commit eb83f4e7842c8dca84c5a70e0b79c19be6c6691a Author: ModelHub XC Date: Sat Sep 19 17:42:13 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: mlfoundations-dev/c1_code_10d_4s_10k Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..9147b15 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,56 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text + + +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zstandard filter=lfs diff=lfs merge=lfs -text +*.tfevents* filter=lfs diff=lfs merge=lfs -text +*.db* filter=lfs diff=lfs merge=lfs -text +*.ark* filter=lfs diff=lfs merge=lfs -text +**/*ckpt*data* filter=lfs diff=lfs merge=lfs -text +**/*ckpt*.meta filter=lfs diff=lfs merge=lfs -text +**/*ckpt*.index filter=lfs diff=lfs merge=lfs -text + +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.gguf* filter=lfs diff=lfs merge=lfs -text +*.ggml filter=lfs diff=lfs merge=lfs -text +*.llamafile* filter=lfs diff=lfs merge=lfs -text +*.pt2 filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text + +vocab.json filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text +model-00001-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +model-00003-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +merges.txt filter=lfs diff=lfs merge=lfs -text +training_args.bin filter=lfs diff=lfs merge=lfs -text +model-00002-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +model-00004-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text \ No newline at end of file diff --git a/README.md b/README.md new file mode 100644 index 0000000..c6d3d54 --- /dev/null +++ b/README.md @@ -0,0 +1,61 @@ +--- +library_name: transformers +license: apache-2.0 +base_model: Qwen/Qwen2.5-7B-Instruct +tags: +- llama-factory +- full +- generated_from_trainer +model-index: +- name: c1_code_10d_4s_10k + results: [] +--- + + + +# c1_code_10d_4s_10k + +This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/c1_code_10d_4s_10k dataset. + +## Model description + +More information needed + +## Intended uses & limitations + +More information needed + +## Training and evaluation data + +More information needed + +## Training procedure + +### Training hyperparameters + +The following hyperparameters were used during training: +- learning_rate: 4e-05 +- train_batch_size: 1 +- eval_batch_size: 8 +- seed: 42 +- distributed_type: multi-GPU +- num_devices: 16 +- gradient_accumulation_steps: 8 +- total_train_batch_size: 128 +- total_eval_batch_size: 128 +- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments +- lr_scheduler_type: cosine +- lr_scheduler_warmup_ratio: 0.1 +- num_epochs: 5.0 + +### Training results + + + +### Framework versions + +- Transformers 4.46.1 +- Pytorch 2.5.1 +- Datasets 3.0.2 +- Tokenizers 0.20.3 diff --git a/added_tokens.json b/added_tokens.json new file mode 100644 index 0000000..482ced4 --- /dev/null +++ b/added_tokens.json @@ -0,0 +1,24 @@ +{ + "": 151658, + "": 151657, + "<|box_end|>": 151649, + "<|box_start|>": 151648, + "<|endoftext|>": 151643, + "<|file_sep|>": 151664, + "<|fim_middle|>": 151660, + "<|fim_pad|>": 151662, + "<|fim_prefix|>": 151659, + "<|fim_suffix|>": 151661, + "<|im_end|>": 151645, + "<|im_start|>": 151644, + "<|image_pad|>": 151655, + "<|object_ref_end|>": 151647, + "<|object_ref_start|>": 151646, + "<|quad_end|>": 151651, + "<|quad_start|>": 151650, + "<|repo_name|>": 151663, + "<|video_pad|>": 151656, + "<|vision_end|>": 151653, + "<|vision_pad|>": 151654, + "<|vision_start|>": 151652 +} diff --git a/all_results.json b/all_results.json new file mode 100644 index 0000000..53d9689 --- /dev/null +++ b/all_results.json @@ -0,0 +1,8 @@ +{ + "epoch": 4.992, + "total_flos": 1.896730808245289e+18, + "train_loss": 0.5385960905979841, + "train_runtime": 18314.8306, + "train_samples_per_second": 2.73, + "train_steps_per_second": 0.021 +} \ No newline at end of file diff --git a/config.json b/config.json new file mode 100644 index 0000000..6c52571 --- /dev/null +++ b/config.json @@ -0,0 +1,29 @@ +{ + "_name_or_path": "Qwen/Qwen2.5-7B-Instruct", + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "bos_token_id": 151643, + "eos_token_id": 151645, + "hidden_act": "silu", + "hidden_size": 3584, + "initializer_range": 0.02, + "intermediate_size": 18944, + "max_position_embeddings": 32768, + "max_window_layers": 28, + "model_type": "qwen2", + "num_attention_heads": 28, + "num_hidden_layers": 28, + "num_key_value_heads": 4, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "sliding_window": null, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.46.1", + "use_cache": false, + "use_sliding_window": false, + "vocab_size": 152064 +} diff --git a/configs.yaml b/configs.yaml new file mode 100644 index 0000000..a03f71d --- /dev/null +++ b/configs.yaml @@ -0,0 +1,37 @@ +assistant_tag: gpt +bf16: true +content_tag: value +cutoff_len: 16384 +dataloader_num_workers: 4 +dataloader_persistent_workers: true +dataloader_pin_memory: true +dataset: mlfoundations-dev/c1_code_10d_4s_10k +dataset_dir: ONLINE +ddp_timeout: 180000000 +deepspeed: dcft/train/zero3.json +do_train: true +enable_liger_kernel: true +finetuning_type: full +global_batch_size: 128 +gradient_accumulation_steps: 8 +hub_model_id: mlfoundations-dev/c1_code_10d_4s_10k +learning_rate: 4.0e-05 +logging_steps: 1 +lr_scheduler_type: cosine +messages: conversations +model_name_or_path: Qwen/Qwen2.5-7B-Instruct +num_train_epochs: 5.0 +output_dir: /scratch/08002/gsmyrnis/checkpoints/c1_code_10d_4s_10k +overwrite_cache: true +per_device_train_batch_size: 1 +plot_loss: true +preprocessing_num_workers: 16 +push_to_db: true +push_to_hub: true +report_to: wandb +role_tag: from +save_strategy: epoch +stage: sft +template: qwen25 +user_tag: human +warmup_ratio: 0.1 diff --git a/configuration.json b/configuration.json new file mode 100644 index 0000000..bbeeda1 --- /dev/null +++ b/configuration.json @@ -0,0 +1 @@ +{"framework": "pytorch", "task": "text-generation", "allow_remote": true} \ No newline at end of file diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000..a753841 --- /dev/null +++ b/generation_config.json @@ -0,0 +1,14 @@ +{ + "bos_token_id": 151643, + "do_sample": true, + "eos_token_id": [ + 151645, + 151643 + ], + "pad_token_id": 151643, + "repetition_penalty": 1.05, + "temperature": 0.7, + "top_k": 20, + "top_p": 0.8, + "transformers_version": "4.46.1" +} diff --git a/merges.txt b/merges.txt new file mode 100644 index 0000000..80c1a19 --- /dev/null +++ b/merges.txt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5 +size 1671853 diff --git a/model-00001-of-00004.safetensors b/model-00001-of-00004.safetensors new file mode 100644 index 0000000..37ca075 --- /dev/null +++ b/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:43c0b55dd14c2b3a5f3025afd2d436ca19bf547f34b366b53b9e4d89faa16be4 +size 4877660776 diff --git a/model-00002-of-00004.safetensors b/model-00002-of-00004.safetensors new file mode 100644 index 0000000..32013aa --- /dev/null +++ b/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0bcbb513128b24b174d7b97a3ade1e400efc7108b8a425cfdeb871b8c56db1b4 +size 4932751008 diff --git a/model-00003-of-00004.safetensors b/model-00003-of-00004.safetensors new file mode 100644 index 0000000..64cf8a3 --- /dev/null +++ b/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ef7da0966d405f8653252b8939466e63a0b8fc6dca2b30d5fa964695a875556 +size 4330865200 diff --git a/model-00004-of-00004.safetensors b/model-00004-of-00004.safetensors new file mode 100644 index 0000000..4e09ed0 --- /dev/null +++ b/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7b564142d1989fd590a7593a65f03ee38cdff3f73d95193dc8c0a2c92b3a07c7 +size 1089994880 diff --git a/model.safetensors.index.json b/model.safetensors.index.json new file mode 100644 index 0000000..6ca5084 --- /dev/null +++ b/model.safetensors.index.json @@ -0,0 +1,346 @@ +{ + "metadata": { + "total_size": 15231233024 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.bias": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.bias": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00003-of-00004.safetensors" + } +} diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000..17305b3 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,31 @@ +{ + "additional_special_tokens": [ + "<|im_start|>", + "<|im_end|>", + "<|object_ref_start|>", + "<|object_ref_end|>", + "<|box_start|>", + "<|box_end|>", + "<|quad_start|>", + "<|quad_end|>", + "<|vision_start|>", + "<|vision_end|>", + "<|vision_pad|>", + "<|image_pad|>", + "<|video_pad|>" + ], + "eos_token": { + "content": "<|endoftext|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|endoftext|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000..51ebb3b --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa +size 11421896 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..b84f53a --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,208 @@ +{ + "add_bos_token": false, + "add_prefix_space": false, + "added_tokens_decoder": { + "151643": { + "content": "<|endoftext|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151644": { + "content": "<|im_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151645": { + "content": "<|im_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151646": { + "content": "<|object_ref_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151647": { + "content": "<|object_ref_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151648": { + "content": "<|box_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151649": { + "content": "<|box_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151650": { + "content": "<|quad_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151651": { + "content": "<|quad_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151652": { + "content": "<|vision_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151653": { + "content": "<|vision_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151654": { + "content": "<|vision_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151655": { + "content": "<|image_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151656": { + "content": "<|video_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151657": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151658": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151659": { + "content": "<|fim_prefix|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151660": { + "content": "<|fim_middle|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151661": { + "content": "<|fim_suffix|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151662": { + "content": "<|fim_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151663": { + "content": "<|repo_name|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151664": { + "content": "<|file_sep|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + } + }, + "additional_special_tokens": [ + "<|im_start|>", + "<|im_end|>", + "<|object_ref_start|>", + "<|object_ref_end|>", + "<|box_start|>", + "<|box_end|>", + "<|quad_start|>", + "<|quad_end|>", + "<|vision_start|>", + "<|vision_end|>", + "<|vision_pad|>", + "<|image_pad|>", + "<|video_pad|>" + ], + "bos_token": null, + "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within XML tags:\\n\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n\\n\\nFor each function call, return a json object with function name and arguments within XML tags:\\n\\n{\\\"name\\\": , \\\"arguments\\\": }\\n<|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n\\n' }}\n {{- message.content }}\n {{- '\\n' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n", + "clean_up_tokenization_spaces": false, + "eos_token": "<|endoftext|>", + "errors": "replace", + "model_max_length": 131072, + "pad_token": "<|endoftext|>", + "padding_side": "right", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +} diff --git a/train_results.json b/train_results.json new file mode 100644 index 0000000..53d9689 --- /dev/null +++ b/train_results.json @@ -0,0 +1,8 @@ +{ + "epoch": 4.992, + "total_flos": 1.896730808245289e+18, + "train_loss": 0.5385960905979841, + "train_runtime": 18314.8306, + "train_samples_per_second": 2.73, + "train_steps_per_second": 0.021 +} \ No newline at end of file diff --git a/trainer_log.jsonl b/trainer_log.jsonl new file mode 100644 index 0000000..28552c1 --- /dev/null +++ b/trainer_log.jsonl @@ -0,0 +1,391 @@ +{"current_steps": 1, "total_steps": 390, "loss": 1.0426, "lr": 1.0256410256410257e-06, "epoch": 0.0128, "percentage": 0.26, "elapsed_time": "0:00:53", "remaining_time": "5:45:29"} +{"current_steps": 2, "total_steps": 390, "loss": 1.0654, "lr": 2.0512820512820513e-06, "epoch": 0.0256, "percentage": 0.51, "elapsed_time": "0:01:38", "remaining_time": "5:19:10"} +{"current_steps": 3, "total_steps": 390, "loss": 1.038, "lr": 3.0769230769230774e-06, "epoch": 0.0384, "percentage": 0.77, "elapsed_time": "0:02:24", "remaining_time": "5:09:38"} +{"current_steps": 4, "total_steps": 390, "loss": 1.0223, "lr": 4.102564102564103e-06, "epoch": 0.0512, "percentage": 1.03, "elapsed_time": "0:03:09", "remaining_time": "5:05:15"} +{"current_steps": 5, "total_steps": 390, "loss": 1.0198, "lr": 5.128205128205128e-06, "epoch": 0.064, "percentage": 1.28, "elapsed_time": "0:03:53", "remaining_time": "4:59:03"} +{"current_steps": 6, "total_steps": 390, "loss": 0.9696, "lr": 6.153846153846155e-06, "epoch": 0.0768, "percentage": 1.54, "elapsed_time": "0:04:38", "remaining_time": "4:56:46"} +{"current_steps": 7, "total_steps": 390, "loss": 0.9539, "lr": 7.17948717948718e-06, "epoch": 0.0896, "percentage": 1.79, "elapsed_time": "0:05:24", "remaining_time": "4:56:00"} +{"current_steps": 8, "total_steps": 390, "loss": 0.9903, "lr": 8.205128205128205e-06, "epoch": 0.1024, "percentage": 2.05, "elapsed_time": "0:06:08", "remaining_time": "4:53:27"} +{"current_steps": 9, "total_steps": 390, "loss": 0.9684, "lr": 9.230769230769232e-06, "epoch": 0.1152, "percentage": 2.31, "elapsed_time": "0:06:54", "remaining_time": "4:52:35"} +{"current_steps": 10, "total_steps": 390, "loss": 0.9146, "lr": 1.0256410256410256e-05, "epoch": 0.128, "percentage": 2.56, "elapsed_time": "0:07:39", "remaining_time": "4:51:09"} +{"current_steps": 11, "total_steps": 390, "loss": 0.9168, "lr": 1.1282051282051283e-05, "epoch": 0.1408, "percentage": 2.82, "elapsed_time": "0:08:24", "remaining_time": "4:49:37"} +{"current_steps": 12, "total_steps": 390, "loss": 0.89, "lr": 1.230769230769231e-05, "epoch": 0.1536, "percentage": 3.08, "elapsed_time": "0:09:10", "remaining_time": "4:48:55"} +{"current_steps": 13, "total_steps": 390, "loss": 0.8776, "lr": 1.3333333333333333e-05, "epoch": 0.1664, "percentage": 3.33, "elapsed_time": "0:09:55", "remaining_time": "4:47:49"} +{"current_steps": 14, "total_steps": 390, "loss": 0.8364, "lr": 1.435897435897436e-05, "epoch": 0.1792, "percentage": 3.59, "elapsed_time": "0:10:41", "remaining_time": "4:47:01"} +{"current_steps": 15, "total_steps": 390, "loss": 0.8212, "lr": 1.5384615384615387e-05, "epoch": 0.192, "percentage": 3.85, "elapsed_time": "0:11:27", "remaining_time": "4:46:30"} +{"current_steps": 16, "total_steps": 390, "loss": 0.8171, "lr": 1.641025641025641e-05, "epoch": 0.2048, "percentage": 4.1, "elapsed_time": "0:12:13", "remaining_time": "4:45:52"} +{"current_steps": 17, "total_steps": 390, "loss": 0.7992, "lr": 1.7435897435897438e-05, "epoch": 0.2176, "percentage": 4.36, "elapsed_time": "0:12:59", "remaining_time": "4:44:52"} +{"current_steps": 18, "total_steps": 390, "loss": 0.7841, "lr": 1.8461538461538465e-05, "epoch": 0.2304, "percentage": 4.62, "elapsed_time": "0:13:45", "remaining_time": "4:44:17"} +{"current_steps": 19, "total_steps": 390, "loss": 0.788, "lr": 1.9487179487179488e-05, "epoch": 0.2432, "percentage": 4.87, "elapsed_time": "0:14:28", "remaining_time": "4:42:45"} +{"current_steps": 20, "total_steps": 390, "loss": 0.7642, "lr": 2.0512820512820512e-05, "epoch": 0.256, "percentage": 5.13, "elapsed_time": "0:15:13", "remaining_time": "4:41:36"} +{"current_steps": 21, "total_steps": 390, "loss": 0.7661, "lr": 2.153846153846154e-05, "epoch": 0.2688, "percentage": 5.38, "elapsed_time": "0:15:59", "remaining_time": "4:41:02"} +{"current_steps": 22, "total_steps": 390, "loss": 0.7551, "lr": 2.2564102564102566e-05, "epoch": 0.2816, "percentage": 5.64, "elapsed_time": "0:16:45", "remaining_time": "4:40:20"} +{"current_steps": 23, "total_steps": 390, "loss": 0.744, "lr": 2.3589743589743593e-05, "epoch": 0.2944, "percentage": 5.9, "elapsed_time": "0:17:30", "remaining_time": "4:39:29"} +{"current_steps": 24, "total_steps": 390, "loss": 0.7488, "lr": 2.461538461538462e-05, "epoch": 0.3072, "percentage": 6.15, "elapsed_time": "0:18:14", "remaining_time": "4:38:17"} +{"current_steps": 25, "total_steps": 390, "loss": 0.7256, "lr": 2.5641025641025646e-05, "epoch": 0.32, "percentage": 6.41, "elapsed_time": "0:18:57", "remaining_time": "4:36:51"} +{"current_steps": 26, "total_steps": 390, "loss": 0.737, "lr": 2.6666666666666667e-05, "epoch": 0.3328, "percentage": 6.67, "elapsed_time": "0:19:43", "remaining_time": "4:36:05"} +{"current_steps": 27, "total_steps": 390, "loss": 0.7581, "lr": 2.7692307692307694e-05, "epoch": 0.3456, "percentage": 6.92, "elapsed_time": "0:20:25", "remaining_time": "4:34:31"} +{"current_steps": 28, "total_steps": 390, "loss": 0.7118, "lr": 2.871794871794872e-05, "epoch": 0.3584, "percentage": 7.18, "elapsed_time": "0:21:07", "remaining_time": "4:33:10"} +{"current_steps": 29, "total_steps": 390, "loss": 0.7304, "lr": 2.9743589743589747e-05, "epoch": 0.3712, "percentage": 7.44, "elapsed_time": "0:21:52", "remaining_time": "4:32:24"} +{"current_steps": 30, "total_steps": 390, "loss": 0.7412, "lr": 3.0769230769230774e-05, "epoch": 0.384, "percentage": 7.69, "elapsed_time": "0:22:39", "remaining_time": "4:31:53"} +{"current_steps": 31, "total_steps": 390, "loss": 0.7142, "lr": 3.1794871794871795e-05, "epoch": 0.3968, "percentage": 7.95, "elapsed_time": "0:23:22", "remaining_time": "4:30:45"} +{"current_steps": 32, "total_steps": 390, "loss": 0.7098, "lr": 3.282051282051282e-05, "epoch": 0.4096, "percentage": 8.21, "elapsed_time": "0:24:07", "remaining_time": "4:29:57"} +{"current_steps": 33, "total_steps": 390, "loss": 0.718, "lr": 3.384615384615385e-05, "epoch": 0.4224, "percentage": 8.46, "elapsed_time": "0:24:54", "remaining_time": "4:29:25"} +{"current_steps": 34, "total_steps": 390, "loss": 0.7281, "lr": 3.4871794871794875e-05, "epoch": 0.4352, "percentage": 8.72, "elapsed_time": "0:25:37", "remaining_time": "4:28:22"} +{"current_steps": 35, "total_steps": 390, "loss": 0.7008, "lr": 3.58974358974359e-05, "epoch": 0.448, "percentage": 8.97, "elapsed_time": "0:26:22", "remaining_time": "4:27:35"} +{"current_steps": 36, "total_steps": 390, "loss": 0.7226, "lr": 3.692307692307693e-05, "epoch": 0.4608, "percentage": 9.23, "elapsed_time": "0:27:09", "remaining_time": "4:27:03"} +{"current_steps": 37, "total_steps": 390, "loss": 0.7272, "lr": 3.794871794871795e-05, "epoch": 0.4736, "percentage": 9.49, "elapsed_time": "0:27:55", "remaining_time": "4:26:21"} +{"current_steps": 38, "total_steps": 390, "loss": 0.7103, "lr": 3.8974358974358976e-05, "epoch": 0.4864, "percentage": 9.74, "elapsed_time": "0:28:39", "remaining_time": "4:25:27"} +{"current_steps": 39, "total_steps": 390, "loss": 0.7066, "lr": 4e-05, "epoch": 0.4992, "percentage": 10.0, "elapsed_time": "0:29:25", "remaining_time": "4:24:47"} +{"current_steps": 40, "total_steps": 390, "loss": 0.7289, "lr": 3.9999198907597046e-05, "epoch": 0.512, "percentage": 10.26, "elapsed_time": "0:30:09", "remaining_time": "4:23:50"} +{"current_steps": 41, "total_steps": 390, "loss": 0.7014, "lr": 3.9996795694563096e-05, "epoch": 0.5248, "percentage": 10.51, "elapsed_time": "0:30:51", "remaining_time": "4:22:40"} +{"current_steps": 42, "total_steps": 390, "loss": 0.6994, "lr": 3.999279055341771e-05, "epoch": 0.5376, "percentage": 10.77, "elapsed_time": "0:31:36", "remaining_time": "4:21:57"} +{"current_steps": 43, "total_steps": 390, "loss": 0.7036, "lr": 3.998718380500971e-05, "epoch": 0.5504, "percentage": 11.03, "elapsed_time": "0:32:22", "remaining_time": "4:21:18"} +{"current_steps": 44, "total_steps": 390, "loss": 0.7017, "lr": 3.997997589849145e-05, "epoch": 0.5632, "percentage": 11.28, "elapsed_time": "0:33:05", "remaining_time": "4:20:10"} +{"current_steps": 45, "total_steps": 390, "loss": 0.7039, "lr": 3.9971167411282835e-05, "epoch": 0.576, "percentage": 11.54, "elapsed_time": "0:33:49", "remaining_time": "4:19:17"} +{"current_steps": 46, "total_steps": 390, "loss": 0.6928, "lr": 3.99607590490251e-05, "epoch": 0.5888, "percentage": 11.79, "elapsed_time": "0:34:32", "remaining_time": "4:18:20"} +{"current_steps": 47, "total_steps": 390, "loss": 0.6926, "lr": 3.9948751645524235e-05, "epoch": 0.6016, "percentage": 12.05, "elapsed_time": "0:35:15", "remaining_time": "4:17:17"} +{"current_steps": 48, "total_steps": 390, "loss": 0.6994, "lr": 3.9935146162684206e-05, "epoch": 0.6144, "percentage": 12.31, "elapsed_time": "0:35:58", "remaining_time": "4:16:20"} +{"current_steps": 49, "total_steps": 390, "loss": 0.6934, "lr": 3.9919943690429906e-05, "epoch": 0.6272, "percentage": 12.56, "elapsed_time": "0:36:41", "remaining_time": "4:15:23"} +{"current_steps": 50, "total_steps": 390, "loss": 0.7097, "lr": 3.9903145446619837e-05, "epoch": 0.64, "percentage": 12.82, "elapsed_time": "0:37:24", "remaining_time": "4:14:25"} +{"current_steps": 51, "total_steps": 390, "loss": 0.6851, "lr": 3.9884752776948564e-05, "epoch": 0.6528, "percentage": 13.08, "elapsed_time": "0:38:10", "remaining_time": "4:13:46"} +{"current_steps": 52, "total_steps": 390, "loss": 0.6924, "lr": 3.9864767154838864e-05, "epoch": 0.6656, "percentage": 13.33, "elapsed_time": "0:38:53", "remaining_time": "4:12:48"} +{"current_steps": 53, "total_steps": 390, "loss": 0.6795, "lr": 3.9843190181323744e-05, "epoch": 0.6784, "percentage": 13.59, "elapsed_time": "0:39:39", "remaining_time": "4:12:10"} +{"current_steps": 54, "total_steps": 390, "loss": 0.6686, "lr": 3.982002358491817e-05, "epoch": 0.6912, "percentage": 13.85, "elapsed_time": "0:40:25", "remaining_time": "4:11:29"} +{"current_steps": 55, "total_steps": 390, "loss": 0.7001, "lr": 3.979526922148058e-05, "epoch": 0.704, "percentage": 14.1, "elapsed_time": "0:41:08", "remaining_time": "4:10:36"} +{"current_steps": 56, "total_steps": 390, "loss": 0.6792, "lr": 3.9768929074064206e-05, "epoch": 0.7168, "percentage": 14.36, "elapsed_time": "0:41:54", "remaining_time": "4:09:54"} +{"current_steps": 57, "total_steps": 390, "loss": 0.6821, "lr": 3.9741005252758255e-05, "epoch": 0.7296, "percentage": 14.62, "elapsed_time": "0:42:40", "remaining_time": "4:09:18"} +{"current_steps": 58, "total_steps": 390, "loss": 0.6715, "lr": 3.971149999451886e-05, "epoch": 0.7424, "percentage": 14.87, "elapsed_time": "0:43:25", "remaining_time": "4:08:33"} +{"current_steps": 59, "total_steps": 390, "loss": 0.6738, "lr": 3.9680415662989806e-05, "epoch": 0.7552, "percentage": 15.13, "elapsed_time": "0:44:09", "remaining_time": "4:07:42"} +{"current_steps": 60, "total_steps": 390, "loss": 0.6776, "lr": 3.9647754748313294e-05, "epoch": 0.768, "percentage": 15.38, "elapsed_time": "0:44:53", "remaining_time": "4:06:54"} +{"current_steps": 61, "total_steps": 390, "loss": 0.6723, "lr": 3.96135198669304e-05, "epoch": 0.7808, "percentage": 15.64, "elapsed_time": "0:45:37", "remaining_time": "4:06:06"} +{"current_steps": 62, "total_steps": 390, "loss": 0.6804, "lr": 3.957771376137144e-05, "epoch": 0.7936, "percentage": 15.9, "elapsed_time": "0:46:20", "remaining_time": "4:05:08"} +{"current_steps": 63, "total_steps": 390, "loss": 0.6866, "lr": 3.954033930003634e-05, "epoch": 0.8064, "percentage": 16.15, "elapsed_time": "0:47:04", "remaining_time": "4:04:21"} +{"current_steps": 64, "total_steps": 390, "loss": 0.6812, "lr": 3.9501399476964806e-05, "epoch": 0.8192, "percentage": 16.41, "elapsed_time": "0:47:51", "remaining_time": "4:03:45"} +{"current_steps": 65, "total_steps": 390, "loss": 0.6734, "lr": 3.946089741159648e-05, "epoch": 0.832, "percentage": 16.67, "elapsed_time": "0:48:34", "remaining_time": "4:02:52"} +{"current_steps": 66, "total_steps": 390, "loss": 0.6962, "lr": 3.9418836348521045e-05, "epoch": 0.8448, "percentage": 16.92, "elapsed_time": "0:49:19", "remaining_time": "4:02:07"} +{"current_steps": 67, "total_steps": 390, "loss": 0.6791, "lr": 3.937521965721831e-05, "epoch": 0.8576, "percentage": 17.18, "elapsed_time": "0:50:02", "remaining_time": "4:01:16"} +{"current_steps": 68, "total_steps": 390, "loss": 0.672, "lr": 3.933005083178828e-05, "epoch": 0.8704, "percentage": 17.44, "elapsed_time": "0:50:43", "remaining_time": "4:00:13"} +{"current_steps": 69, "total_steps": 390, "loss": 0.69, "lr": 3.928333349067125e-05, "epoch": 0.8832, "percentage": 17.69, "elapsed_time": "0:51:28", "remaining_time": "3:59:30"} +{"current_steps": 70, "total_steps": 390, "loss": 0.6679, "lr": 3.923507137635792e-05, "epoch": 0.896, "percentage": 17.95, "elapsed_time": "0:52:13", "remaining_time": "3:58:44"} +{"current_steps": 71, "total_steps": 390, "loss": 0.6576, "lr": 3.9185268355089606e-05, "epoch": 0.9088, "percentage": 18.21, "elapsed_time": "0:52:59", "remaining_time": "3:58:06"} +{"current_steps": 72, "total_steps": 390, "loss": 0.6752, "lr": 3.913392841654851e-05, "epoch": 0.9216, "percentage": 18.46, "elapsed_time": "0:53:44", "remaining_time": "3:57:23"} +{"current_steps": 73, "total_steps": 390, "loss": 0.666, "lr": 3.9081055673538093e-05, "epoch": 0.9344, "percentage": 18.72, "elapsed_time": "0:54:29", "remaining_time": "3:56:37"} +{"current_steps": 74, "total_steps": 390, "loss": 0.6604, "lr": 3.902665436165364e-05, "epoch": 0.9472, "percentage": 18.97, "elapsed_time": "0:55:14", "remaining_time": "3:55:52"} +{"current_steps": 75, "total_steps": 390, "loss": 0.6701, "lr": 3.897072883894291e-05, "epoch": 0.96, "percentage": 19.23, "elapsed_time": "0:55:59", "remaining_time": "3:55:11"} +{"current_steps": 76, "total_steps": 390, "loss": 0.6736, "lr": 3.8913283585557054e-05, "epoch": 0.9728, "percentage": 19.49, "elapsed_time": "0:56:46", "remaining_time": "3:54:34"} +{"current_steps": 77, "total_steps": 390, "loss": 0.6857, "lr": 3.885432320339167e-05, "epoch": 0.9856, "percentage": 19.74, "elapsed_time": "0:57:30", "remaining_time": "3:53:48"} +{"current_steps": 78, "total_steps": 390, "loss": 0.657, "lr": 3.879385241571817e-05, "epoch": 0.9984, "percentage": 20.0, "elapsed_time": "0:58:14", "remaining_time": "3:52:59"} +{"current_steps": 79, "total_steps": 390, "loss": 0.6226, "lr": 3.873187606680543e-05, "epoch": 1.0112, "percentage": 20.26, "elapsed_time": "1:01:02", "remaining_time": "4:00:17"} +{"current_steps": 80, "total_steps": 390, "loss": 0.5975, "lr": 3.866839912153168e-05, "epoch": 1.024, "percentage": 20.51, "elapsed_time": "1:01:46", "remaining_time": "3:59:24"} +{"current_steps": 81, "total_steps": 390, "loss": 0.6125, "lr": 3.860342666498677e-05, "epoch": 1.0368, "percentage": 20.77, "elapsed_time": "1:02:30", "remaining_time": "3:58:28"} +{"current_steps": 82, "total_steps": 390, "loss": 0.6142, "lr": 3.853696390206484e-05, "epoch": 1.0496, "percentage": 21.03, "elapsed_time": "1:03:15", "remaining_time": "3:57:34"} +{"current_steps": 83, "total_steps": 390, "loss": 0.6042, "lr": 3.846901615704734e-05, "epoch": 1.0624, "percentage": 21.28, "elapsed_time": "1:03:59", "remaining_time": "3:56:42"} +{"current_steps": 84, "total_steps": 390, "loss": 0.6097, "lr": 3.839958887317649e-05, "epoch": 1.0752, "percentage": 21.54, "elapsed_time": "1:04:43", "remaining_time": "3:55:46"} +{"current_steps": 85, "total_steps": 390, "loss": 0.6127, "lr": 3.832868761221926e-05, "epoch": 1.088, "percentage": 21.79, "elapsed_time": "1:05:28", "remaining_time": "3:54:56"} +{"current_steps": 86, "total_steps": 390, "loss": 0.6127, "lr": 3.825631805402182e-05, "epoch": 1.1008, "percentage": 22.05, "elapsed_time": "1:06:13", "remaining_time": "3:54:05"} +{"current_steps": 87, "total_steps": 390, "loss": 0.6091, "lr": 3.818248599605448e-05, "epoch": 1.1136, "percentage": 22.31, "elapsed_time": "1:06:55", "remaining_time": "3:53:04"} +{"current_steps": 88, "total_steps": 390, "loss": 0.6156, "lr": 3.810719735294731e-05, "epoch": 1.1264, "percentage": 22.56, "elapsed_time": "1:07:41", "remaining_time": "3:52:19"} +{"current_steps": 89, "total_steps": 390, "loss": 0.6051, "lr": 3.8030458156016326e-05, "epoch": 1.1392, "percentage": 22.82, "elapsed_time": "1:08:26", "remaining_time": "3:51:26"} +{"current_steps": 90, "total_steps": 390, "loss": 0.6066, "lr": 3.795227455278029e-05, "epoch": 1.152, "percentage": 23.08, "elapsed_time": "1:09:09", "remaining_time": "3:50:32"} +{"current_steps": 91, "total_steps": 390, "loss": 0.5973, "lr": 3.787265280646825e-05, "epoch": 1.1648, "percentage": 23.33, "elapsed_time": "1:09:55", "remaining_time": "3:49:44"} +{"current_steps": 92, "total_steps": 390, "loss": 0.609, "lr": 3.7791599295517825e-05, "epoch": 1.1776, "percentage": 23.59, "elapsed_time": "1:10:41", "remaining_time": "3:48:57"} +{"current_steps": 93, "total_steps": 390, "loss": 0.6243, "lr": 3.7709120513064196e-05, "epoch": 1.1904, "percentage": 23.85, "elapsed_time": "1:11:26", "remaining_time": "3:48:08"} +{"current_steps": 94, "total_steps": 390, "loss": 0.6006, "lr": 3.762522306641998e-05, "epoch": 1.2032, "percentage": 24.1, "elapsed_time": "1:12:12", "remaining_time": "3:47:23"} +{"current_steps": 95, "total_steps": 390, "loss": 0.6086, "lr": 3.7539913676545874e-05, "epoch": 1.216, "percentage": 24.36, "elapsed_time": "1:12:57", "remaining_time": "3:46:34"} +{"current_steps": 96, "total_steps": 390, "loss": 0.611, "lr": 3.745319917751229e-05, "epoch": 1.2288000000000001, "percentage": 24.62, "elapsed_time": "1:13:42", "remaining_time": "3:45:45"} +{"current_steps": 97, "total_steps": 390, "loss": 0.5997, "lr": 3.736508651595188e-05, "epoch": 1.2416, "percentage": 24.87, "elapsed_time": "1:14:29", "remaining_time": "3:45:01"} +{"current_steps": 98, "total_steps": 390, "loss": 0.6057, "lr": 3.727558275050301e-05, "epoch": 1.2544, "percentage": 25.13, "elapsed_time": "1:15:15", "remaining_time": "3:44:15"} +{"current_steps": 99, "total_steps": 390, "loss": 0.5939, "lr": 3.718469505124434e-05, "epoch": 1.2671999999999999, "percentage": 25.38, "elapsed_time": "1:16:00", "remaining_time": "3:43:26"} +{"current_steps": 100, "total_steps": 390, "loss": 0.5954, "lr": 3.709243069912041e-05, "epoch": 1.28, "percentage": 25.64, "elapsed_time": "1:16:44", "remaining_time": "3:42:33"} +{"current_steps": 101, "total_steps": 390, "loss": 0.6158, "lr": 3.699879708535838e-05, "epoch": 1.2928, "percentage": 25.9, "elapsed_time": "1:17:30", "remaining_time": "3:41:47"} +{"current_steps": 102, "total_steps": 390, "loss": 0.5978, "lr": 3.69038017108759e-05, "epoch": 1.3056, "percentage": 26.15, "elapsed_time": "1:18:15", "remaining_time": "3:40:57"} +{"current_steps": 103, "total_steps": 390, "loss": 0.5922, "lr": 3.680745218568026e-05, "epoch": 1.3184, "percentage": 26.41, "elapsed_time": "1:18:57", "remaining_time": "3:40:00"} +{"current_steps": 104, "total_steps": 390, "loss": 0.6167, "lr": 3.6709756228258735e-05, "epoch": 1.3312, "percentage": 26.67, "elapsed_time": "1:19:43", "remaining_time": "3:39:13"} +{"current_steps": 105, "total_steps": 390, "loss": 0.6129, "lr": 3.6610721664960236e-05, "epoch": 1.3439999999999999, "percentage": 26.92, "elapsed_time": "1:20:28", "remaining_time": "3:38:27"} +{"current_steps": 106, "total_steps": 390, "loss": 0.596, "lr": 3.65103564293684e-05, "epoch": 1.3568, "percentage": 27.18, "elapsed_time": "1:21:12", "remaining_time": "3:37:35"} +{"current_steps": 107, "total_steps": 390, "loss": 0.5904, "lr": 3.640866856166601e-05, "epoch": 1.3696, "percentage": 27.44, "elapsed_time": "1:21:57", "remaining_time": "3:36:45"} +{"current_steps": 108, "total_steps": 390, "loss": 0.6202, "lr": 3.6305666207990886e-05, "epoch": 1.3824, "percentage": 27.69, "elapsed_time": "1:22:40", "remaining_time": "3:35:52"} +{"current_steps": 109, "total_steps": 390, "loss": 0.6051, "lr": 3.6201357619783336e-05, "epoch": 1.3952, "percentage": 27.95, "elapsed_time": "1:23:25", "remaining_time": "3:35:03"} +{"current_steps": 110, "total_steps": 390, "loss": 0.6219, "lr": 3.609575115312511e-05, "epoch": 1.408, "percentage": 28.21, "elapsed_time": "1:24:10", "remaining_time": "3:34:14"} +{"current_steps": 111, "total_steps": 390, "loss": 0.6054, "lr": 3.598885526807003e-05, "epoch": 1.4208, "percentage": 28.46, "elapsed_time": "1:24:54", "remaining_time": "3:33:24"} +{"current_steps": 112, "total_steps": 390, "loss": 0.5959, "lr": 3.5880678527966224e-05, "epoch": 1.4336, "percentage": 28.72, "elapsed_time": "1:25:37", "remaining_time": "3:32:32"} +{"current_steps": 113, "total_steps": 390, "loss": 0.599, "lr": 3.577122959877017e-05, "epoch": 1.4464000000000001, "percentage": 28.97, "elapsed_time": "1:26:23", "remaining_time": "3:31:46"} +{"current_steps": 114, "total_steps": 390, "loss": 0.6114, "lr": 3.566051724835245e-05, "epoch": 1.4592, "percentage": 29.23, "elapsed_time": "1:27:07", "remaining_time": "3:30:55"} +{"current_steps": 115, "total_steps": 390, "loss": 0.5967, "lr": 3.554855034579532e-05, "epoch": 1.472, "percentage": 29.49, "elapsed_time": "1:27:52", "remaining_time": "3:30:09"} +{"current_steps": 116, "total_steps": 390, "loss": 0.6005, "lr": 3.5435337860682304e-05, "epoch": 1.4848, "percentage": 29.74, "elapsed_time": "1:28:38", "remaining_time": "3:29:22"} +{"current_steps": 117, "total_steps": 390, "loss": 0.5997, "lr": 3.532088886237956e-05, "epoch": 1.4976, "percentage": 30.0, "elapsed_time": "1:29:23", "remaining_time": "3:28:35"} +{"current_steps": 118, "total_steps": 390, "loss": 0.6009, "lr": 3.520521251930941e-05, "epoch": 1.5104, "percentage": 30.26, "elapsed_time": "1:30:08", "remaining_time": "3:27:47"} +{"current_steps": 119, "total_steps": 390, "loss": 0.5893, "lr": 3.5088318098215805e-05, "epoch": 1.5232, "percentage": 30.51, "elapsed_time": "1:30:51", "remaining_time": "3:26:54"} +{"current_steps": 120, "total_steps": 390, "loss": 0.6036, "lr": 3.497021496342203e-05, "epoch": 1.536, "percentage": 30.77, "elapsed_time": "1:31:37", "remaining_time": "3:26:08"} +{"current_steps": 121, "total_steps": 390, "loss": 0.6002, "lr": 3.485091257608047e-05, "epoch": 1.5488, "percentage": 31.03, "elapsed_time": "1:32:22", "remaining_time": "3:25:22"} +{"current_steps": 122, "total_steps": 390, "loss": 0.6137, "lr": 3.473042049341474e-05, "epoch": 1.5615999999999999, "percentage": 31.28, "elapsed_time": "1:33:06", "remaining_time": "3:24:32"} +{"current_steps": 123, "total_steps": 390, "loss": 0.6043, "lr": 3.4608748367954064e-05, "epoch": 1.5744, "percentage": 31.54, "elapsed_time": "1:33:51", "remaining_time": "3:23:43"} +{"current_steps": 124, "total_steps": 390, "loss": 0.5843, "lr": 3.4485905946759965e-05, "epoch": 1.5872000000000002, "percentage": 31.79, "elapsed_time": "1:34:34", "remaining_time": "3:22:53"} +{"current_steps": 125, "total_steps": 390, "loss": 0.5968, "lr": 3.4361903070645484e-05, "epoch": 1.6, "percentage": 32.05, "elapsed_time": "1:35:20", "remaining_time": "3:22:07"} +{"current_steps": 126, "total_steps": 390, "loss": 0.5962, "lr": 3.423674967338681e-05, "epoch": 1.6128, "percentage": 32.31, "elapsed_time": "1:36:05", "remaining_time": "3:21:20"} +{"current_steps": 127, "total_steps": 390, "loss": 0.5949, "lr": 3.411045578092754e-05, "epoch": 1.6256, "percentage": 32.56, "elapsed_time": "1:36:47", "remaining_time": "3:20:27"} +{"current_steps": 128, "total_steps": 390, "loss": 0.5983, "lr": 3.398303151057543e-05, "epoch": 1.6383999999999999, "percentage": 32.82, "elapsed_time": "1:37:31", "remaining_time": "3:19:36"} +{"current_steps": 129, "total_steps": 390, "loss": 0.5918, "lr": 3.385448707019199e-05, "epoch": 1.6512, "percentage": 33.08, "elapsed_time": "1:38:15", "remaining_time": "3:18:48"} +{"current_steps": 130, "total_steps": 390, "loss": 0.606, "lr": 3.372483275737468e-05, "epoch": 1.6640000000000001, "percentage": 33.33, "elapsed_time": "1:39:00", "remaining_time": "3:18:00"} +{"current_steps": 131, "total_steps": 390, "loss": 0.6001, "lr": 3.359407895863199e-05, "epoch": 1.6768, "percentage": 33.59, "elapsed_time": "1:39:44", "remaining_time": "3:17:11"} +{"current_steps": 132, "total_steps": 390, "loss": 0.6024, "lr": 3.34622361485514e-05, "epoch": 1.6896, "percentage": 33.85, "elapsed_time": "1:40:29", "remaining_time": "3:16:25"} +{"current_steps": 133, "total_steps": 390, "loss": 0.5933, "lr": 3.332931488896029e-05, "epoch": 1.7024, "percentage": 34.1, "elapsed_time": "1:41:16", "remaining_time": "3:15:41"} +{"current_steps": 134, "total_steps": 390, "loss": 0.5932, "lr": 3.319532582807977e-05, "epoch": 1.7151999999999998, "percentage": 34.36, "elapsed_time": "1:42:00", "remaining_time": "3:14:52"} +{"current_steps": 135, "total_steps": 390, "loss": 0.6069, "lr": 3.30602796996717e-05, "epoch": 1.728, "percentage": 34.62, "elapsed_time": "1:42:45", "remaining_time": "3:14:05"} +{"current_steps": 136, "total_steps": 390, "loss": 0.5996, "lr": 3.2924187322178865e-05, "epoch": 1.7408000000000001, "percentage": 34.87, "elapsed_time": "1:43:29", "remaining_time": "3:13:17"} +{"current_steps": 137, "total_steps": 390, "loss": 0.609, "lr": 3.278705959785821e-05, "epoch": 1.7536, "percentage": 35.13, "elapsed_time": "1:44:14", "remaining_time": "3:12:29"} +{"current_steps": 138, "total_steps": 390, "loss": 0.5936, "lr": 3.2648907511907544e-05, "epoch": 1.7664, "percentage": 35.38, "elapsed_time": "1:45:00", "remaining_time": "3:11:45"} +{"current_steps": 139, "total_steps": 390, "loss": 0.5788, "lr": 3.250974213158555e-05, "epoch": 1.7792, "percentage": 35.64, "elapsed_time": "1:45:46", "remaining_time": "3:11:00"} +{"current_steps": 140, "total_steps": 390, "loss": 0.5921, "lr": 3.23695746053251e-05, "epoch": 1.792, "percentage": 35.9, "elapsed_time": "1:46:31", "remaining_time": "3:10:12"} +{"current_steps": 141, "total_steps": 390, "loss": 0.6161, "lr": 3.222841616184025e-05, "epoch": 1.8048, "percentage": 36.15, "elapsed_time": "1:47:16", "remaining_time": "3:09:26"} +{"current_steps": 142, "total_steps": 390, "loss": 0.5922, "lr": 3.208627810922665e-05, "epoch": 1.8176, "percentage": 36.41, "elapsed_time": "1:47:58", "remaining_time": "3:08:35"} +{"current_steps": 143, "total_steps": 390, "loss": 0.5768, "lr": 3.194317183405573e-05, "epoch": 1.8304, "percentage": 36.67, "elapsed_time": "1:48:42", "remaining_time": "3:07:46"} +{"current_steps": 144, "total_steps": 390, "loss": 0.5881, "lr": 3.1799108800462466e-05, "epoch": 1.8432, "percentage": 36.92, "elapsed_time": "1:49:27", "remaining_time": "3:06:59"} +{"current_steps": 145, "total_steps": 390, "loss": 0.5979, "lr": 3.1654100549227024e-05, "epoch": 1.8559999999999999, "percentage": 37.18, "elapsed_time": "1:50:12", "remaining_time": "3:06:12"} +{"current_steps": 146, "total_steps": 390, "loss": 0.592, "lr": 3.1508158696850275e-05, "epoch": 1.8688, "percentage": 37.44, "elapsed_time": "1:50:57", "remaining_time": "3:05:26"} +{"current_steps": 147, "total_steps": 390, "loss": 0.5953, "lr": 3.136129493462312e-05, "epoch": 1.8816000000000002, "percentage": 37.69, "elapsed_time": "1:51:41", "remaining_time": "3:04:37"} +{"current_steps": 148, "total_steps": 390, "loss": 0.5995, "lr": 3.121352102768998e-05, "epoch": 1.8944, "percentage": 37.95, "elapsed_time": "1:52:27", "remaining_time": "3:03:52"} +{"current_steps": 149, "total_steps": 390, "loss": 0.5979, "lr": 3.106484881410628e-05, "epoch": 1.9072, "percentage": 38.21, "elapsed_time": "1:53:10", "remaining_time": "3:03:02"} +{"current_steps": 150, "total_steps": 390, "loss": 0.5863, "lr": 3.091529020389009e-05, "epoch": 1.92, "percentage": 38.46, "elapsed_time": "1:53:54", "remaining_time": "3:02:14"} +{"current_steps": 151, "total_steps": 390, "loss": 0.5816, "lr": 3.076485717806808e-05, "epoch": 1.9327999999999999, "percentage": 38.72, "elapsed_time": "1:54:39", "remaining_time": "3:01:28"} +{"current_steps": 152, "total_steps": 390, "loss": 0.5815, "lr": 3.061356178771564e-05, "epoch": 1.9456, "percentage": 38.97, "elapsed_time": "1:55:23", "remaining_time": "3:00:40"} +{"current_steps": 153, "total_steps": 390, "loss": 0.5893, "lr": 3.0461416152991555e-05, "epoch": 1.9584000000000001, "percentage": 39.23, "elapsed_time": "1:56:08", "remaining_time": "2:59:54"} +{"current_steps": 154, "total_steps": 390, "loss": 0.5974, "lr": 3.0308432462167045e-05, "epoch": 1.9712, "percentage": 39.49, "elapsed_time": "1:56:52", "remaining_time": "2:59:05"} +{"current_steps": 155, "total_steps": 390, "loss": 0.5916, "lr": 3.015462297064936e-05, "epoch": 1.984, "percentage": 39.74, "elapsed_time": "1:57:36", "remaining_time": "2:58:18"} +{"current_steps": 156, "total_steps": 390, "loss": 0.5932, "lr": 3.0000000000000004e-05, "epoch": 1.9968, "percentage": 40.0, "elapsed_time": "1:58:19", "remaining_time": "2:57:28"} +{"current_steps": 157, "total_steps": 390, "loss": 0.5451, "lr": 2.98445759369477e-05, "epoch": 2.0096, "percentage": 40.26, "elapsed_time": "2:01:44", "remaining_time": "3:00:41"} +{"current_steps": 158, "total_steps": 390, "loss": 0.5179, "lr": 2.9688363232396056e-05, "epoch": 2.0224, "percentage": 40.51, "elapsed_time": "2:02:30", "remaining_time": "2:59:52"} +{"current_steps": 159, "total_steps": 390, "loss": 0.5069, "lr": 2.9531374400426158e-05, "epoch": 2.0352, "percentage": 40.77, "elapsed_time": "2:03:15", "remaining_time": "2:59:04"} +{"current_steps": 160, "total_steps": 390, "loss": 0.5256, "lr": 2.9373622017294075e-05, "epoch": 2.048, "percentage": 41.03, "elapsed_time": "2:03:58", "remaining_time": "2:58:13"} +{"current_steps": 161, "total_steps": 390, "loss": 0.5169, "lr": 2.9215118720423375e-05, "epoch": 2.0608, "percentage": 41.28, "elapsed_time": "2:04:42", "remaining_time": "2:57:22"} +{"current_steps": 162, "total_steps": 390, "loss": 0.5095, "lr": 2.9055877207392752e-05, "epoch": 2.0736, "percentage": 41.54, "elapsed_time": "2:05:26", "remaining_time": "2:56:32"} +{"current_steps": 163, "total_steps": 390, "loss": 0.5171, "lr": 2.8895910234918828e-05, "epoch": 2.0864, "percentage": 41.79, "elapsed_time": "2:06:12", "remaining_time": "2:55:45"} +{"current_steps": 164, "total_steps": 390, "loss": 0.506, "lr": 2.873523061783426e-05, "epoch": 2.0992, "percentage": 42.05, "elapsed_time": "2:06:57", "remaining_time": "2:54:56"} +{"current_steps": 165, "total_steps": 390, "loss": 0.511, "lr": 2.8573851228061084e-05, "epoch": 2.112, "percentage": 42.31, "elapsed_time": "2:07:40", "remaining_time": "2:54:06"} +{"current_steps": 166, "total_steps": 390, "loss": 0.5179, "lr": 2.8411784993579633e-05, "epoch": 2.1248, "percentage": 42.56, "elapsed_time": "2:08:26", "remaining_time": "2:53:19"} +{"current_steps": 167, "total_steps": 390, "loss": 0.5217, "lr": 2.8249044897392814e-05, "epoch": 2.1376, "percentage": 42.82, "elapsed_time": "2:09:11", "remaining_time": "2:52:30"} +{"current_steps": 168, "total_steps": 390, "loss": 0.5162, "lr": 2.80856439764861e-05, "epoch": 2.1504, "percentage": 43.08, "elapsed_time": "2:09:55", "remaining_time": "2:51:40"} +{"current_steps": 169, "total_steps": 390, "loss": 0.5089, "lr": 2.792159532078314e-05, "epoch": 2.1632, "percentage": 43.33, "elapsed_time": "2:10:39", "remaining_time": "2:50:51"} +{"current_steps": 170, "total_steps": 390, "loss": 0.517, "lr": 2.77569120720971e-05, "epoch": 2.176, "percentage": 43.59, "elapsed_time": "2:11:23", "remaining_time": "2:50:02"} +{"current_steps": 171, "total_steps": 390, "loss": 0.5204, "lr": 2.7591607423077932e-05, "epoch": 2.1888, "percentage": 43.85, "elapsed_time": "2:12:07", "remaining_time": "2:49:12"} +{"current_steps": 172, "total_steps": 390, "loss": 0.5006, "lr": 2.7425694616155474e-05, "epoch": 2.2016, "percentage": 44.1, "elapsed_time": "2:12:51", "remaining_time": "2:48:23"} +{"current_steps": 173, "total_steps": 390, "loss": 0.5018, "lr": 2.7259186942478656e-05, "epoch": 2.2144, "percentage": 44.36, "elapsed_time": "2:13:37", "remaining_time": "2:47:36"} +{"current_steps": 174, "total_steps": 390, "loss": 0.5061, "lr": 2.7092097740850712e-05, "epoch": 2.2272, "percentage": 44.62, "elapsed_time": "2:14:23", "remaining_time": "2:46:49"} +{"current_steps": 175, "total_steps": 390, "loss": 0.5019, "lr": 2.692444039666066e-05, "epoch": 2.24, "percentage": 44.87, "elapsed_time": "2:15:09", "remaining_time": "2:46:02"} +{"current_steps": 176, "total_steps": 390, "loss": 0.5035, "lr": 2.6756228340810946e-05, "epoch": 2.2528, "percentage": 45.13, "elapsed_time": "2:15:52", "remaining_time": "2:45:12"} +{"current_steps": 177, "total_steps": 390, "loss": 0.5084, "lr": 2.6587475048641596e-05, "epoch": 2.2656, "percentage": 45.38, "elapsed_time": "2:16:37", "remaining_time": "2:44:24"} +{"current_steps": 178, "total_steps": 390, "loss": 0.5094, "lr": 2.6418194038850634e-05, "epoch": 2.2784, "percentage": 45.64, "elapsed_time": "2:17:22", "remaining_time": "2:43:36"} +{"current_steps": 179, "total_steps": 390, "loss": 0.5108, "lr": 2.624839887241115e-05, "epoch": 2.2912, "percentage": 45.9, "elapsed_time": "2:18:06", "remaining_time": "2:42:48"} +{"current_steps": 180, "total_steps": 390, "loss": 0.5041, "lr": 2.607810315148494e-05, "epoch": 2.304, "percentage": 46.15, "elapsed_time": "2:18:52", "remaining_time": "2:42:00"} +{"current_steps": 181, "total_steps": 390, "loss": 0.5099, "lr": 2.5907320518332827e-05, "epoch": 2.3168, "percentage": 46.41, "elapsed_time": "2:19:36", "remaining_time": "2:41:12"} +{"current_steps": 182, "total_steps": 390, "loss": 0.5174, "lr": 2.5736064654221808e-05, "epoch": 2.3296, "percentage": 46.67, "elapsed_time": "2:20:22", "remaining_time": "2:40:25"} +{"current_steps": 183, "total_steps": 390, "loss": 0.5046, "lr": 2.5564349278329056e-05, "epoch": 2.3424, "percentage": 46.92, "elapsed_time": "2:21:08", "remaining_time": "2:39:38"} +{"current_steps": 184, "total_steps": 390, "loss": 0.5145, "lr": 2.539218814664288e-05, "epoch": 2.3552, "percentage": 47.18, "elapsed_time": "2:21:53", "remaining_time": "2:38:51"} +{"current_steps": 185, "total_steps": 390, "loss": 0.4889, "lr": 2.521959505086075e-05, "epoch": 2.368, "percentage": 47.44, "elapsed_time": "2:22:37", "remaining_time": "2:38:02"} +{"current_steps": 186, "total_steps": 390, "loss": 0.5229, "lr": 2.5046583817284437e-05, "epoch": 2.3808, "percentage": 47.69, "elapsed_time": "2:23:23", "remaining_time": "2:37:16"} +{"current_steps": 187, "total_steps": 390, "loss": 0.5045, "lr": 2.487316830571244e-05, "epoch": 2.3936, "percentage": 47.95, "elapsed_time": "2:24:09", "remaining_time": "2:36:29"} +{"current_steps": 188, "total_steps": 390, "loss": 0.5007, "lr": 2.4699362408329646e-05, "epoch": 2.4064, "percentage": 48.21, "elapsed_time": "2:24:53", "remaining_time": "2:35:40"} +{"current_steps": 189, "total_steps": 390, "loss": 0.4951, "lr": 2.4525180048594452e-05, "epoch": 2.4192, "percentage": 48.46, "elapsed_time": "2:25:36", "remaining_time": "2:34:51"} +{"current_steps": 190, "total_steps": 390, "loss": 0.5203, "lr": 2.435063518012335e-05, "epoch": 2.432, "percentage": 48.72, "elapsed_time": "2:26:21", "remaining_time": "2:34:04"} +{"current_steps": 191, "total_steps": 390, "loss": 0.5117, "lr": 2.4175741785573177e-05, "epoch": 2.4448, "percentage": 48.97, "elapsed_time": "2:27:07", "remaining_time": "2:33:17"} +{"current_steps": 192, "total_steps": 390, "loss": 0.503, "lr": 2.4000513875520892e-05, "epoch": 2.4576000000000002, "percentage": 49.23, "elapsed_time": "2:27:53", "remaining_time": "2:32:30"} +{"current_steps": 193, "total_steps": 390, "loss": 0.5193, "lr": 2.3824965487341247e-05, "epoch": 2.4704, "percentage": 49.49, "elapsed_time": "2:28:39", "remaining_time": "2:31:44"} +{"current_steps": 194, "total_steps": 390, "loss": 0.5247, "lr": 2.3649110684082258e-05, "epoch": 2.4832, "percentage": 49.74, "elapsed_time": "2:29:22", "remaining_time": "2:30:54"} +{"current_steps": 195, "total_steps": 390, "loss": 0.5066, "lr": 2.3472963553338614e-05, "epoch": 2.496, "percentage": 50.0, "elapsed_time": "2:30:06", "remaining_time": "2:30:06"} +{"current_steps": 196, "total_steps": 390, "loss": 0.5154, "lr": 2.3296538206123134e-05, "epoch": 2.5088, "percentage": 50.26, "elapsed_time": "2:30:48", "remaining_time": "2:29:16"} +{"current_steps": 197, "total_steps": 390, "loss": 0.5199, "lr": 2.311984877573636e-05, "epoch": 2.5216, "percentage": 50.51, "elapsed_time": "2:31:33", "remaining_time": "2:28:28"} +{"current_steps": 198, "total_steps": 390, "loss": 0.5186, "lr": 2.2942909416634326e-05, "epoch": 2.5343999999999998, "percentage": 50.77, "elapsed_time": "2:32:16", "remaining_time": "2:27:39"} +{"current_steps": 199, "total_steps": 390, "loss": 0.5143, "lr": 2.2765734303294666e-05, "epoch": 2.5472, "percentage": 51.03, "elapsed_time": "2:33:02", "remaining_time": "2:26:53"} +{"current_steps": 200, "total_steps": 390, "loss": 0.5132, "lr": 2.2588337629081107e-05, "epoch": 2.56, "percentage": 51.28, "elapsed_time": "2:33:47", "remaining_time": "2:26:06"} +{"current_steps": 201, "total_steps": 390, "loss": 0.5244, "lr": 2.2410733605106462e-05, "epoch": 2.5728, "percentage": 51.54, "elapsed_time": "2:34:33", "remaining_time": "2:25:20"} +{"current_steps": 202, "total_steps": 390, "loss": 0.4999, "lr": 2.2232936459094158e-05, "epoch": 2.5856, "percentage": 51.79, "elapsed_time": "2:35:17", "remaining_time": "2:24:31"} +{"current_steps": 203, "total_steps": 390, "loss": 0.5164, "lr": 2.205496043423849e-05, "epoch": 2.5984, "percentage": 52.05, "elapsed_time": "2:36:01", "remaining_time": "2:23:43"} +{"current_steps": 204, "total_steps": 390, "loss": 0.5052, "lr": 2.1876819788063586e-05, "epoch": 2.6112, "percentage": 52.31, "elapsed_time": "2:36:46", "remaining_time": "2:22:56"} +{"current_steps": 205, "total_steps": 390, "loss": 0.5121, "lr": 2.16985287912813e-05, "epoch": 2.624, "percentage": 52.56, "elapsed_time": "2:37:30", "remaining_time": "2:22:08"} +{"current_steps": 206, "total_steps": 390, "loss": 0.5216, "lr": 2.1520101726647922e-05, "epoch": 2.6368, "percentage": 52.82, "elapsed_time": "2:38:15", "remaining_time": "2:21:21"} +{"current_steps": 207, "total_steps": 390, "loss": 0.52, "lr": 2.1341552887820048e-05, "epoch": 2.6496, "percentage": 53.08, "elapsed_time": "2:39:00", "remaining_time": "2:20:34"} +{"current_steps": 208, "total_steps": 390, "loss": 0.5233, "lr": 2.1162896578209517e-05, "epoch": 2.6624, "percentage": 53.33, "elapsed_time": "2:39:46", "remaining_time": "2:19:48"} +{"current_steps": 209, "total_steps": 390, "loss": 0.5107, "lr": 2.0984147109837564e-05, "epoch": 2.6752000000000002, "percentage": 53.59, "elapsed_time": "2:40:29", "remaining_time": "2:18:59"} +{"current_steps": 210, "total_steps": 390, "loss": 0.5211, "lr": 2.0805318802188307e-05, "epoch": 2.6879999999999997, "percentage": 53.85, "elapsed_time": "2:41:14", "remaining_time": "2:18:12"} +{"current_steps": 211, "total_steps": 390, "loss": 0.5184, "lr": 2.0626425981061608e-05, "epoch": 2.7008, "percentage": 54.1, "elapsed_time": "2:41:59", "remaining_time": "2:17:25"} +{"current_steps": 212, "total_steps": 390, "loss": 0.4971, "lr": 2.0447482977425465e-05, "epoch": 2.7136, "percentage": 54.36, "elapsed_time": "2:42:44", "remaining_time": "2:16:38"} +{"current_steps": 213, "total_steps": 390, "loss": 0.5034, "lr": 2.0268504126267952e-05, "epoch": 2.7264, "percentage": 54.62, "elapsed_time": "2:43:30", "remaining_time": "2:15:52"} +{"current_steps": 214, "total_steps": 390, "loss": 0.5055, "lr": 2.008950376544887e-05, "epoch": 2.7392, "percentage": 54.87, "elapsed_time": "2:44:14", "remaining_time": "2:15:04"} +{"current_steps": 215, "total_steps": 390, "loss": 0.5129, "lr": 1.9910496234551132e-05, "epoch": 2.752, "percentage": 55.13, "elapsed_time": "2:44:59", "remaining_time": "2:14:17"} +{"current_steps": 216, "total_steps": 390, "loss": 0.5142, "lr": 1.9731495873732055e-05, "epoch": 2.7648, "percentage": 55.38, "elapsed_time": "2:45:40", "remaining_time": "2:13:27"} +{"current_steps": 217, "total_steps": 390, "loss": 0.5199, "lr": 1.9552517022574542e-05, "epoch": 2.7776, "percentage": 55.64, "elapsed_time": "2:46:26", "remaining_time": "2:12:41"} +{"current_steps": 218, "total_steps": 390, "loss": 0.5118, "lr": 1.93735740189384e-05, "epoch": 2.7904, "percentage": 55.9, "elapsed_time": "2:47:11", "remaining_time": "2:11:54"} +{"current_steps": 219, "total_steps": 390, "loss": 0.5202, "lr": 1.9194681197811703e-05, "epoch": 2.8032, "percentage": 56.15, "elapsed_time": "2:47:55", "remaining_time": "2:11:07"} +{"current_steps": 220, "total_steps": 390, "loss": 0.5204, "lr": 1.901585289016244e-05, "epoch": 2.816, "percentage": 56.41, "elapsed_time": "2:48:40", "remaining_time": "2:10:20"} +{"current_steps": 221, "total_steps": 390, "loss": 0.5242, "lr": 1.8837103421790486e-05, "epoch": 2.8288, "percentage": 56.67, "elapsed_time": "2:49:22", "remaining_time": "2:09:30"} +{"current_steps": 222, "total_steps": 390, "loss": 0.5121, "lr": 1.8658447112179952e-05, "epoch": 2.8416, "percentage": 56.92, "elapsed_time": "2:50:06", "remaining_time": "2:08:44"} +{"current_steps": 223, "total_steps": 390, "loss": 0.5097, "lr": 1.8479898273352084e-05, "epoch": 2.8544, "percentage": 57.18, "elapsed_time": "2:50:49", "remaining_time": "2:07:55"} +{"current_steps": 224, "total_steps": 390, "loss": 0.5095, "lr": 1.83014712087187e-05, "epoch": 2.8672, "percentage": 57.44, "elapsed_time": "2:51:33", "remaining_time": "2:07:08"} +{"current_steps": 225, "total_steps": 390, "loss": 0.5051, "lr": 1.8123180211936417e-05, "epoch": 2.88, "percentage": 57.69, "elapsed_time": "2:52:17", "remaining_time": "2:06:21"} +{"current_steps": 226, "total_steps": 390, "loss": 0.5182, "lr": 1.794503956576152e-05, "epoch": 2.8928000000000003, "percentage": 57.95, "elapsed_time": "2:53:02", "remaining_time": "2:05:33"} +{"current_steps": 227, "total_steps": 390, "loss": 0.5086, "lr": 1.776706354090585e-05, "epoch": 2.9055999999999997, "percentage": 58.21, "elapsed_time": "2:53:47", "remaining_time": "2:04:47"} +{"current_steps": 228, "total_steps": 390, "loss": 0.524, "lr": 1.758926639489354e-05, "epoch": 2.9184, "percentage": 58.46, "elapsed_time": "2:54:32", "remaining_time": "2:04:01"} +{"current_steps": 229, "total_steps": 390, "loss": 0.5144, "lr": 1.7411662370918893e-05, "epoch": 2.9312, "percentage": 58.72, "elapsed_time": "2:55:16", "remaining_time": "2:03:13"} +{"current_steps": 230, "total_steps": 390, "loss": 0.5072, "lr": 1.7234265696705344e-05, "epoch": 2.944, "percentage": 58.97, "elapsed_time": "2:56:01", "remaining_time": "2:02:27"} +{"current_steps": 231, "total_steps": 390, "loss": 0.5145, "lr": 1.7057090583365678e-05, "epoch": 2.9568, "percentage": 59.23, "elapsed_time": "2:56:47", "remaining_time": "2:01:41"} +{"current_steps": 232, "total_steps": 390, "loss": 0.5029, "lr": 1.6880151224263646e-05, "epoch": 2.9696, "percentage": 59.49, "elapsed_time": "2:57:32", "remaining_time": "2:00:54"} +{"current_steps": 233, "total_steps": 390, "loss": 0.5167, "lr": 1.6703461793876876e-05, "epoch": 2.9824, "percentage": 59.74, "elapsed_time": "2:58:13", "remaining_time": "2:00:05"} +{"current_steps": 234, "total_steps": 390, "loss": 0.5004, "lr": 1.6527036446661396e-05, "epoch": 2.9952, "percentage": 60.0, "elapsed_time": "2:58:57", "remaining_time": "1:59:18"} +{"current_steps": 235, "total_steps": 390, "loss": 0.463, "lr": 1.635088931591775e-05, "epoch": 3.008, "percentage": 60.26, "elapsed_time": "3:02:17", "remaining_time": "2:00:14"} +{"current_steps": 236, "total_steps": 390, "loss": 0.4378, "lr": 1.6175034512658753e-05, "epoch": 3.0208, "percentage": 60.51, "elapsed_time": "3:03:01", "remaining_time": "1:59:26"} +{"current_steps": 237, "total_steps": 390, "loss": 0.4371, "lr": 1.5999486124479115e-05, "epoch": 3.0336, "percentage": 60.77, "elapsed_time": "3:03:45", "remaining_time": "1:58:37"} +{"current_steps": 238, "total_steps": 390, "loss": 0.4426, "lr": 1.5824258214426833e-05, "epoch": 3.0464, "percentage": 61.03, "elapsed_time": "3:04:32", "remaining_time": "1:57:51"} +{"current_steps": 239, "total_steps": 390, "loss": 0.4351, "lr": 1.5649364819876655e-05, "epoch": 3.0592, "percentage": 61.28, "elapsed_time": "3:05:16", "remaining_time": "1:57:03"} +{"current_steps": 240, "total_steps": 390, "loss": 0.4314, "lr": 1.547481995140556e-05, "epoch": 3.072, "percentage": 61.54, "elapsed_time": "3:06:01", "remaining_time": "1:56:15"} +{"current_steps": 241, "total_steps": 390, "loss": 0.4401, "lr": 1.5300637591670357e-05, "epoch": 3.0848, "percentage": 61.79, "elapsed_time": "3:06:44", "remaining_time": "1:55:27"} +{"current_steps": 242, "total_steps": 390, "loss": 0.4337, "lr": 1.5126831694287564e-05, "epoch": 3.0976, "percentage": 62.05, "elapsed_time": "3:07:29", "remaining_time": "1:54:40"} +{"current_steps": 243, "total_steps": 390, "loss": 0.4282, "lr": 1.4953416182715566e-05, "epoch": 3.1104, "percentage": 62.31, "elapsed_time": "3:08:14", "remaining_time": "1:53:52"} +{"current_steps": 244, "total_steps": 390, "loss": 0.4537, "lr": 1.478040494913926e-05, "epoch": 3.1232, "percentage": 62.56, "elapsed_time": "3:08:58", "remaining_time": "1:53:04"} +{"current_steps": 245, "total_steps": 390, "loss": 0.4488, "lr": 1.460781185335713e-05, "epoch": 3.136, "percentage": 62.82, "elapsed_time": "3:09:43", "remaining_time": "1:52:16"} +{"current_steps": 246, "total_steps": 390, "loss": 0.4334, "lr": 1.443565072167095e-05, "epoch": 3.1488, "percentage": 63.08, "elapsed_time": "3:10:27", "remaining_time": "1:51:29"} +{"current_steps": 247, "total_steps": 390, "loss": 0.4443, "lr": 1.4263935345778202e-05, "epoch": 3.1616, "percentage": 63.33, "elapsed_time": "3:11:12", "remaining_time": "1:50:41"} +{"current_steps": 248, "total_steps": 390, "loss": 0.4315, "lr": 1.409267948166718e-05, "epoch": 3.1744, "percentage": 63.59, "elapsed_time": "3:11:56", "remaining_time": "1:49:54"} +{"current_steps": 249, "total_steps": 390, "loss": 0.4449, "lr": 1.3921896848515064e-05, "epoch": 3.1872, "percentage": 63.85, "elapsed_time": "3:12:42", "remaining_time": "1:49:07"} +{"current_steps": 250, "total_steps": 390, "loss": 0.441, "lr": 1.3751601127588849e-05, "epoch": 3.2, "percentage": 64.1, "elapsed_time": "3:13:28", "remaining_time": "1:48:21"} +{"current_steps": 251, "total_steps": 390, "loss": 0.436, "lr": 1.3581805961149371e-05, "epoch": 3.2128, "percentage": 64.36, "elapsed_time": "3:14:13", "remaining_time": "1:47:33"} +{"current_steps": 252, "total_steps": 390, "loss": 0.4353, "lr": 1.341252495135841e-05, "epoch": 3.2256, "percentage": 64.62, "elapsed_time": "3:14:58", "remaining_time": "1:46:46"} +{"current_steps": 253, "total_steps": 390, "loss": 0.4343, "lr": 1.324377165918906e-05, "epoch": 3.2384, "percentage": 64.87, "elapsed_time": "3:15:42", "remaining_time": "1:45:58"} +{"current_steps": 254, "total_steps": 390, "loss": 0.4315, "lr": 1.3075559603339354e-05, "epoch": 3.2512, "percentage": 65.13, "elapsed_time": "3:16:29", "remaining_time": "1:45:12"} +{"current_steps": 255, "total_steps": 390, "loss": 0.4356, "lr": 1.2907902259149287e-05, "epoch": 3.2640000000000002, "percentage": 65.38, "elapsed_time": "3:17:13", "remaining_time": "1:44:25"} +{"current_steps": 256, "total_steps": 390, "loss": 0.4401, "lr": 1.274081305752135e-05, "epoch": 3.2768, "percentage": 65.64, "elapsed_time": "3:17:58", "remaining_time": "1:43:37"} +{"current_steps": 257, "total_steps": 390, "loss": 0.4434, "lr": 1.2574305383844528e-05, "epoch": 3.2896, "percentage": 65.9, "elapsed_time": "3:18:43", "remaining_time": "1:42:50"} +{"current_steps": 258, "total_steps": 390, "loss": 0.4414, "lr": 1.2408392576922075e-05, "epoch": 3.3024, "percentage": 66.15, "elapsed_time": "3:19:29", "remaining_time": "1:42:03"} +{"current_steps": 259, "total_steps": 390, "loss": 0.4475, "lr": 1.2243087927902905e-05, "epoch": 3.3152, "percentage": 66.41, "elapsed_time": "3:20:14", "remaining_time": "1:41:16"} +{"current_steps": 260, "total_steps": 390, "loss": 0.4246, "lr": 1.2078404679216864e-05, "epoch": 3.328, "percentage": 66.67, "elapsed_time": "3:21:00", "remaining_time": "1:40:30"} +{"current_steps": 261, "total_steps": 390, "loss": 0.4303, "lr": 1.1914356023513904e-05, "epoch": 3.3407999999999998, "percentage": 66.92, "elapsed_time": "3:21:44", "remaining_time": "1:39:42"} +{"current_steps": 262, "total_steps": 390, "loss": 0.4366, "lr": 1.1750955102607193e-05, "epoch": 3.3536, "percentage": 67.18, "elapsed_time": "3:22:29", "remaining_time": "1:38:55"} +{"current_steps": 263, "total_steps": 390, "loss": 0.4396, "lr": 1.1588215006420374e-05, "epoch": 3.3664, "percentage": 67.44, "elapsed_time": "3:23:11", "remaining_time": "1:38:07"} +{"current_steps": 264, "total_steps": 390, "loss": 0.4372, "lr": 1.1426148771938915e-05, "epoch": 3.3792, "percentage": 67.69, "elapsed_time": "3:23:56", "remaining_time": "1:37:20"} +{"current_steps": 265, "total_steps": 390, "loss": 0.4366, "lr": 1.1264769382165748e-05, "epoch": 3.392, "percentage": 67.95, "elapsed_time": "3:24:38", "remaining_time": "1:36:31"} +{"current_steps": 266, "total_steps": 390, "loss": 0.4325, "lr": 1.110408976508118e-05, "epoch": 3.4048, "percentage": 68.21, "elapsed_time": "3:25:21", "remaining_time": "1:35:43"} +{"current_steps": 267, "total_steps": 390, "loss": 0.4227, "lr": 1.094412279260726e-05, "epoch": 3.4176, "percentage": 68.46, "elapsed_time": "3:26:07", "remaining_time": "1:34:57"} +{"current_steps": 268, "total_steps": 390, "loss": 0.4413, "lr": 1.0784881279576635e-05, "epoch": 3.4304, "percentage": 68.72, "elapsed_time": "3:26:53", "remaining_time": "1:34:10"} +{"current_steps": 269, "total_steps": 390, "loss": 0.4485, "lr": 1.0626377982705929e-05, "epoch": 3.4432, "percentage": 68.97, "elapsed_time": "3:27:40", "remaining_time": "1:33:24"} +{"current_steps": 270, "total_steps": 390, "loss": 0.4223, "lr": 1.0468625599573842e-05, "epoch": 3.456, "percentage": 69.23, "elapsed_time": "3:28:23", "remaining_time": "1:32:37"} +{"current_steps": 271, "total_steps": 390, "loss": 0.441, "lr": 1.0311636767603952e-05, "epoch": 3.4688, "percentage": 69.49, "elapsed_time": "3:29:06", "remaining_time": "1:31:49"} +{"current_steps": 272, "total_steps": 390, "loss": 0.4388, "lr": 1.0155424063052306e-05, "epoch": 3.4816, "percentage": 69.74, "elapsed_time": "3:29:51", "remaining_time": "1:31:02"} +{"current_steps": 273, "total_steps": 390, "loss": 0.4383, "lr": 1.0000000000000006e-05, "epoch": 3.4944, "percentage": 70.0, "elapsed_time": "3:30:35", "remaining_time": "1:30:15"} +{"current_steps": 274, "total_steps": 390, "loss": 0.4391, "lr": 9.84537702935065e-06, "epoch": 3.5072, "percentage": 70.26, "elapsed_time": "3:31:20", "remaining_time": "1:29:28"} +{"current_steps": 275, "total_steps": 390, "loss": 0.4489, "lr": 9.691567537832964e-06, "epoch": 3.52, "percentage": 70.51, "elapsed_time": "3:32:05", "remaining_time": "1:28:41"} +{"current_steps": 276, "total_steps": 390, "loss": 0.4188, "lr": 9.538583847008452e-06, "epoch": 3.5328, "percentage": 70.77, "elapsed_time": "3:32:50", "remaining_time": "1:27:54"} +{"current_steps": 277, "total_steps": 390, "loss": 0.423, "lr": 9.386438212284372e-06, "epoch": 3.5456, "percentage": 71.03, "elapsed_time": "3:33:36", "remaining_time": "1:27:08"} +{"current_steps": 278, "total_steps": 390, "loss": 0.431, "lr": 9.235142821931928e-06, "epoch": 3.5584, "percentage": 71.28, "elapsed_time": "3:34:20", "remaining_time": "1:26:21"} +{"current_steps": 279, "total_steps": 390, "loss": 0.4381, "lr": 9.084709796109907e-06, "epoch": 3.5712, "percentage": 71.54, "elapsed_time": "3:35:04", "remaining_time": "1:25:34"} +{"current_steps": 280, "total_steps": 390, "loss": 0.4425, "lr": 8.93515118589373e-06, "epoch": 3.584, "percentage": 71.79, "elapsed_time": "3:35:47", "remaining_time": "1:24:46"} +{"current_steps": 281, "total_steps": 390, "loss": 0.4301, "lr": 8.786478972310023e-06, "epoch": 3.5968, "percentage": 72.05, "elapsed_time": "3:36:32", "remaining_time": "1:23:59"} +{"current_steps": 282, "total_steps": 390, "loss": 0.4289, "lr": 8.638705065376887e-06, "epoch": 3.6096, "percentage": 72.31, "elapsed_time": "3:37:15", "remaining_time": "1:23:12"} +{"current_steps": 283, "total_steps": 390, "loss": 0.4311, "lr": 8.491841303149728e-06, "epoch": 3.6224, "percentage": 72.56, "elapsed_time": "3:38:01", "remaining_time": "1:22:26"} +{"current_steps": 284, "total_steps": 390, "loss": 0.4371, "lr": 8.345899450772975e-06, "epoch": 3.6352, "percentage": 72.82, "elapsed_time": "3:38:45", "remaining_time": "1:21:38"} +{"current_steps": 285, "total_steps": 390, "loss": 0.4434, "lr": 8.200891199537549e-06, "epoch": 3.648, "percentage": 73.08, "elapsed_time": "3:39:31", "remaining_time": "1:20:52"} +{"current_steps": 286, "total_steps": 390, "loss": 0.4314, "lr": 8.056828165944282e-06, "epoch": 3.6608, "percentage": 73.33, "elapsed_time": "3:40:16", "remaining_time": "1:20:06"} +{"current_steps": 287, "total_steps": 390, "loss": 0.4393, "lr": 7.913721890773354e-06, "epoch": 3.6736, "percentage": 73.59, "elapsed_time": "3:41:03", "remaining_time": "1:19:20"} +{"current_steps": 288, "total_steps": 390, "loss": 0.4335, "lr": 7.771583838159756e-06, "epoch": 3.6864, "percentage": 73.85, "elapsed_time": "3:41:47", "remaining_time": "1:18:32"} +{"current_steps": 289, "total_steps": 390, "loss": 0.4346, "lr": 7.630425394674903e-06, "epoch": 3.6992000000000003, "percentage": 74.1, "elapsed_time": "3:42:32", "remaining_time": "1:17:46"} +{"current_steps": 290, "total_steps": 390, "loss": 0.4203, "lr": 7.49025786841445e-06, "epoch": 3.7119999999999997, "percentage": 74.36, "elapsed_time": "3:43:18", "remaining_time": "1:17:00"} +{"current_steps": 291, "total_steps": 390, "loss": 0.4461, "lr": 7.3510924880924575e-06, "epoch": 3.7248, "percentage": 74.62, "elapsed_time": "3:44:02", "remaining_time": "1:16:13"} +{"current_steps": 292, "total_steps": 390, "loss": 0.4325, "lr": 7.212940402141808e-06, "epoch": 3.7376, "percentage": 74.87, "elapsed_time": "3:44:47", "remaining_time": "1:15:26"} +{"current_steps": 293, "total_steps": 390, "loss": 0.4248, "lr": 7.075812677821145e-06, "epoch": 3.7504, "percentage": 75.13, "elapsed_time": "3:45:33", "remaining_time": "1:14:40"} +{"current_steps": 294, "total_steps": 390, "loss": 0.455, "lr": 6.939720300328303e-06, "epoch": 3.7632, "percentage": 75.38, "elapsed_time": "3:46:19", "remaining_time": "1:13:53"} +{"current_steps": 295, "total_steps": 390, "loss": 0.4485, "lr": 6.8046741719202385e-06, "epoch": 3.776, "percentage": 75.64, "elapsed_time": "3:47:04", "remaining_time": "1:13:07"} +{"current_steps": 296, "total_steps": 390, "loss": 0.4514, "lr": 6.67068511103971e-06, "epoch": 3.7888, "percentage": 75.9, "elapsed_time": "3:47:50", "remaining_time": "1:12:21"} +{"current_steps": 297, "total_steps": 390, "loss": 0.4467, "lr": 6.537763851448593e-06, "epoch": 3.8016, "percentage": 76.15, "elapsed_time": "3:48:36", "remaining_time": "1:11:35"} +{"current_steps": 298, "total_steps": 390, "loss": 0.4343, "lr": 6.4059210413680175e-06, "epoch": 3.8144, "percentage": 76.41, "elapsed_time": "3:49:20", "remaining_time": "1:10:48"} +{"current_steps": 299, "total_steps": 390, "loss": 0.4271, "lr": 6.275167242625331e-06, "epoch": 3.8272, "percentage": 76.67, "elapsed_time": "3:50:04", "remaining_time": "1:10:01"} +{"current_steps": 300, "total_steps": 390, "loss": 0.4414, "lr": 6.145512929808013e-06, "epoch": 3.84, "percentage": 76.92, "elapsed_time": "3:50:49", "remaining_time": "1:09:14"} +{"current_steps": 301, "total_steps": 390, "loss": 0.4283, "lr": 6.016968489424572e-06, "epoch": 3.8528000000000002, "percentage": 77.18, "elapsed_time": "3:51:33", "remaining_time": "1:08:28"} +{"current_steps": 302, "total_steps": 390, "loss": 0.4249, "lr": 5.889544219072465e-06, "epoch": 3.8656, "percentage": 77.44, "elapsed_time": "3:52:18", "remaining_time": "1:07:41"} +{"current_steps": 303, "total_steps": 390, "loss": 0.4274, "lr": 5.7632503266131925e-06, "epoch": 3.8784, "percentage": 77.69, "elapsed_time": "3:53:01", "remaining_time": "1:06:54"} +{"current_steps": 304, "total_steps": 390, "loss": 0.4265, "lr": 5.638096929354522e-06, "epoch": 3.8912, "percentage": 77.95, "elapsed_time": "3:53:44", "remaining_time": "1:06:07"} +{"current_steps": 305, "total_steps": 390, "loss": 0.4274, "lr": 5.514094053240035e-06, "epoch": 3.904, "percentage": 78.21, "elapsed_time": "3:54:30", "remaining_time": "1:05:21"} +{"current_steps": 306, "total_steps": 390, "loss": 0.4366, "lr": 5.39125163204594e-06, "epoch": 3.9168, "percentage": 78.46, "elapsed_time": "3:55:13", "remaining_time": "1:04:34"} +{"current_steps": 307, "total_steps": 390, "loss": 0.4445, "lr": 5.269579506585259e-06, "epoch": 3.9295999999999998, "percentage": 78.72, "elapsed_time": "3:55:58", "remaining_time": "1:03:47"} +{"current_steps": 308, "total_steps": 390, "loss": 0.4383, "lr": 5.149087423919541e-06, "epoch": 3.9424, "percentage": 78.97, "elapsed_time": "3:56:44", "remaining_time": "1:03:01"} +{"current_steps": 309, "total_steps": 390, "loss": 0.4311, "lr": 5.029785036577976e-06, "epoch": 3.9552, "percentage": 79.23, "elapsed_time": "3:57:29", "remaining_time": "1:02:15"} +{"current_steps": 310, "total_steps": 390, "loss": 0.4481, "lr": 4.911681901784198e-06, "epoch": 3.968, "percentage": 79.49, "elapsed_time": "3:58:14", "remaining_time": "1:01:28"} +{"current_steps": 311, "total_steps": 390, "loss": 0.4391, "lr": 4.794787480690597e-06, "epoch": 3.9808, "percentage": 79.74, "elapsed_time": "3:59:00", "remaining_time": "1:00:42"} +{"current_steps": 312, "total_steps": 390, "loss": 0.4449, "lr": 4.679111137620442e-06, "epoch": 3.9936, "percentage": 80.0, "elapsed_time": "3:59:45", "remaining_time": "0:59:56"} +{"current_steps": 313, "total_steps": 390, "loss": 0.4054, "lr": 4.5646621393177e-06, "epoch": 4.0064, "percentage": 80.26, "elapsed_time": "4:03:05", "remaining_time": "0:59:48"} +{"current_steps": 314, "total_steps": 390, "loss": 0.409, "lr": 4.451449654204685e-06, "epoch": 4.0192, "percentage": 80.51, "elapsed_time": "4:03:49", "remaining_time": "0:59:00"} +{"current_steps": 315, "total_steps": 390, "loss": 0.378, "lr": 4.339482751647557e-06, "epoch": 4.032, "percentage": 80.77, "elapsed_time": "4:04:32", "remaining_time": "0:58:13"} +{"current_steps": 316, "total_steps": 390, "loss": 0.4044, "lr": 4.228770401229824e-06, "epoch": 4.0448, "percentage": 81.03, "elapsed_time": "4:05:17", "remaining_time": "0:57:26"} +{"current_steps": 317, "total_steps": 390, "loss": 0.3691, "lr": 4.119321472033779e-06, "epoch": 4.0576, "percentage": 81.28, "elapsed_time": "4:06:03", "remaining_time": "0:56:39"} +{"current_steps": 318, "total_steps": 390, "loss": 0.3743, "lr": 4.011144731929981e-06, "epoch": 4.0704, "percentage": 81.54, "elapsed_time": "4:06:47", "remaining_time": "0:55:52"} +{"current_steps": 319, "total_steps": 390, "loss": 0.4057, "lr": 3.904248846874894e-06, "epoch": 4.0832, "percentage": 81.79, "elapsed_time": "4:07:32", "remaining_time": "0:55:05"} +{"current_steps": 320, "total_steps": 390, "loss": 0.3841, "lr": 3.7986423802166705e-06, "epoch": 4.096, "percentage": 82.05, "elapsed_time": "4:08:16", "remaining_time": "0:54:18"} +{"current_steps": 321, "total_steps": 390, "loss": 0.391, "lr": 3.694333792009115e-06, "epoch": 4.1088, "percentage": 82.31, "elapsed_time": "4:09:01", "remaining_time": "0:53:31"} +{"current_steps": 322, "total_steps": 390, "loss": 0.391, "lr": 3.5913314383339937e-06, "epoch": 4.1216, "percentage": 82.56, "elapsed_time": "4:09:45", "remaining_time": "0:52:44"} +{"current_steps": 323, "total_steps": 390, "loss": 0.3888, "lr": 3.4896435706316e-06, "epoch": 4.1344, "percentage": 82.82, "elapsed_time": "4:10:31", "remaining_time": "0:51:58"} +{"current_steps": 324, "total_steps": 390, "loss": 0.3897, "lr": 3.3892783350397675e-06, "epoch": 4.1472, "percentage": 83.08, "elapsed_time": "4:11:16", "remaining_time": "0:51:11"} +{"current_steps": 325, "total_steps": 390, "loss": 0.3844, "lr": 3.290243771741275e-06, "epoch": 4.16, "percentage": 83.33, "elapsed_time": "4:12:01", "remaining_time": "0:50:24"} +{"current_steps": 326, "total_steps": 390, "loss": 0.38, "lr": 3.1925478143197418e-06, "epoch": 4.1728, "percentage": 83.59, "elapsed_time": "4:12:46", "remaining_time": "0:49:37"} +{"current_steps": 327, "total_steps": 390, "loss": 0.3915, "lr": 3.0961982891241083e-06, "epoch": 4.1856, "percentage": 83.85, "elapsed_time": "4:13:29", "remaining_time": "0:48:50"} +{"current_steps": 328, "total_steps": 390, "loss": 0.3827, "lr": 3.001202914641628e-06, "epoch": 4.1984, "percentage": 84.1, "elapsed_time": "4:14:14", "remaining_time": "0:48:03"} +{"current_steps": 329, "total_steps": 390, "loss": 0.3802, "lr": 2.907569300879596e-06, "epoch": 4.2112, "percentage": 84.36, "elapsed_time": "4:14:59", "remaining_time": "0:47:16"} +{"current_steps": 330, "total_steps": 390, "loss": 0.3811, "lr": 2.815304948755664e-06, "epoch": 4.224, "percentage": 84.62, "elapsed_time": "4:15:41", "remaining_time": "0:46:29"} +{"current_steps": 331, "total_steps": 390, "loss": 0.3941, "lr": 2.7244172494969978e-06, "epoch": 4.2368, "percentage": 84.87, "elapsed_time": "4:16:25", "remaining_time": "0:45:42"} +{"current_steps": 332, "total_steps": 390, "loss": 0.3836, "lr": 2.6349134840481294e-06, "epoch": 4.2496, "percentage": 85.13, "elapsed_time": "4:17:11", "remaining_time": "0:44:55"} +{"current_steps": 333, "total_steps": 390, "loss": 0.3956, "lr": 2.546800822487714e-06, "epoch": 4.2624, "percentage": 85.38, "elapsed_time": "4:17:56", "remaining_time": "0:44:09"} +{"current_steps": 334, "total_steps": 390, "loss": 0.3984, "lr": 2.4600863234541338e-06, "epoch": 4.2752, "percentage": 85.64, "elapsed_time": "4:18:42", "remaining_time": "0:43:22"} +{"current_steps": 335, "total_steps": 390, "loss": 0.3945, "lr": 2.374776933580025e-06, "epoch": 4.288, "percentage": 85.9, "elapsed_time": "4:19:28", "remaining_time": "0:42:35"} +{"current_steps": 336, "total_steps": 390, "loss": 0.3804, "lr": 2.2908794869358044e-06, "epoch": 4.3008, "percentage": 86.15, "elapsed_time": "4:20:13", "remaining_time": "0:41:49"} +{"current_steps": 337, "total_steps": 390, "loss": 0.3918, "lr": 2.2084007044821764e-06, "epoch": 4.3136, "percentage": 86.41, "elapsed_time": "4:20:56", "remaining_time": "0:41:02"} +{"current_steps": 338, "total_steps": 390, "loss": 0.3926, "lr": 2.127347193531757e-06, "epoch": 4.3264, "percentage": 86.67, "elapsed_time": "4:21:40", "remaining_time": "0:40:15"} +{"current_steps": 339, "total_steps": 390, "loss": 0.3864, "lr": 2.0477254472197237e-06, "epoch": 4.3392, "percentage": 86.92, "elapsed_time": "4:22:26", "remaining_time": "0:39:28"} +{"current_steps": 340, "total_steps": 390, "loss": 0.3812, "lr": 1.96954184398368e-06, "epoch": 4.352, "percentage": 87.18, "elapsed_time": "4:23:10", "remaining_time": "0:38:42"} +{"current_steps": 341, "total_steps": 390, "loss": 0.3853, "lr": 1.8928026470526917e-06, "epoch": 4.3648, "percentage": 87.44, "elapsed_time": "4:23:55", "remaining_time": "0:37:55"} +{"current_steps": 342, "total_steps": 390, "loss": 0.3892, "lr": 1.817514003945524e-06, "epoch": 4.3776, "percentage": 87.69, "elapsed_time": "4:24:39", "remaining_time": "0:37:08"} +{"current_steps": 343, "total_steps": 390, "loss": 0.3746, "lr": 1.743681945978184e-06, "epoch": 4.3904, "percentage": 87.95, "elapsed_time": "4:25:24", "remaining_time": "0:36:22"} +{"current_steps": 344, "total_steps": 390, "loss": 0.3752, "lr": 1.6713123877807413e-06, "epoch": 4.4032, "percentage": 88.21, "elapsed_time": "4:26:07", "remaining_time": "0:35:35"} +{"current_steps": 345, "total_steps": 390, "loss": 0.3938, "lr": 1.6004111268235156e-06, "epoch": 4.416, "percentage": 88.46, "elapsed_time": "4:26:52", "remaining_time": "0:34:48"} +{"current_steps": 346, "total_steps": 390, "loss": 0.381, "lr": 1.5309838429526714e-06, "epoch": 4.4288, "percentage": 88.72, "elapsed_time": "4:27:34", "remaining_time": "0:34:01"} +{"current_steps": 347, "total_steps": 390, "loss": 0.3767, "lr": 1.4630360979351644e-06, "epoch": 4.4416, "percentage": 88.97, "elapsed_time": "4:28:16", "remaining_time": "0:33:14"} +{"current_steps": 348, "total_steps": 390, "loss": 0.3949, "lr": 1.396573335013236e-06, "epoch": 4.4544, "percentage": 89.23, "elapsed_time": "4:29:03", "remaining_time": "0:32:28"} +{"current_steps": 349, "total_steps": 390, "loss": 0.3952, "lr": 1.3316008784683265e-06, "epoch": 4.4672, "percentage": 89.49, "elapsed_time": "4:29:49", "remaining_time": "0:31:41"} +{"current_steps": 350, "total_steps": 390, "loss": 0.4051, "lr": 1.2681239331945695e-06, "epoch": 4.48, "percentage": 89.74, "elapsed_time": "4:30:34", "remaining_time": "0:30:55"} +{"current_steps": 351, "total_steps": 390, "loss": 0.3991, "lr": 1.2061475842818337e-06, "epoch": 4.4928, "percentage": 90.0, "elapsed_time": "4:31:18", "remaining_time": "0:30:08"} +{"current_steps": 352, "total_steps": 390, "loss": 0.3793, "lr": 1.1456767966083393e-06, "epoch": 4.5056, "percentage": 90.26, "elapsed_time": "4:32:04", "remaining_time": "0:29:22"} +{"current_steps": 353, "total_steps": 390, "loss": 0.3937, "lr": 1.086716414442952e-06, "epoch": 4.5184, "percentage": 90.51, "elapsed_time": "4:32:50", "remaining_time": "0:28:35"} +{"current_steps": 354, "total_steps": 390, "loss": 0.391, "lr": 1.0292711610570904e-06, "epoch": 4.5312, "percentage": 90.77, "elapsed_time": "4:33:33", "remaining_time": "0:27:49"} +{"current_steps": 355, "total_steps": 390, "loss": 0.3967, "lr": 9.733456383463658e-07, "epoch": 4.5440000000000005, "percentage": 91.03, "elapsed_time": "4:34:18", "remaining_time": "0:27:02"} +{"current_steps": 356, "total_steps": 390, "loss": 0.4129, "lr": 9.189443264619102e-07, "epoch": 4.5568, "percentage": 91.28, "elapsed_time": "4:35:04", "remaining_time": "0:26:16"} +{"current_steps": 357, "total_steps": 390, "loss": 0.3916, "lr": 8.660715834514977e-07, "epoch": 4.5696, "percentage": 91.54, "elapsed_time": "4:35:50", "remaining_time": "0:25:29"} +{"current_steps": 358, "total_steps": 390, "loss": 0.4004, "lr": 8.147316449103959e-07, "epoch": 4.5824, "percentage": 91.79, "elapsed_time": "4:36:34", "remaining_time": "0:24:43"} +{"current_steps": 359, "total_steps": 390, "loss": 0.4153, "lr": 7.649286236420806e-07, "epoch": 4.5952, "percentage": 92.05, "elapsed_time": "4:37:19", "remaining_time": "0:23:56"} +{"current_steps": 360, "total_steps": 390, "loss": 0.4006, "lr": 7.166665093287539e-07, "epoch": 4.608, "percentage": 92.31, "elapsed_time": "4:38:03", "remaining_time": "0:23:10"} +{"current_steps": 361, "total_steps": 390, "loss": 0.3749, "lr": 6.69949168211721e-07, "epoch": 4.6208, "percentage": 92.56, "elapsed_time": "4:38:45", "remaining_time": "0:22:23"} +{"current_steps": 362, "total_steps": 390, "loss": 0.4018, "lr": 6.247803427816945e-07, "epoch": 4.6336, "percentage": 92.82, "elapsed_time": "4:39:30", "remaining_time": "0:21:37"} +{"current_steps": 363, "total_steps": 390, "loss": 0.3799, "lr": 5.811636514789598e-07, "epoch": 4.6464, "percentage": 93.08, "elapsed_time": "4:40:14", "remaining_time": "0:20:50"} +{"current_steps": 364, "total_steps": 390, "loss": 0.3806, "lr": 5.391025884035239e-07, "epoch": 4.6592, "percentage": 93.33, "elapsed_time": "4:40:58", "remaining_time": "0:20:04"} +{"current_steps": 365, "total_steps": 390, "loss": 0.3777, "lr": 4.986005230351954e-07, "epoch": 4.672, "percentage": 93.59, "elapsed_time": "4:41:43", "remaining_time": "0:19:17"} +{"current_steps": 366, "total_steps": 390, "loss": 0.3986, "lr": 4.5966069996365993e-07, "epoch": 4.6848, "percentage": 93.85, "elapsed_time": "4:42:29", "remaining_time": "0:18:31"} +{"current_steps": 367, "total_steps": 390, "loss": 0.3794, "lr": 4.22286238628562e-07, "epoch": 4.6975999999999996, "percentage": 94.1, "elapsed_time": "4:43:12", "remaining_time": "0:17:44"} +{"current_steps": 368, "total_steps": 390, "loss": 0.3808, "lr": 3.8648013306960664e-07, "epoch": 4.7104, "percentage": 94.36, "elapsed_time": "4:43:57", "remaining_time": "0:16:58"} +{"current_steps": 369, "total_steps": 390, "loss": 0.3912, "lr": 3.522452516867048e-07, "epoch": 4.7232, "percentage": 94.62, "elapsed_time": "4:44:41", "remaining_time": "0:16:12"} +{"current_steps": 370, "total_steps": 390, "loss": 0.3711, "lr": 3.1958433701019697e-07, "epoch": 4.736, "percentage": 94.87, "elapsed_time": "4:45:25", "remaining_time": "0:15:25"} +{"current_steps": 371, "total_steps": 390, "loss": 0.3759, "lr": 2.8850000548115155e-07, "epoch": 4.7488, "percentage": 95.13, "elapsed_time": "4:46:09", "remaining_time": "0:14:39"} +{"current_steps": 372, "total_steps": 390, "loss": 0.3818, "lr": 2.5899474724174313e-07, "epoch": 4.7616, "percentage": 95.38, "elapsed_time": "4:46:53", "remaining_time": "0:13:52"} +{"current_steps": 373, "total_steps": 390, "loss": 0.3909, "lr": 2.3107092593579905e-07, "epoch": 4.7744, "percentage": 95.64, "elapsed_time": "4:47:39", "remaining_time": "0:13:06"} +{"current_steps": 374, "total_steps": 390, "loss": 0.3876, "lr": 2.0473077851942858e-07, "epoch": 4.7872, "percentage": 95.9, "elapsed_time": "4:48:20", "remaining_time": "0:12:20"} +{"current_steps": 375, "total_steps": 390, "loss": 0.3926, "lr": 1.799764150818306e-07, "epoch": 4.8, "percentage": 96.15, "elapsed_time": "4:49:05", "remaining_time": "0:11:33"} +{"current_steps": 376, "total_steps": 390, "loss": 0.3902, "lr": 1.5680981867625566e-07, "epoch": 4.8128, "percentage": 96.41, "elapsed_time": "4:49:49", "remaining_time": "0:10:47"} +{"current_steps": 377, "total_steps": 390, "loss": 0.3818, "lr": 1.3523284516113955e-07, "epoch": 4.8256, "percentage": 96.67, "elapsed_time": "4:50:34", "remaining_time": "0:10:01"} +{"current_steps": 378, "total_steps": 390, "loss": 0.3852, "lr": 1.1524722305144231e-07, "epoch": 4.8384, "percentage": 96.92, "elapsed_time": "4:51:19", "remaining_time": "0:09:14"} +{"current_steps": 379, "total_steps": 390, "loss": 0.3988, "lr": 9.685455338016347e-08, "epoch": 4.8512, "percentage": 97.18, "elapsed_time": "4:52:04", "remaining_time": "0:08:28"} +{"current_steps": 380, "total_steps": 390, "loss": 0.3851, "lr": 8.005630957010014e-08, "epoch": 4.864, "percentage": 97.44, "elapsed_time": "4:52:49", "remaining_time": "0:07:42"} +{"current_steps": 381, "total_steps": 390, "loss": 0.3914, "lr": 6.485383731580142e-08, "epoch": 4.8768, "percentage": 97.69, "elapsed_time": "4:53:35", "remaining_time": "0:06:56"} +{"current_steps": 382, "total_steps": 390, "loss": 0.3845, "lr": 5.1248354475768034e-08, "epoch": 4.8896, "percentage": 97.95, "elapsed_time": "4:54:20", "remaining_time": "0:06:09"} +{"current_steps": 383, "total_steps": 390, "loss": 0.3775, "lr": 3.924095097489922e-08, "epoch": 4.9024, "percentage": 98.21, "elapsed_time": "4:55:06", "remaining_time": "0:05:23"} +{"current_steps": 384, "total_steps": 390, "loss": 0.3839, "lr": 2.8832588717164766e-08, "epoch": 4.9152000000000005, "percentage": 98.46, "elapsed_time": "4:55:51", "remaining_time": "0:04:37"} +{"current_steps": 385, "total_steps": 390, "loss": 0.3927, "lr": 2.0024101508555604e-08, "epoch": 4.928, "percentage": 98.72, "elapsed_time": "4:56:36", "remaining_time": "0:03:51"} +{"current_steps": 386, "total_steps": 390, "loss": 0.3819, "lr": 1.281619499029274e-08, "epoch": 4.9408, "percentage": 98.97, "elapsed_time": "4:57:20", "remaining_time": "0:03:04"} +{"current_steps": 387, "total_steps": 390, "loss": 0.3863, "lr": 7.209446582292501e-09, "epoch": 4.9536, "percentage": 99.23, "elapsed_time": "4:58:04", "remaining_time": "0:02:18"} +{"current_steps": 388, "total_steps": 390, "loss": 0.4027, "lr": 3.2043054369057523e-09, "epoch": 4.9664, "percentage": 99.49, "elapsed_time": "4:58:49", "remaining_time": "0:01:32"} +{"current_steps": 389, "total_steps": 390, "loss": 0.3858, "lr": 8.010924029533406e-10, "epoch": 4.9792, "percentage": 99.74, "elapsed_time": "4:59:34", "remaining_time": "0:00:46"} +{"current_steps": 390, "total_steps": 390, "loss": 0.3772, "lr": 0.0, "epoch": 4.992, "percentage": 100.0, "elapsed_time": "5:00:19", "remaining_time": "0:00:00"} +{"current_steps": 390, "total_steps": 390, "epoch": 4.992, "percentage": 100.0, "elapsed_time": "5:05:13", "remaining_time": "0:00:00"} diff --git a/trainer_state.json b/trainer_state.json new file mode 100644 index 0000000..b1162d2 --- /dev/null +++ b/trainer_state.json @@ -0,0 +1,2772 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 4.992, + "eval_steps": 500, + "global_step": 390, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0128, + "grad_norm": 6.394782504385327, + "learning_rate": 1.0256410256410257e-06, + "loss": 1.0426, + "step": 1 + }, + { + "epoch": 0.0256, + "grad_norm": 6.569259765346459, + "learning_rate": 2.0512820512820513e-06, + "loss": 1.0654, + "step": 2 + }, + { + "epoch": 0.0384, + "grad_norm": 6.230643663019568, + "learning_rate": 3.0769230769230774e-06, + "loss": 1.038, + "step": 3 + }, + { + "epoch": 0.0512, + "grad_norm": 5.728481734524085, + "learning_rate": 4.102564102564103e-06, + "loss": 1.0223, + "step": 4 + }, + { + "epoch": 0.064, + "grad_norm": 4.300311825507104, + "learning_rate": 5.128205128205128e-06, + "loss": 1.0198, + "step": 5 + }, + { + "epoch": 0.0768, + "grad_norm": 2.7305284613406227, + "learning_rate": 6.153846153846155e-06, + "loss": 0.9696, + "step": 6 + }, + { + "epoch": 0.0896, + "grad_norm": 2.4945340392246735, + "learning_rate": 7.17948717948718e-06, + "loss": 0.9539, + "step": 7 + }, + { + "epoch": 0.1024, + "grad_norm": 4.445254891417469, + "learning_rate": 8.205128205128205e-06, + "loss": 0.9903, + "step": 8 + }, + { + "epoch": 0.1152, + "grad_norm": 4.355662620054897, + "learning_rate": 9.230769230769232e-06, + "loss": 0.9684, + "step": 9 + }, + { + "epoch": 0.128, + "grad_norm": 3.994855636728309, + "learning_rate": 1.0256410256410256e-05, + "loss": 0.9146, + "step": 10 + }, + { + "epoch": 0.1408, + "grad_norm": 3.600143840830981, + "learning_rate": 1.1282051282051283e-05, + "loss": 0.9168, + "step": 11 + }, + { + "epoch": 0.1536, + "grad_norm": 2.5072401726977716, + "learning_rate": 1.230769230769231e-05, + "loss": 0.89, + "step": 12 + }, + { + "epoch": 0.1664, + "grad_norm": 1.5695905535048469, + "learning_rate": 1.3333333333333333e-05, + "loss": 0.8776, + "step": 13 + }, + { + "epoch": 0.1792, + "grad_norm": 1.476690203321329, + "learning_rate": 1.435897435897436e-05, + "loss": 0.8364, + "step": 14 + }, + { + "epoch": 0.192, + "grad_norm": 1.2652613857174815, + "learning_rate": 1.5384615384615387e-05, + "loss": 0.8212, + "step": 15 + }, + { + "epoch": 0.2048, + "grad_norm": 1.0424739509209286, + "learning_rate": 1.641025641025641e-05, + "loss": 0.8171, + "step": 16 + }, + { + "epoch": 0.2176, + "grad_norm": 1.1502664604164308, + "learning_rate": 1.7435897435897438e-05, + "loss": 0.7992, + "step": 17 + }, + { + "epoch": 0.2304, + "grad_norm": 1.073919261807942, + "learning_rate": 1.8461538461538465e-05, + "loss": 0.7841, + "step": 18 + }, + { + "epoch": 0.2432, + "grad_norm": 0.9196218647189659, + "learning_rate": 1.9487179487179488e-05, + "loss": 0.788, + "step": 19 + }, + { + "epoch": 0.256, + "grad_norm": 1.0253545230334442, + "learning_rate": 2.0512820512820512e-05, + "loss": 0.7642, + "step": 20 + }, + { + "epoch": 0.2688, + "grad_norm": 1.1328893054322893, + "learning_rate": 2.153846153846154e-05, + "loss": 0.7661, + "step": 21 + }, + { + "epoch": 0.2816, + "grad_norm": 0.8703556504095974, + "learning_rate": 2.2564102564102566e-05, + "loss": 0.7551, + "step": 22 + }, + { + "epoch": 0.2944, + "grad_norm": 1.0207044696700907, + "learning_rate": 2.3589743589743593e-05, + "loss": 0.744, + "step": 23 + }, + { + "epoch": 0.3072, + "grad_norm": 0.7589287619845236, + "learning_rate": 2.461538461538462e-05, + "loss": 0.7488, + "step": 24 + }, + { + "epoch": 0.32, + "grad_norm": 0.740061535551805, + "learning_rate": 2.5641025641025646e-05, + "loss": 0.7256, + "step": 25 + }, + { + "epoch": 0.3328, + "grad_norm": 0.8413097464471732, + "learning_rate": 2.6666666666666667e-05, + "loss": 0.737, + "step": 26 + }, + { + "epoch": 0.3456, + "grad_norm": 0.9536505328257563, + "learning_rate": 2.7692307692307694e-05, + "loss": 0.7581, + "step": 27 + }, + { + "epoch": 0.3584, + "grad_norm": 1.4675472622777184, + "learning_rate": 2.871794871794872e-05, + "loss": 0.7118, + "step": 28 + }, + { + "epoch": 0.3712, + "grad_norm": 0.7463919210078415, + "learning_rate": 2.9743589743589747e-05, + "loss": 0.7304, + "step": 29 + }, + { + "epoch": 0.384, + "grad_norm": 1.5192710771177331, + "learning_rate": 3.0769230769230774e-05, + "loss": 0.7412, + "step": 30 + }, + { + "epoch": 0.3968, + "grad_norm": 0.9332868316358716, + "learning_rate": 3.1794871794871795e-05, + "loss": 0.7142, + "step": 31 + }, + { + "epoch": 0.4096, + "grad_norm": 1.03647672191691, + "learning_rate": 3.282051282051282e-05, + "loss": 0.7098, + "step": 32 + }, + { + "epoch": 0.4224, + "grad_norm": 1.0489540813968616, + "learning_rate": 3.384615384615385e-05, + "loss": 0.718, + "step": 33 + }, + { + "epoch": 0.4352, + "grad_norm": 1.7415718454005513, + "learning_rate": 3.4871794871794875e-05, + "loss": 0.7281, + "step": 34 + }, + { + "epoch": 0.448, + "grad_norm": 1.1834722099310977, + "learning_rate": 3.58974358974359e-05, + "loss": 0.7008, + "step": 35 + }, + { + "epoch": 0.4608, + "grad_norm": 2.474981702846964, + "learning_rate": 3.692307692307693e-05, + "loss": 0.7226, + "step": 36 + }, + { + "epoch": 0.4736, + "grad_norm": 2.4830307714121873, + "learning_rate": 3.794871794871795e-05, + "loss": 0.7272, + "step": 37 + }, + { + "epoch": 0.4864, + "grad_norm": 1.1456942753565234, + "learning_rate": 3.8974358974358976e-05, + "loss": 0.7103, + "step": 38 + }, + { + "epoch": 0.4992, + "grad_norm": 1.6801001628191694, + "learning_rate": 4e-05, + "loss": 0.7066, + "step": 39 + }, + { + "epoch": 0.512, + "grad_norm": 1.0954177872438302, + "learning_rate": 3.9999198907597046e-05, + "loss": 0.7289, + "step": 40 + }, + { + "epoch": 0.5248, + "grad_norm": 1.3266879656505657, + "learning_rate": 3.9996795694563096e-05, + "loss": 0.7014, + "step": 41 + }, + { + "epoch": 0.5376, + "grad_norm": 1.2996108899410874, + "learning_rate": 3.999279055341771e-05, + "loss": 0.6994, + "step": 42 + }, + { + "epoch": 0.5504, + "grad_norm": 1.2080851674726614, + "learning_rate": 3.998718380500971e-05, + "loss": 0.7036, + "step": 43 + }, + { + "epoch": 0.5632, + "grad_norm": 1.000152159295558, + "learning_rate": 3.997997589849145e-05, + "loss": 0.7017, + "step": 44 + }, + { + "epoch": 0.576, + "grad_norm": 0.9903795895140601, + "learning_rate": 3.9971167411282835e-05, + "loss": 0.7039, + "step": 45 + }, + { + "epoch": 0.5888, + "grad_norm": 1.0346243706022227, + "learning_rate": 3.99607590490251e-05, + "loss": 0.6928, + "step": 46 + }, + { + "epoch": 0.6016, + "grad_norm": 1.2310623651435908, + "learning_rate": 3.9948751645524235e-05, + "loss": 0.6926, + "step": 47 + }, + { + "epoch": 0.6144, + "grad_norm": 0.864595484800581, + "learning_rate": 3.9935146162684206e-05, + "loss": 0.6994, + "step": 48 + }, + { + "epoch": 0.6272, + "grad_norm": 0.8355436411996121, + "learning_rate": 3.9919943690429906e-05, + "loss": 0.6934, + "step": 49 + }, + { + "epoch": 0.64, + "grad_norm": 0.9997652645558224, + "learning_rate": 3.9903145446619837e-05, + "loss": 0.7097, + "step": 50 + }, + { + "epoch": 0.6528, + "grad_norm": 1.0298649825320771, + "learning_rate": 3.9884752776948564e-05, + "loss": 0.6851, + "step": 51 + }, + { + "epoch": 0.6656, + "grad_norm": 0.9113000462888259, + "learning_rate": 3.9864767154838864e-05, + "loss": 0.6924, + "step": 52 + }, + { + "epoch": 0.6784, + "grad_norm": 0.7614156810052238, + "learning_rate": 3.9843190181323744e-05, + "loss": 0.6795, + "step": 53 + }, + { + "epoch": 0.6912, + "grad_norm": 0.9253337882696134, + "learning_rate": 3.982002358491817e-05, + "loss": 0.6686, + "step": 54 + }, + { + "epoch": 0.704, + "grad_norm": 0.8701037905779012, + "learning_rate": 3.979526922148058e-05, + "loss": 0.7001, + "step": 55 + }, + { + "epoch": 0.7168, + "grad_norm": 0.7437912252700782, + "learning_rate": 3.9768929074064206e-05, + "loss": 0.6792, + "step": 56 + }, + { + "epoch": 0.7296, + "grad_norm": 0.8094577165731788, + "learning_rate": 3.9741005252758255e-05, + "loss": 0.6821, + "step": 57 + }, + { + "epoch": 0.7424, + "grad_norm": 0.7509802839033517, + "learning_rate": 3.971149999451886e-05, + "loss": 0.6715, + "step": 58 + }, + { + "epoch": 0.7552, + "grad_norm": 0.595632383004398, + "learning_rate": 3.9680415662989806e-05, + "loss": 0.6738, + "step": 59 + }, + { + "epoch": 0.768, + "grad_norm": 0.738106832398729, + "learning_rate": 3.9647754748313294e-05, + "loss": 0.6776, + "step": 60 + }, + { + "epoch": 0.7808, + "grad_norm": 0.9793451997195339, + "learning_rate": 3.96135198669304e-05, + "loss": 0.6723, + "step": 61 + }, + { + "epoch": 0.7936, + "grad_norm": 0.9038400518704891, + "learning_rate": 3.957771376137144e-05, + "loss": 0.6804, + "step": 62 + }, + { + "epoch": 0.8064, + "grad_norm": 0.8940165413819113, + "learning_rate": 3.954033930003634e-05, + "loss": 0.6866, + "step": 63 + }, + { + "epoch": 0.8192, + "grad_norm": 0.6916586084691009, + "learning_rate": 3.9501399476964806e-05, + "loss": 0.6812, + "step": 64 + }, + { + "epoch": 0.832, + "grad_norm": 0.5621936446481481, + "learning_rate": 3.946089741159648e-05, + "loss": 0.6734, + "step": 65 + }, + { + "epoch": 0.8448, + "grad_norm": 0.586800751526733, + "learning_rate": 3.9418836348521045e-05, + "loss": 0.6962, + "step": 66 + }, + { + "epoch": 0.8576, + "grad_norm": 0.6760075379935138, + "learning_rate": 3.937521965721831e-05, + "loss": 0.6791, + "step": 67 + }, + { + "epoch": 0.8704, + "grad_norm": 0.7234942389255747, + "learning_rate": 3.933005083178828e-05, + "loss": 0.672, + "step": 68 + }, + { + "epoch": 0.8832, + "grad_norm": 0.609112508064292, + "learning_rate": 3.928333349067125e-05, + "loss": 0.69, + "step": 69 + }, + { + "epoch": 0.896, + "grad_norm": 0.80856892769616, + "learning_rate": 3.923507137635792e-05, + "loss": 0.6679, + "step": 70 + }, + { + "epoch": 0.9088, + "grad_norm": 0.8135651692515927, + "learning_rate": 3.9185268355089606e-05, + "loss": 0.6576, + "step": 71 + }, + { + "epoch": 0.9216, + "grad_norm": 0.8569740956861722, + "learning_rate": 3.913392841654851e-05, + "loss": 0.6752, + "step": 72 + }, + { + "epoch": 0.9344, + "grad_norm": 0.9734106522191275, + "learning_rate": 3.9081055673538093e-05, + "loss": 0.666, + "step": 73 + }, + { + "epoch": 0.9472, + "grad_norm": 0.9734454565935635, + "learning_rate": 3.902665436165364e-05, + "loss": 0.6604, + "step": 74 + }, + { + "epoch": 0.96, + "grad_norm": 0.8611505130270999, + "learning_rate": 3.897072883894291e-05, + "loss": 0.6701, + "step": 75 + }, + { + "epoch": 0.9728, + "grad_norm": 0.8768484096515019, + "learning_rate": 3.8913283585557054e-05, + "loss": 0.6736, + "step": 76 + }, + { + "epoch": 0.9856, + "grad_norm": 0.6899441671900481, + "learning_rate": 3.885432320339167e-05, + "loss": 0.6857, + "step": 77 + }, + { + "epoch": 0.9984, + "grad_norm": 0.619076353872035, + "learning_rate": 3.879385241571817e-05, + "loss": 0.657, + "step": 78 + }, + { + "epoch": 1.0112, + "grad_norm": 0.6689305394134695, + "learning_rate": 3.873187606680543e-05, + "loss": 0.6226, + "step": 79 + }, + { + "epoch": 1.024, + "grad_norm": 0.5388474756672614, + "learning_rate": 3.866839912153168e-05, + "loss": 0.5975, + "step": 80 + }, + { + "epoch": 1.0368, + "grad_norm": 0.6898767568816427, + "learning_rate": 3.860342666498677e-05, + "loss": 0.6125, + "step": 81 + }, + { + "epoch": 1.0496, + "grad_norm": 0.8180878679575876, + "learning_rate": 3.853696390206484e-05, + "loss": 0.6142, + "step": 82 + }, + { + "epoch": 1.0624, + "grad_norm": 1.0725542408684852, + "learning_rate": 3.846901615704734e-05, + "loss": 0.6042, + "step": 83 + }, + { + "epoch": 1.0752, + "grad_norm": 0.92505515366403, + "learning_rate": 3.839958887317649e-05, + "loss": 0.6097, + "step": 84 + }, + { + "epoch": 1.088, + "grad_norm": 0.7689887829776089, + "learning_rate": 3.832868761221926e-05, + "loss": 0.6127, + "step": 85 + }, + { + "epoch": 1.1008, + "grad_norm": 0.9516898192549433, + "learning_rate": 3.825631805402182e-05, + "loss": 0.6127, + "step": 86 + }, + { + "epoch": 1.1136, + "grad_norm": 0.7950966316561936, + "learning_rate": 3.818248599605448e-05, + "loss": 0.6091, + "step": 87 + }, + { + "epoch": 1.1264, + "grad_norm": 0.6665255828836638, + "learning_rate": 3.810719735294731e-05, + "loss": 0.6156, + "step": 88 + }, + { + "epoch": 1.1392, + "grad_norm": 0.7540335740571178, + "learning_rate": 3.8030458156016326e-05, + "loss": 0.6051, + "step": 89 + }, + { + "epoch": 1.152, + "grad_norm": 0.8629803559587889, + "learning_rate": 3.795227455278029e-05, + "loss": 0.6066, + "step": 90 + }, + { + "epoch": 1.1648, + "grad_norm": 0.6079801774673296, + "learning_rate": 3.787265280646825e-05, + "loss": 0.5973, + "step": 91 + }, + { + "epoch": 1.1776, + "grad_norm": 0.7646030455094784, + "learning_rate": 3.7791599295517825e-05, + "loss": 0.609, + "step": 92 + }, + { + "epoch": 1.1904, + "grad_norm": 0.6181653509538091, + "learning_rate": 3.7709120513064196e-05, + "loss": 0.6243, + "step": 93 + }, + { + "epoch": 1.2032, + "grad_norm": 0.6819268210545444, + "learning_rate": 3.762522306641998e-05, + "loss": 0.6006, + "step": 94 + }, + { + "epoch": 1.216, + "grad_norm": 0.7024059883235171, + "learning_rate": 3.7539913676545874e-05, + "loss": 0.6086, + "step": 95 + }, + { + "epoch": 1.2288000000000001, + "grad_norm": 0.575753668683066, + "learning_rate": 3.745319917751229e-05, + "loss": 0.611, + "step": 96 + }, + { + "epoch": 1.2416, + "grad_norm": 0.8041432384680723, + "learning_rate": 3.736508651595188e-05, + "loss": 0.5997, + "step": 97 + }, + { + "epoch": 1.2544, + "grad_norm": 0.44174254522297657, + "learning_rate": 3.727558275050301e-05, + "loss": 0.6057, + "step": 98 + }, + { + "epoch": 1.2671999999999999, + "grad_norm": 0.6028624132341728, + "learning_rate": 3.718469505124434e-05, + "loss": 0.5939, + "step": 99 + }, + { + "epoch": 1.28, + "grad_norm": 0.4607737998378977, + "learning_rate": 3.709243069912041e-05, + "loss": 0.5954, + "step": 100 + }, + { + "epoch": 1.2928, + "grad_norm": 0.5735717821893057, + "learning_rate": 3.699879708535838e-05, + "loss": 0.6158, + "step": 101 + }, + { + "epoch": 1.3056, + "grad_norm": 0.6925796120242844, + "learning_rate": 3.69038017108759e-05, + "loss": 0.5978, + "step": 102 + }, + { + "epoch": 1.3184, + "grad_norm": 0.7422327396906829, + "learning_rate": 3.680745218568026e-05, + "loss": 0.5922, + "step": 103 + }, + { + "epoch": 1.3312, + "grad_norm": 0.563332719790965, + "learning_rate": 3.6709756228258735e-05, + "loss": 0.6167, + "step": 104 + }, + { + "epoch": 1.3439999999999999, + "grad_norm": 0.5609951130215706, + "learning_rate": 3.6610721664960236e-05, + "loss": 0.6129, + "step": 105 + }, + { + "epoch": 1.3568, + "grad_norm": 0.7534019038629993, + "learning_rate": 3.65103564293684e-05, + "loss": 0.596, + "step": 106 + }, + { + "epoch": 1.3696, + "grad_norm": 0.5055194387609822, + "learning_rate": 3.640866856166601e-05, + "loss": 0.5904, + "step": 107 + }, + { + "epoch": 1.3824, + "grad_norm": 0.5558736753816589, + "learning_rate": 3.6305666207990886e-05, + "loss": 0.6202, + "step": 108 + }, + { + "epoch": 1.3952, + "grad_norm": 0.4617367226090242, + "learning_rate": 3.6201357619783336e-05, + "loss": 0.6051, + "step": 109 + }, + { + "epoch": 1.408, + "grad_norm": 0.5344658374625233, + "learning_rate": 3.609575115312511e-05, + "loss": 0.6219, + "step": 110 + }, + { + "epoch": 1.4208, + "grad_norm": 0.5782265860951042, + "learning_rate": 3.598885526807003e-05, + "loss": 0.6054, + "step": 111 + }, + { + "epoch": 1.4336, + "grad_norm": 0.42442597924731185, + "learning_rate": 3.5880678527966224e-05, + "loss": 0.5959, + "step": 112 + }, + { + "epoch": 1.4464000000000001, + "grad_norm": 0.6114986758491804, + "learning_rate": 3.577122959877017e-05, + "loss": 0.599, + "step": 113 + }, + { + "epoch": 1.4592, + "grad_norm": 0.46032848870907744, + "learning_rate": 3.566051724835245e-05, + "loss": 0.6114, + "step": 114 + }, + { + "epoch": 1.472, + "grad_norm": 0.5009551581292482, + "learning_rate": 3.554855034579532e-05, + "loss": 0.5967, + "step": 115 + }, + { + "epoch": 1.4848, + "grad_norm": 0.499487518866959, + "learning_rate": 3.5435337860682304e-05, + "loss": 0.6005, + "step": 116 + }, + { + "epoch": 1.4976, + "grad_norm": 0.4637149314493999, + "learning_rate": 3.532088886237956e-05, + "loss": 0.5997, + "step": 117 + }, + { + "epoch": 1.5104, + "grad_norm": 0.5997649370066328, + "learning_rate": 3.520521251930941e-05, + "loss": 0.6009, + "step": 118 + }, + { + "epoch": 1.5232, + "grad_norm": 0.49394183241808043, + "learning_rate": 3.5088318098215805e-05, + "loss": 0.5893, + "step": 119 + }, + { + "epoch": 1.536, + "grad_norm": 0.47388103629083245, + "learning_rate": 3.497021496342203e-05, + "loss": 0.6036, + "step": 120 + }, + { + "epoch": 1.5488, + "grad_norm": 0.542093278623551, + "learning_rate": 3.485091257608047e-05, + "loss": 0.6002, + "step": 121 + }, + { + "epoch": 1.5615999999999999, + "grad_norm": 0.6431580839857001, + "learning_rate": 3.473042049341474e-05, + "loss": 0.6137, + "step": 122 + }, + { + "epoch": 1.5744, + "grad_norm": 0.5003369599376405, + "learning_rate": 3.4608748367954064e-05, + "loss": 0.6043, + "step": 123 + }, + { + "epoch": 1.5872000000000002, + "grad_norm": 0.6569135072252106, + "learning_rate": 3.4485905946759965e-05, + "loss": 0.5843, + "step": 124 + }, + { + "epoch": 1.6, + "grad_norm": 0.5280937820334869, + "learning_rate": 3.4361903070645484e-05, + "loss": 0.5968, + "step": 125 + }, + { + "epoch": 1.6128, + "grad_norm": 0.47560785703947545, + "learning_rate": 3.423674967338681e-05, + "loss": 0.5962, + "step": 126 + }, + { + "epoch": 1.6256, + "grad_norm": 0.6169951917734525, + "learning_rate": 3.411045578092754e-05, + "loss": 0.5949, + "step": 127 + }, + { + "epoch": 1.6383999999999999, + "grad_norm": 0.4562836643712705, + "learning_rate": 3.398303151057543e-05, + "loss": 0.5983, + "step": 128 + }, + { + "epoch": 1.6512, + "grad_norm": 0.49312549921360643, + "learning_rate": 3.385448707019199e-05, + "loss": 0.5918, + "step": 129 + }, + { + "epoch": 1.6640000000000001, + "grad_norm": 0.5860158062199761, + "learning_rate": 3.372483275737468e-05, + "loss": 0.606, + "step": 130 + }, + { + "epoch": 1.6768, + "grad_norm": 0.5490285575430652, + "learning_rate": 3.359407895863199e-05, + "loss": 0.6001, + "step": 131 + }, + { + "epoch": 1.6896, + "grad_norm": 0.6348820038195376, + "learning_rate": 3.34622361485514e-05, + "loss": 0.6024, + "step": 132 + }, + { + "epoch": 1.7024, + "grad_norm": 0.42576475419719334, + "learning_rate": 3.332931488896029e-05, + "loss": 0.5933, + "step": 133 + }, + { + "epoch": 1.7151999999999998, + "grad_norm": 0.5900041503972492, + "learning_rate": 3.319532582807977e-05, + "loss": 0.5932, + "step": 134 + }, + { + "epoch": 1.728, + "grad_norm": 0.39127515320743367, + "learning_rate": 3.30602796996717e-05, + "loss": 0.6069, + "step": 135 + }, + { + "epoch": 1.7408000000000001, + "grad_norm": 0.4498915823225618, + "learning_rate": 3.2924187322178865e-05, + "loss": 0.5996, + "step": 136 + }, + { + "epoch": 1.7536, + "grad_norm": 0.42155081455650345, + "learning_rate": 3.278705959785821e-05, + "loss": 0.609, + "step": 137 + }, + { + "epoch": 1.7664, + "grad_norm": 0.4812165543426556, + "learning_rate": 3.2648907511907544e-05, + "loss": 0.5936, + "step": 138 + }, + { + "epoch": 1.7792, + "grad_norm": 0.39696305993195713, + "learning_rate": 3.250974213158555e-05, + "loss": 0.5788, + "step": 139 + }, + { + "epoch": 1.792, + "grad_norm": 0.4365984108187007, + "learning_rate": 3.23695746053251e-05, + "loss": 0.5921, + "step": 140 + }, + { + "epoch": 1.8048, + "grad_norm": 0.5408967696030647, + "learning_rate": 3.222841616184025e-05, + "loss": 0.6161, + "step": 141 + }, + { + "epoch": 1.8176, + "grad_norm": 0.4144438614201316, + "learning_rate": 3.208627810922665e-05, + "loss": 0.5922, + "step": 142 + }, + { + "epoch": 1.8304, + "grad_norm": 0.5737519474487879, + "learning_rate": 3.194317183405573e-05, + "loss": 0.5768, + "step": 143 + }, + { + "epoch": 1.8432, + "grad_norm": 0.47055855879022757, + "learning_rate": 3.1799108800462466e-05, + "loss": 0.5881, + "step": 144 + }, + { + "epoch": 1.8559999999999999, + "grad_norm": 0.5077652807841551, + "learning_rate": 3.1654100549227024e-05, + "loss": 0.5979, + "step": 145 + }, + { + "epoch": 1.8688, + "grad_norm": 0.5925352581007367, + "learning_rate": 3.1508158696850275e-05, + "loss": 0.592, + "step": 146 + }, + { + "epoch": 1.8816000000000002, + "grad_norm": 0.43648166295010277, + "learning_rate": 3.136129493462312e-05, + "loss": 0.5953, + "step": 147 + }, + { + "epoch": 1.8944, + "grad_norm": 0.4711597990350436, + "learning_rate": 3.121352102768998e-05, + "loss": 0.5995, + "step": 148 + }, + { + "epoch": 1.9072, + "grad_norm": 0.4476044182976627, + "learning_rate": 3.106484881410628e-05, + "loss": 0.5979, + "step": 149 + }, + { + "epoch": 1.92, + "grad_norm": 0.38283511609840415, + "learning_rate": 3.091529020389009e-05, + "loss": 0.5863, + "step": 150 + }, + { + "epoch": 1.9327999999999999, + "grad_norm": 0.5371783397006327, + "learning_rate": 3.076485717806808e-05, + "loss": 0.5816, + "step": 151 + }, + { + "epoch": 1.9456, + "grad_norm": 0.45170660642774, + "learning_rate": 3.061356178771564e-05, + "loss": 0.5815, + "step": 152 + }, + { + "epoch": 1.9584000000000001, + "grad_norm": 0.4753466564865614, + "learning_rate": 3.0461416152991555e-05, + "loss": 0.5893, + "step": 153 + }, + { + "epoch": 1.9712, + "grad_norm": 0.597902042974497, + "learning_rate": 3.0308432462167045e-05, + "loss": 0.5974, + "step": 154 + }, + { + "epoch": 1.984, + "grad_norm": 0.4247429930070502, + "learning_rate": 3.015462297064936e-05, + "loss": 0.5916, + "step": 155 + }, + { + "epoch": 1.9968, + "grad_norm": 0.5700650883732693, + "learning_rate": 3.0000000000000004e-05, + "loss": 0.5932, + "step": 156 + }, + { + "epoch": 2.0096, + "grad_norm": 0.46544682012163896, + "learning_rate": 2.98445759369477e-05, + "loss": 0.5451, + "step": 157 + }, + { + "epoch": 2.0224, + "grad_norm": 0.45971846284651463, + "learning_rate": 2.9688363232396056e-05, + "loss": 0.5179, + "step": 158 + }, + { + "epoch": 2.0352, + "grad_norm": 0.6999433468335587, + "learning_rate": 2.9531374400426158e-05, + "loss": 0.5069, + "step": 159 + }, + { + "epoch": 2.048, + "grad_norm": 0.6367425413782538, + "learning_rate": 2.9373622017294075e-05, + "loss": 0.5256, + "step": 160 + }, + { + "epoch": 2.0608, + "grad_norm": 0.6608868802831678, + "learning_rate": 2.9215118720423375e-05, + "loss": 0.5169, + "step": 161 + }, + { + "epoch": 2.0736, + "grad_norm": 0.48788367916038633, + "learning_rate": 2.9055877207392752e-05, + "loss": 0.5095, + "step": 162 + }, + { + "epoch": 2.0864, + "grad_norm": 0.6797298089521543, + "learning_rate": 2.8895910234918828e-05, + "loss": 0.5171, + "step": 163 + }, + { + "epoch": 2.0992, + "grad_norm": 0.6895143600985311, + "learning_rate": 2.873523061783426e-05, + "loss": 0.506, + "step": 164 + }, + { + "epoch": 2.112, + "grad_norm": 0.5092110549661606, + "learning_rate": 2.8573851228061084e-05, + "loss": 0.511, + "step": 165 + }, + { + "epoch": 2.1248, + "grad_norm": 0.6408741928897569, + "learning_rate": 2.8411784993579633e-05, + "loss": 0.5179, + "step": 166 + }, + { + "epoch": 2.1376, + "grad_norm": 0.5003463226866105, + "learning_rate": 2.8249044897392814e-05, + "loss": 0.5217, + "step": 167 + }, + { + "epoch": 2.1504, + "grad_norm": 0.46482236893081896, + "learning_rate": 2.80856439764861e-05, + "loss": 0.5162, + "step": 168 + }, + { + "epoch": 2.1632, + "grad_norm": 0.5535382177980431, + "learning_rate": 2.792159532078314e-05, + "loss": 0.5089, + "step": 169 + }, + { + "epoch": 2.176, + "grad_norm": 0.5069367991468174, + "learning_rate": 2.77569120720971e-05, + "loss": 0.517, + "step": 170 + }, + { + "epoch": 2.1888, + "grad_norm": 0.4895756496702128, + "learning_rate": 2.7591607423077932e-05, + "loss": 0.5204, + "step": 171 + }, + { + "epoch": 2.2016, + "grad_norm": 0.4484537242040031, + "learning_rate": 2.7425694616155474e-05, + "loss": 0.5006, + "step": 172 + }, + { + "epoch": 2.2144, + "grad_norm": 0.39926621412366664, + "learning_rate": 2.7259186942478656e-05, + "loss": 0.5018, + "step": 173 + }, + { + "epoch": 2.2272, + "grad_norm": 0.46792399045602356, + "learning_rate": 2.7092097740850712e-05, + "loss": 0.5061, + "step": 174 + }, + { + "epoch": 2.24, + "grad_norm": 0.4440563705223137, + "learning_rate": 2.692444039666066e-05, + "loss": 0.5019, + "step": 175 + }, + { + "epoch": 2.2528, + "grad_norm": 0.46549376329458764, + "learning_rate": 2.6756228340810946e-05, + "loss": 0.5035, + "step": 176 + }, + { + "epoch": 2.2656, + "grad_norm": 0.38653389290400647, + "learning_rate": 2.6587475048641596e-05, + "loss": 0.5084, + "step": 177 + }, + { + "epoch": 2.2784, + "grad_norm": 0.48544129475234504, + "learning_rate": 2.6418194038850634e-05, + "loss": 0.5094, + "step": 178 + }, + { + "epoch": 2.2912, + "grad_norm": 0.3900570960351546, + "learning_rate": 2.624839887241115e-05, + "loss": 0.5108, + "step": 179 + }, + { + "epoch": 2.304, + "grad_norm": 0.5069783257446727, + "learning_rate": 2.607810315148494e-05, + "loss": 0.5041, + "step": 180 + }, + { + "epoch": 2.3168, + "grad_norm": 0.3945371522760007, + "learning_rate": 2.5907320518332827e-05, + "loss": 0.5099, + "step": 181 + }, + { + "epoch": 2.3296, + "grad_norm": 0.562788336543834, + "learning_rate": 2.5736064654221808e-05, + "loss": 0.5174, + "step": 182 + }, + { + "epoch": 2.3424, + "grad_norm": 0.4545310242783742, + "learning_rate": 2.5564349278329056e-05, + "loss": 0.5046, + "step": 183 + }, + { + "epoch": 2.3552, + "grad_norm": 0.3875509270098935, + "learning_rate": 2.539218814664288e-05, + "loss": 0.5145, + "step": 184 + }, + { + "epoch": 2.368, + "grad_norm": 0.46461018413371363, + "learning_rate": 2.521959505086075e-05, + "loss": 0.4889, + "step": 185 + }, + { + "epoch": 2.3808, + "grad_norm": 0.3464919112988565, + "learning_rate": 2.5046583817284437e-05, + "loss": 0.5229, + "step": 186 + }, + { + "epoch": 2.3936, + "grad_norm": 0.4739230855888146, + "learning_rate": 2.487316830571244e-05, + "loss": 0.5045, + "step": 187 + }, + { + "epoch": 2.4064, + "grad_norm": 0.4527569882340709, + "learning_rate": 2.4699362408329646e-05, + "loss": 0.5007, + "step": 188 + }, + { + "epoch": 2.4192, + "grad_norm": 0.43278250897184295, + "learning_rate": 2.4525180048594452e-05, + "loss": 0.4951, + "step": 189 + }, + { + "epoch": 2.432, + "grad_norm": 0.41300510138939983, + "learning_rate": 2.435063518012335e-05, + "loss": 0.5203, + "step": 190 + }, + { + "epoch": 2.4448, + "grad_norm": 0.37260966158544223, + "learning_rate": 2.4175741785573177e-05, + "loss": 0.5117, + "step": 191 + }, + { + "epoch": 2.4576000000000002, + "grad_norm": 0.36382874275957855, + "learning_rate": 2.4000513875520892e-05, + "loss": 0.503, + "step": 192 + }, + { + "epoch": 2.4704, + "grad_norm": 0.3986579409856395, + "learning_rate": 2.3824965487341247e-05, + "loss": 0.5193, + "step": 193 + }, + { + "epoch": 2.4832, + "grad_norm": 0.34853392834188407, + "learning_rate": 2.3649110684082258e-05, + "loss": 0.5247, + "step": 194 + }, + { + "epoch": 2.496, + "grad_norm": 0.342609875000865, + "learning_rate": 2.3472963553338614e-05, + "loss": 0.5066, + "step": 195 + }, + { + "epoch": 2.5088, + "grad_norm": 0.36160657632661647, + "learning_rate": 2.3296538206123134e-05, + "loss": 0.5154, + "step": 196 + }, + { + "epoch": 2.5216, + "grad_norm": 0.3463567044676736, + "learning_rate": 2.311984877573636e-05, + "loss": 0.5199, + "step": 197 + }, + { + "epoch": 2.5343999999999998, + "grad_norm": 0.4058131993726706, + "learning_rate": 2.2942909416634326e-05, + "loss": 0.5186, + "step": 198 + }, + { + "epoch": 2.5472, + "grad_norm": 0.4057321265113332, + "learning_rate": 2.2765734303294666e-05, + "loss": 0.5143, + "step": 199 + }, + { + "epoch": 2.56, + "grad_norm": 0.3936214661542984, + "learning_rate": 2.2588337629081107e-05, + "loss": 0.5132, + "step": 200 + }, + { + "epoch": 2.5728, + "grad_norm": 0.370528958570038, + "learning_rate": 2.2410733605106462e-05, + "loss": 0.5244, + "step": 201 + }, + { + "epoch": 2.5856, + "grad_norm": 0.4214268915896029, + "learning_rate": 2.2232936459094158e-05, + "loss": 0.4999, + "step": 202 + }, + { + "epoch": 2.5984, + "grad_norm": 0.34631700186838116, + "learning_rate": 2.205496043423849e-05, + "loss": 0.5164, + "step": 203 + }, + { + "epoch": 2.6112, + "grad_norm": 0.40937456281219886, + "learning_rate": 2.1876819788063586e-05, + "loss": 0.5052, + "step": 204 + }, + { + "epoch": 2.624, + "grad_norm": 0.4574105884530299, + "learning_rate": 2.16985287912813e-05, + "loss": 0.5121, + "step": 205 + }, + { + "epoch": 2.6368, + "grad_norm": 0.3923599307357846, + "learning_rate": 2.1520101726647922e-05, + "loss": 0.5216, + "step": 206 + }, + { + "epoch": 2.6496, + "grad_norm": 0.46225848050812934, + "learning_rate": 2.1341552887820048e-05, + "loss": 0.52, + "step": 207 + }, + { + "epoch": 2.6624, + "grad_norm": 0.3813625635848944, + "learning_rate": 2.1162896578209517e-05, + "loss": 0.5233, + "step": 208 + }, + { + "epoch": 2.6752000000000002, + "grad_norm": 0.36829375003858267, + "learning_rate": 2.0984147109837564e-05, + "loss": 0.5107, + "step": 209 + }, + { + "epoch": 2.6879999999999997, + "grad_norm": 0.40407451080108014, + "learning_rate": 2.0805318802188307e-05, + "loss": 0.5211, + "step": 210 + }, + { + "epoch": 2.7008, + "grad_norm": 0.330097221116534, + "learning_rate": 2.0626425981061608e-05, + "loss": 0.5184, + "step": 211 + }, + { + "epoch": 2.7136, + "grad_norm": 0.4569268466278724, + "learning_rate": 2.0447482977425465e-05, + "loss": 0.4971, + "step": 212 + }, + { + "epoch": 2.7264, + "grad_norm": 0.32210414711988283, + "learning_rate": 2.0268504126267952e-05, + "loss": 0.5034, + "step": 213 + }, + { + "epoch": 2.7392, + "grad_norm": 0.4593913215028703, + "learning_rate": 2.008950376544887e-05, + "loss": 0.5055, + "step": 214 + }, + { + "epoch": 2.752, + "grad_norm": 0.32532356865740414, + "learning_rate": 1.9910496234551132e-05, + "loss": 0.5129, + "step": 215 + }, + { + "epoch": 2.7648, + "grad_norm": 0.45486772372147316, + "learning_rate": 1.9731495873732055e-05, + "loss": 0.5142, + "step": 216 + }, + { + "epoch": 2.7776, + "grad_norm": 0.3298842662400382, + "learning_rate": 1.9552517022574542e-05, + "loss": 0.5199, + "step": 217 + }, + { + "epoch": 2.7904, + "grad_norm": 0.4043174303350555, + "learning_rate": 1.93735740189384e-05, + "loss": 0.5118, + "step": 218 + }, + { + "epoch": 2.8032, + "grad_norm": 0.31749811135674905, + "learning_rate": 1.9194681197811703e-05, + "loss": 0.5202, + "step": 219 + }, + { + "epoch": 2.816, + "grad_norm": 0.412159163435511, + "learning_rate": 1.901585289016244e-05, + "loss": 0.5204, + "step": 220 + }, + { + "epoch": 2.8288, + "grad_norm": 0.3107510340812756, + "learning_rate": 1.8837103421790486e-05, + "loss": 0.5242, + "step": 221 + }, + { + "epoch": 2.8416, + "grad_norm": 0.390287835259627, + "learning_rate": 1.8658447112179952e-05, + "loss": 0.5121, + "step": 222 + }, + { + "epoch": 2.8544, + "grad_norm": 0.3287344109428031, + "learning_rate": 1.8479898273352084e-05, + "loss": 0.5097, + "step": 223 + }, + { + "epoch": 2.8672, + "grad_norm": 0.3374507455352951, + "learning_rate": 1.83014712087187e-05, + "loss": 0.5095, + "step": 224 + }, + { + "epoch": 2.88, + "grad_norm": 0.33332837158025963, + "learning_rate": 1.8123180211936417e-05, + "loss": 0.5051, + "step": 225 + }, + { + "epoch": 2.8928000000000003, + "grad_norm": 0.3117179339183091, + "learning_rate": 1.794503956576152e-05, + "loss": 0.5182, + "step": 226 + }, + { + "epoch": 2.9055999999999997, + "grad_norm": 0.3515290686368801, + "learning_rate": 1.776706354090585e-05, + "loss": 0.5086, + "step": 227 + }, + { + "epoch": 2.9184, + "grad_norm": 0.29412008878166757, + "learning_rate": 1.758926639489354e-05, + "loss": 0.524, + "step": 228 + }, + { + "epoch": 2.9312, + "grad_norm": 0.3298778138199897, + "learning_rate": 1.7411662370918893e-05, + "loss": 0.5144, + "step": 229 + }, + { + "epoch": 2.944, + "grad_norm": 0.2821012239713584, + "learning_rate": 1.7234265696705344e-05, + "loss": 0.5072, + "step": 230 + }, + { + "epoch": 2.9568, + "grad_norm": 0.3012660720135622, + "learning_rate": 1.7057090583365678e-05, + "loss": 0.5145, + "step": 231 + }, + { + "epoch": 2.9696, + "grad_norm": 0.28948238104965945, + "learning_rate": 1.6880151224263646e-05, + "loss": 0.5029, + "step": 232 + }, + { + "epoch": 2.9824, + "grad_norm": 0.32162146380244094, + "learning_rate": 1.6703461793876876e-05, + "loss": 0.5167, + "step": 233 + }, + { + "epoch": 2.9952, + "grad_norm": 0.3034401446317415, + "learning_rate": 1.6527036446661396e-05, + "loss": 0.5004, + "step": 234 + }, + { + "epoch": 3.008, + "grad_norm": 0.41046645936186843, + "learning_rate": 1.635088931591775e-05, + "loss": 0.463, + "step": 235 + }, + { + "epoch": 3.0208, + "grad_norm": 0.39836569087962515, + "learning_rate": 1.6175034512658753e-05, + "loss": 0.4378, + "step": 236 + }, + { + "epoch": 3.0336, + "grad_norm": 0.6901150993329158, + "learning_rate": 1.5999486124479115e-05, + "loss": 0.4371, + "step": 237 + }, + { + "epoch": 3.0464, + "grad_norm": 0.41203511413224175, + "learning_rate": 1.5824258214426833e-05, + "loss": 0.4426, + "step": 238 + }, + { + "epoch": 3.0592, + "grad_norm": 0.5062977186349449, + "learning_rate": 1.5649364819876655e-05, + "loss": 0.4351, + "step": 239 + }, + { + "epoch": 3.072, + "grad_norm": 0.3979325140755632, + "learning_rate": 1.547481995140556e-05, + "loss": 0.4314, + "step": 240 + }, + { + "epoch": 3.0848, + "grad_norm": 0.4498545218503761, + "learning_rate": 1.5300637591670357e-05, + "loss": 0.4401, + "step": 241 + }, + { + "epoch": 3.0976, + "grad_norm": 0.42355565040806803, + "learning_rate": 1.5126831694287564e-05, + "loss": 0.4337, + "step": 242 + }, + { + "epoch": 3.1104, + "grad_norm": 0.44462201972834053, + "learning_rate": 1.4953416182715566e-05, + "loss": 0.4282, + "step": 243 + }, + { + "epoch": 3.1232, + "grad_norm": 0.40849174738433486, + "learning_rate": 1.478040494913926e-05, + "loss": 0.4537, + "step": 244 + }, + { + "epoch": 3.136, + "grad_norm": 0.45302408598349836, + "learning_rate": 1.460781185335713e-05, + "loss": 0.4488, + "step": 245 + }, + { + "epoch": 3.1488, + "grad_norm": 0.38161356592555723, + "learning_rate": 1.443565072167095e-05, + "loss": 0.4334, + "step": 246 + }, + { + "epoch": 3.1616, + "grad_norm": 0.4477548848994946, + "learning_rate": 1.4263935345778202e-05, + "loss": 0.4443, + "step": 247 + }, + { + "epoch": 3.1744, + "grad_norm": 0.4083281344276796, + "learning_rate": 1.409267948166718e-05, + "loss": 0.4315, + "step": 248 + }, + { + "epoch": 3.1872, + "grad_norm": 0.3176643943634673, + "learning_rate": 1.3921896848515064e-05, + "loss": 0.4449, + "step": 249 + }, + { + "epoch": 3.2, + "grad_norm": 0.40201283506559043, + "learning_rate": 1.3751601127588849e-05, + "loss": 0.441, + "step": 250 + }, + { + "epoch": 3.2128, + "grad_norm": 0.3383208102497633, + "learning_rate": 1.3581805961149371e-05, + "loss": 0.436, + "step": 251 + }, + { + "epoch": 3.2256, + "grad_norm": 0.3110589456051368, + "learning_rate": 1.341252495135841e-05, + "loss": 0.4353, + "step": 252 + }, + { + "epoch": 3.2384, + "grad_norm": 0.3272197166941022, + "learning_rate": 1.324377165918906e-05, + "loss": 0.4343, + "step": 253 + }, + { + "epoch": 3.2512, + "grad_norm": 0.31641831793934144, + "learning_rate": 1.3075559603339354e-05, + "loss": 0.4315, + "step": 254 + }, + { + "epoch": 3.2640000000000002, + "grad_norm": 0.32780471432228836, + "learning_rate": 1.2907902259149287e-05, + "loss": 0.4356, + "step": 255 + }, + { + "epoch": 3.2768, + "grad_norm": 0.32626163102304695, + "learning_rate": 1.274081305752135e-05, + "loss": 0.4401, + "step": 256 + }, + { + "epoch": 3.2896, + "grad_norm": 0.3169286387534973, + "learning_rate": 1.2574305383844528e-05, + "loss": 0.4434, + "step": 257 + }, + { + "epoch": 3.3024, + "grad_norm": 0.32721144517777084, + "learning_rate": 1.2408392576922075e-05, + "loss": 0.4414, + "step": 258 + }, + { + "epoch": 3.3152, + "grad_norm": 0.27217470559211066, + "learning_rate": 1.2243087927902905e-05, + "loss": 0.4475, + "step": 259 + }, + { + "epoch": 3.328, + "grad_norm": 0.3359690140623319, + "learning_rate": 1.2078404679216864e-05, + "loss": 0.4246, + "step": 260 + }, + { + "epoch": 3.3407999999999998, + "grad_norm": 0.26838112419162513, + "learning_rate": 1.1914356023513904e-05, + "loss": 0.4303, + "step": 261 + }, + { + "epoch": 3.3536, + "grad_norm": 0.29178007189259314, + "learning_rate": 1.1750955102607193e-05, + "loss": 0.4366, + "step": 262 + }, + { + "epoch": 3.3664, + "grad_norm": 0.2791059524097632, + "learning_rate": 1.1588215006420374e-05, + "loss": 0.4396, + "step": 263 + }, + { + "epoch": 3.3792, + "grad_norm": 0.27798085486262086, + "learning_rate": 1.1426148771938915e-05, + "loss": 0.4372, + "step": 264 + }, + { + "epoch": 3.392, + "grad_norm": 0.29048854928627454, + "learning_rate": 1.1264769382165748e-05, + "loss": 0.4366, + "step": 265 + }, + { + "epoch": 3.4048, + "grad_norm": 0.3096664519908522, + "learning_rate": 1.110408976508118e-05, + "loss": 0.4325, + "step": 266 + }, + { + "epoch": 3.4176, + "grad_norm": 0.3053421610217013, + "learning_rate": 1.094412279260726e-05, + "loss": 0.4227, + "step": 267 + }, + { + "epoch": 3.4304, + "grad_norm": 0.2703969052430842, + "learning_rate": 1.0784881279576635e-05, + "loss": 0.4413, + "step": 268 + }, + { + "epoch": 3.4432, + "grad_norm": 0.31743100870335433, + "learning_rate": 1.0626377982705929e-05, + "loss": 0.4485, + "step": 269 + }, + { + "epoch": 3.456, + "grad_norm": 0.2951598049339725, + "learning_rate": 1.0468625599573842e-05, + "loss": 0.4223, + "step": 270 + }, + { + "epoch": 3.4688, + "grad_norm": 0.3014289743968507, + "learning_rate": 1.0311636767603952e-05, + "loss": 0.441, + "step": 271 + }, + { + "epoch": 3.4816, + "grad_norm": 0.2872576850605509, + "learning_rate": 1.0155424063052306e-05, + "loss": 0.4388, + "step": 272 + }, + { + "epoch": 3.4944, + "grad_norm": 0.3003555194020943, + "learning_rate": 1.0000000000000006e-05, + "loss": 0.4383, + "step": 273 + }, + { + "epoch": 3.5072, + "grad_norm": 0.27188339642329284, + "learning_rate": 9.84537702935065e-06, + "loss": 0.4391, + "step": 274 + }, + { + "epoch": 3.52, + "grad_norm": 0.3133892904576033, + "learning_rate": 9.691567537832964e-06, + "loss": 0.4489, + "step": 275 + }, + { + "epoch": 3.5328, + "grad_norm": 0.27326154899036914, + "learning_rate": 9.538583847008452e-06, + "loss": 0.4188, + "step": 276 + }, + { + "epoch": 3.5456, + "grad_norm": 0.2659259632741142, + "learning_rate": 9.386438212284372e-06, + "loss": 0.423, + "step": 277 + }, + { + "epoch": 3.5584, + "grad_norm": 0.28898963859236165, + "learning_rate": 9.235142821931928e-06, + "loss": 0.431, + "step": 278 + }, + { + "epoch": 3.5712, + "grad_norm": 0.27216817344052996, + "learning_rate": 9.084709796109907e-06, + "loss": 0.4381, + "step": 279 + }, + { + "epoch": 3.584, + "grad_norm": 0.2762213585544952, + "learning_rate": 8.93515118589373e-06, + "loss": 0.4425, + "step": 280 + }, + { + "epoch": 3.5968, + "grad_norm": 0.3098087436899742, + "learning_rate": 8.786478972310023e-06, + "loss": 0.4301, + "step": 281 + }, + { + "epoch": 3.6096, + "grad_norm": 0.2772833601212417, + "learning_rate": 8.638705065376887e-06, + "loss": 0.4289, + "step": 282 + }, + { + "epoch": 3.6224, + "grad_norm": 0.28264765343950765, + "learning_rate": 8.491841303149728e-06, + "loss": 0.4311, + "step": 283 + }, + { + "epoch": 3.6352, + "grad_norm": 0.29573961882101746, + "learning_rate": 8.345899450772975e-06, + "loss": 0.4371, + "step": 284 + }, + { + "epoch": 3.648, + "grad_norm": 0.2507671856832434, + "learning_rate": 8.200891199537549e-06, + "loss": 0.4434, + "step": 285 + }, + { + "epoch": 3.6608, + "grad_norm": 0.25755157366812154, + "learning_rate": 8.056828165944282e-06, + "loss": 0.4314, + "step": 286 + }, + { + "epoch": 3.6736, + "grad_norm": 0.26829055240158844, + "learning_rate": 7.913721890773354e-06, + "loss": 0.4393, + "step": 287 + }, + { + "epoch": 3.6864, + "grad_norm": 0.24653841182739372, + "learning_rate": 7.771583838159756e-06, + "loss": 0.4335, + "step": 288 + }, + { + "epoch": 3.6992000000000003, + "grad_norm": 0.27273851728987, + "learning_rate": 7.630425394674903e-06, + "loss": 0.4346, + "step": 289 + }, + { + "epoch": 3.7119999999999997, + "grad_norm": 0.2542223684948622, + "learning_rate": 7.49025786841445e-06, + "loss": 0.4203, + "step": 290 + }, + { + "epoch": 3.7248, + "grad_norm": 0.26861560889688035, + "learning_rate": 7.3510924880924575e-06, + "loss": 0.4461, + "step": 291 + }, + { + "epoch": 3.7376, + "grad_norm": 0.2534116220118216, + "learning_rate": 7.212940402141808e-06, + "loss": 0.4325, + "step": 292 + }, + { + "epoch": 3.7504, + "grad_norm": 0.27795799974837215, + "learning_rate": 7.075812677821145e-06, + "loss": 0.4248, + "step": 293 + }, + { + "epoch": 3.7632, + "grad_norm": 0.2613087055563045, + "learning_rate": 6.939720300328303e-06, + "loss": 0.455, + "step": 294 + }, + { + "epoch": 3.776, + "grad_norm": 0.27299344851818386, + "learning_rate": 6.8046741719202385e-06, + "loss": 0.4485, + "step": 295 + }, + { + "epoch": 3.7888, + "grad_norm": 0.2781875846770507, + "learning_rate": 6.67068511103971e-06, + "loss": 0.4514, + "step": 296 + }, + { + "epoch": 3.8016, + "grad_norm": 0.2584225371356067, + "learning_rate": 6.537763851448593e-06, + "loss": 0.4467, + "step": 297 + }, + { + "epoch": 3.8144, + "grad_norm": 0.3085510141589761, + "learning_rate": 6.4059210413680175e-06, + "loss": 0.4343, + "step": 298 + }, + { + "epoch": 3.8272, + "grad_norm": 0.2715244321951438, + "learning_rate": 6.275167242625331e-06, + "loss": 0.4271, + "step": 299 + }, + { + "epoch": 3.84, + "grad_norm": 0.2738849657890685, + "learning_rate": 6.145512929808013e-06, + "loss": 0.4414, + "step": 300 + }, + { + "epoch": 3.8528000000000002, + "grad_norm": 0.2509332337895888, + "learning_rate": 6.016968489424572e-06, + "loss": 0.4283, + "step": 301 + }, + { + "epoch": 3.8656, + "grad_norm": 0.2585900899489169, + "learning_rate": 5.889544219072465e-06, + "loss": 0.4249, + "step": 302 + }, + { + "epoch": 3.8784, + "grad_norm": 0.2504734736778933, + "learning_rate": 5.7632503266131925e-06, + "loss": 0.4274, + "step": 303 + }, + { + "epoch": 3.8912, + "grad_norm": 0.2343119676208743, + "learning_rate": 5.638096929354522e-06, + "loss": 0.4265, + "step": 304 + }, + { + "epoch": 3.904, + "grad_norm": 0.23914782192599746, + "learning_rate": 5.514094053240035e-06, + "loss": 0.4274, + "step": 305 + }, + { + "epoch": 3.9168, + "grad_norm": 0.24779763263825993, + "learning_rate": 5.39125163204594e-06, + "loss": 0.4366, + "step": 306 + }, + { + "epoch": 3.9295999999999998, + "grad_norm": 0.2387628080118382, + "learning_rate": 5.269579506585259e-06, + "loss": 0.4445, + "step": 307 + }, + { + "epoch": 3.9424, + "grad_norm": 0.23695884317251484, + "learning_rate": 5.149087423919541e-06, + "loss": 0.4383, + "step": 308 + }, + { + "epoch": 3.9552, + "grad_norm": 0.23140809524176137, + "learning_rate": 5.029785036577976e-06, + "loss": 0.4311, + "step": 309 + }, + { + "epoch": 3.968, + "grad_norm": 0.2260624177391602, + "learning_rate": 4.911681901784198e-06, + "loss": 0.4481, + "step": 310 + }, + { + "epoch": 3.9808, + "grad_norm": 0.2452242981234645, + "learning_rate": 4.794787480690597e-06, + "loss": 0.4391, + "step": 311 + }, + { + "epoch": 3.9936, + "grad_norm": 0.23650964042923645, + "learning_rate": 4.679111137620442e-06, + "loss": 0.4449, + "step": 312 + }, + { + "epoch": 4.0064, + "grad_norm": 0.29962007310679284, + "learning_rate": 4.5646621393177e-06, + "loss": 0.4054, + "step": 313 + }, + { + "epoch": 4.0192, + "grad_norm": 0.3312469740936787, + "learning_rate": 4.451449654204685e-06, + "loss": 0.409, + "step": 314 + }, + { + "epoch": 4.032, + "grad_norm": 0.27286209439834047, + "learning_rate": 4.339482751647557e-06, + "loss": 0.378, + "step": 315 + }, + { + "epoch": 4.0448, + "grad_norm": 0.3231896957399875, + "learning_rate": 4.228770401229824e-06, + "loss": 0.4044, + "step": 316 + }, + { + "epoch": 4.0576, + "grad_norm": 0.4227960209011115, + "learning_rate": 4.119321472033779e-06, + "loss": 0.3691, + "step": 317 + }, + { + "epoch": 4.0704, + "grad_norm": 0.3590703865605139, + "learning_rate": 4.011144731929981e-06, + "loss": 0.3743, + "step": 318 + }, + { + "epoch": 4.0832, + "grad_norm": 0.3276315585715536, + "learning_rate": 3.904248846874894e-06, + "loss": 0.4057, + "step": 319 + }, + { + "epoch": 4.096, + "grad_norm": 0.3204463262223722, + "learning_rate": 3.7986423802166705e-06, + "loss": 0.3841, + "step": 320 + }, + { + "epoch": 4.1088, + "grad_norm": 0.3572447564220915, + "learning_rate": 3.694333792009115e-06, + "loss": 0.391, + "step": 321 + }, + { + "epoch": 4.1216, + "grad_norm": 0.30830859475235545, + "learning_rate": 3.5913314383339937e-06, + "loss": 0.391, + "step": 322 + }, + { + "epoch": 4.1344, + "grad_norm": 0.29639654723525105, + "learning_rate": 3.4896435706316e-06, + "loss": 0.3888, + "step": 323 + }, + { + "epoch": 4.1472, + "grad_norm": 0.2866219313421289, + "learning_rate": 3.3892783350397675e-06, + "loss": 0.3897, + "step": 324 + }, + { + "epoch": 4.16, + "grad_norm": 0.2859626125074166, + "learning_rate": 3.290243771741275e-06, + "loss": 0.3844, + "step": 325 + }, + { + "epoch": 4.1728, + "grad_norm": 0.29987145812710686, + "learning_rate": 3.1925478143197418e-06, + "loss": 0.38, + "step": 326 + }, + { + "epoch": 4.1856, + "grad_norm": 0.27714350402080357, + "learning_rate": 3.0961982891241083e-06, + "loss": 0.3915, + "step": 327 + }, + { + "epoch": 4.1984, + "grad_norm": 0.2816921404300732, + "learning_rate": 3.001202914641628e-06, + "loss": 0.3827, + "step": 328 + }, + { + "epoch": 4.2112, + "grad_norm": 0.2624104774293918, + "learning_rate": 2.907569300879596e-06, + "loss": 0.3802, + "step": 329 + }, + { + "epoch": 4.224, + "grad_norm": 0.26216341440755286, + "learning_rate": 2.815304948755664e-06, + "loss": 0.3811, + "step": 330 + }, + { + "epoch": 4.2368, + "grad_norm": 0.2597148815879336, + "learning_rate": 2.7244172494969978e-06, + "loss": 0.3941, + "step": 331 + }, + { + "epoch": 4.2496, + "grad_norm": 0.2589303491965205, + "learning_rate": 2.6349134840481294e-06, + "loss": 0.3836, + "step": 332 + }, + { + "epoch": 4.2624, + "grad_norm": 0.25039699970590135, + "learning_rate": 2.546800822487714e-06, + "loss": 0.3956, + "step": 333 + }, + { + "epoch": 4.2752, + "grad_norm": 0.22867459998419756, + "learning_rate": 2.4600863234541338e-06, + "loss": 0.3984, + "step": 334 + }, + { + "epoch": 4.288, + "grad_norm": 0.22813975032869171, + "learning_rate": 2.374776933580025e-06, + "loss": 0.3945, + "step": 335 + }, + { + "epoch": 4.3008, + "grad_norm": 0.24726113875422293, + "learning_rate": 2.2908794869358044e-06, + "loss": 0.3804, + "step": 336 + }, + { + "epoch": 4.3136, + "grad_norm": 0.23839697981679817, + "learning_rate": 2.2084007044821764e-06, + "loss": 0.3918, + "step": 337 + }, + { + "epoch": 4.3264, + "grad_norm": 0.24105041292682955, + "learning_rate": 2.127347193531757e-06, + "loss": 0.3926, + "step": 338 + }, + { + "epoch": 4.3392, + "grad_norm": 0.22817096036617446, + "learning_rate": 2.0477254472197237e-06, + "loss": 0.3864, + "step": 339 + }, + { + "epoch": 4.352, + "grad_norm": 0.22470269288312994, + "learning_rate": 1.96954184398368e-06, + "loss": 0.3812, + "step": 340 + }, + { + "epoch": 4.3648, + "grad_norm": 0.22124074384761025, + "learning_rate": 1.8928026470526917e-06, + "loss": 0.3853, + "step": 341 + }, + { + "epoch": 4.3776, + "grad_norm": 0.22290829954818844, + "learning_rate": 1.817514003945524e-06, + "loss": 0.3892, + "step": 342 + }, + { + "epoch": 4.3904, + "grad_norm": 0.24016004520116097, + "learning_rate": 1.743681945978184e-06, + "loss": 0.3746, + "step": 343 + }, + { + "epoch": 4.4032, + "grad_norm": 0.21396083544636266, + "learning_rate": 1.6713123877807413e-06, + "loss": 0.3752, + "step": 344 + }, + { + "epoch": 4.416, + "grad_norm": 0.21462260986745693, + "learning_rate": 1.6004111268235156e-06, + "loss": 0.3938, + "step": 345 + }, + { + "epoch": 4.4288, + "grad_norm": 0.21840179140888716, + "learning_rate": 1.5309838429526714e-06, + "loss": 0.381, + "step": 346 + }, + { + "epoch": 4.4416, + "grad_norm": 0.22612770072690125, + "learning_rate": 1.4630360979351644e-06, + "loss": 0.3767, + "step": 347 + }, + { + "epoch": 4.4544, + "grad_norm": 0.22616358980175943, + "learning_rate": 1.396573335013236e-06, + "loss": 0.3949, + "step": 348 + }, + { + "epoch": 4.4672, + "grad_norm": 0.21834471671265632, + "learning_rate": 1.3316008784683265e-06, + "loss": 0.3952, + "step": 349 + }, + { + "epoch": 4.48, + "grad_norm": 0.22384371553459914, + "learning_rate": 1.2681239331945695e-06, + "loss": 0.4051, + "step": 350 + }, + { + "epoch": 4.4928, + "grad_norm": 0.2145005953417142, + "learning_rate": 1.2061475842818337e-06, + "loss": 0.3991, + "step": 351 + }, + { + "epoch": 4.5056, + "grad_norm": 0.2159902095624403, + "learning_rate": 1.1456767966083393e-06, + "loss": 0.3793, + "step": 352 + }, + { + "epoch": 4.5184, + "grad_norm": 0.21070490480560447, + "learning_rate": 1.086716414442952e-06, + "loss": 0.3937, + "step": 353 + }, + { + "epoch": 4.5312, + "grad_norm": 0.220079307692125, + "learning_rate": 1.0292711610570904e-06, + "loss": 0.391, + "step": 354 + }, + { + "epoch": 4.5440000000000005, + "grad_norm": 0.20512973244246804, + "learning_rate": 9.733456383463658e-07, + "loss": 0.3967, + "step": 355 + }, + { + "epoch": 4.5568, + "grad_norm": 0.21311854141756914, + "learning_rate": 9.189443264619102e-07, + "loss": 0.4129, + "step": 356 + }, + { + "epoch": 4.5696, + "grad_norm": 0.21805552103822698, + "learning_rate": 8.660715834514977e-07, + "loss": 0.3916, + "step": 357 + }, + { + "epoch": 4.5824, + "grad_norm": 0.2080489832732226, + "learning_rate": 8.147316449103959e-07, + "loss": 0.4004, + "step": 358 + }, + { + "epoch": 4.5952, + "grad_norm": 0.20739889265925496, + "learning_rate": 7.649286236420806e-07, + "loss": 0.4153, + "step": 359 + }, + { + "epoch": 4.608, + "grad_norm": 0.21067276573842075, + "learning_rate": 7.166665093287539e-07, + "loss": 0.4006, + "step": 360 + }, + { + "epoch": 4.6208, + "grad_norm": 0.22299398200312037, + "learning_rate": 6.69949168211721e-07, + "loss": 0.3749, + "step": 361 + }, + { + "epoch": 4.6336, + "grad_norm": 0.2037159019231344, + "learning_rate": 6.247803427816945e-07, + "loss": 0.4018, + "step": 362 + }, + { + "epoch": 4.6464, + "grad_norm": 0.21263928604401106, + "learning_rate": 5.811636514789598e-07, + "loss": 0.3799, + "step": 363 + }, + { + "epoch": 4.6592, + "grad_norm": 0.2148226409056421, + "learning_rate": 5.391025884035239e-07, + "loss": 0.3806, + "step": 364 + }, + { + "epoch": 4.672, + "grad_norm": 0.20705283380158873, + "learning_rate": 4.986005230351954e-07, + "loss": 0.3777, + "step": 365 + }, + { + "epoch": 4.6848, + "grad_norm": 0.21576036694240572, + "learning_rate": 4.5966069996365993e-07, + "loss": 0.3986, + "step": 366 + }, + { + "epoch": 4.6975999999999996, + "grad_norm": 0.21027087952143309, + "learning_rate": 4.22286238628562e-07, + "loss": 0.3794, + "step": 367 + }, + { + "epoch": 4.7104, + "grad_norm": 0.21565973905897212, + "learning_rate": 3.8648013306960664e-07, + "loss": 0.3808, + "step": 368 + }, + { + "epoch": 4.7232, + "grad_norm": 0.2027994309063992, + "learning_rate": 3.522452516867048e-07, + "loss": 0.3912, + "step": 369 + }, + { + "epoch": 4.736, + "grad_norm": 0.20333862712713335, + "learning_rate": 3.1958433701019697e-07, + "loss": 0.3711, + "step": 370 + }, + { + "epoch": 4.7488, + "grad_norm": 0.20786917102005878, + "learning_rate": 2.8850000548115155e-07, + "loss": 0.3759, + "step": 371 + }, + { + "epoch": 4.7616, + "grad_norm": 0.21090283998936593, + "learning_rate": 2.5899474724174313e-07, + "loss": 0.3818, + "step": 372 + }, + { + "epoch": 4.7744, + "grad_norm": 0.22006312730244917, + "learning_rate": 2.3107092593579905e-07, + "loss": 0.3909, + "step": 373 + }, + { + "epoch": 4.7872, + "grad_norm": 0.20748609787041353, + "learning_rate": 2.0473077851942858e-07, + "loss": 0.3876, + "step": 374 + }, + { + "epoch": 4.8, + "grad_norm": 0.21091568244756467, + "learning_rate": 1.799764150818306e-07, + "loss": 0.3926, + "step": 375 + }, + { + "epoch": 4.8128, + "grad_norm": 0.20541127946636106, + "learning_rate": 1.5680981867625566e-07, + "loss": 0.3902, + "step": 376 + }, + { + "epoch": 4.8256, + "grad_norm": 0.20811933533876514, + "learning_rate": 1.3523284516113955e-07, + "loss": 0.3818, + "step": 377 + }, + { + "epoch": 4.8384, + "grad_norm": 0.20569700919140488, + "learning_rate": 1.1524722305144231e-07, + "loss": 0.3852, + "step": 378 + }, + { + "epoch": 4.8512, + "grad_norm": 0.2069276613444958, + "learning_rate": 9.685455338016347e-08, + "loss": 0.3988, + "step": 379 + }, + { + "epoch": 4.864, + "grad_norm": 0.20911198227882938, + "learning_rate": 8.005630957010014e-08, + "loss": 0.3851, + "step": 380 + }, + { + "epoch": 4.8768, + "grad_norm": 0.20532543240595294, + "learning_rate": 6.485383731580142e-08, + "loss": 0.3914, + "step": 381 + }, + { + "epoch": 4.8896, + "grad_norm": 0.20732465469266587, + "learning_rate": 5.1248354475768034e-08, + "loss": 0.3845, + "step": 382 + }, + { + "epoch": 4.9024, + "grad_norm": 0.21466224187863875, + "learning_rate": 3.924095097489922e-08, + "loss": 0.3775, + "step": 383 + }, + { + "epoch": 4.9152000000000005, + "grad_norm": 0.20921729995840035, + "learning_rate": 2.8832588717164766e-08, + "loss": 0.3839, + "step": 384 + }, + { + "epoch": 4.928, + "grad_norm": 0.20764848352691964, + "learning_rate": 2.0024101508555604e-08, + "loss": 0.3927, + "step": 385 + }, + { + "epoch": 4.9408, + "grad_norm": 0.2119491015744724, + "learning_rate": 1.281619499029274e-08, + "loss": 0.3819, + "step": 386 + }, + { + "epoch": 4.9536, + "grad_norm": 0.21028962864646566, + "learning_rate": 7.209446582292501e-09, + "loss": 0.3863, + "step": 387 + }, + { + "epoch": 4.9664, + "grad_norm": 0.20568189636278641, + "learning_rate": 3.2043054369057523e-09, + "loss": 0.4027, + "step": 388 + }, + { + "epoch": 4.9792, + "grad_norm": 0.20207762958979356, + "learning_rate": 8.010924029533406e-10, + "loss": 0.3858, + "step": 389 + }, + { + "epoch": 4.992, + "grad_norm": 0.20602286936096575, + "learning_rate": 0.0, + "loss": 0.3772, + "step": 390 + }, + { + "epoch": 4.992, + "step": 390, + "total_flos": 1.896730808245289e+18, + "train_loss": 0.5385960905979841, + "train_runtime": 18314.8306, + "train_samples_per_second": 2.73, + "train_steps_per_second": 0.021 + } + ], + "logging_steps": 1, + "max_steps": 390, + "num_input_tokens_seen": 0, + "num_train_epochs": 5, + "save_steps": 500, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 1.896730808245289e+18, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/training_args.bin b/training_args.bin new file mode 100644 index 0000000..9bea874 --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:26710ead36fcb2110c9a0fc3bd89c90d6b2c75f7cc0acfdef3913a83a346b4d5 +size 7224 diff --git a/training_loss.png b/training_loss.png new file mode 100644 index 0000000..77f35f9 Binary files /dev/null and b/training_loss.png differ diff --git a/vocab.json b/vocab.json new file mode 100644 index 0000000..6c49fc6 --- /dev/null +++ b/vocab.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910 +size 2776833