labels shape: torch.Size([1, 21547]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17882 Attention mask shape: torch.Size([1, 1, 17882, 17882]) Position ids shape: torch.Size([1, 17882]) Input IDs shape: torch.Size([1, 17882]) Labels shape: torch.Size([1, 17882]) Final batch size: 1, sequence length: 36128 Attention mask shape: torch.Size([1, 1, 36128, 36128]) Position ids shape: torch.Size([1, 36128]) Input IDs shape: torch.Size([1, 36128]) Labels shape: torch.Size([1, 36128]) Final batch size: 1, sequence length: 34540 Attention mask shape: torch.Size([1, 1, 34540, 34540]) Position ids shape: torch.Size([1, 34540]) Input IDs shape: torch.Size([1, 34540]) Labels shape: torch.Size([1, 34540]) Final batch size: 1, sequence length: 26660 Attention mask shape: torch.Size([1, 1, 26660, 26660]) Position ids shape: torch.Size([1, 26660]) Input IDs shape: torch.Size([1, 26660]) Labels shape: torch.Size([1, 26660]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17379 Attention mask shape: torch.Size([1, 1, 17379, 17379]) Position ids shape: torch.Size([1, 17379]) Input IDs shape: torch.Size([1, 17379]) Labels shape: torch.Size([1, 17379]) Final batch size: 1, sequence length: 27947 Attention mask shape: torch.Size([1, 1, 27947, 27947]) Position ids shape: torch.Size([1, 27947]) Input IDs shape: torch.Size([1, 27947]) Labels shape: torch.Size([1, 27947]) Final batch size: 1, sequence length: 37946 Attention mask shape: torch.Size([1, 1, 37946, 37946]) Position ids shape: torch.Size([1, 37946]) Input IDs shape: torch.Size([1, 37946]) Labels shape: torch.Size([1, 37946]) Final batch size: 1, sequence length: 20843 Attention mask shape: torch.Size([1, 1, 20843, 20843]) Position ids shape: torch.Size([1, 20843]) Input IDs shape: torch.Size([1, 20843]) Labels shape: torch.Size([1, 20843]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21496 Attention mask shape: torch.Size([1, 1, 21496, 21496]) Position ids shape: torch.Size([1, 21496]) Input IDs shape: torch.Size([1, 21496]) Labels shape: torch.Size([1, 21496]) Final batch size: 1, sequence length: 24551 Attention mask shape: torch.Size([1, 1, 24551, 24551]) Position ids shape: torch.Size([1, 24551]) Input IDs shape: torch.Size([1, 24551]) Labels shape: torch.Size([1, 24551]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7681 Attention mask shape: torch.Size([1, 1, 7681, 7681]) Position ids shape: torch.Size([1, 7681]) Input IDs shape: torch.Size([1, 7681]) Labels shape: torch.Size([1, 7681]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36317 Attention mask shape: torch.Size([1, 1, 36317, 36317]) Position ids shape: torch.Size([1, 36317]) Input IDs shape: torch.Size([1, 36317]) Labels shape: torch.Size([1, 36317]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2609, 'grad_norm': 0.17397274372155297, 'learning_rate': 1.1142701927151456e-06, 'num_tokens': -inf, 'epoch': 6.5} Final batch size: 1, sequence length: 5328 Attention mask shape: torch.Size([1, 1, 5328, 5328]) Position ids shape: torch.Size([1, 5328]) Input IDs shape: torch.Size([1, 5328]) Labels shape: torch.Size([1, 5328]) Final batch size: 1, sequence length: 5525 Attention mask shape: torch.Size([1, 1, 5525, 5525]) Position ids shape: torch.Size([1, 5525]) Input IDs shape: torch.Size([1, 5525]) Labels shape: torch.Size([1, 5525]) Final batch size: 1, sequence length: 10273 Attention mask shape: torch.Size([1, 1, 10273, 10273]) Position ids shape: torch.Size([1, 10273]) Input IDs shape: torch.Size([1, 10273]) Labels shape: torch.Size([1, 10273]) Final batch size: 1, sequence length: 11464 Attention mask shape: torch.Size([1, 1, 11464, 11464]) Position ids shape: torch.Size([1, 11464]) Input IDs shape: torch.Size([1, 11464]) Labels shape: torch.Size([1, 11464]) Final batch size: 1, sequence length: 11225 Attention mask shape: torch.Size([1, 1, 11225, 11225]) Position ids shape: torch.Size([1, 11225]) Input IDs shape: torch.Size([1, 11225]) Labels shape: torch.Size([1, 11225]) Final batch size: 1, sequence length: 12419 Attention mask shape: torch.Size([1, 1, 12419, 12419]) Position ids shape: torch.Size([1, 12419]) Input IDs shape: torch.Size([1, 12419]) Labels shape: torch.Size([1, 12419]) Final batch size: 1, sequence length: 10434 Attention mask shape: torch.Size([1, 1, 10434, 10434]) Position ids shape: torch.Size([1, 10434]) Input IDs shape: torch.Size([1, 10434]) Labels shape: torch.Size([1, 10434]) Final batch size: 1, sequence length: 14226 Attention mask shape: torch.Size([1, 1, 14226, 14226]) Position ids shape: torch.Size([1, 14226]) Input IDs shape: torch.Size([1, 14226]) Labels shape: torch.Size([1, 14226]) Final batch size: 1, sequence length: 13657 Attention mask shape: torch.Size([1, 1, 13657, 13657]) Position ids shape: torch.Size([1, 13657]) Input IDs shape: torch.Size([1, 13657]) Labels shape: torch.Size([1, 13657]) Final batch size: 1, sequence length: 9648 Attention mask shape: torch.Size([1, 1, 9648, 9648]) Position ids shape: torch.Size([1, 9648]) Input IDs shape: torch.Size([1, 9648]) Labels shape: torch.Size([1, 9648]) Final batch size: 1, sequence length: 14360 Attention mask shape: torch.Size([1, 1, 14360, 14360]) Position ids shape: torch.Size([1, 14360]) Input IDs shape: torch.Size([1, 14360]) Labels shape: torch.Size([1, 14360]) Final batch size: 1, sequence length: 16398 Attention mask shape: torch.Size([1, 1, 16398, 16398]) Position ids shape: torch.Size([1, 16398]) Input IDs shape: torch.Size([1, 16398]) Labels shape: torch.Size([1, 16398]) Final batch size: 1, sequence length: 13607 Attention mask shape: torch.Size([1, 1, 13607, 13607]) Position ids shape: torch.Size([1, 13607]) Input IDs shape: torch.Size([1, 13607]) Labels shape: torch.Size([1, 13607]) Final batch size: 1, sequence length: 16053 Attention mask shape: torch.Size([1, 1, 16053, 16053]) Position ids shape: torch.Size([1, 16053]) Input IDs shape: torch.Size([1, 16053]) Labels shape: torch.Size([1, 16053]) Final batch size: 1, sequence length: 16756 Attention mask shape: torch.Size([1, 1, 16756, 16756]) Position ids shape: torch.Size([1, 16756]) Input IDs shape: torch.Size([1, 16756]) Labels shape: torch.Size([1, 16756]) Final batch size: 1, sequence length: 16223 Attention mask shape: torch.Size([1, 1, 16223, 16223]) Position ids shape: torch.Size([1, 16223]) Input IDs shape: torch.Size([1, 16223]) Labels shape: torch.Size([1, 16223]) Final batch size: 1, sequence length: 18408 Attention mask shape: torch.Size([1, 1, 18408, 18408]) Position ids shape: torch.Size([1, 18408]) Input IDs shape: torch.Size([1, 18408]) Labels shape: torch.Size([1, 18408]) Final batch size: 1, sequence length: 10017 Attention mask shape: torch.Size([1, 1, 10017, 10017]) Position ids shape: torch.Size([1, 10017]) Input IDs shape: torch.Size([1, 10017]) Labels shape: torch.Size([1, 10017]) Final batch size: 1, sequence length: 19278 Attention mask shape: torch.Size([1, 1, 19278, 19278]) Position ids shape: torch.Size([1, 19278]) Input IDs shape: torch.Size([1, 19278]) Labels shape: torch.Size([1, 19278]) Final batch size: 1, sequence length: 15243 Attention mask shape: torch.Size([1, 1, 15243, 15243]) Position ids shape: torch.Size([1, 15243]) Input IDs shape: torch.Size([1, 15243]) Labels shape: torch.Size([1, 15243]) Final batch size: 1, sequence length: 17117 Attention mask shape: torch.Size([1, 1, 17117, 17117]) Position ids shape: torch.Size([1, 17117]) Input IDs shape: torch.Size([1, 17117]) Labels shape: torch.Size([1, 17117]) Final batch size: 1, sequence length: 17843 Attention mask shape: torch.Size([1, 1, 17843, 17843]) Position ids shape: torch.Size([1, 17843]) Input IDs shape: torch.Size([1, 17843]) Labels shape: torch.Size([1, 17843]) Final batch size: 1, sequence length: 19259 Attention mask shape: torch.Size([1, 1, 19259, 19259]) Position ids shape: torch.Size([1, 19259]) Input IDs shape: torch.Size([1, 19259]) Labels shape: torch.Size([1, 19259]) Final batch size: 1, sequence length: 14437 Attention mask shape: torch.Size([1, 1, 14437, 14437]) Position ids shape: torch.Size([1, 14437]) Input IDs shape: torch.Size([1, 14437]) Labels shape: torch.Size([1, 14437]) Final batch size: 1, sequence length: 19892 Attention mask shape: torch.Size([1, 1, 19892, 19892]) Position ids shape: torch.Size([1, 19892]) Input IDs shape: torch.Size([1, 19892]) Labels shape: torch.Size([1, 19892]) Final batch size: 1, sequence length: 17294 Attention mask shape: torch.Size([1, 1, 17294, 17294]) Position ids shape: torch.Size([1, 17294]) Input IDs shape: torch.Size([1, 17294]) Labels shape: torch.Size([1, 17294]) Final batch size: 1, sequence length: 21455 Attention mask shape: torch.Size([1, 1, 21455, 21455]) Position ids shape: torch.Size([1, 21455]) Input IDs shape: torch.Size([1, 21455]) Labels shape: torch.Size([1, 21455]) Final batch size: 1, sequence length: 19767 Attention mask shape: torch.Size([1, 1, 19767, 19767]) Position ids shape: torch.Size([1, 19767]) Input IDs shape: torch.Size([1, 19767]) Labels shape: torch.Size([1, 19767]) Final batch size: 1, sequence length: 14025 Attention mask shape: torch.Size([1, 1, 14025, 14025]) Position ids shape: torch.Size([1, 14025]) Input IDs shape: torch.Size([1, 14025]) Labels shape: torch.Size([1, 14025]) Final batch size: 1, sequence length: 5734 Attention mask shape: torch.Size([1, 1, 5734, 5734]) Position ids shape: torch.Size([1, 5734]) Input IDs shape: torch.Size([1, 5734]) Labels shape: torch.Size([1, 5734]) Final batch size: 1, sequence length: 20433 Attention mask shape: torch.Size([1, 1, 20433, 20433]) Position ids shape: torch.Size([1, 20433]) Input IDs shape: torch.Size([1, 20433]) Labels shape: torch.Size([1, 20433]) Final batch size: 1, sequence length: 23334 Attention mask shape: torch.Size([1, 1, 23334, 23334]) Position ids shape: torch.Size([1, 23334]) Input IDs shape: torch.Size([1, 23334]) Labels shape: torch.Size([1, 23334]) Final batch size: 1, sequence length: 25405 Attention mask shape: torch.Size([1, 1, 25405, 25405]) Position ids shape: torch.Size([1, 25405]) Input IDs shape: torch.Size([1, 25405]) Labels shape: torch.Size([1, 25405]) Final batch size: 1, sequence length: 6378 Attention mask shape: torch.Size([1, 1, 6378, 6378]) Position ids shape: torch.Size([1, 6378]) Input IDs shape: torch.Size([1, 6378]) Labels shape: torch.Size([1, 6378]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 14429 Attention mask shape: torch.Size([1, 1, 14429, 14429]) Position ids shape: torch.Size([1, 14429]) Input IDs shape: torch.Size([1, 14429]) Labels shape: torch.Size([1, 14429]) Final batch size: 1, sequence length: 25548 Attention mask shape: torch.Size([1, 1, 25548, 25548]) Position ids shape: torch.Size([1, 25548]) Input IDs shape: torch.Size([1, 25548]) Labels shape: torch.Size([1, 25548]) Final batch size: 1, sequence length: 20695 Attention mask shape: torch.Size([1, 1, 20695, 20695]) Position ids shape: torch.Size([1, 20695]) Input IDs shape: torch.Size([1, 20695]) Labels shape: torch.Size([1, 20695]) Final batch size: 1, sequence length: 19492 Attention mask shape: torch.Size([1, 1, 19492, 19492]) Position ids shape: torch.Size([1, 19492]) Input IDs shape: torch.Size([1, 19492]) Labels shape: torch.Size([1, 19492]) Final batch size: 1, sequence length: 25381 Attention mask shape: torch.Size([1, 1, 25381, 25381]) Position ids shape: torch.Size([1, 25381]) Input IDs shape: torch.Size([1, 25381]) Labels shape: torch.Size([1, 25381]) Final batch size: 1, sequence length: 17985 Attention mask shape: torch.Size([1, 1, 17985, 17985]) Position ids shape: torch.Size([1, 17985]) Input IDs shape: torch.Size([1, 17985]) Labels shape: torch.Size([1, 17985]) Final batch size: 1, sequence length: 27659 Attention mask shape: torch.Size([1, 1, 27659, 27659]) Position ids shape: torch.Size([1, 27659]) Input IDs shape: torch.Size([1, 27659]) Labels shape: torch.Size([1, 27659]) Final batch size: 1, sequence length: 13623 Attention mask shape: torch.Size([1, 1, 13623, 13623]) Position ids shape: torch.Size([1, 13623]) Input IDs shape: torch.Size([1, 13623]) Labels shape: torch.Size([1, 13623]) Final batch size: 1, sequence length: 28623 Attention mask shape: torch.Size([1, 1, 28623, 28623]) Position ids shape: torch.Size([1, 28623]) Input IDs shape: torch.Size([1, 28623]) Labels shape: torch.Size([1, 28623]) Final batch size: 1, sequence length: 26063 Attention mask shape: torch.Size([1, 1, 26063, 26063]) Position ids shape: torch.Size([1, 26063]) Input IDs shape: torch.Size([1, 26063]) Labels shape: torch.Size([1, 26063]) Final batch size: 1, sequence length: 28166 Attention mask shape: torch.Size([1, 1, 28166, 28166]) Position ids shape: torch.Size([1, 28166]) Input IDs shape: torch.Size([1, 28166]) Labels shape: torch.Size([1, 28166]) Final batch size: 1, sequence length: 24885 Attention mask shape: torch.Size([1, 1, 24885, 24885]) Position ids shape: torch.Size([1, 24885]) Input IDs shape: torch.Size([1, 24885]) Labels shape: torch.Size([1, 24885]) Final batch size: 1, sequence length: 29561 Attention mask shape: torch.Size([1, 1, 29561, 29561]) Position ids shape: torch.Size([1, 29561]) Input IDs shape: torch.Size([1, 29561]) Labels shape: torch.Size([1, 29561]) Final batch size: 1, sequence length: 30236 Attention mask shape: torch.Size([1, 1, 30236, 30236]) Position ids shape: torch.Size([1, 30236]) Input IDs shape: torch.Size([1, 30236]) Labels shape: torch.Size([1, 30236]) Final batch size: 1, sequence length: 17951 Attention mask shape: torch.Size([1, 1, 17951, 17951]) Position ids shape: torch.Size([1, 17951]) Input IDs shape: torch.Size([1, 17951]) Labels shape: torch.Size([1, 17951]) Final batch size: 1, sequence length: 27484 Attention mask shape: torch.Size([1, 1, 27484, 27484]) Position ids shape: torch.Size([1, 27484]) Input IDs shape: torch.Size([1, 27484]) Labels shape: torch.Size([1, 27484]) Final batch size: 1, sequence length: 22932 Attention mask shape: torch.Size([1, 1, 22932, 22932]) Position ids shape: torch.Size([1, 22932]) Input IDs shape: torch.Size([1, 22932]) Labels shape: torch.Size([1, 22932]) Final batch size: 1, sequence length: 28823 Attention mask shape: torch.Size([1, 1, 28823, 28823]) Position ids shape: torch.Size([1, 28823]) Input IDs shape: torch.Size([1, 28823]) Labels shape: torch.Size([1, 28823]) Final batch size: 1, sequence length: 17634 Attention mask shape: torch.Size([1, 1, 17634, 17634]) Position ids shape: torch.Size([1, 17634]) Input IDs shape: torch.Size([1, 17634]) Labels shape: torch.Size([1, 17634]) Final batch size: 1, sequence length: 29824 Attention mask shape: torch.Size([1, 1, 29824, 29824]) Position ids shape: torch.Size([1, 29824]) Input IDs shape: torch.Size([1, 29824]) Labels shape: torch.Size([1, 29824]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 26639 Attention mask shape: torch.Size([1, 1, 26639, 26639]) Position ids shape: torch.Size([1, 26639]) Input IDs shape: torch.Size([1, 26639]) Labels shape: torch.Size([1, 26639]) Final batch size: 1, sequence length: 28910 Attention mask shape: torch.Size([1, 1, 28910, 28910]) Position ids shape: torch.Size([1, 28910]) Input IDs shape: torch.Size([1, 28910]) Labels shape: torch.Size([1, 28910]) Final batch size: 1, sequence length: 16716 Attention mask shape: torch.Size([1, 1, 16716, 16716]) Position ids shape: torch.Size([1, 16716]) Input IDs shape: torch.Size([1, 16716]) Labels shape: torch.Size([1, 16716]) Final batch size: 1, sequence length: 28206 Attention mask shape: torch.Size([1, 1, 28206, 28206]) Position ids shape: torch.Size([1, 28206]) Input IDs shape: torch.Size([1, 28206]) Labels shape: torch.Size([1, 28206]) Final batch size: 1, sequence length: 32786 Attention mask shape: torch.Size([1, 1, 32786, 32786]) Position ids shape: torch.Size([1, 32786]) Input IDs shape: torch.Size([1, 32786]) Labels shape: torch.Size([1, 32786]) Final batch size: 1, sequence length: 33601 Attention mask shape: torch.Size([1, 1, 33601, 33601]) Position ids shape: torch.Size([1, 33601]) Input IDs shape: torch.Size([1, 33601]) Labels shape: torch.Size([1, 33601]) Final batch size: 1, sequence length: 31464 Attention mask shape: torch.Size([1, 1, 31464, 31464]) Position ids shape: torch.Size([1, 31464]) Input IDs shape: torch.Size([1, 31464]) Labels shape: torch.Size([1, 31464]) Final batch size: 1, sequence length: 25540 Attention mask shape: torch.Size([1, 1, 25540, 25540]) Position ids shape: torch.Size([1, 25540]) Input IDs shape: torch.Size([1, 25540]) Labels shape: torch.Size([1, 25540]) Final batch size: 1, sequence length: 32660 Attention mask shape: torch.Size([1, 1, 32660, 32660]) Position ids shape: torch.Size([1, 32660]) Input IDs shape: torch.Size([1, 32660]) Labels shape: torch.Size([1, 32660]) Final batch size: 1, sequence length: 9029 Attention mask shape: torch.Size([1, 1, 9029, 9029]) Position ids shape: torch.Size([1, 9029]) Input IDs shape: torch.Size([1, 9029]) Labels shape: torch.Size([1, 9029]) Final batch size: 1, sequence length: 17914 Attention mask shape: torch.Size([1, 1, 17914, 17914]) Position ids shape: torch.Size([1, 17914]) Input IDs shape: torch.Size([1, 17914]) Labels shape: torch.Size([1, 17914]) Final batch size: 1, sequence length: 35411 Attention mask shape: torch.Size([1, 1, 35411, 35411]) Position ids shape: torch.Size([1, 35411]) Input IDs shape: torch.Size([1, 35411]) Labels shape: torch.Size([1, 35411]) Final batch size: 1, sequence length: 37555 Attention mask shape: torch.Size([1, 1, 37555, 37555]) Position ids shape: torch.Size([1, 37555]) Input IDs shape: torch.Size([1, 37555]) Labels shape: torch.Size([1, 37555]) Final batch size: 1, sequence length: 10132 Attention mask shape: torch.Size([1, 1, 10132, 10132]) Position ids shape: torch.Size([1, 10132]) Input IDs shape: torch.Size([1, 10132]) Labels shape: torch.Size([1, 10132]) Final batch size: 1, sequence length: 20951 Attention mask shape: torch.Size([1, 1, 20951, 20951]) Position ids shape: torch.Size([1, 20951]) Input IDs shape: torch.Size([1, 20951]) Labels shape: torch.Size([1, 20951]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 36131 Attention mask shape: torch.Size([1, 1, 36131, 36131]) Position ids shape: torch.Size([1, 36131]) Input IDs shape: torch.Size([1, 36131]) Labels shape: torch.Size([1, 36131]) Final batch size: 1, sequence length: 36124 Attention mask shape: torch.Size([1, 1, 36124, 36124]) Position ids shape: torch.Size([1, 36124]) Input IDs shape: torch.Size([1, 36124]) Labels shape: torch.Size([1, 36124]) Final batch size: 1, sequence length: 25529 Attention mask shape: torch.Size([1, 1, 25529, 25529]) Position ids shape: torch.Size([1, 25529]) Input IDs shape: torch.Size([1, 25529]) Labels shape: torch.Size([1, 25529]) Final batch size: 1, sequence length: 36970 Attention mask shape: torch.Size([1, 1, 36970, 36970]) Position ids shape: torch.Size([1, 36970]) Input IDs shape: torch.Size([1, 36970]) Labels shape: torch.Size([1, 36970]) Final batch size: 1, sequence length: 37016 Attention mask shape: torch.Size([1, 1, 37016, 37016]) Position ids shape: torch.Size([1, 37016]) Input IDs shape: torch.Size([1, 37016]) Labels shape: torch.Size([1, 37016]) Final batch size: 1, sequence length: 15513 Attention mask shape: torch.Size([1, 1, 15513, 15513]) Position ids shape: torch.Size([1, 15513]) Input IDs shape: torch.Size([1, 15513]) Labels shape: torch.Size([1, 15513]) Final batch size: 1, sequence length: 32515 Attention mask shape: torch.Size([1, 1, 32515, 32515]) Position ids shape: torch.Size([1, 32515]) Input IDs shape: torch.Size([1, 32515]) Labels shape: torch.Size([1, 32515]) Final batch size: 1, sequence length: 20533 Attention mask shape: torch.Size([1, 1, 20533, 20533]) Position ids shape: torch.Size([1, 20533]) Input IDs shape: torch.Size([1, 20533]) Labels shape: torch.Size([1, 20533]) Final batch size: 1, sequence length: 38712 Attention mask shape: torch.Size([1, 1, 38712, 38712]) Position ids shape: torch.Size([1, 38712]) Input IDs shape: torch.Size([1, 38712]) Labels shape: torch.Size([1, 38712]) Final batch size: 1, sequence length: 35525 Attention mask shape: torch.Size([1, 1, 35525, 35525]) Position ids shape: torch.Size([1, 35525]) Input IDs shape: torch.Size([1, 35525]) Labels shape: torch.Size([1, 35525]) Final batch size: 1, sequence length: 14372 Attention mask shape: torch.Size([1, 1, 14372, 14372]) Position ids shape: torch.Size([1, 14372]) Input IDs shape: torch.Size([1, 14372]) Labels shape: torch.Size([1, 14372]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36292 Attention mask shape: torch.Size([1, 1, 36292, 36292]) Position ids shape: torch.Size([1, 36292]) Input IDs shape: torch.Size([1, 36292]) Labels shape: torch.Size([1, 36292]) Final batch size: 1, sequence length: 33422 Attention mask shape: torch.Size([1, 1, 33422, 33422]) Position ids shape: torch.Size([1, 33422]) Input IDs shape: torch.Size([1, 33422]) Labels shape: torch.Size([1, 33422]) Final batch size: 1, sequence length: 31084 Attention mask shape: torch.Size([1, 1, 31084, 31084]) Position ids shape: torch.Size([1, 31084]) Input IDs shape: torch.Size([1, 31084]) Labels shape: torch.Size([1, 31084]) Final batch size: 1, sequence length: 26562 Attention mask shape: torch.Size([1, 1, 26562, 26562]) Position ids shape: torch.Size([1, 26562]) Input IDs shape: torch.Size([1, 26562]) Labels shape: torch.Size([1, 26562]) Final batch size: 1, sequence length: 38210 Attention mask shape: torch.Size([1, 1, 38210, 38210]) Position ids shape: torch.Size([1, 38210]) Input IDs shape: torch.Size([1, 38210]) Labels shape: torch.Size([1, 38210]) Final batch size: 1, sequence length: 38071 Attention mask shape: torch.Size([1, 1, 38071, 38071]) Position ids shape: torch.Size([1, 38071]) Input IDs shape: torch.Size([1, 38071]) Labels shape: torch.Size([1, 38071]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 35775 Attention mask shape: torch.Size([1, 1, 35775, 35775]) Position ids shape: torch.Size([1, 35775]) Input IDs shape: torch.Size([1, 35775]) Labels shape: torch.Size([1, 35775]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30009 Attention mask shape: torch.Size([1, 1, 30009, 30009]) Position ids shape: torch.Size([1, 30009]) Input IDs shape: torch.Size([1, 30009]) Labels shape: torch.Size([1, 30009]) Final batch size: 1, sequence length: 16409 Attention mask shape: torch.Size([1, 1, 16409, 16409]) Position ids shape: torch.Size([1, 16409]) Input IDs shape: torch.Size([1, 16409]) Labels shape: torch.Size([1, 16409]) Final batch size: 1, sequence length: 28879 Attention mask shape: torch.Size([1, 1, 28879, 28879]) Position ids shape: torch.Size([1, 28879]) Input IDs shape: torch.Size([1, 28879]) Labels shape: torch.Size([1, 28879]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40753 Attention mask shape: torch.Size([1, 1, 40753, 40753]) Position ids shape: torch.Size([1, 40753]) Input IDs shape: torch.Size([1, 40753]) Labels shape: torch.Size([1, 40753]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36185 Attention mask shape: torch.Size([1, 1, 36185, 36185]) Position ids shape: torch.Size([1, 36185]) Input IDs shape: torch.Size([1, 36185]) Labels shape: torch.Size([1, 36185]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36534 Attention mask shape: torch.Size([1, 1, 36534, 36534]) Position ids shape: torch.Size([1, 36534]) Input IDs shape: torch.Size([1, 36534]) Labels shape: torch.Size([1, 36534]) Final batch size: 1, sequence length: 26781 Attention mask shape: torch.Size([1, 1, 26781, 26781]) Position ids shape: torch.Size([1, 26781]) Input IDs shape: torch.Size([1, 26781]) Labels shape: torch.Size([1, 26781]) Final batch size: 1, sequence length: 30653 Attention mask shape: torch.Size([1, 1, 30653, 30653]) Position ids shape: torch.Size([1, 30653]) Input IDs shape: torch.Size([1, 30653]) Labels shape: torch.Size([1, 30653]) Final batch size: 1, sequence length: 15031 Attention mask shape: torch.Size([1, 1, 15031, 15031]) Position ids shape: torch.Size([1, 15031]) Input IDs shape: torch.Size([1, 15031]) Labels shape: torch.Size([1, 15031]) Final batch size: 1, sequence length: 18884 Attention mask shape: torch.Size([1, 1, 18884, 18884]) Position ids shape: torch.Size([1, 18884]) Input IDs shape: torch.Size([1, 18884]) Labels shape: torch.Size([1, 18884]) Final batch size: 1, sequence length: 20307 Attention mask shape: torch.Size([1, 1, 20307, 20307]) Position ids shape: torch.Size([1, 20307]) Input IDs shape: torch.Size([1, 20307]) Labels shape: torch.Size([1, 20307]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25850 Attention mask shape: torch.Size([1, 1, 25850, 25850]) Position ids shape: torch.Size([1, 25850]) Input IDs shape: torch.Size([1, 25850]) Labels shape: torch.Size([1, 25850]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31042 Attention mask shape: torch.Size([1, 1, 31042, 31042]) Position ids shape: torch.Size([1, 31042]) Input IDs shape: torch.Size([1, 31042]) Labels shape: torch.Size([1, 31042]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 37159 Attention mask shape: torch.Size([1, 1, 37159, 37159]) Position ids shape: torch.Size([1, 37159]) Input IDs shape: torch.Size([1, 37159]) Labels shape: torch.Size([1, 37159]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7448 Attention mask shape: torch.Size([1, 1, 7448, 7448]) Position ids shape: torch.Size([1, 7448]) Input IDs shape: torch.Size([1, 7448]) Labels shape: torch.Size([1, 7448]) Final batch size: 1, sequence length: 31448 Attention mask shape: torch.Size([1, 1, 31448, 31448]) Position ids shape: torch.Size([1, 31448]) Input IDs shape: torch.Size([1, 31448]) Labels shape: torch.Size([1, 31448]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40605 Attention mask shape: torch.Size([1, 1, 40605, 40605]) Position ids shape: torch.Size([1, 40605]) Input IDs shape: torch.Size([1, 40605]) Labels shape: torch.Size([1, 40605]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 5405 Attention mask shape: torch.Size([1, 1, 5405, 5405]) Position ids shape: torch.Size([1, 5405]) Input IDs shape: torch.Size([1, 5405]) Labels shape: torch.Size([1, 5405]) {'loss': 0.2484, 'grad_norm': 0.1921352944044531, 'learning_rate': 9.549150281252633e-07, 'num_tokens': -inf, 'epoch': 6.62} Final batch size: 1, sequence length: 4641 Attention mask shape: torch.Size([1, 1, 4641, 4641]) Position ids shape: torch.Size([1, 4641]) Input IDs shape: torch.Size([1, 4641]) Labels shape: torch.Size([1, 4641]) Final batch size: 1, sequence length: 4223 Attention mask shape: torch.Size([1, 1, 4223, 4223]) Position ids shape: torch.Size([1, 4223]) Input IDs shape: torch.Size([1, 4223]) Labels shape: torch.Size([1, 4223]) Final batch size: 1, sequence length: 8177 Attention mask shape: torch.Size([1, 1, 8177, 8177]) Position ids shape: torch.Size([1, 8177]) Input IDs shape: torch.Size([1, 8177]) Labels shape: torch.Size([1, 8177]) Final batch size: 1, sequence length: 7795 Attention mask shape: torch.Size([1, 1, 7795, 7795]) Position ids shape: torch.Size([1, 7795]) Input IDs shape: torch.Size([1, 7795]) Labels shape: torch.Size([1, 7795]) Final batch size: 1, sequence length: 12501 Attention mask shape: torch.Size([1, 1, 12501, 12501]) Position ids shape: torch.Size([1, 12501]) Input IDs shape: torch.Size([1, 12501]) Labels shape: torch.Size([1, 12501]) Final batch size: 1, sequence length: 11974 Attention mask shape: torch.Size([1, 1, 11974, 11974]) Position ids shape: torch.Size([1, 11974]) Input IDs shape: torch.Size([1, 11974]) Labels shape: torch.Size([1, 11974]) Final batch size: 1, sequence length: 13186 Attention mask shape: torch.Size([1, 1, 13186, 13186]) Position ids shape: torch.Size([1, 13186]) Input IDs shape: torch.Size([1, 13186]) Labels shape: torch.Size([1, 13186]) Final batch size: 1, sequence length: 5754 Attention mask shape: torch.Size([1, 1, 5754, 5754]) Position ids shape: torch.Size([1, 5754]) Input IDs shape: torch.Size([1, 5754]) Labels shape: torch.Size([1, 5754]) Final batch size: 1, sequence length: 9021 Attention mask shape: torch.Size([1, 1, 9021, 9021]) Position ids shape: torch.Size([1, 9021]) Input IDs shape: torch.Size([1, 9021]) Labels shape: torch.Size([1, 9021]) Final batch size: 1, sequence length: 9027 Attention mask shape: torch.Size([1, 1, 9027, 9027]) Position ids shape: torch.Size([1, 9027]) Input IDs shape: torch.Size([1, 9027]) Labels shape: torch.Size([1, 9027]) Final batch size: 1, sequence length: 15283 Attention mask shape: torch.Size([1, 1, 15283, 15283]) Position ids shape: torch.Size([1, 15283]) Input IDs shape: torch.Size([1, 15283]) Labels shape: torch.Size([1, 15283]) Final batch size: 1, sequence length: 16330 Attention mask shape: torch.Size([1, 1, 16330, 16330]) Position ids shape: torch.Size([1, 16330]) Input IDs shape: torch.Size([1, 16330]) Labels shape: torch.Size([1, 16330]) Final batch size: 1, sequence length: 16496 Attention mask shape: torch.Size([1, 1, 16496, 16496]) Position ids shape: torch.Size([1, 16496]) Input IDs shape: torch.Size([1, 16496]) Labels shape: torch.Size([1, 16496]) Final batch size: 1, sequence length: 13957 Attention mask shape: torch.Size([1, 1, 13957, 13957]) Position ids shape: torch.Size([1, 13957]) Input IDs shape: torch.Size([1, 13957]) Labels shape: torch.Size([1, 13957]) Final batch size: 1, sequence length: 17092 Attention mask shape: torch.Size([1, 1, 17092, 17092]) Position ids shape: torch.Size([1, 17092]) Input IDs shape: torch.Size([1, 17092]) Labels shape: torch.Size([1, 17092]) Final batch size: 1, sequence length: 16366 Attention mask shape: torch.Size([1, 1, 16366, 16366]) Position ids shape: torch.Size([1, 16366]) Input IDs shape: torch.Size([1, 16366]) Labels shape: torch.Size([1, 16366]) Final batch size: 1, sequence length: 16088 Attention mask shape: torch.Size([1, 1, 16088, 16088]) Position ids shape: torch.Size([1, 16088]) Input IDs shape: torch.Size([1, 16088]) Labels shape: torch.Size([1, 16088]) Final batch size: 1, sequence length: 18958 Attention mask shape: torch.Size([1, 1, 18958, 18958]) Position ids shape: torch.Size([1, 18958]) Input IDs shape: torch.Size([1, 18958]) Labels shape: torch.Size([1, 18958]) Final batch size: 1, sequence length: 16104 Attention mask shape: torch.Size([1, 1, 16104, 16104]) Position ids shape: torch.Size([1, 16104]) Input IDs shape: torch.Size([1, 16104]) Labels shape: torch.Size([1, 16104]) Final batch size: 1, sequence length: 16014 Attention mask shape: torch.Size([1, 1, 16014, 16014]) Position ids shape: torch.Size([1, 16014]) Input IDs shape: torch.Size([1, 16014]) Labels shape: torch.Size([1, 16014]) Final batch size: 1, sequence length: 18415 Attention mask shape: torch.Size([1, 1, 18415, 18415]) Position ids shape: torch.Size([1, 18415]) Input IDs shape: torch.Size([1, 18415]) Labels shape: torch.Size([1, 18415]) Final batch size: 1, sequence length: 18034 Attention mask shape: torch.Size([1, 1, 18034, 18034]) Position ids shape: torch.Size([1, 18034]) Input IDs shape: torch.Size([1, 18034]) Labels shape: torch.Size([1, 18034]) Final batch size: 1, sequence length: 19603 Attention mask shape: torch.Size([1, 1, 19603, 19603]) Position ids shape: torch.Size([1, 19603]) Input IDs shape: torch.Size([1, 19603]) Labels shape: torch.Size([1, 19603]) Final batch size: 1, sequence length: 18938 Attention mask shape: torch.Size([1, 1, 18938, 18938]) Position ids shape: torch.Size([1, 18938]) Input IDs shape: torch.Size([1, 18938]) Labels shape: torch.Size([1, 18938]) Final batch size: 1, sequence length: 21132 Attention mask shape: torch.Size([1, 1, 21132, 21132]) Position ids shape: torch.Size([1, 21132]) Input IDs shape: torch.Size([1, 21132]) Labels shape: torch.Size([1, 21132]) Final batch size: 1, sequence length: 14492 Attention mask shape: torch.Size([1, 1, 14492, 14492]) Position ids shape: torch.Size([1, 14492]) Input IDs shape: torch.Size([1, 14492]) Labels shape: torch.Size([1, 14492]) Final batch size: 1, sequence length: 19034 Attention mask shape: torch.Size([1, 1, 19034, 19034]) Position ids shape: torch.Size([1, 19034]) Input IDs shape: torch.Size([1, 19034]) Labels shape: torch.Size([1, 19034]) Final batch size: 1, sequence length: 21705 Attention mask shape: torch.Size([1, 1, 21705, 21705]) Position ids shape: torch.Size([1, 21705]) Input IDs shape: torch.Size([1, 21705]) Labels shape: torch.Size([1, 21705]) Final batch size: 1, sequence length: 23132 Attention mask shape: torch.Size([1, 1, 23132, 23132]) Position ids shape: torch.Size([1, 23132]) Input IDs shape: torch.Size([1, 23132]) Labels shape: torch.Size([1, 23132]) Final batch size: 1, sequence length: 23102 Attention mask shape: torch.Size([1, 1, 23102, 23102]) Position ids shape: torch.Size([1, 23102]) Input IDs shape: torch.Size([1, 23102]) Labels shape: torch.Size([1, 23102]) Final batch size: 1, sequence length: 13468 Attention mask shape: torch.Size([1, 1, 13468, 13468]) Position ids shape: torch.Size([1, 13468]) Input IDs shape: torch.Size([1, 13468]) Labels shape: torch.Size([1, 13468]) Final batch size: 1, sequence length: 24432 Attention mask shape: torch.Size([1, 1, 24432, 24432]) Position ids shape: torch.Size([1, 24432]) Input IDs shape: torch.Size([1, 24432]) Labels shape: torch.Size([1, 24432]) Final batch size: 1, sequence length: 20726 Attention mask shape: torch.Size([1, 1, 20726, 20726]) Position ids shape: torch.Size([1, 20726]) Input IDs shape: torch.Size([1, 20726]) Labels shape: torch.Size([1, 20726]) Final batch size: 1, sequence length: 21404 Attention mask shape: torch.Size([1, 1, 21404, 21404]) Position ids shape: torch.Size([1, 21404]) Input IDs shape: torch.Size([1, 21404]) Labels shape: torch.Size([1, 21404]) Final batch size: 1, sequence length: 22328 Attention mask shape: torch.Size([1, 1, 22328, 22328]) Position ids shape: torch.Size([1, 22328]) Input IDs shape: torch.Size([1, 22328]) Labels shape: torch.Size([1, 22328]) Final batch size: 1, sequence length: 25351 Attention mask shape: torch.Size([1, 1, 25351, 25351]) Position ids shape: torch.Size([1, 25351]) Input IDs shape: torch.Size([1, 25351]) Labels shape: torch.Size([1, 25351]) Final batch size: 1, sequence length: 16590 Attention mask shape: torch.Size([1, 1, 16590, 16590]) Position ids shape: torch.Size([1, 16590]) Input IDs shape: torch.Size([1, 16590]) Labels shape: torch.Size([1, 16590]) Final batch size: 1, sequence length: 23971 Attention mask shape: torch.Size([1, 1, 23971, 23971]) Position ids shape: torch.Size([1, 23971]) Input IDs shape: torch.Size([1, 23971]) Labels shape: torch.Size([1, 23971]) Final batch size: 1, sequence length: 12756 Attention mask shape: torch.Size([1, 1, 12756, 12756]) Position ids shape: torch.Size([1, 12756]) Input IDs shape: torch.Size([1, 12756]) Labels shape: torch.Size([1, 12756]) Final batch size: 1, sequence length: 23894 Attention mask shape: torch.Size([1, 1, 23894, 23894]) Position ids shape: torch.Size([1, 23894]) Input IDs shape: torch.Size([1, 23894]) Labels shape: torch.Size([1, 23894]) Final batch size: 1, sequence length: 26375 Attention mask shape: torch.Size([1, 1, 26375, 26375]) Position ids shape: torch.Size([1, 26375]) Input IDs shape: torch.Size([1, 26375]) Labels shape: torch.Size([1, 26375]) Final batch size: 1, sequence length: 16291 Attention mask shape: torch.Size([1, 1, 16291, 16291]) Position ids shape: torch.Size([1, 16291]) Input IDs shape: torch.Size([1, 16291]) Labels shape: torch.Size([1, 16291]) Final batch size: 1, sequence length: 17514 Attention mask shape: torch.Size([1, 1, 17514, 17514]) Position ids shape: torch.Size([1, 17514]) Input IDs shape: torch.Size([1, 17514]) Labels shape: torch.Size([1, 17514]) Final batch size: 1, sequence length: 27780 Attention mask shape: torch.Size([1, 1, 27780, 27780]) Position ids shape: torch.Size([1, 27780]) Input IDs shape: torch.Size([1, 27780]) Labels shape: torch.Size([1, 27780]) Final batch size: 1, sequence length: 27785 Attention mask shape: torch.Size([1, 1, 27785, 27785]) Position ids shape: torch.Size([1, 27785]) Input IDs shape: torch.Size([1, 27785]) Labels shape: torch.Size([1, 27785]) Final batch size: 1, sequence length: 22771 Attention mask shape: torch.Size([1, 1, 22771, 22771]) Position ids shape: torch.Size([1, 22771]) Input IDs shape: torch.Size([1, 22771]) Labels shape: torch.Size([1, 22771]) Final batch size: 1, sequence length: 13790 Attention mask shape: torch.Size([1, 1, 13790, 13790]) Position ids shape: torch.Size([1, 13790]) Input IDs shape: torch.Size([1, 13790]) Labels shape: torch.Size([1, 13790]) Final batch size: 1, sequence length: 28845 Attention mask shape: torch.Size([1, 1, 28845, 28845]) Position ids shape: torch.Size([1, 28845]) Input IDs shape: torch.Size([1, 28845]) Labels shape: torch.Size([1, 28845]) Final batch size: 1, sequence length: 26766 Attention mask shape: torch.Size([1, 1, 26766, 26766]) Position ids shape: torch.Size([1, 26766]) Input IDs shape: torch.Size([1, 26766]) Labels shape: torch.Size([1, 26766]) Final batch size: 1, sequence length: 25388 Attention mask shape: torch.Size([1, 1, 25388, 25388]) Position ids shape: torch.Size([1, 25388]) Input IDs shape: torch.Size([1, 25388]) Labels shape: torch.Size([1, 25388]) Final batch size: 1, sequence length: 23810 Attention mask shape: torch.Size([1, 1, 23810, 23810]) Position ids shape: torch.Size([1, 23810]) Input IDs shape: torch.Size([1, 23810]) Labels shape: torch.Size([1, 23810]) Final batch size: 1, sequence length: 20817 Attention mask shape: torch.Size([1, 1, 20817, 20817]) Position ids shape: torch.Size([1, 20817]) Final batch size: 1, sequence length: 30165 Input IDs shape: torch.Size([1, 20817]) Attention mask shape: torch.Size([1, 1, 30165, 30165]) Position ids shape: torch.Size([1, 30165]) Input IDs shape: torch.Size([1, 30165]) Labels shape: torch.Size([1, 20817]) Labels shape: torch.Size([1, 30165]) Final batch size: 1, sequence length: 29168 Attention mask shape: torch.Size([1, 1, 29168, 29168]) Position ids shape: torch.Size([1, 29168]) Input IDs shape: torch.Size([1, 29168]) Labels shape: torch.Size([1, 29168]) Final batch size: 1, sequence length: 26596 Attention mask shape: torch.Size([1, 1, 26596, 26596]) Position ids shape: torch.Size([1, 26596]) Input IDs shape: torch.Size([1, 26596]) Labels shape: torch.Size([1, 26596]) Final batch size: 1, sequence length: 32022 Attention mask shape: torch.Size([1, 1, 32022, 32022]) Position ids shape: torch.Size([1, 32022]) Input IDs shape: torch.Size([1, 32022]) Labels shape: torch.Size([1, 32022]) Final batch size: 1, sequence length: 28574 Attention mask shape: torch.Size([1, 1, 28574, 28574]) Position ids shape: torch.Size([1, 28574]) Input IDs shape: torch.Size([1, 28574]) Labels shape: torch.Size([1, 28574]) Final batch size: 1, sequence length: 28412 Attention mask shape: torch.Size([1, 1, 28412, 28412]) Position ids shape: torch.Size([1, 28412]) Input IDs shape: torch.Size([1, 28412]) Labels shape: torch.Size([1, 28412]) Final batch size: 1, sequence length: 29132 Attention mask shape: torch.Size([1, 1, 29132, 29132]) Position ids shape: torch.Size([1, 29132]) Input IDs shape: torch.Size([1, 29132]) Labels shape: torch.Size([1, 29132]) Final batch size: 1, sequence length: 34457 Attention mask shape: torch.Size([1, 1, 34457, 34457]) Position ids shape: torch.Size([1, 34457]) Input IDs shape: torch.Size([1, 34457]) Labels shape: torch.Size([1, 34457]) Final batch size: 1, sequence length: 29830 Attention mask shape: torch.Size([1, 1, 29830, 29830]) Position ids shape: torch.Size([1, 29830]) Input IDs shape: torch.Size([1, 29830]) Labels shape: torch.Size([1, 29830]) Final batch size: 1, sequence length: 18869 Attention mask shape: torch.Size([1, 1, 18869, 18869]) Position ids shape: torch.Size([1, 18869]) Input IDs shape: torch.Size([1, 18869]) Labels shape: torch.Size([1, 18869]) Final batch size: 1, sequence length: 34326 Attention mask shape: torch.Size([1, 1, 34326, 34326]) Position ids shape: torch.Size([1, 34326]) Input IDs shape: torch.Size([1, 34326]) Labels shape: torch.Size([1, 34326]) Final batch size: 1, sequence length: 22043 Attention mask shape: torch.Size([1, 1, 22043, 22043]) Position ids shape: torch.Size([1, 22043]) Input IDs shape: torch.Size([1, 22043]) Labels shape: torch.Size([1, 22043]) Final batch size: 1, sequence length: 24653 Attention mask shape: torch.Size([1, 1, 24653, 24653]) Final batch size: 1, sequence length: 13986 Position ids shape: torch.Size([1, 24653]) Input IDs shape: torch.Size([1, 24653]) Labels shape: torch.Size([1, 24653]) Attention mask shape: torch.Size([1, 1, 13986, 13986]) Position ids shape: torch.Size([1, 13986]) Input IDs shape: torch.Size([1, 13986]) Labels shape: torch.Size([1, 13986]) Final batch size: 1, sequence length: 26465 Attention mask shape: torch.Size([1, 1, 26465, 26465]) Position ids shape: torch.Size([1, 26465]) Input IDs shape: torch.Size([1, 26465]) Labels shape: torch.Size([1, 26465]) Final batch size: 1, sequence length: 35790 Attention mask shape: torch.Size([1, 1, 35790, 35790]) Position ids shape: torch.Size([1, 35790]) Input IDs shape: torch.Size([1, 35790]) Labels shape: torch.Size([1, 35790]) Final batch size: 1, sequence length: 22014 Attention mask shape: torch.Size([1, 1, 22014, 22014]) Position ids shape: torch.Size([1, 22014]) Input IDs shape: torch.Size([1, 22014]) Labels shape: torch.Size([1, 22014]) Final batch size: 1, sequence length: 38617 Attention mask shape: torch.Size([1, 1, 38617, 38617]) Position ids shape: torch.Size([1, 38617]) Input IDs shape: torch.Size([1, 38617]) Labels shape: torch.Size([1, 38617]) Final batch size: 1, sequence length: 37039 Attention mask shape: torch.Size([1, 1, 37039, 37039]) Position ids shape: torch.Size([1, 37039]) Input IDs shape: torch.Size([1, 37039]) Labels shape: torch.Size([1, 37039]) Final batch size: 1, sequence length: 34874 Attention mask shape: torch.Size([1, 1, 34874, 34874]) Position ids shape: torch.Size([1, 34874]) Input IDs shape: torch.Size([1, 34874]) Labels shape: torch.Size([1, 34874]) Final batch size: 1, sequence length: 36794 Attention mask shape: torch.Size([1, 1, 36794, 36794]) Position ids shape: torch.Size([1, 36794]) Input IDs shape: torch.Size([1, 36794]) Labels shape: torch.Size([1, 36794]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19204 Attention mask shape: torch.Size([1, 1, 19204, 19204]) Position ids shape: torch.Size([1, 19204]) Input IDs shape: torch.Size([1, 19204]) Labels shape: torch.Size([1, 19204]) Final batch size: 1, sequence length: 38919 Attention mask shape: torch.Size([1, 1, 38919, 38919]) Position ids shape: torch.Size([1, 38919]) Input IDs shape: torch.Size([1, 38919]) Labels shape: torch.Size([1, 38919]) Final batch size: 1, sequence length: 35868 Attention mask shape: torch.Size([1, 1, 35868, 35868]) Position ids shape: torch.Size([1, 35868]) Input IDs shape: torch.Size([1, 35868]) Labels shape: torch.Size([1, 35868]) Final batch size: 1, sequence length: 23245 Attention mask shape: torch.Size([1, 1, 23245, 23245]) Position ids shape: torch.Size([1, 23245]) Input IDs shape: torch.Size([1, 23245]) Labels shape: torch.Size([1, 23245]) Final batch size: 1, sequence length: 35803 Attention mask shape: torch.Size([1, 1, 35803, 35803]) Position ids shape: torch.Size([1, 35803]) Input IDs shape: torch.Size([1, 35803]) Labels shape: torch.Size([1, 35803]) Final batch size: 1, sequence length: 25923 Attention mask shape: torch.Size([1, 1, 25923, 25923]) Position ids shape: torch.Size([1, 25923]) Input IDs shape: torch.Size([1, 25923]) Labels shape: torch.Size([1, 25923]) Final batch size: 1, sequence length: 19744 Attention mask shape: torch.Size([1, 1, 19744, 19744]) Final batch size: 1, sequence length: 35999Position ids shape: torch.Size([1, 19744]) Input IDs shape: torch.Size([1, 19744]) Labels shape: torch.Size([1, 19744]) Attention mask shape: torch.Size([1, 1, 35999, 35999]) Position ids shape: torch.Size([1, 35999]) Input IDs shape: torch.Size([1, 35999]) Labels shape: torch.Size([1, 35999]) Final batch size: 1, sequence length: 39150 Attention mask shape: torch.Size([1, 1, 39150, 39150]) Position ids shape: torch.Size([1, 39150]) Input IDs shape: torch.Size([1, 39150]) Labels shape: torch.Size([1, 39150]) Final batch size: 1, sequence length: 31506 Attention mask shape: torch.Size([1, 1, 31506, 31506]) Position ids shape: torch.Size([1, 31506]) Input IDs shape: torch.Size([1, 31506]) Labels shape: torch.Size([1, 31506]) Final batch size: 1, sequence length: 22282 Attention mask shape: torch.Size([1, 1, 22282, 22282]) Position ids shape: torch.Size([1, 22282]) Input IDs shape: torch.Size([1, 22282]) Labels shape: torch.Size([1, 22282]) Final batch size: 1, sequence length: 32974 Attention mask shape: torch.Size([1, 1, 32974, 32974]) Position ids shape: torch.Size([1, 32974]) Input IDs shape: torch.Size([1, 32974]) Labels shape: torch.Size([1, 32974]) Final batch size: 1, sequence length: 28002 Attention mask shape: torch.Size([1, 1, 28002, 28002]) Position ids shape: torch.Size([1, 28002]) Input IDs shape: torch.Size([1, 28002]) Labels shape: torch.Size([1, 28002]) Final batch size: 1, sequence length: 28781 Attention mask shape: torch.Size([1, 1, 28781, 28781]) Position ids shape: torch.Size([1, 28781]) Input IDs shape: torch.Size([1, 28781]) Labels shape: torch.Size([1, 28781]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 39919 Attention mask shape: torch.Size([1, 1, 39919, 39919]) Position ids shape: torch.Size([1, 39919]) Input IDs shape: torch.Size([1, 39919]) Labels shape: torch.Size([1, 39919]) Final batch size: 1, sequence length: 40717 Attention mask shape: torch.Size([1, 1, 40717, 40717]) Position ids shape: torch.Size([1, 40717]) Input IDs shape: torch.Size([1, 40717]) Labels shape: torch.Size([1, 40717]) Final batch size: 1, sequence length: 19846 Attention mask shape: torch.Size([1, 1, 19846, 19846]) Position ids shape: torch.Size([1, 19846]) Input IDs shape: torch.Size([1, 19846]) Labels shape: torch.Size([1, 19846]) Final batch size: 1, sequence length: 40065 Attention mask shape: torch.Size([1, 1, 40065, 40065]) Position ids shape: torch.Size([1, 40065]) Input IDs shape: torch.Size([1, 40065]) Labels shape: torch.Size([1, 40065]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31340 Attention mask shape: torch.Size([1, 1, 31340, 31340]) Position ids shape: torch.Size([1, 31340]) Input IDs shape: torch.Size([1, 31340]) Labels shape: torch.Size([1, 31340]) Final batch size: 1, sequence length: 12418 Attention mask shape: torch.Size([1, 1, 12418, 12418]) Position ids shape: torch.Size([1, 12418]) Input IDs shape: torch.Size([1, 12418]) Labels shape: torch.Size([1, 12418]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 18379 Attention mask shape: torch.Size([1, 1, 18379, 18379]) Position ids shape: torch.Size([1, 18379]) Input IDs shape: torch.Size([1, 18379]) Labels shape: torch.Size([1, 18379]) Final batch size: 1, sequence length: 31359 Attention mask shape: torch.Size([1, 1, 31359, 31359]) Position ids shape: torch.Size([1, 31359]) Input IDs shape: torch.Size([1, 31359]) Labels shape: torch.Size([1, 31359]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7364 Attention mask shape: torch.Size([1, 1, 7364, 7364]) Position ids shape: torch.Size([1, 7364]) Input IDs shape: torch.Size([1, 7364]) Labels shape: torch.Size([1, 7364]) Final batch size: 1, sequence length: 11644 Attention mask shape: torch.Size([1, 1, 11644, 11644]) Position ids shape: torch.Size([1, 11644]) Input IDs shape: torch.Size([1, 11644]) Labels shape: torch.Size([1, 11644]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 9309 Attention mask shape: torch.Size([1, 1, 9309, 9309]) Position ids shape: torch.Size([1, 9309]) Input IDs shape: torch.Size([1, 9309]) Labels shape: torch.Size([1, 9309]) Final batch size: 1, sequence length: 25560 Attention mask shape: torch.Size([1, 1, 25560, 25560]) Position ids shape: torch.Size([1, 25560]) Input IDs shape: torch.Size([1, 25560]) Labels shape: torch.Size([1, 25560]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40874 Attention mask shape: torch.Size([1, 1, 40874, 40874]) Position ids shape: torch.Size([1, 40874]) Input IDs shape: torch.Size([1, 40874]) Labels shape: torch.Size([1, 40874]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 26221 Attention mask shape: torch.Size([1, 1, 26221, 26221]) Position ids shape: torch.Size([1, 26221]) Input IDs shape: torch.Size([1, 26221]) Labels shape: torch.Size([1, 26221]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 29526 Attention mask shape: torch.Size([1, 1, 29526, 29526]) Position ids shape: torch.Size([1, 29526]) Input IDs shape: torch.Size([1, 29526]) Labels shape: torch.Size([1, 29526]) Final batch size: 1, sequence length: 17224 Attention mask shape: torch.Size([1, 1, 17224, 17224]) Position ids shape: torch.Size([1, 17224]) Input IDs shape: torch.Size([1, 17224]) Labels shape: torch.Size([1, 17224]) Final batch size: 1, sequence length: 18654 Attention mask shape: torch.Size([1, 1, 18654, 18654]) Position ids shape: torch.Size([1, 18654]) Input IDs shape: torch.Size([1, 18654]) Labels shape: torch.Size([1, 18654]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25820 Attention mask shape: torch.Size([1, 1, 25820, 25820]) Position ids shape: torch.Size([1, 25820]) Input IDs shape: torch.Size([1, 25820]) Labels shape: torch.Size([1, 25820]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30709 Attention mask shape: torch.Size([1, 1, 30709, 30709]) Position ids shape: torch.Size([1, 30709]) Input IDs shape: torch.Size([1, 30709]) Labels shape: torch.Size([1, 30709]) {'loss': 0.2586, 'grad_norm': 0.16998270147193284, 'learning_rate': 8.066471602728804e-07, 'num_tokens': -inf, 'epoch': 6.75} Final batch size: 1, sequence length: 7461 Attention mask shape: torch.Size([1, 1, 7461, 7461]) Position ids shape: torch.Size([1, 7461]) Input IDs shape: torch.Size([1, 7461]) Labels shape: torch.Size([1, 7461]) Final batch size: 1, sequence length: 12279 Attention mask shape: torch.Size([1, 1, 12279, 12279]) Position ids shape: torch.Size([1, 12279]) Input IDs shape: torch.Size([1, 12279]) Labels shape: torch.Size([1, 12279]) Final batch size: 1, sequence length: 10826 Attention mask shape: torch.Size([1, 1, 10826, 10826]) Position ids shape: torch.Size([1, 10826]) Input IDs shape: torch.Size([1, 10826]) Labels shape: torch.Size([1, 10826]) Final batch size: 1, sequence length: 9858 Attention mask shape: torch.Size([1, 1, 9858, 9858]) Position ids shape: torch.Size([1, 9858]) Input IDs shape: torch.Size([1, 9858]) Labels shape: torch.Size([1, 9858]) Final batch size: 1, sequence length: 13427 Attention mask shape: torch.Size([1, 1, 13427, 13427]) Position ids shape: torch.Size([1, 13427]) Input IDs shape: torch.Size([1, 13427]) Labels shape: torch.Size([1, 13427]) Final batch size: 1, sequence length: 12096 Attention mask shape: torch.Size([1, 1, 12096, 12096]) Position ids shape: torch.Size([1, 12096]) Input IDs shape: torch.Size([1, 12096]) Labels shape: torch.Size([1, 12096]) Final batch size: 1, sequence length: 12960 Attention mask shape: torch.Size([1, 1, 12960, 12960]) Position ids shape: torch.Size([1, 12960]) Input IDs shape: torch.Size([1, 12960]) Labels shape: torch.Size([1, 12960]) Final batch size: 1, sequence length: 15933 Attention mask shape: torch.Size([1, 1, 15933, 15933]) Position ids shape: torch.Size([1, 15933]) Input IDs shape: torch.Size([1, 15933]) Labels shape: torch.Size([1, 15933]) Final batch size: 1, sequence length: 12622 Attention mask shape: torch.Size([1, 1, 12622, 12622]) Position ids shape: torch.Size([1, 12622]) Input IDs shape: torch.Size([1, 12622]) Labels shape: torch.Size([1, 12622]) Final batch size: 1, sequence length: 18214 Attention mask shape: torch.Size([1, 1, 18214, 18214]) Position ids shape: torch.Size([1, 18214]) Input IDs shape: torch.Size([1, 18214]) Labels shape: torch.Size([1, 18214]) Final batch size: 1, sequence length: 13550 Attention mask shape: torch.Size([1, 1, 13550, 13550]) Position ids shape: torch.Size([1, 13550]) Input IDs shape: torch.Size([1, 13550]) Labels shape: torch.Size([1, 13550]) Final batch size: 1, sequence length: 15673 Attention mask shape: torch.Size([1, 1, 15673, 15673]) Position ids shape: torch.Size([1, 15673]) Input IDs shape: torch.Size([1, 15673]) Labels shape: torch.Size([1, 15673]) Final batch size: 1, sequence length: 17861 Attention mask shape: torch.Size([1, 1, 17861, 17861]) Position ids shape: torch.Size([1, 17861]) Input IDs shape: torch.Size([1, 17861]) Labels shape: torch.Size([1, 17861]) Final batch size: 1, sequence length: 13789 Attention mask shape: torch.Size([1, 1, 13789, 13789]) Position ids shape: torch.Size([1, 13789]) Input IDs shape: torch.Size([1, 13789]) Labels shape: torch.Size([1, 13789]) Final batch size: 1, sequence length: 16927 Attention mask shape: torch.Size([1, 1, 16927, 16927]) Position ids shape: torch.Size([1, 16927]) Input IDs shape: torch.Size([1, 16927]) Labels shape: torch.Size([1, 16927]) Final batch size: 1, sequence length: 19899 Attention mask shape: torch.Size([1, 1, 19899, 19899]) Position ids shape: torch.Size([1, 19899]) Input IDs shape: torch.Size([1, 19899]) Labels shape: torch.Size([1, 19899]) Final batch size: 1, sequence length: 16450 Attention mask shape: torch.Size([1, 1, 16450, 16450]) Position ids shape: torch.Size([1, 16450]) Input IDs shape: torch.Size([1, 16450]) Labels shape: torch.Size([1, 16450]) Final batch size: 1, sequence length: 17245 Attention mask shape: torch.Size([1, 1, 17245, 17245]) Position ids shape: torch.Size([1, 17245]) Input IDs shape: torch.Size([1, 17245]) Labels shape: torch.Size([1, 17245]) Final batch size: 1, sequence length: 17934 Attention mask shape: torch.Size([1, 1, 17934, 17934]) Position ids shape: torch.Size([1, 17934]) Input IDs shape: torch.Size([1, 17934]) Labels shape: torch.Size([1, 17934]) Final batch size: 1, sequence length: 23004 Attention mask shape: torch.Size([1, 1, 23004, 23004]) Position ids shape: torch.Size([1, 23004]) Input IDs shape: torch.Size([1, 23004]) Labels shape: torch.Size([1, 23004]) Final batch size: 1, sequence length: 20065 Attention mask shape: torch.Size([1, 1, 20065, 20065]) Position ids shape: torch.Size([1, 20065]) Input IDs shape: torch.Size([1, 20065]) Labels shape: torch.Size([1, 20065]) Final batch size: 1, sequence length: 19953 Attention mask shape: torch.Size([1, 1, 19953, 19953]) Position ids shape: torch.Size([1, 19953]) Input IDs shape: torch.Size([1, 19953]) Labels shape: torch.Size([1, 19953]) Final batch size: 1, sequence length: 21091 Attention mask shape: torch.Size([1, 1, 21091, 21091]) Position ids shape: torch.Size([1, 21091]) Input IDs shape: torch.Size([1, 21091]) Labels shape: torch.Size([1, 21091]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 21675 Attention mask shape: torch.Size([1, 1, 21675, 21675]) Position ids shape: torch.Size([1, 21675]) Input IDs shape: torch.Size([1, 21675]) Labels shape: torch.Size([1, 21675]) Final batch size: 1, sequence length: 16036 Attention mask shape: torch.Size([1, 1, 16036, 16036]) Position ids shape: torch.Size([1, 16036]) Input IDs shape: torch.Size([1, 16036]) Labels shape: torch.Size([1, 16036]) Final batch size: 1, sequence length: 20292 Attention mask shape: torch.Size([1, 1, 20292, 20292]) Position ids shape: torch.Size([1, 20292]) Input IDs shape: torch.Size([1, 20292]) Labels shape: torch.Size([1, 20292]) Final batch size: 1, sequence length: 12728 Attention mask shape: torch.Size([1, 1, 12728, 12728]) Position ids shape: torch.Size([1, 12728]) Input IDs shape: torch.Size([1, 12728]) Labels shape: torch.Size([1, 12728]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 23896 Attention mask shape: torch.Size([1, 1, 23896, 23896]) Position ids shape: torch.Size([1, 23896]) Input IDs shape: torch.Size([1, 23896]) Labels shape: torch.Size([1, 23896]) Final batch size: 1, sequence length: 25600 Attention mask shape: torch.Size([1, 1, 25600, 25600]) Position ids shape: torch.Size([1, 25600]) Input IDs shape: torch.Size([1, 25600]) Labels shape: torch.Size([1, 25600]) Final batch size: 1, sequence length: 19187 Attention mask shape: torch.Size([1, 1, 19187, 19187]) Position ids shape: torch.Size([1, 19187]) Input IDs shape: torch.Size([1, 19187]) Labels shape: torch.Size([1, 19187]) Final batch size: 1, sequence length: 21586 Attention mask shape: torch.Size([1, 1, 21586, 21586]) Position ids shape: torch.Size([1, 21586]) Input IDs shape: torch.Size([1, 21586]) Labels shape: torch.Size([1, 21586]) Final batch size: 1, sequence length: 25035 Attention mask shape: torch.Size([1, 1, 25035, 25035]) Position ids shape: torch.Size([1, 25035]) Input IDs shape: torch.Size([1, 25035]) Labels shape: torch.Size([1, 25035]) Final batch size: 1, sequence length: 25176 Attention mask shape: torch.Size([1, 1, 25176, 25176]) Position ids shape: torch.Size([1, 25176]) Input IDs shape: torch.Size([1, 25176]) Labels shape: torch.Size([1, 25176]) Final batch size: 1, sequence length: 26316 Attention mask shape: torch.Size([1, 1, 26316, 26316]) Position ids shape: torch.Size([1, 26316]) Input IDs shape: torch.Size([1, 26316]) Labels shape: torch.Size([1, 26316]) Final batch size: 1, sequence length: 18832 Attention mask shape: torch.Size([1, 1, 18832, 18832]) Position ids shape: torch.Size([1, 18832]) Input IDs shape: torch.Size([1, 18832]) Labels shape: torch.Size([1, 18832]) Final batch size: 1, sequence length: 24956 Attention mask shape: torch.Size([1, 1, 24956, 24956]) Position ids shape: torch.Size([1, 24956]) Input IDs shape: torch.Size([1, 24956]) Labels shape: torch.Size([1, 24956]) Final batch size: 1, sequence length: 22775 Attention mask shape: torch.Size([1, 1, 22775, 22775]) Position ids shape: torch.Size([1, 22775]) Input IDs shape: torch.Size([1, 22775]) Labels shape: torch.Size([1, 22775]) Final batch size: 1, sequence length: 25435 Attention mask shape: torch.Size([1, 1, 25435, 25435]) Position ids shape: torch.Size([1, 25435]) Input IDs shape: torch.Size([1, 25435]) Labels shape: torch.Size([1, 25435]) Final batch size: 1, sequence length: 14212 Attention mask shape: torch.Size([1, 1, 14212, 14212]) Position ids shape: torch.Size([1, 14212]) Input IDs shape: torch.Size([1, 14212]) Labels shape: torch.Size([1, 14212]) Final batch size: 1, sequence length: 24633 Attention mask shape: torch.Size([1, 1, 24633, 24633]) Position ids shape: torch.Size([1, 24633]) Input IDs shape: torch.Size([1, 24633]) Labels shape: torch.Size([1, 24633]) Final batch size: 1, sequence length: 22778 Attention mask shape: torch.Size([1, 1, 22778, 22778]) Position ids shape: torch.Size([1, 22778]) Input IDs shape: torch.Size([1, 22778]) Labels shape: torch.Size([1, 22778]) Final batch size: 1, sequence length: 3010 Attention mask shape: torch.Size([1, 1, 3010, 3010]) Position ids shape: torch.Size([1, 3010]) Input IDs shape: torch.Size([1, 3010]) Labels shape: torch.Size([1, 3010]) Final batch size: 1, sequence length: 26247 Attention mask shape: torch.Size([1, 1, 26247, 26247]) Position ids shape: torch.Size([1, 26247]) Input IDs shape: torch.Size([1, 26247]) Labels shape: torch.Size([1, 26247]) Final batch size: 1, sequence length: 20582 Attention mask shape: torch.Size([1, 1, 20582, 20582]) Position ids shape: torch.Size([1, 20582]) Input IDs shape: torch.Size([1, 20582]) Labels shape: torch.Size([1, 20582]) Final batch size: 1, sequence length: 24808 Attention mask shape: torch.Size([1, 1, 24808, 24808]) Position ids shape: torch.Size([1, 24808]) Input IDs shape: torch.Size([1, 24808]) Labels shape: torch.Size([1, 24808]) Final batch size: 1, sequence length: 26454 Attention mask shape: torch.Size([1, 1, 26454, 26454]) Position ids shape: torch.Size([1, 26454]) Input IDs shape: torch.Size([1, 26454]) Labels shape: torch.Size([1, 26454]) Final batch size: 1, sequence length: 26500 Attention mask shape: torch.Size([1, 1, 26500, 26500]) Position ids shape: torch.Size([1, 26500]) Input IDs shape: torch.Size([1, 26500]) Labels shape: torch.Size([1, 26500]) Final batch size: 1, sequence length: 28926 Attention mask shape: torch.Size([1, 1, 28926, 28926]) Position ids shape: torch.Size([1, 28926]) Input IDs shape: torch.Size([1, 28926]) Labels shape: torch.Size([1, 28926]) Final batch size: 1, sequence length: 24927 Attention mask shape: torch.Size([1, 1, 24927, 24927]) Position ids shape: torch.Size([1, 24927]) Input IDs shape: torch.Size([1, 24927]) Labels shape: torch.Size([1, 24927]) Final batch size: 1, sequence length: 17763 Attention mask shape: torch.Size([1, 1, 17763, 17763]) Position ids shape: torch.Size([1, 17763]) Input IDs shape: torch.Size([1, 17763]) Labels shape: torch.Size([1, 17763]) Final batch size: 1, sequence length: 27222 Attention mask shape: torch.Size([1, 1, 27222, 27222]) Position ids shape: torch.Size([1, 27222]) Input IDs shape: torch.Size([1, 27222]) Labels shape: torch.Size([1, 27222]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 25325 Attention mask shape: torch.Size([1, 1, 25325, 25325]) Position ids shape: torch.Size([1, 25325]) Input IDs shape: torch.Size([1, 25325]) Labels shape: torch.Size([1, 25325]) Final batch size: 1, sequence length: 24965 Attention mask shape: torch.Size([1, 1, 24965, 24965]) Position ids shape: torch.Size([1, 24965]) Input IDs shape: torch.Size([1, 24965]) Labels shape: torch.Size([1, 24965]) Final batch size: 1, sequence length: 28552 Attention mask shape: torch.Size([1, 1, 28552, 28552]) Position ids shape: torch.Size([1, 28552]) Input IDs shape: torch.Size([1, 28552]) Labels shape: torch.Size([1, 28552]) Final batch size: 1, sequence length: 13946 Attention mask shape: torch.Size([1, 1, 13946, 13946]) Position ids shape: torch.Size([1, 13946]) Input IDs shape: torch.Size([1, 13946]) Labels shape: torch.Size([1, 13946]) Final batch size: 1, sequence length: 16714 Attention mask shape: torch.Size([1, 1, 16714, 16714]) Position ids shape: torch.Size([1, 16714]) Input IDs shape: torch.Size([1, 16714]) Labels shape: torch.Size([1, 16714]) Final batch size: 1, sequence length: 13453 Attention mask shape: torch.Size([1, 1, 13453, 13453]) Position ids shape: torch.Size([1, 13453]) Input IDs shape: torch.Size([1, 13453]) Labels shape: torch.Size([1, 13453]) Final batch size: 1, sequence length: 28644 Attention mask shape: torch.Size([1, 1, 28644, 28644]) Position ids shape: torch.Size([1, 28644]) Input IDs shape: torch.Size([1, 28644]) Labels shape: torch.Size([1, 28644]) Final batch size: 1, sequence length: 6978 Attention mask shape: torch.Size([1, 1, 6978, 6978]) Position ids shape: torch.Size([1, 6978]) Input IDs shape: torch.Size([1, 6978]) Labels shape: torch.Size([1, 6978]) Final batch size: 1, sequence length: 32101 Attention mask shape: torch.Size([1, 1, 32101, 32101]) Position ids shape: torch.Size([1, 32101]) Input IDs shape: torch.Size([1, 32101]) Labels shape: torch.Size([1, 32101]) Final batch size: 1, sequence length: 25979 Attention mask shape: torch.Size([1, 1, 25979, 25979]) Position ids shape: torch.Size([1, 25979]) Input IDs shape: torch.Size([1, 25979]) Labels shape: torch.Size([1, 25979]) Final batch size: 1, sequence length: 12426 Attention mask shape: torch.Size([1, 1, 12426, 12426]) Position ids shape: torch.Size([1, 12426]) Input IDs shape: torch.Size([1, 12426]) Labels shape: torch.Size([1, 12426]) Final batch size: 1, sequence length: 32513 Attention mask shape: torch.Size([1, 1, 32513, 32513]) Position ids shape: torch.Size([1, 32513]) Input IDs shape: torch.Size([1, 32513]) Labels shape: torch.Size([1, 32513]) Final batch size: 1, sequence length: 32100 Attention mask shape: torch.Size([1, 1, 32100, 32100]) Position ids shape: torch.Size([1, 32100]) Input IDs shape: torch.Size([1, 32100]) Labels shape: torch.Size([1, 32100]) Final batch size: 1, sequence length: 16571 Attention mask shape: torch.Size([1, 1, 16571, 16571]) Position ids shape: torch.Size([1, 16571]) Input IDs shape: torch.Size([1, 16571]) Labels shape: torch.Size([1, 16571]) Final batch size: 1, sequence length: 30635 Attention mask shape: torch.Size([1, 1, 30635, 30635]) Position ids shape: torch.Size([1, 30635]) Input IDs shape: torch.Size([1, 30635]) Labels shape: torch.Size([1, 30635]) Final batch size: 1, sequence length: 35058 Attention mask shape: torch.Size([1, 1, 35058, 35058]) Position ids shape: torch.Size([1, 35058]) Input IDs shape: torch.Size([1, 35058]) Labels shape: torch.Size([1, 35058]) Final batch size: 1, sequence length: 17049 Attention mask shape: torch.Size([1, 1, 17049, 17049]) Position ids shape: torch.Size([1, 17049]) Input IDs shape: torch.Size([1, 17049]) Labels shape: torch.Size([1, 17049]) Final batch size: 1, sequence length: 37391 Attention mask shape: torch.Size([1, 1, 37391, 37391]) Position ids shape: torch.Size([1, 37391]) Input IDs shape: torch.Size([1, 37391]) Labels shape: torch.Size([1, 37391]) Final batch size: 1, sequence length: 28377 Attention mask shape: torch.Size([1, 1, 28377, 28377]) Position ids shape: torch.Size([1, 28377]) Input IDs shape: torch.Size([1, 28377]) Labels shape: torch.Size([1, 28377]) Final batch size: 1, sequence length: 31208 Attention mask shape: torch.Size([1, 1, 31208, 31208]) Position ids shape: torch.Size([1, 31208]) Input IDs shape: torch.Size([1, 31208]) Labels shape: torch.Size([1, 31208]) Final batch size: 1, sequence length: 24676 Attention mask shape: torch.Size([1, 1, 24676, 24676]) Position ids shape: torch.Size([1, 24676]) Input IDs shape: torch.Size([1, 24676]) Labels shape: torch.Size([1, 24676]) Final batch size: 1, sequence length: 37381 Attention mask shape: torch.Size([1, 1, 37381, 37381]) Position ids shape: torch.Size([1, 37381]) Input IDs shape: torch.Size([1, 37381]) Labels shape: torch.Size([1, 37381]) Final batch size: 1, sequence length: 31662 Attention mask shape: torch.Size([1, 1, 31662, 31662]) Position ids shape: torch.Size([1, 31662]) Input IDs shape: torch.Size([1, 31662]) Labels shape: torch.Size([1, 31662]) Final batch size: 1, sequence length: 35014 Attention mask shape: torch.Size([1, 1, 35014, 35014]) Position ids shape: torch.Size([1, 35014]) Input IDs shape: torch.Size([1, 35014]) Labels shape: torch.Size([1, 35014]) Final batch size: 1, sequence length: 21290 Attention mask shape: torch.Size([1, 1, 21290, 21290]) Position ids shape: torch.Size([1, 21290]) Input IDs shape: torch.Size([1, 21290]) Labels shape: torch.Size([1, 21290]) Final batch size: 1, sequence length: 27021 Attention mask shape: torch.Size([1, 1, 27021, 27021]) Position ids shape: torch.Size([1, 27021]) Input IDs shape: torch.Size([1, 27021]) Labels shape: torch.Size([1, 27021]) Final batch size: 1, sequence length: 25774 Attention mask shape: torch.Size([1, 1, 25774, 25774]) Position ids shape: torch.Size([1, 25774]) Input IDs shape: torch.Size([1, 25774]) Labels shape: torch.Size([1, 25774]) Final batch size: 1, sequence length: 26789 Attention mask shape: torch.Size([1, 1, 26789, 26789]) Position ids shape: torch.Size([1, 26789]) Input IDs shape: torch.Size([1, 26789]) Labels shape: torch.Size([1, 26789]) Final batch size: 1, sequence length: 25147 Attention mask shape: torch.Size([1, 1, 25147, 25147]) Position ids shape: torch.Size([1, 25147]) Input IDs shape: torch.Size([1, 25147]) Labels shape: torch.Size([1, 25147]) Final batch size: 1, sequence length: 36047 Attention mask shape: torch.Size([1, 1, 36047, 36047]) Position ids shape: torch.Size([1, 36047]) Input IDs shape: torch.Size([1, 36047]) Labels shape: torch.Size([1, 36047]) Final batch size: 1, sequence length: 39474 Attention mask shape: torch.Size([1, 1, 39474, 39474]) Position ids shape: torch.Size([1, 39474]) Input IDs shape: torch.Size([1, 39474]) Labels shape: torch.Size([1, 39474]) Final batch size: 1, sequence length: 25219 Attention mask shape: torch.Size([1, 1, 25219, 25219]) Position ids shape: torch.Size([1, 25219]) Input IDs shape: torch.Size([1, 25219]) Labels shape: torch.Size([1, 25219]) Final batch size: 1, sequence length: 30259 Attention mask shape: torch.Size([1, 1, 30259, 30259]) Position ids shape: torch.Size([1, 30259]) Input IDs shape: torch.Size([1, 30259]) Labels shape: torch.Size([1, 30259]) Final batch size: 1, sequence length: 16816 Attention mask shape: torch.Size([1, 1, 16816, 16816]) Position ids shape: torch.Size([1, 16816]) Input IDs shape: torch.Size([1, 16816]) Labels shape: torch.Size([1, 16816]) Final batch size: 1, sequence length: 38391 Attention mask shape: torch.Size([1, 1, 38391, 38391]) Position ids shape: torch.Size([1, 38391]) Input IDs shape: torch.Size([1, 38391]) Labels shape: torch.Size([1, 38391]) Final batch size: 1, sequence length: 22710 Attention mask shape: torch.Size([1, 1, 22710, 22710]) Position ids shape: torch.Size([1, 22710]) Input IDs shape: torch.Size([1, 22710]) Labels shape: torch.Size([1, 22710]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 26387 Attention mask shape: torch.Size([1, 1, 26387, 26387]) Position ids shape: torch.Size([1, 26387]) Input IDs shape: torch.Size([1, 26387]) Labels shape: torch.Size([1, 26387]) Final batch size: 1, sequence length: 31447 Attention mask shape: torch.Size([1, 1, 31447, 31447]) Position ids shape: torch.Size([1, 31447]) Input IDs shape: torch.Size([1, 31447]) Labels shape: torch.Size([1, 31447]) Final batch size: 1, sequence length: 33621 Attention mask shape: torch.Size([1, 1, 33621, 33621]) Position ids shape: torch.Size([1, 33621]) Input IDs shape: torch.Size([1, 33621]) Labels shape: torch.Size([1, 33621]) Final batch size: 1, sequence length: 35261 Attention mask shape: torch.Size([1, 1, 35261, 35261]) Position ids shape: torch.Size([1, 35261]) Input IDs shape: torch.Size([1, 35261]) Labels shape: torch.Size([1, 35261]) Final batch size: 1, sequence length: 38577 Attention mask shape: torch.Size([1, 1, 38577, 38577]) Position ids shape: torch.Size([1, 38577]) Input IDs shape: torch.Size([1, 38577]) Labels shape: torch.Size([1, 38577]) Final batch size: 1, sequence length: 37720 Attention mask shape: torch.Size([1, 1, 37720, 37720]) Position ids shape: torch.Size([1, 37720]) Input IDs shape: torch.Size([1, 37720]) Labels shape: torch.Size([1, 37720]) Final batch size: 1, sequence length: 21321 Attention mask shape: torch.Size([1, 1, 21321, 21321]) Position ids shape: torch.Size([1, 21321]) Input IDs shape: torch.Size([1, 21321]) Labels shape: torch.Size([1, 21321]) Final batch size: 1, sequence length: 39955 Attention mask shape: torch.Size([1, 1, 39955, 39955]) Position ids shape: torch.Size([1, 39955]) Input IDs shape: torch.Size([1, 39955]) Labels shape: torch.Size([1, 39955]) Final batch size: 1, sequence length: 19441 Attention mask shape: torch.Size([1, 1, 19441, 19441]) Position ids shape: torch.Size([1, 19441]) Input IDs shape: torch.Size([1, 19441]) Labels shape: torch.Size([1, 19441]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40590 Attention mask shape: torch.Size([1, 1, 40590, 40590]) Position ids shape: torch.Size([1, 40590]) Input IDs shape: torch.Size([1, 40590]) Labels shape: torch.Size([1, 40590]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28448 Attention mask shape: torch.Size([1, 1, 28448, 28448]) Position ids shape: torch.Size([1, 28448]) Input IDs shape: torch.Size([1, 28448]) Labels shape: torch.Size([1, 28448]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32129 Attention mask shape: torch.Size([1, 1, 32129, 32129]) Position ids shape: torch.Size([1, 32129]) Input IDs shape: torch.Size([1, 32129]) Labels shape: torch.Size([1, 32129]) Final batch size: 1, sequence length: 36356 Attention mask shape: torch.Size([1, 1, 36356, 36356]) Position ids shape: torch.Size([1, 36356]) Input IDs shape: torch.Size([1, 36356]) Labels shape: torch.Size([1, 36356]) Final batch size: 1, sequence length: 14136 Attention mask shape: torch.Size([1, 1, 14136, 14136]) Position ids shape: torch.Size([1, 14136]) Input IDs shape: torch.Size([1, 14136]) Labels shape: torch.Size([1, 14136]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32079 Attention mask shape: torch.Size([1, 1, 32079, 32079]) Position ids shape: torch.Size([1, 32079]) Input IDs shape: torch.Size([1, 32079]) Labels shape: torch.Size([1, 32079]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33014 Attention mask shape: torch.Size([1, 1, 33014, 33014]) Position ids shape: torch.Size([1, 33014]) Input IDs shape: torch.Size([1, 33014]) Labels shape: torch.Size([1, 33014]) Final batch size: 1, sequence length: 27371 Attention mask shape: torch.Size([1, 1, 27371, 27371]) Position ids shape: torch.Size([1, 27371]) Input IDs shape: torch.Size([1, 27371]) Labels shape: torch.Size([1, 27371]) Final batch size: 1, sequence length: 24623 Attention mask shape: torch.Size([1, 1, 24623, 24623]) Position ids shape: torch.Size([1, 24623]) Input IDs shape: torch.Size([1, 24623]) Labels shape: torch.Size([1, 24623]) Final batch size: 1, sequence length: 30712 Attention mask shape: torch.Size([1, 1, 30712, 30712]) Position ids shape: torch.Size([1, 30712]) Input IDs shape: torch.Size([1, 30712]) Labels shape: torch.Size([1, 30712]) Final batch size: 1, sequence length: 19027 Attention mask shape: torch.Size([1, 1, 19027, 19027]) Position ids shape: torch.Size([1, 19027]) Input IDs shape: torch.Size([1, 19027]) Labels shape: torch.Size([1, 19027]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 23208 Attention mask shape: torch.Size([1, 1, 23208, 23208]) Position ids shape: torch.Size([1, 23208]) Input IDs shape: torch.Size([1, 23208]) Labels shape: torch.Size([1, 23208]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2552, 'grad_norm': 0.1521675624479113, 'learning_rate': 6.698729810778065e-07, 'num_tokens': -inf, 'epoch': 6.88} Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 10507 Attention mask shape: torch.Size([1, 1, 10507, 10507]) Position ids shape: torch.Size([1, 10507]) Input IDs shape: torch.Size([1, 10507]) Labels shape: torch.Size([1, 10507]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 26271 Attention mask shape: torch.Size([1, 1, 26271, 26271]) Position ids shape: torch.Size([1, 26271]) Input IDs shape: torch.Size([1, 26271]) Labels shape: torch.Size([1, 26271]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 31381 Attention mask shape: torch.Size([1, 1, 31381, 31381]) Position ids shape: torch.Size([1, 31381]) Input IDs shape: torch.Size([1, 31381]) Labels shape: torch.Size([1, 31381]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 39519 Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2467, 'grad_norm': 0.16013139776856344, 'learning_rate': 5.449673790581611e-07, 'num_tokens': -inf, 'epoch': 7.0} Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 6362 Attention mask shape: torch.Size([1, 1, 6362, 6362]) Position ids shape: torch.Size([1, 6362]) Input IDs shape: torch.Size([1, 6362]) Labels shape: torch.Size([1, 6362]) Final batch size: 1, sequence length: 10523 Attention mask shape: torch.Size([1, 1, 10523, 10523]) Position ids shape: torch.Size([1, 10523]) Input IDs shape: torch.Size([1, 10523]) Labels shape: torch.Size([1, 10523]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 12385 Attention mask shape: torch.Size([1, 1, 12385, 12385]) Position ids shape: torch.Size([1, 12385]) Input IDs shape: torch.Size([1, 12385]) Labels shape: torch.Size([1, 12385]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 18410 Attention mask shape: torch.Size([1, 1, 18410, 18410]) Position ids shape: torch.Size([1, 18410]) Input IDs shape: torch.Size([1, 18410]) Labels shape: torch.Size([1, 18410]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 22946 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Final batch size: 1, sequence length: 18098 Attention mask shape: torch.Size([1, 1, 18098, 18098]) Position ids shape: torch.Size([1, 18098]) Input IDs shape: torch.Size([1, 18098]) Labels shape: torch.Size([1, 18098]) Final batch size: 1, sequence length: 12656 Attention mask shape: torch.Size([1, 1, 12656, 12656]) Position ids shape: torch.Size([1, 12656]) Input IDs shape: torch.Size([1, 12656]) Labels shape: torch.Size([1, 12656]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 17299 Attention mask shape: torch.Size([1, 1, 17299, 17299]) Position ids shape: torch.Size([1, 17299]) Input IDs shape: torch.Size([1, 17299]) Labels shape: torch.Size([1, 17299]) Final batch size: 1, sequence length: 9091 Attention mask shape: torch.Size([1, 1, 9091, 9091]) Position ids shape: torch.Size([1, 9091]) Input IDs shape: torch.Size([1, 9091]) Labels shape: torch.Size([1, 9091]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 24365 Attention mask shape: torch.Size([1, 1, 24365, 24365]) Position ids shape: torch.Size([1, 24365]) Input IDs shape: torch.Size([1, 24365]) Labels shape: torch.Size([1, 24365]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 18060 Attention mask shape: torch.Size([1, 1, 18060, 18060]) Position ids shape: torch.Size([1, 18060]) Input IDs shape: torch.Size([1, 18060]) Labels shape: torch.Size([1, 18060]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 11819 Attention mask shape: torch.Size([1, 1, 11819, 11819]) Position ids shape: torch.Size([1, 11819]) Input IDs shape: torch.Size([1, 11819]) Labels shape: torch.Size([1, 11819]) Final batch size: 1, sequence length: 6948 Attention mask shape: torch.Size([1, 1, 6948, 6948]) Position ids shape: torch.Size([1, 6948]) Input IDs shape: torch.Size([1, 6948]) Labels shape: torch.Size([1, 6948]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 28841 Attention mask shape: torch.Size([1, 1, 28841, 28841]) Position ids shape: torch.Size([1, 28841]) Input IDs shape: torch.Size([1, 28841]) Labels shape: torch.Size([1, 28841]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 26976 Attention mask shape: torch.Size([1, 1, 26976, 26976]) Position ids shape: torch.Size([1, 26976]) Input IDs shape: torch.Size([1, 26976]) Labels shape: torch.Size([1, 26976]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 18915 Attention mask shape: torch.Size([1, 1, 18915, 18915]) Position ids shape: torch.Size([1, 18915]) Input IDs shape: torch.Size([1, 18915]) Labels shape: torch.Size([1, 18915]) Final batch size: 1, sequence length: 13215 Attention mask shape: torch.Size([1, 1, 13215, 13215]) Position ids shape: torch.Size([1, 13215]) Input IDs shape: torch.Size([1, 13215]) Labels shape: torch.Size([1, 13215]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 27541 Attention mask shape: torch.Size([1, 1, 27541, 27541]) Position ids shape: torch.Size([1, 27541]) Input IDs shape: torch.Size([1, 27541]) Labels shape: torch.Size([1, 27541]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 24880 Attention mask shape: torch.Size([1, 1, 24880, 24880]) Position ids shape: torch.Size([1, 24880]) Input IDs shape: torch.Size([1, 24880]) Labels shape: torch.Size([1, 24880]) Final batch size: 1, sequence length: 25172 Attention mask shape: torch.Size([1, 1, 25172, 25172]) Position ids shape: torch.Size([1, 25172]) Input IDs shape: torch.Size([1, 25172]) Labels shape: torch.Size([1, 25172]) Final batch size: 1, sequence length: 29113 Attention mask shape: torch.Size([1, 1, 29113, 29113]) Position ids shape: torch.Size([1, 29113]) Input IDs shape: torch.Size([1, 29113]) Labels shape: torch.Size([1, 29113]) Final batch size: 1, sequence length: 7722 Attention mask shape: torch.Size([1, 1, 7722, 7722]) Position ids shape: torch.Size([1, 7722]) Input IDs shape: torch.Size([1, 7722]) Labels shape: torch.Size([1, 7722]) Final batch size: 1, sequence length: 21491 Attention mask shape: torch.Size([1, 1, 21491, 21491]) Position ids shape: torch.Size([1, 21491]) Input IDs shape: torch.Size([1, 21491]) Labels shape: torch.Size([1, 21491]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 21596 Attention mask shape: torch.Size([1, 1, 21596, 21596]) Position ids shape: torch.Size([1, 21596]) Input IDs shape: torch.Size([1, 21596]) Labels shape: torch.Size([1, 21596]) Final batch size: 1, sequence length: 21450 Attention mask shape: torch.Size([1, 1, 21450, 21450]) Position ids shape: torch.Size([1, 21450]) Input IDs shape: torch.Size([1, 21450]) Labels shape: torch.Size([1, 21450]) Final batch size: 1, sequence length: 29875 Attention mask shape: torch.Size([1, 1, 29875, 29875]) Position ids shape: torch.Size([1, 29875]) Input IDs shape: torch.Size([1, 29875]) Labels shape: torch.Size([1, 29875]) Final batch size: 1, sequence length: 20363 Attention mask shape: torch.Size([1, 1, 20363, 20363]) Position ids shape: torch.Size([1, 20363]) Input IDs shape: torch.Size([1, 20363]) Labels shape: torch.Size([1, 20363]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 14009 Attention mask shape: torch.Size([1, 1, 14009, 14009]) Position ids shape: torch.Size([1, 14009]) Input IDs shape: torch.Size([1, 14009]) Labels shape: torch.Size([1, 14009]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 20827 Attention mask shape: torch.Size([1, 1, 20827, 20827]) Position ids shape: torch.Size([1, 20827]) Input IDs shape: torch.Size([1, 20827]) Labels shape: torch.Size([1, 20827]) Final batch size: 1, sequence length: 30623 Attention mask shape: torch.Size([1, 1, 30623, 30623]) Position ids shape: torch.Size([1, 30623]) Input IDs shape: torch.Size([1, 30623]) Labels shape: torch.Size([1, 30623]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 17400 Attention mask shape: torch.Size([1, 1, 17400, 17400]) Position ids shape: torch.Size([1, 17400]) Input IDs shape: torch.Size([1, 17400]) Labels shape: torch.Size([1, 17400]) Final batch size: 1, sequence length: 19705 Attention mask shape: torch.Size([1, 1, 19705, 19705]) Position ids shape: torch.Size([1, 19705]) Input IDs shape: torch.Size([1, 19705]) Labels shape: torch.Size([1, 19705]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 28641 Attention mask shape: torch.Size([1, 1, 28641, 28641]) Position ids shape: torch.Size([1, 28641]) Input IDs shape: torch.Size([1, 28641]) Labels shape: torch.Size([1, 28641]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 22098 Attention mask shape: torch.Size([1, 1, 22098, 22098]) Position ids shape: torch.Size([1, 22098]) Input IDs shape: torch.Size([1, 22098]) Labels shape: torch.Size([1, 22098]) Final batch size: 1, sequence length: 22625 Attention mask shape: torch.Size([1, 1, 22625, 22625]) Position ids shape: torch.Size([1, 22625]) Input IDs shape: torch.Size([1, 22625]) Labels shape: torch.Size([1, 22625]) Final batch size: 1, sequence length: 30101 Attention mask shape: torch.Size([1, 1, 30101, 30101]) Position ids shape: torch.Size([1, 30101]) Input IDs shape: torch.Size([1, 30101]) Labels shape: torch.Size([1, 30101]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 27243 Attention mask shape: torch.Size([1, 1, 27243, 27243]) Position ids shape: torch.Size([1, 27243]) Input IDs shape: torch.Size([1, 27243]) Labels shape: torch.Size([1, 27243]) Final batch size: 1, sequence length: 27265 Attention mask shape: torch.Size([1, 1, 27265, 27265]) Position ids shape: torch.Size([1, 27265]) Input IDs shape: torch.Size([1, 27265]) Labels shape: torch.Size([1, 27265]) Final batch size: 1, sequence length: 30066 Attention mask shape: torch.Size([1, 1, 30066, 30066]) Position ids shape: torch.Size([1, 30066]) Input IDs shape: torch.Size([1, 30066]) Labels shape: torch.Size([1, 30066]) Final batch size: 1, sequence length: 24622 Attention mask shape: torch.Size([1, 1, 24622, 24622]) Position ids shape: torch.Size([1, 24622]) Input IDs shape: torch.Size([1, 24622]) Labels shape: torch.Size([1, 24622]) Final batch size: 1, sequence length: 21348 Attention mask shape: torch.Size([1, 1, 21348, 21348]) Position ids shape: torch.Size([1, 21348]) Input IDs shape: torch.Size([1, 21348]) Labels shape: torch.Size([1, 21348]) Final batch size: 1, sequence length: 27441 Attention mask shape: torch.Size([1, 1, 27441, 27441]) Position ids shape: torch.Size([1, 27441]) Input IDs shape: torch.Size([1, 27441]) Labels shape: torch.Size([1, 27441]) Final batch size: 1, sequence length: 39142 Attention mask shape: torch.Size([1, 1, 39142, 39142]) Position ids shape: torch.Size([1, 39142]) Input IDs shape: torch.Size([1, 39142]) Labels shape: torch.Size([1, 39142]) Final batch size: 1, sequence length: 17778 Attention mask shape: torch.Size([1, 1, 17778, 17778]) Position ids shape: torch.Size([1, 17778]) Input IDs shape: torch.Size([1, 17778]) Labels shape: torch.Size([1, 17778]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 18565 Attention mask shape: torch.Size([1, 1, 18565, 18565]) Position ids shape: torch.Size([1, 18565]) Input IDs shape: torch.Size([1, 18565]) Labels shape: torch.Size([1, 18565]) Final batch size: 1, sequence length: 32609 Attention mask shape: torch.Size([1, 1, 32609, 32609]) Position ids shape: torch.Size([1, 32609]) Input IDs shape: torch.Size([1, 32609]) Labels shape: torch.Size([1, 32609]) Final batch size: 1, sequence length: 33367 Attention mask shape: torch.Size([1, 1, 33367, 33367]) Position ids shape: torch.Size([1, 33367]) Input IDs shape: torch.Size([1, 33367]) Labels shape: torch.Size([1, 33367]) Final batch size: 1, sequence length: 26939 Attention mask shape: torch.Size([1, 1, 26939, 26939]) Position ids shape: torch.Size([1, 26939]) Input IDs shape: torch.Size([1, 26939]) Labels shape: torch.Size([1, 26939]) Final batch size: 1, sequence length: 25014 Attention mask shape: torch.Size([1, 1, 25014, 25014]) Position ids shape: torch.Size([1, 25014]) Input IDs shape: torch.Size([1, 25014]) Labels shape: torch.Size([1, 25014]) Final batch size: 1, sequence length: 15656 Attention mask shape: torch.Size([1, 1, 15656, 15656]) Position ids shape: torch.Size([1, 15656]) Input IDs shape: torch.Size([1, 15656]) Labels shape: torch.Size([1, 15656]) Final batch size: 1, sequence length: 30802 Attention mask shape: torch.Size([1, 1, 30802, 30802]) Position ids shape: torch.Size([1, 30802]) Input IDs shape: torch.Size([1, 30802]) Labels shape: torch.Size([1, 30802]) Final batch size: 1, sequence length: 24001 Attention mask shape: torch.Size([1, 1, 24001, 24001]) Position ids shape: torch.Size([1, 24001]) Input IDs shape: torch.Size([1, 24001]) Labels shape: torch.Size([1, 24001]) Final batch size: 1, sequence length: 29632 Attention mask shape: torch.Size([1, 1, 29632, 29632]) Position ids shape: torch.Size([1, 29632]) Input IDs shape: torch.Size([1, 29632]) Labels shape: torch.Size([1, 29632]) Final batch size: 1, sequence length: 36271 Attention mask shape: torch.Size([1, 1, 36271, 36271]) Position ids shape: torch.Size([1, 36271]) Input IDs shape: torch.Size([1, 36271]) Labels shape: torch.Size([1, 36271]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21567 Attention mask shape: torch.Size([1, 1, 21567, 21567]) Position ids shape: torch.Size([1, 21567]) Input IDs shape: torch.Size([1, 21567]) Labels shape: torch.Size([1, 21567]) Final batch size: 1, sequence length: 16587 Attention mask shape: torch.Size([1, 1, 16587, 16587]) Position ids shape: torch.Size([1, 16587]) Input IDs shape: torch.Size([1, 16587]) Labels shape: torch.Size([1, 16587]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38249 Attention mask shape: torch.Size([1, 1, 38249, 38249]) Position ids shape: torch.Size([1, 38249]) Input IDs shape: torch.Size([1, 38249]) Labels shape: torch.Size([1, 38249]) Final batch size: 1, sequence length: 35153 Attention mask shape: torch.Size([1, 1, 35153, 35153]) Position ids shape: torch.Size([1, 35153]) Input IDs shape: torch.Size([1, 35153]) Labels shape: torch.Size([1, 35153]) Final batch size: 1, sequence length: 15217 Attention mask shape: torch.Size([1, 1, 15217, 15217]) Position ids shape: torch.Size([1, 15217]) Input IDs shape: torch.Size([1, 15217]) Labels shape: torch.Size([1, 15217]) Final batch size: 1, sequence length: 19036 Attention mask shape: torch.Size([1, 1, 19036, 19036]) Position ids shape: torch.Size([1, 19036]) Input IDs shape: torch.Size([1, 19036]) Labels shape: torch.Size([1, 19036]) Final batch size: 1, sequence length: 25447 Attention mask shape: torch.Size([1, 1, 25447, 25447]) Position ids shape: torch.Size([1, 25447]) Input IDs shape: torch.Size([1, 25447]) Labels shape: torch.Size([1, 25447]) Final batch size: 1, sequence length: 33839 Attention mask shape: torch.Size([1, 1, 33839, 33839]) Position ids shape: torch.Size([1, 33839]) Input IDs shape: torch.Size([1, 33839]) Labels shape: torch.Size([1, 33839]) Final batch size: 1, sequence length: 40496 Attention mask shape: torch.Size([1, 1, 40496, 40496]) Position ids shape: torch.Size([1, 40496]) Input IDs shape: torch.Size([1, 40496]) Labels shape: torch.Size([1, 40496]) Final batch size: 1, sequence length: 39836 Attention mask shape: torch.Size([1, 1, 39836, 39836]) Position ids shape: torch.Size([1, 39836]) Input IDs shape: torch.Size([1, 39836]) Labels shape: torch.Size([1, 39836]) Final batch size: 1, sequence length: 15508 Attention mask shape: torch.Size([1, 1, 15508, 15508]) Position ids shape: torch.Size([1, 15508]) Input IDs shape: torch.Size([1, 15508]) Labels shape: torch.Size([1, 15508]) Final batch size: 1, sequence length: 26306 Attention mask shape: torch.Size([1, 1, 26306, 26306]) Position ids shape: torch.Size([1, 26306]) Input IDs shape: torch.Size([1, 26306]) Labels shape: torch.Size([1, 26306]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32352 Attention mask shape: torch.Size([1, 1, 32352, 32352]) Position ids shape: torch.Size([1, 32352]) Input IDs shape: torch.Size([1, 32352]) Labels shape: torch.Size([1, 32352]) Final batch size: 1, sequence length: 16677 Attention mask shape: torch.Size([1, 1, 16677, 16677]) Position ids shape: torch.Size([1, 16677]) Input IDs shape: torch.Size([1, 16677]) Labels shape: torch.Size([1, 16677]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 9947 Attention mask shape: torch.Size([1, 1, 9947, 9947]) Position ids shape: torch.Size([1, 9947]) Input IDs shape: torch.Size([1, 9947]) Labels shape: torch.Size([1, 9947]) Final batch size: 1, sequence length: 13509 Attention mask shape: torch.Size([1, 1, 13509, 13509]) Position ids shape: torch.Size([1, 13509]) Input IDs shape: torch.Size([1, 13509]) Labels shape: torch.Size([1, 13509]) Final batch size: 1, sequence length: 21758 Attention mask shape: torch.Size([1, 1, 21758, 21758]) Position ids shape: torch.Size([1, 21758]) Input IDs shape: torch.Size([1, 21758]) Labels shape: torch.Size([1, 21758]) Final batch size: 1, sequence length: 40745 Attention mask shape: torch.Size([1, 1, 40745, 40745]) Position ids shape: torch.Size([1, 40745]) Input IDs shape: torch.Size([1, 40745]) Labels shape: torch.Size([1, 40745]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36511 Attention mask shape: torch.Size([1, 1, 36511, 36511]) Position ids shape: torch.Size([1, 36511]) Input IDs shape: torch.Size([1, 36511]) Labels shape: torch.Size([1, 36511]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 37945 Attention mask shape: torch.Size([1, 1, 37945, 37945]) Position ids shape: torch.Size([1, 37945]) Input IDs shape: torch.Size([1, 37945]) Labels shape: torch.Size([1, 37945]) Final batch size: 1, sequence length: 18606 Attention mask shape: torch.Size([1, 1, 18606, 18606]) Position ids shape: torch.Size([1, 18606]) Input IDs shape: torch.Size([1, 18606]) Labels shape: torch.Size([1, 18606]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 20611 Attention mask shape: torch.Size([1, 1, 20611, 20611]) Position ids shape: torch.Size([1, 20611]) Input IDs shape: torch.Size([1, 20611]) Labels shape: torch.Size([1, 20611]) Final batch size: 1, sequence length: 23258 Attention mask shape: torch.Size([1, 1, 23258, 23258]) Position ids shape: torch.Size([1, 23258]) Input IDs shape: torch.Size([1, 23258]) Labels shape: torch.Size([1, 23258]) Final batch size: 1, sequence length: 21061 Attention mask shape: torch.Size([1, 1, 21061, 21061]) Position ids shape: torch.Size([1, 21061]) Input IDs shape: torch.Size([1, 21061]) Labels shape: torch.Size([1, 21061]) Final batch size: 1, sequence length: 13095 Attention mask shape: torch.Size([1, 1, 13095, 13095]) Position ids shape: torch.Size([1, 13095]) Input IDs shape: torch.Size([1, 13095]) Labels shape: torch.Size([1, 13095]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30109 Attention mask shape: torch.Size([1, 1, 30109, 30109]) Position ids shape: torch.Size([1, 30109]) Input IDs shape: torch.Size([1, 30109]) Labels shape: torch.Size([1, 30109]) Final batch size: 1, sequence length: 13297 Attention mask shape: torch.Size([1, 1, 13297, 13297]) Position ids shape: torch.Size([1, 13297]) Input IDs shape: torch.Size([1, 13297]) Labels shape: torch.Size([1, 13297]) Final batch size: 1, sequence length: 12224 Attention mask shape: torch.Size([1, 1, 12224, 12224]) Position ids shape: torch.Size([1, 12224]) Input IDs shape: torch.Size([1, 12224]) Labels shape: torch.Size([1, 12224]) Final batch size: 1, sequence length: 20422 Attention mask shape: torch.Size([1, 1, 20422, 20422]) Position ids shape: torch.Size([1, 20422]) Input IDs shape: torch.Size([1, 20422]) Labels shape: torch.Size([1, 20422]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34142 Attention mask shape: torch.Size([1, 1, 34142, 34142]) Position ids shape: torch.Size([1, 34142]) Input IDs shape: torch.Size([1, 34142]) Labels shape: torch.Size([1, 34142]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28634 Attention mask shape: torch.Size([1, 1, 28634, 28634]) Position ids shape: torch.Size([1, 28634]) Input IDs shape: torch.Size([1, 28634]) Labels shape: torch.Size([1, 28634]) Final batch size: 1, sequence length: 17373 Attention mask shape: torch.Size([1, 1, 17373, 17373]) Position ids shape: torch.Size([1, 17373]) Input IDs shape: torch.Size([1, 17373]) Labels shape: torch.Size([1, 17373]) Final batch size: 1, sequence length: 32472 Attention mask shape: torch.Size([1, 1, 32472, 32472]) Position ids shape: torch.Size([1, 32472]) Input IDs shape: torch.Size([1, 32472]) Labels shape: torch.Size([1, 32472]) Final batch size: 1, sequence length: 34937 Attention mask shape: torch.Size([1, 1, 34937, 34937]) Position ids shape: torch.Size([1, 34937]) Input IDs shape: torch.Size([1, 34937]) Labels shape: torch.Size([1, 34937]) Final batch size: 1, sequence length: 24298 Attention mask shape: torch.Size([1, 1, 24298, 24298]) Position ids shape: torch.Size([1, 24298]) Input IDs shape: torch.Size([1, 24298]) Labels shape: torch.Size([1, 24298]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21683 Attention mask shape: torch.Size([1, 1, 21683, 21683]) Position ids shape: torch.Size([1, 21683]) Input IDs shape: torch.Size([1, 21683]) Labels shape: torch.Size([1, 21683]) Final batch size: 1, sequence length: 32753 Attention mask shape: torch.Size([1, 1, 32753, 32753]) Position ids shape: torch.Size([1, 32753]) Input IDs shape: torch.Size([1, 32753]) Labels shape: torch.Size([1, 32753]) Final batch size: 1, sequence length: 12653 Attention mask shape: torch.Size([1, 1, 12653, 12653]) Position ids shape: torch.Size([1, 12653]) Input IDs shape: torch.Size([1, 12653]) Labels shape: torch.Size([1, 12653]) Final batch size: 1, sequence length: 22786 Attention mask shape: torch.Size([1, 1, 22786, 22786]) Position ids shape: torch.Size([1, 22786]) Input IDs shape: torch.Size([1, 22786]) Labels shape: torch.Size([1, 22786]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 22623 Attention mask shape: torch.Size([1, 1, 22623, 22623]) Position ids shape: torch.Size([1, 22623]) Input IDs shape: torch.Size([1, 22623]) Labels shape: torch.Size([1, 22623]) Final batch size: 1, sequence length: 14995 Attention mask shape: torch.Size([1, 1, 14995, 14995]) Position ids shape: torch.Size([1, 14995]) Input IDs shape: torch.Size([1, 14995]) Labels shape: torch.Size([1, 14995]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32564 Attention mask shape: torch.Size([1, 1, 32564, 32564]) Position ids shape: torch.Size([1, 32564]) Input IDs shape: torch.Size([1, 32564]) Labels shape: torch.Size([1, 32564]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24424 Attention mask shape: torch.Size([1, 1, 24424, 24424]) Position ids shape: torch.Size([1, 24424]) Input IDs shape: torch.Size([1, 24424]) Labels shape: torch.Size([1, 24424]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19586 Attention mask shape: torch.Size([1, 1, 19586, 19586]) Position ids shape: torch.Size([1, 19586]) Input IDs shape: torch.Size([1, 19586]) Labels shape: torch.Size([1, 19586]) Final batch size: 1, sequence length: 35864 Attention mask shape: torch.Size([1, 1, 35864, 35864]) Position ids shape: torch.Size([1, 35864]) Input IDs shape: torch.Size([1, 35864]) Labels shape: torch.Size([1, 35864]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26006 Attention mask shape: torch.Size([1, 1, 26006, 26006]) Position ids shape: torch.Size([1, 26006]) Input IDs shape: torch.Size([1, 26006]) Labels shape: torch.Size([1, 26006]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32344 Attention mask shape: torch.Size([1, 1, 32344, 32344]) Position ids shape: torch.Size([1, 32344]) Input IDs shape: torch.Size([1, 32344]) Labels shape: torch.Size([1, 32344]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26537 Attention mask shape: torch.Size([1, 1, 26537, 26537]) Position ids shape: torch.Size([1, 26537]) Input IDs shape: torch.Size([1, 26537]) Labels shape: torch.Size([1, 26537]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32919 Attention mask shape: torch.Size([1, 1, 32919, 32919]) Position ids shape: torch.Size([1, 32919]) Input IDs shape: torch.Size([1, 32919]) Labels shape: torch.Size([1, 32919]) Final batch size: 1, sequence length: 37168 Attention mask shape: torch.Size([1, 1, 37168, 37168]) Position ids shape: torch.Size([1, 37168]) Input IDs shape: torch.Size([1, 37168]) Labels shape: torch.Size([1, 37168]) Final batch size: 1, sequence length: 23362 Attention mask shape: torch.Size([1, 1, 23362, 23362]) Position ids shape: torch.Size([1, 23362]) Input IDs shape: torch.Size([1, 23362]) Labels shape: torch.Size([1, 23362]) Final batch size: 1, sequence length: 32514 Attention mask shape: torch.Size([1, 1, 32514, 32514]) Position ids shape: torch.Size([1, 32514]) Input IDs shape: torch.Size([1, 32514]) Labels shape: torch.Size([1, 32514]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28861 Attention mask shape: torch.Size([1, 1, 28861, 28861]) Position ids shape: torch.Size([1, 28861]) Input IDs shape: torch.Size([1, 28861]) Labels shape: torch.Size([1, 28861]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2416, 'grad_norm': 0.17127270931377664, 'learning_rate': 4.322727117869951e-07, 'num_tokens': -inf, 'epoch': 7.12} Final batch size: 1, sequence length: 4858 Attention mask shape: torch.Size([1, 1, 4858, 4858]) Position ids shape: torch.Size([1, 4858]) Input IDs shape: torch.Size([1, 4858]) Labels shape: torch.Size([1, 4858]) Final batch size: 1, sequence length: 7360 Attention mask shape: torch.Size([1, 1, 7360, 7360]) Position ids shape: torch.Size([1, 7360]) Input IDs shape: torch.Size([1, 7360]) Labels shape: torch.Size([1, 7360]) Final batch size: 1, sequence length: 10301 Attention mask shape: torch.Size([1, 1, 10301, 10301]) Position ids shape: torch.Size([1, 10301]) Input IDs shape: torch.Size([1, 10301]) Labels shape: torch.Size([1, 10301]) Final batch size: 1, sequence length: 6316 Attention mask shape: torch.Size([1, 1, 6316, 6316]) Position ids shape: torch.Size([1, 6316]) Input IDs shape: torch.Size([1, 6316]) Labels shape: torch.Size([1, 6316]) Final batch size: 1, sequence length: 11548 Attention mask shape: torch.Size([1, 1, 11548, 11548]) Position ids shape: torch.Size([1, 11548]) Input IDs shape: torch.Size([1, 11548]) Labels shape: torch.Size([1, 11548]) Final batch size: 1, sequence length: 12293 Attention mask shape: torch.Size([1, 1, 12293, 12293]) Position ids shape: torch.Size([1, 12293]) Input IDs shape: torch.Size([1, 12293]) Labels shape: torch.Size([1, 12293]) Final batch size: 1, sequence length: 13355 Attention mask shape: torch.Size([1, 1, 13355, 13355]) Position ids shape: torch.Size([1, 13355]) Input IDs shape: torch.Size([1, 13355]) Labels shape: torch.Size([1, 13355]) Final batch size: 1, sequence length: 14363 Attention mask shape: torch.Size([1, 1, 14363, 14363]) Position ids shape: torch.Size([1, 14363]) Input IDs shape: torch.Size([1, 14363]) Labels shape: torch.Size([1, 14363]) Final batch size: 1, sequence length: 17108 Attention mask shape: torch.Size([1, 1, 17108, 17108]) Position ids shape: torch.Size([1, 17108]) Input IDs shape: torch.Size([1, 17108]) Labels shape: torch.Size([1, 17108]) Final batch size: 1, sequence length: 14587 Attention mask shape: torch.Size([1, 1, 14587, 14587]) Position ids shape: torch.Size([1, 14587]) Input IDs shape: torch.Size([1, 14587]) Labels shape: torch.Size([1, 14587]) Final batch size: 1, sequence length: 11728 Attention mask shape: torch.Size([1, 1, 11728, 11728]) Position ids shape: torch.Size([1, 11728]) Input IDs shape: torch.Size([1, 11728]) Labels shape: torch.Size([1, 11728]) Final batch size: 1, sequence length: 16827 Attention mask shape: torch.Size([1, 1, 16827, 16827]) Position ids shape: torch.Size([1, 16827]) Input IDs shape: torch.Size([1, 16827]) Labels shape: torch.Size([1, 16827]) Final batch size: 1, sequence length: 14704 Attention mask shape: torch.Size([1, 1, 14704, 14704]) Position ids shape: torch.Size([1, 14704]) Input IDs shape: torch.Size([1, 14704]) Labels shape: torch.Size([1, 14704]) Final batch size: 1, sequence length: 18496 Attention mask shape: torch.Size([1, 1, 18496, 18496]) Position ids shape: torch.Size([1, 18496]) Input IDs shape: torch.Size([1, 18496]) Labels shape: torch.Size([1, 18496]) Final batch size: 1, sequence length: 13638 Attention mask shape: torch.Size([1, 1, 13638, 13638]) Position ids shape: torch.Size([1, 13638]) Input IDs shape: torch.Size([1, 13638]) Labels shape: torch.Size([1, 13638]) Final batch size: 1, sequence length: 16536 Attention mask shape: torch.Size([1, 1, 16536, 16536]) Position ids shape: torch.Size([1, 16536]) Input IDs shape: torch.Size([1, 16536]) Labels shape: torch.Size([1, 16536]) Final batch size: 1, sequence length: 17415 Attention mask shape: torch.Size([1, 1, 17415, 17415]) Position ids shape: torch.Size([1, 17415]) Input IDs shape: torch.Size([1, 17415]) Labels shape: torch.Size([1, 17415]) Final batch size: 1, sequence length: 7221 Attention mask shape: torch.Size([1, 1, 7221, 7221]) Position ids shape: torch.Size([1, 7221]) Input IDs shape: torch.Size([1, 7221]) Labels shape: torch.Size([1, 7221]) Final batch size: 1, sequence length: 17220 Attention mask shape: torch.Size([1, 1, 17220, 17220]) Position ids shape: torch.Size([1, 17220]) Input IDs shape: torch.Size([1, 17220]) Labels shape: torch.Size([1, 17220]) Final batch size: 1, sequence length: 18741 Attention mask shape: torch.Size([1, 1, 18741, 18741]) Position ids shape: torch.Size([1, 18741]) Input IDs shape: torch.Size([1, 18741]) Labels shape: torch.Size([1, 18741]) Final batch size: 1, sequence length: 16750 Attention mask shape: torch.Size([1, 1, 16750, 16750]) Position ids shape: torch.Size([1, 16750]) Input IDs shape: torch.Size([1, 16750]) Labels shape: torch.Size([1, 16750]) Final batch size: 1, sequence length: 21420 Attention mask shape: torch.Size([1, 1, 21420, 21420]) Position ids shape: torch.Size([1, 21420]) Input IDs shape: torch.Size([1, 21420]) Labels shape: torch.Size([1, 21420]) Final batch size: 1, sequence length: 19414 Attention mask shape: torch.Size([1, 1, 19414, 19414]) Position ids shape: torch.Size([1, 19414]) Input IDs shape: torch.Size([1, 19414]) Labels shape: torch.Size([1, 19414]) Final batch size: 1, sequence length: 18653 Attention mask shape: torch.Size([1, 1, 18653, 18653]) Position ids shape: torch.Size([1, 18653]) Input IDs shape: torch.Size([1, 18653]) Labels shape: torch.Size([1, 18653]) Final batch size: 1, sequence length: 17733 Attention mask shape: torch.Size([1, 1, 17733, 17733]) Position ids shape: torch.Size([1, 17733]) Input IDs shape: torch.Size([1, 17733]) Labels shape: torch.Size([1, 17733]) Final batch size: 1, sequence length: 20612 Attention mask shape: torch.Size([1, 1, 20612, 20612]) Position ids shape: torch.Size([1, 20612]) Input IDs shape: torch.Size([1, 20612]) Labels shape: torch.Size([1, 20612]) Final batch size: 1, sequence length: 18393 Attention mask shape: torch.Size([1, 1, 18393, 18393]) Position ids shape: torch.Size([1, 18393]) Input IDs shape: torch.Size([1, 18393]) Labels shape: torch.Size([1, 18393]) Final batch size: 1, sequence length: 18927 Attention mask shape: torch.Size([1, 1, 18927, 18927]) Position ids shape: torch.Size([1, 18927]) Input IDs shape: torch.Size([1, 18927]) Labels shape: torch.Size([1, 18927]) Final batch size: 1, sequence length: 22391 Attention mask shape: torch.Size([1, 1, 22391, 22391]) Position ids shape: torch.Size([1, 22391]) Input IDs shape: torch.Size([1, 22391]) Labels shape: torch.Size([1, 22391]) Final batch size: 1, sequence length: 20579 Attention mask shape: torch.Size([1, 1, 20579, 20579]) Position ids shape: torch.Size([1, 20579]) Input IDs shape: torch.Size([1, 20579]) Labels shape: torch.Size([1, 20579]) Final batch size: 1, sequence length: 20933 Attention mask shape: torch.Size([1, 1, 20933, 20933]) Position ids shape: torch.Size([1, 20933]) Input IDs shape: torch.Size([1, 20933]) Labels shape: torch.Size([1, 20933]) Final batch size: 1, sequence length: 22887 Attention mask shape: torch.Size([1, 1, 22887, 22887]) Position ids shape: torch.Size([1, 22887]) Input IDs shape: torch.Size([1, 22887]) Labels shape: torch.Size([1, 22887]) Final batch size: 1, sequence length: 11067 Attention mask shape: torch.Size([1, 1, 11067, 11067]) Position ids shape: torch.Size([1, 11067]) Input IDs shape: torch.Size([1, 11067]) Labels shape: torch.Size([1, 11067]) Final batch size: 1, sequence length: 25747 Attention mask shape: torch.Size([1, 1, 25747, 25747]) Position ids shape: torch.Size([1, 25747]) Input IDs shape: torch.Size([1, 25747]) Labels shape: torch.Size([1, 25747]) Final batch size: 1, sequence length: 24988 Attention mask shape: torch.Size([1, 1, 24988, 24988]) Position ids shape: torch.Size([1, 24988]) Input IDs shape: torch.Size([1, 24988]) Labels shape: torch.Size([1, 24988]) Final batch size: 1, sequence length: 25477 Attention mask shape: torch.Size([1, 1, 25477, 25477]) Position ids shape: torch.Size([1, 25477]) Input IDs shape: torch.Size([1, 25477]) Labels shape: torch.Size([1, 25477]) Final batch size: 1, sequence length: 25651 Attention mask shape: torch.Size([1, 1, 25651, 25651]) Position ids shape: torch.Size([1, 25651]) Input IDs shape: torch.Size([1, 25651]) Labels shape: torch.Size([1, 25651]) Final batch size: 1, sequence length: 15317 Attention mask shape: torch.Size([1, 1, 15317, 15317]) Position ids shape: torch.Size([1, 15317]) Input IDs shape: torch.Size([1, 15317]) Labels shape: torch.Size([1, 15317]) Final batch size: 1, sequence length: 22004 Attention mask shape: torch.Size([1, 1, 22004, 22004]) Position ids shape: torch.Size([1, 22004]) Input IDs shape: torch.Size([1, 22004]) Labels shape: torch.Size([1, 22004]) Final batch size: 1, sequence length: 11184 Attention mask shape: torch.Size([1, 1, 11184, 11184]) Position ids shape: torch.Size([1, 11184]) Input IDs shape: torch.Size([1, 11184]) Labels shape: torch.Size([1, 11184]) Final batch size: 1, sequence length: 26479 Attention mask shape: torch.Size([1, 1, 26479, 26479]) Position ids shape: torch.Size([1, 26479]) Input IDs shape: torch.Size([1, 26479]) Labels shape: torch.Size([1, 26479]) Final batch size: 1, sequence length: 10719 Attention mask shape: torch.Size([1, 1, 10719, 10719]) Position ids shape: torch.Size([1, 10719]) Input IDs shape: torch.Size([1, 10719]) Labels shape: torch.Size([1, 10719]) Final batch size: 1, sequence length: 16915 Attention mask shape: torch.Size([1, 1, 16915, 16915]) Position ids shape: torch.Size([1, 16915]) Input IDs shape: torch.Size([1, 16915]) Labels shape: torch.Size([1, 16915]) Final batch size: 1, sequence length: 19552 Attention mask shape: torch.Size([1, 1, 19552, 19552]) Position ids shape: torch.Size([1, 19552]) Input IDs shape: torch.Size([1, 19552]) Labels shape: torch.Size([1, 19552]) Final batch size: 1, sequence length: 26663 Attention mask shape: torch.Size([1, 1, 26663, 26663]) Position ids shape: torch.Size([1, 26663]) Input IDs shape: torch.Size([1, 26663]) Labels shape: torch.Size([1, 26663]) Final batch size: 1, sequence length: 27334 Attention mask shape: torch.Size([1, 1, 27334, 27334]) Position ids shape: torch.Size([1, 27334]) Input IDs shape: torch.Size([1, 27334]) Labels shape: torch.Size([1, 27334]) Final batch size: 1, sequence length: 19869 Attention mask shape: torch.Size([1, 1, 19869, 19869]) Position ids shape: torch.Size([1, 19869]) Input IDs shape: torch.Size([1, 19869]) Labels shape: torch.Size([1, 19869]) Final batch size: 1, sequence length: 28777 Attention mask shape: torch.Size([1, 1, 28777, 28777]) Position ids shape: torch.Size([1, 28777]) Input IDs shape: torch.Size([1, 28777]) Labels shape: torch.Size([1, 28777]) Final batch size: 1, sequence length: 27447 Attention mask shape: torch.Size([1, 1, 27447, 27447]) Position ids shape: torch.Size([1, 27447]) Input IDs shape: torch.Size([1, 27447]) Labels shape: torch.Size([1, 27447]) Final batch size: 1, sequence length: 27480 Attention mask shape: torch.Size([1, 1, 27480, 27480]) Position ids shape: torch.Size([1, 27480]) Input IDs shape: torch.Size([1, 27480]) Labels shape: torch.Size([1, 27480]) Final batch size: 1, sequence length: 26033 Attention mask shape: torch.Size([1, 1, 26033, 26033]) Position ids shape: torch.Size([1, 26033]) Input IDs shape: torch.Size([1, 26033]) Labels shape: torch.Size([1, 26033]) Final batch size: 1, sequence length: 30981 Attention mask shape: torch.Size([1, 1, 30981, 30981]) Position ids shape: torch.Size([1, 30981]) Input IDs shape: torch.Size([1, 30981]) Labels shape: torch.Size([1, 30981]) Final batch size: 1, sequence length: 17395 Attention mask shape: torch.Size([1, 1, 17395, 17395]) Position ids shape: torch.Size([1, 17395]) Input IDs shape: torch.Size([1, 17395]) Labels shape: torch.Size([1, 17395]) Final batch size: 1, sequence length: 16953 Attention mask shape: torch.Size([1, 1, 16953, 16953]) Position ids shape: torch.Size([1, 16953]) Input IDs shape: torch.Size([1, 16953]) Labels shape: torch.Size([1, 16953]) Final batch size: 1, sequence length: 30031 Attention mask shape: torch.Size([1, 1, 30031, 30031]) Position ids shape: torch.Size([1, 30031]) Input IDs shape: torch.Size([1, 30031]) Labels shape: torch.Size([1, 30031]) Final batch size: 1, sequence length: 18376 Attention mask shape: torch.Size([1, 1, 18376, 18376]) Position ids shape: torch.Size([1, 18376]) Input IDs shape: torch.Size([1, 18376]) Labels shape: torch.Size([1, 18376]) Final batch size: 1, sequence length: 32466 Attention mask shape: torch.Size([1, 1, 32466, 32466]) Position ids shape: torch.Size([1, 32466]) Input IDs shape: torch.Size([1, 32466]) Labels shape: torch.Size([1, 32466]) Final batch size: 1, sequence length: 28060 Attention mask shape: torch.Size([1, 1, 28060, 28060]) Position ids shape: torch.Size([1, 28060]) Input IDs shape: torch.Size([1, 28060]) Labels shape: torch.Size([1, 28060]) Final batch size: 1, sequence length: 30601 Attention mask shape: torch.Size([1, 1, 30601, 30601]) Position ids shape: torch.Size([1, 30601]) Input IDs shape: torch.Size([1, 30601]) Labels shape: torch.Size([1, 30601]) Final batch size: 1, sequence length: 17911 Attention mask shape: torch.Size([1, 1, 17911, 17911]) Position ids shape: torch.Size([1, 17911]) Input IDs shape: torch.Size([1, 17911]) Labels shape: torch.Size([1, 17911]) Final batch size: 1, sequence length: 24782 Attention mask shape: torch.Size([1, 1, 24782, 24782]) Position ids shape: torch.Size([1, 24782]) Input IDs shape: torch.Size([1, 24782]) Labels shape: torch.Size([1, 24782]) Final batch size: 1, sequence length: 12245 Attention mask shape: torch.Size([1, 1, 12245, 12245]) Position ids shape: torch.Size([1, 12245]) Input IDs shape: torch.Size([1, 12245]) Labels shape: torch.Size([1, 12245]) Final batch size: 1, sequence length: 21988 Attention mask shape: torch.Size([1, 1, 21988, 21988]) Position ids shape: torch.Size([1, 21988]) Input IDs shape: torch.Size([1, 21988]) Labels shape: torch.Size([1, 21988]) Final batch size: 1, sequence length: 14873 Attention mask shape: torch.Size([1, 1, 14873, 14873]) Position ids shape: torch.Size([1, 14873]) Input IDs shape: torch.Size([1, 14873]) Labels shape: torch.Size([1, 14873]) Final batch size: 1, sequence length: 33725 Attention mask shape: torch.Size([1, 1, 33725, 33725]) Position ids shape: torch.Size([1, 33725]) Input IDs shape: torch.Size([1, 33725]) Labels shape: torch.Size([1, 33725]) Final batch size: 1, sequence length: 37738 Attention mask shape: torch.Size([1, 1, 37738, 37738]) Position ids shape: torch.Size([1, 37738]) Input IDs shape: torch.Size([1, 37738]) Labels shape: torch.Size([1, 37738]) Final batch size: 1, sequence length: 15924 Attention mask shape: torch.Size([1, 1, 15924, 15924]) Position ids shape: torch.Size([1, 15924]) Input IDs shape: torch.Size([1, 15924]) Labels shape: torch.Size([1, 15924]) Final batch size: 1, sequence length: 27385 Attention mask shape: torch.Size([1, 1, 27385, 27385]) Position ids shape: torch.Size([1, 27385]) Input IDs shape: torch.Size([1, 27385]) Labels shape: torch.Size([1, 27385]) Final batch size: 1, sequence length: 32835 Attention mask shape: torch.Size([1, 1, 32835, 32835]) Position ids shape: torch.Size([1, 32835]) Input IDs shape: torch.Size([1, 32835]) Labels shape: torch.Size([1, 32835]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 17595 Attention mask shape: torch.Size([1, 1, 17595, 17595]) Position ids shape: torch.Size([1, 17595]) Input IDs shape: torch.Size([1, 17595]) Labels shape: torch.Size([1, 17595]) Final batch size: 1, sequence length: 37511 Attention mask shape: torch.Size([1, 1, 37511, 37511]) Position ids shape: torch.Size([1, 37511]) Input IDs shape: torch.Size([1, 37511]) Labels shape: torch.Size([1, 37511]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40397 Attention mask shape: torch.Size([1, 1, 40397, 40397]) Position ids shape: torch.Size([1, 40397]) Input IDs shape: torch.Size([1, 40397]) Labels shape: torch.Size([1, 40397]) Final batch size: 1, sequence length: 36175 Attention mask shape: torch.Size([1, 1, 36175, 36175]) Position ids shape: torch.Size([1, 36175]) Input IDs shape: torch.Size([1, 36175]) Labels shape: torch.Size([1, 36175]) Final batch size: 1, sequence length: 37158 Attention mask shape: torch.Size([1, 1, 37158, 37158]) Position ids shape: torch.Size([1, 37158]) Input IDs shape: torch.Size([1, 37158]) Labels shape: torch.Size([1, 37158]) Final batch size: 1, sequence length: 31879 Attention mask shape: torch.Size([1, 1, 31879, 31879]) Position ids shape: torch.Size([1, 31879]) Input IDs shape: torch.Size([1, 31879]) Labels shape: torch.Size([1, 31879]) Final batch size: 1, sequence length: 36130 Attention mask shape: torch.Size([1, 1, 36130, 36130]) Position ids shape: torch.Size([1, 36130]) Input IDs shape: torch.Size([1, 36130]) Labels shape: torch.Size([1, 36130]) Final batch size: 1, sequence length: 29586 Attention mask shape: torch.Size([1, 1, 29586, 29586]) Position ids shape: torch.Size([1, 29586]) Input IDs shape: torch.Size([1, 29586]) Labels shape: torch.Size([1, 29586]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 37025 Attention mask shape: torch.Size([1, 1, 37025, 37025]) Position ids shape: torch.Size([1, 37025]) Input IDs shape: torch.Size([1, 37025]) Labels shape: torch.Size([1, 37025]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38986 Attention mask shape: torch.Size([1, 1, 38986, 38986]) Position ids shape: torch.Size([1, 38986]) Input IDs shape: torch.Size([1, 38986]) Labels shape: torch.Size([1, 38986]) Final batch size: 1, sequence length: 31348 Attention mask shape: torch.Size([1, 1, 31348, 31348]) Position ids shape: torch.Size([1, 31348]) Input IDs shape: torch.Size([1, 31348]) Labels shape: torch.Size([1, 31348]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40269 Attention mask shape: torch.Size([1, 1, 40269, 40269]) Position ids shape: torch.Size([1, 40269]) Input IDs shape: torch.Size([1, 40269]) Labels shape: torch.Size([1, 40269]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32318 Attention mask shape: torch.Size([1, 1, 32318, 32318]) Position ids shape: torch.Size([1, 32318]) Input IDs shape: torch.Size([1, 32318]) Labels shape: torch.Size([1, 32318]) Final batch size: 1, sequence length: 20509 Attention mask shape: torch.Size([1, 1, 20509, 20509]) Position ids shape: torch.Size([1, 20509]) Input IDs shape: torch.Size([1, 20509]) Labels shape: torch.Size([1, 20509]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40237 Attention mask shape: torch.Size([1, 1, 40237, 40237]) Position ids shape: torch.Size([1, 40237]) Input IDs shape: torch.Size([1, 40237]) Labels shape: torch.Size([1, 40237]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32247 Attention mask shape: torch.Size([1, 1, 32247, 32247]) Position ids shape: torch.Size([1, 32247]) Input IDs shape: torch.Size([1, 32247]) Labels shape: torch.Size([1, 32247]) Final batch size: 1, sequence length: 10198 Attention mask shape: torch.Size([1, 1, 10198, 10198]) Position ids shape: torch.Size([1, 10198]) Input IDs shape: torch.Size([1, 10198]) Labels shape: torch.Size([1, 10198]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17711 Attention mask shape: torch.Size([1, 1, 17711, 17711]) Position ids shape: torch.Size([1, 17711]) Input IDs shape: torch.Size([1, 17711]) Labels shape: torch.Size([1, 17711]) Final batch size: 1, sequence length: 24939 Attention mask shape: torch.Size([1, 1, 24939, 24939]) Position ids shape: torch.Size([1, 24939]) Input IDs shape: torch.Size([1, 24939]) Labels shape: torch.Size([1, 24939]) Final batch size: 1, sequence length: 22205 Attention mask shape: torch.Size([1, 1, 22205, 22205]) Position ids shape: torch.Size([1, 22205]) Input IDs shape: torch.Size([1, 22205]) Labels shape: torch.Size([1, 22205]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28046 Attention mask shape: torch.Size([1, 1, 28046, 28046]) Position ids shape: torch.Size([1, 28046]) Input IDs shape: torch.Size([1, 28046]) Labels shape: torch.Size([1, 28046]) Final batch size: 1, sequence length: 23740 Attention mask shape: torch.Size([1, 1, 23740, 23740]) Position ids shape: torch.Size([1, 23740]) Input IDs shape: torch.Size([1, 23740]) Labels shape: torch.Size([1, 23740]) Final batch size: 1, sequence length: 21225 Attention mask shape: torch.Size([1, 1, 21225, 21225]) Position ids shape: torch.Size([1, 21225]) Input IDs shape: torch.Size([1, 21225]) Labels shape: torch.Size([1, 21225]) Final batch size: 1, sequence length: 36947 Attention mask shape: torch.Size([1, 1, 36947, 36947]) Position ids shape: torch.Size([1, 36947]) Input IDs shape: torch.Size([1, 36947]) Labels shape: torch.Size([1, 36947]) Final batch size: 1, sequence length: 22896 Attention mask shape: torch.Size([1, 1, 22896, 22896]) Position ids shape: torch.Size([1, 22896]) Input IDs shape: torch.Size([1, 22896]) Labels shape: torch.Size([1, 22896]) Final batch size: 1, sequence length: 28631 Attention mask shape: torch.Size([1, 1, 28631, 28631]) Position ids shape: torch.Size([1, 28631]) Input IDs shape: torch.Size([1, 28631]) Labels shape: torch.Size([1, 28631]) Final batch size: 1, sequence length: 27291 Attention mask shape: torch.Size([1, 1, 27291, 27291]) Position ids shape: torch.Size([1, 27291]) Input IDs shape: torch.Size([1, 27291]) Labels shape: torch.Size([1, 27291]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33125 Attention mask shape: torch.Size([1, 1, 33125, 33125]) Position ids shape: torch.Size([1, 33125]) Input IDs shape: torch.Size([1, 33125]) Labels shape: torch.Size([1, 33125]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28149 Attention mask shape: torch.Size([1, 1, 28149, 28149]) Position ids shape: torch.Size([1, 28149]) Input IDs shape: torch.Size([1, 28149]) Labels shape: torch.Size([1, 28149]) Final batch size: 1, sequence length: 38360 Attention mask shape: torch.Size([1, 1, 38360, 38360]) Position ids shape: torch.Size([1, 38360]) Input IDs shape: torch.Size([1, 38360]) Labels shape: torch.Size([1, 38360]) Final batch size: 1, sequence length: 32767 Attention mask shape: torch.Size([1, 1, 32767, 32767]) Position ids shape: torch.Size([1, 32767]) Input IDs shape: torch.Size([1, 32767]) Labels shape: torch.Size([1, 32767]) Final batch size: 1, sequence length: 36567 Attention mask shape: torch.Size([1, 1, 36567, 36567]) Position ids shape: torch.Size([1, 36567]) Input IDs shape: torch.Size([1, 36567]) Labels shape: torch.Size([1, 36567]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34268 Attention mask shape: torch.Size([1, 1, 34268, 34268]) Position ids shape: torch.Size([1, 34268]) Input IDs shape: torch.Size([1, 34268]) Labels shape: torch.Size([1, 34268]) Final batch size: 1, sequence length: 32313 Attention mask shape: torch.Size([1, 1, 32313, 32313]) Position ids shape: torch.Size([1, 32313]) Input IDs shape: torch.Size([1, 32313]) Labels shape: torch.Size([1, 32313]) {'loss': 0.2715, 'grad_norm': 0.1634385934518098, 'learning_rate': 3.320978675139919e-07, 'num_tokens': -inf, 'epoch': 7.25} Final batch size: 1, sequence length: 7998 Attention mask shape: torch.Size([1, 1, 7998, 7998]) Position ids shape: torch.Size([1, 7998]) Input IDs shape: torch.Size([1, 7998]) Labels shape: torch.Size([1, 7998]) Final batch size: 1, sequence length: 7402 Attention mask shape: torch.Size([1, 1, 7402, 7402]) Position ids shape: torch.Size([1, 7402]) Input IDs shape: torch.Size([1, 7402]) Labels shape: torch.Size([1, 7402]) Final batch size: 1, sequence length: 8436 Attention mask shape: torch.Size([1, 1, 8436, 8436]) Position ids shape: torch.Size([1, 8436]) Input IDs shape: torch.Size([1, 8436]) Labels shape: torch.Size([1, 8436]) Final batch size: 1, sequence length: 8655 Attention mask shape: torch.Size([1, 1, 8655, 8655]) Position ids shape: torch.Size([1, 8655]) Input IDs shape: torch.Size([1, 8655]) Labels shape: torch.Size([1, 8655]) Final batch size: 1, sequence length: 10576 Attention mask shape: torch.Size([1, 1, 10576, 10576]) Position ids shape: torch.Size([1, 10576]) Input IDs shape: torch.Size([1, 10576]) Labels shape: torch.Size([1, 10576]) Final batch size: 1, sequence length: 5377 Attention mask shape: torch.Size([1, 1, 5377, 5377]) Position ids shape: torch.Size([1, 5377]) Input IDs shape: torch.Size([1, 5377]) Labels shape: torch.Size([1, 5377]) Final batch size: 1, sequence length: 11709 Attention mask shape: torch.Size([1, 1, 11709, 11709]) Position ids shape: torch.Size([1, 11709]) Input IDs shape: torch.Size([1, 11709]) Labels shape: torch.Size([1, 11709]) Final batch size: 1, sequence length: 7344 Attention mask shape: torch.Size([1, 1, 7344, 7344]) Position ids shape: torch.Size([1, 7344]) Input IDs shape: torch.Size([1, 7344]) Labels shape: torch.Size([1, 7344]) Final batch size: 1, sequence length: 11678 Attention mask shape: torch.Size([1, 1, 11678, 11678]) Position ids shape: torch.Size([1, 11678]) Input IDs shape: torch.Size([1, 11678]) Labels shape: torch.Size([1, 11678]) Final batch size: 1, sequence length: 11365 Attention mask shape: torch.Size([1, 1, 11365, 11365]) Position ids shape: torch.Size([1, 11365]) Input IDs shape: torch.Size([1, 11365]) Labels shape: torch.Size([1, 11365]) Final batch size: 1, sequence length: 14127 Attention mask shape: torch.Size([1, 1, 14127, 14127]) Position ids shape: torch.Size([1, 14127]) Input IDs shape: torch.Size([1, 14127]) Labels shape: torch.Size([1, 14127]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 15079 Attention mask shape: torch.Size([1, 1, 15079, 15079]) Position ids shape: torch.Size([1, 15079]) Input IDs shape: torch.Size([1, 15079]) Labels shape: torch.Size([1, 15079]) Final batch size: 1, sequence length: 11623 Attention mask shape: torch.Size([1, 1, 11623, 11623]) Position ids shape: torch.Size([1, 11623]) Input IDs shape: torch.Size([1, 11623]) Labels shape: torch.Size([1, 11623]) Final batch size: 1, sequence length: 15308 Attention mask shape: torch.Size([1, 1, 15308, 15308]) Position ids shape: torch.Size([1, 15308]) Input IDs shape: torch.Size([1, 15308]) Labels shape: torch.Size([1, 15308]) Final batch size: 1, sequence length: 13639 Attention mask shape: torch.Size([1, 1, 13639, 13639]) Position ids shape: torch.Size([1, 13639]) Input IDs shape: torch.Size([1, 13639]) Labels shape: torch.Size([1, 13639]) Final batch size: 1, sequence length: 14827 Attention mask shape: torch.Size([1, 1, 14827, 14827]) Position ids shape: torch.Size([1, 14827]) Input IDs shape: torch.Size([1, 14827]) Labels shape: torch.Size([1, 14827]) Final batch size: 1, sequence length: 16060 Attention mask shape: torch.Size([1, 1, 16060, 16060]) Position ids shape: torch.Size([1, 16060]) Input IDs shape: torch.Size([1, 16060]) Labels shape: torch.Size([1, 16060]) Final batch size: 1, sequence length: 18953 Attention mask shape: torch.Size([1, 1, 18953, 18953]) Position ids shape: torch.Size([1, 18953]) Input IDs shape: torch.Size([1, 18953]) Labels shape: torch.Size([1, 18953]) Final batch size: 1, sequence length: 19170 Attention mask shape: torch.Size([1, 1, 19170, 19170]) Position ids shape: torch.Size([1, 19170]) Input IDs shape: torch.Size([1, 19170]) Labels shape: torch.Size([1, 19170]) Final batch size: 1, sequence length: 14648 Attention mask shape: torch.Size([1, 1, 14648, 14648]) Position ids shape: torch.Size([1, 14648]) Input IDs shape: torch.Size([1, 14648]) Labels shape: torch.Size([1, 14648]) Final batch size: 1, sequence length: 16535 Attention mask shape: torch.Size([1, 1, 16535, 16535]) Position ids shape: torch.Size([1, 16535]) Input IDs shape: torch.Size([1, 16535]) Labels shape: torch.Size([1, 16535]) Final batch size: 1, sequence length: 15478 Attention mask shape: torch.Size([1, 1, 15478, 15478]) Position ids shape: torch.Size([1, 15478]) Input IDs shape: torch.Size([1, 15478]) Labels shape: torch.Size([1, 15478]) Final batch size: 1, sequence length: 18307 Attention mask shape: torch.Size([1, 1, 18307, 18307]) Position ids shape: torch.Size([1, 18307]) Input IDs shape: torch.Size([1, 18307]) Labels shape: torch.Size([1, 18307]) Final batch size: 1, sequence length: 19138 Attention mask shape: torch.Size([1, 1, 19138, 19138]) Position ids shape: torch.Size([1, 19138]) Input IDs shape: torch.Size([1, 19138]) Labels shape: torch.Size([1, 19138]) Final batch size: 1, sequence length: 20770 Attention mask shape: torch.Size([1, 1, 20770, 20770]) Position ids shape: torch.Size([1, 20770]) Input IDs shape: torch.Size([1, 20770]) Labels shape: torch.Size([1, 20770]) Final batch size: 1, sequence length: 22561 Attention mask shape: torch.Size([1, 1, 22561, 22561]) Position ids shape: torch.Size([1, 22561]) Input IDs shape: torch.Size([1, 22561]) Labels shape: torch.Size([1, 22561]) Final batch size: 1, sequence length: 17058 Attention mask shape: torch.Size([1, 1, 17058, 17058]) Position ids shape: torch.Size([1, 17058]) Input IDs shape: torch.Size([1, 17058]) Labels shape: torch.Size([1, 17058]) Final batch size: 1, sequence length: 12006 Attention mask shape: torch.Size([1, 1, 12006, 12006]) Position ids shape: torch.Size([1, 12006]) Input IDs shape: torch.Size([1, 12006]) Labels shape: torch.Size([1, 12006]) Final batch size: 1, sequence length: 19330 Attention mask shape: torch.Size([1, 1, 19330, 19330]) Position ids shape: torch.Size([1, 19330]) Input IDs shape: torch.Size([1, 19330]) Labels shape: torch.Size([1, 19330]) Final batch size: 1, sequence length: 8839 Attention mask shape: torch.Size([1, 1, 8839, 8839]) Position ids shape: torch.Size([1, 8839]) Input IDs shape: torch.Size([1, 8839]) Labels shape: torch.Size([1, 8839]) Final batch size: 1, sequence length: 19338 Attention mask shape: torch.Size([1, 1, 19338, 19338]) Position ids shape: torch.Size([1, 19338]) Input IDs shape: torch.Size([1, 19338]) Labels shape: torch.Size([1, 19338]) Final batch size: 1, sequence length: 20714 Attention mask shape: torch.Size([1, 1, 20714, 20714]) Position ids shape: torch.Size([1, 20714]) Input IDs shape: torch.Size([1, 20714]) Labels shape: torch.Size([1, 20714]) Final batch size: 1, sequence length: 18836 Attention mask shape: torch.Size([1, 1, 18836, 18836]) Position ids shape: torch.Size([1, 18836]) Input IDs shape: torch.Size([1, 18836]) Labels shape: torch.Size([1, 18836]) Final batch size: 1, sequence length: 18527 Attention mask shape: torch.Size([1, 1, 18527, 18527]) Position ids shape: torch.Size([1, 18527]) Input IDs shape: torch.Size([1, 18527]) Labels shape: torch.Size([1, 18527]) Final batch size: 1, sequence length: 20854 Attention mask shape: torch.Size([1, 1, 20854, 20854]) Position ids shape: torch.Size([1, 20854]) Input IDs shape: torch.Size([1, 20854]) Labels shape: torch.Size([1, 20854]) Final batch size: 1, sequence length: 18325 Attention mask shape: torch.Size([1, 1, 18325, 18325]) Position ids shape: torch.Size([1, 18325]) Input IDs shape: torch.Size([1, 18325]) Labels shape: torch.Size([1, 18325]) Final batch size: 1, sequence length: 18823 Attention mask shape: torch.Size([1, 1, 18823, 18823]) Position ids shape: torch.Size([1, 18823]) Input IDs shape: torch.Size([1, 18823]) Labels shape: torch.Size([1, 18823]) Final batch size: 1, sequence length: 18395 Attention mask shape: torch.Size([1, 1, 18395, 18395]) Position ids shape: torch.Size([1, 18395]) Input IDs shape: torch.Size([1, 18395]) Labels shape: torch.Size([1, 18395]) Final batch size: 1, sequence length: 24248 Attention mask shape: torch.Size([1, 1, 24248, 24248]) Position ids shape: torch.Size([1, 24248]) Input IDs shape: torch.Size([1, 24248]) Labels shape: torch.Size([1, 24248]) Final batch size: 1, sequence length: 21982 Attention mask shape: torch.Size([1, 1, 21982, 21982]) Position ids shape: torch.Size([1, 21982]) Input IDs shape: torch.Size([1, 21982]) Labels shape: torch.Size([1, 21982]) Final batch size: 1, sequence length: 25451 Attention mask shape: torch.Size([1, 1, 25451, 25451]) Position ids shape: torch.Size([1, 25451]) Input IDs shape: torch.Size([1, 25451]) Labels shape: torch.Size([1, 25451]) Final batch size: 1, sequence length: 24433 Attention mask shape: torch.Size([1, 1, 24433, 24433]) Position ids shape: torch.Size([1, 24433]) Input IDs shape: torch.Size([1, 24433]) Labels shape: torch.Size([1, 24433]) Final batch size: 1, sequence length: 15913 Attention mask shape: torch.Size([1, 1, 15913, 15913]) Position ids shape: torch.Size([1, 15913]) Input IDs shape: torch.Size([1, 15913]) Labels shape: torch.Size([1, 15913]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 21405 Attention mask shape: torch.Size([1, 1, 21405, 21405]) Position ids shape: torch.Size([1, 21405]) Input IDs shape: torch.Size([1, 21405]) Labels shape: torch.Size([1, 21405]) Final batch size: 1, sequence length: 29098 Attention mask shape: torch.Size([1, 1, 29098, 29098]) Position ids shape: torch.Size([1, 29098]) Input IDs shape: torch.Size([1, 29098]) Labels shape: torch.Size([1, 29098]) Final batch size: 1, sequence length: 26356 Attention mask shape: torch.Size([1, 1, 26356, 26356]) Position ids shape: torch.Size([1, 26356]) Input IDs shape: torch.Size([1, 26356]) Labels shape: torch.Size([1, 26356]) Final batch size: 1, sequence length: 23560 Attention mask shape: torch.Size([1, 1, 23560, 23560]) Position ids shape: torch.Size([1, 23560]) Input IDs shape: torch.Size([1, 23560]) Labels shape: torch.Size([1, 23560]) Final batch size: 1, sequence length: 23694 Attention mask shape: torch.Size([1, 1, 23694, 23694]) Position ids shape: torch.Size([1, 23694]) Input IDs shape: torch.Size([1, 23694]) Labels shape: torch.Size([1, 23694]) Final batch size: 1, sequence length: 21858 Attention mask shape: torch.Size([1, 1, 21858, 21858]) Position ids shape: torch.Size([1, 21858]) Input IDs shape: torch.Size([1, 21858]) Labels shape: torch.Size([1, 21858]) Final batch size: 1, sequence length: 9380 Attention mask shape: torch.Size([1, 1, 9380, 9380]) Position ids shape: torch.Size([1, 9380]) Input IDs shape: torch.Size([1, 9380]) Labels shape: torch.Size([1, 9380]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 28684 Attention mask shape: torch.Size([1, 1, 28684, 28684]) Position ids shape: torch.Size([1, 28684]) Input IDs shape: torch.Size([1, 28684]) Labels shape: torch.Size([1, 28684]) Final batch size: 1, sequence length: 5801 Attention mask shape: torch.Size([1, 1, 5801, 5801]) Position ids shape: torch.Size([1, 5801]) Input IDs shape: torch.Size([1, 5801]) Labels shape: torch.Size([1, 5801]) Final batch size: 1, sequence length: 31414 Attention mask shape: torch.Size([1, 1, 31414, 31414]) Position ids shape: torch.Size([1, 31414]) Input IDs shape: torch.Size([1, 31414]) Labels shape: torch.Size([1, 31414]) Final batch size: 1, sequence length: 22735 Attention mask shape: torch.Size([1, 1, 22735, 22735]) Position ids shape: torch.Size([1, 22735]) Input IDs shape: torch.Size([1, 22735]) Labels shape: torch.Size([1, 22735]) Final batch size: 1, sequence length: 20559 Attention mask shape: torch.Size([1, 1, 20559, 20559]) Position ids shape: torch.Size([1, 20559]) Input IDs shape: torch.Size([1, 20559]) Labels shape: torch.Size([1, 20559]) Final batch size: 1, sequence length: 19045 Attention mask shape: torch.Size([1, 1, 19045, 19045]) Position ids shape: torch.Size([1, 19045]) Input IDs shape: torch.Size([1, 19045]) Labels shape: torch.Size([1, 19045]) Final batch size: 1, sequence length: 32328 Attention mask shape: torch.Size([1, 1, 32328, 32328]) Position ids shape: torch.Size([1, 32328]) Input IDs shape: torch.Size([1, 32328]) Labels shape: torch.Size([1, 32328]) Final batch size: 1, sequence length: 32885 Attention mask shape: torch.Size([1, 1, 32885, 32885]) Position ids shape: torch.Size([1, 32885]) Input IDs shape: torch.Size([1, 32885]) Labels shape: torch.Size([1, 32885]) Final batch size: 1, sequence length: 15875 Attention mask shape: torch.Size([1, 1, 15875, 15875]) Position ids shape: torch.Size([1, 15875]) Input IDs shape: torch.Size([1, 15875]) Labels shape: torch.Size([1, 15875]) Final batch size: 1, sequence length: 23881 Attention mask shape: torch.Size([1, 1, 23881, 23881]) Position ids shape: torch.Size([1, 23881]) Input IDs shape: torch.Size([1, 23881]) Labels shape: torch.Size([1, 23881]) Final batch size: 1, sequence length: 25674 Final batch size: 1, sequence length: 21615 Attention mask shape: torch.Size([1, 1, 25674, 25674]) Position ids shape: torch.Size([1, 25674]) Input IDs shape: torch.Size([1, 25674]) Attention mask shape: torch.Size([1, 1, 21615, 21615]) Labels shape: torch.Size([1, 25674]) Position ids shape: torch.Size([1, 21615]) Input IDs shape: torch.Size([1, 21615]) Labels shape: torch.Size([1, 21615]) Final batch size: 1, sequence length: 29355 Attention mask shape: torch.Size([1, 1, 29355, 29355]) Position ids shape: torch.Size([1, 29355]) Input IDs shape: torch.Size([1, 29355]) Labels shape: torch.Size([1, 29355]) Final batch size: 1, sequence length: 19239 Attention mask shape: torch.Size([1, 1, 19239, 19239]) Position ids shape: torch.Size([1, 19239]) Input IDs shape: torch.Size([1, 19239]) Labels shape: torch.Size([1, 19239]) Final batch size: 1, sequence length: 27972 Attention mask shape: torch.Size([1, 1, 27972, 27972]) Position ids shape: torch.Size([1, 27972]) Input IDs shape: torch.Size([1, 27972]) Labels shape: torch.Size([1, 27972]) Final batch size: 1, sequence length: 26072 Attention mask shape: torch.Size([1, 1, 26072, 26072]) Position ids shape: torch.Size([1, 26072]) Input IDs shape: torch.Size([1, 26072]) Labels shape: torch.Size([1, 26072]) Final batch size: 1, sequence length: 29150 Attention mask shape: torch.Size([1, 1, 29150, 29150]) Position ids shape: torch.Size([1, 29150]) Input IDs shape: torch.Size([1, 29150]) Labels shape: torch.Size([1, 29150]) Final batch size: 1, sequence length: 16050 Attention mask shape: torch.Size([1, 1, 16050, 16050]) Position ids shape: torch.Size([1, 16050]) Input IDs shape: torch.Size([1, 16050]) Labels shape: torch.Size([1, 16050]) Final batch size: 1, sequence length: 35397 Attention mask shape: torch.Size([1, 1, 35397, 35397]) Position ids shape: torch.Size([1, 35397]) Input IDs shape: torch.Size([1, 35397]) Labels shape: torch.Size([1, 35397]) Final batch size: 1, sequence length: 32529 Attention mask shape: torch.Size([1, 1, 32529, 32529]) Position ids shape: torch.Size([1, 32529]) Input IDs shape: torch.Size([1, 32529]) Labels shape: torch.Size([1, 32529]) Final batch size: 1, sequence length: 17465 Attention mask shape: torch.Size([1, 1, 17465, 17465]) Position ids shape: torch.Size([1, 17465]) Input IDs shape: torch.Size([1, 17465]) Labels shape: torch.Size([1, 17465]) Final batch size: 1, sequence length: 31860 Attention mask shape: torch.Size([1, 1, 31860, 31860]) Position ids shape: torch.Size([1, 31860]) Input IDs shape: torch.Size([1, 31860]) Labels shape: torch.Size([1, 31860]) Final batch size: 1, sequence length: 30687 Attention mask shape: torch.Size([1, 1, 30687, 30687]) Position ids shape: torch.Size([1, 30687]) Input IDs shape: torch.Size([1, 30687]) Labels shape: torch.Size([1, 30687]) Final batch size: 1, sequence length: 30689 Attention mask shape: torch.Size([1, 1, 30689, 30689]) Position ids shape: torch.Size([1, 30689]) Input IDs shape: torch.Size([1, 30689]) Labels shape: torch.Size([1, 30689]) Final batch size: 1, sequence length: 33871 Attention mask shape: torch.Size([1, 1, 33871, 33871]) Position ids shape: torch.Size([1, 33871]) Input IDs shape: torch.Size([1, 33871]) Labels shape: torch.Size([1, 33871]) Final batch size: 1, sequence length: 30346 Attention mask shape: torch.Size([1, 1, 30346, 30346]) Position ids shape: torch.Size([1, 30346]) Input IDs shape: torch.Size([1, 30346]) Labels shape: torch.Size([1, 30346]) Final batch size: 1, sequence length: 26689 Attention mask shape: torch.Size([1, 1, 26689, 26689]) Position ids shape: torch.Size([1, 26689]) Input IDs shape: torch.Size([1, 26689]) Labels shape: torch.Size([1, 26689]) Final batch size: 1, sequence length: 15077 Attention mask shape: torch.Size([1, 1, 15077, 15077]) Position ids shape: torch.Size([1, 15077]) Input IDs shape: torch.Size([1, 15077]) Labels shape: torch.Size([1, 15077]) Final batch size: 1, sequence length: 31712 Attention mask shape: torch.Size([1, 1, 31712, 31712]) Position ids shape: torch.Size([1, 31712]) Input IDs shape: torch.Size([1, 31712]) Labels shape: torch.Size([1, 31712]) Final batch size: 1, sequence length: 22264 Attention mask shape: torch.Size([1, 1, 22264, 22264]) Position ids shape: torch.Size([1, 22264]) Input IDs shape: torch.Size([1, 22264]) Labels shape: torch.Size([1, 22264]) Final batch size: 1, sequence length: 24781 Attention mask shape: torch.Size([1, 1, 24781, 24781]) Position ids shape: torch.Size([1, 24781]) Input IDs shape: torch.Size([1, 24781]) Labels shape: torch.Size([1, 24781]) Final batch size: 1, sequence length: 34921 Attention mask shape: torch.Size([1, 1, 34921, 34921]) Position ids shape: torch.Size([1, 34921]) Input IDs shape: torch.Size([1, 34921]) Labels shape: torch.Size([1, 34921]) Final batch size: 1, sequence length: 32636 Attention mask shape: torch.Size([1, 1, 32636, 32636]) Position ids shape: torch.Size([1, 32636]) Input IDs shape: torch.Size([1, 32636]) Labels shape: torch.Size([1, 32636]) Final batch size: 1, sequence length: 18963 Attention mask shape: torch.Size([1, 1, 18963, 18963]) Position ids shape: torch.Size([1, 18963]) Input IDs shape: torch.Size([1, 18963]) Labels shape: torch.Size([1, 18963]) Final batch size: 1, sequence length: 34581 Attention mask shape: torch.Size([1, 1, 34581, 34581]) Position ids shape: torch.Size([1, 34581]) Input IDs shape: torch.Size([1, 34581]) Labels shape: torch.Size([1, 34581]) Final batch size: 1, sequence length: 34853 Attention mask shape: torch.Size([1, 1, 34853, 34853]) Position ids shape: torch.Size([1, 34853]) Input IDs shape: torch.Size([1, 34853]) Labels shape: torch.Size([1, 34853]) Final batch size: 1, sequence length: 25741 Attention mask shape: torch.Size([1, 1, 25741, 25741]) Position ids shape: torch.Size([1, 25741]) Input IDs shape: torch.Size([1, 25741]) Labels shape: torch.Size([1, 25741]) Final batch size: 1, sequence length: 18177 Attention mask shape: torch.Size([1, 1, 18177, 18177]) Position ids shape: torch.Size([1, 18177]) Input IDs shape: torch.Size([1, 18177]) Labels shape: torch.Size([1, 18177]) Final batch size: 1, sequence length: 28018 Attention mask shape: torch.Size([1, 1, 28018, 28018]) Position ids shape: torch.Size([1, 28018]) Input IDs shape: torch.Size([1, 28018]) Labels shape: torch.Size([1, 28018]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 14976 Attention mask shape: torch.Size([1, 1, 14976, 14976]) Position ids shape: torch.Size([1, 14976]) Input IDs shape: torch.Size([1, 14976]) Labels shape: torch.Size([1, 14976]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17243 Attention mask shape: torch.Size([1, 1, 17243, 17243]) Position ids shape: torch.Size([1, 17243]) Input IDs shape: torch.Size([1, 17243]) Labels shape: torch.Size([1, 17243]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 24002 Attention mask shape: torch.Size([1, 1, 24002, 24002]) Position ids shape: torch.Size([1, 24002]) Input IDs shape: torch.Size([1, 24002]) Labels shape: torch.Size([1, 24002]) Final batch size: 1, sequence length: 26922 Attention mask shape: torch.Size([1, 1, 26922, 26922]) Position ids shape: torch.Size([1, 26922]) Input IDs shape: torch.Size([1, 26922]) Labels shape: torch.Size([1, 26922]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 12218 Attention mask shape: torch.Size([1, 1, 12218, 12218]) Position ids shape: torch.Size([1, 12218]) Input IDs shape: torch.Size([1, 12218]) Labels shape: torch.Size([1, 12218]) Final batch size: 1, sequence length: 31022 Attention mask shape: torch.Size([1, 1, 31022, 31022]) Position ids shape: torch.Size([1, 31022]) Input IDs shape: torch.Size([1, 31022]) Labels shape: torch.Size([1, 31022]) Final batch size: 1, sequence length: 16025 Attention mask shape: torch.Size([1, 1, 16025, 16025]) Position ids shape: torch.Size([1, 16025]) Input IDs shape: torch.Size([1, 16025]) Labels shape: torch.Size([1, 16025]) Final batch size: 1, sequence length: 40488 Attention mask shape: torch.Size([1, 1, 40488, 40488]) Position ids shape: torch.Size([1, 40488]) Input IDs shape: torch.Size([1, 40488]) Labels shape: torch.Size([1, 40488]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 28397 Attention mask shape: torch.Size([1, 1, 28397, 28397]) Position ids shape: torch.Size([1, 28397]) Input IDs shape: torch.Size([1, 28397]) Labels shape: torch.Size([1, 28397]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 11611 Attention mask shape: torch.Size([1, 1, 11611, 11611]) Position ids shape: torch.Size([1, 11611]) Input IDs shape: torch.Size([1, 11611]) Labels shape: torch.Size([1, 11611]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36596 Attention mask shape: torch.Size([1, 1, 36596, 36596]) Position ids shape: torch.Size([1, 36596]) Input IDs shape: torch.Size([1, 36596]) Labels shape: torch.Size([1, 36596]) Final batch size: 1, sequence length: 24043 Attention mask shape: torch.Size([1, 1, 24043, 24043]) Position ids shape: torch.Size([1, 24043]) Input IDs shape: torch.Size([1, 24043]) Labels shape: torch.Size([1, 24043]) Final batch size: 1, sequence length: 28086 Attention mask shape: torch.Size([1, 1, 28086, 28086]) Position ids shape: torch.Size([1, 28086]) Input IDs shape: torch.Size([1, 28086]) Labels shape: torch.Size([1, 28086]) Final batch size: 1, sequence length: 16337 Attention mask shape: torch.Size([1, 1, 16337, 16337]) Position ids shape: torch.Size([1, 16337]) Input IDs shape: torch.Size([1, 16337]) Labels shape: torch.Size([1, 16337]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36334 Attention mask shape: torch.Size([1, 1, 36334, 36334]) Position ids shape: torch.Size([1, 36334]) Input IDs shape: torch.Size([1, 36334]) Labels shape: torch.Size([1, 36334]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 32917 Attention mask shape: torch.Size([1, 1, 32917, 32917]) Position ids shape: torch.Size([1, 32917]) Input IDs shape: torch.Size([1, 32917]) Labels shape: torch.Size([1, 32917]) {'loss': 0.2553, 'grad_norm': 0.1585067514426352, 'learning_rate': 2.447174185242324e-07, 'num_tokens': -inf, 'epoch': 7.38} Final batch size: 1, sequence length: 6215 Attention mask shape: torch.Size([1, 1, 6215, 6215]) Position ids shape: torch.Size([1, 6215]) Input IDs shape: torch.Size([1, 6215]) Labels shape: torch.Size([1, 6215]) Final batch size: 1, sequence length: 8623 Attention mask shape: torch.Size([1, 1, 8623, 8623]) Position ids shape: torch.Size([1, 8623]) Input IDs shape: torch.Size([1, 8623]) Labels shape: torch.Size([1, 8623]) Final batch size: 1, sequence length: 6871 Attention mask shape: torch.Size([1, 1, 6871, 6871]) Position ids shape: torch.Size([1, 6871]) Input IDs shape: torch.Size([1, 6871]) Labels shape: torch.Size([1, 6871]) Final batch size: 1, sequence length: 10107 Attention mask shape: torch.Size([1, 1, 10107, 10107]) Position ids shape: torch.Size([1, 10107]) Input IDs shape: torch.Size([1, 10107]) Labels shape: torch.Size([1, 10107]) Final batch size: 1, sequence length: 5917 Attention mask shape: torch.Size([1, 1, 5917, 5917]) Position ids shape: torch.Size([1, 5917]) Input IDs shape: torch.Size([1, 5917]) Labels shape: torch.Size([1, 5917]) Final batch size: 1, sequence length: 6034 Attention mask shape: torch.Size([1, 1, 6034, 6034]) Position ids shape: torch.Size([1, 6034]) Input IDs shape: torch.Size([1, 6034]) Labels shape: torch.Size([1, 6034]) Final batch size: 1, sequence length: 11616 Attention mask shape: torch.Size([1, 1, 11616, 11616]) Position ids shape: torch.Size([1, 11616]) Input IDs shape: torch.Size([1, 11616]) Labels shape: torch.Size([1, 11616]) Final batch size: 1, sequence length: 5818 Attention mask shape: torch.Size([1, 1, 5818, 5818]) Position ids shape: torch.Size([1, 5818]) Input IDs shape: torch.Size([1, 5818]) Labels shape: torch.Size([1, 5818]) Final batch size: 1, sequence length: 12275 Attention mask shape: torch.Size([1, 1, 12275, 12275]) Position ids shape: torch.Size([1, 12275]) Input IDs shape: torch.Size([1, 12275]) Labels shape: torch.Size([1, 12275]) Final batch size: 1, sequence length: 12317 Attention mask shape: torch.Size([1, 1, 12317, 12317]) Position ids shape: torch.Size([1, 12317]) Input IDs shape: torch.Size([1, 12317]) Labels shape: torch.Size([1, 12317]) Final batch size: 1, sequence length: 13202 Attention mask shape: torch.Size([1, 1, 13202, 13202]) Position ids shape: torch.Size([1, 13202]) Input IDs shape: torch.Size([1, 13202]) Labels shape: torch.Size([1, 13202]) Final batch size: 1, sequence length: 9517 Attention mask shape: torch.Size([1, 1, 9517, 9517]) Position ids shape: torch.Size([1, 9517]) Input IDs shape: torch.Size([1, 9517]) Labels shape: torch.Size([1, 9517]) Final batch size: 1, sequence length: 15029 Attention mask shape: torch.Size([1, 1, 15029, 15029]) Position ids shape: torch.Size([1, 15029]) Input IDs shape: torch.Size([1, 15029]) Labels shape: torch.Size([1, 15029]) Final batch size: 1, sequence length: 8117 Attention mask shape: torch.Size([1, 1, 8117, 8117]) Position ids shape: torch.Size([1, 8117]) Input IDs shape: torch.Size([1, 8117]) Labels shape: torch.Size([1, 8117]) Final batch size: 1, sequence length: 15499 Attention mask shape: torch.Size([1, 1, 15499, 15499]) Position ids shape: torch.Size([1, 15499]) Input IDs shape: torch.Size([1, 15499]) Labels shape: torch.Size([1, 15499]) Final batch size: 1, sequence length: 12813 Attention mask shape: torch.Size([1, 1, 12813, 12813]) Position ids shape: torch.Size([1, 12813]) Input IDs shape: torch.Size([1, 12813]) Labels shape: torch.Size([1, 12813]) Final batch size: 1, sequence length: 13428 Attention mask shape: torch.Size([1, 1, 13428, 13428]) Position ids shape: torch.Size([1, 13428]) Input IDs shape: torch.Size([1, 13428]) Labels shape: torch.Size([1, 13428]) Final batch size: 1, sequence length: 16299 Attention mask shape: torch.Size([1, 1, 16299, 16299]) Position ids shape: torch.Size([1, 16299]) Input IDs shape: torch.Size([1, 16299]) Labels shape: torch.Size([1, 16299]) Final batch size: 1, sequence length: 10136 Attention mask shape: torch.Size([1, 1, 10136, 10136]) Position ids shape: torch.Size([1, 10136]) Input IDs shape: torch.Size([1, 10136]) Labels shape: torch.Size([1, 10136]) Final batch size: 1, sequence length: 11648 Attention mask shape: torch.Size([1, 1, 11648, 11648]) Position ids shape: torch.Size([1, 11648]) Input IDs shape: torch.Size([1, 11648]) Labels shape: torch.Size([1, 11648]) Final batch size: 1, sequence length: 16331 Attention mask shape: torch.Size([1, 1, 16331, 16331]) Position ids shape: torch.Size([1, 16331]) Input IDs shape: torch.Size([1, 16331]) Labels shape: torch.Size([1, 16331]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 17587 Attention mask shape: torch.Size([1, 1, 17587, 17587]) Position ids shape: torch.Size([1, 17587]) Input IDs shape: torch.Size([1, 17587]) Labels shape: torch.Size([1, 17587]) Final batch size: 1, sequence length: 15066 Attention mask shape: torch.Size([1, 1, 15066, 15066]) Position ids shape: torch.Size([1, 15066]) Input IDs shape: torch.Size([1, 15066]) Labels shape: torch.Size([1, 15066]) Final batch size: 1, sequence length: 18414 Attention mask shape: torch.Size([1, 1, 18414, 18414]) Position ids shape: torch.Size([1, 18414]) Input IDs shape: torch.Size([1, 18414]) Labels shape: torch.Size([1, 18414]) Final batch size: 1, sequence length: 12553 Attention mask shape: torch.Size([1, 1, 12553, 12553]) Position ids shape: torch.Size([1, 12553]) Input IDs shape: torch.Size([1, 12553]) Labels shape: torch.Size([1, 12553]) Final batch size: 1, sequence length: 19221 Attention mask shape: torch.Size([1, 1, 19221, 19221]) Position ids shape: torch.Size([1, 19221]) Input IDs shape: torch.Size([1, 19221]) Labels shape: torch.Size([1, 19221]) Final batch size: 1, sequence length: 21028 Attention mask shape: torch.Size([1, 1, 21028, 21028]) Position ids shape: torch.Size([1, 21028]) Input IDs shape: torch.Size([1, 21028]) Labels shape: torch.Size([1, 21028]) Final batch size: 1, sequence length: 21556 Attention mask shape: torch.Size([1, 1, 21556, 21556]) Position ids shape: torch.Size([1, 21556]) Input IDs shape: torch.Size([1, 21556]) Labels shape: torch.Size([1, 21556]) Final batch size: 1, sequence length: 22079 Attention mask shape: torch.Size([1, 1, 22079, 22079]) Position ids shape: torch.Size([1, 22079]) Input IDs shape: torch.Size([1, 22079]) Labels shape: torch.Size([1, 22079]) Final batch size: 1, sequence length: 10269 Attention mask shape: torch.Size([1, 1, 10269, 10269]) Position ids shape: torch.Size([1, 10269]) Input IDs shape: torch.Size([1, 10269]) Labels shape: torch.Size([1, 10269]) Final batch size: 1, sequence length: 22618 Attention mask shape: torch.Size([1, 1, 22618, 22618]) Position ids shape: torch.Size([1, 22618]) Input IDs shape: torch.Size([1, 22618]) Labels shape: torch.Size([1, 22618]) Final batch size: 1, sequence length: 19847 Attention mask shape: torch.Size([1, 1, 19847, 19847]) Position ids shape: torch.Size([1, 19847]) Input IDs shape: torch.Size([1, 19847]) Labels shape: torch.Size([1, 19847]) Final batch size: 1, sequence length: 22138 Attention mask shape: torch.Size([1, 1, 22138, 22138]) Position ids shape: torch.Size([1, 22138]) Input IDs shape: torch.Size([1, 22138]) Labels shape: torch.Size([1, 22138]) Final batch size: 1, sequence length: 19512 Attention mask shape: torch.Size([1, 1, 19512, 19512]) Position ids shape: torch.Size([1, 19512]) Input IDs shape: torch.Size([1, 19512]) Labels shape: torch.Size([1, 19512]) Final batch size: 1, sequence length: 15221 Attention mask shape: torch.Size([1, 1, 15221, 15221]) Position ids shape: torch.Size([1, 15221]) Input IDs shape: torch.Size([1, 15221]) Labels shape: torch.Size([1, 15221]) Final batch size: 1, sequence length: 18438 Attention mask shape: torch.Size([1, 1, 18438, 18438]) Position ids shape: torch.Size([1, 18438]) Input IDs shape: torch.Size([1, 18438]) Labels shape: torch.Size([1, 18438]) Final batch size: 1, sequence length: 23766 Attention mask shape: torch.Size([1, 1, 23766, 23766]) Position ids shape: torch.Size([1, 23766]) Input IDs shape: torch.Size([1, 23766]) Labels shape: torch.Size([1, 23766]) Final batch size: 1, sequence length: 15226 Attention mask shape: torch.Size([1, 1, 15226, 15226]) Position ids shape: torch.Size([1, 15226]) Input IDs shape: torch.Size([1, 15226]) Labels shape: torch.Size([1, 15226]) Final batch size: 1, sequence length: 24499 Attention mask shape: torch.Size([1, 1, 24499, 24499]) Position ids shape: torch.Size([1, 24499]) Input IDs shape: torch.Size([1, 24499]) Labels shape: torch.Size([1, 24499]) Final batch size: 1, sequence length: 18469 Attention mask shape: torch.Size([1, 1, 18469, 18469]) Position ids shape: torch.Size([1, 18469]) Input IDs shape: torch.Size([1, 18469]) Labels shape: torch.Size([1, 18469]) Final batch size: 1, sequence length: 22718 Attention mask shape: torch.Size([1, 1, 22718, 22718]) Position ids shape: torch.Size([1, 22718]) Input IDs shape: torch.Size([1, 22718]) Labels shape: torch.Size([1, 22718]) Final batch size: 1, sequence length: 23973 Attention mask shape: torch.Size([1, 1, 23973, 23973]) Position ids shape: torch.Size([1, 23973]) Input IDs shape: torch.Size([1, 23973]) Labels shape: torch.Size([1, 23973]) Final batch size: 1, sequence length: 7584 Attention mask shape: torch.Size([1, 1, 7584, 7584]) Position ids shape: torch.Size([1, 7584]) Input IDs shape: torch.Size([1, 7584]) Labels shape: torch.Size([1, 7584]) Final batch size: 1, sequence length: 25999 Attention mask shape: torch.Size([1, 1, 25999, 25999]) Position ids shape: torch.Size([1, 25999]) Input IDs shape: torch.Size([1, 25999]) Labels shape: torch.Size([1, 25999]) Final batch size: 1, sequence length: 19428 Attention mask shape: torch.Size([1, 1, 19428, 19428]) Position ids shape: torch.Size([1, 19428]) Input IDs shape: torch.Size([1, 19428]) Labels shape: torch.Size([1, 19428]) Final batch size: 1, sequence length: 24694 Attention mask shape: torch.Size([1, 1, 24694, 24694]) Position ids shape: torch.Size([1, 24694]) Input IDs shape: torch.Size([1, 24694]) Labels shape: torch.Size([1, 24694]) Final batch size: 1, sequence length: 22235 Attention mask shape: torch.Size([1, 1, 22235, 22235]) Position ids shape: torch.Size([1, 22235]) Input IDs shape: torch.Size([1, 22235]) Labels shape: torch.Size([1, 22235]) Final batch size: 1, sequence length: 28497 Attention mask shape: torch.Size([1, 1, 28497, 28497]) Position ids shape: torch.Size([1, 28497]) Input IDs shape: torch.Size([1, 28497]) Labels shape: torch.Size([1, 28497]) Final batch size: 1, sequence length: 15184 Attention mask shape: torch.Size([1, 1, 15184, 15184]) Position ids shape: torch.Size([1, 15184]) Input IDs shape: torch.Size([1, 15184]) Labels shape: torch.Size([1, 15184]) Final batch size: 1, sequence length: 22763 Attention mask shape: torch.Size([1, 1, 22763, 22763]) Position ids shape: torch.Size([1, 22763]) Input IDs shape: torch.Size([1, 22763]) Labels shape: torch.Size([1, 22763]) Final batch size: 1, sequence length: 28749 Attention mask shape: torch.Size([1, 1, 28749, 28749]) Position ids shape: torch.Size([1, 28749]) Input IDs shape: torch.Size([1, 28749]) Labels shape: torch.Size([1, 28749]) Final batch size: 1, sequence length: 17376 Attention mask shape: torch.Size([1, 1, 17376, 17376]) Position ids shape: torch.Size([1, 17376]) Input IDs shape: torch.Size([1, 17376]) Labels shape: torch.Size([1, 17376]) Final batch size: 1, sequence length: 21826 Attention mask shape: torch.Size([1, 1, 21826, 21826]) Position ids shape: torch.Size([1, 21826]) Input IDs shape: torch.Size([1, 21826]) Labels shape: torch.Size([1, 21826]) Final batch size: 1, sequence length: 14784 Attention mask shape: torch.Size([1, 1, 14784, 14784]) Position ids shape: torch.Size([1, 14784]) Input IDs shape: torch.Size([1, 14784]) Labels shape: torch.Size([1, 14784]) Final batch size: 1, sequence length: 25622 Attention mask shape: torch.Size([1, 1, 25622, 25622]) Position ids shape: torch.Size([1, 25622]) Input IDs shape: torch.Size([1, 25622]) Labels shape: torch.Size([1, 25622]) Final batch size: 1, sequence length: 24909 Attention mask shape: torch.Size([1, 1, 24909, 24909]) Position ids shape: torch.Size([1, 24909]) Input IDs shape: torch.Size([1, 24909]) Labels shape: torch.Size([1, 24909]) Final batch size: 1, sequence length: 27566 Attention mask shape: torch.Size([1, 1, 27566, 27566]) Position ids shape: torch.Size([1, 27566]) Input IDs shape: torch.Size([1, 27566]) Labels shape: torch.Size([1, 27566]) Final batch size: 1, sequence length: 30356 Attention mask shape: torch.Size([1, 1, 30356, 30356]) Position ids shape: torch.Size([1, 30356]) Input IDs shape: torch.Size([1, 30356]) Labels shape: torch.Size([1, 30356]) Final batch size: 1, sequence length: 27327 Attention mask shape: torch.Size([1, 1, 27327, 27327]) Position ids shape: torch.Size([1, 27327]) Input IDs shape: torch.Size([1, 27327]) Labels shape: torch.Size([1, 27327]) Final batch size: 1, sequence length: 23851 Attention mask shape: torch.Size([1, 1, 23851, 23851]) Position ids shape: torch.Size([1, 23851]) Input IDs shape: torch.Size([1, 23851]) Labels shape: torch.Size([1, 23851]) Final batch size: 1, sequence length: 22366 Attention mask shape: torch.Size([1, 1, 22366, 22366]) Position ids shape: torch.Size([1, 22366]) Input IDs shape: torch.Size([1, 22366]) Labels shape: torch.Size([1, 22366]) Final batch size: 1, sequence length: 31714 Attention mask shape: torch.Size([1, 1, 31714, 31714]) Position ids shape: torch.Size([1, 31714]) Input IDs shape: torch.Size([1, 31714]) Labels shape: torch.Size([1, 31714]) Final batch size: 1, sequence length: 21034 Attention mask shape: torch.Size([1, 1, 21034, 21034]) Position ids shape: torch.Size([1, 21034]) Input IDs shape: torch.Size([1, 21034]) Labels shape: torch.Size([1, 21034]) Final batch size: 1, sequence length: 11795 Attention mask shape: torch.Size([1, 1, 11795, 11795]) Position ids shape: torch.Size([1, 11795]) Input IDs shape: torch.Size([1, 11795]) Labels shape: torch.Size([1, 11795]) Final batch size: 1, sequence length: 34673 Attention mask shape: torch.Size([1, 1, 34673, 34673]) Position ids shape: torch.Size([1, 34673]) Input IDs shape: torch.Size([1, 34673]) Labels shape: torch.Size([1, 34673]) Final batch size: 1, sequence length: 31903 Attention mask shape: torch.Size([1, 1, 31903, 31903]) Position ids shape: torch.Size([1, 31903]) Input IDs shape: torch.Size([1, 31903]) Labels shape: torch.Size([1, 31903]) Final batch size: 1, sequence length: 27489 Attention mask shape: torch.Size([1, 1, 27489, 27489]) Position ids shape: torch.Size([1, 27489]) Input IDs shape: torch.Size([1, 27489]) Labels shape: torch.Size([1, 27489]) Final batch size: 1, sequence length: 36777 Attention mask shape: torch.Size([1, 1, 36777, 36777]) Position ids shape: torch.Size([1, 36777]) Input IDs shape: torch.Size([1, 36777]) Labels shape: torch.Size([1, 36777]) Final batch size: 1, sequence length: 15588 Attention mask shape: torch.Size([1, 1, 15588, 15588]) Position ids shape: torch.Size([1, 15588]) Input IDs shape: torch.Size([1, 15588]) Labels shape: torch.Size([1, 15588]) Final batch size: 1, sequence length: 37285 Attention mask shape: torch.Size([1, 1, 37285, 37285]) Position ids shape: torch.Size([1, 37285]) Input IDs shape: torch.Size([1, 37285]) Labels shape: torch.Size([1, 37285]) Final batch size: 1, sequence length: 36723 Attention mask shape: torch.Size([1, 1, 36723, 36723]) Position ids shape: torch.Size([1, 36723]) Input IDs shape: torch.Size([1, 36723]) Labels shape: torch.Size([1, 36723]) Final batch size: 1, sequence length: 38848 Attention mask shape: torch.Size([1, 1, 38848, 38848]) Position ids shape: torch.Size([1, 38848]) Input IDs shape: torch.Size([1, 38848]) Labels shape: torch.Size([1, 38848]) Final batch size: 1, sequence length: 28176 Attention mask shape: torch.Size([1, 1, 28176, 28176]) Position ids shape: torch.Size([1, 28176]) Input IDs shape: torch.Size([1, 28176]) Labels shape: torch.Size([1, 28176]) Final batch size: 1, sequence length: 40325 Attention mask shape: torch.Size([1, 1, 40325, 40325]) Position ids shape: torch.Size([1, 40325]) Input IDs shape: torch.Size([1, 40325]) Labels shape: torch.Size([1, 40325]) Final batch size: 1, sequence length: 24801 Attention mask shape: torch.Size([1, 1, 24801, 24801]) Position ids shape: torch.Size([1, 24801]) Input IDs shape: torch.Size([1, 24801]) Labels shape: torch.Size([1, 24801]) Final batch size: 1, sequence length: 25946 Attention mask shape: torch.Size([1, 1, 25946, 25946]) Position ids shape: torch.Size([1, 25946]) Input IDs shape: torch.Size([1, 25946]) Labels shape: torch.Size([1, 25946]) Final batch size: 1, sequence length: 38049 Attention mask shape: torch.Size([1, 1, 38049, 38049]) Position ids shape: torch.Size([1, 38049]) Input IDs shape: torch.Size([1, 38049]) Labels shape: torch.Size([1, 38049]) Final batch size: 1, sequence length: 39253 Attention mask shape: torch.Size([1, 1, 39253, 39253]) Position ids shape: torch.Size([1, 39253]) Input IDs shape: torch.Size([1, 39253]) Labels shape: torch.Size([1, 39253]) Final batch size: 1, sequence length: 6882 Attention mask shape: torch.Size([1, 1, 6882, 6882]) Position ids shape: torch.Size([1, 6882]) Input IDs shape: torch.Size([1, 6882]) Labels shape: torch.Size([1, 6882]) Final batch size: 1, sequence length: 38104 Attention mask shape: torch.Size([1, 1, 38104, 38104]) Position ids shape: torch.Size([1, 38104]) Input IDs shape: torch.Size([1, 38104]) Labels shape: torch.Size([1, 38104]) Final batch size: 1, sequence length: 20755 Attention mask shape: torch.Size([1, 1, 20755, 20755]) Position ids shape: torch.Size([1, 20755]) Input IDs shape: torch.Size([1, 20755]) Labels shape: torch.Size([1, 20755]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 35904 Attention mask shape: torch.Size([1, 1, 35904, 35904]) Position ids shape: torch.Size([1, 35904]) Input IDs shape: torch.Size([1, 35904]) Labels shape: torch.Size([1, 35904]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 34289 Attention mask shape: torch.Size([1, 1, 34289, 34289]) Position ids shape: torch.Size([1, 34289]) Input IDs shape: torch.Size([1, 34289]) Labels shape: torch.Size([1, 34289]) Final batch size: 1, sequence length: 40317 Attention mask shape: torch.Size([1, 1, 40317, 40317]) Position ids shape: torch.Size([1, 40317]) Input IDs shape: torch.Size([1, 40317]) Labels shape: torch.Size([1, 40317]) Final batch size: 1, sequence length: 36786 Attention mask shape: torch.Size([1, 1, 36786, 36786]) Position ids shape: torch.Size([1, 36786]) Input IDs shape: torch.Size([1, 36786]) Labels shape: torch.Size([1, 36786]) Final batch size: 1, sequence length: 40937 Attention mask shape: torch.Size([1, 1, 40937, 40937]) Position ids shape: torch.Size([1, 40937]) Input IDs shape: torch.Size([1, 40937]) Labels shape: torch.Size([1, 40937]) Final batch size: 1, sequence length: 40657 Attention mask shape: torch.Size([1, 1, 40657, 40657]) Position ids shape: torch.Size([1, 40657]) Input IDs shape: torch.Size([1, 40657]) Labels shape: torch.Size([1, 40657]) Final batch size: 1, sequence length: 10469 Attention mask shape: torch.Size([1, 1, 10469, 10469]) Position ids shape: torch.Size([1, 10469]) Input IDs shape: torch.Size([1, 10469]) Labels shape: torch.Size([1, 10469]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19639 Attention mask shape: torch.Size([1, 1, 19639, 19639]) Position ids shape: torch.Size([1, 19639]) Input IDs shape: torch.Size([1, 19639]) Labels shape: torch.Size([1, 19639]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 16649 Attention mask shape: torch.Size([1, 1, 16649, 16649]) Position ids shape: torch.Size([1, 16649]) Input IDs shape: torch.Size([1, 16649]) Labels shape: torch.Size([1, 16649]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19437 Attention mask shape: torch.Size([1, 1, 19437, 19437]) Position ids shape: torch.Size([1, 19437]) Input IDs shape: torch.Size([1, 19437]) Labels shape: torch.Size([1, 19437]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36317 Attention mask shape: torch.Size([1, 1, 36317, 36317]) Position ids shape: torch.Size([1, 36317]) Input IDs shape: torch.Size([1, 36317]) Labels shape: torch.Size([1, 36317]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17882 Attention mask shape: torch.Size([1, 1, 17882, 17882]) Position ids shape: torch.Size([1, 17882]) Input IDs shape: torch.Size([1, 17882]) Labels shape: torch.Size([1, 17882]) Final batch size: 1, sequence length: 36128 Attention mask shape: torch.Size([1, 1, 36128, 36128]) Position ids shape: torch.Size([1, 36128]) Input IDs shape: torch.Size([1, 36128]) Labels shape: torch.Size([1, 36128]) Final batch size: 1, sequence length: 34540 Attention mask shape: torch.Size([1, 1, 34540, 34540]) Position ids shape: torch.Size([1, 34540]) Input IDs shape: torch.Size([1, 34540]) Labels shape: torch.Size([1, 34540]) Final batch size: 1, sequence length: 25963 Attention mask shape: torch.Size([1, 1, 25963, 25963]) Position ids shape: torch.Size([1, 25963]) Input IDs shape: torch.Size([1, 25963]) Labels shape: torch.Size([1, 25963]) Final batch size: 1, sequence length: 26660 Attention mask shape: torch.Size([1, 1, 26660, 26660]) Position ids shape: torch.Size([1, 26660]) Input IDs shape: torch.Size([1, 26660]) Labels shape: torch.Size([1, 26660]) Final batch size: 1, sequence length: 27947 Attention mask shape: torch.Size([1, 1, 27947, 27947]) Position ids shape: torch.Size([1, 27947]) Input IDs shape: torch.Size([1, 27947]) Labels shape: torch.Size([1, 27947]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17379 Attention mask shape: torch.Size([1, 1, 17379, 17379]) Position ids shape: torch.Size([1, 17379]) Input IDs shape: torch.Size([1, 17379]) Labels shape: torch.Size([1, 17379]) Final batch size: 1, sequence length: 38686 Attention mask shape: torch.Size([1, 1, 38686, 38686]) Position ids shape: torch.Size([1, 38686]) Input IDs shape: torch.Size([1, 38686]) Labels shape: torch.Size([1, 38686]) Final batch size: 1, sequence length: 21547 Attention mask shape: torch.Size([1, 1, 21547, 21547]) Position ids shape: torch.Size([1, 21547]) Input IDs shape: torch.Size([1, 21547]) Labels shape: torch.Size([1, 21547]) Final batch size: 1, sequence length: 20843 Attention mask shape: torch.Size([1, 1, 20843, 20843]) Position ids shape: torch.Size([1, 20843]) Input IDs shape: torch.Size([1, 20843]) Labels shape: torch.Size([1, 20843]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 21496 Attention mask shape: torch.Size([1, 1, 21496, 21496]) Position ids shape: torch.Size([1, 21496]) Input IDs shape: torch.Size([1, 21496]) Labels shape: torch.Size([1, 21496]) Final batch size: 1, sequence length: 24551 Attention mask shape: torch.Size([1, 1, 24551, 24551]) Position ids shape: torch.Size([1, 24551]) Input IDs shape: torch.Size([1, 24551]) Labels shape: torch.Size([1, 24551]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 37946 Attention mask shape: torch.Size([1, 1, 37946, 37946]) Position ids shape: torch.Size([1, 37946]) Input IDs shape: torch.Size([1, 37946]) Labels shape: torch.Size([1, 37946]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 7681 Attention mask shape: torch.Size([1, 1, 7681, 7681]) Position ids shape: torch.Size([1, 7681]) Input IDs shape: torch.Size([1, 7681]) Labels shape: torch.Size([1, 7681]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2592, 'grad_norm': 0.15288623816707295, 'learning_rate': 1.7037086855465902e-07, 'num_tokens': -inf, 'epoch': 7.5} Final batch size: 1, sequence length: 5525 Attention mask shape: torch.Size([1, 1, 5525, 5525]) Position ids shape: torch.Size([1, 5525]) Input IDs shape: torch.Size([1, 5525]) Labels shape: torch.Size([1, 5525]) Final batch size: 1, sequence length: 5328 Attention mask shape: torch.Size([1, 1, 5328, 5328]) Position ids shape: torch.Size([1, 5328]) Input IDs shape: torch.Size([1, 5328]) Labels shape: torch.Size([1, 5328]) Final batch size: 1, sequence length: 10273 Attention mask shape: torch.Size([1, 1, 10273, 10273]) Position ids shape: torch.Size([1, 10273]) Input IDs shape: torch.Size([1, 10273]) Labels shape: torch.Size([1, 10273]) Final batch size: 1, sequence length: 11464 Attention mask shape: torch.Size([1, 1, 11464, 11464]) Position ids shape: torch.Size([1, 11464]) Input IDs shape: torch.Size([1, 11464]) Labels shape: torch.Size([1, 11464]) Final batch size: 1, sequence length: 12419 Attention mask shape: torch.Size([1, 1, 12419, 12419]) Position ids shape: torch.Size([1, 12419]) Input IDs shape: torch.Size([1, 12419]) Labels shape: torch.Size([1, 12419]) Final batch size: 1, sequence length: 10434 Attention mask shape: torch.Size([1, 1, 10434, 10434]) Position ids shape: torch.Size([1, 10434]) Input IDs shape: torch.Size([1, 10434]) Labels shape: torch.Size([1, 10434]) Final batch size: 1, sequence length: 13607 Attention mask shape: torch.Size([1, 1, 13607, 13607]) Position ids shape: torch.Size([1, 13607]) Input IDs shape: torch.Size([1, 13607]) Labels shape: torch.Size([1, 13607]) Final batch size: 1, sequence length: 9648 Attention mask shape: torch.Size([1, 1, 9648, 9648]) Position ids shape: torch.Size([1, 9648]) Input IDs shape: torch.Size([1, 9648]) Labels shape: torch.Size([1, 9648]) Final batch size: 1, sequence length: 13657 Attention mask shape: torch.Size([1, 1, 13657, 13657]) Position ids shape: torch.Size([1, 13657]) Input IDs shape: torch.Size([1, 13657]) Labels shape: torch.Size([1, 13657]) Final batch size: 1, sequence length: 14360 Attention mask shape: torch.Size([1, 1, 14360, 14360]) Position ids shape: torch.Size([1, 14360]) Input IDs shape: torch.Size([1, 14360]) Labels shape: torch.Size([1, 14360]) Final batch size: 1, sequence length: 11225 Attention mask shape: torch.Size([1, 1, 11225, 11225]) Position ids shape: torch.Size([1, 11225]) Input IDs shape: torch.Size([1, 11225]) Labels shape: torch.Size([1, 11225]) Final batch size: 1, sequence length: 14226 Attention mask shape: torch.Size([1, 1, 14226, 14226]) Position ids shape: torch.Size([1, 14226]) Input IDs shape: torch.Size([1, 14226]) Labels shape: torch.Size([1, 14226]) Final batch size: 1, sequence length: 16398 Attention mask shape: torch.Size([1, 1, 16398, 16398]) Position ids shape: torch.Size([1, 16398]) Input IDs shape: torch.Size([1, 16398]) Labels shape: torch.Size([1, 16398]) Final batch size: 1, sequence length: 16223 Attention mask shape: torch.Size([1, 1, 16223, 16223]) Position ids shape: torch.Size([1, 16223]) Input IDs shape: torch.Size([1, 16223]) Labels shape: torch.Size([1, 16223]) Final batch size: 1, sequence length: 10017 Attention mask shape: torch.Size([1, 1, 10017, 10017]) Position ids shape: torch.Size([1, 10017]) Input IDs shape: torch.Size([1, 10017]) Labels shape: torch.Size([1, 10017]) Final batch size: 1, sequence length: 18408 Attention mask shape: torch.Size([1, 1, 18408, 18408]) Position ids shape: torch.Size([1, 18408]) Input IDs shape: torch.Size([1, 18408]) Labels shape: torch.Size([1, 18408]) Final batch size: 1, sequence length: 16053 Attention mask shape: torch.Size([1, 1, 16053, 16053]) Position ids shape: torch.Size([1, 16053]) Input IDs shape: torch.Size([1, 16053]) Labels shape: torch.Size([1, 16053]) Final batch size: 1, sequence length: 17117 Attention mask shape: torch.Size([1, 1, 17117, 17117]) Position ids shape: torch.Size([1, 17117]) Input IDs shape: torch.Size([1, 17117]) Labels shape: torch.Size([1, 17117]) Final batch size: 1, sequence length: 15243 Attention mask shape: torch.Size([1, 1, 15243, 15243]) Position ids shape: torch.Size([1, 15243]) Input IDs shape: torch.Size([1, 15243]) Labels shape: torch.Size([1, 15243]) Final batch size: 1, sequence length: 19892 Attention mask shape: torch.Size([1, 1, 19892, 19892]) Position ids shape: torch.Size([1, 19892]) Input IDs shape: torch.Size([1, 19892]) Labels shape: torch.Size([1, 19892]) Final batch size: 1, sequence length: 16756 Attention mask shape: torch.Size([1, 1, 16756, 16756]) Position ids shape: torch.Size([1, 16756]) Input IDs shape: torch.Size([1, 16756]) Labels shape: torch.Size([1, 16756]) Final batch size: 1, sequence length: 19259 Attention mask shape: torch.Size([1, 1, 19259, 19259]) Position ids shape: torch.Size([1, 19259]) Input IDs shape: torch.Size([1, 19259]) Labels shape: torch.Size([1, 19259]) Final batch size: 1, sequence length: 14025 Attention mask shape: torch.Size([1, 1, 14025, 14025]) Position ids shape: torch.Size([1, 14025]) Input IDs shape: torch.Size([1, 14025]) Labels shape: torch.Size([1, 14025]) Final batch size: 1, sequence length: 17843 Attention mask shape: torch.Size([1, 1, 17843, 17843]) Position ids shape: torch.Size([1, 17843]) Input IDs shape: torch.Size([1, 17843]) Labels shape: torch.Size([1, 17843]) Final batch size: 1, sequence length: 19278 Attention mask shape: torch.Size([1, 1, 19278, 19278]) Position ids shape: torch.Size([1, 19278]) Input IDs shape: torch.Size([1, 19278]) Labels shape: torch.Size([1, 19278]) Final batch size: 1, sequence length: 21455 Attention mask shape: torch.Size([1, 1, 21455, 21455]) Position ids shape: torch.Size([1, 21455]) Input IDs shape: torch.Size([1, 21455]) Labels shape: torch.Size([1, 21455]) Final batch size: 1, sequence length: 19767 Attention mask shape: torch.Size([1, 1, 19767, 19767]) Position ids shape: torch.Size([1, 19767]) Input IDs shape: torch.Size([1, 19767]) Labels shape: torch.Size([1, 19767]) Final batch size: 1, sequence length: 17294 Attention mask shape: torch.Size([1, 1, 17294, 17294]) Position ids shape: torch.Size([1, 17294]) Input IDs shape: torch.Size([1, 17294]) Labels shape: torch.Size([1, 17294]) Final batch size: 1, sequence length: 20433 Attention mask shape: torch.Size([1, 1, 20433, 20433]) Position ids shape: torch.Size([1, 20433]) Input IDs shape: torch.Size([1, 20433]) Labels shape: torch.Size([1, 20433]) Final batch size: 1, sequence length: 14437 Attention mask shape: torch.Size([1, 1, 14437, 14437]) Position ids shape: torch.Size([1, 14437]) Input IDs shape: torch.Size([1, 14437]) Labels shape: torch.Size([1, 14437]) Final batch size: 1, sequence length: 20695 Attention mask shape: torch.Size([1, 1, 20695, 20695]) Position ids shape: torch.Size([1, 20695]) Input IDs shape: torch.Size([1, 20695]) Labels shape: torch.Size([1, 20695]) Final batch size: 1, sequence length: 5734 Attention mask shape: torch.Size([1, 1, 5734, 5734]) Position ids shape: torch.Size([1, 5734]) Input IDs shape: torch.Size([1, 5734]) Labels shape: torch.Size([1, 5734]) Final batch size: 1, sequence length: 6378 Attention mask shape: torch.Size([1, 1, 6378, 6378]) Position ids shape: torch.Size([1, 6378]) Input IDs shape: torch.Size([1, 6378]) Labels shape: torch.Size([1, 6378]) Final batch size: 1, sequence length: 19492 Attention mask shape: torch.Size([1, 1, 19492, 19492]) Position ids shape: torch.Size([1, 19492]) Input IDs shape: torch.Size([1, 19492]) Labels shape: torch.Size([1, 19492]) Final batch size: 1, sequence length: 13623 Attention mask shape: torch.Size([1, 1, 13623, 13623]) Position ids shape: torch.Size([1, 13623]) Input IDs shape: torch.Size([1, 13623]) Labels shape: torch.Size([1, 13623]) Final batch size: 1, sequence length: 18377 Attention mask shape: torch.Size([1, 1, 18377, 18377]) Position ids shape: torch.Size([1, 18377]) Input IDs shape: torch.Size([1, 18377]) Labels shape: torch.Size([1, 18377]) Final batch size: 1, sequence length: 25405 Attention mask shape: torch.Size([1, 1, 25405, 25405]) Position ids shape: torch.Size([1, 25405]) Input IDs shape: torch.Size([1, 25405]) Labels shape: torch.Size([1, 25405]) Final batch size: 1, sequence length: 25548 Attention mask shape: torch.Size([1, 1, 25548, 25548]) Position ids shape: torch.Size([1, 25548]) Input IDs shape: torch.Size([1, 25548]) Labels shape: torch.Size([1, 25548]) Final batch size: 1, sequence length: 22932 Attention mask shape: torch.Size([1, 1, 22932, 22932]) Position ids shape: torch.Size([1, 22932]) Input IDs shape: torch.Size([1, 22932]) Labels shape: torch.Size([1, 22932]) Final batch size: 1, sequence length: 24885 Attention mask shape: torch.Size([1, 1, 24885, 24885]) Position ids shape: torch.Size([1, 24885]) Input IDs shape: torch.Size([1, 24885]) Labels shape: torch.Size([1, 24885]) Final batch size: 1, sequence length: 27659 Attention mask shape: torch.Size([1, 1, 27659, 27659]) Position ids shape: torch.Size([1, 27659]) Input IDs shape: torch.Size([1, 27659]) Labels shape: torch.Size([1, 27659]) Final batch size: 1, sequence length: 25381 Attention mask shape: torch.Size([1, 1, 25381, 25381]) Position ids shape: torch.Size([1, 25381]) Input IDs shape: torch.Size([1, 25381]) Labels shape: torch.Size([1, 25381]) Final batch size: 1, sequence length: 17985 Attention mask shape: torch.Size([1, 1, 17985, 17985]) Position ids shape: torch.Size([1, 17985]) Input IDs shape: torch.Size([1, 17985]) Labels shape: torch.Size([1, 17985]) Final batch size: 1, sequence length: 26063 Attention mask shape: torch.Size([1, 1, 26063, 26063]) Position ids shape: torch.Size([1, 26063]) Input IDs shape: torch.Size([1, 26063]) Labels shape: torch.Size([1, 26063]) Final batch size: 1, sequence length: 14429 Attention mask shape: torch.Size([1, 1, 14429, 14429]) Position ids shape: torch.Size([1, 14429]) Input IDs shape: torch.Size([1, 14429]) Labels shape: torch.Size([1, 14429]) Final batch size: 1, sequence length: 26639 Attention mask shape: torch.Size([1, 1, 26639, 26639]) Position ids shape: torch.Size([1, 26639]) Input IDs shape: torch.Size([1, 26639]) Labels shape: torch.Size([1, 26639]) Final batch size: 1, sequence length: 27484 Attention mask shape: torch.Size([1, 1, 27484, 27484]) Position ids shape: torch.Size([1, 27484]) Input IDs shape: torch.Size([1, 27484]) Labels shape: torch.Size([1, 27484]) Final batch size: 1, sequence length: 29561 Attention mask shape: torch.Size([1, 1, 29561, 29561]) Position ids shape: torch.Size([1, 29561]) Input IDs shape: torch.Size([1, 29561]) Labels shape: torch.Size([1, 29561]) Final batch size: 1, sequence length: 28623 Attention mask shape: torch.Size([1, 1, 28623, 28623]) Position ids shape: torch.Size([1, 28623]) Input IDs shape: torch.Size([1, 28623]) Labels shape: torch.Size([1, 28623]) Final batch size: 1, sequence length: 30236 Attention mask shape: torch.Size([1, 1, 30236, 30236]) Position ids shape: torch.Size([1, 30236]) Input IDs shape: torch.Size([1, 30236]) Labels shape: torch.Size([1, 30236]) Final batch size: 1, sequence length: 16716 Attention mask shape: torch.Size([1, 1, 16716, 16716]) Position ids shape: torch.Size([1, 16716]) Input IDs shape: torch.Size([1, 16716]) Labels shape: torch.Size([1, 16716]) Final batch size: 1, sequence length: 28166 Attention mask shape: torch.Size([1, 1, 28166, 28166]) Position ids shape: torch.Size([1, 28166]) Input IDs shape: torch.Size([1, 28166]) Labels shape: torch.Size([1, 28166]) Final batch size: 1, sequence length: 23334 Attention mask shape: torch.Size([1, 1, 23334, 23334]) Position ids shape: torch.Size([1, 23334]) Input IDs shape: torch.Size([1, 23334]) Labels shape: torch.Size([1, 23334]) Final batch size: 1, sequence length: 29824 Attention mask shape: torch.Size([1, 1, 29824, 29824]) Position ids shape: torch.Size([1, 29824]) Input IDs shape: torch.Size([1, 29824]) Labels shape: torch.Size([1, 29824]) Final batch size: 1, sequence length: 28823 Attention mask shape: torch.Size([1, 1, 28823, 28823]) Position ids shape: torch.Size([1, 28823]) Input IDs shape: torch.Size([1, 28823]) Labels shape: torch.Size([1, 28823]) Final batch size: 1, sequence length: 28910 Attention mask shape: torch.Size([1, 1, 28910, 28910]) Position ids shape: torch.Size([1, 28910]) Input IDs shape: torch.Size([1, 28910]) Labels shape: torch.Size([1, 28910]) Final batch size: 1, sequence length: 17951 Attention mask shape: torch.Size([1, 1, 17951, 17951]) Position ids shape: torch.Size([1, 17951]) Input IDs shape: torch.Size([1, 17951]) Labels shape: torch.Size([1, 17951]) Final batch size: 1, sequence length: 28206 Attention mask shape: torch.Size([1, 1, 28206, 28206]) Position ids shape: torch.Size([1, 28206]) Input IDs shape: torch.Size([1, 28206]) Labels shape: torch.Size([1, 28206]) Final batch size: 1, sequence length: 33601 Attention mask shape: torch.Size([1, 1, 33601, 33601]) Position ids shape: torch.Size([1, 33601]) Input IDs shape: torch.Size([1, 33601]) Labels shape: torch.Size([1, 33601]) Final batch size: 1, sequence length: 31464 Attention mask shape: torch.Size([1, 1, 31464, 31464]) Position ids shape: torch.Size([1, 31464]) Input IDs shape: torch.Size([1, 31464]) Labels shape: torch.Size([1, 31464]) Final batch size: 1, sequence length: 32660 Attention mask shape: torch.Size([1, 1, 32660, 32660]) Position ids shape: torch.Size([1, 32660]) Input IDs shape: torch.Size([1, 32660]) Labels shape: torch.Size([1, 32660]) Final batch size: 1, sequence length: 32786 Attention mask shape: torch.Size([1, 1, 32786, 32786]) Position ids shape: torch.Size([1, 32786]) Input IDs shape: torch.Size([1, 32786]) Labels shape: torch.Size([1, 32786]) Final batch size: 1, sequence length: 9029 Attention mask shape: torch.Size([1, 1, 9029, 9029]) Position ids shape: torch.Size([1, 9029]) Input IDs shape: torch.Size([1, 9029]) Labels shape: torch.Size([1, 9029]) Final batch size: 1, sequence length: 20198 Attention mask shape: torch.Size([1, 1, 20198, 20198]) Position ids shape: torch.Size([1, 20198]) Input IDs shape: torch.Size([1, 20198]) Labels shape: torch.Size([1, 20198]) Final batch size: 1, sequence length: 35411 Attention mask shape: torch.Size([1, 1, 35411, 35411]) Position ids shape: torch.Size([1, 35411]) Input IDs shape: torch.Size([1, 35411]) Labels shape: torch.Size([1, 35411]) Final batch size: 1, sequence length: 17914 Attention mask shape: torch.Size([1, 1, 17914, 17914]) Position ids shape: torch.Size([1, 17914]) Input IDs shape: torch.Size([1, 17914]) Labels shape: torch.Size([1, 17914]) Final batch size: 1, sequence length: 25540 Attention mask shape: torch.Size([1, 1, 25540, 25540]) Position ids shape: torch.Size([1, 25540]) Input IDs shape: torch.Size([1, 25540]) Labels shape: torch.Size([1, 25540]) Final batch size: 1, sequence length: 10132 Attention mask shape: torch.Size([1, 1, 10132, 10132]) Position ids shape: torch.Size([1, 10132]) Input IDs shape: torch.Size([1, 10132]) Labels shape: torch.Size([1, 10132]) Final batch size: 1, sequence length: 20951 Attention mask shape: torch.Size([1, 1, 20951, 20951]) Position ids shape: torch.Size([1, 20951]) Input IDs shape: torch.Size([1, 20951]) Labels shape: torch.Size([1, 20951]) Final batch size: 1, sequence length: 17634 Attention mask shape: torch.Size([1, 1, 17634, 17634]) Position ids shape: torch.Size([1, 17634]) Input IDs shape: torch.Size([1, 17634]) Labels shape: torch.Size([1, 17634]) Final batch size: 1, sequence length: 36131 Attention mask shape: torch.Size([1, 1, 36131, 36131]) Position ids shape: torch.Size([1, 36131]) Input IDs shape: torch.Size([1, 36131]) Labels shape: torch.Size([1, 36131]) Final batch size: 1, sequence length: 36970 Attention mask shape: torch.Size([1, 1, 36970, 36970]) Position ids shape: torch.Size([1, 36970]) Input IDs shape: torch.Size([1, 36970]) Labels shape: torch.Size([1, 36970]) Final batch size: 1, sequence length: 38712 Attention mask shape: torch.Size([1, 1, 38712, 38712]) Position ids shape: torch.Size([1, 38712]) Input IDs shape: torch.Size([1, 38712]) Labels shape: torch.Size([1, 38712]) Final batch size: 1, sequence length: 36124 Attention mask shape: torch.Size([1, 1, 36124, 36124]) Position ids shape: torch.Size([1, 36124]) Input IDs shape: torch.Size([1, 36124]) Labels shape: torch.Size([1, 36124]) Final batch size: 1, sequence length: 37555 Attention mask shape: torch.Size([1, 1, 37555, 37555]) Position ids shape: torch.Size([1, 37555]) Input IDs shape: torch.Size([1, 37555]) Labels shape: torch.Size([1, 37555]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 15513 Attention mask shape: torch.Size([1, 1, 15513, 15513]) Position ids shape: torch.Size([1, 15513]) Input IDs shape: torch.Size([1, 15513]) Labels shape: torch.Size([1, 15513]) Final batch size: 1, sequence length: 25529 Attention mask shape: torch.Size([1, 1, 25529, 25529]) Position ids shape: torch.Size([1, 25529]) Input IDs shape: torch.Size([1, 25529]) Labels shape: torch.Size([1, 25529]) Final batch size: 1, sequence length: 20533 Attention mask shape: torch.Size([1, 1, 20533, 20533]) Position ids shape: torch.Size([1, 20533]) Input IDs shape: torch.Size([1, 20533]) Labels shape: torch.Size([1, 20533]) Final batch size: 1, sequence length: 32515 Attention mask shape: torch.Size([1, 1, 32515, 32515]) Position ids shape: torch.Size([1, 32515]) Input IDs shape: torch.Size([1, 32515]) Labels shape: torch.Size([1, 32515]) Final batch size: 1, sequence length: 37016 Attention mask shape: torch.Size([1, 1, 37016, 37016]) Position ids shape: torch.Size([1, 37016]) Input IDs shape: torch.Size([1, 37016]) Labels shape: torch.Size([1, 37016]) Final batch size: 1, sequence length: 35525 Attention mask shape: torch.Size([1, 1, 35525, 35525]) Position ids shape: torch.Size([1, 35525]) Input IDs shape: torch.Size([1, 35525]) Labels shape: torch.Size([1, 35525]) Final batch size: 1, sequence length: 26562 Attention mask shape: torch.Size([1, 1, 26562, 26562]) Position ids shape: torch.Size([1, 26562]) Input IDs shape: torch.Size([1, 26562]) Labels shape: torch.Size([1, 26562]) Final batch size: 1, sequence length: 36292 Attention mask shape: torch.Size([1, 1, 36292, 36292]) Position ids shape: torch.Size([1, 36292]) Input IDs shape: torch.Size([1, 36292]) Labels shape: torch.Size([1, 36292]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31084 Attention mask shape: torch.Size([1, 1, 31084, 31084]) Position ids shape: torch.Size([1, 31084]) Input IDs shape: torch.Size([1, 31084]) Labels shape: torch.Size([1, 31084]) Final batch size: 1, sequence length: 33422 Attention mask shape: torch.Size([1, 1, 33422, 33422]) Position ids shape: torch.Size([1, 33422]) Input IDs shape: torch.Size([1, 33422]) Labels shape: torch.Size([1, 33422]) Final batch size: 1, sequence length: 38210 Attention mask shape: torch.Size([1, 1, 38210, 38210]) Position ids shape: torch.Size([1, 38210]) Input IDs shape: torch.Size([1, 38210]) Labels shape: torch.Size([1, 38210]) Final batch size: 1, sequence length: 14372 Attention mask shape: torch.Size([1, 1, 14372, 14372]) Position ids shape: torch.Size([1, 14372]) Input IDs shape: torch.Size([1, 14372]) Labels shape: torch.Size([1, 14372]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 35775 Attention mask shape: torch.Size([1, 1, 35775, 35775]) Position ids shape: torch.Size([1, 35775]) Input IDs shape: torch.Size([1, 35775]) Labels shape: torch.Size([1, 35775]) Final batch size: 1, sequence length: 28879 Attention mask shape: torch.Size([1, 1, 28879, 28879]) Position ids shape: torch.Size([1, 28879]) Input IDs shape: torch.Size([1, 28879]) Labels shape: torch.Size([1, 28879]) Final batch size: 1, sequence length: 36185 Attention mask shape: torch.Size([1, 1, 36185, 36185]) Position ids shape: torch.Size([1, 36185]) Input IDs shape: torch.Size([1, 36185]) Labels shape: torch.Size([1, 36185]) Final batch size: 1, sequence length: 38071 Attention mask shape: torch.Size([1, 1, 38071, 38071]) Position ids shape: torch.Size([1, 38071]) Input IDs shape: torch.Size([1, 38071]) Labels shape: torch.Size([1, 38071]) Final batch size: 1, sequence length: 30009 Attention mask shape: torch.Size([1, 1, 30009, 30009]) Position ids shape: torch.Size([1, 30009]) Input IDs shape: torch.Size([1, 30009]) Labels shape: torch.Size([1, 30009]) Final batch size: 1, sequence length: 16409 Attention mask shape: torch.Size([1, 1, 16409, 16409]) Position ids shape: torch.Size([1, 16409]) Input IDs shape: torch.Size([1, 16409]) Labels shape: torch.Size([1, 16409]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40753 Attention mask shape: torch.Size([1, 1, 40753, 40753]) Position ids shape: torch.Size([1, 40753]) Input IDs shape: torch.Size([1, 40753]) Labels shape: torch.Size([1, 40753]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30653 Attention mask shape: torch.Size([1, 1, 30653, 30653]) Position ids shape: torch.Size([1, 30653]) Input IDs shape: torch.Size([1, 30653]) Labels shape: torch.Size([1, 30653]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26781 Attention mask shape: torch.Size([1, 1, 26781, 26781]) Position ids shape: torch.Size([1, 26781]) Input IDs shape: torch.Size([1, 26781]) Labels shape: torch.Size([1, 26781]) Final batch size: 1, sequence length: 36534 Attention mask shape: torch.Size([1, 1, 36534, 36534]) Position ids shape: torch.Size([1, 36534]) Input IDs shape: torch.Size([1, 36534]) Labels shape: torch.Size([1, 36534]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 18884 Attention mask shape: torch.Size([1, 1, 18884, 18884]) Position ids shape: torch.Size([1, 18884]) Input IDs shape: torch.Size([1, 18884]) Labels shape: torch.Size([1, 18884]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31042 Attention mask shape: torch.Size([1, 1, 31042, 31042]) Position ids shape: torch.Size([1, 31042]) Input IDs shape: torch.Size([1, 31042]) Labels shape: torch.Size([1, 31042]) Final batch size: 1, sequence length: 25850 Attention mask shape: torch.Size([1, 1, 25850, 25850]) Position ids shape: torch.Size([1, 25850]) Input IDs shape: torch.Size([1, 25850]) Labels shape: torch.Size([1, 25850]) Final batch size: 1, sequence length: 15031 Attention mask shape: torch.Size([1, 1, 15031, 15031]) Position ids shape: torch.Size([1, 15031]) Input IDs shape: torch.Size([1, 15031]) Labels shape: torch.Size([1, 15031]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 37159 Attention mask shape: torch.Size([1, 1, 37159, 37159]) Position ids shape: torch.Size([1, 37159]) Input IDs shape: torch.Size([1, 37159]) Labels shape: torch.Size([1, 37159]) Final batch size: 1, sequence length: 31448 Attention mask shape: torch.Size([1, 1, 31448, 31448]) Position ids shape: torch.Size([1, 31448]) Input IDs shape: torch.Size([1, 31448]) Labels shape: torch.Size([1, 31448]) Final batch size: 1, sequence length: 7448 Attention mask shape: torch.Size([1, 1, 7448, 7448]) Position ids shape: torch.Size([1, 7448]) Input IDs shape: torch.Size([1, 7448]) Labels shape: torch.Size([1, 7448]) Final batch size: 1, sequence length: 20307 Attention mask shape: torch.Size([1, 1, 20307, 20307]) Position ids shape: torch.Size([1, 20307]) Input IDs shape: torch.Size([1, 20307]) Labels shape: torch.Size([1, 20307]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40605 Attention mask shape: torch.Size([1, 1, 40605, 40605]) Position ids shape: torch.Size([1, 40605]) Input IDs shape: torch.Size([1, 40605]) Labels shape: torch.Size([1, 40605]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 5405 Attention mask shape: torch.Size([1, 1, 5405, 5405]) Position ids shape: torch.Size([1, 5405]) Input IDs shape: torch.Size([1, 5405]) Labels shape: torch.Size([1, 5405]) {'loss': 0.247, 'grad_norm': 0.15029786283824884, 'learning_rate': 1.0926199633097156e-07, 'num_tokens': -inf, 'epoch': 7.62} Final batch size: 1, sequence length: 4641 Attention mask shape: torch.Size([1, 1, 4641, 4641]) Position ids shape: torch.Size([1, 4641]) Input IDs shape: torch.Size([1, 4641]) Labels shape: torch.Size([1, 4641]) Final batch size: 1, sequence length: 4223 Attention mask shape: torch.Size([1, 1, 4223, 4223]) Position ids shape: torch.Size([1, 4223]) Input IDs shape: torch.Size([1, 4223]) Labels shape: torch.Size([1, 4223]) Final batch size: 1, sequence length: 7795 Attention mask shape: torch.Size([1, 1, 7795, 7795]) Position ids shape: torch.Size([1, 7795]) Input IDs shape: torch.Size([1, 7795]) Labels shape: torch.Size([1, 7795]) Final batch size: 1, sequence length: 8177 Attention mask shape: torch.Size([1, 1, 8177, 8177]) Position ids shape: torch.Size([1, 8177]) Input IDs shape: torch.Size([1, 8177]) Labels shape: torch.Size([1, 8177]) Final batch size: 1, sequence length: 9021 Attention mask shape: torch.Size([1, 1, 9021, 9021]) Position ids shape: torch.Size([1, 9021]) Input IDs shape: torch.Size([1, 9021]) Labels shape: torch.Size([1, 9021]) Final batch size: 1, sequence length: 9027 Attention mask shape: torch.Size([1, 1, 9027, 9027]) Position ids shape: torch.Size([1, 9027]) Input IDs shape: torch.Size([1, 9027]) Labels shape: torch.Size([1, 9027]) Final batch size: 1, sequence length: 12501 Attention mask shape: torch.Size([1, 1, 12501, 12501]) Position ids shape: torch.Size([1, 12501]) Input IDs shape: torch.Size([1, 12501]) Labels shape: torch.Size([1, 12501]) Final batch size: 1, sequence length: 5754 Attention mask shape: torch.Size([1, 1, 5754, 5754]) Position ids shape: torch.Size([1, 5754]) Input IDs shape: torch.Size([1, 5754]) Labels shape: torch.Size([1, 5754]) Final batch size: 1, sequence length: 13186 Attention mask shape: torch.Size([1, 1, 13186, 13186]) Position ids shape: torch.Size([1, 13186]) Input IDs shape: torch.Size([1, 13186]) Labels shape: torch.Size([1, 13186]) Final batch size: 1, sequence length: 11974 Attention mask shape: torch.Size([1, 1, 11974, 11974]) Position ids shape: torch.Size([1, 11974]) Input IDs shape: torch.Size([1, 11974]) Labels shape: torch.Size([1, 11974]) Final batch size: 1, sequence length: 15283 Attention mask shape: torch.Size([1, 1, 15283, 15283]) Position ids shape: torch.Size([1, 15283]) Input IDs shape: torch.Size([1, 15283]) Labels shape: torch.Size([1, 15283]) Final batch size: 1, sequence length: 16330 Attention mask shape: torch.Size([1, 1, 16330, 16330]) Position ids shape: torch.Size([1, 16330]) Input IDs shape: torch.Size([1, 16330]) Labels shape: torch.Size([1, 16330]) Final batch size: 1, sequence length: 13957 Attention mask shape: torch.Size([1, 1, 13957, 13957]) Position ids shape: torch.Size([1, 13957]) Input IDs shape: torch.Size([1, 13957]) Labels shape: torch.Size([1, 13957]) Final batch size: 1, sequence length: 16104 Attention mask shape: torch.Size([1, 1, 16104, 16104]) Position ids shape: torch.Size([1, 16104]) Input IDs shape: torch.Size([1, 16104]) Labels shape: torch.Size([1, 16104]) Final batch size: 1, sequence length: 16496 Attention mask shape: torch.Size([1, 1, 16496, 16496]) Position ids shape: torch.Size([1, 16496]) Input IDs shape: torch.Size([1, 16496]) Labels shape: torch.Size([1, 16496]) Final batch size: 1, sequence length: 18958 Attention mask shape: torch.Size([1, 1, 18958, 18958]) Position ids shape: torch.Size([1, 18958]) Input IDs shape: torch.Size([1, 18958]) Labels shape: torch.Size([1, 18958]) Final batch size: 1, sequence length: 14492 Attention mask shape: torch.Size([1, 1, 14492, 14492]) Position ids shape: torch.Size([1, 14492]) Input IDs shape: torch.Size([1, 14492]) Labels shape: torch.Size([1, 14492]) Final batch size: 1, sequence length: 16366 Attention mask shape: torch.Size([1, 1, 16366, 16366]) Position ids shape: torch.Size([1, 16366]) Input IDs shape: torch.Size([1, 16366]) Labels shape: torch.Size([1, 16366]) Final batch size: 1, sequence length: 16088 Attention mask shape: torch.Size([1, 1, 16088, 16088]) Position ids shape: torch.Size([1, 16088]) Input IDs shape: torch.Size([1, 16088]) Labels shape: torch.Size([1, 16088]) Final batch size: 1, sequence length: 17092 Attention mask shape: torch.Size([1, 1, 17092, 17092]) Position ids shape: torch.Size([1, 17092]) Input IDs shape: torch.Size([1, 17092]) Labels shape: torch.Size([1, 17092]) Final batch size: 1, sequence length: 19034 Attention mask shape: torch.Size([1, 1, 19034, 19034]) Position ids shape: torch.Size([1, 19034]) Input IDs shape: torch.Size([1, 19034]) Labels shape: torch.Size([1, 19034]) Final batch size: 1, sequence length: 16014 Attention mask shape: torch.Size([1, 1, 16014, 16014]) Position ids shape: torch.Size([1, 16014]) Input IDs shape: torch.Size([1, 16014]) Labels shape: torch.Size([1, 16014]) Final batch size: 1, sequence length: 18938 Attention mask shape: torch.Size([1, 1, 18938, 18938]) Position ids shape: torch.Size([1, 18938]) Input IDs shape: torch.Size([1, 18938]) Labels shape: torch.Size([1, 18938]) Final batch size: 1, sequence length: 21132 Attention mask shape: torch.Size([1, 1, 21132, 21132]) Position ids shape: torch.Size([1, 21132]) Input IDs shape: torch.Size([1, 21132]) Labels shape: torch.Size([1, 21132]) Final batch size: 1, sequence length: 19603 Attention mask shape: torch.Size([1, 1, 19603, 19603]) Position ids shape: torch.Size([1, 19603]) Input IDs shape: torch.Size([1, 19603]) Labels shape: torch.Size([1, 19603]) Final batch size: 1, sequence length: 18415 Attention mask shape: torch.Size([1, 1, 18415, 18415]) Position ids shape: torch.Size([1, 18415]) Input IDs shape: torch.Size([1, 18415]) Labels shape: torch.Size([1, 18415]) Final batch size: 1, sequence length: 18034 Attention mask shape: torch.Size([1, 1, 18034, 18034]) Position ids shape: torch.Size([1, 18034]) Input IDs shape: torch.Size([1, 18034]) Labels shape: torch.Size([1, 18034]) Final batch size: 1, sequence length: 13468 Attention mask shape: torch.Size([1, 1, 13468, 13468]) Position ids shape: torch.Size([1, 13468]) Input IDs shape: torch.Size([1, 13468]) Labels shape: torch.Size([1, 13468]) Final batch size: 1, sequence length: 21705 Attention mask shape: torch.Size([1, 1, 21705, 21705]) Position ids shape: torch.Size([1, 21705]) Input IDs shape: torch.Size([1, 21705]) Labels shape: torch.Size([1, 21705]) Final batch size: 1, sequence length: 23102 Attention mask shape: torch.Size([1, 1, 23102, 23102]) Position ids shape: torch.Size([1, 23102]) Input IDs shape: torch.Size([1, 23102]) Labels shape: torch.Size([1, 23102]) Final batch size: 1, sequence length: 20726 Attention mask shape: torch.Size([1, 1, 20726, 20726]) Position ids shape: torch.Size([1, 20726]) Input IDs shape: torch.Size([1, 20726]) Labels shape: torch.Size([1, 20726]) Final batch size: 1, sequence length: 24432 Attention mask shape: torch.Size([1, 1, 24432, 24432]) Position ids shape: torch.Size([1, 24432]) Input IDs shape: torch.Size([1, 24432]) Labels shape: torch.Size([1, 24432]) Final batch size: 1, sequence length: 23132 Attention mask shape: torch.Size([1, 1, 23132, 23132]) Position ids shape: torch.Size([1, 23132]) Input IDs shape: torch.Size([1, 23132]) Labels shape: torch.Size([1, 23132]) Final batch size: 1, sequence length: 21404 Attention mask shape: torch.Size([1, 1, 21404, 21404]) Position ids shape: torch.Size([1, 21404]) Input IDs shape: torch.Size([1, 21404]) Labels shape: torch.Size([1, 21404]) Final batch size: 1, sequence length: 12756 Attention mask shape: torch.Size([1, 1, 12756, 12756]) Position ids shape: torch.Size([1, 12756]) Input IDs shape: torch.Size([1, 12756]) Labels shape: torch.Size([1, 12756]) Final batch size: 1, sequence length: 23971 Attention mask shape: torch.Size([1, 1, 23971, 23971]) Position ids shape: torch.Size([1, 23971]) Input IDs shape: torch.Size([1, 23971]) Labels shape: torch.Size([1, 23971]) Final batch size: 1, sequence length: 25351 Attention mask shape: torch.Size([1, 1, 25351, 25351]) Position ids shape: torch.Size([1, 25351]) Input IDs shape: torch.Size([1, 25351]) Labels shape: torch.Size([1, 25351]) Final batch size: 1, sequence length: 28412 Attention mask shape: torch.Size([1, 1, 28412, 28412]) Position ids shape: torch.Size([1, 28412]) Input IDs shape: torch.Size([1, 28412]) Labels shape: torch.Size([1, 28412]) Final batch size: 1, sequence length: 16590 Attention mask shape: torch.Size([1, 1, 16590, 16590]) Position ids shape: torch.Size([1, 16590]) Input IDs shape: torch.Size([1, 16590]) Labels shape: torch.Size([1, 16590]) Final batch size: 1, sequence length: 27785 Attention mask shape: torch.Size([1, 1, 27785, 27785]) Position ids shape: torch.Size([1, 27785]) Input IDs shape: torch.Size([1, 27785]) Labels shape: torch.Size([1, 27785]) Final batch size: 1, sequence length: 13790 Attention mask shape: torch.Size([1, 1, 13790, 13790]) Position ids shape: torch.Size([1, 13790]) Input IDs shape: torch.Size([1, 13790]) Labels shape: torch.Size([1, 13790]) Final batch size: 1, sequence length: 26375 Attention mask shape: torch.Size([1, 1, 26375, 26375]) Position ids shape: torch.Size([1, 26375]) Input IDs shape: torch.Size([1, 26375]) Labels shape: torch.Size([1, 26375]) Final batch size: 1, sequence length: 17514 Attention mask shape: torch.Size([1, 1, 17514, 17514]) Position ids shape: torch.Size([1, 17514]) Input IDs shape: torch.Size([1, 17514]) Labels shape: torch.Size([1, 17514]) Final batch size: 1, sequence length: 23894 Attention mask shape: torch.Size([1, 1, 23894, 23894]) Position ids shape: torch.Size([1, 23894]) Input IDs shape: torch.Size([1, 23894]) Labels shape: torch.Size([1, 23894]) Final batch size: 1, sequence length: 27780 Attention mask shape: torch.Size([1, 1, 27780, 27780]) Position ids shape: torch.Size([1, 27780]) Input IDs shape: torch.Size([1, 27780]) Labels shape: torch.Size([1, 27780]) Final batch size: 1, sequence length: 28845 Attention mask shape: torch.Size([1, 1, 28845, 28845]) Position ids shape: torch.Size([1, 28845]) Input IDs shape: torch.Size([1, 28845]) Labels shape: torch.Size([1, 28845]) Final batch size: 1, sequence length: 22771 Attention mask shape: torch.Size([1, 1, 22771, 22771]) Position ids shape: torch.Size([1, 22771]) Input IDs shape: torch.Size([1, 22771]) Labels shape: torch.Size([1, 22771]) Final batch size: 1, sequence length: 26766 Attention mask shape: torch.Size([1, 1, 26766, 26766]) Position ids shape: torch.Size([1, 26766]) Input IDs shape: torch.Size([1, 26766]) Labels shape: torch.Size([1, 26766]) Final batch size: 1, sequence length: 20817 Attention mask shape: torch.Size([1, 1, 20817, 20817]) Position ids shape: torch.Size([1, 20817]) Input IDs shape: torch.Size([1, 20817]) Labels shape: torch.Size([1, 20817]) Final batch size: 1, sequence length: 28574 Attention mask shape: torch.Size([1, 1, 28574, 28574]) Position ids shape: torch.Size([1, 28574]) Input IDs shape: torch.Size([1, 28574]) Labels shape: torch.Size([1, 28574]) Final batch size: 1, sequence length: 26596 Attention mask shape: torch.Size([1, 1, 26596, 26596]) Position ids shape: torch.Size([1, 26596]) Input IDs shape: torch.Size([1, 26596]) Labels shape: torch.Size([1, 26596]) Final batch size: 1, sequence length: 29168 Attention mask shape: torch.Size([1, 1, 29168, 29168]) Position ids shape: torch.Size([1, 29168]) Input IDs shape: torch.Size([1, 29168]) Labels shape: torch.Size([1, 29168]) Final batch size: 1, sequence length: 25388 Attention mask shape: torch.Size([1, 1, 25388, 25388]) Position ids shape: torch.Size([1, 25388]) Input IDs shape: torch.Size([1, 25388]) Labels shape: torch.Size([1, 25388]) Final batch size: 1, sequence length: 30165 Attention mask shape: torch.Size([1, 1, 30165, 30165]) Position ids shape: torch.Size([1, 30165]) Input IDs shape: torch.Size([1, 30165]) Labels shape: torch.Size([1, 30165]) Final batch size: 1, sequence length: 23810 Attention mask shape: torch.Size([1, 1, 23810, 23810]) Position ids shape: torch.Size([1, 23810]) Input IDs shape: torch.Size([1, 23810]) Labels shape: torch.Size([1, 23810]) Final batch size: 1, sequence length: 18869 Attention mask shape: torch.Size([1, 1, 18869, 18869]) Position ids shape: torch.Size([1, 18869]) Input IDs shape: torch.Size([1, 18869]) Labels shape: torch.Size([1, 18869]) Final batch size: 1, sequence length: 32022 Attention mask shape: torch.Size([1, 1, 32022, 32022]) Position ids shape: torch.Size([1, 32022]) Input IDs shape: torch.Size([1, 32022]) Labels shape: torch.Size([1, 32022]) Final batch size: 1, sequence length: 29830 Attention mask shape: torch.Size([1, 1, 29830, 29830]) Position ids shape: torch.Size([1, 29830]) Input IDs shape: torch.Size([1, 29830]) Labels shape: torch.Size([1, 29830]) Final batch size: 1, sequence length: 34326 Attention mask shape: torch.Size([1, 1, 34326, 34326]) Position ids shape: torch.Size([1, 34326]) Input IDs shape: torch.Size([1, 34326]) Labels shape: torch.Size([1, 34326]) Final batch size: 1, sequence length: 16291 Attention mask shape: torch.Size([1, 1, 16291, 16291]) Position ids shape: torch.Size([1, 16291]) Input IDs shape: torch.Size([1, 16291]) Labels shape: torch.Size([1, 16291]) Final batch size: 1, sequence length: 22043 Attention mask shape: torch.Size([1, 1, 22043, 22043]) Position ids shape: torch.Size([1, 22043]) Input IDs shape: torch.Size([1, 22043]) Labels shape: torch.Size([1, 22043]) Final batch size: 1, sequence length: 34457 Attention mask shape: torch.Size([1, 1, 34457, 34457]) Position ids shape: torch.Size([1, 34457]) Input IDs shape: torch.Size([1, 34457]) Labels shape: torch.Size([1, 34457]) Final batch size: 1, sequence length: 13986 Attention mask shape: torch.Size([1, 1, 13986, 13986]) Position ids shape: torch.Size([1, 13986]) Input IDs shape: torch.Size([1, 13986]) Labels shape: torch.Size([1, 13986]) Final batch size: 1, sequence length: 29132 Attention mask shape: torch.Size([1, 1, 29132, 29132]) Position ids shape: torch.Size([1, 29132]) Input IDs shape: torch.Size([1, 29132]) Labels shape: torch.Size([1, 29132]) Final batch size: 1, sequence length: 24653 Attention mask shape: torch.Size([1, 1, 24653, 24653]) Position ids shape: torch.Size([1, 24653]) Input IDs shape: torch.Size([1, 24653]) Labels shape: torch.Size([1, 24653]) Final batch size: 1, sequence length: 26465 Attention mask shape: torch.Size([1, 1, 26465, 26465]) Position ids shape: torch.Size([1, 26465]) Input IDs shape: torch.Size([1, 26465]) Labels shape: torch.Size([1, 26465]) Final batch size: 1, sequence length: 22014 Attention mask shape: torch.Size([1, 1, 22014, 22014]) Position ids shape: torch.Size([1, 22014]) Input IDs shape: torch.Size([1, 22014]) Labels shape: torch.Size([1, 22014]) Final batch size: 1, sequence length: 22328 Attention mask shape: torch.Size([1, 1, 22328, 22328]) Position ids shape: torch.Size([1, 22328]) Input IDs shape: torch.Size([1, 22328]) Labels shape: torch.Size([1, 22328]) Final batch size: 1, sequence length: 23245 Attention mask shape: torch.Size([1, 1, 23245, 23245]) Position ids shape: torch.Size([1, 23245]) Input IDs shape: torch.Size([1, 23245]) Labels shape: torch.Size([1, 23245]) Final batch size: 1, sequence length: 35790 Attention mask shape: torch.Size([1, 1, 35790, 35790]) Position ids shape: torch.Size([1, 35790]) Input IDs shape: torch.Size([1, 35790]) Labels shape: torch.Size([1, 35790]) Final batch size: 1, sequence length: 35868 Attention mask shape: torch.Size([1, 1, 35868, 35868]) Position ids shape: torch.Size([1, 35868]) Input IDs shape: torch.Size([1, 35868]) Labels shape: torch.Size([1, 35868]) Final batch size: 1, sequence length: 38617 Attention mask shape: torch.Size([1, 1, 38617, 38617]) Position ids shape: torch.Size([1, 38617]) Input IDs shape: torch.Size([1, 38617]) Labels shape: torch.Size([1, 38617]) Final batch size: 1, sequence length: 39150 Attention mask shape: torch.Size([1, 1, 39150, 39150]) Position ids shape: torch.Size([1, 39150]) Input IDs shape: torch.Size([1, 39150]) Labels shape: torch.Size([1, 39150]) Final batch size: 1, sequence length: 37039 Attention mask shape: torch.Size([1, 1, 37039, 37039]) Position ids shape: torch.Size([1, 37039]) Input IDs shape: torch.Size([1, 37039]) Labels shape: torch.Size([1, 37039]) Final batch size: 1, sequence length: 40717 Attention mask shape: torch.Size([1, 1, 40717, 40717]) Position ids shape: torch.Size([1, 40717]) Input IDs shape: torch.Size([1, 40717]) Labels shape: torch.Size([1, 40717]) Final batch size: 1, sequence length: 19204 Attention mask shape: torch.Size([1, 1, 19204, 19204]) Position ids shape: torch.Size([1, 19204]) Input IDs shape: torch.Size([1, 19204]) Labels shape: torch.Size([1, 19204]) Final batch size: 1, sequence length: 38919 Attention mask shape: torch.Size([1, 1, 38919, 38919]) Position ids shape: torch.Size([1, 38919]) Input IDs shape: torch.Size([1, 38919]) Labels shape: torch.Size([1, 38919]) Final batch size: 1, sequence length: 35999 Attention mask shape: torch.Size([1, 1, 35999, 35999]) Position ids shape: torch.Size([1, 35999]) Input IDs shape: torch.Size([1, 35999]) Labels shape: torch.Size([1, 35999]) Final batch size: 1, sequence length: 22282 Attention mask shape: torch.Size([1, 1, 22282, 22282]) Position ids shape: torch.Size([1, 22282]) Input IDs shape: torch.Size([1, 22282]) Labels shape: torch.Size([1, 22282]) Final batch size: 1, sequence length: 34874 Attention mask shape: torch.Size([1, 1, 34874, 34874]) Position ids shape: torch.Size([1, 34874]) Input IDs shape: torch.Size([1, 34874]) Labels shape: torch.Size([1, 34874]) Final batch size: 1, sequence length: 32974 Attention mask shape: torch.Size([1, 1, 32974, 32974]) Position ids shape: torch.Size([1, 32974]) Input IDs shape: torch.Size([1, 32974]) Labels shape: torch.Size([1, 32974]) Final batch size: 1, sequence length: 31506 Attention mask shape: torch.Size([1, 1, 31506, 31506]) Position ids shape: torch.Size([1, 31506]) Input IDs shape: torch.Size([1, 31506]) Labels shape: torch.Size([1, 31506]) Final batch size: 1, sequence length: 19744 Attention mask shape: torch.Size([1, 1, 19744, 19744]) Position ids shape: torch.Size([1, 19744]) Input IDs shape: torch.Size([1, 19744]) Labels shape: torch.Size([1, 19744]) Final batch size: 1, sequence length: 28002 Attention mask shape: torch.Size([1, 1, 28002, 28002]) Position ids shape: torch.Size([1, 28002]) Input IDs shape: torch.Size([1, 28002]) Labels shape: torch.Size([1, 28002]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 11644 Attention mask shape: torch.Size([1, 1, 11644, 11644]) Position ids shape: torch.Size([1, 11644]) Input IDs shape: torch.Size([1, 11644]) Labels shape: torch.Size([1, 11644]) Final batch size: 1, sequence length: 35803 Attention mask shape: torch.Size([1, 1, 35803, 35803]) Position ids shape: torch.Size([1, 35803]) Input IDs shape: torch.Size([1, 35803]) Labels shape: torch.Size([1, 35803]) Final batch size: 1, sequence length: 39919 Attention mask shape: torch.Size([1, 1, 39919, 39919]) Position ids shape: torch.Size([1, 39919]) Input IDs shape: torch.Size([1, 39919]) Labels shape: torch.Size([1, 39919]) Final batch size: 1, sequence length: 28781 Attention mask shape: torch.Size([1, 1, 28781, 28781]) Position ids shape: torch.Size([1, 28781]) Input IDs shape: torch.Size([1, 28781]) Labels shape: torch.Size([1, 28781]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 36794 Attention mask shape: torch.Size([1, 1, 36794, 36794]) Position ids shape: torch.Size([1, 36794]) Input IDs shape: torch.Size([1, 36794]) Labels shape: torch.Size([1, 36794]) Final batch size: 1, sequence length: 19846 Attention mask shape: torch.Size([1, 1, 19846, 19846]) Position ids shape: torch.Size([1, 19846]) Input IDs shape: torch.Size([1, 19846]) Labels shape: torch.Size([1, 19846]) Final batch size: 1, sequence length: 31340 Attention mask shape: torch.Size([1, 1, 31340, 31340]) Position ids shape: torch.Size([1, 31340]) Input IDs shape: torch.Size([1, 31340]) Labels shape: torch.Size([1, 31340]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 31359 Attention mask shape: torch.Size([1, 1, 31359, 31359]) Position ids shape: torch.Size([1, 31359]) Input IDs shape: torch.Size([1, 31359]) Labels shape: torch.Size([1, 31359]) Final batch size: 1, sequence length: 18379 Attention mask shape: torch.Size([1, 1, 18379, 18379]) Position ids shape: torch.Size([1, 18379]) Input IDs shape: torch.Size([1, 18379]) Labels shape: torch.Size([1, 18379]) Final batch size: 1, sequence length: 7364 Attention mask shape: torch.Size([1, 1, 7364, 7364]) Position ids shape: torch.Size([1, 7364]) Input IDs shape: torch.Size([1, 7364]) Labels shape: torch.Size([1, 7364]) Final batch size: 1, sequence length: 40065 Attention mask shape: torch.Size([1, 1, 40065, 40065]) Position ids shape: torch.Size([1, 40065]) Input IDs shape: torch.Size([1, 40065]) Labels shape: torch.Size([1, 40065]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 12418 Attention mask shape: torch.Size([1, 1, 12418, 12418]) Position ids shape: torch.Size([1, 12418]) Input IDs shape: torch.Size([1, 12418]) Labels shape: torch.Size([1, 12418]) Final batch size: 1, sequence length: 9309 Attention mask shape: torch.Size([1, 1, 9309, 9309]) Position ids shape: torch.Size([1, 9309]) Input IDs shape: torch.Size([1, 9309]) Labels shape: torch.Size([1, 9309]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26706 Attention mask shape: torch.Size([1, 1, 26706, 26706]) Position ids shape: torch.Size([1, 26706]) Input IDs shape: torch.Size([1, 26706]) Labels shape: torch.Size([1, 26706]) Final batch size: 1, sequence length: 25560 Attention mask shape: torch.Size([1, 1, 25560, 25560]) Position ids shape: torch.Size([1, 25560]) Input IDs shape: torch.Size([1, 25560]) Labels shape: torch.Size([1, 25560]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 26221 Attention mask shape: torch.Size([1, 1, 26221, 26221]) Position ids shape: torch.Size([1, 26221]) Input IDs shape: torch.Size([1, 26221]) Labels shape: torch.Size([1, 26221]) Final batch size: 1, sequence length: 40874 Attention mask shape: torch.Size([1, 1, 40874, 40874]) Position ids shape: torch.Size([1, 40874]) Input IDs shape: torch.Size([1, 40874]) Labels shape: torch.Size([1, 40874]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25923 Attention mask shape: torch.Size([1, 1, 25923, 25923]) Position ids shape: torch.Size([1, 25923]) Input IDs shape: torch.Size([1, 25923]) Labels shape: torch.Size([1, 25923]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 17224 Attention mask shape: torch.Size([1, 1, 17224, 17224]) Position ids shape: torch.Size([1, 17224]) Input IDs shape: torch.Size([1, 17224]) Labels shape: torch.Size([1, 17224]) Final batch size: 1, sequence length: 18654 Attention mask shape: torch.Size([1, 1, 18654, 18654]) Position ids shape: torch.Size([1, 18654]) Input IDs shape: torch.Size([1, 18654]) Labels shape: torch.Size([1, 18654]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 25820 Attention mask shape: torch.Size([1, 1, 25820, 25820]) Position ids shape: torch.Size([1, 25820]) Input IDs shape: torch.Size([1, 25820]) Labels shape: torch.Size([1, 25820]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 30709 Attention mask shape: torch.Size([1, 1, 30709, 30709]) Position ids shape: torch.Size([1, 30709]) Input IDs shape: torch.Size([1, 30709]) Labels shape: torch.Size([1, 30709]) Final batch size: 1, sequence length: 29526 Attention mask shape: torch.Size([1, 1, 29526, 29526]) Position ids shape: torch.Size([1, 29526]) Input IDs shape: torch.Size([1, 29526]) Labels shape: torch.Size([1, 29526]) {'loss': 0.2574, 'grad_norm': 0.14953731539951048, 'learning_rate': 6.15582970243117e-08, 'num_tokens': -inf, 'epoch': 7.75} Final batch size: 1, sequence length: 7461 Attention mask shape: torch.Size([1, 1, 7461, 7461]) Position ids shape: torch.Size([1, 7461]) Input IDs shape: torch.Size([1, 7461]) Labels shape: torch.Size([1, 7461]) Final batch size: 1, sequence length: 10826 Attention mask shape: torch.Size([1, 1, 10826, 10826]) Position ids shape: torch.Size([1, 10826]) Input IDs shape: torch.Size([1, 10826]) Labels shape: torch.Size([1, 10826]) Final batch size: 1, sequence length: 9858 Attention mask shape: torch.Size([1, 1, 9858, 9858]) Position ids shape: torch.Size([1, 9858]) Input IDs shape: torch.Size([1, 9858]) Labels shape: torch.Size([1, 9858]) Final batch size: 1, sequence length: 12960 Attention mask shape: torch.Size([1, 1, 12960, 12960]) Position ids shape: torch.Size([1, 12960]) Input IDs shape: torch.Size([1, 12960]) Labels shape: torch.Size([1, 12960]) Final batch size: 1, sequence length: 12096 Attention mask shape: torch.Size([1, 1, 12096, 12096]) Position ids shape: torch.Size([1, 12096]) Input IDs shape: torch.Size([1, 12096]) Labels shape: torch.Size([1, 12096]) Final batch size: 1, sequence length: 12279 Attention mask shape: torch.Size([1, 1, 12279, 12279]) Position ids shape: torch.Size([1, 12279]) Input IDs shape: torch.Size([1, 12279]) Labels shape: torch.Size([1, 12279]) Final batch size: 1, sequence length: 13427 Attention mask shape: torch.Size([1, 1, 13427, 13427]) Position ids shape: torch.Size([1, 13427]) Input IDs shape: torch.Size([1, 13427]) Labels shape: torch.Size([1, 13427]) Final batch size: 1, sequence length: 15933 Attention mask shape: torch.Size([1, 1, 15933, 15933]) Position ids shape: torch.Size([1, 15933]) Input IDs shape: torch.Size([1, 15933]) Labels shape: torch.Size([1, 15933]) Final batch size: 1, sequence length: 13550 Attention mask shape: torch.Size([1, 1, 13550, 13550]) Position ids shape: torch.Size([1, 13550]) Input IDs shape: torch.Size([1, 13550]) Labels shape: torch.Size([1, 13550]) Final batch size: 1, sequence length: 15673 Attention mask shape: torch.Size([1, 1, 15673, 15673]) Position ids shape: torch.Size([1, 15673]) Input IDs shape: torch.Size([1, 15673]) Labels shape: torch.Size([1, 15673]) Final batch size: 1, sequence length: 12622 Attention mask shape: torch.Size([1, 1, 12622, 12622]) Position ids shape: torch.Size([1, 12622]) Input IDs shape: torch.Size([1, 12622]) Labels shape: torch.Size([1, 12622]) Final batch size: 1, sequence length: 18214 Attention mask shape: torch.Size([1, 1, 18214, 18214]) Position ids shape: torch.Size([1, 18214]) Input IDs shape: torch.Size([1, 18214]) Labels shape: torch.Size([1, 18214]) Final batch size: 1, sequence length: 13789 Attention mask shape: torch.Size([1, 1, 13789, 13789]) Position ids shape: torch.Size([1, 13789]) Input IDs shape: torch.Size([1, 13789]) Labels shape: torch.Size([1, 13789]) Final batch size: 1, sequence length: 16450 Attention mask shape: torch.Size([1, 1, 16450, 16450]) Position ids shape: torch.Size([1, 16450]) Input IDs shape: torch.Size([1, 16450]) Labels shape: torch.Size([1, 16450]) Final batch size: 1, sequence length: 17245 Attention mask shape: torch.Size([1, 1, 17245, 17245]) Position ids shape: torch.Size([1, 17245]) Input IDs shape: torch.Size([1, 17245]) Labels shape: torch.Size([1, 17245]) Final batch size: 1, sequence length: 17934 Attention mask shape: torch.Size([1, 1, 17934, 17934]) Position ids shape: torch.Size([1, 17934]) Input IDs shape: torch.Size([1, 17934]) Labels shape: torch.Size([1, 17934]) Final batch size: 1, sequence length: 17861 Attention mask shape: torch.Size([1, 1, 17861, 17861]) Position ids shape: torch.Size([1, 17861]) Input IDs shape: torch.Size([1, 17861]) Labels shape: torch.Size([1, 17861]) Final batch size: 1, sequence length: 20292 Attention mask shape: torch.Size([1, 1, 20292, 20292]) Position ids shape: torch.Size([1, 20292]) Input IDs shape: torch.Size([1, 20292]) Labels shape: torch.Size([1, 20292]) Final batch size: 1, sequence length: 21408 Attention mask shape: torch.Size([1, 1, 21408, 21408]) Position ids shape: torch.Size([1, 21408]) Input IDs shape: torch.Size([1, 21408]) Labels shape: torch.Size([1, 21408]) Final batch size: 1, sequence length: 16927 Attention mask shape: torch.Size([1, 1, 16927, 16927]) Position ids shape: torch.Size([1, 16927]) Input IDs shape: torch.Size([1, 16927]) Labels shape: torch.Size([1, 16927]) Final batch size: 1, sequence length: 21091 Attention mask shape: torch.Size([1, 1, 21091, 21091]) Position ids shape: torch.Size([1, 21091]) Input IDs shape: torch.Size([1, 21091]) Labels shape: torch.Size([1, 21091]) Final batch size: 1, sequence length: 23004 Attention mask shape: torch.Size([1, 1, 23004, 23004]) Position ids shape: torch.Size([1, 23004]) Input IDs shape: torch.Size([1, 23004]) Labels shape: torch.Size([1, 23004]) Final batch size: 1, sequence length: 21675 Attention mask shape: torch.Size([1, 1, 21675, 21675]) Position ids shape: torch.Size([1, 21675]) Input IDs shape: torch.Size([1, 21675]) Labels shape: torch.Size([1, 21675]) Final batch size: 1, sequence length: 16036 Attention mask shape: torch.Size([1, 1, 16036, 16036]) Position ids shape: torch.Size([1, 16036]) Input IDs shape: torch.Size([1, 16036]) Labels shape: torch.Size([1, 16036]) Final batch size: 1, sequence length: 19899 Attention mask shape: torch.Size([1, 1, 19899, 19899]) Position ids shape: torch.Size([1, 19899]) Input IDs shape: torch.Size([1, 19899]) Labels shape: torch.Size([1, 19899]) Final batch size: 1, sequence length: 23896 Attention mask shape: torch.Size([1, 1, 23896, 23896]) Position ids shape: torch.Size([1, 23896]) Input IDs shape: torch.Size([1, 23896]) Labels shape: torch.Size([1, 23896]) Final batch size: 1, sequence length: 20492 Attention mask shape: torch.Size([1, 1, 20492, 20492]) Position ids shape: torch.Size([1, 20492]) Input IDs shape: torch.Size([1, 20492]) Labels shape: torch.Size([1, 20492]) Final batch size: 1, sequence length: 20065 Attention mask shape: torch.Size([1, 1, 20065, 20065]) Position ids shape: torch.Size([1, 20065]) Input IDs shape: torch.Size([1, 20065]) Labels shape: torch.Size([1, 20065]) Final batch size: 1, sequence length: 12728 Attention mask shape: torch.Size([1, 1, 12728, 12728]) Position ids shape: torch.Size([1, 12728]) Input IDs shape: torch.Size([1, 12728]) Labels shape: torch.Size([1, 12728]) Final batch size: 1, sequence length: 25600 Attention mask shape: torch.Size([1, 1, 25600, 25600]) Position ids shape: torch.Size([1, 25600]) Input IDs shape: torch.Size([1, 25600]) Labels shape: torch.Size([1, 25600]) Final batch size: 1, sequence length: 24956 Attention mask shape: torch.Size([1, 1, 24956, 24956]) Position ids shape: torch.Size([1, 24956]) Input IDs shape: torch.Size([1, 24956]) Labels shape: torch.Size([1, 24956]) Final batch size: 1, sequence length: 19187 Attention mask shape: torch.Size([1, 1, 19187, 19187]) Position ids shape: torch.Size([1, 19187]) Input IDs shape: torch.Size([1, 19187]) Labels shape: torch.Size([1, 19187]) Final batch size: 1, sequence length: 25176 Attention mask shape: torch.Size([1, 1, 25176, 25176]) Position ids shape: torch.Size([1, 25176]) Input IDs shape: torch.Size([1, 25176]) Labels shape: torch.Size([1, 25176]) Final batch size: 1, sequence length: 21586 Attention mask shape: torch.Size([1, 1, 21586, 21586]) Position ids shape: torch.Size([1, 21586]) Input IDs shape: torch.Size([1, 21586]) Labels shape: torch.Size([1, 21586]) Final batch size: 1, sequence length: 24808 Attention mask shape: torch.Size([1, 1, 24808, 24808]) Position ids shape: torch.Size([1, 24808]) Input IDs shape: torch.Size([1, 24808]) Labels shape: torch.Size([1, 24808]) Final batch size: 1, sequence length: 26500 Attention mask shape: torch.Size([1, 1, 26500, 26500]) Position ids shape: torch.Size([1, 26500]) Input IDs shape: torch.Size([1, 26500]) Labels shape: torch.Size([1, 26500]) Final batch size: 1, sequence length: 22778 Attention mask shape: torch.Size([1, 1, 22778, 22778]) Position ids shape: torch.Size([1, 22778]) Input IDs shape: torch.Size([1, 22778]) Labels shape: torch.Size([1, 22778]) Final batch size: 1, sequence length: 18832 Attention mask shape: torch.Size([1, 1, 18832, 18832]) Position ids shape: torch.Size([1, 18832]) Input IDs shape: torch.Size([1, 18832]) Labels shape: torch.Size([1, 18832]) Final batch size: 1, sequence length: 26247 Attention mask shape: torch.Size([1, 1, 26247, 26247]) Position ids shape: torch.Size([1, 26247]) Input IDs shape: torch.Size([1, 26247]) Labels shape: torch.Size([1, 26247]) Final batch size: 1, sequence length: 3010 Attention mask shape: torch.Size([1, 1, 3010, 3010]) Position ids shape: torch.Size([1, 3010]) Input IDs shape: torch.Size([1, 3010]) Labels shape: torch.Size([1, 3010]) Final batch size: 1, sequence length: 25325 Attention mask shape: torch.Size([1, 1, 25325, 25325]) Position ids shape: torch.Size([1, 25325]) Input IDs shape: torch.Size([1, 25325]) Labels shape: torch.Size([1, 25325]) Final batch size: 1, sequence length: 20582 Attention mask shape: torch.Size([1, 1, 20582, 20582]) Position ids shape: torch.Size([1, 20582]) Input IDs shape: torch.Size([1, 20582]) Labels shape: torch.Size([1, 20582]) Final batch size: 1, sequence length: 25035 Attention mask shape: torch.Size([1, 1, 25035, 25035]) Position ids shape: torch.Size([1, 25035]) Input IDs shape: torch.Size([1, 25035]) Labels shape: torch.Size([1, 25035]) Final batch size: 1, sequence length: 26316 Attention mask shape: torch.Size([1, 1, 26316, 26316]) Position ids shape: torch.Size([1, 26316]) Input IDs shape: torch.Size([1, 26316]) Labels shape: torch.Size([1, 26316]) Final batch size: 1, sequence length: 17763 Attention mask shape: torch.Size([1, 1, 17763, 17763]) Position ids shape: torch.Size([1, 17763]) Input IDs shape: torch.Size([1, 17763]) Labels shape: torch.Size([1, 17763]) Final batch size: 1, sequence length: 27222 Attention mask shape: torch.Size([1, 1, 27222, 27222]) Position ids shape: torch.Size([1, 27222]) Input IDs shape: torch.Size([1, 27222]) Labels shape: torch.Size([1, 27222]) Final batch size: 1, sequence length: 25435 Attention mask shape: torch.Size([1, 1, 25435, 25435]) Position ids shape: torch.Size([1, 25435]) Input IDs shape: torch.Size([1, 25435]) Labels shape: torch.Size([1, 25435]) Final batch size: 1, sequence length: 28926 Attention mask shape: torch.Size([1, 1, 28926, 28926]) Position ids shape: torch.Size([1, 28926]) Input IDs shape: torch.Size([1, 28926]) Labels shape: torch.Size([1, 28926]) Final batch size: 1, sequence length: 13946 Attention mask shape: torch.Size([1, 1, 13946, 13946]) Position ids shape: torch.Size([1, 13946]) Input IDs shape: torch.Size([1, 13946]) Labels shape: torch.Size([1, 13946]) Final batch size: 1, sequence length: 22775 Attention mask shape: torch.Size([1, 1, 22775, 22775]) Position ids shape: torch.Size([1, 22775]) Input IDs shape: torch.Size([1, 22775]) Labels shape: torch.Size([1, 22775]) Final batch size: 1, sequence length: 24927 Attention mask shape: torch.Size([1, 1, 24927, 24927]) Position ids shape: torch.Size([1, 24927]) Input IDs shape: torch.Size([1, 24927]) Labels shape: torch.Size([1, 24927]) Final batch size: 1, sequence length: 24965 Attention mask shape: torch.Size([1, 1, 24965, 24965]) Position ids shape: torch.Size([1, 24965]) Input IDs shape: torch.Size([1, 24965]) Labels shape: torch.Size([1, 24965]) Final batch size: 1, sequence length: 14212 Attention mask shape: torch.Size([1, 1, 14212, 14212]) Position ids shape: torch.Size([1, 14212]) Input IDs shape: torch.Size([1, 14212]) Labels shape: torch.Size([1, 14212]) Final batch size: 1, sequence length: 6978 Attention mask shape: torch.Size([1, 1, 6978, 6978]) Position ids shape: torch.Size([1, 6978]) Input IDs shape: torch.Size([1, 6978]) Labels shape: torch.Size([1, 6978]) Final batch size: 1, sequence length: 19953 Attention mask shape: torch.Size([1, 1, 19953, 19953]) Position ids shape: torch.Size([1, 19953]) Input IDs shape: torch.Size([1, 19953]) Labels shape: torch.Size([1, 19953]) Final batch size: 1, sequence length: 16714 Attention mask shape: torch.Size([1, 1, 16714, 16714]) Position ids shape: torch.Size([1, 16714]) Input IDs shape: torch.Size([1, 16714]) Labels shape: torch.Size([1, 16714]) Final batch size: 1, sequence length: 13453 Attention mask shape: torch.Size([1, 1, 13453, 13453]) Position ids shape: torch.Size([1, 13453]) Input IDs shape: torch.Size([1, 13453]) Labels shape: torch.Size([1, 13453]) Final batch size: 1, sequence length: 12426 Attention mask shape: torch.Size([1, 1, 12426, 12426]) Position ids shape: torch.Size([1, 12426]) Input IDs shape: torch.Size([1, 12426]) Labels shape: torch.Size([1, 12426]) Final batch size: 1, sequence length: 26520 Attention mask shape: torch.Size([1, 1, 26520, 26520]) Position ids shape: torch.Size([1, 26520]) Input IDs shape: torch.Size([1, 26520]) Labels shape: torch.Size([1, 26520]) Final batch size: 1, sequence length: 32101 Attention mask shape: torch.Size([1, 1, 32101, 32101]) Position ids shape: torch.Size([1, 32101]) Input IDs shape: torch.Size([1, 32101]) Labels shape: torch.Size([1, 32101]) Final batch size: 1, sequence length: 28644 Attention mask shape: torch.Size([1, 1, 28644, 28644]) Position ids shape: torch.Size([1, 28644]) Input IDs shape: torch.Size([1, 28644]) Labels shape: torch.Size([1, 28644]) Final batch size: 1, sequence length: 24633 Attention mask shape: torch.Size([1, 1, 24633, 24633]) Position ids shape: torch.Size([1, 24633]) Input IDs shape: torch.Size([1, 24633]) Labels shape: torch.Size([1, 24633]) Final batch size: 1, sequence length: 28552 Attention mask shape: torch.Size([1, 1, 28552, 28552]) Position ids shape: torch.Size([1, 28552]) Input IDs shape: torch.Size([1, 28552]) Labels shape: torch.Size([1, 28552]) Final batch size: 1, sequence length: 32513 Attention mask shape: torch.Size([1, 1, 32513, 32513]) Position ids shape: torch.Size([1, 32513]) Input IDs shape: torch.Size([1, 32513]) Labels shape: torch.Size([1, 32513]) Final batch size: 1, sequence length: 25979 Attention mask shape: torch.Size([1, 1, 25979, 25979]) Position ids shape: torch.Size([1, 25979]) Input IDs shape: torch.Size([1, 25979]) Labels shape: torch.Size([1, 25979]) Final batch size: 1, sequence length: 30635 Attention mask shape: torch.Size([1, 1, 30635, 30635]) Position ids shape: torch.Size([1, 30635]) Input IDs shape: torch.Size([1, 30635]) Labels shape: torch.Size([1, 30635]) Final batch size: 1, sequence length: 32100 Attention mask shape: torch.Size([1, 1, 32100, 32100]) Position ids shape: torch.Size([1, 32100]) Input IDs shape: torch.Size([1, 32100]) Labels shape: torch.Size([1, 32100]) Final batch size: 1, sequence length: 35058 Attention mask shape: torch.Size([1, 1, 35058, 35058]) Position ids shape: torch.Size([1, 35058]) Input IDs shape: torch.Size([1, 35058]) Labels shape: torch.Size([1, 35058]) Final batch size: 1, sequence length: 33621 Attention mask shape: torch.Size([1, 1, 33621, 33621]) Position ids shape: torch.Size([1, 33621]) Input IDs shape: torch.Size([1, 33621]) Labels shape: torch.Size([1, 33621]) Final batch size: 1, sequence length: 37381 Attention mask shape: torch.Size([1, 1, 37381, 37381]) Position ids shape: torch.Size([1, 37381]) Input IDs shape: torch.Size([1, 37381]) Labels shape: torch.Size([1, 37381]) Final batch size: 1, sequence length: 17049 Attention mask shape: torch.Size([1, 1, 17049, 17049]) Position ids shape: torch.Size([1, 17049]) Input IDs shape: torch.Size([1, 17049]) Labels shape: torch.Size([1, 17049]) Final batch size: 1, sequence length: 31208 Attention mask shape: torch.Size([1, 1, 31208, 31208]) Position ids shape: torch.Size([1, 31208]) Input IDs shape: torch.Size([1, 31208]) Labels shape: torch.Size([1, 31208]) Final batch size: 1, sequence length: 35014 Attention mask shape: torch.Size([1, 1, 35014, 35014]) Position ids shape: torch.Size([1, 35014]) Input IDs shape: torch.Size([1, 35014]) Labels shape: torch.Size([1, 35014]) Final batch size: 1, sequence length: 16571 Attention mask shape: torch.Size([1, 1, 16571, 16571]) Position ids shape: torch.Size([1, 16571]) Input IDs shape: torch.Size([1, 16571]) Labels shape: torch.Size([1, 16571]) Final batch size: 1, sequence length: 25147 Attention mask shape: torch.Size([1, 1, 25147, 25147]) Position ids shape: torch.Size([1, 25147]) Input IDs shape: torch.Size([1, 25147]) Labels shape: torch.Size([1, 25147]) Final batch size: 1, sequence length: 24676 Attention mask shape: torch.Size([1, 1, 24676, 24676]) Position ids shape: torch.Size([1, 24676]) Input IDs shape: torch.Size([1, 24676]) Labels shape: torch.Size([1, 24676]) Final batch size: 1, sequence length: 28377 Attention mask shape: torch.Size([1, 1, 28377, 28377]) Position ids shape: torch.Size([1, 28377]) Input IDs shape: torch.Size([1, 28377]) Labels shape: torch.Size([1, 28377]) Final batch size: 1, sequence length: 26789 Attention mask shape: torch.Size([1, 1, 26789, 26789]) Position ids shape: torch.Size([1, 26789]) Input IDs shape: torch.Size([1, 26789]) Labels shape: torch.Size([1, 26789]) Final batch size: 1, sequence length: 31662 Attention mask shape: torch.Size([1, 1, 31662, 31662]) Position ids shape: torch.Size([1, 31662]) Input IDs shape: torch.Size([1, 31662]) Labels shape: torch.Size([1, 31662]) Final batch size: 1, sequence length: 36047 Attention mask shape: torch.Size([1, 1, 36047, 36047]) Position ids shape: torch.Size([1, 36047]) Input IDs shape: torch.Size([1, 36047]) Labels shape: torch.Size([1, 36047]) Final batch size: 1, sequence length: 16816 Attention mask shape: torch.Size([1, 1, 16816, 16816]) Position ids shape: torch.Size([1, 16816]) Input IDs shape: torch.Size([1, 16816]) Labels shape: torch.Size([1, 16816]) Final batch size: 1, sequence length: 38391 Attention mask shape: torch.Size([1, 1, 38391, 38391]) Position ids shape: torch.Size([1, 38391]) Input IDs shape: torch.Size([1, 38391]) Labels shape: torch.Size([1, 38391]) Final batch size: 1, sequence length: 39474 Attention mask shape: torch.Size([1, 1, 39474, 39474]) Position ids shape: torch.Size([1, 39474]) Input IDs shape: torch.Size([1, 39474]) Labels shape: torch.Size([1, 39474]) Final batch size: 1, sequence length: 25219 Attention mask shape: torch.Size([1, 1, 25219, 25219]) Position ids shape: torch.Size([1, 25219]) Input IDs shape: torch.Size([1, 25219]) Labels shape: torch.Size([1, 25219]) Final batch size: 1, sequence length: 30259 Attention mask shape: torch.Size([1, 1, 30259, 30259]) Position ids shape: torch.Size([1, 30259]) Input IDs shape: torch.Size([1, 30259]) Labels shape: torch.Size([1, 30259]) Final batch size: 1, sequence length: 37720 Attention mask shape: torch.Size([1, 1, 37720, 37720]) Position ids shape: torch.Size([1, 37720]) Input IDs shape: torch.Size([1, 37720]) Labels shape: torch.Size([1, 37720]) Final batch size: 1, sequence length: 21290 Attention mask shape: torch.Size([1, 1, 21290, 21290]) Position ids shape: torch.Size([1, 21290]) Input IDs shape: torch.Size([1, 21290]) Labels shape: torch.Size([1, 21290]) Final batch size: 1, sequence length: 37391 Attention mask shape: torch.Size([1, 1, 37391, 37391]) Position ids shape: torch.Size([1, 37391]) Input IDs shape: torch.Size([1, 37391]) Labels shape: torch.Size([1, 37391]) Final batch size: 1, sequence length: 22710 Attention mask shape: torch.Size([1, 1, 22710, 22710]) Position ids shape: torch.Size([1, 22710]) Input IDs shape: torch.Size([1, 22710]) Labels shape: torch.Size([1, 22710]) Final batch size: 1, sequence length: 39955 Attention mask shape: torch.Size([1, 1, 39955, 39955]) Position ids shape: torch.Size([1, 39955]) Input IDs shape: torch.Size([1, 39955]) Labels shape: torch.Size([1, 39955]) Final batch size: 1, sequence length: 31447 Attention mask shape: torch.Size([1, 1, 31447, 31447]) Position ids shape: torch.Size([1, 31447]) Input IDs shape: torch.Size([1, 31447]) Labels shape: torch.Size([1, 31447]) Final batch size: 1, sequence length: 30944 Attention mask shape: torch.Size([1, 1, 30944, 30944]) Position ids shape: torch.Size([1, 30944]) Input IDs shape: torch.Size([1, 30944]) Labels shape: torch.Size([1, 30944]) Final batch size: 1, sequence length: 25774 Attention mask shape: torch.Size([1, 1, 25774, 25774]) Position ids shape: torch.Size([1, 25774]) Input IDs shape: torch.Size([1, 25774]) Labels shape: torch.Size([1, 25774]) Final batch size: 1, sequence length: 26454 Attention mask shape: torch.Size([1, 1, 26454, 26454]) Position ids shape: torch.Size([1, 26454]) Input IDs shape: torch.Size([1, 26454]) Labels shape: torch.Size([1, 26454]) Final batch size: 1, sequence length: 21321 Attention mask shape: torch.Size([1, 1, 21321, 21321]) Position ids shape: torch.Size([1, 21321]) Input IDs shape: torch.Size([1, 21321]) Labels shape: torch.Size([1, 21321]) Final batch size: 1, sequence length: 35261 Attention mask shape: torch.Size([1, 1, 35261, 35261]) Position ids shape: torch.Size([1, 35261]) Input IDs shape: torch.Size([1, 35261]) Labels shape: torch.Size([1, 35261]) Final batch size: 1, sequence length: 19441 Attention mask shape: torch.Size([1, 1, 19441, 19441]) Position ids shape: torch.Size([1, 19441]) Input IDs shape: torch.Size([1, 19441]) Labels shape: torch.Size([1, 19441]) Final batch size: 1, sequence length: 26387 Attention mask shape: torch.Size([1, 1, 26387, 26387]) Position ids shape: torch.Size([1, 26387]) Input IDs shape: torch.Size([1, 26387]) Labels shape: torch.Size([1, 26387]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 38577 Attention mask shape: torch.Size([1, 1, 38577, 38577]) Position ids shape: torch.Size([1, 38577]) Input IDs shape: torch.Size([1, 38577]) Labels shape: torch.Size([1, 38577]) Final batch size: 1, sequence length: 32129 Attention mask shape: torch.Size([1, 1, 32129, 32129]) Position ids shape: torch.Size([1, 32129]) Input IDs shape: torch.Size([1, 32129]) Labels shape: torch.Size([1, 32129]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40590 Attention mask shape: torch.Size([1, 1, 40590, 40590]) Position ids shape: torch.Size([1, 40590]) Input IDs shape: torch.Size([1, 40590]) Labels shape: torch.Size([1, 40590]) Final batch size: 1, sequence length: 14136 Attention mask shape: torch.Size([1, 1, 14136, 14136]) Position ids shape: torch.Size([1, 14136]) Input IDs shape: torch.Size([1, 14136]) Labels shape: torch.Size([1, 14136]) Final batch size: 1, sequence length: 36356 Attention mask shape: torch.Size([1, 1, 36356, 36356]) Position ids shape: torch.Size([1, 36356]) Input IDs shape: torch.Size([1, 36356]) Labels shape: torch.Size([1, 36356]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 27021 Attention mask shape: torch.Size([1, 1, 27021, 27021]) Position ids shape: torch.Size([1, 27021]) Input IDs shape: torch.Size([1, 27021]) Labels shape: torch.Size([1, 27021]) Final batch size: 1, sequence length: 32079 Attention mask shape: torch.Size([1, 1, 32079, 32079]) Position ids shape: torch.Size([1, 32079]) Input IDs shape: torch.Size([1, 32079]) Labels shape: torch.Size([1, 32079]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33014 Attention mask shape: torch.Size([1, 1, 33014, 33014]) Position ids shape: torch.Size([1, 33014]) Input IDs shape: torch.Size([1, 33014]) Labels shape: torch.Size([1, 33014]) Final batch size: 1, sequence length: 27371 Attention mask shape: torch.Size([1, 1, 27371, 27371]) Position ids shape: torch.Size([1, 27371]) Input IDs shape: torch.Size([1, 27371]) Labels shape: torch.Size([1, 27371]) Final batch size: 1, sequence length: 28448 Attention mask shape: torch.Size([1, 1, 28448, 28448]) Position ids shape: torch.Size([1, 28448]) Input IDs shape: torch.Size([1, 28448]) Labels shape: torch.Size([1, 28448]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 19027 Attention mask shape: torch.Size([1, 1, 19027, 19027]) Position ids shape: torch.Size([1, 19027]) Input IDs shape: torch.Size([1, 19027]) Labels shape: torch.Size([1, 19027]) Final batch size: 1, sequence length: 24623 Attention mask shape: torch.Size([1, 1, 24623, 24623]) Position ids shape: torch.Size([1, 24623]) Input IDs shape: torch.Size([1, 24623]) Labels shape: torch.Size([1, 24623]) Final batch size: 1, sequence length: 30712 Attention mask shape: torch.Size([1, 1, 30712, 30712]) Position ids shape: torch.Size([1, 30712]) Input IDs shape: torch.Size([1, 30712]) Labels shape: torch.Size([1, 30712]) Final batch size: 1, sequence length: 23208 Attention mask shape: torch.Size([1, 1, 23208, 23208]) Position ids shape: torch.Size([1, 23208]) Input IDs shape: torch.Size([1, 23208]) Labels shape: torch.Size([1, 23208]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.2543, 'grad_norm': 0.1325912662582723, 'learning_rate': 2.7390523158633552e-08, 'num_tokens': -inf, 'epoch': 7.88} Final batch size: 1, sequence length: 6986 Attention mask shape: torch.Size([1, 1, 6986, 6986]) Position ids shape: torch.Size([1, 6986]) Input IDs shape: torch.Size([1, 6986]) Labels shape: torch.Size([1, 6986]) Final batch size: 1, sequence length: 4010 Attention mask shape: torch.Size([1, 1, 4010, 4010]) Position ids shape: torch.Size([1, 4010]) Input IDs shape: torch.Size([1, 4010]) Labels shape: torch.Size([1, 4010]) Final batch size: 1, sequence length: 10937 Attention mask shape: torch.Size([1, 1, 10937, 10937]) Position ids shape: torch.Size([1, 10937]) Input IDs shape: torch.Size([1, 10937]) Labels shape: torch.Size([1, 10937]) Final batch size: 1, sequence length: 10507 Attention mask shape: torch.Size([1, 1, 10507, 10507]) Position ids shape: torch.Size([1, 10507]) Input IDs shape: torch.Size([1, 10507]) Labels shape: torch.Size([1, 10507]) Final batch size: 1, sequence length: 11150 Attention mask shape: torch.Size([1, 1, 11150, 11150]) Position ids shape: torch.Size([1, 11150]) Input IDs shape: torch.Size([1, 11150]) Labels shape: torch.Size([1, 11150]) Final batch size: 1, sequence length: 12259 Attention mask shape: torch.Size([1, 1, 12259, 12259]) Position ids shape: torch.Size([1, 12259]) Input IDs shape: torch.Size([1, 12259]) Labels shape: torch.Size([1, 12259]) Final batch size: 1, sequence length: 11122 Attention mask shape: torch.Size([1, 1, 11122, 11122]) Position ids shape: torch.Size([1, 11122]) Input IDs shape: torch.Size([1, 11122]) Labels shape: torch.Size([1, 11122]) Final batch size: 1, sequence length: 11000 Attention mask shape: torch.Size([1, 1, 11000, 11000]) Position ids shape: torch.Size([1, 11000]) Input IDs shape: torch.Size([1, 11000]) Labels shape: torch.Size([1, 11000]) Final batch size: 1, sequence length: 13264 Attention mask shape: torch.Size([1, 1, 13264, 13264]) Position ids shape: torch.Size([1, 13264]) Input IDs shape: torch.Size([1, 13264]) Labels shape: torch.Size([1, 13264]) Final batch size: 1, sequence length: 16137 Attention mask shape: torch.Size([1, 1, 16137, 16137]) Position ids shape: torch.Size([1, 16137]) Input IDs shape: torch.Size([1, 16137]) Labels shape: torch.Size([1, 16137]) Final batch size: 1, sequence length: 15565 Attention mask shape: torch.Size([1, 1, 15565, 15565]) Position ids shape: torch.Size([1, 15565]) Input IDs shape: torch.Size([1, 15565]) Labels shape: torch.Size([1, 15565]) Final batch size: 1, sequence length: 16187 Attention mask shape: torch.Size([1, 1, 16187, 16187]) Position ids shape: torch.Size([1, 16187]) Input IDs shape: torch.Size([1, 16187]) Labels shape: torch.Size([1, 16187]) Final batch size: 1, sequence length: 18630 Attention mask shape: torch.Size([1, 1, 18630, 18630]) Position ids shape: torch.Size([1, 18630]) Input IDs shape: torch.Size([1, 18630]) Labels shape: torch.Size([1, 18630]) Final batch size: 1, sequence length: 16514 Attention mask shape: torch.Size([1, 1, 16514, 16514]) Position ids shape: torch.Size([1, 16514]) Input IDs shape: torch.Size([1, 16514]) Labels shape: torch.Size([1, 16514]) Final batch size: 1, sequence length: 15617 Attention mask shape: torch.Size([1, 1, 15617, 15617]) Position ids shape: torch.Size([1, 15617]) Input IDs shape: torch.Size([1, 15617]) Labels shape: torch.Size([1, 15617]) Final batch size: 1, sequence length: 16278 Attention mask shape: torch.Size([1, 1, 16278, 16278]) Position ids shape: torch.Size([1, 16278]) Input IDs shape: torch.Size([1, 16278]) Labels shape: torch.Size([1, 16278]) Final batch size: 1, sequence length: 18320 Attention mask shape: torch.Size([1, 1, 18320, 18320]) Position ids shape: torch.Size([1, 18320]) Input IDs shape: torch.Size([1, 18320]) Labels shape: torch.Size([1, 18320]) Final batch size: 1, sequence length: 20916 Attention mask shape: torch.Size([1, 1, 20916, 20916]) Position ids shape: torch.Size([1, 20916]) Input IDs shape: torch.Size([1, 20916]) Labels shape: torch.Size([1, 20916]) Final batch size: 1, sequence length: 20522 Attention mask shape: torch.Size([1, 1, 20522, 20522]) Position ids shape: torch.Size([1, 20522]) Input IDs shape: torch.Size([1, 20522]) Labels shape: torch.Size([1, 20522]) Final batch size: 1, sequence length: 19427 Attention mask shape: torch.Size([1, 1, 19427, 19427]) Position ids shape: torch.Size([1, 19427]) Input IDs shape: torch.Size([1, 19427]) Labels shape: torch.Size([1, 19427]) Final batch size: 1, sequence length: 19679 Attention mask shape: torch.Size([1, 1, 19679, 19679]) Position ids shape: torch.Size([1, 19679]) Input IDs shape: torch.Size([1, 19679]) Labels shape: torch.Size([1, 19679]) Final batch size: 1, sequence length: 20907 Attention mask shape: torch.Size([1, 1, 20907, 20907]) Position ids shape: torch.Size([1, 20907]) Input IDs shape: torch.Size([1, 20907]) Labels shape: torch.Size([1, 20907]) Final batch size: 1, sequence length: 21841 Attention mask shape: torch.Size([1, 1, 21841, 21841]) Position ids shape: torch.Size([1, 21841]) Input IDs shape: torch.Size([1, 21841]) Labels shape: torch.Size([1, 21841]) Final batch size: 1, sequence length: 22742 Attention mask shape: torch.Size([1, 1, 22742, 22742]) Position ids shape: torch.Size([1, 22742]) Input IDs shape: torch.Size([1, 22742]) Labels shape: torch.Size([1, 22742]) Final batch size: 1, sequence length: 19840 Attention mask shape: torch.Size([1, 1, 19840, 19840]) Position ids shape: torch.Size([1, 19840]) Input IDs shape: torch.Size([1, 19840]) Labels shape: torch.Size([1, 19840]) Final batch size: 1, sequence length: 22946 Final batch size: 1, sequence length: 21669 Attention mask shape: torch.Size([1, 1, 22946, 22946]) Position ids shape: torch.Size([1, 22946]) Input IDs shape: torch.Size([1, 22946]) Labels shape: torch.Size([1, 22946]) Attention mask shape: torch.Size([1, 1, 21669, 21669]) Position ids shape: torch.Size([1, 21669]) Input IDs shape: torch.Size([1, 21669]) Labels shape: torch.Size([1, 21669]) Final batch size: 1, sequence length: 18155 Attention mask shape: torch.Size([1, 1, 18155, 18155]) Position ids shape: torch.Size([1, 18155]) Input IDs shape: torch.Size([1, 18155]) Labels shape: torch.Size([1, 18155]) Final batch size: 1, sequence length: 26534 Attention mask shape: torch.Size([1, 1, 26534, 26534]) Position ids shape: torch.Size([1, 26534]) Input IDs shape: torch.Size([1, 26534]) Labels shape: torch.Size([1, 26534]) Final batch size: 1, sequence length: 25735 Attention mask shape: torch.Size([1, 1, 25735, 25735]) Position ids shape: torch.Size([1, 25735]) Input IDs shape: torch.Size([1, 25735]) Labels shape: torch.Size([1, 25735]) Final batch size: 1, sequence length: 27195 Attention mask shape: torch.Size([1, 1, 27195, 27195]) Position ids shape: torch.Size([1, 27195]) Input IDs shape: torch.Size([1, 27195]) Labels shape: torch.Size([1, 27195]) Final batch size: 1, sequence length: 26499 Attention mask shape: torch.Size([1, 1, 26499, 26499]) Position ids shape: torch.Size([1, 26499]) Input IDs shape: torch.Size([1, 26499]) Labels shape: torch.Size([1, 26499]) Final batch size: 1, sequence length: 24414 Attention mask shape: torch.Size([1, 1, 24414, 24414]) Position ids shape: torch.Size([1, 24414]) Input IDs shape: torch.Size([1, 24414]) Labels shape: torch.Size([1, 24414]) Final batch size: 1, sequence length: 26619 Attention mask shape: torch.Size([1, 1, 26619, 26619]) Position ids shape: torch.Size([1, 26619]) Input IDs shape: torch.Size([1, 26619]) Labels shape: torch.Size([1, 26619]) Final batch size: 1, sequence length: 25611 Attention mask shape: torch.Size([1, 1, 25611, 25611]) Position ids shape: torch.Size([1, 25611]) Input IDs shape: torch.Size([1, 25611]) Labels shape: torch.Size([1, 25611]) Final batch size: 1, sequence length: 25197 Attention mask shape: torch.Size([1, 1, 25197, 25197]) Position ids shape: torch.Size([1, 25197]) Input IDs shape: torch.Size([1, 25197]) Labels shape: torch.Size([1, 25197]) Final batch size: 1, sequence length: 26271 Attention mask shape: torch.Size([1, 1, 26271, 26271]) Position ids shape: torch.Size([1, 26271]) Input IDs shape: torch.Size([1, 26271]) Labels shape: torch.Size([1, 26271]) Final batch size: 1, sequence length: 25832 Attention mask shape: torch.Size([1, 1, 25832, 25832]) Position ids shape: torch.Size([1, 25832]) Input IDs shape: torch.Size([1, 25832]) Labels shape: torch.Size([1, 25832]) Final batch size: 1, sequence length: 29625 Attention mask shape: torch.Size([1, 1, 29625, 29625]) Position ids shape: torch.Size([1, 29625]) Input IDs shape: torch.Size([1, 29625]) Labels shape: torch.Size([1, 29625]) Final batch size: 1, sequence length: 31232 Attention mask shape: torch.Size([1, 1, 31232, 31232]) Position ids shape: torch.Size([1, 31232]) Input IDs shape: torch.Size([1, 31232]) Labels shape: torch.Size([1, 31232]) Final batch size: 1, sequence length: 29343 Attention mask shape: torch.Size([1, 1, 29343, 29343]) Position ids shape: torch.Size([1, 29343]) Input IDs shape: torch.Size([1, 29343]) Labels shape: torch.Size([1, 29343]) Final batch size: 1, sequence length: 31381 Attention mask shape: torch.Size([1, 1, 31381, 31381]) Position ids shape: torch.Size([1, 31381]) Input IDs shape: torch.Size([1, 31381]) Labels shape: torch.Size([1, 31381]) Final batch size: 1, sequence length: 32662 Attention mask shape: torch.Size([1, 1, 32662, 32662]) Position ids shape: torch.Size([1, 32662]) Input IDs shape: torch.Size([1, 32662]) Labels shape: torch.Size([1, 32662]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 30220 Attention mask shape: torch.Size([1, 1, 30220, 30220]) Position ids shape: torch.Size([1, 30220]) Input IDs shape: torch.Size([1, 30220]) Labels shape: torch.Size([1, 30220]) Final batch size: 1, sequence length: 32613 Attention mask shape: torch.Size([1, 1, 32613, 32613]) Position ids shape: torch.Size([1, 32613]) Input IDs shape: torch.Size([1, 32613]) Labels shape: torch.Size([1, 32613]) Final batch size: 1, sequence length: 33368 Attention mask shape: torch.Size([1, 1, 33368, 33368]) Position ids shape: torch.Size([1, 33368]) Input IDs shape: torch.Size([1, 33368]) Labels shape: torch.Size([1, 33368]) Final batch size: 1, sequence length: 29742 Attention mask shape: torch.Size([1, 1, 29742, 29742]) Position ids shape: torch.Size([1, 29742]) Input IDs shape: torch.Size([1, 29742]) Labels shape: torch.Size([1, 29742]) Final batch size: 1, sequence length: 35103 Attention mask shape: torch.Size([1, 1, 35103, 35103]) Position ids shape: torch.Size([1, 35103]) Input IDs shape: torch.Size([1, 35103]) Labels shape: torch.Size([1, 35103]) Final batch size: 1, sequence length: 35240 Attention mask shape: torch.Size([1, 1, 35240, 35240]) Position ids shape: torch.Size([1, 35240]) Input IDs shape: torch.Size([1, 35240]) Labels shape: torch.Size([1, 35240]) Final batch size: 1, sequence length: 37394 Attention mask shape: torch.Size([1, 1, 37394, 37394]) Position ids shape: torch.Size([1, 37394]) Input IDs shape: torch.Size([1, 37394]) Labels shape: torch.Size([1, 37394]) Final batch size: 1, sequence length: 36860 Attention mask shape: torch.Size([1, 1, 36860, 36860]) Position ids shape: torch.Size([1, 36860]) Input IDs shape: torch.Size([1, 36860]) Labels shape: torch.Size([1, 36860]) Final batch size: 1, sequence length: 34861 Attention mask shape: torch.Size([1, 1, 34861, 34861]) Position ids shape: torch.Size([1, 34861]) Input IDs shape: torch.Size([1, 34861]) Labels shape: torch.Size([1, 34861]) Final batch size: 1, sequence length: 39519 Attention mask shape: torch.Size([1, 1, 39519, 39519]) Position ids shape: torch.Size([1, 39519]) Input IDs shape: torch.Size([1, 39519]) Labels shape: torch.Size([1, 39519]) Final batch size: 1, sequence length: 37939 Attention mask shape: torch.Size([1, 1, 37939, 37939]) Position ids shape: torch.Size([1, 37939]) Input IDs shape: torch.Size([1, 37939]) Labels shape: torch.Size([1, 37939]) Final batch size: 1, sequence length: 31930 Attention mask shape: torch.Size([1, 1, 31930, 31930]) Position ids shape: torch.Size([1, 31930]) Input IDs shape: torch.Size([1, 31930]) Labels shape: torch.Size([1, 31930]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 33442 Attention mask shape: torch.Size([1, 1, 33442, 33442]) Position ids shape: torch.Size([1, 33442]) Input IDs shape: torch.Size([1, 33442]) Labels shape: torch.Size([1, 33442]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) Final batch size: 1, sequence length: 40960 Attention mask shape: torch.Size([1, 1, 40960, 40960]) Position ids shape: torch.Size([1, 40960]) Input IDs shape: torch.Size([1, 40960]) Labels shape: torch.Size([1, 40960]) {'loss': 0.246, 'grad_norm': 0.14429801382277035, 'learning_rate': 6.852326227130835e-09, 'num_tokens': -inf, 'epoch': 8.0} {'train_runtime': 10438.678, 'train_samples_per_second': 0.739, 'train_steps_per_second': 0.006, 'train_loss': 0.2960809483192861, 'epoch': 8.0} wandb: wandb: 🚀 View run runs/dev/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume at: https://wandb.ai/ligeng-zhu/ThreadWeaver/runs/tw-data-train_final_v2_nb2_mt8192_replaced_fix-8node-resume