1428 lines
1.4 MiB
Plaintext
1428 lines
1.4 MiB
Plaintext
|
==== STARTING EXPERIMENT: qwen3_1.7b_if_fft ====
|
|||
|
|
Log File: saves/qwen3_1.7b/if/fft/qwen3_1.7b_if_fft_20260524_054817.log
|
|||
|
|
HF Hub: https://huggingface.co/KKHYA/qwen3-1.7b-fft-if
|
|||
|
|
Timestamp: 2026-05-24 05:48:17
|
|||
|
|
=====================================
|
|||
|
|
[INFO|2026-05-24 05:48:23] llamafactory.launcher:144 >> Initializing 8 distributed tasks at: 127.0.0.1:36913
|
|||
|
|
W0524 05:48:24.105000 263391 site-packages/torch/distributed/run.py:851]
|
|||
|
|
W0524 05:48:24.105000 263391 site-packages/torch/distributed/run.py:851] *****************************************
|
|||
|
|
W0524 05:48:24.105000 263391 site-packages/torch/distributed/run.py:851] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
|
|||
|
|
W0524 05:48:24.105000 263391 site-packages/torch/distributed/run.py:851] *****************************************
|
|||
|
|
/home/ubuntu/miniconda3/envs/lmfact/lib/python3.11/site-packages/jieba/_compat.py:18: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
|
|||
|
|
import pkg_resources
|
|||
|
|
/home/ubuntu/miniconda3/envs/lmfact/lib/python3.11/site-packages/jieba/_compat.py:18: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
|
|||
|
|
import pkg_resources
|
|||
|
|
/home/ubuntu/miniconda3/envs/lmfact/lib/python3.11/site-packages/jieba/_compat.py:18: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
|
|||
|
|
import pkg_resources
|
|||
|
|
/home/ubuntu/miniconda3/envs/lmfact/lib/python3.11/site-packages/jieba/_compat.py:18: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
|
|||
|
|
import pkg_resources
|
|||
|
|
/home/ubuntu/miniconda3/envs/lmfact/lib/python3.11/site-packages/jieba/_compat.py:18: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
|
|||
|
|
import pkg_resources
|
|||
|
|
/home/ubuntu/miniconda3/envs/lmfact/lib/python3.11/site-packages/jieba/_compat.py:18: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
|
|||
|
|
import pkg_resources
|
|||
|
|
/home/ubuntu/miniconda3/envs/lmfact/lib/python3.11/site-packages/jieba/_compat.py:18: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
|
|||
|
|
import pkg_resources
|
|||
|
|
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
|||
|
|
/home/ubuntu/miniconda3/envs/lmfact/lib/python3.11/site-packages/jieba/_compat.py:18: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
|
|||
|
|
import pkg_resources
|
|||
|
|
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
|||
|
|
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
|||
|
|
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
|||
|
|
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
|||
|
|
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
|||
|
|
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
|||
|
|
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
|||
|
|
[W524 05:48:32.992213791 ProcessGroupNCCL.cpp:932] Warning: TORCH_NCCL_AVOID_RECORD_STREAMS is the default now, this environment variable is thus deprecated. (function operator())
|
|||
|
|
[W524 05:48:33.293666626 ProcessGroupNCCL.cpp:932] Warning: TORCH_NCCL_AVOID_RECORD_STREAMS is the default now, this environment variable is thus deprecated. (function operator())
|
|||
|
|
[W524 05:48:33.308084017 ProcessGroupNCCL.cpp:932] Warning: TORCH_NCCL_AVOID_RECORD_STREAMS is the default now, this environment variable is thus deprecated. (function operator())
|
|||
|
|
[W524 05:48:33.308938238 ProcessGroupNCCL.cpp:932] Warning: TORCH_NCCL_AVOID_RECORD_STREAMS is the default now, this environment variable is thus deprecated. (function operator())
|
|||
|
|
[W524 05:48:33.310524840 ProcessGroupNCCL.cpp:932] Warning: TORCH_NCCL_AVOID_RECORD_STREAMS is the default now, this environment variable is thus deprecated. (function operator())
|
|||
|
|
[W524 05:48:33.339552947 ProcessGroupNCCL.cpp:932] Warning: TORCH_NCCL_AVOID_RECORD_STREAMS is the default now, this environment variable is thus deprecated. (function operator())
|
|||
|
|
[W524 05:48:33.351637395 ProcessGroupNCCL.cpp:932] Warning: TORCH_NCCL_AVOID_RECORD_STREAMS is the default now, this environment variable is thus deprecated. (function operator())
|
|||
|
|
[W524 05:48:33.420535856 ProcessGroupNCCL.cpp:932] Warning: TORCH_NCCL_AVOID_RECORD_STREAMS is the default now, this environment variable is thus deprecated. (function operator())
|
|||
|
|
[INFO|2026-05-24 05:48:36] llamafactory.hparams.parser:505 >> Process rank: 0, world size: 8, device: cuda:0, distributed training: True, compute dtype: torch.bfloat16
|
|||
|
|
[INFO|2026-05-24 05:48:36] llamafactory.hparams.parser:505 >> Process rank: 5, world size: 8, device: cuda:5, distributed training: True, compute dtype: torch.bfloat16
|
|||
|
|
[INFO|2026-05-24 05:48:36] llamafactory.hparams.parser:505 >> Process rank: 2, world size: 8, device: cuda:2, distributed training: True, compute dtype: torch.bfloat16
|
|||
|
|
[INFO|2026-05-24 05:48:36] llamafactory.hparams.parser:505 >> Process rank: 3, world size: 8, device: cuda:3, distributed training: True, compute dtype: torch.bfloat16
|
|||
|
|
[INFO|2026-05-24 05:48:36] llamafactory.hparams.parser:505 >> Process rank: 4, world size: 8, device: cuda:4, distributed training: True, compute dtype: torch.bfloat16
|
|||
|
|
[INFO|2026-05-24 05:48:36] llamafactory.hparams.parser:505 >> Process rank: 7, world size: 8, device: cuda:7, distributed training: True, compute dtype: torch.bfloat16
|
|||
|
|
[INFO|2026-05-24 05:48:36] llamafactory.hparams.parser:505 >> Process rank: 6, world size: 8, device: cuda:6, distributed training: True, compute dtype: torch.bfloat16
|
|||
|
|
[INFO|2026-05-24 05:48:36] llamafactory.hparams.parser:505 >> Process rank: 1, world size: 8, device: cuda:1, distributed training: True, compute dtype: torch.bfloat16
|
|||
|
|
[INFO|configuration_utils.py:670] 2026-05-24 05:48:36,333 >> loading configuration file config.json from cache at /home/ubuntu/.cache/huggingface/hub/models--Qwen--Qwen3-1.7B/snapshots/70d244cc86ccca08cf5af4e1e306ecf908b1ad5e/config.json
|
|||
|
|
[INFO|configuration_utils.py:742] 2026-05-24 05:48:36,335 >> Model config Qwen3Config {
|
|||
|
|
"architectures": [
|
|||
|
|
"Qwen3ForCausalLM"
|
|||
|
|
],
|
|||
|
|
"attention_bias": false,
|
|||
|
|
"attention_dropout": 0.0,
|
|||
|
|
"bos_token_id": 151643,
|
|||
|
|
"dtype": "bfloat16",
|
|||
|
|
"eos_token_id": 151645,
|
|||
|
|
"head_dim": 128,
|
|||
|
|
"hidden_act": "silu",
|
|||
|
|
"hidden_size": 2048,
|
|||
|
|
"initializer_range": 0.02,
|
|||
|
|
"intermediate_size": 6144,
|
|||
|
|
"layer_types": [
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention"
|
|||
|
|
],
|
|||
|
|
"masked_layers": null,
|
|||
|
|
"max_position_embeddings": 40960,
|
|||
|
|
"max_window_layers": 28,
|
|||
|
|
"model_type": "qwen3",
|
|||
|
|
"num_attention_heads": 16,
|
|||
|
|
"num_hidden_layers": 28,
|
|||
|
|
"num_key_value_heads": 8,
|
|||
|
|
"pad_token_id": null,
|
|||
|
|
"rms_norm_eps": 1e-06,
|
|||
|
|
"rope_parameters": {
|
|||
|
|
"rope_theta": 1000000,
|
|||
|
|
"rope_type": "default"
|
|||
|
|
},
|
|||
|
|
"sliding_window": null,
|
|||
|
|
"sparsity_attn": null,
|
|||
|
|
"sparsity_mlp": null,
|
|||
|
|
"subnet_mode": null,
|
|||
|
|
"subnet_type": null,
|
|||
|
|
"threshold_attn": null,
|
|||
|
|
"threshold_mlp": null,
|
|||
|
|
"tie_word_embeddings": true,
|
|||
|
|
"transformers_version": "5.2.0",
|
|||
|
|
"use_cache": true,
|
|||
|
|
"use_sliding_window": false,
|
|||
|
|
"vocab_size": 151936
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
[INFO|configuration_utils.py:670] 2026-05-24 05:48:38,758 >> loading configuration file config.json from cache at /home/ubuntu/.cache/huggingface/hub/models--Qwen--Qwen3-1.7B/snapshots/70d244cc86ccca08cf5af4e1e306ecf908b1ad5e/config.json
|
|||
|
|
[INFO|configuration_utils.py:742] 2026-05-24 05:48:38,759 >> Model config Qwen3Config {
|
|||
|
|
"architectures": [
|
|||
|
|
"Qwen3ForCausalLM"
|
|||
|
|
],
|
|||
|
|
"attention_bias": false,
|
|||
|
|
"attention_dropout": 0.0,
|
|||
|
|
"bos_token_id": 151643,
|
|||
|
|
"dtype": "bfloat16",
|
|||
|
|
"eos_token_id": 151645,
|
|||
|
|
"head_dim": 128,
|
|||
|
|
"hidden_act": "silu",
|
|||
|
|
"hidden_size": 2048,
|
|||
|
|
"initializer_range": 0.02,
|
|||
|
|
"intermediate_size": 6144,
|
|||
|
|
"layer_types": [
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention"
|
|||
|
|
],
|
|||
|
|
"masked_layers": null,
|
|||
|
|
"max_position_embeddings": 40960,
|
|||
|
|
"max_window_layers": 28,
|
|||
|
|
"model_type": "qwen3",
|
|||
|
|
"num_attention_heads": 16,
|
|||
|
|
"num_hidden_layers": 28,
|
|||
|
|
"num_key_value_heads": 8,
|
|||
|
|
"pad_token_id": null,
|
|||
|
|
"rms_norm_eps": 1e-06,
|
|||
|
|
"rope_parameters": {
|
|||
|
|
"rope_theta": 1000000,
|
|||
|
|
"rope_type": "default"
|
|||
|
|
},
|
|||
|
|
"sliding_window": null,
|
|||
|
|
"sparsity_attn": null,
|
|||
|
|
"sparsity_mlp": null,
|
|||
|
|
"subnet_mode": null,
|
|||
|
|
"subnet_type": null,
|
|||
|
|
"threshold_attn": null,
|
|||
|
|
"threshold_mlp": null,
|
|||
|
|
"tie_word_embeddings": true,
|
|||
|
|
"transformers_version": "5.2.0",
|
|||
|
|
"use_cache": true,
|
|||
|
|
"use_sliding_window": false,
|
|||
|
|
"vocab_size": 151936
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
[INFO|configuration_utils.py:670] 2026-05-24 05:48:38,792 >> loading configuration file config.json from cache at /home/ubuntu/.cache/huggingface/hub/models--Qwen--Qwen3-1.7B/snapshots/70d244cc86ccca08cf5af4e1e306ecf908b1ad5e/config.json
|
|||
|
|
[INFO|configuration_utils.py:742] 2026-05-24 05:48:38,793 >> Model config Qwen3Config {
|
|||
|
|
"architectures": [
|
|||
|
|
"Qwen3ForCausalLM"
|
|||
|
|
],
|
|||
|
|
"attention_bias": false,
|
|||
|
|
"attention_dropout": 0.0,
|
|||
|
|
"bos_token_id": 151643,
|
|||
|
|
"dtype": "bfloat16",
|
|||
|
|
"eos_token_id": 151645,
|
|||
|
|
"head_dim": 128,
|
|||
|
|
"hidden_act": "silu",
|
|||
|
|
"hidden_size": 2048,
|
|||
|
|
"initializer_range": 0.02,
|
|||
|
|
"intermediate_size": 6144,
|
|||
|
|
"layer_types": [
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention"
|
|||
|
|
],
|
|||
|
|
"masked_layers": null,
|
|||
|
|
"max_position_embeddings": 40960,
|
|||
|
|
"max_window_layers": 28,
|
|||
|
|
"model_type": "qwen3",
|
|||
|
|
"num_attention_heads": 16,
|
|||
|
|
"num_hidden_layers": 28,
|
|||
|
|
"num_key_value_heads": 8,
|
|||
|
|
"pad_token_id": null,
|
|||
|
|
"rms_norm_eps": 1e-06,
|
|||
|
|
"rope_parameters": {
|
|||
|
|
"rope_theta": 1000000,
|
|||
|
|
"rope_type": "default"
|
|||
|
|
},
|
|||
|
|
"sliding_window": null,
|
|||
|
|
"sparsity_attn": null,
|
|||
|
|
"sparsity_mlp": null,
|
|||
|
|
"subnet_mode": null,
|
|||
|
|
"subnet_type": null,
|
|||
|
|
"threshold_attn": null,
|
|||
|
|
"threshold_mlp": null,
|
|||
|
|
"tie_word_embeddings": true,
|
|||
|
|
"transformers_version": "5.2.0",
|
|||
|
|
"use_cache": true,
|
|||
|
|
"use_sliding_window": false,
|
|||
|
|
"vocab_size": 151936
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
[INFO|2026-05-24 05:48:40] llamafactory.data.loader:144 >> Loading dataset allenai/tulu-3-sft-personas-instruction-following...
|
|||
|
|
Setting num_proc from 16 back to 1 for the train split to disable multiprocessing as it only contains one shard.
|
|||
|
|
Generating train split: 0%| | 0/29980 [00:00<?, ? examples/s]
Generating train split: 50%|█████ | 15000/29980 [00:00<00:00, 138651.32 examples/s]
Generating train split: 100%|██████████| 29980/29980 [00:00<00:00, 168440.08 examples/s]
|
|||
|
|
[INFO|2026-05-24 05:48:41] llamafactory.data.loader:144 >> Sampled 10000 examples from dataset allenai/tulu-3-sft-personas-instruction-following.
|
|||
|
|
Converting format of dataset (num_proc=16): 0%| | 0/10000 [00:00<?, ? examples/s]
Converting format of dataset (num_proc=16): 5%|▍ | 471/10000 [00:00<00:06, 1380.28 examples/s]
Converting format of dataset (num_proc=16): 100%|██████████| 10000/10000 [00:00<00:00, 19785.69 examples/s]
|
|||
|
|
[INFO|2026-05-24 05:48:42] llamafactory.data.loader:144 >> Loading dataset KKHYA/oasst1_converted...
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Setting num_proc from 16 back to 1 for the train split to disable multiprocessing as it only contains one shard.
|
|||
|
|
Generating train split: 0%| | 0/7132 [00:00<?, ? examples/s]
Generating train split: 100%|██████████| 7132/7132 [00:00<00:00, 188075.45 examples/s]
|
|||
|
|
[INFO|2026-05-24 05:48:43] llamafactory.data.loader:144 >> Sampled 7000 examples from dataset KKHYA/oasst1_converted.
|
|||
|
|
Converting format of dataset (num_proc=16): 0%| | 0/7000 [00:00<?, ? examples/s]
Converting format of dataset (num_proc=16): 6%|▋ | 438/7000 [00:00<00:04, 1357.58 examples/s]
Converting format of dataset (num_proc=16): 100%|██████████| 7000/7000 [00:00<00:00, 14380.96 examples/s]
|
|||
|
|
[INFO|2026-05-24 05:48:43] llamafactory.data.loader:144 >> Loading dataset KKHYA/oasst2_converted...
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Setting num_proc from 16 back to 1 for the train split to disable multiprocessing as it only contains one shard.
|
|||
|
|
Generating train split: 0%| | 0/9091 [00:00<?, ? examples/s]
Generating train split: 100%|██████████| 9091/9091 [00:00<00:00, 181843.59 examples/s]
|
|||
|
|
[INFO|2026-05-24 05:48:45] llamafactory.data.loader:144 >> Sampled 9000 examples from dataset KKHYA/oasst2_converted.
|
|||
|
|
Converting format of dataset (num_proc=16): 0%| | 0/9000 [00:00<?, ? examples/s]
Converting format of dataset (num_proc=16): 6%|▋ | 563/9000 [00:00<00:05, 1678.02 examples/s]
Converting format of dataset (num_proc=16): 100%|██████████| 9000/9000 [00:00<00:00, 17867.24 examples/s]
|
|||
|
|
[INFO|2026-05-24 05:48:46] llamafactory.data.loader:144 >> Loading dataset KKHYA/coconot_converted...
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Setting num_proc from 16 back to 1 for the train split to disable multiprocessing as it only contains one shard.
|
|||
|
|
Generating train split: 0%| | 0/10983 [00:00<?, ? examples/s]
Generating train split: 100%|██████████| 10983/10983 [00:00<00:00, 358487.80 examples/s]
|
|||
|
|
Setting num_proc from 16 back to 1 for the test split to disable multiprocessing as it only contains one shard.
|
|||
|
|
Generating test split: 0%| | 0/1001 [00:00<?, ? examples/s]
Generating test split: 100%|██████████| 1001/1001 [00:00<00:00, 348567.73 examples/s]
|
|||
|
|
[INFO|2026-05-24 05:48:47] llamafactory.data.loader:144 >> Sampled 10000 examples from dataset KKHYA/coconot_converted.
|
|||
|
|
Converting format of dataset (num_proc=16): 0%| | 0/10000 [00:00<?, ? examples/s]
Converting format of dataset (num_proc=16): 6%|▋ | 625/10000 [00:00<00:04, 1964.02 examples/s]
Converting format of dataset (num_proc=16): 100%|██████████| 10000/10000 [00:00<00:00, 20333.44 examples/s]
|
|||
|
|
[INFO|2026-05-24 05:48:47] llamafactory.data.loader:144 >> Loading dataset KKHYA/aya_dataset_converted...
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Setting num_proc from 16 back to 1 for the train split to disable multiprocessing as it only contains one shard.
|
|||
|
|
Generating train split: 0%| | 0/202285 [00:00<?, ? examples/s]
Generating train split: 7%|▋ | 15000/202285 [00:00<00:01, 143640.88 examples/s]
Generating train split: 19%|█▉ | 38000/202285 [00:00<00:00, 171031.19 examples/s]
Generating train split: 29%|██▊ | 58000/202285 [00:00<00:00, 179354.20 examples/s]
Generating train split: 39%|███▉ | 79000/202285 [00:00<00:00, 183647.89 examples/s]
Generating train split: 51%|█████ | 103000/202285 [00:00<00:00, 201993.92 examples/s]
Generating train split: 65%|██████▌ | 132000/202285 [00:00<00:00, 193939.96 examples/s]
Generating train split: 79%|███████▊ | 159000/202285 [00:00<00:00, 185568.40 examples/s]
Generating train split: 89%|████████▉ | 181000/202285 [00:00<00:00, 189380.38 examples/s]
Generating train split: 100%|██████████| 202285/202285 [00:01<00:00, 189553.65 examples/s]
|
|||
|
|
Setting num_proc from 16 back to 1 for the test split to disable multiprocessing as it only contains one shard.
|
|||
|
|
Generating test split: 0%| | 0/1750 [00:00<?, ? examples/s]
Generating test split: 100%|██████████| 1750/1750 [00:00<00:00, 170444.73 examples/s]
|
|||
|
|
[INFO|2026-05-24 05:48:53] llamafactory.data.loader:144 >> Sampled 10000 examples from dataset KKHYA/aya_dataset_converted.
|
|||
|
|
Converting format of dataset (num_proc=16): 0%| | 0/10000 [00:00<?, ? examples/s]
Converting format of dataset (num_proc=16): 6%|▌ | 598/10000 [00:00<00:05, 1756.90 examples/s]
Converting format of dataset (num_proc=16): 100%|██████████| 10000/10000 [00:00<00:00, 19372.45 examples/s]
|
|||
|
|
[INFO|2026-05-24 05:48:53] llamafactory.data.loader:144 >> Loading dataset KKHYA/daring_anteater_converted...
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Setting num_proc from 16 to 4 for the train split as it only contains 4 shards.
|
|||
|
|
Generating train split: 0%| | 0/93486 [00:00<?, ? examples/s]
Generating train split: 1%| | 1000/93486 [00:00<00:20, 4509.27 examples/s]
Generating train split: 9%|▊ | 8000/93486 [00:00<00:02, 29408.93 examples/s]
Generating train split: 24%|██▎ | 22000/93486 [00:00<00:01, 60418.24 examples/s]
Generating train split: 40%|███▉ | 37000/93486 [00:00<00:00, 80789.03 examples/s]
Generating train split: 53%|█████▎ | 50000/93486 [00:00<00:00, 89087.59 examples/s]
Generating train split: 65%|██████▌ | 61000/93486 [00:00<00:00, 94539.82 examples/s]
Generating train split: 80%|████████ | 75000/93486 [00:00<00:00, 102598.46 examples/s]
Generating train split: 95%|█████████▍| 88743/93486 [00:01<00:00, 111491.16 examples/s]
Generating train split: 100%|██████████| 93486/93486 [00:01<00:00, 81281.38 examples/s]
|
|||
|
|
[INFO|2026-05-24 05:49:03] llamafactory.data.loader:144 >> Sampled 10000 examples from dataset KKHYA/daring_anteater_converted.
|
|||
|
|
Converting format of dataset (num_proc=16): 0%| | 0/10000 [00:00<?, ? examples/s][WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nTacos from el Barrio\n\nA tray of tacos from el Barrio are pictured at their outdoor seating on Tuesday, May 4, 2021 in Athens, Georgia. (Photo/Lora Yordanova, lorayphoto@gmail.com)\n\nLora Yordanova\n\nThe adobo chicken, fried chicken and barbacoa tacos from el Barrio are only three of their various taco offerings. The tacos all come with a sauce best suited to them that can be drizzled on top. All of the tacos are around $2-3 and each varies slightly in price depending on the meat and toppings included.\n\nThe tacos were made with two corn tortillas to make sure that they didn\'t break, and the ingredients were fresh and delicious. The adobo chicken was particularly good, packing the most flavor in the chicken itself and the pickled onions on top. The barbacoa was probably the driest of the tacos and had the simplest toppings.\n\nEl Barrio Tacos & Tequila on S. Milledge Ave. is open from 11:30 a.m.-9 p.m. Tuesday through Saturday and until 8 p.m. on Sunday.\njson schema:\n{\n"title": "TacoDescription",\n"type": "object",\n"properties": {\n"restaurantName": {"title": "Restaurant Name", "type": "string"},\n"tacoName": {"title": "Taco Name", "type": "array", "items": {"type": "string"}},\n"priceRange": {"title": "Price Range", "type": "string"},\n"tortillaType": {"title": "Tortilla Type", "type": "string"},\n"ingredients": {"title": "Ingredients", "type": "array", "items": {"type": "string"}},\n"tacoSauce": {"title": "Taco Sauce", "type": "string"},\n"openingHours": {"title": "Opening Hours", "type": "string"},\n"location": {"title": "Location", "type": "string"}\n},\n"required": ["restaurantName", "tacoName", "priceRange", "tortillaType", "ingredients", "tacoSauce", "openingHours", "location"]\n}', 'role': 'user'}, {'content': '{\n"restaurantName": "El Barrio Tacos & Tequila",\n"tacoName": ["adobo chicken tacos", "fried chicken tacos", "barbacoa tacos"],\n"priceRange": "$2-3",\n"tortillaType": "corn tortillas",\n"ingredients": ["adobo chicken", "fried chicken", "barbacoa", "pickled onions"],\n"tacoSauce": "a sauce best suited to each taco",\n"openingHours": "11:30 a.m.-9 p.m. Tuesday through Saturday, until 8 p.m. on Sunday",\n"location": "S. Milledge Ave., Athens, Georgia"\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "OregonStateSenateDistrict",\n"type": "object",\n"properties": {\n"districtNumber": {\n"title": "District Number",\n"type": "string"\n},\n"counties": {\n"title": "Counties",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"cities": {\n"title": "Cities",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"representatives": {\n"title": "Representatives",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"party": {\n"title": "Party",\n"type": "string"\n},\n"tenure": {\n"title": "Tenure",\n"type": "string",\n"format": "date-time"\n}\n},\n"required": ["name", "party", "tenure"]\n}\n},\n"electionResults": {\n"title": "Election Results",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"year": {\n"title": "Year",\n"type": "string",\n"format": "date-time"\n},\n"representative": {\n"title": "Representative",\n"type": "string"\n},\n"party": {\n"title": "Party",\n"type": "string"\n}\n},\n"required": ["year", "representative", "party"]\n}\n},\n"districtHistory": {\n"title": "District History",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"timePeriod": {\n"title": "Time Period",\n"type": "string",\n"format": "date-time"\n},\n"coverage": {\n"title": "Coverage",\n"type": "string"\n}\n},\n"required": ["timePeriod", "coverage"]\n}\n}\n},\n"required": ["districtNumber", "counties", "representatives", "electionResults", "districtHistory"]\n}\nparagraph:\nDistrict 10 of the Oregon State Senate comprises parts of Marion and Polk counties, including much of Salem. The last Senator elected to represent this district was a Democrat Deb Patterson (politician), in 2018 Republican Jackie Winters of Salem was elected, she died of lung cancer on May 29, 2019. Denyc Boles, who had been serving as the Representative for District 19 in the Oregon House of Representatives, was appointed on June 25, 2019 as Winters\' replacement. Senator Patterson ousted Denyc Boles in November, 2020 and will fulfill the remainder of Jackie Winters\' term.\n\nElection results\nDistrict boundaries have changed over time. From 1993 until 2003, the district covered parts of east Portland. Since 2003, the district has covered portions of the Salem metropolitan area.\n\nReferences\n\n10\nMarion County, Oregon\nPolk County, Oregon', 'role': 'user'}, {'content': '{\n"districtNumber": "10",\n"counties": ["Marion County, Oregon", "Polk County, Oregon"],\n"cities": ["Salem"],\n"representatives": [\n{\n"name": "Deb Patterson",\n"party": "Democrat",\n"tenure": "2018-11-01"\n},\n{\n"name": "Denyc Boles",\n"party": "Republican",\n"tenure": "2019-06-25"\n},\n{\n"name": "Jackie Winters",\n"party": "Republican",\n"tenure": "2019-01-01 - 2019-05-29"\n}\n],\n"electionResults": [\n{\n"year": "2018",\n"representative": "Deb Patterson",\n"party": "Democrat"\n},\n{\n"year": "2019",\n"representative": "Denyc Boles",\n"party": "Republican"\n},\n{\n"year": "2020",\n"representative": "Deb Patterson",\n"party": "Democrat"\n}\n],\n"districtHistory": [\n{\n"timePeriod": "1993-2003",\n"coverage": "parts of east Portland"\n},\n{\n"timePeriod": "2003-present",\n"coverage": "portions of the Salem metropolitan area"\n}\n]\n}', 'role': 'assistant'}, {'content': 'paragraph:\nScotland and Wales remove Greece from quarantine safe list\n\nQuarantine restrictions will be imposed from tomorrow on people travelling from Greece to Scotland, the Scottish government has announced.\n\nTravellers will be expected to isolate for 14-days if they arrive in the country after 04:00 on Thursday.\n\nMinisters said
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "NewsletterSubscription",\n"type": "object",\n"properties": {\n"subscriberName": {\n"title": "Subscriber Name",\n"type": "string"\n},\n"subscriptionDate": {\n"title": "Subscription Date",\n"type": "string",\n"format": "date"\n},\n"subscribedPublications": {\n"title": "Subscribed Publications",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"privacyNotice": {\n"title": "Privacy Notice",\n"type": "string"\n},\n"unsubscribeInformation": {\n"title": "Unsubscribe Information",\n"type": "string"\n},\n"additionalRecommendations": {\n"title": "Additional Recommendations",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["subscriberName", "subscriptionDate", "subscribedPublications", "privacyNotice", "unsubscribeInformation"]\n}\nparagraph:\nWhen you subscribe we will use the information you provide to send you these newsletters. Sometimes they\'ll include recommendations for other related newsletters or services we offer. OurPrivacy Noticeexplains more about how we use your data, and your rights. You can unsubscribe at any time.\n\nRead More\n\nRelated Articles\n\nRead More\n\nRelated Articles\n\nMichael Lennox, chief executive of Community Pharmacy Somerset, said: "It\'s fantastic to see the launch of the next two pharmacy-led vaccination sites in Somerset and great to see community pharmacy taking an even greater role during the pandemic to support people in Somerset.\n\n"We know that people will really value having the additional choice in location and that pharmacies can make a significant contribution to our Somerset vaccination programmes for Covid."\n\nThe retailer has also recently started to offer vaccinations in other pharmacies including Bluewater, Gloucester and Huddersfield, with an option to expand further to more Boots stores in the future.', 'role': 'user'}, {'content': '{\n"subscriberName": "Michael Lennox",\n"subscriptionDate": "2022-02-16",\n"subscribedPublications": ["Community Pharmacy Somerset Newsletter"],\n"privacyNotice": "OurPrivacy Noticeexplains more about how we use your data, and your rights",\n"unsubscribeInformation": "You can unsubscribe at any time",\n"additionalRecommendations": ["Related Articles", "Related Articles", "Boots Vaccination Services"]\n}', 'role': 'assistant'}, {'content': 'paragraph:\nStarting this Monday morning with temperatures in the 30s across most of the Panhandle. Skies area partly cloudy and under quiet conditions. For this afternoon we are going to be seeing those temperatures increasing into the 60s and 70s, setting us above average. Cloud cover will also increase for today, but winds will stay on the light side. A ridge like pattern will give us plenty of warm temperatures through this week with a mix of 70s after today.\n\nRain has been reduced in the region but still tracking the potential for scattered showers on Friday and on Sunday. Models are not in good agreement so plenty of details to iron out. Stay tuned.\n\nHave a great week,\n\nMeteorologist Maria Pasillas\n\nCopyright 2022 Nexstar Media Inc. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.\njson schema:\n{"title": "WeatherForecast", "type": "object", "properties": {"date": {"title": "Date", "type": "string", "format": "date"}, "startingTemperature": {"title": "Starting Temperature", "type": "number", "minimum": 0}, "maxTemperature": {"title": "Max Temperature", "type": "number", "minimum": 0}, "skyCondition": {"title": "Sky Condition", "type": "string"}, "windCondition": {"title": "Wind Condition", "type": "string"}, "precipitationProb
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{"title": "FootballTeam", "type": "object", "properties": {"teamName": {"title": "Team Name", "type": "string"}, "tournament": {"title": "Tournament", "type": "string"}, "year": {"title": "Year", "type": "number"}, "result": {"title": "Result", "type": "string"}, "starPlayers": {"title": "Star Players", "type": "array", "items": {"type": "object", "properties": {"playerName": {"title": "Player Name", "type": "string"}, "position": {"title": "Position", "type": "string"}, "team": {"title": "Team", "type": "string"}}, "required": ["playerName", "position", "team"]}}, "prediction": {"title": "Prediction", "type": "object", "properties": {"predictorName": {"title": "Predictor Name", "type": "string"}, "predictedPlayer": {"title": "Predicted Player", "type": "string"}, "predictedPerformance": {"title": "Predicted Performance", "type": "string"}}, "required": ["predictorName", "predictedPlayer", "predictedPerformance"]}, "reports": {"title": "Reports", "type": "array", "items": {"type": "object", "properties": {"reportSource": {"title": "Report Source", "type": "string"}, "reportContent": {"title": "Report Content", "type": "string"}}, "required": ["reportSource", "reportContent"]}}, "playerStats": {"title": "Player Stats", "type": "object", "properties": {"playerName": {"title": "Player Name", "type": "string"}, "goals": {"title": "Goals", "type": "number"}, "assists": {"title": "Assists", "type": "number"}, "appearances": {"title": "Appearances", "type": "number"}}, "required": ["playerName", "goals", "assists", "appearances"]}}, "required": ["teamName", "tournament", "year", "result", "starPlayers", "prediction", "reports", "playerStats"]}\nparagraph:\nBrazil are among the tournament favourites to triumph this time round after they were suffered a devastating 7-1 defeat to Germany in the semi-finals in 2014.\n\nThe Selecao boast a host of star players like Philippe Coutinho, Roberto Firmino and superstar Neymar – however, it\'s the Chelsea wideman that Carlos is backing to have a great tournament and even said he\'s up there among the game\'s elite players.\n\n\'Neymar is going to be marvellous in Russia and win the trophy for us,\' Carlos said as per the Metro.\n\n\'But you know who else is going to have a big World Cup? Willian.\n\n\'People talk about [Lionel] Messi, [Cristiano] Ronaldo, Robert Lewandowski and Neymar, but for me, Willian is right up there.\n\n\'He\'s in really good form and always takes responsibility on the pitch.\'\n\nWith Anthony Martial\'s future unclear at Old Trafford there were reports by ESPN that Jose Mourinho has earmarked the Chelsea as a potential replacement for the Frenchman.\n\nMourinho is said to be a keen admirer of the Chelsea star and have previously working together at Stamford Bridge, with the Brazilian impressing last season after bagging 13 goals and 12 assists in 55 appearances in all competitions.', 'role': 'user'}, {'content': '{\n "teamName": "Brazil",\n "tournament": "World Cup",\n "year": 2018,\n "result": "Among the tournament favorites",\n "starPlayers": [\n {\n "playerName": "Philippe Coutinho",\n "position": "Not mentioned",\n "team": "Brazil"\n },\n {\n "playerName": "Roberto Firmino",\n "position": "Not mentioned",\n "team": "Brazil"\n },\n {\n "playerName": "Neymar",\n "position": "Forward",\n "team": "Brazil"\n },\n {\n "playerName": "Willian",\n "position": "Winger",\n "team": "Brazil"\n }\n ],\n "prediction": {\n "predictorName": "Carlos",\n "predictedPlayer": "Neymar",\n "predictedPerf
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "CompanyDescription",\n"type": "object",\n"properties": {\n"companyName": {\n"title": "Company Name",\n"type": "string"\n},\n"companyDescription": {\n"title": "Company Description",\n"type": "string"\n},\n"solutions": {\n"title": "Solutions",\n"type": "object",\n"properties": {\n"solutionName": {\n"title": "Solution Name",\n"type": "string"\n},\n"solutionDescription": {\n"title": "Solution Description",\n"type": "string"\n}\n},\n"required": ["solutionName", "solutionDescription"]\n},\n"platform": {\n"title": "Platform",\n"type": "object",\n"properties": {\n"platformName": {\n"title": "Platform Name",\n"type": "string"\n},\n"platformDescription": {\n"title": "Platform Description",\n"type": "string"\n}\n},\n"required": ["platformName", "platformDescription"]\n},\n"cloudProvider": {\n"title": "Cloud Provider",\n"type": "string"\n}\n},\n"required": ["companyName", "companyDescription", "solutions", "platform", "cloudProvider"]\n}\nparagraph:\nWe unite information across the enterprise, converging front-end sales with back-end supply chain execution. Our software, platform technology and unmatched experience help drive both top-line growth and bottom-line profitability for our customers.\nManhattan Associates designs, builds and delivers leading edge cloud and on-premises solutions so that across the store, through your network or from your fulfillment center, you are ready to reap the rewards of the omnichannel marketplace.\nAbout Manhattan Active® Order Management Systems\nThe Manhattan Active Omni suite is the first of its kind, engineered for an omnichannel world, and built to deliver on the customer experience promise. Born in the cloud, Manhattan Active Omni is a comprehensive portfolio of solutions and services that is always current and fully extensible.\nManhattan Active® runs on Google Cloud.\nManhattan Active® is proud to run on Google Cloud.', 'role': 'user'}, {'content': '{\n"companyName": "Manhattan Associates",\n"companyDescription": "Unites information across the enterprise, converging front-end sales with back-end supply chain execution. The software and platform technology help drive both top-line growth and bottom-line profitability for customers.",\n"solutions": {\n"solutionName": "Manhattan Active® Omni suite",\n"solutionDescription": "The first of its kind, engineered for an omnichannel world, and built to deliver on the customer experience promise. Born in the cloud, Manhattan Active Omni is a comprehensive portfolio of solutions and services that is always current and fully extensible."\n},\n"platform": {\n"platformName": "Manhattan Active®",\n"platformDescription": "Leading edge cloud and on-premises solutions that help you be ready to reap the rewards of the omnichannel marketplace across the store, through your network or from your fulfillment center."\n},\n"cloudProvider": "Google Cloud"\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "Person",\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"occupation": {\n"title": "Occupation",\n"type": "string"\n},\n"spouseName": {\n"title": "Spouse Name",\n"type": "string"\n},\n"location": {\n"title": "Location",\n"type": "string"\n},\n"yearsActive": {\n"title": "Years Active",\n"type": "string"\n},\n"artFocus": {\n"title": "Art Focus",\n"type": "string"\n},\n"artistsCollected": {\n"title": "Artists Collected",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"auctionValue": {\n"title": "Auction Value",\n"type": "number"\n},\n"references": {\n"title": "References",\n"type": "array",\n"items": {\n"t
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "CommitteeMeeting",\n"type": "object",\n"properties": {\n"committeeName": {"title": "Committee Name", "type": "string"},\n"committeeHead": {"title": "Committee Head", "type": "string"},\n"findings": {"title": "Findings", "type": "string"},\n"cases": {"title": "Cases", "type": "string"},\n"investigationType": {"title": "Investigation Type", "type": "string"},\n"witnesses": {"title": "Witnesses", "type": "array", "items": {"type": "string"}},\n"meetingGoal": {"title": "Meeting Goal", "type": "string"}\n},\n"required": ["committeeName", "committeeHead", "findings", "cases", "investigationType", "witnesses", "meetingGoal"]\n}\nparagraph:\nAssembly committee finds Facebook \'complicit\' in Delhi violence\n\nThe committee has also urged the need for an independent investigation leading to the filing of a supplementary chargesheet in the ongoing cases related to Delhi violence.\n\nThis image has been used for representational purpose only\n\nThe Peace and Harmony Committee of Delhi Assembly headed by Raghav Chadha has, prima facie, found Facebook "complicit" in northeast Delhi violence and decided to summon officials of the social media giant in order to examine them on "oath and determine on their culpability".\n\n"The committee through its Chairman Raghav Chadha has prima facie found Facebook complicit in Delhi riots of February 2020 on the premise of incriminatory material produced on record by the witnesses as well as their scathing depositions before the Committee and thus deciding to summon Facebook officials in its next meeting in order to examine them on oath and determine on their culpability," said an official release by the Committee.\n\nThe committee has also urged the need for an independent investigation leading to the filing of a supplementary chargesheet in the ongoing cases related to Delhi violence.\n\nThe committee had earlier called upon other expert witnesses which have deposed before the committee and have contributed some invaluable insights. The committee aims to expeditiously resolve the current issue in hand, it added in the statement.', 'role': 'user'}, {'content': '{\n"committeeName": "Peace and Harmony Committee of Delhi Assembly",\n"committeeHead": "Raghav Chadha",\n"findings": "Facebook \'complicit\' in northeast Delhi violence",\n"cases": "Delhi violence of February 2020",\n"investigationType": "independent investigation leading to the filing of a supplementary chargesheet",\n"witnesses": ["other expert witnesses"],\n"meetingGoal": "expeditiously resolve the current issue in hand"\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "TheaterWorkshop",\n"type": "object",\n"properties": {\n"workshopName": {\n"title": "Workshop Name",\n"type": "string"\n},\n"supportingOrganization": {\n"title": "Supporting Organization",\n"type": "string"\n},\n"sponsor": {\n"title": "Sponsor",\n"type": "string"\n},\n"location": {\n"title": "Location",\n"type": "string"\n},\n"performanceDate": {\n"title": "Performance Date",\n"type": "string",\n"format": "date"\n},\n"performanceTime": {\n"title": "Performance Time",\n"type": "string",\n"format": "time"\n},\n"venue": {\n"title": "Venue",\n"type": "string"\n},\n"ticketPrice": {\n"title": "Ticket Price",\n"type": "number"\n},\n"fundingSource": {\n"title": "Funding Source",\n"type": "string"\n},\n"administeredBy": {\n"title": "Administered By",\n"type": "string"\n}\n},\n"required": [\n"workshopName",\n"supportingOrganization",\n"sponsor",\n"location",\n"performanceDate",\n"performanceTime",\n"venue",\n"ticketPrice",\n"fundingSource",\n"administeredBy"\n]\n}\nparagra
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nVarious campuses of Infosys and Bharat Biotech are also guarded by the CISF\n\nThe Central Industrial Security Force (CISF) is providing security to 64 airports and 11 private establishments, including those of Reliance Industries, Tata Steel and Infosys, Union Minister of State for Home Nityanand Rai said on Wednesday.\n\nThe minister said the deployment of the CISF is considered on the basis of threat perception and the commitment of the management concerned to bear the cost of deployment.\n\n"At present, the CISF is deployed in 11 private sector companies, including in electronics and information technology sector," he said replying to a written question in Rajya Sabha.\n\nRai said the CISF was established under the CISF Act, 1968 for better protection and security of industrial undertakings owned and controlled by central and state governments, joint venture or private industrial undertaking.\n\nAccordingly, the CISF is providing security at 64 airports at present, including those operated under joint ventures and public private partnership (PPP), he said.\njson schema:\n{\n"title": "CISFSecurityDetails",\n"type": "object",\n"properties": {\n"organizationsSecured": {\n"title": "Organizations Secured",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"sectorSecured": {\n"title": "Sector Secured",\n"type": "string"\n},\n"deploymentBasis": {\n"title": "Deployment Basis",\n"type": "object",\n"properties": {\n"criteria": {\n"title": "Criteria",\n"type": "string"\n},\n"managementCommitment": {\n"title": "Management Commitment",\n"type": "string"\n}\n},\n"required": ["criteria", "managementCommitment"]\n},\n"numberOfAirportsSecured": {\n"title": "Number of Airports Secured",\n"type": "number"\n},\n"act": {\n"title": "Act",\n"type": "string"\n}\n},\n"required": ["organizationsSecured", "sectorSecured", "deploymentBasis", "numberOfAirportsSecured", "act"]\n}', 'role': 'user'}, {'content': '{\n"organizationsSecured": ["Reliance Industries", "Tata Steel", "Infosys", "Bharat Biotech"],\n"sectorSecured": "electronics and information technology sector, other private sector companies",\n"deploymentBasis": {\n"criteria": "threat perception",\n"managementCommitment": "commitment of the management concerned to bear the cost of deployment"\n},\n"numberOfAirportsSecured": 64,\n"act": "CISF Act, 1968"\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "Obituary",\n"type": "object",\n"properties": {\n"name": {"title": "Name", "type": "string"},\n"dateOfBirth": {"title": "Date of Birth", "type": "string", "format": "date"},\n"dateOfDeath": {"title": "Date of Death", "type": "string", "format": "date"},\n"hometown": {"title": "Hometown", "type": "string"},\n"education": {"title": "Education", "type": "string"},\n"spouse": {"title": "Spouse", "type": "string"},\n"weddingDate": {"title": "Wedding Date", "type": "string", "format": "date"},\n"career": {"title": "Career", "type": "string"},\n"precededBy": {"title": "Preceded By", "type": "array", "items": {"type": "string"}},\n"survivedBy": {"title": "Survived By", "type": "string"},\n"memorialService": {"title": "Memorial Service", "type": "string"},\n"donationInstructions": {"title": "Donation Instructions", "type": "string", "format": "uri"}\n},\n"required": ["name", "dateOfBirth", "dateOfDeath", "hometown", "education", "spouse", "weddingDate", "career", "precededBy", "survivedBy", "memorialService", "donationInstructions"]\n}\nparagraph:\nWilliam Poirier\n\nOur dear Billy went to his heavenly home on November 19th, 2021; our hearts are broken for the sad loss to our family. William Francis Poirier was born July 14th, 1948 in Yakima to Anne and Vincent Poirier. He graduated from Davis in 1968. Billy was a hard worker, always providing for his wife, Cindy. They were married July 9th, 1977. Billy never knew a stranger, he had so many friends and acquaintances. He subcontracted mail delivery for many years from Selah to Easton, seven days a week. He later worked for Hansen Fruit, and his last job was working for Yakima Specialties, Inc. He was always proud of the jobs he held and had a wonderful work ethic. Bill was preceded in death by his mother Anne, father Vincent, brother Vince Jr., sister Sue, father-in-law Bob Krein, and mother-in-law Frances Krein. He is survived by his loving wife Cynthia and her siblings. There will be a private gathering in loving memory of Bill. We humbly request that any gift donations be given to the National Association of the Deaf (nad.org) in Billy\'s name.', 'role': 'user'}, {'content': '{\n"name": "William Francis Poirier",\n"dateOfBirth": "1948-07-14",\n"dateOfDeath": "2021-11-19",\n"hometown": "Yakima",\n"education": "Davis, 1968",\n"spouse": "Cindy",\n"weddingDate": "1977-07-09",\n"career": "Subcontracted mail delivery, Hansen Fruit, Yakima Specialties, Inc.",\n"precededBy": ["Anne", "Vincent Poirier", "Vince Jr.", "Sue", "Bob Krein", "Frances Krein"],\n"survivedBy": "Cynthia and her siblings",\n"memorialService": "There will be a private gathering",\n"donationInstructions": "nad.org"\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "SIMDeal",\n"type": "object",\n"properties": {\n"providerName": {\n"title": "Provider Name",\n"type": "string"\n},\n"dealName": {\n"title": "Deal Name",\n"type": "string"\n},\n"dataAllowance": {\n"title": "Data Allowance",\n"type": "number"\n},\n"price": {\n"title": "Price",\n"type": "number"\n},\n"contractType": {\n"title": "Contract Type",\n"type": "string"\n},\n"callAllowance": {\n"title": "Call Allowance",\n"type": "string"\n},\n"textAllowance": {\n"title": "Text Allowance",\n"type": "string"\n},\n"roamingAllowance": {\n"title": "Roaming Allowance",\n"type": "string"\n},\n"addOnData": {\n"title": "Add-On Data",\n"type": "number",\n"format": "data-allowance"\n}\n},\n"required": ["providerName", "dealName", "dataAllowance", "price", "contractType", "callAllowance", "textAllowance", "roamingAllowance"]\n}\nparagraph:\nSmarty have released a new SI
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nWant a Mood Boost? Listen To A Sad Song!\n\nScience says that sobbing to sad songs will ultimately help you feel happier overall, so what do you have to lose at this point?\n\nA study, published by Scientific Reports, supports this theory. Researchers explained that by getting out a good cry while listening to sad tunes, you will ultimately trigger feelings of pleasure and feel less stressed afterward.\n\nResearchers tested this hypothesis by asking study participants to fill out short surveys about their reactions while listening to various music.\n\nThey were asked to report how frequently they got goose bumps, felt shivers down their spine, felt like weeping or got a lump in their throat.\n\nResearchers then broke up the participants into two groups based on those who experienced chills and those who experienced tears and asked them to listen to six songs (three of which the subjects picks themselves) that were intended to evoke deep emotions.\n\nAfter monitoring their emotional state, heart rates and physical signs of pleasure, researchers concluded that the participants\' tears resulting from sadness can actually be "psychophysiologically calming," causing sensations of pleasure and even acting as a stress reliever.\njson schema:\n{\n"title": "StudyResults",\n"type": "object",\n"properties": {\n"studyTitle": {\n"title": "Study Title",\n"type": "string"\n},\n"theory": {\n"title": "Theory",\n"type": "string"\n},\n"researchersExplanation": {\n"title": "Researchers\' Explanation",\n"type": "string"\n},\n"testMethod": {\n"title": "Test Method",\n"type": "object",\n"properties": {\n"survey": {\n"title": "Survey",\n"type": "string"\n},\n"groups": {\n"title": "Groups",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["survey", "groups"]\n},\n"songs": {\n"title": "Songs",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"songName": {\n"title": "Song Name",\n"type": "string"\n},\n"emotion": {\n"title": "Intended Emotion",\n"type": "string"\n}\n},\n"required": ["songName", "emotion"]\n}\n},\n"results": {\n"title": "Results",\n"type": "object",\n"properties": {\n"emotionalState": {\n"title": "Emotional State",\n"type": "string"\n},\n"heartRate": {\n"title": "Heart Rate",\n"type": "string"\n},\n"physicalSignsOfPleasure": {\n"title": "Physical Signs of Pleasure",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["emotionalState", "heartRate"]\n}\n}\n}', 'role': 'user'}, {'content': '{\n"title": "StudyResults",\n"type": "object",\n"properties": {\n"studyTitle": {\n"title": "Study Title",\n"type": "string",\n"value": "A study published by Scientific Reports"\n},\n"theory": {\n"title": "Theory",\n"type": "string",\n"value": "Sobbing to sad songs will ultimately help you feel happier overall"\n},\n"researchersExplanation": {\n"title": "Researchers\' Explanation",\n"type": "string",\n"value": "By getting out a good cry while listening to sad tunes, you will ultimately trigger feelings of pleasure and feel less stressed afterward"\n},\n"testMethod": {\n"title": "Test Method",\n"type": "object",\n"properties": {\n"survey": {\n"title": "Survey",\n"type": "string",\n"value": "Participants were asked to fill out short surveys about their reactions while listening to various music"\n},\n"groups": {\n"title": "Groups",\n"type": "array",\n"items": {\n"type": "string"\n},\n"value": ["Participants who experienced chills", "Participants who experienced tears"]\n}\n}\n},\n"songs": {\n"title": "Songs",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"songName": {\n"title": "Song Name",\n"
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "StudySignUp",\n"type": "object",\n"properties": {\n"studyName": {"title": "Study Name", "type": "string"},\n"date": {"title": "Date", "type": "string", "format": "date"},\n"startTime": {"title": "Start Time", "type": "string", "format": "time"},\n"targetGroup": {"title": "Target Group", "type": "object", "properties": {\n"minAge": {"title": "Minimum Age", "type": "number"},\n"maxAge": {"title": "Maximum Age", "type": "number"},\n"group": {"title": "Group", "type": "string"}},\n"required": ["minAge", "maxAge", "group"]},\n"incentive": {"title": "Incentive", "type": "object", "properties": {\n"amount": {"title": "Amount", "type": "number"},\n"type": {"title": "Type", "type": "string"}},\n"required": ["amount", "type"]},\n"contact": {"title": "Contact", "type": "object", "properties": {\n"name": {"title": "Name", "type": "string"},\n"phone": {"title": "Phone", "type": "string"},\n"url": {"title": "URL", "type": "string"}},\n"required": ["name", "phone", "url"]}\n},\n"required": ["studyName", "date", "startTime", "targetGroup", "incentive", "contact"]\n}\nparagraph:\nUCSB Parent Teen Study\n\nApril 18, 2015 from 10:00am\n\nLocal teens (14-18 years old) are invited to participate with their mothers in a Parent-Teen Communication Study that is taking place at UCSB. Participating in this study takes about one hour and involves having conversations with your mother or child about different topics, including relationships, dating, sexuality, sexual orientation, marriage, and families.\n\nEach teen and mother who participates will receive a $15 cash or gift card incentive in appreciation of their time (total $30 combined).\n\nAppointments are available daily and on weekends. Call Audrey at (805) 380-6897 to sign up or learn more. Or, visit: http://ucsbstudy.wix.com/ucsbstudy', 'role': 'user'}, {'content': '{\n"studyName": "Parent-Teen Communication Study",\n"date": "2015-04-18",\n"startTime": "10:00:00",\n"targetGroup": {\n"minAge": 14,\n"maxAge": 18,\n"group": "local teens and their mothers"\n},\n"incentive": {\n"amount": 15,\n"type": "cash or gift card"\n},\n"contact": {\n"name": "Audrey",\n"phone": "(805) 380-6897",\n"url": "http://ucsbstudy.wix.com/ucsbstudy"\n}\n}', 'role': 'assistant'}, {'content': 'paragraph:\nThe original proposal of the measure by the Treasury had boosted the merger appeal of Banco BPM, Italy\'s third-largest bank, fuelling investor bets about a potential takeover by rival UniCredit CRDI.MI.\n\nSpeaking at a news conference after the cabinet approved the spending package, Franco said the cap on the tax incentives would remain at 2% of the assets of the smaller bank involved in a merger.\n\nAt the same news conference, Prime Minister Mario Draghi said he hoped that reductions in the rate of COVID-19 infection would mean that no further economic stimulus would be needed after the 40 billion euro ($50 billion) package approved by the cabinet on Thursday.\n\nTrending Topics\n\nTrending Articles\n\nIn This Story\n\nReuters, the news and media division of Thomson Reuters, is the world\'s largest international multimedia news provider reaching more than one billion people every day. Reuters provides trusted business, financial, national, and international news to professionals via Thomson Reuters desktops, the world\'s media organizations, and directly to consumers at Reuters.com and via Reuters TV.\njson schema:\n{\n"title": "NewsConference",\n"type": "object",\n"properties": {\n"measureProposal": {"title": "Measure Proposal", "type": "string"},\n"bankName1": {"title": "First Bank Name", "type": "string"},\n"bankName2
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nCNN\'s Jim Acosta gave Donald Trump the equivalent of a six-minute eye-roll Saturday as he discussed the one-term president\'s impending speech at the Conservative Political Action Conference, saying, "comparing Trump to a clown is most definitely an insult to clowns." He added, "Please take your fake White House seal, and go play president somewhere else."\n\n"Former president Donald Trump is returning to the CPAC stage tomorrow," Acosta said during an afternoon "Newsroom" broadcast Saturday. "Hide the flags," he added, cuing footage to roll of Trump\'s 2019 and 2020 CPAC speeches that showed him embracing Old Glory.\n\nAfter calling the clips "so disturbing," the CNN host ripped into the Republican: "Now the same man who gripped the Stars and Stripes won\'t let go of something else: The idea that he\'s still president."\n\nHe pointed to Trump\'s recent press conference announcing his lawsuit against Google, Twitter and Facebook, which included flags and a seal on the podium, as "some kind of POTUS Potemkin village with Trump as the pretender-in-chief."\n\n"Yes, he seems to be spending his post-presidency playing president and, as always, he is playing the victim, too, this time claiming disinformation is being directed at him," Acosta added. "What\'s next? Another \'Stop the Steal\' rally on the National Mall? You may find some of the Proud Boys and Oath Keepers are busy — in jail," he spat, referring to the Jan. 6 Capitol riot carried out by Trump supporters who sought to stop the certification of President Joe Biden\'s 2020 election win.\n\nAcosta wrapped up the segment, saying that Trump\'s Sunday CPAC speech will just be another part of his "act."\njson schema:\n{\n"title": "NewsSegment",\n"type": "object",\n"properties": {\n"speakerName": {\n"title": "Speaker Name",\n"type": "string"\n},\n"event": {\n"title": "Event",\n"type": "string"\n},\n"organization": {\n"title": "Organization",\n"type": "string"\n},\n"audience": {\n"title": "Audience",\n"type": "string"\n},\n"topic": {\n"title": "Topic",\n"type": "string"\n},\n"subtopics": {\n"title": "Subtopics",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"opinions": {\n"title": "Opinions",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"reactions": {\n"title": "Reactions",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["speakerName", "event", "organization", "audience", "topic"]\n}', 'role': 'user'}, {'content': '{\n"speakerName": "Jim Acosta",\n"event": "CPAC (Conservative Political Action Conference)",\n"organization": "CNN",\n"audience": "Television and online viewers",\n"topic": "Donald Trump\'s impending speech at CPAC",\n"subtopics": ["Trump\'s impending speech", "Trump\'s press conference", "Capitol riot", "Stop the Steal rally"],\n"opinions": ["comparing Trump to a clown is an insult to clowns", "Trump is still gripping the idea that he\'s still president", "Trump is playing president and playing the victim", "disinformation is being directed at Trump"],\n"reactions": ["Please take your fake White House seal, and go play president somewhere else", "Hide the flags", "so disturbing", "another \'Stop the Steal\' rally on the National Mall", "You may find some of the Proud Boys and Oath Keepers are busy — in jail"]\n}', 'role': 'assistant'}, {'content': 'paragraph:\nMigori leaders urged to encourage education among youth\n\nMigori county women Representative Pamela Odhiambo has challenged leaders to encourage education among the youth in order to promote development among the Kuria community.\n\nCiting FGM as a major hindrance, she urged th
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nThe Lord & Taylor flagship store has debuted on Walmart.com, featuring fashion items from more than 125 of the most well-known brands and we are HERE for it! Now, Walmart will sell items from Vince Camuto, Karl Lagerfeld, Miss Selfridge, La La Anthony, Lucky Brand and more.\n\nThe companies teamed up to start selling women\'s, men\'s and kids\' items, featuring everything from clothing to shoes to accessories.\n\nDenise Incandela, the head of fashion of Walmart U.S. eCommerce, says this is giving the company the change to introduce "new premium brands to our customers."\n\nFor Lord & Taylor, this partnership is helping them to advance their digital presence, RJ Cilley, SVP of Digital at Lord & Taylor says.\n\nAnd the best part? Orders over $35 will be eligible for free two-day shipping.\n\nSo for all the best summer finds, shop till you drop all -- from the comfort of your desk!\n\nThese products were curated by our "Good Morning America" editorial team. "GMA" has affiliate partnerships, so we will get a small share of revenue from your purchases through these links. All product prices are determined by the retailer and subject to change. By visiting these websites, you will leave GoodMorningAmerica.com and any information you share with the retailer will be governed by its website\'s terms and conditions and privacy policies.\njson schema:\n{\n"title": "RetailCollaboration",\n"type": "object",\n"properties": {\n"stores": {\n"title": "Stores Involved",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"storeName": {\n"title": "Store Name",\n"type": "string"\n},\n"flagshipStore": {\n"title": "Flagship Store",\n"type": "boolean"\n}\n},\n"required": ["storeName", "flagshipStore"]\n}\n},\n"productCategories": {\n"title": "Product Categories",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"products": {\n"title": "Products",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"productName": {\n"title": "Product Name",\n"type": "string"\n},\n"brand": {\n"title": "Brand",\n"type": "string"\n}\n},\n"required": ["productName", "brand"]\n}\n},\n"benefits": {\n"title": "Benefits",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"benefitType": {\n"title": "Benefit Type",\n"type": "string"\n},\n"benefitDetails": {\n"title": "Benefit Details",\n"type": "string"\n}\n},\n"required": ["benefitType", "benefitDetails"]\n}\n},\n"partnershipDetails": {\n"title": "Partnership Details",\n"type": "object",\n"properties": {\n"responsibleParty": {\n"title": "Responsible Party",\n"type": "string"\n},\n"role": {\n"title": "Role",\n"type": "string"\n}\n},\n"required": ["responsibleParty", "role"]\n}\n}\n}', 'role': 'user'}, {'content': '{\n"stores": [\n{\n"storeName": "Lord & Taylor",\n"flagshipStore": true\n},\n{\n"storeName": "Walmart",\n"flagshipStore": false\n}\n],\n"productCategories": [\n"women\'s items",\n"men\'s items",\n"kids\' items",\n"clothing",\n"shoes",\n"accessories"\n],\n"products": [\n{\n"productName": "Vince Camuto products",\n"brand": "Vince Camuto"\n},\n{\n"productName": "Karl Lagerfeld products",\n"brand": "Karl Lagerfeld"\n},\n{\n"productName": "Miss Selfridge products",\n"brand": "Miss Selfridge"\n},\n{\n"productName": "La La Anthony products",\n"brand": "La La Anthony"\n},\n{\n"productName": "Lucky Brand products",\n"brand": "Lucky Brand"\n}\n],\n"benefits": [\n{\n"benefitType": "free shipping",\n"benefitDetails": "Orders over $35 will be eligible for free two-day shipping"\n}\n],\n"partnershipDetails": {\n"responsibleParty": "Lord & Taylor",\n"role": "advancing their digital presence"\n}
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "ManagerChange",\n"type": "object",\n"properties": {\n"managerName": {\n"title": "Manager Name",\n"type": "string"\n},\n"teamName": {\n"title": "Team Name",\n"type": "string"\n},\n"predecessorName": {\n"title": "Predecessor Name",\n"type": "string"\n},\n"reasonForChange": {\n"title": "Reason for Change",\n"type": "string"\n},\n"interimManagerName": {\n"title": "Interim Manager Name",\n"type": "string"\n},\n"permanentManagerName": {\n"title": "Permanent Manager Name",\n"type": "string"\n},\n"problemAreas": {\n"title": "Problem Areas",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"boardExpectations": {\n"title": "Board Expectations",\n"type": "string"\n}\n},\n"required": [\n"managerName",\n"teamName",\n"predecessorName",\n"reasonForChange"\n]\n}\nparagraph:\nShareAll sharing options for:Report: Ole Gunnar Solskjaer to be sacked by Manchester United\n\nPer multiple reports, the Manchester United board of directors have decided to sack Ole Gunnar Solskjaer. An official announcement is expected in the coming hours, and the search will begin once again for a new manager on a permanent basis.\n\nManchester United have decided to part company with manager Ole Gunnar Solskjaer, The Athletic understands... #MUFC\n\nIt is rumored that Darren Fletcher or Michael Carrick could take over as interim manager, with the other acting as assistant, while the search for a permanent manager continues. Zinedine Zidane was rumored over the last couple weeks to be an option, but the club may be waiting until the end of the season to decide on a permanent replacement. Any manager that comes in now would likely only be a short term option.\n\nSolskjaer took over from Jose Mourinho after a disastrous start to the 2018/19 season, and impressed enough to earn the permanent job. In his two full seasons in charge he led the team to 3rd and 2nd place respectively in the Premier League, reached the semi-final stage of cup competitions on 5 occasions, and reached the 2021 UEFA Europa League final. Unfortunately he never led the team to silverware, and the collapse of the team to start the 2021/22 campaign is indicative of the holes that remain in the squad despite meaningful work and recruitment in his rebuilding project.\n\nWhoever takes over as manager will indeed find a stronger team on paper than the one Solskjaer took over in 2018, but that doesn\'t mean there isn\'t a lot of work to do. Midfield will remain a problem, as will establishing order in defending, and the attacking problems since adding Cristiano Ronaldo to the mix will need to be sorted. The board clearly believe that the current team should be good enough to challenge for silverware, but a commitment to building a team the new manager wants must be made or they run the risk of doing all of this over again in a couple years time.', 'role': 'user'}, {'content': '{\n"managerName": "To be determined",\n"teamName": "Manchester United",\n"predecessorName": "Ole Gunnar Solskjaer",\n"reasonForChange": "Manchester United\'s board of directors have decided to part company with Solskjaer due to the collapse of the team to start the 2021/22 campaign and his failure to lead the team to silverware in his tenure.",\n"interimManagerName": "Possibly Darren Fletcher or Michael Carrick",\n"permanentManagerName": "Zinedine Zidane (rumored, decision may be made at the end of the season)",\n"problemAreas": [\n"Midfield",\n"Establishing order in defending",\n"Attacking problems since adding Cristiano Ronaldo to the mix"\n],\n"boardExpectations": "The board expects the current team to be good enough to chal
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "CompanyEarnings",\n"type": "object",\n"properties": {\n"companyName": {\n"title": "Company Name",\n"type": "string"\n},\n"stockPrice": {\n"title": "Stock Price",\n"type": "number"\n},\n"date": {\n"title": "Date",\n"type": "string",\n"format": "date"\n},\n"monthlyChange": {\n"title": "Monthly Change",\n"type": "number"\n},\n"marketCapitalization": {\n"title": "Market Capitalization",\n"type": "number"\n},\n"rank": {\n"title": "Rank",\n"type": "number"\n},\n"revenue": {\n"title": "Revenue",\n"type": "number"\n},\n"netProfit": {\n"title": "Net Profit",\n"type": "number"\n},\n"operatingProfitMargin": {\n"title": "Operating Profit Margin",\n"type": "number"\n},\n"salesGrowth": {\n"title": "Sales Growth",\n"type": "number"\n},\n"orderInflow": {\n"title": "Order Inflow",\n"type": "number"\n},\n"orderSegment": {\n"title": "Order Segment",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": [\n"companyName",\n"stockPrice",\n"date",\n"monthlyChange",\n"marketCapitalization",\n"rank",\n"revenue",\n"netProfit",\n"operatingProfitMargin",\n"salesGrowth",\n"orderInflow"\n]\n}\nparagraph:\nLarsen & Toubro (L&T) hit a new high of Rs 1,441, up 2% on BSE in intra-day trade, ahead of its October-December 2017 quarter (Q3FY18) earnings on Wednesday, January 31, 2018. Thus far in January, the stock of engineering & construction major outperformed the market by surging 15% as compared to 7% rise in the S&P BSE Sensex. A strong rally in L&T stock has seen its market capitalisation (market-cap) crossing Rs 2 trillion today. At 11:10 AM; L&T with market-cap stood at Rs 2.01 trillion (Rs 200,614 crore), the BSE data shows. L&T stands at 13th rank in overall market-cap ranking ahead of state-owned companies Indian Oil Corporation and Coal India. Analysts expect strong earnings growth for L&T in Q3FY18. The company announced order intake of Rs 350 billion in Q3FY18 compared to Rs 61 billion in Q3FY17. Order inflow during the quarter has been supported by large as well as base order finalization. Domestic order wins were supported by order finalization in the hydrocarbon and infrastructure segments.\n\nOverseas order finalization was supported by the infrastructure and power segments. Motilal Oswal Securities expect L&T\'s net profit to grow 47% year on year (YoY) to Rs 14.3 billion on revenue growth of 10.1% YoY at Rs 290 billion. Operating profit margin will improve 80bp YoY to 10.4%. "We expect sales growth of 10% YoY at Rs 289 billion, driven by Infrastructure/Services businesses. The company has guided for a 12% sales growth, 25bps margin improvement and has cut order inflow guidance from 12‐14% to flat to minor improvement in FY18. The new order announcement has been very strong this quarter and stands at around Rs400 billion indicating strong inflow trends for the quarter," the brokerage firm Prabhudas Lilladher said in quarterly preview.', 'role': 'user'}, {'content': '{\n"companyName": "Larsen & Toubro",\n"stockPrice": 1441,\n"date": "2018-01-31",\n"monthlyChange": 15,\n"marketCapitalization": 200614000000,\n"rank": 13,\n"revenue": 290000000000,\n"netProfit": 1430000000,\n"operatingProfitMargin": 10.4,\n"salesGrowth": 10.1,\n"orderInflow": 350000000000,\n"orderSegment": [\n"hydrocarbon",\n"infrastructure",\n"overseas infrastructure and power segments"\n]\n}', 'role': 'assistant'}, {'content': 'paragraph:\nValeriy Neverov (; ; born 21 June 1964 in Kharkiv) is a Ukrainian chess grandmaster (1991) and four-time Ukrainian Chess Champion (1983, 1985, 1988 and 1996).\n\nChess career\nIn 1991 Neverov won the Capablanca Memorial in Hav
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nJohn Stamper (born August 30, 1978) is a former American football defensive end who played for the Chicago Bears in 2002. He was drafted in the sixth round of the 2002 NFL Draft. He played college football at University of South Carolina.\n\nReferences \n\n1978 births\nLiving people\nAmerican football defensive ends\nSouth Carolina Gamecocks football players\nChicago Bears players\njson schema:\n{\n"title": "FootballPlayer",\n"type": "object",\n"properties": {\n"playerName": {\n"title": "Player Name",\n"type": "string"\n},\n"birthDate": {\n"title": "Birth Date",\n"type": "string",\n"format": "date"\n},\n"draftYear": {\n"title": "Draft Year",\n"type": "number"\n},\n"collegeTeam": {\n"title": "College Team",\n"type": "string"\n},\n"position": {\n"title": "Position",\n"type": "string"\n},\n"references": {\n"title": "References",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["playerName", "birthDate", "draftYear", "collegeTeam", "position"]\n}', 'role': 'user'}, {'content': '{\n"playerName": "John Stamper",\n"birthDate": "1978-08-30",\n"draftYear": 2002,\n"collegeTeam": "University of South Carolina",\n"position": "defensive end",\n"references": [\n"1978 births",\n"Living people",\n"American football defensive ends",\n"South Carolina Gamecocks football players",\n"Chicago Bears players"\n]\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "BoxerProfile",\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"birthDate": {\n"title": "Birth Date",\n"type": "string",\n"format": "date"\n},\n"deathDate": {\n"title": "Death Date",\n"type": "string",\n"format": "date"\n},\n"sport": {\n"title": "Sport",\n"type": "string"\n},\n"eventCompeted": {\n"title": "Event Competed",\n"type": "string"\n},\n"references": {\n"title": "References",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["name", "birthDate", "sport", "eventCompeted", "references"]\n}\nparagraph:\nZbigniew Piórkowski (2 July 1929 – 10 July 1994) was a Polish boxer. He competed in the men\'s middleweight event at the 1956 Summer Olympics.\n\nReferences\n\n1929 births\n1994 deaths\nPolish male boxers\nOlympic boxers of Poland\nBoxers at the 1956 Summer Olympics\nSportspeople from Łódź\nMiddleweight boxers', 'role': 'user'}, {'content': '{\n"name": "Zbigniew Piórkowski",\n"birthDate": "1929-07-02",\n"deathDate": "1994-07-10",\n"sport": "boxing",\n"eventCompeted": "men\'s middleweight",\n"references": [\n"1929 births",\n"1994 deaths",\n"Polish male boxers",\n"Olympic boxers of Poland",\n"Boxers at the 1956 Summer Olympics",\n"Sportspeople from Łódź",\n"Middleweight boxers"\n]\n}', 'role': 'assistant'}, {'content': 'paragraph:\nTomasz Arceusz (born 11 September 1959 in Poland) is a Polish retired footballer.\n\nReferences\n\nPolish footballers\nLiving people\n1959 births\nAssociation football forwards\nPeople from Płońsk County\nMotor Lublin players\nLegia Warsaw players\nVaasan Palloseura players\nSportspeople from Masovian Voivodeship\njson schema:\n{\n"title": "FootballerProfile",\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"birthDate": {\n"title": "Birth Date",\n"type": "string",\n"format": "date"\n},\n"birthPlace": {\n"title": "Birth Place",\n"type": "string"\n},\n"nationality": {\n"title": "Nationality",\n"type": "string"\n},\n"positionsPlayed": {\n"title": "Positions Played",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"teamsPlayedFor": {\n"title": "Teams Played For",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"r
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "GovernmentDepartment",\n"type": "object",\n"properties": {\n"departmentName": {\n"title": "Department Name",\n"type": "string"\n},\n"responsibilities": {\n"title": "Responsibilities",\n"type": "string"\n},\n"mission": {\n"title": "Mission",\n"type": "string"\n},\n"headOfDepartment": {\n"title": "Head of Department",\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"position": {\n"title": "Position",\n"type": "string"\n}\n},\n"required": ["name", "position"]\n},\n"ministry": {\n"title": "Ministry",\n"type": "string"\n},\n"policingPowers": {\n"title": "Policing Powers",\n"type": "object",\n"properties": {\n"powers": {\n"title": "Powers",\n"type": "string"\n},\n"arrestSuspects": {\n"title": "Arrest Suspects",\n"type": "boolean"\n}\n},\n"required": ["powers", "arrestSuspects"]\n},\n"managedForests": {\n"title": "Managed Forests",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"year": {\n"title": "Year",\n"type": "number"\n}\n},\n"required": ["name", "year"]\n}\n}\n},\n"required": ["departmentName", "responsibilities", "mission", "headOfDepartment", "ministry", "policingPowers", "managedForests"]\n}\nparagraph:\nThe Department of Forest Conservation (Sinhala: වන සංරක්\u200dෂණ දෙපාර්තමේන්තුව Vana Sanrakshana Departhamenthuwa) is a non-ministerial government department responsible for forestry in Sri Lanka. Its mission is to protect and expand Sri Lanka\'s forests and woodlands. The head of the department is the Conservator General, Dr. K.M.A.Bandara. It comes under the purview of the Ministry of Wildlife and Forest Resource Conservation.\n\nIt has limited policing powers in protected forest areas to stop illegal poaching and logging, with the power to arrest suspects.\n\nForests \nSome of the forests managed by the Department of Forest Conservation are:\n\n Hurulu Man and Biosphere Reserve, 1977\n Sinharaja Man and Biosphere Reserve, 1978 \n Kanneliya-Dediyagala-Nakiyadeniya (KDN), 2004\n\nSee also\nLaw enforcement in Sri Lanka\n\nReferences\n\nExternal links\n Department of Forest Conservation\n\nForest Conservation\nForest Conservation\nNature conservation in Sri Lanka\nSri Lanka\nForestry in Sri Lanka\nEnvironmental organisations based in Sri Lanka', 'role': 'user'}, {'content': '{\n"departmentName": "Department of Forest Conservation",\n"responsibilities": "forestry in Sri Lanka",\n"mission": "protect and expand Sri Lanka\'s forests and woodlands",\n"headOfDepartment": {\n"name": "Dr. K.M.A.Bandara",\n"position": "Conservator General"\n},\n"ministry": "Ministry of Wildlife and Forest Resource Conservation",\n"policingPowers": {\n"powers": "limited policing powers in protected forest areas to stop illegal poaching and logging",\n"arrestSuspects": true\n},\n"managedForests": [\n{\n"name": "Hurulu Man and Biosphere Reserve",\n"year": 1977\n},\n{\n"name": "Sinharaja Man and Biosphere Reserve",\n"year": 1978\n},\n{\n"name": "Kanneliya-Dediyagala-Nakiyadeniya (KDN)",\n"year": 2004\n}\n]\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "ZooClosure",\n"type": "object",\n"properties": {\n"location": {"title": "Location", "type": "string"},\n"closureStartDate": {"title": "Closure Start Date", "type": "string", "format": "date"},\n"reason": {\n"title": "Reason",\n"type": "object",\n"properties": {\n"statement": {"title": "Statement", "type": "string"},\n"authority": {"title": "Authority", "type": "string"},\n"department": {"title": "Department", "type": "string"}\n}\n},\
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nNet loss attributable to shareholders for the quarter was $1.14 billion, compared to a profit of $4.06 billion last year.\n\nNet loss per Class A equivalent share attributable to shareholders $692 versus a profit of $2,469 last year.\n\nWarren Buffett has warned about the \'nightmare\' tied to new accounting-rule changes.\nThe rules, which require Berkshire to report unrealized gains or losses in equity investments in net income, helped fuel a $1.14 billion loss in the first quarter, the company said.\n\nBut, operating earnings for the quarter rose to $5.29 billion from $3.56 billion in the previous year.\n\nTotal revenues were $58.47 billion, down from $64.37 billion last year due to a drop in insurance premiums earned. Insurance premiums earned declined to $13.373 billion from the prior year\'s $21.753 billion.\njson schema:\n{\n"title": "QuarterlyFinancialReport",\n"type": "object",\n"properties": {\n"netLoss": {\n"title": "Net Loss",\n"type": "number"\n},\n"netLossLastYear": {\n"title": "Net Loss Last Year",\n"type": "number"\n},\n"netLossChange": {\n"title": "Net Loss Change",\n"type": "string",\n"format": "decimal"\n},\n"earningsPerShare": {\n"title": "Earnings Per Share",\n"type": "number"\n},\n"earningsPerShareLastYear": {\n"title": "Earnings Per Share Last Year",\n"type": "number"\n},\n"earningsPerShareChange": {\n"title": "Earnings Per Share Change",\n"type": "string",\n"format": "decimal"\n},\n"accountingRuleChangesWarning": {\n"title": "Accounting Rule Changes Warning",\n"type": "string"\n},\n"operatingEarnings": {\n"title": "Operating Earnings",\n"type": "number"\n},\n"operatingEarningsChange": {\n"title": "Operating Earnings Change",\n"type": "string",\n"format": "decimal"\n},\n"totalRevenues": {\n"title": "Total Revenues",\n"type": "number"\n},\n"totalRevenuesChange": {\n"title": "Total Revenues Change",\n"type": "string",\n"format": "decimal"\n},\n"insurancePremiumsEarned": {\n"title": "Insurance Premiums Earned",\n"type": "number"\n},\n"insurancePremiumsEarnedChange": {\n"title": "Insurance Premiums Earned Change",\n"type": "string",\n"format": "decimal"\n}\n},\n"required": [\n"netLoss",\n"netLossLastYear",\n"netLossChange",\n"earningsPerShare",\n"earningsPerShareLastYear",\n"earningsPerShareChange",\n"operatingEarnings",\n"operatingEarningsChange",\n"totalRevenues",\n"totalRevenuesChange",\n"insurancePremiumsEarned",\n"insurancePremiumsEarnedChange"\n]\n}', 'role': 'user'}, {'content': '{\n"netLoss": -1140000000,\n"netLossLastYear": 406000000,\n"netLossChange": "-1.14",\n"earningsPerShare": -692,\n"earningsPerShareLastYear": 2469,\n"earningsPerShareChange": "-692",\n"accountingRuleChangesWarning": "Warren Buffett has warned about the \'nightmare\' tied to new accounting-rule changes. The rules, which require Berkshire to report unrealized gains or losses in equity investments in net income, helped fuel a $1.14 billion loss in the first quarter, the company said.",\n"operatingEarnings": 5290000000,\n"operatingEarningsChange": "1.48",\n"totalRevenues": 58470000000,\n"totalRevenuesChange": "-0.09",\n"insurancePremiumsEarned": 13373000000,\n"insurancePremiumsEarnedChange": "-0.39"\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nAlabama Attorney General recognizes St. James as safe school\n\nA Montgomery private school was recognized for its safety initiatives by the Alabama attorney general during a presentation Tuesday.\n\nSt. James School was presented the 2018 Alabama Safe School Initiative Award of Excellence for private schools in the central region by Attorney General Steve Marshall. The award is a part of Safe Schools Initiative program that started in 2002, according to a news release.\n\nA panel of judges met last month to review the submissions of 94 schools throughout the state, with a winner from each of the eight state school districts and a private school from each state region was selected. The number of submissions was twice the number received last year, the news release said.\n\n"In these times, we are all intensely aware of the serious responsibility our schools carry for the safekeeping of our children," Marshall said in the news release. "The Alabama Safe Schools Awards of Excellence is an important tradition of the Attorney General\'s Office, to recognize schools throughout our state for dedication and excellence in keeping students safe. Each day that parents leave their children at school, they are entrusting what they hold most precious into the care of schoolteachers, principals, school resource officers and other staff."\n\nThe awards, he said, are to "honor those who have achieved particularly high standards and set examples that others may follow. We are grateful for their service and commitment to protecting the children of Alabama."\njson schema:\n{\n"title": "SafeSchoolAward",\n"type": "object",\n"properties": {\n"awardName": {"title": "Award Name", "type": "string"},\n"recipient": {"title": "Recipient", "type": "string"},\n"presenter": {"title": "Presenter", "type": "string"},\n"presentationDate": {"title": "Presentation Date", "type": "string", "format": "date"},\n"awardProgramName": {"title": "Award Program Name", "type": "string"},\n"programStartYear": {"title": "Program Start Year", "type": "number"},\n"numberOfSubmissions": {"title": "Number of Submissions", "type": "number"},\n"numberOfWinners": {"title": "Number of Winners", "type": "number"},\n"winningCategory": {"title": "Winning Category", "type": "string"},\n"winningRegion": {"title": "Winning Region", "type": "string"},\n"judgingDate": {"title": "Judging Date", "type": "string", "format": "date"},\n"judgingPanel": {"title": "Judging Panel", "type": "array", "items": {"type": "string"}},\n"quote": {"title": "Quote", "type": "string"}\n},\n"required": ["awardName", "recipient", "presenter", "presentationDate", "awardProgramName", "programStartYear", "numberOfSubmissions", "numberOfWinners", "winningCategory", "winningRegion", "judgingDate", "judgingPanel", "quote"]\n}', 'role': 'user'}, {'content': '{\n"awardName": "Alabama Safe School Initiative Award of Excellence",\n"recipient": "St. James School",\n"presenter": "Attorney General Steve Marshall",\n"presentationDate": "2018-01-01",\n"awardProgramName": "Safe Schools Initiative",\n"programStartYear": 2002,\n"numberOfSubmissions": 94,\n"numberOfWinners": 16,\n"winningCategory": "private schools in the central region",\n"winningRegion": "central region",\n"judgingDate": "2017-12-01",\n"judgingPanel": ["not explicitly mentioned"],\n"quote": "In these times, we are all intensely aware of the serious responsibility our schools carry for the safekeeping of our children\\\\,...\\\\,Each day that parents leave their children at school, they are entrusting what they hold most precious into the care of schoolteachers, principals, sc
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nObituary: Rita DellaVecchia, 81\n\nRita DellaVecchia, 81, of Southington, passed away on Saturday, February 8, 2020 at The Hospital of Central CT, New Britain.\n\nRita was born in Southington on August 24, 1938, the daughter of the late Andrew and Helen (Yawin) DellaVecchia. Prior to her retirement she was a CNA at Southington Care Center. Rita enjoyed doing word puzzles, going out to eat and going to the casino.\n\nShe is survived by a sister Lois Smith and a brother Robert and his wife Sharon DellaVecchia all of Southington. She also leaves many nieces and nephews, cousins and friends. She was predeceased by three brothers Sebastian, Andrew and Larry.\n\nCalling hours will be on Wednesday, Feb. 12th from 10-11am with a Service at 11 am at the DellaVecchia Funeral Home, 211 N. Main St. Southington. Burial will be at Oak Hill Cemetery.\njson schema:\n{\n"title": "Obituary",\n"type": "object",\n"properties": {\n"name": {"title": "Name", "type": "string"},\n"age": {"title": "Age", "type": "number"},\n"hometown": {"title": "Hometown", "type": "string"},\n"dateOfDeath": {"title": "Date of Death", "type": "string", "format": "date"},\n"placeOfDeath": {"title": "Place of Death", "type": "string"},\n"occupation": {"title": "Occupation", "type": "string"},\n"hobbies": {"title": "Hobbies", "type": "array", "items": {"type": "string"}},\n"survivedBy": {"title": "Survived By", "type": "array", "items": {"type": "string"}},\n"predeceasedBy": {"title": "Predeceased By", "type": "array", "items": {"type": "string"}},\n"familyMembers": {\n"title": "Family Members",\n"type": "object",\n"properties": {\n"sister": {"type": "string"},\n"brother": {\n"type": "object",\n"properties": {\n"name": {"type": "string"},\n"spouse": {"type": "string"}\n},\n"required": ["name", "spouse"]\n}\n},\n"required": ["sister", "brother"]\n},\n"callingHours": {"title": "Calling Hours", "type": "object", "properties": {\n"startTime": {"title": "Start Time", "type": "string", "format": "time"},\n"endTime": {"title": "End Time", "type": "string", "format": "time"},\n"location": {"title": "Location", "type": "string"}\n},\n"required": ["startTime", "endTime", "location"]\n},\n"service": {"title": "Service", "type": "object", "properties": {\n"time": {"title": "Time", "type": "string", "format": "time"},\n"location": {"title": "Location", "type": "string"}\n},\n"required": ["time", "location"]\n},\n"burial": {"title": "Burial", "type": "object", "properties": {\n"location": {"title": "Location", "type": "string"}\n},\n"required": ["location"]\n}\n},\n"required": ["name", "age", "hometown", "dateOfDeath", "placeOfDeath", "occupation", "hobbies", "survivedBy", "predeceasedBy", "familyMembers", "callingHours", "service", "burial"]\n}', 'role': 'user'}, {'content': '{\n"name": "Rita DellaVecchia",\n"age": 81,\n"hometown": "Southington",\n"dateOfDeath": "2020-02-08",\n"placeOfDeath": "The Hospital of Central CT, New Britain",\n"occupation": "CNA",\n"hobbies": ["doing word puzzles", "going out to eat", "going to the casino"],\n"survivedBy": ["Lois Smith", "Robert and his wife Sharon DellaVecchia", "many nieces and nephews, cousins and friends"],\n"predeceasedBy": ["Sebastian", "Andrew", "Larry"],\n"familyMembers": {\n"sister": "Lois Smith",\n"brother": {\n"name": "Robert",\n"spouse": "Sharon DellaVecchia"\n}\n},\n"callingHours": {\n"startTime": "10-11am",\n"endTime": "11 am",\n"location": "the DellaVecchia Funeral Home, 211 N. Main St. Southington"\n},\n"service": {\n"time": "11 am",\n"location": "the DellaVecchia Funeral Home, 211 N. Main St. Southington"\n},\n"burial": {\n"location": "Oak H
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{"title": "Obituary", "type": "object", "properties": {"name": {"title": "Name", "type": "string", "description": "The full name of the deceased."}, "dates": {"title": "Dates", "type": "object", "properties": {"deathDate": {"title": "Death Date", "type": "string", "format": "date"}, "birthDate": {"title": "Birth Date", "type": "string", "format": "date"}}, "required": ["deathDate", "birthDate"]}, "survivedBy": {"title": "Survived By", "type": "array", "items": {"type": "object", "properties": {"name": {"title": "Name", "type": "string"}, "relation": {"title": "Relation", "type": "string"}}, "required": ["name", "relation"]}}, "visitation": {"title": "Visitation", "type": "object", "properties": {"location": {"title": "Location", "type": "string"}, "date": {"title": "Date", "type": "string", "format": "date"}, "time": {"title": "Time", "type": "string", "format": "time"}}, "required": ["location", "date", "time"]}, "service": {"title": "Service", "type": "object", "properties": {"location": {"title": "Location", "type": "string"}, "date": {"title": "Date", "type": "string", "format": "date"}, "time": {"title": "Time", "type": "string", "format": "time"}}, "required": ["location", "date", "time"]}, "notifications": {"title": "Notifications", "type": "object", "properties": {"to": {"title": "To", "type": "array", "items": {"type": "string", "format": "email"}}, "from": {"title": "From", "type": "string", "format": "email"}, "message": {"title": "Message", "type": "string", "description": "Optional message to include in the notification."}}, "required": ["to", "from"]}}, "required": ["name", "dates", "survivedBy", "visitation", "service", "notifications"]}\nparagraph:\nEugeniaGibson "Genie" Dougherty\n\nEugenia "Genie" Gibson Dougherty passed away on Sunday, November 24, 2019 with her loving family by her side. She was born on June 18, 1941 to the late Mr. & Mrs. Eugene Gibson. She was married to the late Michael Dougherty and is survived by her beloved daughters Denise Henderson (Kenny Henderson) and Danielle Resseguet (Louis Resseguet), her loving grandchildren Amanda Hunter and Ryan Resseguet, and great granddaughter Demi Mathis, her sister Carol Ann Gibson, nephew Robert Marsh (Angela Marsh), and great nieces Shelby Farrell and Bailey Marsh. Genie graduated from LSU and was a LSU Golden Girl while she attended LSU. She worked and retired as a schoolteacher from the Jefferson Parish School Board. Her favorite pastimes were spending time with her family and watching football. Relatives and friends of the family are invited to a Visitation at Leitz-Eagan Funeral Home, 4747 Veterans Memorial Blvd., Metairie, LA on Monday, December 30th, 2019 from 5:00 pm - 7:00 pm, with a Memorial Service beginning at 7:00 pm. Online messages of condolences and memories may be left at www.leitzeaganfuneralhome.com.\n\nNotify Friends and Family\n\nAdd the email addresses of friends or family members you\'d like to notify about this obituary.\n\nTo: *Invalid To Email\n\nFrom: *Invalid From Email\n\nMessage:\n\nOptional', 'role': 'user'}, {'content': '{\n "name": "Eugenia \\"Genie\\" Gibson Dougherty",\n "dates": {\n "deathDate": "2019-11-24",\n "birthDate": "1941-06-18"\n },\n "survivedBy": [\n {\n "name": "Denise Henderson",\n "relation": "daughter"\n },\n {\n "name": "Danielle Resseguet",\n "relation": "daughter"\n },\n {\n "name": "Amanda Hunter",\n "relation": "grandchild"\n },\n {\n "name": "Ryan Resseguet",\n "relation": "grandchild"\n },\n {\n "name": "Demi Mathis",\n "r
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{"title": "EarningsRelease", "type": "object", "properties": {"companyName": {"title": "Company Name", "type": "string"}, "stockExchange": {"title": "Stock Exchange", "type": "string"}, "stockTicker": {"title": "Stock Ticker", "type": "string"}, "earningsReleaseDate": {"title": "Earnings Release Date", "type": "string", "format": "date"}, "earningsReleaseTime": {"title": "Earnings Release Time", "type": "string", "format": "time"}, "conferenceCallTime": {"title": "Conference Call Time", "type": "string", "format": "time"}, "conferenceCallDate": {"title": "Conference Call Date", "type": "string", "format": "date"}, "callDialInDetails": {"title": "Call Dial-In Details", "type": "object", "properties": {"dialInUrl": {"title": "Dial-In URL", "type": "string", "format": "uri"}, "uniquePin": {"title": "Unique Pin", "type": "string"}}, "required": ["dialInUrl", "uniquePin"]}, "callReplayAvailable": {"title": "Call Replay Available", "type": "boolean"}, "callReplayDuration": {"title": "Call Replay Duration", "type": "integer", "minimum": 1}}, "required": ["companyName", "stockExchange", "stockTicker", "earningsReleaseDate", "conferenceCallTime", "conferenceCallDate", "callDialInDetails", "callReplayAvailable", "callReplayDuration"]}\nparagraph:\nTARRYTOWN, N.Y., Oct. 04, 2022 (GLOBE NEWSWIRE) -- Prestige Consumer Healthcare Inc. (NYSE:PBH) today announced that it will issue its fiscal 2023 second quarter earnings release on Thursday, November 3, 2022 before the opening of the market. The Company will host a conference call to discuss the results that same morning at 8:30 a.m. ET.\n\nTo participate in the live Internet webcast of the conference call, it can be accessed from the Investor Relations page of www.prestigeconsumerhealthcare.com. To participate in the conference call via phone, participants may register for the call here to receive dial-in details and a unique pin. While not required, it is recommended to join 10 minutes prior to the event start.\n\nA conference call replay will be available for approximately one week following completion of the live call and can be accessed on the Company\'s Investor Relations page.', 'role': 'user'}, {'content': '{\n "companyName": "Prestige Consumer Healthcare Inc.",\n "stockExchange": "NYSE",\n "stockTicker": "PBH",\n "earningsReleaseDate": "2022-11-03",\n "earningsReleaseTime": "00:00:00",\n "conferenceCallTime": "08:30:00",\n "conferenceCallDate": "2022-11-03",\n "callDialInDetails": {\n "dialInUrl": "https://www.prestigeconsumerhealthcare.com/investor-relations",\n "uniquePin": "Not provided in the text, participants may register for the call here to receive dial-in details and a unique pin."\n },\n "callReplayAvailable": true,\n "callReplayDuration": 7\n}', 'role': 'assistant'}, {'content': 'paragraph:\nSpacious freshly painted 3 bedroom 2.5 bathroom townhouse condo in a gated resort style community. Upgraded bathrooms and upgraded kitchen with stainless steel appliances, common area tiled. Custom paint. Formal dinning and living area and a family room. 2 assigned parking spaces. The community offers - community pool, children\'s play ground, gym, clubhouse, property management on site and security patrol. Located close to major highways, restaurants, retail stores, city park and much more. Get on I-75 in less than 5 minutes.\n\nNitty Gritty\n\nIncome Requirement: Must have 3x the rent in total household income (before taxes)\n\nBy registering, you give us your express written consent to deliver automated text messages to you at the phone number provided. Consent is not a condition
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nApplications are invited for various project based positions in Dibrugarh University Assam.\n\nDibrugarh University Assam is inviting applications from eligible candidates for the post of Senior Project Research Fellow (SPRF) under the extramural project "Community Behavior And Its Determinants In Relation To Malaria Prevention And Control In Assam And Tripura" implemented by Department of Sociology.\n\nName of post : Senior Project Research Fellow (SPRF)\n\nNo. of posts : 1\n\nName of project : Community Behavior And Its Determinants In Relation To Malaria Prevention And Control In Assam And Tripura\n\nSelection Procedure : A walk-in-interview will be held on 17th August, 2022 from 10 a.m. onwards at Department of Sociology, Dibrugarh University\n\nHow to apply : Candidates may e-mail their application in the prescribed format along with the scanned copies of their mark sheets and testimonials (in Pdf format only) by 16th August, 2022, to the address pranjalsarma@dibru.ac.in\n\nAbout Us\n\nNortheast Now is a multi-app based hyper-regional bilingual news portal. Led by a group of professionals, the digital news platform covers every inch of the eight states of northeast India and the five neighbouring countries. It is the first of its kind new media initiative in the northeast, and is based in Guwahati. As the political dynamics in the northeast is intricate and fluid, Northeast Now is always politically neutral.\njson schema:\n{\n"title": "JobOpening",\n"type": "object",\n"properties": {\n"universityName": {"title": "University Name", "type": "string"},\n"positionName": {"title": "Position Name", "type": "string"},\n"noOfPosts": {"title": "Number of Posts", "type": "number"},\n"projectName": {"title": "Project Name", "type": "string"},\n"selectionProcedure": {"title": "Selection Procedure", "type": "string"},\n"interviewDate": {"title": "Interview Date", "type": "string", "format": "date"},\n"applicationDeadline": {"title": "Application Deadline", "type": "string", "format": "date"},\n"applicationEmail": {"title": "Application Email", "type": "string", "format": "email"},\n"qualifications": {"title": "Qualifications", "type": "array", "items": {"type": "string"}}\n},\n"required": ["universityName", "positionName", "noOfPosts", "projectName", "selectionProcedure", "interviewDate", "applicationDeadline", "applicationEmail"]\n}', 'role': 'user'}, {'content': '{\n"universityName": "Dibrugarh University Assam",\n"positionName": "Senior Project Research Fellow (SPRF)",\n"noOfPosts": 1,\n"projectName": "Community Behavior And Its Determinants In Relation To Malaria Prevention And Control In Assam And Tripura",\n"selectionProcedure": "A walk-in-interview will be held on 17th August, 2022 from 10 a.m. onwards",\n"interviewDate": "2022-08-17",\n"applicationDeadline": "2022-08-16",\n"applicationEmail": "pranjalsarma@dibru.ac.in",\n"qualifications": ["Eligible candidates", "Prescribed format", "Scanned copies of mark sheets and testimonials (in Pdf format only)"]\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "NewsArticle",\n"type": "object",\n"properties": {\n"headline": {"title": "Headline", "type": "string"},\n"source": {"title": "Source", "type": "string"},\n"date": {"title": "Date", "type": "string", "format": "date"},\n"person": {\n"title": "Person",\n"type": "object",\n"properties": {\n"name": {"title": "Name", "type": "string"},\n"position": {"title": "Position", "type": "string"}\n},\n"required": ["name", "position"]\n},\n"location": {\n"title": "Location",\n"type": "object",\n"properties": {\n"country": {"title": "Country",
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nHobbes and the Ideal Citizen\n\nPublished September 9, 2022 at 6:00 AM PDT\n\nWhat do citizens owe the state? What does the state owe its citizens?\n\nSeventeenth century philosopher Thomas Hobbes believed that without government to control our worst impulses, life would be "solitary, poor, nasty, brutish, and short." Consequently, he thought that absolute monarchy is the best form of government. So is Hobbes\' ideal citizen simply someone who is willing to submit to absolute authority, or are there other features the ideal citizen must have? What flaws would make a subject bad, or worse, a threat to peace in the realm? And are there any lessons modern democracies can learn from Hobbes\' political philosophy? The Philosophers submit to Stanford political scientist Alison McQueen, author of Political Realism in Apocalyptic Times. Sunday, September 11 at 11 am.\n\nDevon Strolovitch studied medieval Judeo-Portuguese manuscripts and earned a PhD in Linguistics from Cornell University before coming to KALW. He is the Senior Producer of Philosophy Talk, and since 2007 has hosted Fog City Blues, the weekly digest of Blues in the Bay Area and beyond.\njson schema:\n{\n"title": "RadioProgram",\n"type": "object",\n"properties": {\n"title": {"title": "Title", "type": "string"},\n"publicationDate": {"title": "Publication Date", "type": "string", "format": "date-time"},\n"questions": {\n"title": "Questions",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"questionText": {"title": "Question Text", "type": "string"},\n"author": {"title": "Author", "type": "string"},\n"affiliation": {"title": "Affiliation", "type": "string"},\n"answeredBy": {"title": "Answered By", "type": "string"},\n"airTime": {"title": "Air Time", "type": "string", "format": "date-time"}\n},\n"required": ["questionText", "author", "affiliation", "answeredBy", "airTime"]\n}\n},\n"speakers": {\n"title": "Speakers",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"name": {"title": "Name", "type": "string"},\n"bio": {"title": "Bio", "type": "string"}\n},\n"required": ["name", "bio"]\n}\n}\n},\n"required": ["title", "publicationDate", "questions", "speakers"]\n}', 'role': 'user'}, {'content': '{\n"title": "Hobbes and the Ideal Citizen",\n"publicationDate": "2022-09-09T06:00:00Z",\n"questions": [\n{\n"questionText": "What do citizens owe the state? What does the state owe its citizens?",\n"author": "Stanford political scientist Alison McQueen",\n"affiliation": "Stanford University",\n"answeredBy": "N/A",\n"airTime": "2022-09-11T11:00:00Z"\n}\n],\n"speakers": [\n{\n"name": "Devon Strolovitch",\n"bio": "Senior Producer of Philosophy Talk and host of Fog City Blues, with a background in medieval Judeo-Portuguese manuscripts and a PhD in Linguistics from Cornell University."\n}\n]\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "BirdSpecies",\n"type": "object",\n"properties": {\n"commonName": {\n"title": "Common Name",\n"type": "string"\n},\n"scientificName": {\n"title": "Scientific Name",\n"type": "string"\n},\n"family": {\n"title": "Family",\n"type": "string"\n},\n"habitat": {\n"title": "Habitat",\n"type": "string"\n},\n"distribution": {\n"title": "Distribution",\n"type": "string"\n},\n"subspecies": {\n"title": "Subspecies",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"references": {\n"title": "References",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["commonName", "scientificName", "family", "habitat", "distribution"]\n}\nparagraph:\nThe Peruvian seaside cinclodes or surf cinclodes (Cinclodes taczanowskii) is a species of bird in the family Furnariidae. It is endemic to rocky shorelines in Peru. It is often considered a subspecies of the Chilean seaside cinclodes.\n\nReferences\n\nPeruvian seaside cinclodes\nBirds of Peru\nEndemic birds of Peru\nWestern South American coastal birds\nPeruvian seaside cinclodes\nPeruvian seaside cinclodes\nTaxonomy articles created by Polbot', 'role': 'user'}, {'content': '{\n"commonName": "Peruvian seaside cinclodes",\n"scientificName": "Cinclodes taczanowskii",\n"family": "Furnariidae",\n"habitat": "rocky shorelines",\n"distribution": "endemic to Peru",\n"subspecies": ["Chilean seaside cinclodes"],\n"references": [\n"Birds of Peru",\n"Endemic birds of Peru",\n"Western South American coastal birds",\n"Peruvian seaside cinclodes",\n"Peruvian seaside cinclodes",\n"Taxonomy articles created by Polbot"\n]\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "FootballPlayer",\n"type": "object",\n"properties": {\n"playerName": {\n"title": "Player Name",\n"type": "string"\n},\n"position": {\n"title": "Position",\n"type": "string"\n},\n"collegeTeam": {\n"title": "College Team",\n"type": "string"\n},\n"university": {\n"title": "University",\n"type": "string"\n},\n"alliance": {\n"title": "Alliance",\n"type": "string"\n},\n"selectedAllStarTeam": {\n"title": "Selected All-Star Team",\n"type": "string"\n},\n"references": {\n"title": "References",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["playerName", "position", "collegeTeam", "university", "alliance", "selectedAllStarTeam", "references"]\n}\nparagraph:\nPaul A. "Dutch" Reule was an American football player. He played college football for the Mississippi A & M Aggies of Mississippi A & M University, selected an All-Southern fullback in 1912. He played for the Toledo Maroons.\n\nSee also\n1912 College Football All-Southern Team\n\nReferences\n\nAmerican football fullbacks\nMississippi State Bulldogs football players\nAll-Southern college football players\nToledo Maroons players', 'role': 'user'}, {'content': '{\n"playerName": "Paul A. "Dutch" Reule",\n"position": "fullback",\n"collegeTeam": "Mississippi A & M Aggies",\n"university": "Mississippi A & M University",\n"alliance": "Toledo Maroons",\n"selectedAllStarTeam": "1912 College Football All-Southern Team",\n"references": [\n"American football fullbacks",\n"Mississippi State Bulldogs football players",\n"All-Southern college football players",\n"Toledo Maroons players"\n]\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nReferee Goofy plans to start throwing Disneyland annual passholders in the penalty box if they don\'t show up for Magic Key reservations after the Anaheim theme park quietly let its most die-hard fans slide on a no-show policy for half a year.\n\nDisneyland began enforcing the Magic Key no-show policy this week after a six-month grace period when the penalties for skipping an advance reservation were not imposed, according to Disney officials.\n\nMagic Keyholders will start at zero no-shows this week with no penalties for past infractions.\n\nDisneyland rolled out the Magic Key program in August as a replacement for the popular annual passholder program used by a million fans. Disneyland ended its former four-decade-old annual pass program in January 2021 during the extended coronavirus closures of the parks.\n\nKeyholders are required to make advance reservations as part of a new system designed to reduce crowding and spread demand throughout the year. Magic Key annual passes range in price from $399 to $1,399 — with passholders able to hold two to six reservations at a time depending on their pass level.\n\nKeyholders who are a no-show for three reservations in a 90-day window will be unable to make new park reservations for 30 days beginning the day after the third no-show. Existing park reservations at that time will not be canceled.\n\nBrady MacDonald is a theme park reporter for the Orange County Register and the Southern California News Group. He\'s covered the theme park industry for more than 25 years. He writes about Disney, Universal, Six Flags, SeaWorld, Cedar Fair and Legoland parks in Southern California, across the United States and around the world. As a member of the SCNG Features team, he also writes about entertainment, travel, pop culture, music, restaurants and craft beer.\njson schema:\n{\n"title": "PolicyUpdate",\n"type": "object",\n"properties": {\n"policyName": {\n"title": "Policy Name",\n"type": "string"\n},\n"policyDescription": {\n"title": "Policy Description",\n"type": "string"\n},\n"enforcementDate": {\n"title": "Enforcement Date",\n"type": "string",\n"format": "date"\n},\n"gracePeriod": {\n"title": "Grace Period",\n"type": "object",\n"properties": {\n"startDate": {\n"title": "Start Date",\n"type": "string",\n"format": "date"\n},\n"endDate": {\n"title": "End Date",\n"type": "string",\n"format": "date"\n}\n},\n"required": ["startDate", "endDate"]\n},\n"penalties": {\n"title": "Penalties",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"infraction": {\n"title": "Infraction",\n"type": "string"\n},\n"penalty": {\n"title": "Penalty",\n"type": "string"\n},\n"duration": {\n"title": "Duration",\n"type": "number"\n}\n},\n"required": ["infraction", "penalty", "duration"]\n}\n},\n"contactInfo": {\n"title": "Contact Information",\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"position": {\n"title": "Position",\n"type": "string"\n},\n"publication": {\n"title": "Publication",\n"type": "string"\n}\n},\n"required": ["name", "position", "publication"]\n}\n}\n}', 'role': 'user'}, {'content': '{\n"policyName": "Magic Key no-show policy",\n"policyDescription": "Penalizing Disneyland annual passholders for not showing up for Magic Key reservations",\n"enforcementDate": "Please convert the text \'this week\' to a date",\n"gracePeriod": {\n"startDate": "2021-08-01",\n"endDate": "2022-01-31"\n},\n"penalties": [\n{\n"infraction": "Three no-shows in a 90-day window",\n"penalty": "Unable to make new park reservations for 30 days",\n"duration": 30\n}\n],\n"contactInfo": {\n"nam
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nOceanobacter is a bacteria genus from the family of Oceanospirillaceae with one known species (Oceanobacter kriegii).\n\nReferences\n\nOceanospirillales\nMonotypic bacteria genera\nBacteria genera\njson schema:\n{\n"title": "BacteriaGenus",\n"type": "object",\n"properties": {\n"genusName": {\n"title": "Genus Name",\n"type": "string"\n},\n"family": {\n"title": "Family",\n"type": "string"\n},\n"species": {\n"title": "Species",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"speciesName": {\n"title": "Species Name",\n"type": "string"\n}\n},\n"required": ["speciesName"]\n}\n},\n"references": {\n"title": "References",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["genusName", "family", "species", "references"]\n}', 'role': 'user'}, {'content': '{\n"genusName": "Oceanobacter",\n"family": "Oceanospirillaceae",\n"species": [\n{\n"speciesName": "Oceanobacter kriegii"\n}\n],\n"references": [\n"Oceanospirillales",\n"Monotypic bacteria genera",\n"Bacteria genera"\n]\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nThis magician of the Venezuelan left in shock Sofia Vergara would you look at that! – E! Online Latino\n\nThe star arrives in Colombia America\'s Got Talent of course, it was a big plant. And, despite the sympathy of Sofia, the distance is much relationship Simon Cowellthis does not mean that it is easy to surprise.\n\nIn advance for the next season Chapter 15 total media show, showed an amazing act of a young illusionist.\njson schema:\n{\n"title": "ShowEvent",\n"type": "object",\n"properties": {\n"magicianName": {\n"title": "Magician Name",\n"type": "string"\n},\n"celebrityName": {\n"title": "Celebrity Name",\n"type": "string"\n},\n"eventLocation": {\n"title": "Event Location",\n"type": "string"\n},\n"showName": {\n"title": "Show Name",\n"type": "string"\n},\n"seasonNumber": {\n"title": "Season Number",\n"type": "number"\n},\n"actDescription": {\n"title": "Act Description",\n"type": "string"\n}\n},\n"required": ["magicianName", "celebrityName", "eventLocation", "showName", "seasonNumber", "actDescription"]\n}', 'role': 'user'}, {'content': '{\n"magicianName": "The young illusionist",\n"celebrityName": "Sofia Vergara",\n"eventLocation": "Colombia",\n"showName": "America\'s Got Talent",\n"seasonNumber": 15,\n"actDescription": "showed an amazing act"\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "MothSpecies",\n"type": "object",\n"properties": {\n"scientificName": {\n"title": "Scientific Name",\n"type": "string"\n},\n"family": {\n"title": "Family",\n"type": "string"\n},\n"knownFrom": {\n"title": "Known From",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"references": {\n"title": "References",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"yearDescribed": {\n"title": "Year Described",\n"type": "string",\n"format": "date-year"\n}\n},\n"required": ["scientificName", "family", "knownFrom"]\n}\nparagraph:\nCaloptilia isotoma is a moth of the family Gracillariidae. It is known from Nigeria, Namibia, Zimbabwe and South Africa.\n\nReferences\n\nisotoma\nLepidoptera of Namibia\nLepidoptera of South Africa\nLepidoptera of West Africa\nLepidoptera of Zimbabwe\nMoths of Sub-Saharan Africa\nMoths described in 1914', 'role': 'user'}, {'content': '{\n"scientificName": "Caloptilia isotoma",\n"family": "Gracillariidae",\n"knownFrom": ["Nigeria", "Namibia", "Zimbabwe", "South Africa"],\n"references": ["isotoma", "Lepidoptera of Namibia", "Lepidoptera of South Africa", "Lepidoptera of West Africa", "Lepidoptera of Zimbabwe", "Moths of Sub-Saharan Africa", "Moths described in 1914"],\n"yearDescribed": "1914"\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "CelebrityMarriage",\n"type": "object",\n"properties": {\n"celebrityName": {\n"title": "Celebrity Name",\n"type": "string"\n},\n"spouseName": {\n"title": "Spouse Name",\n"type": "string"\n},\n"marriageDate": {\n"title": "Marriage Date",\n"type": "string",\n"format": "date"\n},\n"marriageImage": {\n"title": "Marriage Image",\n"type": "string",\n"format": "uri"\n},\n"pregnancy": {\n"title": "Pregnancy",\n"type": "boolean"\n},\n"spouseOccupation": {\n"title": "Spouse Occupation",\n"type": "string"\n}\n},\n"required": [\n"celebrityName",\n"spouseName",\n"marriageDate",\n"marriageImage"\n]\n}\nparagraph:\nSurprise! Christina Ricci is married. The actress tied the knot with hairstylist Mark Hampton and revealed the happy news in a pair of cute Instagram selfies over the weekend. The newlyweds look so in love in the close-up snaps, smiling for the camera as Christina holds a small bouquet. Christina and Mark\'s new milestone comes just two months after anothe
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nHüseyin-Ali Pala (born 9 April 1990) is a Turkish-German footballer who plays for Bayburt İdarespor.\n\nExternal links\n\n1990 births\nLiving people\nGerman footballers\nGerman people of Turkish descent\nVfB Stuttgart II players\nStuttgarter Kickers players\nSG Sonnenhof Großaspach players\n3. Liga players\nAssociation football forwards\nPeople from Backnang\nSportspeople from Stuttgart (region)\nFootballers from Baden-Württemberg\njson schema:\n{\n"title": "FootballPlayer",\n"type": "object",\n"properties": {\n"playerName": {\n"title": "Player Name",\n"type": "string"\n},\n"birthDate": {\n"title": "Birth Date",\n"type": "string",\n"format": "date"\n},\n"nationality": {\n"title": "Nationality",\n"type": "string"\n},\n"currentTeam": {\n"title": "Current Team",\n"type": "string"\n},\n"position": {\n"title": "Position",\n"type": "string"\n},\n"placeOfBirth": {\n"title": "Place of Birth",\n"type": "string"\n},\n"league": {\n"title": "League",\n"type": "string"\n}\n},\n"required": ["playerName", "birthDate", "nationality", "currentTeam", "position", "placeOfBirth", "league"]\n}', 'role': 'user'}, {'content': '{\n"playerName": "Hüseyin-Ali Pala",\n"birthDate": "1990-04-09",\n"nationality": "Turkish-German",\n"currentTeam": "Bayburt İdarespor",\n"position": "footballer",\n"placeOfBirth": "Backnang",\n"league": "3. Liga"\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nAn iconic former hotel with planning permission for development on Bideford riverside is being sold at auction.\n\nThe Tantons Hotel in New Road is among the lots in the sixth auction of 2019 by the regional land and property auctioneers Clive Emsons on September 12.\n\nThe iconic Bideford building was damaged by a huge fire in 2011 and has since had conversion works costing more than £1 million.\n\nThe former hotel in North Devon is known to have traded as an inn as far back as 1722.\n\nThe property was trading as Tantons Hotel at the time of the 2011 fire which saw fire crews rescue three guests before fighting to save neighbouring properties.\n\nThe hotel ceased to trade following the fire, and the iconic Grade-II listed building which sits on The Quay opposite Bideford\'s \'Old Bridge\' has now undergone considerable change with the main building converted into eight two and three bedroom flats.\n\nRead More\n\nAt one point the hotel once had 50 bedrooms spread over three and four floors in two buildings.\n\nA ground floor bar and restaurant at the former hotel is available to lease\n\nIt is being auctioned jointly with KLP Kitchener of Exeter.\n\nBritain\'s largest independent regional land and property auctioneer, Clive Emson Auctioneers, raised £19 million across the company from 125 lots catalogued for its July auction series, with a West Country sales rate so far in 2019 of nearly 90%.\n\nIt holds eight rounds of auctions at five locations from Essex to Cornwall on consecutive days every six weeks, this time with a total of 170 lots.\n\nStarting at 11am, the sixth of Clive Emson Auctioneers\' regional West Country auctions in 2019 takes place at the St Mellion International Resort, St Mellion, Cornwall, on Thursday September 12, 2019.\njson schema:\n{\n"title": "PropertyAuction",\n"type": "object",\n"properties": {\n"propertyName": {"title": "Property Name", "type": "string"},\n"location": {"title": "Location", "type": "string"},\n"history": {"title": "History", "type": "string"},\n"damageDetails": {"title": "Damage Details", "type": "string"},\n"conversionDetails": {"title": "Conversion Details", "type": "string"},\n"numberOfBedrooms": {"title": "Number of Bedrooms", "type": "number"},\n"floorDetails": {"title": "Floor Details", "type": "string"},\n"buildingDetails": {"title": "Building Details", "type": "string"},\n"leasedSpace": {"title": "Leased Space", "type": "string"},\n"jointAuctioneer": {"title": "Joint Auctioneer", "type": "string"},\n"auctioneerDetails": {\n"title": "Auctioneer Details",\n"type": "object",\n"properties": {\n"auctionName": {"title": "Auction Name", "type": "string"},\n"salesRate": {"title": "Sales Rate", "type": "number"},\n"lotsCatalogued": {"title": "Lots Catalogued", "type": "number"},\n"auctionDate": {"title": "Auction Date", "type": "string", "format": "date"}\n},\n"required": ["auctionName", "salesRate", "lotsCatalogued", "auctionDate"]\n}\n},\n"required": ["propertyName", "location", "history", "damageDetails", "conversionDetails", "numberOfBedrooms", "floorDetails", "buildingDetails", "leasedSpace", "jointAuctioneer", "auctioneerDetails"]\n}', 'role': 'user'}, {'content': '{\n"propertyName": "Tantons Inn or Tantons Hotel",\n"location": "Bideford riverside, North Devon",\n"history": "known to have traded as an inn as far back as 1722, at one point the hotel once had 50 bedrooms spread over three and four floors in two buildings",\n"damageDetails": "damaged by a huge fire in 2011",\n"conversionDetails": "conversion works costing more than £1 million, the main building converted into eight two and t
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "Manhunt",\n"type": "object",\n"properties": {\n"victimName": {"title": "Victim Name", "type": "string"},\n"victimAge": {"title": "Victim Age", "type": "number"},\n"victimHometown": {"title": "Victim Hometown", "type": "string"},\n"suspectName": {"title": "Suspect Name", "type": "string"},\n"suspectAge": {"title": "Suspect Age", "type": "number"},\n"suspectLastKnownLocation": {"title": "Suspect Last Known Location", "type": "string"},\n"crimeLocation": {"title": "Crime Location", "type": "string"},\n"crimeDate": {"title": "Crime Date", "type": "string", "format": "date"},\n"additionalSuspects": {\n"title": "Additional Suspects",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"suspectAge": {"title": "Suspect Age", "type": "number"},\n"suspectArrestStatus": {"title": "Suspect Arrest Status", "type": "string"},\n"suspectCustodyReason": {"title": "Suspect Custody Reason", "type": "string"}\n},\n"required": ["suspectAge", "suspectArrestStatus", "suspectCustodyReason"]\n}\n},\n"vehicleDescription": {\n"title": "Vehicle Description",\n"type": "object",\n"properties": {\n"vehicleMake": {"title": "Vehicle Make", "type": "string"},\n"vehicleModel": {"title": "Vehicle Model", "type": "string"},\n"vehicleColor": {"title": "Vehicle Color", "type": "string"},\n"vehicleRegistration": {"title": "Vehicle Registration", "type": "string"}\n},\n"required": ["vehicleMake", "vehicleModel", "vehicleColor", "vehicleRegistration"]\n}\n},\n"required": ["victimName", "victimAge", "victimHometown", "suspectName", "suspectAge", "suspectLastKnownLocation", "crimeLocation", "crimeDate", "additionalSuspects", "vehicleDescription"]\n}\nparagraph:\nPolice have launched a manhunt for a 28-year-old wanted on suspicion of the murder of a pensioner in Essex.\n\nEssex police are trying to find Leighton Snook after 83-year-old Donald Ralph was found dead at his home in the village of Aldham, near Colchester, just after 1pm on Tuesday.\n\nRalph\'s car was stolen and officers said they believe the killing was a targeted attack. Police said Snook should not be approached, but if any members of the public see him, they should notify the police immediately.\n\nSnook is from Colchester, but investigators said he has links to Hastings, Suffolk, Hartlepool and County Durham.\n\nA 16-year-old was arrested in Colchester on Wednesday evening and is being held in custody on suspicion of murder.\n\nDetectives are trying to track down Ralph\'s car, a blue Volvo V50 with the registration HD08 YFV, which was last recorded being driven to Hastings on the morning of the murder.\n\nRalph has been described as "very young at heart" and a man who enjoyed being in the countryside. His niece Tina Ralph said: "Don was a very young-at-heart, fit 83-year-old, with boundless energy, enthusiasm and enjoyment of life.\n\n"He was a man who loved all things countryside, particularly fishing and the Norfolk Broads, as well as being an avid bird lover. He loved horse racing and spent many a happy Saturday having a little flutter.\n\n"Don was a unique character, who will be much missed by his wide circle of friends, and all his family."\n\nDCI Scott Egerton of Kent and Essex serious crime directorate said: "We urgently need to speak to Leighton Snook in connection with the murder of Donald Ralph. I would strongly urge him, or anyone who knows of his whereabouts, to contact police.\n\n"I ask members of the public not to approach him but to call police immediately. Anybody who is found to be assisting those involved in Mr Ralph\'s murder in any way will be subject of a thorough inves
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "MilitaryRaid",\n"type": "object",\n"properties": {\n"groupName": {\n"title": "Group Name",\n"type": "string"\n},\n"allies": {\n"title": "Allies",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"target": {\n"title": "Target",\n"type": "string"\n},\n"location": {\n"title": "Location",\n"type": "string"\n},\n"outcome": {\n"title": "Outcome",\n"type": "string"\n},\n"equipmentSeized": {\n"title": "Equipment Seized",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["groupName", "allies", "target", "location", "outcome"]\n}\nparagraph:\nAccording to reports, the Jerusalem Brigade, backed by the Syrian Arab Army (SAA) and National Defense Forces (NDF), raided several ISIS hideouts in an effort to eliminate the remnants of the terrorist group in the Homs countryside.\n\nFollowing one successful raid, the Jerusalem Brigade was able to seize the equipment and supplies left behind by the Islamic State in the Badiya Al-Sham region.\n\nImprove the Discussion\n\nHelp us make this a great place for discussion by always working to improve the discussion in some way, however small. If you are not sure your post adds to the conversation, think over what you want to say and try again later.\n\nThe topics discussed here matter to us, and we want you to act as if they matter to you, too. Be respectful of the topics and the people discussing them, even if you disagree with some of what is being said.', 'role': 'user'}, {'content': '{\n"groupName": "Jerusalem Brigade",\n"allies": ["Syrian Arab Army (SAA)", "National Defense Forces (NDF)"],\n"target": "ISIS hideouts",\n"location": "Homs countryside, Badiya Al-Sham region",\n"outcome": "Successful raid with equipment and supplies seized"\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "CompanyAnnouncement",\n"type": "object",\n"properties": {\n"rightsValue": {\n"title": "Rights Value",\n"type": "number"\n},\n"companyName": {\n"title": "Company Name",\n"type": "string"\n},\n"plannedAction": {\n"title": "Planned Action",\n"type": "string"\n},\n"sourceOfInformation": {\n"title": "Source of Information",\n"type": "object",\n"properties": {\n"peopleIdentified": {\n"title": "People Identified",\n"type": "boolean"\n},\n"internalDeliberations": {\n"title": "Internal Deliberations",\n"type": "string"\n}\n},\n"required": ["peopleIdentified", "internalDeliberations"]\n},\n"investmentInCountry": {\n"title": "Investment in Country",\n"type": "number"\n},\n"businessRationale": {\n"title": "Business Rationale",\n"type": "string"\n}\n},\n"required": ["rightsValue", "companyName", "plannedAction", "sourceOfInformation", "investmentInCountry", "businessRationale"]\n}\nparagraph:\nThe rights had been estimated to fetch an unprecedented $7.7 billion. The US giant is planning to throw in the towel rather than get into a bidding war, the people said, asking not to be identified discussing internal deliberations. While Amazon has already invested more than $6 billion in the country, more spending merely for the online streaming rights to the league didn\'t make business sense, they said.', 'role': 'user'}, {'content': '{\n"rightsValue": 7.7e+09,\n"companyName": "The US giant",\n"plannedAction": "throw in the towel",\n"sourceOfInformation": {\n"peopleIdentified": true,\n"internalDeliberations": "discussing internal deliberations"\n},\n"investmentInCountry": 6e+09,\n"businessRationale": "more spending merely for the online streaming rights to the league didn\'t make business sense"\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "Hotel",\n"type": "object",\n"pr
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "EmailContent",\n"type": "object",\n"properties": {\n"sender": {\n"title": "Sender",\n"type": "string"\n},\n"recipients": {\n"title": "Recipients",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"topic": {\n"title": "Topic",\n"type": "string"\n},\n"content": {\n"title": "Content",\n"type": "string"\n},\n"gifts": {\n"title": "Gifts",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"amount": {\n"title": "Amount",\n"type": "number"\n},\n"year": {\n"title": "Year",\n"type": "integer"\n},\n"purpose": {\n"title": "Purpose",\n"type": "string"\n}\n},\n"required": ["amount", "year", "purpose"]\n}\n},\n"foundation": {\n"title": "Foundation",\n"type": "string"\n},\n"program": {\n"title": "Program",\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"description": {\n"title": "Description",\n"type": "string"\n}\n},\n"required": ["name", "description"]\n}\n},\n"required": ["sender", "recipients", "topic", "content", "gifts", "foundation", "program"]\n}\nparagraph:\n"We are seeking to learn more about the nature of that appointment from Dr. Kosslyn, who no longer works at the University," Lawrence S. Bacow, who is president of the Ivy League school, said in an email to students, alumni and faculty.\n\nThe cash, accepted between 1998 and 2007, supported "various faculty and institutional research activities across the University," Bacow said.\n\nThe largest single gift was $6.5 million Epstein gave in 2003 to support Harvard\'s "Program for Evolutionary Dynamics.\'\n\nThe program, which was founded that year, "is the study of the fundamental mathematical principles that guide evolution," the university\'s website says.\n\n"To date, we have uncovered no gifts received from Epstein or his foundation following his guilty plea," Bacow said.\n\nEpstein\'s gifts funded research and education "and nearly all were spent years ago," the email said, without elaborating.\n\nThe email ends,\n\n"Jeffrey Epstein\'s crimes were repulsive and reprehensible. I profoundly regret Harvard\'s past association with him. Conduct such as his has no place in our society. We act today in recognition of that fact. And we do so knowing that the scourge of sexual assault continues to demand our close attention and concerted action.\n\n"Harvard is not perfect, but you have my commitment as president that we will always strive to be better.', 'role': 'user'}, {'content': '{\n"sender": "Lawrence S. Bacow",\n"recipients": ["students", "alumni", "faculty"],\n"topic": "Nature of appointment with Dr. Kosslyn",\n"content": "Harvard\'s past association with Jeffrey Epstein and the gifts he gave",\n"gifts": [\n{\n"amount": 6500000,\n"year": 2003,\n"purpose": "Support Harvard\'s \'Program for Evolutionary Dynamics\'"\n}\n],\n"foundation": "Not mentioned",\n"program": {\n"name": "Program for Evolutionary Dynamics",\n"description": "Study of the fundamental mathematical principles that guide evolution"\n}\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nNational Voter Registration Day is Tuesday\n\nNational Voter Registration Day is Tuesday, but the last day to register to vote is Oct. 7.\n\nTo register, individuals need to be a United States citizen; a resident of the county where the voter registration application will be made; have no felony convictions; must have not been declared by a court exercising probate jurisdiction to be either totally or partially mentally incapacitated without the right to vote. If you are 17, you can register to vote if you will turn 18 before election day on Nov. 5.\n\nThose interested in learning more information can visit votetexas.gov/register-to-vote/index.html.\njson schema:\n{"title": "VoterRegistrationInfo", "type": "object", "properties": {"eventName": {"title": "Event Name", "type": "string"}, "eventDate": {"title": "Event Date", "type": "string", "format": "date"}, "registrationDeadline": {"title": "Registration Deadline", "type": "string", "format": "date"}, "eligibility": {"title": "Eligibility", "type": "object", "properties": {"citizenship": {"title": "Citizenship", "type": "boolean"}, "countyResidency": {"title": "County Residency", "type": "boolean"}, "felonyConvictions": {"title": "Felony Convictions", "type": "boolean"}, "courtDeclaredIncompetence": {"title": "Court Declared Incompetence", "type": "boolean"}, "age": {"title": "Age", "type": "number", "minimum": 17}}, "required": ["citizenship", "countyResidency", "felonyConvictions", "courtDeclaredIncompetence", "age"]}, "registrationUrl": {"title": "Registration URL", "type": "string", "format": "uri"}}, "required": ["eventName", "eventDate", "registrationDeadline", "eligibility", "registrationUrl"]}', 'role': 'user'}, {'content': '{\n "eventName": "National Voter Registration Day",\n "eventDate": "2023-09-19",\n "registrationDeadline": "2023-10-07",\n "eligibility": {\n "citizenship": true,\n "countyResidency": true,\n "felonyConvictions": false,\n "courtDeclaredIncompetence": false,\n "age": 17\n },\n "registrationUrl": "http://votetexas.gov/register-to-vote/index.html"\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "ArrestInformation",\n"type": "object",\n"properties": {\n"arrestedPersonName": {\n"title": "Arrested Person Name",\n"type": "string"\n},\n"accusedCrime": {\n"title": "Accused Crime",\n"type": "string"\n},\n"crimeSections": {\n"title": "Crime Sections",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"complainantName": {\n"title": "Complainant Name",\n"type": "string"\n},\n"complainantOccupation": {\n"title": "Complainant Occupation",\n"type": "string"\n},\n"relationshipDuration": {\n"title": "Relationship Duration",\n"type": "string"\n},\n"numberOfPregnancies": {\n"title": "Number of Pregnancies",\n"type": "number"\n},\n"threatsDuringRelationship": {\n"title": "Threats During Relationship",\n"type": "string"\n},\n"promiseOfMarriage": {\n"title": "Promise of Marriage",\n"type": "boolean"\n}\n},\n"required": [\n"arrestedPersonName",\n"accusedCrime",\n"crimeSections",\n"complainantName",\n"complainantOccupation",\n"relationshipDuration"\n]\n}\nparagraph:\nFormer AIADMK minister M Manikandan has been arrested in Bengaluru after a Malaysian woman levelled rape charges against him. Manikandan has been booked under various sections of the Indian Penal Code (IPC), including charges of rape, causing miscarriage and cheating.\n\nThe complaint was lodged two weeks ago at the women\'s police station in Adyar, in which the woman stated she is an actor by profession and had been in a relationship with Manikandan for years.\n\nThe Malaysian wo
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "YouTubeChannel",\n"type": "object",\n"properties": {\n"channelName": {\n"title": "Channel Name",\n"type": "string"\n},\n"description": {\n"title": "Description",\n"type": "string"\n},\n"category": {\n"title": "Category",\n"type": "string"\n},\n"influences": {\n"title": "Influences",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"targetAudience": {\n"title": "Target Audience",\n"type": "string"\n},\n"recentVideos": {\n"title": "Recent Videos",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["channelName", "description", "category", "influences", "targetAudience", "recentVideos"]\n}\nparagraph:\nThe Bro Dynasty\n\nHe is a new YouTuber I found him through instagram he is probobly one of my favorites. He is kinda similar to ricegum. He roasts people that are cringey on social media but his main message is to show that anyone in this day can be internet famous. He recently roasted woahhvicky\'s and shelby diggs cringey instagram videos.', 'role': 'user'}, {'content': '{\n"channelName": "The Bro Dynasty",\n"description": "A YouTuber who roasts cringey people on social media to show that anyone can be internet famous.",\n"category": "Entertainment",\n"influences": ["Ricegum"],\n"targetAudience": "People interested in humor and social media trends",\n"recentVideos": ["roasted woahhvicky\'s cringey Instagram video", "roasted shelby diggs cringey Instagram video"]\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
Converting format of dataset (num_proc=16): 4%|▍ | 450/10000 [00:00<00:07, 1322.47 examples/s][WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nLieutenant General Sir Thomas Herbert John Chapman Goodwin (24 May 1871 – 29 September 1960), known as Sir John Goodwin, was a British soldier and medical practitioner, who served as the Governor of Queensland from 1927 to 1932.\n\nEarly life and military career\nGoodwin was born on 24 May 1871 in Kandy, Ceylon (now Sri Lanka) to a British Army surgeon father and an Australian mother. He was educated in England at Newton College, Devon, and undertook medical training at St Mary\'s Hospital, London where he graduated with a Membership of the Royal College of Surgeons and Royal College of Physicians in 1891.\n\nCommissioned a lieutenant in the British Army Medical Department, Goodwin was stationed in India where he saw active service on the North-West Frontier from 1897 to 1898 and was awarded to the Distinguished Service Order.\n\nGovernor of Queensland\nGoodwin served as Governor of Queensland from 13 July 1927 to 7 April 1932.\n\nWilson was a freemason. During his term as governor, he was also Grand Master of the Grand Lodge of Queensland.\n\nReferences\n\n1871 births\n1960 deaths\nCompanions of the Distinguished Service Order\nForeign recipients of the Distinguished Service Medal (United States)\nGovernors of Queensland\nKnights Commander of the Order of St Michael and St George\nKnights Commander of the Order of the Bath\nPeople from Kandy\nRecipients of the Croix de guerre (Belgium)\nRoyal Army Medical Corps officers\nBritish Army lieutenant generals\nBritish people in British Ceylon\nRecipients of the Distinguished Service Medal (US Army)\njson schema:\n{"title": "Person", "type": "object", "properties": {"name": {"title": "Name", "type": "string"}, "birthDate": {"title": "Birth Date", "type": "string", "format": "date"}, "birthPlace": {"title": "Birth Place", "type": "string"}, "education": {"title": "Education", "type": "array", "items": {"type": "object", "properties": {"schoolName": {"title": "School Name", "type": "string"}, "location": {"title": "Location", "type": "string"}}, "required": ["schoolName", "location"]}}, "occupation": {"title": "Occupation", "type": "array", "items": {"type": "object", "properties": {"jobTitle": {"title": "Job Title", "type": "string"}, "organization": {"title": "Organization", "type": "string"}, "achievements": {"title": "Achievements", "type": "array", "items": {"type": "string"}}}, "required": ["jobTitle", "organization", "achievements"]}}, "honors": {"title": "Honors", "type": "array", "items": {"type": "object", "properties": {"honorName": {"title": "Honor Name", "type": "string"}, "year": {"title": "Year", "type": "string", "format": "date"}}, "required": ["honorName", "year"]}}}, "required": ["name", "birthDate", "birthPlace", "education", "occupation", "honors"]}', 'role': 'user'}, {'content': '{\n "name": "Thomas Herbert John Chapman Goodwin",\n "birthDate": "1871-05-24",\n "birthPlace": "Kandy, Ceylon (now Sri Lanka)",\n "education": [\n {\n "schoolName": "Newton College, Devon",\n "location": "England"\n },\n {\n "schoolName": "St Mary\'s Hospital, London",\n "location": "England"\n }\n ],\n "occupation": [\n {\n "jobTitle": "British soldier and medical practitioner",\n "organization": "British Army",\n "achievements": [\n "Awarded the Distinguished Service Order",\n "Served as Governor of Queensland from 1927 to 1932",\n "Grand Master of the Grand Lodge of Queensland"\n ]\n }\n ],\n "honors": [\n
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nAbout this item\nThis Cocktail shaker not only looks attractive and elegant but has also been designed to last. Made from high quality stainless steel with a mirror finish, it is guaranteed not to rust or leak and remain on your bar for many years to come!\nCocktail set is made of the high quality thick stainless steel alloy. Our cocktail mixer set maintains the original flavor of any alcohol beverage and you can mix and enjoy the purest tastes. Cocktail making set has great ability in anti-dents and durable.\nMeasure Cup:With solid 18-8 Stainless steel construction, Rolled top edges for strength and maintaining shape and being dishwasher safe this 15 ml x 30 ml Stainless Steel measure is the only choice for cocktail enthusiasts, home bars and restaurants.\nEasy to clean: Cocktail shaker, inside drawing processing with fine sanding, wear, easy to clean. Cocktail shaker for the aspiring or amateur bartender in all of us.\nIf you are an at-home drink enthusiast or even a professional bartender, this Cocktail Set is perfect for creating great-tasting drinks. It is also a must-have on any outdoor & home bar for you to create a wide variety of drinks including Whiskey, Scotch, Vodka, Tequila, Gin, Rum, Brandy, Sake, milk tea and ice coffee more.\njson schema:\n{"title": "ProductDescription", "type": "object", "properties": {"productName": {"title": "Product Name", "type": "string"}, "material": {"title": "Material", "type": "string"}, "guarantee": {"title": "Guarantee", "type": "string"}, "capacity": {"title": "Capacity", "type": "object", "properties": {"unit": {"title": "Unit", "type": "string"}, "value": {"title": "Value", "type": "number"}}, "required": ["unit", "value"]}, "easyToClean": {"title": "Easy to Clean", "type": "boolean"}, "useCases": {"title": "Use Cases", "type": "array", "items": {"type": "string"}}}, "required": ["productName", "material", "guarantee", "capacity", "easyToClean", "useCases"]}', 'role': 'user'}, {'content': '{\n "productName": "Cocktail Shaker",\n "material": "stainless steel",\n "guarantee": "not to rust or leak, remain on your bar for many years to come",\n "capacity": {\n "unit": "ml",\n "value": 30\n },\n "easyToClean": true,\n "useCases": [\n "Whiskey",\n "Scotch",\n "Vodka",\n "Tequila",\n "Gin",\n "Rum",\n "Brandy",\n "Sake",\n "milk tea",\n "ice coffee"\n ]\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "NewsBroadcast",\n"type": "object",\n"properties": {\n"actions": {\n"title": "Actions",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"description": {\n"title": "Description",\n"type": "string"\n},\n"time": {\n"title": "Time",\n"type": "string",\n"format": "date-time"\n}\n},\n"required": ["name", "description", "time"]\n}\n},\n"source": {\n"title": "Source",\n"type": "string"\n},\n"lastUpdated": {\n"title": "Last Updated",\n"type": "string",\n"format": "date-time"\n},\n"location": {\n"title": "Location",\n"type": "string"\n},\n"copyright": {\n"title": "Copyright",\n"type": "string"\n},\n"newsletterSignup": {\n"title": "Newsletter Signup",\n"type": "object",\n"properties": {\n"text": {\n"title": "Text",\n"type": "string"\n},\n"url": {\n"title": "URL",\n"type": "string",\n"format": "uri"\n}\n},\n"required": ["text", "url"]\n}\n},\n"required": ["actions", "source", "lastUpdated", "location", "copyright", "newsletterSignup"]\n}\nparagraph:\nActions\n\nEverywhere you are: time on the soccer field\n\nVirginia This Morning viewers share photos & videos of what they\'ve been up to.\
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nNew statistics reveal a 50 per cent increase in the number of times the public has acted to prevent a suicide on the railway in the south east.\n\nFollowing the launch of Small Talk Saves Lives by Samaritans in partnership with British Transport Police (BTP), Network Rail and the wider rail industry late last year, new figures show there were 16 interventions by members of the public so far this year in the South-East – a 50 per cent increase compared with 2017.\n\nFive of these interventions were in West Sussex and one was in East Sussex.\n\nThe figures coincide with the launch of a new phase of Small Talk Saves Lives, which emphasises how each of us has the experience we need to help save a life.\n\nSmall talk saves lives\n\nMembers of the public are being encouraged to keep an eye out for someone who may be at risk, using the same small talk we use every day to interrupt someone\'s suicidal thoughts and encourage them to get help.\n\nThat\'s exactly what Network Rail incident controller Ben West, 30, did when he helped save a man\'s life in south London.\n\nBen, who is based at the Three Bridges rail operating centre near Crawley, said: "I was on duty as a mobile operations manager on station patrol and I happened to be in the right place at the right time.\n\n"The Samaritans training teaches you to look for people who are in isolation, people under the influence of drink and drugs, and those with vacant expressions who are not fitting in with the wider scene around them.\n\n"This person ticked all of those boxes, so I went up to them and put myself in a position where I could physically intervene if they tried to jump and then started asking some normal, ice-breaking questions.\n\n"\'How are you today? What train are you trying to get?\' After a few questions I felt confident enough that I had built up a rapport with this person to say \'I know what you\'re going to do, I know you\'re thinking of committing suicide\'.\n\n"He nodded and tears started to go down his face. I said kindly \'why don\'t you come have a chat with me, we\'ll go somewhere private and we can just talk\'."\n\nTV and radio presenter, Gaby Roslin, has got involved with the new Small Talk Saves Lives campaign video and a special station announcement for rail commuters.\n\nGaby said: "The little conversations we have every day can be all that\'s needed to interrupt suicidal thoughts. Once you know that you have the power to make a difference, you\'re more likely to step in and do something.\n\n"I wanted to get involved in the Small Talk Saves Lives campaign after noticing someone in a park and trusting my instincts. Just a few words can have a huge impact."\n\nPassengers can have a key role to play in suicide prevention\n\nSamaritans CEO Ruth Sutherland said: "It\'s really heartening to see more members of the public feeling they have the confidence and knowledge to act if they\'re worried about someone, and we\'re grateful for their support.\n\n"Suicide is preventable and any one of us could have an opportunity to save a life. And a study shows some of us make small talk more than ten times a day.\n\n"A phrase as simple as, \'I can\'t believe this weather\', could be enough to interrupt a person\'s suicidal thoughts. Even if small talk doesn\'t come naturally to you, if something doesn\'t feel right, please try to start a conversation. There\'s no evidence you\'ll make things worse."\n\nJackie Doyle-Price, Minister for Mental Health, Inequalities and Suicide Prevention, added: "It\'s easy to understand why people might feel uncomfortable or shy about approaching a str
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nBest of Threads – WIWT, Mankels, Black Shoes and Podcasts\n\nJune 5, 2017 By Dave I.| Heads up: Buying via our links may result in us getting a commission. Here\'s why.\n\nWhat\'s Dappered Threads? It\'s our forum. Launched in 2011, Threads is a not too big, not too small, just-right-sized forum that doesn\'t put up with snobbery. All thanks to the users and hardworking mods. Once a month, one of the users/mods from Threads provides us with a highlight package of the last few weeks.\n\nStuck for something to wear? The What I Wore Today (WIWT) thread provides an excellent source of inspiration across a variety of styles, not to mention information on sizing and fit. Regular poster Leanderthal has been keeping readers entertained with his manydifferentlooks. Not to be outdone, CJMichaelRay displayed the way to mix and match a Lazio jacket with a Charles Tyrwhitt shirt, Spier & Mackay trousers, and Allen Edmonds McAllisters. The mixing and matching prompted much discussion from jfo2010 and armedferret (of horse-head fame) about the fit of Suitsupply jackets.\n\nWhiskeyburrito259 started off a discussion asking when it became de rigueur for men to show off their ankles. User Deke pointed out that catalogs need to share some of the blame for the popularity of male ankle exposure, while Nandyn rightly indicated pant length for gentlemen of shorter stature might require some people to roll up their pants, but it has to be kept in reason so as to not show off the ankle. Hierophant brought up the point that style/fashion is a personal choice while Shade warned everyone that soon there\'d be a capri trend starting up. So far though, no one has volunteer any photos of their mankels or capris to the group.\n\nAs mentioned on the main site, Massdrop has been killing lately. The recent Allen Edmonds group buy-in has lead to some discussion on the threads. armedferret revived the thread from October, 2016 to show off photos of his Chukkas from the last run. This prompted zorro to comment that he felt the finish on his pair of chukkas could have been a bit better, although Token reassured him that the fit and finish didn\'t look all that bad. Vicious49 garnered envy from those who missed out on the initial purchase by extolling the virtues of his purchase, while Rpd362 posed some questions to the group about sizing. No word yet on how many threads users who missed out on the previous purchase are jumping in on this round.\n\nIf you have a big roadtrip coming up and settling on the right music is going to be an issue, why not give podcasts a go? The what podcasts are you listening to thread has many different listings, suggestions, and mini reviews. DocDave seems to be quite prolific, suggesting different shows to try out. Fencedwall and C.Dubs gave some good feedback on S-Town while JBarwick came to the posting looking for new comedy podcasts. Have a suggestion for him? Go on over to the thread and help him out.\n\nJust to show you that good posts never die, they only recede in to the archives, gochrisgo revived a thread originally posted in May, 2013 on the topic of deciding between Allen Edmonds Fifth Avenue or Park Avenue shoes in black. Readers ryn, srlclark, and ajs116 all chimed in with their go-to shoes for black tie events. Gochrisgo settled on the Fifth Avenues after finding them on sale at the ShoeBank, which differs from the final comments posted in the thread back in June 2013 where the recommendations from ajk170 and Teej were for J&M shoes.\n\nHead on over to Threads and take a spin around. Many thanks to contributor Dave I. for assembling th
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nPolice Arrest Glendale Man After Pursuit Through Streets\n\nA 34-year-old man from Glendale was apprehended Thursday afternoon after leading Burbank Police on a short pursuit.\n\nOn January 30, 2020, at about 1:30 P.M., a Burbank Police Officer contacted the driver of a vehicle for a parking violation in the 200 block of West Elm Avenue. As the officer spoke with the driver, she felt he was possibly under the influence of narcotics, based on initial observations. During the contact, the driver suddenly rolled-up the vehicle\'s windows, locked the car doors, and fled.\n\nThe police officer quickly broadcast information to other patrol units, which allowed our Joint Air Support Unit to locate and track the suspect vehicle from overhead. Police Officers initially pursued the vehicle on the ground but discontinued the chase once speeds became excessive, allowing the police helicopter to follow the suspect from the air.\n\nThe suspect committed numerous traffic violations and drove at unsafe speeds, but remained within Burbank city limits the entire time. At about 1:45 P.M., the suspect pulled into a parking lot in the 400 block of Moss Street and came to a stop. Police Officers approached the vehicle and took the suspect into custody.\n\nThe suspect has since been identified as Yegia Zakaryan, a 34-year-old resident of Glendale. Zakaryan was in possession of several bindles of heroin at the time of his arrest. He was booked on suspicion of evading police and possession of narcotics for sale, as well as an outstanding arrest warrant for drug possession out of Los Angeles.\n\nZakaryan has since been released from custody after posting a $110,000 bond.\njson schema:\n{\n"title": "PoliceReport",\n"type": "object",\n"properties": {\n"eventTitle": {"title": "Event Title", "type": "string"},\n"arrestedName": {"title": "Arrested Name", "type": "string"},\n"age": {"title": "Age", "type": "number"},\n"hometown": {"title": "Hometown", "type": "string"},\n"date": {"title": "Date", "type": "string", "format": "date"},\n"time": {"title": "Time", "type": "string", "format": "time"},\n"location1": {"title": "Location 1", "type": "string"},\n"location2": {"title": "Location 2", "type": "string"},\n"violation": {"title": "Violation", "type": "string"},\n"unit": {"title": "Unit", "type": "string"},\n"officerName": {"title": "Officer Name", "type": "string"},\n"suspicion": {"title": "Suspicion", "type": "string"},\n"action": {"title": "Action", "type": "string"},\n"charge1": {"title": "Charge 1", "type": "string"},\n"charge2": {"title": "Charge 2", "type": "string"},\n"charge3": {"title": "Charge 3", "type": "string"},\n"bail": {"title": "Bail", "type": "number"},\n"status": {"title": "Status", "type": "string"}\n},\n"required": ["eventTitle", "arrestedName", "age", "hometown", "date", "time", "location1", "location2", "violation", "unit", "officerName", "suspicion", "action", "charge1", "charge2", "charge3", "bail", "status"]\n}', 'role': 'user'}, {'content': '{\n"eventTitle": "Police Arrest Glendale Man After Pursuit Through Streets",\n"arrestedName": "Yegia Zakaryan",\n"age": 34,\n"hometown": "Glendale",\n"date": "2020-01-30",\n"time": "13:30",\n"location1": "200 block of West Elm Avenue, Burbank",\n"location2": "400 block of Moss Street, Burbank",\n"violation": "Parking violation, evading police, possession of narcotics for sale",\n"unit": "Burbank Police",\n"officerName": "Not specified in the text",\n"suspicion": "Possibly under the influence of narcotics",\n"action": "Pursuit through streets, arrest",\n"charge1": "Evading police",\n"charge2": "Possession of n
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "Person",\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"lifespan": {\n"title": "Lifespan",\n"type": "object",\n"properties": {\n"birth": {\n"title": "Birth",\n"type": "string",\n"format": "date"\n},\n"death": {\n"title": "Death",\n"type": "string",\n"format": "date"\n}\n},\n"required": ["birth", "death"]\n},\n"occupation": {\n"title": "Occupation",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"aliases": {\n"title": "Aliases",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["name", "lifespan", "occupation"]\n}\nparagraph:\nClaire Adams (1898–1978) was a Canadian actress.\n\nClaire Adams may also refer to:\n\nClaire Adams (British actress), see List of EastEnders characters (2007)\nClaire Adams, musician in Gravenhurst (band)\nClaire Adams, character in Women in the Night', 'role': 'user'}, {'content': '{\n"name": "Claire Adams",\n"lifespan": {\n"birth": "1898",\n"death": "1978"\n},\n"occupation": ["Canadian actress"],\n"aliases": [\n"Claire Adams (British actress)",\n"Claire Adams, musician in Gravenhurst (band)",\n"Claire Adams, character in Women in the Night"\n]\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "CompostingInstructions",\n"type": "object",\n"properties": {\n"compostingMethod": {\n"title": "Composting Method",\n"type": "string"\n},\n"instructions": {\n"title": "Instructions",\n"type": "string"\n},\n"cutPieces": {\n"title": "Cut Pieces",\n"type": "boolean"\n},\n"addLeaves": {\n"title": "Add Leaves",\n"type": "boolean"\n},\n"collectionDeadline": {\n"title": "Collection Deadline",\n"type": "string",\n"format": "date"\n}\n},\n"required": ["compostingMethod", "instructions", "cutPieces", "addLeaves", "collectionDeadline"]\n}\nparagraph:\n"If you cut them into small pieces and add some fall leaves, they will break down much faster than leaving them whole. By next spring, you\'ll have a great, nutrient-rich soil amendment to add to your garden or lawn."\n\nIf you don\'t compost, the pumpkins can be placed into curbside yard-waste carts. The RDCO says yard-waste collection runs until the end of December.', 'role': 'user'}, {'content': '{\n"compostingMethod": "Manual",\n"instructions": "Cut pumpkins into small pieces, add fall leaves and let it break down to create a nutrient-rich soil amendment by next spring.",\n"cutPieces": true,\n"addLeaves": true,\n"collectionDeadline": "December 31"\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "GeographicalRegion",\n"type": "object",\n"properties": {\n"regionName": {\n"title": "Region Name",\n"type": "string"\n},\n"includesCommunes": {\n"title": "Included Communes",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"partOf": {\n"title": "Part of",\n"type": "string"\n},\n"seaName": {\n"title": "Sea Name",\n"type": "string"\n},\n"provinceName": {\n"title": "Province Name",\n"type": "string"\n}\n},\n"required": ["regionName", "includesCommunes", "partOf", "seaName", "provinceName"]\n}\nparagraph:\nTigullio is a traditional region and a gulf in the Metropolitan City of Genoa, Liguria, northern Italy. Part of the Riviera di Levante, it includes the communes of (from West to East) Portofino, Santa Margherita Ligure, Rapallo, Zoagli, Chiavari, Lavagna and Sestri Levante. The name derives from the ancient Ligurian tribe, Tigullii.\n\nItalian Riviera\nLigurian Sea\nProvince of Genoa', 'role': 'user'}, {'content': '{\n"regionName": "Tigullio",\n"includesCommunes": ["Portofino", "Santa Margherita Ligure", "Rapallo", "Zoagli", "Chiavari", "Lavagna", "Sestri Levante"],\n"pa
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "MeetingExpenses",\n"type": "object",\n"properties": {\n"meeting": {\n"title": "Meeting Details",\n"type": "object",\n"properties": {\n"attendees": {\n"title": "Number of Attendees",\n"type": "integer"\n},\n"totalFoodCost": {\n"title": "Total Food Cost",\n"type": "number",\n"format": "currency"\n}\n},\n"required": ["attendees", "totalFoodCost"]\n},\n"expensesPerPerson": {\n"title": "Expenses Per Person",\n"type": "object",\n"properties": {\n"foodCost": {\n"title": "Food Cost Per Person",\n"type": "number",\n"format": "currency"\n}\n},\n"required": ["foodCost"]\n},\n"comparison": {\n"title": "Comparison Details",\n"type": "object",\n"properties": {\n"previousCost": {\n"title": "Previous Cost",\n"type": "number",\n"format": "currency"\n},\n"currentCost": {\n"title": "Current Cost",\n"type": "number",\n"format": "currency"\n}\n},\n"required": ["previousCost", "currentCost"]\n},\n"reactions": {\n"title": "Reactions",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"speaker": {\n"title": "Speaker",\n"type": "string"\n},\n"message": {\n"title": "Message",\n"type": "string"\n}\n},\n"required": ["speaker", "message"]\n}\n}\n},\n"required": ["meeting", "expensesPerPerson", "comparison", "reactions"]\n}\nparagraph:\n"Charley me I still don\'t understand how 8 people spent 10,652,00 on food during a meeting. Ah how? It means each person\'s food (just one meal) cost 1,335.00 Ghana cedis. Really? If you own Ghana as your private company will you spend this amount on a meal for just one meeting?", he queried.\n\n"If we complained during the campaign that NDC spent 20 dollars on Kenkey in Brazil, how then can we support this? 20 dollars is just 100 Ghana. This is 1.335.00 per person per meal. Did they chew golden sausages or the elephant itself came to the meeting?"\n\nReports indicate that Mr. Kwame Owusu is still the Group Chairman of the Luxe Suites Hotel.\n\n"Apart from conflict of interest which is a serious offence let me ask this. How many people ate 135,125.00 Ghana cedis worth of food? Herrrrh they can eat oooo hahahahahaha. And wait, wait, wait. The second invoice…, how many people ate over 10,000.00 Ghana cedis food at one meeting? Come for your medals. You people can eat for Africa. Waaaawolo. Pick pockets have taken the public purse from Nana Addo\'s pocket", A Plus chided.\n\nDo you ever witness news or have a story that should be featured on Pulse Ghana? Submit your stories, pictures and videos to us now via WhatsApp: +233507713497, Social Media @pulseghana: #PulseEyewitness & DM or Email: eyewitness@pulse.com.gh.', 'role': 'user'}, {'content': '{\n"meeting": {\n"attendees": 8,\n"totalFoodCost": 1065200\n},\n"expensesPerPerson": {\n"foodCost": 133500\n},\n"comparison": {\n"previousCost": 20,\n"currentCost": 1335\n},\n"reactions": [\n{\n"speaker": "he",\n"message": "If you own Ghana as your private company will you spend this amount on a meal for just one meeting?"\n},\n{\n"speaker": "A Plus",\n"message": "How many people ate 135,125.00 Ghana cedis worth of food?... You people can eat for Africa. Waaaawolo."\n}\n]\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nEdward Badalian and Daniel Rodriguez are facing charges that include conspiracy, obstruction, entering a restricted building and theft and destruction of government property.\n\nIn a group chat they created on Telegram, Badalian and Rodriguez encouraged others to attend the Jan. 6 "Stop the Steal" rally that preceded the Capitol riot. In one message, Badalian allegedly wrote "we need to violently remove traitors and if they are in key position rapidly replace them with able-bodied patriots."\n\nProsecutors say the Telegram chat group was used to "advocate violence against certain groups and individuals" who "supported" the results of the 2020 presidential election.\n\nDuring the riot, prosecutors allege that Rodriguez threw a flagpole at police, set off a fire extinguisher toward them, and then used a "small, black electroshock weapon" on the back of an officer\'s neck.\n\nIn the days after the riot, Badalian appeared on the right-wing InfoWars program using the pseudonym "Turbo" to discuss video footage that allegedly showed him fighting someone he claimed to be "Antifa" who was disguised as a Trump supporter.\n\nLater in during the call-in program, another person discussing the same video footage referred to him as "Ed" instead of "Turbo." Prosecutors say Badalian and Rodgriguez later went to the home of the person who blurted out Badalian\'s first name and warned them against using his actual name when discussing the riot. They also told the person to delete photos that showed them participating in the riot.\n\nAs it turns out, the person who mistakenly said Badalian\'s first name was salon owner Gina Bisignano, 52, who went viral for spewing homophobic slurs and other hateful rhetoric during an anti-lockdown protest back in December of last year. She was also arrested for participating in the Capitol riot.\njson schema:\n{\n"title": "CriminalCase",\n"type": "object",\n"properties": {\n"defendants": {\n"title": "Defendants",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"charges": {\n"title": "Charges",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["name", "charges"]\n}\n},\n"groupChat": {\n"title": "Group Chat",\n"type": "object",\n"properties": {\n"platform": {\n"title": "Platform",\n"type": "string"\n},\n"messages": {\n"title": "Messages",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"sender": {\n"title": "Sender",\n"type": "string"\n},\n"content": {\n"title": "Content",\n"type": "string"\n}\n},\n"required": ["sender", "content"]\n}\n}\n},\n"required": ["platform", "messages"]\n},\n"incident": {\n"title": "Incident",\n"type": "object",\n"properties": {\n"location": {\n"title": "Location",\n"type": "string"\n},\n"date": {\n"title": "Date",\n"type": "string",\n"format": "date"\n},\n"actions": {\n"title": "Actions",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["location", "date", "actions"]\n},\n"mediaAppearances": {\n"title": "Media Appearances",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"program": {\n"title": "Program",\n"type": "string"\n},\n"alias": {\n"title": "Alias",\n"type": "string"\n},\n"topics": {\n"title": "Topics",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["program", "alias", "topics"]\n}\n},\n"warnings": {\n"title": "Warnings",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"personWarned": {\n"title": "Person Warned",\n"type": "string"\n},\n"reason": {\n"title": "Reason",\n"type": "string"\n}\n},\n"required"
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nPollo Tropical just added beef to its menu because Miami needs more carne\n\nWe heard a rumor that people like ordering the churrasco with balsamic tomatoes and pretending they made the dishes themselves. We endorse this behavior.\n\nIt\'s not just about chicken and pork anymore at Pollo Tropical.\n\nThe Miami-based fast-concept chicken chain has added churrasco with chimichurri to its menu. You can order it as part of a TropiChop bowl or a meal platter with two sides. We recommend getting it with the fried yuca, but then, we always recommend the fried yuca.\n\nAlso added to the menu are citrus marinated chicken wings. They\'re marinated for 24 hours and are worth getting sticky.\njson schema:\n{\n"title": "NewMenuItems",\n"type": "object",\n"properties": {\n"restaurantName": {\n"title": "Restaurant Name",\n"type": "string"\n},\n"newItem1": {\n"title": "New Item 1",\n"type": "object",\n"properties": {\n"itemName": {\n"title": "Item Name",\n"type": "string"\n},\n"description": {\n"title": "Description",\n"type": "string"\n},\n"preparation": {\n"title": "Preparation",\n"type": "string"\n},\n"availableOptions": {\n"title": "Available Options",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": [\n"itemName",\n"description",\n"preparation"\n]\n},\n"newItem2": {\n"title": "New Item 2",\n"type": "object",\n"properties": {\n"itemName": {\n"title": "Item Name",\n"type": "string"\n},\n"description": {\n"title": "Description",\n"type": "string"\n},\n"preparation": {\n"title": "Preparation",\n"type": "string"\n}\n},\n"required": [\n"itemName",\n"description",\n"preparation"\n]\n}\n}\n}', 'role': 'user'}, {'content': '{\n"restaurantName": "Pollo Tropical",\n"newItem1": {\n"itemName": "churrasco with chimichurri",\n"description": "can be ordered as part of a TropiChop bowl or a meal platter with two sides",\n"preparation": "served with chimichurri",\n"availableOptions": ["TropiChop bowl", "meal platter with two sides"]\n},\n"newItem2": {\n"itemName": "citrus marinated chicken wings",\n"description": "marinated for 24 hours",\n"preparation": "can be ordered worth getting sticky"\n}\n}', 'role': 'assistant'}, {'content': 'paragraph:\nBrian Fenimore completes journey from New Hampton to federal court\n\nRepublican-Clipper photo, Phil Conger\nNew federal judge takes oath: Cynthia Norton, chief judge of the U.S. Bankruptcy Court administers the oath of office to new federal bankruptcy judge Brian Fenimore during the investiture ceremony at the federal courthouse in Kansas City. Judge Fenimore\'s son, Will, was holding the Bible. More photos are on page 3.\n\nWith his former New Hampton grade schoolmates and other family and friends proudly looking on, Brian Fenimore took the oath of office as a new federal bankruptcy judge in a ceremony Friday afternoon at the Charles E. Whittaker federal courthouse in Kansas City.\n\nTwo former CIA employees are accusing the Trump administration\'s choice for CIA chief watchdog of being less than candid when he told Congress he didn\'t know about any active whistleblower complaints against him.\njson schema:\n{\n"title": "Ceremony",\n"type": "object",\n"properties": {\n"participant": {\n"title": "Participant",\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"role": {\n"title": "Role",\n"type": "string"\n}\n},\n"required": ["name", "role"]\n},\n"witnesses": {\n"title": "Witnesses",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"relation": {\n"title": "Relation",\n"type": "string"\n}\n},\n"required
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nChurch of the Nativity is a historic Roman Catholic church on the National Register of Historic Places in Menlo Park, California. It was built in 1887 and added to the National Register in 1980.\n\nHistory\nThe church was constructed in 1872 and was moved on log rollers twice before arriving at its current location in 1877. This earned it the nickname of the "Roaming Catholic Church."\n\nIn 1996 the church adopted the tradition of Perpetual Adoration, with members of the church volunteering to pray continuously in an uninterrupted chain, and has continued for over 20 years.\n\nThe current pastor at the Church of the Nativity is the Rev. Msgr. Steven D. Otellini, brother to past Intel CEO Paul Otellini, and, because of the church\'s proximity to Saint Patrick\'s Seminary and University in Menlo Park, weekend assistants often include young or international priests. Nearby are the Vallombrosa Center retreat house and the Corpus Christi Monastery of Dominican sisters. The parish serves a predominantly Anglo Catholic community and has an associated parochial school, Nativity School. With a relatively strong academic reputation locally, students from Nativity School have gone on to such local high schools as Sacred Heart Preparatory (Atherton, California), Saint Francis High School (Mountain View), Mercy High School (Burlingame, California), Bellarmine College Preparatory, and Notre Dame High School (Belmont, California) as well as colleges including Stanford University, Loyola Marymount University, California Polytechnic State University, New York University, University of California, Berkeley, University of California, Davis, and Saint Mary\'s College of California among others.\n\nReferences\n\nExternal links\n\nNativity Catholic School\n\nRoman Catholic churches in California\nNational Register of Historic Places in San Mateo County, California\nCarpenter Gothic church buildings in California\nRoman Catholic churches completed in 1887\nBuildings and structures in San Mateo County, California\nChurches on the National Register of Historic Places in California\n19th-century Roman Catholic church buildings in the United States\njson schema:\n{"title": "HistoricChurch", "type": "object", "properties": {"churchName": {"title": "Church Name", "type": "string"}, "location": {"title": "Location", "type": "object", "properties": {"city": {"title": "City", "type": "string"}, "state": {"title": "State", "type": "string"}, "country": {"title": "Country", "type": "string"}}, "required": ["city", "state", "country"]}, "constructionYear": {"title": "Construction Year", "type": "integer"}, "registerYear": {"title": "Register Year", "type": "integer"}, "nickname": {"title": "Nickname", "type": "string"}, "traditions": {"title": "Traditions", "type": "array", "items": {"type": "object", "properties": {"traditionName": {"title": "Tradition Name", "type": "string"}, "yearAdopted": {"title": "Year Adopted", "type": "integer"}}, "required": ["traditionName", "yearAdopted"]}}, "pastor": {"title": "Pastor", "type": "object", "properties": {"name": {"title": "Name", "type": "string"}, "relatives": {"title": "Relatives", "type": "array", "items": {"type": "string"}}}, "required": ["name"]}, "nearby": {"title": "Nearby", "type": "array", "items": {"type": "object", "properties": {"name": {"title": "Name", "type": "string"}, "type": {"title": "Type", "type": "string"}}, "required": ["name", "type"]}}, "community": {"title": "Community", "type": "string"}, "school": {"title": "School", "type": "object", "properties": {"name": {"title": "Name", "type": "string"}, "reputatio
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nTheMinister of Women Affairs, Pauline Tallen, has assured that her ministry would partner the Nigeria Association of Women Journalists (NAWOJ) on key areas to create a healthier and better society for women and children.\n\nShe gave the assurance when the National President of NAWOJ, Ladi Bala, led a delegation on a courtesy visit to her office on Abuja.\n\nWhile identifying the key areas of collaboration as the girl-child education, gender-based violence, and women participation in politics, the minister noted that her Ministry\'s mandates and that of NAWOJ are similar.\n\nShe called on women groups in the country to speak out against the rising cases of gender-based violence and other ills perpetrated against women and girls in the country.She added that effective representation of women in politics is another major assignment of her ministry, noting that without such representation, there will be no meaningful advancement for them.\n\nThe minister also stressed the need to raise the bar by ensuring that women who are into small-scale businesses are supported in every way necessary to enable them cater for their families.\n\nEarlier in her remarks, Bala, National President of NAWOJ, called on the Federal Government to ensure that the school environment is safe for Nigerian children, in the face of increasing insecurity in the country. She expressed deep concern over cases of abduction of schoolchildren by insurgents in parts of the country.\njson schema:\n{\n"title": "PartnershipAndCollaboration",\n"type": "object",\n"properties": {\n"partnerName": {\n"title": "Partner Name",\n"type": "string"\n},\n"partnerPosition": {\n"title": "Partner Position",\n"type": "string"\n},\n"receivingParty": {\n"title": "Receiving Party",\n"type": "string"\n},\n"purpose": {\n"title": "Purpose",\n"type": "string"\n},\n"keyAreas": {\n"title": "Key Areas",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"area": {\n"title": "Area",\n"type": "string"\n},\n"details": {\n"title": "Details",\n"type": "string"\n}\n}\n},\n"additionalProperties": false\n},\n"callToAction": {\n"title": "Call to Action",\n"type": "string"\n},\n"additionalNeeds": {\n"title": "Additional Needs",\n"type": "string"\n},\n"concerns": {\n"title": "Concerns",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"issue": {\n"title": "Issue",\n"type": "string"\n},\n"details": {\n"title": "Details",\n"type": "string"\n}\n}\n},\n"additionalProperties": false\n}\n},\n"required": [\n"partnerName",\n"partnerPosition",\n"receivingParty",\n"purpose",\n"keyAreas",\n"callToAction"\n]\n}', 'role': 'user'}, {'content': '{\n"partnerName": "Ladi Bala",\n"partnerPosition": "National President of NAWOJ",\n"receivingParty": "Pauline Tallen",\n"purpose": "Minister of Women Affairs",\n"keyAreas": [\n{\n"area": "girl-child education",\n"details": "Creation of a healthier and better society for women and children"\n},\n{\n"area": "gender-based violence",\n"details": "Speaking out against the rising cases and other ills perpetrated against women and girls in the country"\n},\n{\n"area": "women participation in politics",\n"details": "Effective representation of women in politics"\n},\n{\n"area": "support for small-scale businesses",\n"details": "Enabling women to cater for their families"\n}\n],\n"callToAction": "Women groups in the country to speak out against gender-based violence and other ills",\n"additionalNeeds": "Support for women who are into small-scale businesses",\n"concerns": [\n{\n"issue": "Increasing insecurity",\n"details": "Concern over cases of abduction of schoolchildre
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{"title": "EconomicForecast", "type": "object", "properties": {"omicronImpact": {"title": "Omicron Impact", "type": "object", "properties": {"economicImpact": {"title": "Economic Impact", "type": "string"}, "vaccinationRates": {"title": "Vaccination Rates", "type": "string"}, "governmentResponse": {"title": "Government Response", "type": "string"}}, "required": ["economicImpact", "vaccinationRates", "governmentResponse"]}, "labourMarkets": {"title": "Labour Markets", "type": "object", "properties": {"labourShortages": {"title": "Labour Shortages", "type": "string"}, "comparisonPeriod": {"title": "Comparison Period", "type": "string"}}, "required": ["labourShortages", "comparisonPeriod"]}, "inflation": {"title": "Inflation", "type": "object", "properties": {"currentImpact": {"title": "Current Impact", "type": "string"}, "peakInflation": {"title": "Peak Inflation", "type": "number"}, "inflationOutlook": {"title": "Inflation Outlook", "type": "string"}}, "required": ["currentImpact", "peakInflation", "inflationOutlook"]}, "housingMarket": {"title": "Housing Market", "type": "object", "properties": {"priceForecast": {"title": "Price Forecast", "type": "number"}, "declineComparison": {"title": "Decline Comparison", "type": "string"}}, "required": ["priceForecast", "declineComparison"]}, "gdpGrowth": {"title": "GDP Growth", "type": "object", "properties": {"growthForecast": {"title": "Growth Forecast", "type": "number"}, "timePeriod": {"title": "Time Period", "type": "string"}}, "required": ["growthForecast", "timePeriod"]}, "wageGrowth": {"title": "Wage Growth", "type": "object", "properties": {"wageForecast": {"title": "Wage Forecast", "type": "string"}, "drivingFactors": {"title": "Driving Factors", "type": "string"}}, "required": ["wageForecast", "drivingFactors"]}, "asbMission": {"title": "ASB Mission", "type": "string"}}, "required": ["omicronImpact", "labourMarkets", "inflation", "housingMarket", "gdpGrowth", "wageGrowth", "asbMission"]}\nparagraph:\nOmicron\'s economic impact\nhas been far milder than the 2020 and 2021 COVID\noutbreaks\n\nLabour markets are\nexceptionally tight, with labour shortages the most elevated\nsince the 1970s\n\nBorrowers are\ncurrently bearing the brunt of combatting inflation, but\nrate cuts to a more neutral level will happen –\npotentially from 2024\n\nFurther lifts\nin businesses\' capital spending are likely, but this and\nother productivity-boosting measures will take time to have\nan impact\n\nOmicron\'s mild bite\noffset by sluggish global growth Omicron\'s\neconomic impact has been far milder than that of previous\nCOVID outbreaks. High national vaccination rates and a\nfundamental shift in response from the Government means the\neconomy is being far less disrupted by the current outbreak\ncompared with the lockdowns of 2020 and 2021.\n\n"We\nbelieve the experience of recent months is a better\nreflection of the future for where we\'re heading with\nCOVID. Mobility\'s increased, but we anticipate sluggish\nGDP growth over the first half of 2022 as high inflation,\nlabour shortages and weakening global growth start to\nbite," says ASB Chief Economist Nick Tuffley.\n\nThe\nlatest ASB Economic Forecast anticipates growth around 1.5%\nper annum across 2022 and 2023.\n\nJobseekers, rather than employers, are\nin the driver\'s seat as labour shortages continue to bite,\nwith the ASB Economic Forecast indicating wage growth is\nexpected to strengthen across 2022 as businesses seek to\nattract and retain talent.\n\n"The employment\nuncertainties of the first wave of the pandemic are behind\nus," says Mr Tuffl
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nElton John Shows Support for Yola Cover of "Goodbye Yellow Brick Road"\n\nIn the latter part of this week, Yola released a deluxe version of her Grammy-nominated, critically acclaimed album Walk Through Fire featuring two new bonus tracks, including a Dan Auerbach-produced studio recording of Elton John\'s classic "Goodbye Yellow Brick Road," which was personally premiered by Elton John on his social media.\n\nElton John has previously praised Yola in a special edition of his Rocket Hour radio show on Beats, featuring fellow Highwoman, Brandi Carlile. Elton exclaimed, "If you haven\'t heard of Yola… go and see her… you\'ll be delighted at what you hear."\n\nThe deluxe album also includes the song "I Don\'t Wanna Lie," which was co-written by Yola, Auerbach, and legendary keys player Bobby Wood. Musicians who performed on the track included bassist Dave Roe (Johnny Cash) as well as Wood and drummer Gene Chrisman, who played on hits by Dusty Springfield and Elvis Presley.\n\n"Can\'t believe it has only been a year since we announced Walk Through Fire, working with Dan Auerbach and the team has been an absolute dream and so proud of everything we have achieved. So many great songs from that session didn\'t make the final cut, including "I Don\'t Wanna Lie." I\'m a HUGE Elton fan and we\'ve been playing "Goodbye Yellow Brick Road" on the tour and wanted to cut a version for this release which Dan also produced! And then for Elton to personally premiere it, well that is the icing on the cake!\n\nQueen of country-soul Yola recently received four nominations at the 62nd Grammy Awards including Best New Artist and three nominations in the Roots category including Best American Roots Song, Best American Roots Performance (for "Faraway Look") and Best Americana Album for her debut Dan Auerbach-produced album, Walk Through Fire. Auerbach also received a nomination for for Producer of the Year for his work on albums including Walk Through Fire. The 62nd Annual GRAMMY Awards will take place in Los Angeles on Sunday, January 26, 2020, at the STAPLES Center in Los Angeles.\n\nThe album, which was released in February 2019 to tremendous critical acclaim, received support from tastemakers from the New York Times to NPR, which stated the record was "the work of an artist sure to stun audiences for years to come." KCRW made "Faraway Look" their number 1 song of the year. Variety called the album "one of 2019\'s best musical breakouts," while Rolling Stone, included Walk Through Fire in their top 50 albums of 2019 and called Yola "an artist you need to know" and "the voice of 2019."\n\nYola toured the US extensively in 2019, from the Hollywood Bowl to Newport Folk Festival, SXSW, and Farm Aid. She recently performed at Nashville\'s Bridgestone arena, opening for Kacey Musgraves alongside Maggie Rogers, with a special appearance from Harry Styles. Additionally, Yola has performed with The Highwomen and Dolly Parton. She will close the year with a handful of performances, including a sold-out show in London, Holiday Cheer for FUV with Mavis Staples, Nathaniel Rateliff and Mumford And Sons in NYC on December 10, and two New Year\'s performances with Old Crow Medicine Show at the Ryman in Nashville.\n\nYola\'s Walk Through Fire continuing world tour includes shows across the US, Canada, and Europe. Next year\'s dates include Brooklyn\'s Music Hall of Williamsburg, Indianapolis\'s Hi-Fi, Austin\'s Scoot Inn, San Francisco\'s Great American Music Hall, and two nights at the Troubadour in LA; additional shows will be announced shortly. All dates are on sale now. Am
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "CricketInnings",\n"type": "object",\n"properties": {\n"playerName": {"title": "Player Name", "type": "string"},\n"runsScored": {"title": "Runs Scored", "type": "number"},\n"deliveriesFaced": {"title": "Deliveries Faced", "type": "number"},\n"injuriesReceived": {\n"title": "Injuries Received",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"bodyPart": {"title": "Body Part", "type": "string"},\n"details": {"title": "Details", "type": "string"}\n}\n}\n},\n"praiseReceived": {\n"title": "Praise Received",\n"type": "object",\n"properties": {\n"source": {"title": "Source", "type": "string"},\n"comments": {"title": "Comments", "type": "string"}\n}\n}\n},\n"required": ["playerName", "runsScored", "deliveriesFaced"]\n}\nparagraph:\nHyderabad: Cheteshwar Pujara took more than 200 deliveries to score his half-century on the final day of the fourth and deciding Test at the Gabba in Brisbane. However, he took over 10 blows to his body and dodged the Australian bowling attack, tiring them to allow other Indian batters to score runs and pave way for the memorable victory.\n\nFormer Indian skipper Sunil Gavaskar was all praise for the gritty knock by Pujara. "Whatever you say about him, it will be very less. He had put his life for the Indian cricket team. He took so many blows, on the gloves, body and was hit on the helmet as well but he stayed there. Cheteshwar Pujara\'s presence at the other end gives confidence to the stroke-players," he said.\n\n"Because of his presence at the other end, the young players, who are the stroke-players feel the confidence that there is someone who is holding one end and because of that his innings was extremely important," Gavaskar added.', 'role': 'user'}, {'content': '{\n"playerName": "Cheteshwar Pujara",\n"runsScored": 50,\n"deliveriesFaced": 200,\n"injuriesReceived": [\n{\n"bodyPart": "body",\n"details": "Took over 10 blows"\n},\n{\n"bodyPart": "gloves",\n"details": "Hit on the gloves"\n},\n{\n"bodyPart": "helmet",\n"details": "Hit on the helmet"\n}\n],\n"praiseReceived": {\n"source": "Sunil Gavaskar",\n"comments": "Whatever you say about him, it will be very less. He had put his life for the Indian cricket team... Because of his presence at the other end, the young players... feel the confidence that there is someone who is holding one end and because of that his innings was extremely important."\n}\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "NewsArticle",\n"type": "object",\n"properties": {\n"articleTitle": {"title": "Article Title", "type": "string"},\n"artistName": {"title": "Artist Name", "type": "string"},\n"songName": {"title": "Song Name", "type": "string"},\n"albumName": {"title": "Album Name", "type": "string"},\n"releaseDate": {"title": "Release Date", "type": "string", "format": "date"},\n"artistAge": {"title": "Artist Age", "type": "number"},\n"articleBody": {"title": "Article Body", "type": "string"},\n"videoURL": {"title": "Video URL", "type": "string", "format": "uri"},\n"newsSource": {"title": "News Source", "type": "string"},\n"copyrightText": {"title": "Copyright Text", "type": "string"}\n},\n"required": ["articleTitle", "artistName", "songName", "albumName", "releaseDate", "artistAge", "articleBody", "videoURL", "newsSource", "copyrightText"]\n}\nparagraph:\nWillie Nelson Releases \'Me and You\' Video\n\n(Radio.com) Willie Nelson might be pushing 85, but the Texas music legend still has plenty of pep in his step. The \'Red Headed Stranger\' has shared a new song, "Me and You".\n\nThe upbeat song is new advance track from his forthcoming full-l
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nThe U.S. Basketball Writers\' Association will posthumously induct former Denver Post sports reporter Tom Kensler into its Hall of Fame.\n\nKensler died in 2016 from a brain aneurysm. He covered colleges along the Front Range for decades, as well as golf and other sports. Kensler led the Post\'s March Madness coverage for several years and was a mainstay at the Final Four.\n\nKensler won numerous USBWA writing awards throughout his distinguished career.\njson schema:\n{\n"title": "HallOfFameInduction",\n"type": "object",\n"properties": {\n"organization": {\n"title": "Organization",\n"type": "string"\n},\n"inducteeName": {\n"title": "Inductee Name",\n"type": "string"\n},\n"inducteeDeceased": {\n"title": "Inductee Deceased",\n"type": "boolean"\n},\n"deceasedReason": {\n"title": "Deceased Reason",\n"type": "string"\n},\n"coverageDuration": {\n"title": "Coverage Duration",\n"type": "string",\n"format": "date-time"\n},\n"fieldsOfCoverage": {\n"title": "Fields Of Coverage",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"awardsWon": {\n"title": "Awards Won",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": [\n"organization",\n"inducteeName",\n"inducteeDeceased",\n"deceasedReason",\n"coverageDuration",\n"fieldsOfCoverage",\n"awardsWon"\n]\n}', 'role': 'user'}, {'content': '{\n"organization": "U.S. Basketball Writers\' Association",\n"inducteeName": "Tom Kensler",\n"inducteeDeceased": true,\n"deceasedReason": "brain aneurysm",\n"coverageDuration": "decades",\n"fieldsOfCoverage": ["colleges along the Front Range", "golf", "other sports"],\n"awardsWon": ["numerous USBWA writing awards"]\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nGamenauts is an independent game development studio based in Silicon Valley and founded in September 2005 by Stanley Adrianus, formerly of Yahoo! Games.\n\nIn 2012, the company announced a publishing initiative for indie Asian developers, signing up studios from South East Asia such as Kurechii Studio, Menara Games, Artlogic Games and Nerdook Productions to publish their mobile titles for the worldwide market.\n\nGames \n King\'s League: Odyssey (Publisher) - iOS, 2013\n Nuclear Outrun (Publisher) - iOS, 2013 \n Castle Champions (Publisher) - iOS, 2013 \n Ninja Fishing - iOS & Android, 2011\n Stickbound - iOS, 2010\n Wonder Island - PC, 2010\n Cate West: The Velvet Keys - PC, 2009\n Restaurant Rush - PC & Mobile, 2008\n Cate West: The Vanishing Files - PC, DS & Wii, 2008\n Burger Rush PC & Mobile, 2007\n Spacebound - PC, 2006\n\nReferences\n\nExternal links\n Gamenauts Official Website\n\nVideo game companies of the United States\nVideo game development companies\nCasual games\nCompanies based in Silicon Valley\njson schema:\n{\n"title": "GameDevelopmentStudio",\n"type": "object",\n"properties": {\n"name": {"title": "Name", "type": "string"},\n"foundationYear": {"title": "Foundation Year", "type": "number"},\n"founder": {"title": "Founder", "type": "string"},\n"previousCompany": {"title": "Previous Company", "type": "string"},\n"publishingInitiative": {"title": "Publishing Initiative", "type": "object", "properties": {\n"yearAnnounced": {"title": "Year Announced", "type": "number"},\n"studiosSignedUp": {"title": "Studios Signed Up", "type": "array", "items": {"type": "string"}}\n}},\n"gamesPublished": {"title": "Games Published", "type": "array", "items": {\n"type": "object",\n"properties": {\n"gameName": {"title": "Game Name", "type": "string"},\n"platform": {"title": "Platform", "type": "string"},\n"releaseYear": {"title": "Release Year", "type": "number"}\n},\n"required": ["gameName", "platform", "releaseYear"]\n}},\n"gamesDeveloped": {"title": "Games Developed", "type": "array", "items": {\n"type": "object",\n"properties": {\n"gameName": {"title": "Game Name", "type": "string"},\n"platform": {"title": "Platform", "type": "string"},\n"releaseYear": {"title": "Release Year", "type": "number"}\n},\n"required": ["gameName", "platform", "releaseYear"]\n}}\n},\n"required": ["name", "foundationYear", "founder", "gamesPublished", "gamesDeveloped"]\n}', 'role': 'user'}, {'content': '{\n"name": "Gamenauts",\n"foundationYear": 2005,\n"founder": "Stanley Adrianus",\n"previousCompany": "Yahoo! Games",\n"publishingInitiative": {\n"yearAnnounced": 2012,\n"studiosSignedUp": [\n"Kurechii Studio",\n"Menara Games",\n"Artlogic Games",\n"Nerdook Productions"\n]\n},\n"gamesPublished": [\n{\n"gameName": "King\'s League: Odyssey",\n"platform": "iOS",\n"releaseYear": 2013\n},\n{\n"gameName": "Nuclear Outrun",\n"platform": "iOS",\n"releaseYear": 2013\n},\n{\n"gameName": "Castle Champions",\n"platform": "iOS",\n"releaseYear": 2013\n},\n{\n"gameName": "Ninja Fishing",\n"platform": "iOS & Android",\n"releaseYear": 2011\n},\n{\n"gameName": "Stickbound",\n"platform": "iOS",\n"releaseYear": 2010\n},\n{\n"gameName": "Wonder Island",\n"platform": "PC",\n"releaseYear": 2010\n},\n{\n"gameName": "Cate West: The Velvet Keys",\n"platform": "PC",\n"releaseYear": 2009\n},\n{\n"gameName": "Restaurant Rush",\n"platform": "PC & Mobile",\n"releaseYear": 2008\n},\n{\n"gameName": "Cate West: The Vanishing Files",\n"platform": "PC, DS & Wii",\n"releaseYear": 2008\n},\n{\n"gameName": "Burger Rush",\n"platform": "PC & Mobile",\n"releaseYear": 2007\n},\n{\n"gameName": "Spacebound",\n"plat
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{"title": "InjuryReport", "type": "object", "properties": {"game": {"title": "Game", "type": "object", "properties": {"opponent": {"title": "Opponent", "type": "string"}, "venue": {"title": "Venue", "type": "string"}, "kickoffTime": {"title": "Kickoff Time", "type": "string", "format": "date-time"}}, "required": ["opponent", "venue", "kickoffTime"]}, "injuredPlayers": {"title": "Injured Players", "type": "array", "items": {"type": "object", "properties": {"playerName": {"title": "Player Name", "type": "string"}, "injuryType": {"title": "Injury Type", "type": "string"}, "injuryLocation": {"title": "Injury Location", "type": "string"}, "status": {"title": "Status", "type": "string"}}, "required": ["playerName", "injuryType", "injuryLocation", "status"]}}}, "required": ["game", "injuredPlayers"]}\nparagraph:\nAlthough Munster\'s bonus point win against the Bulls was a timely fillip in advance of their clash with Leinster at the Aviva Stadium on Saturday (kick-off 7.35pm), the province are anxiously awaiting scans on a number of players who sustained injuries in that game at Thomond Park last Saturday night.\n\nStephen Archer (ankle), Edwin Edogbo (calf), Craig Casey (groin) and Tadhg Beirne (groin) will all undergo scans and their availability for the match against Leinster will be determined later in the week.\n\nWinger Calvin Nash is also a doubt after he was removed at half-time against the Bulls with a thigh injury.\n\nThis could add to Munster\'s list of outside backs who are currently side lined, and despite Graham Rowntree not ruling out hope of those problems being eased this week, the experienced international trio of Keith Earls (thigh), Andrew Conway (knee) and Simon Zebo (calf) have all again been ruled out.\n\nFurthermore, prop Roman Salanoa suffered a head injury in the final play of the match and will undergo the return to play protocols, while the newly acquired centre Antoine Frisch is rehabbing a calf injury suffered on the Emerging Ireland tour.', 'role': 'user'}, {'content': '{\n "game": {\n "opponent": "Leinster",\n "venue": "Aviva Stadium",\n "kickoffTime": "2023-05-20T19:35:00"\n },\n "injuredPlayers": [\n {\n "playerName": "Stephen Archer",\n "injuryType": "Ankle injury",\n "injuryLocation": "Ankle",\n "status": "Will undergo a scan; availability to be determined later in the week"\n },\n {\n "playerName": "Edwin Edogbo",\n "injuryType": "Calf injury",\n "injuryLocation": "Calf",\n "status": "Will undergo a scan; availability to be determined later in the week"\n },\n {\n "playerName": "Craig Casey",\n "injuryType": "Groin injury",\n "injuryLocation": "Groin",\n "status": "Will undergo a scan; availability to be determined later in the week"\n },\n {\n "playerName": "Tadhg Beirne",\n "injuryType": "Groin injury",\n "injuryLocation": "Groin",\n "status": "Will undergo a scan; availability to be determined later in the week"\n },\n {\n "playerName": "Calvin Nash",\n "injuryType": "Thigh injury",\n "injuryLocation": "Thigh",\n "status": "Removed at half-time against the Bulls; status to be determined"\n },\n {\n "playerName": "Keith Earls",\n "injuryType": "Thigh injury",\n "injuryLocation": "Thigh",\n "status": "Ruled out"\n },\n {\n "playerName": "Andrew Conway",\n "injuryType": "Knee injury",\n "injuryLocation": "Knee",\n "status": "Ruled out"\n },\n {\n "playerName": "Simon Zebo",\n "injuryType": "Calf injury",\n
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nA Super Pink Moon Is Coming in April—And It Will Be This Year\'s Biggest and Brightest\n\nIf you thought the Super Worm Moon in March was impressive, wait until you see April\'s Super Pink Moon—the biggest and brightest moon of the year. Here\'s everything you need to know about the first full moon of spring.\n\nScroll to continue with content\n\nAd\n\nWhy is it called the Pink Moon?\n\nSadly, the name of the moon has nothing to do with its actual color. According to the Farmer\'s Almanac, the full moon is called the Pink Moon because it corresponds with the early spring bloom of phlox subulata, or moss pink, a wildflower native to eastern North America. Instead, it will take on a golden hue near the horizon and fade to bright white.\n\nWhat makes it a supermoon?\n\nApril\'s Super Pink moon is considered a supermoon because of its close proximity to Earth. When the moon reaches the apogee, the point closest to Earth, it appears larger and brighter than a normal full moon. Since the Super Pink Moon will be the closest supermoon to Earth this year, it\'ll be the brightest and biggest full moon of 2020.\n\nWhen can you see the Super Pink Moon?\n\nThe Super Pink Moon will rise on Tuesday, April 7. It\'ll be visible after sunset, and it will reach peak illumination at 10:35 p.m., EDT, according to the Farmer\'s Almanac. The moon will appear full to the naked eye for the next couple of days.\njson schema:\n{\n"title": "SuperPinkMoonInfo",\n"type": "object",\n"properties": {\n"moonName": {\n"title": "Moon Name",\n"type": "string"\n},\n"description": {\n"title": "Description",\n"type": "string"\n},\n"correspondingEvent": {\n"title": "Corresponding Event",\n"type": "string"\n},\n"color": {\n"title": "Color",\n"type": "string"\n},\n"apogeeProximity": {\n"title": "Apogee Proximity",\n"type": "string"\n},\n"appearanceTime": {\n"title": "Appearance Time",\n"type": "string",\n"format": "date-time"\n},\n"additionalInfo": {\n"title": "Additional Info",\n"type": "string"\n}\n},\n"required": ["moonName", "description", "correspondingEvent", "color", "apogeeProximity", "appearanceTime"]\n}', 'role': 'user'}, {'content': '{\n"moonName": "Super Pink Moon",\n"description": "the biggest and brightest moon of the year",\n"correspondingEvent": "the early spring bloom of phlox subulata, or moss pink",\n"color": "It will take on a golden hue near the horizon and fade to bright white.",\n"apogeeProximity": "close proximity to Earth",\n"appearanceTime": "Tuesday, April 7, 10:35 p.m., EDT",\n"additionalInfo": "It\'s considered a supermoon because when the moon reaches the apogee, the point closest to Earth, it appears larger and brighter than a normal full moon."\n}', 'role': 'assistant'}, {'content': 'paragraph:\nGraceREES\n\ndied very peacefully on 19th April 2020 aged 93 years. Retired Nottinghamshire schoolteacher, Sherwood District Councillor and long standing Newark Magistrate. Married to David (who passed away in 2015) for over 60 years, resident of Hockerton for 44 years but since 1999 of Coventry. Much loved mother of Christopher and Jane and devoted grandmother to Isobel, Alistair, Andrew, Laurence and Victoria. Funeral details to be announced. Information via A Pargetter & Son Ltd., Coventry.\n\nNotify Friends and Family\n\nAdd the email addresses of friends or family members you\'d like to notify about this obituary.\n\nTo: *Invalid To Email\n\nFrom: *Invalid From Email\n\nMessage:\n\nOptional\njson schema:\n{\n"title": "Obituary",\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"dateOfDeath": {\n"title": "Date of De
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "JudgeVacancy",\n"type": "object",\n"properties": {\n"vacancyID": {\n"title": "Vacancy ID",\n"type": "string"\n},\n"courtName": {\n"title": "Court Name",\n"type": "string"\n},\n"vacancyDetails": {\n"title": "Vacancy Details",\n"type": "string"\n},\n"nominationDeadline": {\n"title": "Nomination Deadline",\n"type": "string",\n"format": "date-time"\n},\n"nominationMethod": {\n"title": "Nomination Method",\n"type": "object",\n"properties": {\n"email": {\n"title": "Email",\n"type": "string",\n"format": "email"\n},\n"postalMail": {\n"title": "Postal Mail",\n"type": "string"\n}\n},\n"required": ["email", "postalMail"]\n},\n"nominationForm": {\n"title": "Nomination Form",\n"type": "string",\n"format": "uri"\n},\n"eligibilityRequirements": {\n"title": "Eligibility Requirements",\n"type": "object",\n"properties": {\n"minAge": {\n"title": "Minimum Age",\n"type": "number"\n},\n"legalPracticeLicense": {\n"title": "Licensed to Practice Law in Missouri",\n"type": "boolean"\n},\n"usCitizenship": {\n"title": "US Citizenship for at least 15 years",\n"type": "boolean"\n},\n"missouriResidency": {\n"title": "Qualified Missouri Voter for at least nine years",\n"type": "boolean"\n}\n},\n"required": ["minAge", "legalPracticeLicense", "usCitizenship", "missouriResidency"]\n},\n"interviewDetails": {\n"title": "Interview Details",\n"type": "object",\n"properties": {\n"interviewDate": {\n"title": "Interview Date",\n"type": "string",\n"format": "date"\n},\n"interviewLocation": {\n"title": "Interview Location",\n"type": "string"\n},\n"interviewFormat": {\n"title": "Interview Format",\n"type": "string"\n},\n"interviewAccess": {\n"title": "Interview Access",\n"type": "string"\n},\n"applicantInfoRelease": {\n"title": "Applicant Info Release",\n"type": "string"\n}\n},\n"required": ["interviewDate", "interviewLocation", "interviewFormat", "interviewAccess", "applicantInfoRelease"]\n}\n},\n"required": ["vacancyID", "courtName", "vacancyDetails", "nominationDeadline", "nominationMethod", "nominationForm", "eligibilityRequirements", "interviewDetails"]\n}\nparagraph:\nThe Appellate Judicial Commission is accepting applications for judge of the Missouri Court of Appeals, Eastern District, to fill the impending vacancy to be created by the December 31, 2020, retirement of Judge Robert Dowd Jr. after more than four decades of judicial service – the longest of any current sitting judge. You are encouraged to nominate well-qualified candidates for the commission\'s consideration. The Missouri Constitution requires that a judge of the Missouri Court of Appeals be at least 30 years old, licensed to practice law in Missouri, a United States citizen for at least 15 years and a qualified Missouri voter for at least nine years next preceding selection. Nominations should be submitted to the commission by e-mail at EDJudgeVacancy@courts.mo.gov, or by postal mail to Missouri Court of Appeals, Eastern District Judge Vacancy, P.O. Box 150, Jefferson City, MO 65102. The nomination form is available online at https://www.courts.mo.gov/page.jsp?id=168353.\n\nPersons who applied for the Richter vacancy on the Eastern District appeals court who wish to be considered for this vacancy are not required to submit new applications but must notify the commission of their desire to be considered and provide any supplemental materials or new reference letters they wish considered pursuant to the instructions for applicants.\n\nThe commission expects to meet Thursday and Friday, January 14 and 15, 2021, at the Missouri Court of Appeals, Eastern District in the Old Post Office at 815 Olive Str
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nBurnet holding virtual freshmen orientation\n\nIncoming Burnet High School freshmen can learn what it means to be a Bulldog during virtual orientation Aug. 17-19. Photo by Martelle Luedecke/Luedecke Photography\n\nBurnet High School\'s freshmen orientation is online this year due to the COVID-19 pandemic.\n\nThe Class of 2024 can learn what to expect in their first year of high school during the three-day virtual event Aug. 17-19. Each day is themed: Spirit Day on Monday, Pride Day on Tuesday, and Honor Day on Wednesday.\n\nOrientation is via Google Classroom, and incoming freshman will need to join the Class of 2024 Google Classroom. The high school is sending codes to join through parents\' SkyAlert.\n\nStudents can log in to Google Classroom at anytime on those three days and find videos, competitions, and other information. They can also ask questions.\njson schema:\n{\n"title": "VirtualOrientation",\n"type": "object",\n"properties": {\n"schoolName": {"title": "School Name", "type": "string"},\n"classYear": {"title": "Class Year", "type": "string"},\n"orientationType": {"title": "Orientation Type", "type": "string"},\n"dateRange": {"title": "Date Range", "type": "string", "format": "date-time"},\n"daysThemes": {\n"title": "Days Themes",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"day": {"title": "Day", "type": "string"},\n"theme": {"title": "Theme", "type": "string"}\n},\n"required": ["day", "theme"]\n}\n},\n"attendanceMethod": {"title": "Attendance Method", "type": "string"},\n"platform": {"title": "Platform", "type": "string"},\n"accessCode": {"title": "Access Code", "type": "string"},\n"activities": {"title": "Activities", "type": "array", "items": {"type": "string"}}\n},\n"required": ["schoolName", "classYear", "orientationType", "dateRange", "daysThemes", "attendanceMethod", "platform", "accessCode"]\n}', 'role': 'user'}, {'content': '{\n"schoolName": "Burnet High School",\n"classYear": "2024",\n"orientationType": "Virtual",\n"dateRange": "2021-08-17/2021-08-19",\n"daysThemes": [\n{"day": "Monday", "theme": "Spirit Day"},\n{"day": "Tuesday", "theme": "Pride Day"},\n{"day": "Wednesday", "theme": "Honor Day"}\n],\n"attendanceMethod": "Google Classroom",\n"platform": "Google Classroom",\n"accessCode": "The high school is sending codes to join through parents\' SkyAlert.",\n"activities": ["find videos, competitions, and other information", "ask questions"]\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{"title": "EventRegistration", "type": "object", "properties": {"eventName": {"title": "Event Name", "type": "string"}, "startTime": {"title": "Start Time", "type": "string", "format": "time"}, "location": {"title": "Location", "type": "string"}, "ageRange": {"title": "Age Range", "type": "string"}, "challengeName": {"title": "Challenge Name", "type": "string"}, "materials": {"title": "Materials", "type": "array", "items": {"type": "string"}}, "categories": {"title": "Categories", "type": "array", "items": {"type": "string"}}, "registrationDeadline": {"title": "Registration Deadline", "type": "string", "format": "date"}}, "required": ["eventName", "startTime", "location", "ageRange", "challengeName", "materials", "categories"]}\nparagraph:\nFiction to Fashion - Intro Session\n\n6-7pm. Northeast Library. Calling all creative teens (ages 13–18) for our 4th Project Runway-style fashion design challenge. Using unconventional materials like discarded library books, cassette tapes and CDs, we challenge you to make a stunning outfit to send down the runway at Expo 216. You could win "Crowd Favorite," "Most Avant-Garde," "Best Runway Swagger" or "Most Ready to Wear" and take home prizes. Sign up throughout the month of January, intro session is the time to check-in, ask questions, and pick up materials.', 'role': 'user'}, {'content': '{\n "eventName": "Fiction to Fashion - Intro Session",\n "startTime": "18:00",\n "location": "Northeast Library",\n "ageRange": "13-18",\n "challengeName": "Project Runway-style fashion design challenge",\n "materials": ["discarded library books", "cassette tapes", "CDs"],\n "categories": ["Crowd Favorite", "Most Avant-Garde", "Best Runway Swagger", "Most Ready to Wear"],\n "registrationDeadline": "2022-01-31"\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "CrimeReport",\n"type": "object",\n"properties": {\n"crimeLocation": {\n"title": "Crime Location",\n"type": "string"\n},\n"crimeTime": {\n"title": "Crime Time",\n"type": "string",\n"format": "date-time"\n},\n"substanceUsed": {\n"title": "Substance Used",\n"type": "string"\n},\n"victimAge": {\n"title": "Victim Age",\n"type": "number"\n},\n"victimCondition": {\n"title": "Victim Condition",\n"type": "string"\n},\n"witnessAccount": {\n"title": "Witness Account",\n"type": "string"\n},\n"arrestStatus": {\n"title": "Arrest Status",\n"type": "string"\n}\n},\n"required": ["crimeLocation", "crimeTime", "substanceUsed", "victimAge", "victimCondition", "witnessAccount", "arrestStatus"]\n}\nparagraph:\nA teenager was rushed to hospital on Sunday after having a corrosive substance thrown in his eyes in south London.\nPolice were called to Penge Lane, Penge, at around 13.30pm where the 17-year-old was being helped by staff in a nearby restaurant.\nStaff at the Himalayan Kitchen told the Evening Standard the victim screamed "my life is gone" as they tried to wash the liquid from his eyes with bottled water.\nOne worker said: "I was talking to my customers and the man ran up, he was shouting \'my life is gone, call my mum\'."\nThe employee added: "It\'s terrifying, very scary. It is not a nice thing to happen, so cruel."\nA spokesman for the Met said the substance thrown has yet to be analysed but is believed to be ammonia.\nThe victim was taken to a south London Hospital where his condition is not life threatening.\nNo arrests have been made and inquiries continue. (Image: Google Streetview)', 'role': 'user'}, {'content': '{\n"crimeLocation": "Penge Lane, Penge, south London",\n"crimeTime": "Sunday",\n"substanceUsed": "ammonia",\n"victimAge": 17,\n"victimCondition": "not life threatening",\n"witnessAccount": "A man ran up, he was shouting \'my life is gone, call my mum\'. It\'s terrifying, very scary. It is not a nice thing to happen, so cruel.",\n"arrestStatus": "No arrests have been made"\n}', 'role': 'assistant'}, {'content': 'paragraph:\nThe Blood Bank of Delmarva has declared its second blood emergency in three months.\n\nOfficials say blood supplies have dropped to "critical lows" after donations declined and demand increased during Labor Day weekend. The blood bank\'s reserves are less than 50 percent of what they need to be.\n\nSince donated blood only lasts seven days, the supply constantly needs to be replenished.\n\nThe blood bank needs 350 donors every day — 62,000 a year — to make sure the 19 hospitals in the Delmarva Peninsula are supplied to deal with emergencies.\n\nIn late June, the blood bank said it was in a blood emergency because it only had a two-day supply for all blood types — far below the safe minimum.\n\nThe organization\'s officials are asking for all types of blood, especially O negative. This kind of blood is needed for trauma situations and emergency rooms, since it\'s considered to be universal.\njson schema:\n{"title": "BloodEmergency", "type": "object", "properties": {"bloodBankName": {"title": "Blood Bank Name", "type": "string"}, "emergencyStatus": {"title": "Emergency Status", "type": "string"}, "bloodSupplyLevel": {"title": "Blood Supply Level", "type": "number"}, "neededDonorsPerDay": {"title": "Needed Donors Per Day", "type": "number"}, "neededDonorsPerYear": {"title": "Needed Donors Per Year", "type": "number"}, "lastBloodEmergency": {"title": "Last Blood Emergency", "type": "string", "format": "date"}, "bloodTypesInNeed": {"title": "Blood Types In Need", "type": "array", "items": {"type": "string"
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{"title": "RiderInfo", "type": "object", "properties": {"riderName": {"title": "Rider Name", "type": "string"}, "age": {"title": "Age", "type": "number"}, "nationality": {"title": "Nationality", "type": "string"}, "team": {"title": "Team", "type": "string"}, "teamLevel": {"title": "Team Level", "type": "string"}, "events": {"title": "Events", "type": "array", "items": {"type": "string"}}, "trainingProgram": {"title": "Training Program", "type": "string"}, "goals": {"title": "Goals", "type": "string"}, "experience": {"title": "Experience", "type": "string"}, "topRanking": {"title": "Top Ranking", "type": "string"}, "teamLeader": {"title": "Team Leader", "type": "string"}}, "required": ["riderName", "age", "nationality", "team", "teamLevel", "trainingProgram", "goals", "experience", "topRanking", "teamLeader"]}\nparagraph:\n21-year-old B.C. rider added to an international roster of riders for 2021\n\nCanadian enduro racer Evan Wall is stepping into the elite ranks as part of the expanded Orbea Fox Enduro Team in 2021. The 21-year-old rider will be part of the program\'s "National" level, focusing mainly on events within North America.\n\nOrbea Fox Enduro Team\n\n"I am prepping thoroughly with training programs that help me gain confidence and improve my technique. This is the foundation for a strong season," said Wall of his plans ahead of the season. After years of strong results on the Canadian Enduro Series, Wall gained wider recognition after winning the Pinkbike Academy\'s first season in 2020.\n\nExpanded Orbea Fox Enduro Team\n\nEntering its fourth season, the Orbea Fox Enduro Team is expanding its roster. While Evan Wall joins at the national level, Laura Charles joins as an Enduro World Series-level rider.\n\n"It\'s a new challenge, and I\'m certain this experience will help me improve," said Charles. "For my part, I want to contribute to the team by qualifying for a Top 5. That would be incredible."\n\nThe international team will be led once again by Damien Oton, consistently at the top of the EWS\'s overall rankings, and Vid Persak.', 'role': 'user'}, {'content': '{\n "riderName": "Evan Wall",\n "age": 21,\n "nationality": "Canadian",\n "team": "Orbea Fox Enduro Team",\n "teamLevel": "National",\n "events": ["events within North America"],\n "trainingProgram": "training programs that help gain confidence and improve technique",\n "goals": "a strong season",\n "experience": "years of strong results on the Canadian Enduro Series, winner of the Pinkbike Academy\'s first season in 2020",\n "topRanking": "N/A",\n "teamLeader": "Damien Oton"\n}', 'role': 'assistant'}, {'content': 'paragraph:\nKirk Mitchell is a general assignment reporter at The Denver Post who focuses on criminal justice stories. He began working at the newspaper in 1998, after writing for newspapers in Mesa, Ariz., and Twin Falls, Idaho, and The Associated Press in Salt Lake City. Mitchell first started writing the Cold Case blog in Fall 2007, in part because Colorado has more than 1,400 unsolved homicides.\n\nMore in Crime & Courts\n\nThe 10th U.S. Circuit Court of Appeals in Denver on Friday upheld a lower court\'s injunction that prohibits Fort Collins from enforcing a public nudity ban that criminalizes the act of women going topless in public.\n\nDenver oil tycoon Jack Grynberg is trying to convince an Arapahoe County district judge that the octogenarian\'s family owes him as much as $400 million for the lifetime of work he\'s done building a billion-dollar empire for them. A jury last week found for the family in its lawsuit to unseat him.\njson schema
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nOlimpia\'s only loss came at the hand of Peñarol in the middle of the draws. Peñarol got out to a 1-0 lead in the first half with a stoppage-time goal from Juan Ramos.\n\nAfter that, Peñarol came out of half and scored again in the 49th minute with a goal from Federico Carrizo. Olimpia was able to answer one goal from Walter González, but it wasn\'t able to get the game-tying goal.\n\nPeñarol is just one spot above Olimpia in the Group G rankings. The club is 1-0-2 with a total of three points and a negative-one goal differential.\n\nOutside of the team\'s win against Olimpia, Peñarol lost to Colón 2-1 and then to Cerro Porteño 1-0 in its most recent game.\njson schema:\n{"title": "MatchResult", "type": "object", "properties": {"teams": {"title": "Teams", "type": "array", "items": {"type": "object", "properties": {"name": {"title": "Team Name", "type": "string"}, "scores": {"title": "Scores", "type": "array", "items": {"type": "number"}}, "goals": {"title": "Goals", "type": "array", "items": {"type": "object", "properties": {"player": {"title": "Player Name", "type": "string"}, "time": {"title": "Goal Time", "type": "string", "format": "time"}}, "required": ["player", "time"]}}}, "required": ["name", "scores", "goals"]}}, "rankings": {"title": "Rankings", "type": "object", "properties": {"position": {"title": "Position", "type": "number"}, "record": {"title": "Record", "type": "string"}, "points": {"title": "Points", "type": "number"}, "goalDifferential": {"title": "Goal Differential", "type": "number"}}, "required": ["position", "record", "points", "goalDifferential"]}}, "required": ["teams", "rankings"]}', 'role': 'user'}, {'content': '{\n "teams": [\n {\n "name": "Peñarol",\n "scores": [2],\n "goals": [\n {\n "player": "Juan Ramos",\n "time": "45+00"\n },\n {\n "player": "Federico Carrizo",\n "time": "49:00"\n }\n ]\n },\n {\n "name": "Olimpia",\n "scores": [1],\n "goals": [\n {\n "player": "Walter González",\n "time": "xx:xx"\n }\n ]\n }\n ],\n "rankings": {\n "position": 2,\n "record": "1-0-2",\n "points": 3,\n "goalDifferential": -1\n }\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nHypatima spathota is a moth in the family Gelechiidae. It was described by Edward Meyrick in 1913. It is found in Japan, Taiwan, India, Sri Lanka, China, Vietnam and Australia, where it has been recorded from Queensland.\n\nThe wingspan is 15–17\xa0mm. The forewings are dark purple fuscous longitudinally streaked with black and with a dark brown streak above the fold from the base to four-fifths of the wing and a rather broad ochreous-whitish streak along the dorsum from the base to the tornus, then attenuated almost to the apex, including short blackish dashes on each side of the tornus, its upper edge with a short ochreous-whitish dash adjacent at two-fifths, and a similar but transverse mark at two-thirds. There is a fine white longitudinal line above the apical portion. The hindwings are fuscous, paler and thinly scaled anteriorly, with the veins and termen suffused with darker.\n\nThe larvae feed on Mangifera indica and Lannea grandis.\n\nReferences\n\nHypatima\nTaxa named by Edward Meyrick\nMoths described in 1913\njson schema:\n{\n"title": "MothDescription",\n"type": "object",\n"properties": {\n"scientificName": {\n"title": "Scientific Name",\n"type": "string"\n},\n"family": {\n"title": "Family",\n"type": "string"\n},\n"describedBy": {\n"title": "Described By",\n"type": "string"\n},\n"yearDescribed": {\n"title": "Year Described",\n"type": "string",\n"format": "date-year"\n},\n"distribution": {\n"title": "Distribution",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"wingspan": {\n"title": "Wingspan",\n"type": "string"\n},\n"forewingColor": {\n"title": "Forewing Color",\n"type": "string"\n},\n"hindwingColor": {\n"title": "Hindwing Color",\n"type": "string"\n},\n"larvaeFood": {\n"title": "Larvae Food",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"references": {\n"title": "References",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n}\n}', 'role': 'user'}, {'content': '{\n"scientificName": "Hypatima spathota",\n"family": "Gelechiidae",\n"describedBy": "Edward Meyrick",\n"yearDescribed": "1913",\n"distribution": ["Japan", "Taiwan", "India", "Sri Lanka", "China", "Vietnam", "Australia (Queensland)"],\n"wingspan": "15-17 mm",\n"forewingColor": "dark purple fuscous longitudinally streaked with black, with a dark brown streak above the fold and a rather broad ochreous-whitish streak along the dorsum",\n"hindwingColor": "fuscous, paler and thinly scaled anteriorly, with the veins and termen suffused with darker",\n"larvaeFood": ["Mangifera indica", "Lannea grandis"],\n"references": ["Hypatima", "Taxa named by Edward Meyrick", "Moths described in 1913"]\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "EmployeeOfTheYear",\n"type": "object",\n"properties": {\n"employeeName": {"title": "Employee Name", "type": "string"},\n"hometown": {"title": "Hometown", "type": "string"},\n"degree1Major": {"title": "Degree 1 Major", "type": "string"},\n"degree1School": {"title": "Degree 1 School", "type": "string"},\n"degree2Major": {"title": "Degree 2 Major", "type": "string", "format": "nursing"},\n"degree2School": {"title": "Degree 2 School", "type": "string", "format": "university_of_scranton"},\n"yearsAtCompany": {"title": "Years at Company", "type": "number"},\n"department": {"title": "Department", "type": "string"},\n"position": {"title": "Position", "type": "string"},\n"spouseName": {"title": "Spouse Name", "type": "string"},\n"children": {\n"title": "Children",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"name": {"title": "Name", "type": "string"},\n"age": {"title": "Age", "type": "nu
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "ArtShowcase",\n"type": "object",\n"properties": {\n"eventName": {\n"title": "Event Name",\n"type": "string"\n},\n"eventTheme": {\n"title": "Event Theme",\n"type": "string"\n},\n"eventLocation": {\n"title": "Event Location",\n"type": "string"\n},\n"eventDateRange": {\n"title": "Event Date Range",\n"type": "string",\n"format": "date-time"\n},\n"artists": {\n"title": "Artists",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"artistName": {\n"title": "Artist Name",\n"type": "string"\n},\n"artworkName": {\n"title": "Artwork Name",\n"type": "string"\n},\n"artworkType": {\n"title": "Artwork Type",\n"type": "string"\n},\n"artworkDescription": {\n"title": "Artwork Description",\n"type": "string"\n},\n"technique": {\n"title": "Technique",\n"type": "string"\n}\n},\n"required": ["artistName", "artworkName", "artworkType", "artworkDescription", "technique"]\n}\n},\n"sponsors": {\n"title": "Sponsors",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"sponsorName": {\n"title": "Sponsor Name",\n"type": "string"\n},\n"sponsorRole": {\n"title": "Sponsor Role",\n"type": "string"\n}\n},\n"required": ["sponsorName", "sponsorRole"]\n}\n},\n"benefits": {\n"title": "Benefits",\n"type": "object",\n"properties": {\n"artistGrowth": {\n"title": "Artist Growth",\n"type": "string"\n},\n"communityEngagement": {\n"title": "Community Engagement",\n"type": "string"\n}\n},\n"required": ["artistGrowth", "communityEngagement"]\n}\n}\n}\nparagraph:\nLocal artists showcase their artwork to the world\n\nThe Department of Sport, Arts and Culture (DSAC) has named three local artists whose artworks will be showcased during Biennale Arte 2022.\n\nThe festival, which is often described as the Olympics of the art world, will be held in Venice, Italy, from 23 April to 27 November 2022.\n\nThis year\'s festival theme is \'The Milk of Dreams\'.\n\nJohannesburg-based visual artist Phumulani Ntuli, photographer Roger Ballen and visual artist and photographer Lebohang Kganye will exhibit their respective creations – Godide, Theatre of the Apparitions, and B(l)ack to Fairy Tale – at the South African pavilion, which has been sub-themed \'Into the Light\'.\n\nAccording to the department, the theme reflects the richness of creative expression that is unlocked when the artist combines fantasy and re-imagination into their self-identity.\n\n"For his Theatre of the Apparitions showcase, Ballen developed his own technique, coating glass in special paints, etching off the coating and backlighting the glass. Kganye photographed herself autobiographically as the protagonist in westernised fairy tales," says the department.\n\nNtuli\'s work utilises stop-motion animation.\n\nStop motion is an animated filmmaking technique in which objects are manipulated.\n\nSince 2011, the department has been providing monetary sponsorship to the participating artists. This is part of the department\'s mandate to provide market access to artists, says DSAC Director-General Vusumuzi Mkhize.\n\nReaching greater heights\n\nHe says the festival is an opportunity for him to engage with a diverse international art community that includes collectors, art institutions, galleries and the general public.\n\nNtuli adds that it is also a chance to share both old and current South African art, narratives and history with others.\n\nThe backing of the department has allowed him to expand the scope of his work, he says. "I have been planning to expand as an art entrepreneur and have been unable to do this in my own capacity. With the assistance of the department, such business e
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "GuestLectureInformation",\n"type": "object",\n"properties": {\n"guestName": {"title": "Guest Name", "type": "string"},\n"eventName": {"title": "Event Name", "type": "string"},\n"date": {"title": "Date", "type": "string", "format": "date"},\n"location": {"title": "Location", "type": "string"},\n"topic": {"title": "Topic", "type": "string"},\n"participationType": {"title": "Participation Type", "type": "string"},\n"previousExperience": {\n"title": "Previous Experience",\n"type": "object",\n"properties": {\n"guestLectures": {"title": "Guest Lectures", "type": "string"},\n"seminars": {"title": "Seminars", "type": "string"},\n"associationLeadership": {"title": "Association Leadership", "type": "string"},\n"mciInspection": {"title": "MCI Inspection", "type": "string"}\n},\n"required": ["guestLectures", "seminars", "associationLeadership", "mciInspection"]\n}\n},\n"required": ["guestName", "eventName", "date", "location", "topic", "participationType", "previousExperience"]\n}\nparagraph:\nCity Dermatologist to deliver guest lecture at Pune\n\nJanuary 21, 2020\n\nDr. H. Hanumanthappa, Senior Consultant Dermatologist and former Director and Dean; Head, Department of Dermatology, K.R. Hospital, Mysuru, has been invited to deliver a guest lecture on "Diseases Modifying Anti-Rheumatic Drugs in the Treatment of Connective Tissue Disorders" and to participate in a Panel Discussion on "Skin Manifestations of Rheumatic Diseases" during the All India National Conference of Dermatologists (Dermacon- 2020) at Pune from Jan. 30 to Feb. 2. He has delivered guest lectures in National and International Conferences earlier. He has also participated in many Seminars and served as National President of Paediatric Dermatologists\' Association and as MCI Inspector of India.\n\nRecent News\n\nABOUT\n\nMysuru\'s favorite and largest circulated English evening daily has kept the citizens of Mysuru informed and entertained since 1978. Over the past 41 years, Star of Mysore has been the newspaper that Mysureans reach for every evening to know about the happenings in Mysuru city. The newspaper has feature rich articles and dedicated pages targeted at readers across the demographic spectrum of Mysuru city. With a readership of over 2,50,000 Star of Mysore has been the best connection between it\'s readers and their leaders; between advertisers and customers; between Mysuru and Mysureans.', 'role': 'user'}, {'content': '{\n"guestName": "Dr. H. Hanumanthappa",\n"eventName": "All India National Conference of Dermatologists (Dermacon- 2020)",\n"date": "2020-01-30",\n"location": "Pune",\n"topic": "Diseases Modifying Anti-Rheumatic Drugs in the Treatment of Connective Tissue Disorders",\n"participationType": "Guest Lecture and Panel Discussion",\n"previousExperience": {\n"guestLectures": "Yes, in National and International Conferences",\n"seminars": "Not specified",\n"associationLeadership": "Yes, as National President of Paediatric Dermatologists\' Association",\n"mciInspection": "Yes, as MCI Inspector of India"\n}\n}', 'role': 'assistant'}, {'content': 'json schema:\n{"title": "LakeInformation", "type": "object", "properties": {"lakeName": {"title": "Lake Name", "type": "string"}, "location": {"title": "Location", "type": "object", "properties": {"country": {"title": "Country", "type": "string"}, "province": {"title": "Province", "type": "string"}, "nearbyCommunity": {"title": "Nearby Community", "type": "string"}}, "required": ["country", "province", "nearbyCommunity"]}, "features": {"title": "Features", "type": "array", "items": {"type": "object", "properties": {"
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\n"The idea of calling our children, not that these names are wrong or bad, Daniel or Thomas or Samuel doesn\'t sit right with us. It isn\'t us.\n\n"My in-laws are furious. One: because they\'re not traditional names, two: do not bear any affiliation to family members; my fiancé is named after his grandfather, his sister is named after his aunt etc, and three: because they\'re mythical creatures from other cultures.\n\n"That is apparently what they dislike the most. But we really love the names. We already have a girls name picked out too, Valkyrie Lee to pay homage to my Danish grandparents that cared for me when I was a teen.\n\n"We actually came up with this name before I even got pregnant. We like that all the names match in that they are powerful beings that can fly. We want to make our children feel powerful too."\n\nAfter people advised her to change the names, the woman admitted she wasn\'t worried about her kids being bullied.\n\nMORE ON\n\nWe use your sign-up to provide content in ways you\'ve consented to and to improve our understanding of you. This may include adverts from us and 3rd parties based on our understanding. You can unsubscribe at any time. More info\n\nWe use your sign-up to provide content in ways you\'ve consented to and to improve our understanding of you. This may include adverts from us and 3rd parties based on our understanding. You can unsubscribe at any time. More Info.\njson schema:\n{\n"title": "NamingPreferences",\n"type": "object",\n"properties": {\n"preferenceName": {"title": "Preference Name", "type": "string"},\n"names": {\n"title": "Names",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"name": {"title": "Name", "type": "string"},\n"reason": {"title": "Reason", "type": "string"}\n}\n}\n},\n"objections": {\n"title": "Objections",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"response": {\n"title": "Response",\n"type": "string"\n}\n},\n"required": ["preferenceName", "names", "objections", "response"]\n}', 'role': 'user'}, {'content': '{\n"preferenceName": "Names for the children",\n"names": [\n{\n"name": "Not Daniel or Thomas or Samuel",\n"reason": "They don\'t sit right with us, it isn\'t us and they are powerful beings that can fly"\n},\n{\n"name": "Valkyrie Lee",\n"reason": "To pay homage to her Danish grandparents"\n}\n],\n"objections": [\n"Not traditional names",\n"Do not bear any affiliation to family members",\n"Names are mythical creatures from other cultures",\n"Apparently what they dislike the most"\n],\n"response": "After people advised her to change the names, the woman admitted she wasn\'t worried about her kids being bullied."\n}', 'role': 'assistant'}, {'content': 'paragraph:\nInter Milan are all set to bring Lukaku back to the club this summer on a loan deal.\n\nThey are currently deep in negotiations with Chelsea over Lukaku\'s signature and a deal is on the verge of being fully agreed, with the Belgium player expected to take his medicals in Milan at some time next week.\n\nLukaku netted 64 goals in 95 appearances for the Serie A giants in his first two years in Italy.\n\nSpeaking on the sidelines of the Sportsman of the Year award ceremony, Immobile said, "I don\'t think about records. I\'ll certainly realise the extent of what I\'ve done in the world of football once I stop playing. It is clear that we must not rest or give up.\n\n"With Lukaku back in Italy, the level of our league is raised, I\'ll have to continually do my best. Now I only have the next objectives in my head."\njson schema:\n{"title": "TransferNews", "type": "object", "properties": {"tea
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nI\'ve seen two different Minutemen launches from Vandenburg AFB, from Las Vegas. Both of them took place right at sunset, on absolutely clear days. They were both backlit by the sun just below the horizon. I could see them actually appear over the horizon, rise out of the atmosphere, and drop their first stages. The next day, the local paper had front-page photos of the smoke trails. Apparently, if it\'s clear enough, you can see them from a long distance away.\njson schema:\n{\n"title": "MinutemanLaunch",\n"type": "object",\n"properties": {\n"viewingLocation": {"title": "Viewing Location", "type": "string"},\n"numberOfLaunches": {"title": "Number of Launches", "type": "number"},\n"launchTime": {"title": "Launch Time", "type": "string", "format": "date-time"},\n"weatherCondition": {"title": "Weather Condition", "type": "string"},\n"abilityToSeeStages": {"title": "Ability to See Stages", "type": "string"},\n"evidenceOfLaunch": {"title": "Evidence of Launch", "type": "string"},\n"distanceFromLaunchSite": {"title": "Distance from Launch Site", "type": "string"}\n},\n"required": ["viewingLocation", "numberOfLaunches", "launchTime", "weatherCondition", "abilityToSeeStages", "evidenceOfLaunch"]\n}', 'role': 'user'}, {'content': '{\n"viewingLocation": "Las Vegas",\n"numberOfLaunches": 2,\n"launchTime": "Sunset",\n"weatherCondition": "Absolutely clear days",\n"abilityToSeeStages": "Able to see them actually appear over the horizon, rise out of the atmosphere, and drop their first stages",\n"evidenceOfLaunch": "Front-page photos of the smoke trails in the local paper",\n"distanceFromLaunchSite": "A long distance away"\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "BarInfo",\n"type": "object",\n"properties": {\n"barName": {\n"title": "Bar Name",\n"type": "string"\n},\n"location": {\n"title": "Location",\n"type": "string"\n},\n"openingHours": {\n"title": "Opening Hours",\n"type": "object",\n"properties": {\n"Thursday": {\n"type": "string"\n},\n"Friday": {\n"type": "string"\n},\n"Saturday": {\n"type": "string"\n}\n},\n"required": ["Thursday", "Friday", "Saturday"]\n},\n"recommendedReservation": {\n"title": "Recommended Reservation",\n"type": "boolean"\n}\n},\n"required": ["barName", "location", "openingHours", "recommendedReservation"]\n}\nparagraph:\nNacarat Terrace, the bar at the Fairmont Queen Elizabeth Hotel, has announced they will be reopening today, Aug. 20. The beautiful terrasse with great views of downtown Montreal is known for its cocktails, and was voted in the Best Hotel Bar or Restaurant category in this year\'s Best of MTL. Nacarat Terrace will be open from 4 to 10 p.m. on Thursday and Friday, and 2 to 8 p.m. on Saturday. The bar recommends that you make a reservation prior to arrival.\n\nCult MTL is a daily web and monthly print publication focusing on Montreal culture, music, arts, news, politics and city life. Along with interviewing local and visiting artists, promoters and entrepreneurs, we feature analysis and criticism, curate a daily To-Do List and run the annual Best of MTL readers poll.', 'role': 'user'}, {'content': '{\n"barName": "Nacarat Terrace",\n"location": "Fairmont Queen Elizabeth Hotel, Montreal",\n"openingHours": {\n"Thursday": "4 to 10 p.m.",\n"Friday": "4 to 10 p.m.",\n"Saturday": "2 to 8 p.m."\n},\n"recommendedReservation": true\n}', 'role': 'assistant'}].
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{"title": "HorseProfile", "type": "object", "properties": {"horseName": {"title": "Horse Name", "type": "string"}, "age": {"title": "Age", "type": "number"}, "country": {"title": "Country", "type": "string"}, "trainer": {"title": "Trainer", "type": "string"}, "base": {"title": "Base", "type": "string"}, "sire": {"title": "Sire", "type": "string"}, "dam": {"title": "Dam", "type": "string"}, "wins": {"title": "Wins", "type": "number"}, "raceDetails": {"title": "Race Details", "type": "array", "items": {"type": "object", "properties": {"raceDate": {"title": "Race Date", "type": "string", "format": "date"}, "jockey": {"title": "Jockey", "type": "string"}, "prizeMoney": {"title": "Prize Money", "type": "number"}, "opponent": {"title": "Opponent", "type": "string"}, "result": {"title": "Result", "type": "string"}}, "required": ["raceDate", "jockey", "prizeMoney", "opponent", "result"]}}}, "required": ["horseName", "age", "country", "trainer", "base", "sire", "dam", "wins", "raceDetails"]}\nparagraph:\nAsk Nile is a 9yo b Unknown from Ireland trained by S Neville, who is based at Wexford. It is sired by the stallion Ask out of the dam Spirit Of The Nile.\n\nAsk Nile has managed to win 3 races in its career so far. On 14th May 2017 at Killarney, Ask Nile was ridden by B Hayes and scored its most significant win to date, getting the money in the $8,924 Irish Stallion Farms Ebf Beginners Chase, defeating Gwencily Berbas.', 'role': 'user'}, {'content': '{\n "horseName": "Ask Nile",\n "age": 9,\n "country": "Ireland",\n "trainer": "S Neville",\n "base": "Wexford",\n "sire": "Ask",\n "dam": "Spirit Of The Nile",\n "wins": 3,\n "raceDetails": [\n {\n "raceDate": "2017-05-14",\n "jockey": "B Hayes",\n "prizeMoney": 8924,\n "opponent": "Gwencily Berbas",\n "result": "defeated"\n }\n ]\n}', 'role': 'assistant'}, {'content': 'paragraph:\nPost Malone & Lizzo Get Honored With Giant Murals\n\nSo, is everyone getting a mural now? Recently Post Malone AND Lizzo were both honored by having their pictures beautifully painted on the side of buildings. That\'s how you know you\'ve made it!\n\nPosty, who is a Texas native, now has a very colorful mural in East Dallas. If you\'re traveling to Dallas and want to try finding the mural\'s exact location, it\'s at the intersection of Columbia Avenue and North Carroll Avenue. The artist of this masterpiece is Michael McPheeters. I think he nailed the face tats, but if Posty gets more, will he update the mural? Check it out his work below!\n\nIf you\'re trying to take a pic with Lizzo\'s mural, hate to break it to you, but you\'re gonna have to go all the way to Ireland! This bright orange mural is located in Dublin, specifically at the corner of Chatham Row and South William Street. How awesome is this?\njson schema:\n{"title": "MuralHonor", "type": "object", "properties": {"celebrityName": {"title": "Celebrity Name", "type": "string"}, "honorType": {"title": "Honor Type", "type": "string", "enum": ["Mural"]}, "location": {"title": "Location", "type": "object", "properties": {"address": {"title": "Address", "type": "string"}, "city": {"title": "City", "type": "string"}, "country": {"title": "Country", "type": "string"}}, "required": ["address", "city", "country"]}, "artist": {"title": "Artist", "type": "string"}, "imageUrl": {"title": "Image URL", "type": "string", "format": "uri"}}, "required": ["celebrityName", "honorType", "location", "artist"]}', 'role': 'user'}, {'content': '{\n "celebrityName": "Post Malone",\n "honorType": "Mural",\n "location": {\n "address": "intersect
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{"title": "ManufacturingSector", "type": "object", "properties": {"topic": {"title": "Topic", "type": "string"}, "factors": {"title": "Factors", "type": "array", "items": {"type": "object", "properties": {"factorName": {"title": "Factor Name", "type": "string"}, "description": {"title": "Description", "type": "string"}, "examples": {"title": "Examples", "type": "array", "items": {"type": "string"}}}, "required": ["factorName", "description", "examples"]}}, "reasons": {"title": "Reasons", "type": "array", "items": {"type": "object", "properties": {"reasonNumber": {"title": "Reason Number", "type": "integer"}, "description": {"title": "Description", "type": "string"}, "examples": {"title": "Examples", "type": "array", "items": {"type": "string"}}}, "required": ["reasonNumber", "description", "examples"]}}}, "required": ["topic", "factors", "reasons"]}\nparagraph:\nIn recent decades, technology, globalization and Asia\'s rise have all caused significant changes in the manufacturing sector. At least within the confines of Asian and ASEAN countries, this has often meant a race to the bottom for competitiveness based on price, rather than value. With Singapore arguably being better known for financial services and tax incentives, you might assume it would have no future in the global manufacturing sector.\n\nI\'m pleased to say, that\'s a wholly incorrect assumption.\n\nWhile low-cost and repetitive-task human labor has had its heyday, Singapore has shown itself to be a lighthouse of forward-thinking. By fully embracing Industry 4.0, Singapore is moving up the value chain in order to consolidate its position as the world\'s leading industrial hub.\n\nSuch a lofty ambition takes immense proactivity of course, so here are four reasons why Singapore will be the world\'s leading industrial hub.\n\n1. The Singapore government\'s foresight\n\nSingapore has a long history of showing immense foresight into the needs of the nation. The most famous example of this was perhaps the government\'s establishment of the Housing and Development Board (HDB) in 1960, designed to lift the population out of poverty and into their own homes.\n\nIt\'s that culture of foresight that has led to many other progressive insights, including the government\'s recent pledge to grow its manufacturing sector over the next 10 years. Crucially though, this is not a scattergun strategy to build factories and import workers to manufacture low-quality goods.\n\nThe government has robust plans in place to be competitive not in terms of cost, but based on the intellectual property the country can generate. This means rather than low-value manufacturing, Singapore will aim to manufacture in the higher end of the value chain, including sectors like pharmaceuticals, fintech and biotechnology.\n\nSingapore\'s three-pronged legislation for this growth is detailed and well thought out, making provisions for attracting top-tier companies, creating higher-paying manufacturing jobs, and training the local population with extensive tertiary education collaborations.\n\n2. A world-class ecosystem for manufacturing\n\nWhile the government\'s recent but detailed plan for a high-value manufacturing future inspires confidence, you only have to look at how Singapore is performing today to see an upwards trajectory. Singapore is already the world\'s third-largest exporter of high-tech goods, produces four of the world\'s top 10 drugs, and is the fourth-largest producer of refined oil. It\'s little surprise then, that industry giants like Micron, Shell and Merck have not only chosen Singapore as a manuf
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nAmazon India has refused to appear before Joint Committee of Parliament on Data Protection Bill on October 28, BJP MP Meenakshi Lekhi said on Friday.\n\nThe joint committee led by Lekhi has 20 members from the Lok Sabha and 10 from the Rajya Sabha.\n\nLekhi told PTI that Amazon\'s refusal to appear before Parliament committee amounts to breach of privilege.\n\nParliament committee unanimous on action by government against Amazon for refusing to appear, she said.\n\nMeanwhile, the committe also issued summons to Facebook in order to seek its "oral evidence" on the issues of data protection and privacy. Twitter too is said to have been directed for appearance next week.\n\nHowever, the summon to Facebook and Twitter is strictly pertaining to the issue of citizen\'s personal data protection.\n\nThe summons to Twitter assumes significance as these come close on the Centre\'s letter to the microblogging site\'s chief Jack Dorsey.\n\nTaking strong exception to the "misrepresentation" of India\'s map, the government wrote a stern letter to the Twitter CEO, saying that any attempt by the platform to disrespect the sovereignty and integrity of India, which is also reflected by the maps, was totally unacceptable.\n\nLast month, amid severe criticism of Facebook and Union government over the alleged collusion of Facebook India executive Ankhi Das and the BJP, Union Information Technology Minister Ravi Shankar Prasad wrote a hardhitting letter to Facebook CEO Mark Zuckerberg, blaming the Facebook India management of alleged bias against people supporting the right-of-centre ideology.\n\n(With ANI inputs)\n\n(To download our E-paper please click here. The publishers permit sharing of the paper\'s PDF on WhatsApp and other social media platforms.)\njson schema:\n{\n"title": "CommitteeMeeting",\n"type": "object",\n"properties": {\n"companyName": {\n"title": "Company Name",\n"type": "string"\n},\n"refusal": {\n"title": "Refusal",\n"type": "boolean"\n},\n"committeeDetails": {\n"title": "Committee Details",\n"type": "object",\n"properties": {\n"committeeName": {\n"title": "Committee Name",\n"type": "string"\n},\n"totalMembers": {\n"title": "Total Members",\n"type": "number"\n},\n"lokSabhaMembers": {\n"title": "Lok Sabha Members",\n"type": "number"\n},\n"rajyaSabhaMembers": {\n"title": "Rajya Sabha Members",\n"type": "number"\n}\n},\n"required": ["committeeName", "totalMembers", "lokSabhaMembers", "rajyaSabhaMembers"]\n},\n"privilegeBreach": {\n"title": "Privilege Breach",\n"type": "boolean"\n},\n"unanimousAction": {\n"title": "Unanimous Action",\n"type": "string"\n},\n"summonedCompanies": {\n"title": "Summoned Companies",\n"type": "array",\n"items": {\n"type": "object",\n"properties": {\n"companyName": {\n"title": "Company Name",\n"type": "string"\n},\n"summonReason": {\n"title": "Summon Reason",\n"type": "string"\n}\n},\n"required": ["companyName", "summonReason"]\n}\n},\n"letterToCEO": {\n"title": "Letter to CEO",\n"type": "object",\n"properties": {\n"letterContent": {\n"title": "Letter Content",\n"type": "string"\n},\n"exceptionTaken": {\n"title": "Exception Taken",\n"type": "boolean"\n},\n"letterSubject": {\n"title": "Letter Subject",\n"type": "string"\n}\n},\n"required": ["letterContent", "exceptionTaken", "letterSubject"]\n}\n},\n"required": ["companyName", "refusal", "committeeDetails", "privilegeBreach", "unanimousAction", "summonedCompanies", "letterToCEO"]\n}', 'role': 'user'}, {'content': '{\n"companyName": "Amazon",\n"refusal": true,\n"committeeDetails": {\n"committeeName": "Joint Committee of Parliament on Data Protection Bill",\n"totalMembers": 3
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'json schema:\n{\n"title": "ArenaRenovation",\n"type": "object",\n"properties": {\n"renovationID": {\n"title": "Renovation ID",\n"type": "string"\n},\n"renovationFeature": {\n"title": "Renovation Feature",\n"type": "string"\n},\n"location": {\n"title": "Location",\n"type": "string"\n},\n"openingDate": {\n"title": "Opening Date",\n"type": "string",\n"format": "date"\n},\n"size": {\n"title": "Size",\n"type": "number"\n},\n"amenities": {\n"title": "Amenities",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"description": {\n"title": "Description",\n"type": "string"\n}\n},\n"required": ["renovationID", "renovationFeature", "location", "openingDate", "size", "amenities", "description"]\n}\nparagraph:\nQuestion of the Day\n\nWas Mueller nudging Congress to impeach Trump\n\nMonumental Sports and Entertainment has announced a new premium area for Washington Wizards VIP season ticket holders called "The District," a courtside patio area near one baseline at Capital One Arena.\n\nStarting with the 2019-20 season, the patio will be open at all Wizards games. Clocking in at more than 1,000 square feet, it will be the largest such standing-room area in any NBA arena, the Wizards say.\n\n.@MSE has announced the addition of \'The District,\' a courtside patio for all VIP ticket holders beginning with the 2019-20 season. https://t.co/DC8SSj6RjH\n\n"Washington, D.C. is one of the most connected cities in the world and the courtside patio at Capital One Arena will be an ideal location for business development and networking opportunities and perfect for fans who want to socialize at a unique space that serves as a literal extension of the hardwood and our terrific in-venue VIP experiences," Jim Van Stone, Monumental\'s CCO and president of business operations, said in a news release.\n\nThe District will also include four rows of "VIP seating" near the Wizards\' bench.\n\nMonumental began an ambitious renovation project to Capital One Arena last summer between NBA seasons. The new patio, which will include a bar, is part of the second phase of the project. Monumental will debut a new scoreboard with a 360-degree continuous video screen next fall, as well.\n\nLast summer, the $40 million first phase included the addition of a new sound system, DJ stage, 3D projection system and more digital signage.', 'role': 'user'}, {'content': '{\n"renovationID": "Phase2Renovation",\n"renovationFeature": "Courtside patio",\n"location": "Capital One Arena, Washington D.C.",\n"openingDate": "2019-10-01",\n"size": 1000,\n"amenities": ["Bar", "VIP seating", "Business development opportunities", "Networking space"],\n"description": "A new premium area for Washington Wizards VIP season ticket holders called \'The District\', located near one baseline. It is the largest standing-room area in any NBA arena and will include a bar and four rows of VIP seating near the Wizards\' bench."\n}', 'role': 'assistant'}, {'content': 'paragraph:\n[Main Updates]\n\n\'Expedition\' will enable players to temporarily travel to other servers each day from 21:00 ~ 01:00.\n\nRegionalClan List\n\nClan List will now show clans from all adjacent servers within the same regional group.\n\nMagic Square / Secret Peak - Floor expansion\n\n7th Floor will be added in Magic Square and Secret Peak\n\nLong-Term inactive Clans will be deleted.\n\n◈DRACO Updates◈\n\nHYDRA Payment System\n\nNew HYDRA Payment option will be available through the MIR4\'s official PC version\n\n※ Portal / Raids will be restricted from 30 minutes before the maintenance.\n※ Maintenance schedule may change depending on the circumstanc
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nAction related nav\n\nAdvertising\n\nDolphins Add Local Product, Wide Receiver Ricardo Louis\n\nAlain Poupart\n\nLead Writer\n\nThe Dolphins added a local product to their wide receiver corps Monday when they signed Miami Beach High School alum Ricardo Louis.\n\nLouis joins the Dolphins after playing two seasons with the Cleveland Browns after being a fourth-round pick out of Auburn in the 2016 NFL draft. Louis has 45 career receptions: 18 in 2016 and 27 in 2017.\n\nLouis\' biggest play at Auburn became known as the "Prayer at Jordan-Hare." It was a 73-yard touchdown catch with 25 seconds remaining that gave Auburn a 43-38 victory against Georgia and helped them reach the National Championship game against Florida State.\njson schema:\n{\n"title": "NFLPlayerSigning",\n"type": "object",\n"properties": {\n"teamName": {\n"title": "Team Name",\n"type": "string"\n},\n"newPlayerName": {\n"title": "New Player Name",\n"type": "string"\n},\n"playerPosition": {\n"title": "Player Position",\n"type": "string"\n},\n"pastTeam": {\n"title": "Past Team",\n"type": "string"\n},\n"draftRound": {\n"title": "Draft Round",\n"type": "number"\n},\n"numberOfReceptions": {\n"title": "Number of Receptions",\n"type": "number"\n},\n"memorablePlay": {\n"title": "Memorable Play",\n"type": "object",\n"properties": {\n"playName": {\n"title": "Play Name",\n"type": "string"\n},\n"yardage": {\n"title": "Yardage",\n"type": "number"\n},\n"timeRemaining": {\n"title": "Time Remaining",\n"type": "string",\n"format": "time"\n},\n"result": {\n"title": "Result",\n"type": "string"\n}\n},\n"required": ["playName", "yardage", "timeRemaining", "result"]\n}\n},\n"required": ["teamName", "newPlayerName", "playerPosition", "pastTeam", "draftRound", "numberOfReceptions", "memorablePlay"]\n}', 'role': 'user'}, {'content': '{\n"teamName": "Dolphins",\n"newPlayerName": "Ricardo Louis",\n"playerPosition": "wide receiver",\n"pastTeam": "Cleveland Browns",\n"draftRound": 4,\n"numberOfReceptions": 45,\n"memorablePlay": {\n"playName": "The Prayer at Jordan-Hare",\n"yardage": 73,\n"timeRemaining": "25 seconds",\n"result": "Touchdown"\n}\n}', 'role': 'assistant'}, {'content': 'json schema:\n{\n"title": "Person",\n"type": "object",\n"properties": {\n"name": {\n"title": "Name",\n"type": "string"\n},\n"birthYear": {\n"title": "Year of Birth",\n"type": "string",\n"format": "date-time",\n"format": "year"\n},\n"deathYear": {\n"title": "Year of Death",\n"type": "string",\n"format": "date-time",\n"format": "year"\n},\n"occupation": {\n"title": "Occupation",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"politicalCareer": {\n"title": "Political Career",\n"type": "object",\n"properties": {\n"positionsHeld": {\n"title": "Positions Held",\n"type": "array",\n"items": {\n"type": "string"\n}\n},\n"yearsActive": {\n"title": "Years Active",\n"type": "string",\n"format": "date-time",\n"format": "year"\n}\n},\n"required": ["positionsHeld", "yearsActive"]\n},\n"family": {\n"title": "Family",\n"type": "object",\n"properties": {\n"familyMembers": {\n"title": "Family Members",\n"type": "array",\n"items": {\n"type": "string"\n}\n}\n},\n"required": ["familyMembers"]\n}\n},\n"required": ["name", "birthYear", "deathYear", "occupation", "politicalCareer", "family"]\n}\nparagraph:\nWilliam Talbot (ca 1792 – April 28, 1873) was a farmer, fisher and politician in Newfoundland. He served in the Newfoundland House of Assembly.\n\nTalbot spent 59 years in Newfoundland and owned a farm in Harbour Grace. He was elected to the assembly for Conception Bay in 1852 and then was elected for Harbour Main in 1855. Around 1859, he ret
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Invalid role tag in [{'content': 'You are a helpful assistant. Your task is to carefully read the provided paragraph and extract relevant information that corresponds to the keys in the given JSON schema. Once you have identified the necessary information, please populate the JSON schema with the extracted data. In your response, refrain from uttering any extraneous words.', 'role': 'system'}, {'content': 'paragraph:\nThis is how many Afghan refugees Lancashire\'s districts have offered to help so far\n\nLancashire\'s district councils are assessing how many Afghan refugees they can help under a scheme designed specifically to resettle people from the country who helped UK armed forces during the mission in their homeland – with pledges to help 30 families so far made.\n\nThe government has announced that the UK intends to welcome up to 20,000 Afghans over a five-year period – with 5,000 of them set to arrive before the end of the year.\n\nThat commitment was made by ministers following the Taliban takeover over the country this month.\n\nHowever, the UK had already established a programme to offer refuge to those who could be under threat because of their roles in assisting the army over the past two decades.\n\nInternally displaced Afghan families, who fled from Kunduz and Takhar province due to battles between Taliban and Afghan security forces, eat their lunch at the Shahr-e-Naw Park in Kabul on 10th August, 2021\n\nLancashire County Council, which is co-ordinating the region\'s response to the Afghan Locally-Employed Staff Relocation Scheme, says that it has so far received no details from the government about the resettlement of ordinary civilians as part of the UK\'s wider commitment.\n\nCabinet member for community and cultural services Peter Buckley said: "Councils across Lancashire have been asked to make a pledge about the number of families they can help, and so far confirmation has been received from Lancaster, Wyre, South Ribble and West Lancashire.\n\n"Between them, they will be able to house 30 families, with our first family having arrived in Lancaster a few weeks ago.\n\n"We are also set to hear back shortly from the county\'s other district and borough councils – and expect this number to grow.\n\n"When the families arrive, the Home Office will provide funding to help support the families while they integrate over the next 12 months," County Cllr Buckley added.\n\nThe Local Democracy Reporting Service understands that South Ribble has offered to help 10 families and is currently searching for suitable accommodation, while a previous commitment by Chorley to provide the same number of places for refugees in general could now be used for Afghans.\n\nWyre Council told the LDRS that it was not at the stage of agreeing numbers as yet, but added:\n\n"Over the last four years, Wyre Council has been pleased to offer support to 11 Syrian refugee families who we\'ve helped to resettle in the borough. We are currently looking to extend our support of the Government\'s UK Resettlement scheme and provide support for up to five further Syrian households every two years.\n\n"Additionally we are now looking to see how we can support government Afghan relocation schemes as the situation develops. "Our priority is to ensure that any relocated families are safe and settled and do what we can to help them become part of the local community."\n\nThe government\'s Afghan relocations and assistance policy (ARAP) began back in April, as a replacement for previous resettlement arrangements for those who have helped UK forces over the past two decades. It offers three routes to the UK – including urgent relocation, by priority, to those assessed to be at risk of serious threat to life due to their work with the UK. That channel is open to all current or former staff employed directly by the UK government in Afghanistan since 2001, regardless of their role, job, length of service or reason for leaving.\n\nAlso available is relocation, by default, to a spec
|
|||
|
|
[WARNING|2026-05-24 05:49:03] llamafactory.data.converter:149 >> Skipping this abnormal example.
|
|||
|
|
Converting format of dataset (num_proc=16): 100%|██████████| 10000/10000 [00:00<00:00, 17341.09 examples/s]
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 0%| | 0/56000 [00:00<?, ? examples/s]Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Repo card metadata block was not found. Setting CardData to empty.
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 2%|▏ | 1000/56000 [00:02<02:25, 376.82 examples/s]
Running tokenizer on dataset (num_proc=16): 4%|▎ | 2000/56000 [00:03<01:26, 623.00 examples/s]
Running tokenizer on dataset (num_proc=16): 5%|▌ | 3000/56000 [00:03<00:50, 1050.98 examples/s]
Running tokenizer on dataset (num_proc=16): 7%|▋ | 4000/56000 [00:04<00:45, 1152.59 examples/s]
Running tokenizer on dataset (num_proc=16): 9%|▉ | 5000/56000 [00:04<00:32, 1588.14 examples/s]
Running tokenizer on dataset (num_proc=16): 12%|█▏ | 6500/56000 [00:04<00:21, 2330.09 examples/s]
Running tokenizer on dataset (num_proc=16): 13%|█▎ | 7500/56000 [00:05<00:24, 2010.26 examples/s]
Running tokenizer on dataset (num_proc=16): 18%|█▊ | 10000/56000 [00:06<00:15, 2936.84 examples/s]
Running tokenizer on dataset (num_proc=16): 20%|█▉ | 11000/56000 [00:06<00:15, 2821.85 examples/s]
Running tokenizer on dataset (num_proc=16): 23%|██▎ | 13000/56000 [00:06<00:10, 4023.04 examples/s]
Running tokenizer on dataset (num_proc=16): 26%|██▌ | 14500/56000 [00:07<00:13, 3124.91 examples/s]
Running tokenizer on dataset (num_proc=16): 28%|██▊ | 15500/56000 [00:07<00:12, 3235.69 examples/s]
Running tokenizer on dataset (num_proc=16): 29%|██▉ | 16500/56000 [00:07<00:10, 3718.29 examples/s]
Running tokenizer on dataset (num_proc=16): 32%|███▏ | 18000/56000 [00:08<00:13, 2814.57 examples/s]
Running tokenizer on dataset (num_proc=16): 34%|███▍ | 19000/56000 [00:08<00:11, 3099.69 examples/s]
Running tokenizer on dataset (num_proc=16): 37%|███▋ | 20500/56000 [00:09<00:09, 3582.26 examples/s]
Running tokenizer on dataset (num_proc=16): 38%|███▊ | 21500/56000 [00:09<00:12, 2706.74 examples/s]
Running tokenizer on dataset (num_proc=16): 40%|████ | 22500/56000 [00:09<00:10, 3273.76 examples/s]
Running tokenizer on dataset (num_proc=16): 44%|████▍ | 24500/56000 [00:09<00:06, 4784.65 examples/s]
Running tokenizer on dataset (num_proc=16): 46%|████▌ | 25500/56000 [00:10<00:06, 4431.97 examples/s]
Running tokenizer on dataset (num_proc=16): 48%|████▊ | 27000/56000 [00:10<00:05, 5579.47 examples/s]
Running tokenizer on dataset (num_proc=16): 50%|█████ | 28000/56000 [00:10<00:04, 5731.64 examples/s]
Running tokenizer on dataset (num_proc=16): 52%|█████▏ | 29000/56000 [00:10<00:04, 6384.43 examples/s]
Running tokenizer on dataset (num_proc=16): 54%|█████▎ | 30000/56000 [00:10<00:04, 6192.05 examples/s]
Running tokenizer on dataset (num_proc=16): 57%|█████▋ | 32000/56000 [00:10<00:03, 7944.22 examples/s]
Running tokenizer on dataset (num_proc=16): 59%|█████▉ | 33000/56000 [00:11<00:03, 6382.16 examples/s]
Running tokenizer on dataset (num_proc=16): 62%|██████▏ | 34500/56000 [00:11<00:03, 6002.76 examples/s]
Running tokenizer on dataset (num_proc=16): 64%|██████▍ | 36000/56000 [00:11<00:04, 4803.82 examples/s]
Running tokenizer on dataset (num_proc=16): 66%|██████▌ | 37000/56000 [00:12<00:05, 3175.78 examples/s]
Running tokenizer on dataset (num_proc=16): 68%|██████▊ | 38000/56000 [00:13<00:07, 2432.61 examples/s]
Running tokenizer on dataset (num_proc=16): 71%|███████ | 39500/56000 [00:13<00:06, 2507.57 examples/s]
Running tokenizer on dataset (num_proc=16): 72%|███████▏ | 40500/56000 [00:13<00:05, 3071.83 examples/s]
Running tokenizer on dataset (num_proc=16): 74%|███████▍ | 41500/56000 [00:14<00:05, 2793.47 examples/s]
Running tokenizer on dataset (num_proc=16): 76%|███████▌ | 42500/56000 [00:14<00:04, 3113.48 examples/s]
Running tokenizer on dataset (num_proc=16): 77%|███████▋ | 43000/56000 [00:14<00:04, 2694.83 examples/s]
Running tokenizer on dataset (num_proc=16): 7
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 80%|████████ | 45000/56000 [00:15<00:04, 2206.02 examples/s]
Running tokenizer on dataset (num_proc=16): 81%|████████▏ | 45500/56000 [00:16<00:05, 2031.35 examples/s][WARNING|2026-05-24 05:49:21] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:22] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:22] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:22] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:22] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:22] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:22] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 83%|████████▎ | 46500/56000 [00:17<00:08, 1184.62 examples/s][WARNING|2026-05-24 05:49:23] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:23] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:23] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:23] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:23] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:24] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:24] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:24] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:24] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 85%|████████▍ | 47500/56000 [00:19<00:10, 779.36 examples/s] [WARNING|2026-05-24 05:49:25] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:25] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:25] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:25] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 87%|████████▋ | 48500/56000 [00:20<00:08, 881.85 examples/s][WARNING|2026-05-24 05:49:26] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:26] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:26] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:27] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:27] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:27] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:28] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:28] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:28] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 88%|████████▊ | 49500/56000 [00:23<00:10, 648.76 examples/s][WARNING|2026-05-24 05:49:28] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:28] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:28] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:28] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:29] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 90%|█████████ | 50500/56000 [00:24<00:07, 700.19 examples/s][WARNING|2026-05-24 05:49:29] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:29] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 92%|█████████▏| 51500/56000 [00:25<00:05, 831.52 examples/s][WARNING|2026-05-24 05:49:30] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:30] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:30] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:30] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:30] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:30] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:31] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:31] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:31] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:31] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:31] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:32] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:32] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:32] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:32] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:33] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 94%|█████████▍| 52500/56000 [00:28<00:06, 521.22 examples/s][WARNING|2026-05-24 05:49:33] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 96%|█████████▌| 53500/56000 [00:28<00:03, 727.68 examples/s][WARNING|2026-05-24 05:49:33] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:34] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:34] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:34] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:34] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:34] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 97%|█████████▋| 54500/56000 [00:29<00:01, 869.25 examples/s][WARNING|2026-05-24 05:49:34] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:34] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:35] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:35] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:35] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:35] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:36] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
[WARNING|2026-05-24 05:49:36] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 98%|█████████▊| 55000/56000 [00:30<00:01, 660.00 examples/s][WARNING|2026-05-24 05:49:36] llamafactory.data.processor.supervised:149 >> Dropped invalid example: []
|
|||
|
|
Running tokenizer on dataset (num_proc=16): 99%|█████████▉| 55500/56000 [00:31<00:00, 751.65 examples/s]
Running tokenizer on dataset (num_proc=16): 100%|██████████| 56000/56000 [00:31<00:00, 870.56 examples/s]
Running tokenizer on dataset (num_proc=16): 100%|██████████| 56000/56000 [00:34<00:00, 1637.69 examples/s]
|
|||
|
|
training example:
|
|||
|
|
input_ids:
|
|||
|
|
[151644, 872, 198, 2121, 264, 3908, 6240, 30660, 27061, 553, 11044, 2926, 24370, 11, 1855, 264, 18580, 6358, 315, 24931, 8699, 975, 10186, 304, 264, 5099, 27835, 2813, 13, 29279, 697, 6358, 5646, 4236, 1376, 7567, 323, 2326, 4650, 11513, 315, 8699, 975, 13, 98477, 279, 4244, 330, 58242, 1, 323, 330, 9752, 2022, 3263, 576, 6358, 1265, 387, 1378, 42643, 1293, 13, 26642, 697, 14613, 304, 264, 4718, 3561, 448, 1817, 14311, 438, 264, 8651, 4343, 13, 151645, 198, 151644, 77091, 198, 73594, 2236, 198, 515, 220, 330, 27727, 62, 16, 788, 330, 62980, 287, 8699, 975, 10186, 304, 264, 5099, 27835, 2813, 646, 4446, 911, 12114, 7567, 429, 5285, 448, 6481, 6240, 16170, 13, 5512, 11, 8699, 975, 646, 11941, 18379, 9364, 23689, 553, 10004, 24177, 304, 975, 36769, 11, 10693, 7775, 311, 8172, 6584, 323, 4345, 41256, 803, 13444, 13, 1096, 24177, 3545, 3059, 304, 7172, 25148, 11, 438, 8256, 525, 2952, 311, 975, 2337, 862, 16162, 5068, 4115, 2041, 279, 76449, 315, 264, 8606, 5163, 4573, 13, 10440, 11, 5110, 646, 11075, 23839, 2783, 18986, 553, 17719, 279, 1184, 369, 6961, 5163, 3550, 11, 14376, 1495, 389, 18024, 5435, 311, 8016, 11, 29419, 11, 323, 5163, 16720, 13, 20861, 11, 8699, 975, 10186, 1787, 705, 279, 10772, 7314, 11, 27362, 279, 23134, 315, 14988, 7775, 504, 16807, 52901, 10468, 11, 8450, 85544, 264, 803, 16807, 323, 28308, 30927, 13, 35074, 11, 8699, 975, 646, 16792, 311, 264, 38245, 975, 25843, 17590, 11, 892, 646, 18472, 8231, 411, 323, 47539, 7813, 11, 22136, 1293, 9663, 9364, 37131, 13, 17375, 11, 553, 17719, 279, 1184, 369, 7298, 93850, 11, 8699, 975, 11554, 12158, 40229, 11, 43549, 279, 2813, 594, 12499, 42872, 323, 5285, 287, 448, 41588, 21896, 2562, 2750, 10346, 2303, 220, 330, 27727, 62, 17, 788, 330, 19810, 1493, 7567, 11, 1052, 525, 4650, 11513, 429, 1969, 387, 20068, 979, 24931, 8699, 975, 10186, 13, 3776, 3598, 8645, 374, 279, 5214, 315, 53082, 2083, 95293, 323, 20031, 11, 438, 8699, 975, 646, 2990, 311, 15650, 315, 30298, 323, 99077, 4221, 8256, 11, 892, 1231, 5421, 2083, 29195, 323, 10535, 13, 10440, 11, 1052, 374, 279, 4265, 315, 20337, 12966, 975, 5068, 323, 38142, 26, 19680, 1184, 311, 10515, 862, 14830, 311, 5978, 429, 8256, 7146, 26664, 323, 26236, 2041, 2118, 37117, 13, 17375, 11, 279, 6407, 311, 8699, 975, 1231, 4441, 19652, 4344, 304, 2813, 7674, 11, 438, 8606, 2750, 323, 12378, 1231, 1184, 311, 387, 312, 5655, 831, 12852, 311, 4946, 264, 803, 47963, 975, 1614, 13, 2014, 22279, 1493, 11513, 11, 432, 374, 16587, 369, 5110, 311, 2725, 304, 4752, 264, 3746, 4108, 3942, 323, 311, 3410, 2797, 16665, 323, 1824, 369, 8699, 8256, 10040, 532, 73594, 151645, 198]
|
|||
|
|
inputs:
|
|||
|
|
<|im_start|>user
|
|||
|
|
As a young management consultant influenced by Peter Drucker, create a strategic analysis of implementing remote work policies in a mid-sized company. Ensure your analysis includes five key benefits and three potential challenges of remote work. Exclude the words "technology" and "globalization". The analysis should be two paragraphs long. Present your findings in a JSON format with each paragraph as a separate entry.<|im_end|>
|
|||
|
|
<|im_start|>assistant
|
|||
|
|
```json
|
|||
|
|
{
|
|||
|
|
"paragraph_1": "Implementing remote work policies in a mid-sized company can bring about numerous benefits that align with modern management principles. First, remote work can significantly enhance employee satisfaction by offering flexibility in work schedules, allowing individuals to balance professional and personal commitments more effectively. This flexibility often results in increased productivity, as employees are able to work during their peak performance hours without the distractions of a traditional office environment. Second, companies can achieve considerable cost savings by reducing the need for physical office space, cutting down on expenses related to rent, utilities, and office supplies. Third, remote work policies open up the talent pool, enabling the hiring of qualified individuals from diverse geographical locations, thus fostering a more diverse and inclusive workforce. Fourth, remote work can contribute to a healthier work-life integration, which can decrease burnout and turnover rates, promoting long-term employee retention. Finally, by reducing the need for daily commuting, remote work supports environmental sustainability, decreasing the company's carbon footprint and aligning with eco-friendly business values.",
|
|||
|
|
|
|||
|
|
"paragraph_2": "Despite these benefits, there are potential challenges that must be addressed when implementing remote work policies. One major challenge is the risk of diminished team cohesion and collaboration, as remote work can lead to feelings of isolation and detachment among employees, which may impact team dynamics and communication. Second, there is the issue of maintaining consistent work performance and accountability; managers need to adapt their strategies to ensure that employees remain motivated and productive without direct supervision. Finally, the shift to remote work may necessitate changes in company culture, as traditional values and practices may need to be re-evaluated to fit a more decentralized work model. To overcome these challenges, it is crucial for companies to invest in building a strong virtual community and to provide clear expectations and support for remote employees."
|
|||
|
|
}
|
|||
|
|
```<|im_end|>
|
|||
|
|
|
|||
|
|
label_ids:
|
|||
|
|
[-100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, 73594, 2236, 198, 515, 220, 330, 27727, 62, 16, 788, 330, 62980, 287, 8699, 975, 10186, 304, 264, 5099, 27835, 2813, 646, 4446, 911, 12114, 7567, 429, 5285, 448, 6481, 6240, 16170, 13, 5512, 11, 8699, 975, 646, 11941, 18379, 9364, 23689, 553, 10004, 24177, 304, 975, 36769, 11, 10693, 7775, 311, 8172, 6584, 323, 4345, 41256, 803, 13444, 13, 1096, 24177, 3545, 3059, 304, 7172, 25148, 11, 438, 8256, 525, 2952, 311, 975, 2337, 862, 16162, 5068, 4115, 2041, 279, 76449, 315, 264, 8606, 5163, 4573, 13, 10440, 11, 5110, 646, 11075, 23839, 2783, 18986, 553, 17719, 279, 1184, 369, 6961, 5163, 3550, 11, 14376, 1495, 389, 18024, 5435, 311, 8016, 11, 29419, 11, 323, 5163, 16720, 13, 20861, 11, 8699, 975, 10186, 1787, 705, 279, 10772, 7314, 11, 27362, 279, 23134, 315, 14988, 7775, 504, 16807, 52901, 10468, 11, 8450, 85544, 264, 803, 16807, 323, 28308, 30927, 13, 35074, 11, 8699, 975, 646, 16792, 311, 264, 38245, 975, 25843, 17590, 11, 892, 646, 18472, 8231, 411, 323, 47539, 7813, 11, 22136, 1293, 9663, 9364, 37131, 13, 17375, 11, 553, 17719, 279, 1184, 369, 7298, 93850, 11, 8699, 975, 11554, 12158, 40229, 11, 43549, 279, 2813, 594, 12499, 42872, 323, 5285, 287, 448, 41588, 21896, 2562, 2750, 10346, 2303, 220, 330, 27727, 62, 17, 788, 330, 19810, 1493, 7567, 11, 1052, 525, 4650, 11513, 429, 1969, 387, 20068, 979, 24931, 8699, 975, 10186, 13, 3776, 3598, 8645, 374, 279, 5214, 315, 53082, 2083, 95293, 323, 20031, 11, 438, 8699, 975, 646, 2990, 311, 15650, 315, 30298, 323, 99077, 4221, 8256, 11, 892, 1231, 5421, 2083, 29195, 323, 10535, 13, 10440, 11, 1052, 374, 279, 4265, 315, 20337, 12966, 975, 5068, 323, 38142, 26, 19680, 1184, 311, 10515, 862, 14830, 311, 5978, 429, 8256, 7146, 26664, 323, 26236, 2041, 2118, 37117, 13, 17375, 11, 279, 6407, 311, 8699, 975, 1231, 4441, 19652, 4344, 304, 2813, 7674, 11, 438, 8606, 2750, 323, 12378, 1231, 1184, 311, 387, 312, 5655, 831, 12852, 311, 4946, 264, 803, 47963, 975, 1614, 13, 2014, 22279, 1493, 11513, 11, 432, 374, 16587, 369, 5110, 311, 2725, 304, 4752, 264, 3746, 4108, 3942, 323, 311, 3410, 2797, 16665, 323, 1824, 369, 8699, 8256, 10040, 532, 73594, 151645, 198]
|
|||
|
|
labels:
|
|||
|
|
```json
|
|||
|
|
{
|
|||
|
|
"paragraph_1": "Implementing remote work policies in a mid-sized company can bring about numerous benefits that align with modern management principles. First, remote work can significantly enhance employee satisfaction by offering flexibility in work schedules, allowing individuals to balance professional and personal commitments more effectively. This flexibility often results in increased productivity, as employees are able to work during their peak performance hours without the distractions of a traditional office environment. Second, companies can achieve considerable cost savings by reducing the need for physical office space, cutting down on expenses related to rent, utilities, and office supplies. Third, remote work policies open up the talent pool, enabling the hiring of qualified individuals from diverse geographical locations, thus fostering a more diverse and inclusive workforce. Fourth, remote work can contribute to a healthier work-life integration, which can decrease burnout and turnover rates, promoting long-term employee retention. Finally, by reducing the need for daily commuting, remote work supports environmental sustainability, decreasing the company's carbon footprint and aligning with eco-friendly business values.",
|
|||
|
|
|
|||
|
|
"paragraph_2": "Despite these benefits, there are potential challenges that must be addressed when implementing remote work policies. One major challenge is the risk of diminished team cohesion and collaboration, as remote work can lead to feelings of isolation and detachment among employees, which may impact team dynamics and communication. Second, there is the issue of maintaining consistent work performance and accountability; managers need to adapt their strategies to ensure that employees remain motivated and productive without direct supervision. Finally, the shift to remote work may necessitate changes in company culture, as traditional values and practices may need to be re-evaluated to fit a more decentralized work model. To overcome these challenges, it is crucial for companies to invest in building a strong virtual community and to provide clear expectations and support for remote employees."
|
|||
|
|
}
|
|||
|
|
```<|im_end|>
|
|||
|
|
|
|||
|
|
[INFO|configuration_utils.py:670] 2026-05-24 05:49:39,374 >> loading configuration file config.json from cache at /home/ubuntu/.cache/huggingface/hub/models--Qwen--Qwen3-1.7B/snapshots/70d244cc86ccca08cf5af4e1e306ecf908b1ad5e/config.json
|
|||
|
|
[INFO|configuration_utils.py:742] 2026-05-24 05:49:39,375 >> Model config Qwen3Config {
|
|||
|
|
"architectures": [
|
|||
|
|
"Qwen3ForCausalLM"
|
|||
|
|
],
|
|||
|
|
"attention_bias": false,
|
|||
|
|
"attention_dropout": 0.0,
|
|||
|
|
"bos_token_id": 151643,
|
|||
|
|
"dtype": "bfloat16",
|
|||
|
|
"eos_token_id": 151645,
|
|||
|
|
"head_dim": 128,
|
|||
|
|
"hidden_act": "silu",
|
|||
|
|
"hidden_size": 2048,
|
|||
|
|
"initializer_range": 0.02,
|
|||
|
|
"intermediate_size": 6144,
|
|||
|
|
"layer_types": [
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention",
|
|||
|
|
"full_attention"
|
|||
|
|
],
|
|||
|
|
"masked_layers": null,
|
|||
|
|
"max_position_embeddings": 40960,
|
|||
|
|
"max_window_layers": 28,
|
|||
|
|
"model_type": "qwen3",
|
|||
|
|
"num_attention_heads": 16,
|
|||
|
|
"num_hidden_layers": 28,
|
|||
|
|
"num_key_value_heads": 8,
|
|||
|
|
"pad_token_id": null,
|
|||
|
|
"rms_norm_eps": 1e-06,
|
|||
|
|
"rope_parameters": {
|
|||
|
|
"rope_theta": 1000000,
|
|||
|
|
"rope_type": "default"
|
|||
|
|
},
|
|||
|
|
"sliding_window": null,
|
|||
|
|
"sparsity_attn": null,
|
|||
|
|
"sparsity_mlp": null,
|
|||
|
|
"subnet_mode": null,
|
|||
|
|
"subnet_type": null,
|
|||
|
|
"threshold_attn": null,
|
|||
|
|
"threshold_mlp": null,
|
|||
|
|
"tie_word_embeddings": true,
|
|||
|
|
"transformers_version": "5.2.0",
|
|||
|
|
"use_cache": true,
|
|||
|
|
"use_sliding_window": false,
|
|||
|
|
"vocab_size": 151936
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
[INFO|2026-05-24 05:49:39] llamafactory.model.model_utils.kv_cache:144 >> KV cache is disabled during training.
|
|||
|
|
[INFO|modeling_utils.py:710] 2026-05-24 05:49:39,705 >> loading weights file model.safetensors from cache at /home/ubuntu/.cache/huggingface/hub/models--Qwen--Qwen3-1.7B/snapshots/70d244cc86ccca08cf5af4e1e306ecf908b1ad5e/model.safetensors.index.json
|
|||
|
|
Fetching 2 files: 0%| | 0/2 [00:00<?, ?it/s]
Fetching 2 files: 0%| | 0/2 [00:00<?, ?it/s]
Fetching 2 files: 0%| | 0/2 [00:00<?, ?it/s]
Fetching 2 files: 0%| | 0/2 [00:00<?, ?it/s]
Fetching 2 files: 0%| | 0/2 [00:00<?, ?it/s]
Fetching 2 files: 0%| | 0/2 [00:00<?, ?it/s]
Fetching 2 files: 0%| | 0/2 [00:00<?, ?it/s]
Fetching 2 files: 0%| | 0/2 [00:00<?, ?it/s]
Fetching 2 files: 50%|█████ | 1/2 [00:41<00:41, 41.65s/it]
Fetching 2 files: 100%|██████████| 2/2 [00:41<00:00, 20.83s/it]
|
|||
|
|
[INFO|modeling_utils.py:779] 2026-05-24 05:50:21,427 >> Will use dtype=torch.bfloat16 as defined in model's config object
|
|||
|
|
[INFO|configuration_utils.py:1014] 2026-05-24 05:50:21,428 >> Generate config GenerationConfig {
|
|||
|
|
"bos_token_id": 151643,
|
|||
|
|
"eos_token_id": 151645,
|
|||
|
|
"output_attentions": false,
|
|||
|
|
"output_hidden_states": false,
|
|||
|
|
"use_cache": false
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
Fetching 2 files: 50%|█████ | 1/2 [00:40<00:40, 40.73s/it]
Fetching 2 files: 100%|██████████| 2/2 [00:40<00:00, 20.36s/it]
|
|||
|
|
Fetching 2 files: 50%|█████ | 1/2 [00:40<00:40, 40.68s/it]
Fetching 2 files: 100%|██████████| 2/2 [00:40<00:00, 20.34s/it]
|
|||
|
|
Fetching 2 files: 50%|█████ | 1/2 [00:40<00:40, 40.78s/it]
Fetching 2 files: 100%|██████████| 2/2 [00:40<00:00, 20.39s/it]
|
|||
|
|
Fetching 2 files: 50%|█████ | 1/2 [00:40<00:40, 40.59s/it]
Fetching 2 files: 50%|█████ | 1/2 [00:40<00:40, 40.73s/it]
Fetching 2 files: 100%|██████████| 2/2 [00:40<00:00, 20.30s/it]
|
|||
|
|
Fetching 2 files: 100%|██████████| 2/2 [00:40<00:00, 20.36s/it]
|
|||
|
|
Fetching 2 files: 50%|█████ | 1/2 [00:40<00:40, 40.78s/it]
Fetching 2 files: 100%|██████████| 2/2 [00:40<00:00, 20.39s/it]
|
|||
|
|
Fetching 2 files: 50%|█████ | 1/2 [00:40<00:40, 40.78s/it]
Fetching 2 files: 100%|██████████| 2/2 [00:40<00:00, 20.39s/it]
|
|||
|
|
Loading weights: 0%| | 0/311 [00:00<?, ?it/s]
Loading weights: 0%| | 1/311 [00:00<00:00, 6875.91it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 1/311 [00:00<00:00, 4766.25it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 0/311 [00:00<?, ?it/s]
Loading weights: 0%| | 0/311 [00:00<?, ?it/s]
Loading weights: 0%| | 1/311 [00:00<00:00, 6668.21it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 1/311 [00:00<00:00, 6355.01it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 1/311 [00:00<00:00, 4588.95it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 1/311 [00:00<00:00, 4495.50it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 0/311 [00:00<?, ?it/s]
Loading weights: 0%| | 0/311 [00:00<?, ?it/s]
Loading weights: 0%| | 0/311 [00:00<?, ?it/s]
Loading weights: 0%| | 1/311 [00:00<00:00, 6052.39it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 1/311 [00:00<00:00, 4194.30it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 1/311 [00:00<00:00, 6432.98it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 1/311 [00:00<00:00, 4387.35it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 1/311 [00:00<00:00, 6615.62it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 1/311 [00:00<00:00, 4534.38it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 0/311 [00:00<?, ?it/s]
Loading weights: 0%| | 1/311 [00:00<00:00, 4554.08it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 1/311 [00:00<00:00, 3404.47it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 0/311 [00:00<?, ?it/s]
Loading weights: 0%| | 1/311 [00:00<00:00, 5555.37it/s, Materializing param=lm_head.weight]
Loading weights: 0%| | 1/311 [00:00<00:00, 3942.02it/s, Materializing param=lm_head.weight]
Loading weights: 1%| | 2/311 [00:00<00:12, 25.73it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/311 [00:00<00:11, 25.78it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/311 [00:00<00:12, 25.73it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/311 [00:00<00:12, 25.64it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/311 [00:00<00:11, 27.81it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/311 [00:00<00:11, 27.75it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/311 [00:00<00:21, 14.56it/s, Materializing param=lm_head.weight]
Loading weights: 1%| | 2/311 [00:00<00:21, 14.60it/s, Materializing param=lm_head.weight]
Loading weights: 1%| | 2/311 [00:00<00:23, 13.40it/s, Materializing param=lm_head.weight]
Loading weights: 1%| | 2/311 [00:00<00:23, 13.40it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/311 [00:00<00:14, 21.28it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/311 [00:00<00:21, 14.54it/s, Materializing param=lm_head.weight]
Loading weights: 1%| | 2/311 [00:00<00:23, 13.40it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/311 [00:00<00:14, 21.26it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/311 [00:00<00:21, 14.54it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 3/311 [00:00<00:22, 13.40it/s, Materializing param=model.layers.0.input_layernorm.weight]
Loading weights: 1%| | 2/311 [00:00<00:21, 14.54it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/311 [00:00<00:21, 14.56it/s, Materializing param=model.embed_tokens.weight
|
|||
|
|
Loading weights: 35%|███▌ | 110/311 [00:00<00:00, 254.75it/s, Materializing param=model.layers.9.self_attn.q_norm.weight]
Loading weights: 35%|███▌ | 110/311 [00:00<00:00, 254.62it/s, Materializing param=model.layers.9.self_attn.q_norm.weight]
Loading weights: 35%|███▌ | 110/311 [00:00<00:00, 254.75it/s, Materializing param=model.layers.9.self_attn.q_norm.weight]
Loading weights: 35%|███▌ | 110/311 [00:00<00:00, 254.62it/s, Materializing param=model.layers.9.self_attn.q_norm.weight]
Loading weights: 35%|███▌ | 110/311 [00:00<00:00, 252.21it/s, Materializing param=model.layers.9.self_attn.q_norm.weight]
Loading weights: 35%|███▌ | 110/311 [00:00<00:00, 252.25it/s, Materializing param=model.layers.9.self_attn.q_norm.weight]
Loading weights: 35%|███▌ | 110/311 [00:00<00:00, 252.21it/s, Materializing param=model.layers.9.self_attn.q_norm.weight]
Loading weights: 35%|███▌ | 110/311 [00:00<00:00, 252.25it/s, Materializing param=model.layers.9.self_attn.q_norm.weight]
Loading weights: 36%|███▌ | 111/311 [00:00<00:00, 252.21it/s, Materializing param=model.layers.9.self_attn.q_proj.weight]
Loading weights: 36%|███▌ | 111/311 [00:00<00:00, 252.25it/s, Materializing param=model.layers.9.self_attn.q_proj.weight]
Loading weights: 36%|███▌ | 111/311 [00:00<00:00, 254.75it/s, Materializing param=model.layers.9.self_attn.q_proj.weight]
Loading weights: 36%|███▌ | 111/311 [00:00<00:00, 252.21it/s, Materializing param=model.layers.9.self_attn.q_proj.weight]
Loading weights: 36%|███▌ | 111/311 [00:00<00:00, 252.25it/s, Materializing param=model.layers.9.self_attn.q_proj.weight]
Loading weights: 36%|███▌ | 111/311 [00:00<00:00, 254.75it/s, Materializing param=model.layers.9.self_attn.q_proj.weight]
Loading weights: 80%|████████ | 249/311 [00:00<00:00, 517.71it/s, Materializing param=model.layers.22.post_attention_layernorm.weight]
Loading weights: 80%|████████ | 249/311 [00:00<00:00, 581.60it/s, Materializing param=model.layers.22.post_attention_layernorm.weight]
Loading weights: 80%|████████ | 249/311 [00:00<00:00, 517.71it/s, Materializing param=model.layers.22.post_attention_layernorm.weight]
Loading weights: 80%|████████ | 249/311 [00:00<00:00, 516.98it/s, Materializing param=model.layers.22.post_attention_layernorm.weight]
Loading weights: 80%|████████ | 249/311 [00:00<00:00, 581.60it/s, Materializing param=model.layers.22.post_attention_layernorm.weight]
Loading weights: 80%|████████ | 249/311 [00:00<00:00, 516.98it/s, Materializing param=model.layers.22.post_attention_layernorm.weight]
Loading weights: 80%|████████ | 250/311 [00:00<00:00, 517.71it/s, Materializing param=model.layers.22.self_attn.k_norm.weight]
Loading weights: 80%|████████ | 250/311 [00:00<00:00, 517.71it/s, Materializing param=model.layers.22.self_attn.k_norm.weight]
Loading weights: 80%|████████ | 250/311 [00:00<00:00, 581.60it/s, Materializing param=model.layers.22.self_attn.k_norm.weight]
Loading weights: 80%|████████ | 250/311 [00:00<00:00, 516.98it/s, Materializing param=model.layers.22.self_attn.k_norm.weight]
Loading weights: 80%|████████ | 250/311 [00:00<00:00, 581.60it/s, Materializing param=model.layers.22.self_attn.k_norm.weight]
Loading weights: 80%|████████ | 250/311 [00:00<00:00, 516.98it/s, Materializing param=model.layers.22.self_attn.k_norm.weight]
Loading weights: 81%|████████ | 251/311 [00:00<00:00, 517.71it/s, Materializing param=model.layers.22.self_attn.k_proj.weight]
Loading weights: 81%|████████ | 251/311 [00:00<00:00, 517.71it/s, Materializing param=model.layers.22.self_attn.k_proj.weight]
Loading weights: 81%|████████ | 251/311 [00:00<00:00, 581.60it/s, Mate
|
|||
|
|
Loading weights: 37%|███▋ | 114/311 [00:00<00:00, 254.62it/s, Materializing param=model.layers.10.mlp.down_proj.weight]
Loading weights: 37%|███▋ | 114/311 [00:00<00:00, 254.62it/s, Materializing param=model.layers.10.mlp.down_proj.weight]
Loading weights: 37%|███▋ | 115/311 [00:00<00:00, 254.62it/s, Materializing param=model.layers.10.mlp.gate_proj.weight]
Loading weights: 37%|███▋ | 115/311 [00:00<00:00, 254.62it/s, Materializing param=model.layers.10.mlp.gate_proj.weight]
Loading weights: 37%|███▋ | 116/311 [00:00<00:00, 254.62it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 37%|███▋ | 116/311 [00:00<00:00, 254.62it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 37%|███▋ | 116/311 [00:00<00:00, 252.21it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 37%|███▋ | 116/311 [00:00<00:00, 252.25it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 37%|███▋ | 116/311 [00:00<00:00, 254.75it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 37%|███▋ | 116/311 [00:00<00:00, 252.25it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 37%|███▋ | 116/311 [00:00<00:00, 254.75it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 37%|███▋ | 116/311 [00:00<00:00, 252.21it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 517.71it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 516.98it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 516.98it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 517.71it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 581.60it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 581.60it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 38%|███▊ | 117/311 [00:00<00:00, 254.62it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 38%|███▊ | 117/311 [00:00<00:00, 252.25it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 38%|███▊ | 117/311 [00:00<00:00, 252.25it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 38%|███▊ | 117/311 [00:00<00:00, 254.62it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 38%|███▊ | 117/311 [00:00<00:00, 252.21it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 38%|███▊ | 117/311 [00:00<00:00, 254.75it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 38%|███▊ | 117/311 [00:00<00:00, 252.21it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 38%|███▊ | 117/311 [00:00<00:00, 254.75it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 38%|███▊ | 118/311 [00:00<00:00, 252.25it/s, Materializing param=model.layers.10.self_attn.k_norm.weight]
Loading weights: 38%|███▊ | 118/311 [00:00<00:00, 254.62it/s, Materializing param=model.layers.10.self_attn.k_norm.weight]
Loading weights: 38%|███▊ | 118/311 [00:00<00:00, 252.25it/s, Materializing param=model.layers.10.self_attn.k_norm.weight]
Loading weights: 38%|███▊ | 118/3
|
|||
|
|
Loading weights: 100%|█████████▉| 310/311 [00:00<00:00, 568.40it/s, Materializing param=model.layers.27.self_attn.v_proj.weight]
Loading weights: 50%|████▉ | 154/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.13.self_attn.q_norm.weight]
Loading weights: 50%|████▉ | 154/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.13.self_attn.q_norm.weight]
Loading weights: 50%|████▉ | 154/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.13.self_attn.q_norm.weight]
Loading weights: 50%|████▉ | 154/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.13.self_attn.q_norm.weight]
Loading weights: 50%|████▉ | 154/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.13.self_attn.q_norm.weight]
Loading weights: 50%|████▉ | 154/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.13.self_attn.q_norm.weight]
Loading weights: 50%|████▉ | 155/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 50%|████▉ | 155/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 50%|████▉ | 155/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 50%|████▉ | 155/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 50%|████▉ | 155/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 50%|████▉ | 155/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 100%|██████████| 311/311 [00:00<00:00, 568.40it/s, Materializing param=model.norm.weight]
Loading weights: 100%|██████████| 311/311 [00:00<00:00, 568.40it/s, Materializing param=model.norm.weight]
Loading weights: 100%|█████████▉| 310/311 [00:00<00:00, 565.15it/s, Materializing param=model.layers.27.self_attn.v_proj.weight]
Loading weights: 100%|██████████| 311/311 [00:00<00:00, 483.61it/s, Materializing param=model.norm.weight]
Loading weights: 100%|█████████▉| 310/311 [00:00<00:00, 565.15it/s, Materializing param=model.layers.27.self_attn.v_proj.weight]
|
|||
|
|
Loading weights: 100%|██████████| 311/311 [00:00<00:00, 565.15it/s, Materializing param=model.norm.weight]
Loading weights: 100%|██████████| 311/311 [00:00<00:00, 565.15it/s, Materializing param=model.norm.weight]
Loading weights: 100%|██████████| 311/311 [00:00<00:00, 483.53it/s, Materializing param=model.norm.weight]
|
|||
|
|
Loading weights: 50%|█████ | 156/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 50%|█████ | 156/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 50%|█████ | 156/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 50%|█████ | 156/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 50%|█████ | 156/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 50%|█████ | 156/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 50%|████▉ | 154/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.13.self_attn.q_norm.weight]
Loading weights: 50%|████▉ | 154/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.13.self_attn.q_norm.weight]
Loading weights: 50%|████▉ | 155/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 50%|████▉ | 155/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 50%|█████ | 156/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 50%|█████ | 156/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 50%|█████ | 157/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 50%|█████ | 157/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 50%|█████ | 157/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 50%|█████ | 157/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 50%|█████ | 157/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 50%|█████ | 157/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 51%|█████ | 158/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 51%|█████ | 158/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 51%|█████ | 158/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 51%|█████ | 158/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 51%|█████ | 158/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 51%|█████ | 158/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 50%|█████ | 157/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 50%|█████ | 157/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 51%|█████ | 158/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 51%|█████ | 158/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 51%|█████ | 159/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.mlp.gate_proj.weight]
Loading weights: 51%|█████ | 159/311 [00:00<00:00, 314.18it/s, Material
|
|||
|
|
The tied weights mapping and config for this model specifies to tie model.embed_tokens.weight to lm_head.weight, but both are present in the checkpoints, so we will NOT tie them. You should update the config with `tie_word_embeddings=False` to silence this warning
|
|||
|
|
The tied weights mapping and config for this model specifies to tie model.embed_tokens.weight to lm_head.weight, but both are present in the checkpoints, so we will NOT tie them. You should update the config with `tie_word_embeddings=False` to silence this warning
|
|||
|
|
Loading weights: 51%|█████▏ | 160/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 51%|█████▏ | 160/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 51%|█████▏ | 160/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 51%|█████▏ | 160/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 51%|█████▏ | 160/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 51%|█████▏ | 160/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 51%|█████▏ | 160/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 51%|█████▏ | 160/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 52%|█████▏ | 161/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 161/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 161/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 161/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 161/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 161/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 161/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 162/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.self_attn.k_norm.weight]
Loading weights: 52%|█████▏ | 161/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 162/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.self_attn.k_norm.weight]
Loading weights: 52%|█████▏ | 162/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.self_attn.k_norm.weight]
Loading weights: 52%|█████▏ | 162/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.14.self_attn.k_norm.weight]
Loading weights: 52%|█████▏ | 162/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.self_attn.k_norm.weight]
Loading weights: 52%|█████▏ | 162/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.14.self_attn.k_norm.weight]
Loading weights: 52%|█████▏ | 162/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.14.self_attn.k_norm.weight]
Loading weights: 52%|█████▏ | 162/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.14.self_attn.k_norm.weight]
Loading weights: 52%|█████▏ | 163/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.self_attn.k_proj.weight]
Loading weights: 52%|█████▏ | 163/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.14.self_attn.k_proj.weight]
Loading weights: 52%|█████▏ | 163/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.14.self_attn.k_proj.weight]
Loading weights: 52%|█████▏ | 163/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.14.self_attn.k_proj.weight]
Loading weights: 52%|█████▏ | 163/311 [00:00<00:00, 314.14it/s, Ma
|
|||
|
|
Loading weights: 55%|█████▌ | 172/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 55%|█████▌ | 172/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 56%|█████▌ | 173/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.15.self_attn.k_norm.weight]
Loading weights: 56%|█████▌ | 173/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.15.self_attn.k_norm.weight][INFO|configuration_utils.py:1014] 2026-05-24 05:50:22,194 >> Generate config GenerationConfig {
|
|||
|
|
"bos_token_id": 151643,
|
|||
|
|
"do_sample": true,
|
|||
|
|
"eos_token_id": [
|
|||
|
|
151645,
|
|||
|
|
151643
|
|||
|
|
],
|
|||
|
|
"pad_token_id": 151643,
|
|||
|
|
"temperature": 0.6,
|
|||
|
|
"top_k": 20,
|
|||
|
|
"top_p": 0.95
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
Loading weights: 56%|█████▌ | 174/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 56%|█████▌ | 174/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 56%|█████▋ | 175/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 56%|█████▋ | 175/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 55%|█████▌ | 172/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 55%|█████▌ | 172/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 55%|█████▌ | 172/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 56%|█████▌ | 173/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.15.self_attn.k_norm.weight]
Loading weights: 55%|█████▌ | 172/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 56%|█████▌ | 173/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.15.self_attn.k_norm.weight]
Loading weights: 56%|█████▌ | 173/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.15.self_attn.k_norm.weight]
Loading weights: 56%|█████▌ | 173/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.15.self_attn.k_norm.weight]
Loading weights: 56%|█████▌ | 174/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 56%|█████▌ | 174/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 56%|█████▌ | 174/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 55%|█████▌ | 172/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 56%|█████▌ | 174/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 55%|█████▌ | 172/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 56%|█████▌ | 173/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.15.self_attn.k_norm.weight]
Loading weights: 56%|█████▌ | 173/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.15.self_attn.k_norm.weight]
Loading weights: 56%|█████▋ | 175/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 56%|█████▋ | 175/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 56%|█████▋ | 175/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 56%|█████▌ | 174/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 56%|█████▋ | 175/311 [00:00<00:00, 316.24it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 56%|█████▌ | 174/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 56%|█████▋ | 175/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 56%|█████▋ | 175/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 57%|█████▋ | 176/311 [00:00<00:00, 316.10
|
|||
|
|
[INFO|2026-05-24 05:50:22] llamafactory.model.model_utils.checkpointing:144 >> Gradient checkpointing enabled.
|
|||
|
|
[INFO|2026-05-24 05:50:22] llamafactory.model.model_utils.attention:144 >> Using torch SDPA for faster training and inference.
|
|||
|
|
[INFO|2026-05-24 05:50:22] llamafactory.model.adapter:144 >> Upcasting trainable params to float32.
|
|||
|
|
[INFO|2026-05-24 05:50:22] llamafactory.model.adapter:144 >> Fine-tuning method: Full
|
|||
|
|
Loading weights: 62%|██████▏ | 193/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 62%|██████▏ | 193/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.17.mlp.gate_proj.weight]
Loading weights: 62%|██████▏ | 193/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 62%|██████▏ | 193/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 62%|██████▏ | 193/311 [00:00<00:00, 316.10it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 62%|██████▏ | 193/311 [00:00<00:00, 314.18it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 62%|██████▏ | 193/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 62%|██████▏ | 193/311 [00:00<00:00, 314.14it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 62%|██████▏ | 193/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 62%|██████▏ | 194/311 [00:00<00:00, 386.15it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 62%|██████▏ | 194/311 [00:00<00:00, 386.15it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 62%|██████▏ | 194/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 62%|██████▏ | 194/311 [00:00<00:00, 386.15it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 62%|██████▏ | 194/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 62%|██████▏ | 194/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 62%|██████▏ | 194/311 [00:00<00:00, 385.92it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 62%|██████▏ | 194/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 62%|██████▏ | 194/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 62%|██████▏ | 194/311 [00:00<00:00, 385.92it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 63%|██████▎ | 195/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.17.self_attn.k_norm.weight]
Loading weights: 62%|██████▏ | 194/311 [00:00<00:00, 385.92it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 63%|██████▎ | 195/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.17.self_attn.k_norm.weight]
Loading weights: 63%|██████▎ | 195/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.17.self_attn.k_norm.weight]
Loading weights: 63%|██████▎ | 195/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.17.self_attn.k_norm.weight]
Loading weights: 63%|██████▎ | 195/311 [00:00<00:00, 385.92it/s, Materializing param=model.layers.17.self_attn.k_norm.weight]
Loading weights: 63%|██████▎ | 196/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.17.self_attn.k_proj.weight]
Loading weights: 63%|██████▎ | 195/311 [00:00<00:00, 385.92it/s, Materializing param=model.layers.17.self_attn.k_norm.weight]
Loading weights: 63%|██████▎ | 195/311 [00:00<00:00, 386.15it/s, Materializing param=model.layers.17.self_attn.k_norm.weight]
Loading weights: 63%|██████▎
|
|||
|
|
Loading weights: 76%|███████▌ | 237/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 76%|███████▌ | 237/311 [00:00<00:00, 385.92it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 76%|███████▌ | 237/311 [00:00<00:00, 386.15it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 76%|███████▌ | 237/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 76%|███████▌ | 237/311 [00:00<00:00, 386.15it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 76%|███████▌ | 237/311 [00:00<00:00, 385.92it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 76%|███████▌ | 237/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 76%|███████▌ | 237/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 77%|███████▋ | 238/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 238/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 238/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 238/311 [00:00<00:00, 385.92it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 238/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 238/311 [00:00<00:00, 386.15it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 238/311 [00:00<00:00, 385.92it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 238/311 [00:00<00:00, 386.15it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 239/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.21.self_attn.k_norm.weight]
Loading weights: 77%|███████▋ | 239/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.21.self_attn.k_norm.weight]
Loading weights: 77%|███████▋ | 239/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.21.self_attn.k_norm.weight]
Loading weights: 77%|███████▋ | 239/311 [00:00<00:00, 385.92it/s, Materializing param=model.layers.21.self_attn.k_norm.weight]
Loading weights: 77%|███████▋ | 239/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.21.self_attn.k_norm.weight]
Loading weights: 77%|███████▋ | 239/311 [00:00<00:00, 386.15it/s, Materializing param=model.layers.21.self_attn.k_norm.weight]
Loading weights: 77%|███████▋ | 239/311 [00:00<00:00, 385.92it/s, Materializing param=model.layers.21.self_attn.k_norm.weight]
Loading weights: 77%|███████▋ | 239/311 [00:00<00:00, 386.15it/s, Materializing param=model.layers.21.self_attn.k_norm.weight]
Loading weights: 77%|███████▋ | 240/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.21.self_attn.k_proj.weight]
Loading weights: 77%|███████▋ | 240/311 [00:00<00:00, 387.41it/s, Materializing param=model.layers.21.self_attn.k_proj.weight]
Loading weights: 77%|███████▋ | 240/311 [00:00<00:00, 387.52it/s, Materializing param=model.layers.21.self_attn.k_proj.weight]
Loading weights: 77%|███████▋ | 240/311 [00:00<00:00, 385.92it/s, Materializing param=model.layer
|
|||
|
|
Loading weights: 83%|████████▎ | 258/311 [00:00<00:00, 433.31it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]
Loading weights: 83%|████████▎ | 258/311 [00:00<00:00, 434.50it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]
Loading weights: 83%|████████▎ | 258/311 [00:00<00:00, 434.25it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]
Loading weights: 83%|████████▎ | 258/311 [00:00<00:00, 434.50it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]
Loading weights: 83%|████████▎ | 258/311 [00:00<00:00, 434.25it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]
Loading weights: 83%|████████▎ | 258/311 [00:00<00:00, 433.20it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]
Loading weights: 83%|████████▎ | 258/311 [00:00<00:00, 433.31it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]
Loading weights: 83%|████████▎ | 258/311 [00:00<00:00, 433.20it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None, 'pad_token_id': 151643}.
|
|||
|
|
The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None, 'pad_token_id': 151643}.
|
|||
|
|
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 433.20it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 434.25it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 433.31it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 434.50it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 433.20it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 434.25it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 433.31it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 83%|████████▎ | 259/311 [00:00<00:00, 434.50it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 84%|████████▎ | 260/311 [00:00<00:00, 433.20it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 84%|████████▎ | 260/311 [00:00<00:00, 434.50it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 84%|████████▎ | 260/311 [00:00<00:00, 434.25it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 84%|████████▎ | 260/311 [00:00<00:00, 434.50it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 84%|████████▎ | 260/311 [00:00<00:00, 433.31it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 84%|████████▎ | 260/311 [00:00<00:00, 434.25it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 84%|████████▎ | 260/311 [00:00<00:00, 433.31it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 84%|████████▍ | 261/311 [00:00<00:00, 434.50it/s, Materializing param=model.layers.23.self_attn.k_norm.weight]
Loading weights: 84%|████████▍ | 261/311 [00:00<00:00, 434.25it/s, Materializing param=model.layers.23.self_attn.k_norm.weight]
Loading weights: 84%|████████▍ | 261/311 [00:00<00:00, 434.50it/s, Materializing param=model.layers.23.self_attn.k_norm.weight]
Loading weights: 84%|████████▍ | 261/311 [00:00<00:00, 434.25it/s, Materializing param=model.layers.23.self_attn.k_norm.weight]
Loading weights: 84%|████████▍ | 261/311 [00:00<00:00, 433.31it/s, Materializing param=model.layers.23.self_attn.k_norm.weight]
Loading weights: 84%|████████▎ | 260/311 [00:00<00:00, 433.20it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 84%|████████▍ | 261/311 [00:00<00:00, 433.31it/s, Materializing param=model.layers.23.self_attn.k_norm.weight]
Loading weights: 84%|████████▍ | 262/311 [00:00<00:00, 434.50it/s, Materializing param=model.layers.23.self_attn.k_proj.weight]
Loading weights: 84%|████████▍ | 262/311 [00:00<00:00, 434.25it/s, Materializing param=model.layers.23.self_attn.k_proj.weight]
Loading weights: 84%|████████▍ | 262/311 [00:00<00:00, 434.50it/s, Materializing param=model.layers.23.self_attn.k_proj.weight]
Loading weights: 84%|████████▍ | 261/311 [00:00<00:00, 433.20it/s, Materializing param=model.layers.23.self_attn.k_norm.weight]
Loading weights: 84%|████████▍ | 262/311 [00:00<00:00, 434.25it/s, Materializing param=model.layers.23.self_attn.k_proj.weight]
Loading weights: 84%|████████▍ | 262/311 [
|
|||
|
|
Loading weights: 100%|██████████| 311/311 [00:00<00:00, 335.53it/s, Materializing param=model.norm.weight]
Loading weights: 100%|██████████| 311/311 [00:00<00:00, 335.37it/s, Materializing param=model.norm.weight]
|
|||
|
|
|
|||
|
|
|
|||
|
|
The tied weights mapping and config for this model specifies to tie model.embed_tokens.weight to lm_head.weight, but both are present in the checkpoints, so we will NOT tie them. You should update the config with `tie_word_embeddings=False` to silence this warning
|
|||
|
|
The tied weights mapping and config for this model specifies to tie model.embed_tokens.weight to lm_head.weight, but both are present in the checkpoints, so we will NOT tie them. You should update the config with `tie_word_embeddings=False` to silence this warning
|
|||
|
|
The tied weights mapping and config for this model specifies to tie model.embed_tokens.weight to lm_head.weight, but both are present in the checkpoints, so we will NOT tie them. You should update the config with `tie_word_embeddings=False` to silence this warning
|
|||
|
|
The tied weights mapping and config for this model specifies to tie model.embed_tokens.weight to lm_head.weight, but both are present in the checkpoints, so we will NOT tie them. You should update the config with `tie_word_embeddings=False` to silence this warning
|
|||
|
|
The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None, 'pad_token_id': 151643}.
|
|||
|
|
[WARNING|trainer_utils.py:1234] 2026-05-24 05:50:22,623 >> The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None, 'pad_token_id': 151643}.
|
|||
|
|
The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None, 'pad_token_id': 151643}.
|
|||
|
|
The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None, 'pad_token_id': 151643}.
|
|||
|
|
The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None, 'pad_token_id': 151643}.
|
|||
|
|
[INFO|trainer.py:1587] 2026-05-24 05:50:25,166 >> ***** Running training *****
|
|||
|
|
[INFO|trainer.py:1588] 2026-05-24 05:50:25,166 >> Num examples = 55,931
|
|||
|
|
[INFO|trainer.py:1589] 2026-05-24 05:50:25,166 >> Num Epochs = 2
|
|||
|
|
[INFO|trainer.py:1590] 2026-05-24 05:50:25,166 >> Instantaneous batch size per device = 8
|
|||
|
|
[INFO|trainer.py:1593] 2026-05-24 05:50:25,166 >> Total train batch size (w. parallel, distributed & accumulation) = 128
|
|||
|
|
[INFO|trainer.py:1594] 2026-05-24 05:50:25,166 >> Gradient Accumulation steps = 2
|
|||
|
|
[INFO|trainer.py:1595] 2026-05-24 05:50:25,166 >> Total optimization steps = 874
|
|||
|
|
[INFO|trainer.py:1596] 2026-05-24 05:50:25,167 >> Number of trainable parameters = 2,031,739,904
|
|||
|
|
wandb: [wandb.login()] Loaded credentials for https://api.wandb.ai from WANDB_API_KEY.
|
|||
|
|
wandb: Currently logged in as: kkhya (maskmoe) to https://api.wandb.ai. Use `wandb login --relogin` to force relogin
|
|||
|
|
wandb: setting up run j7unaum2
|
|||
|
|
wandb: Tracking run with wandb version 0.27.0
|
|||
|
|
wandb: Run data is saved locally in /home/ubuntu/Mingyuan/lm-factory/wandb/run-20260524_055025-j7unaum2
|
|||
|
|
wandb: Run `wandb offline` to turn off syncing.
|
|||
|
|
wandb: Syncing run qwen3_1.7b_if_fft
|
|||
|
|
wandb: ⭐️ View project at https://wandb.ai/maskmoe/MFT-LM
|
|||
|
|
wandb: 🚀 View run at https://wandb.ai/maskmoe/MFT-LM/runs/j7unaum2
|
|||
|
|
0%| | 0/874 [00:00<?, ?it/s]
0%| | 1/874 [00:03<56:11, 3.86s/it]
0%| | 2/874 [00:05<40:27, 2.78s/it]
0%| | 3/874 [00:08<39:44, 2.74s/it]
0%| | 4/874 [00:11<39:12, 2.70s/it]
1%| | 5/874 [00:13<37:21, 2.58s/it]
1%| | 6/874 [00:16<37:20, 2.58s/it]
1%| | 7/874 [00:18<37:22, 2.59s/it]
1%| | 8/874 [00:21<37:13, 2.58s/it]
1%| | 9/874 [00:24<38:03, 2.64s/it]
1%| | 10/874 [00:26<36:15, 2.52s/it]
{'loss': '1.394', 'grad_norm': '18.22', 'learning_rate': '1.023e-06', 'epoch': '0.02288'}
|
|||
|
|
1%| | 10/874 [00:26<36:15, 2.52s/it]
1%|▏ | 11/874 [00:29<38:59, 2.71s/it]
1%|▏ | 12/874 [00:31<37:51, 2.64s/it]
1%|▏ | 13/874 [00:34<35:45, 2.49s/it]
2%|▏ | 14/874 [00:36<34:35, 2.41s/it]
2%|▏ | 15/874 [00:38<34:13, 2.39s/it]
2%|▏ | 16/874 [00:40<32:38, 2.28s/it]
2%|▏ | 17/874 [00:43<35:21, 2.48s/it]
2%|▏ | 18/874 [00:46<34:50, 2.44s/it]
2%|▏ | 19/874 [00:48<35:04, 2.46s/it]
2%|▏ | 20/874 [00:50<33:39, 2.36s/it]
{'loss': '1.384', 'grad_norm': '14.22', 'learning_rate': '2.159e-06', 'epoch': '0.04577'}
|
|||
|
|
2%|▏ | 20/874 [00:50<33:39, 2.36s/it]
2%|▏ | 21/874 [00:53<37:06, 2.61s/it]
3%|▎ | 22/874 [00:56<37:02, 2.61s/it]
3%|▎ | 23/874 [00:59<37:18, 2.63s/it]
3%|▎ | 24/874 [01:01<34:36, 2.44s/it]
3%|▎ | 25/874 [01:03<34:43, 2.45s/it]
3%|▎ | 26/874 [01:05<32:32, 2.30s/it]
3%|▎ | 27/874 [01:07<33:03, 2.34s/it]
3%|▎ | 28/874 [01:11<36:21, 2.58s/it]
3%|▎ | 29/874 [01:13<34:15, 2.43s/it]
3%|▎ | 30/874 [01:15<35:00, 2.49s/it]
{'loss': '1.34', 'grad_norm': '8.084', 'learning_rate': '3.295e-06', 'epoch': '0.06865'}
|
|||
|
|
3%|▎ | 30/874 [01:15<35:00, 2.49s/it]
4%|▎ | 31/874 [01:18<35:34, 2.53s/it]
4%|▎ | 32/874 [01:21<36:46, 2.62s/it]
4%|▍ | 33/874 [01:23<35:53, 2.56s/it]
4%|▍ | 34/874 [01:26<37:48, 2.70s/it]
4%|▍ | 35/874 [01:28<35:34, 2.54s/it]
4%|▍ | 36/874 [01:30<32:50, 2.35s/it]
4%|▍ | 37/874 [01:33<32:36, 2.34s/it]
4%|▍ | 38/874 [01:36<35:31, 2.55s/it]
4%|▍ | 39/874 [01:38<35:57, 2.58s/it]
5%|▍ | 40/874 [01:40<32:56, 2.37s/it]
{'loss': '1.199', 'grad_norm': '4.825', 'learning_rate': '4.432e-06', 'epoch': '0.09153'}
|
|||
|
|
5%|▍ | 40/874 [01:40<32:56, 2.37s/it]
5%|▍ | 41/874 [01:43<33:19, 2.40s/it]
5%|▍ | 42/874 [01:45<32:18, 2.33s/it]
5%|▍ | 43/874 [01:48<34:41, 2.50s/it]
5%|▌ | 44/874 [01:50<33:57, 2.45s/it]
5%|▌ | 45/874 [01:53<33:42, 2.44s/it]
5%|▌ | 46/874 [01:55<33:00, 2.39s/it]
5%|▌ | 47/874 [01:57<33:33, 2.43s/it]
5%|▌ | 48/874 [02:00<33:19, 2.42s/it]
6%|▌ | 49/874 [02:02<33:05, 2.41s/it]
6%|▌ | 50/874 [02:05<34:08, 2.49s/it]
{'loss': '1.184', 'grad_norm': '2.621', 'learning_rate': '5.568e-06', 'epoch': '0.1144'}
|
|||
|
|
6%|▌ | 50/874 [02:05<34:08, 2.49s/it]
6%|▌ | 51/874 [02:07<33:16, 2.43s/it]
6%|▌ | 52/874 [02:10<34:07, 2.49s/it]
6%|▌ | 53/874 [02:12<34:21, 2.51s/it]
6%|▌ | 54/874 [02:15<35:44, 2.61s/it]
6%|▋ | 55/874 [02:18<35:29, 2.60s/it]
6%|▋ | 56/874 [02:20<34:00, 2.49s/it]
7%|▋ | 57/874 [02:23<38:08, 2.80s/it]
7%|▋ | 58/874 [02:25<34:33, 2.54s/it]
7%|▋ | 59/874 [02:28<34:02, 2.51s/it]
7%|▋ | 60/874 [02:30<32:39, 2.41s/it]
{'loss': '1.151', 'grad_norm': '2.281', 'learning_rate': '6.705e-06', 'epoch': '0.1373'}
|
|||
|
|
7%|▋ | 60/874 [02:30<32:39, 2.41s/it]
7%|▋ | 61/874 [02:33<33:09, 2.45s/it]
7%|▋ | 62/874 [02:35<33:04, 2.44s/it]
7%|▋ | 63/874 [02:38<34:12, 2.53s/it]
7%|▋ | 64/874 [02:40<32:44, 2.42s/it]
7%|▋ | 65/874 [02:43<34:49, 2.58s/it]
8%|▊ | 66/874 [02:46<36:44, 2.73s/it]
8%|▊ | 67/874 [02:48<35:50, 2.66s/it]
8%|▊ | 68/874 [02:51<36:46, 2.74s/it]
8%|▊ | 69/874 [02:54<35:03, 2.61s/it]
8%|▊ | 70/874 [02:56<35:36, 2.66s/it]
{'loss': '1.134', 'grad_norm': '1.84', 'learning_rate': '7.841e-06', 'epoch': '0.1602'}
|
|||
|
|
8%|▊ | 70/874 [02:56<35:36, 2.66s/it]
8%|▊ | 71/874 [02:59<33:56, 2.54s/it]
8%|▊ | 72/874 [03:01<33:09, 2.48s/it]
8%|▊ | 73/874 [03:04<33:23, 2.50s/it]
8%|▊ | 74/874 [03:06<33:14, 2.49s/it]
9%|▊ | 75/874 [03:08<32:14, 2.42s/it]
9%|▊ | 76/874 [03:10<30:46, 2.31s/it]
9%|▉ | 77/874 [03:13<31:44, 2.39s/it]
9%|▉ | 78/874 [03:15<31:27, 2.37s/it]
9%|▉ | 79/874 [03:17<29:55, 2.26s/it]
9%|▉ | 80/874 [03:20<30:55, 2.34s/it]
{'loss': '1.063', 'grad_norm': '1.405', 'learning_rate': '8.977e-06', 'epoch': '0.1831'}
|
|||
|
|
9%|▉ | 80/874 [03:20<30:55, 2.34s/it]
9%|▉ | 81/874 [03:22<30:38, 2.32s/it]
9%|▉ | 82/874 [03:24<30:15, 2.29s/it]
9%|▉ | 83/874 [03:27<30:40, 2.33s/it]
10%|▉ | 84/874 [03:29<31:36, 2.40s/it]
10%|▉ | 85/874 [03:31<30:53, 2.35s/it]
10%|▉ | 86/874 [03:34<30:08, 2.29s/it]
10%|▉ | 87/874 [03:36<30:23, 2.32s/it]
10%|█ | 88/874 [03:38<29:25, 2.25s/it]
10%|█ | 89/874 [03:41<30:40, 2.34s/it]
10%|█ | 90/874 [03:43<29:46, 2.28s/it]
{'loss': '1.056', 'grad_norm': '1.126', 'learning_rate': '1e-05', 'epoch': '0.2059'}
|
|||
|
|
10%|█ | 90/874 [03:43<29:46, 2.28s/it]
10%|█ | 91/874 [03:45<28:52, 2.21s/it]
11%|█ | 92/874 [03:47<29:10, 2.24s/it]
11%|█ | 93/874 [03:50<30:21, 2.33s/it]
11%|█ | 94/874 [03:52<28:58, 2.23s/it]
11%|█ | 95/874 [03:54<30:39, 2.36s/it]
11%|█ | 96/874 [03:56<29:25, 2.27s/it]
11%|█ | 97/874 [03:59<30:19, 2.34s/it]
11%|█ | 98/874 [04:01<31:03, 2.40s/it]
11%|█▏ | 99/874 [04:04<32:47, 2.54s/it]
11%|█▏ | 100/874 [04:06<31:13, 2.42s/it]
{'loss': '1.045', 'grad_norm': '1.186', 'learning_rate': '9.995e-06', 'epoch': '0.2288'}
|
|||
|
|
11%|█▏ | 100/874 [04:06<31:13, 2.42s/it]
12%|█▏ | 101/874 [04:09<31:58, 2.48s/it]
12%|█▏ | 102/874 [04:12<31:59, 2.49s/it]
12%|█▏ | 103/874 [04:14<31:46, 2.47s/it]
12%|█▏ | 104/874 [04:17<31:51, 2.48s/it]
12%|█▏ | 105/874 [04:19<31:30, 2.46s/it]
12%|█▏ | 106/874 [04:22<32:44, 2.56s/it]
12%|█▏ | 107/874 [04:24<30:29, 2.39s/it]
12%|█▏ | 108/874 [04:26<31:03, 2.43s/it]
12%|█▏ | 109/874 [04:29<31:08, 2.44s/it]
13%|█▎ | 110/874 [04:31<29:22, 2.31s/it]
{'loss': '1.07', 'grad_norm': '1.051', 'learning_rate': '9.982e-06', 'epoch': '0.2517'}
|
|||
|
|
13%|█▎ | 110/874 [04:31<29:22, 2.31s/it]
13%|█▎ | 111/874 [04:33<30:11, 2.37s/it]
13%|█▎ | 112/874 [04:35<28:42, 2.26s/it]
13%|█▎ | 113/874 [04:38<31:50, 2.51s/it]
13%|█▎ | 114/874 [04:41<32:36, 2.57s/it]
13%|█▎ | 115/874 [04:43<31:43, 2.51s/it]
13%|█▎ | 116/874 [04:45<28:55, 2.29s/it]
13%|█▎ | 117/874 [04:47<29:03, 2.30s/it]
14%|█▎ | 118/874 [04:50<28:38, 2.27s/it]
14%|█▎ | 119/874 [04:52<27:32, 2.19s/it]
14%|█▎ | 120/874 [04:54<27:10, 2.16s/it]
{'loss': '1.052', 'grad_norm': '0.9556', 'learning_rate': '9.962e-06', 'epoch': '0.2746'}
|
|||
|
|
14%|█▎ | 120/874 [04:54<27:10, 2.16s/it]
14%|█▍ | 121/874 [04:56<26:58, 2.15s/it]
14%|█▍ | 122/874 [04:58<26:34, 2.12s/it]
14%|█▍ | 123/874 [05:01<29:04, 2.32s/it]
14%|█▍ | 124/874 [05:03<27:57, 2.24s/it]
14%|█▍ | 125/874 [05:05<29:10, 2.34s/it]
14%|█▍ | 126/874 [05:08<30:47, 2.47s/it]
15%|█▍ | 127/874 [05:11<32:17, 2.59s/it]
15%|█▍ | 128/874 [05:13<31:09, 2.51s/it]
15%|█▍ | 129/874 [05:16<30:04, 2.42s/it]
15%|█▍ | 130/874 [05:17<28:13, 2.28s/it]
{'loss': '1.053', 'grad_norm': '1.06', 'learning_rate': '9.933e-06', 'epoch': '0.2975'}
|
|||
|
|
15%|█▍ | 130/874 [05:17<28:13, 2.28s/it]
15%|█▍ | 131/874 [05:20<29:24, 2.38s/it]
15%|█▌ | 132/874 [05:22<28:15, 2.29s/it]
15%|█▌ | 133/874 [05:24<27:11, 2.20s/it]
15%|█▌ | 134/874 [05:26<26:00, 2.11s/it]
15%|█▌ | 135/874 [05:28<27:06, 2.20s/it]
16%|█▌ | 136/874 [05:31<29:33, 2.40s/it]
16%|█▌ | 137/874 [05:34<30:17, 2.47s/it]
16%|█▌ | 138/874 [05:36<30:04, 2.45s/it]
16%|█▌ | 139/874 [05:39<28:47, 2.35s/it]
16%|█▌ | 140/874 [05:41<28:08, 2.30s/it]
{'loss': '1.041', 'grad_norm': '0.9935', 'learning_rate': '9.896e-06', 'epoch': '0.3204'}
|
|||
|
|
16%|█▌ | 140/874 [05:41<28:08, 2.30s/it]
16%|█▌ | 141/874 [05:43<26:34, 2.18s/it]
16%|█▌ | 142/874 [05:45<26:57, 2.21s/it]
16%|█▋ | 143/874 [05:48<29:06, 2.39s/it]
16%|█▋ | 144/874 [05:50<28:28, 2.34s/it]
17%|█▋ | 145/874 [05:52<27:39, 2.28s/it]
17%|█▋ | 146/874 [05:55<29:05, 2.40s/it]
17%|█▋ | 147/874 [05:58<32:01, 2.64s/it]
17%|█▋ | 148/874 [06:00<30:25, 2.51s/it]
17%|█▋ | 149/874 [06:02<28:29, 2.36s/it]
17%|█▋ | 150/874 [06:05<29:15, 2.43s/it]
{'loss': '1.015', 'grad_norm': '0.9297', 'learning_rate': '9.852e-06', 'epoch': '0.3432'}
|
|||
|
|
17%|█▋ | 150/874 [06:05<29:15, 2.43s/it]
17%|█▋ | 151/874 [06:07<27:59, 2.32s/it]
17%|█▋ | 152/874 [06:09<28:32, 2.37s/it]
18%|█▊ | 153/874 [06:12<28:42, 2.39s/it]
18%|█▊ | 154/874 [06:14<27:43, 2.31s/it]
18%|█▊ | 155/874 [06:17<29:45, 2.48s/it]
18%|█▊ | 156/874 [06:19<30:43, 2.57s/it]
18%|█▊ | 157/874 [06:22<30:58, 2.59s/it]
18%|█▊ | 158/874 [06:24<29:18, 2.46s/it]
18%|█▊ | 159/874 [06:26<28:25, 2.38s/it]
18%|█▊ | 160/874 [06:28<26:26, 2.22s/it]
{'loss': '1.018', 'grad_norm': '1.093', 'learning_rate': '9.8e-06', 'epoch': '0.3661'}
|
|||
|
|
18%|█▊ | 160/874 [06:28<26:26, 2.22s/it]
18%|█▊ | 161/874 [06:30<25:49, 2.17s/it]
19%|█▊ | 162/874 [06:32<25:28, 2.15s/it]
19%|█▊ | 163/874 [06:35<25:20, 2.14s/it]
19%|█▉ | 164/874 [06:37<26:43, 2.26s/it]
19%|█▉ | 165/874 [06:40<27:03, 2.29s/it]
19%|█▉ | 166/874 [06:41<25:44, 2.18s/it]
19%|█▉ | 167/874 [06:45<29:44, 2.52s/it]
19%|█▉ | 168/874 [06:47<28:28, 2.42s/it]
19%|█▉ | 169/874 [06:49<26:48, 2.28s/it]
19%|█▉ | 170/874 [06:51<25:49, 2.20s/it]
{'loss': '1.061', 'grad_norm': '0.9561', 'learning_rate': '9.74e-06', 'epoch': '0.389'}
|
|||
|
|
19%|█▉ | 170/874 [06:51<25:49, 2.20s/it]
20%|█▉ | 171/874 [06:53<25:30, 2.18s/it]
20%|█▉ | 172/874 [06:56<26:57, 2.30s/it]
20%|█▉ | 173/874 [06:58<27:23, 2.35s/it]
20%|█▉ | 174/874 [07:00<27:23, 2.35s/it]
20%|██ | 175/874 [07:02<25:38, 2.20s/it]
20%|██ | 176/874 [07:04<25:25, 2.18s/it]
20%|██ | 177/874 [07:08<29:18, 2.52s/it]
20%|██ | 178/874 [07:10<28:48, 2.48s/it]
20%|██ | 179/874 [07:13<29:19, 2.53s/it]
21%|██ | 180/874 [07:15<29:18, 2.53s/it]
{'loss': '0.9985', 'grad_norm': '1.008', 'learning_rate': '9.673e-06', 'epoch': '0.4119'}
|
|||
|
|
21%|██ | 180/874 [07:15<29:18, 2.53s/it]
21%|██ | 181/874 [07:18<30:11, 2.61s/it]
21%|██ | 182/874 [07:20<27:08, 2.35s/it]
21%|██ | 183/874 [07:22<27:36, 2.40s/it]
21%|██ | 184/874 [07:25<28:47, 2.50s/it]
21%|██ | 185/874 [07:27<27:22, 2.38s/it]
21%|██▏ | 186/874 [07:30<27:50, 2.43s/it]
21%|██▏ | 187/874 [07:32<28:04, 2.45s/it]
22%|██▏ | 188/874 [07:34<26:46, 2.34s/it]
22%|██▏ | 189/874 [07:37<28:05, 2.46s/it]
22%|██▏ | 190/874 [07:40<30:15, 2.65s/it]
{'loss': '1.06', 'grad_norm': '0.9294', 'learning_rate': '9.598e-06', 'epoch': '0.4348'}
|
|||
|
|
22%|██▏ | 190/874 [07:40<30:15, 2.65s/it]
22%|██▏ | 191/874 [07:42<27:25, 2.41s/it]
22%|██▏ | 192/874 [07:44<26:43, 2.35s/it]
22%|██▏ | 193/874 [07:46<25:35, 2.25s/it]
22%|██▏ | 194/874 [07:49<26:12, 2.31s/it]
22%|██▏ | 195/874 [07:51<26:37, 2.35s/it]
22%|██▏ | 196/874 [07:53<25:20, 2.24s/it]
23%|██▎ | 197/874 [07:56<25:44, 2.28s/it]
23%|██▎ | 198/874 [07:58<25:14, 2.24s/it]
23%|██▎ | 199/874 [08:00<24:27, 2.17s/it]
23%|██▎ | 200/874 [08:02<26:04, 2.32s/it]
{'loss': '1.025', 'grad_norm': '0.9174', 'learning_rate': '9.516e-06', 'epoch': '0.4577'}
|
|||
|
|
23%|██▎ | 200/874 [08:02<26:04, 2.32s/it]
23%|██▎ | 201/874 [08:05<27:05, 2.41s/it]
23%|██▎ | 202/874 [08:07<26:31, 2.37s/it]
23%|██▎ | 203/874 [08:09<25:39, 2.29s/it]
23%|██▎ | 204/874 [08:12<27:02, 2.42s/it]
23%|██▎ | 205/874 [08:14<26:29, 2.38s/it]
24%|██▎ | 206/874 [08:17<26:05, 2.34s/it]
24%|██▎ | 207/874 [08:19<26:30, 2.39s/it]
24%|██▍ | 208/874 [08:21<26:01, 2.35s/it]
24%|██▍ | 209/874 [08:23<25:16, 2.28s/it]
24%|██▍ | 210/874 [08:26<25:31, 2.31s/it]
{'loss': '1.046', 'grad_norm': '0.8785', 'learning_rate': '9.427e-06', 'epoch': '0.4805'}
|
|||
|
|
24%|██▍ | 210/874 [08:26<25:31, 2.31s/it]
24%|██▍ | 211/874 [08:29<28:44, 2.60s/it]
24%|██▍ | 212/874 [08:31<26:54, 2.44s/it]
24%|██▍ | 213/874 [08:33<24:35, 2.23s/it]
24%|██▍ | 214/874 [08:36<26:22, 2.40s/it]
25%|██▍ | 215/874 [08:38<26:07, 2.38s/it]
25%|██▍ | 216/874 [08:40<24:22, 2.22s/it]
25%|██▍ | 217/874 [08:42<24:48, 2.27s/it]
25%|██▍ | 218/874 [08:44<24:20, 2.23s/it]
25%|██▌ | 219/874 [08:47<24:44, 2.27s/it]
25%|██▌ | 220/874 [08:49<25:31, 2.34s/it]
{'loss': '0.9727', 'grad_norm': '0.9212', 'learning_rate': '9.33e-06', 'epoch': '0.5034'}
|
|||
|
|
25%|██▌ | 220/874 [08:49<25:31, 2.34s/it]
25%|██▌ | 221/874 [08:51<24:29, 2.25s/it]
25%|██▌ | 222/874 [08:53<23:00, 2.12s/it]
26%|██▌ | 223/874 [08:55<22:30, 2.07s/it]
26%|██▌ | 224/874 [08:58<23:52, 2.20s/it]
26%|██▌ | 225/874 [09:00<24:07, 2.23s/it]
26%|██▌ | 226/874 [09:03<26:32, 2.46s/it]
26%|██▌ | 227/874 [09:05<24:18, 2.25s/it]
26%|██▌ | 228/874 [09:07<25:30, 2.37s/it]
26%|██▌ | 229/874 [09:09<23:33, 2.19s/it]
26%|██▋ | 230/874 [09:12<25:43, 2.40s/it]
{'loss': '1.001', 'grad_norm': '0.9164', 'learning_rate': '9.227e-06', 'epoch': '0.5263'}
|
|||
|
|
26%|██▋ | 230/874 [09:12<25:43, 2.40s/it]
26%|██▋ | 231/874 [09:15<26:43, 2.49s/it]
27%|██▋ | 232/874 [09:18<28:04, 2.62s/it]
27%|██▋ | 233/874 [09:20<27:10, 2.54s/it]
27%|██▋ | 234/874 [09:22<26:04, 2.45s/it]
27%|██▋ | 235/874 [09:24<24:45, 2.32s/it]
27%|██▋ | 236/874 [09:27<25:03, 2.36s/it]
27%|██▋ | 237/874 [09:29<24:27, 2.30s/it]
27%|██▋ | 238/874 [09:31<23:59, 2.26s/it]
27%|██▋ | 239/874 [09:34<26:17, 2.48s/it]
27%|██▋ | 240/874 [09:37<27:57, 2.65s/it]
{'loss': '1.009', 'grad_norm': '1.038', 'learning_rate': '9.117e-06', 'epoch': '0.5492'}
|
|||
|
|
27%|██▋ | 240/874 [09:37<27:57, 2.65s/it]
28%|██▊ | 241/874 [09:39<26:36, 2.52s/it]
28%|██▊ | 242/874 [09:41<25:09, 2.39s/it]
28%|██▊ | 243/874 [09:44<26:06, 2.48s/it]
28%|██▊ | 244/874 [09:47<27:22, 2.61s/it]
28%|██▊ | 245/874 [09:49<26:37, 2.54s/it]
28%|██▊ | 246/874 [09:51<25:11, 2.41s/it]
28%|██▊ | 247/874 [09:54<24:11, 2.31s/it]
28%|██▊ | 248/874 [09:56<25:48, 2.47s/it]
28%|██▊ | 249/874 [09:59<25:04, 2.41s/it]
29%|██▊ | 250/874 [10:01<24:51, 2.39s/it]
{'loss': '1.033', 'grad_norm': '1.005', 'learning_rate': '9e-06', 'epoch': '0.5721'}
|
|||
|
|
29%|██▊ | 250/874 [10:01<24:51, 2.39s/it]
29%|██▊ | 251/874 [10:04<25:24, 2.45s/it]
29%|██▉ | 252/874 [10:06<25:31, 2.46s/it]
29%|██▉ | 253/874 [10:08<24:06, 2.33s/it]
29%|██▉ | 254/874 [10:10<23:19, 2.26s/it]
29%|██▉ | 255/874 [10:12<22:58, 2.23s/it]
29%|██▉ | 256/874 [10:15<24:00, 2.33s/it]
29%|██▉ | 257/874 [10:17<24:06, 2.34s/it]
30%|██▉ | 258/874 [10:20<26:04, 2.54s/it]
30%|██▉ | 259/874 [10:24<28:18, 2.76s/it]
30%|██▉ | 260/874 [10:26<27:17, 2.67s/it]
{'loss': '1.034', 'grad_norm': '1.084', 'learning_rate': '8.877e-06', 'epoch': '0.595'}
|
|||
|
|
30%|██▉ | 260/874 [10:26<27:17, 2.67s/it]
30%|██▉ | 261/874 [10:29<26:56, 2.64s/it]
30%|██▉ | 262/874 [10:31<25:43, 2.52s/it]
30%|███ | 263/874 [10:33<25:49, 2.54s/it]
30%|███ | 264/874 [10:35<24:10, 2.38s/it]
30%|███ | 265/874 [10:38<23:23, 2.31s/it]
30%|███ | 266/874 [10:39<21:39, 2.14s/it]
31%|███ | 267/874 [10:41<21:53, 2.16s/it]
31%|███ | 268/874 [10:44<23:39, 2.34s/it]
31%|███ | 269/874 [10:47<24:16, 2.41s/it]
31%|███ | 270/874 [10:49<22:51, 2.27s/it]
{'loss': '1.047', 'grad_norm': '1.044', 'learning_rate': '8.748e-06', 'epoch': '0.6178'}
|
|||
|
|
31%|███ | 270/874 [10:49<22:51, 2.27s/it]
31%|███ | 271/874 [10:52<24:14, 2.41s/it]
31%|███ | 272/874 [10:54<23:38, 2.36s/it]
31%|███ | 273/874 [10:56<22:32, 2.25s/it]
31%|███▏ | 274/874 [10:58<22:27, 2.25s/it]
31%|███▏ | 275/874 [11:00<23:12, 2.32s/it]
32%|███▏ | 276/874 [11:03<24:31, 2.46s/it]
32%|███▏ | 277/874 [11:06<24:52, 2.50s/it]
32%|███▏ | 278/874 [11:08<24:43, 2.49s/it]
32%|███▏ | 279/874 [11:11<25:03, 2.53s/it]
32%|███▏ | 280/874 [11:13<24:11, 2.44s/it]
{'loss': '1.008', 'grad_norm': '0.8966', 'learning_rate': '8.612e-06', 'epoch': '0.6407'}
|
|||
|
|
32%|███▏ | 280/874 [11:13<24:11, 2.44s/it]
32%|███▏ | 281/874 [11:16<24:13, 2.45s/it]
32%|███▏ | 282/874 [11:18<23:23, 2.37s/it]
32%|███▏ | 283/874 [11:21<26:39, 2.71s/it]
32%|███▏ | 284/874 [11:24<25:42, 2.61s/it]
33%|███▎ | 285/874 [11:26<24:00, 2.44s/it]
33%|███▎ | 286/874 [11:28<22:22, 2.28s/it]
33%|███▎ | 287/874 [11:30<22:06, 2.26s/it]
33%|███▎ | 288/874 [11:32<21:26, 2.20s/it]
33%|███▎ | 289/874 [11:34<22:23, 2.30s/it]
33%|███▎ | 290/874 [11:37<23:01, 2.37s/it]
{'loss': '1.014', 'grad_norm': '1.051', 'learning_rate': '8.471e-06', 'epoch': '0.6636'}
|
|||
|
|
33%|███▎ | 290/874 [11:37<23:01, 2.37s/it]
33%|███▎ | 291/874 [11:39<22:11, 2.28s/it]
33%|███▎ | 292/874 [11:41<21:55, 2.26s/it]
34%|███▎ | 293/874 [11:44<22:40, 2.34s/it]
34%|███▎ | 294/874 [11:46<22:00, 2.28s/it]
34%|███▍ | 295/874 [11:49<24:30, 2.54s/it]
34%|███▍ | 296/874 [11:52<24:43, 2.57s/it]
34%|███▍ | 297/874 [11:54<24:06, 2.51s/it]
34%|███▍ | 298/874 [11:57<25:22, 2.64s/it]
34%|███▍ | 299/874 [11:59<24:18, 2.54s/it]
34%|███▍ | 300/874 [12:02<24:25, 2.55s/it]
{'loss': '1.048', 'grad_norm': '0.909', 'learning_rate': '8.325e-06', 'epoch': '0.6865'}
|
|||
|
|
34%|███▍ | 300/874 [12:02<24:25, 2.55s/it]
34%|███▍ | 301/874 [12:04<23:51, 2.50s/it]
35%|███▍ | 302/874 [12:06<22:19, 2.34s/it]
35%|███▍ | 303/874 [12:09<24:22, 2.56s/it]
35%|███▍ | 304/874 [12:12<23:48, 2.51s/it]
35%|███▍ | 305/874 [12:14<23:51, 2.52s/it]
35%|███▌ | 306/874 [12:17<23:40, 2.50s/it]
35%|███▌ | 307/874 [12:19<22:29, 2.38s/it]
35%|███▌ | 308/874 [12:21<21:19, 2.26s/it]
35%|███▌ | 309/874 [12:23<21:18, 2.26s/it]
35%|███▌ | 310/874 [12:25<21:31, 2.29s/it]
{'loss': '0.9751', 'grad_norm': '0.8913', 'learning_rate': '8.173e-06', 'epoch': '0.7094'}
|
|||
|
|
35%|███▌ | 310/874 [12:25<21:31, 2.29s/it]
36%|███▌ | 311/874 [12:28<22:34, 2.41s/it]
36%|███▌ | 312/874 [12:31<22:48, 2.44s/it]
36%|███▌ | 313/874 [12:33<22:02, 2.36s/it]
36%|███▌ | 314/874 [12:35<20:55, 2.24s/it]
36%|███▌ | 315/874 [12:37<20:38, 2.22s/it]
36%|███▌ | 316/874 [12:39<19:34, 2.11s/it]
36%|███▋ | 317/874 [12:41<20:00, 2.16s/it]
36%|███▋ | 318/874 [12:43<20:39, 2.23s/it]
36%|███▋ | 319/874 [12:46<20:37, 2.23s/it]
37%|███▋ | 320/874 [12:49<22:20, 2.42s/it]
{'loss': '0.9991', 'grad_norm': '0.9734', 'learning_rate': '8.016e-06', 'epoch': '0.7323'}
|
|||
|
|
37%|███▋ | 320/874 [12:49<22:20, 2.42s/it]
37%|███▋ | 321/874 [12:51<21:54, 2.38s/it]
37%|███▋ | 322/874 [12:53<21:16, 2.31s/it]
37%|███▋ | 323/874 [12:55<20:24, 2.22s/it]
37%|███▋ | 324/874 [12:57<19:30, 2.13s/it]
37%|███▋ | 325/874 [12:59<19:11, 2.10s/it]
37%|███▋ | 326/874 [13:01<19:26, 2.13s/it]
37%|███▋ | 327/874 [13:03<20:01, 2.20s/it]
38%|███▊ | 328/874 [13:06<20:03, 2.20s/it]
38%|███▊ | 329/874 [13:08<20:51, 2.30s/it]
38%|███▊ | 330/874 [13:10<20:20, 2.24s/it]
{'loss': '1.023', 'grad_norm': '0.9568', 'learning_rate': '7.854e-06', 'epoch': '0.7551'}
|
|||
|
|
38%|███▊ | 330/874 [13:10<20:20, 2.24s/it]
38%|███▊ | 331/874 [13:13<20:37, 2.28s/it]
38%|███▊ | 332/874 [13:15<21:11, 2.35s/it]
38%|███▊ | 333/874 [13:18<21:31, 2.39s/it]
38%|███▊ | 334/874 [13:19<19:55, 2.21s/it]
38%|███▊ | 335/874 [13:22<21:07, 2.35s/it]
38%|███▊ | 336/874 [13:24<20:21, 2.27s/it]
39%|███▊ | 337/874 [13:27<21:30, 2.40s/it]
39%|███▊ | 338/874 [13:30<22:32, 2.52s/it]
39%|███▉ | 339/874 [13:33<23:34, 2.64s/it]
39%|███▉ | 340/874 [13:34<21:04, 2.37s/it]
{'loss': '1.032', 'grad_norm': '1.213', 'learning_rate': '7.688e-06', 'epoch': '0.778'}
|
|||
|
|
39%|███▉ | 340/874 [13:34<21:04, 2.37s/it]
39%|███▉ | 341/874 [13:37<20:43, 2.33s/it]
39%|███▉ | 342/874 [13:39<20:15, 2.28s/it]
39%|███▉ | 343/874 [13:41<18:50, 2.13s/it]
39%|███▉ | 344/874 [13:43<19:09, 2.17s/it]
39%|███▉ | 345/874 [13:45<20:11, 2.29s/it]
40%|███▉ | 346/874 [13:47<18:51, 2.14s/it]
40%|███▉ | 347/874 [13:50<20:16, 2.31s/it]
40%|███▉ | 348/874 [13:52<19:23, 2.21s/it]
40%|███▉ | 349/874 [13:54<19:08, 2.19s/it]
40%|████ | 350/874 [13:56<19:02, 2.18s/it]
{'loss': '0.9884', 'grad_norm': '0.8978', 'learning_rate': '7.517e-06', 'epoch': '0.8009'}
|
|||
|
|
40%|████ | 350/874 [13:56<19:02, 2.18s/it]
40%|████ | 351/874 [13:58<18:42, 2.15s/it]
40%|████ | 352/874 [14:01<19:01, 2.19s/it]
40%|████ | 353/874 [14:03<18:37, 2.14s/it]
41%|████ | 354/874 [14:05<20:25, 2.36s/it]
41%|████ | 355/874 [14:08<20:27, 2.37s/it]
41%|████ | 356/874 [14:10<19:31, 2.26s/it]
41%|████ | 357/874 [14:12<20:04, 2.33s/it]
41%|████ | 358/874 [14:15<20:50, 2.42s/it]
41%|████ | 359/874 [14:17<20:39, 2.41s/it]
41%|████ | 360/874 [14:19<19:12, 2.24s/it]
{'loss': '0.9716', 'grad_norm': '1.023', 'learning_rate': '7.343e-06', 'epoch': '0.8238'}
|
|||
|
|
41%|████ | 360/874 [14:19<19:12, 2.24s/it]
41%|████▏ | 361/874 [14:21<18:23, 2.15s/it]
41%|████▏ | 362/874 [14:24<20:18, 2.38s/it]
42%|████▏ | 363/874 [14:26<19:19, 2.27s/it]
42%|████▏ | 364/874 [14:28<18:59, 2.23s/it]
42%|████▏ | 365/874 [14:31<20:13, 2.38s/it]
42%|████▏ | 366/874 [14:33<19:21, 2.29s/it]
42%|████▏ | 367/874 [14:36<20:23, 2.41s/it]
42%|████▏ | 368/874 [14:38<19:52, 2.36s/it]
42%|████▏ | 369/874 [14:41<21:39, 2.57s/it]
42%|████▏ | 370/874 [14:43<20:18, 2.42s/it]
{'loss': '1.001', 'grad_norm': '0.8907', 'learning_rate': '7.164e-06', 'epoch': '0.8467'}
|
|||
|
|
42%|████▏ | 370/874 [14:43<20:18, 2.42s/it]
42%|████▏ | 371/874 [14:46<20:34, 2.45s/it]
43%|████▎ | 372/874 [14:48<21:10, 2.53s/it]
43%|████▎ | 373/874 [14:50<20:02, 2.40s/it]
43%|████▎ | 374/874 [14:53<19:29, 2.34s/it]
43%|████▎ | 375/874 [14:55<18:48, 2.26s/it]
43%|████▎ | 376/874 [14:58<20:25, 2.46s/it]
43%|████▎ | 377/874 [15:00<21:05, 2.55s/it]
43%|████▎ | 378/874 [15:03<20:00, 2.42s/it]
43%|████▎ | 379/874 [15:05<19:10, 2.32s/it]
43%|████▎ | 380/874 [15:07<18:34, 2.26s/it]
{'loss': '1.023', 'grad_norm': '0.9992', 'learning_rate': '6.982e-06', 'epoch': '0.8696'}
|
|||
|
|
43%|████▎ | 380/874 [15:07<18:34, 2.26s/it]
44%|████▎ | 381/874 [15:09<19:17, 2.35s/it]
44%|████▎ | 382/874 [15:12<20:06, 2.45s/it]
44%|████▍ | 383/874 [15:15<20:36, 2.52s/it]
44%|████▍ | 384/874 [15:17<20:29, 2.51s/it]
44%|████▍ | 385/874 [15:19<19:34, 2.40s/it]
44%|████▍ | 386/874 [15:22<20:02, 2.46s/it]
44%|████▍ | 387/874 [15:24<20:01, 2.47s/it]
44%|████▍ | 388/874 [15:27<19:57, 2.46s/it]
45%|████▍ | 389/874 [15:29<18:46, 2.32s/it]
45%|████▍ | 390/874 [15:31<18:02, 2.24s/it]
{'loss': '1.031', 'grad_norm': '0.9423', 'learning_rate': '6.797e-06', 'epoch': '0.8924'}
|
|||
|
|
45%|████▍ | 390/874 [15:31<18:02, 2.24s/it]
45%|████▍ | 391/874 [15:33<18:47, 2.34s/it]
45%|████▍ | 392/874 [15:36<19:27, 2.42s/it]
45%|████▍ | 393/874 [15:39<20:48, 2.59s/it]
45%|████▌ | 394/874 [15:42<21:56, 2.74s/it]
45%|████▌ | 395/874 [15:45<21:15, 2.66s/it]
45%|████▌ | 396/874 [15:46<18:49, 2.36s/it]
45%|████▌ | 397/874 [15:49<18:50, 2.37s/it]
46%|████▌ | 398/874 [15:51<19:28, 2.45s/it]
46%|████▌ | 399/874 [15:54<19:02, 2.41s/it]
46%|████▌ | 400/874 [15:56<19:22, 2.45s/it]
{'loss': '0.9651', 'grad_norm': '0.9312', 'learning_rate': '6.61e-06', 'epoch': '0.9153'}
|
|||
|
|
46%|████▌ | 400/874 [15:56<19:22, 2.45s/it]
46%|████▌ | 401/874 [15:59<19:23, 2.46s/it]
46%|████▌ | 402/874 [16:01<19:12, 2.44s/it]
46%|████▌ | 403/874 [16:03<18:10, 2.32s/it]
46%|████▌ | 404/874 [16:05<18:17, 2.33s/it]
46%|████▋ | 405/874 [16:07<16:43, 2.14s/it]
46%|████▋ | 406/874 [16:10<18:27, 2.37s/it]
47%|████▋ | 407/874 [16:12<16:49, 2.16s/it]
47%|████▋ | 408/874 [16:14<17:40, 2.28s/it]
47%|████▋ | 409/874 [16:16<17:08, 2.21s/it]
47%|████▋ | 410/874 [16:19<18:09, 2.35s/it]
{'loss': '0.9788', 'grad_norm': '0.8995', 'learning_rate': '6.419e-06', 'epoch': '0.9382'}
|
|||
|
|
47%|████▋ | 410/874 [16:19<18:09, 2.35s/it]
47%|████▋ | 411/874 [16:21<18:26, 2.39s/it]
47%|████▋ | 412/874 [16:24<19:27, 2.53s/it]
47%|████▋ | 413/874 [16:26<18:07, 2.36s/it]
47%|████▋ | 414/874 [16:29<18:09, 2.37s/it]
47%|████▋ | 415/874 [16:31<17:28, 2.29s/it]
48%|████▊ | 416/874 [16:33<17:18, 2.27s/it]
48%|████▊ | 417/874 [16:35<16:25, 2.16s/it]
48%|████▊ | 418/874 [16:37<16:32, 2.18s/it]
48%|████▊ | 419/874 [16:39<16:33, 2.18s/it]
48%|████▊ | 420/874 [16:41<16:07, 2.13s/it]
{'loss': '1.017', 'grad_norm': '0.9293', 'learning_rate': '6.226e-06', 'epoch': '0.9611'}
|
|||
|
|
48%|████▊ | 420/874 [16:41<16:07, 2.13s/it]
48%|████▊ | 421/874 [16:44<17:33, 2.33s/it]
48%|████▊ | 422/874 [16:47<18:53, 2.51s/it]
48%|████▊ | 423/874 [16:50<20:16, 2.70s/it]
49%|████▊ | 424/874 [16:52<19:10, 2.56s/it]
49%|████▊ | 425/874 [16:54<17:55, 2.40s/it]
49%|████▊ | 426/874 [16:57<17:47, 2.38s/it]
49%|████▉ | 427/874 [16:59<16:53, 2.27s/it]
49%|████▉ | 428/874 [17:01<16:28, 2.22s/it]
49%|████▉ | 429/874 [17:04<18:00, 2.43s/it]
49%|████▉ | 430/874 [17:06<18:00, 2.43s/it]
{'loss': '1.024', 'grad_norm': '1.006', 'learning_rate': '6.032e-06', 'epoch': '0.984'}
|
|||
|
|
49%|████▉ | 430/874 [17:06<18:00, 2.43s/it]
49%|████▉ | 431/874 [17:08<17:14, 2.33s/it]
49%|████▉ | 432/874 [17:10<16:47, 2.28s/it]
50%|████▉ | 433/874 [17:13<16:34, 2.26s/it]
50%|████▉ | 434/874 [17:15<16:54, 2.31s/it]
50%|████▉ | 435/874 [17:17<15:58, 2.18s/it]
50%|████▉ | 436/874 [17:19<16:17, 2.23s/it]
50%|█████ | 437/874 [17:22<16:11, 2.22s/it]
50%|█████ | 438/874 [17:25<17:54, 2.47s/it]
50%|█████ | 439/874 [17:27<17:17, 2.39s/it]
50%|█████ | 440/874 [17:29<16:29, 2.28s/it]
{'loss': '1.038', 'grad_norm': '1.012', 'learning_rate': '5.835e-06', 'epoch': '1.007'}
|
|||
|
|
50%|█████ | 440/874 [17:29<16:29, 2.28s/it]
50%|█████ | 441/874 [17:31<16:04, 2.23s/it]
51%|█████ | 442/874 [17:33<16:28, 2.29s/it]
51%|█████ | 443/874 [17:36<16:19, 2.27s/it]
51%|█████ | 444/874 [17:38<16:52, 2.35s/it]
51%|█████ | 445/874 [17:40<16:27, 2.30s/it]
51%|█████ | 446/874 [17:43<16:28, 2.31s/it]
51%|█████ | 447/874 [17:45<17:00, 2.39s/it]
51%|█████▏ | 448/874 [17:48<17:25, 2.45s/it]
51%|█████▏ | 449/874 [17:50<16:17, 2.30s/it]
51%|█████▏ | 450/874 [17:52<15:50, 2.24s/it]
{'loss': '0.9823', 'grad_norm': '0.893', 'learning_rate': '5.638e-06', 'epoch': '1.03'}
|
|||
|
|
51%|█████▏ | 450/874 [17:52<15:50, 2.24s/it]
52%|█████▏ | 451/874 [17:54<15:22, 2.18s/it]
52%|█████▏ | 452/874 [17:57<18:13, 2.59s/it]
52%|█████▏ | 453/874 [18:00<17:02, 2.43s/it]
52%|█████▏ | 454/874 [18:02<17:27, 2.49s/it]
52%|█████▏ | 455/874 [18:05<18:50, 2.70s/it]
52%|█████▏ | 456/874 [18:08<18:39, 2.68s/it]
52%|█████▏ | 457/874 [18:10<17:38, 2.54s/it]
52%|█████▏ | 458/874 [18:13<17:44, 2.56s/it]
53%|█████▎ | 459/874 [18:15<17:46, 2.57s/it]
53%|█████▎ | 460/874 [18:18<17:14, 2.50s/it]
{'loss': '1.073', 'grad_norm': '0.9717', 'learning_rate': '5.439e-06', 'epoch': '1.053'}
|
|||
|
|
53%|█████▎ | 460/874 [18:18<17:14, 2.50s/it]
53%|█████▎ | 461/874 [18:20<17:25, 2.53s/it]
53%|█████▎ | 462/874 [18:22<16:01, 2.33s/it]
53%|█████▎ | 463/874 [18:25<16:06, 2.35s/it]
53%|█████▎ | 464/874 [18:27<15:48, 2.31s/it]
53%|█████▎ | 465/874 [18:29<15:26, 2.27s/it]
53%|█████▎ | 466/874 [18:31<15:02, 2.21s/it]
53%|█████▎ | 467/874 [18:34<15:49, 2.33s/it]
54%|█████▎ | 468/874 [18:36<15:51, 2.34s/it]
54%|█████▎ | 469/874 [18:38<15:52, 2.35s/it]
54%|█████▍ | 470/874 [18:40<15:08, 2.25s/it]
{'loss': '1.019', 'grad_norm': '0.9065', 'learning_rate': '5.24e-06', 'epoch': '1.076'}
|
|||
|
|
54%|█████▍ | 470/874 [18:40<15:08, 2.25s/it]
54%|█████▍ | 471/874 [18:43<16:22, 2.44s/it]
54%|█████▍ | 472/874 [18:46<16:44, 2.50s/it]
54%|█████▍ | 473/874 [18:48<15:20, 2.30s/it]
54%|█████▍ | 474/874 [18:50<15:14, 2.29s/it]
54%|█████▍ | 475/874 [18:52<14:28, 2.18s/it]
54%|█████▍ | 476/874 [18:54<14:06, 2.13s/it]
55%|█████▍ | 477/874 [18:56<14:52, 2.25s/it]
55%|█████▍ | 478/874 [18:59<15:41, 2.38s/it]
55%|█████▍ | 479/874 [19:01<15:07, 2.30s/it]
55%|█████▍ | 480/874 [19:05<17:31, 2.67s/it]
{'loss': '1.012', 'grad_norm': '0.8013', 'learning_rate': '5.04e-06', 'epoch': '1.098'}
|
|||
|
|
55%|█████▍ | 480/874 [19:05<17:31, 2.67s/it]
55%|█████▌ | 481/874 [19:07<16:19, 2.49s/it]
55%|█████▌ | 482/874 [19:09<16:09, 2.47s/it]
55%|█████▌ | 483/874 [19:11<15:25, 2.37s/it]
55%|█████▌ | 484/874 [19:13<14:32, 2.24s/it]
55%|█████▌ | 485/874 [19:15<13:48, 2.13s/it]
56%|█████▌ | 486/874 [19:18<14:52, 2.30s/it]
56%|█████▌ | 487/874 [19:20<14:59, 2.32s/it]
56%|█████▌ | 488/874 [19:23<14:41, 2.28s/it]
56%|█████▌ | 489/874 [19:25<15:06, 2.36s/it]
56%|█████▌ | 490/874 [19:27<14:51, 2.32s/it]
{'loss': '0.9734', 'grad_norm': '0.989', 'learning_rate': '4.84e-06', 'epoch': '1.121'}
|
|||
|
|
56%|█████▌ | 490/874 [19:27<14:51, 2.32s/it]
56%|█████▌ | 491/874 [19:29<14:17, 2.24s/it]
56%|█████▋ | 492/874 [19:32<14:19, 2.25s/it]
56%|█████▋ | 493/874 [19:34<14:03, 2.22s/it]
57%|█████▋ | 494/874 [19:37<15:16, 2.41s/it]
57%|█████▋ | 495/874 [19:39<15:24, 2.44s/it]
57%|█████▋ | 496/874 [19:42<15:48, 2.51s/it]
57%|█████▋ | 497/874 [19:45<16:27, 2.62s/it]
57%|█████▋ | 498/874 [19:47<15:28, 2.47s/it]
57%|█████▋ | 499/874 [19:49<15:04, 2.41s/it]
57%|█████▋ | 500/874 [19:51<14:33, 2.34s/it]
{'loss': '1.004', 'grad_norm': '0.9131', 'learning_rate': '4.641e-06', 'epoch': '1.144'}
|
|||
|
|
57%|█████▋ | 500/874 [19:51<14:33, 2.34s/it][INFO|trainer.py:3797] 2026-05-24 06:10:20,259 >> Saving model checkpoint to saves/qwen3_1.7b/if/fft/checkpoint-500
|
|||
|
|
[INFO|configuration_utils.py:432] 2026-05-24 06:10:20,262 >> Configuration saved in saves/qwen3_1.7b/if/fft/checkpoint-500/config.json
|
|||
|
|
[INFO|configuration_utils.py:803] 2026-05-24 06:10:20,263 >> Configuration saved in saves/qwen3_1.7b/if/fft/checkpoint-500/generation_config.json
|
|||
|
|
|
|||
|
|
Writing model shards: 0%| | 0/1 [00:00<?, ?it/s][A
|
|||
|
|
Writing model shards: 100%|██████████| 1/1 [00:03<00:00, 3.49s/it][A
Writing model shards: 100%|██████████| 1/1 [00:03<00:00, 3.49s/it]
|
|||
|
|
[INFO|modeling_utils.py:3380] 2026-05-24 06:10:23,766 >> Model weights saved in saves/qwen3_1.7b/if/fft/checkpoint-500/model.safetensors
|
|||
|
|
[INFO|tokenization_utils_base.py:3224] 2026-05-24 06:10:23,767 >> chat template saved in saves/qwen3_1.7b/if/fft/checkpoint-500/chat_template.jinja
|
|||
|
|
[INFO|tokenization_utils_base.py:2078] 2026-05-24 06:10:23,768 >> tokenizer config file saved in saves/qwen3_1.7b/if/fft/checkpoint-500/tokenizer_config.json
|
|||
|
|
57%|█████▋ | 501/874 [20:00<26:24, 4.25s/it]
57%|█████▋ | 502/874 [20:02<22:48, 3.68s/it]
58%|█████▊ | 503/874 [20:05<20:47, 3.36s/it]
58%|█████▊ | 504/874 [20:07<18:27, 2.99s/it]
58%|█████▊ | 505/874 [20:09<16:15, 2.64s/it]
58%|█████▊ | 506/874 [20:11<15:57, 2.60s/it]
58%|█████▊ | 507/874 [20:13<14:56, 2.44s/it]
58%|█████▊ | 508/874 [20:15<14:07, 2.31s/it]
58%|█████▊ | 509/874 [20:18<14:14, 2.34s/it]
58%|█████▊ | 510/874 [20:20<13:51, 2.28s/it]
{'loss': '1.006', 'grad_norm': '0.872', 'learning_rate': '4.442e-06', 'epoch': '1.167'}
|
|||
|
|
58%|█████▊ | 510/874 [20:20<13:51, 2.28s/it]
58%|█████▊ | 511/874 [20:22<13:40, 2.26s/it]
59%|█████▊ | 512/874 [20:24<13:38, 2.26s/it]
59%|█████▊ | 513/874 [20:27<13:39, 2.27s/it]
59%|█████▉ | 514/874 [20:29<13:09, 2.19s/it]
59%|█████▉ | 515/874 [20:31<12:50, 2.15s/it]
59%|█████▉ | 516/874 [20:34<14:26, 2.42s/it]
59%|█████▉ | 517/874 [20:36<13:32, 2.28s/it]
59%|█████▉ | 518/874 [20:38<13:18, 2.24s/it]
59%|█████▉ | 519/874 [20:40<12:42, 2.15s/it]
59%|█████▉ | 520/874 [20:42<13:20, 2.26s/it]
{'loss': '0.9842', 'grad_norm': '0.8993', 'learning_rate': '4.243e-06', 'epoch': '1.19'}
|
|||
|
|
59%|█████▉ | 520/874 [20:42<13:20, 2.26s/it]
60%|█████▉ | 521/874 [20:45<13:40, 2.33s/it]
60%|█████▉ | 522/874 [20:47<12:49, 2.18s/it]
60%|█████▉ | 523/874 [20:50<13:55, 2.38s/it]
60%|█████▉ | 524/874 [20:52<13:37, 2.34s/it]
60%|██████ | 525/874 [20:54<13:12, 2.27s/it]
60%|██████ | 526/874 [20:56<13:04, 2.25s/it]
60%|██████ | 527/874 [20:58<12:34, 2.17s/it]
60%|██████ | 528/874 [21:01<13:17, 2.31s/it]
61%|██████ | 529/874 [21:03<13:35, 2.36s/it]
61%|██████ | 530/874 [21:05<12:34, 2.19s/it]
{'loss': '0.9782', 'grad_norm': '1.01', 'learning_rate': '4.047e-06', 'epoch': '1.213'}
|
|||
|
|
61%|██████ | 530/874 [21:05<12:34, 2.19s/it]
61%|██████ | 531/874 [21:07<12:48, 2.24s/it]
61%|██████ | 532/874 [21:10<13:20, 2.34s/it]
61%|██████ | 533/874 [21:12<13:04, 2.30s/it]
61%|██████ | 534/874 [21:14<12:27, 2.20s/it]
61%|██████ | 535/874 [21:17<13:08, 2.33s/it]
61%|██████▏ | 536/874 [21:19<13:16, 2.36s/it]
61%|██████▏ | 537/874 [21:22<13:59, 2.49s/it]
62%|██████▏ | 538/874 [21:24<13:34, 2.42s/it]
62%|██████▏ | 539/874 [21:27<13:37, 2.44s/it]
62%|██████▏ | 540/874 [21:29<13:46, 2.48s/it]
{'loss': '0.9991', 'grad_norm': '0.9221', 'learning_rate': '3.851e-06', 'epoch': '1.236'}
|
|||
|
|
62%|██████▏ | 540/874 [21:29<13:46, 2.48s/it]
62%|██████▏ | 541/874 [21:32<13:21, 2.41s/it]
62%|██████▏ | 542/874 [21:34<13:43, 2.48s/it]
62%|██████▏ | 543/874 [21:36<13:15, 2.40s/it]
62%|██████▏ | 544/874 [21:39<13:58, 2.54s/it]
62%|██████▏ | 545/874 [21:42<14:42, 2.68s/it]
62%|██████▏ | 546/874 [21:44<13:34, 2.48s/it]
63%|██████▎ | 547/874 [21:47<14:31, 2.66s/it]
63%|██████▎ | 548/874 [21:49<13:26, 2.47s/it]
63%|██████▎ | 549/874 [21:52<13:16, 2.45s/it]
63%|██████▎ | 550/874 [21:54<12:31, 2.32s/it]
{'loss': '1.003', 'grad_norm': '1.045', 'learning_rate': '3.658e-06', 'epoch': '1.259'}
|
|||
|
|
63%|██████▎ | 550/874 [21:54<12:31, 2.32s/it]
63%|██████▎ | 551/874 [21:56<12:09, 2.26s/it]
63%|██████▎ | 552/874 [21:58<12:23, 2.31s/it]
63%|██████▎ | 553/874 [22:01<12:13, 2.28s/it]
63%|██████▎ | 554/874 [22:03<12:54, 2.42s/it]
64%|██████▎ | 555/874 [22:05<12:10, 2.29s/it]
64%|██████▎ | 556/874 [22:08<11:58, 2.26s/it]
64%|██████▎ | 557/874 [22:11<13:26, 2.54s/it]
64%|██████▍ | 558/874 [22:13<13:38, 2.59s/it]
64%|██████▍ | 559/874 [22:16<13:12, 2.52s/it]
64%|██████▍ | 560/874 [22:19<14:40, 2.80s/it]
{'loss': '0.995', 'grad_norm': '0.8805', 'learning_rate': '3.466e-06', 'epoch': '1.281'}
|
|||
|
|
64%|██████▍ | 560/874 [22:19<14:40, 2.80s/it]
64%|██████▍ | 561/874 [22:22<14:22, 2.76s/it]
64%|██████▍ | 562/874 [22:24<13:24, 2.58s/it]
64%|██████▍ | 563/874 [22:26<13:07, 2.53s/it]
65%|██████▍ | 564/874 [22:29<13:45, 2.66s/it]
65%|██████▍ | 565/874 [22:32<13:12, 2.56s/it]
65%|██████▍ | 566/874 [22:34<12:26, 2.42s/it]
65%|██████▍ | 567/874 [22:36<11:56, 2.34s/it]
65%|██████▍ | 568/874 [22:38<11:42, 2.30s/it]
65%|██████▌ | 569/874 [22:40<11:20, 2.23s/it]
65%|██████▌ | 570/874 [22:42<11:11, 2.21s/it]
{'loss': '1.013', 'grad_norm': '0.944', 'learning_rate': '3.277e-06', 'epoch': '1.304'}
|
|||
|
|
65%|██████▌ | 570/874 [22:42<11:11, 2.21s/it]
65%|██████▌ | 571/874 [22:45<11:01, 2.18s/it]
65%|██████▌ | 572/874 [22:47<10:56, 2.18s/it]
66%|██████▌ | 573/874 [22:49<10:22, 2.07s/it]
66%|██████▌ | 574/874 [22:51<10:57, 2.19s/it]
66%|██████▌ | 575/874 [22:53<11:18, 2.27s/it]
66%|██████▌ | 576/874 [22:56<11:35, 2.33s/it]
66%|██████▌ | 577/874 [22:58<11:16, 2.28s/it]
66%|██████▌ | 578/874 [23:01<12:27, 2.53s/it]
66%|██████▌ | 579/874 [23:03<11:27, 2.33s/it]
66%|██████▋ | 580/874 [23:05<11:15, 2.30s/it]
{'loss': '1.035', 'grad_norm': '0.9838', 'learning_rate': '3.091e-06', 'epoch': '1.327'}
|
|||
|
|
66%|██████▋ | 580/874 [23:05<11:15, 2.30s/it]
66%|██████▋ | 581/874 [23:07<10:40, 2.19s/it]
67%|██████▋ | 582/874 [23:09<10:27, 2.15s/it]
67%|██████▋ | 583/874 [23:12<10:36, 2.19s/it]
67%|██████▋ | 584/874 [23:14<11:06, 2.30s/it]
67%|██████▋ | 585/874 [23:17<11:16, 2.34s/it]
67%|██████▋ | 586/874 [23:19<11:39, 2.43s/it]
67%|██████▋ | 587/874 [23:22<11:38, 2.43s/it]
67%|██████▋ | 588/874 [23:25<12:12, 2.56s/it]
67%|██████▋ | 589/874 [23:27<12:09, 2.56s/it]
68%|██████▊ | 590/874 [23:29<11:17, 2.39s/it]
{'loss': '0.9832', 'grad_norm': '1.039', 'learning_rate': '2.908e-06', 'epoch': '1.35'}
|
|||
|
|
68%|██████▊ | 590/874 [23:29<11:17, 2.39s/it]
68%|██████▊ | 591/874 [23:32<11:40, 2.47s/it]
68%|██████▊ | 592/874 [23:34<11:06, 2.36s/it]
68%|██████▊ | 593/874 [23:37<11:47, 2.52s/it]
68%|██████▊ | 594/874 [23:39<11:55, 2.55s/it]
68%|██████▊ | 595/874 [23:41<11:16, 2.42s/it]
68%|██████▊ | 596/874 [23:44<11:18, 2.44s/it]
68%|██████▊ | 597/874 [23:46<11:15, 2.44s/it]
68%|██████▊ | 598/874 [23:49<10:51, 2.36s/it]
69%|██████▊ | 599/874 [23:51<11:02, 2.41s/it]
69%|██████▊ | 600/874 [23:54<11:20, 2.48s/it]
{'loss': '0.9713', 'grad_norm': '0.8741', 'learning_rate': '2.728e-06', 'epoch': '1.373'}
|
|||
|
|
69%|██████▊ | 600/874 [23:54<11:20, 2.48s/it]
69%|██████▉ | 601/874 [23:56<11:13, 2.47s/it]
69%|██████▉ | 602/874 [23:58<10:08, 2.24s/it]
69%|██████▉ | 603/874 [24:00<09:51, 2.18s/it]
69%|██████▉ | 604/874 [24:02<10:04, 2.24s/it]
69%|██████▉ | 605/874 [24:05<09:59, 2.23s/it]
69%|██████▉ | 606/874 [24:07<10:17, 2.31s/it]
69%|██████▉ | 607/874 [24:10<11:09, 2.51s/it]
70%|██████▉ | 608/874 [24:12<10:39, 2.40s/it]
70%|██████▉ | 609/874 [24:14<10:03, 2.28s/it]
70%|██████▉ | 610/874 [24:16<10:08, 2.31s/it]
{'loss': '1.013', 'grad_norm': '1.017', 'learning_rate': '2.552e-06', 'epoch': '1.396'}
|
|||
|
|
70%|██████▉ | 610/874 [24:16<10:08, 2.31s/it]
70%|██████▉ | 611/874 [24:19<10:34, 2.41s/it]
70%|███████ | 612/874 [24:22<10:49, 2.48s/it]
70%|███████ | 613/874 [24:24<10:19, 2.37s/it]
70%|███████ | 614/874 [24:26<09:56, 2.29s/it]
70%|███████ | 615/874 [24:29<10:12, 2.37s/it]
70%|███████ | 616/874 [24:31<10:14, 2.38s/it]
71%|███████ | 617/874 [24:33<09:44, 2.28s/it]
71%|███████ | 618/874 [24:36<10:03, 2.36s/it]
71%|███████ | 619/874 [24:38<10:28, 2.46s/it]
71%|███████ | 620/874 [24:40<09:46, 2.31s/it]
{'loss': '1.03', 'grad_norm': '1.053', 'learning_rate': '2.38e-06', 'epoch': '1.419'}
|
|||
|
|
71%|███████ | 620/874 [24:40<09:46, 2.31s/it]
71%|███████ | 621/874 [24:43<09:55, 2.35s/it]
71%|███████ | 622/874 [24:45<09:23, 2.24s/it]
71%|███████▏ | 623/874 [24:47<10:06, 2.42s/it]
71%|███████▏ | 624/874 [24:50<09:36, 2.30s/it]
72%|███████▏ | 625/874 [24:52<09:35, 2.31s/it]
72%|███████▏ | 626/874 [24:54<09:52, 2.39s/it]
72%|███████▏ | 627/874 [24:57<09:33, 2.32s/it]
72%|███████▏ | 628/874 [24:59<09:20, 2.28s/it]
72%|███████▏ | 629/874 [25:01<09:07, 2.23s/it]
72%|███████▏ | 630/874 [25:03<08:38, 2.13s/it]
{'loss': '1.045', 'grad_norm': '0.8907', 'learning_rate': '2.212e-06', 'epoch': '1.442'}
|
|||
|
|
72%|███████▏ | 630/874 [25:03<08:38, 2.13s/it]
72%|███████▏ | 631/874 [25:05<08:26, 2.09s/it]
72%|███████▏ | 632/874 [25:07<08:47, 2.18s/it]
72%|███████▏ | 633/874 [25:10<09:06, 2.27s/it]
73%|███████▎ | 634/874 [25:12<09:01, 2.26s/it]
73%|███████▎ | 635/874 [25:14<08:55, 2.24s/it]
73%|███████▎ | 636/874 [25:17<09:21, 2.36s/it]
73%|███████▎ | 637/874 [25:19<08:49, 2.23s/it]
73%|███████▎ | 638/874 [25:21<08:52, 2.26s/it]
73%|███████▎ | 639/874 [25:23<08:50, 2.26s/it]
73%|███████▎ | 640/874 [25:26<08:57, 2.30s/it]
{'loss': '1.01', 'grad_norm': '0.9182', 'learning_rate': '2.048e-06', 'epoch': '1.465'}
|
|||
|
|
73%|███████▎ | 640/874 [25:26<08:57, 2.30s/it]
73%|███████▎ | 641/874 [25:28<09:10, 2.36s/it]
73%|███████▎ | 642/874 [25:31<09:23, 2.43s/it]
74%|███████▎ | 643/874 [25:33<09:37, 2.50s/it]
74%|███████▎ | 644/874 [25:35<08:55, 2.33s/it]
74%|███████▍ | 645/874 [25:37<08:23, 2.20s/it]
74%|███████▍ | 646/874 [25:39<08:19, 2.19s/it]
74%|███████▍ | 647/874 [25:42<08:24, 2.22s/it]
74%|███████▍ | 648/874 [25:44<08:24, 2.23s/it]
74%|███████▍ | 649/874 [25:46<08:42, 2.32s/it]
74%|███████▍ | 650/874 [25:49<08:32, 2.29s/it]
{'loss': '0.9954', 'grad_norm': '0.9592', 'learning_rate': '1.889e-06', 'epoch': '1.487'}
|
|||
|
|
74%|███████▍ | 650/874 [25:49<08:32, 2.29s/it]
74%|███████▍ | 651/874 [25:51<08:19, 2.24s/it]
75%|███████▍ | 652/874 [25:53<08:10, 2.21s/it]
75%|███████▍ | 653/874 [25:55<08:30, 2.31s/it]
75%|███████▍ | 654/874 [25:59<09:31, 2.60s/it]
75%|███████▍ | 655/874 [26:01<08:58, 2.46s/it]
75%|███████▌ | 656/874 [26:03<08:59, 2.48s/it]
75%|███████▌ | 657/874 [26:06<08:51, 2.45s/it]
75%|███████▌ | 658/874 [26:08<08:22, 2.33s/it]
75%|███████▌ | 659/874 [26:10<08:25, 2.35s/it]
76%|███████▌ | 660/874 [26:12<08:15, 2.31s/it]
{'loss': '1.039', 'grad_norm': '0.9536', 'learning_rate': '1.735e-06', 'epoch': '1.51'}
|
|||
|
|
76%|███████▌ | 660/874 [26:12<08:15, 2.31s/it]
76%|███████▌ | 661/874 [26:15<08:11, 2.31s/it]
76%|███████▌ | 662/874 [26:17<08:03, 2.28s/it]
76%|███████▌ | 663/874 [26:19<07:46, 2.21s/it]
76%|███████▌ | 664/874 [26:21<07:40, 2.19s/it]
76%|███████▌ | 665/874 [26:23<07:39, 2.20s/it]
76%|███████▌ | 666/874 [26:26<08:15, 2.38s/it]
76%|███████▋ | 667/874 [26:28<08:03, 2.34s/it]
76%|███████▋ | 668/874 [26:31<07:47, 2.27s/it]
77%|███████▋ | 669/874 [26:33<08:05, 2.37s/it]
77%|███████▋ | 670/874 [26:35<07:33, 2.22s/it]
{'loss': '1.023', 'grad_norm': '0.9851', 'learning_rate': '1.587e-06', 'epoch': '1.533'}
|
|||
|
|
77%|███████▋ | 670/874 [26:35<07:33, 2.22s/it]
77%|███████▋ | 671/874 [26:37<07:26, 2.20s/it]
77%|███████▋ | 672/874 [26:40<07:38, 2.27s/it]
77%|███████▋ | 673/874 [26:42<07:50, 2.34s/it]
77%|███████▋ | 674/874 [26:45<08:18, 2.49s/it]
77%|███████▋ | 675/874 [26:47<07:33, 2.28s/it]
77%|███████▋ | 676/874 [26:49<07:27, 2.26s/it]
77%|███████▋ | 677/874 [26:51<06:51, 2.09s/it]
78%|███████▊ | 678/874 [26:53<06:52, 2.10s/it]
78%|███████▊ | 679/874 [26:55<07:23, 2.27s/it]
78%|███████▊ | 680/874 [26:58<07:21, 2.27s/it]
{'loss': '1', 'grad_norm': '0.912', 'learning_rate': '1.443e-06', 'epoch': '1.556'}
|
|||
|
|
78%|███████▊ | 680/874 [26:58<07:21, 2.27s/it]
78%|███████▊ | 681/874 [27:00<07:02, 2.19s/it]
78%|███████▊ | 682/874 [27:02<06:56, 2.17s/it]
78%|███████▊ | 683/874 [27:05<07:41, 2.41s/it]
78%|███████▊ | 684/874 [27:07<07:40, 2.42s/it]
78%|███████▊ | 685/874 [27:10<07:29, 2.38s/it]
78%|███████▊ | 686/874 [27:12<07:19, 2.34s/it]
79%|███████▊ | 687/874 [27:14<07:21, 2.36s/it]
79%|███████▊ | 688/874 [27:16<07:05, 2.29s/it]
79%|███████▉ | 689/874 [27:19<07:15, 2.35s/it]
79%|███████▉ | 690/874 [27:21<07:23, 2.41s/it]
{'loss': '0.9492', 'grad_norm': '0.8398', 'learning_rate': '1.306e-06', 'epoch': '1.579'}
|
|||
|
|
79%|███████▉ | 690/874 [27:21<07:23, 2.41s/it]
79%|███████▉ | 691/874 [27:24<07:49, 2.57s/it]
79%|███████▉ | 692/874 [27:26<07:11, 2.37s/it]
79%|███████▉ | 693/874 [27:28<07:00, 2.32s/it]
79%|███████▉ | 694/874 [27:30<06:31, 2.17s/it]
80%|███████▉ | 695/874 [27:33<06:37, 2.22s/it]
80%|███████▉ | 696/874 [27:35<06:38, 2.24s/it]
80%|███████▉ | 697/874 [27:37<06:17, 2.13s/it]
80%|███████▉ | 698/874 [27:39<06:09, 2.10s/it]
80%|███████▉ | 699/874 [27:41<06:36, 2.26s/it]
80%|████████ | 700/874 [27:44<06:53, 2.38s/it]
{'loss': '1.022', 'grad_norm': '0.9803', 'learning_rate': '1.174e-06', 'epoch': '1.602'}
|
|||
|
|
80%|████████ | 700/874 [27:44<06:53, 2.38s/it]
80%|████████ | 701/874 [27:46<06:49, 2.37s/it]
80%|████████ | 702/874 [27:49<06:43, 2.35s/it]
80%|████████ | 703/874 [27:51<06:22, 2.24s/it]
81%|████████ | 704/874 [27:53<06:26, 2.28s/it]
81%|████████ | 705/874 [27:56<06:39, 2.36s/it]
81%|████████ | 706/874 [27:58<06:35, 2.36s/it]
81%|████████ | 707/874 [28:00<06:18, 2.27s/it]
81%|████████ | 708/874 [28:02<06:22, 2.31s/it]
81%|████████ | 709/874 [28:05<06:55, 2.52s/it]
81%|████████ | 710/874 [28:07<06:20, 2.32s/it]
{'loss': '1.034', 'grad_norm': '1.007', 'learning_rate': '1.048e-06', 'epoch': '1.625'}
|
|||
|
|
81%|████████ | 710/874 [28:07<06:20, 2.32s/it]
81%|████████▏ | 711/874 [28:09<05:57, 2.19s/it]
81%|████████▏ | 712/874 [28:12<06:10, 2.29s/it]
82%|████████▏ | 713/874 [28:14<06:01, 2.24s/it]
82%|████████▏ | 714/874 [28:16<05:51, 2.20s/it]
82%|████████▏ | 715/874 [28:18<05:40, 2.14s/it]
82%|████████▏ | 716/874 [28:20<05:27, 2.07s/it]
82%|████████▏ | 717/874 [28:22<05:26, 2.08s/it]
82%|████████▏ | 718/874 [28:24<05:35, 2.15s/it]
82%|████████▏ | 719/874 [28:27<05:41, 2.21s/it]
82%|████████▏ | 720/874 [28:29<05:40, 2.21s/it]
{'loss': '0.9451', 'grad_norm': '0.881', 'learning_rate': '9.292e-07', 'epoch': '1.648'}
|
|||
|
|
82%|████████▏ | 720/874 [28:29<05:40, 2.21s/it]
82%|████████▏ | 721/874 [28:31<05:47, 2.27s/it]
83%|████████▎ | 722/874 [28:34<05:56, 2.35s/it]
83%|████████▎ | 723/874 [28:36<05:56, 2.36s/it]
83%|████████▎ | 724/874 [28:38<05:38, 2.25s/it]
83%|████████▎ | 725/874 [28:41<05:49, 2.35s/it]
83%|████████▎ | 726/874 [28:43<05:31, 2.24s/it]
83%|████████▎ | 727/874 [28:46<06:15, 2.55s/it]
83%|████████▎ | 728/874 [28:49<06:27, 2.65s/it]
83%|████████▎ | 729/874 [28:51<05:59, 2.48s/it]
84%|████████▎ | 730/874 [28:53<05:48, 2.42s/it]
{'loss': '1.021', 'grad_norm': '1.014', 'learning_rate': '8.165e-07', 'epoch': '1.67'}
|
|||
|
|
84%|████████▎ | 730/874 [28:53<05:48, 2.42s/it]
84%|████████▎ | 731/874 [28:55<05:34, 2.34s/it]
84%|████████▍ | 732/874 [28:59<06:21, 2.69s/it]
84%|████████▍ | 733/874 [29:01<06:17, 2.68s/it]
84%|████████▍ | 734/874 [29:04<06:08, 2.63s/it]
84%|████████▍ | 735/874 [29:07<06:27, 2.79s/it]
84%|████████▍ | 736/874 [29:09<05:47, 2.52s/it]
84%|████████▍ | 737/874 [29:11<05:10, 2.27s/it]
84%|████████▍ | 738/874 [29:13<05:02, 2.23s/it]
85%|████████▍ | 739/874 [29:15<05:12, 2.31s/it]
85%|████████▍ | 740/874 [29:18<05:14, 2.34s/it]
{'loss': '1.003', 'grad_norm': '0.9183', 'learning_rate': '7.104e-07', 'epoch': '1.693'}
|
|||
|
|
85%|████████▍ | 740/874 [29:18<05:14, 2.34s/it]
85%|████████▍ | 741/874 [29:20<04:54, 2.22s/it]
85%|████████▍ | 742/874 [29:22<05:00, 2.28s/it]
85%|████████▌ | 743/874 [29:24<04:56, 2.26s/it]
85%|████████▌ | 744/874 [29:27<04:58, 2.30s/it]
85%|████████▌ | 745/874 [29:29<05:13, 2.43s/it]
85%|████████▌ | 746/874 [29:32<05:15, 2.46s/it]
85%|████████▌ | 747/874 [29:34<05:02, 2.38s/it]
86%|████████▌ | 748/874 [29:37<04:58, 2.37s/it]
86%|████████▌ | 749/874 [29:39<05:00, 2.40s/it]
86%|████████▌ | 750/874 [29:41<04:37, 2.24s/it]
{'loss': '1.002', 'grad_norm': '0.9808', 'learning_rate': '6.112e-07', 'epoch': '1.716'}
|
|||
|
|
86%|████████▌ | 750/874 [29:41<04:37, 2.24s/it]
86%|████████▌ | 751/874 [29:43<04:28, 2.18s/it]
86%|████████▌ | 752/874 [29:46<04:45, 2.34s/it]
86%|████████▌ | 753/874 [29:47<04:22, 2.17s/it]
86%|████████▋ | 754/874 [29:50<04:21, 2.18s/it]
86%|████████▋ | 755/874 [29:52<04:25, 2.23s/it]
86%|████████▋ | 756/874 [29:54<04:12, 2.14s/it]
87%|████████▋ | 757/874 [29:56<03:58, 2.03s/it]
87%|████████▋ | 758/874 [29:58<04:09, 2.15s/it]
87%|████████▋ | 759/874 [30:01<04:15, 2.22s/it]
87%|████████▋ | 760/874 [30:03<04:24, 2.32s/it]
{'loss': '0.9872', 'grad_norm': '0.8875', 'learning_rate': '5.19e-07', 'epoch': '1.739'}
|
|||
|
|
87%|████████▋ | 760/874 [30:03<04:24, 2.32s/it]
87%|████████▋ | 761/874 [30:05<04:04, 2.16s/it]
87%|████████▋ | 762/874 [30:07<04:03, 2.17s/it]
87%|████████▋ | 763/874 [30:09<03:50, 2.07s/it]
87%|████████▋ | 764/874 [30:11<03:45, 2.05s/it]
88%|████████▊ | 765/874 [30:13<03:46, 2.08s/it]
88%|████████▊ | 766/874 [30:16<03:58, 2.21s/it]
88%|████████▊ | 767/874 [30:18<03:55, 2.20s/it]
88%|████████▊ | 768/874 [30:20<03:50, 2.18s/it]
88%|████████▊ | 769/874 [30:22<04:03, 2.32s/it]
88%|████████▊ | 770/874 [30:25<03:52, 2.23s/it]
{'loss': '0.9818', 'grad_norm': '1.024', 'learning_rate': '4.339e-07', 'epoch': '1.762'}
|
|||
|
|
88%|████████▊ | 770/874 [30:25<03:52, 2.23s/it]
88%|████████▊ | 771/874 [30:27<03:47, 2.21s/it]
88%|████████▊ | 772/874 [30:29<03:36, 2.12s/it]
88%|████████▊ | 773/874 [30:31<03:38, 2.16s/it]
89%|████████▊ | 774/874 [30:33<03:46, 2.27s/it]
89%|████████▊ | 775/874 [30:35<03:34, 2.17s/it]
89%|████████▉ | 776/874 [30:37<03:31, 2.16s/it]
89%|████████▉ | 777/874 [30:39<03:24, 2.11s/it]
89%|████████▉ | 778/874 [30:42<03:45, 2.35s/it]
89%|████████▉ | 779/874 [30:45<03:49, 2.42s/it]
89%|████████▉ | 780/874 [30:47<03:47, 2.42s/it]
{'loss': '1.052', 'grad_norm': '0.9245', 'learning_rate': '3.561e-07', 'epoch': '1.785'}
|
|||
|
|
89%|████████▉ | 780/874 [30:47<03:47, 2.42s/it]
89%|████████▉ | 781/874 [30:50<03:52, 2.50s/it]
89%|████████▉ | 782/874 [30:53<03:52, 2.53s/it]
90%|████████▉ | 783/874 [30:55<03:48, 2.51s/it]
90%|████████▉ | 784/874 [30:57<03:39, 2.44s/it]
90%|████████▉ | 785/874 [31:00<03:31, 2.38s/it]
90%|████████▉ | 786/874 [31:02<03:33, 2.43s/it]
90%|█████████ | 787/874 [31:04<03:22, 2.33s/it]
90%|█████████ | 788/874 [31:07<03:23, 2.36s/it]
90%|█████████ | 789/874 [31:09<03:16, 2.31s/it]
90%|█████████ | 790/874 [31:11<03:09, 2.26s/it]
{'loss': '1.019', 'grad_norm': '0.9501', 'learning_rate': '2.858e-07', 'epoch': '1.808'}
|
|||
|
|
90%|█████████ | 790/874 [31:11<03:09, 2.26s/it]
91%|█████████ | 791/874 [31:14<03:20, 2.41s/it]
91%|█████████ | 792/874 [31:16<03:11, 2.34s/it]
91%|█████████ | 793/874 [31:18<03:01, 2.24s/it]
91%|█████████ | 794/874 [31:20<02:55, 2.19s/it]
91%|█████████ | 795/874 [31:22<02:57, 2.25s/it]
91%|█████████ | 796/874 [31:24<02:42, 2.09s/it]
91%|█████████ | 797/874 [31:27<02:48, 2.19s/it]
91%|█████████▏| 798/874 [31:29<02:49, 2.23s/it]
91%|█████████▏| 799/874 [31:31<02:54, 2.32s/it]
92%|█████████▏| 800/874 [31:35<03:12, 2.61s/it]
{'loss': '0.9678', 'grad_norm': '0.9551', 'learning_rate': '2.23e-07', 'epoch': '1.831'}
|
|||
|
|
92%|█████████▏| 800/874 [31:35<03:12, 2.61s/it]
92%|█████████▏| 801/874 [31:37<03:08, 2.58s/it]
92%|█████████▏| 802/874 [31:39<02:55, 2.43s/it]
92%|█████████▏| 803/874 [31:41<02:39, 2.25s/it]
92%|█████████▏| 804/874 [31:43<02:36, 2.24s/it]
92%|█████████▏| 805/874 [31:46<02:44, 2.38s/it]
92%|█████████▏| 806/874 [31:48<02:36, 2.30s/it]
92%|█████████▏| 807/874 [31:51<02:42, 2.42s/it]
92%|█████████▏| 808/874 [31:54<02:52, 2.61s/it]
93%|█████████▎| 809/874 [31:56<02:34, 2.38s/it]
93%|█████████▎| 810/874 [31:59<02:43, 2.55s/it]
{'loss': '1.034', 'grad_norm': '0.8419', 'learning_rate': '1.678e-07', 'epoch': '1.854'}
|
|||
|
|
93%|█████████▎| 810/874 [31:59<02:43, 2.55s/it]
93%|█████████▎| 811/874 [32:01<02:32, 2.42s/it]
93%|█████████▎| 812/874 [32:03<02:22, 2.30s/it]
93%|█████████▎| 813/874 [32:05<02:26, 2.40s/it]
93%|█████████▎| 814/874 [32:07<02:13, 2.22s/it]
93%|█████████▎| 815/874 [32:09<02:08, 2.19s/it]
93%|█████████▎| 816/874 [32:12<02:06, 2.18s/it]
93%|█████████▎| 817/874 [32:14<02:06, 2.22s/it]
94%|█████████▎| 818/874 [32:16<02:00, 2.16s/it]
94%|█████████▎| 819/874 [32:18<02:01, 2.20s/it]
94%|█████████▍| 820/874 [32:21<02:08, 2.38s/it]
{'loss': '1.053', 'grad_norm': '0.9224', 'learning_rate': '1.203e-07', 'epoch': '1.876'}
|
|||
|
|
94%|█████████▍| 820/874 [32:21<02:08, 2.38s/it]
94%|█████████▍| 821/874 [32:23<02:07, 2.40s/it]
94%|█████████▍| 822/874 [32:26<02:02, 2.36s/it]
94%|█████████▍| 823/874 [32:28<02:00, 2.37s/it]
94%|█████████▍| 824/874 [32:31<02:02, 2.45s/it]
94%|█████████▍| 825/874 [32:33<01:54, 2.33s/it]
95%|█████████▍| 826/874 [32:35<01:49, 2.28s/it]
95%|█████████▍| 827/874 [32:37<01:42, 2.17s/it]
95%|█████████▍| 828/874 [32:39<01:43, 2.26s/it]
95%|█████████▍| 829/874 [32:41<01:40, 2.23s/it]
95%|█████████▍| 830/874 [32:44<01:42, 2.32s/it]
{'loss': '0.9446', 'grad_norm': '0.8703', 'learning_rate': '8.066e-08', 'epoch': '1.899'}
|
|||
|
|
95%|█████████▍| 830/874 [32:44<01:42, 2.32s/it]
95%|█████████▌| 831/874 [32:47<01:47, 2.50s/it]
95%|█████████▌| 832/874 [32:49<01:39, 2.37s/it]
95%|█████████▌| 833/874 [32:51<01:36, 2.36s/it]
95%|█████████▌| 834/874 [32:53<01:31, 2.30s/it]
96%|█████████▌| 835/874 [32:56<01:32, 2.38s/it]
96%|█████████▌| 836/874 [32:59<01:33, 2.46s/it]
96%|█████████▌| 837/874 [33:02<01:40, 2.71s/it]
96%|█████████▌| 838/874 [33:04<01:33, 2.59s/it]
96%|█████████▌| 839/874 [33:07<01:27, 2.50s/it]
96%|█████████▌| 840/874 [33:09<01:23, 2.46s/it]
{'loss': '1.002', 'grad_norm': '0.8677', 'learning_rate': '4.885e-08', 'epoch': '1.922'}
|
|||
|
|
96%|█████████▌| 840/874 [33:09<01:23, 2.46s/it]
96%|█████████▌| 841/874 [33:11<01:17, 2.36s/it]
96%|█████████▋| 842/874 [33:13<01:14, 2.32s/it]
96%|█████████▋| 843/874 [33:15<01:08, 2.20s/it]
97%|█████████▋| 844/874 [33:18<01:10, 2.35s/it]
97%|█████████▋| 845/874 [33:21<01:10, 2.43s/it]
97%|█████████▋| 846/874 [33:22<01:02, 2.25s/it]
97%|█████████▋| 847/874 [33:25<01:00, 2.23s/it]
97%|█████████▋| 848/874 [33:27<01:01, 2.35s/it]
97%|█████████▋| 849/874 [33:30<00:58, 2.36s/it]
97%|█████████▋| 850/874 [33:32<00:54, 2.28s/it]
{'loss': '0.9959', 'grad_norm': '1.043', 'learning_rate': '2.494e-08', 'epoch': '1.945'}
|
|||
|
|
97%|█████████▋| 850/874 [33:32<00:54, 2.28s/it]
97%|█████████▋| 851/874 [33:34<00:53, 2.34s/it]
97%|█████████▋| 852/874 [33:37<00:52, 2.38s/it]
98%|█████████▊| 853/874 [33:39<00:48, 2.31s/it]
98%|█████████▊| 854/874 [33:41<00:44, 2.24s/it]
98%|█████████▊| 855/874 [33:43<00:40, 2.15s/it]
98%|█████████▊| 856/874 [33:45<00:39, 2.22s/it]
98%|█████████▊| 857/874 [33:48<00:41, 2.46s/it]
98%|█████████▊| 858/874 [33:51<00:40, 2.53s/it]
98%|█████████▊| 859/874 [33:53<00:37, 2.53s/it]
98%|█████████▊| 860/874 [33:55<00:33, 2.38s/it]
{'loss': '0.9757', 'grad_norm': '0.9434', 'learning_rate': '8.984e-09', 'epoch': '1.968'}
|
|||
|
|
98%|█████████▊| 860/874 [33:55<00:33, 2.38s/it]
99%|█████████▊| 861/874 [33:58<00:30, 2.32s/it]
99%|█████████▊| 862/874 [34:00<00:26, 2.22s/it]
99%|█████████▊| 863/874 [34:03<00:27, 2.46s/it]
99%|█████████▉| 864/874 [34:05<00:25, 2.55s/it]
99%|█████████▉| 865/874 [34:07<00:20, 2.33s/it]
99%|█████████▉| 866/874 [34:10<00:19, 2.49s/it]
99%|█████████▉| 867/874 [34:12<00:17, 2.46s/it]
99%|█████████▉| 868/874 [34:15<00:15, 2.51s/it]
99%|█████████▉| 869/874 [34:17<00:12, 2.45s/it]
100%|█████████▉| 870/874 [34:19<00:09, 2.29s/it]
{'loss': '0.9688', 'grad_norm': '1.053', 'learning_rate': '9.984e-10', 'epoch': '1.991'}
|
|||
|
|
100%|█████████▉| 870/874 [34:19<00:09, 2.29s/it]
100%|█████████▉| 871/874 [34:21<00:06, 2.26s/it]
100%|█████████▉| 872/874 [34:24<00:04, 2.19s/it]
100%|█████████▉| 873/874 [34:25<00:02, 2.12s/it]
100%|██████████| 874/874 [34:28<00:00, 2.30s/it][INFO|trainer.py:3797] 2026-05-24 06:24:57,254 >> Saving model checkpoint to saves/qwen3_1.7b/if/fft/checkpoint-874
|
|||
|
|
[INFO|configuration_utils.py:432] 2026-05-24 06:24:57,257 >> Configuration saved in saves/qwen3_1.7b/if/fft/checkpoint-874/config.json
|
|||
|
|
[INFO|configuration_utils.py:803] 2026-05-24 06:24:57,257 >> Configuration saved in saves/qwen3_1.7b/if/fft/checkpoint-874/generation_config.json
|
|||
|
|
|
|||
|
|
Writing model shards: 0%| | 0/1 [00:00<?, ?it/s][A
|
|||
|
|
Writing model shards: 100%|██████████| 1/1 [00:03<00:00, 3.61s/it][A
Writing model shards: 100%|██████████| 1/1 [00:03<00:00, 3.61s/it]
|
|||
|
|
[INFO|modeling_utils.py:3380] 2026-05-24 06:25:00,874 >> Model weights saved in saves/qwen3_1.7b/if/fft/checkpoint-874/model.safetensors
|
|||
|
|
[INFO|tokenization_utils_base.py:3224] 2026-05-24 06:25:00,875 >> chat template saved in saves/qwen3_1.7b/if/fft/checkpoint-874/chat_template.jinja
|
|||
|
|
[INFO|tokenization_utils_base.py:2078] 2026-05-24 06:25:00,876 >> tokenizer config file saved in saves/qwen3_1.7b/if/fft/checkpoint-874/tokenizer_config.json
|
|||
|
|
[INFO|trainer.py:1863] 2026-05-24 06:25:01,017 >>
|
|||
|
|
|
|||
|
|
Training completed. Do not forget to share your model on huggingface.co/models =)
|
|||
|
|
|
|||
|
|
|
|||
|
|
{'train_runtime': '2076', 'train_samples_per_second': '53.89', 'train_steps_per_second': '0.421', 'train_loss': '1.031', 'epoch': '2'}
|
|||
|
|
100%|██████████| 874/874 [34:34<00:00, 2.30s/it]
100%|██████████| 874/874 [34:34<00:00, 2.37s/it]
|
|||
|
|
[INFO|trainer.py:3797] 2026-05-24 06:25:02,264 >> Saving model checkpoint to saves/qwen3_1.7b/if/fft
|
|||
|
|
[INFO|configuration_utils.py:432] 2026-05-24 06:25:02,267 >> Configuration saved in saves/qwen3_1.7b/if/fft/config.json
|
|||
|
|
[INFO|configuration_utils.py:803] 2026-05-24 06:25:02,267 >> Configuration saved in saves/qwen3_1.7b/if/fft/generation_config.json
|
|||
|
|
Writing model shards: 0%| | 0/1 [00:00<?, ?it/s]
Writing model shards: 100%|██████████| 1/1 [00:03<00:00, 3.78s/it]
Writing model shards: 100%|██████████| 1/1 [00:03<00:00, 3.78s/it]
|
|||
|
|
[INFO|modeling_utils.py:3380] 2026-05-24 06:25:06,052 >> Model weights saved in saves/qwen3_1.7b/if/fft/model.safetensors
|
|||
|
|
[INFO|tokenization_utils_base.py:3224] 2026-05-24 06:25:06,053 >> chat template saved in saves/qwen3_1.7b/if/fft/chat_template.jinja
|
|||
|
|
[INFO|tokenization_utils_base.py:2078] 2026-05-24 06:25:06,054 >> tokenizer config file saved in saves/qwen3_1.7b/if/fft/tokenizer_config.json
|
|||
|
|
[INFO|trainer.py:3797] 2026-05-24 06:25:08,521 >> Saving model checkpoint to saves/qwen3_1.7b/if/fft
|
|||
|
|
[INFO|configuration_utils.py:432] 2026-05-24 06:25:08,524 >> Configuration saved in saves/qwen3_1.7b/if/fft/config.json
|
|||
|
|
[INFO|configuration_utils.py:803] 2026-05-24 06:25:08,524 >> Configuration saved in saves/qwen3_1.7b/if/fft/generation_config.json
|
|||
|
|
Writing model shards: 0%| | 0/1 [00:00<?, ?it/s]
Writing model shards: 100%|██████████| 1/1 [00:39<00:00, 39.78s/it]
Writing model shards: 100%|██████████| 1/1 [00:39<00:00, 39.78s/it]
|
|||
|
|
[INFO|modeling_utils.py:3380] 2026-05-24 06:25:48,312 >> Model weights saved in saves/qwen3_1.7b/if/fft/model.safetensors
|
|||
|
|
[INFO|tokenization_utils_base.py:3224] 2026-05-24 06:25:48,345 >> chat template saved in saves/qwen3_1.7b/if/fft/chat_template.jinja
|
|||
|
|
[INFO|tokenization_utils_base.py:2078] 2026-05-24 06:25:48,393 >> tokenizer config file saved in saves/qwen3_1.7b/if/fft/tokenizer_config.json
|
|||
|
|
[INFO|modelcard.py:266] 2026-05-24 06:25:48,777 >> Dropping the following result as it does not have all the necessary fields:
|
|||
|
|
{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}
|
|||
|
|
Processing Files (0 / 0) : | | 0.00B / 0.00B
|
|||
|
|
New Data Upload : | | 0.00B / 0.00B [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
Processing Files (1 / 3) : 0%| | 11.4MB / 4.07GB, ???B/s
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 0%| | 29.6kB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 0%| | 29.6kB / 4.06GB [A[A[A
Processing Files (1 / 3) : 0%| | 11.5MB / 4.07GB, 1.08MB/s
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 0%| | 636kB / 4.06GB [A[A[A
Processing Files (1 / 3) : 0%| | 12.1MB / 4.07GB, 1.11MB/s
|
|||
|
|
New Data Upload : 0%| | 607kB / 134MB, 58.9kB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 2%|▏ | 65.6MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 2%|▏ | 77.0MB / 4.07GB, 7.34MB/s
|
|||
|
|
New Data Upload : 33%|███▎ | 65.5MB / 201MB, 6.31MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 3%|▎ | 115MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 3%|▎ | 127MB / 4.07GB, 11.9MB/s
|
|||
|
|
New Data Upload : 57%|█████▋ | 115MB / 201MB, 10.9MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 3%|▎ | 133MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 4%|▎ | 145MB / 4.07GB, 13.4MB/s
|
|||
|
|
New Data Upload : 66%|██████▋ | 133MB / 201MB, 12.4MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 3%|▎ | 142MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 4%|▍ | 154MB / 4.07GB, 13.6MB/s
|
|||
|
|
New Data Upload : 53%|█████▎ | 142MB / 268MB, 12.6MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 6%|▌ | 224MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 6%|▌ | 235MB / 4.07GB, 20.9MB/s
|
|||
|
|
New Data Upload : 67%|██████▋ | 224MB / 335MB, 20.0MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 7%|▋ | 296MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 8%|▊ | 307MB / 4.07GB, 27.1MB/s
|
|||
|
|
New Data Upload : 74%|███████▎ | 296MB / 402MB, 26.2MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 8%|▊ | 339MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 9%|▊ | 350MB / 4.07GB, 30.4MB/s
|
|||
|
|
New Data Upload : 72%|███████▏ | 339MB / 469MB, 29.5MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 10%|▉ | 399MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 10%|█ | 410MB / 4.07GB, 35.2MB/s
|
|||
|
|
New Data Upload : 74%|███████▍ | 399MB / 536MB, 34.3MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 10%|█ | 415MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 10%|█ | 426MB / 4.07GB, 35.9MB/s
|
|||
|
|
New Data Upload : 69%|██████▉ | 415MB / 603MB, 35.0MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 12%|█▏ | 488MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 12%|█▏ | 499MB / 4.07GB, 41.6MB/s
|
|||
|
|
New Data Upload : 73%|███████▎ | 488MB / 670MB, 40.8MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 14%|█▍ | 560MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 14%|█▍ | 572MB / 4.07GB, 47.2MB/s
|
|||
|
|
New Data Upload : 76%|███████▌ | 560MB / 738MB, 46.4MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 16%|█▌ | 634MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 16%|█▌ | 645MB / 4.07GB, 52.8MB/s
|
|||
|
|
New Data Upload : 79%|███████▉ | 634MB / 805MB, 52.0MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 17%|█▋ | 685MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 17%|█▋ | 696MB / 4.07GB, 56.2MB/s
|
|||
|
|
New Data Upload : 79%|███████▊ | 685MB / 872MB, 55.4MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 19%|█▉ | 773MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 19%|█▉ | 784MB / 4.07GB, 62.7MB/s
|
|||
|
|
New Data Upload : 82%|████████▏ | 773MB / 939MB, 62.0MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 21%|██ | 842MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 21%|██ | 853MB / 4.07GB, 67.4MB/s
|
|||
|
|
New Data Upload : 84%|████████▎ | 842MB / 1.01GB, 66.7MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 22%|██▏ | 897MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 22%|██▏ | 909MB / 4.07GB, 70.8MB/s
|
|||
|
|
New Data Upload : 84%|████████▎ | 897MB / 1.07GB, 70.1MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 24%|██▍ | 967MB / 4.06GB [A[A[A
Processing Files (1 / 3) : 24%|██▍ | 979MB / 4.07GB, 75.3MB/s
|
|||
|
|
New Data Upload : 85%|████████▍ | 967MB / 1.14GB, 74.6MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 26%|██▌ | 1.05GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 26%|██▌ | 1.06GB / 4.07GB, 80.8MB/s
|
|||
|
|
New Data Upload : 87%|████████▋ | 1.05GB / 1.21GB, 80.0MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 28%|██▊ | 1.12GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 28%|██▊ | 1.14GB / 4.07GB, 85.6MB/s
|
|||
|
|
New Data Upload : 88%|████████▊ | 1.12GB / 1.27GB, 84.8MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 29%|██▉ | 1.19GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 30%|██▉ | 1.20GB / 4.07GB, 89.6MB/s
|
|||
|
|
New Data Upload : 89%|████████▉ | 1.19GB / 1.34GB, 88.8MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 31%|███ | 1.27GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 31%|███▏ | 1.28GB / 4.07GB, 94.2MB/s
|
|||
|
|
New Data Upload : 90%|████████▉ | 1.27GB / 1.41GB, 93.4MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 33%|███▎ | 1.35GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 33%|███▎ | 1.36GB / 4.07GB, 99.0MB/s
|
|||
|
|
New Data Upload : 91%|█████████▏| 1.35GB / 1.48GB, 98.3MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 35%|███▌ | 1.42GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 35%|███▌ | 1.44GB / 4.07GB, 103MB/s
|
|||
|
|
New Data Upload : 92%|█████████▏| 1.42GB / 1.54GB, 103MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 36%|███▋ | 1.48GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 37%|███▋ | 1.49GB / 4.07GB, 106MB/s
|
|||
|
|
New Data Upload : 92%|█████████▏| 1.48GB / 1.61GB, 105MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 37%|███▋ | 1.52GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 38%|███▊ | 1.53GB / 4.07GB, 107MB/s
|
|||
|
|
New Data Upload : 91%|█████████ | 1.52GB / 1.68GB, 106MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 39%|███▉ | 1.59GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 39%|███▉ | 1.60GB / 4.07GB, 111MB/s
|
|||
|
|
New Data Upload : 91%|█████████ | 1.59GB / 1.74GB, 110MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 41%|████ | 1.66GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 41%|████ | 1.68GB / 4.07GB, 115MB/s
|
|||
|
|
New Data Upload : 92%|█████████▏| 1.66GB / 1.81GB, 114MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 43%|████▎ | 1.73GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 43%|████▎ | 1.74GB / 4.07GB, 118MB/s
|
|||
|
|
New Data Upload : 92%|█████████▏| 1.73GB / 1.88GB, 117MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 43%|████▎ | 1.76GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 43%|████▎ | 1.77GB / 4.07GB, 118MB/s
|
|||
|
|
New Data Upload : 90%|█████████ | 1.76GB / 1.94GB, 117MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 45%|████▌ | 1.84GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 45%|████▌ | 1.85GB / 4.07GB, 122MB/s
|
|||
|
|
New Data Upload : 91%|█████████▏| 1.84GB / 2.01GB, 121MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 47%|████▋ | 1.91GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 47%|████▋ | 1.92GB / 4.07GB, 125MB/s
|
|||
|
|
New Data Upload : 92%|█████████▏| 1.91GB / 2.08GB, 124MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 49%|████▉ | 1.98GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 49%|████▉ | 1.99GB / 4.07GB, 129MB/s
|
|||
|
|
New Data Upload : 92%|█████████▏| 1.98GB / 2.15GB, 128MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 51%|█████ | 2.06GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 51%|█████ | 2.08GB / 4.07GB, 133MB/s
|
|||
|
|
New Data Upload : 93%|█████████▎| 2.06GB / 2.21GB, 132MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 53%|█████▎ | 2.15GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 53%|█████▎ | 2.17GB / 4.07GB, 137MB/s
|
|||
|
|
New Data Upload : 94%|█████████▍| 2.15GB / 2.28GB, 137MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 55%|█████▍ | 2.22GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 55%|█████▍ | 2.23GB / 4.07GB, 140MB/s
|
|||
|
|
New Data Upload : 94%|█████████▍| 2.22GB / 2.35GB, 139MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 55%|█████▌ | 2.25GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 56%|█████▌ | 2.26GB / 4.07GB, 140MB/s
|
|||
|
|
New Data Upload : 93%|█████████▎| 2.25GB / 2.41GB, 139MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 57%|█████▋ | 2.33GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 57%|█████▋ | 2.34GB / 4.07GB, 143MB/s
|
|||
|
|
New Data Upload : 94%|█████████▍| 2.33GB / 2.48GB, 143MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 59%|█████▉ | 2.41GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 59%|█████▉ | 2.42GB / 4.07GB, 147MB/s
|
|||
|
|
New Data Upload : 95%|█████████▍| 2.41GB / 2.55GB, 146MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 61%|██████ | 2.48GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 61%|██████ | 2.49GB / 4.07GB, 149MB/s
|
|||
|
|
New Data Upload : 95%|█████████▍| 2.48GB / 2.62GB, 149MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 62%|██████▏ | 2.52GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 62%|██████▏ | 2.53GB / 4.07GB, 150MB/s
|
|||
|
|
New Data Upload : 94%|█████████▍| 2.52GB / 2.68GB, 149MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 64%|██████▎ | 2.58GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 64%|██████▎ | 2.60GB / 4.07GB, 152MB/s
|
|||
|
|
New Data Upload : 94%|█████████▍| 2.58GB / 2.75GB, 151MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 66%|██████▌ | 2.67GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 66%|██████▌ | 2.68GB / 4.07GB, 156MB/s
|
|||
|
|
New Data Upload : 95%|█████████▍| 2.67GB / 2.82GB, 155MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 67%|██████▋ | 2.74GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 68%|██████▊ | 2.75GB / 4.07GB, 158MB/s
|
|||
|
|
New Data Upload : 95%|█████████▌| 2.74GB / 2.88GB, 158MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 69%|██████▉ | 2.80GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 69%|██████▉ | 2.81GB / 4.07GB, 160MB/s
|
|||
|
|
New Data Upload : 95%|█████████▍| 2.80GB / 2.95GB, 159MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 71%|███████ | 2.88GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 71%|███████ | 2.89GB / 4.07GB, 163MB/s
|
|||
|
|
New Data Upload : 95%|█████████▌| 2.88GB / 3.02GB, 163MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 73%|███████▎ | 2.95GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 73%|███████▎ | 2.96GB / 4.07GB, 165MB/s
|
|||
|
|
New Data Upload : 96%|█████████▌| 2.95GB / 3.08GB, 165MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 75%|███████▌ | 3.06GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 75%|███████▌ | 3.07GB / 4.07GB, 170MB/s
|
|||
|
|
New Data Upload : 97%|█████████▋| 3.06GB / 3.15GB, 170MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 76%|███████▋ | 3.10GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 76%|███████▋ | 3.11GB / 4.07GB, 170MB/s
|
|||
|
|
New Data Upload : 96%|█████████▋| 3.10GB / 3.22GB, 170MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 79%|███████▊ | 3.19GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 79%|███████▊ | 3.21GB / 4.07GB, 174MB/s
|
|||
|
|
New Data Upload : 97%|█████████▋| 3.19GB / 3.29GB, 174MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 80%|████████ | 3.26GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 80%|████████ | 3.27GB / 4.07GB, 176MB/s
|
|||
|
|
New Data Upload : 97%|█████████▋| 3.26GB / 3.35GB, 176MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 81%|████████ | 3.30GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 81%|████████▏ | 3.31GB / 4.07GB, 176MB/s
|
|||
|
|
New Data Upload : 96%|█████████▋| 3.30GB / 3.42GB, 175MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 83%|████████▎ | 3.36GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 83%|████████▎ | 3.37GB / 4.07GB, 177MB/s
|
|||
|
|
New Data Upload : 96%|█████████▋| 3.36GB / 3.49GB, 177MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 85%|████████▍ | 3.45GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 85%|████████▍ | 3.46GB / 4.07GB, 181MB/s
|
|||
|
|
New Data Upload : 97%|█████████▋| 3.45GB / 3.55GB, 180MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 87%|████████▋ | 3.53GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 87%|████████▋ | 3.54GB / 4.07GB, 183MB/s
|
|||
|
|
New Data Upload : 97%|█████████▋| 3.53GB / 3.62GB, 183MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 88%|████████▊ | 3.60GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 89%|████████▊ | 3.61GB / 4.07GB, 185MB/s
|
|||
|
|
New Data Upload : 97%|█████████▋| 3.59GB / 3.69GB, 184MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 89%|████████▉ | 3.63GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 89%|████████▉ | 3.64GB / 4.07GB, 184MB/s
|
|||
|
|
New Data Upload : 97%|█████████▋| 3.63GB / 3.76GB, 184MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 90%|█████████ | 3.67GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 90%|█████████ | 3.68GB / 4.07GB, 183MB/s
|
|||
|
|
New Data Upload : 96%|█████████▌| 3.67GB / 3.82GB, 183MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 93%|█████████▎| 3.77GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 93%|█████████▎| 3.78GB / 4.07GB, 188MB/s
|
|||
|
|
New Data Upload : 97%|█████████▋| 3.77GB / 3.89GB, 187MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 94%|█████████▍| 3.82GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 94%|█████████▍| 3.83GB / 4.07GB, 188MB/s
|
|||
|
|
New Data Upload : 97%|█████████▋| 3.82GB / 3.96GB, 188MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 96%|█████████▌| 3.90GB / 4.06GB [A[A[A
Processing Files (1 / 3) : 96%|█████████▌| 3.91GB / 4.07GB, 190MB/s
|
|||
|
|
New Data Upload : 97%|█████████▋| 3.89GB / 4.02GB, 190MB/s [A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 42%|████▏ | 3.01kB / 7.18kB [A[A[A[A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 98%|█████████▊| 3.99GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 42%|████▏ | 3.01kB / 7.18kB [A[A[A[A
Processing Files (1 / 3) : 98%|█████████▊| 4.01GB / 4.07GB, 194MB/s
|
|||
|
|
New Data Upload : 98%|█████████▊| 3.99GB / 4.06GB, 194MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 100%|█████████▉| 4.06GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 99%|█████████▉| 7.14kB / 7.18kB [A[A[A[A
Processing Files (1 / 3) : 100%|█████████▉| 4.07GB / 4.07GB, 195MB/s
|
|||
|
|
New Data Upload : 100%|█████████▉| 4.06GB / 4.06GB, 195MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 100%|█████████▉| 4.06GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A[A[A
Processing Files (2 / 3) : 100%|█████████▉| 4.07GB / 4.07GB, 188MB/s
|
|||
|
|
New Data Upload : 100%|█████████▉| 4.06GB / 4.06GB, 188MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 100%|█████████▉| 4.06GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A[A[A
Processing Files (2 / 3) : 100%|█████████▉| 4.07GB / 4.07GB, 185MB/s
|
|||
|
|
New Data Upload : 100%|█████████▉| 4.06GB / 4.06GB, 184MB/s [A
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 100%|██████████| 4.06GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A[A[A
Processing Files (3 / 3) : 100%|██████████| 4.07GB / 4.07GB, 181MB/s
|
|||
|
|
New Data Upload : 100%|██████████| 4.06GB / 4.06GB, 181MB/s [A
Processing Files (3 / 3) : 100%|██████████| 4.07GB / 4.07GB, 181MB/s
|
|||
|
|
New Data Upload : 100%|██████████| 4.06GB / 4.06GB, 181MB/s
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB
|
|||
|
|
.../if/fft/model.safetensors: 100%|██████████| 4.06GB / 4.06GB
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB
|
|||
|
|
***** train metrics *****
|
|||
|
|
epoch = 2.0
|
|||
|
|
total_flos = 2153705312GF
|
|||
|
|
train_loss = 1.0308
|
|||
|
|
train_runtime = 0:34:35.84
|
|||
|
|
train_samples_per_second = 53.887
|
|||
|
|
train_steps_per_second = 0.421
|
|||
|
|
Figure saved at: saves/qwen3_1.7b/if/fft/training_loss.png
|
|||
|
|
[WARNING|2026-05-24 06:26:11] llamafactory.extras.ploting:149 >> No metric eval_loss to plot.
|
|||
|
|
[WARNING|2026-05-24 06:26:11] llamafactory.extras.ploting:149 >> No metric eval_accuracy to plot.
|
|||
|
|
[INFO|trainer.py:3797] 2026-05-24 06:26:12,567 >> Saving model checkpoint to saves/qwen3_1.7b/if/fft
|
|||
|
|
[INFO|configuration_utils.py:432] 2026-05-24 06:26:12,612 >> Configuration saved in saves/qwen3_1.7b/if/fft/config.json
|
|||
|
|
[INFO|configuration_utils.py:803] 2026-05-24 06:26:12,661 >> Configuration saved in saves/qwen3_1.7b/if/fft/generation_config.json
|
|||
|
|
Writing model shards: 0%| | 0/1 [00:00<?, ?it/s]
Writing model shards: 100%|██████████| 1/1 [00:36<00:00, 36.39s/it]
Writing model shards: 100%|██████████| 1/1 [00:36<00:00, 36.39s/it]
|
|||
|
|
[INFO|modeling_utils.py:3380] 2026-05-24 06:26:49,059 >> Model weights saved in saves/qwen3_1.7b/if/fft/model.safetensors
|
|||
|
|
[INFO|tokenization_utils_base.py:3224] 2026-05-24 06:26:49,060 >> chat template saved in saves/qwen3_1.7b/if/fft/chat_template.jinja
|
|||
|
|
[INFO|tokenization_utils_base.py:2078] 2026-05-24 06:26:49,060 >> tokenizer config file saved in saves/qwen3_1.7b/if/fft/tokenizer_config.json
|
|||
|
|
[INFO|modelcard.py:266] 2026-05-24 06:26:49,239 >> Dropping the following result as it does not have all the necessary fields:
|
|||
|
|
{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}
|
|||
|
|
Processing Files (0 / 0) : | | 0.00B / 0.00B
|
|||
|
|
New Data Upload : | | 0.00B / 0.00B [A
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 9%|▉ | 368MB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 9%|▉ | 368MB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
Processing Files (2 / 3) : 9%|▉ | 379MB / 4.07GB, ???B/s
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 18%|█▊ | 736MB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
Processing Files (2 / 3) : 18%|█▊ | 747MB / 4.07GB, ???B/s
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 27%|██▋ | 1.11GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
Processing Files (2 / 3) : 28%|██▊ | 1.12GB / 4.07GB, 108MB/s
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 37%|███▋ | 1.49GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
Processing Files (2 / 3) : 37%|███▋ | 1.50GB / 4.07GB, 142MB/s
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 46%|████▌ | 1.86GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
Processing Files (2 / 3) : 46%|████▌ | 1.88GB / 4.07GB, 175MB/s
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 55%|█████▌ | 2.24GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
Processing Files (2 / 3) : 55%|█████▌ | 2.25GB / 4.07GB, 206MB/s
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 64%|██████▍ | 2.62GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
Processing Files (2 / 3) : 64%|██████▍ | 2.63GB / 4.07GB, 237MB/s
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 74%|███████▎ | 2.99GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
Processing Files (2 / 3) : 74%|███████▎ | 3.00GB / 4.07GB, 267MB/s
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 83%|████████▎ | 3.36GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
Processing Files (2 / 3) : 83%|████████▎ | 3.37GB / 4.07GB, 295MB/s
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 92%|█████████▏| 3.73GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
Processing Files (2 / 3) : 92%|█████████▏| 3.74GB / 4.07GB, 323MB/s
|
|||
|
|
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB [A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
.../if/fft/model.safetensors: 100%|██████████| 4.06GB / 4.06GB [A[A[A
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB [A[A[A[A
Processing Files (3 / 3) : 100%|██████████| 4.07GB / 4.07GB, 346MB/s
Processing Files (3 / 3) : 100%|██████████| 4.07GB / 4.07GB, 346MB/s
|
|||
|
|
New Data Upload : | | 0.00B / 0.00B, 0.00B/s
|
|||
|
|
.../if/fft/training_args.bin: 100%|██████████| 7.18kB / 7.18kB
|
|||
|
|
.../if/fft/model.safetensors: 100%|██████████| 4.06GB / 4.06GB
|
|||
|
|
....7b/if/fft/tokenizer.json: 100%|██████████| 11.4MB / 11.4MB
|