commit dc501e8c90362f570b0cddc57ceba87e2b11d865 Author: ModelHub XC Date: Tue Sep 15 10:29:12 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: mlfoundations-dev/d1_math_multiple_languages_10k Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..d6420f8 --- /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 + +tokenizer.json filter=lfs diff=lfs merge=lfs -text +vocab.json filter=lfs diff=lfs merge=lfs -text +model-00002-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +training_args.bin filter=lfs diff=lfs merge=lfs -text +merges.txt filter=lfs diff=lfs merge=lfs -text +model-00003-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +model-00004-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +model-00001-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..2475373 --- /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: d1_math_multiple_languages_10k + results: [] +--- + + + +# d1_math_multiple_languages_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/d1_math_multiple_languages_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: 4 +- gradient_accumulation_steps: 32 +- total_train_batch_size: 128 +- total_eval_batch_size: 32 +- optimizer: Use 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 2.6.0+cu124 +- Datasets 3.1.0 +- 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..d2e988b --- /dev/null +++ b/all_results.json @@ -0,0 +1,8 @@ +{ + "epoch": 4.992, + "total_flos": 1.9952099917069025e+18, + "train_loss": 0.36405690262715024, + "train_runtime": 60425.7754, + "train_samples_per_second": 0.827, + "train_steps_per_second": 0.006 +} \ 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..f1f0588 --- /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/d1_math_multiple_languages_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: 32 +hub_model_id: mlfoundations-dev/d1_math_multiple_languages_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: /data/cat/ws/ryma833h-dcft/checkpoints/d1_math_multiple_languages_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..47a4f61 --- /dev/null +++ b/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0c2e5043d4ccdbeca1b2770d91796b58913f5f3e58c7ba4e14d41c8b234b802c +size 4877660776 diff --git a/model-00002-of-00004.safetensors b/model-00002-of-00004.safetensors new file mode 100644 index 0000000..446cd64 --- /dev/null +++ b/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0181f33cac45e8dd801c05ae112599e7542e5cad98faaac1509c19cf432e4832 +size 4932751008 diff --git a/model-00003-of-00004.safetensors b/model-00003-of-00004.safetensors new file mode 100644 index 0000000..c415223 --- /dev/null +++ b/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:70319c45d39a500a7b4181788e0294a54fa9d2a5d48a95d5d885b396ac6eeadf +size 4330865200 diff --git a/model-00004-of-00004.safetensors b/model-00004-of-00004.safetensors new file mode 100644 index 0000000..259222e --- /dev/null +++ b/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:481fe207993934cef59b187e6512ffad2e2a38a6a044a6db0b31a1983dede03f +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..d2e988b --- /dev/null +++ b/train_results.json @@ -0,0 +1,8 @@ +{ + "epoch": 4.992, + "total_flos": 1.9952099917069025e+18, + "train_loss": 0.36405690262715024, + "train_runtime": 60425.7754, + "train_samples_per_second": 0.827, + "train_steps_per_second": 0.006 +} \ No newline at end of file diff --git a/trainer_log.jsonl b/trainer_log.jsonl new file mode 100644 index 0000000..92bdaba --- /dev/null +++ b/trainer_log.jsonl @@ -0,0 +1,391 @@ +{"current_steps": 1, "total_steps": 390, "loss": 0.8325, "lr": 1.0256410256410257e-06, "epoch": 0.0128, "percentage": 0.26, "elapsed_time": "0:02:34", "remaining_time": "16:43:19"} +{"current_steps": 2, "total_steps": 390, "loss": 0.8289, "lr": 2.0512820512820513e-06, "epoch": 0.0256, "percentage": 0.51, "elapsed_time": "0:05:13", "remaining_time": "16:54:48"} +{"current_steps": 3, "total_steps": 390, "loss": 0.8078, "lr": 3.0769230769230774e-06, "epoch": 0.0384, "percentage": 0.77, "elapsed_time": "0:07:57", "remaining_time": "17:06:48"} +{"current_steps": 4, "total_steps": 390, "loss": 0.8006, "lr": 4.102564102564103e-06, "epoch": 0.0512, "percentage": 1.03, "elapsed_time": "0:10:26", "remaining_time": "16:47:05"} +{"current_steps": 5, "total_steps": 390, "loss": 0.7421, "lr": 5.128205128205128e-06, "epoch": 0.064, "percentage": 1.28, "elapsed_time": "0:12:44", "remaining_time": "16:20:59"} +{"current_steps": 6, "total_steps": 390, "loss": 0.7318, "lr": 6.153846153846155e-06, "epoch": 0.0768, "percentage": 1.54, "elapsed_time": "0:15:17", "remaining_time": "16:19:05"} +{"current_steps": 7, "total_steps": 390, "loss": 0.7258, "lr": 7.17948717948718e-06, "epoch": 0.0896, "percentage": 1.79, "elapsed_time": "0:17:46", "remaining_time": "16:12:45"} +{"current_steps": 8, "total_steps": 390, "loss": 0.7241, "lr": 8.205128205128205e-06, "epoch": 0.1024, "percentage": 2.05, "elapsed_time": "0:20:24", "remaining_time": "16:14:31"} +{"current_steps": 9, "total_steps": 390, "loss": 0.7218, "lr": 9.230769230769232e-06, "epoch": 0.1152, "percentage": 2.31, "elapsed_time": "0:22:52", "remaining_time": "16:08:13"} +{"current_steps": 10, "total_steps": 390, "loss": 0.6697, "lr": 1.0256410256410256e-05, "epoch": 0.128, "percentage": 2.56, "elapsed_time": "0:25:34", "remaining_time": "16:12:03"} +{"current_steps": 11, "total_steps": 390, "loss": 0.6733, "lr": 1.1282051282051283e-05, "epoch": 0.1408, "percentage": 2.82, "elapsed_time": "0:28:13", "remaining_time": "16:12:42"} +{"current_steps": 12, "total_steps": 390, "loss": 0.6395, "lr": 1.230769230769231e-05, "epoch": 0.1536, "percentage": 3.08, "elapsed_time": "0:30:49", "remaining_time": "16:11:04"} +{"current_steps": 13, "total_steps": 390, "loss": 0.6393, "lr": 1.3333333333333333e-05, "epoch": 0.1664, "percentage": 3.33, "elapsed_time": "0:33:33", "remaining_time": "16:13:13"} +{"current_steps": 14, "total_steps": 390, "loss": 0.6228, "lr": 1.435897435897436e-05, "epoch": 0.1792, "percentage": 3.59, "elapsed_time": "0:36:14", "remaining_time": "16:13:15"} +{"current_steps": 15, "total_steps": 390, "loss": 0.5948, "lr": 1.5384615384615387e-05, "epoch": 0.192, "percentage": 3.85, "elapsed_time": "0:38:47", "remaining_time": "16:09:52"} +{"current_steps": 16, "total_steps": 390, "loss": 0.5971, "lr": 1.641025641025641e-05, "epoch": 0.2048, "percentage": 4.1, "elapsed_time": "0:41:15", "remaining_time": "16:04:22"} +{"current_steps": 17, "total_steps": 390, "loss": 0.5931, "lr": 1.7435897435897438e-05, "epoch": 0.2176, "percentage": 4.36, "elapsed_time": "0:43:46", "remaining_time": "16:00:38"} +{"current_steps": 18, "total_steps": 390, "loss": 0.5712, "lr": 1.8461538461538465e-05, "epoch": 0.2304, "percentage": 4.62, "elapsed_time": "0:46:18", "remaining_time": "15:57:07"} +{"current_steps": 19, "total_steps": 390, "loss": 0.5674, "lr": 1.9487179487179488e-05, "epoch": 0.2432, "percentage": 4.87, "elapsed_time": "0:49:00", "remaining_time": "15:56:56"} +{"current_steps": 20, "total_steps": 390, "loss": 0.5393, "lr": 2.0512820512820512e-05, "epoch": 0.256, "percentage": 5.13, "elapsed_time": "0:51:27", "remaining_time": "15:51:51"} +{"current_steps": 21, "total_steps": 390, "loss": 0.551, "lr": 2.153846153846154e-05, "epoch": 0.2688, "percentage": 5.38, "elapsed_time": "0:53:57", "remaining_time": "15:48:13"} +{"current_steps": 22, "total_steps": 390, "loss": 0.5431, "lr": 2.2564102564102566e-05, "epoch": 0.2816, "percentage": 5.64, "elapsed_time": "0:56:28", "remaining_time": "15:44:36"} +{"current_steps": 23, "total_steps": 390, "loss": 0.531, "lr": 2.3589743589743593e-05, "epoch": 0.2944, "percentage": 5.9, "elapsed_time": "0:59:00", "remaining_time": "15:41:28"} +{"current_steps": 24, "total_steps": 390, "loss": 0.537, "lr": 2.461538461538462e-05, "epoch": 0.3072, "percentage": 6.15, "elapsed_time": "1:01:35", "remaining_time": "15:39:13"} +{"current_steps": 25, "total_steps": 390, "loss": 0.5243, "lr": 2.5641025641025646e-05, "epoch": 0.32, "percentage": 6.41, "elapsed_time": "1:04:04", "remaining_time": "15:35:30"} +{"current_steps": 26, "total_steps": 390, "loss": 0.5256, "lr": 2.6666666666666667e-05, "epoch": 0.3328, "percentage": 6.67, "elapsed_time": "1:06:32", "remaining_time": "15:31:37"} +{"current_steps": 27, "total_steps": 390, "loss": 0.527, "lr": 2.7692307692307694e-05, "epoch": 0.3456, "percentage": 6.92, "elapsed_time": "1:09:16", "remaining_time": "15:31:17"} +{"current_steps": 28, "total_steps": 390, "loss": 0.5243, "lr": 2.871794871794872e-05, "epoch": 0.3584, "percentage": 7.18, "elapsed_time": "1:11:36", "remaining_time": "15:25:42"} +{"current_steps": 29, "total_steps": 390, "loss": 0.5213, "lr": 2.9743589743589747e-05, "epoch": 0.3712, "percentage": 7.44, "elapsed_time": "1:14:10", "remaining_time": "15:23:22"} +{"current_steps": 30, "total_steps": 390, "loss": 0.5113, "lr": 3.0769230769230774e-05, "epoch": 0.384, "percentage": 7.69, "elapsed_time": "1:16:30", "remaining_time": "15:18:04"} +{"current_steps": 31, "total_steps": 390, "loss": 0.5088, "lr": 3.1794871794871795e-05, "epoch": 0.3968, "percentage": 7.95, "elapsed_time": "1:19:02", "remaining_time": "15:15:21"} +{"current_steps": 32, "total_steps": 390, "loss": 0.5194, "lr": 3.282051282051282e-05, "epoch": 0.4096, "percentage": 8.21, "elapsed_time": "1:21:23", "remaining_time": "15:10:37"} +{"current_steps": 33, "total_steps": 390, "loss": 0.5193, "lr": 3.384615384615385e-05, "epoch": 0.4224, "percentage": 8.46, "elapsed_time": "1:24:14", "remaining_time": "15:11:21"} +{"current_steps": 34, "total_steps": 390, "loss": 0.5039, "lr": 3.4871794871794875e-05, "epoch": 0.4352, "percentage": 8.72, "elapsed_time": "1:26:50", "remaining_time": "15:09:15"} +{"current_steps": 35, "total_steps": 390, "loss": 0.5234, "lr": 3.58974358974359e-05, "epoch": 0.448, "percentage": 8.97, "elapsed_time": "1:29:14", "remaining_time": "15:05:07"} +{"current_steps": 36, "total_steps": 390, "loss": 0.5138, "lr": 3.692307692307693e-05, "epoch": 0.4608, "percentage": 9.23, "elapsed_time": "1:31:47", "remaining_time": "15:02:40"} +{"current_steps": 37, "total_steps": 390, "loss": 0.5164, "lr": 3.794871794871795e-05, "epoch": 0.4736, "percentage": 9.49, "elapsed_time": "1:34:21", "remaining_time": "15:00:15"} +{"current_steps": 38, "total_steps": 390, "loss": 0.5086, "lr": 3.8974358974358976e-05, "epoch": 0.4864, "percentage": 9.74, "elapsed_time": "1:36:53", "remaining_time": "14:57:27"} +{"current_steps": 39, "total_steps": 390, "loss": 0.5108, "lr": 4e-05, "epoch": 0.4992, "percentage": 10.0, "elapsed_time": "1:39:44", "remaining_time": "14:57:36"} +{"current_steps": 40, "total_steps": 390, "loss": 0.4799, "lr": 3.9999198907597046e-05, "epoch": 0.512, "percentage": 10.26, "elapsed_time": "1:42:15", "remaining_time": "14:54:43"} +{"current_steps": 41, "total_steps": 390, "loss": 0.5216, "lr": 3.9996795694563096e-05, "epoch": 0.5248, "percentage": 10.51, "elapsed_time": "1:44:32", "remaining_time": "14:49:55"} +{"current_steps": 42, "total_steps": 390, "loss": 0.4935, "lr": 3.999279055341771e-05, "epoch": 0.5376, "percentage": 10.77, "elapsed_time": "1:47:15", "remaining_time": "14:48:42"} +{"current_steps": 43, "total_steps": 390, "loss": 0.5215, "lr": 3.998718380500971e-05, "epoch": 0.5504, "percentage": 11.03, "elapsed_time": "1:50:01", "remaining_time": "14:47:50"} +{"current_steps": 44, "total_steps": 390, "loss": 0.5134, "lr": 3.997997589849145e-05, "epoch": 0.5632, "percentage": 11.28, "elapsed_time": "1:52:37", "remaining_time": "14:45:37"} +{"current_steps": 45, "total_steps": 390, "loss": 0.4923, "lr": 3.9971167411282835e-05, "epoch": 0.576, "percentage": 11.54, "elapsed_time": "1:55:04", "remaining_time": "14:42:12"} +{"current_steps": 46, "total_steps": 390, "loss": 0.4989, "lr": 3.99607590490251e-05, "epoch": 0.5888, "percentage": 11.79, "elapsed_time": "1:57:28", "remaining_time": "14:38:30"} +{"current_steps": 47, "total_steps": 390, "loss": 0.4942, "lr": 3.9948751645524235e-05, "epoch": 0.6016, "percentage": 12.05, "elapsed_time": "2:00:01", "remaining_time": "14:35:58"} +{"current_steps": 48, "total_steps": 390, "loss": 0.4992, "lr": 3.9935146162684206e-05, "epoch": 0.6144, "percentage": 12.31, "elapsed_time": "2:02:38", "remaining_time": "14:33:51"} +{"current_steps": 49, "total_steps": 390, "loss": 0.5025, "lr": 3.9919943690429906e-05, "epoch": 0.6272, "percentage": 12.56, "elapsed_time": "2:05:24", "remaining_time": "14:32:44"} +{"current_steps": 50, "total_steps": 390, "loss": 0.4828, "lr": 3.9903145446619837e-05, "epoch": 0.64, "percentage": 12.82, "elapsed_time": "2:07:54", "remaining_time": "14:29:44"} +{"current_steps": 51, "total_steps": 390, "loss": 0.4791, "lr": 3.9884752776948564e-05, "epoch": 0.6528, "percentage": 13.08, "elapsed_time": "2:10:31", "remaining_time": "14:27:35"} +{"current_steps": 52, "total_steps": 390, "loss": 0.4788, "lr": 3.9864767154838864e-05, "epoch": 0.6656, "percentage": 13.33, "elapsed_time": "2:13:15", "remaining_time": "14:26:09"} +{"current_steps": 53, "total_steps": 390, "loss": 0.4821, "lr": 3.9843190181323744e-05, "epoch": 0.6784, "percentage": 13.59, "elapsed_time": "2:15:42", "remaining_time": "14:22:52"} +{"current_steps": 54, "total_steps": 390, "loss": 0.4838, "lr": 3.982002358491817e-05, "epoch": 0.6912, "percentage": 13.85, "elapsed_time": "2:18:25", "remaining_time": "14:21:20"} +{"current_steps": 55, "total_steps": 390, "loss": 0.4977, "lr": 3.979526922148058e-05, "epoch": 0.704, "percentage": 14.1, "elapsed_time": "2:21:12", "remaining_time": "14:20:06"} +{"current_steps": 56, "total_steps": 390, "loss": 0.476, "lr": 3.9768929074064206e-05, "epoch": 0.7168, "percentage": 14.36, "elapsed_time": "2:23:44", "remaining_time": "14:17:21"} +{"current_steps": 57, "total_steps": 390, "loss": 0.4851, "lr": 3.9741005252758255e-05, "epoch": 0.7296, "percentage": 14.62, "elapsed_time": "2:26:21", "remaining_time": "14:15:00"} +{"current_steps": 58, "total_steps": 390, "loss": 0.4758, "lr": 3.971149999451886e-05, "epoch": 0.7424, "percentage": 14.87, "elapsed_time": "2:29:02", "remaining_time": "14:13:05"} +{"current_steps": 59, "total_steps": 390, "loss": 0.4708, "lr": 3.9680415662989806e-05, "epoch": 0.7552, "percentage": 15.13, "elapsed_time": "2:31:36", "remaining_time": "14:10:32"} +{"current_steps": 60, "total_steps": 390, "loss": 0.4745, "lr": 3.9647754748313294e-05, "epoch": 0.768, "percentage": 15.38, "elapsed_time": "2:34:16", "remaining_time": "14:08:30"} +{"current_steps": 61, "total_steps": 390, "loss": 0.4855, "lr": 3.96135198669304e-05, "epoch": 0.7808, "percentage": 15.64, "elapsed_time": "2:36:54", "remaining_time": "14:06:16"} +{"current_steps": 62, "total_steps": 390, "loss": 0.4616, "lr": 3.957771376137144e-05, "epoch": 0.7936, "percentage": 15.9, "elapsed_time": "2:39:20", "remaining_time": "14:02:58"} +{"current_steps": 63, "total_steps": 390, "loss": 0.4751, "lr": 3.954033930003634e-05, "epoch": 0.8064, "percentage": 16.15, "elapsed_time": "2:41:48", "remaining_time": "13:59:52"} +{"current_steps": 64, "total_steps": 390, "loss": 0.4767, "lr": 3.9501399476964806e-05, "epoch": 0.8192, "percentage": 16.41, "elapsed_time": "2:44:21", "remaining_time": "13:57:11"} +{"current_steps": 65, "total_steps": 390, "loss": 0.474, "lr": 3.946089741159648e-05, "epoch": 0.832, "percentage": 16.67, "elapsed_time": "2:46:58", "remaining_time": "13:54:54"} +{"current_steps": 66, "total_steps": 390, "loss": 0.4749, "lr": 3.9418836348521045e-05, "epoch": 0.8448, "percentage": 16.92, "elapsed_time": "2:49:36", "remaining_time": "13:52:39"} +{"current_steps": 67, "total_steps": 390, "loss": 0.4748, "lr": 3.937521965721831e-05, "epoch": 0.8576, "percentage": 17.18, "elapsed_time": "2:52:20", "remaining_time": "13:50:49"} +{"current_steps": 68, "total_steps": 390, "loss": 0.4673, "lr": 3.933005083178828e-05, "epoch": 0.8704, "percentage": 17.44, "elapsed_time": "2:54:51", "remaining_time": "13:48:01"} +{"current_steps": 69, "total_steps": 390, "loss": 0.4767, "lr": 3.928333349067125e-05, "epoch": 0.8832, "percentage": 17.69, "elapsed_time": "2:57:17", "remaining_time": "13:44:49"} +{"current_steps": 70, "total_steps": 390, "loss": 0.4679, "lr": 3.923507137635792e-05, "epoch": 0.896, "percentage": 17.95, "elapsed_time": "2:59:47", "remaining_time": "13:41:53"} +{"current_steps": 71, "total_steps": 390, "loss": 0.465, "lr": 3.9185268355089606e-05, "epoch": 0.9088, "percentage": 18.21, "elapsed_time": "3:02:20", "remaining_time": "13:39:16"} +{"current_steps": 72, "total_steps": 390, "loss": 0.4735, "lr": 3.913392841654851e-05, "epoch": 0.9216, "percentage": 18.46, "elapsed_time": "3:05:00", "remaining_time": "13:37:08"} +{"current_steps": 73, "total_steps": 390, "loss": 0.4697, "lr": 3.9081055673538093e-05, "epoch": 0.9344, "percentage": 18.72, "elapsed_time": "3:07:40", "remaining_time": "13:34:58"} +{"current_steps": 74, "total_steps": 390, "loss": 0.4752, "lr": 3.902665436165364e-05, "epoch": 0.9472, "percentage": 18.97, "elapsed_time": "3:10:05", "remaining_time": "13:31:43"} +{"current_steps": 75, "total_steps": 390, "loss": 0.4484, "lr": 3.897072883894291e-05, "epoch": 0.96, "percentage": 19.23, "elapsed_time": "3:12:34", "remaining_time": "13:28:50"} +{"current_steps": 76, "total_steps": 390, "loss": 0.4584, "lr": 3.8913283585557054e-05, "epoch": 0.9728, "percentage": 19.49, "elapsed_time": "3:15:08", "remaining_time": "13:26:13"} +{"current_steps": 77, "total_steps": 390, "loss": 0.4695, "lr": 3.885432320339167e-05, "epoch": 0.9856, "percentage": 19.74, "elapsed_time": "3:17:39", "remaining_time": "13:23:26"} +{"current_steps": 78, "total_steps": 390, "loss": 0.4659, "lr": 3.879385241571817e-05, "epoch": 0.9984, "percentage": 20.0, "elapsed_time": "3:20:08", "remaining_time": "13:20:34"} +{"current_steps": 79, "total_steps": 390, "loss": 0.4383, "lr": 3.873187606680543e-05, "epoch": 1.0112, "percentage": 20.26, "elapsed_time": "3:23:26", "remaining_time": "13:20:54"} +{"current_steps": 80, "total_steps": 390, "loss": 0.4235, "lr": 3.866839912153168e-05, "epoch": 1.024, "percentage": 20.51, "elapsed_time": "3:26:07", "remaining_time": "13:18:42"} +{"current_steps": 81, "total_steps": 390, "loss": 0.426, "lr": 3.860342666498677e-05, "epoch": 1.0368, "percentage": 20.77, "elapsed_time": "3:28:51", "remaining_time": "13:16:45"} +{"current_steps": 82, "total_steps": 390, "loss": 0.4293, "lr": 3.853696390206484e-05, "epoch": 1.0496, "percentage": 21.03, "elapsed_time": "3:31:20", "remaining_time": "13:13:48"} +{"current_steps": 83, "total_steps": 390, "loss": 0.4277, "lr": 3.846901615704734e-05, "epoch": 1.0624, "percentage": 21.28, "elapsed_time": "3:33:58", "remaining_time": "13:11:26"} +{"current_steps": 84, "total_steps": 390, "loss": 0.4136, "lr": 3.839958887317649e-05, "epoch": 1.0752, "percentage": 21.54, "elapsed_time": "3:36:35", "remaining_time": "13:09:01"} +{"current_steps": 85, "total_steps": 390, "loss": 0.4203, "lr": 3.832868761221926e-05, "epoch": 1.088, "percentage": 21.79, "elapsed_time": "3:39:02", "remaining_time": "13:05:57"} +{"current_steps": 86, "total_steps": 390, "loss": 0.4133, "lr": 3.825631805402182e-05, "epoch": 1.1008, "percentage": 22.05, "elapsed_time": "3:41:41", "remaining_time": "13:03:40"} +{"current_steps": 87, "total_steps": 390, "loss": 0.4189, "lr": 3.818248599605448e-05, "epoch": 1.1136, "percentage": 22.31, "elapsed_time": "3:44:11", "remaining_time": "13:00:47"} +{"current_steps": 88, "total_steps": 390, "loss": 0.4319, "lr": 3.810719735294731e-05, "epoch": 1.1264, "percentage": 22.56, "elapsed_time": "3:46:44", "remaining_time": "12:58:07"} +{"current_steps": 89, "total_steps": 390, "loss": 0.4138, "lr": 3.8030458156016326e-05, "epoch": 1.1392, "percentage": 22.82, "elapsed_time": "3:49:12", "remaining_time": "12:55:11"} +{"current_steps": 90, "total_steps": 390, "loss": 0.4087, "lr": 3.795227455278029e-05, "epoch": 1.152, "percentage": 23.08, "elapsed_time": "3:51:52", "remaining_time": "12:52:54"} +{"current_steps": 91, "total_steps": 390, "loss": 0.4158, "lr": 3.787265280646825e-05, "epoch": 1.1648, "percentage": 23.33, "elapsed_time": "3:54:30", "remaining_time": "12:50:31"} +{"current_steps": 92, "total_steps": 390, "loss": 0.4163, "lr": 3.7791599295517825e-05, "epoch": 1.1776, "percentage": 23.59, "elapsed_time": "3:57:15", "remaining_time": "12:48:30"} +{"current_steps": 93, "total_steps": 390, "loss": 0.4166, "lr": 3.7709120513064196e-05, "epoch": 1.1904, "percentage": 23.85, "elapsed_time": "3:59:53", "remaining_time": "12:46:06"} +{"current_steps": 94, "total_steps": 390, "loss": 0.4103, "lr": 3.762522306641998e-05, "epoch": 1.2032, "percentage": 24.1, "elapsed_time": "4:02:33", "remaining_time": "12:43:46"} +{"current_steps": 95, "total_steps": 390, "loss": 0.3998, "lr": 3.7539913676545874e-05, "epoch": 1.216, "percentage": 24.36, "elapsed_time": "4:05:06", "remaining_time": "12:41:06"} +{"current_steps": 96, "total_steps": 390, "loss": 0.4097, "lr": 3.745319917751229e-05, "epoch": 1.2288000000000001, "percentage": 24.62, "elapsed_time": "4:07:32", "remaining_time": "12:38:04"} +{"current_steps": 97, "total_steps": 390, "loss": 0.4228, "lr": 3.736508651595188e-05, "epoch": 1.2416, "percentage": 24.87, "elapsed_time": "4:10:13", "remaining_time": "12:35:51"} +{"current_steps": 98, "total_steps": 390, "loss": 0.4124, "lr": 3.727558275050301e-05, "epoch": 1.2544, "percentage": 25.13, "elapsed_time": "4:12:53", "remaining_time": "12:33:29"} +{"current_steps": 99, "total_steps": 390, "loss": 0.404, "lr": 3.718469505124434e-05, "epoch": 1.2671999999999999, "percentage": 25.38, "elapsed_time": "4:15:16", "remaining_time": "12:30:21"} +{"current_steps": 100, "total_steps": 390, "loss": 0.4187, "lr": 3.709243069912041e-05, "epoch": 1.28, "percentage": 25.64, "elapsed_time": "4:18:03", "remaining_time": "12:28:22"} +{"current_steps": 101, "total_steps": 390, "loss": 0.4139, "lr": 3.699879708535838e-05, "epoch": 1.2928, "percentage": 25.9, "elapsed_time": "4:20:25", "remaining_time": "12:25:10"} +{"current_steps": 102, "total_steps": 390, "loss": 0.4159, "lr": 3.69038017108759e-05, "epoch": 1.3056, "percentage": 26.15, "elapsed_time": "4:23:16", "remaining_time": "12:23:20"} +{"current_steps": 103, "total_steps": 390, "loss": 0.4181, "lr": 3.680745218568026e-05, "epoch": 1.3184, "percentage": 26.41, "elapsed_time": "4:25:48", "remaining_time": "12:20:38"} +{"current_steps": 104, "total_steps": 390, "loss": 0.4046, "lr": 3.6709756228258735e-05, "epoch": 1.3312, "percentage": 26.67, "elapsed_time": "4:28:22", "remaining_time": "12:18:01"} +{"current_steps": 105, "total_steps": 390, "loss": 0.4263, "lr": 3.6610721664960236e-05, "epoch": 1.3439999999999999, "percentage": 26.92, "elapsed_time": "4:31:06", "remaining_time": "12:15:51"} +{"current_steps": 106, "total_steps": 390, "loss": 0.417, "lr": 3.65103564293684e-05, "epoch": 1.3568, "percentage": 27.18, "elapsed_time": "4:33:44", "remaining_time": "12:13:24"} +{"current_steps": 107, "total_steps": 390, "loss": 0.426, "lr": 3.640866856166601e-05, "epoch": 1.3696, "percentage": 27.44, "elapsed_time": "4:36:17", "remaining_time": "12:10:45"} +{"current_steps": 108, "total_steps": 390, "loss": 0.4097, "lr": 3.6305666207990886e-05, "epoch": 1.3824, "percentage": 27.69, "elapsed_time": "4:38:44", "remaining_time": "12:07:50"} +{"current_steps": 109, "total_steps": 390, "loss": 0.4068, "lr": 3.6201357619783336e-05, "epoch": 1.3952, "percentage": 27.95, "elapsed_time": "4:41:08", "remaining_time": "12:04:47"} +{"current_steps": 110, "total_steps": 390, "loss": 0.415, "lr": 3.609575115312511e-05, "epoch": 1.408, "percentage": 28.21, "elapsed_time": "4:43:48", "remaining_time": "12:02:26"} +{"current_steps": 111, "total_steps": 390, "loss": 0.4101, "lr": 3.598885526807003e-05, "epoch": 1.4208, "percentage": 28.46, "elapsed_time": "4:46:31", "remaining_time": "12:00:12"} +{"current_steps": 112, "total_steps": 390, "loss": 0.4213, "lr": 3.5880678527966224e-05, "epoch": 1.4336, "percentage": 28.72, "elapsed_time": "4:49:11", "remaining_time": "11:57:49"} +{"current_steps": 113, "total_steps": 390, "loss": 0.3989, "lr": 3.577122959877017e-05, "epoch": 1.4464000000000001, "percentage": 28.97, "elapsed_time": "4:51:50", "remaining_time": "11:55:23"} +{"current_steps": 114, "total_steps": 390, "loss": 0.416, "lr": 3.566051724835245e-05, "epoch": 1.4592, "percentage": 29.23, "elapsed_time": "4:54:07", "remaining_time": "11:52:04"} +{"current_steps": 115, "total_steps": 390, "loss": 0.4225, "lr": 3.554855034579532e-05, "epoch": 1.472, "percentage": 29.49, "elapsed_time": "4:56:49", "remaining_time": "11:49:47"} +{"current_steps": 116, "total_steps": 390, "loss": 0.4153, "lr": 3.5435337860682304e-05, "epoch": 1.4848, "percentage": 29.74, "elapsed_time": "4:59:28", "remaining_time": "11:47:23"} +{"current_steps": 117, "total_steps": 390, "loss": 0.4108, "lr": 3.532088886237956e-05, "epoch": 1.4976, "percentage": 30.0, "elapsed_time": "5:02:01", "remaining_time": "11:44:44"} +{"current_steps": 118, "total_steps": 390, "loss": 0.4124, "lr": 3.520521251930941e-05, "epoch": 1.5104, "percentage": 30.26, "elapsed_time": "5:04:30", "remaining_time": "11:41:54"} +{"current_steps": 119, "total_steps": 390, "loss": 0.4081, "lr": 3.5088318098215805e-05, "epoch": 1.5232, "percentage": 30.51, "elapsed_time": "5:07:09", "remaining_time": "11:39:29"} +{"current_steps": 120, "total_steps": 390, "loss": 0.4018, "lr": 3.497021496342203e-05, "epoch": 1.536, "percentage": 30.77, "elapsed_time": "5:09:43", "remaining_time": "11:36:53"} +{"current_steps": 121, "total_steps": 390, "loss": 0.3943, "lr": 3.485091257608047e-05, "epoch": 1.5488, "percentage": 31.03, "elapsed_time": "5:12:09", "remaining_time": "11:33:58"} +{"current_steps": 122, "total_steps": 390, "loss": 0.4001, "lr": 3.473042049341474e-05, "epoch": 1.5615999999999999, "percentage": 31.28, "elapsed_time": "5:14:33", "remaining_time": "11:30:59"} +{"current_steps": 123, "total_steps": 390, "loss": 0.4002, "lr": 3.4608748367954064e-05, "epoch": 1.5744, "percentage": 31.54, "elapsed_time": "5:17:07", "remaining_time": "11:28:22"} +{"current_steps": 124, "total_steps": 390, "loss": 0.4078, "lr": 3.4485905946759965e-05, "epoch": 1.5872000000000002, "percentage": 31.79, "elapsed_time": "5:19:28", "remaining_time": "11:25:19"} +{"current_steps": 125, "total_steps": 390, "loss": 0.4025, "lr": 3.4361903070645484e-05, "epoch": 1.6, "percentage": 32.05, "elapsed_time": "5:22:09", "remaining_time": "11:22:58"} +{"current_steps": 126, "total_steps": 390, "loss": 0.4062, "lr": 3.423674967338681e-05, "epoch": 1.6128, "percentage": 32.31, "elapsed_time": "5:24:44", "remaining_time": "11:20:23"} +{"current_steps": 127, "total_steps": 390, "loss": 0.4177, "lr": 3.411045578092754e-05, "epoch": 1.6256, "percentage": 32.56, "elapsed_time": "5:27:11", "remaining_time": "11:17:33"} +{"current_steps": 128, "total_steps": 390, "loss": 0.4166, "lr": 3.398303151057543e-05, "epoch": 1.6383999999999999, "percentage": 32.82, "elapsed_time": "5:29:50", "remaining_time": "11:15:08"} +{"current_steps": 129, "total_steps": 390, "loss": 0.4095, "lr": 3.385448707019199e-05, "epoch": 1.6512, "percentage": 33.08, "elapsed_time": "5:32:15", "remaining_time": "11:12:14"} +{"current_steps": 130, "total_steps": 390, "loss": 0.4088, "lr": 3.372483275737468e-05, "epoch": 1.6640000000000001, "percentage": 33.33, "elapsed_time": "5:34:54", "remaining_time": "11:09:49"} +{"current_steps": 131, "total_steps": 390, "loss": 0.4203, "lr": 3.359407895863199e-05, "epoch": 1.6768, "percentage": 33.59, "elapsed_time": "5:37:18", "remaining_time": "11:06:54"} +{"current_steps": 132, "total_steps": 390, "loss": 0.4049, "lr": 3.34622361485514e-05, "epoch": 1.6896, "percentage": 33.85, "elapsed_time": "5:39:56", "remaining_time": "11:04:25"} +{"current_steps": 133, "total_steps": 390, "loss": 0.4093, "lr": 3.332931488896029e-05, "epoch": 1.7024, "percentage": 34.1, "elapsed_time": "5:42:26", "remaining_time": "11:01:42"} +{"current_steps": 134, "total_steps": 390, "loss": 0.4161, "lr": 3.319532582807977e-05, "epoch": 1.7151999999999998, "percentage": 34.36, "elapsed_time": "5:45:11", "remaining_time": "10:59:27"} +{"current_steps": 135, "total_steps": 390, "loss": 0.4142, "lr": 3.30602796996717e-05, "epoch": 1.728, "percentage": 34.62, "elapsed_time": "5:47:44", "remaining_time": "10:56:50"} +{"current_steps": 136, "total_steps": 390, "loss": 0.4162, "lr": 3.2924187322178865e-05, "epoch": 1.7408000000000001, "percentage": 34.87, "elapsed_time": "5:50:21", "remaining_time": "10:54:20"} +{"current_steps": 137, "total_steps": 390, "loss": 0.4251, "lr": 3.278705959785821e-05, "epoch": 1.7536, "percentage": 35.13, "elapsed_time": "5:52:53", "remaining_time": "10:51:41"} +{"current_steps": 138, "total_steps": 390, "loss": 0.4018, "lr": 3.2648907511907544e-05, "epoch": 1.7664, "percentage": 35.38, "elapsed_time": "5:55:19", "remaining_time": "10:48:51"} +{"current_steps": 139, "total_steps": 390, "loss": 0.4096, "lr": 3.250974213158555e-05, "epoch": 1.7792, "percentage": 35.64, "elapsed_time": "5:57:49", "remaining_time": "10:46:08"} +{"current_steps": 140, "total_steps": 390, "loss": 0.4142, "lr": 3.23695746053251e-05, "epoch": 1.792, "percentage": 35.9, "elapsed_time": "6:00:24", "remaining_time": "10:43:34"} +{"current_steps": 141, "total_steps": 390, "loss": 0.418, "lr": 3.222841616184025e-05, "epoch": 1.8048, "percentage": 36.15, "elapsed_time": "6:03:07", "remaining_time": "10:41:15"} +{"current_steps": 142, "total_steps": 390, "loss": 0.4028, "lr": 3.208627810922665e-05, "epoch": 1.8176, "percentage": 36.41, "elapsed_time": "6:05:40", "remaining_time": "10:38:38"} +{"current_steps": 143, "total_steps": 390, "loss": 0.4132, "lr": 3.194317183405573e-05, "epoch": 1.8304, "percentage": 36.67, "elapsed_time": "6:07:58", "remaining_time": "10:35:36"} +{"current_steps": 144, "total_steps": 390, "loss": 0.4031, "lr": 3.1799108800462466e-05, "epoch": 1.8432, "percentage": 36.92, "elapsed_time": "6:10:28", "remaining_time": "10:32:54"} +{"current_steps": 145, "total_steps": 390, "loss": 0.4162, "lr": 3.1654100549227024e-05, "epoch": 1.8559999999999999, "percentage": 37.18, "elapsed_time": "6:13:00", "remaining_time": "10:30:16"} +{"current_steps": 146, "total_steps": 390, "loss": 0.4015, "lr": 3.1508158696850275e-05, "epoch": 1.8688, "percentage": 37.44, "elapsed_time": "6:15:39", "remaining_time": "10:27:48"} +{"current_steps": 147, "total_steps": 390, "loss": 0.4051, "lr": 3.136129493462312e-05, "epoch": 1.8816000000000002, "percentage": 37.69, "elapsed_time": "6:18:08", "remaining_time": "10:25:05"} +{"current_steps": 148, "total_steps": 390, "loss": 0.4121, "lr": 3.121352102768998e-05, "epoch": 1.8944, "percentage": 37.95, "elapsed_time": "6:20:44", "remaining_time": "10:22:33"} +{"current_steps": 149, "total_steps": 390, "loss": 0.4126, "lr": 3.106484881410628e-05, "epoch": 1.9072, "percentage": 38.21, "elapsed_time": "6:23:31", "remaining_time": "10:20:20"} +{"current_steps": 150, "total_steps": 390, "loss": 0.4131, "lr": 3.091529020389009e-05, "epoch": 1.92, "percentage": 38.46, "elapsed_time": "6:26:03", "remaining_time": "10:17:41"} +{"current_steps": 151, "total_steps": 390, "loss": 0.4148, "lr": 3.076485717806808e-05, "epoch": 1.9327999999999999, "percentage": 38.72, "elapsed_time": "6:28:43", "remaining_time": "10:15:15"} +{"current_steps": 152, "total_steps": 390, "loss": 0.3936, "lr": 3.061356178771564e-05, "epoch": 1.9456, "percentage": 38.97, "elapsed_time": "6:31:32", "remaining_time": "10:13:04"} +{"current_steps": 153, "total_steps": 390, "loss": 0.4075, "lr": 3.0461416152991555e-05, "epoch": 1.9584000000000001, "percentage": 39.23, "elapsed_time": "6:34:03", "remaining_time": "10:10:23"} +{"current_steps": 154, "total_steps": 390, "loss": 0.4057, "lr": 3.0308432462167045e-05, "epoch": 1.9712, "percentage": 39.49, "elapsed_time": "6:36:35", "remaining_time": "10:07:45"} +{"current_steps": 155, "total_steps": 390, "loss": 0.4026, "lr": 3.015462297064936e-05, "epoch": 1.984, "percentage": 39.74, "elapsed_time": "6:39:14", "remaining_time": "10:05:17"} +{"current_steps": 156, "total_steps": 390, "loss": 0.3867, "lr": 3.0000000000000004e-05, "epoch": 1.9968, "percentage": 40.0, "elapsed_time": "6:41:29", "remaining_time": "10:02:14"} +{"current_steps": 157, "total_steps": 390, "loss": 0.3827, "lr": 2.98445759369477e-05, "epoch": 2.0096, "percentage": 40.26, "elapsed_time": "6:44:50", "remaining_time": "10:00:49"} +{"current_steps": 158, "total_steps": 390, "loss": 0.3576, "lr": 2.9688363232396056e-05, "epoch": 2.0224, "percentage": 40.51, "elapsed_time": "6:47:28", "remaining_time": "9:58:18"} +{"current_steps": 159, "total_steps": 390, "loss": 0.358, "lr": 2.9531374400426158e-05, "epoch": 2.0352, "percentage": 40.77, "elapsed_time": "6:50:05", "remaining_time": "9:55:46"} +{"current_steps": 160, "total_steps": 390, "loss": 0.3447, "lr": 2.9373622017294075e-05, "epoch": 2.048, "percentage": 41.03, "elapsed_time": "6:52:36", "remaining_time": "9:53:07"} +{"current_steps": 161, "total_steps": 390, "loss": 0.3417, "lr": 2.9215118720423375e-05, "epoch": 2.0608, "percentage": 41.28, "elapsed_time": "6:55:11", "remaining_time": "9:50:32"} +{"current_steps": 162, "total_steps": 390, "loss": 0.3474, "lr": 2.9055877207392752e-05, "epoch": 2.0736, "percentage": 41.54, "elapsed_time": "6:57:48", "remaining_time": "9:48:01"} +{"current_steps": 163, "total_steps": 390, "loss": 0.3546, "lr": 2.8895910234918828e-05, "epoch": 2.0864, "percentage": 41.79, "elapsed_time": "7:00:13", "remaining_time": "9:45:13"} +{"current_steps": 164, "total_steps": 390, "loss": 0.3483, "lr": 2.873523061783426e-05, "epoch": 2.0992, "percentage": 42.05, "elapsed_time": "7:02:43", "remaining_time": "9:42:31"} +{"current_steps": 165, "total_steps": 390, "loss": 0.341, "lr": 2.8573851228061084e-05, "epoch": 2.112, "percentage": 42.31, "elapsed_time": "7:05:24", "remaining_time": "9:40:05"} +{"current_steps": 166, "total_steps": 390, "loss": 0.3389, "lr": 2.8411784993579633e-05, "epoch": 2.1248, "percentage": 42.56, "elapsed_time": "7:07:45", "remaining_time": "9:37:12"} +{"current_steps": 167, "total_steps": 390, "loss": 0.3473, "lr": 2.8249044897392814e-05, "epoch": 2.1376, "percentage": 42.82, "elapsed_time": "7:10:16", "remaining_time": "9:34:33"} +{"current_steps": 168, "total_steps": 390, "loss": 0.3393, "lr": 2.80856439764861e-05, "epoch": 2.1504, "percentage": 43.08, "elapsed_time": "7:12:51", "remaining_time": "9:31:59"} +{"current_steps": 169, "total_steps": 390, "loss": 0.3409, "lr": 2.792159532078314e-05, "epoch": 2.1632, "percentage": 43.33, "elapsed_time": "7:15:24", "remaining_time": "9:29:22"} +{"current_steps": 170, "total_steps": 390, "loss": 0.3339, "lr": 2.77569120720971e-05, "epoch": 2.176, "percentage": 43.59, "elapsed_time": "7:18:02", "remaining_time": "9:26:53"} +{"current_steps": 171, "total_steps": 390, "loss": 0.3441, "lr": 2.7591607423077932e-05, "epoch": 2.1888, "percentage": 43.85, "elapsed_time": "7:20:39", "remaining_time": "9:24:21"} +{"current_steps": 172, "total_steps": 390, "loss": 0.3413, "lr": 2.7425694616155474e-05, "epoch": 2.2016, "percentage": 44.1, "elapsed_time": "7:23:17", "remaining_time": "9:21:50"} +{"current_steps": 173, "total_steps": 390, "loss": 0.3455, "lr": 2.7259186942478656e-05, "epoch": 2.2144, "percentage": 44.36, "elapsed_time": "7:25:46", "remaining_time": "9:19:08"} +{"current_steps": 174, "total_steps": 390, "loss": 0.3481, "lr": 2.7092097740850712e-05, "epoch": 2.2272, "percentage": 44.62, "elapsed_time": "7:28:20", "remaining_time": "9:16:33"} +{"current_steps": 175, "total_steps": 390, "loss": 0.3337, "lr": 2.692444039666066e-05, "epoch": 2.24, "percentage": 44.87, "elapsed_time": "7:30:54", "remaining_time": "9:13:58"} +{"current_steps": 176, "total_steps": 390, "loss": 0.3423, "lr": 2.6756228340810946e-05, "epoch": 2.2528, "percentage": 45.13, "elapsed_time": "7:33:20", "remaining_time": "9:11:13"} +{"current_steps": 177, "total_steps": 390, "loss": 0.3466, "lr": 2.6587475048641596e-05, "epoch": 2.2656, "percentage": 45.38, "elapsed_time": "7:36:11", "remaining_time": "9:08:58"} +{"current_steps": 178, "total_steps": 390, "loss": 0.3401, "lr": 2.6418194038850634e-05, "epoch": 2.2784, "percentage": 45.64, "elapsed_time": "7:38:55", "remaining_time": "9:06:35"} +{"current_steps": 179, "total_steps": 390, "loss": 0.3355, "lr": 2.624839887241115e-05, "epoch": 2.2912, "percentage": 45.9, "elapsed_time": "7:41:17", "remaining_time": "9:03:45"} +{"current_steps": 180, "total_steps": 390, "loss": 0.3384, "lr": 2.607810315148494e-05, "epoch": 2.304, "percentage": 46.15, "elapsed_time": "7:43:58", "remaining_time": "9:01:17"} +{"current_steps": 181, "total_steps": 390, "loss": 0.3434, "lr": 2.5907320518332827e-05, "epoch": 2.3168, "percentage": 46.41, "elapsed_time": "7:46:34", "remaining_time": "8:58:45"} +{"current_steps": 182, "total_steps": 390, "loss": 0.347, "lr": 2.5736064654221808e-05, "epoch": 2.3296, "percentage": 46.67, "elapsed_time": "7:49:07", "remaining_time": "8:56:08"} +{"current_steps": 183, "total_steps": 390, "loss": 0.3434, "lr": 2.5564349278329056e-05, "epoch": 2.3424, "percentage": 46.92, "elapsed_time": "7:51:38", "remaining_time": "8:53:29"} +{"current_steps": 184, "total_steps": 390, "loss": 0.336, "lr": 2.539218814664288e-05, "epoch": 2.3552, "percentage": 47.18, "elapsed_time": "7:54:12", "remaining_time": "8:50:54"} +{"current_steps": 185, "total_steps": 390, "loss": 0.3431, "lr": 2.521959505086075e-05, "epoch": 2.368, "percentage": 47.44, "elapsed_time": "7:56:55", "remaining_time": "8:48:28"} +{"current_steps": 186, "total_steps": 390, "loss": 0.3373, "lr": 2.5046583817284437e-05, "epoch": 2.3808, "percentage": 47.69, "elapsed_time": "7:59:30", "remaining_time": "8:45:54"} +{"current_steps": 187, "total_steps": 390, "loss": 0.3389, "lr": 2.487316830571244e-05, "epoch": 2.3936, "percentage": 47.95, "elapsed_time": "8:02:16", "remaining_time": "8:43:32"} +{"current_steps": 188, "total_steps": 390, "loss": 0.3303, "lr": 2.4699362408329646e-05, "epoch": 2.4064, "percentage": 48.21, "elapsed_time": "8:04:44", "remaining_time": "8:40:50"} +{"current_steps": 189, "total_steps": 390, "loss": 0.3502, "lr": 2.4525180048594452e-05, "epoch": 2.4192, "percentage": 48.46, "elapsed_time": "8:07:20", "remaining_time": "8:38:16"} +{"current_steps": 190, "total_steps": 390, "loss": 0.342, "lr": 2.435063518012335e-05, "epoch": 2.432, "percentage": 48.72, "elapsed_time": "8:09:44", "remaining_time": "8:35:31"} +{"current_steps": 191, "total_steps": 390, "loss": 0.3486, "lr": 2.4175741785573177e-05, "epoch": 2.4448, "percentage": 48.97, "elapsed_time": "8:12:30", "remaining_time": "8:33:08"} +{"current_steps": 192, "total_steps": 390, "loss": 0.339, "lr": 2.4000513875520892e-05, "epoch": 2.4576000000000002, "percentage": 49.23, "elapsed_time": "8:15:03", "remaining_time": "8:30:32"} +{"current_steps": 193, "total_steps": 390, "loss": 0.3346, "lr": 2.3824965487341247e-05, "epoch": 2.4704, "percentage": 49.49, "elapsed_time": "8:17:28", "remaining_time": "8:27:46"} +{"current_steps": 194, "total_steps": 390, "loss": 0.3413, "lr": 2.3649110684082258e-05, "epoch": 2.4832, "percentage": 49.74, "elapsed_time": "8:19:59", "remaining_time": "8:25:09"} +{"current_steps": 195, "total_steps": 390, "loss": 0.3415, "lr": 2.3472963553338614e-05, "epoch": 2.496, "percentage": 50.0, "elapsed_time": "8:22:51", "remaining_time": "8:22:51"} +{"current_steps": 196, "total_steps": 390, "loss": 0.3424, "lr": 2.3296538206123134e-05, "epoch": 2.5088, "percentage": 50.26, "elapsed_time": "8:25:38", "remaining_time": "8:20:29"} +{"current_steps": 197, "total_steps": 390, "loss": 0.3308, "lr": 2.311984877573636e-05, "epoch": 2.5216, "percentage": 50.51, "elapsed_time": "8:28:10", "remaining_time": "8:17:51"} +{"current_steps": 198, "total_steps": 390, "loss": 0.3457, "lr": 2.2942909416634326e-05, "epoch": 2.5343999999999998, "percentage": 50.77, "elapsed_time": "8:30:52", "remaining_time": "8:15:24"} +{"current_steps": 199, "total_steps": 390, "loss": 0.3428, "lr": 2.2765734303294666e-05, "epoch": 2.5472, "percentage": 51.03, "elapsed_time": "8:33:16", "remaining_time": "8:12:38"} +{"current_steps": 200, "total_steps": 390, "loss": 0.3406, "lr": 2.2588337629081107e-05, "epoch": 2.56, "percentage": 51.28, "elapsed_time": "8:35:59", "remaining_time": "8:10:11"} +{"current_steps": 201, "total_steps": 390, "loss": 0.3452, "lr": 2.2410733605106462e-05, "epoch": 2.5728, "percentage": 51.54, "elapsed_time": "8:38:31", "remaining_time": "8:07:34"} +{"current_steps": 202, "total_steps": 390, "loss": 0.3399, "lr": 2.2232936459094158e-05, "epoch": 2.5856, "percentage": 51.79, "elapsed_time": "8:41:03", "remaining_time": "8:04:56"} +{"current_steps": 203, "total_steps": 390, "loss": 0.3355, "lr": 2.205496043423849e-05, "epoch": 2.5984, "percentage": 52.05, "elapsed_time": "8:43:20", "remaining_time": "8:02:05"} +{"current_steps": 204, "total_steps": 390, "loss": 0.3394, "lr": 2.1876819788063586e-05, "epoch": 2.6112, "percentage": 52.31, "elapsed_time": "8:45:57", "remaining_time": "7:59:32"} +{"current_steps": 205, "total_steps": 390, "loss": 0.3458, "lr": 2.16985287912813e-05, "epoch": 2.624, "percentage": 52.56, "elapsed_time": "8:48:35", "remaining_time": "7:57:00"} +{"current_steps": 206, "total_steps": 390, "loss": 0.3427, "lr": 2.1520101726647922e-05, "epoch": 2.6368, "percentage": 52.82, "elapsed_time": "8:51:19", "remaining_time": "7:54:35"} +{"current_steps": 207, "total_steps": 390, "loss": 0.341, "lr": 2.1341552887820048e-05, "epoch": 2.6496, "percentage": 53.08, "elapsed_time": "8:53:41", "remaining_time": "7:51:49"} +{"current_steps": 208, "total_steps": 390, "loss": 0.3451, "lr": 2.1162896578209517e-05, "epoch": 2.6624, "percentage": 53.33, "elapsed_time": "8:56:15", "remaining_time": "7:49:13"} +{"current_steps": 209, "total_steps": 390, "loss": 0.341, "lr": 2.0984147109837564e-05, "epoch": 2.6752000000000002, "percentage": 53.59, "elapsed_time": "8:58:46", "remaining_time": "7:46:35"} +{"current_steps": 210, "total_steps": 390, "loss": 0.3499, "lr": 2.0805318802188307e-05, "epoch": 2.6879999999999997, "percentage": 53.85, "elapsed_time": "9:01:30", "remaining_time": "7:44:09"} +{"current_steps": 211, "total_steps": 390, "loss": 0.337, "lr": 2.0626425981061608e-05, "epoch": 2.7008, "percentage": 54.1, "elapsed_time": "9:04:15", "remaining_time": "7:41:42"} +{"current_steps": 212, "total_steps": 390, "loss": 0.3364, "lr": 2.0447482977425465e-05, "epoch": 2.7136, "percentage": 54.36, "elapsed_time": "9:06:48", "remaining_time": "7:39:06"} +{"current_steps": 213, "total_steps": 390, "loss": 0.339, "lr": 2.0268504126267952e-05, "epoch": 2.7264, "percentage": 54.62, "elapsed_time": "9:09:19", "remaining_time": "7:36:28"} +{"current_steps": 214, "total_steps": 390, "loss": 0.3425, "lr": 2.008950376544887e-05, "epoch": 2.7392, "percentage": 54.87, "elapsed_time": "9:11:54", "remaining_time": "7:33:53"} +{"current_steps": 215, "total_steps": 390, "loss": 0.3402, "lr": 1.9910496234551132e-05, "epoch": 2.752, "percentage": 55.13, "elapsed_time": "9:14:21", "remaining_time": "7:31:13"} +{"current_steps": 216, "total_steps": 390, "loss": 0.3531, "lr": 1.9731495873732055e-05, "epoch": 2.7648, "percentage": 55.38, "elapsed_time": "9:16:52", "remaining_time": "7:28:35"} +{"current_steps": 217, "total_steps": 390, "loss": 0.343, "lr": 1.9552517022574542e-05, "epoch": 2.7776, "percentage": 55.64, "elapsed_time": "9:19:11", "remaining_time": "7:25:48"} +{"current_steps": 218, "total_steps": 390, "loss": 0.3449, "lr": 1.93735740189384e-05, "epoch": 2.7904, "percentage": 55.9, "elapsed_time": "9:21:50", "remaining_time": "7:23:17"} +{"current_steps": 219, "total_steps": 390, "loss": 0.3569, "lr": 1.9194681197811703e-05, "epoch": 2.8032, "percentage": 56.15, "elapsed_time": "9:24:26", "remaining_time": "7:20:44"} +{"current_steps": 220, "total_steps": 390, "loss": 0.3338, "lr": 1.901585289016244e-05, "epoch": 2.816, "percentage": 56.41, "elapsed_time": "9:26:59", "remaining_time": "7:18:08"} +{"current_steps": 221, "total_steps": 390, "loss": 0.3316, "lr": 1.8837103421790486e-05, "epoch": 2.8288, "percentage": 56.67, "elapsed_time": "9:29:23", "remaining_time": "7:15:25"} +{"current_steps": 222, "total_steps": 390, "loss": 0.3326, "lr": 1.8658447112179952e-05, "epoch": 2.8416, "percentage": 56.92, "elapsed_time": "9:32:05", "remaining_time": "7:12:56"} +{"current_steps": 223, "total_steps": 390, "loss": 0.3269, "lr": 1.8479898273352084e-05, "epoch": 2.8544, "percentage": 57.18, "elapsed_time": "9:34:37", "remaining_time": "7:10:19"} +{"current_steps": 224, "total_steps": 390, "loss": 0.3437, "lr": 1.83014712087187e-05, "epoch": 2.8672, "percentage": 57.44, "elapsed_time": "9:37:28", "remaining_time": "7:07:56"} +{"current_steps": 225, "total_steps": 390, "loss": 0.3317, "lr": 1.8123180211936417e-05, "epoch": 2.88, "percentage": 57.69, "elapsed_time": "9:39:56", "remaining_time": "7:05:17"} +{"current_steps": 226, "total_steps": 390, "loss": 0.3391, "lr": 1.794503956576152e-05, "epoch": 2.8928000000000003, "percentage": 57.95, "elapsed_time": "9:42:31", "remaining_time": "7:02:42"} +{"current_steps": 227, "total_steps": 390, "loss": 0.349, "lr": 1.776706354090585e-05, "epoch": 2.9055999999999997, "percentage": 58.21, "elapsed_time": "9:45:06", "remaining_time": "7:00:08"} +{"current_steps": 228, "total_steps": 390, "loss": 0.3435, "lr": 1.758926639489354e-05, "epoch": 2.9184, "percentage": 58.46, "elapsed_time": "9:47:43", "remaining_time": "6:57:35"} +{"current_steps": 229, "total_steps": 390, "loss": 0.3367, "lr": 1.7411662370918893e-05, "epoch": 2.9312, "percentage": 58.72, "elapsed_time": "9:50:09", "remaining_time": "6:54:54"} +{"current_steps": 230, "total_steps": 390, "loss": 0.3398, "lr": 1.7234265696705344e-05, "epoch": 2.944, "percentage": 58.97, "elapsed_time": "9:52:43", "remaining_time": "6:52:19"} +{"current_steps": 231, "total_steps": 390, "loss": 0.3401, "lr": 1.7057090583365678e-05, "epoch": 2.9568, "percentage": 59.23, "elapsed_time": "9:55:04", "remaining_time": "6:49:35"} +{"current_steps": 232, "total_steps": 390, "loss": 0.3468, "lr": 1.6880151224263646e-05, "epoch": 2.9696, "percentage": 59.49, "elapsed_time": "9:57:36", "remaining_time": "6:46:59"} +{"current_steps": 233, "total_steps": 390, "loss": 0.3455, "lr": 1.6703461793876876e-05, "epoch": 2.9824, "percentage": 59.74, "elapsed_time": "10:00:00", "remaining_time": "6:44:17"} +{"current_steps": 234, "total_steps": 390, "loss": 0.3425, "lr": 1.6527036446661396e-05, "epoch": 2.9952, "percentage": 60.0, "elapsed_time": "10:02:32", "remaining_time": "6:41:41"} +{"current_steps": 235, "total_steps": 390, "loss": 0.312, "lr": 1.635088931591775e-05, "epoch": 3.008, "percentage": 60.26, "elapsed_time": "10:05:58", "remaining_time": "6:39:40"} +{"current_steps": 236, "total_steps": 390, "loss": 0.2867, "lr": 1.6175034512658753e-05, "epoch": 3.0208, "percentage": 60.51, "elapsed_time": "10:08:37", "remaining_time": "6:37:09"} +{"current_steps": 237, "total_steps": 390, "loss": 0.2916, "lr": 1.5999486124479115e-05, "epoch": 3.0336, "percentage": 60.77, "elapsed_time": "10:11:20", "remaining_time": "6:34:39"} +{"current_steps": 238, "total_steps": 390, "loss": 0.2841, "lr": 1.5824258214426833e-05, "epoch": 3.0464, "percentage": 61.03, "elapsed_time": "10:13:59", "remaining_time": "6:32:07"} +{"current_steps": 239, "total_steps": 390, "loss": 0.2853, "lr": 1.5649364819876655e-05, "epoch": 3.0592, "percentage": 61.28, "elapsed_time": "10:16:35", "remaining_time": "6:29:33"} +{"current_steps": 240, "total_steps": 390, "loss": 0.2738, "lr": 1.547481995140556e-05, "epoch": 3.072, "percentage": 61.54, "elapsed_time": "10:18:53", "remaining_time": "6:26:48"} +{"current_steps": 241, "total_steps": 390, "loss": 0.2873, "lr": 1.5300637591670357e-05, "epoch": 3.0848, "percentage": 61.79, "elapsed_time": "10:21:35", "remaining_time": "6:24:18"} +{"current_steps": 242, "total_steps": 390, "loss": 0.2693, "lr": 1.5126831694287564e-05, "epoch": 3.0976, "percentage": 62.05, "elapsed_time": "10:23:49", "remaining_time": "6:21:30"} +{"current_steps": 243, "total_steps": 390, "loss": 0.2759, "lr": 1.4953416182715566e-05, "epoch": 3.1104, "percentage": 62.31, "elapsed_time": "10:26:28", "remaining_time": "6:18:58"} +{"current_steps": 244, "total_steps": 390, "loss": 0.2836, "lr": 1.478040494913926e-05, "epoch": 3.1232, "percentage": 62.56, "elapsed_time": "10:29:08", "remaining_time": "6:16:27"} +{"current_steps": 245, "total_steps": 390, "loss": 0.2819, "lr": 1.460781185335713e-05, "epoch": 3.136, "percentage": 62.82, "elapsed_time": "10:31:45", "remaining_time": "6:13:53"} +{"current_steps": 246, "total_steps": 390, "loss": 0.2752, "lr": 1.443565072167095e-05, "epoch": 3.1488, "percentage": 63.08, "elapsed_time": "10:34:12", "remaining_time": "6:11:14"} +{"current_steps": 247, "total_steps": 390, "loss": 0.2796, "lr": 1.4263935345778202e-05, "epoch": 3.1616, "percentage": 63.33, "elapsed_time": "10:36:46", "remaining_time": "6:08:39"} +{"current_steps": 248, "total_steps": 390, "loss": 0.2891, "lr": 1.409267948166718e-05, "epoch": 3.1744, "percentage": 63.59, "elapsed_time": "10:39:22", "remaining_time": "6:06:05"} +{"current_steps": 249, "total_steps": 390, "loss": 0.2909, "lr": 1.3921896848515064e-05, "epoch": 3.1872, "percentage": 63.85, "elapsed_time": "10:42:04", "remaining_time": "6:03:34"} +{"current_steps": 250, "total_steps": 390, "loss": 0.2779, "lr": 1.3751601127588849e-05, "epoch": 3.2, "percentage": 64.1, "elapsed_time": "10:44:34", "remaining_time": "6:00:57"} +{"current_steps": 251, "total_steps": 390, "loss": 0.2724, "lr": 1.3581805961149371e-05, "epoch": 3.2128, "percentage": 64.36, "elapsed_time": "10:47:05", "remaining_time": "5:58:21"} +{"current_steps": 252, "total_steps": 390, "loss": 0.2861, "lr": 1.341252495135841e-05, "epoch": 3.2256, "percentage": 64.62, "elapsed_time": "10:49:49", "remaining_time": "5:55:51"} +{"current_steps": 253, "total_steps": 390, "loss": 0.2749, "lr": 1.324377165918906e-05, "epoch": 3.2384, "percentage": 64.87, "elapsed_time": "10:52:30", "remaining_time": "5:53:20"} +{"current_steps": 254, "total_steps": 390, "loss": 0.2798, "lr": 1.3075559603339354e-05, "epoch": 3.2512, "percentage": 65.13, "elapsed_time": "10:55:05", "remaining_time": "5:50:45"} +{"current_steps": 255, "total_steps": 390, "loss": 0.2761, "lr": 1.2907902259149287e-05, "epoch": 3.2640000000000002, "percentage": 65.38, "elapsed_time": "10:57:31", "remaining_time": "5:48:06"} +{"current_steps": 256, "total_steps": 390, "loss": 0.2813, "lr": 1.274081305752135e-05, "epoch": 3.2768, "percentage": 65.64, "elapsed_time": "10:59:59", "remaining_time": "5:45:27"} +{"current_steps": 257, "total_steps": 390, "loss": 0.2861, "lr": 1.2574305383844528e-05, "epoch": 3.2896, "percentage": 65.9, "elapsed_time": "11:02:37", "remaining_time": "5:42:54"} +{"current_steps": 258, "total_steps": 390, "loss": 0.2769, "lr": 1.2408392576922075e-05, "epoch": 3.3024, "percentage": 66.15, "elapsed_time": "11:05:04", "remaining_time": "5:40:16"} +{"current_steps": 259, "total_steps": 390, "loss": 0.2824, "lr": 1.2243087927902905e-05, "epoch": 3.3152, "percentage": 66.41, "elapsed_time": "11:07:36", "remaining_time": "5:37:40"} +{"current_steps": 260, "total_steps": 390, "loss": 0.2847, "lr": 1.2078404679216864e-05, "epoch": 3.328, "percentage": 66.67, "elapsed_time": "11:10:03", "remaining_time": "5:35:01"} +{"current_steps": 261, "total_steps": 390, "loss": 0.2777, "lr": 1.1914356023513904e-05, "epoch": 3.3407999999999998, "percentage": 66.92, "elapsed_time": "11:12:45", "remaining_time": "5:32:30"} +{"current_steps": 262, "total_steps": 390, "loss": 0.2719, "lr": 1.1750955102607193e-05, "epoch": 3.3536, "percentage": 67.18, "elapsed_time": "11:15:13", "remaining_time": "5:29:53"} +{"current_steps": 263, "total_steps": 390, "loss": 0.2788, "lr": 1.1588215006420374e-05, "epoch": 3.3664, "percentage": 67.44, "elapsed_time": "11:17:47", "remaining_time": "5:27:17"} +{"current_steps": 264, "total_steps": 390, "loss": 0.2746, "lr": 1.1426148771938915e-05, "epoch": 3.3792, "percentage": 67.69, "elapsed_time": "11:20:09", "remaining_time": "5:24:37"} +{"current_steps": 265, "total_steps": 390, "loss": 0.2816, "lr": 1.1264769382165748e-05, "epoch": 3.392, "percentage": 67.95, "elapsed_time": "11:22:43", "remaining_time": "5:22:02"} +{"current_steps": 266, "total_steps": 390, "loss": 0.2819, "lr": 1.110408976508118e-05, "epoch": 3.4048, "percentage": 68.21, "elapsed_time": "11:25:16", "remaining_time": "5:19:27"} +{"current_steps": 267, "total_steps": 390, "loss": 0.2756, "lr": 1.094412279260726e-05, "epoch": 3.4176, "percentage": 68.46, "elapsed_time": "11:27:43", "remaining_time": "5:16:48"} +{"current_steps": 268, "total_steps": 390, "loss": 0.2689, "lr": 1.0784881279576635e-05, "epoch": 3.4304, "percentage": 68.72, "elapsed_time": "11:30:20", "remaining_time": "5:14:15"} +{"current_steps": 269, "total_steps": 390, "loss": 0.2807, "lr": 1.0626377982705929e-05, "epoch": 3.4432, "percentage": 68.97, "elapsed_time": "11:32:50", "remaining_time": "5:11:39"} +{"current_steps": 270, "total_steps": 390, "loss": 0.2849, "lr": 1.0468625599573842e-05, "epoch": 3.456, "percentage": 69.23, "elapsed_time": "11:35:22", "remaining_time": "5:09:03"} +{"current_steps": 271, "total_steps": 390, "loss": 0.2825, "lr": 1.0311636767603952e-05, "epoch": 3.4688, "percentage": 69.49, "elapsed_time": "11:37:55", "remaining_time": "5:06:28"} +{"current_steps": 272, "total_steps": 390, "loss": 0.2911, "lr": 1.0155424063052306e-05, "epoch": 3.4816, "percentage": 69.74, "elapsed_time": "11:40:42", "remaining_time": "5:03:59"} +{"current_steps": 273, "total_steps": 390, "loss": 0.2897, "lr": 1.0000000000000006e-05, "epoch": 3.4944, "percentage": 70.0, "elapsed_time": "11:43:11", "remaining_time": "5:01:21"} +{"current_steps": 274, "total_steps": 390, "loss": 0.281, "lr": 9.84537702935065e-06, "epoch": 3.5072, "percentage": 70.26, "elapsed_time": "11:45:38", "remaining_time": "4:58:44"} +{"current_steps": 275, "total_steps": 390, "loss": 0.2905, "lr": 9.691567537832964e-06, "epoch": 3.52, "percentage": 70.51, "elapsed_time": "11:48:13", "remaining_time": "4:56:09"} +{"current_steps": 276, "total_steps": 390, "loss": 0.2971, "lr": 9.538583847008452e-06, "epoch": 3.5328, "percentage": 70.77, "elapsed_time": "11:50:54", "remaining_time": "4:53:38"} +{"current_steps": 277, "total_steps": 390, "loss": 0.2777, "lr": 9.386438212284372e-06, "epoch": 3.5456, "percentage": 71.03, "elapsed_time": "11:53:29", "remaining_time": "4:51:03"} +{"current_steps": 278, "total_steps": 390, "loss": 0.2789, "lr": 9.235142821931928e-06, "epoch": 3.5584, "percentage": 71.28, "elapsed_time": "11:55:55", "remaining_time": "4:48:25"} +{"current_steps": 279, "total_steps": 390, "loss": 0.2843, "lr": 9.084709796109907e-06, "epoch": 3.5712, "percentage": 71.54, "elapsed_time": "11:58:25", "remaining_time": "4:45:49"} +{"current_steps": 280, "total_steps": 390, "loss": 0.2822, "lr": 8.93515118589373e-06, "epoch": 3.584, "percentage": 71.79, "elapsed_time": "12:01:07", "remaining_time": "4:43:17"} +{"current_steps": 281, "total_steps": 390, "loss": 0.288, "lr": 8.786478972310023e-06, "epoch": 3.5968, "percentage": 72.05, "elapsed_time": "12:03:40", "remaining_time": "4:40:42"} +{"current_steps": 282, "total_steps": 390, "loss": 0.2787, "lr": 8.638705065376887e-06, "epoch": 3.6096, "percentage": 72.31, "elapsed_time": "12:06:19", "remaining_time": "4:38:09"} +{"current_steps": 283, "total_steps": 390, "loss": 0.2878, "lr": 8.491841303149728e-06, "epoch": 3.6224, "percentage": 72.56, "elapsed_time": "12:08:51", "remaining_time": "4:35:34"} +{"current_steps": 284, "total_steps": 390, "loss": 0.2897, "lr": 8.345899450772975e-06, "epoch": 3.6352, "percentage": 72.82, "elapsed_time": "12:11:39", "remaining_time": "4:33:04"} +{"current_steps": 285, "total_steps": 390, "loss": 0.2726, "lr": 8.200891199537549e-06, "epoch": 3.648, "percentage": 73.08, "elapsed_time": "12:14:02", "remaining_time": "4:30:26"} +{"current_steps": 286, "total_steps": 390, "loss": 0.2833, "lr": 8.056828165944282e-06, "epoch": 3.6608, "percentage": 73.33, "elapsed_time": "12:16:45", "remaining_time": "4:27:54"} +{"current_steps": 287, "total_steps": 390, "loss": 0.2811, "lr": 7.913721890773354e-06, "epoch": 3.6736, "percentage": 73.59, "elapsed_time": "12:19:20", "remaining_time": "4:25:20"} +{"current_steps": 288, "total_steps": 390, "loss": 0.2826, "lr": 7.771583838159756e-06, "epoch": 3.6864, "percentage": 73.85, "elapsed_time": "12:21:56", "remaining_time": "4:22:46"} +{"current_steps": 289, "total_steps": 390, "loss": 0.2826, "lr": 7.630425394674903e-06, "epoch": 3.6992000000000003, "percentage": 74.1, "elapsed_time": "12:24:42", "remaining_time": "4:20:15"} +{"current_steps": 290, "total_steps": 390, "loss": 0.2853, "lr": 7.49025786841445e-06, "epoch": 3.7119999999999997, "percentage": 74.36, "elapsed_time": "12:27:12", "remaining_time": "4:17:39"} +{"current_steps": 291, "total_steps": 390, "loss": 0.2841, "lr": 7.3510924880924575e-06, "epoch": 3.7248, "percentage": 74.62, "elapsed_time": "12:29:54", "remaining_time": "4:15:07"} +{"current_steps": 292, "total_steps": 390, "loss": 0.2835, "lr": 7.212940402141808e-06, "epoch": 3.7376, "percentage": 74.87, "elapsed_time": "12:32:39", "remaining_time": "4:12:36"} +{"current_steps": 293, "total_steps": 390, "loss": 0.2843, "lr": 7.075812677821145e-06, "epoch": 3.7504, "percentage": 75.13, "elapsed_time": "12:35:02", "remaining_time": "4:09:57"} +{"current_steps": 294, "total_steps": 390, "loss": 0.2688, "lr": 6.939720300328303e-06, "epoch": 3.7632, "percentage": 75.38, "elapsed_time": "12:37:37", "remaining_time": "4:07:23"} +{"current_steps": 295, "total_steps": 390, "loss": 0.2754, "lr": 6.8046741719202385e-06, "epoch": 3.776, "percentage": 75.64, "elapsed_time": "12:40:08", "remaining_time": "4:04:47"} +{"current_steps": 296, "total_steps": 390, "loss": 0.28, "lr": 6.67068511103971e-06, "epoch": 3.7888, "percentage": 75.9, "elapsed_time": "12:42:45", "remaining_time": "4:02:13"} +{"current_steps": 297, "total_steps": 390, "loss": 0.2797, "lr": 6.537763851448593e-06, "epoch": 3.8016, "percentage": 76.15, "elapsed_time": "12:45:22", "remaining_time": "3:59:39"} +{"current_steps": 298, "total_steps": 390, "loss": 0.2783, "lr": 6.4059210413680175e-06, "epoch": 3.8144, "percentage": 76.41, "elapsed_time": "12:47:52", "remaining_time": "3:57:03"} +{"current_steps": 299, "total_steps": 390, "loss": 0.2806, "lr": 6.275167242625331e-06, "epoch": 3.8272, "percentage": 76.67, "elapsed_time": "12:50:35", "remaining_time": "3:54:31"} +{"current_steps": 300, "total_steps": 390, "loss": 0.2823, "lr": 6.145512929808013e-06, "epoch": 3.84, "percentage": 76.92, "elapsed_time": "12:53:18", "remaining_time": "3:51:59"} +{"current_steps": 301, "total_steps": 390, "loss": 0.2694, "lr": 6.016968489424572e-06, "epoch": 3.8528000000000002, "percentage": 77.18, "elapsed_time": "12:55:52", "remaining_time": "3:49:24"} +{"current_steps": 302, "total_steps": 390, "loss": 0.2815, "lr": 5.889544219072465e-06, "epoch": 3.8656, "percentage": 77.44, "elapsed_time": "12:58:31", "remaining_time": "3:46:51"} +{"current_steps": 303, "total_steps": 390, "loss": 0.281, "lr": 5.7632503266131925e-06, "epoch": 3.8784, "percentage": 77.69, "elapsed_time": "13:01:04", "remaining_time": "3:44:16"} +{"current_steps": 304, "total_steps": 390, "loss": 0.2698, "lr": 5.638096929354522e-06, "epoch": 3.8912, "percentage": 77.95, "elapsed_time": "13:03:37", "remaining_time": "3:41:41"} +{"current_steps": 305, "total_steps": 390, "loss": 0.2814, "lr": 5.514094053240035e-06, "epoch": 3.904, "percentage": 78.21, "elapsed_time": "13:06:09", "remaining_time": "3:39:05"} +{"current_steps": 306, "total_steps": 390, "loss": 0.281, "lr": 5.39125163204594e-06, "epoch": 3.9168, "percentage": 78.46, "elapsed_time": "13:08:47", "remaining_time": "3:36:31"} +{"current_steps": 307, "total_steps": 390, "loss": 0.2784, "lr": 5.269579506585259e-06, "epoch": 3.9295999999999998, "percentage": 78.72, "elapsed_time": "13:11:26", "remaining_time": "3:33:58"} +{"current_steps": 308, "total_steps": 390, "loss": 0.2684, "lr": 5.149087423919541e-06, "epoch": 3.9424, "percentage": 78.97, "elapsed_time": "13:13:59", "remaining_time": "3:31:23"} +{"current_steps": 309, "total_steps": 390, "loss": 0.2812, "lr": 5.029785036577976e-06, "epoch": 3.9552, "percentage": 79.23, "elapsed_time": "13:16:36", "remaining_time": "3:28:49"} +{"current_steps": 310, "total_steps": 390, "loss": 0.2717, "lr": 4.911681901784198e-06, "epoch": 3.968, "percentage": 79.49, "elapsed_time": "13:19:06", "remaining_time": "3:26:13"} +{"current_steps": 311, "total_steps": 390, "loss": 0.2738, "lr": 4.794787480690597e-06, "epoch": 3.9808, "percentage": 79.74, "elapsed_time": "13:21:41", "remaining_time": "3:23:38"} +{"current_steps": 312, "total_steps": 390, "loss": 0.2782, "lr": 4.679111137620442e-06, "epoch": 3.9936, "percentage": 80.0, "elapsed_time": "13:24:18", "remaining_time": "3:21:04"} +{"current_steps": 313, "total_steps": 390, "loss": 0.2671, "lr": 4.5646621393177e-06, "epoch": 4.0064, "percentage": 80.26, "elapsed_time": "13:27:42", "remaining_time": "3:18:42"} +{"current_steps": 314, "total_steps": 390, "loss": 0.2487, "lr": 4.451449654204685e-06, "epoch": 4.0192, "percentage": 80.51, "elapsed_time": "13:30:17", "remaining_time": "3:16:07"} +{"current_steps": 315, "total_steps": 390, "loss": 0.2463, "lr": 4.339482751647557e-06, "epoch": 4.032, "percentage": 80.77, "elapsed_time": "13:32:54", "remaining_time": "3:13:32"} +{"current_steps": 316, "total_steps": 390, "loss": 0.2385, "lr": 4.228770401229824e-06, "epoch": 4.0448, "percentage": 81.03, "elapsed_time": "13:35:29", "remaining_time": "3:10:58"} +{"current_steps": 317, "total_steps": 390, "loss": 0.2463, "lr": 4.119321472033779e-06, "epoch": 4.0576, "percentage": 81.28, "elapsed_time": "13:38:03", "remaining_time": "3:08:23"} +{"current_steps": 318, "total_steps": 390, "loss": 0.2441, "lr": 4.011144731929981e-06, "epoch": 4.0704, "percentage": 81.54, "elapsed_time": "13:40:35", "remaining_time": "3:05:47"} +{"current_steps": 319, "total_steps": 390, "loss": 0.2356, "lr": 3.904248846874894e-06, "epoch": 4.0832, "percentage": 81.79, "elapsed_time": "13:42:57", "remaining_time": "3:03:10"} +{"current_steps": 320, "total_steps": 390, "loss": 0.2515, "lr": 3.7986423802166705e-06, "epoch": 4.096, "percentage": 82.05, "elapsed_time": "13:45:24", "remaining_time": "3:00:33"} +{"current_steps": 321, "total_steps": 390, "loss": 0.2536, "lr": 3.694333792009115e-06, "epoch": 4.1088, "percentage": 82.31, "elapsed_time": "13:48:01", "remaining_time": "2:57:59"} +{"current_steps": 322, "total_steps": 390, "loss": 0.2503, "lr": 3.5913314383339937e-06, "epoch": 4.1216, "percentage": 82.56, "elapsed_time": "13:50:42", "remaining_time": "2:55:25"} +{"current_steps": 323, "total_steps": 390, "loss": 0.2481, "lr": 3.4896435706316e-06, "epoch": 4.1344, "percentage": 82.82, "elapsed_time": "13:53:09", "remaining_time": "2:52:49"} +{"current_steps": 324, "total_steps": 390, "loss": 0.234, "lr": 3.3892783350397675e-06, "epoch": 4.1472, "percentage": 83.08, "elapsed_time": "13:55:36", "remaining_time": "2:50:12"} +{"current_steps": 325, "total_steps": 390, "loss": 0.2427, "lr": 3.290243771741275e-06, "epoch": 4.16, "percentage": 83.33, "elapsed_time": "13:58:12", "remaining_time": "2:47:38"} +{"current_steps": 326, "total_steps": 390, "loss": 0.2453, "lr": 3.1925478143197418e-06, "epoch": 4.1728, "percentage": 83.59, "elapsed_time": "14:00:52", "remaining_time": "2:45:04"} +{"current_steps": 327, "total_steps": 390, "loss": 0.2476, "lr": 3.0961982891241083e-06, "epoch": 4.1856, "percentage": 83.85, "elapsed_time": "14:03:26", "remaining_time": "2:42:29"} +{"current_steps": 328, "total_steps": 390, "loss": 0.2529, "lr": 3.001202914641628e-06, "epoch": 4.1984, "percentage": 84.1, "elapsed_time": "14:05:55", "remaining_time": "2:39:54"} +{"current_steps": 329, "total_steps": 390, "loss": 0.2365, "lr": 2.907569300879596e-06, "epoch": 4.2112, "percentage": 84.36, "elapsed_time": "14:08:35", "remaining_time": "2:37:20"} +{"current_steps": 330, "total_steps": 390, "loss": 0.245, "lr": 2.815304948755664e-06, "epoch": 4.224, "percentage": 84.62, "elapsed_time": "14:11:18", "remaining_time": "2:34:46"} +{"current_steps": 331, "total_steps": 390, "loss": 0.2228, "lr": 2.7244172494969978e-06, "epoch": 4.2368, "percentage": 84.87, "elapsed_time": "14:13:44", "remaining_time": "2:32:10"} +{"current_steps": 332, "total_steps": 390, "loss": 0.2534, "lr": 2.6349134840481294e-06, "epoch": 4.2496, "percentage": 85.13, "elapsed_time": "14:16:26", "remaining_time": "2:29:37"} +{"current_steps": 333, "total_steps": 390, "loss": 0.2333, "lr": 2.546800822487714e-06, "epoch": 4.2624, "percentage": 85.38, "elapsed_time": "14:18:44", "remaining_time": "2:26:59"} +{"current_steps": 334, "total_steps": 390, "loss": 0.2506, "lr": 2.4600863234541338e-06, "epoch": 4.2752, "percentage": 85.64, "elapsed_time": "14:21:27", "remaining_time": "2:24:26"} +{"current_steps": 335, "total_steps": 390, "loss": 0.2418, "lr": 2.374776933580025e-06, "epoch": 4.288, "percentage": 85.9, "elapsed_time": "14:24:04", "remaining_time": "2:21:51"} +{"current_steps": 336, "total_steps": 390, "loss": 0.2348, "lr": 2.2908794869358044e-06, "epoch": 4.3008, "percentage": 86.15, "elapsed_time": "14:26:43", "remaining_time": "2:19:17"} +{"current_steps": 337, "total_steps": 390, "loss": 0.2472, "lr": 2.2084007044821764e-06, "epoch": 4.3136, "percentage": 86.41, "elapsed_time": "14:29:18", "remaining_time": "2:16:43"} +{"current_steps": 338, "total_steps": 390, "loss": 0.2399, "lr": 2.127347193531757e-06, "epoch": 4.3264, "percentage": 86.67, "elapsed_time": "14:31:48", "remaining_time": "2:14:07"} +{"current_steps": 339, "total_steps": 390, "loss": 0.2273, "lr": 2.0477254472197237e-06, "epoch": 4.3392, "percentage": 86.92, "elapsed_time": "14:34:21", "remaining_time": "2:11:32"} +{"current_steps": 340, "total_steps": 390, "loss": 0.2388, "lr": 1.96954184398368e-06, "epoch": 4.352, "percentage": 87.18, "elapsed_time": "14:37:13", "remaining_time": "2:09:00"} +{"current_steps": 341, "total_steps": 390, "loss": 0.2371, "lr": 1.8928026470526917e-06, "epoch": 4.3648, "percentage": 87.44, "elapsed_time": "14:39:49", "remaining_time": "2:06:25"} +{"current_steps": 342, "total_steps": 390, "loss": 0.24, "lr": 1.817514003945524e-06, "epoch": 4.3776, "percentage": 87.69, "elapsed_time": "14:42:13", "remaining_time": "2:03:49"} +{"current_steps": 343, "total_steps": 390, "loss": 0.2425, "lr": 1.743681945978184e-06, "epoch": 4.3904, "percentage": 87.95, "elapsed_time": "14:44:39", "remaining_time": "2:01:13"} +{"current_steps": 344, "total_steps": 390, "loss": 0.2355, "lr": 1.6713123877807413e-06, "epoch": 4.4032, "percentage": 88.21, "elapsed_time": "14:47:17", "remaining_time": "1:58:38"} +{"current_steps": 345, "total_steps": 390, "loss": 0.2399, "lr": 1.6004111268235156e-06, "epoch": 4.416, "percentage": 88.46, "elapsed_time": "14:49:47", "remaining_time": "1:56:03"} +{"current_steps": 346, "total_steps": 390, "loss": 0.2421, "lr": 1.5309838429526714e-06, "epoch": 4.4288, "percentage": 88.72, "elapsed_time": "14:52:19", "remaining_time": "1:53:28"} +{"current_steps": 347, "total_steps": 390, "loss": 0.2324, "lr": 1.4630360979351644e-06, "epoch": 4.4416, "percentage": 88.97, "elapsed_time": "14:54:50", "remaining_time": "1:50:53"} +{"current_steps": 348, "total_steps": 390, "loss": 0.2296, "lr": 1.396573335013236e-06, "epoch": 4.4544, "percentage": 89.23, "elapsed_time": "14:57:25", "remaining_time": "1:48:18"} +{"current_steps": 349, "total_steps": 390, "loss": 0.2455, "lr": 1.3316008784683265e-06, "epoch": 4.4672, "percentage": 89.49, "elapsed_time": "14:59:40", "remaining_time": "1:45:41"} +{"current_steps": 350, "total_steps": 390, "loss": 0.2392, "lr": 1.2681239331945695e-06, "epoch": 4.48, "percentage": 89.74, "elapsed_time": "15:02:23", "remaining_time": "1:43:07"} +{"current_steps": 351, "total_steps": 390, "loss": 0.2372, "lr": 1.2061475842818337e-06, "epoch": 4.4928, "percentage": 90.0, "elapsed_time": "15:04:56", "remaining_time": "1:40:32"} +{"current_steps": 352, "total_steps": 390, "loss": 0.2429, "lr": 1.1456767966083393e-06, "epoch": 4.5056, "percentage": 90.26, "elapsed_time": "15:07:23", "remaining_time": "1:37:57"} +{"current_steps": 353, "total_steps": 390, "loss": 0.2424, "lr": 1.086716414442952e-06, "epoch": 4.5184, "percentage": 90.51, "elapsed_time": "15:09:47", "remaining_time": "1:35:21"} +{"current_steps": 354, "total_steps": 390, "loss": 0.2405, "lr": 1.0292711610570904e-06, "epoch": 4.5312, "percentage": 90.77, "elapsed_time": "15:12:28", "remaining_time": "1:32:47"} +{"current_steps": 355, "total_steps": 390, "loss": 0.2403, "lr": 9.733456383463658e-07, "epoch": 4.5440000000000005, "percentage": 91.03, "elapsed_time": "15:15:05", "remaining_time": "1:30:13"} +{"current_steps": 356, "total_steps": 390, "loss": 0.2358, "lr": 9.189443264619102e-07, "epoch": 4.5568, "percentage": 91.28, "elapsed_time": "15:17:42", "remaining_time": "1:27:38"} +{"current_steps": 357, "total_steps": 390, "loss": 0.242, "lr": 8.660715834514977e-07, "epoch": 4.5696, "percentage": 91.54, "elapsed_time": "15:20:15", "remaining_time": "1:25:03"} +{"current_steps": 358, "total_steps": 390, "loss": 0.2353, "lr": 8.147316449103959e-07, "epoch": 4.5824, "percentage": 91.79, "elapsed_time": "15:22:44", "remaining_time": "1:22:28"} +{"current_steps": 359, "total_steps": 390, "loss": 0.2411, "lr": 7.649286236420806e-07, "epoch": 4.5952, "percentage": 92.05, "elapsed_time": "15:25:09", "remaining_time": "1:19:53"} +{"current_steps": 360, "total_steps": 390, "loss": 0.2324, "lr": 7.166665093287539e-07, "epoch": 4.608, "percentage": 92.31, "elapsed_time": "15:27:45", "remaining_time": "1:17:18"} +{"current_steps": 361, "total_steps": 390, "loss": 0.2437, "lr": 6.69949168211721e-07, "epoch": 4.6208, "percentage": 92.56, "elapsed_time": "15:30:07", "remaining_time": "1:14:43"} +{"current_steps": 362, "total_steps": 390, "loss": 0.2455, "lr": 6.247803427816945e-07, "epoch": 4.6336, "percentage": 92.82, "elapsed_time": "15:32:50", "remaining_time": "1:12:09"} +{"current_steps": 363, "total_steps": 390, "loss": 0.2258, "lr": 5.811636514789598e-07, "epoch": 4.6464, "percentage": 93.08, "elapsed_time": "15:35:25", "remaining_time": "1:09:34"} +{"current_steps": 364, "total_steps": 390, "loss": 0.239, "lr": 5.391025884035239e-07, "epoch": 4.6592, "percentage": 93.33, "elapsed_time": "15:37:54", "remaining_time": "1:06:59"} +{"current_steps": 365, "total_steps": 390, "loss": 0.2269, "lr": 4.986005230351954e-07, "epoch": 4.672, "percentage": 93.59, "elapsed_time": "15:40:15", "remaining_time": "1:04:24"} +{"current_steps": 366, "total_steps": 390, "loss": 0.2508, "lr": 4.5966069996365993e-07, "epoch": 4.6848, "percentage": 93.85, "elapsed_time": "15:42:47", "remaining_time": "1:01:49"} +{"current_steps": 367, "total_steps": 390, "loss": 0.2368, "lr": 4.22286238628562e-07, "epoch": 4.6975999999999996, "percentage": 94.1, "elapsed_time": "15:45:19", "remaining_time": "0:59:14"} +{"current_steps": 368, "total_steps": 390, "loss": 0.2477, "lr": 3.8648013306960664e-07, "epoch": 4.7104, "percentage": 94.36, "elapsed_time": "15:47:55", "remaining_time": "0:56:40"} +{"current_steps": 369, "total_steps": 390, "loss": 0.2469, "lr": 3.522452516867048e-07, "epoch": 4.7232, "percentage": 94.62, "elapsed_time": "15:50:30", "remaining_time": "0:54:05"} +{"current_steps": 370, "total_steps": 390, "loss": 0.2421, "lr": 3.1958433701019697e-07, "epoch": 4.736, "percentage": 94.87, "elapsed_time": "15:53:03", "remaining_time": "0:51:31"} +{"current_steps": 371, "total_steps": 390, "loss": 0.2441, "lr": 2.8850000548115155e-07, "epoch": 4.7488, "percentage": 95.13, "elapsed_time": "15:55:32", "remaining_time": "0:48:56"} +{"current_steps": 372, "total_steps": 390, "loss": 0.2423, "lr": 2.5899474724174313e-07, "epoch": 4.7616, "percentage": 95.38, "elapsed_time": "15:58:16", "remaining_time": "0:46:22"} +{"current_steps": 373, "total_steps": 390, "loss": 0.259, "lr": 2.3107092593579905e-07, "epoch": 4.7744, "percentage": 95.64, "elapsed_time": "16:00:39", "remaining_time": "0:43:47"} +{"current_steps": 374, "total_steps": 390, "loss": 0.2478, "lr": 2.0473077851942858e-07, "epoch": 4.7872, "percentage": 95.9, "elapsed_time": "16:03:25", "remaining_time": "0:41:12"} +{"current_steps": 375, "total_steps": 390, "loss": 0.2484, "lr": 1.799764150818306e-07, "epoch": 4.8, "percentage": 96.15, "elapsed_time": "16:06:08", "remaining_time": "0:38:38"} +{"current_steps": 376, "total_steps": 390, "loss": 0.2365, "lr": 1.5680981867625566e-07, "epoch": 4.8128, "percentage": 96.41, "elapsed_time": "16:08:42", "remaining_time": "0:36:04"} +{"current_steps": 377, "total_steps": 390, "loss": 0.2543, "lr": 1.3523284516113955e-07, "epoch": 4.8256, "percentage": 96.67, "elapsed_time": "16:11:20", "remaining_time": "0:33:29"} +{"current_steps": 378, "total_steps": 390, "loss": 0.2449, "lr": 1.1524722305144231e-07, "epoch": 4.8384, "percentage": 96.92, "elapsed_time": "16:14:05", "remaining_time": "0:30:55"} +{"current_steps": 379, "total_steps": 390, "loss": 0.2415, "lr": 9.685455338016347e-08, "epoch": 4.8512, "percentage": 97.18, "elapsed_time": "16:16:45", "remaining_time": "0:28:20"} +{"current_steps": 380, "total_steps": 390, "loss": 0.2429, "lr": 8.005630957010014e-08, "epoch": 4.864, "percentage": 97.44, "elapsed_time": "16:19:22", "remaining_time": "0:25:46"} +{"current_steps": 381, "total_steps": 390, "loss": 0.2505, "lr": 6.485383731580142e-08, "epoch": 4.8768, "percentage": 97.69, "elapsed_time": "16:21:56", "remaining_time": "0:23:11"} +{"current_steps": 382, "total_steps": 390, "loss": 0.234, "lr": 5.1248354475768034e-08, "epoch": 4.8896, "percentage": 97.95, "elapsed_time": "16:24:25", "remaining_time": "0:20:36"} +{"current_steps": 383, "total_steps": 390, "loss": 0.2415, "lr": 3.924095097489922e-08, "epoch": 4.9024, "percentage": 98.21, "elapsed_time": "16:27:06", "remaining_time": "0:18:02"} +{"current_steps": 384, "total_steps": 390, "loss": 0.2261, "lr": 2.8832588717164766e-08, "epoch": 4.9152000000000005, "percentage": 98.46, "elapsed_time": "16:29:35", "remaining_time": "0:15:27"} +{"current_steps": 385, "total_steps": 390, "loss": 0.2434, "lr": 2.0024101508555604e-08, "epoch": 4.928, "percentage": 98.72, "elapsed_time": "16:32:08", "remaining_time": "0:12:53"} +{"current_steps": 386, "total_steps": 390, "loss": 0.2454, "lr": 1.281619499029274e-08, "epoch": 4.9408, "percentage": 98.97, "elapsed_time": "16:34:56", "remaining_time": "0:10:18"} +{"current_steps": 387, "total_steps": 390, "loss": 0.2519, "lr": 7.209446582292501e-09, "epoch": 4.9536, "percentage": 99.23, "elapsed_time": "16:37:31", "remaining_time": "0:07:43"} +{"current_steps": 388, "total_steps": 390, "loss": 0.2388, "lr": 3.2043054369057523e-09, "epoch": 4.9664, "percentage": 99.49, "elapsed_time": "16:40:05", "remaining_time": "0:05:09"} +{"current_steps": 389, "total_steps": 390, "loss": 0.2355, "lr": 8.010924029533406e-10, "epoch": 4.9792, "percentage": 99.74, "elapsed_time": "16:42:39", "remaining_time": "0:02:34"} +{"current_steps": 390, "total_steps": 390, "loss": 0.2438, "lr": 0.0, "epoch": 4.992, "percentage": 100.0, "elapsed_time": "16:45:15", "remaining_time": "0:00:00"} +{"current_steps": 390, "total_steps": 390, "epoch": 4.992, "percentage": 100.0, "elapsed_time": "16:47:04", "remaining_time": "0:00:00"} diff --git a/trainer_state.json b/trainer_state.json new file mode 100644 index 0000000..540132a --- /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": 6.464605218515887, + "learning_rate": 1.0256410256410257e-06, + "loss": 0.8325, + "step": 1 + }, + { + "epoch": 0.0256, + "grad_norm": 6.380708677323359, + "learning_rate": 2.0512820512820513e-06, + "loss": 0.8289, + "step": 2 + }, + { + "epoch": 0.0384, + "grad_norm": 6.0732844198848, + "learning_rate": 3.0769230769230774e-06, + "loss": 0.8078, + "step": 3 + }, + { + "epoch": 0.0512, + "grad_norm": 5.719294948449857, + "learning_rate": 4.102564102564103e-06, + "loss": 0.8006, + "step": 4 + }, + { + "epoch": 0.064, + "grad_norm": 4.0447104367675575, + "learning_rate": 5.128205128205128e-06, + "loss": 0.7421, + "step": 5 + }, + { + "epoch": 0.0768, + "grad_norm": 2.3143768442428203, + "learning_rate": 6.153846153846155e-06, + "loss": 0.7318, + "step": 6 + }, + { + "epoch": 0.0896, + "grad_norm": 2.107989614101519, + "learning_rate": 7.17948717948718e-06, + "loss": 0.7258, + "step": 7 + }, + { + "epoch": 0.1024, + "grad_norm": 3.9720378553941766, + "learning_rate": 8.205128205128205e-06, + "loss": 0.7241, + "step": 8 + }, + { + "epoch": 0.1152, + "grad_norm": 4.270610484139302, + "learning_rate": 9.230769230769232e-06, + "loss": 0.7218, + "step": 9 + }, + { + "epoch": 0.128, + "grad_norm": 4.148966868142642, + "learning_rate": 1.0256410256410256e-05, + "loss": 0.6697, + "step": 10 + }, + { + "epoch": 0.1408, + "grad_norm": 3.7021017185511176, + "learning_rate": 1.1282051282051283e-05, + "loss": 0.6733, + "step": 11 + }, + { + "epoch": 0.1536, + "grad_norm": 2.142866919387794, + "learning_rate": 1.230769230769231e-05, + "loss": 0.6395, + "step": 12 + }, + { + "epoch": 0.1664, + "grad_norm": 1.707078933942607, + "learning_rate": 1.3333333333333333e-05, + "loss": 0.6393, + "step": 13 + }, + { + "epoch": 0.1792, + "grad_norm": 2.147594996913022, + "learning_rate": 1.435897435897436e-05, + "loss": 0.6228, + "step": 14 + }, + { + "epoch": 0.192, + "grad_norm": 1.5939068186916545, + "learning_rate": 1.5384615384615387e-05, + "loss": 0.5948, + "step": 15 + }, + { + "epoch": 0.2048, + "grad_norm": 1.0835969430668775, + "learning_rate": 1.641025641025641e-05, + "loss": 0.5971, + "step": 16 + }, + { + "epoch": 0.2176, + "grad_norm": 1.2234307086927172, + "learning_rate": 1.7435897435897438e-05, + "loss": 0.5931, + "step": 17 + }, + { + "epoch": 0.2304, + "grad_norm": 0.8928903139357364, + "learning_rate": 1.8461538461538465e-05, + "loss": 0.5712, + "step": 18 + }, + { + "epoch": 0.2432, + "grad_norm": 0.9673561701509722, + "learning_rate": 1.9487179487179488e-05, + "loss": 0.5674, + "step": 19 + }, + { + "epoch": 0.256, + "grad_norm": 0.8468004286837573, + "learning_rate": 2.0512820512820512e-05, + "loss": 0.5393, + "step": 20 + }, + { + "epoch": 0.2688, + "grad_norm": 0.8301118890092068, + "learning_rate": 2.153846153846154e-05, + "loss": 0.551, + "step": 21 + }, + { + "epoch": 0.2816, + "grad_norm": 0.9353297733467271, + "learning_rate": 2.2564102564102566e-05, + "loss": 0.5431, + "step": 22 + }, + { + "epoch": 0.2944, + "grad_norm": 0.7191992110388274, + "learning_rate": 2.3589743589743593e-05, + "loss": 0.531, + "step": 23 + }, + { + "epoch": 0.3072, + "grad_norm": 0.7987803101137527, + "learning_rate": 2.461538461538462e-05, + "loss": 0.537, + "step": 24 + }, + { + "epoch": 0.32, + "grad_norm": 0.9000823684030751, + "learning_rate": 2.5641025641025646e-05, + "loss": 0.5243, + "step": 25 + }, + { + "epoch": 0.3328, + "grad_norm": 0.639728529078365, + "learning_rate": 2.6666666666666667e-05, + "loss": 0.5256, + "step": 26 + }, + { + "epoch": 0.3456, + "grad_norm": 0.7310823848630157, + "learning_rate": 2.7692307692307694e-05, + "loss": 0.527, + "step": 27 + }, + { + "epoch": 0.3584, + "grad_norm": 0.7617952383130974, + "learning_rate": 2.871794871794872e-05, + "loss": 0.5243, + "step": 28 + }, + { + "epoch": 0.3712, + "grad_norm": 0.6087182615528468, + "learning_rate": 2.9743589743589747e-05, + "loss": 0.5213, + "step": 29 + }, + { + "epoch": 0.384, + "grad_norm": 0.6138496664446432, + "learning_rate": 3.0769230769230774e-05, + "loss": 0.5113, + "step": 30 + }, + { + "epoch": 0.3968, + "grad_norm": 0.5842736668905947, + "learning_rate": 3.1794871794871795e-05, + "loss": 0.5088, + "step": 31 + }, + { + "epoch": 0.4096, + "grad_norm": 0.6842984878708783, + "learning_rate": 3.282051282051282e-05, + "loss": 0.5194, + "step": 32 + }, + { + "epoch": 0.4224, + "grad_norm": 0.7772980067344607, + "learning_rate": 3.384615384615385e-05, + "loss": 0.5193, + "step": 33 + }, + { + "epoch": 0.4352, + "grad_norm": 0.6889382493737749, + "learning_rate": 3.4871794871794875e-05, + "loss": 0.5039, + "step": 34 + }, + { + "epoch": 0.448, + "grad_norm": 5.907781646563153, + "learning_rate": 3.58974358974359e-05, + "loss": 0.5234, + "step": 35 + }, + { + "epoch": 0.4608, + "grad_norm": 1.833998012641166, + "learning_rate": 3.692307692307693e-05, + "loss": 0.5138, + "step": 36 + }, + { + "epoch": 0.4736, + "grad_norm": 1.358088445040064, + "learning_rate": 3.794871794871795e-05, + "loss": 0.5164, + "step": 37 + }, + { + "epoch": 0.4864, + "grad_norm": 0.9665299887078758, + "learning_rate": 3.8974358974358976e-05, + "loss": 0.5086, + "step": 38 + }, + { + "epoch": 0.4992, + "grad_norm": 1.8462558706888752, + "learning_rate": 4e-05, + "loss": 0.5108, + "step": 39 + }, + { + "epoch": 0.512, + "grad_norm": 1.0717311224964012, + "learning_rate": 3.9999198907597046e-05, + "loss": 0.4799, + "step": 40 + }, + { + "epoch": 0.5248, + "grad_norm": 2.1436041861152764, + "learning_rate": 3.9996795694563096e-05, + "loss": 0.5216, + "step": 41 + }, + { + "epoch": 0.5376, + "grad_norm": 1.8190734353201568, + "learning_rate": 3.999279055341771e-05, + "loss": 0.4935, + "step": 42 + }, + { + "epoch": 0.5504, + "grad_norm": 1.3379628057404362, + "learning_rate": 3.998718380500971e-05, + "loss": 0.5215, + "step": 43 + }, + { + "epoch": 0.5632, + "grad_norm": 1.7052354362588542, + "learning_rate": 3.997997589849145e-05, + "loss": 0.5134, + "step": 44 + }, + { + "epoch": 0.576, + "grad_norm": 1.0266630844521263, + "learning_rate": 3.9971167411282835e-05, + "loss": 0.4923, + "step": 45 + }, + { + "epoch": 0.5888, + "grad_norm": 1.3058420762275704, + "learning_rate": 3.99607590490251e-05, + "loss": 0.4989, + "step": 46 + }, + { + "epoch": 0.6016, + "grad_norm": 1.0731875195296174, + "learning_rate": 3.9948751645524235e-05, + "loss": 0.4942, + "step": 47 + }, + { + "epoch": 0.6144, + "grad_norm": 0.8898272327272633, + "learning_rate": 3.9935146162684206e-05, + "loss": 0.4992, + "step": 48 + }, + { + "epoch": 0.6272, + "grad_norm": 1.1040031356310187, + "learning_rate": 3.9919943690429906e-05, + "loss": 0.5025, + "step": 49 + }, + { + "epoch": 0.64, + "grad_norm": 0.8174445533574496, + "learning_rate": 3.9903145446619837e-05, + "loss": 0.4828, + "step": 50 + }, + { + "epoch": 0.6528, + "grad_norm": 0.8673864554018242, + "learning_rate": 3.9884752776948564e-05, + "loss": 0.4791, + "step": 51 + }, + { + "epoch": 0.6656, + "grad_norm": 0.796148569165424, + "learning_rate": 3.9864767154838864e-05, + "loss": 0.4788, + "step": 52 + }, + { + "epoch": 0.6784, + "grad_norm": 0.6568989662572859, + "learning_rate": 3.9843190181323744e-05, + "loss": 0.4821, + "step": 53 + }, + { + "epoch": 0.6912, + "grad_norm": 0.8104547268528312, + "learning_rate": 3.982002358491817e-05, + "loss": 0.4838, + "step": 54 + }, + { + "epoch": 0.704, + "grad_norm": 0.6205373039374902, + "learning_rate": 3.979526922148058e-05, + "loss": 0.4977, + "step": 55 + }, + { + "epoch": 0.7168, + "grad_norm": 0.7152522003873999, + "learning_rate": 3.9768929074064206e-05, + "loss": 0.476, + "step": 56 + }, + { + "epoch": 0.7296, + "grad_norm": 0.5502148010990727, + "learning_rate": 3.9741005252758255e-05, + "loss": 0.4851, + "step": 57 + }, + { + "epoch": 0.7424, + "grad_norm": 0.653670757127795, + "learning_rate": 3.971149999451886e-05, + "loss": 0.4758, + "step": 58 + }, + { + "epoch": 0.7552, + "grad_norm": 0.6467489669688496, + "learning_rate": 3.9680415662989806e-05, + "loss": 0.4708, + "step": 59 + }, + { + "epoch": 0.768, + "grad_norm": 0.7583281936365043, + "learning_rate": 3.9647754748313294e-05, + "loss": 0.4745, + "step": 60 + }, + { + "epoch": 0.7808, + "grad_norm": 0.638240853677551, + "learning_rate": 3.96135198669304e-05, + "loss": 0.4855, + "step": 61 + }, + { + "epoch": 0.7936, + "grad_norm": 0.5490847741875153, + "learning_rate": 3.957771376137144e-05, + "loss": 0.4616, + "step": 62 + }, + { + "epoch": 0.8064, + "grad_norm": 0.5887261961332173, + "learning_rate": 3.954033930003634e-05, + "loss": 0.4751, + "step": 63 + }, + { + "epoch": 0.8192, + "grad_norm": 0.5124953087552118, + "learning_rate": 3.9501399476964806e-05, + "loss": 0.4767, + "step": 64 + }, + { + "epoch": 0.832, + "grad_norm": 0.59821449783325, + "learning_rate": 3.946089741159648e-05, + "loss": 0.474, + "step": 65 + }, + { + "epoch": 0.8448, + "grad_norm": 0.6120167845677714, + "learning_rate": 3.9418836348521045e-05, + "loss": 0.4749, + "step": 66 + }, + { + "epoch": 0.8576, + "grad_norm": 0.5235563985592369, + "learning_rate": 3.937521965721831e-05, + "loss": 0.4748, + "step": 67 + }, + { + "epoch": 0.8704, + "grad_norm": 0.5350857351039692, + "learning_rate": 3.933005083178828e-05, + "loss": 0.4673, + "step": 68 + }, + { + "epoch": 0.8832, + "grad_norm": 0.6142423277054795, + "learning_rate": 3.928333349067125e-05, + "loss": 0.4767, + "step": 69 + }, + { + "epoch": 0.896, + "grad_norm": 0.590492201343943, + "learning_rate": 3.923507137635792e-05, + "loss": 0.4679, + "step": 70 + }, + { + "epoch": 0.9088, + "grad_norm": 0.47391904061487716, + "learning_rate": 3.9185268355089606e-05, + "loss": 0.465, + "step": 71 + }, + { + "epoch": 0.9216, + "grad_norm": 0.5617478093615034, + "learning_rate": 3.913392841654851e-05, + "loss": 0.4735, + "step": 72 + }, + { + "epoch": 0.9344, + "grad_norm": 0.47206063489404443, + "learning_rate": 3.9081055673538093e-05, + "loss": 0.4697, + "step": 73 + }, + { + "epoch": 0.9472, + "grad_norm": 0.5078461387459143, + "learning_rate": 3.902665436165364e-05, + "loss": 0.4752, + "step": 74 + }, + { + "epoch": 0.96, + "grad_norm": 0.5108689431705631, + "learning_rate": 3.897072883894291e-05, + "loss": 0.4484, + "step": 75 + }, + { + "epoch": 0.9728, + "grad_norm": 0.5288717566865853, + "learning_rate": 3.8913283585557054e-05, + "loss": 0.4584, + "step": 76 + }, + { + "epoch": 0.9856, + "grad_norm": 0.5106573654152327, + "learning_rate": 3.885432320339167e-05, + "loss": 0.4695, + "step": 77 + }, + { + "epoch": 0.9984, + "grad_norm": 0.4978461545439326, + "learning_rate": 3.879385241571817e-05, + "loss": 0.4659, + "step": 78 + }, + { + "epoch": 1.0112, + "grad_norm": 0.5374431536494356, + "learning_rate": 3.873187606680543e-05, + "loss": 0.4383, + "step": 79 + }, + { + "epoch": 1.024, + "grad_norm": 0.466905661747252, + "learning_rate": 3.866839912153168e-05, + "loss": 0.4235, + "step": 80 + }, + { + "epoch": 1.0368, + "grad_norm": 0.5524070933573034, + "learning_rate": 3.860342666498677e-05, + "loss": 0.426, + "step": 81 + }, + { + "epoch": 1.0496, + "grad_norm": 0.5552329501082961, + "learning_rate": 3.853696390206484e-05, + "loss": 0.4293, + "step": 82 + }, + { + "epoch": 1.0624, + "grad_norm": 0.6365882933312345, + "learning_rate": 3.846901615704734e-05, + "loss": 0.4277, + "step": 83 + }, + { + "epoch": 1.0752, + "grad_norm": 0.6792621794760965, + "learning_rate": 3.839958887317649e-05, + "loss": 0.4136, + "step": 84 + }, + { + "epoch": 1.088, + "grad_norm": 0.7086948729956156, + "learning_rate": 3.832868761221926e-05, + "loss": 0.4203, + "step": 85 + }, + { + "epoch": 1.1008, + "grad_norm": 0.5914674949977854, + "learning_rate": 3.825631805402182e-05, + "loss": 0.4133, + "step": 86 + }, + { + "epoch": 1.1136, + "grad_norm": 0.6457869945763296, + "learning_rate": 3.818248599605448e-05, + "loss": 0.4189, + "step": 87 + }, + { + "epoch": 1.1264, + "grad_norm": 0.7794506477481747, + "learning_rate": 3.810719735294731e-05, + "loss": 0.4319, + "step": 88 + }, + { + "epoch": 1.1392, + "grad_norm": 0.6157599617077694, + "learning_rate": 3.8030458156016326e-05, + "loss": 0.4138, + "step": 89 + }, + { + "epoch": 1.152, + "grad_norm": 0.5668112614414113, + "learning_rate": 3.795227455278029e-05, + "loss": 0.4087, + "step": 90 + }, + { + "epoch": 1.1648, + "grad_norm": 0.7534233323894747, + "learning_rate": 3.787265280646825e-05, + "loss": 0.4158, + "step": 91 + }, + { + "epoch": 1.1776, + "grad_norm": 0.8557762634468793, + "learning_rate": 3.7791599295517825e-05, + "loss": 0.4163, + "step": 92 + }, + { + "epoch": 1.1904, + "grad_norm": 0.5514378853304944, + "learning_rate": 3.7709120513064196e-05, + "loss": 0.4166, + "step": 93 + }, + { + "epoch": 1.2032, + "grad_norm": 0.6176747491282486, + "learning_rate": 3.762522306641998e-05, + "loss": 0.4103, + "step": 94 + }, + { + "epoch": 1.216, + "grad_norm": 0.7020404535006932, + "learning_rate": 3.7539913676545874e-05, + "loss": 0.3998, + "step": 95 + }, + { + "epoch": 1.2288000000000001, + "grad_norm": 0.36522429007528184, + "learning_rate": 3.745319917751229e-05, + "loss": 0.4097, + "step": 96 + }, + { + "epoch": 1.2416, + "grad_norm": 0.5900634589160236, + "learning_rate": 3.736508651595188e-05, + "loss": 0.4228, + "step": 97 + }, + { + "epoch": 1.2544, + "grad_norm": 0.4613833081509036, + "learning_rate": 3.727558275050301e-05, + "loss": 0.4124, + "step": 98 + }, + { + "epoch": 1.2671999999999999, + "grad_norm": 0.4846729996977901, + "learning_rate": 3.718469505124434e-05, + "loss": 0.404, + "step": 99 + }, + { + "epoch": 1.28, + "grad_norm": 0.4683152886505894, + "learning_rate": 3.709243069912041e-05, + "loss": 0.4187, + "step": 100 + }, + { + "epoch": 1.2928, + "grad_norm": 0.3717197380076313, + "learning_rate": 3.699879708535838e-05, + "loss": 0.4139, + "step": 101 + }, + { + "epoch": 1.3056, + "grad_norm": 0.4302258236247511, + "learning_rate": 3.69038017108759e-05, + "loss": 0.4159, + "step": 102 + }, + { + "epoch": 1.3184, + "grad_norm": 0.3982962645464779, + "learning_rate": 3.680745218568026e-05, + "loss": 0.4181, + "step": 103 + }, + { + "epoch": 1.3312, + "grad_norm": 0.40200214235666887, + "learning_rate": 3.6709756228258735e-05, + "loss": 0.4046, + "step": 104 + }, + { + "epoch": 1.3439999999999999, + "grad_norm": 0.44402430474888865, + "learning_rate": 3.6610721664960236e-05, + "loss": 0.4263, + "step": 105 + }, + { + "epoch": 1.3568, + "grad_norm": 0.38820837732347074, + "learning_rate": 3.65103564293684e-05, + "loss": 0.417, + "step": 106 + }, + { + "epoch": 1.3696, + "grad_norm": 0.5213458360363288, + "learning_rate": 3.640866856166601e-05, + "loss": 0.426, + "step": 107 + }, + { + "epoch": 1.3824, + "grad_norm": 0.4308788707291437, + "learning_rate": 3.6305666207990886e-05, + "loss": 0.4097, + "step": 108 + }, + { + "epoch": 1.3952, + "grad_norm": 0.4644795634740679, + "learning_rate": 3.6201357619783336e-05, + "loss": 0.4068, + "step": 109 + }, + { + "epoch": 1.408, + "grad_norm": 0.44267654288374053, + "learning_rate": 3.609575115312511e-05, + "loss": 0.415, + "step": 110 + }, + { + "epoch": 1.4208, + "grad_norm": 0.4714173326270217, + "learning_rate": 3.598885526807003e-05, + "loss": 0.4101, + "step": 111 + }, + { + "epoch": 1.4336, + "grad_norm": 0.4805820675277153, + "learning_rate": 3.5880678527966224e-05, + "loss": 0.4213, + "step": 112 + }, + { + "epoch": 1.4464000000000001, + "grad_norm": 0.39641366601935735, + "learning_rate": 3.577122959877017e-05, + "loss": 0.3989, + "step": 113 + }, + { + "epoch": 1.4592, + "grad_norm": 0.3759641748474472, + "learning_rate": 3.566051724835245e-05, + "loss": 0.416, + "step": 114 + }, + { + "epoch": 1.472, + "grad_norm": 0.3694282519820903, + "learning_rate": 3.554855034579532e-05, + "loss": 0.4225, + "step": 115 + }, + { + "epoch": 1.4848, + "grad_norm": 0.4024973231579404, + "learning_rate": 3.5435337860682304e-05, + "loss": 0.4153, + "step": 116 + }, + { + "epoch": 1.4976, + "grad_norm": 0.37364740883863296, + "learning_rate": 3.532088886237956e-05, + "loss": 0.4108, + "step": 117 + }, + { + "epoch": 1.5104, + "grad_norm": 0.36906546048784605, + "learning_rate": 3.520521251930941e-05, + "loss": 0.4124, + "step": 118 + }, + { + "epoch": 1.5232, + "grad_norm": 0.3748555038416534, + "learning_rate": 3.5088318098215805e-05, + "loss": 0.4081, + "step": 119 + }, + { + "epoch": 1.536, + "grad_norm": 0.3582549157315094, + "learning_rate": 3.497021496342203e-05, + "loss": 0.4018, + "step": 120 + }, + { + "epoch": 1.5488, + "grad_norm": 0.3887317377098141, + "learning_rate": 3.485091257608047e-05, + "loss": 0.3943, + "step": 121 + }, + { + "epoch": 1.5615999999999999, + "grad_norm": 0.4585279897491825, + "learning_rate": 3.473042049341474e-05, + "loss": 0.4001, + "step": 122 + }, + { + "epoch": 1.5744, + "grad_norm": 0.37694477337454396, + "learning_rate": 3.4608748367954064e-05, + "loss": 0.4002, + "step": 123 + }, + { + "epoch": 1.5872000000000002, + "grad_norm": 0.39032755502498523, + "learning_rate": 3.4485905946759965e-05, + "loss": 0.4078, + "step": 124 + }, + { + "epoch": 1.6, + "grad_norm": 0.4322350994624392, + "learning_rate": 3.4361903070645484e-05, + "loss": 0.4025, + "step": 125 + }, + { + "epoch": 1.6128, + "grad_norm": 0.39632556623882664, + "learning_rate": 3.423674967338681e-05, + "loss": 0.4062, + "step": 126 + }, + { + "epoch": 1.6256, + "grad_norm": 0.41948444365675525, + "learning_rate": 3.411045578092754e-05, + "loss": 0.4177, + "step": 127 + }, + { + "epoch": 1.6383999999999999, + "grad_norm": 0.4799383655202944, + "learning_rate": 3.398303151057543e-05, + "loss": 0.4166, + "step": 128 + }, + { + "epoch": 1.6512, + "grad_norm": 0.47130088781959734, + "learning_rate": 3.385448707019199e-05, + "loss": 0.4095, + "step": 129 + }, + { + "epoch": 1.6640000000000001, + "grad_norm": 0.38829379832198224, + "learning_rate": 3.372483275737468e-05, + "loss": 0.4088, + "step": 130 + }, + { + "epoch": 1.6768, + "grad_norm": 0.48425044375595544, + "learning_rate": 3.359407895863199e-05, + "loss": 0.4203, + "step": 131 + }, + { + "epoch": 1.6896, + "grad_norm": 0.5022610534269328, + "learning_rate": 3.34622361485514e-05, + "loss": 0.4049, + "step": 132 + }, + { + "epoch": 1.7024, + "grad_norm": 0.43114127314574263, + "learning_rate": 3.332931488896029e-05, + "loss": 0.4093, + "step": 133 + }, + { + "epoch": 1.7151999999999998, + "grad_norm": 0.5129917681391781, + "learning_rate": 3.319532582807977e-05, + "loss": 0.4161, + "step": 134 + }, + { + "epoch": 1.728, + "grad_norm": 0.4853215798673184, + "learning_rate": 3.30602796996717e-05, + "loss": 0.4142, + "step": 135 + }, + { + "epoch": 1.7408000000000001, + "grad_norm": 0.5395589871823466, + "learning_rate": 3.2924187322178865e-05, + "loss": 0.4162, + "step": 136 + }, + { + "epoch": 1.7536, + "grad_norm": 0.40618095402028487, + "learning_rate": 3.278705959785821e-05, + "loss": 0.4251, + "step": 137 + }, + { + "epoch": 1.7664, + "grad_norm": 0.4746670426025735, + "learning_rate": 3.2648907511907544e-05, + "loss": 0.4018, + "step": 138 + }, + { + "epoch": 1.7792, + "grad_norm": 0.38240123980896445, + "learning_rate": 3.250974213158555e-05, + "loss": 0.4096, + "step": 139 + }, + { + "epoch": 1.792, + "grad_norm": 0.4587704974203193, + "learning_rate": 3.23695746053251e-05, + "loss": 0.4142, + "step": 140 + }, + { + "epoch": 1.8048, + "grad_norm": 0.4138827354677887, + "learning_rate": 3.222841616184025e-05, + "loss": 0.418, + "step": 141 + }, + { + "epoch": 1.8176, + "grad_norm": 0.45435877514938233, + "learning_rate": 3.208627810922665e-05, + "loss": 0.4028, + "step": 142 + }, + { + "epoch": 1.8304, + "grad_norm": 0.4647800854837698, + "learning_rate": 3.194317183405573e-05, + "loss": 0.4132, + "step": 143 + }, + { + "epoch": 1.8432, + "grad_norm": 0.36420373665662603, + "learning_rate": 3.1799108800462466e-05, + "loss": 0.4031, + "step": 144 + }, + { + "epoch": 1.8559999999999999, + "grad_norm": 0.4666313110580902, + "learning_rate": 3.1654100549227024e-05, + "loss": 0.4162, + "step": 145 + }, + { + "epoch": 1.8688, + "grad_norm": 0.39777290497002804, + "learning_rate": 3.1508158696850275e-05, + "loss": 0.4015, + "step": 146 + }, + { + "epoch": 1.8816000000000002, + "grad_norm": 0.41964971235728266, + "learning_rate": 3.136129493462312e-05, + "loss": 0.4051, + "step": 147 + }, + { + "epoch": 1.8944, + "grad_norm": 0.42487362870525114, + "learning_rate": 3.121352102768998e-05, + "loss": 0.4121, + "step": 148 + }, + { + "epoch": 1.9072, + "grad_norm": 0.34253627904746686, + "learning_rate": 3.106484881410628e-05, + "loss": 0.4126, + "step": 149 + }, + { + "epoch": 1.92, + "grad_norm": 0.3822674020304617, + "learning_rate": 3.091529020389009e-05, + "loss": 0.4131, + "step": 150 + }, + { + "epoch": 1.9327999999999999, + "grad_norm": 0.40435918024987844, + "learning_rate": 3.076485717806808e-05, + "loss": 0.4148, + "step": 151 + }, + { + "epoch": 1.9456, + "grad_norm": 0.3499098381687515, + "learning_rate": 3.061356178771564e-05, + "loss": 0.3936, + "step": 152 + }, + { + "epoch": 1.9584000000000001, + "grad_norm": 0.4076701390830936, + "learning_rate": 3.0461416152991555e-05, + "loss": 0.4075, + "step": 153 + }, + { + "epoch": 1.9712, + "grad_norm": 0.40087378483854863, + "learning_rate": 3.0308432462167045e-05, + "loss": 0.4057, + "step": 154 + }, + { + "epoch": 1.984, + "grad_norm": 0.41603083979984834, + "learning_rate": 3.015462297064936e-05, + "loss": 0.4026, + "step": 155 + }, + { + "epoch": 1.9968, + "grad_norm": 0.3504122558670022, + "learning_rate": 3.0000000000000004e-05, + "loss": 0.3867, + "step": 156 + }, + { + "epoch": 2.0096, + "grad_norm": 0.5272796148664116, + "learning_rate": 2.98445759369477e-05, + "loss": 0.3827, + "step": 157 + }, + { + "epoch": 2.0224, + "grad_norm": 0.44531053711387253, + "learning_rate": 2.9688363232396056e-05, + "loss": 0.3576, + "step": 158 + }, + { + "epoch": 2.0352, + "grad_norm": 0.5959308354780575, + "learning_rate": 2.9531374400426158e-05, + "loss": 0.358, + "step": 159 + }, + { + "epoch": 2.048, + "grad_norm": 0.5693188685590685, + "learning_rate": 2.9373622017294075e-05, + "loss": 0.3447, + "step": 160 + }, + { + "epoch": 2.0608, + "grad_norm": 0.5042613043316267, + "learning_rate": 2.9215118720423375e-05, + "loss": 0.3417, + "step": 161 + }, + { + "epoch": 2.0736, + "grad_norm": 0.553051231213321, + "learning_rate": 2.9055877207392752e-05, + "loss": 0.3474, + "step": 162 + }, + { + "epoch": 2.0864, + "grad_norm": 0.4557734083518579, + "learning_rate": 2.8895910234918828e-05, + "loss": 0.3546, + "step": 163 + }, + { + "epoch": 2.0992, + "grad_norm": 0.5416631636159364, + "learning_rate": 2.873523061783426e-05, + "loss": 0.3483, + "step": 164 + }, + { + "epoch": 2.112, + "grad_norm": 0.5389527047734003, + "learning_rate": 2.8573851228061084e-05, + "loss": 0.341, + "step": 165 + }, + { + "epoch": 2.1248, + "grad_norm": 0.4105182765148673, + "learning_rate": 2.8411784993579633e-05, + "loss": 0.3389, + "step": 166 + }, + { + "epoch": 2.1376, + "grad_norm": 0.56488782807409, + "learning_rate": 2.8249044897392814e-05, + "loss": 0.3473, + "step": 167 + }, + { + "epoch": 2.1504, + "grad_norm": 0.3973328446053959, + "learning_rate": 2.80856439764861e-05, + "loss": 0.3393, + "step": 168 + }, + { + "epoch": 2.1632, + "grad_norm": 0.48655927946324257, + "learning_rate": 2.792159532078314e-05, + "loss": 0.3409, + "step": 169 + }, + { + "epoch": 2.176, + "grad_norm": 0.3562939309608415, + "learning_rate": 2.77569120720971e-05, + "loss": 0.3339, + "step": 170 + }, + { + "epoch": 2.1888, + "grad_norm": 0.3865349086129511, + "learning_rate": 2.7591607423077932e-05, + "loss": 0.3441, + "step": 171 + }, + { + "epoch": 2.2016, + "grad_norm": 0.38384389039352235, + "learning_rate": 2.7425694616155474e-05, + "loss": 0.3413, + "step": 172 + }, + { + "epoch": 2.2144, + "grad_norm": 0.32530141741650587, + "learning_rate": 2.7259186942478656e-05, + "loss": 0.3455, + "step": 173 + }, + { + "epoch": 2.2272, + "grad_norm": 0.3686673261248661, + "learning_rate": 2.7092097740850712e-05, + "loss": 0.3481, + "step": 174 + }, + { + "epoch": 2.24, + "grad_norm": 0.3143044879080098, + "learning_rate": 2.692444039666066e-05, + "loss": 0.3337, + "step": 175 + }, + { + "epoch": 2.2528, + "grad_norm": 0.31619399123559355, + "learning_rate": 2.6756228340810946e-05, + "loss": 0.3423, + "step": 176 + }, + { + "epoch": 2.2656, + "grad_norm": 0.3373730218823642, + "learning_rate": 2.6587475048641596e-05, + "loss": 0.3466, + "step": 177 + }, + { + "epoch": 2.2784, + "grad_norm": 0.3409092059873659, + "learning_rate": 2.6418194038850634e-05, + "loss": 0.3401, + "step": 178 + }, + { + "epoch": 2.2912, + "grad_norm": 0.3262535081367611, + "learning_rate": 2.624839887241115e-05, + "loss": 0.3355, + "step": 179 + }, + { + "epoch": 2.304, + "grad_norm": 0.3598189947105674, + "learning_rate": 2.607810315148494e-05, + "loss": 0.3384, + "step": 180 + }, + { + "epoch": 2.3168, + "grad_norm": 0.3082225835028636, + "learning_rate": 2.5907320518332827e-05, + "loss": 0.3434, + "step": 181 + }, + { + "epoch": 2.3296, + "grad_norm": 0.37794853145832014, + "learning_rate": 2.5736064654221808e-05, + "loss": 0.347, + "step": 182 + }, + { + "epoch": 2.3424, + "grad_norm": 0.3441245055128099, + "learning_rate": 2.5564349278329056e-05, + "loss": 0.3434, + "step": 183 + }, + { + "epoch": 2.3552, + "grad_norm": 0.34812376784545485, + "learning_rate": 2.539218814664288e-05, + "loss": 0.336, + "step": 184 + }, + { + "epoch": 2.368, + "grad_norm": 0.3611431416358283, + "learning_rate": 2.521959505086075e-05, + "loss": 0.3431, + "step": 185 + }, + { + "epoch": 2.3808, + "grad_norm": 0.4222133844001757, + "learning_rate": 2.5046583817284437e-05, + "loss": 0.3373, + "step": 186 + }, + { + "epoch": 2.3936, + "grad_norm": 0.31914592506777995, + "learning_rate": 2.487316830571244e-05, + "loss": 0.3389, + "step": 187 + }, + { + "epoch": 2.4064, + "grad_norm": 0.31451553979055025, + "learning_rate": 2.4699362408329646e-05, + "loss": 0.3303, + "step": 188 + }, + { + "epoch": 2.4192, + "grad_norm": 0.37003768460022973, + "learning_rate": 2.4525180048594452e-05, + "loss": 0.3502, + "step": 189 + }, + { + "epoch": 2.432, + "grad_norm": 0.3239873799914562, + "learning_rate": 2.435063518012335e-05, + "loss": 0.342, + "step": 190 + }, + { + "epoch": 2.4448, + "grad_norm": 0.31332443912294144, + "learning_rate": 2.4175741785573177e-05, + "loss": 0.3486, + "step": 191 + }, + { + "epoch": 2.4576000000000002, + "grad_norm": 0.3394957631642956, + "learning_rate": 2.4000513875520892e-05, + "loss": 0.339, + "step": 192 + }, + { + "epoch": 2.4704, + "grad_norm": 0.30179131299529555, + "learning_rate": 2.3824965487341247e-05, + "loss": 0.3346, + "step": 193 + }, + { + "epoch": 2.4832, + "grad_norm": 0.311970573179333, + "learning_rate": 2.3649110684082258e-05, + "loss": 0.3413, + "step": 194 + }, + { + "epoch": 2.496, + "grad_norm": 0.30637190878979437, + "learning_rate": 2.3472963553338614e-05, + "loss": 0.3415, + "step": 195 + }, + { + "epoch": 2.5088, + "grad_norm": 0.3262722361133359, + "learning_rate": 2.3296538206123134e-05, + "loss": 0.3424, + "step": 196 + }, + { + "epoch": 2.5216, + "grad_norm": 0.30807470653995994, + "learning_rate": 2.311984877573636e-05, + "loss": 0.3308, + "step": 197 + }, + { + "epoch": 2.5343999999999998, + "grad_norm": 0.334927287671144, + "learning_rate": 2.2942909416634326e-05, + "loss": 0.3457, + "step": 198 + }, + { + "epoch": 2.5472, + "grad_norm": 0.28967982177015333, + "learning_rate": 2.2765734303294666e-05, + "loss": 0.3428, + "step": 199 + }, + { + "epoch": 2.56, + "grad_norm": 0.36898835911684036, + "learning_rate": 2.2588337629081107e-05, + "loss": 0.3406, + "step": 200 + }, + { + "epoch": 2.5728, + "grad_norm": 0.28145488084817705, + "learning_rate": 2.2410733605106462e-05, + "loss": 0.3452, + "step": 201 + }, + { + "epoch": 2.5856, + "grad_norm": 0.3941864938083129, + "learning_rate": 2.2232936459094158e-05, + "loss": 0.3399, + "step": 202 + }, + { + "epoch": 2.5984, + "grad_norm": 0.31259268252215316, + "learning_rate": 2.205496043423849e-05, + "loss": 0.3355, + "step": 203 + }, + { + "epoch": 2.6112, + "grad_norm": 0.37615122244959603, + "learning_rate": 2.1876819788063586e-05, + "loss": 0.3394, + "step": 204 + }, + { + "epoch": 2.624, + "grad_norm": 0.2940297985807267, + "learning_rate": 2.16985287912813e-05, + "loss": 0.3458, + "step": 205 + }, + { + "epoch": 2.6368, + "grad_norm": 0.4070811437427959, + "learning_rate": 2.1520101726647922e-05, + "loss": 0.3427, + "step": 206 + }, + { + "epoch": 2.6496, + "grad_norm": 0.32272328892363317, + "learning_rate": 2.1341552887820048e-05, + "loss": 0.341, + "step": 207 + }, + { + "epoch": 2.6624, + "grad_norm": 0.4474845128230661, + "learning_rate": 2.1162896578209517e-05, + "loss": 0.3451, + "step": 208 + }, + { + "epoch": 2.6752000000000002, + "grad_norm": 0.3499793256711866, + "learning_rate": 2.0984147109837564e-05, + "loss": 0.341, + "step": 209 + }, + { + "epoch": 2.6879999999999997, + "grad_norm": 0.39384147774941897, + "learning_rate": 2.0805318802188307e-05, + "loss": 0.3499, + "step": 210 + }, + { + "epoch": 2.7008, + "grad_norm": 0.35193869889879736, + "learning_rate": 2.0626425981061608e-05, + "loss": 0.337, + "step": 211 + }, + { + "epoch": 2.7136, + "grad_norm": 0.3726248118838429, + "learning_rate": 2.0447482977425465e-05, + "loss": 0.3364, + "step": 212 + }, + { + "epoch": 2.7264, + "grad_norm": 0.2919938332617151, + "learning_rate": 2.0268504126267952e-05, + "loss": 0.339, + "step": 213 + }, + { + "epoch": 2.7392, + "grad_norm": 0.3339209869689389, + "learning_rate": 2.008950376544887e-05, + "loss": 0.3425, + "step": 214 + }, + { + "epoch": 2.752, + "grad_norm": 0.29640305737764455, + "learning_rate": 1.9910496234551132e-05, + "loss": 0.3402, + "step": 215 + }, + { + "epoch": 2.7648, + "grad_norm": 0.30254409880624555, + "learning_rate": 1.9731495873732055e-05, + "loss": 0.3531, + "step": 216 + }, + { + "epoch": 2.7776, + "grad_norm": 0.30911087112319563, + "learning_rate": 1.9552517022574542e-05, + "loss": 0.343, + "step": 217 + }, + { + "epoch": 2.7904, + "grad_norm": 0.2563063478891904, + "learning_rate": 1.93735740189384e-05, + "loss": 0.3449, + "step": 218 + }, + { + "epoch": 2.8032, + "grad_norm": 0.2705467137959849, + "learning_rate": 1.9194681197811703e-05, + "loss": 0.3569, + "step": 219 + }, + { + "epoch": 2.816, + "grad_norm": 0.2682899014111958, + "learning_rate": 1.901585289016244e-05, + "loss": 0.3338, + "step": 220 + }, + { + "epoch": 2.8288, + "grad_norm": 0.27996514808139306, + "learning_rate": 1.8837103421790486e-05, + "loss": 0.3316, + "step": 221 + }, + { + "epoch": 2.8416, + "grad_norm": 0.25978948174639804, + "learning_rate": 1.8658447112179952e-05, + "loss": 0.3326, + "step": 222 + }, + { + "epoch": 2.8544, + "grad_norm": 0.2564459029629428, + "learning_rate": 1.8479898273352084e-05, + "loss": 0.3269, + "step": 223 + }, + { + "epoch": 2.8672, + "grad_norm": 0.2854711673452058, + "learning_rate": 1.83014712087187e-05, + "loss": 0.3437, + "step": 224 + }, + { + "epoch": 2.88, + "grad_norm": 0.27320772787317565, + "learning_rate": 1.8123180211936417e-05, + "loss": 0.3317, + "step": 225 + }, + { + "epoch": 2.8928000000000003, + "grad_norm": 0.2966590212320321, + "learning_rate": 1.794503956576152e-05, + "loss": 0.3391, + "step": 226 + }, + { + "epoch": 2.9055999999999997, + "grad_norm": 0.31906514403005093, + "learning_rate": 1.776706354090585e-05, + "loss": 0.349, + "step": 227 + }, + { + "epoch": 2.9184, + "grad_norm": 0.2749646812104898, + "learning_rate": 1.758926639489354e-05, + "loss": 0.3435, + "step": 228 + }, + { + "epoch": 2.9312, + "grad_norm": 0.34925561830239105, + "learning_rate": 1.7411662370918893e-05, + "loss": 0.3367, + "step": 229 + }, + { + "epoch": 2.944, + "grad_norm": 0.3258150309203499, + "learning_rate": 1.7234265696705344e-05, + "loss": 0.3398, + "step": 230 + }, + { + "epoch": 2.9568, + "grad_norm": 0.31046565035409324, + "learning_rate": 1.7057090583365678e-05, + "loss": 0.3401, + "step": 231 + }, + { + "epoch": 2.9696, + "grad_norm": 0.27678310267277134, + "learning_rate": 1.6880151224263646e-05, + "loss": 0.3468, + "step": 232 + }, + { + "epoch": 2.9824, + "grad_norm": 0.26358082043532155, + "learning_rate": 1.6703461793876876e-05, + "loss": 0.3455, + "step": 233 + }, + { + "epoch": 2.9952, + "grad_norm": 0.2782015930914342, + "learning_rate": 1.6527036446661396e-05, + "loss": 0.3425, + "step": 234 + }, + { + "epoch": 3.008, + "grad_norm": 0.31433339064037064, + "learning_rate": 1.635088931591775e-05, + "loss": 0.312, + "step": 235 + }, + { + "epoch": 3.0208, + "grad_norm": 0.31630189155401184, + "learning_rate": 1.6175034512658753e-05, + "loss": 0.2867, + "step": 236 + }, + { + "epoch": 3.0336, + "grad_norm": 0.7008268080615292, + "learning_rate": 1.5999486124479115e-05, + "loss": 0.2916, + "step": 237 + }, + { + "epoch": 3.0464, + "grad_norm": 0.46688092430277195, + "learning_rate": 1.5824258214426833e-05, + "loss": 0.2841, + "step": 238 + }, + { + "epoch": 3.0592, + "grad_norm": 0.4372201691892296, + "learning_rate": 1.5649364819876655e-05, + "loss": 0.2853, + "step": 239 + }, + { + "epoch": 3.072, + "grad_norm": 0.28891039619475767, + "learning_rate": 1.547481995140556e-05, + "loss": 0.2738, + "step": 240 + }, + { + "epoch": 3.0848, + "grad_norm": 0.41172928619364885, + "learning_rate": 1.5300637591670357e-05, + "loss": 0.2873, + "step": 241 + }, + { + "epoch": 3.0976, + "grad_norm": 0.33342033959666423, + "learning_rate": 1.5126831694287564e-05, + "loss": 0.2693, + "step": 242 + }, + { + "epoch": 3.1104, + "grad_norm": 0.3409937922538487, + "learning_rate": 1.4953416182715566e-05, + "loss": 0.2759, + "step": 243 + }, + { + "epoch": 3.1232, + "grad_norm": 0.320100558341601, + "learning_rate": 1.478040494913926e-05, + "loss": 0.2836, + "step": 244 + }, + { + "epoch": 3.136, + "grad_norm": 0.29809795708940967, + "learning_rate": 1.460781185335713e-05, + "loss": 0.2819, + "step": 245 + }, + { + "epoch": 3.1488, + "grad_norm": 0.29250467761203774, + "learning_rate": 1.443565072167095e-05, + "loss": 0.2752, + "step": 246 + }, + { + "epoch": 3.1616, + "grad_norm": 0.28423813424974437, + "learning_rate": 1.4263935345778202e-05, + "loss": 0.2796, + "step": 247 + }, + { + "epoch": 3.1744, + "grad_norm": 0.29305956201911165, + "learning_rate": 1.409267948166718e-05, + "loss": 0.2891, + "step": 248 + }, + { + "epoch": 3.1872, + "grad_norm": 0.29150469848488814, + "learning_rate": 1.3921896848515064e-05, + "loss": 0.2909, + "step": 249 + }, + { + "epoch": 3.2, + "grad_norm": 0.27886889202149356, + "learning_rate": 1.3751601127588849e-05, + "loss": 0.2779, + "step": 250 + }, + { + "epoch": 3.2128, + "grad_norm": 0.2755947418181061, + "learning_rate": 1.3581805961149371e-05, + "loss": 0.2724, + "step": 251 + }, + { + "epoch": 3.2256, + "grad_norm": 0.2673816477671389, + "learning_rate": 1.341252495135841e-05, + "loss": 0.2861, + "step": 252 + }, + { + "epoch": 3.2384, + "grad_norm": 0.2635382764399931, + "learning_rate": 1.324377165918906e-05, + "loss": 0.2749, + "step": 253 + }, + { + "epoch": 3.2512, + "grad_norm": 0.2566074019105406, + "learning_rate": 1.3075559603339354e-05, + "loss": 0.2798, + "step": 254 + }, + { + "epoch": 3.2640000000000002, + "grad_norm": 0.26266710866870446, + "learning_rate": 1.2907902259149287e-05, + "loss": 0.2761, + "step": 255 + }, + { + "epoch": 3.2768, + "grad_norm": 0.2805453653484367, + "learning_rate": 1.274081305752135e-05, + "loss": 0.2813, + "step": 256 + }, + { + "epoch": 3.2896, + "grad_norm": 0.2582997259789758, + "learning_rate": 1.2574305383844528e-05, + "loss": 0.2861, + "step": 257 + }, + { + "epoch": 3.3024, + "grad_norm": 0.2711321120184753, + "learning_rate": 1.2408392576922075e-05, + "loss": 0.2769, + "step": 258 + }, + { + "epoch": 3.3152, + "grad_norm": 0.26481288833671507, + "learning_rate": 1.2243087927902905e-05, + "loss": 0.2824, + "step": 259 + }, + { + "epoch": 3.328, + "grad_norm": 0.28066151058647587, + "learning_rate": 1.2078404679216864e-05, + "loss": 0.2847, + "step": 260 + }, + { + "epoch": 3.3407999999999998, + "grad_norm": 0.27305714376519247, + "learning_rate": 1.1914356023513904e-05, + "loss": 0.2777, + "step": 261 + }, + { + "epoch": 3.3536, + "grad_norm": 0.2559955512720122, + "learning_rate": 1.1750955102607193e-05, + "loss": 0.2719, + "step": 262 + }, + { + "epoch": 3.3664, + "grad_norm": 0.24283413587059144, + "learning_rate": 1.1588215006420374e-05, + "loss": 0.2788, + "step": 263 + }, + { + "epoch": 3.3792, + "grad_norm": 0.2502212680798441, + "learning_rate": 1.1426148771938915e-05, + "loss": 0.2746, + "step": 264 + }, + { + "epoch": 3.392, + "grad_norm": 0.26092754422583947, + "learning_rate": 1.1264769382165748e-05, + "loss": 0.2816, + "step": 265 + }, + { + "epoch": 3.4048, + "grad_norm": 0.2282597423607946, + "learning_rate": 1.110408976508118e-05, + "loss": 0.2819, + "step": 266 + }, + { + "epoch": 3.4176, + "grad_norm": 0.2711529173262311, + "learning_rate": 1.094412279260726e-05, + "loss": 0.2756, + "step": 267 + }, + { + "epoch": 3.4304, + "grad_norm": 0.23736304306624106, + "learning_rate": 1.0784881279576635e-05, + "loss": 0.2689, + "step": 268 + }, + { + "epoch": 3.4432, + "grad_norm": 0.26229924143448996, + "learning_rate": 1.0626377982705929e-05, + "loss": 0.2807, + "step": 269 + }, + { + "epoch": 3.456, + "grad_norm": 0.24160423907741094, + "learning_rate": 1.0468625599573842e-05, + "loss": 0.2849, + "step": 270 + }, + { + "epoch": 3.4688, + "grad_norm": 0.24781523917685258, + "learning_rate": 1.0311636767603952e-05, + "loss": 0.2825, + "step": 271 + }, + { + "epoch": 3.4816, + "grad_norm": 0.23837646582381714, + "learning_rate": 1.0155424063052306e-05, + "loss": 0.2911, + "step": 272 + }, + { + "epoch": 3.4944, + "grad_norm": 0.22393081371619156, + "learning_rate": 1.0000000000000006e-05, + "loss": 0.2897, + "step": 273 + }, + { + "epoch": 3.5072, + "grad_norm": 0.2429111872222838, + "learning_rate": 9.84537702935065e-06, + "loss": 0.281, + "step": 274 + }, + { + "epoch": 3.52, + "grad_norm": 0.2254481115341149, + "learning_rate": 9.691567537832964e-06, + "loss": 0.2905, + "step": 275 + }, + { + "epoch": 3.5328, + "grad_norm": 0.2349724201722084, + "learning_rate": 9.538583847008452e-06, + "loss": 0.2971, + "step": 276 + }, + { + "epoch": 3.5456, + "grad_norm": 0.23371285739416045, + "learning_rate": 9.386438212284372e-06, + "loss": 0.2777, + "step": 277 + }, + { + "epoch": 3.5584, + "grad_norm": 0.23792960032240412, + "learning_rate": 9.235142821931928e-06, + "loss": 0.2789, + "step": 278 + }, + { + "epoch": 3.5712, + "grad_norm": 0.22122647180575977, + "learning_rate": 9.084709796109907e-06, + "loss": 0.2843, + "step": 279 + }, + { + "epoch": 3.584, + "grad_norm": 0.2438872273005057, + "learning_rate": 8.93515118589373e-06, + "loss": 0.2822, + "step": 280 + }, + { + "epoch": 3.5968, + "grad_norm": 0.23342124760266283, + "learning_rate": 8.786478972310023e-06, + "loss": 0.288, + "step": 281 + }, + { + "epoch": 3.6096, + "grad_norm": 0.23582335353818698, + "learning_rate": 8.638705065376887e-06, + "loss": 0.2787, + "step": 282 + }, + { + "epoch": 3.6224, + "grad_norm": 0.24209907403389272, + "learning_rate": 8.491841303149728e-06, + "loss": 0.2878, + "step": 283 + }, + { + "epoch": 3.6352, + "grad_norm": 0.21384750896331084, + "learning_rate": 8.345899450772975e-06, + "loss": 0.2897, + "step": 284 + }, + { + "epoch": 3.648, + "grad_norm": 0.2195417663551771, + "learning_rate": 8.200891199537549e-06, + "loss": 0.2726, + "step": 285 + }, + { + "epoch": 3.6608, + "grad_norm": 0.2231301243500821, + "learning_rate": 8.056828165944282e-06, + "loss": 0.2833, + "step": 286 + }, + { + "epoch": 3.6736, + "grad_norm": 0.2149364649666386, + "learning_rate": 7.913721890773354e-06, + "loss": 0.2811, + "step": 287 + }, + { + "epoch": 3.6864, + "grad_norm": 0.21357423713236115, + "learning_rate": 7.771583838159756e-06, + "loss": 0.2826, + "step": 288 + }, + { + "epoch": 3.6992000000000003, + "grad_norm": 0.24825207098701313, + "learning_rate": 7.630425394674903e-06, + "loss": 0.2826, + "step": 289 + }, + { + "epoch": 3.7119999999999997, + "grad_norm": 0.21415212818461488, + "learning_rate": 7.49025786841445e-06, + "loss": 0.2853, + "step": 290 + }, + { + "epoch": 3.7248, + "grad_norm": 0.22015828134494975, + "learning_rate": 7.3510924880924575e-06, + "loss": 0.2841, + "step": 291 + }, + { + "epoch": 3.7376, + "grad_norm": 0.22224706450011542, + "learning_rate": 7.212940402141808e-06, + "loss": 0.2835, + "step": 292 + }, + { + "epoch": 3.7504, + "grad_norm": 0.21581961016679654, + "learning_rate": 7.075812677821145e-06, + "loss": 0.2843, + "step": 293 + }, + { + "epoch": 3.7632, + "grad_norm": 0.21835107436796022, + "learning_rate": 6.939720300328303e-06, + "loss": 0.2688, + "step": 294 + }, + { + "epoch": 3.776, + "grad_norm": 0.212157439367985, + "learning_rate": 6.8046741719202385e-06, + "loss": 0.2754, + "step": 295 + }, + { + "epoch": 3.7888, + "grad_norm": 0.21531406072877302, + "learning_rate": 6.67068511103971e-06, + "loss": 0.28, + "step": 296 + }, + { + "epoch": 3.8016, + "grad_norm": 0.2334993523584025, + "learning_rate": 6.537763851448593e-06, + "loss": 0.2797, + "step": 297 + }, + { + "epoch": 3.8144, + "grad_norm": 0.21250002898841552, + "learning_rate": 6.4059210413680175e-06, + "loss": 0.2783, + "step": 298 + }, + { + "epoch": 3.8272, + "grad_norm": 0.2050927549154722, + "learning_rate": 6.275167242625331e-06, + "loss": 0.2806, + "step": 299 + }, + { + "epoch": 3.84, + "grad_norm": 0.23528957358515223, + "learning_rate": 6.145512929808013e-06, + "loss": 0.2823, + "step": 300 + }, + { + "epoch": 3.8528000000000002, + "grad_norm": 0.21866030120409596, + "learning_rate": 6.016968489424572e-06, + "loss": 0.2694, + "step": 301 + }, + { + "epoch": 3.8656, + "grad_norm": 0.1986886641313073, + "learning_rate": 5.889544219072465e-06, + "loss": 0.2815, + "step": 302 + }, + { + "epoch": 3.8784, + "grad_norm": 0.20407358585815805, + "learning_rate": 5.7632503266131925e-06, + "loss": 0.281, + "step": 303 + }, + { + "epoch": 3.8912, + "grad_norm": 0.21795566932605281, + "learning_rate": 5.638096929354522e-06, + "loss": 0.2698, + "step": 304 + }, + { + "epoch": 3.904, + "grad_norm": 0.20828825849742277, + "learning_rate": 5.514094053240035e-06, + "loss": 0.2814, + "step": 305 + }, + { + "epoch": 3.9168, + "grad_norm": 0.20552858430321327, + "learning_rate": 5.39125163204594e-06, + "loss": 0.281, + "step": 306 + }, + { + "epoch": 3.9295999999999998, + "grad_norm": 0.18563231992381732, + "learning_rate": 5.269579506585259e-06, + "loss": 0.2784, + "step": 307 + }, + { + "epoch": 3.9424, + "grad_norm": 0.2181494958934182, + "learning_rate": 5.149087423919541e-06, + "loss": 0.2684, + "step": 308 + }, + { + "epoch": 3.9552, + "grad_norm": 0.20575475264490437, + "learning_rate": 5.029785036577976e-06, + "loss": 0.2812, + "step": 309 + }, + { + "epoch": 3.968, + "grad_norm": 0.20762835949111869, + "learning_rate": 4.911681901784198e-06, + "loss": 0.2717, + "step": 310 + }, + { + "epoch": 3.9808, + "grad_norm": 0.19160088442220838, + "learning_rate": 4.794787480690597e-06, + "loss": 0.2738, + "step": 311 + }, + { + "epoch": 3.9936, + "grad_norm": 0.2024203895067496, + "learning_rate": 4.679111137620442e-06, + "loss": 0.2782, + "step": 312 + }, + { + "epoch": 4.0064, + "grad_norm": 0.2483219477733269, + "learning_rate": 4.5646621393177e-06, + "loss": 0.2671, + "step": 313 + }, + { + "epoch": 4.0192, + "grad_norm": 0.31427588349183117, + "learning_rate": 4.451449654204685e-06, + "loss": 0.2487, + "step": 314 + }, + { + "epoch": 4.032, + "grad_norm": 0.2210039512386439, + "learning_rate": 4.339482751647557e-06, + "loss": 0.2463, + "step": 315 + }, + { + "epoch": 4.0448, + "grad_norm": 0.26443806180928325, + "learning_rate": 4.228770401229824e-06, + "loss": 0.2385, + "step": 316 + }, + { + "epoch": 4.0576, + "grad_norm": 0.3912919431430795, + "learning_rate": 4.119321472033779e-06, + "loss": 0.2463, + "step": 317 + }, + { + "epoch": 4.0704, + "grad_norm": 0.25929046643811793, + "learning_rate": 4.011144731929981e-06, + "loss": 0.2441, + "step": 318 + }, + { + "epoch": 4.0832, + "grad_norm": 0.24349599096142088, + "learning_rate": 3.904248846874894e-06, + "loss": 0.2356, + "step": 319 + }, + { + "epoch": 4.096, + "grad_norm": 0.2900612365803001, + "learning_rate": 3.7986423802166705e-06, + "loss": 0.2515, + "step": 320 + }, + { + "epoch": 4.1088, + "grad_norm": 0.2873414105221977, + "learning_rate": 3.694333792009115e-06, + "loss": 0.2536, + "step": 321 + }, + { + "epoch": 4.1216, + "grad_norm": 0.23751942576573712, + "learning_rate": 3.5913314383339937e-06, + "loss": 0.2503, + "step": 322 + }, + { + "epoch": 4.1344, + "grad_norm": 0.21072311310845263, + "learning_rate": 3.4896435706316e-06, + "loss": 0.2481, + "step": 323 + }, + { + "epoch": 4.1472, + "grad_norm": 0.2434813470084433, + "learning_rate": 3.3892783350397675e-06, + "loss": 0.234, + "step": 324 + }, + { + "epoch": 4.16, + "grad_norm": 0.268913173192014, + "learning_rate": 3.290243771741275e-06, + "loss": 0.2427, + "step": 325 + }, + { + "epoch": 4.1728, + "grad_norm": 0.2518464525686985, + "learning_rate": 3.1925478143197418e-06, + "loss": 0.2453, + "step": 326 + }, + { + "epoch": 4.1856, + "grad_norm": 0.2202620097200464, + "learning_rate": 3.0961982891241083e-06, + "loss": 0.2476, + "step": 327 + }, + { + "epoch": 4.1984, + "grad_norm": 0.20611433865734524, + "learning_rate": 3.001202914641628e-06, + "loss": 0.2529, + "step": 328 + }, + { + "epoch": 4.2112, + "grad_norm": 0.23978958306869538, + "learning_rate": 2.907569300879596e-06, + "loss": 0.2365, + "step": 329 + }, + { + "epoch": 4.224, + "grad_norm": 0.24793303268384181, + "learning_rate": 2.815304948755664e-06, + "loss": 0.245, + "step": 330 + }, + { + "epoch": 4.2368, + "grad_norm": 0.2211020129342562, + "learning_rate": 2.7244172494969978e-06, + "loss": 0.2228, + "step": 331 + }, + { + "epoch": 4.2496, + "grad_norm": 0.21422382276077329, + "learning_rate": 2.6349134840481294e-06, + "loss": 0.2534, + "step": 332 + }, + { + "epoch": 4.2624, + "grad_norm": 0.23353733968476206, + "learning_rate": 2.546800822487714e-06, + "loss": 0.2333, + "step": 333 + }, + { + "epoch": 4.2752, + "grad_norm": 0.22380233207256295, + "learning_rate": 2.4600863234541338e-06, + "loss": 0.2506, + "step": 334 + }, + { + "epoch": 4.288, + "grad_norm": 0.22667996547433356, + "learning_rate": 2.374776933580025e-06, + "loss": 0.2418, + "step": 335 + }, + { + "epoch": 4.3008, + "grad_norm": 0.2083067132416159, + "learning_rate": 2.2908794869358044e-06, + "loss": 0.2348, + "step": 336 + }, + { + "epoch": 4.3136, + "grad_norm": 0.1899808874315056, + "learning_rate": 2.2084007044821764e-06, + "loss": 0.2472, + "step": 337 + }, + { + "epoch": 4.3264, + "grad_norm": 0.19416358137566275, + "learning_rate": 2.127347193531757e-06, + "loss": 0.2399, + "step": 338 + }, + { + "epoch": 4.3392, + "grad_norm": 0.19459645408383455, + "learning_rate": 2.0477254472197237e-06, + "loss": 0.2273, + "step": 339 + }, + { + "epoch": 4.352, + "grad_norm": 0.18919120751562665, + "learning_rate": 1.96954184398368e-06, + "loss": 0.2388, + "step": 340 + }, + { + "epoch": 4.3648, + "grad_norm": 0.19581399230843038, + "learning_rate": 1.8928026470526917e-06, + "loss": 0.2371, + "step": 341 + }, + { + "epoch": 4.3776, + "grad_norm": 0.19539124986378142, + "learning_rate": 1.817514003945524e-06, + "loss": 0.24, + "step": 342 + }, + { + "epoch": 4.3904, + "grad_norm": 0.18798319840625727, + "learning_rate": 1.743681945978184e-06, + "loss": 0.2425, + "step": 343 + }, + { + "epoch": 4.4032, + "grad_norm": 0.19148378298433658, + "learning_rate": 1.6713123877807413e-06, + "loss": 0.2355, + "step": 344 + }, + { + "epoch": 4.416, + "grad_norm": 0.19209574289693548, + "learning_rate": 1.6004111268235156e-06, + "loss": 0.2399, + "step": 345 + }, + { + "epoch": 4.4288, + "grad_norm": 0.19840558163677857, + "learning_rate": 1.5309838429526714e-06, + "loss": 0.2421, + "step": 346 + }, + { + "epoch": 4.4416, + "grad_norm": 0.1806706307524296, + "learning_rate": 1.4630360979351644e-06, + "loss": 0.2324, + "step": 347 + }, + { + "epoch": 4.4544, + "grad_norm": 0.17851900712531546, + "learning_rate": 1.396573335013236e-06, + "loss": 0.2296, + "step": 348 + }, + { + "epoch": 4.4672, + "grad_norm": 0.1851790531424452, + "learning_rate": 1.3316008784683265e-06, + "loss": 0.2455, + "step": 349 + }, + { + "epoch": 4.48, + "grad_norm": 0.18420836933280219, + "learning_rate": 1.2681239331945695e-06, + "loss": 0.2392, + "step": 350 + }, + { + "epoch": 4.4928, + "grad_norm": 0.182888054979642, + "learning_rate": 1.2061475842818337e-06, + "loss": 0.2372, + "step": 351 + }, + { + "epoch": 4.5056, + "grad_norm": 0.18199196177578558, + "learning_rate": 1.1456767966083393e-06, + "loss": 0.2429, + "step": 352 + }, + { + "epoch": 4.5184, + "grad_norm": 0.18285968384881784, + "learning_rate": 1.086716414442952e-06, + "loss": 0.2424, + "step": 353 + }, + { + "epoch": 4.5312, + "grad_norm": 0.18120132437676076, + "learning_rate": 1.0292711610570904e-06, + "loss": 0.2405, + "step": 354 + }, + { + "epoch": 4.5440000000000005, + "grad_norm": 0.18256647144707736, + "learning_rate": 9.733456383463658e-07, + "loss": 0.2403, + "step": 355 + }, + { + "epoch": 4.5568, + "grad_norm": 0.1806229959243068, + "learning_rate": 9.189443264619102e-07, + "loss": 0.2358, + "step": 356 + }, + { + "epoch": 4.5696, + "grad_norm": 0.17903500347908813, + "learning_rate": 8.660715834514977e-07, + "loss": 0.242, + "step": 357 + }, + { + "epoch": 4.5824, + "grad_norm": 0.18322902020731868, + "learning_rate": 8.147316449103959e-07, + "loss": 0.2353, + "step": 358 + }, + { + "epoch": 4.5952, + "grad_norm": 0.18920089983891739, + "learning_rate": 7.649286236420806e-07, + "loss": 0.2411, + "step": 359 + }, + { + "epoch": 4.608, + "grad_norm": 0.1715008809199515, + "learning_rate": 7.166665093287539e-07, + "loss": 0.2324, + "step": 360 + }, + { + "epoch": 4.6208, + "grad_norm": 0.18710310075634906, + "learning_rate": 6.69949168211721e-07, + "loss": 0.2437, + "step": 361 + }, + { + "epoch": 4.6336, + "grad_norm": 0.18506851044805525, + "learning_rate": 6.247803427816945e-07, + "loss": 0.2455, + "step": 362 + }, + { + "epoch": 4.6464, + "grad_norm": 0.18094845128973183, + "learning_rate": 5.811636514789598e-07, + "loss": 0.2258, + "step": 363 + }, + { + "epoch": 4.6592, + "grad_norm": 0.17435332481727803, + "learning_rate": 5.391025884035239e-07, + "loss": 0.239, + "step": 364 + }, + { + "epoch": 4.672, + "grad_norm": 0.17695486262206614, + "learning_rate": 4.986005230351954e-07, + "loss": 0.2269, + "step": 365 + }, + { + "epoch": 4.6848, + "grad_norm": 0.18275683820670355, + "learning_rate": 4.5966069996365993e-07, + "loss": 0.2508, + "step": 366 + }, + { + "epoch": 4.6975999999999996, + "grad_norm": 0.1837979328938096, + "learning_rate": 4.22286238628562e-07, + "loss": 0.2368, + "step": 367 + }, + { + "epoch": 4.7104, + "grad_norm": 0.18220151570131113, + "learning_rate": 3.8648013306960664e-07, + "loss": 0.2477, + "step": 368 + }, + { + "epoch": 4.7232, + "grad_norm": 0.1730991293089666, + "learning_rate": 3.522452516867048e-07, + "loss": 0.2469, + "step": 369 + }, + { + "epoch": 4.736, + "grad_norm": 0.19236348216935706, + "learning_rate": 3.1958433701019697e-07, + "loss": 0.2421, + "step": 370 + }, + { + "epoch": 4.7488, + "grad_norm": 0.1706653957534717, + "learning_rate": 2.8850000548115155e-07, + "loss": 0.2441, + "step": 371 + }, + { + "epoch": 4.7616, + "grad_norm": 0.18911082173661192, + "learning_rate": 2.5899474724174313e-07, + "loss": 0.2423, + "step": 372 + }, + { + "epoch": 4.7744, + "grad_norm": 0.17618675679611007, + "learning_rate": 2.3107092593579905e-07, + "loss": 0.259, + "step": 373 + }, + { + "epoch": 4.7872, + "grad_norm": 0.17961036651229495, + "learning_rate": 2.0473077851942858e-07, + "loss": 0.2478, + "step": 374 + }, + { + "epoch": 4.8, + "grad_norm": 0.17730836222535062, + "learning_rate": 1.799764150818306e-07, + "loss": 0.2484, + "step": 375 + }, + { + "epoch": 4.8128, + "grad_norm": 0.17676500581389382, + "learning_rate": 1.5680981867625566e-07, + "loss": 0.2365, + "step": 376 + }, + { + "epoch": 4.8256, + "grad_norm": 0.17587484900010034, + "learning_rate": 1.3523284516113955e-07, + "loss": 0.2543, + "step": 377 + }, + { + "epoch": 4.8384, + "grad_norm": 0.17111307611066914, + "learning_rate": 1.1524722305144231e-07, + "loss": 0.2449, + "step": 378 + }, + { + "epoch": 4.8512, + "grad_norm": 0.16914105389943437, + "learning_rate": 9.685455338016347e-08, + "loss": 0.2415, + "step": 379 + }, + { + "epoch": 4.864, + "grad_norm": 0.17406515355425717, + "learning_rate": 8.005630957010014e-08, + "loss": 0.2429, + "step": 380 + }, + { + "epoch": 4.8768, + "grad_norm": 0.17327135001293076, + "learning_rate": 6.485383731580142e-08, + "loss": 0.2505, + "step": 381 + }, + { + "epoch": 4.8896, + "grad_norm": 0.1762049065101459, + "learning_rate": 5.1248354475768034e-08, + "loss": 0.234, + "step": 382 + }, + { + "epoch": 4.9024, + "grad_norm": 0.17124756759793125, + "learning_rate": 3.924095097489922e-08, + "loss": 0.2415, + "step": 383 + }, + { + "epoch": 4.9152000000000005, + "grad_norm": 0.18444523509295194, + "learning_rate": 2.8832588717164766e-08, + "loss": 0.2261, + "step": 384 + }, + { + "epoch": 4.928, + "grad_norm": 0.18310827178900155, + "learning_rate": 2.0024101508555604e-08, + "loss": 0.2434, + "step": 385 + }, + { + "epoch": 4.9408, + "grad_norm": 0.17186707504148238, + "learning_rate": 1.281619499029274e-08, + "loss": 0.2454, + "step": 386 + }, + { + "epoch": 4.9536, + "grad_norm": 0.17818375445377446, + "learning_rate": 7.209446582292501e-09, + "loss": 0.2519, + "step": 387 + }, + { + "epoch": 4.9664, + "grad_norm": 0.16976076854341357, + "learning_rate": 3.2043054369057523e-09, + "loss": 0.2388, + "step": 388 + }, + { + "epoch": 4.9792, + "grad_norm": 0.17562050527830075, + "learning_rate": 8.010924029533406e-10, + "loss": 0.2355, + "step": 389 + }, + { + "epoch": 4.992, + "grad_norm": 0.24606256837815751, + "learning_rate": 0.0, + "loss": 0.2438, + "step": 390 + }, + { + "epoch": 4.992, + "step": 390, + "total_flos": 1.9952099917069025e+18, + "train_loss": 0.36405690262715024, + "train_runtime": 60425.7754, + "train_samples_per_second": 0.827, + "train_steps_per_second": 0.006 + } + ], + "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": 1.9952099917069025e+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..a10246f --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c6090b506d21c2646621d11337cbdab1e30838ef97cd9660e082687626f900c4 +size 7352 diff --git a/training_loss.png b/training_loss.png new file mode 100644 index 0000000..7937ac1 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