diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..a6f0260516b93f4e03d2ed9e663d80b544f2af0a 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,3 +33,10 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text +checkpoint-1276/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-1595/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-1914/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-319/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-638/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-957/tokenizer.json filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000000000000000000000000000000000000..a8325c13a455ba067f3f3892fc751081d947ceeb --- /dev/null +++ b/README.md @@ -0,0 +1,146 @@ +--- +library_name: peft +license: other +base_model: nvidia/Llama-3_3-Nemotron-Super-49B-v1 +tags: +- generated_from_trainer +datasets: +- ugaoo/subset_each5k_multimedqa +model-index: +- name: out/subset_each5k_multimedqa + results: [] +--- + + + +[Built with Axolotl](https://github.com/axolotl-ai-cloud/axolotl) +
See axolotl config + +axolotl version: `0.8.0.dev0` +```yaml +base_model: nvidia/Llama-3_3-Nemotron-Super-49B-v1 +model_type: AutoModelForCausalLM +tokenizer_type: AutoTokenizer +trust_remote_code: true + +load_in_8bit: false +load_in_4bit: true +strict: false + +datasets: + - path: ugaoo/subset_each5k_multimedqa + type: alpaca +val_set_size: 0 +output_dir: ./out/subset_each5k_multimedqa + +sequence_len: 4000 +sample_packing: true +pad_to_sequence_len: true + +adapter: qlora +lora_r: 256 +lora_alpha: 512 +lora_dropout: 0.05 +lora_target_linear: true +lora_target_modules: + - q_proj + - k_proj + - v_proj + - o_proj + - up_proj + - down_proj + - gate_proj +lora_modules_to_save: + - embed_tokens + - lm_head + +wandb_project: cosmosearch +wandb_entity: +wandb_watch: +wandb_name: subset_each5k_multimedqa_Super-49B +wandb_log_model: + +gradient_accumulation_steps: 3 +micro_batch_size: 4 +num_epochs: 6 +optimizer: adamw_torch +lr_scheduler: cosine +learning_rate: 5e-6 + +train_on_inputs: false +group_by_length: false +bf16: auto +fp16: false +tf32: false + +gradient_checkpointing: true +early_stopping_patience: +resume_from_checkpoint: +logging_steps: 1 +xformers_attention: +flash_attention: true + +warmup_steps: 100 +evals_per_epoch: 6 +eval_table_size: +saves_per_epoch: 1 +debug: +deepspeed: +weight_decay: 0.0 +fsdp: +fsdp_config: +save_total_limit: 6 +special_tokens: + pad_token: <|end_of_text|> + +``` + +

+ +# out/subset_each5k_multimedqa + +This model is a fine-tuned version of [nvidia/Llama-3_3-Nemotron-Super-49B-v1](https://huggingface.co/nvidia/Llama-3_3-Nemotron-Super-49B-v1) on the ugaoo/subset_each5k_multimedqa 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: 5e-06 +- train_batch_size: 4 +- eval_batch_size: 4 +- seed: 42 +- distributed_type: multi-GPU +- num_devices: 2 +- gradient_accumulation_steps: 3 +- total_train_batch_size: 24 +- total_eval_batch_size: 8 +- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments +- lr_scheduler_type: cosine +- lr_scheduler_warmup_steps: 100 +- num_epochs: 6.0 + +### Training results + + + +### Framework versions + +- PEFT 0.15.0 +- Transformers 4.49.0 +- Pytorch 2.5.1+cu124 +- Datasets 3.4.1 +- Tokenizers 0.21.1 \ No newline at end of file diff --git a/adapter_config.json b/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..1edb33780e2306c6b19fd727be8e9b8b35f237c4 --- /dev/null +++ b/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3_3-Nemotron-Super-49B-v1", + "bias": "none", + "corda_config": null, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "k_proj", + "down_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/adapter_model.safetensors b/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7d16d638ba5f1f0d6e17a6a0445d8d989c8ed984 --- /dev/null +++ b/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8cb98185ecb57fad2b2cd35f0c47cae364129464746319ff1be6d8581c7a3a9 +size 9016826528 diff --git a/checkpoint-1276/README.md b/checkpoint-1276/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f4a3934800eeb082a0cb833d7b6af4f68eed3615 --- /dev/null +++ b/checkpoint-1276/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3_3-Nemotron-Super-49B-v1 +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.15.0 \ No newline at end of file diff --git a/checkpoint-1276/adapter_config.json b/checkpoint-1276/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..1edb33780e2306c6b19fd727be8e9b8b35f237c4 --- /dev/null +++ b/checkpoint-1276/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3_3-Nemotron-Super-49B-v1", + "bias": "none", + "corda_config": null, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "k_proj", + "down_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-1276/adapter_model.safetensors b/checkpoint-1276/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4b8be07df270fd4e87735aa91354c7f51137f5d6 --- /dev/null +++ b/checkpoint-1276/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d7bde0757e769babf9dc6c014a0ce0896f1088d912615723a24277598b70c49d +size 9016826528 diff --git a/checkpoint-1276/global_step1274/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-1276/global_step1274/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..07cf87de2769e0bd9e822fc60bf88d57b3ae7a7b --- /dev/null +++ b/checkpoint-1276/global_step1274/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:222bc048054940785017ef071226858bec2645b03f930461055002b65ba56e4f +size 27050164444 diff --git a/checkpoint-1276/global_step1274/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-1276/global_step1274/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..db37057b05cf3281fbe85b3adf2d1cf698afb0f7 --- /dev/null +++ b/checkpoint-1276/global_step1274/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5fdc005932a6d5446bc0f918fc68de05979fd84a3d4b1bda5efb335c34ececda +size 27050169884 diff --git a/checkpoint-1276/global_step1274/mp_rank_00_model_states.pt b/checkpoint-1276/global_step1274/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..0cb6e95641a13a1b52784e3eeafc59fb42aea0df --- /dev/null +++ b/checkpoint-1276/global_step1274/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:545433629b5dbe126a1b377a8dc22685cf8ad4ec5132093787c83469d43c9015 +size 9776788601 diff --git a/checkpoint-1276/latest b/checkpoint-1276/latest new file mode 100644 index 0000000000000000000000000000000000000000..d33d77b6ed79554e8b85a17b6b59c2f172ea9f9e --- /dev/null +++ b/checkpoint-1276/latest @@ -0,0 +1 @@ +global_step1274 \ No newline at end of file diff --git a/checkpoint-1276/rng_state_0.pth b/checkpoint-1276/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..0d88332a38b69b81b1016c4fa69dbe65f599099a --- /dev/null +++ b/checkpoint-1276/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:73180648d25cd446f1429dc27f4aa551fb5f30d042b0bc24192e218ac88eca4f +size 14512 diff --git a/checkpoint-1276/rng_state_1.pth b/checkpoint-1276/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..eb29cd6ceb687a018b46728f9e475936553186c4 --- /dev/null +++ b/checkpoint-1276/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a036b9448290510bd0a8a9d696995f9fa767f1e306257df7a231be1150d26a2 +size 14512 diff --git a/checkpoint-1276/scheduler.pt b/checkpoint-1276/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..809cdd5871eb1dfef67f89f3b3d929c507653f8a --- /dev/null +++ b/checkpoint-1276/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:561b6ce97e25a940d610e551fa7945082006574c3b6c0ab9444224cd88cf7e29 +size 1064 diff --git a/checkpoint-1276/special_tokens_map.json b/checkpoint-1276/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-1276/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-1276/tokenizer.json b/checkpoint-1276/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-1276/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-1276/tokenizer_config.json b/checkpoint-1276/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..edd01b980c1db496ea102a51c972ee8f5d1a2c74 --- /dev/null +++ b/checkpoint-1276/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "chat_template": "{{- bos_token }}{%- if messages[0]['role'] == 'system' %}{%- set system_message = messages[0]['content']|trim %}{%- set messages = messages[1:] %}{%- else %}{%- set system_message = \"\" %}{%- endif %}{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}{{- system_message }}{{- \"<|eot_id|>\" }}{%- for message in messages %}{%- if message['role'] == 'assistant' and '' in message['content'] %}{%- set content = message['content'].split('')[-1].lstrip() %}{%- else %}{%- set content = message['content'] %}{%- endif %}{{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n' + content | trim + '<|eot_id|>' }}{%- endfor %}{%- if add_generation_prompt %}{{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}{%- endif %}", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-1276/trainer_state.json b/checkpoint-1276/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..41253ad9b2e2c904758595f32b00239fc143f3ed --- /dev/null +++ b/checkpoint-1276/trainer_state.json @@ -0,0 +1,8965 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 3.989561586638831, + "eval_steps": 500, + "global_step": 1276, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.003131524008350731, + "grad_norm": 13.917898178100586, + "learning_rate": 5.0000000000000004e-08, + "loss": 4.1051, + "step": 1 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 17.327869415283203, + "learning_rate": 1.0000000000000001e-07, + "loss": 4.1048, + "step": 2 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 14.063946723937988, + "learning_rate": 1.5000000000000002e-07, + "loss": 4.0741, + "step": 3 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 16.817699432373047, + "learning_rate": 2.0000000000000002e-07, + "loss": 4.2002, + "step": 4 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 14.47036361694336, + "learning_rate": 2.5000000000000004e-07, + "loss": 4.2652, + "step": 5 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 14.474193572998047, + "learning_rate": 3.0000000000000004e-07, + "loss": 4.0888, + "step": 6 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 14.865458488464355, + "learning_rate": 3.5000000000000004e-07, + "loss": 4.0014, + "step": 7 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 15.338888168334961, + "learning_rate": 4.0000000000000003e-07, + "loss": 4.13, + "step": 8 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 15.154336929321289, + "learning_rate": 4.5000000000000003e-07, + "loss": 4.2493, + "step": 9 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 15.919597625732422, + "learning_rate": 5.000000000000001e-07, + "loss": 4.0535, + "step": 10 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 14.981926918029785, + "learning_rate": 5.5e-07, + "loss": 3.9064, + "step": 11 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 13.36101245880127, + "learning_rate": 6.000000000000001e-07, + "loss": 4.1939, + "step": 12 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 15.58773422241211, + "learning_rate": 6.5e-07, + "loss": 4.18, + "step": 13 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 13.560139656066895, + "learning_rate": 7.000000000000001e-07, + "loss": 3.9414, + "step": 14 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 12.307971954345703, + "learning_rate": 7.5e-07, + "loss": 3.8836, + "step": 15 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 14.533182144165039, + "learning_rate": 8.000000000000001e-07, + "loss": 4.1551, + "step": 16 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 13.453729629516602, + "learning_rate": 8.500000000000001e-07, + "loss": 4.0048, + "step": 17 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 13.45992374420166, + "learning_rate": 9.000000000000001e-07, + "loss": 4.0745, + "step": 18 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 11.857145309448242, + "learning_rate": 9.500000000000001e-07, + "loss": 3.9871, + "step": 19 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 11.872294425964355, + "learning_rate": 1.0000000000000002e-06, + "loss": 3.8959, + "step": 20 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.969825744628906, + "learning_rate": 1.0500000000000001e-06, + "loss": 4.0308, + "step": 21 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 12.33769416809082, + "learning_rate": 1.1e-06, + "loss": 3.9341, + "step": 22 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 12.669405937194824, + "learning_rate": 1.1500000000000002e-06, + "loss": 3.8511, + "step": 23 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 10.677213668823242, + "learning_rate": 1.2000000000000002e-06, + "loss": 3.7764, + "step": 24 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 10.366402626037598, + "learning_rate": 1.25e-06, + "loss": 3.5291, + "step": 25 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 11.211421012878418, + "learning_rate": 1.3e-06, + "loss": 3.5765, + "step": 26 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 11.313716888427734, + "learning_rate": 1.3500000000000002e-06, + "loss": 3.4849, + "step": 27 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 10.41294002532959, + "learning_rate": 1.4000000000000001e-06, + "loss": 3.2653, + "step": 28 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 10.40064525604248, + "learning_rate": 1.45e-06, + "loss": 3.3384, + "step": 29 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 10.05427074432373, + "learning_rate": 1.5e-06, + "loss": 3.2257, + "step": 30 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 9.583163261413574, + "learning_rate": 1.5500000000000002e-06, + "loss": 3.1371, + "step": 31 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 10.09977912902832, + "learning_rate": 1.6000000000000001e-06, + "loss": 3.0658, + "step": 32 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 9.271486282348633, + "learning_rate": 1.6500000000000003e-06, + "loss": 2.9693, + "step": 33 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 10.687992095947266, + "learning_rate": 1.7000000000000002e-06, + "loss": 2.95, + "step": 34 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 8.762290000915527, + "learning_rate": 1.75e-06, + "loss": 2.8286, + "step": 35 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 10.13785171508789, + "learning_rate": 1.8000000000000001e-06, + "loss": 2.3664, + "step": 36 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 18.301353454589844, + "learning_rate": 1.85e-06, + "loss": 2.5533, + "step": 37 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 11.490377426147461, + "learning_rate": 1.9000000000000002e-06, + "loss": 2.6133, + "step": 38 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 15.614163398742676, + "learning_rate": 1.9500000000000004e-06, + "loss": 2.3596, + "step": 39 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 17.757442474365234, + "learning_rate": 2.0000000000000003e-06, + "loss": 2.3491, + "step": 40 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 17.18431854248047, + "learning_rate": 2.05e-06, + "loss": 2.2361, + "step": 41 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 16.149789810180664, + "learning_rate": 2.1000000000000002e-06, + "loss": 2.1457, + "step": 42 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 15.256914138793945, + "learning_rate": 2.15e-06, + "loss": 2.12, + "step": 43 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 15.537406921386719, + "learning_rate": 2.2e-06, + "loss": 2.1877, + "step": 44 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 7.947713851928711, + "learning_rate": 2.25e-06, + "loss": 2.1648, + "step": 45 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 8.818676948547363, + "learning_rate": 2.3000000000000004e-06, + "loss": 2.134, + "step": 46 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 5.175768852233887, + "learning_rate": 2.35e-06, + "loss": 2.0796, + "step": 47 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 6.750611305236816, + "learning_rate": 2.4000000000000003e-06, + "loss": 1.9174, + "step": 48 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 6.2147979736328125, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.8065, + "step": 49 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 13.291611671447754, + "learning_rate": 2.5e-06, + "loss": 1.7061, + "step": 50 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 7.251201629638672, + "learning_rate": 2.55e-06, + "loss": 1.7924, + "step": 51 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 5.2126054763793945, + "learning_rate": 2.6e-06, + "loss": 1.6735, + "step": 52 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 5.435528755187988, + "learning_rate": 2.6500000000000005e-06, + "loss": 1.6265, + "step": 53 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 4.505807399749756, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.4851, + "step": 54 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 5.128388404846191, + "learning_rate": 2.7500000000000004e-06, + "loss": 1.5832, + "step": 55 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 16.935827255249023, + "learning_rate": 2.8000000000000003e-06, + "loss": 1.6553, + "step": 56 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 3.664458990097046, + "learning_rate": 2.85e-06, + "loss": 1.5, + "step": 57 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 7.763802528381348, + "learning_rate": 2.9e-06, + "loss": 1.367, + "step": 58 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 3.2216155529022217, + "learning_rate": 2.95e-06, + "loss": 1.3863, + "step": 59 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 4.384445667266846, + "learning_rate": 3e-06, + "loss": 1.4247, + "step": 60 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 4.8080878257751465, + "learning_rate": 3.05e-06, + "loss": 1.3257, + "step": 61 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 4.154761791229248, + "learning_rate": 3.1000000000000004e-06, + "loss": 1.321, + "step": 62 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 6.4742112159729, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.2823, + "step": 63 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 2.583422899246216, + "learning_rate": 3.2000000000000003e-06, + "loss": 1.2136, + "step": 64 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 4.1933488845825195, + "learning_rate": 3.2500000000000002e-06, + "loss": 1.1855, + "step": 65 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 4.11049747467041, + "learning_rate": 3.3000000000000006e-06, + "loss": 1.2389, + "step": 66 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 2.264458417892456, + "learning_rate": 3.3500000000000005e-06, + "loss": 1.0651, + "step": 67 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 2.5408174991607666, + "learning_rate": 3.4000000000000005e-06, + "loss": 1.1389, + "step": 68 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 7.82421350479126, + "learning_rate": 3.45e-06, + "loss": 1.0956, + "step": 69 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 3.070939064025879, + "learning_rate": 3.5e-06, + "loss": 1.0451, + "step": 70 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 2.6310527324676514, + "learning_rate": 3.5500000000000003e-06, + "loss": 1.0538, + "step": 71 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 7.630155563354492, + "learning_rate": 3.6000000000000003e-06, + "loss": 1.0052, + "step": 72 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 6.950636863708496, + "learning_rate": 3.65e-06, + "loss": 1.0473, + "step": 73 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 2.2703945636749268, + "learning_rate": 3.7e-06, + "loss": 1.0576, + "step": 74 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 3.3817710876464844, + "learning_rate": 3.7500000000000005e-06, + "loss": 1.0177, + "step": 75 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 7.266414642333984, + "learning_rate": 3.8000000000000005e-06, + "loss": 1.0645, + "step": 76 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 5.782608509063721, + "learning_rate": 3.85e-06, + "loss": 1.0162, + "step": 77 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 2.7938575744628906, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9664, + "step": 78 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 6.681935787200928, + "learning_rate": 3.95e-06, + "loss": 0.953, + "step": 79 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 2.253279209136963, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9568, + "step": 80 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 1.4875826835632324, + "learning_rate": 4.05e-06, + "loss": 0.9448, + "step": 81 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 2.4987940788269043, + "learning_rate": 4.1e-06, + "loss": 0.9393, + "step": 82 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 4.712948322296143, + "learning_rate": 4.15e-06, + "loss": 0.9532, + "step": 83 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 6.9030632972717285, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.96, + "step": 84 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 3.4780967235565186, + "learning_rate": 4.25e-06, + "loss": 0.8993, + "step": 85 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.526064395904541, + "learning_rate": 4.3e-06, + "loss": 0.9021, + "step": 86 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 10.727686882019043, + "learning_rate": 4.350000000000001e-06, + "loss": 0.856, + "step": 87 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 12.483160972595215, + "learning_rate": 4.4e-06, + "loss": 0.9357, + "step": 88 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 6.544492244720459, + "learning_rate": 4.450000000000001e-06, + "loss": 0.9168, + "step": 89 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 1.178139567375183, + "learning_rate": 4.5e-06, + "loss": 0.8748, + "step": 90 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 1.711506962776184, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8425, + "step": 91 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 3.281747341156006, + "learning_rate": 4.600000000000001e-06, + "loss": 0.8491, + "step": 92 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 2.2964377403259277, + "learning_rate": 4.65e-06, + "loss": 0.8038, + "step": 93 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 1.959700345993042, + "learning_rate": 4.7e-06, + "loss": 0.8439, + "step": 94 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 3.979384183883667, + "learning_rate": 4.75e-06, + "loss": 0.8839, + "step": 95 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 1.4721262454986572, + "learning_rate": 4.800000000000001e-06, + "loss": 0.845, + "step": 96 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 2.862248659133911, + "learning_rate": 4.85e-06, + "loss": 0.7748, + "step": 97 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 3.7439088821411133, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8145, + "step": 98 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 1.6654618978500366, + "learning_rate": 4.95e-06, + "loss": 0.8326, + "step": 99 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 7.8437581062316895, + "learning_rate": 5e-06, + "loss": 0.8666, + "step": 100 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 6.429738521575928, + "learning_rate": 4.999996250830422e-06, + "loss": 0.836, + "step": 101 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 2.6017794609069824, + "learning_rate": 4.9999850033329326e-06, + "loss": 0.7785, + "step": 102 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 1.0575449466705322, + "learning_rate": 4.999966257541265e-06, + "loss": 0.7639, + "step": 103 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 2.6932010650634766, + "learning_rate": 4.999940013511647e-06, + "loss": 0.8214, + "step": 104 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 2.925288438796997, + "learning_rate": 4.999906271322792e-06, + "loss": 0.8797, + "step": 105 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.3570607900619507, + "learning_rate": 4.9998650310759035e-06, + "loss": 0.792, + "step": 106 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 5.126713752746582, + "learning_rate": 4.999816292894676e-06, + "loss": 0.8352, + "step": 107 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 1.8966432809829712, + "learning_rate": 4.99976005692529e-06, + "loss": 0.7663, + "step": 108 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 1.3100829124450684, + "learning_rate": 4.999696323336418e-06, + "loss": 0.771, + "step": 109 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 2.4025354385375977, + "learning_rate": 4.999625092319218e-06, + "loss": 0.7618, + "step": 110 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.130232810974121, + "learning_rate": 4.999546364087334e-06, + "loss": 0.7705, + "step": 111 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 3.430262327194214, + "learning_rate": 4.999460138876901e-06, + "loss": 0.77, + "step": 112 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 1.1272103786468506, + "learning_rate": 4.999366416946536e-06, + "loss": 0.7133, + "step": 113 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 1.1740471124649048, + "learning_rate": 4.999265198577342e-06, + "loss": 0.7684, + "step": 114 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.3138248920440674, + "learning_rate": 4.999156484072907e-06, + "loss": 0.7888, + "step": 115 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 1.061711311340332, + "learning_rate": 4.999040273759304e-06, + "loss": 0.7484, + "step": 116 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.4682390689849854, + "learning_rate": 4.998916567985083e-06, + "loss": 0.7296, + "step": 117 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 2.884068250656128, + "learning_rate": 4.998785367121284e-06, + "loss": 0.7662, + "step": 118 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9812761545181274, + "learning_rate": 4.9986466715614205e-06, + "loss": 0.7307, + "step": 119 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 2.2237496376037598, + "learning_rate": 4.998500481721484e-06, + "loss": 0.6761, + "step": 120 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 1.4004178047180176, + "learning_rate": 4.998346798039952e-06, + "loss": 0.7505, + "step": 121 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 5.54975700378418, + "learning_rate": 4.99818562097777e-06, + "loss": 0.7615, + "step": 122 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 6.17140531539917, + "learning_rate": 4.9980169510183624e-06, + "loss": 0.7002, + "step": 123 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 4.974380016326904, + "learning_rate": 4.997840788667628e-06, + "loss": 0.7449, + "step": 124 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 1.4133399724960327, + "learning_rate": 4.997657134453937e-06, + "loss": 0.7442, + "step": 125 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 1.868915319442749, + "learning_rate": 4.9974659889281295e-06, + "loss": 0.7104, + "step": 126 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 1.2599350214004517, + "learning_rate": 4.997267352663514e-06, + "loss": 0.7385, + "step": 127 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.4353641271591187, + "learning_rate": 4.997061226255869e-06, + "loss": 0.7081, + "step": 128 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 3.2492141723632812, + "learning_rate": 4.996847610323437e-06, + "loss": 0.7859, + "step": 129 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 9.599719047546387, + "learning_rate": 4.996626505506923e-06, + "loss": 0.7241, + "step": 130 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 10.053650856018066, + "learning_rate": 4.996397912469494e-06, + "loss": 0.6841, + "step": 131 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 1.323876976966858, + "learning_rate": 4.996161831896777e-06, + "loss": 0.7317, + "step": 132 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 1.4180598258972168, + "learning_rate": 4.9959182644968594e-06, + "loss": 0.692, + "step": 133 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 1.2194396257400513, + "learning_rate": 4.99566721100028e-06, + "loss": 0.7068, + "step": 134 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0984960794448853, + "learning_rate": 4.995408672160031e-06, + "loss": 0.6946, + "step": 135 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 1.9341071844100952, + "learning_rate": 4.995142648751561e-06, + "loss": 0.7467, + "step": 136 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 1.9960932731628418, + "learning_rate": 4.9948691415727594e-06, + "loss": 0.7379, + "step": 137 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.8743917942047119, + "learning_rate": 4.994588151443968e-06, + "loss": 0.66, + "step": 138 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8655261993408203, + "learning_rate": 4.99429967920797e-06, + "loss": 0.6646, + "step": 139 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 5.462070941925049, + "learning_rate": 4.994003725729992e-06, + "loss": 0.643, + "step": 140 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 2.1401469707489014, + "learning_rate": 4.993700291897695e-06, + "loss": 0.6639, + "step": 141 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 1.8219833374023438, + "learning_rate": 4.9933893786211815e-06, + "loss": 0.6673, + "step": 142 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 1.641079306602478, + "learning_rate": 4.993070986832984e-06, + "loss": 0.658, + "step": 143 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 1.1739819049835205, + "learning_rate": 4.992745117488066e-06, + "loss": 0.6826, + "step": 144 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 2.309185743331909, + "learning_rate": 4.9924117715638185e-06, + "loss": 0.6536, + "step": 145 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 1.09304940700531, + "learning_rate": 4.99207095006006e-06, + "loss": 0.721, + "step": 146 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.9056984186172485, + "learning_rate": 4.991722653999025e-06, + "loss": 0.7019, + "step": 147 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.8440625667572021, + "learning_rate": 4.991366884425374e-06, + "loss": 0.707, + "step": 148 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 1.2244676351547241, + "learning_rate": 4.991003642406177e-06, + "loss": 0.6407, + "step": 149 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9258589744567871, + "learning_rate": 4.99063292903092e-06, + "loss": 0.6954, + "step": 150 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 4.176390647888184, + "learning_rate": 4.990254745411496e-06, + "loss": 0.6812, + "step": 151 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 1.4322530031204224, + "learning_rate": 4.989869092682205e-06, + "loss": 0.6808, + "step": 152 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.8017717003822327, + "learning_rate": 4.989475971999748e-06, + "loss": 0.687, + "step": 153 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 1.5641374588012695, + "learning_rate": 4.989075384543228e-06, + "loss": 0.6599, + "step": 154 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 1.1522141695022583, + "learning_rate": 4.98866733151414e-06, + "loss": 0.6546, + "step": 155 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8593171238899231, + "learning_rate": 4.988251814136372e-06, + "loss": 0.6857, + "step": 156 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 2.668159246444702, + "learning_rate": 4.9878288336562e-06, + "loss": 0.661, + "step": 157 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9953671097755432, + "learning_rate": 4.987398391342285e-06, + "loss": 0.6512, + "step": 158 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.042872667312622, + "learning_rate": 4.986960488485667e-06, + "loss": 0.6311, + "step": 159 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.9070663452148438, + "learning_rate": 4.9865151263997645e-06, + "loss": 0.675, + "step": 160 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.8460433483123779, + "learning_rate": 4.986062306420367e-06, + "loss": 0.6635, + "step": 161 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 1.2639834880828857, + "learning_rate": 4.985602029905635e-06, + "loss": 0.6327, + "step": 162 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.8775074481964111, + "learning_rate": 4.985134298236091e-06, + "loss": 0.644, + "step": 163 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 1.2031961679458618, + "learning_rate": 4.98465911281462e-06, + "loss": 0.6254, + "step": 164 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.892494797706604, + "learning_rate": 4.984176475066463e-06, + "loss": 0.7122, + "step": 165 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 2.7122485637664795, + "learning_rate": 4.983686386439212e-06, + "loss": 0.6679, + "step": 166 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9344426989555359, + "learning_rate": 4.983188848402806e-06, + "loss": 0.6319, + "step": 167 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 1.4093577861785889, + "learning_rate": 4.982683862449531e-06, + "loss": 0.6425, + "step": 168 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 1.1285009384155273, + "learning_rate": 4.982171430094007e-06, + "loss": 0.6298, + "step": 169 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.952778935432434, + "learning_rate": 4.981651552873193e-06, + "loss": 0.7066, + "step": 170 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 5.133765697479248, + "learning_rate": 4.981124232346374e-06, + "loss": 0.6634, + "step": 171 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9770542979240417, + "learning_rate": 4.980589470095161e-06, + "loss": 0.7121, + "step": 172 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.8414323925971985, + "learning_rate": 4.980047267723487e-06, + "loss": 0.6397, + "step": 173 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 1.9173879623413086, + "learning_rate": 4.979497626857596e-06, + "loss": 0.6228, + "step": 174 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 1.0823363065719604, + "learning_rate": 4.978940549146048e-06, + "loss": 0.6475, + "step": 175 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 3.715353488922119, + "learning_rate": 4.978376036259706e-06, + "loss": 0.7127, + "step": 176 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.981584370136261, + "learning_rate": 4.9778040898917325e-06, + "loss": 0.6468, + "step": 177 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.70566987991333, + "learning_rate": 4.977224711757587e-06, + "loss": 0.6476, + "step": 178 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9217923283576965, + "learning_rate": 4.976637903595019e-06, + "loss": 0.6731, + "step": 179 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.8994677662849426, + "learning_rate": 4.976043667164063e-06, + "loss": 0.6562, + "step": 180 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 1.1613017320632935, + "learning_rate": 4.975442004247034e-06, + "loss": 0.6417, + "step": 181 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 1.6041977405548096, + "learning_rate": 4.974832916648521e-06, + "loss": 0.6029, + "step": 182 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.7978405952453613, + "learning_rate": 4.974216406195383e-06, + "loss": 0.6269, + "step": 183 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.6021920442581177, + "learning_rate": 4.973592474736739e-06, + "loss": 0.6149, + "step": 184 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8973568677902222, + "learning_rate": 4.972961124143971e-06, + "loss": 0.6648, + "step": 185 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 1.9432591199874878, + "learning_rate": 4.972322356310711e-06, + "loss": 0.6299, + "step": 186 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 4.457028388977051, + "learning_rate": 4.971676173152839e-06, + "loss": 0.656, + "step": 187 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 2.0989716053009033, + "learning_rate": 4.971022576608473e-06, + "loss": 0.6539, + "step": 188 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 1.0646967887878418, + "learning_rate": 4.97036156863797e-06, + "loss": 0.6727, + "step": 189 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.6522265672683716, + "learning_rate": 4.969693151223914e-06, + "loss": 0.6643, + "step": 190 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.7503505945205688, + "learning_rate": 4.969017326371115e-06, + "loss": 0.6402, + "step": 191 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 1.2341989278793335, + "learning_rate": 4.968334096106597e-06, + "loss": 0.6413, + "step": 192 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 3.089054584503174, + "learning_rate": 4.967643462479597e-06, + "loss": 0.6825, + "step": 193 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 2.711623430252075, + "learning_rate": 4.966945427561557e-06, + "loss": 0.65, + "step": 194 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 4.641184329986572, + "learning_rate": 4.966239993446118e-06, + "loss": 0.6229, + "step": 195 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.7984074354171753, + "learning_rate": 4.965527162249114e-06, + "loss": 0.6473, + "step": 196 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 1.1643115282058716, + "learning_rate": 4.964806936108566e-06, + "loss": 0.6404, + "step": 197 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 2.1877920627593994, + "learning_rate": 4.9640793171846725e-06, + "loss": 0.6185, + "step": 198 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.7970566749572754, + "learning_rate": 4.963344307659807e-06, + "loss": 0.634, + "step": 199 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 1.6014361381530762, + "learning_rate": 4.96260190973851e-06, + "loss": 0.6562, + "step": 200 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.8743320107460022, + "learning_rate": 4.961852125647482e-06, + "loss": 0.6133, + "step": 201 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 1.9526551961898804, + "learning_rate": 4.961094957635578e-06, + "loss": 0.6451, + "step": 202 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 3.6597347259521484, + "learning_rate": 4.960330407973798e-06, + "loss": 0.6386, + "step": 203 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 1.7180207967758179, + "learning_rate": 4.959558478955283e-06, + "loss": 0.6688, + "step": 204 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 0.9058470129966736, + "learning_rate": 4.958779172895308e-06, + "loss": 0.6161, + "step": 205 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 1.0031033754348755, + "learning_rate": 4.957992492131274e-06, + "loss": 0.6437, + "step": 206 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 1.5846725702285767, + "learning_rate": 4.9571984390226985e-06, + "loss": 0.6332, + "step": 207 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 1.9951609373092651, + "learning_rate": 4.956397015951215e-06, + "loss": 0.636, + "step": 208 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 1.4122583866119385, + "learning_rate": 4.95558822532056e-06, + "loss": 0.6586, + "step": 209 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.2243481874465942, + "learning_rate": 4.954772069556568e-06, + "loss": 0.6313, + "step": 210 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.8756356835365295, + "learning_rate": 4.953948551107164e-06, + "loss": 0.6406, + "step": 211 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 2.9979734420776367, + "learning_rate": 4.953117672442356e-06, + "loss": 0.5803, + "step": 212 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 2.1859359741210938, + "learning_rate": 4.952279436054229e-06, + "loss": 0.6607, + "step": 213 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.6929755806922913, + "learning_rate": 4.9514338444569346e-06, + "loss": 0.5989, + "step": 214 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 1.0361783504486084, + "learning_rate": 4.950580900186685e-06, + "loss": 0.6654, + "step": 215 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 1.210898518562317, + "learning_rate": 4.9497206058017475e-06, + "loss": 0.6213, + "step": 216 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 1.200990080833435, + "learning_rate": 4.948852963882434e-06, + "loss": 0.6654, + "step": 217 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 1.481831669807434, + "learning_rate": 4.947977977031093e-06, + "loss": 0.6474, + "step": 218 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.9883334636688232, + "learning_rate": 4.947095647872103e-06, + "loss": 0.6735, + "step": 219 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.7436536550521851, + "learning_rate": 4.946205979051868e-06, + "loss": 0.6456, + "step": 220 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9057570099830627, + "learning_rate": 4.945308973238802e-06, + "loss": 0.6228, + "step": 221 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 1.341081142425537, + "learning_rate": 4.944404633123324e-06, + "loss": 0.6417, + "step": 222 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.7958157062530518, + "learning_rate": 4.943492961417859e-06, + "loss": 0.6494, + "step": 223 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.216025471687317, + "learning_rate": 4.9425739608568106e-06, + "loss": 0.6566, + "step": 224 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.9774854779243469, + "learning_rate": 4.9416476341965735e-06, + "loss": 0.6171, + "step": 225 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 2.1562681198120117, + "learning_rate": 4.940713984215512e-06, + "loss": 0.629, + "step": 226 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 1.9521286487579346, + "learning_rate": 4.9397730137139556e-06, + "loss": 0.6475, + "step": 227 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 1.5749104022979736, + "learning_rate": 4.9388247255141895e-06, + "loss": 0.6053, + "step": 228 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 1.2008254528045654, + "learning_rate": 4.937869122460449e-06, + "loss": 0.6052, + "step": 229 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 1.0774102210998535, + "learning_rate": 4.93690620741891e-06, + "loss": 0.6099, + "step": 230 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 1.0929996967315674, + "learning_rate": 4.935935983277675e-06, + "loss": 0.6363, + "step": 231 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.8830653429031372, + "learning_rate": 4.934958452946774e-06, + "loss": 0.6136, + "step": 232 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 3.591218948364258, + "learning_rate": 4.933973619358147e-06, + "loss": 0.5962, + "step": 233 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 2.5797672271728516, + "learning_rate": 4.932981485465643e-06, + "loss": 0.6405, + "step": 234 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 1.0467664003372192, + "learning_rate": 4.9319820542450025e-06, + "loss": 0.6155, + "step": 235 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.8099795579910278, + "learning_rate": 4.930975328693856e-06, + "loss": 0.5615, + "step": 236 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.8906702995300293, + "learning_rate": 4.92996131183171e-06, + "loss": 0.6501, + "step": 237 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 1.0871416330337524, + "learning_rate": 4.928940006699944e-06, + "loss": 0.6282, + "step": 238 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 1.3209614753723145, + "learning_rate": 4.927911416361792e-06, + "loss": 0.598, + "step": 239 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.2252682447433472, + "learning_rate": 4.926875543902344e-06, + "loss": 0.6433, + "step": 240 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0569007396697998, + "learning_rate": 4.9258323924285285e-06, + "loss": 0.5927, + "step": 241 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9309014081954956, + "learning_rate": 4.924781965069106e-06, + "loss": 0.5927, + "step": 242 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0200378894805908, + "learning_rate": 4.923724264974662e-06, + "loss": 0.6064, + "step": 243 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 1.0533075332641602, + "learning_rate": 4.922659295317593e-06, + "loss": 0.6373, + "step": 244 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 0.7889382839202881, + "learning_rate": 4.921587059292102e-06, + "loss": 0.5887, + "step": 245 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.7943588495254517, + "learning_rate": 4.920507560114183e-06, + "loss": 0.593, + "step": 246 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.8247205018997192, + "learning_rate": 4.919420801021617e-06, + "loss": 0.6151, + "step": 247 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 0.9979158043861389, + "learning_rate": 4.91832678527396e-06, + "loss": 0.6019, + "step": 248 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9346868991851807, + "learning_rate": 4.917225516152532e-06, + "loss": 0.6098, + "step": 249 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.7487881183624268, + "learning_rate": 4.916116996960408e-06, + "loss": 0.5965, + "step": 250 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.821576714515686, + "learning_rate": 4.915001231022411e-06, + "loss": 0.6483, + "step": 251 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 1.0413196086883545, + "learning_rate": 4.913878221685096e-06, + "loss": 0.6108, + "step": 252 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9560331702232361, + "learning_rate": 4.912747972316745e-06, + "loss": 0.5758, + "step": 253 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.8964638113975525, + "learning_rate": 4.911610486307356e-06, + "loss": 0.6432, + "step": 254 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.8418346047401428, + "learning_rate": 4.910465767068631e-06, + "loss": 0.6027, + "step": 255 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 1.792371153831482, + "learning_rate": 4.909313818033966e-06, + "loss": 0.6198, + "step": 256 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 1.036665439605713, + "learning_rate": 4.908154642658446e-06, + "loss": 0.6255, + "step": 257 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.7592151165008545, + "learning_rate": 4.906988244418823e-06, + "loss": 0.6035, + "step": 258 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.8843073844909668, + "learning_rate": 4.90581462681352e-06, + "loss": 0.6299, + "step": 259 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9489964246749878, + "learning_rate": 4.9046337933626086e-06, + "loss": 0.5869, + "step": 260 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.851691722869873, + "learning_rate": 4.903445747607806e-06, + "loss": 0.603, + "step": 261 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 1.3722106218338013, + "learning_rate": 4.902250493112458e-06, + "loss": 0.5939, + "step": 262 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 1.1002827882766724, + "learning_rate": 4.901048033461537e-06, + "loss": 0.6452, + "step": 263 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 0.8428632020950317, + "learning_rate": 4.89983837226162e-06, + "loss": 0.5956, + "step": 264 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.7666584849357605, + "learning_rate": 4.898621513140889e-06, + "loss": 0.6067, + "step": 265 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.8413611054420471, + "learning_rate": 4.897397459749113e-06, + "loss": 0.5985, + "step": 266 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 2.3374335765838623, + "learning_rate": 4.896166215757638e-06, + "loss": 0.5885, + "step": 267 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 2.236640214920044, + "learning_rate": 4.894927784859377e-06, + "loss": 0.6408, + "step": 268 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.9715856313705444, + "learning_rate": 4.893682170768802e-06, + "loss": 0.5954, + "step": 269 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0249912738800049, + "learning_rate": 4.892429377221928e-06, + "loss": 0.6186, + "step": 270 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 1.255426049232483, + "learning_rate": 4.891169407976302e-06, + "loss": 0.6351, + "step": 271 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.9339559674263, + "learning_rate": 4.889902266810995e-06, + "loss": 0.5944, + "step": 272 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 1.2473429441452026, + "learning_rate": 4.888627957526589e-06, + "loss": 0.544, + "step": 273 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 1.0589442253112793, + "learning_rate": 4.887346483945166e-06, + "loss": 0.5543, + "step": 274 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9844024777412415, + "learning_rate": 4.886057849910294e-06, + "loss": 0.5941, + "step": 275 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 2.88578200340271, + "learning_rate": 4.8847620592870196e-06, + "loss": 0.6124, + "step": 276 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 0.7496054172515869, + "learning_rate": 4.8834591159618524e-06, + "loss": 0.6006, + "step": 277 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 0.7403052449226379, + "learning_rate": 4.88214902384276e-06, + "loss": 0.5911, + "step": 278 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9003771543502808, + "learning_rate": 4.880831786859146e-06, + "loss": 0.6347, + "step": 279 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0345501899719238, + "learning_rate": 4.879507408961847e-06, + "loss": 0.6111, + "step": 280 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.4385879039764404, + "learning_rate": 4.878175894123116e-06, + "loss": 0.6454, + "step": 281 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.8469482064247131, + "learning_rate": 4.8768372463366145e-06, + "loss": 0.6163, + "step": 282 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.8859589695930481, + "learning_rate": 4.875491469617395e-06, + "loss": 0.6144, + "step": 283 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 1.8436834812164307, + "learning_rate": 4.874138568001895e-06, + "loss": 0.6275, + "step": 284 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.6646101474761963, + "learning_rate": 4.87277854554792e-06, + "loss": 0.615, + "step": 285 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 1.0070925951004028, + "learning_rate": 4.871411406334633e-06, + "loss": 0.5898, + "step": 286 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9785194993019104, + "learning_rate": 4.870037154462545e-06, + "loss": 0.5992, + "step": 287 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.7244889736175537, + "learning_rate": 4.868655794053497e-06, + "loss": 0.6078, + "step": 288 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.4496444463729858, + "learning_rate": 4.8672673292506535e-06, + "loss": 0.5855, + "step": 289 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 1.8514957427978516, + "learning_rate": 4.865871764218486e-06, + "loss": 0.5707, + "step": 290 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.8439773321151733, + "learning_rate": 4.864469103142763e-06, + "loss": 0.5562, + "step": 291 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.8146086931228638, + "learning_rate": 4.8630593502305355e-06, + "loss": 0.6161, + "step": 292 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.8920315504074097, + "learning_rate": 4.861642509710126e-06, + "loss": 0.6139, + "step": 293 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.4980088472366333, + "learning_rate": 4.860218585831116e-06, + "loss": 0.6187, + "step": 294 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 0.9910127520561218, + "learning_rate": 4.8587875828643285e-06, + "loss": 0.5852, + "step": 295 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.819600522518158, + "learning_rate": 4.857349505101823e-06, + "loss": 0.6172, + "step": 296 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 1.1059772968292236, + "learning_rate": 4.855904356856878e-06, + "loss": 0.5868, + "step": 297 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 1.2362196445465088, + "learning_rate": 4.854452142463977e-06, + "loss": 0.625, + "step": 298 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.9956470727920532, + "learning_rate": 4.852992866278799e-06, + "loss": 0.5923, + "step": 299 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 0.864109218120575, + "learning_rate": 4.851526532678203e-06, + "loss": 0.6315, + "step": 300 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 0.8900614380836487, + "learning_rate": 4.850053146060217e-06, + "loss": 0.6128, + "step": 301 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.927254855632782, + "learning_rate": 4.84857271084402e-06, + "loss": 0.5955, + "step": 302 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0046517848968506, + "learning_rate": 4.847085231469935e-06, + "loss": 0.6134, + "step": 303 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.734597384929657, + "learning_rate": 4.8455907123994125e-06, + "loss": 0.5927, + "step": 304 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.7338348031044006, + "learning_rate": 4.844089158115016e-06, + "loss": 0.5897, + "step": 305 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9163988828659058, + "learning_rate": 4.8425805731204106e-06, + "loss": 0.6051, + "step": 306 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 1.050246238708496, + "learning_rate": 4.84106496194035e-06, + "loss": 0.5751, + "step": 307 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 0.7637603878974915, + "learning_rate": 4.83954232912066e-06, + "loss": 0.5677, + "step": 308 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.7110525965690613, + "learning_rate": 4.838012679228229e-06, + "loss": 0.6051, + "step": 309 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.7662068605422974, + "learning_rate": 4.836476016850988e-06, + "loss": 0.59, + "step": 310 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 0.8907375335693359, + "learning_rate": 4.834932346597906e-06, + "loss": 0.5792, + "step": 311 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.8939849138259888, + "learning_rate": 4.833381673098966e-06, + "loss": 0.6062, + "step": 312 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 0.8878788948059082, + "learning_rate": 4.8318240010051595e-06, + "loss": 0.5694, + "step": 313 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 1.2523870468139648, + "learning_rate": 4.830259334988468e-06, + "loss": 0.5809, + "step": 314 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 1.0836797952651978, + "learning_rate": 4.82868767974185e-06, + "loss": 0.5949, + "step": 315 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.7985473871231079, + "learning_rate": 4.827109039979226e-06, + "loss": 0.6057, + "step": 316 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 1.042951226234436, + "learning_rate": 4.825523420435469e-06, + "loss": 0.6004, + "step": 317 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.7845115661621094, + "learning_rate": 4.823930825866381e-06, + "loss": 0.6161, + "step": 318 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.931854784488678, + "learning_rate": 4.82233126104869e-06, + "loss": 0.5912, + "step": 319 + }, + { + "epoch": 1.0, + "grad_norm": 0.931854784488678, + "learning_rate": 4.8207247307800275e-06, + "loss": 0.1914, + "step": 320 + }, + { + "epoch": 1.0031315240083507, + "grad_norm": 0.751028835773468, + "learning_rate": 4.819111239878916e-06, + "loss": 0.5802, + "step": 321 + }, + { + "epoch": 1.0062630480167014, + "grad_norm": 1.4943569898605347, + "learning_rate": 4.817490793184758e-06, + "loss": 0.613, + "step": 322 + }, + { + "epoch": 1.0093945720250521, + "grad_norm": 2.296318531036377, + "learning_rate": 4.815863395557816e-06, + "loss": 0.5453, + "step": 323 + }, + { + "epoch": 1.0125260960334028, + "grad_norm": 0.760101318359375, + "learning_rate": 4.814229051879202e-06, + "loss": 0.5302, + "step": 324 + }, + { + "epoch": 1.0156576200417538, + "grad_norm": 0.8145846128463745, + "learning_rate": 4.812587767050861e-06, + "loss": 0.5831, + "step": 325 + }, + { + "epoch": 1.0187891440501045, + "grad_norm": 0.9169796109199524, + "learning_rate": 4.8109395459955565e-06, + "loss": 0.5756, + "step": 326 + }, + { + "epoch": 1.0219206680584552, + "grad_norm": 0.8791524171829224, + "learning_rate": 4.809284393656858e-06, + "loss": 0.5988, + "step": 327 + }, + { + "epoch": 1.0250521920668059, + "grad_norm": 1.0184170007705688, + "learning_rate": 4.807622314999122e-06, + "loss": 0.5476, + "step": 328 + }, + { + "epoch": 1.0281837160751566, + "grad_norm": 0.8095184564590454, + "learning_rate": 4.8059533150074805e-06, + "loss": 0.5723, + "step": 329 + }, + { + "epoch": 1.0313152400835073, + "grad_norm": 0.7621930241584778, + "learning_rate": 4.804277398687826e-06, + "loss": 0.5841, + "step": 330 + }, + { + "epoch": 1.034446764091858, + "grad_norm": 3.729628324508667, + "learning_rate": 4.802594571066791e-06, + "loss": 0.5639, + "step": 331 + }, + { + "epoch": 1.0375782881002087, + "grad_norm": 1.6502974033355713, + "learning_rate": 4.800904837191743e-06, + "loss": 0.6024, + "step": 332 + }, + { + "epoch": 1.0407098121085594, + "grad_norm": 0.8031198978424072, + "learning_rate": 4.799208202130762e-06, + "loss": 0.5305, + "step": 333 + }, + { + "epoch": 1.0438413361169103, + "grad_norm": 0.939644992351532, + "learning_rate": 4.797504670972623e-06, + "loss": 0.5446, + "step": 334 + }, + { + "epoch": 1.046972860125261, + "grad_norm": 1.0589954853057861, + "learning_rate": 4.795794248826789e-06, + "loss": 0.5366, + "step": 335 + }, + { + "epoch": 1.0501043841336117, + "grad_norm": 0.9089614748954773, + "learning_rate": 4.794076940823391e-06, + "loss": 0.5795, + "step": 336 + }, + { + "epoch": 1.0532359081419624, + "grad_norm": 0.7732561230659485, + "learning_rate": 4.792352752113212e-06, + "loss": 0.5765, + "step": 337 + }, + { + "epoch": 1.0563674321503131, + "grad_norm": 1.811553955078125, + "learning_rate": 4.790621687867672e-06, + "loss": 0.561, + "step": 338 + }, + { + "epoch": 1.0594989561586639, + "grad_norm": 1.1930758953094482, + "learning_rate": 4.788883753278813e-06, + "loss": 0.5, + "step": 339 + }, + { + "epoch": 1.0626304801670146, + "grad_norm": 0.9551813006401062, + "learning_rate": 4.787138953559285e-06, + "loss": 0.5228, + "step": 340 + }, + { + "epoch": 1.0657620041753653, + "grad_norm": 0.9609586596488953, + "learning_rate": 4.785387293942329e-06, + "loss": 0.5827, + "step": 341 + }, + { + "epoch": 1.068893528183716, + "grad_norm": 0.8403449654579163, + "learning_rate": 4.783628779681759e-06, + "loss": 0.5585, + "step": 342 + }, + { + "epoch": 1.072025052192067, + "grad_norm": 0.9108251929283142, + "learning_rate": 4.7818634160519496e-06, + "loss": 0.6077, + "step": 343 + }, + { + "epoch": 1.0751565762004176, + "grad_norm": 0.9476898908615112, + "learning_rate": 4.780091208347819e-06, + "loss": 0.5493, + "step": 344 + }, + { + "epoch": 1.0782881002087683, + "grad_norm": 1.1943707466125488, + "learning_rate": 4.778312161884813e-06, + "loss": 0.5736, + "step": 345 + }, + { + "epoch": 1.081419624217119, + "grad_norm": 3.1342639923095703, + "learning_rate": 4.77652628199889e-06, + "loss": 0.5765, + "step": 346 + }, + { + "epoch": 1.0845511482254697, + "grad_norm": 2.7982125282287598, + "learning_rate": 4.7747335740465015e-06, + "loss": 0.6003, + "step": 347 + }, + { + "epoch": 1.0876826722338204, + "grad_norm": 1.5068914890289307, + "learning_rate": 4.7729340434045815e-06, + "loss": 0.5033, + "step": 348 + }, + { + "epoch": 1.0908141962421711, + "grad_norm": 0.8273429274559021, + "learning_rate": 4.771127695470527e-06, + "loss": 0.5309, + "step": 349 + }, + { + "epoch": 1.0939457202505218, + "grad_norm": 1.104974389076233, + "learning_rate": 4.76931453566218e-06, + "loss": 0.5244, + "step": 350 + }, + { + "epoch": 1.0970772442588728, + "grad_norm": 1.096509337425232, + "learning_rate": 4.7674945694178166e-06, + "loss": 0.5585, + "step": 351 + }, + { + "epoch": 1.1002087682672235, + "grad_norm": 1.0238200426101685, + "learning_rate": 4.765667802196127e-06, + "loss": 0.5589, + "step": 352 + }, + { + "epoch": 1.1033402922755742, + "grad_norm": 0.7515526413917542, + "learning_rate": 4.763834239476197e-06, + "loss": 0.5304, + "step": 353 + }, + { + "epoch": 1.1064718162839249, + "grad_norm": 1.0282566547393799, + "learning_rate": 4.761993886757499e-06, + "loss": 0.5476, + "step": 354 + }, + { + "epoch": 1.1096033402922756, + "grad_norm": 0.9962708950042725, + "learning_rate": 4.760146749559868e-06, + "loss": 0.5117, + "step": 355 + }, + { + "epoch": 1.1127348643006263, + "grad_norm": 0.7851671576499939, + "learning_rate": 4.758292833423488e-06, + "loss": 0.5542, + "step": 356 + }, + { + "epoch": 1.115866388308977, + "grad_norm": 0.8857759237289429, + "learning_rate": 4.756432143908876e-06, + "loss": 0.544, + "step": 357 + }, + { + "epoch": 1.1189979123173277, + "grad_norm": 0.9402740597724915, + "learning_rate": 4.7545646865968645e-06, + "loss": 0.5656, + "step": 358 + }, + { + "epoch": 1.1221294363256784, + "grad_norm": 0.8210407495498657, + "learning_rate": 4.752690467088584e-06, + "loss": 0.5733, + "step": 359 + }, + { + "epoch": 1.1252609603340291, + "grad_norm": 0.795684278011322, + "learning_rate": 4.750809491005449e-06, + "loss": 0.5678, + "step": 360 + }, + { + "epoch": 1.12839248434238, + "grad_norm": 0.8712463974952698, + "learning_rate": 4.748921763989139e-06, + "loss": 0.5777, + "step": 361 + }, + { + "epoch": 1.1315240083507307, + "grad_norm": 0.9810119867324829, + "learning_rate": 4.747027291701578e-06, + "loss": 0.5511, + "step": 362 + }, + { + "epoch": 1.1346555323590815, + "grad_norm": 0.81117844581604, + "learning_rate": 4.745126079824926e-06, + "loss": 0.5038, + "step": 363 + }, + { + "epoch": 1.1377870563674322, + "grad_norm": 0.7631494402885437, + "learning_rate": 4.743218134061556e-06, + "loss": 0.6272, + "step": 364 + }, + { + "epoch": 1.1409185803757829, + "grad_norm": 0.7601696252822876, + "learning_rate": 4.741303460134038e-06, + "loss": 0.571, + "step": 365 + }, + { + "epoch": 1.1440501043841336, + "grad_norm": 1.7977744340896606, + "learning_rate": 4.7393820637851205e-06, + "loss": 0.538, + "step": 366 + }, + { + "epoch": 1.1471816283924843, + "grad_norm": 2.022578001022339, + "learning_rate": 4.737453950777718e-06, + "loss": 0.5822, + "step": 367 + }, + { + "epoch": 1.150313152400835, + "grad_norm": 0.7586764693260193, + "learning_rate": 4.735519126894885e-06, + "loss": 0.5986, + "step": 368 + }, + { + "epoch": 1.153444676409186, + "grad_norm": 0.8970286846160889, + "learning_rate": 4.733577597939812e-06, + "loss": 0.542, + "step": 369 + }, + { + "epoch": 1.1565762004175366, + "grad_norm": 0.8546352982521057, + "learning_rate": 4.731629369735793e-06, + "loss": 0.5832, + "step": 370 + }, + { + "epoch": 1.1597077244258873, + "grad_norm": 0.9266164898872375, + "learning_rate": 4.72967444812622e-06, + "loss": 0.551, + "step": 371 + }, + { + "epoch": 1.162839248434238, + "grad_norm": 1.0413658618927002, + "learning_rate": 4.7277128389745595e-06, + "loss": 0.5866, + "step": 372 + }, + { + "epoch": 1.1659707724425887, + "grad_norm": 0.9312199950218201, + "learning_rate": 4.7257445481643334e-06, + "loss": 0.5723, + "step": 373 + }, + { + "epoch": 1.1691022964509394, + "grad_norm": 0.7389806509017944, + "learning_rate": 4.723769581599109e-06, + "loss": 0.5209, + "step": 374 + }, + { + "epoch": 1.1722338204592901, + "grad_norm": 3.053169012069702, + "learning_rate": 4.721787945202472e-06, + "loss": 0.6094, + "step": 375 + }, + { + "epoch": 1.1753653444676408, + "grad_norm": 1.288589596748352, + "learning_rate": 4.719799644918017e-06, + "loss": 0.5616, + "step": 376 + }, + { + "epoch": 1.1784968684759916, + "grad_norm": 0.7675042152404785, + "learning_rate": 4.717804686709323e-06, + "loss": 0.4963, + "step": 377 + }, + { + "epoch": 1.1816283924843423, + "grad_norm": 0.7246491312980652, + "learning_rate": 4.715803076559938e-06, + "loss": 0.5273, + "step": 378 + }, + { + "epoch": 1.1847599164926932, + "grad_norm": 0.8193361759185791, + "learning_rate": 4.713794820473366e-06, + "loss": 0.6107, + "step": 379 + }, + { + "epoch": 1.187891440501044, + "grad_norm": 0.9498510360717773, + "learning_rate": 4.711779924473037e-06, + "loss": 0.5421, + "step": 380 + }, + { + "epoch": 1.1910229645093946, + "grad_norm": 1.0479756593704224, + "learning_rate": 4.709758394602305e-06, + "loss": 0.5257, + "step": 381 + }, + { + "epoch": 1.1941544885177453, + "grad_norm": 0.907866895198822, + "learning_rate": 4.707730236924413e-06, + "loss": 0.5289, + "step": 382 + }, + { + "epoch": 1.197286012526096, + "grad_norm": 0.8861165642738342, + "learning_rate": 4.705695457522488e-06, + "loss": 0.5727, + "step": 383 + }, + { + "epoch": 1.2004175365344467, + "grad_norm": 0.7467761039733887, + "learning_rate": 4.703654062499516e-06, + "loss": 0.5602, + "step": 384 + }, + { + "epoch": 1.2035490605427974, + "grad_norm": 0.7456198334693909, + "learning_rate": 4.701606057978325e-06, + "loss": 0.5345, + "step": 385 + }, + { + "epoch": 1.2066805845511483, + "grad_norm": 1.9976060390472412, + "learning_rate": 4.699551450101571e-06, + "loss": 0.5504, + "step": 386 + }, + { + "epoch": 1.209812108559499, + "grad_norm": 1.5253807306289673, + "learning_rate": 4.697490245031709e-06, + "loss": 0.5568, + "step": 387 + }, + { + "epoch": 1.2129436325678498, + "grad_norm": 1.0786075592041016, + "learning_rate": 4.6954224489509885e-06, + "loss": 0.5564, + "step": 388 + }, + { + "epoch": 1.2160751565762005, + "grad_norm": 0.8385995030403137, + "learning_rate": 4.693348068061422e-06, + "loss": 0.5341, + "step": 389 + }, + { + "epoch": 1.2192066805845512, + "grad_norm": 0.8184949159622192, + "learning_rate": 4.691267108584774e-06, + "loss": 0.5614, + "step": 390 + }, + { + "epoch": 1.2223382045929019, + "grad_norm": 0.9964898824691772, + "learning_rate": 4.68917957676254e-06, + "loss": 0.5589, + "step": 391 + }, + { + "epoch": 1.2254697286012526, + "grad_norm": 1.0168914794921875, + "learning_rate": 4.687085478855931e-06, + "loss": 0.5892, + "step": 392 + }, + { + "epoch": 1.2286012526096033, + "grad_norm": 0.8841140866279602, + "learning_rate": 4.684984821145846e-06, + "loss": 0.5327, + "step": 393 + }, + { + "epoch": 1.231732776617954, + "grad_norm": 0.834431529045105, + "learning_rate": 4.682877609932866e-06, + "loss": 0.5594, + "step": 394 + }, + { + "epoch": 1.2348643006263047, + "grad_norm": 0.7256641983985901, + "learning_rate": 4.6807638515372234e-06, + "loss": 0.5443, + "step": 395 + }, + { + "epoch": 1.2379958246346556, + "grad_norm": 0.765096127986908, + "learning_rate": 4.678643552298788e-06, + "loss": 0.5439, + "step": 396 + }, + { + "epoch": 1.2411273486430063, + "grad_norm": 0.8760455846786499, + "learning_rate": 4.676516718577051e-06, + "loss": 0.5485, + "step": 397 + }, + { + "epoch": 1.244258872651357, + "grad_norm": 2.7111501693725586, + "learning_rate": 4.674383356751099e-06, + "loss": 0.5696, + "step": 398 + }, + { + "epoch": 1.2473903966597077, + "grad_norm": 1.0521738529205322, + "learning_rate": 4.672243473219601e-06, + "loss": 0.5503, + "step": 399 + }, + { + "epoch": 1.2505219206680585, + "grad_norm": 0.8909669518470764, + "learning_rate": 4.670097074400785e-06, + "loss": 0.5183, + "step": 400 + }, + { + "epoch": 1.2536534446764092, + "grad_norm": 0.7483847737312317, + "learning_rate": 4.667944166732424e-06, + "loss": 0.5669, + "step": 401 + }, + { + "epoch": 1.2567849686847599, + "grad_norm": 1.146997094154358, + "learning_rate": 4.665784756671808e-06, + "loss": 0.5464, + "step": 402 + }, + { + "epoch": 1.2599164926931106, + "grad_norm": 0.8998096585273743, + "learning_rate": 4.663618850695733e-06, + "loss": 0.5502, + "step": 403 + }, + { + "epoch": 1.2630480167014615, + "grad_norm": 0.8882688283920288, + "learning_rate": 4.6614464553004795e-06, + "loss": 0.5507, + "step": 404 + }, + { + "epoch": 1.2661795407098122, + "grad_norm": 0.8310684561729431, + "learning_rate": 4.659267577001789e-06, + "loss": 0.5164, + "step": 405 + }, + { + "epoch": 1.269311064718163, + "grad_norm": 0.9286114573478699, + "learning_rate": 4.657082222334851e-06, + "loss": 0.4813, + "step": 406 + }, + { + "epoch": 1.2724425887265136, + "grad_norm": 1.2394906282424927, + "learning_rate": 4.654890397854275e-06, + "loss": 0.5837, + "step": 407 + }, + { + "epoch": 1.2755741127348643, + "grad_norm": 4.00585412979126, + "learning_rate": 4.652692110134079e-06, + "loss": 0.5453, + "step": 408 + }, + { + "epoch": 1.278705636743215, + "grad_norm": 1.1667803525924683, + "learning_rate": 4.650487365767667e-06, + "loss": 0.5652, + "step": 409 + }, + { + "epoch": 1.2818371607515657, + "grad_norm": 0.9351289868354797, + "learning_rate": 4.648276171367807e-06, + "loss": 0.5576, + "step": 410 + }, + { + "epoch": 1.2849686847599164, + "grad_norm": 0.8107728958129883, + "learning_rate": 4.646058533566614e-06, + "loss": 0.5821, + "step": 411 + }, + { + "epoch": 1.2881002087682671, + "grad_norm": 0.7293011546134949, + "learning_rate": 4.643834459015525e-06, + "loss": 0.5363, + "step": 412 + }, + { + "epoch": 1.2912317327766178, + "grad_norm": 0.7550690770149231, + "learning_rate": 4.641603954385289e-06, + "loss": 0.53, + "step": 413 + }, + { + "epoch": 1.2943632567849686, + "grad_norm": 0.7626177072525024, + "learning_rate": 4.639367026365938e-06, + "loss": 0.5307, + "step": 414 + }, + { + "epoch": 1.2974947807933195, + "grad_norm": 1.0841096639633179, + "learning_rate": 4.637123681666769e-06, + "loss": 0.5162, + "step": 415 + }, + { + "epoch": 1.3006263048016702, + "grad_norm": 0.8814271092414856, + "learning_rate": 4.634873927016326e-06, + "loss": 0.5369, + "step": 416 + }, + { + "epoch": 1.303757828810021, + "grad_norm": 0.7402971982955933, + "learning_rate": 4.632617769162378e-06, + "loss": 0.5846, + "step": 417 + }, + { + "epoch": 1.3068893528183716, + "grad_norm": 0.8106061220169067, + "learning_rate": 4.6303552148719e-06, + "loss": 0.5289, + "step": 418 + }, + { + "epoch": 1.3100208768267223, + "grad_norm": 0.9241361618041992, + "learning_rate": 4.628086270931053e-06, + "loss": 0.5714, + "step": 419 + }, + { + "epoch": 1.313152400835073, + "grad_norm": 0.950332522392273, + "learning_rate": 4.625810944145159e-06, + "loss": 0.5817, + "step": 420 + }, + { + "epoch": 1.316283924843424, + "grad_norm": 0.9037718772888184, + "learning_rate": 4.623529241338689e-06, + "loss": 0.5902, + "step": 421 + }, + { + "epoch": 1.3194154488517746, + "grad_norm": 1.2110658884048462, + "learning_rate": 4.621241169355234e-06, + "loss": 0.561, + "step": 422 + }, + { + "epoch": 1.3225469728601253, + "grad_norm": 0.8582742214202881, + "learning_rate": 4.618946735057491e-06, + "loss": 0.5003, + "step": 423 + }, + { + "epoch": 1.325678496868476, + "grad_norm": 0.9203405976295471, + "learning_rate": 4.6166459453272386e-06, + "loss": 0.5639, + "step": 424 + }, + { + "epoch": 1.3288100208768268, + "grad_norm": 0.933721125125885, + "learning_rate": 4.614338807065317e-06, + "loss": 0.5766, + "step": 425 + }, + { + "epoch": 1.3319415448851775, + "grad_norm": 0.8435131311416626, + "learning_rate": 4.612025327191608e-06, + "loss": 0.5656, + "step": 426 + }, + { + "epoch": 1.3350730688935282, + "grad_norm": 0.795796811580658, + "learning_rate": 4.609705512645015e-06, + "loss": 0.4996, + "step": 427 + }, + { + "epoch": 1.3382045929018789, + "grad_norm": 0.8168228268623352, + "learning_rate": 4.6073793703834404e-06, + "loss": 0.5465, + "step": 428 + }, + { + "epoch": 1.3413361169102296, + "grad_norm": 0.8795569539070129, + "learning_rate": 4.605046907383765e-06, + "loss": 0.5407, + "step": 429 + }, + { + "epoch": 1.3444676409185803, + "grad_norm": 0.8504094481468201, + "learning_rate": 4.6027081306418295e-06, + "loss": 0.5589, + "step": 430 + }, + { + "epoch": 1.347599164926931, + "grad_norm": 1.485202431678772, + "learning_rate": 4.600363047172409e-06, + "loss": 0.5515, + "step": 431 + }, + { + "epoch": 1.350730688935282, + "grad_norm": 1.1156851053237915, + "learning_rate": 4.598011664009197e-06, + "loss": 0.5681, + "step": 432 + }, + { + "epoch": 1.3538622129436326, + "grad_norm": 0.8666876554489136, + "learning_rate": 4.595653988204779e-06, + "loss": 0.5451, + "step": 433 + }, + { + "epoch": 1.3569937369519833, + "grad_norm": 0.8192381858825684, + "learning_rate": 4.593290026830619e-06, + "loss": 0.5632, + "step": 434 + }, + { + "epoch": 1.360125260960334, + "grad_norm": 0.7994804978370667, + "learning_rate": 4.590919786977029e-06, + "loss": 0.5181, + "step": 435 + }, + { + "epoch": 1.3632567849686847, + "grad_norm": 0.8038607835769653, + "learning_rate": 4.5885432757531535e-06, + "loss": 0.5385, + "step": 436 + }, + { + "epoch": 1.3663883089770354, + "grad_norm": 0.7677503824234009, + "learning_rate": 4.586160500286948e-06, + "loss": 0.5455, + "step": 437 + }, + { + "epoch": 1.3695198329853862, + "grad_norm": 0.8293285369873047, + "learning_rate": 4.583771467725157e-06, + "loss": 0.5401, + "step": 438 + }, + { + "epoch": 1.372651356993737, + "grad_norm": 0.8607680797576904, + "learning_rate": 4.581376185233289e-06, + "loss": 0.5782, + "step": 439 + }, + { + "epoch": 1.3757828810020878, + "grad_norm": 0.8847081065177917, + "learning_rate": 4.578974659995601e-06, + "loss": 0.572, + "step": 440 + }, + { + "epoch": 1.3789144050104385, + "grad_norm": 0.7669641971588135, + "learning_rate": 4.576566899215075e-06, + "loss": 0.5655, + "step": 441 + }, + { + "epoch": 1.3820459290187892, + "grad_norm": 0.8514629006385803, + "learning_rate": 4.5741529101133904e-06, + "loss": 0.5218, + "step": 442 + }, + { + "epoch": 1.38517745302714, + "grad_norm": 0.8719842433929443, + "learning_rate": 4.5717326999309145e-06, + "loss": 0.5579, + "step": 443 + }, + { + "epoch": 1.3883089770354906, + "grad_norm": 1.1142809391021729, + "learning_rate": 4.569306275926667e-06, + "loss": 0.5535, + "step": 444 + }, + { + "epoch": 1.3914405010438413, + "grad_norm": 0.7392387986183167, + "learning_rate": 4.566873645378309e-06, + "loss": 0.5335, + "step": 445 + }, + { + "epoch": 1.394572025052192, + "grad_norm": 0.9066658616065979, + "learning_rate": 4.564434815582117e-06, + "loss": 0.5286, + "step": 446 + }, + { + "epoch": 1.3977035490605427, + "grad_norm": 0.8648932576179504, + "learning_rate": 4.561989793852959e-06, + "loss": 0.5008, + "step": 447 + }, + { + "epoch": 1.4008350730688934, + "grad_norm": 0.7768712043762207, + "learning_rate": 4.559538587524276e-06, + "loss": 0.5727, + "step": 448 + }, + { + "epoch": 1.4039665970772441, + "grad_norm": 0.7851182222366333, + "learning_rate": 4.557081203948059e-06, + "loss": 0.5731, + "step": 449 + }, + { + "epoch": 1.407098121085595, + "grad_norm": 0.8959861397743225, + "learning_rate": 4.5546176504948255e-06, + "loss": 0.5587, + "step": 450 + }, + { + "epoch": 1.4102296450939458, + "grad_norm": 1.0538026094436646, + "learning_rate": 4.552147934553601e-06, + "loss": 0.5808, + "step": 451 + }, + { + "epoch": 1.4133611691022965, + "grad_norm": 0.9887629151344299, + "learning_rate": 4.54967206353189e-06, + "loss": 0.5658, + "step": 452 + }, + { + "epoch": 1.4164926931106472, + "grad_norm": 0.9579302072525024, + "learning_rate": 4.547190044855663e-06, + "loss": 0.5092, + "step": 453 + }, + { + "epoch": 1.4196242171189979, + "grad_norm": 0.6993522047996521, + "learning_rate": 4.544701885969326e-06, + "loss": 0.5233, + "step": 454 + }, + { + "epoch": 1.4227557411273486, + "grad_norm": 0.8197568655014038, + "learning_rate": 4.542207594335703e-06, + "loss": 0.553, + "step": 455 + }, + { + "epoch": 1.4258872651356993, + "grad_norm": 2.921947717666626, + "learning_rate": 4.53970717743601e-06, + "loss": 0.4857, + "step": 456 + }, + { + "epoch": 1.4290187891440502, + "grad_norm": 1.3547242879867554, + "learning_rate": 4.53720064276984e-06, + "loss": 0.5676, + "step": 457 + }, + { + "epoch": 1.432150313152401, + "grad_norm": 1.4175567626953125, + "learning_rate": 4.534687997855131e-06, + "loss": 0.5164, + "step": 458 + }, + { + "epoch": 1.4352818371607516, + "grad_norm": 1.378146767616272, + "learning_rate": 4.532169250228145e-06, + "loss": 0.5429, + "step": 459 + }, + { + "epoch": 1.4384133611691023, + "grad_norm": 0.7811698317527771, + "learning_rate": 4.529644407443456e-06, + "loss": 0.524, + "step": 460 + }, + { + "epoch": 1.441544885177453, + "grad_norm": 1.1481678485870361, + "learning_rate": 4.527113477073914e-06, + "loss": 0.5513, + "step": 461 + }, + { + "epoch": 1.4446764091858038, + "grad_norm": 0.8450161218643188, + "learning_rate": 4.5245764667106266e-06, + "loss": 0.5632, + "step": 462 + }, + { + "epoch": 1.4478079331941545, + "grad_norm": 1.1582145690917969, + "learning_rate": 4.522033383962941e-06, + "loss": 0.5834, + "step": 463 + }, + { + "epoch": 1.4509394572025052, + "grad_norm": 1.0403447151184082, + "learning_rate": 4.519484236458416e-06, + "loss": 0.506, + "step": 464 + }, + { + "epoch": 1.4540709812108559, + "grad_norm": 0.7894920706748962, + "learning_rate": 4.516929031842799e-06, + "loss": 0.5526, + "step": 465 + }, + { + "epoch": 1.4572025052192066, + "grad_norm": 0.8092262744903564, + "learning_rate": 4.51436777778001e-06, + "loss": 0.5619, + "step": 466 + }, + { + "epoch": 1.4603340292275573, + "grad_norm": 0.9773806929588318, + "learning_rate": 4.511800481952106e-06, + "loss": 0.5179, + "step": 467 + }, + { + "epoch": 1.4634655532359082, + "grad_norm": 1.018676519393921, + "learning_rate": 4.509227152059271e-06, + "loss": 0.5415, + "step": 468 + }, + { + "epoch": 1.466597077244259, + "grad_norm": 0.7457838654518127, + "learning_rate": 4.506647795819784e-06, + "loss": 0.5473, + "step": 469 + }, + { + "epoch": 1.4697286012526096, + "grad_norm": 0.7826436161994934, + "learning_rate": 4.50406242097e-06, + "loss": 0.5526, + "step": 470 + }, + { + "epoch": 1.4728601252609603, + "grad_norm": 0.9492483139038086, + "learning_rate": 4.501471035264328e-06, + "loss": 0.5179, + "step": 471 + }, + { + "epoch": 1.475991649269311, + "grad_norm": 0.93398517370224, + "learning_rate": 4.4988736464752005e-06, + "loss": 0.5195, + "step": 472 + }, + { + "epoch": 1.4791231732776617, + "grad_norm": 0.8396487832069397, + "learning_rate": 4.496270262393061e-06, + "loss": 0.5447, + "step": 473 + }, + { + "epoch": 1.4822546972860124, + "grad_norm": 0.7450584173202515, + "learning_rate": 4.4936608908263315e-06, + "loss": 0.5207, + "step": 474 + }, + { + "epoch": 1.4853862212943634, + "grad_norm": 0.7887717485427856, + "learning_rate": 4.491045539601392e-06, + "loss": 0.523, + "step": 475 + }, + { + "epoch": 1.488517745302714, + "grad_norm": 1.2051388025283813, + "learning_rate": 4.48842421656256e-06, + "loss": 0.5402, + "step": 476 + }, + { + "epoch": 1.4916492693110648, + "grad_norm": 2.3103389739990234, + "learning_rate": 4.485796929572063e-06, + "loss": 0.5588, + "step": 477 + }, + { + "epoch": 1.4947807933194155, + "grad_norm": 0.7473112344741821, + "learning_rate": 4.483163686510016e-06, + "loss": 0.5731, + "step": 478 + }, + { + "epoch": 1.4979123173277662, + "grad_norm": 0.7545126676559448, + "learning_rate": 4.480524495274399e-06, + "loss": 0.5536, + "step": 479 + }, + { + "epoch": 1.501043841336117, + "grad_norm": 0.7801297903060913, + "learning_rate": 4.477879363781033e-06, + "loss": 0.5696, + "step": 480 + }, + { + "epoch": 1.5041753653444676, + "grad_norm": 0.7740563750267029, + "learning_rate": 4.475228299963554e-06, + "loss": 0.5526, + "step": 481 + }, + { + "epoch": 1.5073068893528183, + "grad_norm": 0.8600060343742371, + "learning_rate": 4.4725713117733936e-06, + "loss": 0.5051, + "step": 482 + }, + { + "epoch": 1.510438413361169, + "grad_norm": 0.6934283971786499, + "learning_rate": 4.46990840717975e-06, + "loss": 0.5564, + "step": 483 + }, + { + "epoch": 1.5135699373695197, + "grad_norm": 0.8927920460700989, + "learning_rate": 4.46723959416957e-06, + "loss": 0.5529, + "step": 484 + }, + { + "epoch": 1.5167014613778704, + "grad_norm": 0.9570988416671753, + "learning_rate": 4.464564880747517e-06, + "loss": 0.5661, + "step": 485 + }, + { + "epoch": 1.5198329853862211, + "grad_norm": 0.7229202389717102, + "learning_rate": 4.461884274935956e-06, + "loss": 0.5964, + "step": 486 + }, + { + "epoch": 1.522964509394572, + "grad_norm": 0.7367239594459534, + "learning_rate": 4.4591977847749225e-06, + "loss": 0.5455, + "step": 487 + }, + { + "epoch": 1.5260960334029228, + "grad_norm": 0.8062120676040649, + "learning_rate": 4.456505418322103e-06, + "loss": 0.5735, + "step": 488 + }, + { + "epoch": 1.5292275574112735, + "grad_norm": 0.8854482769966125, + "learning_rate": 4.453807183652808e-06, + "loss": 0.5421, + "step": 489 + }, + { + "epoch": 1.5323590814196242, + "grad_norm": 0.7518959045410156, + "learning_rate": 4.451103088859951e-06, + "loss": 0.5083, + "step": 490 + }, + { + "epoch": 1.535490605427975, + "grad_norm": 0.8621206879615784, + "learning_rate": 4.448393142054016e-06, + "loss": 0.4712, + "step": 491 + }, + { + "epoch": 1.5386221294363258, + "grad_norm": 1.0618741512298584, + "learning_rate": 4.445677351363046e-06, + "loss": 0.5808, + "step": 492 + }, + { + "epoch": 1.5417536534446765, + "grad_norm": 0.8261345028877258, + "learning_rate": 4.442955724932607e-06, + "loss": 0.5625, + "step": 493 + }, + { + "epoch": 1.5448851774530272, + "grad_norm": 0.7067139744758606, + "learning_rate": 4.440228270925772e-06, + "loss": 0.5661, + "step": 494 + }, + { + "epoch": 1.548016701461378, + "grad_norm": 0.9234416484832764, + "learning_rate": 4.437494997523091e-06, + "loss": 0.5428, + "step": 495 + }, + { + "epoch": 1.5511482254697286, + "grad_norm": 0.9273470044136047, + "learning_rate": 4.434755912922567e-06, + "loss": 0.5388, + "step": 496 + }, + { + "epoch": 1.5542797494780793, + "grad_norm": 1.0163263082504272, + "learning_rate": 4.4320110253396345e-06, + "loss": 0.5409, + "step": 497 + }, + { + "epoch": 1.55741127348643, + "grad_norm": 0.9542096853256226, + "learning_rate": 4.429260343007133e-06, + "loss": 0.5329, + "step": 498 + }, + { + "epoch": 1.5605427974947808, + "grad_norm": 0.8076801896095276, + "learning_rate": 4.426503874175283e-06, + "loss": 0.5616, + "step": 499 + }, + { + "epoch": 1.5636743215031315, + "grad_norm": 1.0063767433166504, + "learning_rate": 4.423741627111658e-06, + "loss": 0.5369, + "step": 500 + }, + { + "epoch": 1.5668058455114822, + "grad_norm": 1.040286898612976, + "learning_rate": 4.420973610101166e-06, + "loss": 0.5474, + "step": 501 + }, + { + "epoch": 1.5699373695198329, + "grad_norm": 0.7832860946655273, + "learning_rate": 4.4181998314460164e-06, + "loss": 0.5486, + "step": 502 + }, + { + "epoch": 1.5730688935281836, + "grad_norm": 0.8162257075309753, + "learning_rate": 4.415420299465706e-06, + "loss": 0.5054, + "step": 503 + }, + { + "epoch": 1.5762004175365343, + "grad_norm": 0.9108433127403259, + "learning_rate": 4.4126350224969814e-06, + "loss": 0.5399, + "step": 504 + }, + { + "epoch": 1.5793319415448852, + "grad_norm": 0.8002520799636841, + "learning_rate": 4.409844008893824e-06, + "loss": 0.5485, + "step": 505 + }, + { + "epoch": 1.582463465553236, + "grad_norm": 0.8543248772621155, + "learning_rate": 4.407047267027423e-06, + "loss": 0.4984, + "step": 506 + }, + { + "epoch": 1.5855949895615866, + "grad_norm": 0.7154155373573303, + "learning_rate": 4.404244805286141e-06, + "loss": 0.5392, + "step": 507 + }, + { + "epoch": 1.5887265135699373, + "grad_norm": 0.818553626537323, + "learning_rate": 4.401436632075504e-06, + "loss": 0.5178, + "step": 508 + }, + { + "epoch": 1.5918580375782883, + "grad_norm": 0.7535017728805542, + "learning_rate": 4.398622755818167e-06, + "loss": 0.5446, + "step": 509 + }, + { + "epoch": 1.594989561586639, + "grad_norm": 0.9328975677490234, + "learning_rate": 4.395803184953889e-06, + "loss": 0.5546, + "step": 510 + }, + { + "epoch": 1.5981210855949897, + "grad_norm": 0.7960026860237122, + "learning_rate": 4.392977927939508e-06, + "loss": 0.5451, + "step": 511 + }, + { + "epoch": 1.6012526096033404, + "grad_norm": 0.9686267971992493, + "learning_rate": 4.3901469932489195e-06, + "loss": 0.5198, + "step": 512 + }, + { + "epoch": 1.604384133611691, + "grad_norm": 0.903137743473053, + "learning_rate": 4.387310389373047e-06, + "loss": 0.5395, + "step": 513 + }, + { + "epoch": 1.6075156576200418, + "grad_norm": 1.0728516578674316, + "learning_rate": 4.384468124819816e-06, + "loss": 0.5843, + "step": 514 + }, + { + "epoch": 1.6106471816283925, + "grad_norm": 1.0245436429977417, + "learning_rate": 4.3816202081141345e-06, + "loss": 0.5672, + "step": 515 + }, + { + "epoch": 1.6137787056367432, + "grad_norm": 0.9672732353210449, + "learning_rate": 4.378766647797858e-06, + "loss": 0.5369, + "step": 516 + }, + { + "epoch": 1.616910229645094, + "grad_norm": 0.9149513840675354, + "learning_rate": 4.375907452429774e-06, + "loss": 0.4628, + "step": 517 + }, + { + "epoch": 1.6200417536534446, + "grad_norm": 0.7543843984603882, + "learning_rate": 4.373042630585567e-06, + "loss": 0.5344, + "step": 518 + }, + { + "epoch": 1.6231732776617953, + "grad_norm": 0.7589017152786255, + "learning_rate": 4.370172190857801e-06, + "loss": 0.5672, + "step": 519 + }, + { + "epoch": 1.626304801670146, + "grad_norm": 0.803040623664856, + "learning_rate": 4.367296141855887e-06, + "loss": 0.5313, + "step": 520 + }, + { + "epoch": 1.6294363256784967, + "grad_norm": 0.8305794596672058, + "learning_rate": 4.3644144922060625e-06, + "loss": 0.5754, + "step": 521 + }, + { + "epoch": 1.6325678496868476, + "grad_norm": 1.0086486339569092, + "learning_rate": 4.361527250551361e-06, + "loss": 0.5433, + "step": 522 + }, + { + "epoch": 1.6356993736951984, + "grad_norm": 0.7217550277709961, + "learning_rate": 4.35863442555159e-06, + "loss": 0.524, + "step": 523 + }, + { + "epoch": 1.638830897703549, + "grad_norm": 0.7788524627685547, + "learning_rate": 4.355736025883303e-06, + "loss": 0.536, + "step": 524 + }, + { + "epoch": 1.6419624217118998, + "grad_norm": 0.8460550904273987, + "learning_rate": 4.352832060239774e-06, + "loss": 0.5381, + "step": 525 + }, + { + "epoch": 1.6450939457202505, + "grad_norm": 0.7571215033531189, + "learning_rate": 4.3499225373309675e-06, + "loss": 0.541, + "step": 526 + }, + { + "epoch": 1.6482254697286014, + "grad_norm": 0.7343226671218872, + "learning_rate": 4.347007465883523e-06, + "loss": 0.5147, + "step": 527 + }, + { + "epoch": 1.651356993736952, + "grad_norm": 0.7271892428398132, + "learning_rate": 4.3440868546407165e-06, + "loss": 0.5311, + "step": 528 + }, + { + "epoch": 1.6544885177453028, + "grad_norm": 0.8166136741638184, + "learning_rate": 4.341160712362442e-06, + "loss": 0.5379, + "step": 529 + }, + { + "epoch": 1.6576200417536535, + "grad_norm": 1.5985233783721924, + "learning_rate": 4.338229047825182e-06, + "loss": 0.5782, + "step": 530 + }, + { + "epoch": 1.6607515657620042, + "grad_norm": 0.7835702896118164, + "learning_rate": 4.3352918698219835e-06, + "loss": 0.525, + "step": 531 + }, + { + "epoch": 1.663883089770355, + "grad_norm": 0.7278687953948975, + "learning_rate": 4.332349187162428e-06, + "loss": 0.5266, + "step": 532 + }, + { + "epoch": 1.6670146137787056, + "grad_norm": 0.8240190148353577, + "learning_rate": 4.329401008672608e-06, + "loss": 0.5515, + "step": 533 + }, + { + "epoch": 1.6701461377870563, + "grad_norm": 0.9447080492973328, + "learning_rate": 4.326447343195102e-06, + "loss": 0.5596, + "step": 534 + }, + { + "epoch": 1.673277661795407, + "grad_norm": 0.7827372550964355, + "learning_rate": 4.323488199588944e-06, + "loss": 0.5466, + "step": 535 + }, + { + "epoch": 1.6764091858037578, + "grad_norm": 0.9252517223358154, + "learning_rate": 4.320523586729599e-06, + "loss": 0.5433, + "step": 536 + }, + { + "epoch": 1.6795407098121085, + "grad_norm": 0.9437504410743713, + "learning_rate": 4.317553513508934e-06, + "loss": 0.5552, + "step": 537 + }, + { + "epoch": 1.6826722338204592, + "grad_norm": 0.8972746133804321, + "learning_rate": 4.3145779888351986e-06, + "loss": 0.5259, + "step": 538 + }, + { + "epoch": 1.6858037578288099, + "grad_norm": 0.8017446994781494, + "learning_rate": 4.311597021632988e-06, + "loss": 0.5263, + "step": 539 + }, + { + "epoch": 1.6889352818371608, + "grad_norm": 0.7875497341156006, + "learning_rate": 4.3086106208432235e-06, + "loss": 0.5316, + "step": 540 + }, + { + "epoch": 1.6920668058455115, + "grad_norm": 0.8204905986785889, + "learning_rate": 4.305618795423125e-06, + "loss": 0.5506, + "step": 541 + }, + { + "epoch": 1.6951983298538622, + "grad_norm": 0.888359785079956, + "learning_rate": 4.30262155434618e-06, + "loss": 0.4825, + "step": 542 + }, + { + "epoch": 1.698329853862213, + "grad_norm": 1.1026058197021484, + "learning_rate": 4.29961890660212e-06, + "loss": 0.5321, + "step": 543 + }, + { + "epoch": 1.7014613778705638, + "grad_norm": 0.7662535905838013, + "learning_rate": 4.2966108611968945e-06, + "loss": 0.5432, + "step": 544 + }, + { + "epoch": 1.7045929018789145, + "grad_norm": 1.1951749324798584, + "learning_rate": 4.293597427152641e-06, + "loss": 0.5123, + "step": 545 + }, + { + "epoch": 1.7077244258872653, + "grad_norm": 1.303183913230896, + "learning_rate": 4.290578613507661e-06, + "loss": 0.5346, + "step": 546 + }, + { + "epoch": 1.710855949895616, + "grad_norm": 0.7653357982635498, + "learning_rate": 4.287554429316387e-06, + "loss": 0.5397, + "step": 547 + }, + { + "epoch": 1.7139874739039667, + "grad_norm": 0.796215295791626, + "learning_rate": 4.284524883649366e-06, + "loss": 0.5421, + "step": 548 + }, + { + "epoch": 1.7171189979123174, + "grad_norm": 0.7599332332611084, + "learning_rate": 4.281489985593219e-06, + "loss": 0.5289, + "step": 549 + }, + { + "epoch": 1.720250521920668, + "grad_norm": 0.8029115796089172, + "learning_rate": 4.2784497442506265e-06, + "loss": 0.5409, + "step": 550 + }, + { + "epoch": 1.7233820459290188, + "grad_norm": 0.7194099426269531, + "learning_rate": 4.275404168740291e-06, + "loss": 0.5327, + "step": 551 + }, + { + "epoch": 1.7265135699373695, + "grad_norm": 0.7960740923881531, + "learning_rate": 4.272353268196917e-06, + "loss": 0.4896, + "step": 552 + }, + { + "epoch": 1.7296450939457202, + "grad_norm": 0.9572116732597351, + "learning_rate": 4.269297051771178e-06, + "loss": 0.5402, + "step": 553 + }, + { + "epoch": 1.732776617954071, + "grad_norm": 1.3604938983917236, + "learning_rate": 4.266235528629695e-06, + "loss": 0.5792, + "step": 554 + }, + { + "epoch": 1.7359081419624216, + "grad_norm": 2.067286729812622, + "learning_rate": 4.263168707955002e-06, + "loss": 0.5033, + "step": 555 + }, + { + "epoch": 1.7390396659707723, + "grad_norm": 0.8031097054481506, + "learning_rate": 4.260096598945523e-06, + "loss": 0.5117, + "step": 556 + }, + { + "epoch": 1.742171189979123, + "grad_norm": 1.0241729021072388, + "learning_rate": 4.257019210815546e-06, + "loss": 0.5359, + "step": 557 + }, + { + "epoch": 1.745302713987474, + "grad_norm": 0.7625218629837036, + "learning_rate": 4.25393655279519e-06, + "loss": 0.5625, + "step": 558 + }, + { + "epoch": 1.7484342379958246, + "grad_norm": 0.8603503704071045, + "learning_rate": 4.250848634130381e-06, + "loss": 0.5043, + "step": 559 + }, + { + "epoch": 1.7515657620041754, + "grad_norm": 0.9543750286102295, + "learning_rate": 4.247755464082824e-06, + "loss": 0.5364, + "step": 560 + }, + { + "epoch": 1.754697286012526, + "grad_norm": 0.9707463979721069, + "learning_rate": 4.244657051929973e-06, + "loss": 0.5184, + "step": 561 + }, + { + "epoch": 1.757828810020877, + "grad_norm": 0.7491432428359985, + "learning_rate": 4.241553406965008e-06, + "loss": 0.559, + "step": 562 + }, + { + "epoch": 1.7609603340292277, + "grad_norm": 0.7444972991943359, + "learning_rate": 4.238444538496801e-06, + "loss": 0.5327, + "step": 563 + }, + { + "epoch": 1.7640918580375784, + "grad_norm": 2.7108678817749023, + "learning_rate": 4.235330455849892e-06, + "loss": 0.55, + "step": 564 + }, + { + "epoch": 1.767223382045929, + "grad_norm": 1.6716049909591675, + "learning_rate": 4.232211168364459e-06, + "loss": 0.5093, + "step": 565 + }, + { + "epoch": 1.7703549060542798, + "grad_norm": 0.7023475170135498, + "learning_rate": 4.229086685396295e-06, + "loss": 0.569, + "step": 566 + }, + { + "epoch": 1.7734864300626305, + "grad_norm": 0.8596265316009521, + "learning_rate": 4.225957016316771e-06, + "loss": 0.5128, + "step": 567 + }, + { + "epoch": 1.7766179540709812, + "grad_norm": 0.8110849857330322, + "learning_rate": 4.222822170512816e-06, + "loss": 0.5142, + "step": 568 + }, + { + "epoch": 1.779749478079332, + "grad_norm": 0.7583725452423096, + "learning_rate": 4.219682157386884e-06, + "loss": 0.5584, + "step": 569 + }, + { + "epoch": 1.7828810020876826, + "grad_norm": 0.787811279296875, + "learning_rate": 4.21653698635693e-06, + "loss": 0.5068, + "step": 570 + }, + { + "epoch": 1.7860125260960333, + "grad_norm": 0.8298993110656738, + "learning_rate": 4.213386666856375e-06, + "loss": 0.5496, + "step": 571 + }, + { + "epoch": 1.789144050104384, + "grad_norm": 0.8999841213226318, + "learning_rate": 4.210231208334087e-06, + "loss": 0.5454, + "step": 572 + }, + { + "epoch": 1.7922755741127347, + "grad_norm": 4.264521598815918, + "learning_rate": 4.207070620254345e-06, + "loss": 0.5486, + "step": 573 + }, + { + "epoch": 1.7954070981210855, + "grad_norm": 0.8517448306083679, + "learning_rate": 4.203904912096812e-06, + "loss": 0.5566, + "step": 574 + }, + { + "epoch": 1.7985386221294362, + "grad_norm": 0.9230182766914368, + "learning_rate": 4.200734093356511e-06, + "loss": 0.4964, + "step": 575 + }, + { + "epoch": 1.801670146137787, + "grad_norm": 1.224039912223816, + "learning_rate": 4.197558173543791e-06, + "loss": 0.5356, + "step": 576 + }, + { + "epoch": 1.8048016701461378, + "grad_norm": 0.9998573660850525, + "learning_rate": 4.194377162184301e-06, + "loss": 0.5334, + "step": 577 + }, + { + "epoch": 1.8079331941544885, + "grad_norm": 0.865521252155304, + "learning_rate": 4.191191068818963e-06, + "loss": 0.5036, + "step": 578 + }, + { + "epoch": 1.8110647181628392, + "grad_norm": 0.8048138618469238, + "learning_rate": 4.18799990300394e-06, + "loss": 0.4979, + "step": 579 + }, + { + "epoch": 1.8141962421711901, + "grad_norm": 0.717815637588501, + "learning_rate": 4.184803674310609e-06, + "loss": 0.5623, + "step": 580 + }, + { + "epoch": 1.8173277661795408, + "grad_norm": 0.8403327465057373, + "learning_rate": 4.1816023923255335e-06, + "loss": 0.5055, + "step": 581 + }, + { + "epoch": 1.8204592901878915, + "grad_norm": 0.7298995852470398, + "learning_rate": 4.178396066650432e-06, + "loss": 0.5641, + "step": 582 + }, + { + "epoch": 1.8235908141962422, + "grad_norm": 0.9469727873802185, + "learning_rate": 4.1751847069021516e-06, + "loss": 0.5557, + "step": 583 + }, + { + "epoch": 1.826722338204593, + "grad_norm": 0.8641784191131592, + "learning_rate": 4.1719683227126386e-06, + "loss": 0.5153, + "step": 584 + }, + { + "epoch": 1.8298538622129437, + "grad_norm": 0.7316668629646301, + "learning_rate": 4.168746923728908e-06, + "loss": 0.4988, + "step": 585 + }, + { + "epoch": 1.8329853862212944, + "grad_norm": 0.8795468807220459, + "learning_rate": 4.165520519613017e-06, + "loss": 0.5483, + "step": 586 + }, + { + "epoch": 1.836116910229645, + "grad_norm": 0.7323560118675232, + "learning_rate": 4.162289120042034e-06, + "loss": 0.5194, + "step": 587 + }, + { + "epoch": 1.8392484342379958, + "grad_norm": 0.8217021822929382, + "learning_rate": 4.159052734708013e-06, + "loss": 0.532, + "step": 588 + }, + { + "epoch": 1.8423799582463465, + "grad_norm": 0.7669674754142761, + "learning_rate": 4.155811373317958e-06, + "loss": 0.541, + "step": 589 + }, + { + "epoch": 1.8455114822546972, + "grad_norm": 0.8312156200408936, + "learning_rate": 4.152565045593801e-06, + "loss": 0.5298, + "step": 590 + }, + { + "epoch": 1.848643006263048, + "grad_norm": 0.8967565298080444, + "learning_rate": 4.1493137612723665e-06, + "loss": 0.51, + "step": 591 + }, + { + "epoch": 1.8517745302713986, + "grad_norm": 0.8706664443016052, + "learning_rate": 4.14605753010535e-06, + "loss": 0.4941, + "step": 592 + }, + { + "epoch": 1.8549060542797495, + "grad_norm": 0.7585753798484802, + "learning_rate": 4.14279636185928e-06, + "loss": 0.5161, + "step": 593 + }, + { + "epoch": 1.8580375782881002, + "grad_norm": 0.7495241165161133, + "learning_rate": 4.1395302663154954e-06, + "loss": 0.5388, + "step": 594 + }, + { + "epoch": 1.861169102296451, + "grad_norm": 1.0746862888336182, + "learning_rate": 4.136259253270114e-06, + "loss": 0.4976, + "step": 595 + }, + { + "epoch": 1.8643006263048016, + "grad_norm": 0.872309684753418, + "learning_rate": 4.132983332534e-06, + "loss": 0.559, + "step": 596 + }, + { + "epoch": 1.8674321503131524, + "grad_norm": 0.8759891986846924, + "learning_rate": 4.1297025139327405e-06, + "loss": 0.5436, + "step": 597 + }, + { + "epoch": 1.8705636743215033, + "grad_norm": 1.1044493913650513, + "learning_rate": 4.126416807306611e-06, + "loss": 0.5476, + "step": 598 + }, + { + "epoch": 1.873695198329854, + "grad_norm": 0.8340442180633545, + "learning_rate": 4.123126222510549e-06, + "loss": 0.4592, + "step": 599 + }, + { + "epoch": 1.8768267223382047, + "grad_norm": 0.8331449031829834, + "learning_rate": 4.119830769414123e-06, + "loss": 0.5219, + "step": 600 + }, + { + "epoch": 1.8799582463465554, + "grad_norm": 1.0862973928451538, + "learning_rate": 4.116530457901503e-06, + "loss": 0.5159, + "step": 601 + }, + { + "epoch": 1.883089770354906, + "grad_norm": 0.8524414300918579, + "learning_rate": 4.113225297871431e-06, + "loss": 0.5502, + "step": 602 + }, + { + "epoch": 1.8862212943632568, + "grad_norm": 1.4945416450500488, + "learning_rate": 4.10991529923719e-06, + "loss": 0.5627, + "step": 603 + }, + { + "epoch": 1.8893528183716075, + "grad_norm": 1.5518157482147217, + "learning_rate": 4.10660047192658e-06, + "loss": 0.5517, + "step": 604 + }, + { + "epoch": 1.8924843423799582, + "grad_norm": 2.56638765335083, + "learning_rate": 4.103280825881878e-06, + "loss": 0.5422, + "step": 605 + }, + { + "epoch": 1.895615866388309, + "grad_norm": 0.867254912853241, + "learning_rate": 4.099956371059817e-06, + "loss": 0.4991, + "step": 606 + }, + { + "epoch": 1.8987473903966596, + "grad_norm": 0.9555892944335938, + "learning_rate": 4.096627117431554e-06, + "loss": 0.5339, + "step": 607 + }, + { + "epoch": 1.9018789144050103, + "grad_norm": 0.7905483245849609, + "learning_rate": 4.093293074982638e-06, + "loss": 0.5168, + "step": 608 + }, + { + "epoch": 1.905010438413361, + "grad_norm": 0.7500227093696594, + "learning_rate": 4.089954253712981e-06, + "loss": 0.5096, + "step": 609 + }, + { + "epoch": 1.9081419624217117, + "grad_norm": 0.8458324074745178, + "learning_rate": 4.086610663636828e-06, + "loss": 0.5296, + "step": 610 + }, + { + "epoch": 1.9112734864300627, + "grad_norm": 0.7392706871032715, + "learning_rate": 4.08326231478273e-06, + "loss": 0.5305, + "step": 611 + }, + { + "epoch": 1.9144050104384134, + "grad_norm": 0.8113343715667725, + "learning_rate": 4.079909217193508e-06, + "loss": 0.5044, + "step": 612 + }, + { + "epoch": 1.917536534446764, + "grad_norm": 0.7637801766395569, + "learning_rate": 4.076551380926226e-06, + "loss": 0.5298, + "step": 613 + }, + { + "epoch": 1.9206680584551148, + "grad_norm": 1.0523375272750854, + "learning_rate": 4.073188816052164e-06, + "loss": 0.5111, + "step": 614 + }, + { + "epoch": 1.9237995824634657, + "grad_norm": 0.8224868774414062, + "learning_rate": 4.069821532656781e-06, + "loss": 0.5178, + "step": 615 + }, + { + "epoch": 1.9269311064718164, + "grad_norm": 0.7270777821540833, + "learning_rate": 4.066449540839693e-06, + "loss": 0.5307, + "step": 616 + }, + { + "epoch": 1.9300626304801671, + "grad_norm": 0.7214602828025818, + "learning_rate": 4.063072850714631e-06, + "loss": 0.5171, + "step": 617 + }, + { + "epoch": 1.9331941544885178, + "grad_norm": 0.7333671450614929, + "learning_rate": 4.059691472409426e-06, + "loss": 0.56, + "step": 618 + }, + { + "epoch": 1.9363256784968685, + "grad_norm": 0.9166824221611023, + "learning_rate": 4.056305416065964e-06, + "loss": 0.5388, + "step": 619 + }, + { + "epoch": 1.9394572025052192, + "grad_norm": 0.7743303775787354, + "learning_rate": 4.052914691840167e-06, + "loss": 0.5134, + "step": 620 + }, + { + "epoch": 1.94258872651357, + "grad_norm": 0.704097330570221, + "learning_rate": 4.0495193099019524e-06, + "loss": 0.4926, + "step": 621 + }, + { + "epoch": 1.9457202505219207, + "grad_norm": 0.8508503437042236, + "learning_rate": 4.046119280435212e-06, + "loss": 0.5008, + "step": 622 + }, + { + "epoch": 1.9488517745302714, + "grad_norm": 0.725933313369751, + "learning_rate": 4.042714613637775e-06, + "loss": 0.5549, + "step": 623 + }, + { + "epoch": 1.951983298538622, + "grad_norm": 0.8919175863265991, + "learning_rate": 4.039305319721381e-06, + "loss": 0.5183, + "step": 624 + }, + { + "epoch": 1.9551148225469728, + "grad_norm": 0.827919065952301, + "learning_rate": 4.035891408911644e-06, + "loss": 0.5624, + "step": 625 + }, + { + "epoch": 1.9582463465553235, + "grad_norm": 0.7415187358856201, + "learning_rate": 4.032472891448032e-06, + "loss": 0.5454, + "step": 626 + }, + { + "epoch": 1.9613778705636742, + "grad_norm": 0.7675788998603821, + "learning_rate": 4.029049777583824e-06, + "loss": 0.5361, + "step": 627 + }, + { + "epoch": 1.964509394572025, + "grad_norm": 0.8464030623435974, + "learning_rate": 4.025622077586088e-06, + "loss": 0.5295, + "step": 628 + }, + { + "epoch": 1.9676409185803758, + "grad_norm": 0.7641633749008179, + "learning_rate": 4.022189801735646e-06, + "loss": 0.55, + "step": 629 + }, + { + "epoch": 1.9707724425887265, + "grad_norm": 0.7813227772712708, + "learning_rate": 4.018752960327048e-06, + "loss": 0.5587, + "step": 630 + }, + { + "epoch": 1.9739039665970772, + "grad_norm": 0.7576701641082764, + "learning_rate": 4.015311563668533e-06, + "loss": 0.5413, + "step": 631 + }, + { + "epoch": 1.977035490605428, + "grad_norm": 0.6949650049209595, + "learning_rate": 4.011865622082004e-06, + "loss": 0.5344, + "step": 632 + }, + { + "epoch": 1.9801670146137789, + "grad_norm": 0.9009145498275757, + "learning_rate": 4.008415145902997e-06, + "loss": 0.5233, + "step": 633 + }, + { + "epoch": 1.9832985386221296, + "grad_norm": 0.7635822892189026, + "learning_rate": 4.004960145480651e-06, + "loss": 0.4981, + "step": 634 + }, + { + "epoch": 1.9864300626304803, + "grad_norm": 0.8916334509849548, + "learning_rate": 4.0015006311776685e-06, + "loss": 0.5311, + "step": 635 + }, + { + "epoch": 1.989561586638831, + "grad_norm": 0.7197673320770264, + "learning_rate": 3.998036613370295e-06, + "loss": 0.5361, + "step": 636 + }, + { + "epoch": 1.9926931106471817, + "grad_norm": 0.8391228914260864, + "learning_rate": 3.994568102448284e-06, + "loss": 0.5473, + "step": 637 + }, + { + "epoch": 1.9958246346555324, + "grad_norm": 0.9371750950813293, + "learning_rate": 3.991095108814862e-06, + "loss": 0.5303, + "step": 638 + }, + { + "epoch": 1.998956158663883, + "grad_norm": 0.8929619789123535, + "learning_rate": 3.9876176428867046e-06, + "loss": 0.533, + "step": 639 + }, + { + "epoch": 2.0, + "grad_norm": 0.8929619789123535, + "learning_rate": 3.9841357150938984e-06, + "loss": 0.1831, + "step": 640 + }, + { + "epoch": 2.0031315240083507, + "grad_norm": 0.8802503347396851, + "learning_rate": 3.9806493358799135e-06, + "loss": 0.493, + "step": 641 + }, + { + "epoch": 2.0062630480167014, + "grad_norm": 0.802759051322937, + "learning_rate": 3.977158515701571e-06, + "loss": 0.498, + "step": 642 + }, + { + "epoch": 2.009394572025052, + "grad_norm": 1.0235401391983032, + "learning_rate": 3.973663265029013e-06, + "loss": 0.4887, + "step": 643 + }, + { + "epoch": 2.012526096033403, + "grad_norm": 0.7219089865684509, + "learning_rate": 3.97016359434567e-06, + "loss": 0.4628, + "step": 644 + }, + { + "epoch": 2.0156576200417535, + "grad_norm": 0.7887073755264282, + "learning_rate": 3.966659514148229e-06, + "loss": 0.525, + "step": 645 + }, + { + "epoch": 2.0187891440501042, + "grad_norm": 0.7960914969444275, + "learning_rate": 3.963151034946602e-06, + "loss": 0.4643, + "step": 646 + }, + { + "epoch": 2.021920668058455, + "grad_norm": 0.7902271151542664, + "learning_rate": 3.959638167263895e-06, + "loss": 0.4922, + "step": 647 + }, + { + "epoch": 2.0250521920668056, + "grad_norm": 0.9501478672027588, + "learning_rate": 3.956120921636379e-06, + "loss": 0.5285, + "step": 648 + }, + { + "epoch": 2.028183716075157, + "grad_norm": 0.9510527849197388, + "learning_rate": 3.952599308613454e-06, + "loss": 0.4909, + "step": 649 + }, + { + "epoch": 2.0313152400835075, + "grad_norm": 0.9408219456672668, + "learning_rate": 3.949073338757619e-06, + "loss": 0.4912, + "step": 650 + }, + { + "epoch": 2.034446764091858, + "grad_norm": 0.7148041725158691, + "learning_rate": 3.945543022644441e-06, + "loss": 0.4792, + "step": 651 + }, + { + "epoch": 2.037578288100209, + "grad_norm": 0.7737464904785156, + "learning_rate": 3.942008370862522e-06, + "loss": 0.4694, + "step": 652 + }, + { + "epoch": 2.0407098121085596, + "grad_norm": 0.8405889868736267, + "learning_rate": 3.938469394013472e-06, + "loss": 0.5048, + "step": 653 + }, + { + "epoch": 2.0438413361169103, + "grad_norm": 0.7896456718444824, + "learning_rate": 3.934926102711869e-06, + "loss": 0.4882, + "step": 654 + }, + { + "epoch": 2.046972860125261, + "grad_norm": 0.9290387034416199, + "learning_rate": 3.931378507585231e-06, + "loss": 0.503, + "step": 655 + }, + { + "epoch": 2.0501043841336117, + "grad_norm": 0.7386118769645691, + "learning_rate": 3.927826619273991e-06, + "loss": 0.4918, + "step": 656 + }, + { + "epoch": 2.0532359081419624, + "grad_norm": 0.9878676533699036, + "learning_rate": 3.92427044843145e-06, + "loss": 0.4958, + "step": 657 + }, + { + "epoch": 2.056367432150313, + "grad_norm": 1.0111151933670044, + "learning_rate": 3.92071000572376e-06, + "loss": 0.4886, + "step": 658 + }, + { + "epoch": 2.059498956158664, + "grad_norm": 0.8612061738967896, + "learning_rate": 3.917145301829884e-06, + "loss": 0.5216, + "step": 659 + }, + { + "epoch": 2.0626304801670146, + "grad_norm": 0.7458518743515015, + "learning_rate": 3.913576347441564e-06, + "loss": 0.4807, + "step": 660 + }, + { + "epoch": 2.0657620041753653, + "grad_norm": 0.7775886058807373, + "learning_rate": 3.910003153263294e-06, + "loss": 0.4837, + "step": 661 + }, + { + "epoch": 2.068893528183716, + "grad_norm": 0.7144196629524231, + "learning_rate": 3.906425730012282e-06, + "loss": 0.5081, + "step": 662 + }, + { + "epoch": 2.0720250521920667, + "grad_norm": 0.844971776008606, + "learning_rate": 3.9028440884184215e-06, + "loss": 0.474, + "step": 663 + }, + { + "epoch": 2.0751565762004174, + "grad_norm": 0.9709283113479614, + "learning_rate": 3.899258239224257e-06, + "loss": 0.503, + "step": 664 + }, + { + "epoch": 2.078288100208768, + "grad_norm": 1.1325515508651733, + "learning_rate": 3.895668193184954e-06, + "loss": 0.5058, + "step": 665 + }, + { + "epoch": 2.081419624217119, + "grad_norm": 0.7211254239082336, + "learning_rate": 3.892073961068266e-06, + "loss": 0.4982, + "step": 666 + }, + { + "epoch": 2.08455114822547, + "grad_norm": 0.8975517153739929, + "learning_rate": 3.888475553654502e-06, + "loss": 0.4699, + "step": 667 + }, + { + "epoch": 2.0876826722338206, + "grad_norm": 0.8270771503448486, + "learning_rate": 3.884872981736493e-06, + "loss": 0.4586, + "step": 668 + }, + { + "epoch": 2.0908141962421714, + "grad_norm": 0.8606625199317932, + "learning_rate": 3.881266256119561e-06, + "loss": 0.5299, + "step": 669 + }, + { + "epoch": 2.093945720250522, + "grad_norm": 0.9013976454734802, + "learning_rate": 3.877655387621488e-06, + "loss": 0.4887, + "step": 670 + }, + { + "epoch": 2.0970772442588728, + "grad_norm": 0.7603903412818909, + "learning_rate": 3.8740403870724795e-06, + "loss": 0.4992, + "step": 671 + }, + { + "epoch": 2.1002087682672235, + "grad_norm": 1.0432350635528564, + "learning_rate": 3.870421265315137e-06, + "loss": 0.5035, + "step": 672 + }, + { + "epoch": 2.103340292275574, + "grad_norm": 0.7727136611938477, + "learning_rate": 3.8667980332044195e-06, + "loss": 0.5006, + "step": 673 + }, + { + "epoch": 2.106471816283925, + "grad_norm": 0.9764307141304016, + "learning_rate": 3.863170701607618e-06, + "loss": 0.5061, + "step": 674 + }, + { + "epoch": 2.1096033402922756, + "grad_norm": 0.747818648815155, + "learning_rate": 3.859539281404317e-06, + "loss": 0.4761, + "step": 675 + }, + { + "epoch": 2.1127348643006263, + "grad_norm": 0.7254915237426758, + "learning_rate": 3.855903783486364e-06, + "loss": 0.5166, + "step": 676 + }, + { + "epoch": 2.115866388308977, + "grad_norm": 0.7678592801094055, + "learning_rate": 3.852264218757839e-06, + "loss": 0.5122, + "step": 677 + }, + { + "epoch": 2.1189979123173277, + "grad_norm": 0.8140144348144531, + "learning_rate": 3.8486205981350165e-06, + "loss": 0.4551, + "step": 678 + }, + { + "epoch": 2.1221294363256784, + "grad_norm": 0.9417359232902527, + "learning_rate": 3.844972932546338e-06, + "loss": 0.4748, + "step": 679 + }, + { + "epoch": 2.125260960334029, + "grad_norm": 0.8035290241241455, + "learning_rate": 3.841321232932378e-06, + "loss": 0.5079, + "step": 680 + }, + { + "epoch": 2.12839248434238, + "grad_norm": 0.8300641775131226, + "learning_rate": 3.837665510245809e-06, + "loss": 0.5018, + "step": 681 + }, + { + "epoch": 2.1315240083507305, + "grad_norm": 0.8293547034263611, + "learning_rate": 3.8340057754513715e-06, + "loss": 0.5042, + "step": 682 + }, + { + "epoch": 2.1346555323590812, + "grad_norm": 0.7780007719993591, + "learning_rate": 3.8303420395258365e-06, + "loss": 0.5048, + "step": 683 + }, + { + "epoch": 2.137787056367432, + "grad_norm": 0.7519420385360718, + "learning_rate": 3.8266743134579785e-06, + "loss": 0.5108, + "step": 684 + }, + { + "epoch": 2.140918580375783, + "grad_norm": 0.7872384190559387, + "learning_rate": 3.8230026082485404e-06, + "loss": 0.4924, + "step": 685 + }, + { + "epoch": 2.144050104384134, + "grad_norm": 0.7479491829872131, + "learning_rate": 3.819326934910197e-06, + "loss": 0.5184, + "step": 686 + }, + { + "epoch": 2.1471816283924845, + "grad_norm": 0.8438299298286438, + "learning_rate": 3.815647304467527e-06, + "loss": 0.4791, + "step": 687 + }, + { + "epoch": 2.150313152400835, + "grad_norm": 0.7923721671104431, + "learning_rate": 3.8119637279569773e-06, + "loss": 0.5305, + "step": 688 + }, + { + "epoch": 2.153444676409186, + "grad_norm": 0.7856534719467163, + "learning_rate": 3.80827621642683e-06, + "loss": 0.5063, + "step": 689 + }, + { + "epoch": 2.1565762004175366, + "grad_norm": 0.8544500470161438, + "learning_rate": 3.8045847809371706e-06, + "loss": 0.4989, + "step": 690 + }, + { + "epoch": 2.1597077244258873, + "grad_norm": 0.865390956401825, + "learning_rate": 3.800889432559852e-06, + "loss": 0.4931, + "step": 691 + }, + { + "epoch": 2.162839248434238, + "grad_norm": 0.9809399247169495, + "learning_rate": 3.797190182378466e-06, + "loss": 0.4785, + "step": 692 + }, + { + "epoch": 2.1659707724425887, + "grad_norm": 0.7954536080360413, + "learning_rate": 3.793487041488304e-06, + "loss": 0.4847, + "step": 693 + }, + { + "epoch": 2.1691022964509394, + "grad_norm": 0.754704475402832, + "learning_rate": 3.7897800209963298e-06, + "loss": 0.5125, + "step": 694 + }, + { + "epoch": 2.17223382045929, + "grad_norm": 0.7319822311401367, + "learning_rate": 3.7860691320211414e-06, + "loss": 0.477, + "step": 695 + }, + { + "epoch": 2.175365344467641, + "grad_norm": 0.8198635578155518, + "learning_rate": 3.7823543856929403e-06, + "loss": 0.4764, + "step": 696 + }, + { + "epoch": 2.1784968684759916, + "grad_norm": 0.708933413028717, + "learning_rate": 3.7786357931534987e-06, + "loss": 0.4948, + "step": 697 + }, + { + "epoch": 2.1816283924843423, + "grad_norm": 0.8493193984031677, + "learning_rate": 3.774913365556123e-06, + "loss": 0.5271, + "step": 698 + }, + { + "epoch": 2.184759916492693, + "grad_norm": 0.7999475002288818, + "learning_rate": 3.771187114065622e-06, + "loss": 0.4804, + "step": 699 + }, + { + "epoch": 2.1878914405010437, + "grad_norm": 0.8366796970367432, + "learning_rate": 3.7674570498582776e-06, + "loss": 0.457, + "step": 700 + }, + { + "epoch": 2.1910229645093944, + "grad_norm": 0.7935530543327332, + "learning_rate": 3.7637231841218015e-06, + "loss": 0.5001, + "step": 701 + }, + { + "epoch": 2.1941544885177455, + "grad_norm": 0.7700153589248657, + "learning_rate": 3.7599855280553125e-06, + "loss": 0.5091, + "step": 702 + }, + { + "epoch": 2.1972860125260962, + "grad_norm": 0.7991652488708496, + "learning_rate": 3.756244092869294e-06, + "loss": 0.4955, + "step": 703 + }, + { + "epoch": 2.200417536534447, + "grad_norm": 0.720051646232605, + "learning_rate": 3.752498889785567e-06, + "loss": 0.4902, + "step": 704 + }, + { + "epoch": 2.2035490605427976, + "grad_norm": 0.7312369346618652, + "learning_rate": 3.748749930037252e-06, + "loss": 0.4935, + "step": 705 + }, + { + "epoch": 2.2066805845511483, + "grad_norm": 0.8418563008308411, + "learning_rate": 3.744997224868739e-06, + "loss": 0.5186, + "step": 706 + }, + { + "epoch": 2.209812108559499, + "grad_norm": 0.8324081301689148, + "learning_rate": 3.741240785535649e-06, + "loss": 0.482, + "step": 707 + }, + { + "epoch": 2.2129436325678498, + "grad_norm": 0.8051855564117432, + "learning_rate": 3.737480623304805e-06, + "loss": 0.4663, + "step": 708 + }, + { + "epoch": 2.2160751565762005, + "grad_norm": 0.9464184641838074, + "learning_rate": 3.7337167494541948e-06, + "loss": 0.451, + "step": 709 + }, + { + "epoch": 2.219206680584551, + "grad_norm": 1.0227075815200806, + "learning_rate": 3.729949175272942e-06, + "loss": 0.4428, + "step": 710 + }, + { + "epoch": 2.222338204592902, + "grad_norm": 0.7930364012718201, + "learning_rate": 3.7261779120612633e-06, + "loss": 0.5132, + "step": 711 + }, + { + "epoch": 2.2254697286012526, + "grad_norm": 0.9033688306808472, + "learning_rate": 3.7224029711304444e-06, + "loss": 0.476, + "step": 712 + }, + { + "epoch": 2.2286012526096033, + "grad_norm": 0.8026887774467468, + "learning_rate": 3.7186243638028007e-06, + "loss": 0.4959, + "step": 713 + }, + { + "epoch": 2.231732776617954, + "grad_norm": 0.9391745328903198, + "learning_rate": 3.714842101411642e-06, + "loss": 0.4962, + "step": 714 + }, + { + "epoch": 2.2348643006263047, + "grad_norm": 0.7774361371994019, + "learning_rate": 3.711056195301245e-06, + "loss": 0.4748, + "step": 715 + }, + { + "epoch": 2.2379958246346554, + "grad_norm": 0.9278722405433655, + "learning_rate": 3.7072666568268115e-06, + "loss": 0.5074, + "step": 716 + }, + { + "epoch": 2.241127348643006, + "grad_norm": 0.771423876285553, + "learning_rate": 3.7034734973544406e-06, + "loss": 0.5072, + "step": 717 + }, + { + "epoch": 2.244258872651357, + "grad_norm": 0.8707448244094849, + "learning_rate": 3.6996767282610892e-06, + "loss": 0.4851, + "step": 718 + }, + { + "epoch": 2.2473903966597075, + "grad_norm": 0.7641019821166992, + "learning_rate": 3.695876360934543e-06, + "loss": 0.4941, + "step": 719 + }, + { + "epoch": 2.2505219206680582, + "grad_norm": 0.7647167444229126, + "learning_rate": 3.69207240677338e-06, + "loss": 0.5225, + "step": 720 + }, + { + "epoch": 2.2536534446764094, + "grad_norm": 0.9108865261077881, + "learning_rate": 3.6882648771869345e-06, + "loss": 0.454, + "step": 721 + }, + { + "epoch": 2.25678496868476, + "grad_norm": 0.86728835105896, + "learning_rate": 3.6844537835952666e-06, + "loss": 0.4461, + "step": 722 + }, + { + "epoch": 2.259916492693111, + "grad_norm": 1.1055282354354858, + "learning_rate": 3.6806391374291238e-06, + "loss": 0.4618, + "step": 723 + }, + { + "epoch": 2.2630480167014615, + "grad_norm": 0.7591858506202698, + "learning_rate": 3.6768209501299116e-06, + "loss": 0.4901, + "step": 724 + }, + { + "epoch": 2.266179540709812, + "grad_norm": 0.7966359257698059, + "learning_rate": 3.6729992331496554e-06, + "loss": 0.5171, + "step": 725 + }, + { + "epoch": 2.269311064718163, + "grad_norm": 0.983161211013794, + "learning_rate": 3.6691739979509672e-06, + "loss": 0.4949, + "step": 726 + }, + { + "epoch": 2.2724425887265136, + "grad_norm": 0.9200037121772766, + "learning_rate": 3.6653452560070106e-06, + "loss": 0.5234, + "step": 727 + }, + { + "epoch": 2.2755741127348643, + "grad_norm": 1.0288461446762085, + "learning_rate": 3.6615130188014685e-06, + "loss": 0.4713, + "step": 728 + }, + { + "epoch": 2.278705636743215, + "grad_norm": 0.7325463891029358, + "learning_rate": 3.6576772978285065e-06, + "loss": 0.527, + "step": 729 + }, + { + "epoch": 2.2818371607515657, + "grad_norm": 1.0045446157455444, + "learning_rate": 3.6538381045927395e-06, + "loss": 0.5139, + "step": 730 + }, + { + "epoch": 2.2849686847599164, + "grad_norm": 0.7391849756240845, + "learning_rate": 3.6499954506091963e-06, + "loss": 0.4829, + "step": 731 + }, + { + "epoch": 2.288100208768267, + "grad_norm": 0.7808229923248291, + "learning_rate": 3.646149347403286e-06, + "loss": 0.4831, + "step": 732 + }, + { + "epoch": 2.291231732776618, + "grad_norm": 0.7056961059570312, + "learning_rate": 3.6422998065107628e-06, + "loss": 0.5066, + "step": 733 + }, + { + "epoch": 2.2943632567849686, + "grad_norm": 0.7498443126678467, + "learning_rate": 3.6384468394776935e-06, + "loss": 0.4724, + "step": 734 + }, + { + "epoch": 2.2974947807933193, + "grad_norm": 0.8511576056480408, + "learning_rate": 3.634590457860418e-06, + "loss": 0.5286, + "step": 735 + }, + { + "epoch": 2.30062630480167, + "grad_norm": 0.873635470867157, + "learning_rate": 3.63073067322552e-06, + "loss": 0.4751, + "step": 736 + }, + { + "epoch": 2.3037578288100207, + "grad_norm": 0.7427377104759216, + "learning_rate": 3.626867497149788e-06, + "loss": 0.475, + "step": 737 + }, + { + "epoch": 2.306889352818372, + "grad_norm": 1.0591017007827759, + "learning_rate": 3.623000941220186e-06, + "loss": 0.4591, + "step": 738 + }, + { + "epoch": 2.3100208768267225, + "grad_norm": 0.8767879009246826, + "learning_rate": 3.6191310170338114e-06, + "loss": 0.4673, + "step": 739 + }, + { + "epoch": 2.3131524008350732, + "grad_norm": 0.9156234860420227, + "learning_rate": 3.615257736197866e-06, + "loss": 0.4622, + "step": 740 + }, + { + "epoch": 2.316283924843424, + "grad_norm": 0.6743756532669067, + "learning_rate": 3.611381110329619e-06, + "loss": 0.4723, + "step": 741 + }, + { + "epoch": 2.3194154488517746, + "grad_norm": 0.8655558228492737, + "learning_rate": 3.6075011510563732e-06, + "loss": 0.471, + "step": 742 + }, + { + "epoch": 2.3225469728601253, + "grad_norm": 0.7652033567428589, + "learning_rate": 3.603617870015429e-06, + "loss": 0.5155, + "step": 743 + }, + { + "epoch": 2.325678496868476, + "grad_norm": 0.7970699667930603, + "learning_rate": 3.599731278854049e-06, + "loss": 0.4507, + "step": 744 + }, + { + "epoch": 2.3288100208768268, + "grad_norm": 0.7538278698921204, + "learning_rate": 3.5958413892294253e-06, + "loss": 0.5093, + "step": 745 + }, + { + "epoch": 2.3319415448851775, + "grad_norm": 0.735996663570404, + "learning_rate": 3.5919482128086414e-06, + "loss": 0.5008, + "step": 746 + }, + { + "epoch": 2.335073068893528, + "grad_norm": 0.7643904685974121, + "learning_rate": 3.588051761268642e-06, + "loss": 0.5072, + "step": 747 + }, + { + "epoch": 2.338204592901879, + "grad_norm": 0.7646260857582092, + "learning_rate": 3.584152046296191e-06, + "loss": 0.4578, + "step": 748 + }, + { + "epoch": 2.3413361169102296, + "grad_norm": 0.7873825430870056, + "learning_rate": 3.5802490795878446e-06, + "loss": 0.5249, + "step": 749 + }, + { + "epoch": 2.3444676409185803, + "grad_norm": 1.095333218574524, + "learning_rate": 3.5763428728499095e-06, + "loss": 0.4913, + "step": 750 + }, + { + "epoch": 2.347599164926931, + "grad_norm": 1.3425395488739014, + "learning_rate": 3.5724334377984107e-06, + "loss": 0.5317, + "step": 751 + }, + { + "epoch": 2.3507306889352817, + "grad_norm": 0.7151113748550415, + "learning_rate": 3.568520786159055e-06, + "loss": 0.5135, + "step": 752 + }, + { + "epoch": 2.3538622129436324, + "grad_norm": 0.8072878122329712, + "learning_rate": 3.5646049296672004e-06, + "loss": 0.4863, + "step": 753 + }, + { + "epoch": 2.356993736951983, + "grad_norm": 0.8040189743041992, + "learning_rate": 3.5606858800678123e-06, + "loss": 0.4668, + "step": 754 + }, + { + "epoch": 2.3601252609603343, + "grad_norm": 0.7749765515327454, + "learning_rate": 3.5567636491154385e-06, + "loss": 0.4681, + "step": 755 + }, + { + "epoch": 2.3632567849686845, + "grad_norm": 0.773013710975647, + "learning_rate": 3.5528382485741638e-06, + "loss": 0.5012, + "step": 756 + }, + { + "epoch": 2.3663883089770357, + "grad_norm": 0.7017714381217957, + "learning_rate": 3.5489096902175835e-06, + "loss": 0.5019, + "step": 757 + }, + { + "epoch": 2.3695198329853864, + "grad_norm": 1.132458209991455, + "learning_rate": 3.5449779858287625e-06, + "loss": 0.5131, + "step": 758 + }, + { + "epoch": 2.372651356993737, + "grad_norm": 0.7624574899673462, + "learning_rate": 3.541043147200202e-06, + "loss": 0.4856, + "step": 759 + }, + { + "epoch": 2.375782881002088, + "grad_norm": 0.9078478217124939, + "learning_rate": 3.5371051861338036e-06, + "loss": 0.4337, + "step": 760 + }, + { + "epoch": 2.3789144050104385, + "grad_norm": 0.8608354330062866, + "learning_rate": 3.5331641144408344e-06, + "loss": 0.5053, + "step": 761 + }, + { + "epoch": 2.382045929018789, + "grad_norm": 0.775047779083252, + "learning_rate": 3.529219943941892e-06, + "loss": 0.4779, + "step": 762 + }, + { + "epoch": 2.38517745302714, + "grad_norm": 0.7775866389274597, + "learning_rate": 3.525272686466866e-06, + "loss": 0.4979, + "step": 763 + }, + { + "epoch": 2.3883089770354906, + "grad_norm": 0.9386464357376099, + "learning_rate": 3.521322353854908e-06, + "loss": 0.5222, + "step": 764 + }, + { + "epoch": 2.3914405010438413, + "grad_norm": 0.874109148979187, + "learning_rate": 3.517368957954391e-06, + "loss": 0.4681, + "step": 765 + }, + { + "epoch": 2.394572025052192, + "grad_norm": 0.824588418006897, + "learning_rate": 3.5134125106228766e-06, + "loss": 0.4955, + "step": 766 + }, + { + "epoch": 2.3977035490605427, + "grad_norm": 0.8790764808654785, + "learning_rate": 3.5094530237270774e-06, + "loss": 0.4722, + "step": 767 + }, + { + "epoch": 2.4008350730688934, + "grad_norm": 1.1399786472320557, + "learning_rate": 3.5054905091428253e-06, + "loss": 0.4771, + "step": 768 + }, + { + "epoch": 2.403966597077244, + "grad_norm": 1.2586532831192017, + "learning_rate": 3.50152497875503e-06, + "loss": 0.4849, + "step": 769 + }, + { + "epoch": 2.407098121085595, + "grad_norm": 0.7706464529037476, + "learning_rate": 3.4975564444576487e-06, + "loss": 0.477, + "step": 770 + }, + { + "epoch": 2.4102296450939455, + "grad_norm": 0.7695909142494202, + "learning_rate": 3.4935849181536484e-06, + "loss": 0.4695, + "step": 771 + }, + { + "epoch": 2.4133611691022967, + "grad_norm": 0.7744433283805847, + "learning_rate": 3.489610411754969e-06, + "loss": 0.499, + "step": 772 + }, + { + "epoch": 2.416492693110647, + "grad_norm": 0.9265744686126709, + "learning_rate": 3.48563293718249e-06, + "loss": 0.481, + "step": 773 + }, + { + "epoch": 2.419624217118998, + "grad_norm": 1.0680506229400635, + "learning_rate": 3.481652506365992e-06, + "loss": 0.4898, + "step": 774 + }, + { + "epoch": 2.422755741127349, + "grad_norm": 0.721493661403656, + "learning_rate": 3.477669131244122e-06, + "loss": 0.4813, + "step": 775 + }, + { + "epoch": 2.4258872651356995, + "grad_norm": 0.7993559837341309, + "learning_rate": 3.4736828237643616e-06, + "loss": 0.5179, + "step": 776 + }, + { + "epoch": 2.4290187891440502, + "grad_norm": 0.8148090839385986, + "learning_rate": 3.4696935958829837e-06, + "loss": 0.4753, + "step": 777 + }, + { + "epoch": 2.432150313152401, + "grad_norm": 0.8006406426429749, + "learning_rate": 3.465701459565022e-06, + "loss": 0.501, + "step": 778 + }, + { + "epoch": 2.4352818371607516, + "grad_norm": 0.9307970404624939, + "learning_rate": 3.4617064267842327e-06, + "loss": 0.487, + "step": 779 + }, + { + "epoch": 2.4384133611691023, + "grad_norm": 0.7192814946174622, + "learning_rate": 3.45770850952306e-06, + "loss": 0.4769, + "step": 780 + }, + { + "epoch": 2.441544885177453, + "grad_norm": 0.7386271953582764, + "learning_rate": 3.4537077197726023e-06, + "loss": 0.4726, + "step": 781 + }, + { + "epoch": 2.4446764091858038, + "grad_norm": 0.8006314039230347, + "learning_rate": 3.449704069532567e-06, + "loss": 0.494, + "step": 782 + }, + { + "epoch": 2.4478079331941545, + "grad_norm": 0.7466752529144287, + "learning_rate": 3.4456975708112477e-06, + "loss": 0.4778, + "step": 783 + }, + { + "epoch": 2.450939457202505, + "grad_norm": 0.8348856568336487, + "learning_rate": 3.4416882356254777e-06, + "loss": 0.4766, + "step": 784 + }, + { + "epoch": 2.454070981210856, + "grad_norm": 0.754851758480072, + "learning_rate": 3.4376760760005994e-06, + "loss": 0.4673, + "step": 785 + }, + { + "epoch": 2.4572025052192066, + "grad_norm": 0.7854018807411194, + "learning_rate": 3.433661103970427e-06, + "loss": 0.4954, + "step": 786 + }, + { + "epoch": 2.4603340292275573, + "grad_norm": 0.7238256931304932, + "learning_rate": 3.4296433315772084e-06, + "loss": 0.496, + "step": 787 + }, + { + "epoch": 2.463465553235908, + "grad_norm": 0.7007659673690796, + "learning_rate": 3.4256227708715915e-06, + "loss": 0.4793, + "step": 788 + }, + { + "epoch": 2.4665970772442587, + "grad_norm": 0.7234371900558472, + "learning_rate": 3.421599433912588e-06, + "loss": 0.4935, + "step": 789 + }, + { + "epoch": 2.4697286012526094, + "grad_norm": 0.7537544965744019, + "learning_rate": 3.4175733327675355e-06, + "loss": 0.5194, + "step": 790 + }, + { + "epoch": 2.4728601252609606, + "grad_norm": 0.7608047127723694, + "learning_rate": 3.4135444795120633e-06, + "loss": 0.4793, + "step": 791 + }, + { + "epoch": 2.4759916492693113, + "grad_norm": 0.7847898006439209, + "learning_rate": 3.4095128862300542e-06, + "loss": 0.4877, + "step": 792 + }, + { + "epoch": 2.479123173277662, + "grad_norm": 0.8002011179924011, + "learning_rate": 3.405478565013609e-06, + "loss": 0.4927, + "step": 793 + }, + { + "epoch": 2.4822546972860127, + "grad_norm": 0.8200219869613647, + "learning_rate": 3.401441527963013e-06, + "loss": 0.4997, + "step": 794 + }, + { + "epoch": 2.4853862212943634, + "grad_norm": 0.7220162749290466, + "learning_rate": 3.3974017871866938e-06, + "loss": 0.4668, + "step": 795 + }, + { + "epoch": 2.488517745302714, + "grad_norm": 0.8022251129150391, + "learning_rate": 3.3933593548011912e-06, + "loss": 0.5179, + "step": 796 + }, + { + "epoch": 2.491649269311065, + "grad_norm": 0.7914465069770813, + "learning_rate": 3.389314242931115e-06, + "loss": 0.4943, + "step": 797 + }, + { + "epoch": 2.4947807933194155, + "grad_norm": 1.1399403810501099, + "learning_rate": 3.385266463709116e-06, + "loss": 0.4896, + "step": 798 + }, + { + "epoch": 2.497912317327766, + "grad_norm": 0.8098909854888916, + "learning_rate": 3.38121602927584e-06, + "loss": 0.4904, + "step": 799 + }, + { + "epoch": 2.501043841336117, + "grad_norm": 0.7434052228927612, + "learning_rate": 3.377162951779902e-06, + "loss": 0.4864, + "step": 800 + }, + { + "epoch": 2.5041753653444676, + "grad_norm": 0.7397809624671936, + "learning_rate": 3.3731072433778407e-06, + "loss": 0.486, + "step": 801 + }, + { + "epoch": 2.5073068893528183, + "grad_norm": 0.99027019739151, + "learning_rate": 3.3690489162340867e-06, + "loss": 0.5011, + "step": 802 + }, + { + "epoch": 2.510438413361169, + "grad_norm": 0.8443610668182373, + "learning_rate": 3.3649879825209246e-06, + "loss": 0.455, + "step": 803 + }, + { + "epoch": 2.5135699373695197, + "grad_norm": 0.755649983882904, + "learning_rate": 3.3609244544184604e-06, + "loss": 0.4563, + "step": 804 + }, + { + "epoch": 2.5167014613778704, + "grad_norm": 0.728018045425415, + "learning_rate": 3.3568583441145765e-06, + "loss": 0.471, + "step": 805 + }, + { + "epoch": 2.519832985386221, + "grad_norm": 0.7777130603790283, + "learning_rate": 3.352789663804904e-06, + "loss": 0.4667, + "step": 806 + }, + { + "epoch": 2.522964509394572, + "grad_norm": 0.7545619606971741, + "learning_rate": 3.3487184256927785e-06, + "loss": 0.4915, + "step": 807 + }, + { + "epoch": 2.526096033402923, + "grad_norm": 0.8374579548835754, + "learning_rate": 3.3446446419892127e-06, + "loss": 0.485, + "step": 808 + }, + { + "epoch": 2.5292275574112733, + "grad_norm": 0.7354666590690613, + "learning_rate": 3.340568324912849e-06, + "loss": 0.5254, + "step": 809 + }, + { + "epoch": 2.5323590814196244, + "grad_norm": 0.7581545114517212, + "learning_rate": 3.3364894866899324e-06, + "loss": 0.4483, + "step": 810 + }, + { + "epoch": 2.535490605427975, + "grad_norm": 0.8077559471130371, + "learning_rate": 3.3324081395542662e-06, + "loss": 0.5022, + "step": 811 + }, + { + "epoch": 2.538622129436326, + "grad_norm": 0.8827865719795227, + "learning_rate": 3.3283242957471806e-06, + "loss": 0.4909, + "step": 812 + }, + { + "epoch": 2.5417536534446765, + "grad_norm": 0.9139482378959656, + "learning_rate": 3.3242379675174953e-06, + "loss": 0.5205, + "step": 813 + }, + { + "epoch": 2.5448851774530272, + "grad_norm": 0.7616812586784363, + "learning_rate": 3.3201491671214797e-06, + "loss": 0.4744, + "step": 814 + }, + { + "epoch": 2.548016701461378, + "grad_norm": 0.987173318862915, + "learning_rate": 3.3160579068228183e-06, + "loss": 0.4876, + "step": 815 + }, + { + "epoch": 2.5511482254697286, + "grad_norm": 1.259137749671936, + "learning_rate": 3.311964198892574e-06, + "loss": 0.454, + "step": 816 + }, + { + "epoch": 2.5542797494780793, + "grad_norm": 0.7866336703300476, + "learning_rate": 3.3078680556091513e-06, + "loss": 0.5107, + "step": 817 + }, + { + "epoch": 2.55741127348643, + "grad_norm": 0.9311352372169495, + "learning_rate": 3.303769489258258e-06, + "loss": 0.4843, + "step": 818 + }, + { + "epoch": 2.5605427974947808, + "grad_norm": 0.8556346893310547, + "learning_rate": 3.299668512132872e-06, + "loss": 0.5017, + "step": 819 + }, + { + "epoch": 2.5636743215031315, + "grad_norm": 2.810598373413086, + "learning_rate": 3.2955651365331988e-06, + "loss": 0.5223, + "step": 820 + }, + { + "epoch": 2.566805845511482, + "grad_norm": 1.0120766162872314, + "learning_rate": 3.29145937476664e-06, + "loss": 0.4959, + "step": 821 + }, + { + "epoch": 2.569937369519833, + "grad_norm": 0.751412034034729, + "learning_rate": 3.287351239147752e-06, + "loss": 0.4941, + "step": 822 + }, + { + "epoch": 2.5730688935281836, + "grad_norm": 1.9308148622512817, + "learning_rate": 3.2832407419982136e-06, + "loss": 0.4965, + "step": 823 + }, + { + "epoch": 2.5762004175365343, + "grad_norm": 0.9215649962425232, + "learning_rate": 3.279127895646786e-06, + "loss": 0.5071, + "step": 824 + }, + { + "epoch": 2.5793319415448854, + "grad_norm": 0.7599574327468872, + "learning_rate": 3.2750127124292754e-06, + "loss": 0.5191, + "step": 825 + }, + { + "epoch": 2.5824634655532357, + "grad_norm": 0.8234940767288208, + "learning_rate": 3.270895204688496e-06, + "loss": 0.4947, + "step": 826 + }, + { + "epoch": 2.585594989561587, + "grad_norm": 0.8401572704315186, + "learning_rate": 3.266775384774238e-06, + "loss": 0.4547, + "step": 827 + }, + { + "epoch": 2.588726513569937, + "grad_norm": 0.8927991986274719, + "learning_rate": 3.262653265043223e-06, + "loss": 0.4296, + "step": 828 + }, + { + "epoch": 2.5918580375782883, + "grad_norm": 0.8009241223335266, + "learning_rate": 3.2585288578590716e-06, + "loss": 0.4578, + "step": 829 + }, + { + "epoch": 2.594989561586639, + "grad_norm": 0.7982021570205688, + "learning_rate": 3.2544021755922663e-06, + "loss": 0.4961, + "step": 830 + }, + { + "epoch": 2.5981210855949897, + "grad_norm": 0.7096095681190491, + "learning_rate": 3.2502732306201112e-06, + "loss": 0.4975, + "step": 831 + }, + { + "epoch": 2.6012526096033404, + "grad_norm": 1.1092045307159424, + "learning_rate": 3.246142035326699e-06, + "loss": 0.4705, + "step": 832 + }, + { + "epoch": 2.604384133611691, + "grad_norm": 0.785799503326416, + "learning_rate": 3.24200860210287e-06, + "loss": 0.479, + "step": 833 + }, + { + "epoch": 2.607515657620042, + "grad_norm": 0.7315773367881775, + "learning_rate": 3.2378729433461804e-06, + "loss": 0.5036, + "step": 834 + }, + { + "epoch": 2.6106471816283925, + "grad_norm": 0.7840189337730408, + "learning_rate": 3.233735071460856e-06, + "loss": 0.4967, + "step": 835 + }, + { + "epoch": 2.613778705636743, + "grad_norm": 0.7186565399169922, + "learning_rate": 3.2295949988577655e-06, + "loss": 0.4889, + "step": 836 + }, + { + "epoch": 2.616910229645094, + "grad_norm": 0.766054093837738, + "learning_rate": 3.2254527379543747e-06, + "loss": 0.539, + "step": 837 + }, + { + "epoch": 2.6200417536534446, + "grad_norm": 0.7705381512641907, + "learning_rate": 3.2213083011747165e-06, + "loss": 0.4968, + "step": 838 + }, + { + "epoch": 2.6231732776617953, + "grad_norm": 1.3530604839324951, + "learning_rate": 3.217161700949346e-06, + "loss": 0.52, + "step": 839 + }, + { + "epoch": 2.626304801670146, + "grad_norm": 0.737389862537384, + "learning_rate": 3.2130129497153107e-06, + "loss": 0.4823, + "step": 840 + }, + { + "epoch": 2.6294363256784967, + "grad_norm": 0.9121193885803223, + "learning_rate": 3.2088620599161064e-06, + "loss": 0.4592, + "step": 841 + }, + { + "epoch": 2.632567849686848, + "grad_norm": 0.8869616389274597, + "learning_rate": 3.2047090440016464e-06, + "loss": 0.5001, + "step": 842 + }, + { + "epoch": 2.635699373695198, + "grad_norm": 0.8447219133377075, + "learning_rate": 3.200553914428219e-06, + "loss": 0.4969, + "step": 843 + }, + { + "epoch": 2.6388308977035493, + "grad_norm": 0.8877657055854797, + "learning_rate": 3.1963966836584524e-06, + "loss": 0.4718, + "step": 844 + }, + { + "epoch": 2.6419624217118995, + "grad_norm": 1.045272946357727, + "learning_rate": 3.192237364161277e-06, + "loss": 0.4864, + "step": 845 + }, + { + "epoch": 2.6450939457202507, + "grad_norm": 0.8485913276672363, + "learning_rate": 3.1880759684118876e-06, + "loss": 0.4688, + "step": 846 + }, + { + "epoch": 2.6482254697286014, + "grad_norm": 0.7328930497169495, + "learning_rate": 3.183912508891709e-06, + "loss": 0.4728, + "step": 847 + }, + { + "epoch": 2.651356993736952, + "grad_norm": 0.7377315759658813, + "learning_rate": 3.179746998088351e-06, + "loss": 0.4672, + "step": 848 + }, + { + "epoch": 2.654488517745303, + "grad_norm": 0.8017002940177917, + "learning_rate": 3.1755794484955817e-06, + "loss": 0.4884, + "step": 849 + }, + { + "epoch": 2.6576200417536535, + "grad_norm": 1.045470952987671, + "learning_rate": 3.171409872613278e-06, + "loss": 0.4789, + "step": 850 + }, + { + "epoch": 2.6607515657620042, + "grad_norm": 0.8823987245559692, + "learning_rate": 3.1672382829473997e-06, + "loss": 0.5117, + "step": 851 + }, + { + "epoch": 2.663883089770355, + "grad_norm": 0.7395204901695251, + "learning_rate": 3.163064692009944e-06, + "loss": 0.5476, + "step": 852 + }, + { + "epoch": 2.6670146137787056, + "grad_norm": 0.7778941988945007, + "learning_rate": 3.1588891123189103e-06, + "loss": 0.5092, + "step": 853 + }, + { + "epoch": 2.6701461377870563, + "grad_norm": 0.8072531819343567, + "learning_rate": 3.1547115563982643e-06, + "loss": 0.4961, + "step": 854 + }, + { + "epoch": 2.673277661795407, + "grad_norm": 0.9018139243125916, + "learning_rate": 3.1505320367778993e-06, + "loss": 0.4624, + "step": 855 + }, + { + "epoch": 2.6764091858037578, + "grad_norm": 0.8554450869560242, + "learning_rate": 3.1463505659935957e-06, + "loss": 0.4971, + "step": 856 + }, + { + "epoch": 2.6795407098121085, + "grad_norm": 0.7727259397506714, + "learning_rate": 3.14216715658699e-06, + "loss": 0.4544, + "step": 857 + }, + { + "epoch": 2.682672233820459, + "grad_norm": 0.9253409504890442, + "learning_rate": 3.137981821105529e-06, + "loss": 0.4893, + "step": 858 + }, + { + "epoch": 2.68580375782881, + "grad_norm": 0.8809456825256348, + "learning_rate": 3.1337945721024403e-06, + "loss": 0.5242, + "step": 859 + }, + { + "epoch": 2.6889352818371606, + "grad_norm": 0.981755256652832, + "learning_rate": 3.129605422136689e-06, + "loss": 0.4686, + "step": 860 + }, + { + "epoch": 2.6920668058455117, + "grad_norm": 1.1278467178344727, + "learning_rate": 3.1254143837729412e-06, + "loss": 0.4813, + "step": 861 + }, + { + "epoch": 2.695198329853862, + "grad_norm": 0.8529123663902283, + "learning_rate": 3.1212214695815285e-06, + "loss": 0.4723, + "step": 862 + }, + { + "epoch": 2.698329853862213, + "grad_norm": 0.7764189839363098, + "learning_rate": 3.1170266921384075e-06, + "loss": 0.4777, + "step": 863 + }, + { + "epoch": 2.701461377870564, + "grad_norm": 0.7364740967750549, + "learning_rate": 3.112830064025124e-06, + "loss": 0.4975, + "step": 864 + }, + { + "epoch": 2.7045929018789145, + "grad_norm": 0.7594549059867859, + "learning_rate": 3.108631597828774e-06, + "loss": 0.5083, + "step": 865 + }, + { + "epoch": 2.7077244258872653, + "grad_norm": 0.7337073683738708, + "learning_rate": 3.104431306141968e-06, + "loss": 0.4778, + "step": 866 + }, + { + "epoch": 2.710855949895616, + "grad_norm": 0.7709932327270508, + "learning_rate": 3.1002292015627894e-06, + "loss": 0.4754, + "step": 867 + }, + { + "epoch": 2.7139874739039667, + "grad_norm": 0.8001313209533691, + "learning_rate": 3.0960252966947605e-06, + "loss": 0.4489, + "step": 868 + }, + { + "epoch": 2.7171189979123174, + "grad_norm": 0.8280592560768127, + "learning_rate": 3.091819604146804e-06, + "loss": 0.4606, + "step": 869 + }, + { + "epoch": 2.720250521920668, + "grad_norm": 0.7463534474372864, + "learning_rate": 3.0876121365332024e-06, + "loss": 0.5168, + "step": 870 + }, + { + "epoch": 2.723382045929019, + "grad_norm": 0.9011222124099731, + "learning_rate": 3.0834029064735636e-06, + "loss": 0.5163, + "step": 871 + }, + { + "epoch": 2.7265135699373695, + "grad_norm": 0.7811456322669983, + "learning_rate": 3.0791919265927827e-06, + "loss": 0.5004, + "step": 872 + }, + { + "epoch": 2.72964509394572, + "grad_norm": 0.9251837134361267, + "learning_rate": 3.0749792095210003e-06, + "loss": 0.5081, + "step": 873 + }, + { + "epoch": 2.732776617954071, + "grad_norm": 0.8347085118293762, + "learning_rate": 3.0707647678935695e-06, + "loss": 0.4793, + "step": 874 + }, + { + "epoch": 2.7359081419624216, + "grad_norm": 0.9766442179679871, + "learning_rate": 3.0665486143510153e-06, + "loss": 0.493, + "step": 875 + }, + { + "epoch": 2.7390396659707723, + "grad_norm": 0.7692548036575317, + "learning_rate": 3.0623307615389975e-06, + "loss": 0.4874, + "step": 876 + }, + { + "epoch": 2.742171189979123, + "grad_norm": 0.7714599370956421, + "learning_rate": 3.0581112221082727e-06, + "loss": 0.4929, + "step": 877 + }, + { + "epoch": 2.745302713987474, + "grad_norm": 0.7797786593437195, + "learning_rate": 3.053890008714655e-06, + "loss": 0.4359, + "step": 878 + }, + { + "epoch": 2.7484342379958244, + "grad_norm": 5.118397235870361, + "learning_rate": 3.049667134018981e-06, + "loss": 0.4634, + "step": 879 + }, + { + "epoch": 2.7515657620041756, + "grad_norm": 0.7684539556503296, + "learning_rate": 3.04544261068707e-06, + "loss": 0.4688, + "step": 880 + }, + { + "epoch": 2.754697286012526, + "grad_norm": 0.8678610920906067, + "learning_rate": 3.0412164513896846e-06, + "loss": 0.5213, + "step": 881 + }, + { + "epoch": 2.757828810020877, + "grad_norm": 0.80293869972229, + "learning_rate": 3.0369886688024954e-06, + "loss": 0.4392, + "step": 882 + }, + { + "epoch": 2.7609603340292277, + "grad_norm": 0.7438644766807556, + "learning_rate": 3.0327592756060412e-06, + "loss": 0.528, + "step": 883 + }, + { + "epoch": 2.7640918580375784, + "grad_norm": 0.7701645493507385, + "learning_rate": 3.0285282844856917e-06, + "loss": 0.504, + "step": 884 + }, + { + "epoch": 2.767223382045929, + "grad_norm": 0.7113856673240662, + "learning_rate": 3.024295708131611e-06, + "loss": 0.4819, + "step": 885 + }, + { + "epoch": 2.77035490605428, + "grad_norm": 1.2697532176971436, + "learning_rate": 3.020061559238714e-06, + "loss": 0.5009, + "step": 886 + }, + { + "epoch": 2.7734864300626305, + "grad_norm": 1.0299439430236816, + "learning_rate": 3.015825850506636e-06, + "loss": 0.4707, + "step": 887 + }, + { + "epoch": 2.776617954070981, + "grad_norm": 0.9703660607337952, + "learning_rate": 3.011588594639688e-06, + "loss": 0.4102, + "step": 888 + }, + { + "epoch": 2.779749478079332, + "grad_norm": 0.7357314825057983, + "learning_rate": 3.0073498043468247e-06, + "loss": 0.4649, + "step": 889 + }, + { + "epoch": 2.7828810020876826, + "grad_norm": 0.7815471291542053, + "learning_rate": 3.0031094923415993e-06, + "loss": 0.469, + "step": 890 + }, + { + "epoch": 2.7860125260960333, + "grad_norm": 0.7856019139289856, + "learning_rate": 2.9988676713421318e-06, + "loss": 0.4241, + "step": 891 + }, + { + "epoch": 2.789144050104384, + "grad_norm": 0.7668167352676392, + "learning_rate": 2.994624354071066e-06, + "loss": 0.5309, + "step": 892 + }, + { + "epoch": 2.7922755741127347, + "grad_norm": 0.7485945820808411, + "learning_rate": 2.990379553255535e-06, + "loss": 0.5173, + "step": 893 + }, + { + "epoch": 2.7954070981210855, + "grad_norm": 0.8065824508666992, + "learning_rate": 2.986133281627123e-06, + "loss": 0.4995, + "step": 894 + }, + { + "epoch": 2.798538622129436, + "grad_norm": 0.7156995534896851, + "learning_rate": 2.9818855519218217e-06, + "loss": 0.4642, + "step": 895 + }, + { + "epoch": 2.801670146137787, + "grad_norm": 0.9115403890609741, + "learning_rate": 2.97763637688e-06, + "loss": 0.4799, + "step": 896 + }, + { + "epoch": 2.804801670146138, + "grad_norm": 0.7466689944267273, + "learning_rate": 2.9733857692463584e-06, + "loss": 0.4942, + "step": 897 + }, + { + "epoch": 2.8079331941544883, + "grad_norm": 0.7484914064407349, + "learning_rate": 2.9691337417698974e-06, + "loss": 0.4618, + "step": 898 + }, + { + "epoch": 2.8110647181628394, + "grad_norm": 0.816704511642456, + "learning_rate": 2.9648803072038736e-06, + "loss": 0.4748, + "step": 899 + }, + { + "epoch": 2.81419624217119, + "grad_norm": 0.7627584934234619, + "learning_rate": 2.9606254783057666e-06, + "loss": 0.4667, + "step": 900 + }, + { + "epoch": 2.817327766179541, + "grad_norm": 0.7341011166572571, + "learning_rate": 2.9563692678372342e-06, + "loss": 0.4802, + "step": 901 + }, + { + "epoch": 2.8204592901878915, + "grad_norm": 1.2541382312774658, + "learning_rate": 2.952111688564082e-06, + "loss": 0.5231, + "step": 902 + }, + { + "epoch": 2.8235908141962422, + "grad_norm": 0.7172819375991821, + "learning_rate": 2.9478527532562184e-06, + "loss": 0.4488, + "step": 903 + }, + { + "epoch": 2.826722338204593, + "grad_norm": 0.774529218673706, + "learning_rate": 2.943592474687621e-06, + "loss": 0.4964, + "step": 904 + }, + { + "epoch": 2.8298538622129437, + "grad_norm": 0.7315672636032104, + "learning_rate": 2.939330865636294e-06, + "loss": 0.4817, + "step": 905 + }, + { + "epoch": 2.8329853862212944, + "grad_norm": 0.7698234915733337, + "learning_rate": 2.9350679388842347e-06, + "loss": 0.5075, + "step": 906 + }, + { + "epoch": 2.836116910229645, + "grad_norm": 0.7717766761779785, + "learning_rate": 2.93080370721739e-06, + "loss": 0.4789, + "step": 907 + }, + { + "epoch": 2.8392484342379958, + "grad_norm": 0.7383570075035095, + "learning_rate": 2.926538183425622e-06, + "loss": 0.4992, + "step": 908 + }, + { + "epoch": 2.8423799582463465, + "grad_norm": 0.7858864068984985, + "learning_rate": 2.92227138030267e-06, + "loss": 0.4993, + "step": 909 + }, + { + "epoch": 2.845511482254697, + "grad_norm": 0.8220369219779968, + "learning_rate": 2.9180033106461076e-06, + "loss": 0.4929, + "step": 910 + }, + { + "epoch": 2.848643006263048, + "grad_norm": 0.7507152557373047, + "learning_rate": 2.9137339872573086e-06, + "loss": 0.4394, + "step": 911 + }, + { + "epoch": 2.8517745302713986, + "grad_norm": 0.7935269474983215, + "learning_rate": 2.9094634229414063e-06, + "loss": 0.4656, + "step": 912 + }, + { + "epoch": 2.8549060542797493, + "grad_norm": 0.9187721610069275, + "learning_rate": 2.9051916305072576e-06, + "loss": 0.4918, + "step": 913 + }, + { + "epoch": 2.8580375782881005, + "grad_norm": 0.8699706792831421, + "learning_rate": 2.9009186227674e-06, + "loss": 0.5106, + "step": 914 + }, + { + "epoch": 2.8611691022964507, + "grad_norm": 0.7175673246383667, + "learning_rate": 2.896644412538021e-06, + "loss": 0.5105, + "step": 915 + }, + { + "epoch": 2.864300626304802, + "grad_norm": 0.8563990592956543, + "learning_rate": 2.892369012638909e-06, + "loss": 0.4993, + "step": 916 + }, + { + "epoch": 2.867432150313152, + "grad_norm": 0.7891882658004761, + "learning_rate": 2.8880924358934246e-06, + "loss": 0.4983, + "step": 917 + }, + { + "epoch": 2.8705636743215033, + "grad_norm": 0.9247110486030579, + "learning_rate": 2.8838146951284575e-06, + "loss": 0.4789, + "step": 918 + }, + { + "epoch": 2.873695198329854, + "grad_norm": 0.7523055672645569, + "learning_rate": 2.879535803174387e-06, + "loss": 0.4982, + "step": 919 + }, + { + "epoch": 2.8768267223382047, + "grad_norm": 0.8096909523010254, + "learning_rate": 2.8752557728650467e-06, + "loss": 0.4958, + "step": 920 + }, + { + "epoch": 2.8799582463465554, + "grad_norm": 2.3476874828338623, + "learning_rate": 2.870974617037684e-06, + "loss": 0.491, + "step": 921 + }, + { + "epoch": 2.883089770354906, + "grad_norm": 0.8388578295707703, + "learning_rate": 2.8666923485329224e-06, + "loss": 0.5275, + "step": 922 + }, + { + "epoch": 2.886221294363257, + "grad_norm": 0.8162729144096375, + "learning_rate": 2.8624089801947234e-06, + "loss": 0.4776, + "step": 923 + }, + { + "epoch": 2.8893528183716075, + "grad_norm": 0.7306103110313416, + "learning_rate": 2.858124524870345e-06, + "loss": 0.4814, + "step": 924 + }, + { + "epoch": 2.892484342379958, + "grad_norm": 0.8736817836761475, + "learning_rate": 2.853838995410307e-06, + "loss": 0.5097, + "step": 925 + }, + { + "epoch": 2.895615866388309, + "grad_norm": 0.7771823406219482, + "learning_rate": 2.8495524046683525e-06, + "loss": 0.4806, + "step": 926 + }, + { + "epoch": 2.8987473903966596, + "grad_norm": 0.9421334862709045, + "learning_rate": 2.845264765501404e-06, + "loss": 0.5055, + "step": 927 + }, + { + "epoch": 2.9018789144050103, + "grad_norm": 0.8403921127319336, + "learning_rate": 2.8409760907695314e-06, + "loss": 0.4775, + "step": 928 + }, + { + "epoch": 2.905010438413361, + "grad_norm": 0.8095362186431885, + "learning_rate": 2.836686393335909e-06, + "loss": 0.4532, + "step": 929 + }, + { + "epoch": 2.9081419624217117, + "grad_norm": 0.7340645790100098, + "learning_rate": 2.8323956860667813e-06, + "loss": 0.4835, + "step": 930 + }, + { + "epoch": 2.911273486430063, + "grad_norm": 0.6970911026000977, + "learning_rate": 2.828103981831417e-06, + "loss": 0.4999, + "step": 931 + }, + { + "epoch": 2.914405010438413, + "grad_norm": 0.8136418461799622, + "learning_rate": 2.8238112935020794e-06, + "loss": 0.5038, + "step": 932 + }, + { + "epoch": 2.9175365344467643, + "grad_norm": 0.9045608043670654, + "learning_rate": 2.8195176339539816e-06, + "loss": 0.486, + "step": 933 + }, + { + "epoch": 2.9206680584551146, + "grad_norm": 1.14940345287323, + "learning_rate": 2.815223016065249e-06, + "loss": 0.5079, + "step": 934 + }, + { + "epoch": 2.9237995824634657, + "grad_norm": 0.7411190867424011, + "learning_rate": 2.8109274527168826e-06, + "loss": 0.4564, + "step": 935 + }, + { + "epoch": 2.9269311064718164, + "grad_norm": 0.8903455138206482, + "learning_rate": 2.806630956792719e-06, + "loss": 0.451, + "step": 936 + }, + { + "epoch": 2.930062630480167, + "grad_norm": 0.7865445017814636, + "learning_rate": 2.8023335411793904e-06, + "loss": 0.4658, + "step": 937 + }, + { + "epoch": 2.933194154488518, + "grad_norm": 0.8185790777206421, + "learning_rate": 2.798035218766292e-06, + "loss": 0.4776, + "step": 938 + }, + { + "epoch": 2.9363256784968685, + "grad_norm": 0.7516276836395264, + "learning_rate": 2.793736002445531e-06, + "loss": 0.4447, + "step": 939 + }, + { + "epoch": 2.9394572025052192, + "grad_norm": 0.738080620765686, + "learning_rate": 2.789435905111903e-06, + "loss": 0.4832, + "step": 940 + }, + { + "epoch": 2.94258872651357, + "grad_norm": 0.7971507906913757, + "learning_rate": 2.785134939662843e-06, + "loss": 0.4835, + "step": 941 + }, + { + "epoch": 2.9457202505219207, + "grad_norm": 0.7529093623161316, + "learning_rate": 2.78083311899839e-06, + "loss": 0.4759, + "step": 942 + }, + { + "epoch": 2.9488517745302714, + "grad_norm": 0.8222358226776123, + "learning_rate": 2.7765304560211482e-06, + "loss": 0.4365, + "step": 943 + }, + { + "epoch": 2.951983298538622, + "grad_norm": 0.729945182800293, + "learning_rate": 2.7722269636362462e-06, + "loss": 0.5026, + "step": 944 + }, + { + "epoch": 2.9551148225469728, + "grad_norm": 0.7287900447845459, + "learning_rate": 2.767922654751306e-06, + "loss": 0.4916, + "step": 945 + }, + { + "epoch": 2.9582463465553235, + "grad_norm": 0.869637131690979, + "learning_rate": 2.763617542276391e-06, + "loss": 0.5018, + "step": 946 + }, + { + "epoch": 2.961377870563674, + "grad_norm": 1.004909634590149, + "learning_rate": 2.7593116391239806e-06, + "loss": 0.5152, + "step": 947 + }, + { + "epoch": 2.964509394572025, + "grad_norm": 0.8263046145439148, + "learning_rate": 2.7550049582089235e-06, + "loss": 0.5249, + "step": 948 + }, + { + "epoch": 2.9676409185803756, + "grad_norm": 0.7963895797729492, + "learning_rate": 2.750697512448401e-06, + "loss": 0.5084, + "step": 949 + }, + { + "epoch": 2.9707724425887267, + "grad_norm": 0.7211249470710754, + "learning_rate": 2.7463893147618893e-06, + "loss": 0.4691, + "step": 950 + }, + { + "epoch": 2.973903966597077, + "grad_norm": 0.8010216951370239, + "learning_rate": 2.742080378071118e-06, + "loss": 0.5026, + "step": 951 + }, + { + "epoch": 2.977035490605428, + "grad_norm": 0.780078649520874, + "learning_rate": 2.7377707153000356e-06, + "loss": 0.4758, + "step": 952 + }, + { + "epoch": 2.980167014613779, + "grad_norm": 0.7728193998336792, + "learning_rate": 2.7334603393747684e-06, + "loss": 0.488, + "step": 953 + }, + { + "epoch": 2.9832985386221296, + "grad_norm": 0.836329996585846, + "learning_rate": 2.7291492632235777e-06, + "loss": 0.456, + "step": 954 + }, + { + "epoch": 2.9864300626304803, + "grad_norm": 0.7241990566253662, + "learning_rate": 2.724837499776831e-06, + "loss": 0.4953, + "step": 955 + }, + { + "epoch": 2.989561586638831, + "grad_norm": 0.7595076560974121, + "learning_rate": 2.7205250619669527e-06, + "loss": 0.446, + "step": 956 + }, + { + "epoch": 2.9926931106471817, + "grad_norm": 0.8177686333656311, + "learning_rate": 2.716211962728392e-06, + "loss": 0.5057, + "step": 957 + }, + { + "epoch": 2.9958246346555324, + "grad_norm": 0.7506977915763855, + "learning_rate": 2.71189821499758e-06, + "loss": 0.4821, + "step": 958 + }, + { + "epoch": 2.998956158663883, + "grad_norm": 0.8085163235664368, + "learning_rate": 2.7075838317128943e-06, + "loss": 0.5002, + "step": 959 + }, + { + "epoch": 3.0, + "grad_norm": 3.3674418926239014, + "learning_rate": 2.7032688258146207e-06, + "loss": 0.183, + "step": 960 + }, + { + "epoch": 3.0031315240083507, + "grad_norm": 1.066100835800171, + "learning_rate": 2.698953210244908e-06, + "loss": 0.4427, + "step": 961 + }, + { + "epoch": 3.0062630480167014, + "grad_norm": 0.7690210938453674, + "learning_rate": 2.6946369979477365e-06, + "loss": 0.4638, + "step": 962 + }, + { + "epoch": 3.009394572025052, + "grad_norm": 0.8114679455757141, + "learning_rate": 2.690320201868876e-06, + "loss": 0.4373, + "step": 963 + }, + { + "epoch": 3.012526096033403, + "grad_norm": 0.7680971622467041, + "learning_rate": 2.686002834955847e-06, + "loss": 0.4477, + "step": 964 + }, + { + "epoch": 3.0156576200417535, + "grad_norm": 0.7194678783416748, + "learning_rate": 2.6816849101578808e-06, + "loss": 0.455, + "step": 965 + }, + { + "epoch": 3.0187891440501042, + "grad_norm": 0.6890467405319214, + "learning_rate": 2.6773664404258854e-06, + "loss": 0.4246, + "step": 966 + }, + { + "epoch": 3.021920668058455, + "grad_norm": 0.8064301013946533, + "learning_rate": 2.6730474387123987e-06, + "loss": 0.497, + "step": 967 + }, + { + "epoch": 3.0250521920668056, + "grad_norm": 0.8164849281311035, + "learning_rate": 2.668727917971559e-06, + "loss": 0.457, + "step": 968 + }, + { + "epoch": 3.028183716075157, + "grad_norm": 0.7793440818786621, + "learning_rate": 2.6644078911590565e-06, + "loss": 0.4632, + "step": 969 + }, + { + "epoch": 3.0313152400835075, + "grad_norm": 0.7208535671234131, + "learning_rate": 2.6600873712321033e-06, + "loss": 0.4786, + "step": 970 + }, + { + "epoch": 3.034446764091858, + "grad_norm": 0.8905500769615173, + "learning_rate": 2.655766371149389e-06, + "loss": 0.4317, + "step": 971 + }, + { + "epoch": 3.037578288100209, + "grad_norm": 0.7537338733673096, + "learning_rate": 2.6514449038710418e-06, + "loss": 0.4783, + "step": 972 + }, + { + "epoch": 3.0407098121085596, + "grad_norm": 0.7901656031608582, + "learning_rate": 2.6471229823585937e-06, + "loss": 0.4762, + "step": 973 + }, + { + "epoch": 3.0438413361169103, + "grad_norm": 0.7427340745925903, + "learning_rate": 2.6428006195749373e-06, + "loss": 0.4782, + "step": 974 + }, + { + "epoch": 3.046972860125261, + "grad_norm": 0.8364680409431458, + "learning_rate": 2.6384778284842905e-06, + "loss": 0.4551, + "step": 975 + }, + { + "epoch": 3.0501043841336117, + "grad_norm": 1.1049500703811646, + "learning_rate": 2.634154622052155e-06, + "loss": 0.4451, + "step": 976 + }, + { + "epoch": 3.0532359081419624, + "grad_norm": 1.0797414779663086, + "learning_rate": 2.6298310132452757e-06, + "loss": 0.4512, + "step": 977 + }, + { + "epoch": 3.056367432150313, + "grad_norm": 0.8832088708877563, + "learning_rate": 2.62550701503161e-06, + "loss": 0.4569, + "step": 978 + }, + { + "epoch": 3.059498956158664, + "grad_norm": 0.748951256275177, + "learning_rate": 2.621182640380277e-06, + "loss": 0.4535, + "step": 979 + }, + { + "epoch": 3.0626304801670146, + "grad_norm": 0.776386022567749, + "learning_rate": 2.616857902261529e-06, + "loss": 0.4582, + "step": 980 + }, + { + "epoch": 3.0657620041753653, + "grad_norm": 0.730057418346405, + "learning_rate": 2.6125328136467074e-06, + "loss": 0.4581, + "step": 981 + }, + { + "epoch": 3.068893528183716, + "grad_norm": 0.7918877005577087, + "learning_rate": 2.6082073875082046e-06, + "loss": 0.4355, + "step": 982 + }, + { + "epoch": 3.0720250521920667, + "grad_norm": 0.803987443447113, + "learning_rate": 2.6038816368194265e-06, + "loss": 0.455, + "step": 983 + }, + { + "epoch": 3.0751565762004174, + "grad_norm": 1.2970365285873413, + "learning_rate": 2.599555574554749e-06, + "loss": 0.448, + "step": 984 + }, + { + "epoch": 3.078288100208768, + "grad_norm": 0.7814194560050964, + "learning_rate": 2.595229213689487e-06, + "loss": 0.4197, + "step": 985 + }, + { + "epoch": 3.081419624217119, + "grad_norm": 0.9865803122520447, + "learning_rate": 2.5909025671998483e-06, + "loss": 0.4649, + "step": 986 + }, + { + "epoch": 3.08455114822547, + "grad_norm": 0.7769168615341187, + "learning_rate": 2.586575648062898e-06, + "loss": 0.4287, + "step": 987 + }, + { + "epoch": 3.0876826722338206, + "grad_norm": 1.454988956451416, + "learning_rate": 2.582248469256519e-06, + "loss": 0.4548, + "step": 988 + }, + { + "epoch": 3.0908141962421714, + "grad_norm": 0.7656151056289673, + "learning_rate": 2.577921043759372e-06, + "loss": 0.4323, + "step": 989 + }, + { + "epoch": 3.093945720250522, + "grad_norm": 0.7989768385887146, + "learning_rate": 2.5735933845508598e-06, + "loss": 0.4616, + "step": 990 + }, + { + "epoch": 3.0970772442588728, + "grad_norm": 0.8711255192756653, + "learning_rate": 2.5692655046110855e-06, + "loss": 0.4923, + "step": 991 + }, + { + "epoch": 3.1002087682672235, + "grad_norm": 0.7333446145057678, + "learning_rate": 2.564937416920813e-06, + "loss": 0.4297, + "step": 992 + }, + { + "epoch": 3.103340292275574, + "grad_norm": 0.7611051201820374, + "learning_rate": 2.5606091344614297e-06, + "loss": 0.4276, + "step": 993 + }, + { + "epoch": 3.106471816283925, + "grad_norm": 0.9742305278778076, + "learning_rate": 2.5562806702149083e-06, + "loss": 0.4597, + "step": 994 + }, + { + "epoch": 3.1096033402922756, + "grad_norm": 0.8043314218521118, + "learning_rate": 2.551952037163765e-06, + "loss": 0.459, + "step": 995 + }, + { + "epoch": 3.1127348643006263, + "grad_norm": 0.7217118740081787, + "learning_rate": 2.5476232482910253e-06, + "loss": 0.4514, + "step": 996 + }, + { + "epoch": 3.115866388308977, + "grad_norm": 1.1410573720932007, + "learning_rate": 2.5432943165801765e-06, + "loss": 0.4942, + "step": 997 + }, + { + "epoch": 3.1189979123173277, + "grad_norm": 0.7188895344734192, + "learning_rate": 2.5389652550151416e-06, + "loss": 0.4641, + "step": 998 + }, + { + "epoch": 3.1221294363256784, + "grad_norm": 0.7223486304283142, + "learning_rate": 2.5346360765802276e-06, + "loss": 0.4382, + "step": 999 + }, + { + "epoch": 3.125260960334029, + "grad_norm": 0.8269757032394409, + "learning_rate": 2.5303067942600933e-06, + "loss": 0.4502, + "step": 1000 + }, + { + "epoch": 3.12839248434238, + "grad_norm": 0.7081235647201538, + "learning_rate": 2.5259774210397104e-06, + "loss": 0.4389, + "step": 1001 + }, + { + "epoch": 3.1315240083507305, + "grad_norm": 0.7493525147438049, + "learning_rate": 2.5216479699043224e-06, + "loss": 0.4697, + "step": 1002 + }, + { + "epoch": 3.1346555323590812, + "grad_norm": 0.7633835077285767, + "learning_rate": 2.5173184538394054e-06, + "loss": 0.4535, + "step": 1003 + }, + { + "epoch": 3.137787056367432, + "grad_norm": 0.7765631079673767, + "learning_rate": 2.5129888858306333e-06, + "loss": 0.4616, + "step": 1004 + }, + { + "epoch": 3.140918580375783, + "grad_norm": 0.7026706337928772, + "learning_rate": 2.508659278863832e-06, + "loss": 0.4535, + "step": 1005 + }, + { + "epoch": 3.144050104384134, + "grad_norm": 0.8816738128662109, + "learning_rate": 2.5043296459249466e-06, + "loss": 0.4274, + "step": 1006 + }, + { + "epoch": 3.1471816283924845, + "grad_norm": 0.8072274327278137, + "learning_rate": 2.5e-06, + "loss": 0.4566, + "step": 1007 + }, + { + "epoch": 3.150313152400835, + "grad_norm": 0.7775781750679016, + "learning_rate": 2.4956703540750542e-06, + "loss": 0.4248, + "step": 1008 + }, + { + "epoch": 3.153444676409186, + "grad_norm": 0.9271306991577148, + "learning_rate": 2.4913407211361686e-06, + "loss": 0.4251, + "step": 1009 + }, + { + "epoch": 3.1565762004175366, + "grad_norm": 0.763867974281311, + "learning_rate": 2.487011114169368e-06, + "loss": 0.4525, + "step": 1010 + }, + { + "epoch": 3.1597077244258873, + "grad_norm": 0.8754820227622986, + "learning_rate": 2.4826815461605955e-06, + "loss": 0.4524, + "step": 1011 + }, + { + "epoch": 3.162839248434238, + "grad_norm": 0.8261513113975525, + "learning_rate": 2.4783520300956784e-06, + "loss": 0.4303, + "step": 1012 + }, + { + "epoch": 3.1659707724425887, + "grad_norm": 0.8629854917526245, + "learning_rate": 2.4740225789602905e-06, + "loss": 0.4628, + "step": 1013 + }, + { + "epoch": 3.1691022964509394, + "grad_norm": 0.7281518578529358, + "learning_rate": 2.469693205739907e-06, + "loss": 0.461, + "step": 1014 + }, + { + "epoch": 3.17223382045929, + "grad_norm": 0.757644772529602, + "learning_rate": 2.465363923419774e-06, + "loss": 0.4455, + "step": 1015 + }, + { + "epoch": 3.175365344467641, + "grad_norm": 0.8021314740180969, + "learning_rate": 2.4610347449848592e-06, + "loss": 0.4755, + "step": 1016 + }, + { + "epoch": 3.1784968684759916, + "grad_norm": 0.7467564940452576, + "learning_rate": 2.456705683419824e-06, + "loss": 0.4419, + "step": 1017 + }, + { + "epoch": 3.1816283924843423, + "grad_norm": 0.8575125336647034, + "learning_rate": 2.452376751708976e-06, + "loss": 0.4577, + "step": 1018 + }, + { + "epoch": 3.184759916492693, + "grad_norm": 0.8101590871810913, + "learning_rate": 2.448047962836235e-06, + "loss": 0.4546, + "step": 1019 + }, + { + "epoch": 3.1878914405010437, + "grad_norm": 0.8146190643310547, + "learning_rate": 2.443719329785093e-06, + "loss": 0.4219, + "step": 1020 + }, + { + "epoch": 3.1910229645093944, + "grad_norm": 1.2800556421279907, + "learning_rate": 2.4393908655385708e-06, + "loss": 0.4502, + "step": 1021 + }, + { + "epoch": 3.1941544885177455, + "grad_norm": 0.7362221479415894, + "learning_rate": 2.4350625830791875e-06, + "loss": 0.4802, + "step": 1022 + }, + { + "epoch": 3.1972860125260962, + "grad_norm": 0.8307201862335205, + "learning_rate": 2.430734495388915e-06, + "loss": 0.4808, + "step": 1023 + }, + { + "epoch": 3.200417536534447, + "grad_norm": 0.8467394709587097, + "learning_rate": 2.42640661544914e-06, + "loss": 0.4598, + "step": 1024 + }, + { + "epoch": 3.2035490605427976, + "grad_norm": 0.9570673108100891, + "learning_rate": 2.422078956240629e-06, + "loss": 0.4341, + "step": 1025 + }, + { + "epoch": 3.2066805845511483, + "grad_norm": 0.7976422309875488, + "learning_rate": 2.4177515307434824e-06, + "loss": 0.4285, + "step": 1026 + }, + { + "epoch": 3.209812108559499, + "grad_norm": 0.8241607546806335, + "learning_rate": 2.413424351937103e-06, + "loss": 0.4563, + "step": 1027 + }, + { + "epoch": 3.2129436325678498, + "grad_norm": 0.7191072702407837, + "learning_rate": 2.4090974328001526e-06, + "loss": 0.456, + "step": 1028 + }, + { + "epoch": 3.2160751565762005, + "grad_norm": 0.8112174868583679, + "learning_rate": 2.4047707863105133e-06, + "loss": 0.4592, + "step": 1029 + }, + { + "epoch": 3.219206680584551, + "grad_norm": 0.7170486450195312, + "learning_rate": 2.4004444254452522e-06, + "loss": 0.4559, + "step": 1030 + }, + { + "epoch": 3.222338204592902, + "grad_norm": 0.7092299461364746, + "learning_rate": 2.3961183631805748e-06, + "loss": 0.4504, + "step": 1031 + }, + { + "epoch": 3.2254697286012526, + "grad_norm": 0.8104662299156189, + "learning_rate": 2.391792612491796e-06, + "loss": 0.4327, + "step": 1032 + }, + { + "epoch": 3.2286012526096033, + "grad_norm": 0.8512858152389526, + "learning_rate": 2.387467186353293e-06, + "loss": 0.4506, + "step": 1033 + }, + { + "epoch": 3.231732776617954, + "grad_norm": 0.7987692952156067, + "learning_rate": 2.3831420977384715e-06, + "loss": 0.4257, + "step": 1034 + }, + { + "epoch": 3.2348643006263047, + "grad_norm": 0.7805537581443787, + "learning_rate": 2.3788173596197244e-06, + "loss": 0.4692, + "step": 1035 + }, + { + "epoch": 3.2379958246346554, + "grad_norm": 0.737304151058197, + "learning_rate": 2.374492984968392e-06, + "loss": 0.4308, + "step": 1036 + }, + { + "epoch": 3.241127348643006, + "grad_norm": 0.8113856315612793, + "learning_rate": 2.3701689867547247e-06, + "loss": 0.4668, + "step": 1037 + }, + { + "epoch": 3.244258872651357, + "grad_norm": 1.507103443145752, + "learning_rate": 2.3658453779478464e-06, + "loss": 0.4527, + "step": 1038 + }, + { + "epoch": 3.2473903966597075, + "grad_norm": 0.7973915338516235, + "learning_rate": 2.3615221715157095e-06, + "loss": 0.4741, + "step": 1039 + }, + { + "epoch": 3.2505219206680582, + "grad_norm": 0.8206940293312073, + "learning_rate": 2.3571993804250635e-06, + "loss": 0.4582, + "step": 1040 + }, + { + "epoch": 3.2536534446764094, + "grad_norm": 0.8519637584686279, + "learning_rate": 2.3528770176414076e-06, + "loss": 0.4412, + "step": 1041 + }, + { + "epoch": 3.25678496868476, + "grad_norm": 0.7542241811752319, + "learning_rate": 2.348555096128959e-06, + "loss": 0.4861, + "step": 1042 + }, + { + "epoch": 3.259916492693111, + "grad_norm": 0.8074842691421509, + "learning_rate": 2.3442336288506125e-06, + "loss": 0.4707, + "step": 1043 + }, + { + "epoch": 3.2630480167014615, + "grad_norm": 0.8164265751838684, + "learning_rate": 2.3399126287678975e-06, + "loss": 0.4417, + "step": 1044 + }, + { + "epoch": 3.266179540709812, + "grad_norm": 0.7689628005027771, + "learning_rate": 2.3355921088409435e-06, + "loss": 0.4815, + "step": 1045 + }, + { + "epoch": 3.269311064718163, + "grad_norm": 0.7709240913391113, + "learning_rate": 2.3312720820284423e-06, + "loss": 0.4444, + "step": 1046 + }, + { + "epoch": 3.2724425887265136, + "grad_norm": 0.7340330481529236, + "learning_rate": 2.326952561287602e-06, + "loss": 0.4694, + "step": 1047 + }, + { + "epoch": 3.2755741127348643, + "grad_norm": 0.7785805463790894, + "learning_rate": 2.3226335595741154e-06, + "loss": 0.4273, + "step": 1048 + }, + { + "epoch": 3.278705636743215, + "grad_norm": 0.9968108534812927, + "learning_rate": 2.3183150898421196e-06, + "loss": 0.4782, + "step": 1049 + }, + { + "epoch": 3.2818371607515657, + "grad_norm": 0.7823553681373596, + "learning_rate": 2.3139971650441533e-06, + "loss": 0.444, + "step": 1050 + }, + { + "epoch": 3.2849686847599164, + "grad_norm": 0.7317377924919128, + "learning_rate": 2.3096797981311252e-06, + "loss": 0.4692, + "step": 1051 + }, + { + "epoch": 3.288100208768267, + "grad_norm": 0.8546518683433533, + "learning_rate": 2.3053630020522643e-06, + "loss": 0.4632, + "step": 1052 + }, + { + "epoch": 3.291231732776618, + "grad_norm": 1.2284396886825562, + "learning_rate": 2.301046789755093e-06, + "loss": 0.4409, + "step": 1053 + }, + { + "epoch": 3.2943632567849686, + "grad_norm": 0.8000460863113403, + "learning_rate": 2.2967311741853797e-06, + "loss": 0.4456, + "step": 1054 + }, + { + "epoch": 3.2974947807933193, + "grad_norm": 0.7689793109893799, + "learning_rate": 2.2924161682871053e-06, + "loss": 0.45, + "step": 1055 + }, + { + "epoch": 3.30062630480167, + "grad_norm": 0.8032956719398499, + "learning_rate": 2.288101785002421e-06, + "loss": 0.4817, + "step": 1056 + }, + { + "epoch": 3.3037578288100207, + "grad_norm": 0.6831309795379639, + "learning_rate": 2.283788037271609e-06, + "loss": 0.4502, + "step": 1057 + }, + { + "epoch": 3.306889352818372, + "grad_norm": 0.8581221103668213, + "learning_rate": 2.279474938033048e-06, + "loss": 0.4569, + "step": 1058 + }, + { + "epoch": 3.3100208768267225, + "grad_norm": 0.7937221527099609, + "learning_rate": 2.2751625002231696e-06, + "loss": 0.4451, + "step": 1059 + }, + { + "epoch": 3.3131524008350732, + "grad_norm": 0.8095264434814453, + "learning_rate": 2.270850736776422e-06, + "loss": 0.4462, + "step": 1060 + }, + { + "epoch": 3.316283924843424, + "grad_norm": 0.9141370058059692, + "learning_rate": 2.2665396606252332e-06, + "loss": 0.419, + "step": 1061 + }, + { + "epoch": 3.3194154488517746, + "grad_norm": 0.8648553490638733, + "learning_rate": 2.262229284699965e-06, + "loss": 0.4562, + "step": 1062 + }, + { + "epoch": 3.3225469728601253, + "grad_norm": 0.7716917395591736, + "learning_rate": 2.2579196219288825e-06, + "loss": 0.4734, + "step": 1063 + }, + { + "epoch": 3.325678496868476, + "grad_norm": 0.8074535727500916, + "learning_rate": 2.2536106852381116e-06, + "loss": 0.4272, + "step": 1064 + }, + { + "epoch": 3.3288100208768268, + "grad_norm": 0.8989127278327942, + "learning_rate": 2.249302487551599e-06, + "loss": 0.4728, + "step": 1065 + }, + { + "epoch": 3.3319415448851775, + "grad_norm": 0.7662765383720398, + "learning_rate": 2.2449950417910777e-06, + "loss": 0.4457, + "step": 1066 + }, + { + "epoch": 3.335073068893528, + "grad_norm": 0.7789275050163269, + "learning_rate": 2.24068836087602e-06, + "loss": 0.3919, + "step": 1067 + }, + { + "epoch": 3.338204592901879, + "grad_norm": 0.7359098196029663, + "learning_rate": 2.2363824577236097e-06, + "loss": 0.4876, + "step": 1068 + }, + { + "epoch": 3.3413361169102296, + "grad_norm": 0.9960948824882507, + "learning_rate": 2.232077345248695e-06, + "loss": 0.4894, + "step": 1069 + }, + { + "epoch": 3.3444676409185803, + "grad_norm": 0.7446064352989197, + "learning_rate": 2.2277730363637537e-06, + "loss": 0.4765, + "step": 1070 + }, + { + "epoch": 3.347599164926931, + "grad_norm": 0.7674328088760376, + "learning_rate": 2.2234695439788534e-06, + "loss": 0.4468, + "step": 1071 + }, + { + "epoch": 3.3507306889352817, + "grad_norm": 0.8957347869873047, + "learning_rate": 2.2191668810016105e-06, + "loss": 0.4733, + "step": 1072 + }, + { + "epoch": 3.3538622129436324, + "grad_norm": 0.9110277891159058, + "learning_rate": 2.2148650603371573e-06, + "loss": 0.4399, + "step": 1073 + }, + { + "epoch": 3.356993736951983, + "grad_norm": 0.9158220291137695, + "learning_rate": 2.2105640948880976e-06, + "loss": 0.4609, + "step": 1074 + }, + { + "epoch": 3.3601252609603343, + "grad_norm": 0.7630184888839722, + "learning_rate": 2.206263997554469e-06, + "loss": 0.4674, + "step": 1075 + }, + { + "epoch": 3.3632567849686845, + "grad_norm": 0.7975273728370667, + "learning_rate": 2.20196478123371e-06, + "loss": 0.4478, + "step": 1076 + }, + { + "epoch": 3.3663883089770357, + "grad_norm": 0.8825351595878601, + "learning_rate": 2.19766645882061e-06, + "loss": 0.4687, + "step": 1077 + }, + { + "epoch": 3.3695198329853864, + "grad_norm": 0.8907671570777893, + "learning_rate": 2.1933690432072817e-06, + "loss": 0.4223, + "step": 1078 + }, + { + "epoch": 3.372651356993737, + "grad_norm": 0.7449545860290527, + "learning_rate": 2.189072547283118e-06, + "loss": 0.4595, + "step": 1079 + }, + { + "epoch": 3.375782881002088, + "grad_norm": 0.8460972309112549, + "learning_rate": 2.184776983934751e-06, + "loss": 0.4443, + "step": 1080 + }, + { + "epoch": 3.3789144050104385, + "grad_norm": 0.7524845600128174, + "learning_rate": 2.1804823660460196e-06, + "loss": 0.4235, + "step": 1081 + }, + { + "epoch": 3.382045929018789, + "grad_norm": 0.8448389768600464, + "learning_rate": 2.176188706497921e-06, + "loss": 0.4387, + "step": 1082 + }, + { + "epoch": 3.38517745302714, + "grad_norm": 0.7701981663703918, + "learning_rate": 2.1718960181685838e-06, + "loss": 0.4257, + "step": 1083 + }, + { + "epoch": 3.3883089770354906, + "grad_norm": 0.8178983330726624, + "learning_rate": 2.167604313933219e-06, + "loss": 0.4983, + "step": 1084 + }, + { + "epoch": 3.3914405010438413, + "grad_norm": 0.7477235198020935, + "learning_rate": 2.163313606664091e-06, + "loss": 0.4559, + "step": 1085 + }, + { + "epoch": 3.394572025052192, + "grad_norm": 0.8127962350845337, + "learning_rate": 2.1590239092304694e-06, + "loss": 0.453, + "step": 1086 + }, + { + "epoch": 3.3977035490605427, + "grad_norm": 0.7462339997291565, + "learning_rate": 2.1547352344985966e-06, + "loss": 0.4697, + "step": 1087 + }, + { + "epoch": 3.4008350730688934, + "grad_norm": 0.9641384482383728, + "learning_rate": 2.1504475953316483e-06, + "loss": 0.4495, + "step": 1088 + }, + { + "epoch": 3.403966597077244, + "grad_norm": 0.7612512707710266, + "learning_rate": 2.146161004589693e-06, + "loss": 0.4579, + "step": 1089 + }, + { + "epoch": 3.407098121085595, + "grad_norm": 0.7547829747200012, + "learning_rate": 2.141875475129655e-06, + "loss": 0.4334, + "step": 1090 + }, + { + "epoch": 3.4102296450939455, + "grad_norm": 0.8036953806877136, + "learning_rate": 2.137591019805278e-06, + "loss": 0.4466, + "step": 1091 + }, + { + "epoch": 3.4133611691022967, + "grad_norm": 0.7319284081459045, + "learning_rate": 2.1333076514670784e-06, + "loss": 0.4942, + "step": 1092 + }, + { + "epoch": 3.416492693110647, + "grad_norm": 0.8278589248657227, + "learning_rate": 2.1290253829623165e-06, + "loss": 0.4554, + "step": 1093 + }, + { + "epoch": 3.419624217118998, + "grad_norm": 0.733059287071228, + "learning_rate": 2.124744227134954e-06, + "loss": 0.4187, + "step": 1094 + }, + { + "epoch": 3.422755741127349, + "grad_norm": 0.8222727179527283, + "learning_rate": 2.1204641968256136e-06, + "loss": 0.4587, + "step": 1095 + }, + { + "epoch": 3.4258872651356995, + "grad_norm": 0.8296732902526855, + "learning_rate": 2.1161853048715438e-06, + "loss": 0.4868, + "step": 1096 + }, + { + "epoch": 3.4290187891440502, + "grad_norm": 0.7309690713882446, + "learning_rate": 2.1119075641065758e-06, + "loss": 0.4594, + "step": 1097 + }, + { + "epoch": 3.432150313152401, + "grad_norm": 1.4901788234710693, + "learning_rate": 2.1076309873610916e-06, + "loss": 0.4216, + "step": 1098 + }, + { + "epoch": 3.4352818371607516, + "grad_norm": 0.7993581891059875, + "learning_rate": 2.1033555874619794e-06, + "loss": 0.4842, + "step": 1099 + }, + { + "epoch": 3.4384133611691023, + "grad_norm": 0.8846752643585205, + "learning_rate": 2.0990813772325995e-06, + "loss": 0.4395, + "step": 1100 + }, + { + "epoch": 3.441544885177453, + "grad_norm": 1.0796778202056885, + "learning_rate": 2.0948083694927436e-06, + "loss": 0.4573, + "step": 1101 + }, + { + "epoch": 3.4446764091858038, + "grad_norm": 0.74623042345047, + "learning_rate": 2.090536577058595e-06, + "loss": 0.4563, + "step": 1102 + }, + { + "epoch": 3.4478079331941545, + "grad_norm": 0.8245521783828735, + "learning_rate": 2.086266012742692e-06, + "loss": 0.4433, + "step": 1103 + }, + { + "epoch": 3.450939457202505, + "grad_norm": 0.8003777265548706, + "learning_rate": 2.081996689353893e-06, + "loss": 0.4599, + "step": 1104 + }, + { + "epoch": 3.454070981210856, + "grad_norm": 0.8309001922607422, + "learning_rate": 2.0777286196973302e-06, + "loss": 0.485, + "step": 1105 + }, + { + "epoch": 3.4572025052192066, + "grad_norm": 0.8299122452735901, + "learning_rate": 2.0734618165743782e-06, + "loss": 0.4685, + "step": 1106 + }, + { + "epoch": 3.4603340292275573, + "grad_norm": 0.9347029328346252, + "learning_rate": 2.069196292782611e-06, + "loss": 0.4615, + "step": 1107 + }, + { + "epoch": 3.463465553235908, + "grad_norm": 0.7146593332290649, + "learning_rate": 2.064932061115766e-06, + "loss": 0.4433, + "step": 1108 + }, + { + "epoch": 3.4665970772442587, + "grad_norm": 0.7674420475959778, + "learning_rate": 2.0606691343637063e-06, + "loss": 0.4444, + "step": 1109 + }, + { + "epoch": 3.4697286012526094, + "grad_norm": 0.7925504446029663, + "learning_rate": 2.05640752531238e-06, + "loss": 0.4631, + "step": 1110 + }, + { + "epoch": 3.4728601252609606, + "grad_norm": 0.7755677700042725, + "learning_rate": 2.0521472467437825e-06, + "loss": 0.4709, + "step": 1111 + }, + { + "epoch": 3.4759916492693113, + "grad_norm": 0.8535795211791992, + "learning_rate": 2.0478883114359187e-06, + "loss": 0.442, + "step": 1112 + }, + { + "epoch": 3.479123173277662, + "grad_norm": 0.723953127861023, + "learning_rate": 2.043630732162767e-06, + "loss": 0.4782, + "step": 1113 + }, + { + "epoch": 3.4822546972860127, + "grad_norm": 0.7817316651344299, + "learning_rate": 2.0393745216942343e-06, + "loss": 0.4841, + "step": 1114 + }, + { + "epoch": 3.4853862212943634, + "grad_norm": 0.8878781795501709, + "learning_rate": 2.0351196927961268e-06, + "loss": 0.4673, + "step": 1115 + }, + { + "epoch": 3.488517745302714, + "grad_norm": 0.8645241856575012, + "learning_rate": 2.030866258230104e-06, + "loss": 0.432, + "step": 1116 + }, + { + "epoch": 3.491649269311065, + "grad_norm": 0.7294583320617676, + "learning_rate": 2.026614230753643e-06, + "loss": 0.4683, + "step": 1117 + }, + { + "epoch": 3.4947807933194155, + "grad_norm": 0.7412407994270325, + "learning_rate": 2.022363623120001e-06, + "loss": 0.4523, + "step": 1118 + }, + { + "epoch": 3.497912317327766, + "grad_norm": 0.8559291362762451, + "learning_rate": 2.0181144480781787e-06, + "loss": 0.4309, + "step": 1119 + }, + { + "epoch": 3.501043841336117, + "grad_norm": 0.7442825436592102, + "learning_rate": 2.0138667183728775e-06, + "loss": 0.4096, + "step": 1120 + }, + { + "epoch": 3.5041753653444676, + "grad_norm": 0.7605662941932678, + "learning_rate": 2.0096204467444645e-06, + "loss": 0.4404, + "step": 1121 + }, + { + "epoch": 3.5073068893528183, + "grad_norm": 0.7984277009963989, + "learning_rate": 2.005375645928935e-06, + "loss": 0.4661, + "step": 1122 + }, + { + "epoch": 3.510438413361169, + "grad_norm": 1.1044552326202393, + "learning_rate": 2.001132328657869e-06, + "loss": 0.4185, + "step": 1123 + }, + { + "epoch": 3.5135699373695197, + "grad_norm": 0.8210328817367554, + "learning_rate": 1.996890507658401e-06, + "loss": 0.4746, + "step": 1124 + }, + { + "epoch": 3.5167014613778704, + "grad_norm": 0.7302148342132568, + "learning_rate": 1.9926501956531758e-06, + "loss": 0.4333, + "step": 1125 + }, + { + "epoch": 3.519832985386221, + "grad_norm": 0.7713826894760132, + "learning_rate": 1.9884114053603114e-06, + "loss": 0.4485, + "step": 1126 + }, + { + "epoch": 3.522964509394572, + "grad_norm": 0.7386549711227417, + "learning_rate": 1.984174149493365e-06, + "loss": 0.4678, + "step": 1127 + }, + { + "epoch": 3.526096033402923, + "grad_norm": 0.8006004095077515, + "learning_rate": 1.979938440761287e-06, + "loss": 0.4755, + "step": 1128 + }, + { + "epoch": 3.5292275574112733, + "grad_norm": 1.0635333061218262, + "learning_rate": 1.97570429186839e-06, + "loss": 0.4985, + "step": 1129 + }, + { + "epoch": 3.5323590814196244, + "grad_norm": 0.7470075488090515, + "learning_rate": 1.9714717155143083e-06, + "loss": 0.4307, + "step": 1130 + }, + { + "epoch": 3.535490605427975, + "grad_norm": 0.8314558863639832, + "learning_rate": 1.967240724393959e-06, + "loss": 0.4894, + "step": 1131 + }, + { + "epoch": 3.538622129436326, + "grad_norm": 0.761443018913269, + "learning_rate": 1.963011331197506e-06, + "loss": 0.4653, + "step": 1132 + }, + { + "epoch": 3.5417536534446765, + "grad_norm": 0.7483212351799011, + "learning_rate": 1.9587835486103163e-06, + "loss": 0.4456, + "step": 1133 + }, + { + "epoch": 3.5448851774530272, + "grad_norm": 0.7794159054756165, + "learning_rate": 1.9545573893129306e-06, + "loss": 0.4707, + "step": 1134 + }, + { + "epoch": 3.548016701461378, + "grad_norm": 0.8016185760498047, + "learning_rate": 1.950332865981019e-06, + "loss": 0.4547, + "step": 1135 + }, + { + "epoch": 3.5511482254697286, + "grad_norm": 0.8089869618415833, + "learning_rate": 1.9461099912853453e-06, + "loss": 0.4499, + "step": 1136 + }, + { + "epoch": 3.5542797494780793, + "grad_norm": 0.7774782180786133, + "learning_rate": 1.9418887778917286e-06, + "loss": 0.4531, + "step": 1137 + }, + { + "epoch": 3.55741127348643, + "grad_norm": 0.7793645262718201, + "learning_rate": 1.937669238461003e-06, + "loss": 0.4553, + "step": 1138 + }, + { + "epoch": 3.5605427974947808, + "grad_norm": 0.8139959573745728, + "learning_rate": 1.933451385648985e-06, + "loss": 0.458, + "step": 1139 + }, + { + "epoch": 3.5636743215031315, + "grad_norm": 0.7517053484916687, + "learning_rate": 1.929235232106431e-06, + "loss": 0.4779, + "step": 1140 + }, + { + "epoch": 3.566805845511482, + "grad_norm": 0.8851562142372131, + "learning_rate": 1.925020790479e-06, + "loss": 0.425, + "step": 1141 + }, + { + "epoch": 3.569937369519833, + "grad_norm": 0.8129401803016663, + "learning_rate": 1.920808073407218e-06, + "loss": 0.4616, + "step": 1142 + }, + { + "epoch": 3.5730688935281836, + "grad_norm": 0.7110117077827454, + "learning_rate": 1.916597093526437e-06, + "loss": 0.4748, + "step": 1143 + }, + { + "epoch": 3.5762004175365343, + "grad_norm": 0.8268555402755737, + "learning_rate": 1.912387863466798e-06, + "loss": 0.4752, + "step": 1144 + }, + { + "epoch": 3.5793319415448854, + "grad_norm": 1.1036733388900757, + "learning_rate": 1.9081803958531967e-06, + "loss": 0.4879, + "step": 1145 + }, + { + "epoch": 3.5824634655532357, + "grad_norm": 0.8561109304428101, + "learning_rate": 1.9039747033052395e-06, + "loss": 0.4409, + "step": 1146 + }, + { + "epoch": 3.585594989561587, + "grad_norm": 0.7597541809082031, + "learning_rate": 1.8997707984372119e-06, + "loss": 0.4518, + "step": 1147 + }, + { + "epoch": 3.588726513569937, + "grad_norm": 0.7225353121757507, + "learning_rate": 1.8955686938580329e-06, + "loss": 0.4735, + "step": 1148 + }, + { + "epoch": 3.5918580375782883, + "grad_norm": 0.9409791231155396, + "learning_rate": 1.8913684021712264e-06, + "loss": 0.4249, + "step": 1149 + }, + { + "epoch": 3.594989561586639, + "grad_norm": 0.8757275342941284, + "learning_rate": 1.8871699359748763e-06, + "loss": 0.4479, + "step": 1150 + }, + { + "epoch": 3.5981210855949897, + "grad_norm": 0.8090003728866577, + "learning_rate": 1.882973307861593e-06, + "loss": 0.4846, + "step": 1151 + }, + { + "epoch": 3.6012526096033404, + "grad_norm": 0.7568825483322144, + "learning_rate": 1.8787785304184726e-06, + "loss": 0.4301, + "step": 1152 + }, + { + "epoch": 3.604384133611691, + "grad_norm": 0.8233815431594849, + "learning_rate": 1.8745856162270592e-06, + "loss": 0.4838, + "step": 1153 + }, + { + "epoch": 3.607515657620042, + "grad_norm": 0.7817628979682922, + "learning_rate": 1.8703945778633121e-06, + "loss": 0.4669, + "step": 1154 + }, + { + "epoch": 3.6106471816283925, + "grad_norm": 0.821304202079773, + "learning_rate": 1.8662054278975605e-06, + "loss": 0.4536, + "step": 1155 + }, + { + "epoch": 3.613778705636743, + "grad_norm": 0.9304089546203613, + "learning_rate": 1.8620181788944712e-06, + "loss": 0.4489, + "step": 1156 + }, + { + "epoch": 3.616910229645094, + "grad_norm": 0.8202670216560364, + "learning_rate": 1.8578328434130114e-06, + "loss": 0.4309, + "step": 1157 + }, + { + "epoch": 3.6200417536534446, + "grad_norm": 0.8890257477760315, + "learning_rate": 1.8536494340064051e-06, + "loss": 0.4736, + "step": 1158 + }, + { + "epoch": 3.6231732776617953, + "grad_norm": 0.7940590381622314, + "learning_rate": 1.8494679632221013e-06, + "loss": 0.4468, + "step": 1159 + }, + { + "epoch": 3.626304801670146, + "grad_norm": 0.8388273119926453, + "learning_rate": 1.845288443601736e-06, + "loss": 0.4753, + "step": 1160 + }, + { + "epoch": 3.6294363256784967, + "grad_norm": 0.9392285346984863, + "learning_rate": 1.84111088768109e-06, + "loss": 0.4736, + "step": 1161 + }, + { + "epoch": 3.632567849686848, + "grad_norm": 0.7411681413650513, + "learning_rate": 1.8369353079900576e-06, + "loss": 0.4303, + "step": 1162 + }, + { + "epoch": 3.635699373695198, + "grad_norm": 0.8722569942474365, + "learning_rate": 1.8327617170526014e-06, + "loss": 0.4604, + "step": 1163 + }, + { + "epoch": 3.6388308977035493, + "grad_norm": 0.786891758441925, + "learning_rate": 1.8285901273867229e-06, + "loss": 0.4756, + "step": 1164 + }, + { + "epoch": 3.6419624217118995, + "grad_norm": 0.8159083724021912, + "learning_rate": 1.824420551504419e-06, + "loss": 0.4675, + "step": 1165 + }, + { + "epoch": 3.6450939457202507, + "grad_norm": 0.8271334767341614, + "learning_rate": 1.8202530019116487e-06, + "loss": 0.4311, + "step": 1166 + }, + { + "epoch": 3.6482254697286014, + "grad_norm": 0.7617189288139343, + "learning_rate": 1.816087491108292e-06, + "loss": 0.4522, + "step": 1167 + }, + { + "epoch": 3.651356993736952, + "grad_norm": 0.7248172760009766, + "learning_rate": 1.8119240315881126e-06, + "loss": 0.465, + "step": 1168 + }, + { + "epoch": 3.654488517745303, + "grad_norm": 0.8606911897659302, + "learning_rate": 1.8077626358387235e-06, + "loss": 0.4524, + "step": 1169 + }, + { + "epoch": 3.6576200417536535, + "grad_norm": 0.8571308851242065, + "learning_rate": 1.8036033163415484e-06, + "loss": 0.4625, + "step": 1170 + }, + { + "epoch": 3.6607515657620042, + "grad_norm": 0.7239511609077454, + "learning_rate": 1.7994460855717812e-06, + "loss": 0.5025, + "step": 1171 + }, + { + "epoch": 3.663883089770355, + "grad_norm": 0.7958929538726807, + "learning_rate": 1.7952909559983544e-06, + "loss": 0.4382, + "step": 1172 + }, + { + "epoch": 3.6670146137787056, + "grad_norm": 0.7920124530792236, + "learning_rate": 1.7911379400838947e-06, + "loss": 0.4393, + "step": 1173 + }, + { + "epoch": 3.6701461377870563, + "grad_norm": 0.8072578310966492, + "learning_rate": 1.7869870502846903e-06, + "loss": 0.4627, + "step": 1174 + }, + { + "epoch": 3.673277661795407, + "grad_norm": 0.8586218357086182, + "learning_rate": 1.7828382990506543e-06, + "loss": 0.4456, + "step": 1175 + }, + { + "epoch": 3.6764091858037578, + "grad_norm": 0.8741613030433655, + "learning_rate": 1.7786916988252845e-06, + "loss": 0.4613, + "step": 1176 + }, + { + "epoch": 3.6795407098121085, + "grad_norm": 0.7691352367401123, + "learning_rate": 1.774547262045626e-06, + "loss": 0.4641, + "step": 1177 + }, + { + "epoch": 3.682672233820459, + "grad_norm": 0.7866089940071106, + "learning_rate": 1.7704050011422357e-06, + "loss": 0.4308, + "step": 1178 + }, + { + "epoch": 3.68580375782881, + "grad_norm": 0.9934884309768677, + "learning_rate": 1.7662649285391447e-06, + "loss": 0.4434, + "step": 1179 + }, + { + "epoch": 3.6889352818371606, + "grad_norm": 0.794385552406311, + "learning_rate": 1.7621270566538204e-06, + "loss": 0.4481, + "step": 1180 + }, + { + "epoch": 3.6920668058455117, + "grad_norm": 0.7573548555374146, + "learning_rate": 1.7579913978971296e-06, + "loss": 0.4525, + "step": 1181 + }, + { + "epoch": 3.695198329853862, + "grad_norm": 0.7073976993560791, + "learning_rate": 1.7538579646733023e-06, + "loss": 0.4214, + "step": 1182 + }, + { + "epoch": 3.698329853862213, + "grad_norm": 0.8009579181671143, + "learning_rate": 1.7497267693798902e-06, + "loss": 0.4542, + "step": 1183 + }, + { + "epoch": 3.701461377870564, + "grad_norm": 1.4488778114318848, + "learning_rate": 1.7455978244077348e-06, + "loss": 0.443, + "step": 1184 + }, + { + "epoch": 3.7045929018789145, + "grad_norm": 1.0529266595840454, + "learning_rate": 1.7414711421409292e-06, + "loss": 0.4908, + "step": 1185 + }, + { + "epoch": 3.7077244258872653, + "grad_norm": 0.757431149482727, + "learning_rate": 1.7373467349567775e-06, + "loss": 0.4454, + "step": 1186 + }, + { + "epoch": 3.710855949895616, + "grad_norm": 0.7086379528045654, + "learning_rate": 1.733224615225763e-06, + "loss": 0.4292, + "step": 1187 + }, + { + "epoch": 3.7139874739039667, + "grad_norm": 0.7454110383987427, + "learning_rate": 1.7291047953115049e-06, + "loss": 0.4421, + "step": 1188 + }, + { + "epoch": 3.7171189979123174, + "grad_norm": 0.804027795791626, + "learning_rate": 1.7249872875707257e-06, + "loss": 0.4819, + "step": 1189 + }, + { + "epoch": 3.720250521920668, + "grad_norm": 0.8159645199775696, + "learning_rate": 1.7208721043532146e-06, + "loss": 0.4628, + "step": 1190 + }, + { + "epoch": 3.723382045929019, + "grad_norm": 0.8451672196388245, + "learning_rate": 1.7167592580017866e-06, + "loss": 0.4773, + "step": 1191 + }, + { + "epoch": 3.7265135699373695, + "grad_norm": 0.920553982257843, + "learning_rate": 1.7126487608522492e-06, + "loss": 0.4787, + "step": 1192 + }, + { + "epoch": 3.72964509394572, + "grad_norm": 0.9169708490371704, + "learning_rate": 1.7085406252333613e-06, + "loss": 0.4543, + "step": 1193 + }, + { + "epoch": 3.732776617954071, + "grad_norm": 0.7245096564292908, + "learning_rate": 1.7044348634668023e-06, + "loss": 0.4599, + "step": 1194 + }, + { + "epoch": 3.7359081419624216, + "grad_norm": 0.835832417011261, + "learning_rate": 1.7003314878671284e-06, + "loss": 0.4794, + "step": 1195 + }, + { + "epoch": 3.7390396659707723, + "grad_norm": 0.8455896973609924, + "learning_rate": 1.696230510741742e-06, + "loss": 0.4329, + "step": 1196 + }, + { + "epoch": 3.742171189979123, + "grad_norm": 0.743611752986908, + "learning_rate": 1.692131944390849e-06, + "loss": 0.4455, + "step": 1197 + }, + { + "epoch": 3.745302713987474, + "grad_norm": 1.1334915161132812, + "learning_rate": 1.6880358011074272e-06, + "loss": 0.4673, + "step": 1198 + }, + { + "epoch": 3.7484342379958244, + "grad_norm": 0.6935724020004272, + "learning_rate": 1.6839420931771828e-06, + "loss": 0.4686, + "step": 1199 + }, + { + "epoch": 3.7515657620041756, + "grad_norm": 0.962006151676178, + "learning_rate": 1.6798508328785213e-06, + "loss": 0.4474, + "step": 1200 + }, + { + "epoch": 3.754697286012526, + "grad_norm": 1.1499714851379395, + "learning_rate": 1.6757620324825047e-06, + "loss": 0.4659, + "step": 1201 + }, + { + "epoch": 3.757828810020877, + "grad_norm": 0.7689645886421204, + "learning_rate": 1.6716757042528192e-06, + "loss": 0.4552, + "step": 1202 + }, + { + "epoch": 3.7609603340292277, + "grad_norm": 0.7482030391693115, + "learning_rate": 1.6675918604457352e-06, + "loss": 0.4747, + "step": 1203 + }, + { + "epoch": 3.7640918580375784, + "grad_norm": 0.7727032899856567, + "learning_rate": 1.6635105133100686e-06, + "loss": 0.4508, + "step": 1204 + }, + { + "epoch": 3.767223382045929, + "grad_norm": 0.8722149133682251, + "learning_rate": 1.6594316750871514e-06, + "loss": 0.4685, + "step": 1205 + }, + { + "epoch": 3.77035490605428, + "grad_norm": 0.771304726600647, + "learning_rate": 1.6553553580107884e-06, + "loss": 0.4418, + "step": 1206 + }, + { + "epoch": 3.7734864300626305, + "grad_norm": 0.767315149307251, + "learning_rate": 1.6512815743072214e-06, + "loss": 0.4532, + "step": 1207 + }, + { + "epoch": 3.776617954070981, + "grad_norm": 0.8825518488883972, + "learning_rate": 1.6472103361950976e-06, + "loss": 0.468, + "step": 1208 + }, + { + "epoch": 3.779749478079332, + "grad_norm": 0.8887981176376343, + "learning_rate": 1.6431416558854243e-06, + "loss": 0.4264, + "step": 1209 + }, + { + "epoch": 3.7828810020876826, + "grad_norm": 0.8399733304977417, + "learning_rate": 1.63907554558154e-06, + "loss": 0.4405, + "step": 1210 + }, + { + "epoch": 3.7860125260960333, + "grad_norm": 0.8112586140632629, + "learning_rate": 1.6350120174790751e-06, + "loss": 0.445, + "step": 1211 + }, + { + "epoch": 3.789144050104384, + "grad_norm": 0.860775351524353, + "learning_rate": 1.6309510837659137e-06, + "loss": 0.4557, + "step": 1212 + }, + { + "epoch": 3.7922755741127347, + "grad_norm": 0.8522343039512634, + "learning_rate": 1.626892756622161e-06, + "loss": 0.481, + "step": 1213 + }, + { + "epoch": 3.7954070981210855, + "grad_norm": 0.7927511930465698, + "learning_rate": 1.6228370482200988e-06, + "loss": 0.4328, + "step": 1214 + }, + { + "epoch": 3.798538622129436, + "grad_norm": 0.7350064516067505, + "learning_rate": 1.6187839707241604e-06, + "loss": 0.4604, + "step": 1215 + }, + { + "epoch": 3.801670146137787, + "grad_norm": 0.8363698124885559, + "learning_rate": 1.6147335362908847e-06, + "loss": 0.4271, + "step": 1216 + }, + { + "epoch": 3.804801670146138, + "grad_norm": 1.080613613128662, + "learning_rate": 1.610685757068885e-06, + "loss": 0.447, + "step": 1217 + }, + { + "epoch": 3.8079331941544883, + "grad_norm": 1.1507478952407837, + "learning_rate": 1.6066406451988104e-06, + "loss": 0.4664, + "step": 1218 + }, + { + "epoch": 3.8110647181628394, + "grad_norm": 0.7778187990188599, + "learning_rate": 1.6025982128133073e-06, + "loss": 0.456, + "step": 1219 + }, + { + "epoch": 3.81419624217119, + "grad_norm": 0.8383583426475525, + "learning_rate": 1.5985584720369876e-06, + "loss": 0.4684, + "step": 1220 + }, + { + "epoch": 3.817327766179541, + "grad_norm": 0.7743321061134338, + "learning_rate": 1.5945214349863914e-06, + "loss": 0.4567, + "step": 1221 + }, + { + "epoch": 3.8204592901878915, + "grad_norm": 0.8020774126052856, + "learning_rate": 1.5904871137699462e-06, + "loss": 0.4175, + "step": 1222 + }, + { + "epoch": 3.8235908141962422, + "grad_norm": 0.790318489074707, + "learning_rate": 1.5864555204879375e-06, + "loss": 0.469, + "step": 1223 + }, + { + "epoch": 3.826722338204593, + "grad_norm": 0.8583689332008362, + "learning_rate": 1.5824266672324652e-06, + "loss": 0.4931, + "step": 1224 + }, + { + "epoch": 3.8298538622129437, + "grad_norm": 0.7788206934928894, + "learning_rate": 1.5784005660874125e-06, + "loss": 0.4643, + "step": 1225 + }, + { + "epoch": 3.8329853862212944, + "grad_norm": 0.8385717868804932, + "learning_rate": 1.574377229128409e-06, + "loss": 0.4567, + "step": 1226 + }, + { + "epoch": 3.836116910229645, + "grad_norm": 0.8447727560997009, + "learning_rate": 1.5703566684227922e-06, + "loss": 0.42, + "step": 1227 + }, + { + "epoch": 3.8392484342379958, + "grad_norm": 0.7286496758460999, + "learning_rate": 1.5663388960295742e-06, + "loss": 0.4603, + "step": 1228 + }, + { + "epoch": 3.8423799582463465, + "grad_norm": 0.8493947982788086, + "learning_rate": 1.562323923999401e-06, + "loss": 0.4731, + "step": 1229 + }, + { + "epoch": 3.845511482254697, + "grad_norm": 0.8641151785850525, + "learning_rate": 1.5583117643745233e-06, + "loss": 0.4491, + "step": 1230 + }, + { + "epoch": 3.848643006263048, + "grad_norm": 0.9493702054023743, + "learning_rate": 1.5543024291887532e-06, + "loss": 0.411, + "step": 1231 + }, + { + "epoch": 3.8517745302713986, + "grad_norm": 0.7246205806732178, + "learning_rate": 1.5502959304674337e-06, + "loss": 0.4569, + "step": 1232 + }, + { + "epoch": 3.8549060542797493, + "grad_norm": 0.7576872110366821, + "learning_rate": 1.5462922802273994e-06, + "loss": 0.4548, + "step": 1233 + }, + { + "epoch": 3.8580375782881005, + "grad_norm": 0.7710747718811035, + "learning_rate": 1.5422914904769404e-06, + "loss": 0.447, + "step": 1234 + }, + { + "epoch": 3.8611691022964507, + "grad_norm": 0.7661204934120178, + "learning_rate": 1.5382935732157677e-06, + "loss": 0.4601, + "step": 1235 + }, + { + "epoch": 3.864300626304802, + "grad_norm": 0.7133070826530457, + "learning_rate": 1.5342985404349788e-06, + "loss": 0.4245, + "step": 1236 + }, + { + "epoch": 3.867432150313152, + "grad_norm": 0.7716459631919861, + "learning_rate": 1.5303064041170163e-06, + "loss": 0.4543, + "step": 1237 + }, + { + "epoch": 3.8705636743215033, + "grad_norm": 0.737501859664917, + "learning_rate": 1.5263171762356388e-06, + "loss": 0.4405, + "step": 1238 + }, + { + "epoch": 3.873695198329854, + "grad_norm": 0.7885998487472534, + "learning_rate": 1.5223308687558786e-06, + "loss": 0.4412, + "step": 1239 + }, + { + "epoch": 3.8768267223382047, + "grad_norm": 0.7516661882400513, + "learning_rate": 1.5183474936340092e-06, + "loss": 0.4185, + "step": 1240 + }, + { + "epoch": 3.8799582463465554, + "grad_norm": 0.8790446519851685, + "learning_rate": 1.5143670628175111e-06, + "loss": 0.444, + "step": 1241 + }, + { + "epoch": 3.883089770354906, + "grad_norm": 0.7695789337158203, + "learning_rate": 1.5103895882450315e-06, + "loss": 0.4615, + "step": 1242 + }, + { + "epoch": 3.886221294363257, + "grad_norm": 0.7859196662902832, + "learning_rate": 1.506415081846353e-06, + "loss": 0.4379, + "step": 1243 + }, + { + "epoch": 3.8893528183716075, + "grad_norm": 0.7771942615509033, + "learning_rate": 1.5024435555423522e-06, + "loss": 0.5013, + "step": 1244 + }, + { + "epoch": 3.892484342379958, + "grad_norm": 0.7415695190429688, + "learning_rate": 1.498475021244971e-06, + "loss": 0.4861, + "step": 1245 + }, + { + "epoch": 3.895615866388309, + "grad_norm": 0.7909391522407532, + "learning_rate": 1.4945094908571755e-06, + "loss": 0.4599, + "step": 1246 + }, + { + "epoch": 3.8987473903966596, + "grad_norm": 0.7749060988426208, + "learning_rate": 1.490546976272923e-06, + "loss": 0.3986, + "step": 1247 + }, + { + "epoch": 3.9018789144050103, + "grad_norm": 0.810681164264679, + "learning_rate": 1.4865874893771248e-06, + "loss": 0.4495, + "step": 1248 + }, + { + "epoch": 3.905010438413361, + "grad_norm": 0.8018531799316406, + "learning_rate": 1.4826310420456103e-06, + "loss": 0.4426, + "step": 1249 + }, + { + "epoch": 3.9081419624217117, + "grad_norm": 0.756064236164093, + "learning_rate": 1.4786776461450924e-06, + "loss": 0.4474, + "step": 1250 + }, + { + "epoch": 3.911273486430063, + "grad_norm": 0.7581740021705627, + "learning_rate": 1.4747273135331347e-06, + "loss": 0.4494, + "step": 1251 + }, + { + "epoch": 3.914405010438413, + "grad_norm": 0.7666076421737671, + "learning_rate": 1.4707800560581086e-06, + "loss": 0.4593, + "step": 1252 + }, + { + "epoch": 3.9175365344467643, + "grad_norm": 0.7339973449707031, + "learning_rate": 1.4668358855591664e-06, + "loss": 0.4682, + "step": 1253 + }, + { + "epoch": 3.9206680584551146, + "grad_norm": 0.8504599928855896, + "learning_rate": 1.4628948138661974e-06, + "loss": 0.4504, + "step": 1254 + }, + { + "epoch": 3.9237995824634657, + "grad_norm": 0.8332642912864685, + "learning_rate": 1.4589568527997985e-06, + "loss": 0.5024, + "step": 1255 + }, + { + "epoch": 3.9269311064718164, + "grad_norm": 0.7813694477081299, + "learning_rate": 1.4550220141712384e-06, + "loss": 0.4547, + "step": 1256 + }, + { + "epoch": 3.930062630480167, + "grad_norm": 0.842258632183075, + "learning_rate": 1.451090309782417e-06, + "loss": 0.4584, + "step": 1257 + }, + { + "epoch": 3.933194154488518, + "grad_norm": 0.8159133791923523, + "learning_rate": 1.4471617514258373e-06, + "loss": 0.4538, + "step": 1258 + }, + { + "epoch": 3.9363256784968685, + "grad_norm": 0.8117021322250366, + "learning_rate": 1.4432363508845626e-06, + "loss": 0.4315, + "step": 1259 + }, + { + "epoch": 3.9394572025052192, + "grad_norm": 0.8087465167045593, + "learning_rate": 1.4393141199321881e-06, + "loss": 0.4367, + "step": 1260 + }, + { + "epoch": 3.94258872651357, + "grad_norm": 0.7954697012901306, + "learning_rate": 1.435395070332801e-06, + "loss": 0.4515, + "step": 1261 + }, + { + "epoch": 3.9457202505219207, + "grad_norm": 0.7305286526679993, + "learning_rate": 1.4314792138409454e-06, + "loss": 0.4879, + "step": 1262 + }, + { + "epoch": 3.9488517745302714, + "grad_norm": 0.8883433938026428, + "learning_rate": 1.4275665622015908e-06, + "loss": 0.4489, + "step": 1263 + }, + { + "epoch": 3.951983298538622, + "grad_norm": 0.8176298141479492, + "learning_rate": 1.4236571271500909e-06, + "loss": 0.4583, + "step": 1264 + }, + { + "epoch": 3.9551148225469728, + "grad_norm": 0.8042430281639099, + "learning_rate": 1.4197509204121563e-06, + "loss": 0.4277, + "step": 1265 + }, + { + "epoch": 3.9582463465553235, + "grad_norm": 0.8153829574584961, + "learning_rate": 1.4158479537038095e-06, + "loss": 0.4389, + "step": 1266 + }, + { + "epoch": 3.961377870563674, + "grad_norm": 0.7908188104629517, + "learning_rate": 1.4119482387313588e-06, + "loss": 0.4421, + "step": 1267 + }, + { + "epoch": 3.964509394572025, + "grad_norm": 0.831758975982666, + "learning_rate": 1.4080517871913596e-06, + "loss": 0.4308, + "step": 1268 + }, + { + "epoch": 3.9676409185803756, + "grad_norm": 0.8191989064216614, + "learning_rate": 1.4041586107705758e-06, + "loss": 0.4654, + "step": 1269 + }, + { + "epoch": 3.9707724425887267, + "grad_norm": 0.9455055594444275, + "learning_rate": 1.4002687211459524e-06, + "loss": 0.4668, + "step": 1270 + }, + { + "epoch": 3.973903966597077, + "grad_norm": 0.9271034002304077, + "learning_rate": 1.396382129984572e-06, + "loss": 0.4414, + "step": 1271 + }, + { + "epoch": 3.977035490605428, + "grad_norm": 0.7652955651283264, + "learning_rate": 1.392498848943627e-06, + "loss": 0.4575, + "step": 1272 + }, + { + "epoch": 3.980167014613779, + "grad_norm": 0.7850046157836914, + "learning_rate": 1.3886188896703816e-06, + "loss": 0.4554, + "step": 1273 + }, + { + "epoch": 3.9832985386221296, + "grad_norm": 0.7194349765777588, + "learning_rate": 1.3847422638021357e-06, + "loss": 0.437, + "step": 1274 + }, + { + "epoch": 3.9864300626304803, + "grad_norm": 0.8726270198822021, + "learning_rate": 1.3808689829661899e-06, + "loss": 0.4657, + "step": 1275 + }, + { + "epoch": 3.989561586638831, + "grad_norm": 0.7741451263427734, + "learning_rate": 1.3769990587798146e-06, + "loss": 0.3931, + "step": 1276 + } + ], + "logging_steps": 1, + "max_steps": 1914, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 319, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3.866690661704217e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-1276/training_args.bin b/checkpoint-1276/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8067ee9c1c0bc752bdfd00cfcaf1a6e717d2356b --- /dev/null +++ b/checkpoint-1276/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c650156a192cae0a2070c4323ee8a93e9b52fb76041d59ae0633b98389585727 +size 7928 diff --git a/checkpoint-1276/zero_to_fp32.py b/checkpoint-1276/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-1276/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-1595/README.md b/checkpoint-1595/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f4a3934800eeb082a0cb833d7b6af4f68eed3615 --- /dev/null +++ b/checkpoint-1595/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3_3-Nemotron-Super-49B-v1 +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.15.0 \ No newline at end of file diff --git a/checkpoint-1595/adapter_config.json b/checkpoint-1595/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..1edb33780e2306c6b19fd727be8e9b8b35f237c4 --- /dev/null +++ b/checkpoint-1595/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3_3-Nemotron-Super-49B-v1", + "bias": "none", + "corda_config": null, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "k_proj", + "down_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-1595/adapter_model.safetensors b/checkpoint-1595/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..060917ffa3d5b31b004e46e40378c582bde06737 --- /dev/null +++ b/checkpoint-1595/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d35d441689ced9e852803471814dd913950ba023247e90cf0f39c559f8737edd +size 9016826528 diff --git a/checkpoint-1595/global_step1592/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-1595/global_step1592/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..e7782778faa1f579e9feebecfad2740a3b47cef8 --- /dev/null +++ b/checkpoint-1595/global_step1592/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eef7ae5af7ab5ad410fdb8eb5b22f8645e4ffeb24143f04db6f6f04e7a5fde3e +size 27050164444 diff --git a/checkpoint-1595/global_step1592/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-1595/global_step1592/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..61eef1ac5e0373f31da6117a6e419c60dd409a21 --- /dev/null +++ b/checkpoint-1595/global_step1592/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97198b3e57b90b1b2b2e567d1cc90e29b870c31b282dd01bf137d73411429cf2 +size 27050169884 diff --git a/checkpoint-1595/global_step1592/mp_rank_00_model_states.pt b/checkpoint-1595/global_step1592/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..81c8370e3a784f0ecc677f2a9cf82fa72be43217 --- /dev/null +++ b/checkpoint-1595/global_step1592/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cd85f629390b3f992baab88f003d5b350091934f627a5849cf1530fad54eec7d +size 9776788601 diff --git a/checkpoint-1595/latest b/checkpoint-1595/latest new file mode 100644 index 0000000000000000000000000000000000000000..e3a83295a735ddc825ec0b9a9572d40b9610afc8 --- /dev/null +++ b/checkpoint-1595/latest @@ -0,0 +1 @@ +global_step1592 \ No newline at end of file diff --git a/checkpoint-1595/rng_state_0.pth b/checkpoint-1595/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..f7a7da06359fa62040befa6b2c55fdda15e3f700 --- /dev/null +++ b/checkpoint-1595/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8385fc4f5dbe1eb501284b9c0e740cdc8c640953df92a6173e9527f6939ac71 +size 14512 diff --git a/checkpoint-1595/rng_state_1.pth b/checkpoint-1595/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..5bf28c27329b2bc1b484e1eea5340aec233d9718 --- /dev/null +++ b/checkpoint-1595/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5a1edd2f0a0a5dac1c9d5d46cf82665e001446d5153cc015216e691b160d490b +size 14512 diff --git a/checkpoint-1595/scheduler.pt b/checkpoint-1595/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..a28eaf1a08372c2f3bb865836d82087559999807 --- /dev/null +++ b/checkpoint-1595/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c004aebd915715db3fac26f19f3aa0aa57fed482d012d2785763c8c5661bd279 +size 1064 diff --git a/checkpoint-1595/special_tokens_map.json b/checkpoint-1595/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-1595/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-1595/tokenizer.json b/checkpoint-1595/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-1595/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-1595/tokenizer_config.json b/checkpoint-1595/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..edd01b980c1db496ea102a51c972ee8f5d1a2c74 --- /dev/null +++ b/checkpoint-1595/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "chat_template": "{{- bos_token }}{%- if messages[0]['role'] == 'system' %}{%- set system_message = messages[0]['content']|trim %}{%- set messages = messages[1:] %}{%- else %}{%- set system_message = \"\" %}{%- endif %}{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}{{- system_message }}{{- \"<|eot_id|>\" }}{%- for message in messages %}{%- if message['role'] == 'assistant' and '' in message['content'] %}{%- set content = message['content'].split('')[-1].lstrip() %}{%- else %}{%- set content = message['content'] %}{%- endif %}{{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n' + content | trim + '<|eot_id|>' }}{%- endfor %}{%- if add_generation_prompt %}{{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}{%- endif %}", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-1595/trainer_state.json b/checkpoint-1595/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..3c48c1f5ff248851b77fbaa112a7cbceb308354e --- /dev/null +++ b/checkpoint-1595/trainer_state.json @@ -0,0 +1,11198 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 4.98643006263048, + "eval_steps": 500, + "global_step": 1595, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.003131524008350731, + "grad_norm": 13.917898178100586, + "learning_rate": 5.0000000000000004e-08, + "loss": 4.1051, + "step": 1 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 17.327869415283203, + "learning_rate": 1.0000000000000001e-07, + "loss": 4.1048, + "step": 2 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 14.063946723937988, + "learning_rate": 1.5000000000000002e-07, + "loss": 4.0741, + "step": 3 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 16.817699432373047, + "learning_rate": 2.0000000000000002e-07, + "loss": 4.2002, + "step": 4 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 14.47036361694336, + "learning_rate": 2.5000000000000004e-07, + "loss": 4.2652, + "step": 5 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 14.474193572998047, + "learning_rate": 3.0000000000000004e-07, + "loss": 4.0888, + "step": 6 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 14.865458488464355, + "learning_rate": 3.5000000000000004e-07, + "loss": 4.0014, + "step": 7 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 15.338888168334961, + "learning_rate": 4.0000000000000003e-07, + "loss": 4.13, + "step": 8 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 15.154336929321289, + "learning_rate": 4.5000000000000003e-07, + "loss": 4.2493, + "step": 9 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 15.919597625732422, + "learning_rate": 5.000000000000001e-07, + "loss": 4.0535, + "step": 10 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 14.981926918029785, + "learning_rate": 5.5e-07, + "loss": 3.9064, + "step": 11 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 13.36101245880127, + "learning_rate": 6.000000000000001e-07, + "loss": 4.1939, + "step": 12 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 15.58773422241211, + "learning_rate": 6.5e-07, + "loss": 4.18, + "step": 13 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 13.560139656066895, + "learning_rate": 7.000000000000001e-07, + "loss": 3.9414, + "step": 14 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 12.307971954345703, + "learning_rate": 7.5e-07, + "loss": 3.8836, + "step": 15 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 14.533182144165039, + "learning_rate": 8.000000000000001e-07, + "loss": 4.1551, + "step": 16 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 13.453729629516602, + "learning_rate": 8.500000000000001e-07, + "loss": 4.0048, + "step": 17 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 13.45992374420166, + "learning_rate": 9.000000000000001e-07, + "loss": 4.0745, + "step": 18 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 11.857145309448242, + "learning_rate": 9.500000000000001e-07, + "loss": 3.9871, + "step": 19 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 11.872294425964355, + "learning_rate": 1.0000000000000002e-06, + "loss": 3.8959, + "step": 20 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.969825744628906, + "learning_rate": 1.0500000000000001e-06, + "loss": 4.0308, + "step": 21 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 12.33769416809082, + "learning_rate": 1.1e-06, + "loss": 3.9341, + "step": 22 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 12.669405937194824, + "learning_rate": 1.1500000000000002e-06, + "loss": 3.8511, + "step": 23 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 10.677213668823242, + "learning_rate": 1.2000000000000002e-06, + "loss": 3.7764, + "step": 24 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 10.366402626037598, + "learning_rate": 1.25e-06, + "loss": 3.5291, + "step": 25 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 11.211421012878418, + "learning_rate": 1.3e-06, + "loss": 3.5765, + "step": 26 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 11.313716888427734, + "learning_rate": 1.3500000000000002e-06, + "loss": 3.4849, + "step": 27 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 10.41294002532959, + "learning_rate": 1.4000000000000001e-06, + "loss": 3.2653, + "step": 28 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 10.40064525604248, + "learning_rate": 1.45e-06, + "loss": 3.3384, + "step": 29 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 10.05427074432373, + "learning_rate": 1.5e-06, + "loss": 3.2257, + "step": 30 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 9.583163261413574, + "learning_rate": 1.5500000000000002e-06, + "loss": 3.1371, + "step": 31 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 10.09977912902832, + "learning_rate": 1.6000000000000001e-06, + "loss": 3.0658, + "step": 32 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 9.271486282348633, + "learning_rate": 1.6500000000000003e-06, + "loss": 2.9693, + "step": 33 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 10.687992095947266, + "learning_rate": 1.7000000000000002e-06, + "loss": 2.95, + "step": 34 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 8.762290000915527, + "learning_rate": 1.75e-06, + "loss": 2.8286, + "step": 35 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 10.13785171508789, + "learning_rate": 1.8000000000000001e-06, + "loss": 2.3664, + "step": 36 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 18.301353454589844, + "learning_rate": 1.85e-06, + "loss": 2.5533, + "step": 37 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 11.490377426147461, + "learning_rate": 1.9000000000000002e-06, + "loss": 2.6133, + "step": 38 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 15.614163398742676, + "learning_rate": 1.9500000000000004e-06, + "loss": 2.3596, + "step": 39 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 17.757442474365234, + "learning_rate": 2.0000000000000003e-06, + "loss": 2.3491, + "step": 40 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 17.18431854248047, + "learning_rate": 2.05e-06, + "loss": 2.2361, + "step": 41 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 16.149789810180664, + "learning_rate": 2.1000000000000002e-06, + "loss": 2.1457, + "step": 42 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 15.256914138793945, + "learning_rate": 2.15e-06, + "loss": 2.12, + "step": 43 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 15.537406921386719, + "learning_rate": 2.2e-06, + "loss": 2.1877, + "step": 44 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 7.947713851928711, + "learning_rate": 2.25e-06, + "loss": 2.1648, + "step": 45 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 8.818676948547363, + "learning_rate": 2.3000000000000004e-06, + "loss": 2.134, + "step": 46 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 5.175768852233887, + "learning_rate": 2.35e-06, + "loss": 2.0796, + "step": 47 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 6.750611305236816, + "learning_rate": 2.4000000000000003e-06, + "loss": 1.9174, + "step": 48 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 6.2147979736328125, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.8065, + "step": 49 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 13.291611671447754, + "learning_rate": 2.5e-06, + "loss": 1.7061, + "step": 50 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 7.251201629638672, + "learning_rate": 2.55e-06, + "loss": 1.7924, + "step": 51 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 5.2126054763793945, + "learning_rate": 2.6e-06, + "loss": 1.6735, + "step": 52 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 5.435528755187988, + "learning_rate": 2.6500000000000005e-06, + "loss": 1.6265, + "step": 53 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 4.505807399749756, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.4851, + "step": 54 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 5.128388404846191, + "learning_rate": 2.7500000000000004e-06, + "loss": 1.5832, + "step": 55 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 16.935827255249023, + "learning_rate": 2.8000000000000003e-06, + "loss": 1.6553, + "step": 56 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 3.664458990097046, + "learning_rate": 2.85e-06, + "loss": 1.5, + "step": 57 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 7.763802528381348, + "learning_rate": 2.9e-06, + "loss": 1.367, + "step": 58 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 3.2216155529022217, + "learning_rate": 2.95e-06, + "loss": 1.3863, + "step": 59 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 4.384445667266846, + "learning_rate": 3e-06, + "loss": 1.4247, + "step": 60 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 4.8080878257751465, + "learning_rate": 3.05e-06, + "loss": 1.3257, + "step": 61 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 4.154761791229248, + "learning_rate": 3.1000000000000004e-06, + "loss": 1.321, + "step": 62 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 6.4742112159729, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.2823, + "step": 63 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 2.583422899246216, + "learning_rate": 3.2000000000000003e-06, + "loss": 1.2136, + "step": 64 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 4.1933488845825195, + "learning_rate": 3.2500000000000002e-06, + "loss": 1.1855, + "step": 65 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 4.11049747467041, + "learning_rate": 3.3000000000000006e-06, + "loss": 1.2389, + "step": 66 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 2.264458417892456, + "learning_rate": 3.3500000000000005e-06, + "loss": 1.0651, + "step": 67 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 2.5408174991607666, + "learning_rate": 3.4000000000000005e-06, + "loss": 1.1389, + "step": 68 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 7.82421350479126, + "learning_rate": 3.45e-06, + "loss": 1.0956, + "step": 69 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 3.070939064025879, + "learning_rate": 3.5e-06, + "loss": 1.0451, + "step": 70 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 2.6310527324676514, + "learning_rate": 3.5500000000000003e-06, + "loss": 1.0538, + "step": 71 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 7.630155563354492, + "learning_rate": 3.6000000000000003e-06, + "loss": 1.0052, + "step": 72 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 6.950636863708496, + "learning_rate": 3.65e-06, + "loss": 1.0473, + "step": 73 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 2.2703945636749268, + "learning_rate": 3.7e-06, + "loss": 1.0576, + "step": 74 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 3.3817710876464844, + "learning_rate": 3.7500000000000005e-06, + "loss": 1.0177, + "step": 75 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 7.266414642333984, + "learning_rate": 3.8000000000000005e-06, + "loss": 1.0645, + "step": 76 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 5.782608509063721, + "learning_rate": 3.85e-06, + "loss": 1.0162, + "step": 77 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 2.7938575744628906, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9664, + "step": 78 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 6.681935787200928, + "learning_rate": 3.95e-06, + "loss": 0.953, + "step": 79 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 2.253279209136963, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9568, + "step": 80 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 1.4875826835632324, + "learning_rate": 4.05e-06, + "loss": 0.9448, + "step": 81 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 2.4987940788269043, + "learning_rate": 4.1e-06, + "loss": 0.9393, + "step": 82 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 4.712948322296143, + "learning_rate": 4.15e-06, + "loss": 0.9532, + "step": 83 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 6.9030632972717285, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.96, + "step": 84 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 3.4780967235565186, + "learning_rate": 4.25e-06, + "loss": 0.8993, + "step": 85 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.526064395904541, + "learning_rate": 4.3e-06, + "loss": 0.9021, + "step": 86 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 10.727686882019043, + "learning_rate": 4.350000000000001e-06, + "loss": 0.856, + "step": 87 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 12.483160972595215, + "learning_rate": 4.4e-06, + "loss": 0.9357, + "step": 88 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 6.544492244720459, + "learning_rate": 4.450000000000001e-06, + "loss": 0.9168, + "step": 89 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 1.178139567375183, + "learning_rate": 4.5e-06, + "loss": 0.8748, + "step": 90 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 1.711506962776184, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8425, + "step": 91 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 3.281747341156006, + "learning_rate": 4.600000000000001e-06, + "loss": 0.8491, + "step": 92 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 2.2964377403259277, + "learning_rate": 4.65e-06, + "loss": 0.8038, + "step": 93 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 1.959700345993042, + "learning_rate": 4.7e-06, + "loss": 0.8439, + "step": 94 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 3.979384183883667, + "learning_rate": 4.75e-06, + "loss": 0.8839, + "step": 95 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 1.4721262454986572, + "learning_rate": 4.800000000000001e-06, + "loss": 0.845, + "step": 96 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 2.862248659133911, + "learning_rate": 4.85e-06, + "loss": 0.7748, + "step": 97 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 3.7439088821411133, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8145, + "step": 98 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 1.6654618978500366, + "learning_rate": 4.95e-06, + "loss": 0.8326, + "step": 99 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 7.8437581062316895, + "learning_rate": 5e-06, + "loss": 0.8666, + "step": 100 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 6.429738521575928, + "learning_rate": 4.999996250830422e-06, + "loss": 0.836, + "step": 101 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 2.6017794609069824, + "learning_rate": 4.9999850033329326e-06, + "loss": 0.7785, + "step": 102 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 1.0575449466705322, + "learning_rate": 4.999966257541265e-06, + "loss": 0.7639, + "step": 103 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 2.6932010650634766, + "learning_rate": 4.999940013511647e-06, + "loss": 0.8214, + "step": 104 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 2.925288438796997, + "learning_rate": 4.999906271322792e-06, + "loss": 0.8797, + "step": 105 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.3570607900619507, + "learning_rate": 4.9998650310759035e-06, + "loss": 0.792, + "step": 106 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 5.126713752746582, + "learning_rate": 4.999816292894676e-06, + "loss": 0.8352, + "step": 107 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 1.8966432809829712, + "learning_rate": 4.99976005692529e-06, + "loss": 0.7663, + "step": 108 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 1.3100829124450684, + "learning_rate": 4.999696323336418e-06, + "loss": 0.771, + "step": 109 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 2.4025354385375977, + "learning_rate": 4.999625092319218e-06, + "loss": 0.7618, + "step": 110 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.130232810974121, + "learning_rate": 4.999546364087334e-06, + "loss": 0.7705, + "step": 111 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 3.430262327194214, + "learning_rate": 4.999460138876901e-06, + "loss": 0.77, + "step": 112 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 1.1272103786468506, + "learning_rate": 4.999366416946536e-06, + "loss": 0.7133, + "step": 113 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 1.1740471124649048, + "learning_rate": 4.999265198577342e-06, + "loss": 0.7684, + "step": 114 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.3138248920440674, + "learning_rate": 4.999156484072907e-06, + "loss": 0.7888, + "step": 115 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 1.061711311340332, + "learning_rate": 4.999040273759304e-06, + "loss": 0.7484, + "step": 116 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.4682390689849854, + "learning_rate": 4.998916567985083e-06, + "loss": 0.7296, + "step": 117 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 2.884068250656128, + "learning_rate": 4.998785367121284e-06, + "loss": 0.7662, + "step": 118 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9812761545181274, + "learning_rate": 4.9986466715614205e-06, + "loss": 0.7307, + "step": 119 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 2.2237496376037598, + "learning_rate": 4.998500481721484e-06, + "loss": 0.6761, + "step": 120 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 1.4004178047180176, + "learning_rate": 4.998346798039952e-06, + "loss": 0.7505, + "step": 121 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 5.54975700378418, + "learning_rate": 4.99818562097777e-06, + "loss": 0.7615, + "step": 122 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 6.17140531539917, + "learning_rate": 4.9980169510183624e-06, + "loss": 0.7002, + "step": 123 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 4.974380016326904, + "learning_rate": 4.997840788667628e-06, + "loss": 0.7449, + "step": 124 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 1.4133399724960327, + "learning_rate": 4.997657134453937e-06, + "loss": 0.7442, + "step": 125 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 1.868915319442749, + "learning_rate": 4.9974659889281295e-06, + "loss": 0.7104, + "step": 126 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 1.2599350214004517, + "learning_rate": 4.997267352663514e-06, + "loss": 0.7385, + "step": 127 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.4353641271591187, + "learning_rate": 4.997061226255869e-06, + "loss": 0.7081, + "step": 128 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 3.2492141723632812, + "learning_rate": 4.996847610323437e-06, + "loss": 0.7859, + "step": 129 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 9.599719047546387, + "learning_rate": 4.996626505506923e-06, + "loss": 0.7241, + "step": 130 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 10.053650856018066, + "learning_rate": 4.996397912469494e-06, + "loss": 0.6841, + "step": 131 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 1.323876976966858, + "learning_rate": 4.996161831896777e-06, + "loss": 0.7317, + "step": 132 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 1.4180598258972168, + "learning_rate": 4.9959182644968594e-06, + "loss": 0.692, + "step": 133 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 1.2194396257400513, + "learning_rate": 4.99566721100028e-06, + "loss": 0.7068, + "step": 134 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0984960794448853, + "learning_rate": 4.995408672160031e-06, + "loss": 0.6946, + "step": 135 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 1.9341071844100952, + "learning_rate": 4.995142648751561e-06, + "loss": 0.7467, + "step": 136 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 1.9960932731628418, + "learning_rate": 4.9948691415727594e-06, + "loss": 0.7379, + "step": 137 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.8743917942047119, + "learning_rate": 4.994588151443968e-06, + "loss": 0.66, + "step": 138 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8655261993408203, + "learning_rate": 4.99429967920797e-06, + "loss": 0.6646, + "step": 139 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 5.462070941925049, + "learning_rate": 4.994003725729992e-06, + "loss": 0.643, + "step": 140 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 2.1401469707489014, + "learning_rate": 4.993700291897695e-06, + "loss": 0.6639, + "step": 141 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 1.8219833374023438, + "learning_rate": 4.9933893786211815e-06, + "loss": 0.6673, + "step": 142 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 1.641079306602478, + "learning_rate": 4.993070986832984e-06, + "loss": 0.658, + "step": 143 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 1.1739819049835205, + "learning_rate": 4.992745117488066e-06, + "loss": 0.6826, + "step": 144 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 2.309185743331909, + "learning_rate": 4.9924117715638185e-06, + "loss": 0.6536, + "step": 145 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 1.09304940700531, + "learning_rate": 4.99207095006006e-06, + "loss": 0.721, + "step": 146 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.9056984186172485, + "learning_rate": 4.991722653999025e-06, + "loss": 0.7019, + "step": 147 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.8440625667572021, + "learning_rate": 4.991366884425374e-06, + "loss": 0.707, + "step": 148 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 1.2244676351547241, + "learning_rate": 4.991003642406177e-06, + "loss": 0.6407, + "step": 149 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9258589744567871, + "learning_rate": 4.99063292903092e-06, + "loss": 0.6954, + "step": 150 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 4.176390647888184, + "learning_rate": 4.990254745411496e-06, + "loss": 0.6812, + "step": 151 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 1.4322530031204224, + "learning_rate": 4.989869092682205e-06, + "loss": 0.6808, + "step": 152 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.8017717003822327, + "learning_rate": 4.989475971999748e-06, + "loss": 0.687, + "step": 153 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 1.5641374588012695, + "learning_rate": 4.989075384543228e-06, + "loss": 0.6599, + "step": 154 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 1.1522141695022583, + "learning_rate": 4.98866733151414e-06, + "loss": 0.6546, + "step": 155 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8593171238899231, + "learning_rate": 4.988251814136372e-06, + "loss": 0.6857, + "step": 156 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 2.668159246444702, + "learning_rate": 4.9878288336562e-06, + "loss": 0.661, + "step": 157 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9953671097755432, + "learning_rate": 4.987398391342285e-06, + "loss": 0.6512, + "step": 158 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.042872667312622, + "learning_rate": 4.986960488485667e-06, + "loss": 0.6311, + "step": 159 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.9070663452148438, + "learning_rate": 4.9865151263997645e-06, + "loss": 0.675, + "step": 160 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.8460433483123779, + "learning_rate": 4.986062306420367e-06, + "loss": 0.6635, + "step": 161 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 1.2639834880828857, + "learning_rate": 4.985602029905635e-06, + "loss": 0.6327, + "step": 162 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.8775074481964111, + "learning_rate": 4.985134298236091e-06, + "loss": 0.644, + "step": 163 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 1.2031961679458618, + "learning_rate": 4.98465911281462e-06, + "loss": 0.6254, + "step": 164 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.892494797706604, + "learning_rate": 4.984176475066463e-06, + "loss": 0.7122, + "step": 165 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 2.7122485637664795, + "learning_rate": 4.983686386439212e-06, + "loss": 0.6679, + "step": 166 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9344426989555359, + "learning_rate": 4.983188848402806e-06, + "loss": 0.6319, + "step": 167 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 1.4093577861785889, + "learning_rate": 4.982683862449531e-06, + "loss": 0.6425, + "step": 168 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 1.1285009384155273, + "learning_rate": 4.982171430094007e-06, + "loss": 0.6298, + "step": 169 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.952778935432434, + "learning_rate": 4.981651552873193e-06, + "loss": 0.7066, + "step": 170 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 5.133765697479248, + "learning_rate": 4.981124232346374e-06, + "loss": 0.6634, + "step": 171 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9770542979240417, + "learning_rate": 4.980589470095161e-06, + "loss": 0.7121, + "step": 172 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.8414323925971985, + "learning_rate": 4.980047267723487e-06, + "loss": 0.6397, + "step": 173 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 1.9173879623413086, + "learning_rate": 4.979497626857596e-06, + "loss": 0.6228, + "step": 174 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 1.0823363065719604, + "learning_rate": 4.978940549146048e-06, + "loss": 0.6475, + "step": 175 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 3.715353488922119, + "learning_rate": 4.978376036259706e-06, + "loss": 0.7127, + "step": 176 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.981584370136261, + "learning_rate": 4.9778040898917325e-06, + "loss": 0.6468, + "step": 177 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.70566987991333, + "learning_rate": 4.977224711757587e-06, + "loss": 0.6476, + "step": 178 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9217923283576965, + "learning_rate": 4.976637903595019e-06, + "loss": 0.6731, + "step": 179 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.8994677662849426, + "learning_rate": 4.976043667164063e-06, + "loss": 0.6562, + "step": 180 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 1.1613017320632935, + "learning_rate": 4.975442004247034e-06, + "loss": 0.6417, + "step": 181 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 1.6041977405548096, + "learning_rate": 4.974832916648521e-06, + "loss": 0.6029, + "step": 182 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.7978405952453613, + "learning_rate": 4.974216406195383e-06, + "loss": 0.6269, + "step": 183 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.6021920442581177, + "learning_rate": 4.973592474736739e-06, + "loss": 0.6149, + "step": 184 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8973568677902222, + "learning_rate": 4.972961124143971e-06, + "loss": 0.6648, + "step": 185 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 1.9432591199874878, + "learning_rate": 4.972322356310711e-06, + "loss": 0.6299, + "step": 186 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 4.457028388977051, + "learning_rate": 4.971676173152839e-06, + "loss": 0.656, + "step": 187 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 2.0989716053009033, + "learning_rate": 4.971022576608473e-06, + "loss": 0.6539, + "step": 188 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 1.0646967887878418, + "learning_rate": 4.97036156863797e-06, + "loss": 0.6727, + "step": 189 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.6522265672683716, + "learning_rate": 4.969693151223914e-06, + "loss": 0.6643, + "step": 190 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.7503505945205688, + "learning_rate": 4.969017326371115e-06, + "loss": 0.6402, + "step": 191 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 1.2341989278793335, + "learning_rate": 4.968334096106597e-06, + "loss": 0.6413, + "step": 192 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 3.089054584503174, + "learning_rate": 4.967643462479597e-06, + "loss": 0.6825, + "step": 193 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 2.711623430252075, + "learning_rate": 4.966945427561557e-06, + "loss": 0.65, + "step": 194 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 4.641184329986572, + "learning_rate": 4.966239993446118e-06, + "loss": 0.6229, + "step": 195 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.7984074354171753, + "learning_rate": 4.965527162249114e-06, + "loss": 0.6473, + "step": 196 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 1.1643115282058716, + "learning_rate": 4.964806936108566e-06, + "loss": 0.6404, + "step": 197 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 2.1877920627593994, + "learning_rate": 4.9640793171846725e-06, + "loss": 0.6185, + "step": 198 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.7970566749572754, + "learning_rate": 4.963344307659807e-06, + "loss": 0.634, + "step": 199 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 1.6014361381530762, + "learning_rate": 4.96260190973851e-06, + "loss": 0.6562, + "step": 200 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.8743320107460022, + "learning_rate": 4.961852125647482e-06, + "loss": 0.6133, + "step": 201 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 1.9526551961898804, + "learning_rate": 4.961094957635578e-06, + "loss": 0.6451, + "step": 202 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 3.6597347259521484, + "learning_rate": 4.960330407973798e-06, + "loss": 0.6386, + "step": 203 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 1.7180207967758179, + "learning_rate": 4.959558478955283e-06, + "loss": 0.6688, + "step": 204 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 0.9058470129966736, + "learning_rate": 4.958779172895308e-06, + "loss": 0.6161, + "step": 205 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 1.0031033754348755, + "learning_rate": 4.957992492131274e-06, + "loss": 0.6437, + "step": 206 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 1.5846725702285767, + "learning_rate": 4.9571984390226985e-06, + "loss": 0.6332, + "step": 207 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 1.9951609373092651, + "learning_rate": 4.956397015951215e-06, + "loss": 0.636, + "step": 208 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 1.4122583866119385, + "learning_rate": 4.95558822532056e-06, + "loss": 0.6586, + "step": 209 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.2243481874465942, + "learning_rate": 4.954772069556568e-06, + "loss": 0.6313, + "step": 210 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.8756356835365295, + "learning_rate": 4.953948551107164e-06, + "loss": 0.6406, + "step": 211 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 2.9979734420776367, + "learning_rate": 4.953117672442356e-06, + "loss": 0.5803, + "step": 212 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 2.1859359741210938, + "learning_rate": 4.952279436054229e-06, + "loss": 0.6607, + "step": 213 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.6929755806922913, + "learning_rate": 4.9514338444569346e-06, + "loss": 0.5989, + "step": 214 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 1.0361783504486084, + "learning_rate": 4.950580900186685e-06, + "loss": 0.6654, + "step": 215 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 1.210898518562317, + "learning_rate": 4.9497206058017475e-06, + "loss": 0.6213, + "step": 216 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 1.200990080833435, + "learning_rate": 4.948852963882434e-06, + "loss": 0.6654, + "step": 217 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 1.481831669807434, + "learning_rate": 4.947977977031093e-06, + "loss": 0.6474, + "step": 218 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.9883334636688232, + "learning_rate": 4.947095647872103e-06, + "loss": 0.6735, + "step": 219 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.7436536550521851, + "learning_rate": 4.946205979051868e-06, + "loss": 0.6456, + "step": 220 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9057570099830627, + "learning_rate": 4.945308973238802e-06, + "loss": 0.6228, + "step": 221 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 1.341081142425537, + "learning_rate": 4.944404633123324e-06, + "loss": 0.6417, + "step": 222 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.7958157062530518, + "learning_rate": 4.943492961417859e-06, + "loss": 0.6494, + "step": 223 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.216025471687317, + "learning_rate": 4.9425739608568106e-06, + "loss": 0.6566, + "step": 224 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.9774854779243469, + "learning_rate": 4.9416476341965735e-06, + "loss": 0.6171, + "step": 225 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 2.1562681198120117, + "learning_rate": 4.940713984215512e-06, + "loss": 0.629, + "step": 226 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 1.9521286487579346, + "learning_rate": 4.9397730137139556e-06, + "loss": 0.6475, + "step": 227 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 1.5749104022979736, + "learning_rate": 4.9388247255141895e-06, + "loss": 0.6053, + "step": 228 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 1.2008254528045654, + "learning_rate": 4.937869122460449e-06, + "loss": 0.6052, + "step": 229 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 1.0774102210998535, + "learning_rate": 4.93690620741891e-06, + "loss": 0.6099, + "step": 230 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 1.0929996967315674, + "learning_rate": 4.935935983277675e-06, + "loss": 0.6363, + "step": 231 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.8830653429031372, + "learning_rate": 4.934958452946774e-06, + "loss": 0.6136, + "step": 232 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 3.591218948364258, + "learning_rate": 4.933973619358147e-06, + "loss": 0.5962, + "step": 233 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 2.5797672271728516, + "learning_rate": 4.932981485465643e-06, + "loss": 0.6405, + "step": 234 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 1.0467664003372192, + "learning_rate": 4.9319820542450025e-06, + "loss": 0.6155, + "step": 235 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.8099795579910278, + "learning_rate": 4.930975328693856e-06, + "loss": 0.5615, + "step": 236 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.8906702995300293, + "learning_rate": 4.92996131183171e-06, + "loss": 0.6501, + "step": 237 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 1.0871416330337524, + "learning_rate": 4.928940006699944e-06, + "loss": 0.6282, + "step": 238 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 1.3209614753723145, + "learning_rate": 4.927911416361792e-06, + "loss": 0.598, + "step": 239 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.2252682447433472, + "learning_rate": 4.926875543902344e-06, + "loss": 0.6433, + "step": 240 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0569007396697998, + "learning_rate": 4.9258323924285285e-06, + "loss": 0.5927, + "step": 241 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9309014081954956, + "learning_rate": 4.924781965069106e-06, + "loss": 0.5927, + "step": 242 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0200378894805908, + "learning_rate": 4.923724264974662e-06, + "loss": 0.6064, + "step": 243 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 1.0533075332641602, + "learning_rate": 4.922659295317593e-06, + "loss": 0.6373, + "step": 244 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 0.7889382839202881, + "learning_rate": 4.921587059292102e-06, + "loss": 0.5887, + "step": 245 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.7943588495254517, + "learning_rate": 4.920507560114183e-06, + "loss": 0.593, + "step": 246 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.8247205018997192, + "learning_rate": 4.919420801021617e-06, + "loss": 0.6151, + "step": 247 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 0.9979158043861389, + "learning_rate": 4.91832678527396e-06, + "loss": 0.6019, + "step": 248 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9346868991851807, + "learning_rate": 4.917225516152532e-06, + "loss": 0.6098, + "step": 249 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.7487881183624268, + "learning_rate": 4.916116996960408e-06, + "loss": 0.5965, + "step": 250 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.821576714515686, + "learning_rate": 4.915001231022411e-06, + "loss": 0.6483, + "step": 251 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 1.0413196086883545, + "learning_rate": 4.913878221685096e-06, + "loss": 0.6108, + "step": 252 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9560331702232361, + "learning_rate": 4.912747972316745e-06, + "loss": 0.5758, + "step": 253 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.8964638113975525, + "learning_rate": 4.911610486307356e-06, + "loss": 0.6432, + "step": 254 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.8418346047401428, + "learning_rate": 4.910465767068631e-06, + "loss": 0.6027, + "step": 255 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 1.792371153831482, + "learning_rate": 4.909313818033966e-06, + "loss": 0.6198, + "step": 256 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 1.036665439605713, + "learning_rate": 4.908154642658446e-06, + "loss": 0.6255, + "step": 257 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.7592151165008545, + "learning_rate": 4.906988244418823e-06, + "loss": 0.6035, + "step": 258 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.8843073844909668, + "learning_rate": 4.90581462681352e-06, + "loss": 0.6299, + "step": 259 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9489964246749878, + "learning_rate": 4.9046337933626086e-06, + "loss": 0.5869, + "step": 260 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.851691722869873, + "learning_rate": 4.903445747607806e-06, + "loss": 0.603, + "step": 261 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 1.3722106218338013, + "learning_rate": 4.902250493112458e-06, + "loss": 0.5939, + "step": 262 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 1.1002827882766724, + "learning_rate": 4.901048033461537e-06, + "loss": 0.6452, + "step": 263 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 0.8428632020950317, + "learning_rate": 4.89983837226162e-06, + "loss": 0.5956, + "step": 264 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.7666584849357605, + "learning_rate": 4.898621513140889e-06, + "loss": 0.6067, + "step": 265 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.8413611054420471, + "learning_rate": 4.897397459749113e-06, + "loss": 0.5985, + "step": 266 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 2.3374335765838623, + "learning_rate": 4.896166215757638e-06, + "loss": 0.5885, + "step": 267 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 2.236640214920044, + "learning_rate": 4.894927784859377e-06, + "loss": 0.6408, + "step": 268 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.9715856313705444, + "learning_rate": 4.893682170768802e-06, + "loss": 0.5954, + "step": 269 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0249912738800049, + "learning_rate": 4.892429377221928e-06, + "loss": 0.6186, + "step": 270 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 1.255426049232483, + "learning_rate": 4.891169407976302e-06, + "loss": 0.6351, + "step": 271 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.9339559674263, + "learning_rate": 4.889902266810995e-06, + "loss": 0.5944, + "step": 272 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 1.2473429441452026, + "learning_rate": 4.888627957526589e-06, + "loss": 0.544, + "step": 273 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 1.0589442253112793, + "learning_rate": 4.887346483945166e-06, + "loss": 0.5543, + "step": 274 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9844024777412415, + "learning_rate": 4.886057849910294e-06, + "loss": 0.5941, + "step": 275 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 2.88578200340271, + "learning_rate": 4.8847620592870196e-06, + "loss": 0.6124, + "step": 276 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 0.7496054172515869, + "learning_rate": 4.8834591159618524e-06, + "loss": 0.6006, + "step": 277 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 0.7403052449226379, + "learning_rate": 4.88214902384276e-06, + "loss": 0.5911, + "step": 278 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9003771543502808, + "learning_rate": 4.880831786859146e-06, + "loss": 0.6347, + "step": 279 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0345501899719238, + "learning_rate": 4.879507408961847e-06, + "loss": 0.6111, + "step": 280 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.4385879039764404, + "learning_rate": 4.878175894123116e-06, + "loss": 0.6454, + "step": 281 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.8469482064247131, + "learning_rate": 4.8768372463366145e-06, + "loss": 0.6163, + "step": 282 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.8859589695930481, + "learning_rate": 4.875491469617395e-06, + "loss": 0.6144, + "step": 283 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 1.8436834812164307, + "learning_rate": 4.874138568001895e-06, + "loss": 0.6275, + "step": 284 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.6646101474761963, + "learning_rate": 4.87277854554792e-06, + "loss": 0.615, + "step": 285 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 1.0070925951004028, + "learning_rate": 4.871411406334633e-06, + "loss": 0.5898, + "step": 286 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9785194993019104, + "learning_rate": 4.870037154462545e-06, + "loss": 0.5992, + "step": 287 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.7244889736175537, + "learning_rate": 4.868655794053497e-06, + "loss": 0.6078, + "step": 288 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.4496444463729858, + "learning_rate": 4.8672673292506535e-06, + "loss": 0.5855, + "step": 289 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 1.8514957427978516, + "learning_rate": 4.865871764218486e-06, + "loss": 0.5707, + "step": 290 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.8439773321151733, + "learning_rate": 4.864469103142763e-06, + "loss": 0.5562, + "step": 291 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.8146086931228638, + "learning_rate": 4.8630593502305355e-06, + "loss": 0.6161, + "step": 292 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.8920315504074097, + "learning_rate": 4.861642509710126e-06, + "loss": 0.6139, + "step": 293 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.4980088472366333, + "learning_rate": 4.860218585831116e-06, + "loss": 0.6187, + "step": 294 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 0.9910127520561218, + "learning_rate": 4.8587875828643285e-06, + "loss": 0.5852, + "step": 295 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.819600522518158, + "learning_rate": 4.857349505101823e-06, + "loss": 0.6172, + "step": 296 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 1.1059772968292236, + "learning_rate": 4.855904356856878e-06, + "loss": 0.5868, + "step": 297 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 1.2362196445465088, + "learning_rate": 4.854452142463977e-06, + "loss": 0.625, + "step": 298 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.9956470727920532, + "learning_rate": 4.852992866278799e-06, + "loss": 0.5923, + "step": 299 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 0.864109218120575, + "learning_rate": 4.851526532678203e-06, + "loss": 0.6315, + "step": 300 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 0.8900614380836487, + "learning_rate": 4.850053146060217e-06, + "loss": 0.6128, + "step": 301 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.927254855632782, + "learning_rate": 4.84857271084402e-06, + "loss": 0.5955, + "step": 302 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0046517848968506, + "learning_rate": 4.847085231469935e-06, + "loss": 0.6134, + "step": 303 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.734597384929657, + "learning_rate": 4.8455907123994125e-06, + "loss": 0.5927, + "step": 304 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.7338348031044006, + "learning_rate": 4.844089158115016e-06, + "loss": 0.5897, + "step": 305 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9163988828659058, + "learning_rate": 4.8425805731204106e-06, + "loss": 0.6051, + "step": 306 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 1.050246238708496, + "learning_rate": 4.84106496194035e-06, + "loss": 0.5751, + "step": 307 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 0.7637603878974915, + "learning_rate": 4.83954232912066e-06, + "loss": 0.5677, + "step": 308 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.7110525965690613, + "learning_rate": 4.838012679228229e-06, + "loss": 0.6051, + "step": 309 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.7662068605422974, + "learning_rate": 4.836476016850988e-06, + "loss": 0.59, + "step": 310 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 0.8907375335693359, + "learning_rate": 4.834932346597906e-06, + "loss": 0.5792, + "step": 311 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.8939849138259888, + "learning_rate": 4.833381673098966e-06, + "loss": 0.6062, + "step": 312 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 0.8878788948059082, + "learning_rate": 4.8318240010051595e-06, + "loss": 0.5694, + "step": 313 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 1.2523870468139648, + "learning_rate": 4.830259334988468e-06, + "loss": 0.5809, + "step": 314 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 1.0836797952651978, + "learning_rate": 4.82868767974185e-06, + "loss": 0.5949, + "step": 315 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.7985473871231079, + "learning_rate": 4.827109039979226e-06, + "loss": 0.6057, + "step": 316 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 1.042951226234436, + "learning_rate": 4.825523420435469e-06, + "loss": 0.6004, + "step": 317 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.7845115661621094, + "learning_rate": 4.823930825866381e-06, + "loss": 0.6161, + "step": 318 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.931854784488678, + "learning_rate": 4.82233126104869e-06, + "loss": 0.5912, + "step": 319 + }, + { + "epoch": 1.0, + "grad_norm": 0.931854784488678, + "learning_rate": 4.8207247307800275e-06, + "loss": 0.1914, + "step": 320 + }, + { + "epoch": 1.0031315240083507, + "grad_norm": 0.751028835773468, + "learning_rate": 4.819111239878916e-06, + "loss": 0.5802, + "step": 321 + }, + { + "epoch": 1.0062630480167014, + "grad_norm": 1.4943569898605347, + "learning_rate": 4.817490793184758e-06, + "loss": 0.613, + "step": 322 + }, + { + "epoch": 1.0093945720250521, + "grad_norm": 2.296318531036377, + "learning_rate": 4.815863395557816e-06, + "loss": 0.5453, + "step": 323 + }, + { + "epoch": 1.0125260960334028, + "grad_norm": 0.760101318359375, + "learning_rate": 4.814229051879202e-06, + "loss": 0.5302, + "step": 324 + }, + { + "epoch": 1.0156576200417538, + "grad_norm": 0.8145846128463745, + "learning_rate": 4.812587767050861e-06, + "loss": 0.5831, + "step": 325 + }, + { + "epoch": 1.0187891440501045, + "grad_norm": 0.9169796109199524, + "learning_rate": 4.8109395459955565e-06, + "loss": 0.5756, + "step": 326 + }, + { + "epoch": 1.0219206680584552, + "grad_norm": 0.8791524171829224, + "learning_rate": 4.809284393656858e-06, + "loss": 0.5988, + "step": 327 + }, + { + "epoch": 1.0250521920668059, + "grad_norm": 1.0184170007705688, + "learning_rate": 4.807622314999122e-06, + "loss": 0.5476, + "step": 328 + }, + { + "epoch": 1.0281837160751566, + "grad_norm": 0.8095184564590454, + "learning_rate": 4.8059533150074805e-06, + "loss": 0.5723, + "step": 329 + }, + { + "epoch": 1.0313152400835073, + "grad_norm": 0.7621930241584778, + "learning_rate": 4.804277398687826e-06, + "loss": 0.5841, + "step": 330 + }, + { + "epoch": 1.034446764091858, + "grad_norm": 3.729628324508667, + "learning_rate": 4.802594571066791e-06, + "loss": 0.5639, + "step": 331 + }, + { + "epoch": 1.0375782881002087, + "grad_norm": 1.6502974033355713, + "learning_rate": 4.800904837191743e-06, + "loss": 0.6024, + "step": 332 + }, + { + "epoch": 1.0407098121085594, + "grad_norm": 0.8031198978424072, + "learning_rate": 4.799208202130762e-06, + "loss": 0.5305, + "step": 333 + }, + { + "epoch": 1.0438413361169103, + "grad_norm": 0.939644992351532, + "learning_rate": 4.797504670972623e-06, + "loss": 0.5446, + "step": 334 + }, + { + "epoch": 1.046972860125261, + "grad_norm": 1.0589954853057861, + "learning_rate": 4.795794248826789e-06, + "loss": 0.5366, + "step": 335 + }, + { + "epoch": 1.0501043841336117, + "grad_norm": 0.9089614748954773, + "learning_rate": 4.794076940823391e-06, + "loss": 0.5795, + "step": 336 + }, + { + "epoch": 1.0532359081419624, + "grad_norm": 0.7732561230659485, + "learning_rate": 4.792352752113212e-06, + "loss": 0.5765, + "step": 337 + }, + { + "epoch": 1.0563674321503131, + "grad_norm": 1.811553955078125, + "learning_rate": 4.790621687867672e-06, + "loss": 0.561, + "step": 338 + }, + { + "epoch": 1.0594989561586639, + "grad_norm": 1.1930758953094482, + "learning_rate": 4.788883753278813e-06, + "loss": 0.5, + "step": 339 + }, + { + "epoch": 1.0626304801670146, + "grad_norm": 0.9551813006401062, + "learning_rate": 4.787138953559285e-06, + "loss": 0.5228, + "step": 340 + }, + { + "epoch": 1.0657620041753653, + "grad_norm": 0.9609586596488953, + "learning_rate": 4.785387293942329e-06, + "loss": 0.5827, + "step": 341 + }, + { + "epoch": 1.068893528183716, + "grad_norm": 0.8403449654579163, + "learning_rate": 4.783628779681759e-06, + "loss": 0.5585, + "step": 342 + }, + { + "epoch": 1.072025052192067, + "grad_norm": 0.9108251929283142, + "learning_rate": 4.7818634160519496e-06, + "loss": 0.6077, + "step": 343 + }, + { + "epoch": 1.0751565762004176, + "grad_norm": 0.9476898908615112, + "learning_rate": 4.780091208347819e-06, + "loss": 0.5493, + "step": 344 + }, + { + "epoch": 1.0782881002087683, + "grad_norm": 1.1943707466125488, + "learning_rate": 4.778312161884813e-06, + "loss": 0.5736, + "step": 345 + }, + { + "epoch": 1.081419624217119, + "grad_norm": 3.1342639923095703, + "learning_rate": 4.77652628199889e-06, + "loss": 0.5765, + "step": 346 + }, + { + "epoch": 1.0845511482254697, + "grad_norm": 2.7982125282287598, + "learning_rate": 4.7747335740465015e-06, + "loss": 0.6003, + "step": 347 + }, + { + "epoch": 1.0876826722338204, + "grad_norm": 1.5068914890289307, + "learning_rate": 4.7729340434045815e-06, + "loss": 0.5033, + "step": 348 + }, + { + "epoch": 1.0908141962421711, + "grad_norm": 0.8273429274559021, + "learning_rate": 4.771127695470527e-06, + "loss": 0.5309, + "step": 349 + }, + { + "epoch": 1.0939457202505218, + "grad_norm": 1.104974389076233, + "learning_rate": 4.76931453566218e-06, + "loss": 0.5244, + "step": 350 + }, + { + "epoch": 1.0970772442588728, + "grad_norm": 1.096509337425232, + "learning_rate": 4.7674945694178166e-06, + "loss": 0.5585, + "step": 351 + }, + { + "epoch": 1.1002087682672235, + "grad_norm": 1.0238200426101685, + "learning_rate": 4.765667802196127e-06, + "loss": 0.5589, + "step": 352 + }, + { + "epoch": 1.1033402922755742, + "grad_norm": 0.7515526413917542, + "learning_rate": 4.763834239476197e-06, + "loss": 0.5304, + "step": 353 + }, + { + "epoch": 1.1064718162839249, + "grad_norm": 1.0282566547393799, + "learning_rate": 4.761993886757499e-06, + "loss": 0.5476, + "step": 354 + }, + { + "epoch": 1.1096033402922756, + "grad_norm": 0.9962708950042725, + "learning_rate": 4.760146749559868e-06, + "loss": 0.5117, + "step": 355 + }, + { + "epoch": 1.1127348643006263, + "grad_norm": 0.7851671576499939, + "learning_rate": 4.758292833423488e-06, + "loss": 0.5542, + "step": 356 + }, + { + "epoch": 1.115866388308977, + "grad_norm": 0.8857759237289429, + "learning_rate": 4.756432143908876e-06, + "loss": 0.544, + "step": 357 + }, + { + "epoch": 1.1189979123173277, + "grad_norm": 0.9402740597724915, + "learning_rate": 4.7545646865968645e-06, + "loss": 0.5656, + "step": 358 + }, + { + "epoch": 1.1221294363256784, + "grad_norm": 0.8210407495498657, + "learning_rate": 4.752690467088584e-06, + "loss": 0.5733, + "step": 359 + }, + { + "epoch": 1.1252609603340291, + "grad_norm": 0.795684278011322, + "learning_rate": 4.750809491005449e-06, + "loss": 0.5678, + "step": 360 + }, + { + "epoch": 1.12839248434238, + "grad_norm": 0.8712463974952698, + "learning_rate": 4.748921763989139e-06, + "loss": 0.5777, + "step": 361 + }, + { + "epoch": 1.1315240083507307, + "grad_norm": 0.9810119867324829, + "learning_rate": 4.747027291701578e-06, + "loss": 0.5511, + "step": 362 + }, + { + "epoch": 1.1346555323590815, + "grad_norm": 0.81117844581604, + "learning_rate": 4.745126079824926e-06, + "loss": 0.5038, + "step": 363 + }, + { + "epoch": 1.1377870563674322, + "grad_norm": 0.7631494402885437, + "learning_rate": 4.743218134061556e-06, + "loss": 0.6272, + "step": 364 + }, + { + "epoch": 1.1409185803757829, + "grad_norm": 0.7601696252822876, + "learning_rate": 4.741303460134038e-06, + "loss": 0.571, + "step": 365 + }, + { + "epoch": 1.1440501043841336, + "grad_norm": 1.7977744340896606, + "learning_rate": 4.7393820637851205e-06, + "loss": 0.538, + "step": 366 + }, + { + "epoch": 1.1471816283924843, + "grad_norm": 2.022578001022339, + "learning_rate": 4.737453950777718e-06, + "loss": 0.5822, + "step": 367 + }, + { + "epoch": 1.150313152400835, + "grad_norm": 0.7586764693260193, + "learning_rate": 4.735519126894885e-06, + "loss": 0.5986, + "step": 368 + }, + { + "epoch": 1.153444676409186, + "grad_norm": 0.8970286846160889, + "learning_rate": 4.733577597939812e-06, + "loss": 0.542, + "step": 369 + }, + { + "epoch": 1.1565762004175366, + "grad_norm": 0.8546352982521057, + "learning_rate": 4.731629369735793e-06, + "loss": 0.5832, + "step": 370 + }, + { + "epoch": 1.1597077244258873, + "grad_norm": 0.9266164898872375, + "learning_rate": 4.72967444812622e-06, + "loss": 0.551, + "step": 371 + }, + { + "epoch": 1.162839248434238, + "grad_norm": 1.0413658618927002, + "learning_rate": 4.7277128389745595e-06, + "loss": 0.5866, + "step": 372 + }, + { + "epoch": 1.1659707724425887, + "grad_norm": 0.9312199950218201, + "learning_rate": 4.7257445481643334e-06, + "loss": 0.5723, + "step": 373 + }, + { + "epoch": 1.1691022964509394, + "grad_norm": 0.7389806509017944, + "learning_rate": 4.723769581599109e-06, + "loss": 0.5209, + "step": 374 + }, + { + "epoch": 1.1722338204592901, + "grad_norm": 3.053169012069702, + "learning_rate": 4.721787945202472e-06, + "loss": 0.6094, + "step": 375 + }, + { + "epoch": 1.1753653444676408, + "grad_norm": 1.288589596748352, + "learning_rate": 4.719799644918017e-06, + "loss": 0.5616, + "step": 376 + }, + { + "epoch": 1.1784968684759916, + "grad_norm": 0.7675042152404785, + "learning_rate": 4.717804686709323e-06, + "loss": 0.4963, + "step": 377 + }, + { + "epoch": 1.1816283924843423, + "grad_norm": 0.7246491312980652, + "learning_rate": 4.715803076559938e-06, + "loss": 0.5273, + "step": 378 + }, + { + "epoch": 1.1847599164926932, + "grad_norm": 0.8193361759185791, + "learning_rate": 4.713794820473366e-06, + "loss": 0.6107, + "step": 379 + }, + { + "epoch": 1.187891440501044, + "grad_norm": 0.9498510360717773, + "learning_rate": 4.711779924473037e-06, + "loss": 0.5421, + "step": 380 + }, + { + "epoch": 1.1910229645093946, + "grad_norm": 1.0479756593704224, + "learning_rate": 4.709758394602305e-06, + "loss": 0.5257, + "step": 381 + }, + { + "epoch": 1.1941544885177453, + "grad_norm": 0.907866895198822, + "learning_rate": 4.707730236924413e-06, + "loss": 0.5289, + "step": 382 + }, + { + "epoch": 1.197286012526096, + "grad_norm": 0.8861165642738342, + "learning_rate": 4.705695457522488e-06, + "loss": 0.5727, + "step": 383 + }, + { + "epoch": 1.2004175365344467, + "grad_norm": 0.7467761039733887, + "learning_rate": 4.703654062499516e-06, + "loss": 0.5602, + "step": 384 + }, + { + "epoch": 1.2035490605427974, + "grad_norm": 0.7456198334693909, + "learning_rate": 4.701606057978325e-06, + "loss": 0.5345, + "step": 385 + }, + { + "epoch": 1.2066805845511483, + "grad_norm": 1.9976060390472412, + "learning_rate": 4.699551450101571e-06, + "loss": 0.5504, + "step": 386 + }, + { + "epoch": 1.209812108559499, + "grad_norm": 1.5253807306289673, + "learning_rate": 4.697490245031709e-06, + "loss": 0.5568, + "step": 387 + }, + { + "epoch": 1.2129436325678498, + "grad_norm": 1.0786075592041016, + "learning_rate": 4.6954224489509885e-06, + "loss": 0.5564, + "step": 388 + }, + { + "epoch": 1.2160751565762005, + "grad_norm": 0.8385995030403137, + "learning_rate": 4.693348068061422e-06, + "loss": 0.5341, + "step": 389 + }, + { + "epoch": 1.2192066805845512, + "grad_norm": 0.8184949159622192, + "learning_rate": 4.691267108584774e-06, + "loss": 0.5614, + "step": 390 + }, + { + "epoch": 1.2223382045929019, + "grad_norm": 0.9964898824691772, + "learning_rate": 4.68917957676254e-06, + "loss": 0.5589, + "step": 391 + }, + { + "epoch": 1.2254697286012526, + "grad_norm": 1.0168914794921875, + "learning_rate": 4.687085478855931e-06, + "loss": 0.5892, + "step": 392 + }, + { + "epoch": 1.2286012526096033, + "grad_norm": 0.8841140866279602, + "learning_rate": 4.684984821145846e-06, + "loss": 0.5327, + "step": 393 + }, + { + "epoch": 1.231732776617954, + "grad_norm": 0.834431529045105, + "learning_rate": 4.682877609932866e-06, + "loss": 0.5594, + "step": 394 + }, + { + "epoch": 1.2348643006263047, + "grad_norm": 0.7256641983985901, + "learning_rate": 4.6807638515372234e-06, + "loss": 0.5443, + "step": 395 + }, + { + "epoch": 1.2379958246346556, + "grad_norm": 0.765096127986908, + "learning_rate": 4.678643552298788e-06, + "loss": 0.5439, + "step": 396 + }, + { + "epoch": 1.2411273486430063, + "grad_norm": 0.8760455846786499, + "learning_rate": 4.676516718577051e-06, + "loss": 0.5485, + "step": 397 + }, + { + "epoch": 1.244258872651357, + "grad_norm": 2.7111501693725586, + "learning_rate": 4.674383356751099e-06, + "loss": 0.5696, + "step": 398 + }, + { + "epoch": 1.2473903966597077, + "grad_norm": 1.0521738529205322, + "learning_rate": 4.672243473219601e-06, + "loss": 0.5503, + "step": 399 + }, + { + "epoch": 1.2505219206680585, + "grad_norm": 0.8909669518470764, + "learning_rate": 4.670097074400785e-06, + "loss": 0.5183, + "step": 400 + }, + { + "epoch": 1.2536534446764092, + "grad_norm": 0.7483847737312317, + "learning_rate": 4.667944166732424e-06, + "loss": 0.5669, + "step": 401 + }, + { + "epoch": 1.2567849686847599, + "grad_norm": 1.146997094154358, + "learning_rate": 4.665784756671808e-06, + "loss": 0.5464, + "step": 402 + }, + { + "epoch": 1.2599164926931106, + "grad_norm": 0.8998096585273743, + "learning_rate": 4.663618850695733e-06, + "loss": 0.5502, + "step": 403 + }, + { + "epoch": 1.2630480167014615, + "grad_norm": 0.8882688283920288, + "learning_rate": 4.6614464553004795e-06, + "loss": 0.5507, + "step": 404 + }, + { + "epoch": 1.2661795407098122, + "grad_norm": 0.8310684561729431, + "learning_rate": 4.659267577001789e-06, + "loss": 0.5164, + "step": 405 + }, + { + "epoch": 1.269311064718163, + "grad_norm": 0.9286114573478699, + "learning_rate": 4.657082222334851e-06, + "loss": 0.4813, + "step": 406 + }, + { + "epoch": 1.2724425887265136, + "grad_norm": 1.2394906282424927, + "learning_rate": 4.654890397854275e-06, + "loss": 0.5837, + "step": 407 + }, + { + "epoch": 1.2755741127348643, + "grad_norm": 4.00585412979126, + "learning_rate": 4.652692110134079e-06, + "loss": 0.5453, + "step": 408 + }, + { + "epoch": 1.278705636743215, + "grad_norm": 1.1667803525924683, + "learning_rate": 4.650487365767667e-06, + "loss": 0.5652, + "step": 409 + }, + { + "epoch": 1.2818371607515657, + "grad_norm": 0.9351289868354797, + "learning_rate": 4.648276171367807e-06, + "loss": 0.5576, + "step": 410 + }, + { + "epoch": 1.2849686847599164, + "grad_norm": 0.8107728958129883, + "learning_rate": 4.646058533566614e-06, + "loss": 0.5821, + "step": 411 + }, + { + "epoch": 1.2881002087682671, + "grad_norm": 0.7293011546134949, + "learning_rate": 4.643834459015525e-06, + "loss": 0.5363, + "step": 412 + }, + { + "epoch": 1.2912317327766178, + "grad_norm": 0.7550690770149231, + "learning_rate": 4.641603954385289e-06, + "loss": 0.53, + "step": 413 + }, + { + "epoch": 1.2943632567849686, + "grad_norm": 0.7626177072525024, + "learning_rate": 4.639367026365938e-06, + "loss": 0.5307, + "step": 414 + }, + { + "epoch": 1.2974947807933195, + "grad_norm": 1.0841096639633179, + "learning_rate": 4.637123681666769e-06, + "loss": 0.5162, + "step": 415 + }, + { + "epoch": 1.3006263048016702, + "grad_norm": 0.8814271092414856, + "learning_rate": 4.634873927016326e-06, + "loss": 0.5369, + "step": 416 + }, + { + "epoch": 1.303757828810021, + "grad_norm": 0.7402971982955933, + "learning_rate": 4.632617769162378e-06, + "loss": 0.5846, + "step": 417 + }, + { + "epoch": 1.3068893528183716, + "grad_norm": 0.8106061220169067, + "learning_rate": 4.6303552148719e-06, + "loss": 0.5289, + "step": 418 + }, + { + "epoch": 1.3100208768267223, + "grad_norm": 0.9241361618041992, + "learning_rate": 4.628086270931053e-06, + "loss": 0.5714, + "step": 419 + }, + { + "epoch": 1.313152400835073, + "grad_norm": 0.950332522392273, + "learning_rate": 4.625810944145159e-06, + "loss": 0.5817, + "step": 420 + }, + { + "epoch": 1.316283924843424, + "grad_norm": 0.9037718772888184, + "learning_rate": 4.623529241338689e-06, + "loss": 0.5902, + "step": 421 + }, + { + "epoch": 1.3194154488517746, + "grad_norm": 1.2110658884048462, + "learning_rate": 4.621241169355234e-06, + "loss": 0.561, + "step": 422 + }, + { + "epoch": 1.3225469728601253, + "grad_norm": 0.8582742214202881, + "learning_rate": 4.618946735057491e-06, + "loss": 0.5003, + "step": 423 + }, + { + "epoch": 1.325678496868476, + "grad_norm": 0.9203405976295471, + "learning_rate": 4.6166459453272386e-06, + "loss": 0.5639, + "step": 424 + }, + { + "epoch": 1.3288100208768268, + "grad_norm": 0.933721125125885, + "learning_rate": 4.614338807065317e-06, + "loss": 0.5766, + "step": 425 + }, + { + "epoch": 1.3319415448851775, + "grad_norm": 0.8435131311416626, + "learning_rate": 4.612025327191608e-06, + "loss": 0.5656, + "step": 426 + }, + { + "epoch": 1.3350730688935282, + "grad_norm": 0.795796811580658, + "learning_rate": 4.609705512645015e-06, + "loss": 0.4996, + "step": 427 + }, + { + "epoch": 1.3382045929018789, + "grad_norm": 0.8168228268623352, + "learning_rate": 4.6073793703834404e-06, + "loss": 0.5465, + "step": 428 + }, + { + "epoch": 1.3413361169102296, + "grad_norm": 0.8795569539070129, + "learning_rate": 4.605046907383765e-06, + "loss": 0.5407, + "step": 429 + }, + { + "epoch": 1.3444676409185803, + "grad_norm": 0.8504094481468201, + "learning_rate": 4.6027081306418295e-06, + "loss": 0.5589, + "step": 430 + }, + { + "epoch": 1.347599164926931, + "grad_norm": 1.485202431678772, + "learning_rate": 4.600363047172409e-06, + "loss": 0.5515, + "step": 431 + }, + { + "epoch": 1.350730688935282, + "grad_norm": 1.1156851053237915, + "learning_rate": 4.598011664009197e-06, + "loss": 0.5681, + "step": 432 + }, + { + "epoch": 1.3538622129436326, + "grad_norm": 0.8666876554489136, + "learning_rate": 4.595653988204779e-06, + "loss": 0.5451, + "step": 433 + }, + { + "epoch": 1.3569937369519833, + "grad_norm": 0.8192381858825684, + "learning_rate": 4.593290026830619e-06, + "loss": 0.5632, + "step": 434 + }, + { + "epoch": 1.360125260960334, + "grad_norm": 0.7994804978370667, + "learning_rate": 4.590919786977029e-06, + "loss": 0.5181, + "step": 435 + }, + { + "epoch": 1.3632567849686847, + "grad_norm": 0.8038607835769653, + "learning_rate": 4.5885432757531535e-06, + "loss": 0.5385, + "step": 436 + }, + { + "epoch": 1.3663883089770354, + "grad_norm": 0.7677503824234009, + "learning_rate": 4.586160500286948e-06, + "loss": 0.5455, + "step": 437 + }, + { + "epoch": 1.3695198329853862, + "grad_norm": 0.8293285369873047, + "learning_rate": 4.583771467725157e-06, + "loss": 0.5401, + "step": 438 + }, + { + "epoch": 1.372651356993737, + "grad_norm": 0.8607680797576904, + "learning_rate": 4.581376185233289e-06, + "loss": 0.5782, + "step": 439 + }, + { + "epoch": 1.3757828810020878, + "grad_norm": 0.8847081065177917, + "learning_rate": 4.578974659995601e-06, + "loss": 0.572, + "step": 440 + }, + { + "epoch": 1.3789144050104385, + "grad_norm": 0.7669641971588135, + "learning_rate": 4.576566899215075e-06, + "loss": 0.5655, + "step": 441 + }, + { + "epoch": 1.3820459290187892, + "grad_norm": 0.8514629006385803, + "learning_rate": 4.5741529101133904e-06, + "loss": 0.5218, + "step": 442 + }, + { + "epoch": 1.38517745302714, + "grad_norm": 0.8719842433929443, + "learning_rate": 4.5717326999309145e-06, + "loss": 0.5579, + "step": 443 + }, + { + "epoch": 1.3883089770354906, + "grad_norm": 1.1142809391021729, + "learning_rate": 4.569306275926667e-06, + "loss": 0.5535, + "step": 444 + }, + { + "epoch": 1.3914405010438413, + "grad_norm": 0.7392387986183167, + "learning_rate": 4.566873645378309e-06, + "loss": 0.5335, + "step": 445 + }, + { + "epoch": 1.394572025052192, + "grad_norm": 0.9066658616065979, + "learning_rate": 4.564434815582117e-06, + "loss": 0.5286, + "step": 446 + }, + { + "epoch": 1.3977035490605427, + "grad_norm": 0.8648932576179504, + "learning_rate": 4.561989793852959e-06, + "loss": 0.5008, + "step": 447 + }, + { + "epoch": 1.4008350730688934, + "grad_norm": 0.7768712043762207, + "learning_rate": 4.559538587524276e-06, + "loss": 0.5727, + "step": 448 + }, + { + "epoch": 1.4039665970772441, + "grad_norm": 0.7851182222366333, + "learning_rate": 4.557081203948059e-06, + "loss": 0.5731, + "step": 449 + }, + { + "epoch": 1.407098121085595, + "grad_norm": 0.8959861397743225, + "learning_rate": 4.5546176504948255e-06, + "loss": 0.5587, + "step": 450 + }, + { + "epoch": 1.4102296450939458, + "grad_norm": 1.0538026094436646, + "learning_rate": 4.552147934553601e-06, + "loss": 0.5808, + "step": 451 + }, + { + "epoch": 1.4133611691022965, + "grad_norm": 0.9887629151344299, + "learning_rate": 4.54967206353189e-06, + "loss": 0.5658, + "step": 452 + }, + { + "epoch": 1.4164926931106472, + "grad_norm": 0.9579302072525024, + "learning_rate": 4.547190044855663e-06, + "loss": 0.5092, + "step": 453 + }, + { + "epoch": 1.4196242171189979, + "grad_norm": 0.6993522047996521, + "learning_rate": 4.544701885969326e-06, + "loss": 0.5233, + "step": 454 + }, + { + "epoch": 1.4227557411273486, + "grad_norm": 0.8197568655014038, + "learning_rate": 4.542207594335703e-06, + "loss": 0.553, + "step": 455 + }, + { + "epoch": 1.4258872651356993, + "grad_norm": 2.921947717666626, + "learning_rate": 4.53970717743601e-06, + "loss": 0.4857, + "step": 456 + }, + { + "epoch": 1.4290187891440502, + "grad_norm": 1.3547242879867554, + "learning_rate": 4.53720064276984e-06, + "loss": 0.5676, + "step": 457 + }, + { + "epoch": 1.432150313152401, + "grad_norm": 1.4175567626953125, + "learning_rate": 4.534687997855131e-06, + "loss": 0.5164, + "step": 458 + }, + { + "epoch": 1.4352818371607516, + "grad_norm": 1.378146767616272, + "learning_rate": 4.532169250228145e-06, + "loss": 0.5429, + "step": 459 + }, + { + "epoch": 1.4384133611691023, + "grad_norm": 0.7811698317527771, + "learning_rate": 4.529644407443456e-06, + "loss": 0.524, + "step": 460 + }, + { + "epoch": 1.441544885177453, + "grad_norm": 1.1481678485870361, + "learning_rate": 4.527113477073914e-06, + "loss": 0.5513, + "step": 461 + }, + { + "epoch": 1.4446764091858038, + "grad_norm": 0.8450161218643188, + "learning_rate": 4.5245764667106266e-06, + "loss": 0.5632, + "step": 462 + }, + { + "epoch": 1.4478079331941545, + "grad_norm": 1.1582145690917969, + "learning_rate": 4.522033383962941e-06, + "loss": 0.5834, + "step": 463 + }, + { + "epoch": 1.4509394572025052, + "grad_norm": 1.0403447151184082, + "learning_rate": 4.519484236458416e-06, + "loss": 0.506, + "step": 464 + }, + { + "epoch": 1.4540709812108559, + "grad_norm": 0.7894920706748962, + "learning_rate": 4.516929031842799e-06, + "loss": 0.5526, + "step": 465 + }, + { + "epoch": 1.4572025052192066, + "grad_norm": 0.8092262744903564, + "learning_rate": 4.51436777778001e-06, + "loss": 0.5619, + "step": 466 + }, + { + "epoch": 1.4603340292275573, + "grad_norm": 0.9773806929588318, + "learning_rate": 4.511800481952106e-06, + "loss": 0.5179, + "step": 467 + }, + { + "epoch": 1.4634655532359082, + "grad_norm": 1.018676519393921, + "learning_rate": 4.509227152059271e-06, + "loss": 0.5415, + "step": 468 + }, + { + "epoch": 1.466597077244259, + "grad_norm": 0.7457838654518127, + "learning_rate": 4.506647795819784e-06, + "loss": 0.5473, + "step": 469 + }, + { + "epoch": 1.4697286012526096, + "grad_norm": 0.7826436161994934, + "learning_rate": 4.50406242097e-06, + "loss": 0.5526, + "step": 470 + }, + { + "epoch": 1.4728601252609603, + "grad_norm": 0.9492483139038086, + "learning_rate": 4.501471035264328e-06, + "loss": 0.5179, + "step": 471 + }, + { + "epoch": 1.475991649269311, + "grad_norm": 0.93398517370224, + "learning_rate": 4.4988736464752005e-06, + "loss": 0.5195, + "step": 472 + }, + { + "epoch": 1.4791231732776617, + "grad_norm": 0.8396487832069397, + "learning_rate": 4.496270262393061e-06, + "loss": 0.5447, + "step": 473 + }, + { + "epoch": 1.4822546972860124, + "grad_norm": 0.7450584173202515, + "learning_rate": 4.4936608908263315e-06, + "loss": 0.5207, + "step": 474 + }, + { + "epoch": 1.4853862212943634, + "grad_norm": 0.7887717485427856, + "learning_rate": 4.491045539601392e-06, + "loss": 0.523, + "step": 475 + }, + { + "epoch": 1.488517745302714, + "grad_norm": 1.2051388025283813, + "learning_rate": 4.48842421656256e-06, + "loss": 0.5402, + "step": 476 + }, + { + "epoch": 1.4916492693110648, + "grad_norm": 2.3103389739990234, + "learning_rate": 4.485796929572063e-06, + "loss": 0.5588, + "step": 477 + }, + { + "epoch": 1.4947807933194155, + "grad_norm": 0.7473112344741821, + "learning_rate": 4.483163686510016e-06, + "loss": 0.5731, + "step": 478 + }, + { + "epoch": 1.4979123173277662, + "grad_norm": 0.7545126676559448, + "learning_rate": 4.480524495274399e-06, + "loss": 0.5536, + "step": 479 + }, + { + "epoch": 1.501043841336117, + "grad_norm": 0.7801297903060913, + "learning_rate": 4.477879363781033e-06, + "loss": 0.5696, + "step": 480 + }, + { + "epoch": 1.5041753653444676, + "grad_norm": 0.7740563750267029, + "learning_rate": 4.475228299963554e-06, + "loss": 0.5526, + "step": 481 + }, + { + "epoch": 1.5073068893528183, + "grad_norm": 0.8600060343742371, + "learning_rate": 4.4725713117733936e-06, + "loss": 0.5051, + "step": 482 + }, + { + "epoch": 1.510438413361169, + "grad_norm": 0.6934283971786499, + "learning_rate": 4.46990840717975e-06, + "loss": 0.5564, + "step": 483 + }, + { + "epoch": 1.5135699373695197, + "grad_norm": 0.8927920460700989, + "learning_rate": 4.46723959416957e-06, + "loss": 0.5529, + "step": 484 + }, + { + "epoch": 1.5167014613778704, + "grad_norm": 0.9570988416671753, + "learning_rate": 4.464564880747517e-06, + "loss": 0.5661, + "step": 485 + }, + { + "epoch": 1.5198329853862211, + "grad_norm": 0.7229202389717102, + "learning_rate": 4.461884274935956e-06, + "loss": 0.5964, + "step": 486 + }, + { + "epoch": 1.522964509394572, + "grad_norm": 0.7367239594459534, + "learning_rate": 4.4591977847749225e-06, + "loss": 0.5455, + "step": 487 + }, + { + "epoch": 1.5260960334029228, + "grad_norm": 0.8062120676040649, + "learning_rate": 4.456505418322103e-06, + "loss": 0.5735, + "step": 488 + }, + { + "epoch": 1.5292275574112735, + "grad_norm": 0.8854482769966125, + "learning_rate": 4.453807183652808e-06, + "loss": 0.5421, + "step": 489 + }, + { + "epoch": 1.5323590814196242, + "grad_norm": 0.7518959045410156, + "learning_rate": 4.451103088859951e-06, + "loss": 0.5083, + "step": 490 + }, + { + "epoch": 1.535490605427975, + "grad_norm": 0.8621206879615784, + "learning_rate": 4.448393142054016e-06, + "loss": 0.4712, + "step": 491 + }, + { + "epoch": 1.5386221294363258, + "grad_norm": 1.0618741512298584, + "learning_rate": 4.445677351363046e-06, + "loss": 0.5808, + "step": 492 + }, + { + "epoch": 1.5417536534446765, + "grad_norm": 0.8261345028877258, + "learning_rate": 4.442955724932607e-06, + "loss": 0.5625, + "step": 493 + }, + { + "epoch": 1.5448851774530272, + "grad_norm": 0.7067139744758606, + "learning_rate": 4.440228270925772e-06, + "loss": 0.5661, + "step": 494 + }, + { + "epoch": 1.548016701461378, + "grad_norm": 0.9234416484832764, + "learning_rate": 4.437494997523091e-06, + "loss": 0.5428, + "step": 495 + }, + { + "epoch": 1.5511482254697286, + "grad_norm": 0.9273470044136047, + "learning_rate": 4.434755912922567e-06, + "loss": 0.5388, + "step": 496 + }, + { + "epoch": 1.5542797494780793, + "grad_norm": 1.0163263082504272, + "learning_rate": 4.4320110253396345e-06, + "loss": 0.5409, + "step": 497 + }, + { + "epoch": 1.55741127348643, + "grad_norm": 0.9542096853256226, + "learning_rate": 4.429260343007133e-06, + "loss": 0.5329, + "step": 498 + }, + { + "epoch": 1.5605427974947808, + "grad_norm": 0.8076801896095276, + "learning_rate": 4.426503874175283e-06, + "loss": 0.5616, + "step": 499 + }, + { + "epoch": 1.5636743215031315, + "grad_norm": 1.0063767433166504, + "learning_rate": 4.423741627111658e-06, + "loss": 0.5369, + "step": 500 + }, + { + "epoch": 1.5668058455114822, + "grad_norm": 1.040286898612976, + "learning_rate": 4.420973610101166e-06, + "loss": 0.5474, + "step": 501 + }, + { + "epoch": 1.5699373695198329, + "grad_norm": 0.7832860946655273, + "learning_rate": 4.4181998314460164e-06, + "loss": 0.5486, + "step": 502 + }, + { + "epoch": 1.5730688935281836, + "grad_norm": 0.8162257075309753, + "learning_rate": 4.415420299465706e-06, + "loss": 0.5054, + "step": 503 + }, + { + "epoch": 1.5762004175365343, + "grad_norm": 0.9108433127403259, + "learning_rate": 4.4126350224969814e-06, + "loss": 0.5399, + "step": 504 + }, + { + "epoch": 1.5793319415448852, + "grad_norm": 0.8002520799636841, + "learning_rate": 4.409844008893824e-06, + "loss": 0.5485, + "step": 505 + }, + { + "epoch": 1.582463465553236, + "grad_norm": 0.8543248772621155, + "learning_rate": 4.407047267027423e-06, + "loss": 0.4984, + "step": 506 + }, + { + "epoch": 1.5855949895615866, + "grad_norm": 0.7154155373573303, + "learning_rate": 4.404244805286141e-06, + "loss": 0.5392, + "step": 507 + }, + { + "epoch": 1.5887265135699373, + "grad_norm": 0.818553626537323, + "learning_rate": 4.401436632075504e-06, + "loss": 0.5178, + "step": 508 + }, + { + "epoch": 1.5918580375782883, + "grad_norm": 0.7535017728805542, + "learning_rate": 4.398622755818167e-06, + "loss": 0.5446, + "step": 509 + }, + { + "epoch": 1.594989561586639, + "grad_norm": 0.9328975677490234, + "learning_rate": 4.395803184953889e-06, + "loss": 0.5546, + "step": 510 + }, + { + "epoch": 1.5981210855949897, + "grad_norm": 0.7960026860237122, + "learning_rate": 4.392977927939508e-06, + "loss": 0.5451, + "step": 511 + }, + { + "epoch": 1.6012526096033404, + "grad_norm": 0.9686267971992493, + "learning_rate": 4.3901469932489195e-06, + "loss": 0.5198, + "step": 512 + }, + { + "epoch": 1.604384133611691, + "grad_norm": 0.903137743473053, + "learning_rate": 4.387310389373047e-06, + "loss": 0.5395, + "step": 513 + }, + { + "epoch": 1.6075156576200418, + "grad_norm": 1.0728516578674316, + "learning_rate": 4.384468124819816e-06, + "loss": 0.5843, + "step": 514 + }, + { + "epoch": 1.6106471816283925, + "grad_norm": 1.0245436429977417, + "learning_rate": 4.3816202081141345e-06, + "loss": 0.5672, + "step": 515 + }, + { + "epoch": 1.6137787056367432, + "grad_norm": 0.9672732353210449, + "learning_rate": 4.378766647797858e-06, + "loss": 0.5369, + "step": 516 + }, + { + "epoch": 1.616910229645094, + "grad_norm": 0.9149513840675354, + "learning_rate": 4.375907452429774e-06, + "loss": 0.4628, + "step": 517 + }, + { + "epoch": 1.6200417536534446, + "grad_norm": 0.7543843984603882, + "learning_rate": 4.373042630585567e-06, + "loss": 0.5344, + "step": 518 + }, + { + "epoch": 1.6231732776617953, + "grad_norm": 0.7589017152786255, + "learning_rate": 4.370172190857801e-06, + "loss": 0.5672, + "step": 519 + }, + { + "epoch": 1.626304801670146, + "grad_norm": 0.803040623664856, + "learning_rate": 4.367296141855887e-06, + "loss": 0.5313, + "step": 520 + }, + { + "epoch": 1.6294363256784967, + "grad_norm": 0.8305794596672058, + "learning_rate": 4.3644144922060625e-06, + "loss": 0.5754, + "step": 521 + }, + { + "epoch": 1.6325678496868476, + "grad_norm": 1.0086486339569092, + "learning_rate": 4.361527250551361e-06, + "loss": 0.5433, + "step": 522 + }, + { + "epoch": 1.6356993736951984, + "grad_norm": 0.7217550277709961, + "learning_rate": 4.35863442555159e-06, + "loss": 0.524, + "step": 523 + }, + { + "epoch": 1.638830897703549, + "grad_norm": 0.7788524627685547, + "learning_rate": 4.355736025883303e-06, + "loss": 0.536, + "step": 524 + }, + { + "epoch": 1.6419624217118998, + "grad_norm": 0.8460550904273987, + "learning_rate": 4.352832060239774e-06, + "loss": 0.5381, + "step": 525 + }, + { + "epoch": 1.6450939457202505, + "grad_norm": 0.7571215033531189, + "learning_rate": 4.3499225373309675e-06, + "loss": 0.541, + "step": 526 + }, + { + "epoch": 1.6482254697286014, + "grad_norm": 0.7343226671218872, + "learning_rate": 4.347007465883523e-06, + "loss": 0.5147, + "step": 527 + }, + { + "epoch": 1.651356993736952, + "grad_norm": 0.7271892428398132, + "learning_rate": 4.3440868546407165e-06, + "loss": 0.5311, + "step": 528 + }, + { + "epoch": 1.6544885177453028, + "grad_norm": 0.8166136741638184, + "learning_rate": 4.341160712362442e-06, + "loss": 0.5379, + "step": 529 + }, + { + "epoch": 1.6576200417536535, + "grad_norm": 1.5985233783721924, + "learning_rate": 4.338229047825182e-06, + "loss": 0.5782, + "step": 530 + }, + { + "epoch": 1.6607515657620042, + "grad_norm": 0.7835702896118164, + "learning_rate": 4.3352918698219835e-06, + "loss": 0.525, + "step": 531 + }, + { + "epoch": 1.663883089770355, + "grad_norm": 0.7278687953948975, + "learning_rate": 4.332349187162428e-06, + "loss": 0.5266, + "step": 532 + }, + { + "epoch": 1.6670146137787056, + "grad_norm": 0.8240190148353577, + "learning_rate": 4.329401008672608e-06, + "loss": 0.5515, + "step": 533 + }, + { + "epoch": 1.6701461377870563, + "grad_norm": 0.9447080492973328, + "learning_rate": 4.326447343195102e-06, + "loss": 0.5596, + "step": 534 + }, + { + "epoch": 1.673277661795407, + "grad_norm": 0.7827372550964355, + "learning_rate": 4.323488199588944e-06, + "loss": 0.5466, + "step": 535 + }, + { + "epoch": 1.6764091858037578, + "grad_norm": 0.9252517223358154, + "learning_rate": 4.320523586729599e-06, + "loss": 0.5433, + "step": 536 + }, + { + "epoch": 1.6795407098121085, + "grad_norm": 0.9437504410743713, + "learning_rate": 4.317553513508934e-06, + "loss": 0.5552, + "step": 537 + }, + { + "epoch": 1.6826722338204592, + "grad_norm": 0.8972746133804321, + "learning_rate": 4.3145779888351986e-06, + "loss": 0.5259, + "step": 538 + }, + { + "epoch": 1.6858037578288099, + "grad_norm": 0.8017446994781494, + "learning_rate": 4.311597021632988e-06, + "loss": 0.5263, + "step": 539 + }, + { + "epoch": 1.6889352818371608, + "grad_norm": 0.7875497341156006, + "learning_rate": 4.3086106208432235e-06, + "loss": 0.5316, + "step": 540 + }, + { + "epoch": 1.6920668058455115, + "grad_norm": 0.8204905986785889, + "learning_rate": 4.305618795423125e-06, + "loss": 0.5506, + "step": 541 + }, + { + "epoch": 1.6951983298538622, + "grad_norm": 0.888359785079956, + "learning_rate": 4.30262155434618e-06, + "loss": 0.4825, + "step": 542 + }, + { + "epoch": 1.698329853862213, + "grad_norm": 1.1026058197021484, + "learning_rate": 4.29961890660212e-06, + "loss": 0.5321, + "step": 543 + }, + { + "epoch": 1.7014613778705638, + "grad_norm": 0.7662535905838013, + "learning_rate": 4.2966108611968945e-06, + "loss": 0.5432, + "step": 544 + }, + { + "epoch": 1.7045929018789145, + "grad_norm": 1.1951749324798584, + "learning_rate": 4.293597427152641e-06, + "loss": 0.5123, + "step": 545 + }, + { + "epoch": 1.7077244258872653, + "grad_norm": 1.303183913230896, + "learning_rate": 4.290578613507661e-06, + "loss": 0.5346, + "step": 546 + }, + { + "epoch": 1.710855949895616, + "grad_norm": 0.7653357982635498, + "learning_rate": 4.287554429316387e-06, + "loss": 0.5397, + "step": 547 + }, + { + "epoch": 1.7139874739039667, + "grad_norm": 0.796215295791626, + "learning_rate": 4.284524883649366e-06, + "loss": 0.5421, + "step": 548 + }, + { + "epoch": 1.7171189979123174, + "grad_norm": 0.7599332332611084, + "learning_rate": 4.281489985593219e-06, + "loss": 0.5289, + "step": 549 + }, + { + "epoch": 1.720250521920668, + "grad_norm": 0.8029115796089172, + "learning_rate": 4.2784497442506265e-06, + "loss": 0.5409, + "step": 550 + }, + { + "epoch": 1.7233820459290188, + "grad_norm": 0.7194099426269531, + "learning_rate": 4.275404168740291e-06, + "loss": 0.5327, + "step": 551 + }, + { + "epoch": 1.7265135699373695, + "grad_norm": 0.7960740923881531, + "learning_rate": 4.272353268196917e-06, + "loss": 0.4896, + "step": 552 + }, + { + "epoch": 1.7296450939457202, + "grad_norm": 0.9572116732597351, + "learning_rate": 4.269297051771178e-06, + "loss": 0.5402, + "step": 553 + }, + { + "epoch": 1.732776617954071, + "grad_norm": 1.3604938983917236, + "learning_rate": 4.266235528629695e-06, + "loss": 0.5792, + "step": 554 + }, + { + "epoch": 1.7359081419624216, + "grad_norm": 2.067286729812622, + "learning_rate": 4.263168707955002e-06, + "loss": 0.5033, + "step": 555 + }, + { + "epoch": 1.7390396659707723, + "grad_norm": 0.8031097054481506, + "learning_rate": 4.260096598945523e-06, + "loss": 0.5117, + "step": 556 + }, + { + "epoch": 1.742171189979123, + "grad_norm": 1.0241729021072388, + "learning_rate": 4.257019210815546e-06, + "loss": 0.5359, + "step": 557 + }, + { + "epoch": 1.745302713987474, + "grad_norm": 0.7625218629837036, + "learning_rate": 4.25393655279519e-06, + "loss": 0.5625, + "step": 558 + }, + { + "epoch": 1.7484342379958246, + "grad_norm": 0.8603503704071045, + "learning_rate": 4.250848634130381e-06, + "loss": 0.5043, + "step": 559 + }, + { + "epoch": 1.7515657620041754, + "grad_norm": 0.9543750286102295, + "learning_rate": 4.247755464082824e-06, + "loss": 0.5364, + "step": 560 + }, + { + "epoch": 1.754697286012526, + "grad_norm": 0.9707463979721069, + "learning_rate": 4.244657051929973e-06, + "loss": 0.5184, + "step": 561 + }, + { + "epoch": 1.757828810020877, + "grad_norm": 0.7491432428359985, + "learning_rate": 4.241553406965008e-06, + "loss": 0.559, + "step": 562 + }, + { + "epoch": 1.7609603340292277, + "grad_norm": 0.7444972991943359, + "learning_rate": 4.238444538496801e-06, + "loss": 0.5327, + "step": 563 + }, + { + "epoch": 1.7640918580375784, + "grad_norm": 2.7108678817749023, + "learning_rate": 4.235330455849892e-06, + "loss": 0.55, + "step": 564 + }, + { + "epoch": 1.767223382045929, + "grad_norm": 1.6716049909591675, + "learning_rate": 4.232211168364459e-06, + "loss": 0.5093, + "step": 565 + }, + { + "epoch": 1.7703549060542798, + "grad_norm": 0.7023475170135498, + "learning_rate": 4.229086685396295e-06, + "loss": 0.569, + "step": 566 + }, + { + "epoch": 1.7734864300626305, + "grad_norm": 0.8596265316009521, + "learning_rate": 4.225957016316771e-06, + "loss": 0.5128, + "step": 567 + }, + { + "epoch": 1.7766179540709812, + "grad_norm": 0.8110849857330322, + "learning_rate": 4.222822170512816e-06, + "loss": 0.5142, + "step": 568 + }, + { + "epoch": 1.779749478079332, + "grad_norm": 0.7583725452423096, + "learning_rate": 4.219682157386884e-06, + "loss": 0.5584, + "step": 569 + }, + { + "epoch": 1.7828810020876826, + "grad_norm": 0.787811279296875, + "learning_rate": 4.21653698635693e-06, + "loss": 0.5068, + "step": 570 + }, + { + "epoch": 1.7860125260960333, + "grad_norm": 0.8298993110656738, + "learning_rate": 4.213386666856375e-06, + "loss": 0.5496, + "step": 571 + }, + { + "epoch": 1.789144050104384, + "grad_norm": 0.8999841213226318, + "learning_rate": 4.210231208334087e-06, + "loss": 0.5454, + "step": 572 + }, + { + "epoch": 1.7922755741127347, + "grad_norm": 4.264521598815918, + "learning_rate": 4.207070620254345e-06, + "loss": 0.5486, + "step": 573 + }, + { + "epoch": 1.7954070981210855, + "grad_norm": 0.8517448306083679, + "learning_rate": 4.203904912096812e-06, + "loss": 0.5566, + "step": 574 + }, + { + "epoch": 1.7985386221294362, + "grad_norm": 0.9230182766914368, + "learning_rate": 4.200734093356511e-06, + "loss": 0.4964, + "step": 575 + }, + { + "epoch": 1.801670146137787, + "grad_norm": 1.224039912223816, + "learning_rate": 4.197558173543791e-06, + "loss": 0.5356, + "step": 576 + }, + { + "epoch": 1.8048016701461378, + "grad_norm": 0.9998573660850525, + "learning_rate": 4.194377162184301e-06, + "loss": 0.5334, + "step": 577 + }, + { + "epoch": 1.8079331941544885, + "grad_norm": 0.865521252155304, + "learning_rate": 4.191191068818963e-06, + "loss": 0.5036, + "step": 578 + }, + { + "epoch": 1.8110647181628392, + "grad_norm": 0.8048138618469238, + "learning_rate": 4.18799990300394e-06, + "loss": 0.4979, + "step": 579 + }, + { + "epoch": 1.8141962421711901, + "grad_norm": 0.717815637588501, + "learning_rate": 4.184803674310609e-06, + "loss": 0.5623, + "step": 580 + }, + { + "epoch": 1.8173277661795408, + "grad_norm": 0.8403327465057373, + "learning_rate": 4.1816023923255335e-06, + "loss": 0.5055, + "step": 581 + }, + { + "epoch": 1.8204592901878915, + "grad_norm": 0.7298995852470398, + "learning_rate": 4.178396066650432e-06, + "loss": 0.5641, + "step": 582 + }, + { + "epoch": 1.8235908141962422, + "grad_norm": 0.9469727873802185, + "learning_rate": 4.1751847069021516e-06, + "loss": 0.5557, + "step": 583 + }, + { + "epoch": 1.826722338204593, + "grad_norm": 0.8641784191131592, + "learning_rate": 4.1719683227126386e-06, + "loss": 0.5153, + "step": 584 + }, + { + "epoch": 1.8298538622129437, + "grad_norm": 0.7316668629646301, + "learning_rate": 4.168746923728908e-06, + "loss": 0.4988, + "step": 585 + }, + { + "epoch": 1.8329853862212944, + "grad_norm": 0.8795468807220459, + "learning_rate": 4.165520519613017e-06, + "loss": 0.5483, + "step": 586 + }, + { + "epoch": 1.836116910229645, + "grad_norm": 0.7323560118675232, + "learning_rate": 4.162289120042034e-06, + "loss": 0.5194, + "step": 587 + }, + { + "epoch": 1.8392484342379958, + "grad_norm": 0.8217021822929382, + "learning_rate": 4.159052734708013e-06, + "loss": 0.532, + "step": 588 + }, + { + "epoch": 1.8423799582463465, + "grad_norm": 0.7669674754142761, + "learning_rate": 4.155811373317958e-06, + "loss": 0.541, + "step": 589 + }, + { + "epoch": 1.8455114822546972, + "grad_norm": 0.8312156200408936, + "learning_rate": 4.152565045593801e-06, + "loss": 0.5298, + "step": 590 + }, + { + "epoch": 1.848643006263048, + "grad_norm": 0.8967565298080444, + "learning_rate": 4.1493137612723665e-06, + "loss": 0.51, + "step": 591 + }, + { + "epoch": 1.8517745302713986, + "grad_norm": 0.8706664443016052, + "learning_rate": 4.14605753010535e-06, + "loss": 0.4941, + "step": 592 + }, + { + "epoch": 1.8549060542797495, + "grad_norm": 0.7585753798484802, + "learning_rate": 4.14279636185928e-06, + "loss": 0.5161, + "step": 593 + }, + { + "epoch": 1.8580375782881002, + "grad_norm": 0.7495241165161133, + "learning_rate": 4.1395302663154954e-06, + "loss": 0.5388, + "step": 594 + }, + { + "epoch": 1.861169102296451, + "grad_norm": 1.0746862888336182, + "learning_rate": 4.136259253270114e-06, + "loss": 0.4976, + "step": 595 + }, + { + "epoch": 1.8643006263048016, + "grad_norm": 0.872309684753418, + "learning_rate": 4.132983332534e-06, + "loss": 0.559, + "step": 596 + }, + { + "epoch": 1.8674321503131524, + "grad_norm": 0.8759891986846924, + "learning_rate": 4.1297025139327405e-06, + "loss": 0.5436, + "step": 597 + }, + { + "epoch": 1.8705636743215033, + "grad_norm": 1.1044493913650513, + "learning_rate": 4.126416807306611e-06, + "loss": 0.5476, + "step": 598 + }, + { + "epoch": 1.873695198329854, + "grad_norm": 0.8340442180633545, + "learning_rate": 4.123126222510549e-06, + "loss": 0.4592, + "step": 599 + }, + { + "epoch": 1.8768267223382047, + "grad_norm": 0.8331449031829834, + "learning_rate": 4.119830769414123e-06, + "loss": 0.5219, + "step": 600 + }, + { + "epoch": 1.8799582463465554, + "grad_norm": 1.0862973928451538, + "learning_rate": 4.116530457901503e-06, + "loss": 0.5159, + "step": 601 + }, + { + "epoch": 1.883089770354906, + "grad_norm": 0.8524414300918579, + "learning_rate": 4.113225297871431e-06, + "loss": 0.5502, + "step": 602 + }, + { + "epoch": 1.8862212943632568, + "grad_norm": 1.4945416450500488, + "learning_rate": 4.10991529923719e-06, + "loss": 0.5627, + "step": 603 + }, + { + "epoch": 1.8893528183716075, + "grad_norm": 1.5518157482147217, + "learning_rate": 4.10660047192658e-06, + "loss": 0.5517, + "step": 604 + }, + { + "epoch": 1.8924843423799582, + "grad_norm": 2.56638765335083, + "learning_rate": 4.103280825881878e-06, + "loss": 0.5422, + "step": 605 + }, + { + "epoch": 1.895615866388309, + "grad_norm": 0.867254912853241, + "learning_rate": 4.099956371059817e-06, + "loss": 0.4991, + "step": 606 + }, + { + "epoch": 1.8987473903966596, + "grad_norm": 0.9555892944335938, + "learning_rate": 4.096627117431554e-06, + "loss": 0.5339, + "step": 607 + }, + { + "epoch": 1.9018789144050103, + "grad_norm": 0.7905483245849609, + "learning_rate": 4.093293074982638e-06, + "loss": 0.5168, + "step": 608 + }, + { + "epoch": 1.905010438413361, + "grad_norm": 0.7500227093696594, + "learning_rate": 4.089954253712981e-06, + "loss": 0.5096, + "step": 609 + }, + { + "epoch": 1.9081419624217117, + "grad_norm": 0.8458324074745178, + "learning_rate": 4.086610663636828e-06, + "loss": 0.5296, + "step": 610 + }, + { + "epoch": 1.9112734864300627, + "grad_norm": 0.7392706871032715, + "learning_rate": 4.08326231478273e-06, + "loss": 0.5305, + "step": 611 + }, + { + "epoch": 1.9144050104384134, + "grad_norm": 0.8113343715667725, + "learning_rate": 4.079909217193508e-06, + "loss": 0.5044, + "step": 612 + }, + { + "epoch": 1.917536534446764, + "grad_norm": 0.7637801766395569, + "learning_rate": 4.076551380926226e-06, + "loss": 0.5298, + "step": 613 + }, + { + "epoch": 1.9206680584551148, + "grad_norm": 1.0523375272750854, + "learning_rate": 4.073188816052164e-06, + "loss": 0.5111, + "step": 614 + }, + { + "epoch": 1.9237995824634657, + "grad_norm": 0.8224868774414062, + "learning_rate": 4.069821532656781e-06, + "loss": 0.5178, + "step": 615 + }, + { + "epoch": 1.9269311064718164, + "grad_norm": 0.7270777821540833, + "learning_rate": 4.066449540839693e-06, + "loss": 0.5307, + "step": 616 + }, + { + "epoch": 1.9300626304801671, + "grad_norm": 0.7214602828025818, + "learning_rate": 4.063072850714631e-06, + "loss": 0.5171, + "step": 617 + }, + { + "epoch": 1.9331941544885178, + "grad_norm": 0.7333671450614929, + "learning_rate": 4.059691472409426e-06, + "loss": 0.56, + "step": 618 + }, + { + "epoch": 1.9363256784968685, + "grad_norm": 0.9166824221611023, + "learning_rate": 4.056305416065964e-06, + "loss": 0.5388, + "step": 619 + }, + { + "epoch": 1.9394572025052192, + "grad_norm": 0.7743303775787354, + "learning_rate": 4.052914691840167e-06, + "loss": 0.5134, + "step": 620 + }, + { + "epoch": 1.94258872651357, + "grad_norm": 0.704097330570221, + "learning_rate": 4.0495193099019524e-06, + "loss": 0.4926, + "step": 621 + }, + { + "epoch": 1.9457202505219207, + "grad_norm": 0.8508503437042236, + "learning_rate": 4.046119280435212e-06, + "loss": 0.5008, + "step": 622 + }, + { + "epoch": 1.9488517745302714, + "grad_norm": 0.725933313369751, + "learning_rate": 4.042714613637775e-06, + "loss": 0.5549, + "step": 623 + }, + { + "epoch": 1.951983298538622, + "grad_norm": 0.8919175863265991, + "learning_rate": 4.039305319721381e-06, + "loss": 0.5183, + "step": 624 + }, + { + "epoch": 1.9551148225469728, + "grad_norm": 0.827919065952301, + "learning_rate": 4.035891408911644e-06, + "loss": 0.5624, + "step": 625 + }, + { + "epoch": 1.9582463465553235, + "grad_norm": 0.7415187358856201, + "learning_rate": 4.032472891448032e-06, + "loss": 0.5454, + "step": 626 + }, + { + "epoch": 1.9613778705636742, + "grad_norm": 0.7675788998603821, + "learning_rate": 4.029049777583824e-06, + "loss": 0.5361, + "step": 627 + }, + { + "epoch": 1.964509394572025, + "grad_norm": 0.8464030623435974, + "learning_rate": 4.025622077586088e-06, + "loss": 0.5295, + "step": 628 + }, + { + "epoch": 1.9676409185803758, + "grad_norm": 0.7641633749008179, + "learning_rate": 4.022189801735646e-06, + "loss": 0.55, + "step": 629 + }, + { + "epoch": 1.9707724425887265, + "grad_norm": 0.7813227772712708, + "learning_rate": 4.018752960327048e-06, + "loss": 0.5587, + "step": 630 + }, + { + "epoch": 1.9739039665970772, + "grad_norm": 0.7576701641082764, + "learning_rate": 4.015311563668533e-06, + "loss": 0.5413, + "step": 631 + }, + { + "epoch": 1.977035490605428, + "grad_norm": 0.6949650049209595, + "learning_rate": 4.011865622082004e-06, + "loss": 0.5344, + "step": 632 + }, + { + "epoch": 1.9801670146137789, + "grad_norm": 0.9009145498275757, + "learning_rate": 4.008415145902997e-06, + "loss": 0.5233, + "step": 633 + }, + { + "epoch": 1.9832985386221296, + "grad_norm": 0.7635822892189026, + "learning_rate": 4.004960145480651e-06, + "loss": 0.4981, + "step": 634 + }, + { + "epoch": 1.9864300626304803, + "grad_norm": 0.8916334509849548, + "learning_rate": 4.0015006311776685e-06, + "loss": 0.5311, + "step": 635 + }, + { + "epoch": 1.989561586638831, + "grad_norm": 0.7197673320770264, + "learning_rate": 3.998036613370295e-06, + "loss": 0.5361, + "step": 636 + }, + { + "epoch": 1.9926931106471817, + "grad_norm": 0.8391228914260864, + "learning_rate": 3.994568102448284e-06, + "loss": 0.5473, + "step": 637 + }, + { + "epoch": 1.9958246346555324, + "grad_norm": 0.9371750950813293, + "learning_rate": 3.991095108814862e-06, + "loss": 0.5303, + "step": 638 + }, + { + "epoch": 1.998956158663883, + "grad_norm": 0.8929619789123535, + "learning_rate": 3.9876176428867046e-06, + "loss": 0.533, + "step": 639 + }, + { + "epoch": 2.0, + "grad_norm": 0.8929619789123535, + "learning_rate": 3.9841357150938984e-06, + "loss": 0.1831, + "step": 640 + }, + { + "epoch": 2.0031315240083507, + "grad_norm": 0.8802503347396851, + "learning_rate": 3.9806493358799135e-06, + "loss": 0.493, + "step": 641 + }, + { + "epoch": 2.0062630480167014, + "grad_norm": 0.802759051322937, + "learning_rate": 3.977158515701571e-06, + "loss": 0.498, + "step": 642 + }, + { + "epoch": 2.009394572025052, + "grad_norm": 1.0235401391983032, + "learning_rate": 3.973663265029013e-06, + "loss": 0.4887, + "step": 643 + }, + { + "epoch": 2.012526096033403, + "grad_norm": 0.7219089865684509, + "learning_rate": 3.97016359434567e-06, + "loss": 0.4628, + "step": 644 + }, + { + "epoch": 2.0156576200417535, + "grad_norm": 0.7887073755264282, + "learning_rate": 3.966659514148229e-06, + "loss": 0.525, + "step": 645 + }, + { + "epoch": 2.0187891440501042, + "grad_norm": 0.7960914969444275, + "learning_rate": 3.963151034946602e-06, + "loss": 0.4643, + "step": 646 + }, + { + "epoch": 2.021920668058455, + "grad_norm": 0.7902271151542664, + "learning_rate": 3.959638167263895e-06, + "loss": 0.4922, + "step": 647 + }, + { + "epoch": 2.0250521920668056, + "grad_norm": 0.9501478672027588, + "learning_rate": 3.956120921636379e-06, + "loss": 0.5285, + "step": 648 + }, + { + "epoch": 2.028183716075157, + "grad_norm": 0.9510527849197388, + "learning_rate": 3.952599308613454e-06, + "loss": 0.4909, + "step": 649 + }, + { + "epoch": 2.0313152400835075, + "grad_norm": 0.9408219456672668, + "learning_rate": 3.949073338757619e-06, + "loss": 0.4912, + "step": 650 + }, + { + "epoch": 2.034446764091858, + "grad_norm": 0.7148041725158691, + "learning_rate": 3.945543022644441e-06, + "loss": 0.4792, + "step": 651 + }, + { + "epoch": 2.037578288100209, + "grad_norm": 0.7737464904785156, + "learning_rate": 3.942008370862522e-06, + "loss": 0.4694, + "step": 652 + }, + { + "epoch": 2.0407098121085596, + "grad_norm": 0.8405889868736267, + "learning_rate": 3.938469394013472e-06, + "loss": 0.5048, + "step": 653 + }, + { + "epoch": 2.0438413361169103, + "grad_norm": 0.7896456718444824, + "learning_rate": 3.934926102711869e-06, + "loss": 0.4882, + "step": 654 + }, + { + "epoch": 2.046972860125261, + "grad_norm": 0.9290387034416199, + "learning_rate": 3.931378507585231e-06, + "loss": 0.503, + "step": 655 + }, + { + "epoch": 2.0501043841336117, + "grad_norm": 0.7386118769645691, + "learning_rate": 3.927826619273991e-06, + "loss": 0.4918, + "step": 656 + }, + { + "epoch": 2.0532359081419624, + "grad_norm": 0.9878676533699036, + "learning_rate": 3.92427044843145e-06, + "loss": 0.4958, + "step": 657 + }, + { + "epoch": 2.056367432150313, + "grad_norm": 1.0111151933670044, + "learning_rate": 3.92071000572376e-06, + "loss": 0.4886, + "step": 658 + }, + { + "epoch": 2.059498956158664, + "grad_norm": 0.8612061738967896, + "learning_rate": 3.917145301829884e-06, + "loss": 0.5216, + "step": 659 + }, + { + "epoch": 2.0626304801670146, + "grad_norm": 0.7458518743515015, + "learning_rate": 3.913576347441564e-06, + "loss": 0.4807, + "step": 660 + }, + { + "epoch": 2.0657620041753653, + "grad_norm": 0.7775886058807373, + "learning_rate": 3.910003153263294e-06, + "loss": 0.4837, + "step": 661 + }, + { + "epoch": 2.068893528183716, + "grad_norm": 0.7144196629524231, + "learning_rate": 3.906425730012282e-06, + "loss": 0.5081, + "step": 662 + }, + { + "epoch": 2.0720250521920667, + "grad_norm": 0.844971776008606, + "learning_rate": 3.9028440884184215e-06, + "loss": 0.474, + "step": 663 + }, + { + "epoch": 2.0751565762004174, + "grad_norm": 0.9709283113479614, + "learning_rate": 3.899258239224257e-06, + "loss": 0.503, + "step": 664 + }, + { + "epoch": 2.078288100208768, + "grad_norm": 1.1325515508651733, + "learning_rate": 3.895668193184954e-06, + "loss": 0.5058, + "step": 665 + }, + { + "epoch": 2.081419624217119, + "grad_norm": 0.7211254239082336, + "learning_rate": 3.892073961068266e-06, + "loss": 0.4982, + "step": 666 + }, + { + "epoch": 2.08455114822547, + "grad_norm": 0.8975517153739929, + "learning_rate": 3.888475553654502e-06, + "loss": 0.4699, + "step": 667 + }, + { + "epoch": 2.0876826722338206, + "grad_norm": 0.8270771503448486, + "learning_rate": 3.884872981736493e-06, + "loss": 0.4586, + "step": 668 + }, + { + "epoch": 2.0908141962421714, + "grad_norm": 0.8606625199317932, + "learning_rate": 3.881266256119561e-06, + "loss": 0.5299, + "step": 669 + }, + { + "epoch": 2.093945720250522, + "grad_norm": 0.9013976454734802, + "learning_rate": 3.877655387621488e-06, + "loss": 0.4887, + "step": 670 + }, + { + "epoch": 2.0970772442588728, + "grad_norm": 0.7603903412818909, + "learning_rate": 3.8740403870724795e-06, + "loss": 0.4992, + "step": 671 + }, + { + "epoch": 2.1002087682672235, + "grad_norm": 1.0432350635528564, + "learning_rate": 3.870421265315137e-06, + "loss": 0.5035, + "step": 672 + }, + { + "epoch": 2.103340292275574, + "grad_norm": 0.7727136611938477, + "learning_rate": 3.8667980332044195e-06, + "loss": 0.5006, + "step": 673 + }, + { + "epoch": 2.106471816283925, + "grad_norm": 0.9764307141304016, + "learning_rate": 3.863170701607618e-06, + "loss": 0.5061, + "step": 674 + }, + { + "epoch": 2.1096033402922756, + "grad_norm": 0.747818648815155, + "learning_rate": 3.859539281404317e-06, + "loss": 0.4761, + "step": 675 + }, + { + "epoch": 2.1127348643006263, + "grad_norm": 0.7254915237426758, + "learning_rate": 3.855903783486364e-06, + "loss": 0.5166, + "step": 676 + }, + { + "epoch": 2.115866388308977, + "grad_norm": 0.7678592801094055, + "learning_rate": 3.852264218757839e-06, + "loss": 0.5122, + "step": 677 + }, + { + "epoch": 2.1189979123173277, + "grad_norm": 0.8140144348144531, + "learning_rate": 3.8486205981350165e-06, + "loss": 0.4551, + "step": 678 + }, + { + "epoch": 2.1221294363256784, + "grad_norm": 0.9417359232902527, + "learning_rate": 3.844972932546338e-06, + "loss": 0.4748, + "step": 679 + }, + { + "epoch": 2.125260960334029, + "grad_norm": 0.8035290241241455, + "learning_rate": 3.841321232932378e-06, + "loss": 0.5079, + "step": 680 + }, + { + "epoch": 2.12839248434238, + "grad_norm": 0.8300641775131226, + "learning_rate": 3.837665510245809e-06, + "loss": 0.5018, + "step": 681 + }, + { + "epoch": 2.1315240083507305, + "grad_norm": 0.8293547034263611, + "learning_rate": 3.8340057754513715e-06, + "loss": 0.5042, + "step": 682 + }, + { + "epoch": 2.1346555323590812, + "grad_norm": 0.7780007719993591, + "learning_rate": 3.8303420395258365e-06, + "loss": 0.5048, + "step": 683 + }, + { + "epoch": 2.137787056367432, + "grad_norm": 0.7519420385360718, + "learning_rate": 3.8266743134579785e-06, + "loss": 0.5108, + "step": 684 + }, + { + "epoch": 2.140918580375783, + "grad_norm": 0.7872384190559387, + "learning_rate": 3.8230026082485404e-06, + "loss": 0.4924, + "step": 685 + }, + { + "epoch": 2.144050104384134, + "grad_norm": 0.7479491829872131, + "learning_rate": 3.819326934910197e-06, + "loss": 0.5184, + "step": 686 + }, + { + "epoch": 2.1471816283924845, + "grad_norm": 0.8438299298286438, + "learning_rate": 3.815647304467527e-06, + "loss": 0.4791, + "step": 687 + }, + { + "epoch": 2.150313152400835, + "grad_norm": 0.7923721671104431, + "learning_rate": 3.8119637279569773e-06, + "loss": 0.5305, + "step": 688 + }, + { + "epoch": 2.153444676409186, + "grad_norm": 0.7856534719467163, + "learning_rate": 3.80827621642683e-06, + "loss": 0.5063, + "step": 689 + }, + { + "epoch": 2.1565762004175366, + "grad_norm": 0.8544500470161438, + "learning_rate": 3.8045847809371706e-06, + "loss": 0.4989, + "step": 690 + }, + { + "epoch": 2.1597077244258873, + "grad_norm": 0.865390956401825, + "learning_rate": 3.800889432559852e-06, + "loss": 0.4931, + "step": 691 + }, + { + "epoch": 2.162839248434238, + "grad_norm": 0.9809399247169495, + "learning_rate": 3.797190182378466e-06, + "loss": 0.4785, + "step": 692 + }, + { + "epoch": 2.1659707724425887, + "grad_norm": 0.7954536080360413, + "learning_rate": 3.793487041488304e-06, + "loss": 0.4847, + "step": 693 + }, + { + "epoch": 2.1691022964509394, + "grad_norm": 0.754704475402832, + "learning_rate": 3.7897800209963298e-06, + "loss": 0.5125, + "step": 694 + }, + { + "epoch": 2.17223382045929, + "grad_norm": 0.7319822311401367, + "learning_rate": 3.7860691320211414e-06, + "loss": 0.477, + "step": 695 + }, + { + "epoch": 2.175365344467641, + "grad_norm": 0.8198635578155518, + "learning_rate": 3.7823543856929403e-06, + "loss": 0.4764, + "step": 696 + }, + { + "epoch": 2.1784968684759916, + "grad_norm": 0.708933413028717, + "learning_rate": 3.7786357931534987e-06, + "loss": 0.4948, + "step": 697 + }, + { + "epoch": 2.1816283924843423, + "grad_norm": 0.8493193984031677, + "learning_rate": 3.774913365556123e-06, + "loss": 0.5271, + "step": 698 + }, + { + "epoch": 2.184759916492693, + "grad_norm": 0.7999475002288818, + "learning_rate": 3.771187114065622e-06, + "loss": 0.4804, + "step": 699 + }, + { + "epoch": 2.1878914405010437, + "grad_norm": 0.8366796970367432, + "learning_rate": 3.7674570498582776e-06, + "loss": 0.457, + "step": 700 + }, + { + "epoch": 2.1910229645093944, + "grad_norm": 0.7935530543327332, + "learning_rate": 3.7637231841218015e-06, + "loss": 0.5001, + "step": 701 + }, + { + "epoch": 2.1941544885177455, + "grad_norm": 0.7700153589248657, + "learning_rate": 3.7599855280553125e-06, + "loss": 0.5091, + "step": 702 + }, + { + "epoch": 2.1972860125260962, + "grad_norm": 0.7991652488708496, + "learning_rate": 3.756244092869294e-06, + "loss": 0.4955, + "step": 703 + }, + { + "epoch": 2.200417536534447, + "grad_norm": 0.720051646232605, + "learning_rate": 3.752498889785567e-06, + "loss": 0.4902, + "step": 704 + }, + { + "epoch": 2.2035490605427976, + "grad_norm": 0.7312369346618652, + "learning_rate": 3.748749930037252e-06, + "loss": 0.4935, + "step": 705 + }, + { + "epoch": 2.2066805845511483, + "grad_norm": 0.8418563008308411, + "learning_rate": 3.744997224868739e-06, + "loss": 0.5186, + "step": 706 + }, + { + "epoch": 2.209812108559499, + "grad_norm": 0.8324081301689148, + "learning_rate": 3.741240785535649e-06, + "loss": 0.482, + "step": 707 + }, + { + "epoch": 2.2129436325678498, + "grad_norm": 0.8051855564117432, + "learning_rate": 3.737480623304805e-06, + "loss": 0.4663, + "step": 708 + }, + { + "epoch": 2.2160751565762005, + "grad_norm": 0.9464184641838074, + "learning_rate": 3.7337167494541948e-06, + "loss": 0.451, + "step": 709 + }, + { + "epoch": 2.219206680584551, + "grad_norm": 1.0227075815200806, + "learning_rate": 3.729949175272942e-06, + "loss": 0.4428, + "step": 710 + }, + { + "epoch": 2.222338204592902, + "grad_norm": 0.7930364012718201, + "learning_rate": 3.7261779120612633e-06, + "loss": 0.5132, + "step": 711 + }, + { + "epoch": 2.2254697286012526, + "grad_norm": 0.9033688306808472, + "learning_rate": 3.7224029711304444e-06, + "loss": 0.476, + "step": 712 + }, + { + "epoch": 2.2286012526096033, + "grad_norm": 0.8026887774467468, + "learning_rate": 3.7186243638028007e-06, + "loss": 0.4959, + "step": 713 + }, + { + "epoch": 2.231732776617954, + "grad_norm": 0.9391745328903198, + "learning_rate": 3.714842101411642e-06, + "loss": 0.4962, + "step": 714 + }, + { + "epoch": 2.2348643006263047, + "grad_norm": 0.7774361371994019, + "learning_rate": 3.711056195301245e-06, + "loss": 0.4748, + "step": 715 + }, + { + "epoch": 2.2379958246346554, + "grad_norm": 0.9278722405433655, + "learning_rate": 3.7072666568268115e-06, + "loss": 0.5074, + "step": 716 + }, + { + "epoch": 2.241127348643006, + "grad_norm": 0.771423876285553, + "learning_rate": 3.7034734973544406e-06, + "loss": 0.5072, + "step": 717 + }, + { + "epoch": 2.244258872651357, + "grad_norm": 0.8707448244094849, + "learning_rate": 3.6996767282610892e-06, + "loss": 0.4851, + "step": 718 + }, + { + "epoch": 2.2473903966597075, + "grad_norm": 0.7641019821166992, + "learning_rate": 3.695876360934543e-06, + "loss": 0.4941, + "step": 719 + }, + { + "epoch": 2.2505219206680582, + "grad_norm": 0.7647167444229126, + "learning_rate": 3.69207240677338e-06, + "loss": 0.5225, + "step": 720 + }, + { + "epoch": 2.2536534446764094, + "grad_norm": 0.9108865261077881, + "learning_rate": 3.6882648771869345e-06, + "loss": 0.454, + "step": 721 + }, + { + "epoch": 2.25678496868476, + "grad_norm": 0.86728835105896, + "learning_rate": 3.6844537835952666e-06, + "loss": 0.4461, + "step": 722 + }, + { + "epoch": 2.259916492693111, + "grad_norm": 1.1055282354354858, + "learning_rate": 3.6806391374291238e-06, + "loss": 0.4618, + "step": 723 + }, + { + "epoch": 2.2630480167014615, + "grad_norm": 0.7591858506202698, + "learning_rate": 3.6768209501299116e-06, + "loss": 0.4901, + "step": 724 + }, + { + "epoch": 2.266179540709812, + "grad_norm": 0.7966359257698059, + "learning_rate": 3.6729992331496554e-06, + "loss": 0.5171, + "step": 725 + }, + { + "epoch": 2.269311064718163, + "grad_norm": 0.983161211013794, + "learning_rate": 3.6691739979509672e-06, + "loss": 0.4949, + "step": 726 + }, + { + "epoch": 2.2724425887265136, + "grad_norm": 0.9200037121772766, + "learning_rate": 3.6653452560070106e-06, + "loss": 0.5234, + "step": 727 + }, + { + "epoch": 2.2755741127348643, + "grad_norm": 1.0288461446762085, + "learning_rate": 3.6615130188014685e-06, + "loss": 0.4713, + "step": 728 + }, + { + "epoch": 2.278705636743215, + "grad_norm": 0.7325463891029358, + "learning_rate": 3.6576772978285065e-06, + "loss": 0.527, + "step": 729 + }, + { + "epoch": 2.2818371607515657, + "grad_norm": 1.0045446157455444, + "learning_rate": 3.6538381045927395e-06, + "loss": 0.5139, + "step": 730 + }, + { + "epoch": 2.2849686847599164, + "grad_norm": 0.7391849756240845, + "learning_rate": 3.6499954506091963e-06, + "loss": 0.4829, + "step": 731 + }, + { + "epoch": 2.288100208768267, + "grad_norm": 0.7808229923248291, + "learning_rate": 3.646149347403286e-06, + "loss": 0.4831, + "step": 732 + }, + { + "epoch": 2.291231732776618, + "grad_norm": 0.7056961059570312, + "learning_rate": 3.6422998065107628e-06, + "loss": 0.5066, + "step": 733 + }, + { + "epoch": 2.2943632567849686, + "grad_norm": 0.7498443126678467, + "learning_rate": 3.6384468394776935e-06, + "loss": 0.4724, + "step": 734 + }, + { + "epoch": 2.2974947807933193, + "grad_norm": 0.8511576056480408, + "learning_rate": 3.634590457860418e-06, + "loss": 0.5286, + "step": 735 + }, + { + "epoch": 2.30062630480167, + "grad_norm": 0.873635470867157, + "learning_rate": 3.63073067322552e-06, + "loss": 0.4751, + "step": 736 + }, + { + "epoch": 2.3037578288100207, + "grad_norm": 0.7427377104759216, + "learning_rate": 3.626867497149788e-06, + "loss": 0.475, + "step": 737 + }, + { + "epoch": 2.306889352818372, + "grad_norm": 1.0591017007827759, + "learning_rate": 3.623000941220186e-06, + "loss": 0.4591, + "step": 738 + }, + { + "epoch": 2.3100208768267225, + "grad_norm": 0.8767879009246826, + "learning_rate": 3.6191310170338114e-06, + "loss": 0.4673, + "step": 739 + }, + { + "epoch": 2.3131524008350732, + "grad_norm": 0.9156234860420227, + "learning_rate": 3.615257736197866e-06, + "loss": 0.4622, + "step": 740 + }, + { + "epoch": 2.316283924843424, + "grad_norm": 0.6743756532669067, + "learning_rate": 3.611381110329619e-06, + "loss": 0.4723, + "step": 741 + }, + { + "epoch": 2.3194154488517746, + "grad_norm": 0.8655558228492737, + "learning_rate": 3.6075011510563732e-06, + "loss": 0.471, + "step": 742 + }, + { + "epoch": 2.3225469728601253, + "grad_norm": 0.7652033567428589, + "learning_rate": 3.603617870015429e-06, + "loss": 0.5155, + "step": 743 + }, + { + "epoch": 2.325678496868476, + "grad_norm": 0.7970699667930603, + "learning_rate": 3.599731278854049e-06, + "loss": 0.4507, + "step": 744 + }, + { + "epoch": 2.3288100208768268, + "grad_norm": 0.7538278698921204, + "learning_rate": 3.5958413892294253e-06, + "loss": 0.5093, + "step": 745 + }, + { + "epoch": 2.3319415448851775, + "grad_norm": 0.735996663570404, + "learning_rate": 3.5919482128086414e-06, + "loss": 0.5008, + "step": 746 + }, + { + "epoch": 2.335073068893528, + "grad_norm": 0.7643904685974121, + "learning_rate": 3.588051761268642e-06, + "loss": 0.5072, + "step": 747 + }, + { + "epoch": 2.338204592901879, + "grad_norm": 0.7646260857582092, + "learning_rate": 3.584152046296191e-06, + "loss": 0.4578, + "step": 748 + }, + { + "epoch": 2.3413361169102296, + "grad_norm": 0.7873825430870056, + "learning_rate": 3.5802490795878446e-06, + "loss": 0.5249, + "step": 749 + }, + { + "epoch": 2.3444676409185803, + "grad_norm": 1.095333218574524, + "learning_rate": 3.5763428728499095e-06, + "loss": 0.4913, + "step": 750 + }, + { + "epoch": 2.347599164926931, + "grad_norm": 1.3425395488739014, + "learning_rate": 3.5724334377984107e-06, + "loss": 0.5317, + "step": 751 + }, + { + "epoch": 2.3507306889352817, + "grad_norm": 0.7151113748550415, + "learning_rate": 3.568520786159055e-06, + "loss": 0.5135, + "step": 752 + }, + { + "epoch": 2.3538622129436324, + "grad_norm": 0.8072878122329712, + "learning_rate": 3.5646049296672004e-06, + "loss": 0.4863, + "step": 753 + }, + { + "epoch": 2.356993736951983, + "grad_norm": 0.8040189743041992, + "learning_rate": 3.5606858800678123e-06, + "loss": 0.4668, + "step": 754 + }, + { + "epoch": 2.3601252609603343, + "grad_norm": 0.7749765515327454, + "learning_rate": 3.5567636491154385e-06, + "loss": 0.4681, + "step": 755 + }, + { + "epoch": 2.3632567849686845, + "grad_norm": 0.773013710975647, + "learning_rate": 3.5528382485741638e-06, + "loss": 0.5012, + "step": 756 + }, + { + "epoch": 2.3663883089770357, + "grad_norm": 0.7017714381217957, + "learning_rate": 3.5489096902175835e-06, + "loss": 0.5019, + "step": 757 + }, + { + "epoch": 2.3695198329853864, + "grad_norm": 1.132458209991455, + "learning_rate": 3.5449779858287625e-06, + "loss": 0.5131, + "step": 758 + }, + { + "epoch": 2.372651356993737, + "grad_norm": 0.7624574899673462, + "learning_rate": 3.541043147200202e-06, + "loss": 0.4856, + "step": 759 + }, + { + "epoch": 2.375782881002088, + "grad_norm": 0.9078478217124939, + "learning_rate": 3.5371051861338036e-06, + "loss": 0.4337, + "step": 760 + }, + { + "epoch": 2.3789144050104385, + "grad_norm": 0.8608354330062866, + "learning_rate": 3.5331641144408344e-06, + "loss": 0.5053, + "step": 761 + }, + { + "epoch": 2.382045929018789, + "grad_norm": 0.775047779083252, + "learning_rate": 3.529219943941892e-06, + "loss": 0.4779, + "step": 762 + }, + { + "epoch": 2.38517745302714, + "grad_norm": 0.7775866389274597, + "learning_rate": 3.525272686466866e-06, + "loss": 0.4979, + "step": 763 + }, + { + "epoch": 2.3883089770354906, + "grad_norm": 0.9386464357376099, + "learning_rate": 3.521322353854908e-06, + "loss": 0.5222, + "step": 764 + }, + { + "epoch": 2.3914405010438413, + "grad_norm": 0.874109148979187, + "learning_rate": 3.517368957954391e-06, + "loss": 0.4681, + "step": 765 + }, + { + "epoch": 2.394572025052192, + "grad_norm": 0.824588418006897, + "learning_rate": 3.5134125106228766e-06, + "loss": 0.4955, + "step": 766 + }, + { + "epoch": 2.3977035490605427, + "grad_norm": 0.8790764808654785, + "learning_rate": 3.5094530237270774e-06, + "loss": 0.4722, + "step": 767 + }, + { + "epoch": 2.4008350730688934, + "grad_norm": 1.1399786472320557, + "learning_rate": 3.5054905091428253e-06, + "loss": 0.4771, + "step": 768 + }, + { + "epoch": 2.403966597077244, + "grad_norm": 1.2586532831192017, + "learning_rate": 3.50152497875503e-06, + "loss": 0.4849, + "step": 769 + }, + { + "epoch": 2.407098121085595, + "grad_norm": 0.7706464529037476, + "learning_rate": 3.4975564444576487e-06, + "loss": 0.477, + "step": 770 + }, + { + "epoch": 2.4102296450939455, + "grad_norm": 0.7695909142494202, + "learning_rate": 3.4935849181536484e-06, + "loss": 0.4695, + "step": 771 + }, + { + "epoch": 2.4133611691022967, + "grad_norm": 0.7744433283805847, + "learning_rate": 3.489610411754969e-06, + "loss": 0.499, + "step": 772 + }, + { + "epoch": 2.416492693110647, + "grad_norm": 0.9265744686126709, + "learning_rate": 3.48563293718249e-06, + "loss": 0.481, + "step": 773 + }, + { + "epoch": 2.419624217118998, + "grad_norm": 1.0680506229400635, + "learning_rate": 3.481652506365992e-06, + "loss": 0.4898, + "step": 774 + }, + { + "epoch": 2.422755741127349, + "grad_norm": 0.721493661403656, + "learning_rate": 3.477669131244122e-06, + "loss": 0.4813, + "step": 775 + }, + { + "epoch": 2.4258872651356995, + "grad_norm": 0.7993559837341309, + "learning_rate": 3.4736828237643616e-06, + "loss": 0.5179, + "step": 776 + }, + { + "epoch": 2.4290187891440502, + "grad_norm": 0.8148090839385986, + "learning_rate": 3.4696935958829837e-06, + "loss": 0.4753, + "step": 777 + }, + { + "epoch": 2.432150313152401, + "grad_norm": 0.8006406426429749, + "learning_rate": 3.465701459565022e-06, + "loss": 0.501, + "step": 778 + }, + { + "epoch": 2.4352818371607516, + "grad_norm": 0.9307970404624939, + "learning_rate": 3.4617064267842327e-06, + "loss": 0.487, + "step": 779 + }, + { + "epoch": 2.4384133611691023, + "grad_norm": 0.7192814946174622, + "learning_rate": 3.45770850952306e-06, + "loss": 0.4769, + "step": 780 + }, + { + "epoch": 2.441544885177453, + "grad_norm": 0.7386271953582764, + "learning_rate": 3.4537077197726023e-06, + "loss": 0.4726, + "step": 781 + }, + { + "epoch": 2.4446764091858038, + "grad_norm": 0.8006314039230347, + "learning_rate": 3.449704069532567e-06, + "loss": 0.494, + "step": 782 + }, + { + "epoch": 2.4478079331941545, + "grad_norm": 0.7466752529144287, + "learning_rate": 3.4456975708112477e-06, + "loss": 0.4778, + "step": 783 + }, + { + "epoch": 2.450939457202505, + "grad_norm": 0.8348856568336487, + "learning_rate": 3.4416882356254777e-06, + "loss": 0.4766, + "step": 784 + }, + { + "epoch": 2.454070981210856, + "grad_norm": 0.754851758480072, + "learning_rate": 3.4376760760005994e-06, + "loss": 0.4673, + "step": 785 + }, + { + "epoch": 2.4572025052192066, + "grad_norm": 0.7854018807411194, + "learning_rate": 3.433661103970427e-06, + "loss": 0.4954, + "step": 786 + }, + { + "epoch": 2.4603340292275573, + "grad_norm": 0.7238256931304932, + "learning_rate": 3.4296433315772084e-06, + "loss": 0.496, + "step": 787 + }, + { + "epoch": 2.463465553235908, + "grad_norm": 0.7007659673690796, + "learning_rate": 3.4256227708715915e-06, + "loss": 0.4793, + "step": 788 + }, + { + "epoch": 2.4665970772442587, + "grad_norm": 0.7234371900558472, + "learning_rate": 3.421599433912588e-06, + "loss": 0.4935, + "step": 789 + }, + { + "epoch": 2.4697286012526094, + "grad_norm": 0.7537544965744019, + "learning_rate": 3.4175733327675355e-06, + "loss": 0.5194, + "step": 790 + }, + { + "epoch": 2.4728601252609606, + "grad_norm": 0.7608047127723694, + "learning_rate": 3.4135444795120633e-06, + "loss": 0.4793, + "step": 791 + }, + { + "epoch": 2.4759916492693113, + "grad_norm": 0.7847898006439209, + "learning_rate": 3.4095128862300542e-06, + "loss": 0.4877, + "step": 792 + }, + { + "epoch": 2.479123173277662, + "grad_norm": 0.8002011179924011, + "learning_rate": 3.405478565013609e-06, + "loss": 0.4927, + "step": 793 + }, + { + "epoch": 2.4822546972860127, + "grad_norm": 0.8200219869613647, + "learning_rate": 3.401441527963013e-06, + "loss": 0.4997, + "step": 794 + }, + { + "epoch": 2.4853862212943634, + "grad_norm": 0.7220162749290466, + "learning_rate": 3.3974017871866938e-06, + "loss": 0.4668, + "step": 795 + }, + { + "epoch": 2.488517745302714, + "grad_norm": 0.8022251129150391, + "learning_rate": 3.3933593548011912e-06, + "loss": 0.5179, + "step": 796 + }, + { + "epoch": 2.491649269311065, + "grad_norm": 0.7914465069770813, + "learning_rate": 3.389314242931115e-06, + "loss": 0.4943, + "step": 797 + }, + { + "epoch": 2.4947807933194155, + "grad_norm": 1.1399403810501099, + "learning_rate": 3.385266463709116e-06, + "loss": 0.4896, + "step": 798 + }, + { + "epoch": 2.497912317327766, + "grad_norm": 0.8098909854888916, + "learning_rate": 3.38121602927584e-06, + "loss": 0.4904, + "step": 799 + }, + { + "epoch": 2.501043841336117, + "grad_norm": 0.7434052228927612, + "learning_rate": 3.377162951779902e-06, + "loss": 0.4864, + "step": 800 + }, + { + "epoch": 2.5041753653444676, + "grad_norm": 0.7397809624671936, + "learning_rate": 3.3731072433778407e-06, + "loss": 0.486, + "step": 801 + }, + { + "epoch": 2.5073068893528183, + "grad_norm": 0.99027019739151, + "learning_rate": 3.3690489162340867e-06, + "loss": 0.5011, + "step": 802 + }, + { + "epoch": 2.510438413361169, + "grad_norm": 0.8443610668182373, + "learning_rate": 3.3649879825209246e-06, + "loss": 0.455, + "step": 803 + }, + { + "epoch": 2.5135699373695197, + "grad_norm": 0.755649983882904, + "learning_rate": 3.3609244544184604e-06, + "loss": 0.4563, + "step": 804 + }, + { + "epoch": 2.5167014613778704, + "grad_norm": 0.728018045425415, + "learning_rate": 3.3568583441145765e-06, + "loss": 0.471, + "step": 805 + }, + { + "epoch": 2.519832985386221, + "grad_norm": 0.7777130603790283, + "learning_rate": 3.352789663804904e-06, + "loss": 0.4667, + "step": 806 + }, + { + "epoch": 2.522964509394572, + "grad_norm": 0.7545619606971741, + "learning_rate": 3.3487184256927785e-06, + "loss": 0.4915, + "step": 807 + }, + { + "epoch": 2.526096033402923, + "grad_norm": 0.8374579548835754, + "learning_rate": 3.3446446419892127e-06, + "loss": 0.485, + "step": 808 + }, + { + "epoch": 2.5292275574112733, + "grad_norm": 0.7354666590690613, + "learning_rate": 3.340568324912849e-06, + "loss": 0.5254, + "step": 809 + }, + { + "epoch": 2.5323590814196244, + "grad_norm": 0.7581545114517212, + "learning_rate": 3.3364894866899324e-06, + "loss": 0.4483, + "step": 810 + }, + { + "epoch": 2.535490605427975, + "grad_norm": 0.8077559471130371, + "learning_rate": 3.3324081395542662e-06, + "loss": 0.5022, + "step": 811 + }, + { + "epoch": 2.538622129436326, + "grad_norm": 0.8827865719795227, + "learning_rate": 3.3283242957471806e-06, + "loss": 0.4909, + "step": 812 + }, + { + "epoch": 2.5417536534446765, + "grad_norm": 0.9139482378959656, + "learning_rate": 3.3242379675174953e-06, + "loss": 0.5205, + "step": 813 + }, + { + "epoch": 2.5448851774530272, + "grad_norm": 0.7616812586784363, + "learning_rate": 3.3201491671214797e-06, + "loss": 0.4744, + "step": 814 + }, + { + "epoch": 2.548016701461378, + "grad_norm": 0.987173318862915, + "learning_rate": 3.3160579068228183e-06, + "loss": 0.4876, + "step": 815 + }, + { + "epoch": 2.5511482254697286, + "grad_norm": 1.259137749671936, + "learning_rate": 3.311964198892574e-06, + "loss": 0.454, + "step": 816 + }, + { + "epoch": 2.5542797494780793, + "grad_norm": 0.7866336703300476, + "learning_rate": 3.3078680556091513e-06, + "loss": 0.5107, + "step": 817 + }, + { + "epoch": 2.55741127348643, + "grad_norm": 0.9311352372169495, + "learning_rate": 3.303769489258258e-06, + "loss": 0.4843, + "step": 818 + }, + { + "epoch": 2.5605427974947808, + "grad_norm": 0.8556346893310547, + "learning_rate": 3.299668512132872e-06, + "loss": 0.5017, + "step": 819 + }, + { + "epoch": 2.5636743215031315, + "grad_norm": 2.810598373413086, + "learning_rate": 3.2955651365331988e-06, + "loss": 0.5223, + "step": 820 + }, + { + "epoch": 2.566805845511482, + "grad_norm": 1.0120766162872314, + "learning_rate": 3.29145937476664e-06, + "loss": 0.4959, + "step": 821 + }, + { + "epoch": 2.569937369519833, + "grad_norm": 0.751412034034729, + "learning_rate": 3.287351239147752e-06, + "loss": 0.4941, + "step": 822 + }, + { + "epoch": 2.5730688935281836, + "grad_norm": 1.9308148622512817, + "learning_rate": 3.2832407419982136e-06, + "loss": 0.4965, + "step": 823 + }, + { + "epoch": 2.5762004175365343, + "grad_norm": 0.9215649962425232, + "learning_rate": 3.279127895646786e-06, + "loss": 0.5071, + "step": 824 + }, + { + "epoch": 2.5793319415448854, + "grad_norm": 0.7599574327468872, + "learning_rate": 3.2750127124292754e-06, + "loss": 0.5191, + "step": 825 + }, + { + "epoch": 2.5824634655532357, + "grad_norm": 0.8234940767288208, + "learning_rate": 3.270895204688496e-06, + "loss": 0.4947, + "step": 826 + }, + { + "epoch": 2.585594989561587, + "grad_norm": 0.8401572704315186, + "learning_rate": 3.266775384774238e-06, + "loss": 0.4547, + "step": 827 + }, + { + "epoch": 2.588726513569937, + "grad_norm": 0.8927991986274719, + "learning_rate": 3.262653265043223e-06, + "loss": 0.4296, + "step": 828 + }, + { + "epoch": 2.5918580375782883, + "grad_norm": 0.8009241223335266, + "learning_rate": 3.2585288578590716e-06, + "loss": 0.4578, + "step": 829 + }, + { + "epoch": 2.594989561586639, + "grad_norm": 0.7982021570205688, + "learning_rate": 3.2544021755922663e-06, + "loss": 0.4961, + "step": 830 + }, + { + "epoch": 2.5981210855949897, + "grad_norm": 0.7096095681190491, + "learning_rate": 3.2502732306201112e-06, + "loss": 0.4975, + "step": 831 + }, + { + "epoch": 2.6012526096033404, + "grad_norm": 1.1092045307159424, + "learning_rate": 3.246142035326699e-06, + "loss": 0.4705, + "step": 832 + }, + { + "epoch": 2.604384133611691, + "grad_norm": 0.785799503326416, + "learning_rate": 3.24200860210287e-06, + "loss": 0.479, + "step": 833 + }, + { + "epoch": 2.607515657620042, + "grad_norm": 0.7315773367881775, + "learning_rate": 3.2378729433461804e-06, + "loss": 0.5036, + "step": 834 + }, + { + "epoch": 2.6106471816283925, + "grad_norm": 0.7840189337730408, + "learning_rate": 3.233735071460856e-06, + "loss": 0.4967, + "step": 835 + }, + { + "epoch": 2.613778705636743, + "grad_norm": 0.7186565399169922, + "learning_rate": 3.2295949988577655e-06, + "loss": 0.4889, + "step": 836 + }, + { + "epoch": 2.616910229645094, + "grad_norm": 0.766054093837738, + "learning_rate": 3.2254527379543747e-06, + "loss": 0.539, + "step": 837 + }, + { + "epoch": 2.6200417536534446, + "grad_norm": 0.7705381512641907, + "learning_rate": 3.2213083011747165e-06, + "loss": 0.4968, + "step": 838 + }, + { + "epoch": 2.6231732776617953, + "grad_norm": 1.3530604839324951, + "learning_rate": 3.217161700949346e-06, + "loss": 0.52, + "step": 839 + }, + { + "epoch": 2.626304801670146, + "grad_norm": 0.737389862537384, + "learning_rate": 3.2130129497153107e-06, + "loss": 0.4823, + "step": 840 + }, + { + "epoch": 2.6294363256784967, + "grad_norm": 0.9121193885803223, + "learning_rate": 3.2088620599161064e-06, + "loss": 0.4592, + "step": 841 + }, + { + "epoch": 2.632567849686848, + "grad_norm": 0.8869616389274597, + "learning_rate": 3.2047090440016464e-06, + "loss": 0.5001, + "step": 842 + }, + { + "epoch": 2.635699373695198, + "grad_norm": 0.8447219133377075, + "learning_rate": 3.200553914428219e-06, + "loss": 0.4969, + "step": 843 + }, + { + "epoch": 2.6388308977035493, + "grad_norm": 0.8877657055854797, + "learning_rate": 3.1963966836584524e-06, + "loss": 0.4718, + "step": 844 + }, + { + "epoch": 2.6419624217118995, + "grad_norm": 1.045272946357727, + "learning_rate": 3.192237364161277e-06, + "loss": 0.4864, + "step": 845 + }, + { + "epoch": 2.6450939457202507, + "grad_norm": 0.8485913276672363, + "learning_rate": 3.1880759684118876e-06, + "loss": 0.4688, + "step": 846 + }, + { + "epoch": 2.6482254697286014, + "grad_norm": 0.7328930497169495, + "learning_rate": 3.183912508891709e-06, + "loss": 0.4728, + "step": 847 + }, + { + "epoch": 2.651356993736952, + "grad_norm": 0.7377315759658813, + "learning_rate": 3.179746998088351e-06, + "loss": 0.4672, + "step": 848 + }, + { + "epoch": 2.654488517745303, + "grad_norm": 0.8017002940177917, + "learning_rate": 3.1755794484955817e-06, + "loss": 0.4884, + "step": 849 + }, + { + "epoch": 2.6576200417536535, + "grad_norm": 1.045470952987671, + "learning_rate": 3.171409872613278e-06, + "loss": 0.4789, + "step": 850 + }, + { + "epoch": 2.6607515657620042, + "grad_norm": 0.8823987245559692, + "learning_rate": 3.1672382829473997e-06, + "loss": 0.5117, + "step": 851 + }, + { + "epoch": 2.663883089770355, + "grad_norm": 0.7395204901695251, + "learning_rate": 3.163064692009944e-06, + "loss": 0.5476, + "step": 852 + }, + { + "epoch": 2.6670146137787056, + "grad_norm": 0.7778941988945007, + "learning_rate": 3.1588891123189103e-06, + "loss": 0.5092, + "step": 853 + }, + { + "epoch": 2.6701461377870563, + "grad_norm": 0.8072531819343567, + "learning_rate": 3.1547115563982643e-06, + "loss": 0.4961, + "step": 854 + }, + { + "epoch": 2.673277661795407, + "grad_norm": 0.9018139243125916, + "learning_rate": 3.1505320367778993e-06, + "loss": 0.4624, + "step": 855 + }, + { + "epoch": 2.6764091858037578, + "grad_norm": 0.8554450869560242, + "learning_rate": 3.1463505659935957e-06, + "loss": 0.4971, + "step": 856 + }, + { + "epoch": 2.6795407098121085, + "grad_norm": 0.7727259397506714, + "learning_rate": 3.14216715658699e-06, + "loss": 0.4544, + "step": 857 + }, + { + "epoch": 2.682672233820459, + "grad_norm": 0.9253409504890442, + "learning_rate": 3.137981821105529e-06, + "loss": 0.4893, + "step": 858 + }, + { + "epoch": 2.68580375782881, + "grad_norm": 0.8809456825256348, + "learning_rate": 3.1337945721024403e-06, + "loss": 0.5242, + "step": 859 + }, + { + "epoch": 2.6889352818371606, + "grad_norm": 0.981755256652832, + "learning_rate": 3.129605422136689e-06, + "loss": 0.4686, + "step": 860 + }, + { + "epoch": 2.6920668058455117, + "grad_norm": 1.1278467178344727, + "learning_rate": 3.1254143837729412e-06, + "loss": 0.4813, + "step": 861 + }, + { + "epoch": 2.695198329853862, + "grad_norm": 0.8529123663902283, + "learning_rate": 3.1212214695815285e-06, + "loss": 0.4723, + "step": 862 + }, + { + "epoch": 2.698329853862213, + "grad_norm": 0.7764189839363098, + "learning_rate": 3.1170266921384075e-06, + "loss": 0.4777, + "step": 863 + }, + { + "epoch": 2.701461377870564, + "grad_norm": 0.7364740967750549, + "learning_rate": 3.112830064025124e-06, + "loss": 0.4975, + "step": 864 + }, + { + "epoch": 2.7045929018789145, + "grad_norm": 0.7594549059867859, + "learning_rate": 3.108631597828774e-06, + "loss": 0.5083, + "step": 865 + }, + { + "epoch": 2.7077244258872653, + "grad_norm": 0.7337073683738708, + "learning_rate": 3.104431306141968e-06, + "loss": 0.4778, + "step": 866 + }, + { + "epoch": 2.710855949895616, + "grad_norm": 0.7709932327270508, + "learning_rate": 3.1002292015627894e-06, + "loss": 0.4754, + "step": 867 + }, + { + "epoch": 2.7139874739039667, + "grad_norm": 0.8001313209533691, + "learning_rate": 3.0960252966947605e-06, + "loss": 0.4489, + "step": 868 + }, + { + "epoch": 2.7171189979123174, + "grad_norm": 0.8280592560768127, + "learning_rate": 3.091819604146804e-06, + "loss": 0.4606, + "step": 869 + }, + { + "epoch": 2.720250521920668, + "grad_norm": 0.7463534474372864, + "learning_rate": 3.0876121365332024e-06, + "loss": 0.5168, + "step": 870 + }, + { + "epoch": 2.723382045929019, + "grad_norm": 0.9011222124099731, + "learning_rate": 3.0834029064735636e-06, + "loss": 0.5163, + "step": 871 + }, + { + "epoch": 2.7265135699373695, + "grad_norm": 0.7811456322669983, + "learning_rate": 3.0791919265927827e-06, + "loss": 0.5004, + "step": 872 + }, + { + "epoch": 2.72964509394572, + "grad_norm": 0.9251837134361267, + "learning_rate": 3.0749792095210003e-06, + "loss": 0.5081, + "step": 873 + }, + { + "epoch": 2.732776617954071, + "grad_norm": 0.8347085118293762, + "learning_rate": 3.0707647678935695e-06, + "loss": 0.4793, + "step": 874 + }, + { + "epoch": 2.7359081419624216, + "grad_norm": 0.9766442179679871, + "learning_rate": 3.0665486143510153e-06, + "loss": 0.493, + "step": 875 + }, + { + "epoch": 2.7390396659707723, + "grad_norm": 0.7692548036575317, + "learning_rate": 3.0623307615389975e-06, + "loss": 0.4874, + "step": 876 + }, + { + "epoch": 2.742171189979123, + "grad_norm": 0.7714599370956421, + "learning_rate": 3.0581112221082727e-06, + "loss": 0.4929, + "step": 877 + }, + { + "epoch": 2.745302713987474, + "grad_norm": 0.7797786593437195, + "learning_rate": 3.053890008714655e-06, + "loss": 0.4359, + "step": 878 + }, + { + "epoch": 2.7484342379958244, + "grad_norm": 5.118397235870361, + "learning_rate": 3.049667134018981e-06, + "loss": 0.4634, + "step": 879 + }, + { + "epoch": 2.7515657620041756, + "grad_norm": 0.7684539556503296, + "learning_rate": 3.04544261068707e-06, + "loss": 0.4688, + "step": 880 + }, + { + "epoch": 2.754697286012526, + "grad_norm": 0.8678610920906067, + "learning_rate": 3.0412164513896846e-06, + "loss": 0.5213, + "step": 881 + }, + { + "epoch": 2.757828810020877, + "grad_norm": 0.80293869972229, + "learning_rate": 3.0369886688024954e-06, + "loss": 0.4392, + "step": 882 + }, + { + "epoch": 2.7609603340292277, + "grad_norm": 0.7438644766807556, + "learning_rate": 3.0327592756060412e-06, + "loss": 0.528, + "step": 883 + }, + { + "epoch": 2.7640918580375784, + "grad_norm": 0.7701645493507385, + "learning_rate": 3.0285282844856917e-06, + "loss": 0.504, + "step": 884 + }, + { + "epoch": 2.767223382045929, + "grad_norm": 0.7113856673240662, + "learning_rate": 3.024295708131611e-06, + "loss": 0.4819, + "step": 885 + }, + { + "epoch": 2.77035490605428, + "grad_norm": 1.2697532176971436, + "learning_rate": 3.020061559238714e-06, + "loss": 0.5009, + "step": 886 + }, + { + "epoch": 2.7734864300626305, + "grad_norm": 1.0299439430236816, + "learning_rate": 3.015825850506636e-06, + "loss": 0.4707, + "step": 887 + }, + { + "epoch": 2.776617954070981, + "grad_norm": 0.9703660607337952, + "learning_rate": 3.011588594639688e-06, + "loss": 0.4102, + "step": 888 + }, + { + "epoch": 2.779749478079332, + "grad_norm": 0.7357314825057983, + "learning_rate": 3.0073498043468247e-06, + "loss": 0.4649, + "step": 889 + }, + { + "epoch": 2.7828810020876826, + "grad_norm": 0.7815471291542053, + "learning_rate": 3.0031094923415993e-06, + "loss": 0.469, + "step": 890 + }, + { + "epoch": 2.7860125260960333, + "grad_norm": 0.7856019139289856, + "learning_rate": 2.9988676713421318e-06, + "loss": 0.4241, + "step": 891 + }, + { + "epoch": 2.789144050104384, + "grad_norm": 0.7668167352676392, + "learning_rate": 2.994624354071066e-06, + "loss": 0.5309, + "step": 892 + }, + { + "epoch": 2.7922755741127347, + "grad_norm": 0.7485945820808411, + "learning_rate": 2.990379553255535e-06, + "loss": 0.5173, + "step": 893 + }, + { + "epoch": 2.7954070981210855, + "grad_norm": 0.8065824508666992, + "learning_rate": 2.986133281627123e-06, + "loss": 0.4995, + "step": 894 + }, + { + "epoch": 2.798538622129436, + "grad_norm": 0.7156995534896851, + "learning_rate": 2.9818855519218217e-06, + "loss": 0.4642, + "step": 895 + }, + { + "epoch": 2.801670146137787, + "grad_norm": 0.9115403890609741, + "learning_rate": 2.97763637688e-06, + "loss": 0.4799, + "step": 896 + }, + { + "epoch": 2.804801670146138, + "grad_norm": 0.7466689944267273, + "learning_rate": 2.9733857692463584e-06, + "loss": 0.4942, + "step": 897 + }, + { + "epoch": 2.8079331941544883, + "grad_norm": 0.7484914064407349, + "learning_rate": 2.9691337417698974e-06, + "loss": 0.4618, + "step": 898 + }, + { + "epoch": 2.8110647181628394, + "grad_norm": 0.816704511642456, + "learning_rate": 2.9648803072038736e-06, + "loss": 0.4748, + "step": 899 + }, + { + "epoch": 2.81419624217119, + "grad_norm": 0.7627584934234619, + "learning_rate": 2.9606254783057666e-06, + "loss": 0.4667, + "step": 900 + }, + { + "epoch": 2.817327766179541, + "grad_norm": 0.7341011166572571, + "learning_rate": 2.9563692678372342e-06, + "loss": 0.4802, + "step": 901 + }, + { + "epoch": 2.8204592901878915, + "grad_norm": 1.2541382312774658, + "learning_rate": 2.952111688564082e-06, + "loss": 0.5231, + "step": 902 + }, + { + "epoch": 2.8235908141962422, + "grad_norm": 0.7172819375991821, + "learning_rate": 2.9478527532562184e-06, + "loss": 0.4488, + "step": 903 + }, + { + "epoch": 2.826722338204593, + "grad_norm": 0.774529218673706, + "learning_rate": 2.943592474687621e-06, + "loss": 0.4964, + "step": 904 + }, + { + "epoch": 2.8298538622129437, + "grad_norm": 0.7315672636032104, + "learning_rate": 2.939330865636294e-06, + "loss": 0.4817, + "step": 905 + }, + { + "epoch": 2.8329853862212944, + "grad_norm": 0.7698234915733337, + "learning_rate": 2.9350679388842347e-06, + "loss": 0.5075, + "step": 906 + }, + { + "epoch": 2.836116910229645, + "grad_norm": 0.7717766761779785, + "learning_rate": 2.93080370721739e-06, + "loss": 0.4789, + "step": 907 + }, + { + "epoch": 2.8392484342379958, + "grad_norm": 0.7383570075035095, + "learning_rate": 2.926538183425622e-06, + "loss": 0.4992, + "step": 908 + }, + { + "epoch": 2.8423799582463465, + "grad_norm": 0.7858864068984985, + "learning_rate": 2.92227138030267e-06, + "loss": 0.4993, + "step": 909 + }, + { + "epoch": 2.845511482254697, + "grad_norm": 0.8220369219779968, + "learning_rate": 2.9180033106461076e-06, + "loss": 0.4929, + "step": 910 + }, + { + "epoch": 2.848643006263048, + "grad_norm": 0.7507152557373047, + "learning_rate": 2.9137339872573086e-06, + "loss": 0.4394, + "step": 911 + }, + { + "epoch": 2.8517745302713986, + "grad_norm": 0.7935269474983215, + "learning_rate": 2.9094634229414063e-06, + "loss": 0.4656, + "step": 912 + }, + { + "epoch": 2.8549060542797493, + "grad_norm": 0.9187721610069275, + "learning_rate": 2.9051916305072576e-06, + "loss": 0.4918, + "step": 913 + }, + { + "epoch": 2.8580375782881005, + "grad_norm": 0.8699706792831421, + "learning_rate": 2.9009186227674e-06, + "loss": 0.5106, + "step": 914 + }, + { + "epoch": 2.8611691022964507, + "grad_norm": 0.7175673246383667, + "learning_rate": 2.896644412538021e-06, + "loss": 0.5105, + "step": 915 + }, + { + "epoch": 2.864300626304802, + "grad_norm": 0.8563990592956543, + "learning_rate": 2.892369012638909e-06, + "loss": 0.4993, + "step": 916 + }, + { + "epoch": 2.867432150313152, + "grad_norm": 0.7891882658004761, + "learning_rate": 2.8880924358934246e-06, + "loss": 0.4983, + "step": 917 + }, + { + "epoch": 2.8705636743215033, + "grad_norm": 0.9247110486030579, + "learning_rate": 2.8838146951284575e-06, + "loss": 0.4789, + "step": 918 + }, + { + "epoch": 2.873695198329854, + "grad_norm": 0.7523055672645569, + "learning_rate": 2.879535803174387e-06, + "loss": 0.4982, + "step": 919 + }, + { + "epoch": 2.8768267223382047, + "grad_norm": 0.8096909523010254, + "learning_rate": 2.8752557728650467e-06, + "loss": 0.4958, + "step": 920 + }, + { + "epoch": 2.8799582463465554, + "grad_norm": 2.3476874828338623, + "learning_rate": 2.870974617037684e-06, + "loss": 0.491, + "step": 921 + }, + { + "epoch": 2.883089770354906, + "grad_norm": 0.8388578295707703, + "learning_rate": 2.8666923485329224e-06, + "loss": 0.5275, + "step": 922 + }, + { + "epoch": 2.886221294363257, + "grad_norm": 0.8162729144096375, + "learning_rate": 2.8624089801947234e-06, + "loss": 0.4776, + "step": 923 + }, + { + "epoch": 2.8893528183716075, + "grad_norm": 0.7306103110313416, + "learning_rate": 2.858124524870345e-06, + "loss": 0.4814, + "step": 924 + }, + { + "epoch": 2.892484342379958, + "grad_norm": 0.8736817836761475, + "learning_rate": 2.853838995410307e-06, + "loss": 0.5097, + "step": 925 + }, + { + "epoch": 2.895615866388309, + "grad_norm": 0.7771823406219482, + "learning_rate": 2.8495524046683525e-06, + "loss": 0.4806, + "step": 926 + }, + { + "epoch": 2.8987473903966596, + "grad_norm": 0.9421334862709045, + "learning_rate": 2.845264765501404e-06, + "loss": 0.5055, + "step": 927 + }, + { + "epoch": 2.9018789144050103, + "grad_norm": 0.8403921127319336, + "learning_rate": 2.8409760907695314e-06, + "loss": 0.4775, + "step": 928 + }, + { + "epoch": 2.905010438413361, + "grad_norm": 0.8095362186431885, + "learning_rate": 2.836686393335909e-06, + "loss": 0.4532, + "step": 929 + }, + { + "epoch": 2.9081419624217117, + "grad_norm": 0.7340645790100098, + "learning_rate": 2.8323956860667813e-06, + "loss": 0.4835, + "step": 930 + }, + { + "epoch": 2.911273486430063, + "grad_norm": 0.6970911026000977, + "learning_rate": 2.828103981831417e-06, + "loss": 0.4999, + "step": 931 + }, + { + "epoch": 2.914405010438413, + "grad_norm": 0.8136418461799622, + "learning_rate": 2.8238112935020794e-06, + "loss": 0.5038, + "step": 932 + }, + { + "epoch": 2.9175365344467643, + "grad_norm": 0.9045608043670654, + "learning_rate": 2.8195176339539816e-06, + "loss": 0.486, + "step": 933 + }, + { + "epoch": 2.9206680584551146, + "grad_norm": 1.14940345287323, + "learning_rate": 2.815223016065249e-06, + "loss": 0.5079, + "step": 934 + }, + { + "epoch": 2.9237995824634657, + "grad_norm": 0.7411190867424011, + "learning_rate": 2.8109274527168826e-06, + "loss": 0.4564, + "step": 935 + }, + { + "epoch": 2.9269311064718164, + "grad_norm": 0.8903455138206482, + "learning_rate": 2.806630956792719e-06, + "loss": 0.451, + "step": 936 + }, + { + "epoch": 2.930062630480167, + "grad_norm": 0.7865445017814636, + "learning_rate": 2.8023335411793904e-06, + "loss": 0.4658, + "step": 937 + }, + { + "epoch": 2.933194154488518, + "grad_norm": 0.8185790777206421, + "learning_rate": 2.798035218766292e-06, + "loss": 0.4776, + "step": 938 + }, + { + "epoch": 2.9363256784968685, + "grad_norm": 0.7516276836395264, + "learning_rate": 2.793736002445531e-06, + "loss": 0.4447, + "step": 939 + }, + { + "epoch": 2.9394572025052192, + "grad_norm": 0.738080620765686, + "learning_rate": 2.789435905111903e-06, + "loss": 0.4832, + "step": 940 + }, + { + "epoch": 2.94258872651357, + "grad_norm": 0.7971507906913757, + "learning_rate": 2.785134939662843e-06, + "loss": 0.4835, + "step": 941 + }, + { + "epoch": 2.9457202505219207, + "grad_norm": 0.7529093623161316, + "learning_rate": 2.78083311899839e-06, + "loss": 0.4759, + "step": 942 + }, + { + "epoch": 2.9488517745302714, + "grad_norm": 0.8222358226776123, + "learning_rate": 2.7765304560211482e-06, + "loss": 0.4365, + "step": 943 + }, + { + "epoch": 2.951983298538622, + "grad_norm": 0.729945182800293, + "learning_rate": 2.7722269636362462e-06, + "loss": 0.5026, + "step": 944 + }, + { + "epoch": 2.9551148225469728, + "grad_norm": 0.7287900447845459, + "learning_rate": 2.767922654751306e-06, + "loss": 0.4916, + "step": 945 + }, + { + "epoch": 2.9582463465553235, + "grad_norm": 0.869637131690979, + "learning_rate": 2.763617542276391e-06, + "loss": 0.5018, + "step": 946 + }, + { + "epoch": 2.961377870563674, + "grad_norm": 1.004909634590149, + "learning_rate": 2.7593116391239806e-06, + "loss": 0.5152, + "step": 947 + }, + { + "epoch": 2.964509394572025, + "grad_norm": 0.8263046145439148, + "learning_rate": 2.7550049582089235e-06, + "loss": 0.5249, + "step": 948 + }, + { + "epoch": 2.9676409185803756, + "grad_norm": 0.7963895797729492, + "learning_rate": 2.750697512448401e-06, + "loss": 0.5084, + "step": 949 + }, + { + "epoch": 2.9707724425887267, + "grad_norm": 0.7211249470710754, + "learning_rate": 2.7463893147618893e-06, + "loss": 0.4691, + "step": 950 + }, + { + "epoch": 2.973903966597077, + "grad_norm": 0.8010216951370239, + "learning_rate": 2.742080378071118e-06, + "loss": 0.5026, + "step": 951 + }, + { + "epoch": 2.977035490605428, + "grad_norm": 0.780078649520874, + "learning_rate": 2.7377707153000356e-06, + "loss": 0.4758, + "step": 952 + }, + { + "epoch": 2.980167014613779, + "grad_norm": 0.7728193998336792, + "learning_rate": 2.7334603393747684e-06, + "loss": 0.488, + "step": 953 + }, + { + "epoch": 2.9832985386221296, + "grad_norm": 0.836329996585846, + "learning_rate": 2.7291492632235777e-06, + "loss": 0.456, + "step": 954 + }, + { + "epoch": 2.9864300626304803, + "grad_norm": 0.7241990566253662, + "learning_rate": 2.724837499776831e-06, + "loss": 0.4953, + "step": 955 + }, + { + "epoch": 2.989561586638831, + "grad_norm": 0.7595076560974121, + "learning_rate": 2.7205250619669527e-06, + "loss": 0.446, + "step": 956 + }, + { + "epoch": 2.9926931106471817, + "grad_norm": 0.8177686333656311, + "learning_rate": 2.716211962728392e-06, + "loss": 0.5057, + "step": 957 + }, + { + "epoch": 2.9958246346555324, + "grad_norm": 0.7506977915763855, + "learning_rate": 2.71189821499758e-06, + "loss": 0.4821, + "step": 958 + }, + { + "epoch": 2.998956158663883, + "grad_norm": 0.8085163235664368, + "learning_rate": 2.7075838317128943e-06, + "loss": 0.5002, + "step": 959 + }, + { + "epoch": 3.0, + "grad_norm": 3.3674418926239014, + "learning_rate": 2.7032688258146207e-06, + "loss": 0.183, + "step": 960 + }, + { + "epoch": 3.0031315240083507, + "grad_norm": 1.066100835800171, + "learning_rate": 2.698953210244908e-06, + "loss": 0.4427, + "step": 961 + }, + { + "epoch": 3.0062630480167014, + "grad_norm": 0.7690210938453674, + "learning_rate": 2.6946369979477365e-06, + "loss": 0.4638, + "step": 962 + }, + { + "epoch": 3.009394572025052, + "grad_norm": 0.8114679455757141, + "learning_rate": 2.690320201868876e-06, + "loss": 0.4373, + "step": 963 + }, + { + "epoch": 3.012526096033403, + "grad_norm": 0.7680971622467041, + "learning_rate": 2.686002834955847e-06, + "loss": 0.4477, + "step": 964 + }, + { + "epoch": 3.0156576200417535, + "grad_norm": 0.7194678783416748, + "learning_rate": 2.6816849101578808e-06, + "loss": 0.455, + "step": 965 + }, + { + "epoch": 3.0187891440501042, + "grad_norm": 0.6890467405319214, + "learning_rate": 2.6773664404258854e-06, + "loss": 0.4246, + "step": 966 + }, + { + "epoch": 3.021920668058455, + "grad_norm": 0.8064301013946533, + "learning_rate": 2.6730474387123987e-06, + "loss": 0.497, + "step": 967 + }, + { + "epoch": 3.0250521920668056, + "grad_norm": 0.8164849281311035, + "learning_rate": 2.668727917971559e-06, + "loss": 0.457, + "step": 968 + }, + { + "epoch": 3.028183716075157, + "grad_norm": 0.7793440818786621, + "learning_rate": 2.6644078911590565e-06, + "loss": 0.4632, + "step": 969 + }, + { + "epoch": 3.0313152400835075, + "grad_norm": 0.7208535671234131, + "learning_rate": 2.6600873712321033e-06, + "loss": 0.4786, + "step": 970 + }, + { + "epoch": 3.034446764091858, + "grad_norm": 0.8905500769615173, + "learning_rate": 2.655766371149389e-06, + "loss": 0.4317, + "step": 971 + }, + { + "epoch": 3.037578288100209, + "grad_norm": 0.7537338733673096, + "learning_rate": 2.6514449038710418e-06, + "loss": 0.4783, + "step": 972 + }, + { + "epoch": 3.0407098121085596, + "grad_norm": 0.7901656031608582, + "learning_rate": 2.6471229823585937e-06, + "loss": 0.4762, + "step": 973 + }, + { + "epoch": 3.0438413361169103, + "grad_norm": 0.7427340745925903, + "learning_rate": 2.6428006195749373e-06, + "loss": 0.4782, + "step": 974 + }, + { + "epoch": 3.046972860125261, + "grad_norm": 0.8364680409431458, + "learning_rate": 2.6384778284842905e-06, + "loss": 0.4551, + "step": 975 + }, + { + "epoch": 3.0501043841336117, + "grad_norm": 1.1049500703811646, + "learning_rate": 2.634154622052155e-06, + "loss": 0.4451, + "step": 976 + }, + { + "epoch": 3.0532359081419624, + "grad_norm": 1.0797414779663086, + "learning_rate": 2.6298310132452757e-06, + "loss": 0.4512, + "step": 977 + }, + { + "epoch": 3.056367432150313, + "grad_norm": 0.8832088708877563, + "learning_rate": 2.62550701503161e-06, + "loss": 0.4569, + "step": 978 + }, + { + "epoch": 3.059498956158664, + "grad_norm": 0.748951256275177, + "learning_rate": 2.621182640380277e-06, + "loss": 0.4535, + "step": 979 + }, + { + "epoch": 3.0626304801670146, + "grad_norm": 0.776386022567749, + "learning_rate": 2.616857902261529e-06, + "loss": 0.4582, + "step": 980 + }, + { + "epoch": 3.0657620041753653, + "grad_norm": 0.730057418346405, + "learning_rate": 2.6125328136467074e-06, + "loss": 0.4581, + "step": 981 + }, + { + "epoch": 3.068893528183716, + "grad_norm": 0.7918877005577087, + "learning_rate": 2.6082073875082046e-06, + "loss": 0.4355, + "step": 982 + }, + { + "epoch": 3.0720250521920667, + "grad_norm": 0.803987443447113, + "learning_rate": 2.6038816368194265e-06, + "loss": 0.455, + "step": 983 + }, + { + "epoch": 3.0751565762004174, + "grad_norm": 1.2970365285873413, + "learning_rate": 2.599555574554749e-06, + "loss": 0.448, + "step": 984 + }, + { + "epoch": 3.078288100208768, + "grad_norm": 0.7814194560050964, + "learning_rate": 2.595229213689487e-06, + "loss": 0.4197, + "step": 985 + }, + { + "epoch": 3.081419624217119, + "grad_norm": 0.9865803122520447, + "learning_rate": 2.5909025671998483e-06, + "loss": 0.4649, + "step": 986 + }, + { + "epoch": 3.08455114822547, + "grad_norm": 0.7769168615341187, + "learning_rate": 2.586575648062898e-06, + "loss": 0.4287, + "step": 987 + }, + { + "epoch": 3.0876826722338206, + "grad_norm": 1.454988956451416, + "learning_rate": 2.582248469256519e-06, + "loss": 0.4548, + "step": 988 + }, + { + "epoch": 3.0908141962421714, + "grad_norm": 0.7656151056289673, + "learning_rate": 2.577921043759372e-06, + "loss": 0.4323, + "step": 989 + }, + { + "epoch": 3.093945720250522, + "grad_norm": 0.7989768385887146, + "learning_rate": 2.5735933845508598e-06, + "loss": 0.4616, + "step": 990 + }, + { + "epoch": 3.0970772442588728, + "grad_norm": 0.8711255192756653, + "learning_rate": 2.5692655046110855e-06, + "loss": 0.4923, + "step": 991 + }, + { + "epoch": 3.1002087682672235, + "grad_norm": 0.7333446145057678, + "learning_rate": 2.564937416920813e-06, + "loss": 0.4297, + "step": 992 + }, + { + "epoch": 3.103340292275574, + "grad_norm": 0.7611051201820374, + "learning_rate": 2.5606091344614297e-06, + "loss": 0.4276, + "step": 993 + }, + { + "epoch": 3.106471816283925, + "grad_norm": 0.9742305278778076, + "learning_rate": 2.5562806702149083e-06, + "loss": 0.4597, + "step": 994 + }, + { + "epoch": 3.1096033402922756, + "grad_norm": 0.8043314218521118, + "learning_rate": 2.551952037163765e-06, + "loss": 0.459, + "step": 995 + }, + { + "epoch": 3.1127348643006263, + "grad_norm": 0.7217118740081787, + "learning_rate": 2.5476232482910253e-06, + "loss": 0.4514, + "step": 996 + }, + { + "epoch": 3.115866388308977, + "grad_norm": 1.1410573720932007, + "learning_rate": 2.5432943165801765e-06, + "loss": 0.4942, + "step": 997 + }, + { + "epoch": 3.1189979123173277, + "grad_norm": 0.7188895344734192, + "learning_rate": 2.5389652550151416e-06, + "loss": 0.4641, + "step": 998 + }, + { + "epoch": 3.1221294363256784, + "grad_norm": 0.7223486304283142, + "learning_rate": 2.5346360765802276e-06, + "loss": 0.4382, + "step": 999 + }, + { + "epoch": 3.125260960334029, + "grad_norm": 0.8269757032394409, + "learning_rate": 2.5303067942600933e-06, + "loss": 0.4502, + "step": 1000 + }, + { + "epoch": 3.12839248434238, + "grad_norm": 0.7081235647201538, + "learning_rate": 2.5259774210397104e-06, + "loss": 0.4389, + "step": 1001 + }, + { + "epoch": 3.1315240083507305, + "grad_norm": 0.7493525147438049, + "learning_rate": 2.5216479699043224e-06, + "loss": 0.4697, + "step": 1002 + }, + { + "epoch": 3.1346555323590812, + "grad_norm": 0.7633835077285767, + "learning_rate": 2.5173184538394054e-06, + "loss": 0.4535, + "step": 1003 + }, + { + "epoch": 3.137787056367432, + "grad_norm": 0.7765631079673767, + "learning_rate": 2.5129888858306333e-06, + "loss": 0.4616, + "step": 1004 + }, + { + "epoch": 3.140918580375783, + "grad_norm": 0.7026706337928772, + "learning_rate": 2.508659278863832e-06, + "loss": 0.4535, + "step": 1005 + }, + { + "epoch": 3.144050104384134, + "grad_norm": 0.8816738128662109, + "learning_rate": 2.5043296459249466e-06, + "loss": 0.4274, + "step": 1006 + }, + { + "epoch": 3.1471816283924845, + "grad_norm": 0.8072274327278137, + "learning_rate": 2.5e-06, + "loss": 0.4566, + "step": 1007 + }, + { + "epoch": 3.150313152400835, + "grad_norm": 0.7775781750679016, + "learning_rate": 2.4956703540750542e-06, + "loss": 0.4248, + "step": 1008 + }, + { + "epoch": 3.153444676409186, + "grad_norm": 0.9271306991577148, + "learning_rate": 2.4913407211361686e-06, + "loss": 0.4251, + "step": 1009 + }, + { + "epoch": 3.1565762004175366, + "grad_norm": 0.763867974281311, + "learning_rate": 2.487011114169368e-06, + "loss": 0.4525, + "step": 1010 + }, + { + "epoch": 3.1597077244258873, + "grad_norm": 0.8754820227622986, + "learning_rate": 2.4826815461605955e-06, + "loss": 0.4524, + "step": 1011 + }, + { + "epoch": 3.162839248434238, + "grad_norm": 0.8261513113975525, + "learning_rate": 2.4783520300956784e-06, + "loss": 0.4303, + "step": 1012 + }, + { + "epoch": 3.1659707724425887, + "grad_norm": 0.8629854917526245, + "learning_rate": 2.4740225789602905e-06, + "loss": 0.4628, + "step": 1013 + }, + { + "epoch": 3.1691022964509394, + "grad_norm": 0.7281518578529358, + "learning_rate": 2.469693205739907e-06, + "loss": 0.461, + "step": 1014 + }, + { + "epoch": 3.17223382045929, + "grad_norm": 0.757644772529602, + "learning_rate": 2.465363923419774e-06, + "loss": 0.4455, + "step": 1015 + }, + { + "epoch": 3.175365344467641, + "grad_norm": 0.8021314740180969, + "learning_rate": 2.4610347449848592e-06, + "loss": 0.4755, + "step": 1016 + }, + { + "epoch": 3.1784968684759916, + "grad_norm": 0.7467564940452576, + "learning_rate": 2.456705683419824e-06, + "loss": 0.4419, + "step": 1017 + }, + { + "epoch": 3.1816283924843423, + "grad_norm": 0.8575125336647034, + "learning_rate": 2.452376751708976e-06, + "loss": 0.4577, + "step": 1018 + }, + { + "epoch": 3.184759916492693, + "grad_norm": 0.8101590871810913, + "learning_rate": 2.448047962836235e-06, + "loss": 0.4546, + "step": 1019 + }, + { + "epoch": 3.1878914405010437, + "grad_norm": 0.8146190643310547, + "learning_rate": 2.443719329785093e-06, + "loss": 0.4219, + "step": 1020 + }, + { + "epoch": 3.1910229645093944, + "grad_norm": 1.2800556421279907, + "learning_rate": 2.4393908655385708e-06, + "loss": 0.4502, + "step": 1021 + }, + { + "epoch": 3.1941544885177455, + "grad_norm": 0.7362221479415894, + "learning_rate": 2.4350625830791875e-06, + "loss": 0.4802, + "step": 1022 + }, + { + "epoch": 3.1972860125260962, + "grad_norm": 0.8307201862335205, + "learning_rate": 2.430734495388915e-06, + "loss": 0.4808, + "step": 1023 + }, + { + "epoch": 3.200417536534447, + "grad_norm": 0.8467394709587097, + "learning_rate": 2.42640661544914e-06, + "loss": 0.4598, + "step": 1024 + }, + { + "epoch": 3.2035490605427976, + "grad_norm": 0.9570673108100891, + "learning_rate": 2.422078956240629e-06, + "loss": 0.4341, + "step": 1025 + }, + { + "epoch": 3.2066805845511483, + "grad_norm": 0.7976422309875488, + "learning_rate": 2.4177515307434824e-06, + "loss": 0.4285, + "step": 1026 + }, + { + "epoch": 3.209812108559499, + "grad_norm": 0.8241607546806335, + "learning_rate": 2.413424351937103e-06, + "loss": 0.4563, + "step": 1027 + }, + { + "epoch": 3.2129436325678498, + "grad_norm": 0.7191072702407837, + "learning_rate": 2.4090974328001526e-06, + "loss": 0.456, + "step": 1028 + }, + { + "epoch": 3.2160751565762005, + "grad_norm": 0.8112174868583679, + "learning_rate": 2.4047707863105133e-06, + "loss": 0.4592, + "step": 1029 + }, + { + "epoch": 3.219206680584551, + "grad_norm": 0.7170486450195312, + "learning_rate": 2.4004444254452522e-06, + "loss": 0.4559, + "step": 1030 + }, + { + "epoch": 3.222338204592902, + "grad_norm": 0.7092299461364746, + "learning_rate": 2.3961183631805748e-06, + "loss": 0.4504, + "step": 1031 + }, + { + "epoch": 3.2254697286012526, + "grad_norm": 0.8104662299156189, + "learning_rate": 2.391792612491796e-06, + "loss": 0.4327, + "step": 1032 + }, + { + "epoch": 3.2286012526096033, + "grad_norm": 0.8512858152389526, + "learning_rate": 2.387467186353293e-06, + "loss": 0.4506, + "step": 1033 + }, + { + "epoch": 3.231732776617954, + "grad_norm": 0.7987692952156067, + "learning_rate": 2.3831420977384715e-06, + "loss": 0.4257, + "step": 1034 + }, + { + "epoch": 3.2348643006263047, + "grad_norm": 0.7805537581443787, + "learning_rate": 2.3788173596197244e-06, + "loss": 0.4692, + "step": 1035 + }, + { + "epoch": 3.2379958246346554, + "grad_norm": 0.737304151058197, + "learning_rate": 2.374492984968392e-06, + "loss": 0.4308, + "step": 1036 + }, + { + "epoch": 3.241127348643006, + "grad_norm": 0.8113856315612793, + "learning_rate": 2.3701689867547247e-06, + "loss": 0.4668, + "step": 1037 + }, + { + "epoch": 3.244258872651357, + "grad_norm": 1.507103443145752, + "learning_rate": 2.3658453779478464e-06, + "loss": 0.4527, + "step": 1038 + }, + { + "epoch": 3.2473903966597075, + "grad_norm": 0.7973915338516235, + "learning_rate": 2.3615221715157095e-06, + "loss": 0.4741, + "step": 1039 + }, + { + "epoch": 3.2505219206680582, + "grad_norm": 0.8206940293312073, + "learning_rate": 2.3571993804250635e-06, + "loss": 0.4582, + "step": 1040 + }, + { + "epoch": 3.2536534446764094, + "grad_norm": 0.8519637584686279, + "learning_rate": 2.3528770176414076e-06, + "loss": 0.4412, + "step": 1041 + }, + { + "epoch": 3.25678496868476, + "grad_norm": 0.7542241811752319, + "learning_rate": 2.348555096128959e-06, + "loss": 0.4861, + "step": 1042 + }, + { + "epoch": 3.259916492693111, + "grad_norm": 0.8074842691421509, + "learning_rate": 2.3442336288506125e-06, + "loss": 0.4707, + "step": 1043 + }, + { + "epoch": 3.2630480167014615, + "grad_norm": 0.8164265751838684, + "learning_rate": 2.3399126287678975e-06, + "loss": 0.4417, + "step": 1044 + }, + { + "epoch": 3.266179540709812, + "grad_norm": 0.7689628005027771, + "learning_rate": 2.3355921088409435e-06, + "loss": 0.4815, + "step": 1045 + }, + { + "epoch": 3.269311064718163, + "grad_norm": 0.7709240913391113, + "learning_rate": 2.3312720820284423e-06, + "loss": 0.4444, + "step": 1046 + }, + { + "epoch": 3.2724425887265136, + "grad_norm": 0.7340330481529236, + "learning_rate": 2.326952561287602e-06, + "loss": 0.4694, + "step": 1047 + }, + { + "epoch": 3.2755741127348643, + "grad_norm": 0.7785805463790894, + "learning_rate": 2.3226335595741154e-06, + "loss": 0.4273, + "step": 1048 + }, + { + "epoch": 3.278705636743215, + "grad_norm": 0.9968108534812927, + "learning_rate": 2.3183150898421196e-06, + "loss": 0.4782, + "step": 1049 + }, + { + "epoch": 3.2818371607515657, + "grad_norm": 0.7823553681373596, + "learning_rate": 2.3139971650441533e-06, + "loss": 0.444, + "step": 1050 + }, + { + "epoch": 3.2849686847599164, + "grad_norm": 0.7317377924919128, + "learning_rate": 2.3096797981311252e-06, + "loss": 0.4692, + "step": 1051 + }, + { + "epoch": 3.288100208768267, + "grad_norm": 0.8546518683433533, + "learning_rate": 2.3053630020522643e-06, + "loss": 0.4632, + "step": 1052 + }, + { + "epoch": 3.291231732776618, + "grad_norm": 1.2284396886825562, + "learning_rate": 2.301046789755093e-06, + "loss": 0.4409, + "step": 1053 + }, + { + "epoch": 3.2943632567849686, + "grad_norm": 0.8000460863113403, + "learning_rate": 2.2967311741853797e-06, + "loss": 0.4456, + "step": 1054 + }, + { + "epoch": 3.2974947807933193, + "grad_norm": 0.7689793109893799, + "learning_rate": 2.2924161682871053e-06, + "loss": 0.45, + "step": 1055 + }, + { + "epoch": 3.30062630480167, + "grad_norm": 0.8032956719398499, + "learning_rate": 2.288101785002421e-06, + "loss": 0.4817, + "step": 1056 + }, + { + "epoch": 3.3037578288100207, + "grad_norm": 0.6831309795379639, + "learning_rate": 2.283788037271609e-06, + "loss": 0.4502, + "step": 1057 + }, + { + "epoch": 3.306889352818372, + "grad_norm": 0.8581221103668213, + "learning_rate": 2.279474938033048e-06, + "loss": 0.4569, + "step": 1058 + }, + { + "epoch": 3.3100208768267225, + "grad_norm": 0.7937221527099609, + "learning_rate": 2.2751625002231696e-06, + "loss": 0.4451, + "step": 1059 + }, + { + "epoch": 3.3131524008350732, + "grad_norm": 0.8095264434814453, + "learning_rate": 2.270850736776422e-06, + "loss": 0.4462, + "step": 1060 + }, + { + "epoch": 3.316283924843424, + "grad_norm": 0.9141370058059692, + "learning_rate": 2.2665396606252332e-06, + "loss": 0.419, + "step": 1061 + }, + { + "epoch": 3.3194154488517746, + "grad_norm": 0.8648553490638733, + "learning_rate": 2.262229284699965e-06, + "loss": 0.4562, + "step": 1062 + }, + { + "epoch": 3.3225469728601253, + "grad_norm": 0.7716917395591736, + "learning_rate": 2.2579196219288825e-06, + "loss": 0.4734, + "step": 1063 + }, + { + "epoch": 3.325678496868476, + "grad_norm": 0.8074535727500916, + "learning_rate": 2.2536106852381116e-06, + "loss": 0.4272, + "step": 1064 + }, + { + "epoch": 3.3288100208768268, + "grad_norm": 0.8989127278327942, + "learning_rate": 2.249302487551599e-06, + "loss": 0.4728, + "step": 1065 + }, + { + "epoch": 3.3319415448851775, + "grad_norm": 0.7662765383720398, + "learning_rate": 2.2449950417910777e-06, + "loss": 0.4457, + "step": 1066 + }, + { + "epoch": 3.335073068893528, + "grad_norm": 0.7789275050163269, + "learning_rate": 2.24068836087602e-06, + "loss": 0.3919, + "step": 1067 + }, + { + "epoch": 3.338204592901879, + "grad_norm": 0.7359098196029663, + "learning_rate": 2.2363824577236097e-06, + "loss": 0.4876, + "step": 1068 + }, + { + "epoch": 3.3413361169102296, + "grad_norm": 0.9960948824882507, + "learning_rate": 2.232077345248695e-06, + "loss": 0.4894, + "step": 1069 + }, + { + "epoch": 3.3444676409185803, + "grad_norm": 0.7446064352989197, + "learning_rate": 2.2277730363637537e-06, + "loss": 0.4765, + "step": 1070 + }, + { + "epoch": 3.347599164926931, + "grad_norm": 0.7674328088760376, + "learning_rate": 2.2234695439788534e-06, + "loss": 0.4468, + "step": 1071 + }, + { + "epoch": 3.3507306889352817, + "grad_norm": 0.8957347869873047, + "learning_rate": 2.2191668810016105e-06, + "loss": 0.4733, + "step": 1072 + }, + { + "epoch": 3.3538622129436324, + "grad_norm": 0.9110277891159058, + "learning_rate": 2.2148650603371573e-06, + "loss": 0.4399, + "step": 1073 + }, + { + "epoch": 3.356993736951983, + "grad_norm": 0.9158220291137695, + "learning_rate": 2.2105640948880976e-06, + "loss": 0.4609, + "step": 1074 + }, + { + "epoch": 3.3601252609603343, + "grad_norm": 0.7630184888839722, + "learning_rate": 2.206263997554469e-06, + "loss": 0.4674, + "step": 1075 + }, + { + "epoch": 3.3632567849686845, + "grad_norm": 0.7975273728370667, + "learning_rate": 2.20196478123371e-06, + "loss": 0.4478, + "step": 1076 + }, + { + "epoch": 3.3663883089770357, + "grad_norm": 0.8825351595878601, + "learning_rate": 2.19766645882061e-06, + "loss": 0.4687, + "step": 1077 + }, + { + "epoch": 3.3695198329853864, + "grad_norm": 0.8907671570777893, + "learning_rate": 2.1933690432072817e-06, + "loss": 0.4223, + "step": 1078 + }, + { + "epoch": 3.372651356993737, + "grad_norm": 0.7449545860290527, + "learning_rate": 2.189072547283118e-06, + "loss": 0.4595, + "step": 1079 + }, + { + "epoch": 3.375782881002088, + "grad_norm": 0.8460972309112549, + "learning_rate": 2.184776983934751e-06, + "loss": 0.4443, + "step": 1080 + }, + { + "epoch": 3.3789144050104385, + "grad_norm": 0.7524845600128174, + "learning_rate": 2.1804823660460196e-06, + "loss": 0.4235, + "step": 1081 + }, + { + "epoch": 3.382045929018789, + "grad_norm": 0.8448389768600464, + "learning_rate": 2.176188706497921e-06, + "loss": 0.4387, + "step": 1082 + }, + { + "epoch": 3.38517745302714, + "grad_norm": 0.7701981663703918, + "learning_rate": 2.1718960181685838e-06, + "loss": 0.4257, + "step": 1083 + }, + { + "epoch": 3.3883089770354906, + "grad_norm": 0.8178983330726624, + "learning_rate": 2.167604313933219e-06, + "loss": 0.4983, + "step": 1084 + }, + { + "epoch": 3.3914405010438413, + "grad_norm": 0.7477235198020935, + "learning_rate": 2.163313606664091e-06, + "loss": 0.4559, + "step": 1085 + }, + { + "epoch": 3.394572025052192, + "grad_norm": 0.8127962350845337, + "learning_rate": 2.1590239092304694e-06, + "loss": 0.453, + "step": 1086 + }, + { + "epoch": 3.3977035490605427, + "grad_norm": 0.7462339997291565, + "learning_rate": 2.1547352344985966e-06, + "loss": 0.4697, + "step": 1087 + }, + { + "epoch": 3.4008350730688934, + "grad_norm": 0.9641384482383728, + "learning_rate": 2.1504475953316483e-06, + "loss": 0.4495, + "step": 1088 + }, + { + "epoch": 3.403966597077244, + "grad_norm": 0.7612512707710266, + "learning_rate": 2.146161004589693e-06, + "loss": 0.4579, + "step": 1089 + }, + { + "epoch": 3.407098121085595, + "grad_norm": 0.7547829747200012, + "learning_rate": 2.141875475129655e-06, + "loss": 0.4334, + "step": 1090 + }, + { + "epoch": 3.4102296450939455, + "grad_norm": 0.8036953806877136, + "learning_rate": 2.137591019805278e-06, + "loss": 0.4466, + "step": 1091 + }, + { + "epoch": 3.4133611691022967, + "grad_norm": 0.7319284081459045, + "learning_rate": 2.1333076514670784e-06, + "loss": 0.4942, + "step": 1092 + }, + { + "epoch": 3.416492693110647, + "grad_norm": 0.8278589248657227, + "learning_rate": 2.1290253829623165e-06, + "loss": 0.4554, + "step": 1093 + }, + { + "epoch": 3.419624217118998, + "grad_norm": 0.733059287071228, + "learning_rate": 2.124744227134954e-06, + "loss": 0.4187, + "step": 1094 + }, + { + "epoch": 3.422755741127349, + "grad_norm": 0.8222727179527283, + "learning_rate": 2.1204641968256136e-06, + "loss": 0.4587, + "step": 1095 + }, + { + "epoch": 3.4258872651356995, + "grad_norm": 0.8296732902526855, + "learning_rate": 2.1161853048715438e-06, + "loss": 0.4868, + "step": 1096 + }, + { + "epoch": 3.4290187891440502, + "grad_norm": 0.7309690713882446, + "learning_rate": 2.1119075641065758e-06, + "loss": 0.4594, + "step": 1097 + }, + { + "epoch": 3.432150313152401, + "grad_norm": 1.4901788234710693, + "learning_rate": 2.1076309873610916e-06, + "loss": 0.4216, + "step": 1098 + }, + { + "epoch": 3.4352818371607516, + "grad_norm": 0.7993581891059875, + "learning_rate": 2.1033555874619794e-06, + "loss": 0.4842, + "step": 1099 + }, + { + "epoch": 3.4384133611691023, + "grad_norm": 0.8846752643585205, + "learning_rate": 2.0990813772325995e-06, + "loss": 0.4395, + "step": 1100 + }, + { + "epoch": 3.441544885177453, + "grad_norm": 1.0796778202056885, + "learning_rate": 2.0948083694927436e-06, + "loss": 0.4573, + "step": 1101 + }, + { + "epoch": 3.4446764091858038, + "grad_norm": 0.74623042345047, + "learning_rate": 2.090536577058595e-06, + "loss": 0.4563, + "step": 1102 + }, + { + "epoch": 3.4478079331941545, + "grad_norm": 0.8245521783828735, + "learning_rate": 2.086266012742692e-06, + "loss": 0.4433, + "step": 1103 + }, + { + "epoch": 3.450939457202505, + "grad_norm": 0.8003777265548706, + "learning_rate": 2.081996689353893e-06, + "loss": 0.4599, + "step": 1104 + }, + { + "epoch": 3.454070981210856, + "grad_norm": 0.8309001922607422, + "learning_rate": 2.0777286196973302e-06, + "loss": 0.485, + "step": 1105 + }, + { + "epoch": 3.4572025052192066, + "grad_norm": 0.8299122452735901, + "learning_rate": 2.0734618165743782e-06, + "loss": 0.4685, + "step": 1106 + }, + { + "epoch": 3.4603340292275573, + "grad_norm": 0.9347029328346252, + "learning_rate": 2.069196292782611e-06, + "loss": 0.4615, + "step": 1107 + }, + { + "epoch": 3.463465553235908, + "grad_norm": 0.7146593332290649, + "learning_rate": 2.064932061115766e-06, + "loss": 0.4433, + "step": 1108 + }, + { + "epoch": 3.4665970772442587, + "grad_norm": 0.7674420475959778, + "learning_rate": 2.0606691343637063e-06, + "loss": 0.4444, + "step": 1109 + }, + { + "epoch": 3.4697286012526094, + "grad_norm": 0.7925504446029663, + "learning_rate": 2.05640752531238e-06, + "loss": 0.4631, + "step": 1110 + }, + { + "epoch": 3.4728601252609606, + "grad_norm": 0.7755677700042725, + "learning_rate": 2.0521472467437825e-06, + "loss": 0.4709, + "step": 1111 + }, + { + "epoch": 3.4759916492693113, + "grad_norm": 0.8535795211791992, + "learning_rate": 2.0478883114359187e-06, + "loss": 0.442, + "step": 1112 + }, + { + "epoch": 3.479123173277662, + "grad_norm": 0.723953127861023, + "learning_rate": 2.043630732162767e-06, + "loss": 0.4782, + "step": 1113 + }, + { + "epoch": 3.4822546972860127, + "grad_norm": 0.7817316651344299, + "learning_rate": 2.0393745216942343e-06, + "loss": 0.4841, + "step": 1114 + }, + { + "epoch": 3.4853862212943634, + "grad_norm": 0.8878781795501709, + "learning_rate": 2.0351196927961268e-06, + "loss": 0.4673, + "step": 1115 + }, + { + "epoch": 3.488517745302714, + "grad_norm": 0.8645241856575012, + "learning_rate": 2.030866258230104e-06, + "loss": 0.432, + "step": 1116 + }, + { + "epoch": 3.491649269311065, + "grad_norm": 0.7294583320617676, + "learning_rate": 2.026614230753643e-06, + "loss": 0.4683, + "step": 1117 + }, + { + "epoch": 3.4947807933194155, + "grad_norm": 0.7412407994270325, + "learning_rate": 2.022363623120001e-06, + "loss": 0.4523, + "step": 1118 + }, + { + "epoch": 3.497912317327766, + "grad_norm": 0.8559291362762451, + "learning_rate": 2.0181144480781787e-06, + "loss": 0.4309, + "step": 1119 + }, + { + "epoch": 3.501043841336117, + "grad_norm": 0.7442825436592102, + "learning_rate": 2.0138667183728775e-06, + "loss": 0.4096, + "step": 1120 + }, + { + "epoch": 3.5041753653444676, + "grad_norm": 0.7605662941932678, + "learning_rate": 2.0096204467444645e-06, + "loss": 0.4404, + "step": 1121 + }, + { + "epoch": 3.5073068893528183, + "grad_norm": 0.7984277009963989, + "learning_rate": 2.005375645928935e-06, + "loss": 0.4661, + "step": 1122 + }, + { + "epoch": 3.510438413361169, + "grad_norm": 1.1044552326202393, + "learning_rate": 2.001132328657869e-06, + "loss": 0.4185, + "step": 1123 + }, + { + "epoch": 3.5135699373695197, + "grad_norm": 0.8210328817367554, + "learning_rate": 1.996890507658401e-06, + "loss": 0.4746, + "step": 1124 + }, + { + "epoch": 3.5167014613778704, + "grad_norm": 0.7302148342132568, + "learning_rate": 1.9926501956531758e-06, + "loss": 0.4333, + "step": 1125 + }, + { + "epoch": 3.519832985386221, + "grad_norm": 0.7713826894760132, + "learning_rate": 1.9884114053603114e-06, + "loss": 0.4485, + "step": 1126 + }, + { + "epoch": 3.522964509394572, + "grad_norm": 0.7386549711227417, + "learning_rate": 1.984174149493365e-06, + "loss": 0.4678, + "step": 1127 + }, + { + "epoch": 3.526096033402923, + "grad_norm": 0.8006004095077515, + "learning_rate": 1.979938440761287e-06, + "loss": 0.4755, + "step": 1128 + }, + { + "epoch": 3.5292275574112733, + "grad_norm": 1.0635333061218262, + "learning_rate": 1.97570429186839e-06, + "loss": 0.4985, + "step": 1129 + }, + { + "epoch": 3.5323590814196244, + "grad_norm": 0.7470075488090515, + "learning_rate": 1.9714717155143083e-06, + "loss": 0.4307, + "step": 1130 + }, + { + "epoch": 3.535490605427975, + "grad_norm": 0.8314558863639832, + "learning_rate": 1.967240724393959e-06, + "loss": 0.4894, + "step": 1131 + }, + { + "epoch": 3.538622129436326, + "grad_norm": 0.761443018913269, + "learning_rate": 1.963011331197506e-06, + "loss": 0.4653, + "step": 1132 + }, + { + "epoch": 3.5417536534446765, + "grad_norm": 0.7483212351799011, + "learning_rate": 1.9587835486103163e-06, + "loss": 0.4456, + "step": 1133 + }, + { + "epoch": 3.5448851774530272, + "grad_norm": 0.7794159054756165, + "learning_rate": 1.9545573893129306e-06, + "loss": 0.4707, + "step": 1134 + }, + { + "epoch": 3.548016701461378, + "grad_norm": 0.8016185760498047, + "learning_rate": 1.950332865981019e-06, + "loss": 0.4547, + "step": 1135 + }, + { + "epoch": 3.5511482254697286, + "grad_norm": 0.8089869618415833, + "learning_rate": 1.9461099912853453e-06, + "loss": 0.4499, + "step": 1136 + }, + { + "epoch": 3.5542797494780793, + "grad_norm": 0.7774782180786133, + "learning_rate": 1.9418887778917286e-06, + "loss": 0.4531, + "step": 1137 + }, + { + "epoch": 3.55741127348643, + "grad_norm": 0.7793645262718201, + "learning_rate": 1.937669238461003e-06, + "loss": 0.4553, + "step": 1138 + }, + { + "epoch": 3.5605427974947808, + "grad_norm": 0.8139959573745728, + "learning_rate": 1.933451385648985e-06, + "loss": 0.458, + "step": 1139 + }, + { + "epoch": 3.5636743215031315, + "grad_norm": 0.7517053484916687, + "learning_rate": 1.929235232106431e-06, + "loss": 0.4779, + "step": 1140 + }, + { + "epoch": 3.566805845511482, + "grad_norm": 0.8851562142372131, + "learning_rate": 1.925020790479e-06, + "loss": 0.425, + "step": 1141 + }, + { + "epoch": 3.569937369519833, + "grad_norm": 0.8129401803016663, + "learning_rate": 1.920808073407218e-06, + "loss": 0.4616, + "step": 1142 + }, + { + "epoch": 3.5730688935281836, + "grad_norm": 0.7110117077827454, + "learning_rate": 1.916597093526437e-06, + "loss": 0.4748, + "step": 1143 + }, + { + "epoch": 3.5762004175365343, + "grad_norm": 0.8268555402755737, + "learning_rate": 1.912387863466798e-06, + "loss": 0.4752, + "step": 1144 + }, + { + "epoch": 3.5793319415448854, + "grad_norm": 1.1036733388900757, + "learning_rate": 1.9081803958531967e-06, + "loss": 0.4879, + "step": 1145 + }, + { + "epoch": 3.5824634655532357, + "grad_norm": 0.8561109304428101, + "learning_rate": 1.9039747033052395e-06, + "loss": 0.4409, + "step": 1146 + }, + { + "epoch": 3.585594989561587, + "grad_norm": 0.7597541809082031, + "learning_rate": 1.8997707984372119e-06, + "loss": 0.4518, + "step": 1147 + }, + { + "epoch": 3.588726513569937, + "grad_norm": 0.7225353121757507, + "learning_rate": 1.8955686938580329e-06, + "loss": 0.4735, + "step": 1148 + }, + { + "epoch": 3.5918580375782883, + "grad_norm": 0.9409791231155396, + "learning_rate": 1.8913684021712264e-06, + "loss": 0.4249, + "step": 1149 + }, + { + "epoch": 3.594989561586639, + "grad_norm": 0.8757275342941284, + "learning_rate": 1.8871699359748763e-06, + "loss": 0.4479, + "step": 1150 + }, + { + "epoch": 3.5981210855949897, + "grad_norm": 0.8090003728866577, + "learning_rate": 1.882973307861593e-06, + "loss": 0.4846, + "step": 1151 + }, + { + "epoch": 3.6012526096033404, + "grad_norm": 0.7568825483322144, + "learning_rate": 1.8787785304184726e-06, + "loss": 0.4301, + "step": 1152 + }, + { + "epoch": 3.604384133611691, + "grad_norm": 0.8233815431594849, + "learning_rate": 1.8745856162270592e-06, + "loss": 0.4838, + "step": 1153 + }, + { + "epoch": 3.607515657620042, + "grad_norm": 0.7817628979682922, + "learning_rate": 1.8703945778633121e-06, + "loss": 0.4669, + "step": 1154 + }, + { + "epoch": 3.6106471816283925, + "grad_norm": 0.821304202079773, + "learning_rate": 1.8662054278975605e-06, + "loss": 0.4536, + "step": 1155 + }, + { + "epoch": 3.613778705636743, + "grad_norm": 0.9304089546203613, + "learning_rate": 1.8620181788944712e-06, + "loss": 0.4489, + "step": 1156 + }, + { + "epoch": 3.616910229645094, + "grad_norm": 0.8202670216560364, + "learning_rate": 1.8578328434130114e-06, + "loss": 0.4309, + "step": 1157 + }, + { + "epoch": 3.6200417536534446, + "grad_norm": 0.8890257477760315, + "learning_rate": 1.8536494340064051e-06, + "loss": 0.4736, + "step": 1158 + }, + { + "epoch": 3.6231732776617953, + "grad_norm": 0.7940590381622314, + "learning_rate": 1.8494679632221013e-06, + "loss": 0.4468, + "step": 1159 + }, + { + "epoch": 3.626304801670146, + "grad_norm": 0.8388273119926453, + "learning_rate": 1.845288443601736e-06, + "loss": 0.4753, + "step": 1160 + }, + { + "epoch": 3.6294363256784967, + "grad_norm": 0.9392285346984863, + "learning_rate": 1.84111088768109e-06, + "loss": 0.4736, + "step": 1161 + }, + { + "epoch": 3.632567849686848, + "grad_norm": 0.7411681413650513, + "learning_rate": 1.8369353079900576e-06, + "loss": 0.4303, + "step": 1162 + }, + { + "epoch": 3.635699373695198, + "grad_norm": 0.8722569942474365, + "learning_rate": 1.8327617170526014e-06, + "loss": 0.4604, + "step": 1163 + }, + { + "epoch": 3.6388308977035493, + "grad_norm": 0.786891758441925, + "learning_rate": 1.8285901273867229e-06, + "loss": 0.4756, + "step": 1164 + }, + { + "epoch": 3.6419624217118995, + "grad_norm": 0.8159083724021912, + "learning_rate": 1.824420551504419e-06, + "loss": 0.4675, + "step": 1165 + }, + { + "epoch": 3.6450939457202507, + "grad_norm": 0.8271334767341614, + "learning_rate": 1.8202530019116487e-06, + "loss": 0.4311, + "step": 1166 + }, + { + "epoch": 3.6482254697286014, + "grad_norm": 0.7617189288139343, + "learning_rate": 1.816087491108292e-06, + "loss": 0.4522, + "step": 1167 + }, + { + "epoch": 3.651356993736952, + "grad_norm": 0.7248172760009766, + "learning_rate": 1.8119240315881126e-06, + "loss": 0.465, + "step": 1168 + }, + { + "epoch": 3.654488517745303, + "grad_norm": 0.8606911897659302, + "learning_rate": 1.8077626358387235e-06, + "loss": 0.4524, + "step": 1169 + }, + { + "epoch": 3.6576200417536535, + "grad_norm": 0.8571308851242065, + "learning_rate": 1.8036033163415484e-06, + "loss": 0.4625, + "step": 1170 + }, + { + "epoch": 3.6607515657620042, + "grad_norm": 0.7239511609077454, + "learning_rate": 1.7994460855717812e-06, + "loss": 0.5025, + "step": 1171 + }, + { + "epoch": 3.663883089770355, + "grad_norm": 0.7958929538726807, + "learning_rate": 1.7952909559983544e-06, + "loss": 0.4382, + "step": 1172 + }, + { + "epoch": 3.6670146137787056, + "grad_norm": 0.7920124530792236, + "learning_rate": 1.7911379400838947e-06, + "loss": 0.4393, + "step": 1173 + }, + { + "epoch": 3.6701461377870563, + "grad_norm": 0.8072578310966492, + "learning_rate": 1.7869870502846903e-06, + "loss": 0.4627, + "step": 1174 + }, + { + "epoch": 3.673277661795407, + "grad_norm": 0.8586218357086182, + "learning_rate": 1.7828382990506543e-06, + "loss": 0.4456, + "step": 1175 + }, + { + "epoch": 3.6764091858037578, + "grad_norm": 0.8741613030433655, + "learning_rate": 1.7786916988252845e-06, + "loss": 0.4613, + "step": 1176 + }, + { + "epoch": 3.6795407098121085, + "grad_norm": 0.7691352367401123, + "learning_rate": 1.774547262045626e-06, + "loss": 0.4641, + "step": 1177 + }, + { + "epoch": 3.682672233820459, + "grad_norm": 0.7866089940071106, + "learning_rate": 1.7704050011422357e-06, + "loss": 0.4308, + "step": 1178 + }, + { + "epoch": 3.68580375782881, + "grad_norm": 0.9934884309768677, + "learning_rate": 1.7662649285391447e-06, + "loss": 0.4434, + "step": 1179 + }, + { + "epoch": 3.6889352818371606, + "grad_norm": 0.794385552406311, + "learning_rate": 1.7621270566538204e-06, + "loss": 0.4481, + "step": 1180 + }, + { + "epoch": 3.6920668058455117, + "grad_norm": 0.7573548555374146, + "learning_rate": 1.7579913978971296e-06, + "loss": 0.4525, + "step": 1181 + }, + { + "epoch": 3.695198329853862, + "grad_norm": 0.7073976993560791, + "learning_rate": 1.7538579646733023e-06, + "loss": 0.4214, + "step": 1182 + }, + { + "epoch": 3.698329853862213, + "grad_norm": 0.8009579181671143, + "learning_rate": 1.7497267693798902e-06, + "loss": 0.4542, + "step": 1183 + }, + { + "epoch": 3.701461377870564, + "grad_norm": 1.4488778114318848, + "learning_rate": 1.7455978244077348e-06, + "loss": 0.443, + "step": 1184 + }, + { + "epoch": 3.7045929018789145, + "grad_norm": 1.0529266595840454, + "learning_rate": 1.7414711421409292e-06, + "loss": 0.4908, + "step": 1185 + }, + { + "epoch": 3.7077244258872653, + "grad_norm": 0.757431149482727, + "learning_rate": 1.7373467349567775e-06, + "loss": 0.4454, + "step": 1186 + }, + { + "epoch": 3.710855949895616, + "grad_norm": 0.7086379528045654, + "learning_rate": 1.733224615225763e-06, + "loss": 0.4292, + "step": 1187 + }, + { + "epoch": 3.7139874739039667, + "grad_norm": 0.7454110383987427, + "learning_rate": 1.7291047953115049e-06, + "loss": 0.4421, + "step": 1188 + }, + { + "epoch": 3.7171189979123174, + "grad_norm": 0.804027795791626, + "learning_rate": 1.7249872875707257e-06, + "loss": 0.4819, + "step": 1189 + }, + { + "epoch": 3.720250521920668, + "grad_norm": 0.8159645199775696, + "learning_rate": 1.7208721043532146e-06, + "loss": 0.4628, + "step": 1190 + }, + { + "epoch": 3.723382045929019, + "grad_norm": 0.8451672196388245, + "learning_rate": 1.7167592580017866e-06, + "loss": 0.4773, + "step": 1191 + }, + { + "epoch": 3.7265135699373695, + "grad_norm": 0.920553982257843, + "learning_rate": 1.7126487608522492e-06, + "loss": 0.4787, + "step": 1192 + }, + { + "epoch": 3.72964509394572, + "grad_norm": 0.9169708490371704, + "learning_rate": 1.7085406252333613e-06, + "loss": 0.4543, + "step": 1193 + }, + { + "epoch": 3.732776617954071, + "grad_norm": 0.7245096564292908, + "learning_rate": 1.7044348634668023e-06, + "loss": 0.4599, + "step": 1194 + }, + { + "epoch": 3.7359081419624216, + "grad_norm": 0.835832417011261, + "learning_rate": 1.7003314878671284e-06, + "loss": 0.4794, + "step": 1195 + }, + { + "epoch": 3.7390396659707723, + "grad_norm": 0.8455896973609924, + "learning_rate": 1.696230510741742e-06, + "loss": 0.4329, + "step": 1196 + }, + { + "epoch": 3.742171189979123, + "grad_norm": 0.743611752986908, + "learning_rate": 1.692131944390849e-06, + "loss": 0.4455, + "step": 1197 + }, + { + "epoch": 3.745302713987474, + "grad_norm": 1.1334915161132812, + "learning_rate": 1.6880358011074272e-06, + "loss": 0.4673, + "step": 1198 + }, + { + "epoch": 3.7484342379958244, + "grad_norm": 0.6935724020004272, + "learning_rate": 1.6839420931771828e-06, + "loss": 0.4686, + "step": 1199 + }, + { + "epoch": 3.7515657620041756, + "grad_norm": 0.962006151676178, + "learning_rate": 1.6798508328785213e-06, + "loss": 0.4474, + "step": 1200 + }, + { + "epoch": 3.754697286012526, + "grad_norm": 1.1499714851379395, + "learning_rate": 1.6757620324825047e-06, + "loss": 0.4659, + "step": 1201 + }, + { + "epoch": 3.757828810020877, + "grad_norm": 0.7689645886421204, + "learning_rate": 1.6716757042528192e-06, + "loss": 0.4552, + "step": 1202 + }, + { + "epoch": 3.7609603340292277, + "grad_norm": 0.7482030391693115, + "learning_rate": 1.6675918604457352e-06, + "loss": 0.4747, + "step": 1203 + }, + { + "epoch": 3.7640918580375784, + "grad_norm": 0.7727032899856567, + "learning_rate": 1.6635105133100686e-06, + "loss": 0.4508, + "step": 1204 + }, + { + "epoch": 3.767223382045929, + "grad_norm": 0.8722149133682251, + "learning_rate": 1.6594316750871514e-06, + "loss": 0.4685, + "step": 1205 + }, + { + "epoch": 3.77035490605428, + "grad_norm": 0.771304726600647, + "learning_rate": 1.6553553580107884e-06, + "loss": 0.4418, + "step": 1206 + }, + { + "epoch": 3.7734864300626305, + "grad_norm": 0.767315149307251, + "learning_rate": 1.6512815743072214e-06, + "loss": 0.4532, + "step": 1207 + }, + { + "epoch": 3.776617954070981, + "grad_norm": 0.8825518488883972, + "learning_rate": 1.6472103361950976e-06, + "loss": 0.468, + "step": 1208 + }, + { + "epoch": 3.779749478079332, + "grad_norm": 0.8887981176376343, + "learning_rate": 1.6431416558854243e-06, + "loss": 0.4264, + "step": 1209 + }, + { + "epoch": 3.7828810020876826, + "grad_norm": 0.8399733304977417, + "learning_rate": 1.63907554558154e-06, + "loss": 0.4405, + "step": 1210 + }, + { + "epoch": 3.7860125260960333, + "grad_norm": 0.8112586140632629, + "learning_rate": 1.6350120174790751e-06, + "loss": 0.445, + "step": 1211 + }, + { + "epoch": 3.789144050104384, + "grad_norm": 0.860775351524353, + "learning_rate": 1.6309510837659137e-06, + "loss": 0.4557, + "step": 1212 + }, + { + "epoch": 3.7922755741127347, + "grad_norm": 0.8522343039512634, + "learning_rate": 1.626892756622161e-06, + "loss": 0.481, + "step": 1213 + }, + { + "epoch": 3.7954070981210855, + "grad_norm": 0.7927511930465698, + "learning_rate": 1.6228370482200988e-06, + "loss": 0.4328, + "step": 1214 + }, + { + "epoch": 3.798538622129436, + "grad_norm": 0.7350064516067505, + "learning_rate": 1.6187839707241604e-06, + "loss": 0.4604, + "step": 1215 + }, + { + "epoch": 3.801670146137787, + "grad_norm": 0.8363698124885559, + "learning_rate": 1.6147335362908847e-06, + "loss": 0.4271, + "step": 1216 + }, + { + "epoch": 3.804801670146138, + "grad_norm": 1.080613613128662, + "learning_rate": 1.610685757068885e-06, + "loss": 0.447, + "step": 1217 + }, + { + "epoch": 3.8079331941544883, + "grad_norm": 1.1507478952407837, + "learning_rate": 1.6066406451988104e-06, + "loss": 0.4664, + "step": 1218 + }, + { + "epoch": 3.8110647181628394, + "grad_norm": 0.7778187990188599, + "learning_rate": 1.6025982128133073e-06, + "loss": 0.456, + "step": 1219 + }, + { + "epoch": 3.81419624217119, + "grad_norm": 0.8383583426475525, + "learning_rate": 1.5985584720369876e-06, + "loss": 0.4684, + "step": 1220 + }, + { + "epoch": 3.817327766179541, + "grad_norm": 0.7743321061134338, + "learning_rate": 1.5945214349863914e-06, + "loss": 0.4567, + "step": 1221 + }, + { + "epoch": 3.8204592901878915, + "grad_norm": 0.8020774126052856, + "learning_rate": 1.5904871137699462e-06, + "loss": 0.4175, + "step": 1222 + }, + { + "epoch": 3.8235908141962422, + "grad_norm": 0.790318489074707, + "learning_rate": 1.5864555204879375e-06, + "loss": 0.469, + "step": 1223 + }, + { + "epoch": 3.826722338204593, + "grad_norm": 0.8583689332008362, + "learning_rate": 1.5824266672324652e-06, + "loss": 0.4931, + "step": 1224 + }, + { + "epoch": 3.8298538622129437, + "grad_norm": 0.7788206934928894, + "learning_rate": 1.5784005660874125e-06, + "loss": 0.4643, + "step": 1225 + }, + { + "epoch": 3.8329853862212944, + "grad_norm": 0.8385717868804932, + "learning_rate": 1.574377229128409e-06, + "loss": 0.4567, + "step": 1226 + }, + { + "epoch": 3.836116910229645, + "grad_norm": 0.8447727560997009, + "learning_rate": 1.5703566684227922e-06, + "loss": 0.42, + "step": 1227 + }, + { + "epoch": 3.8392484342379958, + "grad_norm": 0.7286496758460999, + "learning_rate": 1.5663388960295742e-06, + "loss": 0.4603, + "step": 1228 + }, + { + "epoch": 3.8423799582463465, + "grad_norm": 0.8493947982788086, + "learning_rate": 1.562323923999401e-06, + "loss": 0.4731, + "step": 1229 + }, + { + "epoch": 3.845511482254697, + "grad_norm": 0.8641151785850525, + "learning_rate": 1.5583117643745233e-06, + "loss": 0.4491, + "step": 1230 + }, + { + "epoch": 3.848643006263048, + "grad_norm": 0.9493702054023743, + "learning_rate": 1.5543024291887532e-06, + "loss": 0.411, + "step": 1231 + }, + { + "epoch": 3.8517745302713986, + "grad_norm": 0.7246205806732178, + "learning_rate": 1.5502959304674337e-06, + "loss": 0.4569, + "step": 1232 + }, + { + "epoch": 3.8549060542797493, + "grad_norm": 0.7576872110366821, + "learning_rate": 1.5462922802273994e-06, + "loss": 0.4548, + "step": 1233 + }, + { + "epoch": 3.8580375782881005, + "grad_norm": 0.7710747718811035, + "learning_rate": 1.5422914904769404e-06, + "loss": 0.447, + "step": 1234 + }, + { + "epoch": 3.8611691022964507, + "grad_norm": 0.7661204934120178, + "learning_rate": 1.5382935732157677e-06, + "loss": 0.4601, + "step": 1235 + }, + { + "epoch": 3.864300626304802, + "grad_norm": 0.7133070826530457, + "learning_rate": 1.5342985404349788e-06, + "loss": 0.4245, + "step": 1236 + }, + { + "epoch": 3.867432150313152, + "grad_norm": 0.7716459631919861, + "learning_rate": 1.5303064041170163e-06, + "loss": 0.4543, + "step": 1237 + }, + { + "epoch": 3.8705636743215033, + "grad_norm": 0.737501859664917, + "learning_rate": 1.5263171762356388e-06, + "loss": 0.4405, + "step": 1238 + }, + { + "epoch": 3.873695198329854, + "grad_norm": 0.7885998487472534, + "learning_rate": 1.5223308687558786e-06, + "loss": 0.4412, + "step": 1239 + }, + { + "epoch": 3.8768267223382047, + "grad_norm": 0.7516661882400513, + "learning_rate": 1.5183474936340092e-06, + "loss": 0.4185, + "step": 1240 + }, + { + "epoch": 3.8799582463465554, + "grad_norm": 0.8790446519851685, + "learning_rate": 1.5143670628175111e-06, + "loss": 0.444, + "step": 1241 + }, + { + "epoch": 3.883089770354906, + "grad_norm": 0.7695789337158203, + "learning_rate": 1.5103895882450315e-06, + "loss": 0.4615, + "step": 1242 + }, + { + "epoch": 3.886221294363257, + "grad_norm": 0.7859196662902832, + "learning_rate": 1.506415081846353e-06, + "loss": 0.4379, + "step": 1243 + }, + { + "epoch": 3.8893528183716075, + "grad_norm": 0.7771942615509033, + "learning_rate": 1.5024435555423522e-06, + "loss": 0.5013, + "step": 1244 + }, + { + "epoch": 3.892484342379958, + "grad_norm": 0.7415695190429688, + "learning_rate": 1.498475021244971e-06, + "loss": 0.4861, + "step": 1245 + }, + { + "epoch": 3.895615866388309, + "grad_norm": 0.7909391522407532, + "learning_rate": 1.4945094908571755e-06, + "loss": 0.4599, + "step": 1246 + }, + { + "epoch": 3.8987473903966596, + "grad_norm": 0.7749060988426208, + "learning_rate": 1.490546976272923e-06, + "loss": 0.3986, + "step": 1247 + }, + { + "epoch": 3.9018789144050103, + "grad_norm": 0.810681164264679, + "learning_rate": 1.4865874893771248e-06, + "loss": 0.4495, + "step": 1248 + }, + { + "epoch": 3.905010438413361, + "grad_norm": 0.8018531799316406, + "learning_rate": 1.4826310420456103e-06, + "loss": 0.4426, + "step": 1249 + }, + { + "epoch": 3.9081419624217117, + "grad_norm": 0.756064236164093, + "learning_rate": 1.4786776461450924e-06, + "loss": 0.4474, + "step": 1250 + }, + { + "epoch": 3.911273486430063, + "grad_norm": 0.7581740021705627, + "learning_rate": 1.4747273135331347e-06, + "loss": 0.4494, + "step": 1251 + }, + { + "epoch": 3.914405010438413, + "grad_norm": 0.7666076421737671, + "learning_rate": 1.4707800560581086e-06, + "loss": 0.4593, + "step": 1252 + }, + { + "epoch": 3.9175365344467643, + "grad_norm": 0.7339973449707031, + "learning_rate": 1.4668358855591664e-06, + "loss": 0.4682, + "step": 1253 + }, + { + "epoch": 3.9206680584551146, + "grad_norm": 0.8504599928855896, + "learning_rate": 1.4628948138661974e-06, + "loss": 0.4504, + "step": 1254 + }, + { + "epoch": 3.9237995824634657, + "grad_norm": 0.8332642912864685, + "learning_rate": 1.4589568527997985e-06, + "loss": 0.5024, + "step": 1255 + }, + { + "epoch": 3.9269311064718164, + "grad_norm": 0.7813694477081299, + "learning_rate": 1.4550220141712384e-06, + "loss": 0.4547, + "step": 1256 + }, + { + "epoch": 3.930062630480167, + "grad_norm": 0.842258632183075, + "learning_rate": 1.451090309782417e-06, + "loss": 0.4584, + "step": 1257 + }, + { + "epoch": 3.933194154488518, + "grad_norm": 0.8159133791923523, + "learning_rate": 1.4471617514258373e-06, + "loss": 0.4538, + "step": 1258 + }, + { + "epoch": 3.9363256784968685, + "grad_norm": 0.8117021322250366, + "learning_rate": 1.4432363508845626e-06, + "loss": 0.4315, + "step": 1259 + }, + { + "epoch": 3.9394572025052192, + "grad_norm": 0.8087465167045593, + "learning_rate": 1.4393141199321881e-06, + "loss": 0.4367, + "step": 1260 + }, + { + "epoch": 3.94258872651357, + "grad_norm": 0.7954697012901306, + "learning_rate": 1.435395070332801e-06, + "loss": 0.4515, + "step": 1261 + }, + { + "epoch": 3.9457202505219207, + "grad_norm": 0.7305286526679993, + "learning_rate": 1.4314792138409454e-06, + "loss": 0.4879, + "step": 1262 + }, + { + "epoch": 3.9488517745302714, + "grad_norm": 0.8883433938026428, + "learning_rate": 1.4275665622015908e-06, + "loss": 0.4489, + "step": 1263 + }, + { + "epoch": 3.951983298538622, + "grad_norm": 0.8176298141479492, + "learning_rate": 1.4236571271500909e-06, + "loss": 0.4583, + "step": 1264 + }, + { + "epoch": 3.9551148225469728, + "grad_norm": 0.8042430281639099, + "learning_rate": 1.4197509204121563e-06, + "loss": 0.4277, + "step": 1265 + }, + { + "epoch": 3.9582463465553235, + "grad_norm": 0.8153829574584961, + "learning_rate": 1.4158479537038095e-06, + "loss": 0.4389, + "step": 1266 + }, + { + "epoch": 3.961377870563674, + "grad_norm": 0.7908188104629517, + "learning_rate": 1.4119482387313588e-06, + "loss": 0.4421, + "step": 1267 + }, + { + "epoch": 3.964509394572025, + "grad_norm": 0.831758975982666, + "learning_rate": 1.4080517871913596e-06, + "loss": 0.4308, + "step": 1268 + }, + { + "epoch": 3.9676409185803756, + "grad_norm": 0.8191989064216614, + "learning_rate": 1.4041586107705758e-06, + "loss": 0.4654, + "step": 1269 + }, + { + "epoch": 3.9707724425887267, + "grad_norm": 0.9455055594444275, + "learning_rate": 1.4002687211459524e-06, + "loss": 0.4668, + "step": 1270 + }, + { + "epoch": 3.973903966597077, + "grad_norm": 0.9271034002304077, + "learning_rate": 1.396382129984572e-06, + "loss": 0.4414, + "step": 1271 + }, + { + "epoch": 3.977035490605428, + "grad_norm": 0.7652955651283264, + "learning_rate": 1.392498848943627e-06, + "loss": 0.4575, + "step": 1272 + }, + { + "epoch": 3.980167014613779, + "grad_norm": 0.7850046157836914, + "learning_rate": 1.3886188896703816e-06, + "loss": 0.4554, + "step": 1273 + }, + { + "epoch": 3.9832985386221296, + "grad_norm": 0.7194349765777588, + "learning_rate": 1.3847422638021357e-06, + "loss": 0.437, + "step": 1274 + }, + { + "epoch": 3.9864300626304803, + "grad_norm": 0.8726270198822021, + "learning_rate": 1.3808689829661899e-06, + "loss": 0.4657, + "step": 1275 + }, + { + "epoch": 3.989561586638831, + "grad_norm": 0.7741451263427734, + "learning_rate": 1.3769990587798146e-06, + "loss": 0.3931, + "step": 1276 + }, + { + "epoch": 3.9926931106471817, + "grad_norm": 0.9160324931144714, + "learning_rate": 1.3731325028502116e-06, + "loss": 0.4358, + "step": 1277 + }, + { + "epoch": 3.9958246346555324, + "grad_norm": 0.7819761037826538, + "learning_rate": 1.3692693267744806e-06, + "loss": 0.4767, + "step": 1278 + }, + { + "epoch": 3.998956158663883, + "grad_norm": 0.7595860958099365, + "learning_rate": 1.365409542139583e-06, + "loss": 0.4217, + "step": 1279 + }, + { + "epoch": 4.0, + "grad_norm": 0.7595860958099365, + "learning_rate": 1.361553160522307e-06, + "loss": 0.1478, + "step": 1280 + }, + { + "epoch": 4.003131524008351, + "grad_norm": 1.0793898105621338, + "learning_rate": 1.357700193489237e-06, + "loss": 0.4453, + "step": 1281 + }, + { + "epoch": 4.006263048016701, + "grad_norm": 0.8228668570518494, + "learning_rate": 1.3538506525967148e-06, + "loss": 0.3981, + "step": 1282 + }, + { + "epoch": 4.009394572025053, + "grad_norm": 0.7946518063545227, + "learning_rate": 1.3500045493908044e-06, + "loss": 0.423, + "step": 1283 + }, + { + "epoch": 4.012526096033403, + "grad_norm": 0.7924147248268127, + "learning_rate": 1.3461618954072614e-06, + "loss": 0.4211, + "step": 1284 + }, + { + "epoch": 4.015657620041754, + "grad_norm": 0.891967236995697, + "learning_rate": 1.3423227021714937e-06, + "loss": 0.4365, + "step": 1285 + }, + { + "epoch": 4.018789144050104, + "grad_norm": 0.9970533847808838, + "learning_rate": 1.3384869811985323e-06, + "loss": 0.4053, + "step": 1286 + }, + { + "epoch": 4.021920668058455, + "grad_norm": 1.001489281654358, + "learning_rate": 1.33465474399299e-06, + "loss": 0.4382, + "step": 1287 + }, + { + "epoch": 4.025052192066806, + "grad_norm": 0.7584173083305359, + "learning_rate": 1.3308260020490332e-06, + "loss": 0.441, + "step": 1288 + }, + { + "epoch": 4.028183716075157, + "grad_norm": 0.7444385290145874, + "learning_rate": 1.327000766850345e-06, + "loss": 0.4172, + "step": 1289 + }, + { + "epoch": 4.031315240083507, + "grad_norm": 0.8264429569244385, + "learning_rate": 1.3231790498700886e-06, + "loss": 0.3956, + "step": 1290 + }, + { + "epoch": 4.034446764091858, + "grad_norm": 0.7339959740638733, + "learning_rate": 1.319360862570877e-06, + "loss": 0.4431, + "step": 1291 + }, + { + "epoch": 4.0375782881002085, + "grad_norm": 0.9837627410888672, + "learning_rate": 1.3155462164047345e-06, + "loss": 0.4439, + "step": 1292 + }, + { + "epoch": 4.04070981210856, + "grad_norm": 0.8792996406555176, + "learning_rate": 1.3117351228130657e-06, + "loss": 0.4664, + "step": 1293 + }, + { + "epoch": 4.04384133611691, + "grad_norm": 0.8377962112426758, + "learning_rate": 1.3079275932266205e-06, + "loss": 0.4267, + "step": 1294 + }, + { + "epoch": 4.046972860125261, + "grad_norm": 0.755979597568512, + "learning_rate": 1.3041236390654577e-06, + "loss": 0.4394, + "step": 1295 + }, + { + "epoch": 4.050104384133611, + "grad_norm": 0.7994539737701416, + "learning_rate": 1.3003232717389114e-06, + "loss": 0.4547, + "step": 1296 + }, + { + "epoch": 4.053235908141962, + "grad_norm": 0.8308882117271423, + "learning_rate": 1.2965265026455598e-06, + "loss": 0.408, + "step": 1297 + }, + { + "epoch": 4.056367432150314, + "grad_norm": 0.8120949268341064, + "learning_rate": 1.292733343173188e-06, + "loss": 0.3524, + "step": 1298 + }, + { + "epoch": 4.059498956158664, + "grad_norm": 0.7639186978340149, + "learning_rate": 1.288943804698755e-06, + "loss": 0.4677, + "step": 1299 + }, + { + "epoch": 4.062630480167015, + "grad_norm": 0.8439323306083679, + "learning_rate": 1.2851578985883586e-06, + "loss": 0.436, + "step": 1300 + }, + { + "epoch": 4.065762004175365, + "grad_norm": 0.7270680069923401, + "learning_rate": 1.2813756361972001e-06, + "loss": 0.4519, + "step": 1301 + }, + { + "epoch": 4.068893528183716, + "grad_norm": 0.7746233940124512, + "learning_rate": 1.2775970288695554e-06, + "loss": 0.4503, + "step": 1302 + }, + { + "epoch": 4.072025052192067, + "grad_norm": 0.7725366950035095, + "learning_rate": 1.2738220879387376e-06, + "loss": 0.4467, + "step": 1303 + }, + { + "epoch": 4.075156576200418, + "grad_norm": 0.7371346950531006, + "learning_rate": 1.2700508247270597e-06, + "loss": 0.4144, + "step": 1304 + }, + { + "epoch": 4.078288100208768, + "grad_norm": 0.778403639793396, + "learning_rate": 1.2662832505458057e-06, + "loss": 0.4345, + "step": 1305 + }, + { + "epoch": 4.081419624217119, + "grad_norm": 0.8683139085769653, + "learning_rate": 1.2625193766951957e-06, + "loss": 0.4345, + "step": 1306 + }, + { + "epoch": 4.0845511482254695, + "grad_norm": 0.7849321365356445, + "learning_rate": 1.2587592144643513e-06, + "loss": 0.4505, + "step": 1307 + }, + { + "epoch": 4.087682672233821, + "grad_norm": 0.9351589679718018, + "learning_rate": 1.2550027751312618e-06, + "loss": 0.4173, + "step": 1308 + }, + { + "epoch": 4.090814196242171, + "grad_norm": 0.7938134074211121, + "learning_rate": 1.2512500699627489e-06, + "loss": 0.4061, + "step": 1309 + }, + { + "epoch": 4.093945720250522, + "grad_norm": 0.8015687465667725, + "learning_rate": 1.2475011102144337e-06, + "loss": 0.423, + "step": 1310 + }, + { + "epoch": 4.097077244258872, + "grad_norm": 0.8477723598480225, + "learning_rate": 1.2437559071307062e-06, + "loss": 0.4475, + "step": 1311 + }, + { + "epoch": 4.1002087682672235, + "grad_norm": 0.7643985152244568, + "learning_rate": 1.2400144719446885e-06, + "loss": 0.4613, + "step": 1312 + }, + { + "epoch": 4.103340292275574, + "grad_norm": 0.7612454295158386, + "learning_rate": 1.2362768158781985e-06, + "loss": 0.4333, + "step": 1313 + }, + { + "epoch": 4.106471816283925, + "grad_norm": 0.9039669036865234, + "learning_rate": 1.2325429501417232e-06, + "loss": 0.4091, + "step": 1314 + }, + { + "epoch": 4.109603340292275, + "grad_norm": 0.8423281908035278, + "learning_rate": 1.228812885934378e-06, + "loss": 0.4482, + "step": 1315 + }, + { + "epoch": 4.112734864300626, + "grad_norm": 0.8980326652526855, + "learning_rate": 1.2250866344438782e-06, + "loss": 0.4354, + "step": 1316 + }, + { + "epoch": 4.115866388308977, + "grad_norm": 0.7539350986480713, + "learning_rate": 1.221364206846502e-06, + "loss": 0.4332, + "step": 1317 + }, + { + "epoch": 4.118997912317328, + "grad_norm": 0.855925977230072, + "learning_rate": 1.2176456143070597e-06, + "loss": 0.4198, + "step": 1318 + }, + { + "epoch": 4.122129436325679, + "grad_norm": 0.7506774663925171, + "learning_rate": 1.2139308679788594e-06, + "loss": 0.4484, + "step": 1319 + }, + { + "epoch": 4.125260960334029, + "grad_norm": 0.820524275302887, + "learning_rate": 1.2102199790036709e-06, + "loss": 0.4208, + "step": 1320 + }, + { + "epoch": 4.12839248434238, + "grad_norm": 0.8255232572555542, + "learning_rate": 1.2065129585116968e-06, + "loss": 0.4431, + "step": 1321 + }, + { + "epoch": 4.1315240083507305, + "grad_norm": 0.7454650402069092, + "learning_rate": 1.2028098176215347e-06, + "loss": 0.3833, + "step": 1322 + }, + { + "epoch": 4.134655532359082, + "grad_norm": 0.7500002980232239, + "learning_rate": 1.199110567440148e-06, + "loss": 0.4486, + "step": 1323 + }, + { + "epoch": 4.137787056367432, + "grad_norm": 0.783704400062561, + "learning_rate": 1.1954152190628304e-06, + "loss": 0.4535, + "step": 1324 + }, + { + "epoch": 4.140918580375783, + "grad_norm": 0.7665114402770996, + "learning_rate": 1.1917237835731713e-06, + "loss": 0.4575, + "step": 1325 + }, + { + "epoch": 4.144050104384133, + "grad_norm": 0.7992947697639465, + "learning_rate": 1.1880362720430235e-06, + "loss": 0.4294, + "step": 1326 + }, + { + "epoch": 4.1471816283924845, + "grad_norm": 0.8273386359214783, + "learning_rate": 1.1843526955324736e-06, + "loss": 0.4578, + "step": 1327 + }, + { + "epoch": 4.150313152400835, + "grad_norm": 0.7992604970932007, + "learning_rate": 1.1806730650898029e-06, + "loss": 0.4125, + "step": 1328 + }, + { + "epoch": 4.153444676409186, + "grad_norm": 0.7935183644294739, + "learning_rate": 1.1769973917514612e-06, + "loss": 0.4537, + "step": 1329 + }, + { + "epoch": 4.156576200417536, + "grad_norm": 0.8761199116706848, + "learning_rate": 1.173325686542022e-06, + "loss": 0.4433, + "step": 1330 + }, + { + "epoch": 4.159707724425887, + "grad_norm": 0.7764279246330261, + "learning_rate": 1.1696579604741643e-06, + "loss": 0.4165, + "step": 1331 + }, + { + "epoch": 4.162839248434238, + "grad_norm": 0.9479373693466187, + "learning_rate": 1.165994224548629e-06, + "loss": 0.42, + "step": 1332 + }, + { + "epoch": 4.165970772442589, + "grad_norm": 1.1482913494110107, + "learning_rate": 1.162334489754191e-06, + "loss": 0.4545, + "step": 1333 + }, + { + "epoch": 4.16910229645094, + "grad_norm": 0.825964093208313, + "learning_rate": 1.1586787670676228e-06, + "loss": 0.4423, + "step": 1334 + }, + { + "epoch": 4.17223382045929, + "grad_norm": 0.8352368474006653, + "learning_rate": 1.1550270674536626e-06, + "loss": 0.4509, + "step": 1335 + }, + { + "epoch": 4.175365344467641, + "grad_norm": 0.8211585879325867, + "learning_rate": 1.1513794018649846e-06, + "loss": 0.4482, + "step": 1336 + }, + { + "epoch": 4.1784968684759916, + "grad_norm": 0.7863073945045471, + "learning_rate": 1.1477357812421628e-06, + "loss": 0.4318, + "step": 1337 + }, + { + "epoch": 4.181628392484343, + "grad_norm": 0.8023143410682678, + "learning_rate": 1.1440962165136366e-06, + "loss": 0.4269, + "step": 1338 + }, + { + "epoch": 4.184759916492693, + "grad_norm": 0.8101564049720764, + "learning_rate": 1.1404607185956843e-06, + "loss": 0.4284, + "step": 1339 + }, + { + "epoch": 4.187891440501044, + "grad_norm": 0.755865216255188, + "learning_rate": 1.1368292983923826e-06, + "loss": 0.4195, + "step": 1340 + }, + { + "epoch": 4.191022964509394, + "grad_norm": 0.7991048097610474, + "learning_rate": 1.1332019667955805e-06, + "loss": 0.3774, + "step": 1341 + }, + { + "epoch": 4.1941544885177455, + "grad_norm": 0.7407388091087341, + "learning_rate": 1.1295787346848641e-06, + "loss": 0.4321, + "step": 1342 + }, + { + "epoch": 4.197286012526096, + "grad_norm": 0.8769527673721313, + "learning_rate": 1.1259596129275205e-06, + "loss": 0.4349, + "step": 1343 + }, + { + "epoch": 4.200417536534447, + "grad_norm": 0.7908845543861389, + "learning_rate": 1.1223446123785128e-06, + "loss": 0.4595, + "step": 1344 + }, + { + "epoch": 4.203549060542797, + "grad_norm": 0.810578465461731, + "learning_rate": 1.1187337438804394e-06, + "loss": 0.4464, + "step": 1345 + }, + { + "epoch": 4.206680584551148, + "grad_norm": 0.8122991323471069, + "learning_rate": 1.1151270182635082e-06, + "loss": 0.4437, + "step": 1346 + }, + { + "epoch": 4.209812108559499, + "grad_norm": 0.8065986633300781, + "learning_rate": 1.1115244463454988e-06, + "loss": 0.4198, + "step": 1347 + }, + { + "epoch": 4.21294363256785, + "grad_norm": 0.7652766108512878, + "learning_rate": 1.107926038931734e-06, + "loss": 0.4141, + "step": 1348 + }, + { + "epoch": 4.2160751565762, + "grad_norm": 0.7936646938323975, + "learning_rate": 1.1043318068150457e-06, + "loss": 0.43, + "step": 1349 + }, + { + "epoch": 4.219206680584551, + "grad_norm": 0.8732025027275085, + "learning_rate": 1.1007417607757435e-06, + "loss": 0.4243, + "step": 1350 + }, + { + "epoch": 4.222338204592901, + "grad_norm": 0.8516861200332642, + "learning_rate": 1.0971559115815795e-06, + "loss": 0.4352, + "step": 1351 + }, + { + "epoch": 4.225469728601253, + "grad_norm": 0.7985629439353943, + "learning_rate": 1.093574269987718e-06, + "loss": 0.4193, + "step": 1352 + }, + { + "epoch": 4.228601252609604, + "grad_norm": 0.7350221872329712, + "learning_rate": 1.0899968467367056e-06, + "loss": 0.4466, + "step": 1353 + }, + { + "epoch": 4.231732776617954, + "grad_norm": 0.7626602649688721, + "learning_rate": 1.086423652558436e-06, + "loss": 0.399, + "step": 1354 + }, + { + "epoch": 4.234864300626305, + "grad_norm": 0.7596367597579956, + "learning_rate": 1.082854698170117e-06, + "loss": 0.4023, + "step": 1355 + }, + { + "epoch": 4.237995824634655, + "grad_norm": 0.7802876830101013, + "learning_rate": 1.0792899942762406e-06, + "loss": 0.4351, + "step": 1356 + }, + { + "epoch": 4.241127348643007, + "grad_norm": 1.0809392929077148, + "learning_rate": 1.0757295515685504e-06, + "loss": 0.4271, + "step": 1357 + }, + { + "epoch": 4.244258872651357, + "grad_norm": 0.8581835031509399, + "learning_rate": 1.0721733807260094e-06, + "loss": 0.4332, + "step": 1358 + }, + { + "epoch": 4.247390396659708, + "grad_norm": 0.884496808052063, + "learning_rate": 1.0686214924147686e-06, + "loss": 0.4317, + "step": 1359 + }, + { + "epoch": 4.250521920668058, + "grad_norm": 0.7651242017745972, + "learning_rate": 1.0650738972881325e-06, + "loss": 0.4614, + "step": 1360 + }, + { + "epoch": 4.253653444676409, + "grad_norm": 0.7798812985420227, + "learning_rate": 1.0615306059865286e-06, + "loss": 0.4088, + "step": 1361 + }, + { + "epoch": 4.25678496868476, + "grad_norm": 0.9011983871459961, + "learning_rate": 1.0579916291374776e-06, + "loss": 0.4242, + "step": 1362 + }, + { + "epoch": 4.259916492693111, + "grad_norm": 0.7490668892860413, + "learning_rate": 1.0544569773555602e-06, + "loss": 0.4369, + "step": 1363 + }, + { + "epoch": 4.263048016701461, + "grad_norm": 0.7936168909072876, + "learning_rate": 1.0509266612423819e-06, + "loss": 0.4321, + "step": 1364 + }, + { + "epoch": 4.266179540709812, + "grad_norm": 0.8428751826286316, + "learning_rate": 1.047400691386547e-06, + "loss": 0.4391, + "step": 1365 + }, + { + "epoch": 4.2693110647181625, + "grad_norm": 0.7458098530769348, + "learning_rate": 1.0438790783636213e-06, + "loss": 0.4789, + "step": 1366 + }, + { + "epoch": 4.272442588726514, + "grad_norm": 0.8869938254356384, + "learning_rate": 1.0403618327361056e-06, + "loss": 0.4181, + "step": 1367 + }, + { + "epoch": 4.275574112734864, + "grad_norm": 0.7711350321769714, + "learning_rate": 1.0368489650533989e-06, + "loss": 0.4334, + "step": 1368 + }, + { + "epoch": 4.278705636743215, + "grad_norm": 0.7643832564353943, + "learning_rate": 1.0333404858517712e-06, + "loss": 0.4444, + "step": 1369 + }, + { + "epoch": 4.281837160751566, + "grad_norm": 0.925422191619873, + "learning_rate": 1.02983640565433e-06, + "loss": 0.4134, + "step": 1370 + }, + { + "epoch": 4.284968684759916, + "grad_norm": 1.4223377704620361, + "learning_rate": 1.0263367349709866e-06, + "loss": 0.4441, + "step": 1371 + }, + { + "epoch": 4.288100208768268, + "grad_norm": 0.9713383316993713, + "learning_rate": 1.0228414842984297e-06, + "loss": 0.418, + "step": 1372 + }, + { + "epoch": 4.291231732776618, + "grad_norm": 0.8378648161888123, + "learning_rate": 1.0193506641200874e-06, + "loss": 0.4017, + "step": 1373 + }, + { + "epoch": 4.294363256784969, + "grad_norm": 0.9343763589859009, + "learning_rate": 1.0158642849061018e-06, + "loss": 0.4468, + "step": 1374 + }, + { + "epoch": 4.297494780793319, + "grad_norm": 0.7863892912864685, + "learning_rate": 1.012382357113296e-06, + "loss": 0.4235, + "step": 1375 + }, + { + "epoch": 4.30062630480167, + "grad_norm": 0.8109874725341797, + "learning_rate": 1.0089048911851388e-06, + "loss": 0.4356, + "step": 1376 + }, + { + "epoch": 4.303757828810021, + "grad_norm": 0.7978389859199524, + "learning_rate": 1.005431897551717e-06, + "loss": 0.4517, + "step": 1377 + }, + { + "epoch": 4.306889352818372, + "grad_norm": 0.8213954567909241, + "learning_rate": 1.001963386629705e-06, + "loss": 0.4263, + "step": 1378 + }, + { + "epoch": 4.310020876826722, + "grad_norm": 0.7367708683013916, + "learning_rate": 9.984993688223319e-07, + "loss": 0.4467, + "step": 1379 + }, + { + "epoch": 4.313152400835073, + "grad_norm": 0.8723400831222534, + "learning_rate": 9.950398545193508e-07, + "loss": 0.4517, + "step": 1380 + }, + { + "epoch": 4.3162839248434235, + "grad_norm": 0.7927054166793823, + "learning_rate": 9.915848540970033e-07, + "loss": 0.4755, + "step": 1381 + }, + { + "epoch": 4.319415448851775, + "grad_norm": 0.8598108291625977, + "learning_rate": 9.88134377917997e-07, + "loss": 0.4539, + "step": 1382 + }, + { + "epoch": 4.322546972860125, + "grad_norm": 1.0660723447799683, + "learning_rate": 9.84688436331468e-07, + "loss": 0.4541, + "step": 1383 + }, + { + "epoch": 4.325678496868476, + "grad_norm": 0.840942919254303, + "learning_rate": 9.81247039672953e-07, + "loss": 0.4239, + "step": 1384 + }, + { + "epoch": 4.328810020876826, + "grad_norm": 0.8142675757408142, + "learning_rate": 9.778101982643549e-07, + "loss": 0.4259, + "step": 1385 + }, + { + "epoch": 4.3319415448851775, + "grad_norm": 0.7841728329658508, + "learning_rate": 9.743779224139133e-07, + "loss": 0.432, + "step": 1386 + }, + { + "epoch": 4.335073068893529, + "grad_norm": 0.9642646908760071, + "learning_rate": 9.709502224161769e-07, + "loss": 0.4423, + "step": 1387 + }, + { + "epoch": 4.338204592901879, + "grad_norm": 0.7967872619628906, + "learning_rate": 9.675271085519692e-07, + "loss": 0.4428, + "step": 1388 + }, + { + "epoch": 4.34133611691023, + "grad_norm": 0.7652244567871094, + "learning_rate": 9.64108591088356e-07, + "loss": 0.4607, + "step": 1389 + }, + { + "epoch": 4.34446764091858, + "grad_norm": 0.8806408643722534, + "learning_rate": 9.606946802786204e-07, + "loss": 0.3859, + "step": 1390 + }, + { + "epoch": 4.347599164926931, + "grad_norm": 0.8638584613800049, + "learning_rate": 9.572853863622252e-07, + "loss": 0.3758, + "step": 1391 + }, + { + "epoch": 4.350730688935282, + "grad_norm": 0.7262263894081116, + "learning_rate": 9.538807195647882e-07, + "loss": 0.4435, + "step": 1392 + }, + { + "epoch": 4.353862212943633, + "grad_norm": 0.785175621509552, + "learning_rate": 9.504806900980482e-07, + "loss": 0.4037, + "step": 1393 + }, + { + "epoch": 4.356993736951983, + "grad_norm": 0.8310196995735168, + "learning_rate": 9.470853081598338e-07, + "loss": 0.4434, + "step": 1394 + }, + { + "epoch": 4.360125260960334, + "grad_norm": 0.7547351717948914, + "learning_rate": 9.436945839340364e-07, + "loss": 0.4318, + "step": 1395 + }, + { + "epoch": 4.3632567849686845, + "grad_norm": 0.816460907459259, + "learning_rate": 9.403085275905746e-07, + "loss": 0.4173, + "step": 1396 + }, + { + "epoch": 4.366388308977036, + "grad_norm": 0.8159666657447815, + "learning_rate": 9.369271492853696e-07, + "loss": 0.4506, + "step": 1397 + }, + { + "epoch": 4.369519832985386, + "grad_norm": 0.8234115839004517, + "learning_rate": 9.335504591603084e-07, + "loss": 0.4186, + "step": 1398 + }, + { + "epoch": 4.372651356993737, + "grad_norm": 0.9084867238998413, + "learning_rate": 9.301784673432187e-07, + "loss": 0.4414, + "step": 1399 + }, + { + "epoch": 4.375782881002087, + "grad_norm": 0.7826300263404846, + "learning_rate": 9.268111839478369e-07, + "loss": 0.4415, + "step": 1400 + }, + { + "epoch": 4.3789144050104385, + "grad_norm": 0.7957226634025574, + "learning_rate": 9.23448619073774e-07, + "loss": 0.4307, + "step": 1401 + }, + { + "epoch": 4.382045929018789, + "grad_norm": 0.7768528461456299, + "learning_rate": 9.200907828064931e-07, + "loss": 0.4107, + "step": 1402 + }, + { + "epoch": 4.38517745302714, + "grad_norm": 0.878063440322876, + "learning_rate": 9.167376852172702e-07, + "loss": 0.4232, + "step": 1403 + }, + { + "epoch": 4.388308977035491, + "grad_norm": 0.7961450219154358, + "learning_rate": 9.133893363631713e-07, + "loss": 0.449, + "step": 1404 + }, + { + "epoch": 4.391440501043841, + "grad_norm": 0.8734915256500244, + "learning_rate": 9.100457462870197e-07, + "loss": 0.4392, + "step": 1405 + }, + { + "epoch": 4.3945720250521925, + "grad_norm": 0.7659426331520081, + "learning_rate": 9.06706925017363e-07, + "loss": 0.4391, + "step": 1406 + }, + { + "epoch": 4.397703549060543, + "grad_norm": 0.949674665927887, + "learning_rate": 9.033728825684466e-07, + "loss": 0.4054, + "step": 1407 + }, + { + "epoch": 4.400835073068894, + "grad_norm": 0.8209534287452698, + "learning_rate": 9.000436289401832e-07, + "loss": 0.3973, + "step": 1408 + }, + { + "epoch": 4.403966597077244, + "grad_norm": 0.9697580933570862, + "learning_rate": 8.967191741181225e-07, + "loss": 0.4228, + "step": 1409 + }, + { + "epoch": 4.407098121085595, + "grad_norm": 0.7681871652603149, + "learning_rate": 8.933995280734217e-07, + "loss": 0.4183, + "step": 1410 + }, + { + "epoch": 4.4102296450939455, + "grad_norm": 0.7434380054473877, + "learning_rate": 8.900847007628103e-07, + "loss": 0.4068, + "step": 1411 + }, + { + "epoch": 4.413361169102297, + "grad_norm": 0.9835489988327026, + "learning_rate": 8.867747021285697e-07, + "loss": 0.4356, + "step": 1412 + }, + { + "epoch": 4.416492693110647, + "grad_norm": 0.7727295160293579, + "learning_rate": 8.834695420984971e-07, + "loss": 0.4201, + "step": 1413 + }, + { + "epoch": 4.419624217118998, + "grad_norm": 0.8018465042114258, + "learning_rate": 8.801692305858775e-07, + "loss": 0.4471, + "step": 1414 + }, + { + "epoch": 4.422755741127348, + "grad_norm": 0.8191575407981873, + "learning_rate": 8.76873777489452e-07, + "loss": 0.4438, + "step": 1415 + }, + { + "epoch": 4.4258872651356995, + "grad_norm": 0.7755978107452393, + "learning_rate": 8.735831926933896e-07, + "loss": 0.4441, + "step": 1416 + }, + { + "epoch": 4.42901878914405, + "grad_norm": 0.8469848036766052, + "learning_rate": 8.702974860672603e-07, + "loss": 0.4134, + "step": 1417 + }, + { + "epoch": 4.432150313152401, + "grad_norm": 0.7887343168258667, + "learning_rate": 8.67016667466001e-07, + "loss": 0.4247, + "step": 1418 + }, + { + "epoch": 4.435281837160751, + "grad_norm": 0.7888761758804321, + "learning_rate": 8.637407467298867e-07, + "loss": 0.4313, + "step": 1419 + }, + { + "epoch": 4.438413361169102, + "grad_norm": 0.8109477162361145, + "learning_rate": 8.60469733684505e-07, + "loss": 0.4472, + "step": 1420 + }, + { + "epoch": 4.4415448851774535, + "grad_norm": 0.8876966238021851, + "learning_rate": 8.572036381407203e-07, + "loss": 0.4395, + "step": 1421 + }, + { + "epoch": 4.444676409185804, + "grad_norm": 0.7669360637664795, + "learning_rate": 8.539424698946503e-07, + "loss": 0.3943, + "step": 1422 + }, + { + "epoch": 4.447807933194154, + "grad_norm": 0.7810583710670471, + "learning_rate": 8.506862387276338e-07, + "loss": 0.4189, + "step": 1423 + }, + { + "epoch": 4.450939457202505, + "grad_norm": 0.9363948702812195, + "learning_rate": 8.474349544061996e-07, + "loss": 0.4245, + "step": 1424 + }, + { + "epoch": 4.454070981210856, + "grad_norm": 0.8147290349006653, + "learning_rate": 8.441886266820418e-07, + "loss": 0.4381, + "step": 1425 + }, + { + "epoch": 4.457202505219207, + "grad_norm": 0.9227074384689331, + "learning_rate": 8.409472652919873e-07, + "loss": 0.4289, + "step": 1426 + }, + { + "epoch": 4.460334029227558, + "grad_norm": 0.8162581920623779, + "learning_rate": 8.377108799579661e-07, + "loss": 0.4565, + "step": 1427 + }, + { + "epoch": 4.463465553235908, + "grad_norm": 0.7634908556938171, + "learning_rate": 8.344794803869835e-07, + "loss": 0.4402, + "step": 1428 + }, + { + "epoch": 4.466597077244259, + "grad_norm": 0.7960299253463745, + "learning_rate": 8.312530762710924e-07, + "loss": 0.4533, + "step": 1429 + }, + { + "epoch": 4.469728601252609, + "grad_norm": 0.8322834968566895, + "learning_rate": 8.280316772873617e-07, + "loss": 0.4282, + "step": 1430 + }, + { + "epoch": 4.4728601252609606, + "grad_norm": 0.7285633683204651, + "learning_rate": 8.248152930978498e-07, + "loss": 0.4287, + "step": 1431 + }, + { + "epoch": 4.475991649269311, + "grad_norm": 0.760178804397583, + "learning_rate": 8.21603933349569e-07, + "loss": 0.4243, + "step": 1432 + }, + { + "epoch": 4.479123173277662, + "grad_norm": 0.8069238662719727, + "learning_rate": 8.183976076744671e-07, + "loss": 0.4352, + "step": 1433 + }, + { + "epoch": 4.482254697286012, + "grad_norm": 0.8068020343780518, + "learning_rate": 8.151963256893911e-07, + "loss": 0.4143, + "step": 1434 + }, + { + "epoch": 4.485386221294363, + "grad_norm": 0.7954995036125183, + "learning_rate": 8.120000969960606e-07, + "loss": 0.4122, + "step": 1435 + }, + { + "epoch": 4.488517745302714, + "grad_norm": 0.791763961315155, + "learning_rate": 8.088089311810379e-07, + "loss": 0.4016, + "step": 1436 + }, + { + "epoch": 4.491649269311065, + "grad_norm": 0.8935474157333374, + "learning_rate": 8.056228378156994e-07, + "loss": 0.4392, + "step": 1437 + }, + { + "epoch": 4.494780793319415, + "grad_norm": 0.791764497756958, + "learning_rate": 8.024418264562094e-07, + "loss": 0.4433, + "step": 1438 + }, + { + "epoch": 4.497912317327766, + "grad_norm": 0.8151891827583313, + "learning_rate": 7.992659066434899e-07, + "loss": 0.4161, + "step": 1439 + }, + { + "epoch": 4.5010438413361165, + "grad_norm": 0.9389538168907166, + "learning_rate": 7.960950879031884e-07, + "loss": 0.4405, + "step": 1440 + }, + { + "epoch": 4.504175365344468, + "grad_norm": 0.7620337009429932, + "learning_rate": 7.929293797456561e-07, + "loss": 0.4443, + "step": 1441 + }, + { + "epoch": 4.507306889352819, + "grad_norm": 0.7784422039985657, + "learning_rate": 7.897687916659133e-07, + "loss": 0.4688, + "step": 1442 + }, + { + "epoch": 4.510438413361169, + "grad_norm": 0.759872555732727, + "learning_rate": 7.866133331436249e-07, + "loss": 0.4349, + "step": 1443 + }, + { + "epoch": 4.51356993736952, + "grad_norm": 0.8815323114395142, + "learning_rate": 7.834630136430713e-07, + "loss": 0.4729, + "step": 1444 + }, + { + "epoch": 4.51670146137787, + "grad_norm": 0.7574021816253662, + "learning_rate": 7.803178426131161e-07, + "loss": 0.4263, + "step": 1445 + }, + { + "epoch": 4.519832985386222, + "grad_norm": 0.7910036444664001, + "learning_rate": 7.771778294871849e-07, + "loss": 0.424, + "step": 1446 + }, + { + "epoch": 4.522964509394572, + "grad_norm": 0.8388727903366089, + "learning_rate": 7.740429836832295e-07, + "loss": 0.4495, + "step": 1447 + }, + { + "epoch": 4.526096033402923, + "grad_norm": 0.7924241423606873, + "learning_rate": 7.709133146037059e-07, + "loss": 0.4157, + "step": 1448 + }, + { + "epoch": 4.529227557411273, + "grad_norm": 0.8103229403495789, + "learning_rate": 7.677888316355411e-07, + "loss": 0.4321, + "step": 1449 + }, + { + "epoch": 4.532359081419624, + "grad_norm": 0.8175046443939209, + "learning_rate": 7.64669544150109e-07, + "loss": 0.4375, + "step": 1450 + }, + { + "epoch": 4.535490605427975, + "grad_norm": 0.8193264603614807, + "learning_rate": 7.615554615032e-07, + "loss": 0.4576, + "step": 1451 + }, + { + "epoch": 4.538622129436326, + "grad_norm": 0.7646095156669617, + "learning_rate": 7.584465930349924e-07, + "loss": 0.4095, + "step": 1452 + }, + { + "epoch": 4.541753653444676, + "grad_norm": 0.8025760054588318, + "learning_rate": 7.553429480700275e-07, + "loss": 0.4387, + "step": 1453 + }, + { + "epoch": 4.544885177453027, + "grad_norm": 0.7339401841163635, + "learning_rate": 7.522445359171768e-07, + "loss": 0.4263, + "step": 1454 + }, + { + "epoch": 4.5480167014613775, + "grad_norm": 0.764500617980957, + "learning_rate": 7.491513658696189e-07, + "loss": 0.4694, + "step": 1455 + }, + { + "epoch": 4.551148225469729, + "grad_norm": 0.7974144220352173, + "learning_rate": 7.460634472048104e-07, + "loss": 0.4335, + "step": 1456 + }, + { + "epoch": 4.554279749478079, + "grad_norm": 0.7994454503059387, + "learning_rate": 7.429807891844546e-07, + "loss": 0.4155, + "step": 1457 + }, + { + "epoch": 4.55741127348643, + "grad_norm": 0.8552598357200623, + "learning_rate": 7.39903401054477e-07, + "loss": 0.428, + "step": 1458 + }, + { + "epoch": 4.560542797494781, + "grad_norm": 0.8604694604873657, + "learning_rate": 7.368312920449985e-07, + "loss": 0.4562, + "step": 1459 + }, + { + "epoch": 4.5636743215031315, + "grad_norm": 0.7699970602989197, + "learning_rate": 7.337644713703049e-07, + "loss": 0.4322, + "step": 1460 + }, + { + "epoch": 4.566805845511483, + "grad_norm": 0.9426809549331665, + "learning_rate": 7.307029482288227e-07, + "loss": 0.387, + "step": 1461 + }, + { + "epoch": 4.569937369519833, + "grad_norm": 0.8332602381706238, + "learning_rate": 7.276467318030841e-07, + "loss": 0.4381, + "step": 1462 + }, + { + "epoch": 4.573068893528184, + "grad_norm": 0.7953102588653564, + "learning_rate": 7.245958312597095e-07, + "loss": 0.3994, + "step": 1463 + }, + { + "epoch": 4.576200417536534, + "grad_norm": 0.7888246178627014, + "learning_rate": 7.215502557493743e-07, + "loss": 0.4191, + "step": 1464 + }, + { + "epoch": 4.579331941544885, + "grad_norm": 1.00043523311615, + "learning_rate": 7.185100144067816e-07, + "loss": 0.4321, + "step": 1465 + }, + { + "epoch": 4.582463465553236, + "grad_norm": 0.8209455609321594, + "learning_rate": 7.154751163506354e-07, + "loss": 0.3841, + "step": 1466 + }, + { + "epoch": 4.585594989561587, + "grad_norm": 0.7574297785758972, + "learning_rate": 7.124455706836131e-07, + "loss": 0.452, + "step": 1467 + }, + { + "epoch": 4.588726513569937, + "grad_norm": 0.8509530425071716, + "learning_rate": 7.094213864923397e-07, + "loss": 0.4225, + "step": 1468 + }, + { + "epoch": 4.591858037578288, + "grad_norm": 0.7906093597412109, + "learning_rate": 7.064025728473589e-07, + "loss": 0.418, + "step": 1469 + }, + { + "epoch": 4.5949895615866385, + "grad_norm": 0.7408525943756104, + "learning_rate": 7.033891388031056e-07, + "loss": 0.4366, + "step": 1470 + }, + { + "epoch": 4.59812108559499, + "grad_norm": 0.766342043876648, + "learning_rate": 7.003810933978805e-07, + "loss": 0.4315, + "step": 1471 + }, + { + "epoch": 4.60125260960334, + "grad_norm": 0.7852875590324402, + "learning_rate": 6.973784456538207e-07, + "loss": 0.4332, + "step": 1472 + }, + { + "epoch": 4.604384133611691, + "grad_norm": 0.7975596785545349, + "learning_rate": 6.943812045768753e-07, + "loss": 0.4563, + "step": 1473 + }, + { + "epoch": 4.607515657620041, + "grad_norm": 0.755770742893219, + "learning_rate": 6.913893791567767e-07, + "loss": 0.417, + "step": 1474 + }, + { + "epoch": 4.6106471816283925, + "grad_norm": 0.8083193302154541, + "learning_rate": 6.884029783670123e-07, + "loss": 0.4285, + "step": 1475 + }, + { + "epoch": 4.613778705636744, + "grad_norm": 0.8146732449531555, + "learning_rate": 6.854220111648022e-07, + "loss": 0.4587, + "step": 1476 + }, + { + "epoch": 4.616910229645094, + "grad_norm": 0.7958261370658875, + "learning_rate": 6.82446486491066e-07, + "loss": 0.4584, + "step": 1477 + }, + { + "epoch": 4.620041753653445, + "grad_norm": 0.9063606262207031, + "learning_rate": 6.79476413270402e-07, + "loss": 0.4056, + "step": 1478 + }, + { + "epoch": 4.623173277661795, + "grad_norm": 0.8176311254501343, + "learning_rate": 6.76511800411056e-07, + "loss": 0.3987, + "step": 1479 + }, + { + "epoch": 4.6263048016701465, + "grad_norm": 0.7843388319015503, + "learning_rate": 6.735526568048975e-07, + "loss": 0.4541, + "step": 1480 + }, + { + "epoch": 4.629436325678497, + "grad_norm": 0.7491182684898376, + "learning_rate": 6.705989913273922e-07, + "loss": 0.4218, + "step": 1481 + }, + { + "epoch": 4.632567849686848, + "grad_norm": 0.8836645483970642, + "learning_rate": 6.676508128375728e-07, + "loss": 0.4153, + "step": 1482 + }, + { + "epoch": 4.635699373695198, + "grad_norm": 0.7614302635192871, + "learning_rate": 6.647081301780175e-07, + "loss": 0.4557, + "step": 1483 + }, + { + "epoch": 4.638830897703549, + "grad_norm": 0.8144665360450745, + "learning_rate": 6.617709521748181e-07, + "loss": 0.4103, + "step": 1484 + }, + { + "epoch": 4.6419624217118995, + "grad_norm": 0.757859468460083, + "learning_rate": 6.588392876375579e-07, + "loss": 0.4498, + "step": 1485 + }, + { + "epoch": 4.645093945720251, + "grad_norm": 0.8298365473747253, + "learning_rate": 6.559131453592837e-07, + "loss": 0.4318, + "step": 1486 + }, + { + "epoch": 4.648225469728601, + "grad_norm": 0.9429208636283875, + "learning_rate": 6.529925341164781e-07, + "loss": 0.4272, + "step": 1487 + }, + { + "epoch": 4.651356993736952, + "grad_norm": 0.7761420011520386, + "learning_rate": 6.500774626690329e-07, + "loss": 0.4224, + "step": 1488 + }, + { + "epoch": 4.654488517745302, + "grad_norm": 0.8003484606742859, + "learning_rate": 6.471679397602273e-07, + "loss": 0.4333, + "step": 1489 + }, + { + "epoch": 4.6576200417536535, + "grad_norm": 0.8728703856468201, + "learning_rate": 6.44263974116697e-07, + "loss": 0.4412, + "step": 1490 + }, + { + "epoch": 4.660751565762004, + "grad_norm": 0.7923367023468018, + "learning_rate": 6.41365574448411e-07, + "loss": 0.4401, + "step": 1491 + }, + { + "epoch": 4.663883089770355, + "grad_norm": 0.8110585808753967, + "learning_rate": 6.384727494486398e-07, + "loss": 0.4152, + "step": 1492 + }, + { + "epoch": 4.667014613778706, + "grad_norm": 0.7865241765975952, + "learning_rate": 6.355855077939385e-07, + "loss": 0.4608, + "step": 1493 + }, + { + "epoch": 4.670146137787056, + "grad_norm": 0.8125724196434021, + "learning_rate": 6.327038581441136e-07, + "loss": 0.419, + "step": 1494 + }, + { + "epoch": 4.673277661795407, + "grad_norm": 0.8024583458900452, + "learning_rate": 6.298278091422003e-07, + "loss": 0.4575, + "step": 1495 + }, + { + "epoch": 4.676409185803758, + "grad_norm": 0.7569037675857544, + "learning_rate": 6.269573694144343e-07, + "loss": 0.4257, + "step": 1496 + }, + { + "epoch": 4.679540709812109, + "grad_norm": 0.783733069896698, + "learning_rate": 6.240925475702275e-07, + "loss": 0.4361, + "step": 1497 + }, + { + "epoch": 4.682672233820459, + "grad_norm": 0.7900946736335754, + "learning_rate": 6.212333522021424e-07, + "loss": 0.4372, + "step": 1498 + }, + { + "epoch": 4.68580375782881, + "grad_norm": 0.7943285703659058, + "learning_rate": 6.183797918858667e-07, + "loss": 0.4337, + "step": 1499 + }, + { + "epoch": 4.688935281837161, + "grad_norm": 0.7793418765068054, + "learning_rate": 6.155318751801842e-07, + "loss": 0.4264, + "step": 1500 + }, + { + "epoch": 4.692066805845512, + "grad_norm": 1.0244885683059692, + "learning_rate": 6.126896106269533e-07, + "loss": 0.446, + "step": 1501 + }, + { + "epoch": 4.695198329853862, + "grad_norm": 0.8742548823356628, + "learning_rate": 6.098530067510808e-07, + "loss": 0.3953, + "step": 1502 + }, + { + "epoch": 4.698329853862213, + "grad_norm": 0.7844028472900391, + "learning_rate": 6.070220720604919e-07, + "loss": 0.402, + "step": 1503 + }, + { + "epoch": 4.701461377870563, + "grad_norm": 0.7485696077346802, + "learning_rate": 6.041968150461119e-07, + "loss": 0.4435, + "step": 1504 + }, + { + "epoch": 4.7045929018789145, + "grad_norm": 0.8526521921157837, + "learning_rate": 6.01377244181833e-07, + "loss": 0.4565, + "step": 1505 + }, + { + "epoch": 4.707724425887265, + "grad_norm": 0.7832267880439758, + "learning_rate": 5.985633679244957e-07, + "loss": 0.4489, + "step": 1506 + }, + { + "epoch": 4.710855949895616, + "grad_norm": 0.8092924952507019, + "learning_rate": 5.957551947138599e-07, + "loss": 0.4011, + "step": 1507 + }, + { + "epoch": 4.713987473903966, + "grad_norm": 0.8008314371109009, + "learning_rate": 5.92952732972579e-07, + "loss": 0.4557, + "step": 1508 + }, + { + "epoch": 4.717118997912317, + "grad_norm": 0.8025342226028442, + "learning_rate": 5.901559911061758e-07, + "loss": 0.38, + "step": 1509 + }, + { + "epoch": 4.7202505219206685, + "grad_norm": 0.8290171027183533, + "learning_rate": 5.873649775030188e-07, + "loss": 0.4019, + "step": 1510 + }, + { + "epoch": 4.723382045929019, + "grad_norm": 0.8971850872039795, + "learning_rate": 5.845797005342943e-07, + "loss": 0.4354, + "step": 1511 + }, + { + "epoch": 4.726513569937369, + "grad_norm": 0.8087455630302429, + "learning_rate": 5.818001685539843e-07, + "loss": 0.4537, + "step": 1512 + }, + { + "epoch": 4.72964509394572, + "grad_norm": 0.822287380695343, + "learning_rate": 5.790263898988355e-07, + "loss": 0.4692, + "step": 1513 + }, + { + "epoch": 4.732776617954071, + "grad_norm": 1.007889986038208, + "learning_rate": 5.762583728883428e-07, + "loss": 0.4317, + "step": 1514 + }, + { + "epoch": 4.735908141962422, + "grad_norm": 0.7517583966255188, + "learning_rate": 5.734961258247177e-07, + "loss": 0.4168, + "step": 1515 + }, + { + "epoch": 4.739039665970773, + "grad_norm": 1.1469851732254028, + "learning_rate": 5.707396569928675e-07, + "loss": 0.4412, + "step": 1516 + }, + { + "epoch": 4.742171189979123, + "grad_norm": 0.8889291286468506, + "learning_rate": 5.679889746603667e-07, + "loss": 0.4133, + "step": 1517 + }, + { + "epoch": 4.745302713987474, + "grad_norm": 0.7792209386825562, + "learning_rate": 5.652440870774342e-07, + "loss": 0.4256, + "step": 1518 + }, + { + "epoch": 4.748434237995824, + "grad_norm": 0.7862168550491333, + "learning_rate": 5.625050024769099e-07, + "loss": 0.395, + "step": 1519 + }, + { + "epoch": 4.751565762004176, + "grad_norm": 0.8305843472480774, + "learning_rate": 5.597717290742285e-07, + "loss": 0.4172, + "step": 1520 + }, + { + "epoch": 4.754697286012526, + "grad_norm": 0.8393134474754333, + "learning_rate": 5.570442750673932e-07, + "loss": 0.4438, + "step": 1521 + }, + { + "epoch": 4.757828810020877, + "grad_norm": 0.8481642603874207, + "learning_rate": 5.543226486369555e-07, + "loss": 0.4251, + "step": 1522 + }, + { + "epoch": 4.760960334029227, + "grad_norm": 0.8482305407524109, + "learning_rate": 5.516068579459849e-07, + "loss": 0.4372, + "step": 1523 + }, + { + "epoch": 4.764091858037578, + "grad_norm": 0.7428895235061646, + "learning_rate": 5.488969111400502e-07, + "loss": 0.4017, + "step": 1524 + }, + { + "epoch": 4.767223382045929, + "grad_norm": 0.9066843390464783, + "learning_rate": 5.461928163471922e-07, + "loss": 0.4263, + "step": 1525 + }, + { + "epoch": 4.77035490605428, + "grad_norm": 0.8754342794418335, + "learning_rate": 5.434945816778972e-07, + "loss": 0.4086, + "step": 1526 + }, + { + "epoch": 4.773486430062631, + "grad_norm": 0.7580026388168335, + "learning_rate": 5.408022152250783e-07, + "loss": 0.4331, + "step": 1527 + }, + { + "epoch": 4.776617954070981, + "grad_norm": 0.7960991859436035, + "learning_rate": 5.381157250640451e-07, + "loss": 0.4482, + "step": 1528 + }, + { + "epoch": 4.7797494780793315, + "grad_norm": 0.8199532628059387, + "learning_rate": 5.354351192524842e-07, + "loss": 0.4558, + "step": 1529 + }, + { + "epoch": 4.782881002087683, + "grad_norm": 0.8457065224647522, + "learning_rate": 5.327604058304312e-07, + "loss": 0.4711, + "step": 1530 + }, + { + "epoch": 4.786012526096034, + "grad_norm": 0.9078546762466431, + "learning_rate": 5.300915928202499e-07, + "loss": 0.4341, + "step": 1531 + }, + { + "epoch": 4.789144050104384, + "grad_norm": 0.7782992720603943, + "learning_rate": 5.274286882266072e-07, + "loss": 0.4115, + "step": 1532 + }, + { + "epoch": 4.792275574112735, + "grad_norm": 0.894313395023346, + "learning_rate": 5.247717000364463e-07, + "loss": 0.4415, + "step": 1533 + }, + { + "epoch": 4.7954070981210855, + "grad_norm": 0.8233445882797241, + "learning_rate": 5.221206362189682e-07, + "loss": 0.4198, + "step": 1534 + }, + { + "epoch": 4.798538622129437, + "grad_norm": 0.7654063105583191, + "learning_rate": 5.194755047256017e-07, + "loss": 0.4364, + "step": 1535 + }, + { + "epoch": 4.801670146137787, + "grad_norm": 0.7879475951194763, + "learning_rate": 5.168363134899845e-07, + "loss": 0.4366, + "step": 1536 + }, + { + "epoch": 4.804801670146138, + "grad_norm": 0.7754428386688232, + "learning_rate": 5.142030704279377e-07, + "loss": 0.4365, + "step": 1537 + }, + { + "epoch": 4.807933194154488, + "grad_norm": 0.758696436882019, + "learning_rate": 5.11575783437441e-07, + "loss": 0.4442, + "step": 1538 + }, + { + "epoch": 4.811064718162839, + "grad_norm": 0.7870402336120605, + "learning_rate": 5.089544603986085e-07, + "loss": 0.4298, + "step": 1539 + }, + { + "epoch": 4.81419624217119, + "grad_norm": 0.7607670426368713, + "learning_rate": 5.063391091736691e-07, + "loss": 0.435, + "step": 1540 + }, + { + "epoch": 4.817327766179541, + "grad_norm": 0.8175606727600098, + "learning_rate": 5.037297376069388e-07, + "loss": 0.4325, + "step": 1541 + }, + { + "epoch": 4.820459290187891, + "grad_norm": 0.9083423018455505, + "learning_rate": 5.011263535248001e-07, + "loss": 0.3847, + "step": 1542 + }, + { + "epoch": 4.823590814196242, + "grad_norm": 0.7435231804847717, + "learning_rate": 4.985289647356731e-07, + "loss": 0.4818, + "step": 1543 + }, + { + "epoch": 4.826722338204593, + "grad_norm": 1.5190348625183105, + "learning_rate": 4.9593757903e-07, + "loss": 0.4068, + "step": 1544 + }, + { + "epoch": 4.829853862212944, + "grad_norm": 0.9122394919395447, + "learning_rate": 4.933522041802166e-07, + "loss": 0.4701, + "step": 1545 + }, + { + "epoch": 4.832985386221294, + "grad_norm": 0.9334122538566589, + "learning_rate": 4.9077284794073e-07, + "loss": 0.4178, + "step": 1546 + }, + { + "epoch": 4.836116910229645, + "grad_norm": 0.8561045527458191, + "learning_rate": 4.881995180478949e-07, + "loss": 0.4373, + "step": 1547 + }, + { + "epoch": 4.839248434237996, + "grad_norm": 0.8061811923980713, + "learning_rate": 4.85632222219991e-07, + "loss": 0.4414, + "step": 1548 + }, + { + "epoch": 4.8423799582463465, + "grad_norm": 0.7866719961166382, + "learning_rate": 4.830709681572008e-07, + "loss": 0.438, + "step": 1549 + }, + { + "epoch": 4.845511482254698, + "grad_norm": 0.8633884787559509, + "learning_rate": 4.805157635415852e-07, + "loss": 0.4639, + "step": 1550 + }, + { + "epoch": 4.848643006263048, + "grad_norm": 1.125654935836792, + "learning_rate": 4.779666160370596e-07, + "loss": 0.4364, + "step": 1551 + }, + { + "epoch": 4.851774530271399, + "grad_norm": 0.8371828198432922, + "learning_rate": 4.7542353328937473e-07, + "loss": 0.4436, + "step": 1552 + }, + { + "epoch": 4.854906054279749, + "grad_norm": 0.8166123628616333, + "learning_rate": 4.7288652292608736e-07, + "loss": 0.4382, + "step": 1553 + }, + { + "epoch": 4.8580375782881005, + "grad_norm": 0.8664330244064331, + "learning_rate": 4.7035559255654393e-07, + "loss": 0.4314, + "step": 1554 + }, + { + "epoch": 4.861169102296451, + "grad_norm": 0.8051781058311462, + "learning_rate": 4.6783074977185485e-07, + "loss": 0.4522, + "step": 1555 + }, + { + "epoch": 4.864300626304802, + "grad_norm": 0.7385572791099548, + "learning_rate": 4.653120021448701e-07, + "loss": 0.4254, + "step": 1556 + }, + { + "epoch": 4.867432150313152, + "grad_norm": 0.7310816645622253, + "learning_rate": 4.6279935723016033e-07, + "loss": 0.4277, + "step": 1557 + }, + { + "epoch": 4.870563674321503, + "grad_norm": 0.7682080864906311, + "learning_rate": 4.602928225639899e-07, + "loss": 0.422, + "step": 1558 + }, + { + "epoch": 4.8736951983298535, + "grad_norm": 0.8093141317367554, + "learning_rate": 4.577924056642985e-07, + "loss": 0.4376, + "step": 1559 + }, + { + "epoch": 4.876826722338205, + "grad_norm": 0.8006629943847656, + "learning_rate": 4.5529811403067503e-07, + "loss": 0.4483, + "step": 1560 + }, + { + "epoch": 4.879958246346555, + "grad_norm": 0.7647503614425659, + "learning_rate": 4.528099551443377e-07, + "loss": 0.418, + "step": 1561 + }, + { + "epoch": 4.883089770354906, + "grad_norm": 0.8467709422111511, + "learning_rate": 4.503279364681104e-07, + "loss": 0.4272, + "step": 1562 + }, + { + "epoch": 4.886221294363256, + "grad_norm": 0.8667131662368774, + "learning_rate": 4.4785206544640055e-07, + "loss": 0.4347, + "step": 1563 + }, + { + "epoch": 4.8893528183716075, + "grad_norm": 0.8670737147331238, + "learning_rate": 4.453823495051748e-07, + "loss": 0.4313, + "step": 1564 + }, + { + "epoch": 4.892484342379959, + "grad_norm": 0.8131253719329834, + "learning_rate": 4.4291879605194157e-07, + "loss": 0.4335, + "step": 1565 + }, + { + "epoch": 4.895615866388309, + "grad_norm": 0.9633654356002808, + "learning_rate": 4.404614124757242e-07, + "loss": 0.4637, + "step": 1566 + }, + { + "epoch": 4.89874739039666, + "grad_norm": 0.9058681726455688, + "learning_rate": 4.3801020614704174e-07, + "loss": 0.4503, + "step": 1567 + }, + { + "epoch": 4.90187891440501, + "grad_norm": 0.7923601865768433, + "learning_rate": 4.3556518441788405e-07, + "loss": 0.4526, + "step": 1568 + }, + { + "epoch": 4.9050104384133615, + "grad_norm": 0.7733371257781982, + "learning_rate": 4.331263546216913e-07, + "loss": 0.437, + "step": 1569 + }, + { + "epoch": 4.908141962421712, + "grad_norm": 0.7897522449493408, + "learning_rate": 4.306937240733336e-07, + "loss": 0.4213, + "step": 1570 + }, + { + "epoch": 4.911273486430063, + "grad_norm": 0.7857375741004944, + "learning_rate": 4.282673000690865e-07, + "loss": 0.4576, + "step": 1571 + }, + { + "epoch": 4.914405010438413, + "grad_norm": 0.7968363761901855, + "learning_rate": 4.258470898866102e-07, + "loss": 0.4383, + "step": 1572 + }, + { + "epoch": 4.917536534446764, + "grad_norm": 0.762876033782959, + "learning_rate": 4.2343310078492626e-07, + "loss": 0.4129, + "step": 1573 + }, + { + "epoch": 4.920668058455115, + "grad_norm": 0.9070468544960022, + "learning_rate": 4.21025340004399e-07, + "loss": 0.3994, + "step": 1574 + }, + { + "epoch": 4.923799582463466, + "grad_norm": 0.8299228549003601, + "learning_rate": 4.186238147667113e-07, + "loss": 0.4461, + "step": 1575 + }, + { + "epoch": 4.926931106471816, + "grad_norm": 0.8604512214660645, + "learning_rate": 4.162285322748441e-07, + "loss": 0.4447, + "step": 1576 + }, + { + "epoch": 4.930062630480167, + "grad_norm": 0.8441461324691772, + "learning_rate": 4.1383949971305207e-07, + "loss": 0.4547, + "step": 1577 + }, + { + "epoch": 4.933194154488517, + "grad_norm": 0.7836902737617493, + "learning_rate": 4.114567242468473e-07, + "loss": 0.4311, + "step": 1578 + }, + { + "epoch": 4.9363256784968685, + "grad_norm": 0.7957454919815063, + "learning_rate": 4.0908021302297205e-07, + "loss": 0.4074, + "step": 1579 + }, + { + "epoch": 4.939457202505219, + "grad_norm": 0.772109866142273, + "learning_rate": 4.0670997316938216e-07, + "loss": 0.4696, + "step": 1580 + }, + { + "epoch": 4.94258872651357, + "grad_norm": 1.3906599283218384, + "learning_rate": 4.043460117952214e-07, + "loss": 0.4379, + "step": 1581 + }, + { + "epoch": 4.945720250521921, + "grad_norm": 0.9287625551223755, + "learning_rate": 4.0198833599080415e-07, + "loss": 0.4576, + "step": 1582 + }, + { + "epoch": 4.948851774530271, + "grad_norm": 0.8610300421714783, + "learning_rate": 3.99636952827592e-07, + "loss": 0.4158, + "step": 1583 + }, + { + "epoch": 4.9519832985386225, + "grad_norm": 0.7845875024795532, + "learning_rate": 3.972918693581712e-07, + "loss": 0.4428, + "step": 1584 + }, + { + "epoch": 4.955114822546973, + "grad_norm": 0.8758688569068909, + "learning_rate": 3.949530926162354e-07, + "loss": 0.4113, + "step": 1585 + }, + { + "epoch": 4.958246346555324, + "grad_norm": 0.797717809677124, + "learning_rate": 3.926206296165599e-07, + "loss": 0.4181, + "step": 1586 + }, + { + "epoch": 4.961377870563674, + "grad_norm": 0.7839358448982239, + "learning_rate": 3.9029448735498495e-07, + "loss": 0.4218, + "step": 1587 + }, + { + "epoch": 4.964509394572025, + "grad_norm": 0.8518044352531433, + "learning_rate": 3.8797467280839216e-07, + "loss": 0.4425, + "step": 1588 + }, + { + "epoch": 4.967640918580376, + "grad_norm": 0.7976752519607544, + "learning_rate": 3.856611929346837e-07, + "loss": 0.4736, + "step": 1589 + }, + { + "epoch": 4.970772442588727, + "grad_norm": 0.9000734090805054, + "learning_rate": 3.8335405467276184e-07, + "loss": 0.4282, + "step": 1590 + }, + { + "epoch": 4.973903966597077, + "grad_norm": 0.8424035906791687, + "learning_rate": 3.810532649425089e-07, + "loss": 0.4443, + "step": 1591 + }, + { + "epoch": 4.977035490605428, + "grad_norm": 0.8014306426048279, + "learning_rate": 3.7875883064476564e-07, + "loss": 0.4617, + "step": 1592 + }, + { + "epoch": 4.980167014613778, + "grad_norm": 0.7423905730247498, + "learning_rate": 3.76470758661312e-07, + "loss": 0.468, + "step": 1593 + }, + { + "epoch": 4.98329853862213, + "grad_norm": 0.7726851105690002, + "learning_rate": 3.741890558548414e-07, + "loss": 0.3824, + "step": 1594 + }, + { + "epoch": 4.98643006263048, + "grad_norm": 0.7915946245193481, + "learning_rate": 3.719137290689479e-07, + "loss": 0.4091, + "step": 1595 + } + ], + "logging_steps": 1, + "max_steps": 1914, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 319, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 4.832478095742704e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-1595/training_args.bin b/checkpoint-1595/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8067ee9c1c0bc752bdfd00cfcaf1a6e717d2356b --- /dev/null +++ b/checkpoint-1595/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c650156a192cae0a2070c4323ee8a93e9b52fb76041d59ae0633b98389585727 +size 7928 diff --git a/checkpoint-1595/zero_to_fp32.py b/checkpoint-1595/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-1595/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-1914/README.md b/checkpoint-1914/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f4a3934800eeb082a0cb833d7b6af4f68eed3615 --- /dev/null +++ b/checkpoint-1914/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3_3-Nemotron-Super-49B-v1 +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.15.0 \ No newline at end of file diff --git a/checkpoint-1914/adapter_config.json b/checkpoint-1914/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..1edb33780e2306c6b19fd727be8e9b8b35f237c4 --- /dev/null +++ b/checkpoint-1914/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3_3-Nemotron-Super-49B-v1", + "bias": "none", + "corda_config": null, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "k_proj", + "down_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-1914/adapter_model.safetensors b/checkpoint-1914/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7d16d638ba5f1f0d6e17a6a0445d8d989c8ed984 --- /dev/null +++ b/checkpoint-1914/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8cb98185ecb57fad2b2cd35f0c47cae364129464746319ff1be6d8581c7a3a9 +size 9016826528 diff --git a/checkpoint-1914/global_step1910/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-1914/global_step1910/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..0bdf3673ea7afbd7b9f53dff3f76cdd1c715d052 --- /dev/null +++ b/checkpoint-1914/global_step1910/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d69c4a6a0d43cfa67e8b570fd1fd08d5a97fbcdbbd74c051b9246440c74c3a4e +size 27050164444 diff --git a/checkpoint-1914/global_step1910/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-1914/global_step1910/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..d472eae6fe89d7db48ba709625dcc331dc9031a7 --- /dev/null +++ b/checkpoint-1914/global_step1910/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:55607f1577a54d3930469747c642afe4e3b759952305d0b7ccc74ffdc9ff4d18 +size 27050169884 diff --git a/checkpoint-1914/global_step1910/mp_rank_00_model_states.pt b/checkpoint-1914/global_step1910/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a820bbbcb6acfb8adba9a85c54a1d58fb399e496 --- /dev/null +++ b/checkpoint-1914/global_step1910/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2febc93df50ccbf08ff22419a2955756d96439de06bfba0e7145eea9b610077f +size 9776788601 diff --git a/checkpoint-1914/latest b/checkpoint-1914/latest new file mode 100644 index 0000000000000000000000000000000000000000..44970e72d9272500910003990fd0742fc709640f --- /dev/null +++ b/checkpoint-1914/latest @@ -0,0 +1 @@ +global_step1910 \ No newline at end of file diff --git a/checkpoint-1914/rng_state_0.pth b/checkpoint-1914/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..20f781331db199355d352dd623f2cfa1bd655144 --- /dev/null +++ b/checkpoint-1914/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c76958bd6fb582aa7128013f6648323584e6359a68e13537634a12b640e33b3b +size 14512 diff --git a/checkpoint-1914/rng_state_1.pth b/checkpoint-1914/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..6b6d09621a9f9c529ded1d3367386d4b7dbfa3d6 --- /dev/null +++ b/checkpoint-1914/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0468a880821a2b59a3405b34372a9bfe1bffa49d6f45e70fe1a332aaf2506fd4 +size 14512 diff --git a/checkpoint-1914/scheduler.pt b/checkpoint-1914/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..ce8debe0515e0d91db0a76ec2fd8617dad934be3 --- /dev/null +++ b/checkpoint-1914/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:518502dfa64b1b4e7140cd5c1d3de6ab9c448c69ea1aa3d6aa99b8b0f388bb09 +size 1064 diff --git a/checkpoint-1914/special_tokens_map.json b/checkpoint-1914/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-1914/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-1914/tokenizer.json b/checkpoint-1914/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-1914/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-1914/tokenizer_config.json b/checkpoint-1914/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..edd01b980c1db496ea102a51c972ee8f5d1a2c74 --- /dev/null +++ b/checkpoint-1914/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "chat_template": "{{- bos_token }}{%- if messages[0]['role'] == 'system' %}{%- set system_message = messages[0]['content']|trim %}{%- set messages = messages[1:] %}{%- else %}{%- set system_message = \"\" %}{%- endif %}{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}{{- system_message }}{{- \"<|eot_id|>\" }}{%- for message in messages %}{%- if message['role'] == 'assistant' and '' in message['content'] %}{%- set content = message['content'].split('')[-1].lstrip() %}{%- else %}{%- set content = message['content'] %}{%- endif %}{{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n' + content | trim + '<|eot_id|>' }}{%- endfor %}{%- if add_generation_prompt %}{{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}{%- endif %}", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-1914/trainer_state.json b/checkpoint-1914/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..52e4d90e074c683cf3ff19209008804a30cc5990 --- /dev/null +++ b/checkpoint-1914/trainer_state.json @@ -0,0 +1,13431 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 5.98329853862213, + "eval_steps": 500, + "global_step": 1914, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.003131524008350731, + "grad_norm": 13.917898178100586, + "learning_rate": 5.0000000000000004e-08, + "loss": 4.1051, + "step": 1 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 17.327869415283203, + "learning_rate": 1.0000000000000001e-07, + "loss": 4.1048, + "step": 2 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 14.063946723937988, + "learning_rate": 1.5000000000000002e-07, + "loss": 4.0741, + "step": 3 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 16.817699432373047, + "learning_rate": 2.0000000000000002e-07, + "loss": 4.2002, + "step": 4 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 14.47036361694336, + "learning_rate": 2.5000000000000004e-07, + "loss": 4.2652, + "step": 5 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 14.474193572998047, + "learning_rate": 3.0000000000000004e-07, + "loss": 4.0888, + "step": 6 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 14.865458488464355, + "learning_rate": 3.5000000000000004e-07, + "loss": 4.0014, + "step": 7 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 15.338888168334961, + "learning_rate": 4.0000000000000003e-07, + "loss": 4.13, + "step": 8 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 15.154336929321289, + "learning_rate": 4.5000000000000003e-07, + "loss": 4.2493, + "step": 9 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 15.919597625732422, + "learning_rate": 5.000000000000001e-07, + "loss": 4.0535, + "step": 10 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 14.981926918029785, + "learning_rate": 5.5e-07, + "loss": 3.9064, + "step": 11 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 13.36101245880127, + "learning_rate": 6.000000000000001e-07, + "loss": 4.1939, + "step": 12 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 15.58773422241211, + "learning_rate": 6.5e-07, + "loss": 4.18, + "step": 13 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 13.560139656066895, + "learning_rate": 7.000000000000001e-07, + "loss": 3.9414, + "step": 14 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 12.307971954345703, + "learning_rate": 7.5e-07, + "loss": 3.8836, + "step": 15 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 14.533182144165039, + "learning_rate": 8.000000000000001e-07, + "loss": 4.1551, + "step": 16 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 13.453729629516602, + "learning_rate": 8.500000000000001e-07, + "loss": 4.0048, + "step": 17 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 13.45992374420166, + "learning_rate": 9.000000000000001e-07, + "loss": 4.0745, + "step": 18 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 11.857145309448242, + "learning_rate": 9.500000000000001e-07, + "loss": 3.9871, + "step": 19 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 11.872294425964355, + "learning_rate": 1.0000000000000002e-06, + "loss": 3.8959, + "step": 20 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.969825744628906, + "learning_rate": 1.0500000000000001e-06, + "loss": 4.0308, + "step": 21 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 12.33769416809082, + "learning_rate": 1.1e-06, + "loss": 3.9341, + "step": 22 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 12.669405937194824, + "learning_rate": 1.1500000000000002e-06, + "loss": 3.8511, + "step": 23 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 10.677213668823242, + "learning_rate": 1.2000000000000002e-06, + "loss": 3.7764, + "step": 24 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 10.366402626037598, + "learning_rate": 1.25e-06, + "loss": 3.5291, + "step": 25 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 11.211421012878418, + "learning_rate": 1.3e-06, + "loss": 3.5765, + "step": 26 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 11.313716888427734, + "learning_rate": 1.3500000000000002e-06, + "loss": 3.4849, + "step": 27 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 10.41294002532959, + "learning_rate": 1.4000000000000001e-06, + "loss": 3.2653, + "step": 28 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 10.40064525604248, + "learning_rate": 1.45e-06, + "loss": 3.3384, + "step": 29 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 10.05427074432373, + "learning_rate": 1.5e-06, + "loss": 3.2257, + "step": 30 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 9.583163261413574, + "learning_rate": 1.5500000000000002e-06, + "loss": 3.1371, + "step": 31 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 10.09977912902832, + "learning_rate": 1.6000000000000001e-06, + "loss": 3.0658, + "step": 32 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 9.271486282348633, + "learning_rate": 1.6500000000000003e-06, + "loss": 2.9693, + "step": 33 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 10.687992095947266, + "learning_rate": 1.7000000000000002e-06, + "loss": 2.95, + "step": 34 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 8.762290000915527, + "learning_rate": 1.75e-06, + "loss": 2.8286, + "step": 35 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 10.13785171508789, + "learning_rate": 1.8000000000000001e-06, + "loss": 2.3664, + "step": 36 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 18.301353454589844, + "learning_rate": 1.85e-06, + "loss": 2.5533, + "step": 37 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 11.490377426147461, + "learning_rate": 1.9000000000000002e-06, + "loss": 2.6133, + "step": 38 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 15.614163398742676, + "learning_rate": 1.9500000000000004e-06, + "loss": 2.3596, + "step": 39 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 17.757442474365234, + "learning_rate": 2.0000000000000003e-06, + "loss": 2.3491, + "step": 40 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 17.18431854248047, + "learning_rate": 2.05e-06, + "loss": 2.2361, + "step": 41 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 16.149789810180664, + "learning_rate": 2.1000000000000002e-06, + "loss": 2.1457, + "step": 42 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 15.256914138793945, + "learning_rate": 2.15e-06, + "loss": 2.12, + "step": 43 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 15.537406921386719, + "learning_rate": 2.2e-06, + "loss": 2.1877, + "step": 44 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 7.947713851928711, + "learning_rate": 2.25e-06, + "loss": 2.1648, + "step": 45 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 8.818676948547363, + "learning_rate": 2.3000000000000004e-06, + "loss": 2.134, + "step": 46 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 5.175768852233887, + "learning_rate": 2.35e-06, + "loss": 2.0796, + "step": 47 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 6.750611305236816, + "learning_rate": 2.4000000000000003e-06, + "loss": 1.9174, + "step": 48 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 6.2147979736328125, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.8065, + "step": 49 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 13.291611671447754, + "learning_rate": 2.5e-06, + "loss": 1.7061, + "step": 50 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 7.251201629638672, + "learning_rate": 2.55e-06, + "loss": 1.7924, + "step": 51 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 5.2126054763793945, + "learning_rate": 2.6e-06, + "loss": 1.6735, + "step": 52 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 5.435528755187988, + "learning_rate": 2.6500000000000005e-06, + "loss": 1.6265, + "step": 53 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 4.505807399749756, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.4851, + "step": 54 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 5.128388404846191, + "learning_rate": 2.7500000000000004e-06, + "loss": 1.5832, + "step": 55 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 16.935827255249023, + "learning_rate": 2.8000000000000003e-06, + "loss": 1.6553, + "step": 56 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 3.664458990097046, + "learning_rate": 2.85e-06, + "loss": 1.5, + "step": 57 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 7.763802528381348, + "learning_rate": 2.9e-06, + "loss": 1.367, + "step": 58 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 3.2216155529022217, + "learning_rate": 2.95e-06, + "loss": 1.3863, + "step": 59 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 4.384445667266846, + "learning_rate": 3e-06, + "loss": 1.4247, + "step": 60 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 4.8080878257751465, + "learning_rate": 3.05e-06, + "loss": 1.3257, + "step": 61 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 4.154761791229248, + "learning_rate": 3.1000000000000004e-06, + "loss": 1.321, + "step": 62 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 6.4742112159729, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.2823, + "step": 63 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 2.583422899246216, + "learning_rate": 3.2000000000000003e-06, + "loss": 1.2136, + "step": 64 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 4.1933488845825195, + "learning_rate": 3.2500000000000002e-06, + "loss": 1.1855, + "step": 65 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 4.11049747467041, + "learning_rate": 3.3000000000000006e-06, + "loss": 1.2389, + "step": 66 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 2.264458417892456, + "learning_rate": 3.3500000000000005e-06, + "loss": 1.0651, + "step": 67 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 2.5408174991607666, + "learning_rate": 3.4000000000000005e-06, + "loss": 1.1389, + "step": 68 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 7.82421350479126, + "learning_rate": 3.45e-06, + "loss": 1.0956, + "step": 69 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 3.070939064025879, + "learning_rate": 3.5e-06, + "loss": 1.0451, + "step": 70 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 2.6310527324676514, + "learning_rate": 3.5500000000000003e-06, + "loss": 1.0538, + "step": 71 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 7.630155563354492, + "learning_rate": 3.6000000000000003e-06, + "loss": 1.0052, + "step": 72 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 6.950636863708496, + "learning_rate": 3.65e-06, + "loss": 1.0473, + "step": 73 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 2.2703945636749268, + "learning_rate": 3.7e-06, + "loss": 1.0576, + "step": 74 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 3.3817710876464844, + "learning_rate": 3.7500000000000005e-06, + "loss": 1.0177, + "step": 75 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 7.266414642333984, + "learning_rate": 3.8000000000000005e-06, + "loss": 1.0645, + "step": 76 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 5.782608509063721, + "learning_rate": 3.85e-06, + "loss": 1.0162, + "step": 77 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 2.7938575744628906, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9664, + "step": 78 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 6.681935787200928, + "learning_rate": 3.95e-06, + "loss": 0.953, + "step": 79 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 2.253279209136963, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9568, + "step": 80 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 1.4875826835632324, + "learning_rate": 4.05e-06, + "loss": 0.9448, + "step": 81 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 2.4987940788269043, + "learning_rate": 4.1e-06, + "loss": 0.9393, + "step": 82 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 4.712948322296143, + "learning_rate": 4.15e-06, + "loss": 0.9532, + "step": 83 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 6.9030632972717285, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.96, + "step": 84 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 3.4780967235565186, + "learning_rate": 4.25e-06, + "loss": 0.8993, + "step": 85 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.526064395904541, + "learning_rate": 4.3e-06, + "loss": 0.9021, + "step": 86 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 10.727686882019043, + "learning_rate": 4.350000000000001e-06, + "loss": 0.856, + "step": 87 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 12.483160972595215, + "learning_rate": 4.4e-06, + "loss": 0.9357, + "step": 88 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 6.544492244720459, + "learning_rate": 4.450000000000001e-06, + "loss": 0.9168, + "step": 89 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 1.178139567375183, + "learning_rate": 4.5e-06, + "loss": 0.8748, + "step": 90 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 1.711506962776184, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8425, + "step": 91 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 3.281747341156006, + "learning_rate": 4.600000000000001e-06, + "loss": 0.8491, + "step": 92 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 2.2964377403259277, + "learning_rate": 4.65e-06, + "loss": 0.8038, + "step": 93 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 1.959700345993042, + "learning_rate": 4.7e-06, + "loss": 0.8439, + "step": 94 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 3.979384183883667, + "learning_rate": 4.75e-06, + "loss": 0.8839, + "step": 95 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 1.4721262454986572, + "learning_rate": 4.800000000000001e-06, + "loss": 0.845, + "step": 96 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 2.862248659133911, + "learning_rate": 4.85e-06, + "loss": 0.7748, + "step": 97 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 3.7439088821411133, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8145, + "step": 98 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 1.6654618978500366, + "learning_rate": 4.95e-06, + "loss": 0.8326, + "step": 99 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 7.8437581062316895, + "learning_rate": 5e-06, + "loss": 0.8666, + "step": 100 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 6.429738521575928, + "learning_rate": 4.999996250830422e-06, + "loss": 0.836, + "step": 101 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 2.6017794609069824, + "learning_rate": 4.9999850033329326e-06, + "loss": 0.7785, + "step": 102 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 1.0575449466705322, + "learning_rate": 4.999966257541265e-06, + "loss": 0.7639, + "step": 103 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 2.6932010650634766, + "learning_rate": 4.999940013511647e-06, + "loss": 0.8214, + "step": 104 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 2.925288438796997, + "learning_rate": 4.999906271322792e-06, + "loss": 0.8797, + "step": 105 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.3570607900619507, + "learning_rate": 4.9998650310759035e-06, + "loss": 0.792, + "step": 106 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 5.126713752746582, + "learning_rate": 4.999816292894676e-06, + "loss": 0.8352, + "step": 107 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 1.8966432809829712, + "learning_rate": 4.99976005692529e-06, + "loss": 0.7663, + "step": 108 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 1.3100829124450684, + "learning_rate": 4.999696323336418e-06, + "loss": 0.771, + "step": 109 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 2.4025354385375977, + "learning_rate": 4.999625092319218e-06, + "loss": 0.7618, + "step": 110 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.130232810974121, + "learning_rate": 4.999546364087334e-06, + "loss": 0.7705, + "step": 111 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 3.430262327194214, + "learning_rate": 4.999460138876901e-06, + "loss": 0.77, + "step": 112 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 1.1272103786468506, + "learning_rate": 4.999366416946536e-06, + "loss": 0.7133, + "step": 113 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 1.1740471124649048, + "learning_rate": 4.999265198577342e-06, + "loss": 0.7684, + "step": 114 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.3138248920440674, + "learning_rate": 4.999156484072907e-06, + "loss": 0.7888, + "step": 115 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 1.061711311340332, + "learning_rate": 4.999040273759304e-06, + "loss": 0.7484, + "step": 116 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.4682390689849854, + "learning_rate": 4.998916567985083e-06, + "loss": 0.7296, + "step": 117 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 2.884068250656128, + "learning_rate": 4.998785367121284e-06, + "loss": 0.7662, + "step": 118 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9812761545181274, + "learning_rate": 4.9986466715614205e-06, + "loss": 0.7307, + "step": 119 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 2.2237496376037598, + "learning_rate": 4.998500481721484e-06, + "loss": 0.6761, + "step": 120 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 1.4004178047180176, + "learning_rate": 4.998346798039952e-06, + "loss": 0.7505, + "step": 121 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 5.54975700378418, + "learning_rate": 4.99818562097777e-06, + "loss": 0.7615, + "step": 122 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 6.17140531539917, + "learning_rate": 4.9980169510183624e-06, + "loss": 0.7002, + "step": 123 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 4.974380016326904, + "learning_rate": 4.997840788667628e-06, + "loss": 0.7449, + "step": 124 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 1.4133399724960327, + "learning_rate": 4.997657134453937e-06, + "loss": 0.7442, + "step": 125 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 1.868915319442749, + "learning_rate": 4.9974659889281295e-06, + "loss": 0.7104, + "step": 126 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 1.2599350214004517, + "learning_rate": 4.997267352663514e-06, + "loss": 0.7385, + "step": 127 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.4353641271591187, + "learning_rate": 4.997061226255869e-06, + "loss": 0.7081, + "step": 128 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 3.2492141723632812, + "learning_rate": 4.996847610323437e-06, + "loss": 0.7859, + "step": 129 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 9.599719047546387, + "learning_rate": 4.996626505506923e-06, + "loss": 0.7241, + "step": 130 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 10.053650856018066, + "learning_rate": 4.996397912469494e-06, + "loss": 0.6841, + "step": 131 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 1.323876976966858, + "learning_rate": 4.996161831896777e-06, + "loss": 0.7317, + "step": 132 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 1.4180598258972168, + "learning_rate": 4.9959182644968594e-06, + "loss": 0.692, + "step": 133 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 1.2194396257400513, + "learning_rate": 4.99566721100028e-06, + "loss": 0.7068, + "step": 134 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0984960794448853, + "learning_rate": 4.995408672160031e-06, + "loss": 0.6946, + "step": 135 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 1.9341071844100952, + "learning_rate": 4.995142648751561e-06, + "loss": 0.7467, + "step": 136 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 1.9960932731628418, + "learning_rate": 4.9948691415727594e-06, + "loss": 0.7379, + "step": 137 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.8743917942047119, + "learning_rate": 4.994588151443968e-06, + "loss": 0.66, + "step": 138 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8655261993408203, + "learning_rate": 4.99429967920797e-06, + "loss": 0.6646, + "step": 139 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 5.462070941925049, + "learning_rate": 4.994003725729992e-06, + "loss": 0.643, + "step": 140 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 2.1401469707489014, + "learning_rate": 4.993700291897695e-06, + "loss": 0.6639, + "step": 141 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 1.8219833374023438, + "learning_rate": 4.9933893786211815e-06, + "loss": 0.6673, + "step": 142 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 1.641079306602478, + "learning_rate": 4.993070986832984e-06, + "loss": 0.658, + "step": 143 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 1.1739819049835205, + "learning_rate": 4.992745117488066e-06, + "loss": 0.6826, + "step": 144 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 2.309185743331909, + "learning_rate": 4.9924117715638185e-06, + "loss": 0.6536, + "step": 145 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 1.09304940700531, + "learning_rate": 4.99207095006006e-06, + "loss": 0.721, + "step": 146 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.9056984186172485, + "learning_rate": 4.991722653999025e-06, + "loss": 0.7019, + "step": 147 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.8440625667572021, + "learning_rate": 4.991366884425374e-06, + "loss": 0.707, + "step": 148 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 1.2244676351547241, + "learning_rate": 4.991003642406177e-06, + "loss": 0.6407, + "step": 149 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9258589744567871, + "learning_rate": 4.99063292903092e-06, + "loss": 0.6954, + "step": 150 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 4.176390647888184, + "learning_rate": 4.990254745411496e-06, + "loss": 0.6812, + "step": 151 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 1.4322530031204224, + "learning_rate": 4.989869092682205e-06, + "loss": 0.6808, + "step": 152 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.8017717003822327, + "learning_rate": 4.989475971999748e-06, + "loss": 0.687, + "step": 153 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 1.5641374588012695, + "learning_rate": 4.989075384543228e-06, + "loss": 0.6599, + "step": 154 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 1.1522141695022583, + "learning_rate": 4.98866733151414e-06, + "loss": 0.6546, + "step": 155 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8593171238899231, + "learning_rate": 4.988251814136372e-06, + "loss": 0.6857, + "step": 156 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 2.668159246444702, + "learning_rate": 4.9878288336562e-06, + "loss": 0.661, + "step": 157 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9953671097755432, + "learning_rate": 4.987398391342285e-06, + "loss": 0.6512, + "step": 158 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.042872667312622, + "learning_rate": 4.986960488485667e-06, + "loss": 0.6311, + "step": 159 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.9070663452148438, + "learning_rate": 4.9865151263997645e-06, + "loss": 0.675, + "step": 160 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.8460433483123779, + "learning_rate": 4.986062306420367e-06, + "loss": 0.6635, + "step": 161 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 1.2639834880828857, + "learning_rate": 4.985602029905635e-06, + "loss": 0.6327, + "step": 162 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.8775074481964111, + "learning_rate": 4.985134298236091e-06, + "loss": 0.644, + "step": 163 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 1.2031961679458618, + "learning_rate": 4.98465911281462e-06, + "loss": 0.6254, + "step": 164 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.892494797706604, + "learning_rate": 4.984176475066463e-06, + "loss": 0.7122, + "step": 165 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 2.7122485637664795, + "learning_rate": 4.983686386439212e-06, + "loss": 0.6679, + "step": 166 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9344426989555359, + "learning_rate": 4.983188848402806e-06, + "loss": 0.6319, + "step": 167 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 1.4093577861785889, + "learning_rate": 4.982683862449531e-06, + "loss": 0.6425, + "step": 168 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 1.1285009384155273, + "learning_rate": 4.982171430094007e-06, + "loss": 0.6298, + "step": 169 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.952778935432434, + "learning_rate": 4.981651552873193e-06, + "loss": 0.7066, + "step": 170 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 5.133765697479248, + "learning_rate": 4.981124232346374e-06, + "loss": 0.6634, + "step": 171 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9770542979240417, + "learning_rate": 4.980589470095161e-06, + "loss": 0.7121, + "step": 172 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.8414323925971985, + "learning_rate": 4.980047267723487e-06, + "loss": 0.6397, + "step": 173 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 1.9173879623413086, + "learning_rate": 4.979497626857596e-06, + "loss": 0.6228, + "step": 174 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 1.0823363065719604, + "learning_rate": 4.978940549146048e-06, + "loss": 0.6475, + "step": 175 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 3.715353488922119, + "learning_rate": 4.978376036259706e-06, + "loss": 0.7127, + "step": 176 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.981584370136261, + "learning_rate": 4.9778040898917325e-06, + "loss": 0.6468, + "step": 177 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.70566987991333, + "learning_rate": 4.977224711757587e-06, + "loss": 0.6476, + "step": 178 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9217923283576965, + "learning_rate": 4.976637903595019e-06, + "loss": 0.6731, + "step": 179 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.8994677662849426, + "learning_rate": 4.976043667164063e-06, + "loss": 0.6562, + "step": 180 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 1.1613017320632935, + "learning_rate": 4.975442004247034e-06, + "loss": 0.6417, + "step": 181 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 1.6041977405548096, + "learning_rate": 4.974832916648521e-06, + "loss": 0.6029, + "step": 182 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.7978405952453613, + "learning_rate": 4.974216406195383e-06, + "loss": 0.6269, + "step": 183 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.6021920442581177, + "learning_rate": 4.973592474736739e-06, + "loss": 0.6149, + "step": 184 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8973568677902222, + "learning_rate": 4.972961124143971e-06, + "loss": 0.6648, + "step": 185 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 1.9432591199874878, + "learning_rate": 4.972322356310711e-06, + "loss": 0.6299, + "step": 186 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 4.457028388977051, + "learning_rate": 4.971676173152839e-06, + "loss": 0.656, + "step": 187 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 2.0989716053009033, + "learning_rate": 4.971022576608473e-06, + "loss": 0.6539, + "step": 188 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 1.0646967887878418, + "learning_rate": 4.97036156863797e-06, + "loss": 0.6727, + "step": 189 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.6522265672683716, + "learning_rate": 4.969693151223914e-06, + "loss": 0.6643, + "step": 190 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.7503505945205688, + "learning_rate": 4.969017326371115e-06, + "loss": 0.6402, + "step": 191 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 1.2341989278793335, + "learning_rate": 4.968334096106597e-06, + "loss": 0.6413, + "step": 192 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 3.089054584503174, + "learning_rate": 4.967643462479597e-06, + "loss": 0.6825, + "step": 193 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 2.711623430252075, + "learning_rate": 4.966945427561557e-06, + "loss": 0.65, + "step": 194 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 4.641184329986572, + "learning_rate": 4.966239993446118e-06, + "loss": 0.6229, + "step": 195 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.7984074354171753, + "learning_rate": 4.965527162249114e-06, + "loss": 0.6473, + "step": 196 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 1.1643115282058716, + "learning_rate": 4.964806936108566e-06, + "loss": 0.6404, + "step": 197 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 2.1877920627593994, + "learning_rate": 4.9640793171846725e-06, + "loss": 0.6185, + "step": 198 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.7970566749572754, + "learning_rate": 4.963344307659807e-06, + "loss": 0.634, + "step": 199 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 1.6014361381530762, + "learning_rate": 4.96260190973851e-06, + "loss": 0.6562, + "step": 200 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.8743320107460022, + "learning_rate": 4.961852125647482e-06, + "loss": 0.6133, + "step": 201 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 1.9526551961898804, + "learning_rate": 4.961094957635578e-06, + "loss": 0.6451, + "step": 202 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 3.6597347259521484, + "learning_rate": 4.960330407973798e-06, + "loss": 0.6386, + "step": 203 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 1.7180207967758179, + "learning_rate": 4.959558478955283e-06, + "loss": 0.6688, + "step": 204 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 0.9058470129966736, + "learning_rate": 4.958779172895308e-06, + "loss": 0.6161, + "step": 205 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 1.0031033754348755, + "learning_rate": 4.957992492131274e-06, + "loss": 0.6437, + "step": 206 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 1.5846725702285767, + "learning_rate": 4.9571984390226985e-06, + "loss": 0.6332, + "step": 207 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 1.9951609373092651, + "learning_rate": 4.956397015951215e-06, + "loss": 0.636, + "step": 208 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 1.4122583866119385, + "learning_rate": 4.95558822532056e-06, + "loss": 0.6586, + "step": 209 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.2243481874465942, + "learning_rate": 4.954772069556568e-06, + "loss": 0.6313, + "step": 210 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.8756356835365295, + "learning_rate": 4.953948551107164e-06, + "loss": 0.6406, + "step": 211 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 2.9979734420776367, + "learning_rate": 4.953117672442356e-06, + "loss": 0.5803, + "step": 212 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 2.1859359741210938, + "learning_rate": 4.952279436054229e-06, + "loss": 0.6607, + "step": 213 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.6929755806922913, + "learning_rate": 4.9514338444569346e-06, + "loss": 0.5989, + "step": 214 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 1.0361783504486084, + "learning_rate": 4.950580900186685e-06, + "loss": 0.6654, + "step": 215 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 1.210898518562317, + "learning_rate": 4.9497206058017475e-06, + "loss": 0.6213, + "step": 216 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 1.200990080833435, + "learning_rate": 4.948852963882434e-06, + "loss": 0.6654, + "step": 217 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 1.481831669807434, + "learning_rate": 4.947977977031093e-06, + "loss": 0.6474, + "step": 218 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.9883334636688232, + "learning_rate": 4.947095647872103e-06, + "loss": 0.6735, + "step": 219 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.7436536550521851, + "learning_rate": 4.946205979051868e-06, + "loss": 0.6456, + "step": 220 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9057570099830627, + "learning_rate": 4.945308973238802e-06, + "loss": 0.6228, + "step": 221 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 1.341081142425537, + "learning_rate": 4.944404633123324e-06, + "loss": 0.6417, + "step": 222 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.7958157062530518, + "learning_rate": 4.943492961417859e-06, + "loss": 0.6494, + "step": 223 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.216025471687317, + "learning_rate": 4.9425739608568106e-06, + "loss": 0.6566, + "step": 224 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.9774854779243469, + "learning_rate": 4.9416476341965735e-06, + "loss": 0.6171, + "step": 225 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 2.1562681198120117, + "learning_rate": 4.940713984215512e-06, + "loss": 0.629, + "step": 226 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 1.9521286487579346, + "learning_rate": 4.9397730137139556e-06, + "loss": 0.6475, + "step": 227 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 1.5749104022979736, + "learning_rate": 4.9388247255141895e-06, + "loss": 0.6053, + "step": 228 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 1.2008254528045654, + "learning_rate": 4.937869122460449e-06, + "loss": 0.6052, + "step": 229 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 1.0774102210998535, + "learning_rate": 4.93690620741891e-06, + "loss": 0.6099, + "step": 230 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 1.0929996967315674, + "learning_rate": 4.935935983277675e-06, + "loss": 0.6363, + "step": 231 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.8830653429031372, + "learning_rate": 4.934958452946774e-06, + "loss": 0.6136, + "step": 232 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 3.591218948364258, + "learning_rate": 4.933973619358147e-06, + "loss": 0.5962, + "step": 233 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 2.5797672271728516, + "learning_rate": 4.932981485465643e-06, + "loss": 0.6405, + "step": 234 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 1.0467664003372192, + "learning_rate": 4.9319820542450025e-06, + "loss": 0.6155, + "step": 235 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.8099795579910278, + "learning_rate": 4.930975328693856e-06, + "loss": 0.5615, + "step": 236 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.8906702995300293, + "learning_rate": 4.92996131183171e-06, + "loss": 0.6501, + "step": 237 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 1.0871416330337524, + "learning_rate": 4.928940006699944e-06, + "loss": 0.6282, + "step": 238 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 1.3209614753723145, + "learning_rate": 4.927911416361792e-06, + "loss": 0.598, + "step": 239 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.2252682447433472, + "learning_rate": 4.926875543902344e-06, + "loss": 0.6433, + "step": 240 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0569007396697998, + "learning_rate": 4.9258323924285285e-06, + "loss": 0.5927, + "step": 241 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9309014081954956, + "learning_rate": 4.924781965069106e-06, + "loss": 0.5927, + "step": 242 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0200378894805908, + "learning_rate": 4.923724264974662e-06, + "loss": 0.6064, + "step": 243 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 1.0533075332641602, + "learning_rate": 4.922659295317593e-06, + "loss": 0.6373, + "step": 244 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 0.7889382839202881, + "learning_rate": 4.921587059292102e-06, + "loss": 0.5887, + "step": 245 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.7943588495254517, + "learning_rate": 4.920507560114183e-06, + "loss": 0.593, + "step": 246 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.8247205018997192, + "learning_rate": 4.919420801021617e-06, + "loss": 0.6151, + "step": 247 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 0.9979158043861389, + "learning_rate": 4.91832678527396e-06, + "loss": 0.6019, + "step": 248 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9346868991851807, + "learning_rate": 4.917225516152532e-06, + "loss": 0.6098, + "step": 249 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.7487881183624268, + "learning_rate": 4.916116996960408e-06, + "loss": 0.5965, + "step": 250 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.821576714515686, + "learning_rate": 4.915001231022411e-06, + "loss": 0.6483, + "step": 251 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 1.0413196086883545, + "learning_rate": 4.913878221685096e-06, + "loss": 0.6108, + "step": 252 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9560331702232361, + "learning_rate": 4.912747972316745e-06, + "loss": 0.5758, + "step": 253 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.8964638113975525, + "learning_rate": 4.911610486307356e-06, + "loss": 0.6432, + "step": 254 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.8418346047401428, + "learning_rate": 4.910465767068631e-06, + "loss": 0.6027, + "step": 255 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 1.792371153831482, + "learning_rate": 4.909313818033966e-06, + "loss": 0.6198, + "step": 256 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 1.036665439605713, + "learning_rate": 4.908154642658446e-06, + "loss": 0.6255, + "step": 257 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.7592151165008545, + "learning_rate": 4.906988244418823e-06, + "loss": 0.6035, + "step": 258 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.8843073844909668, + "learning_rate": 4.90581462681352e-06, + "loss": 0.6299, + "step": 259 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9489964246749878, + "learning_rate": 4.9046337933626086e-06, + "loss": 0.5869, + "step": 260 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.851691722869873, + "learning_rate": 4.903445747607806e-06, + "loss": 0.603, + "step": 261 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 1.3722106218338013, + "learning_rate": 4.902250493112458e-06, + "loss": 0.5939, + "step": 262 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 1.1002827882766724, + "learning_rate": 4.901048033461537e-06, + "loss": 0.6452, + "step": 263 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 0.8428632020950317, + "learning_rate": 4.89983837226162e-06, + "loss": 0.5956, + "step": 264 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.7666584849357605, + "learning_rate": 4.898621513140889e-06, + "loss": 0.6067, + "step": 265 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.8413611054420471, + "learning_rate": 4.897397459749113e-06, + "loss": 0.5985, + "step": 266 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 2.3374335765838623, + "learning_rate": 4.896166215757638e-06, + "loss": 0.5885, + "step": 267 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 2.236640214920044, + "learning_rate": 4.894927784859377e-06, + "loss": 0.6408, + "step": 268 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.9715856313705444, + "learning_rate": 4.893682170768802e-06, + "loss": 0.5954, + "step": 269 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0249912738800049, + "learning_rate": 4.892429377221928e-06, + "loss": 0.6186, + "step": 270 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 1.255426049232483, + "learning_rate": 4.891169407976302e-06, + "loss": 0.6351, + "step": 271 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.9339559674263, + "learning_rate": 4.889902266810995e-06, + "loss": 0.5944, + "step": 272 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 1.2473429441452026, + "learning_rate": 4.888627957526589e-06, + "loss": 0.544, + "step": 273 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 1.0589442253112793, + "learning_rate": 4.887346483945166e-06, + "loss": 0.5543, + "step": 274 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9844024777412415, + "learning_rate": 4.886057849910294e-06, + "loss": 0.5941, + "step": 275 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 2.88578200340271, + "learning_rate": 4.8847620592870196e-06, + "loss": 0.6124, + "step": 276 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 0.7496054172515869, + "learning_rate": 4.8834591159618524e-06, + "loss": 0.6006, + "step": 277 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 0.7403052449226379, + "learning_rate": 4.88214902384276e-06, + "loss": 0.5911, + "step": 278 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9003771543502808, + "learning_rate": 4.880831786859146e-06, + "loss": 0.6347, + "step": 279 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0345501899719238, + "learning_rate": 4.879507408961847e-06, + "loss": 0.6111, + "step": 280 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.4385879039764404, + "learning_rate": 4.878175894123116e-06, + "loss": 0.6454, + "step": 281 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.8469482064247131, + "learning_rate": 4.8768372463366145e-06, + "loss": 0.6163, + "step": 282 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.8859589695930481, + "learning_rate": 4.875491469617395e-06, + "loss": 0.6144, + "step": 283 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 1.8436834812164307, + "learning_rate": 4.874138568001895e-06, + "loss": 0.6275, + "step": 284 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.6646101474761963, + "learning_rate": 4.87277854554792e-06, + "loss": 0.615, + "step": 285 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 1.0070925951004028, + "learning_rate": 4.871411406334633e-06, + "loss": 0.5898, + "step": 286 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9785194993019104, + "learning_rate": 4.870037154462545e-06, + "loss": 0.5992, + "step": 287 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.7244889736175537, + "learning_rate": 4.868655794053497e-06, + "loss": 0.6078, + "step": 288 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.4496444463729858, + "learning_rate": 4.8672673292506535e-06, + "loss": 0.5855, + "step": 289 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 1.8514957427978516, + "learning_rate": 4.865871764218486e-06, + "loss": 0.5707, + "step": 290 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.8439773321151733, + "learning_rate": 4.864469103142763e-06, + "loss": 0.5562, + "step": 291 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.8146086931228638, + "learning_rate": 4.8630593502305355e-06, + "loss": 0.6161, + "step": 292 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.8920315504074097, + "learning_rate": 4.861642509710126e-06, + "loss": 0.6139, + "step": 293 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.4980088472366333, + "learning_rate": 4.860218585831116e-06, + "loss": 0.6187, + "step": 294 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 0.9910127520561218, + "learning_rate": 4.8587875828643285e-06, + "loss": 0.5852, + "step": 295 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.819600522518158, + "learning_rate": 4.857349505101823e-06, + "loss": 0.6172, + "step": 296 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 1.1059772968292236, + "learning_rate": 4.855904356856878e-06, + "loss": 0.5868, + "step": 297 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 1.2362196445465088, + "learning_rate": 4.854452142463977e-06, + "loss": 0.625, + "step": 298 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.9956470727920532, + "learning_rate": 4.852992866278799e-06, + "loss": 0.5923, + "step": 299 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 0.864109218120575, + "learning_rate": 4.851526532678203e-06, + "loss": 0.6315, + "step": 300 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 0.8900614380836487, + "learning_rate": 4.850053146060217e-06, + "loss": 0.6128, + "step": 301 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.927254855632782, + "learning_rate": 4.84857271084402e-06, + "loss": 0.5955, + "step": 302 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0046517848968506, + "learning_rate": 4.847085231469935e-06, + "loss": 0.6134, + "step": 303 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.734597384929657, + "learning_rate": 4.8455907123994125e-06, + "loss": 0.5927, + "step": 304 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.7338348031044006, + "learning_rate": 4.844089158115016e-06, + "loss": 0.5897, + "step": 305 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9163988828659058, + "learning_rate": 4.8425805731204106e-06, + "loss": 0.6051, + "step": 306 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 1.050246238708496, + "learning_rate": 4.84106496194035e-06, + "loss": 0.5751, + "step": 307 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 0.7637603878974915, + "learning_rate": 4.83954232912066e-06, + "loss": 0.5677, + "step": 308 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.7110525965690613, + "learning_rate": 4.838012679228229e-06, + "loss": 0.6051, + "step": 309 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.7662068605422974, + "learning_rate": 4.836476016850988e-06, + "loss": 0.59, + "step": 310 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 0.8907375335693359, + "learning_rate": 4.834932346597906e-06, + "loss": 0.5792, + "step": 311 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.8939849138259888, + "learning_rate": 4.833381673098966e-06, + "loss": 0.6062, + "step": 312 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 0.8878788948059082, + "learning_rate": 4.8318240010051595e-06, + "loss": 0.5694, + "step": 313 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 1.2523870468139648, + "learning_rate": 4.830259334988468e-06, + "loss": 0.5809, + "step": 314 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 1.0836797952651978, + "learning_rate": 4.82868767974185e-06, + "loss": 0.5949, + "step": 315 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.7985473871231079, + "learning_rate": 4.827109039979226e-06, + "loss": 0.6057, + "step": 316 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 1.042951226234436, + "learning_rate": 4.825523420435469e-06, + "loss": 0.6004, + "step": 317 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.7845115661621094, + "learning_rate": 4.823930825866381e-06, + "loss": 0.6161, + "step": 318 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.931854784488678, + "learning_rate": 4.82233126104869e-06, + "loss": 0.5912, + "step": 319 + }, + { + "epoch": 1.0, + "grad_norm": 0.931854784488678, + "learning_rate": 4.8207247307800275e-06, + "loss": 0.1914, + "step": 320 + }, + { + "epoch": 1.0031315240083507, + "grad_norm": 0.751028835773468, + "learning_rate": 4.819111239878916e-06, + "loss": 0.5802, + "step": 321 + }, + { + "epoch": 1.0062630480167014, + "grad_norm": 1.4943569898605347, + "learning_rate": 4.817490793184758e-06, + "loss": 0.613, + "step": 322 + }, + { + "epoch": 1.0093945720250521, + "grad_norm": 2.296318531036377, + "learning_rate": 4.815863395557816e-06, + "loss": 0.5453, + "step": 323 + }, + { + "epoch": 1.0125260960334028, + "grad_norm": 0.760101318359375, + "learning_rate": 4.814229051879202e-06, + "loss": 0.5302, + "step": 324 + }, + { + "epoch": 1.0156576200417538, + "grad_norm": 0.8145846128463745, + "learning_rate": 4.812587767050861e-06, + "loss": 0.5831, + "step": 325 + }, + { + "epoch": 1.0187891440501045, + "grad_norm": 0.9169796109199524, + "learning_rate": 4.8109395459955565e-06, + "loss": 0.5756, + "step": 326 + }, + { + "epoch": 1.0219206680584552, + "grad_norm": 0.8791524171829224, + "learning_rate": 4.809284393656858e-06, + "loss": 0.5988, + "step": 327 + }, + { + "epoch": 1.0250521920668059, + "grad_norm": 1.0184170007705688, + "learning_rate": 4.807622314999122e-06, + "loss": 0.5476, + "step": 328 + }, + { + "epoch": 1.0281837160751566, + "grad_norm": 0.8095184564590454, + "learning_rate": 4.8059533150074805e-06, + "loss": 0.5723, + "step": 329 + }, + { + "epoch": 1.0313152400835073, + "grad_norm": 0.7621930241584778, + "learning_rate": 4.804277398687826e-06, + "loss": 0.5841, + "step": 330 + }, + { + "epoch": 1.034446764091858, + "grad_norm": 3.729628324508667, + "learning_rate": 4.802594571066791e-06, + "loss": 0.5639, + "step": 331 + }, + { + "epoch": 1.0375782881002087, + "grad_norm": 1.6502974033355713, + "learning_rate": 4.800904837191743e-06, + "loss": 0.6024, + "step": 332 + }, + { + "epoch": 1.0407098121085594, + "grad_norm": 0.8031198978424072, + "learning_rate": 4.799208202130762e-06, + "loss": 0.5305, + "step": 333 + }, + { + "epoch": 1.0438413361169103, + "grad_norm": 0.939644992351532, + "learning_rate": 4.797504670972623e-06, + "loss": 0.5446, + "step": 334 + }, + { + "epoch": 1.046972860125261, + "grad_norm": 1.0589954853057861, + "learning_rate": 4.795794248826789e-06, + "loss": 0.5366, + "step": 335 + }, + { + "epoch": 1.0501043841336117, + "grad_norm": 0.9089614748954773, + "learning_rate": 4.794076940823391e-06, + "loss": 0.5795, + "step": 336 + }, + { + "epoch": 1.0532359081419624, + "grad_norm": 0.7732561230659485, + "learning_rate": 4.792352752113212e-06, + "loss": 0.5765, + "step": 337 + }, + { + "epoch": 1.0563674321503131, + "grad_norm": 1.811553955078125, + "learning_rate": 4.790621687867672e-06, + "loss": 0.561, + "step": 338 + }, + { + "epoch": 1.0594989561586639, + "grad_norm": 1.1930758953094482, + "learning_rate": 4.788883753278813e-06, + "loss": 0.5, + "step": 339 + }, + { + "epoch": 1.0626304801670146, + "grad_norm": 0.9551813006401062, + "learning_rate": 4.787138953559285e-06, + "loss": 0.5228, + "step": 340 + }, + { + "epoch": 1.0657620041753653, + "grad_norm": 0.9609586596488953, + "learning_rate": 4.785387293942329e-06, + "loss": 0.5827, + "step": 341 + }, + { + "epoch": 1.068893528183716, + "grad_norm": 0.8403449654579163, + "learning_rate": 4.783628779681759e-06, + "loss": 0.5585, + "step": 342 + }, + { + "epoch": 1.072025052192067, + "grad_norm": 0.9108251929283142, + "learning_rate": 4.7818634160519496e-06, + "loss": 0.6077, + "step": 343 + }, + { + "epoch": 1.0751565762004176, + "grad_norm": 0.9476898908615112, + "learning_rate": 4.780091208347819e-06, + "loss": 0.5493, + "step": 344 + }, + { + "epoch": 1.0782881002087683, + "grad_norm": 1.1943707466125488, + "learning_rate": 4.778312161884813e-06, + "loss": 0.5736, + "step": 345 + }, + { + "epoch": 1.081419624217119, + "grad_norm": 3.1342639923095703, + "learning_rate": 4.77652628199889e-06, + "loss": 0.5765, + "step": 346 + }, + { + "epoch": 1.0845511482254697, + "grad_norm": 2.7982125282287598, + "learning_rate": 4.7747335740465015e-06, + "loss": 0.6003, + "step": 347 + }, + { + "epoch": 1.0876826722338204, + "grad_norm": 1.5068914890289307, + "learning_rate": 4.7729340434045815e-06, + "loss": 0.5033, + "step": 348 + }, + { + "epoch": 1.0908141962421711, + "grad_norm": 0.8273429274559021, + "learning_rate": 4.771127695470527e-06, + "loss": 0.5309, + "step": 349 + }, + { + "epoch": 1.0939457202505218, + "grad_norm": 1.104974389076233, + "learning_rate": 4.76931453566218e-06, + "loss": 0.5244, + "step": 350 + }, + { + "epoch": 1.0970772442588728, + "grad_norm": 1.096509337425232, + "learning_rate": 4.7674945694178166e-06, + "loss": 0.5585, + "step": 351 + }, + { + "epoch": 1.1002087682672235, + "grad_norm": 1.0238200426101685, + "learning_rate": 4.765667802196127e-06, + "loss": 0.5589, + "step": 352 + }, + { + "epoch": 1.1033402922755742, + "grad_norm": 0.7515526413917542, + "learning_rate": 4.763834239476197e-06, + "loss": 0.5304, + "step": 353 + }, + { + "epoch": 1.1064718162839249, + "grad_norm": 1.0282566547393799, + "learning_rate": 4.761993886757499e-06, + "loss": 0.5476, + "step": 354 + }, + { + "epoch": 1.1096033402922756, + "grad_norm": 0.9962708950042725, + "learning_rate": 4.760146749559868e-06, + "loss": 0.5117, + "step": 355 + }, + { + "epoch": 1.1127348643006263, + "grad_norm": 0.7851671576499939, + "learning_rate": 4.758292833423488e-06, + "loss": 0.5542, + "step": 356 + }, + { + "epoch": 1.115866388308977, + "grad_norm": 0.8857759237289429, + "learning_rate": 4.756432143908876e-06, + "loss": 0.544, + "step": 357 + }, + { + "epoch": 1.1189979123173277, + "grad_norm": 0.9402740597724915, + "learning_rate": 4.7545646865968645e-06, + "loss": 0.5656, + "step": 358 + }, + { + "epoch": 1.1221294363256784, + "grad_norm": 0.8210407495498657, + "learning_rate": 4.752690467088584e-06, + "loss": 0.5733, + "step": 359 + }, + { + "epoch": 1.1252609603340291, + "grad_norm": 0.795684278011322, + "learning_rate": 4.750809491005449e-06, + "loss": 0.5678, + "step": 360 + }, + { + "epoch": 1.12839248434238, + "grad_norm": 0.8712463974952698, + "learning_rate": 4.748921763989139e-06, + "loss": 0.5777, + "step": 361 + }, + { + "epoch": 1.1315240083507307, + "grad_norm": 0.9810119867324829, + "learning_rate": 4.747027291701578e-06, + "loss": 0.5511, + "step": 362 + }, + { + "epoch": 1.1346555323590815, + "grad_norm": 0.81117844581604, + "learning_rate": 4.745126079824926e-06, + "loss": 0.5038, + "step": 363 + }, + { + "epoch": 1.1377870563674322, + "grad_norm": 0.7631494402885437, + "learning_rate": 4.743218134061556e-06, + "loss": 0.6272, + "step": 364 + }, + { + "epoch": 1.1409185803757829, + "grad_norm": 0.7601696252822876, + "learning_rate": 4.741303460134038e-06, + "loss": 0.571, + "step": 365 + }, + { + "epoch": 1.1440501043841336, + "grad_norm": 1.7977744340896606, + "learning_rate": 4.7393820637851205e-06, + "loss": 0.538, + "step": 366 + }, + { + "epoch": 1.1471816283924843, + "grad_norm": 2.022578001022339, + "learning_rate": 4.737453950777718e-06, + "loss": 0.5822, + "step": 367 + }, + { + "epoch": 1.150313152400835, + "grad_norm": 0.7586764693260193, + "learning_rate": 4.735519126894885e-06, + "loss": 0.5986, + "step": 368 + }, + { + "epoch": 1.153444676409186, + "grad_norm": 0.8970286846160889, + "learning_rate": 4.733577597939812e-06, + "loss": 0.542, + "step": 369 + }, + { + "epoch": 1.1565762004175366, + "grad_norm": 0.8546352982521057, + "learning_rate": 4.731629369735793e-06, + "loss": 0.5832, + "step": 370 + }, + { + "epoch": 1.1597077244258873, + "grad_norm": 0.9266164898872375, + "learning_rate": 4.72967444812622e-06, + "loss": 0.551, + "step": 371 + }, + { + "epoch": 1.162839248434238, + "grad_norm": 1.0413658618927002, + "learning_rate": 4.7277128389745595e-06, + "loss": 0.5866, + "step": 372 + }, + { + "epoch": 1.1659707724425887, + "grad_norm": 0.9312199950218201, + "learning_rate": 4.7257445481643334e-06, + "loss": 0.5723, + "step": 373 + }, + { + "epoch": 1.1691022964509394, + "grad_norm": 0.7389806509017944, + "learning_rate": 4.723769581599109e-06, + "loss": 0.5209, + "step": 374 + }, + { + "epoch": 1.1722338204592901, + "grad_norm": 3.053169012069702, + "learning_rate": 4.721787945202472e-06, + "loss": 0.6094, + "step": 375 + }, + { + "epoch": 1.1753653444676408, + "grad_norm": 1.288589596748352, + "learning_rate": 4.719799644918017e-06, + "loss": 0.5616, + "step": 376 + }, + { + "epoch": 1.1784968684759916, + "grad_norm": 0.7675042152404785, + "learning_rate": 4.717804686709323e-06, + "loss": 0.4963, + "step": 377 + }, + { + "epoch": 1.1816283924843423, + "grad_norm": 0.7246491312980652, + "learning_rate": 4.715803076559938e-06, + "loss": 0.5273, + "step": 378 + }, + { + "epoch": 1.1847599164926932, + "grad_norm": 0.8193361759185791, + "learning_rate": 4.713794820473366e-06, + "loss": 0.6107, + "step": 379 + }, + { + "epoch": 1.187891440501044, + "grad_norm": 0.9498510360717773, + "learning_rate": 4.711779924473037e-06, + "loss": 0.5421, + "step": 380 + }, + { + "epoch": 1.1910229645093946, + "grad_norm": 1.0479756593704224, + "learning_rate": 4.709758394602305e-06, + "loss": 0.5257, + "step": 381 + }, + { + "epoch": 1.1941544885177453, + "grad_norm": 0.907866895198822, + "learning_rate": 4.707730236924413e-06, + "loss": 0.5289, + "step": 382 + }, + { + "epoch": 1.197286012526096, + "grad_norm": 0.8861165642738342, + "learning_rate": 4.705695457522488e-06, + "loss": 0.5727, + "step": 383 + }, + { + "epoch": 1.2004175365344467, + "grad_norm": 0.7467761039733887, + "learning_rate": 4.703654062499516e-06, + "loss": 0.5602, + "step": 384 + }, + { + "epoch": 1.2035490605427974, + "grad_norm": 0.7456198334693909, + "learning_rate": 4.701606057978325e-06, + "loss": 0.5345, + "step": 385 + }, + { + "epoch": 1.2066805845511483, + "grad_norm": 1.9976060390472412, + "learning_rate": 4.699551450101571e-06, + "loss": 0.5504, + "step": 386 + }, + { + "epoch": 1.209812108559499, + "grad_norm": 1.5253807306289673, + "learning_rate": 4.697490245031709e-06, + "loss": 0.5568, + "step": 387 + }, + { + "epoch": 1.2129436325678498, + "grad_norm": 1.0786075592041016, + "learning_rate": 4.6954224489509885e-06, + "loss": 0.5564, + "step": 388 + }, + { + "epoch": 1.2160751565762005, + "grad_norm": 0.8385995030403137, + "learning_rate": 4.693348068061422e-06, + "loss": 0.5341, + "step": 389 + }, + { + "epoch": 1.2192066805845512, + "grad_norm": 0.8184949159622192, + "learning_rate": 4.691267108584774e-06, + "loss": 0.5614, + "step": 390 + }, + { + "epoch": 1.2223382045929019, + "grad_norm": 0.9964898824691772, + "learning_rate": 4.68917957676254e-06, + "loss": 0.5589, + "step": 391 + }, + { + "epoch": 1.2254697286012526, + "grad_norm": 1.0168914794921875, + "learning_rate": 4.687085478855931e-06, + "loss": 0.5892, + "step": 392 + }, + { + "epoch": 1.2286012526096033, + "grad_norm": 0.8841140866279602, + "learning_rate": 4.684984821145846e-06, + "loss": 0.5327, + "step": 393 + }, + { + "epoch": 1.231732776617954, + "grad_norm": 0.834431529045105, + "learning_rate": 4.682877609932866e-06, + "loss": 0.5594, + "step": 394 + }, + { + "epoch": 1.2348643006263047, + "grad_norm": 0.7256641983985901, + "learning_rate": 4.6807638515372234e-06, + "loss": 0.5443, + "step": 395 + }, + { + "epoch": 1.2379958246346556, + "grad_norm": 0.765096127986908, + "learning_rate": 4.678643552298788e-06, + "loss": 0.5439, + "step": 396 + }, + { + "epoch": 1.2411273486430063, + "grad_norm": 0.8760455846786499, + "learning_rate": 4.676516718577051e-06, + "loss": 0.5485, + "step": 397 + }, + { + "epoch": 1.244258872651357, + "grad_norm": 2.7111501693725586, + "learning_rate": 4.674383356751099e-06, + "loss": 0.5696, + "step": 398 + }, + { + "epoch": 1.2473903966597077, + "grad_norm": 1.0521738529205322, + "learning_rate": 4.672243473219601e-06, + "loss": 0.5503, + "step": 399 + }, + { + "epoch": 1.2505219206680585, + "grad_norm": 0.8909669518470764, + "learning_rate": 4.670097074400785e-06, + "loss": 0.5183, + "step": 400 + }, + { + "epoch": 1.2536534446764092, + "grad_norm": 0.7483847737312317, + "learning_rate": 4.667944166732424e-06, + "loss": 0.5669, + "step": 401 + }, + { + "epoch": 1.2567849686847599, + "grad_norm": 1.146997094154358, + "learning_rate": 4.665784756671808e-06, + "loss": 0.5464, + "step": 402 + }, + { + "epoch": 1.2599164926931106, + "grad_norm": 0.8998096585273743, + "learning_rate": 4.663618850695733e-06, + "loss": 0.5502, + "step": 403 + }, + { + "epoch": 1.2630480167014615, + "grad_norm": 0.8882688283920288, + "learning_rate": 4.6614464553004795e-06, + "loss": 0.5507, + "step": 404 + }, + { + "epoch": 1.2661795407098122, + "grad_norm": 0.8310684561729431, + "learning_rate": 4.659267577001789e-06, + "loss": 0.5164, + "step": 405 + }, + { + "epoch": 1.269311064718163, + "grad_norm": 0.9286114573478699, + "learning_rate": 4.657082222334851e-06, + "loss": 0.4813, + "step": 406 + }, + { + "epoch": 1.2724425887265136, + "grad_norm": 1.2394906282424927, + "learning_rate": 4.654890397854275e-06, + "loss": 0.5837, + "step": 407 + }, + { + "epoch": 1.2755741127348643, + "grad_norm": 4.00585412979126, + "learning_rate": 4.652692110134079e-06, + "loss": 0.5453, + "step": 408 + }, + { + "epoch": 1.278705636743215, + "grad_norm": 1.1667803525924683, + "learning_rate": 4.650487365767667e-06, + "loss": 0.5652, + "step": 409 + }, + { + "epoch": 1.2818371607515657, + "grad_norm": 0.9351289868354797, + "learning_rate": 4.648276171367807e-06, + "loss": 0.5576, + "step": 410 + }, + { + "epoch": 1.2849686847599164, + "grad_norm": 0.8107728958129883, + "learning_rate": 4.646058533566614e-06, + "loss": 0.5821, + "step": 411 + }, + { + "epoch": 1.2881002087682671, + "grad_norm": 0.7293011546134949, + "learning_rate": 4.643834459015525e-06, + "loss": 0.5363, + "step": 412 + }, + { + "epoch": 1.2912317327766178, + "grad_norm": 0.7550690770149231, + "learning_rate": 4.641603954385289e-06, + "loss": 0.53, + "step": 413 + }, + { + "epoch": 1.2943632567849686, + "grad_norm": 0.7626177072525024, + "learning_rate": 4.639367026365938e-06, + "loss": 0.5307, + "step": 414 + }, + { + "epoch": 1.2974947807933195, + "grad_norm": 1.0841096639633179, + "learning_rate": 4.637123681666769e-06, + "loss": 0.5162, + "step": 415 + }, + { + "epoch": 1.3006263048016702, + "grad_norm": 0.8814271092414856, + "learning_rate": 4.634873927016326e-06, + "loss": 0.5369, + "step": 416 + }, + { + "epoch": 1.303757828810021, + "grad_norm": 0.7402971982955933, + "learning_rate": 4.632617769162378e-06, + "loss": 0.5846, + "step": 417 + }, + { + "epoch": 1.3068893528183716, + "grad_norm": 0.8106061220169067, + "learning_rate": 4.6303552148719e-06, + "loss": 0.5289, + "step": 418 + }, + { + "epoch": 1.3100208768267223, + "grad_norm": 0.9241361618041992, + "learning_rate": 4.628086270931053e-06, + "loss": 0.5714, + "step": 419 + }, + { + "epoch": 1.313152400835073, + "grad_norm": 0.950332522392273, + "learning_rate": 4.625810944145159e-06, + "loss": 0.5817, + "step": 420 + }, + { + "epoch": 1.316283924843424, + "grad_norm": 0.9037718772888184, + "learning_rate": 4.623529241338689e-06, + "loss": 0.5902, + "step": 421 + }, + { + "epoch": 1.3194154488517746, + "grad_norm": 1.2110658884048462, + "learning_rate": 4.621241169355234e-06, + "loss": 0.561, + "step": 422 + }, + { + "epoch": 1.3225469728601253, + "grad_norm": 0.8582742214202881, + "learning_rate": 4.618946735057491e-06, + "loss": 0.5003, + "step": 423 + }, + { + "epoch": 1.325678496868476, + "grad_norm": 0.9203405976295471, + "learning_rate": 4.6166459453272386e-06, + "loss": 0.5639, + "step": 424 + }, + { + "epoch": 1.3288100208768268, + "grad_norm": 0.933721125125885, + "learning_rate": 4.614338807065317e-06, + "loss": 0.5766, + "step": 425 + }, + { + "epoch": 1.3319415448851775, + "grad_norm": 0.8435131311416626, + "learning_rate": 4.612025327191608e-06, + "loss": 0.5656, + "step": 426 + }, + { + "epoch": 1.3350730688935282, + "grad_norm": 0.795796811580658, + "learning_rate": 4.609705512645015e-06, + "loss": 0.4996, + "step": 427 + }, + { + "epoch": 1.3382045929018789, + "grad_norm": 0.8168228268623352, + "learning_rate": 4.6073793703834404e-06, + "loss": 0.5465, + "step": 428 + }, + { + "epoch": 1.3413361169102296, + "grad_norm": 0.8795569539070129, + "learning_rate": 4.605046907383765e-06, + "loss": 0.5407, + "step": 429 + }, + { + "epoch": 1.3444676409185803, + "grad_norm": 0.8504094481468201, + "learning_rate": 4.6027081306418295e-06, + "loss": 0.5589, + "step": 430 + }, + { + "epoch": 1.347599164926931, + "grad_norm": 1.485202431678772, + "learning_rate": 4.600363047172409e-06, + "loss": 0.5515, + "step": 431 + }, + { + "epoch": 1.350730688935282, + "grad_norm": 1.1156851053237915, + "learning_rate": 4.598011664009197e-06, + "loss": 0.5681, + "step": 432 + }, + { + "epoch": 1.3538622129436326, + "grad_norm": 0.8666876554489136, + "learning_rate": 4.595653988204779e-06, + "loss": 0.5451, + "step": 433 + }, + { + "epoch": 1.3569937369519833, + "grad_norm": 0.8192381858825684, + "learning_rate": 4.593290026830619e-06, + "loss": 0.5632, + "step": 434 + }, + { + "epoch": 1.360125260960334, + "grad_norm": 0.7994804978370667, + "learning_rate": 4.590919786977029e-06, + "loss": 0.5181, + "step": 435 + }, + { + "epoch": 1.3632567849686847, + "grad_norm": 0.8038607835769653, + "learning_rate": 4.5885432757531535e-06, + "loss": 0.5385, + "step": 436 + }, + { + "epoch": 1.3663883089770354, + "grad_norm": 0.7677503824234009, + "learning_rate": 4.586160500286948e-06, + "loss": 0.5455, + "step": 437 + }, + { + "epoch": 1.3695198329853862, + "grad_norm": 0.8293285369873047, + "learning_rate": 4.583771467725157e-06, + "loss": 0.5401, + "step": 438 + }, + { + "epoch": 1.372651356993737, + "grad_norm": 0.8607680797576904, + "learning_rate": 4.581376185233289e-06, + "loss": 0.5782, + "step": 439 + }, + { + "epoch": 1.3757828810020878, + "grad_norm": 0.8847081065177917, + "learning_rate": 4.578974659995601e-06, + "loss": 0.572, + "step": 440 + }, + { + "epoch": 1.3789144050104385, + "grad_norm": 0.7669641971588135, + "learning_rate": 4.576566899215075e-06, + "loss": 0.5655, + "step": 441 + }, + { + "epoch": 1.3820459290187892, + "grad_norm": 0.8514629006385803, + "learning_rate": 4.5741529101133904e-06, + "loss": 0.5218, + "step": 442 + }, + { + "epoch": 1.38517745302714, + "grad_norm": 0.8719842433929443, + "learning_rate": 4.5717326999309145e-06, + "loss": 0.5579, + "step": 443 + }, + { + "epoch": 1.3883089770354906, + "grad_norm": 1.1142809391021729, + "learning_rate": 4.569306275926667e-06, + "loss": 0.5535, + "step": 444 + }, + { + "epoch": 1.3914405010438413, + "grad_norm": 0.7392387986183167, + "learning_rate": 4.566873645378309e-06, + "loss": 0.5335, + "step": 445 + }, + { + "epoch": 1.394572025052192, + "grad_norm": 0.9066658616065979, + "learning_rate": 4.564434815582117e-06, + "loss": 0.5286, + "step": 446 + }, + { + "epoch": 1.3977035490605427, + "grad_norm": 0.8648932576179504, + "learning_rate": 4.561989793852959e-06, + "loss": 0.5008, + "step": 447 + }, + { + "epoch": 1.4008350730688934, + "grad_norm": 0.7768712043762207, + "learning_rate": 4.559538587524276e-06, + "loss": 0.5727, + "step": 448 + }, + { + "epoch": 1.4039665970772441, + "grad_norm": 0.7851182222366333, + "learning_rate": 4.557081203948059e-06, + "loss": 0.5731, + "step": 449 + }, + { + "epoch": 1.407098121085595, + "grad_norm": 0.8959861397743225, + "learning_rate": 4.5546176504948255e-06, + "loss": 0.5587, + "step": 450 + }, + { + "epoch": 1.4102296450939458, + "grad_norm": 1.0538026094436646, + "learning_rate": 4.552147934553601e-06, + "loss": 0.5808, + "step": 451 + }, + { + "epoch": 1.4133611691022965, + "grad_norm": 0.9887629151344299, + "learning_rate": 4.54967206353189e-06, + "loss": 0.5658, + "step": 452 + }, + { + "epoch": 1.4164926931106472, + "grad_norm": 0.9579302072525024, + "learning_rate": 4.547190044855663e-06, + "loss": 0.5092, + "step": 453 + }, + { + "epoch": 1.4196242171189979, + "grad_norm": 0.6993522047996521, + "learning_rate": 4.544701885969326e-06, + "loss": 0.5233, + "step": 454 + }, + { + "epoch": 1.4227557411273486, + "grad_norm": 0.8197568655014038, + "learning_rate": 4.542207594335703e-06, + "loss": 0.553, + "step": 455 + }, + { + "epoch": 1.4258872651356993, + "grad_norm": 2.921947717666626, + "learning_rate": 4.53970717743601e-06, + "loss": 0.4857, + "step": 456 + }, + { + "epoch": 1.4290187891440502, + "grad_norm": 1.3547242879867554, + "learning_rate": 4.53720064276984e-06, + "loss": 0.5676, + "step": 457 + }, + { + "epoch": 1.432150313152401, + "grad_norm": 1.4175567626953125, + "learning_rate": 4.534687997855131e-06, + "loss": 0.5164, + "step": 458 + }, + { + "epoch": 1.4352818371607516, + "grad_norm": 1.378146767616272, + "learning_rate": 4.532169250228145e-06, + "loss": 0.5429, + "step": 459 + }, + { + "epoch": 1.4384133611691023, + "grad_norm": 0.7811698317527771, + "learning_rate": 4.529644407443456e-06, + "loss": 0.524, + "step": 460 + }, + { + "epoch": 1.441544885177453, + "grad_norm": 1.1481678485870361, + "learning_rate": 4.527113477073914e-06, + "loss": 0.5513, + "step": 461 + }, + { + "epoch": 1.4446764091858038, + "grad_norm": 0.8450161218643188, + "learning_rate": 4.5245764667106266e-06, + "loss": 0.5632, + "step": 462 + }, + { + "epoch": 1.4478079331941545, + "grad_norm": 1.1582145690917969, + "learning_rate": 4.522033383962941e-06, + "loss": 0.5834, + "step": 463 + }, + { + "epoch": 1.4509394572025052, + "grad_norm": 1.0403447151184082, + "learning_rate": 4.519484236458416e-06, + "loss": 0.506, + "step": 464 + }, + { + "epoch": 1.4540709812108559, + "grad_norm": 0.7894920706748962, + "learning_rate": 4.516929031842799e-06, + "loss": 0.5526, + "step": 465 + }, + { + "epoch": 1.4572025052192066, + "grad_norm": 0.8092262744903564, + "learning_rate": 4.51436777778001e-06, + "loss": 0.5619, + "step": 466 + }, + { + "epoch": 1.4603340292275573, + "grad_norm": 0.9773806929588318, + "learning_rate": 4.511800481952106e-06, + "loss": 0.5179, + "step": 467 + }, + { + "epoch": 1.4634655532359082, + "grad_norm": 1.018676519393921, + "learning_rate": 4.509227152059271e-06, + "loss": 0.5415, + "step": 468 + }, + { + "epoch": 1.466597077244259, + "grad_norm": 0.7457838654518127, + "learning_rate": 4.506647795819784e-06, + "loss": 0.5473, + "step": 469 + }, + { + "epoch": 1.4697286012526096, + "grad_norm": 0.7826436161994934, + "learning_rate": 4.50406242097e-06, + "loss": 0.5526, + "step": 470 + }, + { + "epoch": 1.4728601252609603, + "grad_norm": 0.9492483139038086, + "learning_rate": 4.501471035264328e-06, + "loss": 0.5179, + "step": 471 + }, + { + "epoch": 1.475991649269311, + "grad_norm": 0.93398517370224, + "learning_rate": 4.4988736464752005e-06, + "loss": 0.5195, + "step": 472 + }, + { + "epoch": 1.4791231732776617, + "grad_norm": 0.8396487832069397, + "learning_rate": 4.496270262393061e-06, + "loss": 0.5447, + "step": 473 + }, + { + "epoch": 1.4822546972860124, + "grad_norm": 0.7450584173202515, + "learning_rate": 4.4936608908263315e-06, + "loss": 0.5207, + "step": 474 + }, + { + "epoch": 1.4853862212943634, + "grad_norm": 0.7887717485427856, + "learning_rate": 4.491045539601392e-06, + "loss": 0.523, + "step": 475 + }, + { + "epoch": 1.488517745302714, + "grad_norm": 1.2051388025283813, + "learning_rate": 4.48842421656256e-06, + "loss": 0.5402, + "step": 476 + }, + { + "epoch": 1.4916492693110648, + "grad_norm": 2.3103389739990234, + "learning_rate": 4.485796929572063e-06, + "loss": 0.5588, + "step": 477 + }, + { + "epoch": 1.4947807933194155, + "grad_norm": 0.7473112344741821, + "learning_rate": 4.483163686510016e-06, + "loss": 0.5731, + "step": 478 + }, + { + "epoch": 1.4979123173277662, + "grad_norm": 0.7545126676559448, + "learning_rate": 4.480524495274399e-06, + "loss": 0.5536, + "step": 479 + }, + { + "epoch": 1.501043841336117, + "grad_norm": 0.7801297903060913, + "learning_rate": 4.477879363781033e-06, + "loss": 0.5696, + "step": 480 + }, + { + "epoch": 1.5041753653444676, + "grad_norm": 0.7740563750267029, + "learning_rate": 4.475228299963554e-06, + "loss": 0.5526, + "step": 481 + }, + { + "epoch": 1.5073068893528183, + "grad_norm": 0.8600060343742371, + "learning_rate": 4.4725713117733936e-06, + "loss": 0.5051, + "step": 482 + }, + { + "epoch": 1.510438413361169, + "grad_norm": 0.6934283971786499, + "learning_rate": 4.46990840717975e-06, + "loss": 0.5564, + "step": 483 + }, + { + "epoch": 1.5135699373695197, + "grad_norm": 0.8927920460700989, + "learning_rate": 4.46723959416957e-06, + "loss": 0.5529, + "step": 484 + }, + { + "epoch": 1.5167014613778704, + "grad_norm": 0.9570988416671753, + "learning_rate": 4.464564880747517e-06, + "loss": 0.5661, + "step": 485 + }, + { + "epoch": 1.5198329853862211, + "grad_norm": 0.7229202389717102, + "learning_rate": 4.461884274935956e-06, + "loss": 0.5964, + "step": 486 + }, + { + "epoch": 1.522964509394572, + "grad_norm": 0.7367239594459534, + "learning_rate": 4.4591977847749225e-06, + "loss": 0.5455, + "step": 487 + }, + { + "epoch": 1.5260960334029228, + "grad_norm": 0.8062120676040649, + "learning_rate": 4.456505418322103e-06, + "loss": 0.5735, + "step": 488 + }, + { + "epoch": 1.5292275574112735, + "grad_norm": 0.8854482769966125, + "learning_rate": 4.453807183652808e-06, + "loss": 0.5421, + "step": 489 + }, + { + "epoch": 1.5323590814196242, + "grad_norm": 0.7518959045410156, + "learning_rate": 4.451103088859951e-06, + "loss": 0.5083, + "step": 490 + }, + { + "epoch": 1.535490605427975, + "grad_norm": 0.8621206879615784, + "learning_rate": 4.448393142054016e-06, + "loss": 0.4712, + "step": 491 + }, + { + "epoch": 1.5386221294363258, + "grad_norm": 1.0618741512298584, + "learning_rate": 4.445677351363046e-06, + "loss": 0.5808, + "step": 492 + }, + { + "epoch": 1.5417536534446765, + "grad_norm": 0.8261345028877258, + "learning_rate": 4.442955724932607e-06, + "loss": 0.5625, + "step": 493 + }, + { + "epoch": 1.5448851774530272, + "grad_norm": 0.7067139744758606, + "learning_rate": 4.440228270925772e-06, + "loss": 0.5661, + "step": 494 + }, + { + "epoch": 1.548016701461378, + "grad_norm": 0.9234416484832764, + "learning_rate": 4.437494997523091e-06, + "loss": 0.5428, + "step": 495 + }, + { + "epoch": 1.5511482254697286, + "grad_norm": 0.9273470044136047, + "learning_rate": 4.434755912922567e-06, + "loss": 0.5388, + "step": 496 + }, + { + "epoch": 1.5542797494780793, + "grad_norm": 1.0163263082504272, + "learning_rate": 4.4320110253396345e-06, + "loss": 0.5409, + "step": 497 + }, + { + "epoch": 1.55741127348643, + "grad_norm": 0.9542096853256226, + "learning_rate": 4.429260343007133e-06, + "loss": 0.5329, + "step": 498 + }, + { + "epoch": 1.5605427974947808, + "grad_norm": 0.8076801896095276, + "learning_rate": 4.426503874175283e-06, + "loss": 0.5616, + "step": 499 + }, + { + "epoch": 1.5636743215031315, + "grad_norm": 1.0063767433166504, + "learning_rate": 4.423741627111658e-06, + "loss": 0.5369, + "step": 500 + }, + { + "epoch": 1.5668058455114822, + "grad_norm": 1.040286898612976, + "learning_rate": 4.420973610101166e-06, + "loss": 0.5474, + "step": 501 + }, + { + "epoch": 1.5699373695198329, + "grad_norm": 0.7832860946655273, + "learning_rate": 4.4181998314460164e-06, + "loss": 0.5486, + "step": 502 + }, + { + "epoch": 1.5730688935281836, + "grad_norm": 0.8162257075309753, + "learning_rate": 4.415420299465706e-06, + "loss": 0.5054, + "step": 503 + }, + { + "epoch": 1.5762004175365343, + "grad_norm": 0.9108433127403259, + "learning_rate": 4.4126350224969814e-06, + "loss": 0.5399, + "step": 504 + }, + { + "epoch": 1.5793319415448852, + "grad_norm": 0.8002520799636841, + "learning_rate": 4.409844008893824e-06, + "loss": 0.5485, + "step": 505 + }, + { + "epoch": 1.582463465553236, + "grad_norm": 0.8543248772621155, + "learning_rate": 4.407047267027423e-06, + "loss": 0.4984, + "step": 506 + }, + { + "epoch": 1.5855949895615866, + "grad_norm": 0.7154155373573303, + "learning_rate": 4.404244805286141e-06, + "loss": 0.5392, + "step": 507 + }, + { + "epoch": 1.5887265135699373, + "grad_norm": 0.818553626537323, + "learning_rate": 4.401436632075504e-06, + "loss": 0.5178, + "step": 508 + }, + { + "epoch": 1.5918580375782883, + "grad_norm": 0.7535017728805542, + "learning_rate": 4.398622755818167e-06, + "loss": 0.5446, + "step": 509 + }, + { + "epoch": 1.594989561586639, + "grad_norm": 0.9328975677490234, + "learning_rate": 4.395803184953889e-06, + "loss": 0.5546, + "step": 510 + }, + { + "epoch": 1.5981210855949897, + "grad_norm": 0.7960026860237122, + "learning_rate": 4.392977927939508e-06, + "loss": 0.5451, + "step": 511 + }, + { + "epoch": 1.6012526096033404, + "grad_norm": 0.9686267971992493, + "learning_rate": 4.3901469932489195e-06, + "loss": 0.5198, + "step": 512 + }, + { + "epoch": 1.604384133611691, + "grad_norm": 0.903137743473053, + "learning_rate": 4.387310389373047e-06, + "loss": 0.5395, + "step": 513 + }, + { + "epoch": 1.6075156576200418, + "grad_norm": 1.0728516578674316, + "learning_rate": 4.384468124819816e-06, + "loss": 0.5843, + "step": 514 + }, + { + "epoch": 1.6106471816283925, + "grad_norm": 1.0245436429977417, + "learning_rate": 4.3816202081141345e-06, + "loss": 0.5672, + "step": 515 + }, + { + "epoch": 1.6137787056367432, + "grad_norm": 0.9672732353210449, + "learning_rate": 4.378766647797858e-06, + "loss": 0.5369, + "step": 516 + }, + { + "epoch": 1.616910229645094, + "grad_norm": 0.9149513840675354, + "learning_rate": 4.375907452429774e-06, + "loss": 0.4628, + "step": 517 + }, + { + "epoch": 1.6200417536534446, + "grad_norm": 0.7543843984603882, + "learning_rate": 4.373042630585567e-06, + "loss": 0.5344, + "step": 518 + }, + { + "epoch": 1.6231732776617953, + "grad_norm": 0.7589017152786255, + "learning_rate": 4.370172190857801e-06, + "loss": 0.5672, + "step": 519 + }, + { + "epoch": 1.626304801670146, + "grad_norm": 0.803040623664856, + "learning_rate": 4.367296141855887e-06, + "loss": 0.5313, + "step": 520 + }, + { + "epoch": 1.6294363256784967, + "grad_norm": 0.8305794596672058, + "learning_rate": 4.3644144922060625e-06, + "loss": 0.5754, + "step": 521 + }, + { + "epoch": 1.6325678496868476, + "grad_norm": 1.0086486339569092, + "learning_rate": 4.361527250551361e-06, + "loss": 0.5433, + "step": 522 + }, + { + "epoch": 1.6356993736951984, + "grad_norm": 0.7217550277709961, + "learning_rate": 4.35863442555159e-06, + "loss": 0.524, + "step": 523 + }, + { + "epoch": 1.638830897703549, + "grad_norm": 0.7788524627685547, + "learning_rate": 4.355736025883303e-06, + "loss": 0.536, + "step": 524 + }, + { + "epoch": 1.6419624217118998, + "grad_norm": 0.8460550904273987, + "learning_rate": 4.352832060239774e-06, + "loss": 0.5381, + "step": 525 + }, + { + "epoch": 1.6450939457202505, + "grad_norm": 0.7571215033531189, + "learning_rate": 4.3499225373309675e-06, + "loss": 0.541, + "step": 526 + }, + { + "epoch": 1.6482254697286014, + "grad_norm": 0.7343226671218872, + "learning_rate": 4.347007465883523e-06, + "loss": 0.5147, + "step": 527 + }, + { + "epoch": 1.651356993736952, + "grad_norm": 0.7271892428398132, + "learning_rate": 4.3440868546407165e-06, + "loss": 0.5311, + "step": 528 + }, + { + "epoch": 1.6544885177453028, + "grad_norm": 0.8166136741638184, + "learning_rate": 4.341160712362442e-06, + "loss": 0.5379, + "step": 529 + }, + { + "epoch": 1.6576200417536535, + "grad_norm": 1.5985233783721924, + "learning_rate": 4.338229047825182e-06, + "loss": 0.5782, + "step": 530 + }, + { + "epoch": 1.6607515657620042, + "grad_norm": 0.7835702896118164, + "learning_rate": 4.3352918698219835e-06, + "loss": 0.525, + "step": 531 + }, + { + "epoch": 1.663883089770355, + "grad_norm": 0.7278687953948975, + "learning_rate": 4.332349187162428e-06, + "loss": 0.5266, + "step": 532 + }, + { + "epoch": 1.6670146137787056, + "grad_norm": 0.8240190148353577, + "learning_rate": 4.329401008672608e-06, + "loss": 0.5515, + "step": 533 + }, + { + "epoch": 1.6701461377870563, + "grad_norm": 0.9447080492973328, + "learning_rate": 4.326447343195102e-06, + "loss": 0.5596, + "step": 534 + }, + { + "epoch": 1.673277661795407, + "grad_norm": 0.7827372550964355, + "learning_rate": 4.323488199588944e-06, + "loss": 0.5466, + "step": 535 + }, + { + "epoch": 1.6764091858037578, + "grad_norm": 0.9252517223358154, + "learning_rate": 4.320523586729599e-06, + "loss": 0.5433, + "step": 536 + }, + { + "epoch": 1.6795407098121085, + "grad_norm": 0.9437504410743713, + "learning_rate": 4.317553513508934e-06, + "loss": 0.5552, + "step": 537 + }, + { + "epoch": 1.6826722338204592, + "grad_norm": 0.8972746133804321, + "learning_rate": 4.3145779888351986e-06, + "loss": 0.5259, + "step": 538 + }, + { + "epoch": 1.6858037578288099, + "grad_norm": 0.8017446994781494, + "learning_rate": 4.311597021632988e-06, + "loss": 0.5263, + "step": 539 + }, + { + "epoch": 1.6889352818371608, + "grad_norm": 0.7875497341156006, + "learning_rate": 4.3086106208432235e-06, + "loss": 0.5316, + "step": 540 + }, + { + "epoch": 1.6920668058455115, + "grad_norm": 0.8204905986785889, + "learning_rate": 4.305618795423125e-06, + "loss": 0.5506, + "step": 541 + }, + { + "epoch": 1.6951983298538622, + "grad_norm": 0.888359785079956, + "learning_rate": 4.30262155434618e-06, + "loss": 0.4825, + "step": 542 + }, + { + "epoch": 1.698329853862213, + "grad_norm": 1.1026058197021484, + "learning_rate": 4.29961890660212e-06, + "loss": 0.5321, + "step": 543 + }, + { + "epoch": 1.7014613778705638, + "grad_norm": 0.7662535905838013, + "learning_rate": 4.2966108611968945e-06, + "loss": 0.5432, + "step": 544 + }, + { + "epoch": 1.7045929018789145, + "grad_norm": 1.1951749324798584, + "learning_rate": 4.293597427152641e-06, + "loss": 0.5123, + "step": 545 + }, + { + "epoch": 1.7077244258872653, + "grad_norm": 1.303183913230896, + "learning_rate": 4.290578613507661e-06, + "loss": 0.5346, + "step": 546 + }, + { + "epoch": 1.710855949895616, + "grad_norm": 0.7653357982635498, + "learning_rate": 4.287554429316387e-06, + "loss": 0.5397, + "step": 547 + }, + { + "epoch": 1.7139874739039667, + "grad_norm": 0.796215295791626, + "learning_rate": 4.284524883649366e-06, + "loss": 0.5421, + "step": 548 + }, + { + "epoch": 1.7171189979123174, + "grad_norm": 0.7599332332611084, + "learning_rate": 4.281489985593219e-06, + "loss": 0.5289, + "step": 549 + }, + { + "epoch": 1.720250521920668, + "grad_norm": 0.8029115796089172, + "learning_rate": 4.2784497442506265e-06, + "loss": 0.5409, + "step": 550 + }, + { + "epoch": 1.7233820459290188, + "grad_norm": 0.7194099426269531, + "learning_rate": 4.275404168740291e-06, + "loss": 0.5327, + "step": 551 + }, + { + "epoch": 1.7265135699373695, + "grad_norm": 0.7960740923881531, + "learning_rate": 4.272353268196917e-06, + "loss": 0.4896, + "step": 552 + }, + { + "epoch": 1.7296450939457202, + "grad_norm": 0.9572116732597351, + "learning_rate": 4.269297051771178e-06, + "loss": 0.5402, + "step": 553 + }, + { + "epoch": 1.732776617954071, + "grad_norm": 1.3604938983917236, + "learning_rate": 4.266235528629695e-06, + "loss": 0.5792, + "step": 554 + }, + { + "epoch": 1.7359081419624216, + "grad_norm": 2.067286729812622, + "learning_rate": 4.263168707955002e-06, + "loss": 0.5033, + "step": 555 + }, + { + "epoch": 1.7390396659707723, + "grad_norm": 0.8031097054481506, + "learning_rate": 4.260096598945523e-06, + "loss": 0.5117, + "step": 556 + }, + { + "epoch": 1.742171189979123, + "grad_norm": 1.0241729021072388, + "learning_rate": 4.257019210815546e-06, + "loss": 0.5359, + "step": 557 + }, + { + "epoch": 1.745302713987474, + "grad_norm": 0.7625218629837036, + "learning_rate": 4.25393655279519e-06, + "loss": 0.5625, + "step": 558 + }, + { + "epoch": 1.7484342379958246, + "grad_norm": 0.8603503704071045, + "learning_rate": 4.250848634130381e-06, + "loss": 0.5043, + "step": 559 + }, + { + "epoch": 1.7515657620041754, + "grad_norm": 0.9543750286102295, + "learning_rate": 4.247755464082824e-06, + "loss": 0.5364, + "step": 560 + }, + { + "epoch": 1.754697286012526, + "grad_norm": 0.9707463979721069, + "learning_rate": 4.244657051929973e-06, + "loss": 0.5184, + "step": 561 + }, + { + "epoch": 1.757828810020877, + "grad_norm": 0.7491432428359985, + "learning_rate": 4.241553406965008e-06, + "loss": 0.559, + "step": 562 + }, + { + "epoch": 1.7609603340292277, + "grad_norm": 0.7444972991943359, + "learning_rate": 4.238444538496801e-06, + "loss": 0.5327, + "step": 563 + }, + { + "epoch": 1.7640918580375784, + "grad_norm": 2.7108678817749023, + "learning_rate": 4.235330455849892e-06, + "loss": 0.55, + "step": 564 + }, + { + "epoch": 1.767223382045929, + "grad_norm": 1.6716049909591675, + "learning_rate": 4.232211168364459e-06, + "loss": 0.5093, + "step": 565 + }, + { + "epoch": 1.7703549060542798, + "grad_norm": 0.7023475170135498, + "learning_rate": 4.229086685396295e-06, + "loss": 0.569, + "step": 566 + }, + { + "epoch": 1.7734864300626305, + "grad_norm": 0.8596265316009521, + "learning_rate": 4.225957016316771e-06, + "loss": 0.5128, + "step": 567 + }, + { + "epoch": 1.7766179540709812, + "grad_norm": 0.8110849857330322, + "learning_rate": 4.222822170512816e-06, + "loss": 0.5142, + "step": 568 + }, + { + "epoch": 1.779749478079332, + "grad_norm": 0.7583725452423096, + "learning_rate": 4.219682157386884e-06, + "loss": 0.5584, + "step": 569 + }, + { + "epoch": 1.7828810020876826, + "grad_norm": 0.787811279296875, + "learning_rate": 4.21653698635693e-06, + "loss": 0.5068, + "step": 570 + }, + { + "epoch": 1.7860125260960333, + "grad_norm": 0.8298993110656738, + "learning_rate": 4.213386666856375e-06, + "loss": 0.5496, + "step": 571 + }, + { + "epoch": 1.789144050104384, + "grad_norm": 0.8999841213226318, + "learning_rate": 4.210231208334087e-06, + "loss": 0.5454, + "step": 572 + }, + { + "epoch": 1.7922755741127347, + "grad_norm": 4.264521598815918, + "learning_rate": 4.207070620254345e-06, + "loss": 0.5486, + "step": 573 + }, + { + "epoch": 1.7954070981210855, + "grad_norm": 0.8517448306083679, + "learning_rate": 4.203904912096812e-06, + "loss": 0.5566, + "step": 574 + }, + { + "epoch": 1.7985386221294362, + "grad_norm": 0.9230182766914368, + "learning_rate": 4.200734093356511e-06, + "loss": 0.4964, + "step": 575 + }, + { + "epoch": 1.801670146137787, + "grad_norm": 1.224039912223816, + "learning_rate": 4.197558173543791e-06, + "loss": 0.5356, + "step": 576 + }, + { + "epoch": 1.8048016701461378, + "grad_norm": 0.9998573660850525, + "learning_rate": 4.194377162184301e-06, + "loss": 0.5334, + "step": 577 + }, + { + "epoch": 1.8079331941544885, + "grad_norm": 0.865521252155304, + "learning_rate": 4.191191068818963e-06, + "loss": 0.5036, + "step": 578 + }, + { + "epoch": 1.8110647181628392, + "grad_norm": 0.8048138618469238, + "learning_rate": 4.18799990300394e-06, + "loss": 0.4979, + "step": 579 + }, + { + "epoch": 1.8141962421711901, + "grad_norm": 0.717815637588501, + "learning_rate": 4.184803674310609e-06, + "loss": 0.5623, + "step": 580 + }, + { + "epoch": 1.8173277661795408, + "grad_norm": 0.8403327465057373, + "learning_rate": 4.1816023923255335e-06, + "loss": 0.5055, + "step": 581 + }, + { + "epoch": 1.8204592901878915, + "grad_norm": 0.7298995852470398, + "learning_rate": 4.178396066650432e-06, + "loss": 0.5641, + "step": 582 + }, + { + "epoch": 1.8235908141962422, + "grad_norm": 0.9469727873802185, + "learning_rate": 4.1751847069021516e-06, + "loss": 0.5557, + "step": 583 + }, + { + "epoch": 1.826722338204593, + "grad_norm": 0.8641784191131592, + "learning_rate": 4.1719683227126386e-06, + "loss": 0.5153, + "step": 584 + }, + { + "epoch": 1.8298538622129437, + "grad_norm": 0.7316668629646301, + "learning_rate": 4.168746923728908e-06, + "loss": 0.4988, + "step": 585 + }, + { + "epoch": 1.8329853862212944, + "grad_norm": 0.8795468807220459, + "learning_rate": 4.165520519613017e-06, + "loss": 0.5483, + "step": 586 + }, + { + "epoch": 1.836116910229645, + "grad_norm": 0.7323560118675232, + "learning_rate": 4.162289120042034e-06, + "loss": 0.5194, + "step": 587 + }, + { + "epoch": 1.8392484342379958, + "grad_norm": 0.8217021822929382, + "learning_rate": 4.159052734708013e-06, + "loss": 0.532, + "step": 588 + }, + { + "epoch": 1.8423799582463465, + "grad_norm": 0.7669674754142761, + "learning_rate": 4.155811373317958e-06, + "loss": 0.541, + "step": 589 + }, + { + "epoch": 1.8455114822546972, + "grad_norm": 0.8312156200408936, + "learning_rate": 4.152565045593801e-06, + "loss": 0.5298, + "step": 590 + }, + { + "epoch": 1.848643006263048, + "grad_norm": 0.8967565298080444, + "learning_rate": 4.1493137612723665e-06, + "loss": 0.51, + "step": 591 + }, + { + "epoch": 1.8517745302713986, + "grad_norm": 0.8706664443016052, + "learning_rate": 4.14605753010535e-06, + "loss": 0.4941, + "step": 592 + }, + { + "epoch": 1.8549060542797495, + "grad_norm": 0.7585753798484802, + "learning_rate": 4.14279636185928e-06, + "loss": 0.5161, + "step": 593 + }, + { + "epoch": 1.8580375782881002, + "grad_norm": 0.7495241165161133, + "learning_rate": 4.1395302663154954e-06, + "loss": 0.5388, + "step": 594 + }, + { + "epoch": 1.861169102296451, + "grad_norm": 1.0746862888336182, + "learning_rate": 4.136259253270114e-06, + "loss": 0.4976, + "step": 595 + }, + { + "epoch": 1.8643006263048016, + "grad_norm": 0.872309684753418, + "learning_rate": 4.132983332534e-06, + "loss": 0.559, + "step": 596 + }, + { + "epoch": 1.8674321503131524, + "grad_norm": 0.8759891986846924, + "learning_rate": 4.1297025139327405e-06, + "loss": 0.5436, + "step": 597 + }, + { + "epoch": 1.8705636743215033, + "grad_norm": 1.1044493913650513, + "learning_rate": 4.126416807306611e-06, + "loss": 0.5476, + "step": 598 + }, + { + "epoch": 1.873695198329854, + "grad_norm": 0.8340442180633545, + "learning_rate": 4.123126222510549e-06, + "loss": 0.4592, + "step": 599 + }, + { + "epoch": 1.8768267223382047, + "grad_norm": 0.8331449031829834, + "learning_rate": 4.119830769414123e-06, + "loss": 0.5219, + "step": 600 + }, + { + "epoch": 1.8799582463465554, + "grad_norm": 1.0862973928451538, + "learning_rate": 4.116530457901503e-06, + "loss": 0.5159, + "step": 601 + }, + { + "epoch": 1.883089770354906, + "grad_norm": 0.8524414300918579, + "learning_rate": 4.113225297871431e-06, + "loss": 0.5502, + "step": 602 + }, + { + "epoch": 1.8862212943632568, + "grad_norm": 1.4945416450500488, + "learning_rate": 4.10991529923719e-06, + "loss": 0.5627, + "step": 603 + }, + { + "epoch": 1.8893528183716075, + "grad_norm": 1.5518157482147217, + "learning_rate": 4.10660047192658e-06, + "loss": 0.5517, + "step": 604 + }, + { + "epoch": 1.8924843423799582, + "grad_norm": 2.56638765335083, + "learning_rate": 4.103280825881878e-06, + "loss": 0.5422, + "step": 605 + }, + { + "epoch": 1.895615866388309, + "grad_norm": 0.867254912853241, + "learning_rate": 4.099956371059817e-06, + "loss": 0.4991, + "step": 606 + }, + { + "epoch": 1.8987473903966596, + "grad_norm": 0.9555892944335938, + "learning_rate": 4.096627117431554e-06, + "loss": 0.5339, + "step": 607 + }, + { + "epoch": 1.9018789144050103, + "grad_norm": 0.7905483245849609, + "learning_rate": 4.093293074982638e-06, + "loss": 0.5168, + "step": 608 + }, + { + "epoch": 1.905010438413361, + "grad_norm": 0.7500227093696594, + "learning_rate": 4.089954253712981e-06, + "loss": 0.5096, + "step": 609 + }, + { + "epoch": 1.9081419624217117, + "grad_norm": 0.8458324074745178, + "learning_rate": 4.086610663636828e-06, + "loss": 0.5296, + "step": 610 + }, + { + "epoch": 1.9112734864300627, + "grad_norm": 0.7392706871032715, + "learning_rate": 4.08326231478273e-06, + "loss": 0.5305, + "step": 611 + }, + { + "epoch": 1.9144050104384134, + "grad_norm": 0.8113343715667725, + "learning_rate": 4.079909217193508e-06, + "loss": 0.5044, + "step": 612 + }, + { + "epoch": 1.917536534446764, + "grad_norm": 0.7637801766395569, + "learning_rate": 4.076551380926226e-06, + "loss": 0.5298, + "step": 613 + }, + { + "epoch": 1.9206680584551148, + "grad_norm": 1.0523375272750854, + "learning_rate": 4.073188816052164e-06, + "loss": 0.5111, + "step": 614 + }, + { + "epoch": 1.9237995824634657, + "grad_norm": 0.8224868774414062, + "learning_rate": 4.069821532656781e-06, + "loss": 0.5178, + "step": 615 + }, + { + "epoch": 1.9269311064718164, + "grad_norm": 0.7270777821540833, + "learning_rate": 4.066449540839693e-06, + "loss": 0.5307, + "step": 616 + }, + { + "epoch": 1.9300626304801671, + "grad_norm": 0.7214602828025818, + "learning_rate": 4.063072850714631e-06, + "loss": 0.5171, + "step": 617 + }, + { + "epoch": 1.9331941544885178, + "grad_norm": 0.7333671450614929, + "learning_rate": 4.059691472409426e-06, + "loss": 0.56, + "step": 618 + }, + { + "epoch": 1.9363256784968685, + "grad_norm": 0.9166824221611023, + "learning_rate": 4.056305416065964e-06, + "loss": 0.5388, + "step": 619 + }, + { + "epoch": 1.9394572025052192, + "grad_norm": 0.7743303775787354, + "learning_rate": 4.052914691840167e-06, + "loss": 0.5134, + "step": 620 + }, + { + "epoch": 1.94258872651357, + "grad_norm": 0.704097330570221, + "learning_rate": 4.0495193099019524e-06, + "loss": 0.4926, + "step": 621 + }, + { + "epoch": 1.9457202505219207, + "grad_norm": 0.8508503437042236, + "learning_rate": 4.046119280435212e-06, + "loss": 0.5008, + "step": 622 + }, + { + "epoch": 1.9488517745302714, + "grad_norm": 0.725933313369751, + "learning_rate": 4.042714613637775e-06, + "loss": 0.5549, + "step": 623 + }, + { + "epoch": 1.951983298538622, + "grad_norm": 0.8919175863265991, + "learning_rate": 4.039305319721381e-06, + "loss": 0.5183, + "step": 624 + }, + { + "epoch": 1.9551148225469728, + "grad_norm": 0.827919065952301, + "learning_rate": 4.035891408911644e-06, + "loss": 0.5624, + "step": 625 + }, + { + "epoch": 1.9582463465553235, + "grad_norm": 0.7415187358856201, + "learning_rate": 4.032472891448032e-06, + "loss": 0.5454, + "step": 626 + }, + { + "epoch": 1.9613778705636742, + "grad_norm": 0.7675788998603821, + "learning_rate": 4.029049777583824e-06, + "loss": 0.5361, + "step": 627 + }, + { + "epoch": 1.964509394572025, + "grad_norm": 0.8464030623435974, + "learning_rate": 4.025622077586088e-06, + "loss": 0.5295, + "step": 628 + }, + { + "epoch": 1.9676409185803758, + "grad_norm": 0.7641633749008179, + "learning_rate": 4.022189801735646e-06, + "loss": 0.55, + "step": 629 + }, + { + "epoch": 1.9707724425887265, + "grad_norm": 0.7813227772712708, + "learning_rate": 4.018752960327048e-06, + "loss": 0.5587, + "step": 630 + }, + { + "epoch": 1.9739039665970772, + "grad_norm": 0.7576701641082764, + "learning_rate": 4.015311563668533e-06, + "loss": 0.5413, + "step": 631 + }, + { + "epoch": 1.977035490605428, + "grad_norm": 0.6949650049209595, + "learning_rate": 4.011865622082004e-06, + "loss": 0.5344, + "step": 632 + }, + { + "epoch": 1.9801670146137789, + "grad_norm": 0.9009145498275757, + "learning_rate": 4.008415145902997e-06, + "loss": 0.5233, + "step": 633 + }, + { + "epoch": 1.9832985386221296, + "grad_norm": 0.7635822892189026, + "learning_rate": 4.004960145480651e-06, + "loss": 0.4981, + "step": 634 + }, + { + "epoch": 1.9864300626304803, + "grad_norm": 0.8916334509849548, + "learning_rate": 4.0015006311776685e-06, + "loss": 0.5311, + "step": 635 + }, + { + "epoch": 1.989561586638831, + "grad_norm": 0.7197673320770264, + "learning_rate": 3.998036613370295e-06, + "loss": 0.5361, + "step": 636 + }, + { + "epoch": 1.9926931106471817, + "grad_norm": 0.8391228914260864, + "learning_rate": 3.994568102448284e-06, + "loss": 0.5473, + "step": 637 + }, + { + "epoch": 1.9958246346555324, + "grad_norm": 0.9371750950813293, + "learning_rate": 3.991095108814862e-06, + "loss": 0.5303, + "step": 638 + }, + { + "epoch": 1.998956158663883, + "grad_norm": 0.8929619789123535, + "learning_rate": 3.9876176428867046e-06, + "loss": 0.533, + "step": 639 + }, + { + "epoch": 2.0, + "grad_norm": 0.8929619789123535, + "learning_rate": 3.9841357150938984e-06, + "loss": 0.1831, + "step": 640 + }, + { + "epoch": 2.0031315240083507, + "grad_norm": 0.8802503347396851, + "learning_rate": 3.9806493358799135e-06, + "loss": 0.493, + "step": 641 + }, + { + "epoch": 2.0062630480167014, + "grad_norm": 0.802759051322937, + "learning_rate": 3.977158515701571e-06, + "loss": 0.498, + "step": 642 + }, + { + "epoch": 2.009394572025052, + "grad_norm": 1.0235401391983032, + "learning_rate": 3.973663265029013e-06, + "loss": 0.4887, + "step": 643 + }, + { + "epoch": 2.012526096033403, + "grad_norm": 0.7219089865684509, + "learning_rate": 3.97016359434567e-06, + "loss": 0.4628, + "step": 644 + }, + { + "epoch": 2.0156576200417535, + "grad_norm": 0.7887073755264282, + "learning_rate": 3.966659514148229e-06, + "loss": 0.525, + "step": 645 + }, + { + "epoch": 2.0187891440501042, + "grad_norm": 0.7960914969444275, + "learning_rate": 3.963151034946602e-06, + "loss": 0.4643, + "step": 646 + }, + { + "epoch": 2.021920668058455, + "grad_norm": 0.7902271151542664, + "learning_rate": 3.959638167263895e-06, + "loss": 0.4922, + "step": 647 + }, + { + "epoch": 2.0250521920668056, + "grad_norm": 0.9501478672027588, + "learning_rate": 3.956120921636379e-06, + "loss": 0.5285, + "step": 648 + }, + { + "epoch": 2.028183716075157, + "grad_norm": 0.9510527849197388, + "learning_rate": 3.952599308613454e-06, + "loss": 0.4909, + "step": 649 + }, + { + "epoch": 2.0313152400835075, + "grad_norm": 0.9408219456672668, + "learning_rate": 3.949073338757619e-06, + "loss": 0.4912, + "step": 650 + }, + { + "epoch": 2.034446764091858, + "grad_norm": 0.7148041725158691, + "learning_rate": 3.945543022644441e-06, + "loss": 0.4792, + "step": 651 + }, + { + "epoch": 2.037578288100209, + "grad_norm": 0.7737464904785156, + "learning_rate": 3.942008370862522e-06, + "loss": 0.4694, + "step": 652 + }, + { + "epoch": 2.0407098121085596, + "grad_norm": 0.8405889868736267, + "learning_rate": 3.938469394013472e-06, + "loss": 0.5048, + "step": 653 + }, + { + "epoch": 2.0438413361169103, + "grad_norm": 0.7896456718444824, + "learning_rate": 3.934926102711869e-06, + "loss": 0.4882, + "step": 654 + }, + { + "epoch": 2.046972860125261, + "grad_norm": 0.9290387034416199, + "learning_rate": 3.931378507585231e-06, + "loss": 0.503, + "step": 655 + }, + { + "epoch": 2.0501043841336117, + "grad_norm": 0.7386118769645691, + "learning_rate": 3.927826619273991e-06, + "loss": 0.4918, + "step": 656 + }, + { + "epoch": 2.0532359081419624, + "grad_norm": 0.9878676533699036, + "learning_rate": 3.92427044843145e-06, + "loss": 0.4958, + "step": 657 + }, + { + "epoch": 2.056367432150313, + "grad_norm": 1.0111151933670044, + "learning_rate": 3.92071000572376e-06, + "loss": 0.4886, + "step": 658 + }, + { + "epoch": 2.059498956158664, + "grad_norm": 0.8612061738967896, + "learning_rate": 3.917145301829884e-06, + "loss": 0.5216, + "step": 659 + }, + { + "epoch": 2.0626304801670146, + "grad_norm": 0.7458518743515015, + "learning_rate": 3.913576347441564e-06, + "loss": 0.4807, + "step": 660 + }, + { + "epoch": 2.0657620041753653, + "grad_norm": 0.7775886058807373, + "learning_rate": 3.910003153263294e-06, + "loss": 0.4837, + "step": 661 + }, + { + "epoch": 2.068893528183716, + "grad_norm": 0.7144196629524231, + "learning_rate": 3.906425730012282e-06, + "loss": 0.5081, + "step": 662 + }, + { + "epoch": 2.0720250521920667, + "grad_norm": 0.844971776008606, + "learning_rate": 3.9028440884184215e-06, + "loss": 0.474, + "step": 663 + }, + { + "epoch": 2.0751565762004174, + "grad_norm": 0.9709283113479614, + "learning_rate": 3.899258239224257e-06, + "loss": 0.503, + "step": 664 + }, + { + "epoch": 2.078288100208768, + "grad_norm": 1.1325515508651733, + "learning_rate": 3.895668193184954e-06, + "loss": 0.5058, + "step": 665 + }, + { + "epoch": 2.081419624217119, + "grad_norm": 0.7211254239082336, + "learning_rate": 3.892073961068266e-06, + "loss": 0.4982, + "step": 666 + }, + { + "epoch": 2.08455114822547, + "grad_norm": 0.8975517153739929, + "learning_rate": 3.888475553654502e-06, + "loss": 0.4699, + "step": 667 + }, + { + "epoch": 2.0876826722338206, + "grad_norm": 0.8270771503448486, + "learning_rate": 3.884872981736493e-06, + "loss": 0.4586, + "step": 668 + }, + { + "epoch": 2.0908141962421714, + "grad_norm": 0.8606625199317932, + "learning_rate": 3.881266256119561e-06, + "loss": 0.5299, + "step": 669 + }, + { + "epoch": 2.093945720250522, + "grad_norm": 0.9013976454734802, + "learning_rate": 3.877655387621488e-06, + "loss": 0.4887, + "step": 670 + }, + { + "epoch": 2.0970772442588728, + "grad_norm": 0.7603903412818909, + "learning_rate": 3.8740403870724795e-06, + "loss": 0.4992, + "step": 671 + }, + { + "epoch": 2.1002087682672235, + "grad_norm": 1.0432350635528564, + "learning_rate": 3.870421265315137e-06, + "loss": 0.5035, + "step": 672 + }, + { + "epoch": 2.103340292275574, + "grad_norm": 0.7727136611938477, + "learning_rate": 3.8667980332044195e-06, + "loss": 0.5006, + "step": 673 + }, + { + "epoch": 2.106471816283925, + "grad_norm": 0.9764307141304016, + "learning_rate": 3.863170701607618e-06, + "loss": 0.5061, + "step": 674 + }, + { + "epoch": 2.1096033402922756, + "grad_norm": 0.747818648815155, + "learning_rate": 3.859539281404317e-06, + "loss": 0.4761, + "step": 675 + }, + { + "epoch": 2.1127348643006263, + "grad_norm": 0.7254915237426758, + "learning_rate": 3.855903783486364e-06, + "loss": 0.5166, + "step": 676 + }, + { + "epoch": 2.115866388308977, + "grad_norm": 0.7678592801094055, + "learning_rate": 3.852264218757839e-06, + "loss": 0.5122, + "step": 677 + }, + { + "epoch": 2.1189979123173277, + "grad_norm": 0.8140144348144531, + "learning_rate": 3.8486205981350165e-06, + "loss": 0.4551, + "step": 678 + }, + { + "epoch": 2.1221294363256784, + "grad_norm": 0.9417359232902527, + "learning_rate": 3.844972932546338e-06, + "loss": 0.4748, + "step": 679 + }, + { + "epoch": 2.125260960334029, + "grad_norm": 0.8035290241241455, + "learning_rate": 3.841321232932378e-06, + "loss": 0.5079, + "step": 680 + }, + { + "epoch": 2.12839248434238, + "grad_norm": 0.8300641775131226, + "learning_rate": 3.837665510245809e-06, + "loss": 0.5018, + "step": 681 + }, + { + "epoch": 2.1315240083507305, + "grad_norm": 0.8293547034263611, + "learning_rate": 3.8340057754513715e-06, + "loss": 0.5042, + "step": 682 + }, + { + "epoch": 2.1346555323590812, + "grad_norm": 0.7780007719993591, + "learning_rate": 3.8303420395258365e-06, + "loss": 0.5048, + "step": 683 + }, + { + "epoch": 2.137787056367432, + "grad_norm": 0.7519420385360718, + "learning_rate": 3.8266743134579785e-06, + "loss": 0.5108, + "step": 684 + }, + { + "epoch": 2.140918580375783, + "grad_norm": 0.7872384190559387, + "learning_rate": 3.8230026082485404e-06, + "loss": 0.4924, + "step": 685 + }, + { + "epoch": 2.144050104384134, + "grad_norm": 0.7479491829872131, + "learning_rate": 3.819326934910197e-06, + "loss": 0.5184, + "step": 686 + }, + { + "epoch": 2.1471816283924845, + "grad_norm": 0.8438299298286438, + "learning_rate": 3.815647304467527e-06, + "loss": 0.4791, + "step": 687 + }, + { + "epoch": 2.150313152400835, + "grad_norm": 0.7923721671104431, + "learning_rate": 3.8119637279569773e-06, + "loss": 0.5305, + "step": 688 + }, + { + "epoch": 2.153444676409186, + "grad_norm": 0.7856534719467163, + "learning_rate": 3.80827621642683e-06, + "loss": 0.5063, + "step": 689 + }, + { + "epoch": 2.1565762004175366, + "grad_norm": 0.8544500470161438, + "learning_rate": 3.8045847809371706e-06, + "loss": 0.4989, + "step": 690 + }, + { + "epoch": 2.1597077244258873, + "grad_norm": 0.865390956401825, + "learning_rate": 3.800889432559852e-06, + "loss": 0.4931, + "step": 691 + }, + { + "epoch": 2.162839248434238, + "grad_norm": 0.9809399247169495, + "learning_rate": 3.797190182378466e-06, + "loss": 0.4785, + "step": 692 + }, + { + "epoch": 2.1659707724425887, + "grad_norm": 0.7954536080360413, + "learning_rate": 3.793487041488304e-06, + "loss": 0.4847, + "step": 693 + }, + { + "epoch": 2.1691022964509394, + "grad_norm": 0.754704475402832, + "learning_rate": 3.7897800209963298e-06, + "loss": 0.5125, + "step": 694 + }, + { + "epoch": 2.17223382045929, + "grad_norm": 0.7319822311401367, + "learning_rate": 3.7860691320211414e-06, + "loss": 0.477, + "step": 695 + }, + { + "epoch": 2.175365344467641, + "grad_norm": 0.8198635578155518, + "learning_rate": 3.7823543856929403e-06, + "loss": 0.4764, + "step": 696 + }, + { + "epoch": 2.1784968684759916, + "grad_norm": 0.708933413028717, + "learning_rate": 3.7786357931534987e-06, + "loss": 0.4948, + "step": 697 + }, + { + "epoch": 2.1816283924843423, + "grad_norm": 0.8493193984031677, + "learning_rate": 3.774913365556123e-06, + "loss": 0.5271, + "step": 698 + }, + { + "epoch": 2.184759916492693, + "grad_norm": 0.7999475002288818, + "learning_rate": 3.771187114065622e-06, + "loss": 0.4804, + "step": 699 + }, + { + "epoch": 2.1878914405010437, + "grad_norm": 0.8366796970367432, + "learning_rate": 3.7674570498582776e-06, + "loss": 0.457, + "step": 700 + }, + { + "epoch": 2.1910229645093944, + "grad_norm": 0.7935530543327332, + "learning_rate": 3.7637231841218015e-06, + "loss": 0.5001, + "step": 701 + }, + { + "epoch": 2.1941544885177455, + "grad_norm": 0.7700153589248657, + "learning_rate": 3.7599855280553125e-06, + "loss": 0.5091, + "step": 702 + }, + { + "epoch": 2.1972860125260962, + "grad_norm": 0.7991652488708496, + "learning_rate": 3.756244092869294e-06, + "loss": 0.4955, + "step": 703 + }, + { + "epoch": 2.200417536534447, + "grad_norm": 0.720051646232605, + "learning_rate": 3.752498889785567e-06, + "loss": 0.4902, + "step": 704 + }, + { + "epoch": 2.2035490605427976, + "grad_norm": 0.7312369346618652, + "learning_rate": 3.748749930037252e-06, + "loss": 0.4935, + "step": 705 + }, + { + "epoch": 2.2066805845511483, + "grad_norm": 0.8418563008308411, + "learning_rate": 3.744997224868739e-06, + "loss": 0.5186, + "step": 706 + }, + { + "epoch": 2.209812108559499, + "grad_norm": 0.8324081301689148, + "learning_rate": 3.741240785535649e-06, + "loss": 0.482, + "step": 707 + }, + { + "epoch": 2.2129436325678498, + "grad_norm": 0.8051855564117432, + "learning_rate": 3.737480623304805e-06, + "loss": 0.4663, + "step": 708 + }, + { + "epoch": 2.2160751565762005, + "grad_norm": 0.9464184641838074, + "learning_rate": 3.7337167494541948e-06, + "loss": 0.451, + "step": 709 + }, + { + "epoch": 2.219206680584551, + "grad_norm": 1.0227075815200806, + "learning_rate": 3.729949175272942e-06, + "loss": 0.4428, + "step": 710 + }, + { + "epoch": 2.222338204592902, + "grad_norm": 0.7930364012718201, + "learning_rate": 3.7261779120612633e-06, + "loss": 0.5132, + "step": 711 + }, + { + "epoch": 2.2254697286012526, + "grad_norm": 0.9033688306808472, + "learning_rate": 3.7224029711304444e-06, + "loss": 0.476, + "step": 712 + }, + { + "epoch": 2.2286012526096033, + "grad_norm": 0.8026887774467468, + "learning_rate": 3.7186243638028007e-06, + "loss": 0.4959, + "step": 713 + }, + { + "epoch": 2.231732776617954, + "grad_norm": 0.9391745328903198, + "learning_rate": 3.714842101411642e-06, + "loss": 0.4962, + "step": 714 + }, + { + "epoch": 2.2348643006263047, + "grad_norm": 0.7774361371994019, + "learning_rate": 3.711056195301245e-06, + "loss": 0.4748, + "step": 715 + }, + { + "epoch": 2.2379958246346554, + "grad_norm": 0.9278722405433655, + "learning_rate": 3.7072666568268115e-06, + "loss": 0.5074, + "step": 716 + }, + { + "epoch": 2.241127348643006, + "grad_norm": 0.771423876285553, + "learning_rate": 3.7034734973544406e-06, + "loss": 0.5072, + "step": 717 + }, + { + "epoch": 2.244258872651357, + "grad_norm": 0.8707448244094849, + "learning_rate": 3.6996767282610892e-06, + "loss": 0.4851, + "step": 718 + }, + { + "epoch": 2.2473903966597075, + "grad_norm": 0.7641019821166992, + "learning_rate": 3.695876360934543e-06, + "loss": 0.4941, + "step": 719 + }, + { + "epoch": 2.2505219206680582, + "grad_norm": 0.7647167444229126, + "learning_rate": 3.69207240677338e-06, + "loss": 0.5225, + "step": 720 + }, + { + "epoch": 2.2536534446764094, + "grad_norm": 0.9108865261077881, + "learning_rate": 3.6882648771869345e-06, + "loss": 0.454, + "step": 721 + }, + { + "epoch": 2.25678496868476, + "grad_norm": 0.86728835105896, + "learning_rate": 3.6844537835952666e-06, + "loss": 0.4461, + "step": 722 + }, + { + "epoch": 2.259916492693111, + "grad_norm": 1.1055282354354858, + "learning_rate": 3.6806391374291238e-06, + "loss": 0.4618, + "step": 723 + }, + { + "epoch": 2.2630480167014615, + "grad_norm": 0.7591858506202698, + "learning_rate": 3.6768209501299116e-06, + "loss": 0.4901, + "step": 724 + }, + { + "epoch": 2.266179540709812, + "grad_norm": 0.7966359257698059, + "learning_rate": 3.6729992331496554e-06, + "loss": 0.5171, + "step": 725 + }, + { + "epoch": 2.269311064718163, + "grad_norm": 0.983161211013794, + "learning_rate": 3.6691739979509672e-06, + "loss": 0.4949, + "step": 726 + }, + { + "epoch": 2.2724425887265136, + "grad_norm": 0.9200037121772766, + "learning_rate": 3.6653452560070106e-06, + "loss": 0.5234, + "step": 727 + }, + { + "epoch": 2.2755741127348643, + "grad_norm": 1.0288461446762085, + "learning_rate": 3.6615130188014685e-06, + "loss": 0.4713, + "step": 728 + }, + { + "epoch": 2.278705636743215, + "grad_norm": 0.7325463891029358, + "learning_rate": 3.6576772978285065e-06, + "loss": 0.527, + "step": 729 + }, + { + "epoch": 2.2818371607515657, + "grad_norm": 1.0045446157455444, + "learning_rate": 3.6538381045927395e-06, + "loss": 0.5139, + "step": 730 + }, + { + "epoch": 2.2849686847599164, + "grad_norm": 0.7391849756240845, + "learning_rate": 3.6499954506091963e-06, + "loss": 0.4829, + "step": 731 + }, + { + "epoch": 2.288100208768267, + "grad_norm": 0.7808229923248291, + "learning_rate": 3.646149347403286e-06, + "loss": 0.4831, + "step": 732 + }, + { + "epoch": 2.291231732776618, + "grad_norm": 0.7056961059570312, + "learning_rate": 3.6422998065107628e-06, + "loss": 0.5066, + "step": 733 + }, + { + "epoch": 2.2943632567849686, + "grad_norm": 0.7498443126678467, + "learning_rate": 3.6384468394776935e-06, + "loss": 0.4724, + "step": 734 + }, + { + "epoch": 2.2974947807933193, + "grad_norm": 0.8511576056480408, + "learning_rate": 3.634590457860418e-06, + "loss": 0.5286, + "step": 735 + }, + { + "epoch": 2.30062630480167, + "grad_norm": 0.873635470867157, + "learning_rate": 3.63073067322552e-06, + "loss": 0.4751, + "step": 736 + }, + { + "epoch": 2.3037578288100207, + "grad_norm": 0.7427377104759216, + "learning_rate": 3.626867497149788e-06, + "loss": 0.475, + "step": 737 + }, + { + "epoch": 2.306889352818372, + "grad_norm": 1.0591017007827759, + "learning_rate": 3.623000941220186e-06, + "loss": 0.4591, + "step": 738 + }, + { + "epoch": 2.3100208768267225, + "grad_norm": 0.8767879009246826, + "learning_rate": 3.6191310170338114e-06, + "loss": 0.4673, + "step": 739 + }, + { + "epoch": 2.3131524008350732, + "grad_norm": 0.9156234860420227, + "learning_rate": 3.615257736197866e-06, + "loss": 0.4622, + "step": 740 + }, + { + "epoch": 2.316283924843424, + "grad_norm": 0.6743756532669067, + "learning_rate": 3.611381110329619e-06, + "loss": 0.4723, + "step": 741 + }, + { + "epoch": 2.3194154488517746, + "grad_norm": 0.8655558228492737, + "learning_rate": 3.6075011510563732e-06, + "loss": 0.471, + "step": 742 + }, + { + "epoch": 2.3225469728601253, + "grad_norm": 0.7652033567428589, + "learning_rate": 3.603617870015429e-06, + "loss": 0.5155, + "step": 743 + }, + { + "epoch": 2.325678496868476, + "grad_norm": 0.7970699667930603, + "learning_rate": 3.599731278854049e-06, + "loss": 0.4507, + "step": 744 + }, + { + "epoch": 2.3288100208768268, + "grad_norm": 0.7538278698921204, + "learning_rate": 3.5958413892294253e-06, + "loss": 0.5093, + "step": 745 + }, + { + "epoch": 2.3319415448851775, + "grad_norm": 0.735996663570404, + "learning_rate": 3.5919482128086414e-06, + "loss": 0.5008, + "step": 746 + }, + { + "epoch": 2.335073068893528, + "grad_norm": 0.7643904685974121, + "learning_rate": 3.588051761268642e-06, + "loss": 0.5072, + "step": 747 + }, + { + "epoch": 2.338204592901879, + "grad_norm": 0.7646260857582092, + "learning_rate": 3.584152046296191e-06, + "loss": 0.4578, + "step": 748 + }, + { + "epoch": 2.3413361169102296, + "grad_norm": 0.7873825430870056, + "learning_rate": 3.5802490795878446e-06, + "loss": 0.5249, + "step": 749 + }, + { + "epoch": 2.3444676409185803, + "grad_norm": 1.095333218574524, + "learning_rate": 3.5763428728499095e-06, + "loss": 0.4913, + "step": 750 + }, + { + "epoch": 2.347599164926931, + "grad_norm": 1.3425395488739014, + "learning_rate": 3.5724334377984107e-06, + "loss": 0.5317, + "step": 751 + }, + { + "epoch": 2.3507306889352817, + "grad_norm": 0.7151113748550415, + "learning_rate": 3.568520786159055e-06, + "loss": 0.5135, + "step": 752 + }, + { + "epoch": 2.3538622129436324, + "grad_norm": 0.8072878122329712, + "learning_rate": 3.5646049296672004e-06, + "loss": 0.4863, + "step": 753 + }, + { + "epoch": 2.356993736951983, + "grad_norm": 0.8040189743041992, + "learning_rate": 3.5606858800678123e-06, + "loss": 0.4668, + "step": 754 + }, + { + "epoch": 2.3601252609603343, + "grad_norm": 0.7749765515327454, + "learning_rate": 3.5567636491154385e-06, + "loss": 0.4681, + "step": 755 + }, + { + "epoch": 2.3632567849686845, + "grad_norm": 0.773013710975647, + "learning_rate": 3.5528382485741638e-06, + "loss": 0.5012, + "step": 756 + }, + { + "epoch": 2.3663883089770357, + "grad_norm": 0.7017714381217957, + "learning_rate": 3.5489096902175835e-06, + "loss": 0.5019, + "step": 757 + }, + { + "epoch": 2.3695198329853864, + "grad_norm": 1.132458209991455, + "learning_rate": 3.5449779858287625e-06, + "loss": 0.5131, + "step": 758 + }, + { + "epoch": 2.372651356993737, + "grad_norm": 0.7624574899673462, + "learning_rate": 3.541043147200202e-06, + "loss": 0.4856, + "step": 759 + }, + { + "epoch": 2.375782881002088, + "grad_norm": 0.9078478217124939, + "learning_rate": 3.5371051861338036e-06, + "loss": 0.4337, + "step": 760 + }, + { + "epoch": 2.3789144050104385, + "grad_norm": 0.8608354330062866, + "learning_rate": 3.5331641144408344e-06, + "loss": 0.5053, + "step": 761 + }, + { + "epoch": 2.382045929018789, + "grad_norm": 0.775047779083252, + "learning_rate": 3.529219943941892e-06, + "loss": 0.4779, + "step": 762 + }, + { + "epoch": 2.38517745302714, + "grad_norm": 0.7775866389274597, + "learning_rate": 3.525272686466866e-06, + "loss": 0.4979, + "step": 763 + }, + { + "epoch": 2.3883089770354906, + "grad_norm": 0.9386464357376099, + "learning_rate": 3.521322353854908e-06, + "loss": 0.5222, + "step": 764 + }, + { + "epoch": 2.3914405010438413, + "grad_norm": 0.874109148979187, + "learning_rate": 3.517368957954391e-06, + "loss": 0.4681, + "step": 765 + }, + { + "epoch": 2.394572025052192, + "grad_norm": 0.824588418006897, + "learning_rate": 3.5134125106228766e-06, + "loss": 0.4955, + "step": 766 + }, + { + "epoch": 2.3977035490605427, + "grad_norm": 0.8790764808654785, + "learning_rate": 3.5094530237270774e-06, + "loss": 0.4722, + "step": 767 + }, + { + "epoch": 2.4008350730688934, + "grad_norm": 1.1399786472320557, + "learning_rate": 3.5054905091428253e-06, + "loss": 0.4771, + "step": 768 + }, + { + "epoch": 2.403966597077244, + "grad_norm": 1.2586532831192017, + "learning_rate": 3.50152497875503e-06, + "loss": 0.4849, + "step": 769 + }, + { + "epoch": 2.407098121085595, + "grad_norm": 0.7706464529037476, + "learning_rate": 3.4975564444576487e-06, + "loss": 0.477, + "step": 770 + }, + { + "epoch": 2.4102296450939455, + "grad_norm": 0.7695909142494202, + "learning_rate": 3.4935849181536484e-06, + "loss": 0.4695, + "step": 771 + }, + { + "epoch": 2.4133611691022967, + "grad_norm": 0.7744433283805847, + "learning_rate": 3.489610411754969e-06, + "loss": 0.499, + "step": 772 + }, + { + "epoch": 2.416492693110647, + "grad_norm": 0.9265744686126709, + "learning_rate": 3.48563293718249e-06, + "loss": 0.481, + "step": 773 + }, + { + "epoch": 2.419624217118998, + "grad_norm": 1.0680506229400635, + "learning_rate": 3.481652506365992e-06, + "loss": 0.4898, + "step": 774 + }, + { + "epoch": 2.422755741127349, + "grad_norm": 0.721493661403656, + "learning_rate": 3.477669131244122e-06, + "loss": 0.4813, + "step": 775 + }, + { + "epoch": 2.4258872651356995, + "grad_norm": 0.7993559837341309, + "learning_rate": 3.4736828237643616e-06, + "loss": 0.5179, + "step": 776 + }, + { + "epoch": 2.4290187891440502, + "grad_norm": 0.8148090839385986, + "learning_rate": 3.4696935958829837e-06, + "loss": 0.4753, + "step": 777 + }, + { + "epoch": 2.432150313152401, + "grad_norm": 0.8006406426429749, + "learning_rate": 3.465701459565022e-06, + "loss": 0.501, + "step": 778 + }, + { + "epoch": 2.4352818371607516, + "grad_norm": 0.9307970404624939, + "learning_rate": 3.4617064267842327e-06, + "loss": 0.487, + "step": 779 + }, + { + "epoch": 2.4384133611691023, + "grad_norm": 0.7192814946174622, + "learning_rate": 3.45770850952306e-06, + "loss": 0.4769, + "step": 780 + }, + { + "epoch": 2.441544885177453, + "grad_norm": 0.7386271953582764, + "learning_rate": 3.4537077197726023e-06, + "loss": 0.4726, + "step": 781 + }, + { + "epoch": 2.4446764091858038, + "grad_norm": 0.8006314039230347, + "learning_rate": 3.449704069532567e-06, + "loss": 0.494, + "step": 782 + }, + { + "epoch": 2.4478079331941545, + "grad_norm": 0.7466752529144287, + "learning_rate": 3.4456975708112477e-06, + "loss": 0.4778, + "step": 783 + }, + { + "epoch": 2.450939457202505, + "grad_norm": 0.8348856568336487, + "learning_rate": 3.4416882356254777e-06, + "loss": 0.4766, + "step": 784 + }, + { + "epoch": 2.454070981210856, + "grad_norm": 0.754851758480072, + "learning_rate": 3.4376760760005994e-06, + "loss": 0.4673, + "step": 785 + }, + { + "epoch": 2.4572025052192066, + "grad_norm": 0.7854018807411194, + "learning_rate": 3.433661103970427e-06, + "loss": 0.4954, + "step": 786 + }, + { + "epoch": 2.4603340292275573, + "grad_norm": 0.7238256931304932, + "learning_rate": 3.4296433315772084e-06, + "loss": 0.496, + "step": 787 + }, + { + "epoch": 2.463465553235908, + "grad_norm": 0.7007659673690796, + "learning_rate": 3.4256227708715915e-06, + "loss": 0.4793, + "step": 788 + }, + { + "epoch": 2.4665970772442587, + "grad_norm": 0.7234371900558472, + "learning_rate": 3.421599433912588e-06, + "loss": 0.4935, + "step": 789 + }, + { + "epoch": 2.4697286012526094, + "grad_norm": 0.7537544965744019, + "learning_rate": 3.4175733327675355e-06, + "loss": 0.5194, + "step": 790 + }, + { + "epoch": 2.4728601252609606, + "grad_norm": 0.7608047127723694, + "learning_rate": 3.4135444795120633e-06, + "loss": 0.4793, + "step": 791 + }, + { + "epoch": 2.4759916492693113, + "grad_norm": 0.7847898006439209, + "learning_rate": 3.4095128862300542e-06, + "loss": 0.4877, + "step": 792 + }, + { + "epoch": 2.479123173277662, + "grad_norm": 0.8002011179924011, + "learning_rate": 3.405478565013609e-06, + "loss": 0.4927, + "step": 793 + }, + { + "epoch": 2.4822546972860127, + "grad_norm": 0.8200219869613647, + "learning_rate": 3.401441527963013e-06, + "loss": 0.4997, + "step": 794 + }, + { + "epoch": 2.4853862212943634, + "grad_norm": 0.7220162749290466, + "learning_rate": 3.3974017871866938e-06, + "loss": 0.4668, + "step": 795 + }, + { + "epoch": 2.488517745302714, + "grad_norm": 0.8022251129150391, + "learning_rate": 3.3933593548011912e-06, + "loss": 0.5179, + "step": 796 + }, + { + "epoch": 2.491649269311065, + "grad_norm": 0.7914465069770813, + "learning_rate": 3.389314242931115e-06, + "loss": 0.4943, + "step": 797 + }, + { + "epoch": 2.4947807933194155, + "grad_norm": 1.1399403810501099, + "learning_rate": 3.385266463709116e-06, + "loss": 0.4896, + "step": 798 + }, + { + "epoch": 2.497912317327766, + "grad_norm": 0.8098909854888916, + "learning_rate": 3.38121602927584e-06, + "loss": 0.4904, + "step": 799 + }, + { + "epoch": 2.501043841336117, + "grad_norm": 0.7434052228927612, + "learning_rate": 3.377162951779902e-06, + "loss": 0.4864, + "step": 800 + }, + { + "epoch": 2.5041753653444676, + "grad_norm": 0.7397809624671936, + "learning_rate": 3.3731072433778407e-06, + "loss": 0.486, + "step": 801 + }, + { + "epoch": 2.5073068893528183, + "grad_norm": 0.99027019739151, + "learning_rate": 3.3690489162340867e-06, + "loss": 0.5011, + "step": 802 + }, + { + "epoch": 2.510438413361169, + "grad_norm": 0.8443610668182373, + "learning_rate": 3.3649879825209246e-06, + "loss": 0.455, + "step": 803 + }, + { + "epoch": 2.5135699373695197, + "grad_norm": 0.755649983882904, + "learning_rate": 3.3609244544184604e-06, + "loss": 0.4563, + "step": 804 + }, + { + "epoch": 2.5167014613778704, + "grad_norm": 0.728018045425415, + "learning_rate": 3.3568583441145765e-06, + "loss": 0.471, + "step": 805 + }, + { + "epoch": 2.519832985386221, + "grad_norm": 0.7777130603790283, + "learning_rate": 3.352789663804904e-06, + "loss": 0.4667, + "step": 806 + }, + { + "epoch": 2.522964509394572, + "grad_norm": 0.7545619606971741, + "learning_rate": 3.3487184256927785e-06, + "loss": 0.4915, + "step": 807 + }, + { + "epoch": 2.526096033402923, + "grad_norm": 0.8374579548835754, + "learning_rate": 3.3446446419892127e-06, + "loss": 0.485, + "step": 808 + }, + { + "epoch": 2.5292275574112733, + "grad_norm": 0.7354666590690613, + "learning_rate": 3.340568324912849e-06, + "loss": 0.5254, + "step": 809 + }, + { + "epoch": 2.5323590814196244, + "grad_norm": 0.7581545114517212, + "learning_rate": 3.3364894866899324e-06, + "loss": 0.4483, + "step": 810 + }, + { + "epoch": 2.535490605427975, + "grad_norm": 0.8077559471130371, + "learning_rate": 3.3324081395542662e-06, + "loss": 0.5022, + "step": 811 + }, + { + "epoch": 2.538622129436326, + "grad_norm": 0.8827865719795227, + "learning_rate": 3.3283242957471806e-06, + "loss": 0.4909, + "step": 812 + }, + { + "epoch": 2.5417536534446765, + "grad_norm": 0.9139482378959656, + "learning_rate": 3.3242379675174953e-06, + "loss": 0.5205, + "step": 813 + }, + { + "epoch": 2.5448851774530272, + "grad_norm": 0.7616812586784363, + "learning_rate": 3.3201491671214797e-06, + "loss": 0.4744, + "step": 814 + }, + { + "epoch": 2.548016701461378, + "grad_norm": 0.987173318862915, + "learning_rate": 3.3160579068228183e-06, + "loss": 0.4876, + "step": 815 + }, + { + "epoch": 2.5511482254697286, + "grad_norm": 1.259137749671936, + "learning_rate": 3.311964198892574e-06, + "loss": 0.454, + "step": 816 + }, + { + "epoch": 2.5542797494780793, + "grad_norm": 0.7866336703300476, + "learning_rate": 3.3078680556091513e-06, + "loss": 0.5107, + "step": 817 + }, + { + "epoch": 2.55741127348643, + "grad_norm": 0.9311352372169495, + "learning_rate": 3.303769489258258e-06, + "loss": 0.4843, + "step": 818 + }, + { + "epoch": 2.5605427974947808, + "grad_norm": 0.8556346893310547, + "learning_rate": 3.299668512132872e-06, + "loss": 0.5017, + "step": 819 + }, + { + "epoch": 2.5636743215031315, + "grad_norm": 2.810598373413086, + "learning_rate": 3.2955651365331988e-06, + "loss": 0.5223, + "step": 820 + }, + { + "epoch": 2.566805845511482, + "grad_norm": 1.0120766162872314, + "learning_rate": 3.29145937476664e-06, + "loss": 0.4959, + "step": 821 + }, + { + "epoch": 2.569937369519833, + "grad_norm": 0.751412034034729, + "learning_rate": 3.287351239147752e-06, + "loss": 0.4941, + "step": 822 + }, + { + "epoch": 2.5730688935281836, + "grad_norm": 1.9308148622512817, + "learning_rate": 3.2832407419982136e-06, + "loss": 0.4965, + "step": 823 + }, + { + "epoch": 2.5762004175365343, + "grad_norm": 0.9215649962425232, + "learning_rate": 3.279127895646786e-06, + "loss": 0.5071, + "step": 824 + }, + { + "epoch": 2.5793319415448854, + "grad_norm": 0.7599574327468872, + "learning_rate": 3.2750127124292754e-06, + "loss": 0.5191, + "step": 825 + }, + { + "epoch": 2.5824634655532357, + "grad_norm": 0.8234940767288208, + "learning_rate": 3.270895204688496e-06, + "loss": 0.4947, + "step": 826 + }, + { + "epoch": 2.585594989561587, + "grad_norm": 0.8401572704315186, + "learning_rate": 3.266775384774238e-06, + "loss": 0.4547, + "step": 827 + }, + { + "epoch": 2.588726513569937, + "grad_norm": 0.8927991986274719, + "learning_rate": 3.262653265043223e-06, + "loss": 0.4296, + "step": 828 + }, + { + "epoch": 2.5918580375782883, + "grad_norm": 0.8009241223335266, + "learning_rate": 3.2585288578590716e-06, + "loss": 0.4578, + "step": 829 + }, + { + "epoch": 2.594989561586639, + "grad_norm": 0.7982021570205688, + "learning_rate": 3.2544021755922663e-06, + "loss": 0.4961, + "step": 830 + }, + { + "epoch": 2.5981210855949897, + "grad_norm": 0.7096095681190491, + "learning_rate": 3.2502732306201112e-06, + "loss": 0.4975, + "step": 831 + }, + { + "epoch": 2.6012526096033404, + "grad_norm": 1.1092045307159424, + "learning_rate": 3.246142035326699e-06, + "loss": 0.4705, + "step": 832 + }, + { + "epoch": 2.604384133611691, + "grad_norm": 0.785799503326416, + "learning_rate": 3.24200860210287e-06, + "loss": 0.479, + "step": 833 + }, + { + "epoch": 2.607515657620042, + "grad_norm": 0.7315773367881775, + "learning_rate": 3.2378729433461804e-06, + "loss": 0.5036, + "step": 834 + }, + { + "epoch": 2.6106471816283925, + "grad_norm": 0.7840189337730408, + "learning_rate": 3.233735071460856e-06, + "loss": 0.4967, + "step": 835 + }, + { + "epoch": 2.613778705636743, + "grad_norm": 0.7186565399169922, + "learning_rate": 3.2295949988577655e-06, + "loss": 0.4889, + "step": 836 + }, + { + "epoch": 2.616910229645094, + "grad_norm": 0.766054093837738, + "learning_rate": 3.2254527379543747e-06, + "loss": 0.539, + "step": 837 + }, + { + "epoch": 2.6200417536534446, + "grad_norm": 0.7705381512641907, + "learning_rate": 3.2213083011747165e-06, + "loss": 0.4968, + "step": 838 + }, + { + "epoch": 2.6231732776617953, + "grad_norm": 1.3530604839324951, + "learning_rate": 3.217161700949346e-06, + "loss": 0.52, + "step": 839 + }, + { + "epoch": 2.626304801670146, + "grad_norm": 0.737389862537384, + "learning_rate": 3.2130129497153107e-06, + "loss": 0.4823, + "step": 840 + }, + { + "epoch": 2.6294363256784967, + "grad_norm": 0.9121193885803223, + "learning_rate": 3.2088620599161064e-06, + "loss": 0.4592, + "step": 841 + }, + { + "epoch": 2.632567849686848, + "grad_norm": 0.8869616389274597, + "learning_rate": 3.2047090440016464e-06, + "loss": 0.5001, + "step": 842 + }, + { + "epoch": 2.635699373695198, + "grad_norm": 0.8447219133377075, + "learning_rate": 3.200553914428219e-06, + "loss": 0.4969, + "step": 843 + }, + { + "epoch": 2.6388308977035493, + "grad_norm": 0.8877657055854797, + "learning_rate": 3.1963966836584524e-06, + "loss": 0.4718, + "step": 844 + }, + { + "epoch": 2.6419624217118995, + "grad_norm": 1.045272946357727, + "learning_rate": 3.192237364161277e-06, + "loss": 0.4864, + "step": 845 + }, + { + "epoch": 2.6450939457202507, + "grad_norm": 0.8485913276672363, + "learning_rate": 3.1880759684118876e-06, + "loss": 0.4688, + "step": 846 + }, + { + "epoch": 2.6482254697286014, + "grad_norm": 0.7328930497169495, + "learning_rate": 3.183912508891709e-06, + "loss": 0.4728, + "step": 847 + }, + { + "epoch": 2.651356993736952, + "grad_norm": 0.7377315759658813, + "learning_rate": 3.179746998088351e-06, + "loss": 0.4672, + "step": 848 + }, + { + "epoch": 2.654488517745303, + "grad_norm": 0.8017002940177917, + "learning_rate": 3.1755794484955817e-06, + "loss": 0.4884, + "step": 849 + }, + { + "epoch": 2.6576200417536535, + "grad_norm": 1.045470952987671, + "learning_rate": 3.171409872613278e-06, + "loss": 0.4789, + "step": 850 + }, + { + "epoch": 2.6607515657620042, + "grad_norm": 0.8823987245559692, + "learning_rate": 3.1672382829473997e-06, + "loss": 0.5117, + "step": 851 + }, + { + "epoch": 2.663883089770355, + "grad_norm": 0.7395204901695251, + "learning_rate": 3.163064692009944e-06, + "loss": 0.5476, + "step": 852 + }, + { + "epoch": 2.6670146137787056, + "grad_norm": 0.7778941988945007, + "learning_rate": 3.1588891123189103e-06, + "loss": 0.5092, + "step": 853 + }, + { + "epoch": 2.6701461377870563, + "grad_norm": 0.8072531819343567, + "learning_rate": 3.1547115563982643e-06, + "loss": 0.4961, + "step": 854 + }, + { + "epoch": 2.673277661795407, + "grad_norm": 0.9018139243125916, + "learning_rate": 3.1505320367778993e-06, + "loss": 0.4624, + "step": 855 + }, + { + "epoch": 2.6764091858037578, + "grad_norm": 0.8554450869560242, + "learning_rate": 3.1463505659935957e-06, + "loss": 0.4971, + "step": 856 + }, + { + "epoch": 2.6795407098121085, + "grad_norm": 0.7727259397506714, + "learning_rate": 3.14216715658699e-06, + "loss": 0.4544, + "step": 857 + }, + { + "epoch": 2.682672233820459, + "grad_norm": 0.9253409504890442, + "learning_rate": 3.137981821105529e-06, + "loss": 0.4893, + "step": 858 + }, + { + "epoch": 2.68580375782881, + "grad_norm": 0.8809456825256348, + "learning_rate": 3.1337945721024403e-06, + "loss": 0.5242, + "step": 859 + }, + { + "epoch": 2.6889352818371606, + "grad_norm": 0.981755256652832, + "learning_rate": 3.129605422136689e-06, + "loss": 0.4686, + "step": 860 + }, + { + "epoch": 2.6920668058455117, + "grad_norm": 1.1278467178344727, + "learning_rate": 3.1254143837729412e-06, + "loss": 0.4813, + "step": 861 + }, + { + "epoch": 2.695198329853862, + "grad_norm": 0.8529123663902283, + "learning_rate": 3.1212214695815285e-06, + "loss": 0.4723, + "step": 862 + }, + { + "epoch": 2.698329853862213, + "grad_norm": 0.7764189839363098, + "learning_rate": 3.1170266921384075e-06, + "loss": 0.4777, + "step": 863 + }, + { + "epoch": 2.701461377870564, + "grad_norm": 0.7364740967750549, + "learning_rate": 3.112830064025124e-06, + "loss": 0.4975, + "step": 864 + }, + { + "epoch": 2.7045929018789145, + "grad_norm": 0.7594549059867859, + "learning_rate": 3.108631597828774e-06, + "loss": 0.5083, + "step": 865 + }, + { + "epoch": 2.7077244258872653, + "grad_norm": 0.7337073683738708, + "learning_rate": 3.104431306141968e-06, + "loss": 0.4778, + "step": 866 + }, + { + "epoch": 2.710855949895616, + "grad_norm": 0.7709932327270508, + "learning_rate": 3.1002292015627894e-06, + "loss": 0.4754, + "step": 867 + }, + { + "epoch": 2.7139874739039667, + "grad_norm": 0.8001313209533691, + "learning_rate": 3.0960252966947605e-06, + "loss": 0.4489, + "step": 868 + }, + { + "epoch": 2.7171189979123174, + "grad_norm": 0.8280592560768127, + "learning_rate": 3.091819604146804e-06, + "loss": 0.4606, + "step": 869 + }, + { + "epoch": 2.720250521920668, + "grad_norm": 0.7463534474372864, + "learning_rate": 3.0876121365332024e-06, + "loss": 0.5168, + "step": 870 + }, + { + "epoch": 2.723382045929019, + "grad_norm": 0.9011222124099731, + "learning_rate": 3.0834029064735636e-06, + "loss": 0.5163, + "step": 871 + }, + { + "epoch": 2.7265135699373695, + "grad_norm": 0.7811456322669983, + "learning_rate": 3.0791919265927827e-06, + "loss": 0.5004, + "step": 872 + }, + { + "epoch": 2.72964509394572, + "grad_norm": 0.9251837134361267, + "learning_rate": 3.0749792095210003e-06, + "loss": 0.5081, + "step": 873 + }, + { + "epoch": 2.732776617954071, + "grad_norm": 0.8347085118293762, + "learning_rate": 3.0707647678935695e-06, + "loss": 0.4793, + "step": 874 + }, + { + "epoch": 2.7359081419624216, + "grad_norm": 0.9766442179679871, + "learning_rate": 3.0665486143510153e-06, + "loss": 0.493, + "step": 875 + }, + { + "epoch": 2.7390396659707723, + "grad_norm": 0.7692548036575317, + "learning_rate": 3.0623307615389975e-06, + "loss": 0.4874, + "step": 876 + }, + { + "epoch": 2.742171189979123, + "grad_norm": 0.7714599370956421, + "learning_rate": 3.0581112221082727e-06, + "loss": 0.4929, + "step": 877 + }, + { + "epoch": 2.745302713987474, + "grad_norm": 0.7797786593437195, + "learning_rate": 3.053890008714655e-06, + "loss": 0.4359, + "step": 878 + }, + { + "epoch": 2.7484342379958244, + "grad_norm": 5.118397235870361, + "learning_rate": 3.049667134018981e-06, + "loss": 0.4634, + "step": 879 + }, + { + "epoch": 2.7515657620041756, + "grad_norm": 0.7684539556503296, + "learning_rate": 3.04544261068707e-06, + "loss": 0.4688, + "step": 880 + }, + { + "epoch": 2.754697286012526, + "grad_norm": 0.8678610920906067, + "learning_rate": 3.0412164513896846e-06, + "loss": 0.5213, + "step": 881 + }, + { + "epoch": 2.757828810020877, + "grad_norm": 0.80293869972229, + "learning_rate": 3.0369886688024954e-06, + "loss": 0.4392, + "step": 882 + }, + { + "epoch": 2.7609603340292277, + "grad_norm": 0.7438644766807556, + "learning_rate": 3.0327592756060412e-06, + "loss": 0.528, + "step": 883 + }, + { + "epoch": 2.7640918580375784, + "grad_norm": 0.7701645493507385, + "learning_rate": 3.0285282844856917e-06, + "loss": 0.504, + "step": 884 + }, + { + "epoch": 2.767223382045929, + "grad_norm": 0.7113856673240662, + "learning_rate": 3.024295708131611e-06, + "loss": 0.4819, + "step": 885 + }, + { + "epoch": 2.77035490605428, + "grad_norm": 1.2697532176971436, + "learning_rate": 3.020061559238714e-06, + "loss": 0.5009, + "step": 886 + }, + { + "epoch": 2.7734864300626305, + "grad_norm": 1.0299439430236816, + "learning_rate": 3.015825850506636e-06, + "loss": 0.4707, + "step": 887 + }, + { + "epoch": 2.776617954070981, + "grad_norm": 0.9703660607337952, + "learning_rate": 3.011588594639688e-06, + "loss": 0.4102, + "step": 888 + }, + { + "epoch": 2.779749478079332, + "grad_norm": 0.7357314825057983, + "learning_rate": 3.0073498043468247e-06, + "loss": 0.4649, + "step": 889 + }, + { + "epoch": 2.7828810020876826, + "grad_norm": 0.7815471291542053, + "learning_rate": 3.0031094923415993e-06, + "loss": 0.469, + "step": 890 + }, + { + "epoch": 2.7860125260960333, + "grad_norm": 0.7856019139289856, + "learning_rate": 2.9988676713421318e-06, + "loss": 0.4241, + "step": 891 + }, + { + "epoch": 2.789144050104384, + "grad_norm": 0.7668167352676392, + "learning_rate": 2.994624354071066e-06, + "loss": 0.5309, + "step": 892 + }, + { + "epoch": 2.7922755741127347, + "grad_norm": 0.7485945820808411, + "learning_rate": 2.990379553255535e-06, + "loss": 0.5173, + "step": 893 + }, + { + "epoch": 2.7954070981210855, + "grad_norm": 0.8065824508666992, + "learning_rate": 2.986133281627123e-06, + "loss": 0.4995, + "step": 894 + }, + { + "epoch": 2.798538622129436, + "grad_norm": 0.7156995534896851, + "learning_rate": 2.9818855519218217e-06, + "loss": 0.4642, + "step": 895 + }, + { + "epoch": 2.801670146137787, + "grad_norm": 0.9115403890609741, + "learning_rate": 2.97763637688e-06, + "loss": 0.4799, + "step": 896 + }, + { + "epoch": 2.804801670146138, + "grad_norm": 0.7466689944267273, + "learning_rate": 2.9733857692463584e-06, + "loss": 0.4942, + "step": 897 + }, + { + "epoch": 2.8079331941544883, + "grad_norm": 0.7484914064407349, + "learning_rate": 2.9691337417698974e-06, + "loss": 0.4618, + "step": 898 + }, + { + "epoch": 2.8110647181628394, + "grad_norm": 0.816704511642456, + "learning_rate": 2.9648803072038736e-06, + "loss": 0.4748, + "step": 899 + }, + { + "epoch": 2.81419624217119, + "grad_norm": 0.7627584934234619, + "learning_rate": 2.9606254783057666e-06, + "loss": 0.4667, + "step": 900 + }, + { + "epoch": 2.817327766179541, + "grad_norm": 0.7341011166572571, + "learning_rate": 2.9563692678372342e-06, + "loss": 0.4802, + "step": 901 + }, + { + "epoch": 2.8204592901878915, + "grad_norm": 1.2541382312774658, + "learning_rate": 2.952111688564082e-06, + "loss": 0.5231, + "step": 902 + }, + { + "epoch": 2.8235908141962422, + "grad_norm": 0.7172819375991821, + "learning_rate": 2.9478527532562184e-06, + "loss": 0.4488, + "step": 903 + }, + { + "epoch": 2.826722338204593, + "grad_norm": 0.774529218673706, + "learning_rate": 2.943592474687621e-06, + "loss": 0.4964, + "step": 904 + }, + { + "epoch": 2.8298538622129437, + "grad_norm": 0.7315672636032104, + "learning_rate": 2.939330865636294e-06, + "loss": 0.4817, + "step": 905 + }, + { + "epoch": 2.8329853862212944, + "grad_norm": 0.7698234915733337, + "learning_rate": 2.9350679388842347e-06, + "loss": 0.5075, + "step": 906 + }, + { + "epoch": 2.836116910229645, + "grad_norm": 0.7717766761779785, + "learning_rate": 2.93080370721739e-06, + "loss": 0.4789, + "step": 907 + }, + { + "epoch": 2.8392484342379958, + "grad_norm": 0.7383570075035095, + "learning_rate": 2.926538183425622e-06, + "loss": 0.4992, + "step": 908 + }, + { + "epoch": 2.8423799582463465, + "grad_norm": 0.7858864068984985, + "learning_rate": 2.92227138030267e-06, + "loss": 0.4993, + "step": 909 + }, + { + "epoch": 2.845511482254697, + "grad_norm": 0.8220369219779968, + "learning_rate": 2.9180033106461076e-06, + "loss": 0.4929, + "step": 910 + }, + { + "epoch": 2.848643006263048, + "grad_norm": 0.7507152557373047, + "learning_rate": 2.9137339872573086e-06, + "loss": 0.4394, + "step": 911 + }, + { + "epoch": 2.8517745302713986, + "grad_norm": 0.7935269474983215, + "learning_rate": 2.9094634229414063e-06, + "loss": 0.4656, + "step": 912 + }, + { + "epoch": 2.8549060542797493, + "grad_norm": 0.9187721610069275, + "learning_rate": 2.9051916305072576e-06, + "loss": 0.4918, + "step": 913 + }, + { + "epoch": 2.8580375782881005, + "grad_norm": 0.8699706792831421, + "learning_rate": 2.9009186227674e-06, + "loss": 0.5106, + "step": 914 + }, + { + "epoch": 2.8611691022964507, + "grad_norm": 0.7175673246383667, + "learning_rate": 2.896644412538021e-06, + "loss": 0.5105, + "step": 915 + }, + { + "epoch": 2.864300626304802, + "grad_norm": 0.8563990592956543, + "learning_rate": 2.892369012638909e-06, + "loss": 0.4993, + "step": 916 + }, + { + "epoch": 2.867432150313152, + "grad_norm": 0.7891882658004761, + "learning_rate": 2.8880924358934246e-06, + "loss": 0.4983, + "step": 917 + }, + { + "epoch": 2.8705636743215033, + "grad_norm": 0.9247110486030579, + "learning_rate": 2.8838146951284575e-06, + "loss": 0.4789, + "step": 918 + }, + { + "epoch": 2.873695198329854, + "grad_norm": 0.7523055672645569, + "learning_rate": 2.879535803174387e-06, + "loss": 0.4982, + "step": 919 + }, + { + "epoch": 2.8768267223382047, + "grad_norm": 0.8096909523010254, + "learning_rate": 2.8752557728650467e-06, + "loss": 0.4958, + "step": 920 + }, + { + "epoch": 2.8799582463465554, + "grad_norm": 2.3476874828338623, + "learning_rate": 2.870974617037684e-06, + "loss": 0.491, + "step": 921 + }, + { + "epoch": 2.883089770354906, + "grad_norm": 0.8388578295707703, + "learning_rate": 2.8666923485329224e-06, + "loss": 0.5275, + "step": 922 + }, + { + "epoch": 2.886221294363257, + "grad_norm": 0.8162729144096375, + "learning_rate": 2.8624089801947234e-06, + "loss": 0.4776, + "step": 923 + }, + { + "epoch": 2.8893528183716075, + "grad_norm": 0.7306103110313416, + "learning_rate": 2.858124524870345e-06, + "loss": 0.4814, + "step": 924 + }, + { + "epoch": 2.892484342379958, + "grad_norm": 0.8736817836761475, + "learning_rate": 2.853838995410307e-06, + "loss": 0.5097, + "step": 925 + }, + { + "epoch": 2.895615866388309, + "grad_norm": 0.7771823406219482, + "learning_rate": 2.8495524046683525e-06, + "loss": 0.4806, + "step": 926 + }, + { + "epoch": 2.8987473903966596, + "grad_norm": 0.9421334862709045, + "learning_rate": 2.845264765501404e-06, + "loss": 0.5055, + "step": 927 + }, + { + "epoch": 2.9018789144050103, + "grad_norm": 0.8403921127319336, + "learning_rate": 2.8409760907695314e-06, + "loss": 0.4775, + "step": 928 + }, + { + "epoch": 2.905010438413361, + "grad_norm": 0.8095362186431885, + "learning_rate": 2.836686393335909e-06, + "loss": 0.4532, + "step": 929 + }, + { + "epoch": 2.9081419624217117, + "grad_norm": 0.7340645790100098, + "learning_rate": 2.8323956860667813e-06, + "loss": 0.4835, + "step": 930 + }, + { + "epoch": 2.911273486430063, + "grad_norm": 0.6970911026000977, + "learning_rate": 2.828103981831417e-06, + "loss": 0.4999, + "step": 931 + }, + { + "epoch": 2.914405010438413, + "grad_norm": 0.8136418461799622, + "learning_rate": 2.8238112935020794e-06, + "loss": 0.5038, + "step": 932 + }, + { + "epoch": 2.9175365344467643, + "grad_norm": 0.9045608043670654, + "learning_rate": 2.8195176339539816e-06, + "loss": 0.486, + "step": 933 + }, + { + "epoch": 2.9206680584551146, + "grad_norm": 1.14940345287323, + "learning_rate": 2.815223016065249e-06, + "loss": 0.5079, + "step": 934 + }, + { + "epoch": 2.9237995824634657, + "grad_norm": 0.7411190867424011, + "learning_rate": 2.8109274527168826e-06, + "loss": 0.4564, + "step": 935 + }, + { + "epoch": 2.9269311064718164, + "grad_norm": 0.8903455138206482, + "learning_rate": 2.806630956792719e-06, + "loss": 0.451, + "step": 936 + }, + { + "epoch": 2.930062630480167, + "grad_norm": 0.7865445017814636, + "learning_rate": 2.8023335411793904e-06, + "loss": 0.4658, + "step": 937 + }, + { + "epoch": 2.933194154488518, + "grad_norm": 0.8185790777206421, + "learning_rate": 2.798035218766292e-06, + "loss": 0.4776, + "step": 938 + }, + { + "epoch": 2.9363256784968685, + "grad_norm": 0.7516276836395264, + "learning_rate": 2.793736002445531e-06, + "loss": 0.4447, + "step": 939 + }, + { + "epoch": 2.9394572025052192, + "grad_norm": 0.738080620765686, + "learning_rate": 2.789435905111903e-06, + "loss": 0.4832, + "step": 940 + }, + { + "epoch": 2.94258872651357, + "grad_norm": 0.7971507906913757, + "learning_rate": 2.785134939662843e-06, + "loss": 0.4835, + "step": 941 + }, + { + "epoch": 2.9457202505219207, + "grad_norm": 0.7529093623161316, + "learning_rate": 2.78083311899839e-06, + "loss": 0.4759, + "step": 942 + }, + { + "epoch": 2.9488517745302714, + "grad_norm": 0.8222358226776123, + "learning_rate": 2.7765304560211482e-06, + "loss": 0.4365, + "step": 943 + }, + { + "epoch": 2.951983298538622, + "grad_norm": 0.729945182800293, + "learning_rate": 2.7722269636362462e-06, + "loss": 0.5026, + "step": 944 + }, + { + "epoch": 2.9551148225469728, + "grad_norm": 0.7287900447845459, + "learning_rate": 2.767922654751306e-06, + "loss": 0.4916, + "step": 945 + }, + { + "epoch": 2.9582463465553235, + "grad_norm": 0.869637131690979, + "learning_rate": 2.763617542276391e-06, + "loss": 0.5018, + "step": 946 + }, + { + "epoch": 2.961377870563674, + "grad_norm": 1.004909634590149, + "learning_rate": 2.7593116391239806e-06, + "loss": 0.5152, + "step": 947 + }, + { + "epoch": 2.964509394572025, + "grad_norm": 0.8263046145439148, + "learning_rate": 2.7550049582089235e-06, + "loss": 0.5249, + "step": 948 + }, + { + "epoch": 2.9676409185803756, + "grad_norm": 0.7963895797729492, + "learning_rate": 2.750697512448401e-06, + "loss": 0.5084, + "step": 949 + }, + { + "epoch": 2.9707724425887267, + "grad_norm": 0.7211249470710754, + "learning_rate": 2.7463893147618893e-06, + "loss": 0.4691, + "step": 950 + }, + { + "epoch": 2.973903966597077, + "grad_norm": 0.8010216951370239, + "learning_rate": 2.742080378071118e-06, + "loss": 0.5026, + "step": 951 + }, + { + "epoch": 2.977035490605428, + "grad_norm": 0.780078649520874, + "learning_rate": 2.7377707153000356e-06, + "loss": 0.4758, + "step": 952 + }, + { + "epoch": 2.980167014613779, + "grad_norm": 0.7728193998336792, + "learning_rate": 2.7334603393747684e-06, + "loss": 0.488, + "step": 953 + }, + { + "epoch": 2.9832985386221296, + "grad_norm": 0.836329996585846, + "learning_rate": 2.7291492632235777e-06, + "loss": 0.456, + "step": 954 + }, + { + "epoch": 2.9864300626304803, + "grad_norm": 0.7241990566253662, + "learning_rate": 2.724837499776831e-06, + "loss": 0.4953, + "step": 955 + }, + { + "epoch": 2.989561586638831, + "grad_norm": 0.7595076560974121, + "learning_rate": 2.7205250619669527e-06, + "loss": 0.446, + "step": 956 + }, + { + "epoch": 2.9926931106471817, + "grad_norm": 0.8177686333656311, + "learning_rate": 2.716211962728392e-06, + "loss": 0.5057, + "step": 957 + }, + { + "epoch": 2.9958246346555324, + "grad_norm": 0.7506977915763855, + "learning_rate": 2.71189821499758e-06, + "loss": 0.4821, + "step": 958 + }, + { + "epoch": 2.998956158663883, + "grad_norm": 0.8085163235664368, + "learning_rate": 2.7075838317128943e-06, + "loss": 0.5002, + "step": 959 + }, + { + "epoch": 3.0, + "grad_norm": 3.3674418926239014, + "learning_rate": 2.7032688258146207e-06, + "loss": 0.183, + "step": 960 + }, + { + "epoch": 3.0031315240083507, + "grad_norm": 1.066100835800171, + "learning_rate": 2.698953210244908e-06, + "loss": 0.4427, + "step": 961 + }, + { + "epoch": 3.0062630480167014, + "grad_norm": 0.7690210938453674, + "learning_rate": 2.6946369979477365e-06, + "loss": 0.4638, + "step": 962 + }, + { + "epoch": 3.009394572025052, + "grad_norm": 0.8114679455757141, + "learning_rate": 2.690320201868876e-06, + "loss": 0.4373, + "step": 963 + }, + { + "epoch": 3.012526096033403, + "grad_norm": 0.7680971622467041, + "learning_rate": 2.686002834955847e-06, + "loss": 0.4477, + "step": 964 + }, + { + "epoch": 3.0156576200417535, + "grad_norm": 0.7194678783416748, + "learning_rate": 2.6816849101578808e-06, + "loss": 0.455, + "step": 965 + }, + { + "epoch": 3.0187891440501042, + "grad_norm": 0.6890467405319214, + "learning_rate": 2.6773664404258854e-06, + "loss": 0.4246, + "step": 966 + }, + { + "epoch": 3.021920668058455, + "grad_norm": 0.8064301013946533, + "learning_rate": 2.6730474387123987e-06, + "loss": 0.497, + "step": 967 + }, + { + "epoch": 3.0250521920668056, + "grad_norm": 0.8164849281311035, + "learning_rate": 2.668727917971559e-06, + "loss": 0.457, + "step": 968 + }, + { + "epoch": 3.028183716075157, + "grad_norm": 0.7793440818786621, + "learning_rate": 2.6644078911590565e-06, + "loss": 0.4632, + "step": 969 + }, + { + "epoch": 3.0313152400835075, + "grad_norm": 0.7208535671234131, + "learning_rate": 2.6600873712321033e-06, + "loss": 0.4786, + "step": 970 + }, + { + "epoch": 3.034446764091858, + "grad_norm": 0.8905500769615173, + "learning_rate": 2.655766371149389e-06, + "loss": 0.4317, + "step": 971 + }, + { + "epoch": 3.037578288100209, + "grad_norm": 0.7537338733673096, + "learning_rate": 2.6514449038710418e-06, + "loss": 0.4783, + "step": 972 + }, + { + "epoch": 3.0407098121085596, + "grad_norm": 0.7901656031608582, + "learning_rate": 2.6471229823585937e-06, + "loss": 0.4762, + "step": 973 + }, + { + "epoch": 3.0438413361169103, + "grad_norm": 0.7427340745925903, + "learning_rate": 2.6428006195749373e-06, + "loss": 0.4782, + "step": 974 + }, + { + "epoch": 3.046972860125261, + "grad_norm": 0.8364680409431458, + "learning_rate": 2.6384778284842905e-06, + "loss": 0.4551, + "step": 975 + }, + { + "epoch": 3.0501043841336117, + "grad_norm": 1.1049500703811646, + "learning_rate": 2.634154622052155e-06, + "loss": 0.4451, + "step": 976 + }, + { + "epoch": 3.0532359081419624, + "grad_norm": 1.0797414779663086, + "learning_rate": 2.6298310132452757e-06, + "loss": 0.4512, + "step": 977 + }, + { + "epoch": 3.056367432150313, + "grad_norm": 0.8832088708877563, + "learning_rate": 2.62550701503161e-06, + "loss": 0.4569, + "step": 978 + }, + { + "epoch": 3.059498956158664, + "grad_norm": 0.748951256275177, + "learning_rate": 2.621182640380277e-06, + "loss": 0.4535, + "step": 979 + }, + { + "epoch": 3.0626304801670146, + "grad_norm": 0.776386022567749, + "learning_rate": 2.616857902261529e-06, + "loss": 0.4582, + "step": 980 + }, + { + "epoch": 3.0657620041753653, + "grad_norm": 0.730057418346405, + "learning_rate": 2.6125328136467074e-06, + "loss": 0.4581, + "step": 981 + }, + { + "epoch": 3.068893528183716, + "grad_norm": 0.7918877005577087, + "learning_rate": 2.6082073875082046e-06, + "loss": 0.4355, + "step": 982 + }, + { + "epoch": 3.0720250521920667, + "grad_norm": 0.803987443447113, + "learning_rate": 2.6038816368194265e-06, + "loss": 0.455, + "step": 983 + }, + { + "epoch": 3.0751565762004174, + "grad_norm": 1.2970365285873413, + "learning_rate": 2.599555574554749e-06, + "loss": 0.448, + "step": 984 + }, + { + "epoch": 3.078288100208768, + "grad_norm": 0.7814194560050964, + "learning_rate": 2.595229213689487e-06, + "loss": 0.4197, + "step": 985 + }, + { + "epoch": 3.081419624217119, + "grad_norm": 0.9865803122520447, + "learning_rate": 2.5909025671998483e-06, + "loss": 0.4649, + "step": 986 + }, + { + "epoch": 3.08455114822547, + "grad_norm": 0.7769168615341187, + "learning_rate": 2.586575648062898e-06, + "loss": 0.4287, + "step": 987 + }, + { + "epoch": 3.0876826722338206, + "grad_norm": 1.454988956451416, + "learning_rate": 2.582248469256519e-06, + "loss": 0.4548, + "step": 988 + }, + { + "epoch": 3.0908141962421714, + "grad_norm": 0.7656151056289673, + "learning_rate": 2.577921043759372e-06, + "loss": 0.4323, + "step": 989 + }, + { + "epoch": 3.093945720250522, + "grad_norm": 0.7989768385887146, + "learning_rate": 2.5735933845508598e-06, + "loss": 0.4616, + "step": 990 + }, + { + "epoch": 3.0970772442588728, + "grad_norm": 0.8711255192756653, + "learning_rate": 2.5692655046110855e-06, + "loss": 0.4923, + "step": 991 + }, + { + "epoch": 3.1002087682672235, + "grad_norm": 0.7333446145057678, + "learning_rate": 2.564937416920813e-06, + "loss": 0.4297, + "step": 992 + }, + { + "epoch": 3.103340292275574, + "grad_norm": 0.7611051201820374, + "learning_rate": 2.5606091344614297e-06, + "loss": 0.4276, + "step": 993 + }, + { + "epoch": 3.106471816283925, + "grad_norm": 0.9742305278778076, + "learning_rate": 2.5562806702149083e-06, + "loss": 0.4597, + "step": 994 + }, + { + "epoch": 3.1096033402922756, + "grad_norm": 0.8043314218521118, + "learning_rate": 2.551952037163765e-06, + "loss": 0.459, + "step": 995 + }, + { + "epoch": 3.1127348643006263, + "grad_norm": 0.7217118740081787, + "learning_rate": 2.5476232482910253e-06, + "loss": 0.4514, + "step": 996 + }, + { + "epoch": 3.115866388308977, + "grad_norm": 1.1410573720932007, + "learning_rate": 2.5432943165801765e-06, + "loss": 0.4942, + "step": 997 + }, + { + "epoch": 3.1189979123173277, + "grad_norm": 0.7188895344734192, + "learning_rate": 2.5389652550151416e-06, + "loss": 0.4641, + "step": 998 + }, + { + "epoch": 3.1221294363256784, + "grad_norm": 0.7223486304283142, + "learning_rate": 2.5346360765802276e-06, + "loss": 0.4382, + "step": 999 + }, + { + "epoch": 3.125260960334029, + "grad_norm": 0.8269757032394409, + "learning_rate": 2.5303067942600933e-06, + "loss": 0.4502, + "step": 1000 + }, + { + "epoch": 3.12839248434238, + "grad_norm": 0.7081235647201538, + "learning_rate": 2.5259774210397104e-06, + "loss": 0.4389, + "step": 1001 + }, + { + "epoch": 3.1315240083507305, + "grad_norm": 0.7493525147438049, + "learning_rate": 2.5216479699043224e-06, + "loss": 0.4697, + "step": 1002 + }, + { + "epoch": 3.1346555323590812, + "grad_norm": 0.7633835077285767, + "learning_rate": 2.5173184538394054e-06, + "loss": 0.4535, + "step": 1003 + }, + { + "epoch": 3.137787056367432, + "grad_norm": 0.7765631079673767, + "learning_rate": 2.5129888858306333e-06, + "loss": 0.4616, + "step": 1004 + }, + { + "epoch": 3.140918580375783, + "grad_norm": 0.7026706337928772, + "learning_rate": 2.508659278863832e-06, + "loss": 0.4535, + "step": 1005 + }, + { + "epoch": 3.144050104384134, + "grad_norm": 0.8816738128662109, + "learning_rate": 2.5043296459249466e-06, + "loss": 0.4274, + "step": 1006 + }, + { + "epoch": 3.1471816283924845, + "grad_norm": 0.8072274327278137, + "learning_rate": 2.5e-06, + "loss": 0.4566, + "step": 1007 + }, + { + "epoch": 3.150313152400835, + "grad_norm": 0.7775781750679016, + "learning_rate": 2.4956703540750542e-06, + "loss": 0.4248, + "step": 1008 + }, + { + "epoch": 3.153444676409186, + "grad_norm": 0.9271306991577148, + "learning_rate": 2.4913407211361686e-06, + "loss": 0.4251, + "step": 1009 + }, + { + "epoch": 3.1565762004175366, + "grad_norm": 0.763867974281311, + "learning_rate": 2.487011114169368e-06, + "loss": 0.4525, + "step": 1010 + }, + { + "epoch": 3.1597077244258873, + "grad_norm": 0.8754820227622986, + "learning_rate": 2.4826815461605955e-06, + "loss": 0.4524, + "step": 1011 + }, + { + "epoch": 3.162839248434238, + "grad_norm": 0.8261513113975525, + "learning_rate": 2.4783520300956784e-06, + "loss": 0.4303, + "step": 1012 + }, + { + "epoch": 3.1659707724425887, + "grad_norm": 0.8629854917526245, + "learning_rate": 2.4740225789602905e-06, + "loss": 0.4628, + "step": 1013 + }, + { + "epoch": 3.1691022964509394, + "grad_norm": 0.7281518578529358, + "learning_rate": 2.469693205739907e-06, + "loss": 0.461, + "step": 1014 + }, + { + "epoch": 3.17223382045929, + "grad_norm": 0.757644772529602, + "learning_rate": 2.465363923419774e-06, + "loss": 0.4455, + "step": 1015 + }, + { + "epoch": 3.175365344467641, + "grad_norm": 0.8021314740180969, + "learning_rate": 2.4610347449848592e-06, + "loss": 0.4755, + "step": 1016 + }, + { + "epoch": 3.1784968684759916, + "grad_norm": 0.7467564940452576, + "learning_rate": 2.456705683419824e-06, + "loss": 0.4419, + "step": 1017 + }, + { + "epoch": 3.1816283924843423, + "grad_norm": 0.8575125336647034, + "learning_rate": 2.452376751708976e-06, + "loss": 0.4577, + "step": 1018 + }, + { + "epoch": 3.184759916492693, + "grad_norm": 0.8101590871810913, + "learning_rate": 2.448047962836235e-06, + "loss": 0.4546, + "step": 1019 + }, + { + "epoch": 3.1878914405010437, + "grad_norm": 0.8146190643310547, + "learning_rate": 2.443719329785093e-06, + "loss": 0.4219, + "step": 1020 + }, + { + "epoch": 3.1910229645093944, + "grad_norm": 1.2800556421279907, + "learning_rate": 2.4393908655385708e-06, + "loss": 0.4502, + "step": 1021 + }, + { + "epoch": 3.1941544885177455, + "grad_norm": 0.7362221479415894, + "learning_rate": 2.4350625830791875e-06, + "loss": 0.4802, + "step": 1022 + }, + { + "epoch": 3.1972860125260962, + "grad_norm": 0.8307201862335205, + "learning_rate": 2.430734495388915e-06, + "loss": 0.4808, + "step": 1023 + }, + { + "epoch": 3.200417536534447, + "grad_norm": 0.8467394709587097, + "learning_rate": 2.42640661544914e-06, + "loss": 0.4598, + "step": 1024 + }, + { + "epoch": 3.2035490605427976, + "grad_norm": 0.9570673108100891, + "learning_rate": 2.422078956240629e-06, + "loss": 0.4341, + "step": 1025 + }, + { + "epoch": 3.2066805845511483, + "grad_norm": 0.7976422309875488, + "learning_rate": 2.4177515307434824e-06, + "loss": 0.4285, + "step": 1026 + }, + { + "epoch": 3.209812108559499, + "grad_norm": 0.8241607546806335, + "learning_rate": 2.413424351937103e-06, + "loss": 0.4563, + "step": 1027 + }, + { + "epoch": 3.2129436325678498, + "grad_norm": 0.7191072702407837, + "learning_rate": 2.4090974328001526e-06, + "loss": 0.456, + "step": 1028 + }, + { + "epoch": 3.2160751565762005, + "grad_norm": 0.8112174868583679, + "learning_rate": 2.4047707863105133e-06, + "loss": 0.4592, + "step": 1029 + }, + { + "epoch": 3.219206680584551, + "grad_norm": 0.7170486450195312, + "learning_rate": 2.4004444254452522e-06, + "loss": 0.4559, + "step": 1030 + }, + { + "epoch": 3.222338204592902, + "grad_norm": 0.7092299461364746, + "learning_rate": 2.3961183631805748e-06, + "loss": 0.4504, + "step": 1031 + }, + { + "epoch": 3.2254697286012526, + "grad_norm": 0.8104662299156189, + "learning_rate": 2.391792612491796e-06, + "loss": 0.4327, + "step": 1032 + }, + { + "epoch": 3.2286012526096033, + "grad_norm": 0.8512858152389526, + "learning_rate": 2.387467186353293e-06, + "loss": 0.4506, + "step": 1033 + }, + { + "epoch": 3.231732776617954, + "grad_norm": 0.7987692952156067, + "learning_rate": 2.3831420977384715e-06, + "loss": 0.4257, + "step": 1034 + }, + { + "epoch": 3.2348643006263047, + "grad_norm": 0.7805537581443787, + "learning_rate": 2.3788173596197244e-06, + "loss": 0.4692, + "step": 1035 + }, + { + "epoch": 3.2379958246346554, + "grad_norm": 0.737304151058197, + "learning_rate": 2.374492984968392e-06, + "loss": 0.4308, + "step": 1036 + }, + { + "epoch": 3.241127348643006, + "grad_norm": 0.8113856315612793, + "learning_rate": 2.3701689867547247e-06, + "loss": 0.4668, + "step": 1037 + }, + { + "epoch": 3.244258872651357, + "grad_norm": 1.507103443145752, + "learning_rate": 2.3658453779478464e-06, + "loss": 0.4527, + "step": 1038 + }, + { + "epoch": 3.2473903966597075, + "grad_norm": 0.7973915338516235, + "learning_rate": 2.3615221715157095e-06, + "loss": 0.4741, + "step": 1039 + }, + { + "epoch": 3.2505219206680582, + "grad_norm": 0.8206940293312073, + "learning_rate": 2.3571993804250635e-06, + "loss": 0.4582, + "step": 1040 + }, + { + "epoch": 3.2536534446764094, + "grad_norm": 0.8519637584686279, + "learning_rate": 2.3528770176414076e-06, + "loss": 0.4412, + "step": 1041 + }, + { + "epoch": 3.25678496868476, + "grad_norm": 0.7542241811752319, + "learning_rate": 2.348555096128959e-06, + "loss": 0.4861, + "step": 1042 + }, + { + "epoch": 3.259916492693111, + "grad_norm": 0.8074842691421509, + "learning_rate": 2.3442336288506125e-06, + "loss": 0.4707, + "step": 1043 + }, + { + "epoch": 3.2630480167014615, + "grad_norm": 0.8164265751838684, + "learning_rate": 2.3399126287678975e-06, + "loss": 0.4417, + "step": 1044 + }, + { + "epoch": 3.266179540709812, + "grad_norm": 0.7689628005027771, + "learning_rate": 2.3355921088409435e-06, + "loss": 0.4815, + "step": 1045 + }, + { + "epoch": 3.269311064718163, + "grad_norm": 0.7709240913391113, + "learning_rate": 2.3312720820284423e-06, + "loss": 0.4444, + "step": 1046 + }, + { + "epoch": 3.2724425887265136, + "grad_norm": 0.7340330481529236, + "learning_rate": 2.326952561287602e-06, + "loss": 0.4694, + "step": 1047 + }, + { + "epoch": 3.2755741127348643, + "grad_norm": 0.7785805463790894, + "learning_rate": 2.3226335595741154e-06, + "loss": 0.4273, + "step": 1048 + }, + { + "epoch": 3.278705636743215, + "grad_norm": 0.9968108534812927, + "learning_rate": 2.3183150898421196e-06, + "loss": 0.4782, + "step": 1049 + }, + { + "epoch": 3.2818371607515657, + "grad_norm": 0.7823553681373596, + "learning_rate": 2.3139971650441533e-06, + "loss": 0.444, + "step": 1050 + }, + { + "epoch": 3.2849686847599164, + "grad_norm": 0.7317377924919128, + "learning_rate": 2.3096797981311252e-06, + "loss": 0.4692, + "step": 1051 + }, + { + "epoch": 3.288100208768267, + "grad_norm": 0.8546518683433533, + "learning_rate": 2.3053630020522643e-06, + "loss": 0.4632, + "step": 1052 + }, + { + "epoch": 3.291231732776618, + "grad_norm": 1.2284396886825562, + "learning_rate": 2.301046789755093e-06, + "loss": 0.4409, + "step": 1053 + }, + { + "epoch": 3.2943632567849686, + "grad_norm": 0.8000460863113403, + "learning_rate": 2.2967311741853797e-06, + "loss": 0.4456, + "step": 1054 + }, + { + "epoch": 3.2974947807933193, + "grad_norm": 0.7689793109893799, + "learning_rate": 2.2924161682871053e-06, + "loss": 0.45, + "step": 1055 + }, + { + "epoch": 3.30062630480167, + "grad_norm": 0.8032956719398499, + "learning_rate": 2.288101785002421e-06, + "loss": 0.4817, + "step": 1056 + }, + { + "epoch": 3.3037578288100207, + "grad_norm": 0.6831309795379639, + "learning_rate": 2.283788037271609e-06, + "loss": 0.4502, + "step": 1057 + }, + { + "epoch": 3.306889352818372, + "grad_norm": 0.8581221103668213, + "learning_rate": 2.279474938033048e-06, + "loss": 0.4569, + "step": 1058 + }, + { + "epoch": 3.3100208768267225, + "grad_norm": 0.7937221527099609, + "learning_rate": 2.2751625002231696e-06, + "loss": 0.4451, + "step": 1059 + }, + { + "epoch": 3.3131524008350732, + "grad_norm": 0.8095264434814453, + "learning_rate": 2.270850736776422e-06, + "loss": 0.4462, + "step": 1060 + }, + { + "epoch": 3.316283924843424, + "grad_norm": 0.9141370058059692, + "learning_rate": 2.2665396606252332e-06, + "loss": 0.419, + "step": 1061 + }, + { + "epoch": 3.3194154488517746, + "grad_norm": 0.8648553490638733, + "learning_rate": 2.262229284699965e-06, + "loss": 0.4562, + "step": 1062 + }, + { + "epoch": 3.3225469728601253, + "grad_norm": 0.7716917395591736, + "learning_rate": 2.2579196219288825e-06, + "loss": 0.4734, + "step": 1063 + }, + { + "epoch": 3.325678496868476, + "grad_norm": 0.8074535727500916, + "learning_rate": 2.2536106852381116e-06, + "loss": 0.4272, + "step": 1064 + }, + { + "epoch": 3.3288100208768268, + "grad_norm": 0.8989127278327942, + "learning_rate": 2.249302487551599e-06, + "loss": 0.4728, + "step": 1065 + }, + { + "epoch": 3.3319415448851775, + "grad_norm": 0.7662765383720398, + "learning_rate": 2.2449950417910777e-06, + "loss": 0.4457, + "step": 1066 + }, + { + "epoch": 3.335073068893528, + "grad_norm": 0.7789275050163269, + "learning_rate": 2.24068836087602e-06, + "loss": 0.3919, + "step": 1067 + }, + { + "epoch": 3.338204592901879, + "grad_norm": 0.7359098196029663, + "learning_rate": 2.2363824577236097e-06, + "loss": 0.4876, + "step": 1068 + }, + { + "epoch": 3.3413361169102296, + "grad_norm": 0.9960948824882507, + "learning_rate": 2.232077345248695e-06, + "loss": 0.4894, + "step": 1069 + }, + { + "epoch": 3.3444676409185803, + "grad_norm": 0.7446064352989197, + "learning_rate": 2.2277730363637537e-06, + "loss": 0.4765, + "step": 1070 + }, + { + "epoch": 3.347599164926931, + "grad_norm": 0.7674328088760376, + "learning_rate": 2.2234695439788534e-06, + "loss": 0.4468, + "step": 1071 + }, + { + "epoch": 3.3507306889352817, + "grad_norm": 0.8957347869873047, + "learning_rate": 2.2191668810016105e-06, + "loss": 0.4733, + "step": 1072 + }, + { + "epoch": 3.3538622129436324, + "grad_norm": 0.9110277891159058, + "learning_rate": 2.2148650603371573e-06, + "loss": 0.4399, + "step": 1073 + }, + { + "epoch": 3.356993736951983, + "grad_norm": 0.9158220291137695, + "learning_rate": 2.2105640948880976e-06, + "loss": 0.4609, + "step": 1074 + }, + { + "epoch": 3.3601252609603343, + "grad_norm": 0.7630184888839722, + "learning_rate": 2.206263997554469e-06, + "loss": 0.4674, + "step": 1075 + }, + { + "epoch": 3.3632567849686845, + "grad_norm": 0.7975273728370667, + "learning_rate": 2.20196478123371e-06, + "loss": 0.4478, + "step": 1076 + }, + { + "epoch": 3.3663883089770357, + "grad_norm": 0.8825351595878601, + "learning_rate": 2.19766645882061e-06, + "loss": 0.4687, + "step": 1077 + }, + { + "epoch": 3.3695198329853864, + "grad_norm": 0.8907671570777893, + "learning_rate": 2.1933690432072817e-06, + "loss": 0.4223, + "step": 1078 + }, + { + "epoch": 3.372651356993737, + "grad_norm": 0.7449545860290527, + "learning_rate": 2.189072547283118e-06, + "loss": 0.4595, + "step": 1079 + }, + { + "epoch": 3.375782881002088, + "grad_norm": 0.8460972309112549, + "learning_rate": 2.184776983934751e-06, + "loss": 0.4443, + "step": 1080 + }, + { + "epoch": 3.3789144050104385, + "grad_norm": 0.7524845600128174, + "learning_rate": 2.1804823660460196e-06, + "loss": 0.4235, + "step": 1081 + }, + { + "epoch": 3.382045929018789, + "grad_norm": 0.8448389768600464, + "learning_rate": 2.176188706497921e-06, + "loss": 0.4387, + "step": 1082 + }, + { + "epoch": 3.38517745302714, + "grad_norm": 0.7701981663703918, + "learning_rate": 2.1718960181685838e-06, + "loss": 0.4257, + "step": 1083 + }, + { + "epoch": 3.3883089770354906, + "grad_norm": 0.8178983330726624, + "learning_rate": 2.167604313933219e-06, + "loss": 0.4983, + "step": 1084 + }, + { + "epoch": 3.3914405010438413, + "grad_norm": 0.7477235198020935, + "learning_rate": 2.163313606664091e-06, + "loss": 0.4559, + "step": 1085 + }, + { + "epoch": 3.394572025052192, + "grad_norm": 0.8127962350845337, + "learning_rate": 2.1590239092304694e-06, + "loss": 0.453, + "step": 1086 + }, + { + "epoch": 3.3977035490605427, + "grad_norm": 0.7462339997291565, + "learning_rate": 2.1547352344985966e-06, + "loss": 0.4697, + "step": 1087 + }, + { + "epoch": 3.4008350730688934, + "grad_norm": 0.9641384482383728, + "learning_rate": 2.1504475953316483e-06, + "loss": 0.4495, + "step": 1088 + }, + { + "epoch": 3.403966597077244, + "grad_norm": 0.7612512707710266, + "learning_rate": 2.146161004589693e-06, + "loss": 0.4579, + "step": 1089 + }, + { + "epoch": 3.407098121085595, + "grad_norm": 0.7547829747200012, + "learning_rate": 2.141875475129655e-06, + "loss": 0.4334, + "step": 1090 + }, + { + "epoch": 3.4102296450939455, + "grad_norm": 0.8036953806877136, + "learning_rate": 2.137591019805278e-06, + "loss": 0.4466, + "step": 1091 + }, + { + "epoch": 3.4133611691022967, + "grad_norm": 0.7319284081459045, + "learning_rate": 2.1333076514670784e-06, + "loss": 0.4942, + "step": 1092 + }, + { + "epoch": 3.416492693110647, + "grad_norm": 0.8278589248657227, + "learning_rate": 2.1290253829623165e-06, + "loss": 0.4554, + "step": 1093 + }, + { + "epoch": 3.419624217118998, + "grad_norm": 0.733059287071228, + "learning_rate": 2.124744227134954e-06, + "loss": 0.4187, + "step": 1094 + }, + { + "epoch": 3.422755741127349, + "grad_norm": 0.8222727179527283, + "learning_rate": 2.1204641968256136e-06, + "loss": 0.4587, + "step": 1095 + }, + { + "epoch": 3.4258872651356995, + "grad_norm": 0.8296732902526855, + "learning_rate": 2.1161853048715438e-06, + "loss": 0.4868, + "step": 1096 + }, + { + "epoch": 3.4290187891440502, + "grad_norm": 0.7309690713882446, + "learning_rate": 2.1119075641065758e-06, + "loss": 0.4594, + "step": 1097 + }, + { + "epoch": 3.432150313152401, + "grad_norm": 1.4901788234710693, + "learning_rate": 2.1076309873610916e-06, + "loss": 0.4216, + "step": 1098 + }, + { + "epoch": 3.4352818371607516, + "grad_norm": 0.7993581891059875, + "learning_rate": 2.1033555874619794e-06, + "loss": 0.4842, + "step": 1099 + }, + { + "epoch": 3.4384133611691023, + "grad_norm": 0.8846752643585205, + "learning_rate": 2.0990813772325995e-06, + "loss": 0.4395, + "step": 1100 + }, + { + "epoch": 3.441544885177453, + "grad_norm": 1.0796778202056885, + "learning_rate": 2.0948083694927436e-06, + "loss": 0.4573, + "step": 1101 + }, + { + "epoch": 3.4446764091858038, + "grad_norm": 0.74623042345047, + "learning_rate": 2.090536577058595e-06, + "loss": 0.4563, + "step": 1102 + }, + { + "epoch": 3.4478079331941545, + "grad_norm": 0.8245521783828735, + "learning_rate": 2.086266012742692e-06, + "loss": 0.4433, + "step": 1103 + }, + { + "epoch": 3.450939457202505, + "grad_norm": 0.8003777265548706, + "learning_rate": 2.081996689353893e-06, + "loss": 0.4599, + "step": 1104 + }, + { + "epoch": 3.454070981210856, + "grad_norm": 0.8309001922607422, + "learning_rate": 2.0777286196973302e-06, + "loss": 0.485, + "step": 1105 + }, + { + "epoch": 3.4572025052192066, + "grad_norm": 0.8299122452735901, + "learning_rate": 2.0734618165743782e-06, + "loss": 0.4685, + "step": 1106 + }, + { + "epoch": 3.4603340292275573, + "grad_norm": 0.9347029328346252, + "learning_rate": 2.069196292782611e-06, + "loss": 0.4615, + "step": 1107 + }, + { + "epoch": 3.463465553235908, + "grad_norm": 0.7146593332290649, + "learning_rate": 2.064932061115766e-06, + "loss": 0.4433, + "step": 1108 + }, + { + "epoch": 3.4665970772442587, + "grad_norm": 0.7674420475959778, + "learning_rate": 2.0606691343637063e-06, + "loss": 0.4444, + "step": 1109 + }, + { + "epoch": 3.4697286012526094, + "grad_norm": 0.7925504446029663, + "learning_rate": 2.05640752531238e-06, + "loss": 0.4631, + "step": 1110 + }, + { + "epoch": 3.4728601252609606, + "grad_norm": 0.7755677700042725, + "learning_rate": 2.0521472467437825e-06, + "loss": 0.4709, + "step": 1111 + }, + { + "epoch": 3.4759916492693113, + "grad_norm": 0.8535795211791992, + "learning_rate": 2.0478883114359187e-06, + "loss": 0.442, + "step": 1112 + }, + { + "epoch": 3.479123173277662, + "grad_norm": 0.723953127861023, + "learning_rate": 2.043630732162767e-06, + "loss": 0.4782, + "step": 1113 + }, + { + "epoch": 3.4822546972860127, + "grad_norm": 0.7817316651344299, + "learning_rate": 2.0393745216942343e-06, + "loss": 0.4841, + "step": 1114 + }, + { + "epoch": 3.4853862212943634, + "grad_norm": 0.8878781795501709, + "learning_rate": 2.0351196927961268e-06, + "loss": 0.4673, + "step": 1115 + }, + { + "epoch": 3.488517745302714, + "grad_norm": 0.8645241856575012, + "learning_rate": 2.030866258230104e-06, + "loss": 0.432, + "step": 1116 + }, + { + "epoch": 3.491649269311065, + "grad_norm": 0.7294583320617676, + "learning_rate": 2.026614230753643e-06, + "loss": 0.4683, + "step": 1117 + }, + { + "epoch": 3.4947807933194155, + "grad_norm": 0.7412407994270325, + "learning_rate": 2.022363623120001e-06, + "loss": 0.4523, + "step": 1118 + }, + { + "epoch": 3.497912317327766, + "grad_norm": 0.8559291362762451, + "learning_rate": 2.0181144480781787e-06, + "loss": 0.4309, + "step": 1119 + }, + { + "epoch": 3.501043841336117, + "grad_norm": 0.7442825436592102, + "learning_rate": 2.0138667183728775e-06, + "loss": 0.4096, + "step": 1120 + }, + { + "epoch": 3.5041753653444676, + "grad_norm": 0.7605662941932678, + "learning_rate": 2.0096204467444645e-06, + "loss": 0.4404, + "step": 1121 + }, + { + "epoch": 3.5073068893528183, + "grad_norm": 0.7984277009963989, + "learning_rate": 2.005375645928935e-06, + "loss": 0.4661, + "step": 1122 + }, + { + "epoch": 3.510438413361169, + "grad_norm": 1.1044552326202393, + "learning_rate": 2.001132328657869e-06, + "loss": 0.4185, + "step": 1123 + }, + { + "epoch": 3.5135699373695197, + "grad_norm": 0.8210328817367554, + "learning_rate": 1.996890507658401e-06, + "loss": 0.4746, + "step": 1124 + }, + { + "epoch": 3.5167014613778704, + "grad_norm": 0.7302148342132568, + "learning_rate": 1.9926501956531758e-06, + "loss": 0.4333, + "step": 1125 + }, + { + "epoch": 3.519832985386221, + "grad_norm": 0.7713826894760132, + "learning_rate": 1.9884114053603114e-06, + "loss": 0.4485, + "step": 1126 + }, + { + "epoch": 3.522964509394572, + "grad_norm": 0.7386549711227417, + "learning_rate": 1.984174149493365e-06, + "loss": 0.4678, + "step": 1127 + }, + { + "epoch": 3.526096033402923, + "grad_norm": 0.8006004095077515, + "learning_rate": 1.979938440761287e-06, + "loss": 0.4755, + "step": 1128 + }, + { + "epoch": 3.5292275574112733, + "grad_norm": 1.0635333061218262, + "learning_rate": 1.97570429186839e-06, + "loss": 0.4985, + "step": 1129 + }, + { + "epoch": 3.5323590814196244, + "grad_norm": 0.7470075488090515, + "learning_rate": 1.9714717155143083e-06, + "loss": 0.4307, + "step": 1130 + }, + { + "epoch": 3.535490605427975, + "grad_norm": 0.8314558863639832, + "learning_rate": 1.967240724393959e-06, + "loss": 0.4894, + "step": 1131 + }, + { + "epoch": 3.538622129436326, + "grad_norm": 0.761443018913269, + "learning_rate": 1.963011331197506e-06, + "loss": 0.4653, + "step": 1132 + }, + { + "epoch": 3.5417536534446765, + "grad_norm": 0.7483212351799011, + "learning_rate": 1.9587835486103163e-06, + "loss": 0.4456, + "step": 1133 + }, + { + "epoch": 3.5448851774530272, + "grad_norm": 0.7794159054756165, + "learning_rate": 1.9545573893129306e-06, + "loss": 0.4707, + "step": 1134 + }, + { + "epoch": 3.548016701461378, + "grad_norm": 0.8016185760498047, + "learning_rate": 1.950332865981019e-06, + "loss": 0.4547, + "step": 1135 + }, + { + "epoch": 3.5511482254697286, + "grad_norm": 0.8089869618415833, + "learning_rate": 1.9461099912853453e-06, + "loss": 0.4499, + "step": 1136 + }, + { + "epoch": 3.5542797494780793, + "grad_norm": 0.7774782180786133, + "learning_rate": 1.9418887778917286e-06, + "loss": 0.4531, + "step": 1137 + }, + { + "epoch": 3.55741127348643, + "grad_norm": 0.7793645262718201, + "learning_rate": 1.937669238461003e-06, + "loss": 0.4553, + "step": 1138 + }, + { + "epoch": 3.5605427974947808, + "grad_norm": 0.8139959573745728, + "learning_rate": 1.933451385648985e-06, + "loss": 0.458, + "step": 1139 + }, + { + "epoch": 3.5636743215031315, + "grad_norm": 0.7517053484916687, + "learning_rate": 1.929235232106431e-06, + "loss": 0.4779, + "step": 1140 + }, + { + "epoch": 3.566805845511482, + "grad_norm": 0.8851562142372131, + "learning_rate": 1.925020790479e-06, + "loss": 0.425, + "step": 1141 + }, + { + "epoch": 3.569937369519833, + "grad_norm": 0.8129401803016663, + "learning_rate": 1.920808073407218e-06, + "loss": 0.4616, + "step": 1142 + }, + { + "epoch": 3.5730688935281836, + "grad_norm": 0.7110117077827454, + "learning_rate": 1.916597093526437e-06, + "loss": 0.4748, + "step": 1143 + }, + { + "epoch": 3.5762004175365343, + "grad_norm": 0.8268555402755737, + "learning_rate": 1.912387863466798e-06, + "loss": 0.4752, + "step": 1144 + }, + { + "epoch": 3.5793319415448854, + "grad_norm": 1.1036733388900757, + "learning_rate": 1.9081803958531967e-06, + "loss": 0.4879, + "step": 1145 + }, + { + "epoch": 3.5824634655532357, + "grad_norm": 0.8561109304428101, + "learning_rate": 1.9039747033052395e-06, + "loss": 0.4409, + "step": 1146 + }, + { + "epoch": 3.585594989561587, + "grad_norm": 0.7597541809082031, + "learning_rate": 1.8997707984372119e-06, + "loss": 0.4518, + "step": 1147 + }, + { + "epoch": 3.588726513569937, + "grad_norm": 0.7225353121757507, + "learning_rate": 1.8955686938580329e-06, + "loss": 0.4735, + "step": 1148 + }, + { + "epoch": 3.5918580375782883, + "grad_norm": 0.9409791231155396, + "learning_rate": 1.8913684021712264e-06, + "loss": 0.4249, + "step": 1149 + }, + { + "epoch": 3.594989561586639, + "grad_norm": 0.8757275342941284, + "learning_rate": 1.8871699359748763e-06, + "loss": 0.4479, + "step": 1150 + }, + { + "epoch": 3.5981210855949897, + "grad_norm": 0.8090003728866577, + "learning_rate": 1.882973307861593e-06, + "loss": 0.4846, + "step": 1151 + }, + { + "epoch": 3.6012526096033404, + "grad_norm": 0.7568825483322144, + "learning_rate": 1.8787785304184726e-06, + "loss": 0.4301, + "step": 1152 + }, + { + "epoch": 3.604384133611691, + "grad_norm": 0.8233815431594849, + "learning_rate": 1.8745856162270592e-06, + "loss": 0.4838, + "step": 1153 + }, + { + "epoch": 3.607515657620042, + "grad_norm": 0.7817628979682922, + "learning_rate": 1.8703945778633121e-06, + "loss": 0.4669, + "step": 1154 + }, + { + "epoch": 3.6106471816283925, + "grad_norm": 0.821304202079773, + "learning_rate": 1.8662054278975605e-06, + "loss": 0.4536, + "step": 1155 + }, + { + "epoch": 3.613778705636743, + "grad_norm": 0.9304089546203613, + "learning_rate": 1.8620181788944712e-06, + "loss": 0.4489, + "step": 1156 + }, + { + "epoch": 3.616910229645094, + "grad_norm": 0.8202670216560364, + "learning_rate": 1.8578328434130114e-06, + "loss": 0.4309, + "step": 1157 + }, + { + "epoch": 3.6200417536534446, + "grad_norm": 0.8890257477760315, + "learning_rate": 1.8536494340064051e-06, + "loss": 0.4736, + "step": 1158 + }, + { + "epoch": 3.6231732776617953, + "grad_norm": 0.7940590381622314, + "learning_rate": 1.8494679632221013e-06, + "loss": 0.4468, + "step": 1159 + }, + { + "epoch": 3.626304801670146, + "grad_norm": 0.8388273119926453, + "learning_rate": 1.845288443601736e-06, + "loss": 0.4753, + "step": 1160 + }, + { + "epoch": 3.6294363256784967, + "grad_norm": 0.9392285346984863, + "learning_rate": 1.84111088768109e-06, + "loss": 0.4736, + "step": 1161 + }, + { + "epoch": 3.632567849686848, + "grad_norm": 0.7411681413650513, + "learning_rate": 1.8369353079900576e-06, + "loss": 0.4303, + "step": 1162 + }, + { + "epoch": 3.635699373695198, + "grad_norm": 0.8722569942474365, + "learning_rate": 1.8327617170526014e-06, + "loss": 0.4604, + "step": 1163 + }, + { + "epoch": 3.6388308977035493, + "grad_norm": 0.786891758441925, + "learning_rate": 1.8285901273867229e-06, + "loss": 0.4756, + "step": 1164 + }, + { + "epoch": 3.6419624217118995, + "grad_norm": 0.8159083724021912, + "learning_rate": 1.824420551504419e-06, + "loss": 0.4675, + "step": 1165 + }, + { + "epoch": 3.6450939457202507, + "grad_norm": 0.8271334767341614, + "learning_rate": 1.8202530019116487e-06, + "loss": 0.4311, + "step": 1166 + }, + { + "epoch": 3.6482254697286014, + "grad_norm": 0.7617189288139343, + "learning_rate": 1.816087491108292e-06, + "loss": 0.4522, + "step": 1167 + }, + { + "epoch": 3.651356993736952, + "grad_norm": 0.7248172760009766, + "learning_rate": 1.8119240315881126e-06, + "loss": 0.465, + "step": 1168 + }, + { + "epoch": 3.654488517745303, + "grad_norm": 0.8606911897659302, + "learning_rate": 1.8077626358387235e-06, + "loss": 0.4524, + "step": 1169 + }, + { + "epoch": 3.6576200417536535, + "grad_norm": 0.8571308851242065, + "learning_rate": 1.8036033163415484e-06, + "loss": 0.4625, + "step": 1170 + }, + { + "epoch": 3.6607515657620042, + "grad_norm": 0.7239511609077454, + "learning_rate": 1.7994460855717812e-06, + "loss": 0.5025, + "step": 1171 + }, + { + "epoch": 3.663883089770355, + "grad_norm": 0.7958929538726807, + "learning_rate": 1.7952909559983544e-06, + "loss": 0.4382, + "step": 1172 + }, + { + "epoch": 3.6670146137787056, + "grad_norm": 0.7920124530792236, + "learning_rate": 1.7911379400838947e-06, + "loss": 0.4393, + "step": 1173 + }, + { + "epoch": 3.6701461377870563, + "grad_norm": 0.8072578310966492, + "learning_rate": 1.7869870502846903e-06, + "loss": 0.4627, + "step": 1174 + }, + { + "epoch": 3.673277661795407, + "grad_norm": 0.8586218357086182, + "learning_rate": 1.7828382990506543e-06, + "loss": 0.4456, + "step": 1175 + }, + { + "epoch": 3.6764091858037578, + "grad_norm": 0.8741613030433655, + "learning_rate": 1.7786916988252845e-06, + "loss": 0.4613, + "step": 1176 + }, + { + "epoch": 3.6795407098121085, + "grad_norm": 0.7691352367401123, + "learning_rate": 1.774547262045626e-06, + "loss": 0.4641, + "step": 1177 + }, + { + "epoch": 3.682672233820459, + "grad_norm": 0.7866089940071106, + "learning_rate": 1.7704050011422357e-06, + "loss": 0.4308, + "step": 1178 + }, + { + "epoch": 3.68580375782881, + "grad_norm": 0.9934884309768677, + "learning_rate": 1.7662649285391447e-06, + "loss": 0.4434, + "step": 1179 + }, + { + "epoch": 3.6889352818371606, + "grad_norm": 0.794385552406311, + "learning_rate": 1.7621270566538204e-06, + "loss": 0.4481, + "step": 1180 + }, + { + "epoch": 3.6920668058455117, + "grad_norm": 0.7573548555374146, + "learning_rate": 1.7579913978971296e-06, + "loss": 0.4525, + "step": 1181 + }, + { + "epoch": 3.695198329853862, + "grad_norm": 0.7073976993560791, + "learning_rate": 1.7538579646733023e-06, + "loss": 0.4214, + "step": 1182 + }, + { + "epoch": 3.698329853862213, + "grad_norm": 0.8009579181671143, + "learning_rate": 1.7497267693798902e-06, + "loss": 0.4542, + "step": 1183 + }, + { + "epoch": 3.701461377870564, + "grad_norm": 1.4488778114318848, + "learning_rate": 1.7455978244077348e-06, + "loss": 0.443, + "step": 1184 + }, + { + "epoch": 3.7045929018789145, + "grad_norm": 1.0529266595840454, + "learning_rate": 1.7414711421409292e-06, + "loss": 0.4908, + "step": 1185 + }, + { + "epoch": 3.7077244258872653, + "grad_norm": 0.757431149482727, + "learning_rate": 1.7373467349567775e-06, + "loss": 0.4454, + "step": 1186 + }, + { + "epoch": 3.710855949895616, + "grad_norm": 0.7086379528045654, + "learning_rate": 1.733224615225763e-06, + "loss": 0.4292, + "step": 1187 + }, + { + "epoch": 3.7139874739039667, + "grad_norm": 0.7454110383987427, + "learning_rate": 1.7291047953115049e-06, + "loss": 0.4421, + "step": 1188 + }, + { + "epoch": 3.7171189979123174, + "grad_norm": 0.804027795791626, + "learning_rate": 1.7249872875707257e-06, + "loss": 0.4819, + "step": 1189 + }, + { + "epoch": 3.720250521920668, + "grad_norm": 0.8159645199775696, + "learning_rate": 1.7208721043532146e-06, + "loss": 0.4628, + "step": 1190 + }, + { + "epoch": 3.723382045929019, + "grad_norm": 0.8451672196388245, + "learning_rate": 1.7167592580017866e-06, + "loss": 0.4773, + "step": 1191 + }, + { + "epoch": 3.7265135699373695, + "grad_norm": 0.920553982257843, + "learning_rate": 1.7126487608522492e-06, + "loss": 0.4787, + "step": 1192 + }, + { + "epoch": 3.72964509394572, + "grad_norm": 0.9169708490371704, + "learning_rate": 1.7085406252333613e-06, + "loss": 0.4543, + "step": 1193 + }, + { + "epoch": 3.732776617954071, + "grad_norm": 0.7245096564292908, + "learning_rate": 1.7044348634668023e-06, + "loss": 0.4599, + "step": 1194 + }, + { + "epoch": 3.7359081419624216, + "grad_norm": 0.835832417011261, + "learning_rate": 1.7003314878671284e-06, + "loss": 0.4794, + "step": 1195 + }, + { + "epoch": 3.7390396659707723, + "grad_norm": 0.8455896973609924, + "learning_rate": 1.696230510741742e-06, + "loss": 0.4329, + "step": 1196 + }, + { + "epoch": 3.742171189979123, + "grad_norm": 0.743611752986908, + "learning_rate": 1.692131944390849e-06, + "loss": 0.4455, + "step": 1197 + }, + { + "epoch": 3.745302713987474, + "grad_norm": 1.1334915161132812, + "learning_rate": 1.6880358011074272e-06, + "loss": 0.4673, + "step": 1198 + }, + { + "epoch": 3.7484342379958244, + "grad_norm": 0.6935724020004272, + "learning_rate": 1.6839420931771828e-06, + "loss": 0.4686, + "step": 1199 + }, + { + "epoch": 3.7515657620041756, + "grad_norm": 0.962006151676178, + "learning_rate": 1.6798508328785213e-06, + "loss": 0.4474, + "step": 1200 + }, + { + "epoch": 3.754697286012526, + "grad_norm": 1.1499714851379395, + "learning_rate": 1.6757620324825047e-06, + "loss": 0.4659, + "step": 1201 + }, + { + "epoch": 3.757828810020877, + "grad_norm": 0.7689645886421204, + "learning_rate": 1.6716757042528192e-06, + "loss": 0.4552, + "step": 1202 + }, + { + "epoch": 3.7609603340292277, + "grad_norm": 0.7482030391693115, + "learning_rate": 1.6675918604457352e-06, + "loss": 0.4747, + "step": 1203 + }, + { + "epoch": 3.7640918580375784, + "grad_norm": 0.7727032899856567, + "learning_rate": 1.6635105133100686e-06, + "loss": 0.4508, + "step": 1204 + }, + { + "epoch": 3.767223382045929, + "grad_norm": 0.8722149133682251, + "learning_rate": 1.6594316750871514e-06, + "loss": 0.4685, + "step": 1205 + }, + { + "epoch": 3.77035490605428, + "grad_norm": 0.771304726600647, + "learning_rate": 1.6553553580107884e-06, + "loss": 0.4418, + "step": 1206 + }, + { + "epoch": 3.7734864300626305, + "grad_norm": 0.767315149307251, + "learning_rate": 1.6512815743072214e-06, + "loss": 0.4532, + "step": 1207 + }, + { + "epoch": 3.776617954070981, + "grad_norm": 0.8825518488883972, + "learning_rate": 1.6472103361950976e-06, + "loss": 0.468, + "step": 1208 + }, + { + "epoch": 3.779749478079332, + "grad_norm": 0.8887981176376343, + "learning_rate": 1.6431416558854243e-06, + "loss": 0.4264, + "step": 1209 + }, + { + "epoch": 3.7828810020876826, + "grad_norm": 0.8399733304977417, + "learning_rate": 1.63907554558154e-06, + "loss": 0.4405, + "step": 1210 + }, + { + "epoch": 3.7860125260960333, + "grad_norm": 0.8112586140632629, + "learning_rate": 1.6350120174790751e-06, + "loss": 0.445, + "step": 1211 + }, + { + "epoch": 3.789144050104384, + "grad_norm": 0.860775351524353, + "learning_rate": 1.6309510837659137e-06, + "loss": 0.4557, + "step": 1212 + }, + { + "epoch": 3.7922755741127347, + "grad_norm": 0.8522343039512634, + "learning_rate": 1.626892756622161e-06, + "loss": 0.481, + "step": 1213 + }, + { + "epoch": 3.7954070981210855, + "grad_norm": 0.7927511930465698, + "learning_rate": 1.6228370482200988e-06, + "loss": 0.4328, + "step": 1214 + }, + { + "epoch": 3.798538622129436, + "grad_norm": 0.7350064516067505, + "learning_rate": 1.6187839707241604e-06, + "loss": 0.4604, + "step": 1215 + }, + { + "epoch": 3.801670146137787, + "grad_norm": 0.8363698124885559, + "learning_rate": 1.6147335362908847e-06, + "loss": 0.4271, + "step": 1216 + }, + { + "epoch": 3.804801670146138, + "grad_norm": 1.080613613128662, + "learning_rate": 1.610685757068885e-06, + "loss": 0.447, + "step": 1217 + }, + { + "epoch": 3.8079331941544883, + "grad_norm": 1.1507478952407837, + "learning_rate": 1.6066406451988104e-06, + "loss": 0.4664, + "step": 1218 + }, + { + "epoch": 3.8110647181628394, + "grad_norm": 0.7778187990188599, + "learning_rate": 1.6025982128133073e-06, + "loss": 0.456, + "step": 1219 + }, + { + "epoch": 3.81419624217119, + "grad_norm": 0.8383583426475525, + "learning_rate": 1.5985584720369876e-06, + "loss": 0.4684, + "step": 1220 + }, + { + "epoch": 3.817327766179541, + "grad_norm": 0.7743321061134338, + "learning_rate": 1.5945214349863914e-06, + "loss": 0.4567, + "step": 1221 + }, + { + "epoch": 3.8204592901878915, + "grad_norm": 0.8020774126052856, + "learning_rate": 1.5904871137699462e-06, + "loss": 0.4175, + "step": 1222 + }, + { + "epoch": 3.8235908141962422, + "grad_norm": 0.790318489074707, + "learning_rate": 1.5864555204879375e-06, + "loss": 0.469, + "step": 1223 + }, + { + "epoch": 3.826722338204593, + "grad_norm": 0.8583689332008362, + "learning_rate": 1.5824266672324652e-06, + "loss": 0.4931, + "step": 1224 + }, + { + "epoch": 3.8298538622129437, + "grad_norm": 0.7788206934928894, + "learning_rate": 1.5784005660874125e-06, + "loss": 0.4643, + "step": 1225 + }, + { + "epoch": 3.8329853862212944, + "grad_norm": 0.8385717868804932, + "learning_rate": 1.574377229128409e-06, + "loss": 0.4567, + "step": 1226 + }, + { + "epoch": 3.836116910229645, + "grad_norm": 0.8447727560997009, + "learning_rate": 1.5703566684227922e-06, + "loss": 0.42, + "step": 1227 + }, + { + "epoch": 3.8392484342379958, + "grad_norm": 0.7286496758460999, + "learning_rate": 1.5663388960295742e-06, + "loss": 0.4603, + "step": 1228 + }, + { + "epoch": 3.8423799582463465, + "grad_norm": 0.8493947982788086, + "learning_rate": 1.562323923999401e-06, + "loss": 0.4731, + "step": 1229 + }, + { + "epoch": 3.845511482254697, + "grad_norm": 0.8641151785850525, + "learning_rate": 1.5583117643745233e-06, + "loss": 0.4491, + "step": 1230 + }, + { + "epoch": 3.848643006263048, + "grad_norm": 0.9493702054023743, + "learning_rate": 1.5543024291887532e-06, + "loss": 0.411, + "step": 1231 + }, + { + "epoch": 3.8517745302713986, + "grad_norm": 0.7246205806732178, + "learning_rate": 1.5502959304674337e-06, + "loss": 0.4569, + "step": 1232 + }, + { + "epoch": 3.8549060542797493, + "grad_norm": 0.7576872110366821, + "learning_rate": 1.5462922802273994e-06, + "loss": 0.4548, + "step": 1233 + }, + { + "epoch": 3.8580375782881005, + "grad_norm": 0.7710747718811035, + "learning_rate": 1.5422914904769404e-06, + "loss": 0.447, + "step": 1234 + }, + { + "epoch": 3.8611691022964507, + "grad_norm": 0.7661204934120178, + "learning_rate": 1.5382935732157677e-06, + "loss": 0.4601, + "step": 1235 + }, + { + "epoch": 3.864300626304802, + "grad_norm": 0.7133070826530457, + "learning_rate": 1.5342985404349788e-06, + "loss": 0.4245, + "step": 1236 + }, + { + "epoch": 3.867432150313152, + "grad_norm": 0.7716459631919861, + "learning_rate": 1.5303064041170163e-06, + "loss": 0.4543, + "step": 1237 + }, + { + "epoch": 3.8705636743215033, + "grad_norm": 0.737501859664917, + "learning_rate": 1.5263171762356388e-06, + "loss": 0.4405, + "step": 1238 + }, + { + "epoch": 3.873695198329854, + "grad_norm": 0.7885998487472534, + "learning_rate": 1.5223308687558786e-06, + "loss": 0.4412, + "step": 1239 + }, + { + "epoch": 3.8768267223382047, + "grad_norm": 0.7516661882400513, + "learning_rate": 1.5183474936340092e-06, + "loss": 0.4185, + "step": 1240 + }, + { + "epoch": 3.8799582463465554, + "grad_norm": 0.8790446519851685, + "learning_rate": 1.5143670628175111e-06, + "loss": 0.444, + "step": 1241 + }, + { + "epoch": 3.883089770354906, + "grad_norm": 0.7695789337158203, + "learning_rate": 1.5103895882450315e-06, + "loss": 0.4615, + "step": 1242 + }, + { + "epoch": 3.886221294363257, + "grad_norm": 0.7859196662902832, + "learning_rate": 1.506415081846353e-06, + "loss": 0.4379, + "step": 1243 + }, + { + "epoch": 3.8893528183716075, + "grad_norm": 0.7771942615509033, + "learning_rate": 1.5024435555423522e-06, + "loss": 0.5013, + "step": 1244 + }, + { + "epoch": 3.892484342379958, + "grad_norm": 0.7415695190429688, + "learning_rate": 1.498475021244971e-06, + "loss": 0.4861, + "step": 1245 + }, + { + "epoch": 3.895615866388309, + "grad_norm": 0.7909391522407532, + "learning_rate": 1.4945094908571755e-06, + "loss": 0.4599, + "step": 1246 + }, + { + "epoch": 3.8987473903966596, + "grad_norm": 0.7749060988426208, + "learning_rate": 1.490546976272923e-06, + "loss": 0.3986, + "step": 1247 + }, + { + "epoch": 3.9018789144050103, + "grad_norm": 0.810681164264679, + "learning_rate": 1.4865874893771248e-06, + "loss": 0.4495, + "step": 1248 + }, + { + "epoch": 3.905010438413361, + "grad_norm": 0.8018531799316406, + "learning_rate": 1.4826310420456103e-06, + "loss": 0.4426, + "step": 1249 + }, + { + "epoch": 3.9081419624217117, + "grad_norm": 0.756064236164093, + "learning_rate": 1.4786776461450924e-06, + "loss": 0.4474, + "step": 1250 + }, + { + "epoch": 3.911273486430063, + "grad_norm": 0.7581740021705627, + "learning_rate": 1.4747273135331347e-06, + "loss": 0.4494, + "step": 1251 + }, + { + "epoch": 3.914405010438413, + "grad_norm": 0.7666076421737671, + "learning_rate": 1.4707800560581086e-06, + "loss": 0.4593, + "step": 1252 + }, + { + "epoch": 3.9175365344467643, + "grad_norm": 0.7339973449707031, + "learning_rate": 1.4668358855591664e-06, + "loss": 0.4682, + "step": 1253 + }, + { + "epoch": 3.9206680584551146, + "grad_norm": 0.8504599928855896, + "learning_rate": 1.4628948138661974e-06, + "loss": 0.4504, + "step": 1254 + }, + { + "epoch": 3.9237995824634657, + "grad_norm": 0.8332642912864685, + "learning_rate": 1.4589568527997985e-06, + "loss": 0.5024, + "step": 1255 + }, + { + "epoch": 3.9269311064718164, + "grad_norm": 0.7813694477081299, + "learning_rate": 1.4550220141712384e-06, + "loss": 0.4547, + "step": 1256 + }, + { + "epoch": 3.930062630480167, + "grad_norm": 0.842258632183075, + "learning_rate": 1.451090309782417e-06, + "loss": 0.4584, + "step": 1257 + }, + { + "epoch": 3.933194154488518, + "grad_norm": 0.8159133791923523, + "learning_rate": 1.4471617514258373e-06, + "loss": 0.4538, + "step": 1258 + }, + { + "epoch": 3.9363256784968685, + "grad_norm": 0.8117021322250366, + "learning_rate": 1.4432363508845626e-06, + "loss": 0.4315, + "step": 1259 + }, + { + "epoch": 3.9394572025052192, + "grad_norm": 0.8087465167045593, + "learning_rate": 1.4393141199321881e-06, + "loss": 0.4367, + "step": 1260 + }, + { + "epoch": 3.94258872651357, + "grad_norm": 0.7954697012901306, + "learning_rate": 1.435395070332801e-06, + "loss": 0.4515, + "step": 1261 + }, + { + "epoch": 3.9457202505219207, + "grad_norm": 0.7305286526679993, + "learning_rate": 1.4314792138409454e-06, + "loss": 0.4879, + "step": 1262 + }, + { + "epoch": 3.9488517745302714, + "grad_norm": 0.8883433938026428, + "learning_rate": 1.4275665622015908e-06, + "loss": 0.4489, + "step": 1263 + }, + { + "epoch": 3.951983298538622, + "grad_norm": 0.8176298141479492, + "learning_rate": 1.4236571271500909e-06, + "loss": 0.4583, + "step": 1264 + }, + { + "epoch": 3.9551148225469728, + "grad_norm": 0.8042430281639099, + "learning_rate": 1.4197509204121563e-06, + "loss": 0.4277, + "step": 1265 + }, + { + "epoch": 3.9582463465553235, + "grad_norm": 0.8153829574584961, + "learning_rate": 1.4158479537038095e-06, + "loss": 0.4389, + "step": 1266 + }, + { + "epoch": 3.961377870563674, + "grad_norm": 0.7908188104629517, + "learning_rate": 1.4119482387313588e-06, + "loss": 0.4421, + "step": 1267 + }, + { + "epoch": 3.964509394572025, + "grad_norm": 0.831758975982666, + "learning_rate": 1.4080517871913596e-06, + "loss": 0.4308, + "step": 1268 + }, + { + "epoch": 3.9676409185803756, + "grad_norm": 0.8191989064216614, + "learning_rate": 1.4041586107705758e-06, + "loss": 0.4654, + "step": 1269 + }, + { + "epoch": 3.9707724425887267, + "grad_norm": 0.9455055594444275, + "learning_rate": 1.4002687211459524e-06, + "loss": 0.4668, + "step": 1270 + }, + { + "epoch": 3.973903966597077, + "grad_norm": 0.9271034002304077, + "learning_rate": 1.396382129984572e-06, + "loss": 0.4414, + "step": 1271 + }, + { + "epoch": 3.977035490605428, + "grad_norm": 0.7652955651283264, + "learning_rate": 1.392498848943627e-06, + "loss": 0.4575, + "step": 1272 + }, + { + "epoch": 3.980167014613779, + "grad_norm": 0.7850046157836914, + "learning_rate": 1.3886188896703816e-06, + "loss": 0.4554, + "step": 1273 + }, + { + "epoch": 3.9832985386221296, + "grad_norm": 0.7194349765777588, + "learning_rate": 1.3847422638021357e-06, + "loss": 0.437, + "step": 1274 + }, + { + "epoch": 3.9864300626304803, + "grad_norm": 0.8726270198822021, + "learning_rate": 1.3808689829661899e-06, + "loss": 0.4657, + "step": 1275 + }, + { + "epoch": 3.989561586638831, + "grad_norm": 0.7741451263427734, + "learning_rate": 1.3769990587798146e-06, + "loss": 0.3931, + "step": 1276 + }, + { + "epoch": 3.9926931106471817, + "grad_norm": 0.9160324931144714, + "learning_rate": 1.3731325028502116e-06, + "loss": 0.4358, + "step": 1277 + }, + { + "epoch": 3.9958246346555324, + "grad_norm": 0.7819761037826538, + "learning_rate": 1.3692693267744806e-06, + "loss": 0.4767, + "step": 1278 + }, + { + "epoch": 3.998956158663883, + "grad_norm": 0.7595860958099365, + "learning_rate": 1.365409542139583e-06, + "loss": 0.4217, + "step": 1279 + }, + { + "epoch": 4.0, + "grad_norm": 0.7595860958099365, + "learning_rate": 1.361553160522307e-06, + "loss": 0.1478, + "step": 1280 + }, + { + "epoch": 4.003131524008351, + "grad_norm": 1.0793898105621338, + "learning_rate": 1.357700193489237e-06, + "loss": 0.4453, + "step": 1281 + }, + { + "epoch": 4.006263048016701, + "grad_norm": 0.8228668570518494, + "learning_rate": 1.3538506525967148e-06, + "loss": 0.3981, + "step": 1282 + }, + { + "epoch": 4.009394572025053, + "grad_norm": 0.7946518063545227, + "learning_rate": 1.3500045493908044e-06, + "loss": 0.423, + "step": 1283 + }, + { + "epoch": 4.012526096033403, + "grad_norm": 0.7924147248268127, + "learning_rate": 1.3461618954072614e-06, + "loss": 0.4211, + "step": 1284 + }, + { + "epoch": 4.015657620041754, + "grad_norm": 0.891967236995697, + "learning_rate": 1.3423227021714937e-06, + "loss": 0.4365, + "step": 1285 + }, + { + "epoch": 4.018789144050104, + "grad_norm": 0.9970533847808838, + "learning_rate": 1.3384869811985323e-06, + "loss": 0.4053, + "step": 1286 + }, + { + "epoch": 4.021920668058455, + "grad_norm": 1.001489281654358, + "learning_rate": 1.33465474399299e-06, + "loss": 0.4382, + "step": 1287 + }, + { + "epoch": 4.025052192066806, + "grad_norm": 0.7584173083305359, + "learning_rate": 1.3308260020490332e-06, + "loss": 0.441, + "step": 1288 + }, + { + "epoch": 4.028183716075157, + "grad_norm": 0.7444385290145874, + "learning_rate": 1.327000766850345e-06, + "loss": 0.4172, + "step": 1289 + }, + { + "epoch": 4.031315240083507, + "grad_norm": 0.8264429569244385, + "learning_rate": 1.3231790498700886e-06, + "loss": 0.3956, + "step": 1290 + }, + { + "epoch": 4.034446764091858, + "grad_norm": 0.7339959740638733, + "learning_rate": 1.319360862570877e-06, + "loss": 0.4431, + "step": 1291 + }, + { + "epoch": 4.0375782881002085, + "grad_norm": 0.9837627410888672, + "learning_rate": 1.3155462164047345e-06, + "loss": 0.4439, + "step": 1292 + }, + { + "epoch": 4.04070981210856, + "grad_norm": 0.8792996406555176, + "learning_rate": 1.3117351228130657e-06, + "loss": 0.4664, + "step": 1293 + }, + { + "epoch": 4.04384133611691, + "grad_norm": 0.8377962112426758, + "learning_rate": 1.3079275932266205e-06, + "loss": 0.4267, + "step": 1294 + }, + { + "epoch": 4.046972860125261, + "grad_norm": 0.755979597568512, + "learning_rate": 1.3041236390654577e-06, + "loss": 0.4394, + "step": 1295 + }, + { + "epoch": 4.050104384133611, + "grad_norm": 0.7994539737701416, + "learning_rate": 1.3003232717389114e-06, + "loss": 0.4547, + "step": 1296 + }, + { + "epoch": 4.053235908141962, + "grad_norm": 0.8308882117271423, + "learning_rate": 1.2965265026455598e-06, + "loss": 0.408, + "step": 1297 + }, + { + "epoch": 4.056367432150314, + "grad_norm": 0.8120949268341064, + "learning_rate": 1.292733343173188e-06, + "loss": 0.3524, + "step": 1298 + }, + { + "epoch": 4.059498956158664, + "grad_norm": 0.7639186978340149, + "learning_rate": 1.288943804698755e-06, + "loss": 0.4677, + "step": 1299 + }, + { + "epoch": 4.062630480167015, + "grad_norm": 0.8439323306083679, + "learning_rate": 1.2851578985883586e-06, + "loss": 0.436, + "step": 1300 + }, + { + "epoch": 4.065762004175365, + "grad_norm": 0.7270680069923401, + "learning_rate": 1.2813756361972001e-06, + "loss": 0.4519, + "step": 1301 + }, + { + "epoch": 4.068893528183716, + "grad_norm": 0.7746233940124512, + "learning_rate": 1.2775970288695554e-06, + "loss": 0.4503, + "step": 1302 + }, + { + "epoch": 4.072025052192067, + "grad_norm": 0.7725366950035095, + "learning_rate": 1.2738220879387376e-06, + "loss": 0.4467, + "step": 1303 + }, + { + "epoch": 4.075156576200418, + "grad_norm": 0.7371346950531006, + "learning_rate": 1.2700508247270597e-06, + "loss": 0.4144, + "step": 1304 + }, + { + "epoch": 4.078288100208768, + "grad_norm": 0.778403639793396, + "learning_rate": 1.2662832505458057e-06, + "loss": 0.4345, + "step": 1305 + }, + { + "epoch": 4.081419624217119, + "grad_norm": 0.8683139085769653, + "learning_rate": 1.2625193766951957e-06, + "loss": 0.4345, + "step": 1306 + }, + { + "epoch": 4.0845511482254695, + "grad_norm": 0.7849321365356445, + "learning_rate": 1.2587592144643513e-06, + "loss": 0.4505, + "step": 1307 + }, + { + "epoch": 4.087682672233821, + "grad_norm": 0.9351589679718018, + "learning_rate": 1.2550027751312618e-06, + "loss": 0.4173, + "step": 1308 + }, + { + "epoch": 4.090814196242171, + "grad_norm": 0.7938134074211121, + "learning_rate": 1.2512500699627489e-06, + "loss": 0.4061, + "step": 1309 + }, + { + "epoch": 4.093945720250522, + "grad_norm": 0.8015687465667725, + "learning_rate": 1.2475011102144337e-06, + "loss": 0.423, + "step": 1310 + }, + { + "epoch": 4.097077244258872, + "grad_norm": 0.8477723598480225, + "learning_rate": 1.2437559071307062e-06, + "loss": 0.4475, + "step": 1311 + }, + { + "epoch": 4.1002087682672235, + "grad_norm": 0.7643985152244568, + "learning_rate": 1.2400144719446885e-06, + "loss": 0.4613, + "step": 1312 + }, + { + "epoch": 4.103340292275574, + "grad_norm": 0.7612454295158386, + "learning_rate": 1.2362768158781985e-06, + "loss": 0.4333, + "step": 1313 + }, + { + "epoch": 4.106471816283925, + "grad_norm": 0.9039669036865234, + "learning_rate": 1.2325429501417232e-06, + "loss": 0.4091, + "step": 1314 + }, + { + "epoch": 4.109603340292275, + "grad_norm": 0.8423281908035278, + "learning_rate": 1.228812885934378e-06, + "loss": 0.4482, + "step": 1315 + }, + { + "epoch": 4.112734864300626, + "grad_norm": 0.8980326652526855, + "learning_rate": 1.2250866344438782e-06, + "loss": 0.4354, + "step": 1316 + }, + { + "epoch": 4.115866388308977, + "grad_norm": 0.7539350986480713, + "learning_rate": 1.221364206846502e-06, + "loss": 0.4332, + "step": 1317 + }, + { + "epoch": 4.118997912317328, + "grad_norm": 0.855925977230072, + "learning_rate": 1.2176456143070597e-06, + "loss": 0.4198, + "step": 1318 + }, + { + "epoch": 4.122129436325679, + "grad_norm": 0.7506774663925171, + "learning_rate": 1.2139308679788594e-06, + "loss": 0.4484, + "step": 1319 + }, + { + "epoch": 4.125260960334029, + "grad_norm": 0.820524275302887, + "learning_rate": 1.2102199790036709e-06, + "loss": 0.4208, + "step": 1320 + }, + { + "epoch": 4.12839248434238, + "grad_norm": 0.8255232572555542, + "learning_rate": 1.2065129585116968e-06, + "loss": 0.4431, + "step": 1321 + }, + { + "epoch": 4.1315240083507305, + "grad_norm": 0.7454650402069092, + "learning_rate": 1.2028098176215347e-06, + "loss": 0.3833, + "step": 1322 + }, + { + "epoch": 4.134655532359082, + "grad_norm": 0.7500002980232239, + "learning_rate": 1.199110567440148e-06, + "loss": 0.4486, + "step": 1323 + }, + { + "epoch": 4.137787056367432, + "grad_norm": 0.783704400062561, + "learning_rate": 1.1954152190628304e-06, + "loss": 0.4535, + "step": 1324 + }, + { + "epoch": 4.140918580375783, + "grad_norm": 0.7665114402770996, + "learning_rate": 1.1917237835731713e-06, + "loss": 0.4575, + "step": 1325 + }, + { + "epoch": 4.144050104384133, + "grad_norm": 0.7992947697639465, + "learning_rate": 1.1880362720430235e-06, + "loss": 0.4294, + "step": 1326 + }, + { + "epoch": 4.1471816283924845, + "grad_norm": 0.8273386359214783, + "learning_rate": 1.1843526955324736e-06, + "loss": 0.4578, + "step": 1327 + }, + { + "epoch": 4.150313152400835, + "grad_norm": 0.7992604970932007, + "learning_rate": 1.1806730650898029e-06, + "loss": 0.4125, + "step": 1328 + }, + { + "epoch": 4.153444676409186, + "grad_norm": 0.7935183644294739, + "learning_rate": 1.1769973917514612e-06, + "loss": 0.4537, + "step": 1329 + }, + { + "epoch": 4.156576200417536, + "grad_norm": 0.8761199116706848, + "learning_rate": 1.173325686542022e-06, + "loss": 0.4433, + "step": 1330 + }, + { + "epoch": 4.159707724425887, + "grad_norm": 0.7764279246330261, + "learning_rate": 1.1696579604741643e-06, + "loss": 0.4165, + "step": 1331 + }, + { + "epoch": 4.162839248434238, + "grad_norm": 0.9479373693466187, + "learning_rate": 1.165994224548629e-06, + "loss": 0.42, + "step": 1332 + }, + { + "epoch": 4.165970772442589, + "grad_norm": 1.1482913494110107, + "learning_rate": 1.162334489754191e-06, + "loss": 0.4545, + "step": 1333 + }, + { + "epoch": 4.16910229645094, + "grad_norm": 0.825964093208313, + "learning_rate": 1.1586787670676228e-06, + "loss": 0.4423, + "step": 1334 + }, + { + "epoch": 4.17223382045929, + "grad_norm": 0.8352368474006653, + "learning_rate": 1.1550270674536626e-06, + "loss": 0.4509, + "step": 1335 + }, + { + "epoch": 4.175365344467641, + "grad_norm": 0.8211585879325867, + "learning_rate": 1.1513794018649846e-06, + "loss": 0.4482, + "step": 1336 + }, + { + "epoch": 4.1784968684759916, + "grad_norm": 0.7863073945045471, + "learning_rate": 1.1477357812421628e-06, + "loss": 0.4318, + "step": 1337 + }, + { + "epoch": 4.181628392484343, + "grad_norm": 0.8023143410682678, + "learning_rate": 1.1440962165136366e-06, + "loss": 0.4269, + "step": 1338 + }, + { + "epoch": 4.184759916492693, + "grad_norm": 0.8101564049720764, + "learning_rate": 1.1404607185956843e-06, + "loss": 0.4284, + "step": 1339 + }, + { + "epoch": 4.187891440501044, + "grad_norm": 0.755865216255188, + "learning_rate": 1.1368292983923826e-06, + "loss": 0.4195, + "step": 1340 + }, + { + "epoch": 4.191022964509394, + "grad_norm": 0.7991048097610474, + "learning_rate": 1.1332019667955805e-06, + "loss": 0.3774, + "step": 1341 + }, + { + "epoch": 4.1941544885177455, + "grad_norm": 0.7407388091087341, + "learning_rate": 1.1295787346848641e-06, + "loss": 0.4321, + "step": 1342 + }, + { + "epoch": 4.197286012526096, + "grad_norm": 0.8769527673721313, + "learning_rate": 1.1259596129275205e-06, + "loss": 0.4349, + "step": 1343 + }, + { + "epoch": 4.200417536534447, + "grad_norm": 0.7908845543861389, + "learning_rate": 1.1223446123785128e-06, + "loss": 0.4595, + "step": 1344 + }, + { + "epoch": 4.203549060542797, + "grad_norm": 0.810578465461731, + "learning_rate": 1.1187337438804394e-06, + "loss": 0.4464, + "step": 1345 + }, + { + "epoch": 4.206680584551148, + "grad_norm": 0.8122991323471069, + "learning_rate": 1.1151270182635082e-06, + "loss": 0.4437, + "step": 1346 + }, + { + "epoch": 4.209812108559499, + "grad_norm": 0.8065986633300781, + "learning_rate": 1.1115244463454988e-06, + "loss": 0.4198, + "step": 1347 + }, + { + "epoch": 4.21294363256785, + "grad_norm": 0.7652766108512878, + "learning_rate": 1.107926038931734e-06, + "loss": 0.4141, + "step": 1348 + }, + { + "epoch": 4.2160751565762, + "grad_norm": 0.7936646938323975, + "learning_rate": 1.1043318068150457e-06, + "loss": 0.43, + "step": 1349 + }, + { + "epoch": 4.219206680584551, + "grad_norm": 0.8732025027275085, + "learning_rate": 1.1007417607757435e-06, + "loss": 0.4243, + "step": 1350 + }, + { + "epoch": 4.222338204592901, + "grad_norm": 0.8516861200332642, + "learning_rate": 1.0971559115815795e-06, + "loss": 0.4352, + "step": 1351 + }, + { + "epoch": 4.225469728601253, + "grad_norm": 0.7985629439353943, + "learning_rate": 1.093574269987718e-06, + "loss": 0.4193, + "step": 1352 + }, + { + "epoch": 4.228601252609604, + "grad_norm": 0.7350221872329712, + "learning_rate": 1.0899968467367056e-06, + "loss": 0.4466, + "step": 1353 + }, + { + "epoch": 4.231732776617954, + "grad_norm": 0.7626602649688721, + "learning_rate": 1.086423652558436e-06, + "loss": 0.399, + "step": 1354 + }, + { + "epoch": 4.234864300626305, + "grad_norm": 0.7596367597579956, + "learning_rate": 1.082854698170117e-06, + "loss": 0.4023, + "step": 1355 + }, + { + "epoch": 4.237995824634655, + "grad_norm": 0.7802876830101013, + "learning_rate": 1.0792899942762406e-06, + "loss": 0.4351, + "step": 1356 + }, + { + "epoch": 4.241127348643007, + "grad_norm": 1.0809392929077148, + "learning_rate": 1.0757295515685504e-06, + "loss": 0.4271, + "step": 1357 + }, + { + "epoch": 4.244258872651357, + "grad_norm": 0.8581835031509399, + "learning_rate": 1.0721733807260094e-06, + "loss": 0.4332, + "step": 1358 + }, + { + "epoch": 4.247390396659708, + "grad_norm": 0.884496808052063, + "learning_rate": 1.0686214924147686e-06, + "loss": 0.4317, + "step": 1359 + }, + { + "epoch": 4.250521920668058, + "grad_norm": 0.7651242017745972, + "learning_rate": 1.0650738972881325e-06, + "loss": 0.4614, + "step": 1360 + }, + { + "epoch": 4.253653444676409, + "grad_norm": 0.7798812985420227, + "learning_rate": 1.0615306059865286e-06, + "loss": 0.4088, + "step": 1361 + }, + { + "epoch": 4.25678496868476, + "grad_norm": 0.9011983871459961, + "learning_rate": 1.0579916291374776e-06, + "loss": 0.4242, + "step": 1362 + }, + { + "epoch": 4.259916492693111, + "grad_norm": 0.7490668892860413, + "learning_rate": 1.0544569773555602e-06, + "loss": 0.4369, + "step": 1363 + }, + { + "epoch": 4.263048016701461, + "grad_norm": 0.7936168909072876, + "learning_rate": 1.0509266612423819e-06, + "loss": 0.4321, + "step": 1364 + }, + { + "epoch": 4.266179540709812, + "grad_norm": 0.8428751826286316, + "learning_rate": 1.047400691386547e-06, + "loss": 0.4391, + "step": 1365 + }, + { + "epoch": 4.2693110647181625, + "grad_norm": 0.7458098530769348, + "learning_rate": 1.0438790783636213e-06, + "loss": 0.4789, + "step": 1366 + }, + { + "epoch": 4.272442588726514, + "grad_norm": 0.8869938254356384, + "learning_rate": 1.0403618327361056e-06, + "loss": 0.4181, + "step": 1367 + }, + { + "epoch": 4.275574112734864, + "grad_norm": 0.7711350321769714, + "learning_rate": 1.0368489650533989e-06, + "loss": 0.4334, + "step": 1368 + }, + { + "epoch": 4.278705636743215, + "grad_norm": 0.7643832564353943, + "learning_rate": 1.0333404858517712e-06, + "loss": 0.4444, + "step": 1369 + }, + { + "epoch": 4.281837160751566, + "grad_norm": 0.925422191619873, + "learning_rate": 1.02983640565433e-06, + "loss": 0.4134, + "step": 1370 + }, + { + "epoch": 4.284968684759916, + "grad_norm": 1.4223377704620361, + "learning_rate": 1.0263367349709866e-06, + "loss": 0.4441, + "step": 1371 + }, + { + "epoch": 4.288100208768268, + "grad_norm": 0.9713383316993713, + "learning_rate": 1.0228414842984297e-06, + "loss": 0.418, + "step": 1372 + }, + { + "epoch": 4.291231732776618, + "grad_norm": 0.8378648161888123, + "learning_rate": 1.0193506641200874e-06, + "loss": 0.4017, + "step": 1373 + }, + { + "epoch": 4.294363256784969, + "grad_norm": 0.9343763589859009, + "learning_rate": 1.0158642849061018e-06, + "loss": 0.4468, + "step": 1374 + }, + { + "epoch": 4.297494780793319, + "grad_norm": 0.7863892912864685, + "learning_rate": 1.012382357113296e-06, + "loss": 0.4235, + "step": 1375 + }, + { + "epoch": 4.30062630480167, + "grad_norm": 0.8109874725341797, + "learning_rate": 1.0089048911851388e-06, + "loss": 0.4356, + "step": 1376 + }, + { + "epoch": 4.303757828810021, + "grad_norm": 0.7978389859199524, + "learning_rate": 1.005431897551717e-06, + "loss": 0.4517, + "step": 1377 + }, + { + "epoch": 4.306889352818372, + "grad_norm": 0.8213954567909241, + "learning_rate": 1.001963386629705e-06, + "loss": 0.4263, + "step": 1378 + }, + { + "epoch": 4.310020876826722, + "grad_norm": 0.7367708683013916, + "learning_rate": 9.984993688223319e-07, + "loss": 0.4467, + "step": 1379 + }, + { + "epoch": 4.313152400835073, + "grad_norm": 0.8723400831222534, + "learning_rate": 9.950398545193508e-07, + "loss": 0.4517, + "step": 1380 + }, + { + "epoch": 4.3162839248434235, + "grad_norm": 0.7927054166793823, + "learning_rate": 9.915848540970033e-07, + "loss": 0.4755, + "step": 1381 + }, + { + "epoch": 4.319415448851775, + "grad_norm": 0.8598108291625977, + "learning_rate": 9.88134377917997e-07, + "loss": 0.4539, + "step": 1382 + }, + { + "epoch": 4.322546972860125, + "grad_norm": 1.0660723447799683, + "learning_rate": 9.84688436331468e-07, + "loss": 0.4541, + "step": 1383 + }, + { + "epoch": 4.325678496868476, + "grad_norm": 0.840942919254303, + "learning_rate": 9.81247039672953e-07, + "loss": 0.4239, + "step": 1384 + }, + { + "epoch": 4.328810020876826, + "grad_norm": 0.8142675757408142, + "learning_rate": 9.778101982643549e-07, + "loss": 0.4259, + "step": 1385 + }, + { + "epoch": 4.3319415448851775, + "grad_norm": 0.7841728329658508, + "learning_rate": 9.743779224139133e-07, + "loss": 0.432, + "step": 1386 + }, + { + "epoch": 4.335073068893529, + "grad_norm": 0.9642646908760071, + "learning_rate": 9.709502224161769e-07, + "loss": 0.4423, + "step": 1387 + }, + { + "epoch": 4.338204592901879, + "grad_norm": 0.7967872619628906, + "learning_rate": 9.675271085519692e-07, + "loss": 0.4428, + "step": 1388 + }, + { + "epoch": 4.34133611691023, + "grad_norm": 0.7652244567871094, + "learning_rate": 9.64108591088356e-07, + "loss": 0.4607, + "step": 1389 + }, + { + "epoch": 4.34446764091858, + "grad_norm": 0.8806408643722534, + "learning_rate": 9.606946802786204e-07, + "loss": 0.3859, + "step": 1390 + }, + { + "epoch": 4.347599164926931, + "grad_norm": 0.8638584613800049, + "learning_rate": 9.572853863622252e-07, + "loss": 0.3758, + "step": 1391 + }, + { + "epoch": 4.350730688935282, + "grad_norm": 0.7262263894081116, + "learning_rate": 9.538807195647882e-07, + "loss": 0.4435, + "step": 1392 + }, + { + "epoch": 4.353862212943633, + "grad_norm": 0.785175621509552, + "learning_rate": 9.504806900980482e-07, + "loss": 0.4037, + "step": 1393 + }, + { + "epoch": 4.356993736951983, + "grad_norm": 0.8310196995735168, + "learning_rate": 9.470853081598338e-07, + "loss": 0.4434, + "step": 1394 + }, + { + "epoch": 4.360125260960334, + "grad_norm": 0.7547351717948914, + "learning_rate": 9.436945839340364e-07, + "loss": 0.4318, + "step": 1395 + }, + { + "epoch": 4.3632567849686845, + "grad_norm": 0.816460907459259, + "learning_rate": 9.403085275905746e-07, + "loss": 0.4173, + "step": 1396 + }, + { + "epoch": 4.366388308977036, + "grad_norm": 0.8159666657447815, + "learning_rate": 9.369271492853696e-07, + "loss": 0.4506, + "step": 1397 + }, + { + "epoch": 4.369519832985386, + "grad_norm": 0.8234115839004517, + "learning_rate": 9.335504591603084e-07, + "loss": 0.4186, + "step": 1398 + }, + { + "epoch": 4.372651356993737, + "grad_norm": 0.9084867238998413, + "learning_rate": 9.301784673432187e-07, + "loss": 0.4414, + "step": 1399 + }, + { + "epoch": 4.375782881002087, + "grad_norm": 0.7826300263404846, + "learning_rate": 9.268111839478369e-07, + "loss": 0.4415, + "step": 1400 + }, + { + "epoch": 4.3789144050104385, + "grad_norm": 0.7957226634025574, + "learning_rate": 9.23448619073774e-07, + "loss": 0.4307, + "step": 1401 + }, + { + "epoch": 4.382045929018789, + "grad_norm": 0.7768528461456299, + "learning_rate": 9.200907828064931e-07, + "loss": 0.4107, + "step": 1402 + }, + { + "epoch": 4.38517745302714, + "grad_norm": 0.878063440322876, + "learning_rate": 9.167376852172702e-07, + "loss": 0.4232, + "step": 1403 + }, + { + "epoch": 4.388308977035491, + "grad_norm": 0.7961450219154358, + "learning_rate": 9.133893363631713e-07, + "loss": 0.449, + "step": 1404 + }, + { + "epoch": 4.391440501043841, + "grad_norm": 0.8734915256500244, + "learning_rate": 9.100457462870197e-07, + "loss": 0.4392, + "step": 1405 + }, + { + "epoch": 4.3945720250521925, + "grad_norm": 0.7659426331520081, + "learning_rate": 9.06706925017363e-07, + "loss": 0.4391, + "step": 1406 + }, + { + "epoch": 4.397703549060543, + "grad_norm": 0.949674665927887, + "learning_rate": 9.033728825684466e-07, + "loss": 0.4054, + "step": 1407 + }, + { + "epoch": 4.400835073068894, + "grad_norm": 0.8209534287452698, + "learning_rate": 9.000436289401832e-07, + "loss": 0.3973, + "step": 1408 + }, + { + "epoch": 4.403966597077244, + "grad_norm": 0.9697580933570862, + "learning_rate": 8.967191741181225e-07, + "loss": 0.4228, + "step": 1409 + }, + { + "epoch": 4.407098121085595, + "grad_norm": 0.7681871652603149, + "learning_rate": 8.933995280734217e-07, + "loss": 0.4183, + "step": 1410 + }, + { + "epoch": 4.4102296450939455, + "grad_norm": 0.7434380054473877, + "learning_rate": 8.900847007628103e-07, + "loss": 0.4068, + "step": 1411 + }, + { + "epoch": 4.413361169102297, + "grad_norm": 0.9835489988327026, + "learning_rate": 8.867747021285697e-07, + "loss": 0.4356, + "step": 1412 + }, + { + "epoch": 4.416492693110647, + "grad_norm": 0.7727295160293579, + "learning_rate": 8.834695420984971e-07, + "loss": 0.4201, + "step": 1413 + }, + { + "epoch": 4.419624217118998, + "grad_norm": 0.8018465042114258, + "learning_rate": 8.801692305858775e-07, + "loss": 0.4471, + "step": 1414 + }, + { + "epoch": 4.422755741127348, + "grad_norm": 0.8191575407981873, + "learning_rate": 8.76873777489452e-07, + "loss": 0.4438, + "step": 1415 + }, + { + "epoch": 4.4258872651356995, + "grad_norm": 0.7755978107452393, + "learning_rate": 8.735831926933896e-07, + "loss": 0.4441, + "step": 1416 + }, + { + "epoch": 4.42901878914405, + "grad_norm": 0.8469848036766052, + "learning_rate": 8.702974860672603e-07, + "loss": 0.4134, + "step": 1417 + }, + { + "epoch": 4.432150313152401, + "grad_norm": 0.7887343168258667, + "learning_rate": 8.67016667466001e-07, + "loss": 0.4247, + "step": 1418 + }, + { + "epoch": 4.435281837160751, + "grad_norm": 0.7888761758804321, + "learning_rate": 8.637407467298867e-07, + "loss": 0.4313, + "step": 1419 + }, + { + "epoch": 4.438413361169102, + "grad_norm": 0.8109477162361145, + "learning_rate": 8.60469733684505e-07, + "loss": 0.4472, + "step": 1420 + }, + { + "epoch": 4.4415448851774535, + "grad_norm": 0.8876966238021851, + "learning_rate": 8.572036381407203e-07, + "loss": 0.4395, + "step": 1421 + }, + { + "epoch": 4.444676409185804, + "grad_norm": 0.7669360637664795, + "learning_rate": 8.539424698946503e-07, + "loss": 0.3943, + "step": 1422 + }, + { + "epoch": 4.447807933194154, + "grad_norm": 0.7810583710670471, + "learning_rate": 8.506862387276338e-07, + "loss": 0.4189, + "step": 1423 + }, + { + "epoch": 4.450939457202505, + "grad_norm": 0.9363948702812195, + "learning_rate": 8.474349544061996e-07, + "loss": 0.4245, + "step": 1424 + }, + { + "epoch": 4.454070981210856, + "grad_norm": 0.8147290349006653, + "learning_rate": 8.441886266820418e-07, + "loss": 0.4381, + "step": 1425 + }, + { + "epoch": 4.457202505219207, + "grad_norm": 0.9227074384689331, + "learning_rate": 8.409472652919873e-07, + "loss": 0.4289, + "step": 1426 + }, + { + "epoch": 4.460334029227558, + "grad_norm": 0.8162581920623779, + "learning_rate": 8.377108799579661e-07, + "loss": 0.4565, + "step": 1427 + }, + { + "epoch": 4.463465553235908, + "grad_norm": 0.7634908556938171, + "learning_rate": 8.344794803869835e-07, + "loss": 0.4402, + "step": 1428 + }, + { + "epoch": 4.466597077244259, + "grad_norm": 0.7960299253463745, + "learning_rate": 8.312530762710924e-07, + "loss": 0.4533, + "step": 1429 + }, + { + "epoch": 4.469728601252609, + "grad_norm": 0.8322834968566895, + "learning_rate": 8.280316772873617e-07, + "loss": 0.4282, + "step": 1430 + }, + { + "epoch": 4.4728601252609606, + "grad_norm": 0.7285633683204651, + "learning_rate": 8.248152930978498e-07, + "loss": 0.4287, + "step": 1431 + }, + { + "epoch": 4.475991649269311, + "grad_norm": 0.760178804397583, + "learning_rate": 8.21603933349569e-07, + "loss": 0.4243, + "step": 1432 + }, + { + "epoch": 4.479123173277662, + "grad_norm": 0.8069238662719727, + "learning_rate": 8.183976076744671e-07, + "loss": 0.4352, + "step": 1433 + }, + { + "epoch": 4.482254697286012, + "grad_norm": 0.8068020343780518, + "learning_rate": 8.151963256893911e-07, + "loss": 0.4143, + "step": 1434 + }, + { + "epoch": 4.485386221294363, + "grad_norm": 0.7954995036125183, + "learning_rate": 8.120000969960606e-07, + "loss": 0.4122, + "step": 1435 + }, + { + "epoch": 4.488517745302714, + "grad_norm": 0.791763961315155, + "learning_rate": 8.088089311810379e-07, + "loss": 0.4016, + "step": 1436 + }, + { + "epoch": 4.491649269311065, + "grad_norm": 0.8935474157333374, + "learning_rate": 8.056228378156994e-07, + "loss": 0.4392, + "step": 1437 + }, + { + "epoch": 4.494780793319415, + "grad_norm": 0.791764497756958, + "learning_rate": 8.024418264562094e-07, + "loss": 0.4433, + "step": 1438 + }, + { + "epoch": 4.497912317327766, + "grad_norm": 0.8151891827583313, + "learning_rate": 7.992659066434899e-07, + "loss": 0.4161, + "step": 1439 + }, + { + "epoch": 4.5010438413361165, + "grad_norm": 0.9389538168907166, + "learning_rate": 7.960950879031884e-07, + "loss": 0.4405, + "step": 1440 + }, + { + "epoch": 4.504175365344468, + "grad_norm": 0.7620337009429932, + "learning_rate": 7.929293797456561e-07, + "loss": 0.4443, + "step": 1441 + }, + { + "epoch": 4.507306889352819, + "grad_norm": 0.7784422039985657, + "learning_rate": 7.897687916659133e-07, + "loss": 0.4688, + "step": 1442 + }, + { + "epoch": 4.510438413361169, + "grad_norm": 0.759872555732727, + "learning_rate": 7.866133331436249e-07, + "loss": 0.4349, + "step": 1443 + }, + { + "epoch": 4.51356993736952, + "grad_norm": 0.8815323114395142, + "learning_rate": 7.834630136430713e-07, + "loss": 0.4729, + "step": 1444 + }, + { + "epoch": 4.51670146137787, + "grad_norm": 0.7574021816253662, + "learning_rate": 7.803178426131161e-07, + "loss": 0.4263, + "step": 1445 + }, + { + "epoch": 4.519832985386222, + "grad_norm": 0.7910036444664001, + "learning_rate": 7.771778294871849e-07, + "loss": 0.424, + "step": 1446 + }, + { + "epoch": 4.522964509394572, + "grad_norm": 0.8388727903366089, + "learning_rate": 7.740429836832295e-07, + "loss": 0.4495, + "step": 1447 + }, + { + "epoch": 4.526096033402923, + "grad_norm": 0.7924241423606873, + "learning_rate": 7.709133146037059e-07, + "loss": 0.4157, + "step": 1448 + }, + { + "epoch": 4.529227557411273, + "grad_norm": 0.8103229403495789, + "learning_rate": 7.677888316355411e-07, + "loss": 0.4321, + "step": 1449 + }, + { + "epoch": 4.532359081419624, + "grad_norm": 0.8175046443939209, + "learning_rate": 7.64669544150109e-07, + "loss": 0.4375, + "step": 1450 + }, + { + "epoch": 4.535490605427975, + "grad_norm": 0.8193264603614807, + "learning_rate": 7.615554615032e-07, + "loss": 0.4576, + "step": 1451 + }, + { + "epoch": 4.538622129436326, + "grad_norm": 0.7646095156669617, + "learning_rate": 7.584465930349924e-07, + "loss": 0.4095, + "step": 1452 + }, + { + "epoch": 4.541753653444676, + "grad_norm": 0.8025760054588318, + "learning_rate": 7.553429480700275e-07, + "loss": 0.4387, + "step": 1453 + }, + { + "epoch": 4.544885177453027, + "grad_norm": 0.7339401841163635, + "learning_rate": 7.522445359171768e-07, + "loss": 0.4263, + "step": 1454 + }, + { + "epoch": 4.5480167014613775, + "grad_norm": 0.764500617980957, + "learning_rate": 7.491513658696189e-07, + "loss": 0.4694, + "step": 1455 + }, + { + "epoch": 4.551148225469729, + "grad_norm": 0.7974144220352173, + "learning_rate": 7.460634472048104e-07, + "loss": 0.4335, + "step": 1456 + }, + { + "epoch": 4.554279749478079, + "grad_norm": 0.7994454503059387, + "learning_rate": 7.429807891844546e-07, + "loss": 0.4155, + "step": 1457 + }, + { + "epoch": 4.55741127348643, + "grad_norm": 0.8552598357200623, + "learning_rate": 7.39903401054477e-07, + "loss": 0.428, + "step": 1458 + }, + { + "epoch": 4.560542797494781, + "grad_norm": 0.8604694604873657, + "learning_rate": 7.368312920449985e-07, + "loss": 0.4562, + "step": 1459 + }, + { + "epoch": 4.5636743215031315, + "grad_norm": 0.7699970602989197, + "learning_rate": 7.337644713703049e-07, + "loss": 0.4322, + "step": 1460 + }, + { + "epoch": 4.566805845511483, + "grad_norm": 0.9426809549331665, + "learning_rate": 7.307029482288227e-07, + "loss": 0.387, + "step": 1461 + }, + { + "epoch": 4.569937369519833, + "grad_norm": 0.8332602381706238, + "learning_rate": 7.276467318030841e-07, + "loss": 0.4381, + "step": 1462 + }, + { + "epoch": 4.573068893528184, + "grad_norm": 0.7953102588653564, + "learning_rate": 7.245958312597095e-07, + "loss": 0.3994, + "step": 1463 + }, + { + "epoch": 4.576200417536534, + "grad_norm": 0.7888246178627014, + "learning_rate": 7.215502557493743e-07, + "loss": 0.4191, + "step": 1464 + }, + { + "epoch": 4.579331941544885, + "grad_norm": 1.00043523311615, + "learning_rate": 7.185100144067816e-07, + "loss": 0.4321, + "step": 1465 + }, + { + "epoch": 4.582463465553236, + "grad_norm": 0.8209455609321594, + "learning_rate": 7.154751163506354e-07, + "loss": 0.3841, + "step": 1466 + }, + { + "epoch": 4.585594989561587, + "grad_norm": 0.7574297785758972, + "learning_rate": 7.124455706836131e-07, + "loss": 0.452, + "step": 1467 + }, + { + "epoch": 4.588726513569937, + "grad_norm": 0.8509530425071716, + "learning_rate": 7.094213864923397e-07, + "loss": 0.4225, + "step": 1468 + }, + { + "epoch": 4.591858037578288, + "grad_norm": 0.7906093597412109, + "learning_rate": 7.064025728473589e-07, + "loss": 0.418, + "step": 1469 + }, + { + "epoch": 4.5949895615866385, + "grad_norm": 0.7408525943756104, + "learning_rate": 7.033891388031056e-07, + "loss": 0.4366, + "step": 1470 + }, + { + "epoch": 4.59812108559499, + "grad_norm": 0.766342043876648, + "learning_rate": 7.003810933978805e-07, + "loss": 0.4315, + "step": 1471 + }, + { + "epoch": 4.60125260960334, + "grad_norm": 0.7852875590324402, + "learning_rate": 6.973784456538207e-07, + "loss": 0.4332, + "step": 1472 + }, + { + "epoch": 4.604384133611691, + "grad_norm": 0.7975596785545349, + "learning_rate": 6.943812045768753e-07, + "loss": 0.4563, + "step": 1473 + }, + { + "epoch": 4.607515657620041, + "grad_norm": 0.755770742893219, + "learning_rate": 6.913893791567767e-07, + "loss": 0.417, + "step": 1474 + }, + { + "epoch": 4.6106471816283925, + "grad_norm": 0.8083193302154541, + "learning_rate": 6.884029783670123e-07, + "loss": 0.4285, + "step": 1475 + }, + { + "epoch": 4.613778705636744, + "grad_norm": 0.8146732449531555, + "learning_rate": 6.854220111648022e-07, + "loss": 0.4587, + "step": 1476 + }, + { + "epoch": 4.616910229645094, + "grad_norm": 0.7958261370658875, + "learning_rate": 6.82446486491066e-07, + "loss": 0.4584, + "step": 1477 + }, + { + "epoch": 4.620041753653445, + "grad_norm": 0.9063606262207031, + "learning_rate": 6.79476413270402e-07, + "loss": 0.4056, + "step": 1478 + }, + { + "epoch": 4.623173277661795, + "grad_norm": 0.8176311254501343, + "learning_rate": 6.76511800411056e-07, + "loss": 0.3987, + "step": 1479 + }, + { + "epoch": 4.6263048016701465, + "grad_norm": 0.7843388319015503, + "learning_rate": 6.735526568048975e-07, + "loss": 0.4541, + "step": 1480 + }, + { + "epoch": 4.629436325678497, + "grad_norm": 0.7491182684898376, + "learning_rate": 6.705989913273922e-07, + "loss": 0.4218, + "step": 1481 + }, + { + "epoch": 4.632567849686848, + "grad_norm": 0.8836645483970642, + "learning_rate": 6.676508128375728e-07, + "loss": 0.4153, + "step": 1482 + }, + { + "epoch": 4.635699373695198, + "grad_norm": 0.7614302635192871, + "learning_rate": 6.647081301780175e-07, + "loss": 0.4557, + "step": 1483 + }, + { + "epoch": 4.638830897703549, + "grad_norm": 0.8144665360450745, + "learning_rate": 6.617709521748181e-07, + "loss": 0.4103, + "step": 1484 + }, + { + "epoch": 4.6419624217118995, + "grad_norm": 0.757859468460083, + "learning_rate": 6.588392876375579e-07, + "loss": 0.4498, + "step": 1485 + }, + { + "epoch": 4.645093945720251, + "grad_norm": 0.8298365473747253, + "learning_rate": 6.559131453592837e-07, + "loss": 0.4318, + "step": 1486 + }, + { + "epoch": 4.648225469728601, + "grad_norm": 0.9429208636283875, + "learning_rate": 6.529925341164781e-07, + "loss": 0.4272, + "step": 1487 + }, + { + "epoch": 4.651356993736952, + "grad_norm": 0.7761420011520386, + "learning_rate": 6.500774626690329e-07, + "loss": 0.4224, + "step": 1488 + }, + { + "epoch": 4.654488517745302, + "grad_norm": 0.8003484606742859, + "learning_rate": 6.471679397602273e-07, + "loss": 0.4333, + "step": 1489 + }, + { + "epoch": 4.6576200417536535, + "grad_norm": 0.8728703856468201, + "learning_rate": 6.44263974116697e-07, + "loss": 0.4412, + "step": 1490 + }, + { + "epoch": 4.660751565762004, + "grad_norm": 0.7923367023468018, + "learning_rate": 6.41365574448411e-07, + "loss": 0.4401, + "step": 1491 + }, + { + "epoch": 4.663883089770355, + "grad_norm": 0.8110585808753967, + "learning_rate": 6.384727494486398e-07, + "loss": 0.4152, + "step": 1492 + }, + { + "epoch": 4.667014613778706, + "grad_norm": 0.7865241765975952, + "learning_rate": 6.355855077939385e-07, + "loss": 0.4608, + "step": 1493 + }, + { + "epoch": 4.670146137787056, + "grad_norm": 0.8125724196434021, + "learning_rate": 6.327038581441136e-07, + "loss": 0.419, + "step": 1494 + }, + { + "epoch": 4.673277661795407, + "grad_norm": 0.8024583458900452, + "learning_rate": 6.298278091422003e-07, + "loss": 0.4575, + "step": 1495 + }, + { + "epoch": 4.676409185803758, + "grad_norm": 0.7569037675857544, + "learning_rate": 6.269573694144343e-07, + "loss": 0.4257, + "step": 1496 + }, + { + "epoch": 4.679540709812109, + "grad_norm": 0.783733069896698, + "learning_rate": 6.240925475702275e-07, + "loss": 0.4361, + "step": 1497 + }, + { + "epoch": 4.682672233820459, + "grad_norm": 0.7900946736335754, + "learning_rate": 6.212333522021424e-07, + "loss": 0.4372, + "step": 1498 + }, + { + "epoch": 4.68580375782881, + "grad_norm": 0.7943285703659058, + "learning_rate": 6.183797918858667e-07, + "loss": 0.4337, + "step": 1499 + }, + { + "epoch": 4.688935281837161, + "grad_norm": 0.7793418765068054, + "learning_rate": 6.155318751801842e-07, + "loss": 0.4264, + "step": 1500 + }, + { + "epoch": 4.692066805845512, + "grad_norm": 1.0244885683059692, + "learning_rate": 6.126896106269533e-07, + "loss": 0.446, + "step": 1501 + }, + { + "epoch": 4.695198329853862, + "grad_norm": 0.8742548823356628, + "learning_rate": 6.098530067510808e-07, + "loss": 0.3953, + "step": 1502 + }, + { + "epoch": 4.698329853862213, + "grad_norm": 0.7844028472900391, + "learning_rate": 6.070220720604919e-07, + "loss": 0.402, + "step": 1503 + }, + { + "epoch": 4.701461377870563, + "grad_norm": 0.7485696077346802, + "learning_rate": 6.041968150461119e-07, + "loss": 0.4435, + "step": 1504 + }, + { + "epoch": 4.7045929018789145, + "grad_norm": 0.8526521921157837, + "learning_rate": 6.01377244181833e-07, + "loss": 0.4565, + "step": 1505 + }, + { + "epoch": 4.707724425887265, + "grad_norm": 0.7832267880439758, + "learning_rate": 5.985633679244957e-07, + "loss": 0.4489, + "step": 1506 + }, + { + "epoch": 4.710855949895616, + "grad_norm": 0.8092924952507019, + "learning_rate": 5.957551947138599e-07, + "loss": 0.4011, + "step": 1507 + }, + { + "epoch": 4.713987473903966, + "grad_norm": 0.8008314371109009, + "learning_rate": 5.92952732972579e-07, + "loss": 0.4557, + "step": 1508 + }, + { + "epoch": 4.717118997912317, + "grad_norm": 0.8025342226028442, + "learning_rate": 5.901559911061758e-07, + "loss": 0.38, + "step": 1509 + }, + { + "epoch": 4.7202505219206685, + "grad_norm": 0.8290171027183533, + "learning_rate": 5.873649775030188e-07, + "loss": 0.4019, + "step": 1510 + }, + { + "epoch": 4.723382045929019, + "grad_norm": 0.8971850872039795, + "learning_rate": 5.845797005342943e-07, + "loss": 0.4354, + "step": 1511 + }, + { + "epoch": 4.726513569937369, + "grad_norm": 0.8087455630302429, + "learning_rate": 5.818001685539843e-07, + "loss": 0.4537, + "step": 1512 + }, + { + "epoch": 4.72964509394572, + "grad_norm": 0.822287380695343, + "learning_rate": 5.790263898988355e-07, + "loss": 0.4692, + "step": 1513 + }, + { + "epoch": 4.732776617954071, + "grad_norm": 1.007889986038208, + "learning_rate": 5.762583728883428e-07, + "loss": 0.4317, + "step": 1514 + }, + { + "epoch": 4.735908141962422, + "grad_norm": 0.7517583966255188, + "learning_rate": 5.734961258247177e-07, + "loss": 0.4168, + "step": 1515 + }, + { + "epoch": 4.739039665970773, + "grad_norm": 1.1469851732254028, + "learning_rate": 5.707396569928675e-07, + "loss": 0.4412, + "step": 1516 + }, + { + "epoch": 4.742171189979123, + "grad_norm": 0.8889291286468506, + "learning_rate": 5.679889746603667e-07, + "loss": 0.4133, + "step": 1517 + }, + { + "epoch": 4.745302713987474, + "grad_norm": 0.7792209386825562, + "learning_rate": 5.652440870774342e-07, + "loss": 0.4256, + "step": 1518 + }, + { + "epoch": 4.748434237995824, + "grad_norm": 0.7862168550491333, + "learning_rate": 5.625050024769099e-07, + "loss": 0.395, + "step": 1519 + }, + { + "epoch": 4.751565762004176, + "grad_norm": 0.8305843472480774, + "learning_rate": 5.597717290742285e-07, + "loss": 0.4172, + "step": 1520 + }, + { + "epoch": 4.754697286012526, + "grad_norm": 0.8393134474754333, + "learning_rate": 5.570442750673932e-07, + "loss": 0.4438, + "step": 1521 + }, + { + "epoch": 4.757828810020877, + "grad_norm": 0.8481642603874207, + "learning_rate": 5.543226486369555e-07, + "loss": 0.4251, + "step": 1522 + }, + { + "epoch": 4.760960334029227, + "grad_norm": 0.8482305407524109, + "learning_rate": 5.516068579459849e-07, + "loss": 0.4372, + "step": 1523 + }, + { + "epoch": 4.764091858037578, + "grad_norm": 0.7428895235061646, + "learning_rate": 5.488969111400502e-07, + "loss": 0.4017, + "step": 1524 + }, + { + "epoch": 4.767223382045929, + "grad_norm": 0.9066843390464783, + "learning_rate": 5.461928163471922e-07, + "loss": 0.4263, + "step": 1525 + }, + { + "epoch": 4.77035490605428, + "grad_norm": 0.8754342794418335, + "learning_rate": 5.434945816778972e-07, + "loss": 0.4086, + "step": 1526 + }, + { + "epoch": 4.773486430062631, + "grad_norm": 0.7580026388168335, + "learning_rate": 5.408022152250783e-07, + "loss": 0.4331, + "step": 1527 + }, + { + "epoch": 4.776617954070981, + "grad_norm": 0.7960991859436035, + "learning_rate": 5.381157250640451e-07, + "loss": 0.4482, + "step": 1528 + }, + { + "epoch": 4.7797494780793315, + "grad_norm": 0.8199532628059387, + "learning_rate": 5.354351192524842e-07, + "loss": 0.4558, + "step": 1529 + }, + { + "epoch": 4.782881002087683, + "grad_norm": 0.8457065224647522, + "learning_rate": 5.327604058304312e-07, + "loss": 0.4711, + "step": 1530 + }, + { + "epoch": 4.786012526096034, + "grad_norm": 0.9078546762466431, + "learning_rate": 5.300915928202499e-07, + "loss": 0.4341, + "step": 1531 + }, + { + "epoch": 4.789144050104384, + "grad_norm": 0.7782992720603943, + "learning_rate": 5.274286882266072e-07, + "loss": 0.4115, + "step": 1532 + }, + { + "epoch": 4.792275574112735, + "grad_norm": 0.894313395023346, + "learning_rate": 5.247717000364463e-07, + "loss": 0.4415, + "step": 1533 + }, + { + "epoch": 4.7954070981210855, + "grad_norm": 0.8233445882797241, + "learning_rate": 5.221206362189682e-07, + "loss": 0.4198, + "step": 1534 + }, + { + "epoch": 4.798538622129437, + "grad_norm": 0.7654063105583191, + "learning_rate": 5.194755047256017e-07, + "loss": 0.4364, + "step": 1535 + }, + { + "epoch": 4.801670146137787, + "grad_norm": 0.7879475951194763, + "learning_rate": 5.168363134899845e-07, + "loss": 0.4366, + "step": 1536 + }, + { + "epoch": 4.804801670146138, + "grad_norm": 0.7754428386688232, + "learning_rate": 5.142030704279377e-07, + "loss": 0.4365, + "step": 1537 + }, + { + "epoch": 4.807933194154488, + "grad_norm": 0.758696436882019, + "learning_rate": 5.11575783437441e-07, + "loss": 0.4442, + "step": 1538 + }, + { + "epoch": 4.811064718162839, + "grad_norm": 0.7870402336120605, + "learning_rate": 5.089544603986085e-07, + "loss": 0.4298, + "step": 1539 + }, + { + "epoch": 4.81419624217119, + "grad_norm": 0.7607670426368713, + "learning_rate": 5.063391091736691e-07, + "loss": 0.435, + "step": 1540 + }, + { + "epoch": 4.817327766179541, + "grad_norm": 0.8175606727600098, + "learning_rate": 5.037297376069388e-07, + "loss": 0.4325, + "step": 1541 + }, + { + "epoch": 4.820459290187891, + "grad_norm": 0.9083423018455505, + "learning_rate": 5.011263535248001e-07, + "loss": 0.3847, + "step": 1542 + }, + { + "epoch": 4.823590814196242, + "grad_norm": 0.7435231804847717, + "learning_rate": 4.985289647356731e-07, + "loss": 0.4818, + "step": 1543 + }, + { + "epoch": 4.826722338204593, + "grad_norm": 1.5190348625183105, + "learning_rate": 4.9593757903e-07, + "loss": 0.4068, + "step": 1544 + }, + { + "epoch": 4.829853862212944, + "grad_norm": 0.9122394919395447, + "learning_rate": 4.933522041802166e-07, + "loss": 0.4701, + "step": 1545 + }, + { + "epoch": 4.832985386221294, + "grad_norm": 0.9334122538566589, + "learning_rate": 4.9077284794073e-07, + "loss": 0.4178, + "step": 1546 + }, + { + "epoch": 4.836116910229645, + "grad_norm": 0.8561045527458191, + "learning_rate": 4.881995180478949e-07, + "loss": 0.4373, + "step": 1547 + }, + { + "epoch": 4.839248434237996, + "grad_norm": 0.8061811923980713, + "learning_rate": 4.85632222219991e-07, + "loss": 0.4414, + "step": 1548 + }, + { + "epoch": 4.8423799582463465, + "grad_norm": 0.7866719961166382, + "learning_rate": 4.830709681572008e-07, + "loss": 0.438, + "step": 1549 + }, + { + "epoch": 4.845511482254698, + "grad_norm": 0.8633884787559509, + "learning_rate": 4.805157635415852e-07, + "loss": 0.4639, + "step": 1550 + }, + { + "epoch": 4.848643006263048, + "grad_norm": 1.125654935836792, + "learning_rate": 4.779666160370596e-07, + "loss": 0.4364, + "step": 1551 + }, + { + "epoch": 4.851774530271399, + "grad_norm": 0.8371828198432922, + "learning_rate": 4.7542353328937473e-07, + "loss": 0.4436, + "step": 1552 + }, + { + "epoch": 4.854906054279749, + "grad_norm": 0.8166123628616333, + "learning_rate": 4.7288652292608736e-07, + "loss": 0.4382, + "step": 1553 + }, + { + "epoch": 4.8580375782881005, + "grad_norm": 0.8664330244064331, + "learning_rate": 4.7035559255654393e-07, + "loss": 0.4314, + "step": 1554 + }, + { + "epoch": 4.861169102296451, + "grad_norm": 0.8051781058311462, + "learning_rate": 4.6783074977185485e-07, + "loss": 0.4522, + "step": 1555 + }, + { + "epoch": 4.864300626304802, + "grad_norm": 0.7385572791099548, + "learning_rate": 4.653120021448701e-07, + "loss": 0.4254, + "step": 1556 + }, + { + "epoch": 4.867432150313152, + "grad_norm": 0.7310816645622253, + "learning_rate": 4.6279935723016033e-07, + "loss": 0.4277, + "step": 1557 + }, + { + "epoch": 4.870563674321503, + "grad_norm": 0.7682080864906311, + "learning_rate": 4.602928225639899e-07, + "loss": 0.422, + "step": 1558 + }, + { + "epoch": 4.8736951983298535, + "grad_norm": 0.8093141317367554, + "learning_rate": 4.577924056642985e-07, + "loss": 0.4376, + "step": 1559 + }, + { + "epoch": 4.876826722338205, + "grad_norm": 0.8006629943847656, + "learning_rate": 4.5529811403067503e-07, + "loss": 0.4483, + "step": 1560 + }, + { + "epoch": 4.879958246346555, + "grad_norm": 0.7647503614425659, + "learning_rate": 4.528099551443377e-07, + "loss": 0.418, + "step": 1561 + }, + { + "epoch": 4.883089770354906, + "grad_norm": 0.8467709422111511, + "learning_rate": 4.503279364681104e-07, + "loss": 0.4272, + "step": 1562 + }, + { + "epoch": 4.886221294363256, + "grad_norm": 0.8667131662368774, + "learning_rate": 4.4785206544640055e-07, + "loss": 0.4347, + "step": 1563 + }, + { + "epoch": 4.8893528183716075, + "grad_norm": 0.8670737147331238, + "learning_rate": 4.453823495051748e-07, + "loss": 0.4313, + "step": 1564 + }, + { + "epoch": 4.892484342379959, + "grad_norm": 0.8131253719329834, + "learning_rate": 4.4291879605194157e-07, + "loss": 0.4335, + "step": 1565 + }, + { + "epoch": 4.895615866388309, + "grad_norm": 0.9633654356002808, + "learning_rate": 4.404614124757242e-07, + "loss": 0.4637, + "step": 1566 + }, + { + "epoch": 4.89874739039666, + "grad_norm": 0.9058681726455688, + "learning_rate": 4.3801020614704174e-07, + "loss": 0.4503, + "step": 1567 + }, + { + "epoch": 4.90187891440501, + "grad_norm": 0.7923601865768433, + "learning_rate": 4.3556518441788405e-07, + "loss": 0.4526, + "step": 1568 + }, + { + "epoch": 4.9050104384133615, + "grad_norm": 0.7733371257781982, + "learning_rate": 4.331263546216913e-07, + "loss": 0.437, + "step": 1569 + }, + { + "epoch": 4.908141962421712, + "grad_norm": 0.7897522449493408, + "learning_rate": 4.306937240733336e-07, + "loss": 0.4213, + "step": 1570 + }, + { + "epoch": 4.911273486430063, + "grad_norm": 0.7857375741004944, + "learning_rate": 4.282673000690865e-07, + "loss": 0.4576, + "step": 1571 + }, + { + "epoch": 4.914405010438413, + "grad_norm": 0.7968363761901855, + "learning_rate": 4.258470898866102e-07, + "loss": 0.4383, + "step": 1572 + }, + { + "epoch": 4.917536534446764, + "grad_norm": 0.762876033782959, + "learning_rate": 4.2343310078492626e-07, + "loss": 0.4129, + "step": 1573 + }, + { + "epoch": 4.920668058455115, + "grad_norm": 0.9070468544960022, + "learning_rate": 4.21025340004399e-07, + "loss": 0.3994, + "step": 1574 + }, + { + "epoch": 4.923799582463466, + "grad_norm": 0.8299228549003601, + "learning_rate": 4.186238147667113e-07, + "loss": 0.4461, + "step": 1575 + }, + { + "epoch": 4.926931106471816, + "grad_norm": 0.8604512214660645, + "learning_rate": 4.162285322748441e-07, + "loss": 0.4447, + "step": 1576 + }, + { + "epoch": 4.930062630480167, + "grad_norm": 0.8441461324691772, + "learning_rate": 4.1383949971305207e-07, + "loss": 0.4547, + "step": 1577 + }, + { + "epoch": 4.933194154488517, + "grad_norm": 0.7836902737617493, + "learning_rate": 4.114567242468473e-07, + "loss": 0.4311, + "step": 1578 + }, + { + "epoch": 4.9363256784968685, + "grad_norm": 0.7957454919815063, + "learning_rate": 4.0908021302297205e-07, + "loss": 0.4074, + "step": 1579 + }, + { + "epoch": 4.939457202505219, + "grad_norm": 0.772109866142273, + "learning_rate": 4.0670997316938216e-07, + "loss": 0.4696, + "step": 1580 + }, + { + "epoch": 4.94258872651357, + "grad_norm": 1.3906599283218384, + "learning_rate": 4.043460117952214e-07, + "loss": 0.4379, + "step": 1581 + }, + { + "epoch": 4.945720250521921, + "grad_norm": 0.9287625551223755, + "learning_rate": 4.0198833599080415e-07, + "loss": 0.4576, + "step": 1582 + }, + { + "epoch": 4.948851774530271, + "grad_norm": 0.8610300421714783, + "learning_rate": 3.99636952827592e-07, + "loss": 0.4158, + "step": 1583 + }, + { + "epoch": 4.9519832985386225, + "grad_norm": 0.7845875024795532, + "learning_rate": 3.972918693581712e-07, + "loss": 0.4428, + "step": 1584 + }, + { + "epoch": 4.955114822546973, + "grad_norm": 0.8758688569068909, + "learning_rate": 3.949530926162354e-07, + "loss": 0.4113, + "step": 1585 + }, + { + "epoch": 4.958246346555324, + "grad_norm": 0.797717809677124, + "learning_rate": 3.926206296165599e-07, + "loss": 0.4181, + "step": 1586 + }, + { + "epoch": 4.961377870563674, + "grad_norm": 0.7839358448982239, + "learning_rate": 3.9029448735498495e-07, + "loss": 0.4218, + "step": 1587 + }, + { + "epoch": 4.964509394572025, + "grad_norm": 0.8518044352531433, + "learning_rate": 3.8797467280839216e-07, + "loss": 0.4425, + "step": 1588 + }, + { + "epoch": 4.967640918580376, + "grad_norm": 0.7976752519607544, + "learning_rate": 3.856611929346837e-07, + "loss": 0.4736, + "step": 1589 + }, + { + "epoch": 4.970772442588727, + "grad_norm": 0.9000734090805054, + "learning_rate": 3.8335405467276184e-07, + "loss": 0.4282, + "step": 1590 + }, + { + "epoch": 4.973903966597077, + "grad_norm": 0.8424035906791687, + "learning_rate": 3.810532649425089e-07, + "loss": 0.4443, + "step": 1591 + }, + { + "epoch": 4.977035490605428, + "grad_norm": 0.8014306426048279, + "learning_rate": 3.7875883064476564e-07, + "loss": 0.4617, + "step": 1592 + }, + { + "epoch": 4.980167014613778, + "grad_norm": 0.7423905730247498, + "learning_rate": 3.76470758661312e-07, + "loss": 0.468, + "step": 1593 + }, + { + "epoch": 4.98329853862213, + "grad_norm": 0.7726851105690002, + "learning_rate": 3.741890558548414e-07, + "loss": 0.3824, + "step": 1594 + }, + { + "epoch": 4.98643006263048, + "grad_norm": 0.7915946245193481, + "learning_rate": 3.719137290689479e-07, + "loss": 0.4091, + "step": 1595 + }, + { + "epoch": 4.989561586638831, + "grad_norm": 0.8000501394271851, + "learning_rate": 3.696447851280999e-07, + "loss": 0.4523, + "step": 1596 + }, + { + "epoch": 4.992693110647181, + "grad_norm": 0.8696115612983704, + "learning_rate": 3.6738223083762235e-07, + "loss": 0.4383, + "step": 1597 + }, + { + "epoch": 4.995824634655532, + "grad_norm": 0.781044065952301, + "learning_rate": 3.65126072983675e-07, + "loss": 0.4506, + "step": 1598 + }, + { + "epoch": 4.9989561586638835, + "grad_norm": 0.7926247715950012, + "learning_rate": 3.628763183332315e-07, + "loss": 0.4322, + "step": 1599 + }, + { + "epoch": 5.0, + "grad_norm": 0.7926247715950012, + "learning_rate": 3.606329736340622e-07, + "loss": 0.1457, + "step": 1600 + }, + { + "epoch": 5.003131524008351, + "grad_norm": 0.7916404008865356, + "learning_rate": 3.583960456147109e-07, + "loss": 0.431, + "step": 1601 + }, + { + "epoch": 5.006263048016701, + "grad_norm": 0.7794471383094788, + "learning_rate": 3.5616554098447496e-07, + "loss": 0.4054, + "step": 1602 + }, + { + "epoch": 5.009394572025053, + "grad_norm": 0.8006277680397034, + "learning_rate": 3.5394146643338735e-07, + "loss": 0.4363, + "step": 1603 + }, + { + "epoch": 5.012526096033403, + "grad_norm": 0.9790255427360535, + "learning_rate": 3.5172382863219277e-07, + "loss": 0.4169, + "step": 1604 + }, + { + "epoch": 5.015657620041754, + "grad_norm": 1.0182846784591675, + "learning_rate": 3.4951263423233265e-07, + "loss": 0.4704, + "step": 1605 + }, + { + "epoch": 5.018789144050104, + "grad_norm": 0.7919366359710693, + "learning_rate": 3.47307889865921e-07, + "loss": 0.4567, + "step": 1606 + }, + { + "epoch": 5.021920668058455, + "grad_norm": 0.7617084383964539, + "learning_rate": 3.4510960214572565e-07, + "loss": 0.3865, + "step": 1607 + }, + { + "epoch": 5.025052192066806, + "grad_norm": 0.7226077318191528, + "learning_rate": 3.429177776651502e-07, + "loss": 0.3847, + "step": 1608 + }, + { + "epoch": 5.028183716075157, + "grad_norm": 0.9526088833808899, + "learning_rate": 3.407324229982112e-07, + "loss": 0.4116, + "step": 1609 + }, + { + "epoch": 5.031315240083507, + "grad_norm": 0.891015350818634, + "learning_rate": 3.385535446995214e-07, + "loss": 0.4026, + "step": 1610 + }, + { + "epoch": 5.034446764091858, + "grad_norm": 0.8143556714057922, + "learning_rate": 3.3638114930426735e-07, + "loss": 0.3868, + "step": 1611 + }, + { + "epoch": 5.0375782881002085, + "grad_norm": 0.7985808849334717, + "learning_rate": 3.3421524332819285e-07, + "loss": 0.4176, + "step": 1612 + }, + { + "epoch": 5.04070981210856, + "grad_norm": 0.7742509245872498, + "learning_rate": 3.320558332675772e-07, + "loss": 0.3822, + "step": 1613 + }, + { + "epoch": 5.04384133611691, + "grad_norm": 0.8089630603790283, + "learning_rate": 3.2990292559921533e-07, + "loss": 0.3907, + "step": 1614 + }, + { + "epoch": 5.046972860125261, + "grad_norm": 0.7523915767669678, + "learning_rate": 3.2775652678039974e-07, + "loss": 0.4232, + "step": 1615 + }, + { + "epoch": 5.050104384133611, + "grad_norm": 0.852554202079773, + "learning_rate": 3.2561664324890134e-07, + "loss": 0.4127, + "step": 1616 + }, + { + "epoch": 5.053235908141962, + "grad_norm": 0.8473255038261414, + "learning_rate": 3.2348328142294936e-07, + "loss": 0.4323, + "step": 1617 + }, + { + "epoch": 5.056367432150314, + "grad_norm": 0.8101459741592407, + "learning_rate": 3.213564477012121e-07, + "loss": 0.4208, + "step": 1618 + }, + { + "epoch": 5.059498956158664, + "grad_norm": 0.7732040882110596, + "learning_rate": 3.192361484627776e-07, + "loss": 0.4312, + "step": 1619 + }, + { + "epoch": 5.062630480167015, + "grad_norm": 0.7684810757637024, + "learning_rate": 3.171223900671344e-07, + "loss": 0.4223, + "step": 1620 + }, + { + "epoch": 5.065762004175365, + "grad_norm": 1.0066255331039429, + "learning_rate": 3.1501517885415364e-07, + "loss": 0.4284, + "step": 1621 + }, + { + "epoch": 5.068893528183716, + "grad_norm": 0.7965579032897949, + "learning_rate": 3.129145211440701e-07, + "loss": 0.4041, + "step": 1622 + }, + { + "epoch": 5.072025052192067, + "grad_norm": 0.7432489395141602, + "learning_rate": 3.1082042323746063e-07, + "loss": 0.3852, + "step": 1623 + }, + { + "epoch": 5.075156576200418, + "grad_norm": 0.894244909286499, + "learning_rate": 3.0873289141522714e-07, + "loss": 0.4479, + "step": 1624 + }, + { + "epoch": 5.078288100208768, + "grad_norm": 0.7385063767433167, + "learning_rate": 3.0665193193857887e-07, + "loss": 0.4048, + "step": 1625 + }, + { + "epoch": 5.081419624217119, + "grad_norm": 0.9110557436943054, + "learning_rate": 3.045775510490115e-07, + "loss": 0.4375, + "step": 1626 + }, + { + "epoch": 5.0845511482254695, + "grad_norm": 0.7882060408592224, + "learning_rate": 3.0250975496829067e-07, + "loss": 0.4421, + "step": 1627 + }, + { + "epoch": 5.087682672233821, + "grad_norm": 0.9643567800521851, + "learning_rate": 3.0044854989843027e-07, + "loss": 0.4161, + "step": 1628 + }, + { + "epoch": 5.090814196242171, + "grad_norm": 0.8286798000335693, + "learning_rate": 2.983939420216753e-07, + "loss": 0.4013, + "step": 1629 + }, + { + "epoch": 5.093945720250522, + "grad_norm": 0.7984923124313354, + "learning_rate": 2.9634593750048515e-07, + "loss": 0.3759, + "step": 1630 + }, + { + "epoch": 5.097077244258872, + "grad_norm": 0.7450951337814331, + "learning_rate": 2.9430454247751344e-07, + "loss": 0.4233, + "step": 1631 + }, + { + "epoch": 5.1002087682672235, + "grad_norm": 0.7780002355575562, + "learning_rate": 2.92269763075588e-07, + "loss": 0.4299, + "step": 1632 + }, + { + "epoch": 5.103340292275574, + "grad_norm": 0.7606906890869141, + "learning_rate": 2.9024160539769614e-07, + "loss": 0.42, + "step": 1633 + }, + { + "epoch": 5.106471816283925, + "grad_norm": 0.8596094250679016, + "learning_rate": 2.882200755269626e-07, + "loss": 0.3855, + "step": 1634 + }, + { + "epoch": 5.109603340292275, + "grad_norm": 0.7705757021903992, + "learning_rate": 2.8620517952663464e-07, + "loss": 0.4214, + "step": 1635 + }, + { + "epoch": 5.112734864300626, + "grad_norm": 0.9210963845252991, + "learning_rate": 2.8419692344006186e-07, + "loss": 0.4137, + "step": 1636 + }, + { + "epoch": 5.115866388308977, + "grad_norm": 0.8594221472740173, + "learning_rate": 2.821953132906777e-07, + "loss": 0.4409, + "step": 1637 + }, + { + "epoch": 5.118997912317328, + "grad_norm": 0.8145505785942078, + "learning_rate": 2.8020035508198374e-07, + "loss": 0.4206, + "step": 1638 + }, + { + "epoch": 5.122129436325679, + "grad_norm": 0.8511978983879089, + "learning_rate": 2.782120547975281e-07, + "loss": 0.4069, + "step": 1639 + }, + { + "epoch": 5.125260960334029, + "grad_norm": 0.7916121482849121, + "learning_rate": 2.762304184008918e-07, + "loss": 0.4053, + "step": 1640 + }, + { + "epoch": 5.12839248434238, + "grad_norm": 1.0378046035766602, + "learning_rate": 2.7425545183566686e-07, + "loss": 0.4058, + "step": 1641 + }, + { + "epoch": 5.1315240083507305, + "grad_norm": 0.7503740787506104, + "learning_rate": 2.722871610254413e-07, + "loss": 0.4146, + "step": 1642 + }, + { + "epoch": 5.134655532359082, + "grad_norm": 0.8860176801681519, + "learning_rate": 2.703255518737802e-07, + "loss": 0.4192, + "step": 1643 + }, + { + "epoch": 5.137787056367432, + "grad_norm": 0.7693402171134949, + "learning_rate": 2.683706302642075e-07, + "loss": 0.4222, + "step": 1644 + }, + { + "epoch": 5.140918580375783, + "grad_norm": 0.8174161911010742, + "learning_rate": 2.6642240206018884e-07, + "loss": 0.4267, + "step": 1645 + }, + { + "epoch": 5.144050104384133, + "grad_norm": 0.7899668216705322, + "learning_rate": 2.644808731051152e-07, + "loss": 0.3892, + "step": 1646 + }, + { + "epoch": 5.1471816283924845, + "grad_norm": 0.8949608206748962, + "learning_rate": 2.625460492222831e-07, + "loss": 0.3932, + "step": 1647 + }, + { + "epoch": 5.150313152400835, + "grad_norm": 0.841679036617279, + "learning_rate": 2.6061793621488e-07, + "loss": 0.426, + "step": 1648 + }, + { + "epoch": 5.153444676409186, + "grad_norm": 0.8701457977294922, + "learning_rate": 2.5869653986596297e-07, + "loss": 0.4113, + "step": 1649 + }, + { + "epoch": 5.156576200417536, + "grad_norm": 0.8086332678794861, + "learning_rate": 2.567818659384441e-07, + "loss": 0.4324, + "step": 1650 + }, + { + "epoch": 5.159707724425887, + "grad_norm": 0.8881601095199585, + "learning_rate": 2.548739201750741e-07, + "loss": 0.4159, + "step": 1651 + }, + { + "epoch": 5.162839248434238, + "grad_norm": 0.7634018659591675, + "learning_rate": 2.529727082984229e-07, + "loss": 0.3897, + "step": 1652 + }, + { + "epoch": 5.165970772442589, + "grad_norm": 0.8098715543746948, + "learning_rate": 2.510782360108624e-07, + "loss": 0.42, + "step": 1653 + }, + { + "epoch": 5.16910229645094, + "grad_norm": 0.9650408029556274, + "learning_rate": 2.491905089945515e-07, + "loss": 0.4377, + "step": 1654 + }, + { + "epoch": 5.17223382045929, + "grad_norm": 0.8331677317619324, + "learning_rate": 2.473095329114164e-07, + "loss": 0.4066, + "step": 1655 + }, + { + "epoch": 5.175365344467641, + "grad_norm": 0.8330480456352234, + "learning_rate": 2.454353134031362e-07, + "loss": 0.4159, + "step": 1656 + }, + { + "epoch": 5.1784968684759916, + "grad_norm": 0.7608641386032104, + "learning_rate": 2.435678560911245e-07, + "loss": 0.4265, + "step": 1657 + }, + { + "epoch": 5.181628392484343, + "grad_norm": 0.8353298306465149, + "learning_rate": 2.417071665765122e-07, + "loss": 0.4457, + "step": 1658 + }, + { + "epoch": 5.184759916492693, + "grad_norm": 0.8479558825492859, + "learning_rate": 2.398532504401324e-07, + "loss": 0.4121, + "step": 1659 + }, + { + "epoch": 5.187891440501044, + "grad_norm": 0.914951503276825, + "learning_rate": 2.3800611324250101e-07, + "loss": 0.3989, + "step": 1660 + }, + { + "epoch": 5.191022964509394, + "grad_norm": 0.7943673133850098, + "learning_rate": 2.3616576052380324e-07, + "loss": 0.4235, + "step": 1661 + }, + { + "epoch": 5.1941544885177455, + "grad_norm": 0.8725146055221558, + "learning_rate": 2.34332197803874e-07, + "loss": 0.4337, + "step": 1662 + }, + { + "epoch": 5.197286012526096, + "grad_norm": 0.7821521162986755, + "learning_rate": 2.325054305821836e-07, + "loss": 0.4089, + "step": 1663 + }, + { + "epoch": 5.200417536534447, + "grad_norm": 0.8576532602310181, + "learning_rate": 2.3068546433782063e-07, + "loss": 0.4292, + "step": 1664 + }, + { + "epoch": 5.203549060542797, + "grad_norm": 0.9175107479095459, + "learning_rate": 2.2887230452947384e-07, + "loss": 0.4333, + "step": 1665 + }, + { + "epoch": 5.206680584551148, + "grad_norm": 0.8123425245285034, + "learning_rate": 2.270659565954189e-07, + "loss": 0.448, + "step": 1666 + }, + { + "epoch": 5.209812108559499, + "grad_norm": 0.7988130450248718, + "learning_rate": 2.2526642595349867e-07, + "loss": 0.3868, + "step": 1667 + }, + { + "epoch": 5.21294363256785, + "grad_norm": 0.7928944230079651, + "learning_rate": 2.234737180011104e-07, + "loss": 0.4048, + "step": 1668 + }, + { + "epoch": 5.2160751565762, + "grad_norm": 0.8115013837814331, + "learning_rate": 2.2168783811518712e-07, + "loss": 0.3962, + "step": 1669 + }, + { + "epoch": 5.219206680584551, + "grad_norm": 0.8003987073898315, + "learning_rate": 2.1990879165218164e-07, + "loss": 0.4315, + "step": 1670 + }, + { + "epoch": 5.222338204592901, + "grad_norm": 0.8416216373443604, + "learning_rate": 2.1813658394805093e-07, + "loss": 0.4344, + "step": 1671 + }, + { + "epoch": 5.225469728601253, + "grad_norm": 0.8418934941291809, + "learning_rate": 2.1637122031824137e-07, + "loss": 0.4099, + "step": 1672 + }, + { + "epoch": 5.228601252609604, + "grad_norm": 0.907768726348877, + "learning_rate": 2.146127060576711e-07, + "loss": 0.4316, + "step": 1673 + }, + { + "epoch": 5.231732776617954, + "grad_norm": 0.8959410786628723, + "learning_rate": 2.1286104644071548e-07, + "loss": 0.4037, + "step": 1674 + }, + { + "epoch": 5.234864300626305, + "grad_norm": 0.9213347434997559, + "learning_rate": 2.1111624672118774e-07, + "loss": 0.4207, + "step": 1675 + }, + { + "epoch": 5.237995824634655, + "grad_norm": 0.8055871725082397, + "learning_rate": 2.0937831213232894e-07, + "loss": 0.4432, + "step": 1676 + }, + { + "epoch": 5.241127348643007, + "grad_norm": 1.0330266952514648, + "learning_rate": 2.076472478867883e-07, + "loss": 0.4398, + "step": 1677 + }, + { + "epoch": 5.244258872651357, + "grad_norm": 0.8400726914405823, + "learning_rate": 2.0592305917660898e-07, + "loss": 0.4617, + "step": 1678 + }, + { + "epoch": 5.247390396659708, + "grad_norm": 0.8515186905860901, + "learning_rate": 2.0420575117321124e-07, + "loss": 0.4366, + "step": 1679 + }, + { + "epoch": 5.250521920668058, + "grad_norm": 0.7332753539085388, + "learning_rate": 2.0249532902737734e-07, + "loss": 0.4142, + "step": 1680 + }, + { + "epoch": 5.253653444676409, + "grad_norm": 0.7820742130279541, + "learning_rate": 2.0079179786923862e-07, + "loss": 0.4278, + "step": 1681 + }, + { + "epoch": 5.25678496868476, + "grad_norm": 0.7951577305793762, + "learning_rate": 1.9909516280825687e-07, + "loss": 0.4393, + "step": 1682 + }, + { + "epoch": 5.259916492693111, + "grad_norm": 0.9419519305229187, + "learning_rate": 1.9740542893320892e-07, + "loss": 0.43, + "step": 1683 + }, + { + "epoch": 5.263048016701461, + "grad_norm": 0.837685227394104, + "learning_rate": 1.9572260131217547e-07, + "loss": 0.3954, + "step": 1684 + }, + { + "epoch": 5.266179540709812, + "grad_norm": 2.381927013397217, + "learning_rate": 1.940466849925199e-07, + "loss": 0.4168, + "step": 1685 + }, + { + "epoch": 5.2693110647181625, + "grad_norm": 0.8350804448127747, + "learning_rate": 1.923776850008785e-07, + "loss": 0.4456, + "step": 1686 + }, + { + "epoch": 5.272442588726514, + "grad_norm": 0.835779070854187, + "learning_rate": 1.9071560634314313e-07, + "loss": 0.4207, + "step": 1687 + }, + { + "epoch": 5.275574112734864, + "grad_norm": 0.794666588306427, + "learning_rate": 1.8906045400444412e-07, + "loss": 0.4318, + "step": 1688 + }, + { + "epoch": 5.278705636743215, + "grad_norm": 0.780330240726471, + "learning_rate": 1.874122329491404e-07, + "loss": 0.4024, + "step": 1689 + }, + { + "epoch": 5.281837160751566, + "grad_norm": 0.8204572796821594, + "learning_rate": 1.8577094812079894e-07, + "loss": 0.431, + "step": 1690 + }, + { + "epoch": 5.284968684759916, + "grad_norm": 0.8642128705978394, + "learning_rate": 1.8413660444218468e-07, + "loss": 0.4083, + "step": 1691 + }, + { + "epoch": 5.288100208768268, + "grad_norm": 0.818459689617157, + "learning_rate": 1.825092068152423e-07, + "loss": 0.4101, + "step": 1692 + }, + { + "epoch": 5.291231732776618, + "grad_norm": 0.8250467777252197, + "learning_rate": 1.8088876012108376e-07, + "loss": 0.4261, + "step": 1693 + }, + { + "epoch": 5.294363256784969, + "grad_norm": 1.001344919204712, + "learning_rate": 1.7927526921997312e-07, + "loss": 0.4107, + "step": 1694 + }, + { + "epoch": 5.297494780793319, + "grad_norm": 0.8144769668579102, + "learning_rate": 1.776687389513107e-07, + "loss": 0.42, + "step": 1695 + }, + { + "epoch": 5.30062630480167, + "grad_norm": 0.8407689332962036, + "learning_rate": 1.7606917413361956e-07, + "loss": 0.4483, + "step": 1696 + }, + { + "epoch": 5.303757828810021, + "grad_norm": 0.8678401112556458, + "learning_rate": 1.7447657956453228e-07, + "loss": 0.4815, + "step": 1697 + }, + { + "epoch": 5.306889352818372, + "grad_norm": 0.7716734409332275, + "learning_rate": 1.7289096002077398e-07, + "loss": 0.4267, + "step": 1698 + }, + { + "epoch": 5.310020876826722, + "grad_norm": 0.9155471920967102, + "learning_rate": 1.713123202581507e-07, + "loss": 0.4544, + "step": 1699 + }, + { + "epoch": 5.313152400835073, + "grad_norm": 0.7895417213439941, + "learning_rate": 1.6974066501153268e-07, + "loss": 0.4166, + "step": 1700 + }, + { + "epoch": 5.3162839248434235, + "grad_norm": 0.8002907037734985, + "learning_rate": 1.681759989948409e-07, + "loss": 0.43, + "step": 1701 + }, + { + "epoch": 5.319415448851775, + "grad_norm": 0.8847978115081787, + "learning_rate": 1.6661832690103435e-07, + "loss": 0.4231, + "step": 1702 + }, + { + "epoch": 5.322546972860125, + "grad_norm": 0.8419442176818848, + "learning_rate": 1.650676534020948e-07, + "loss": 0.4103, + "step": 1703 + }, + { + "epoch": 5.325678496868476, + "grad_norm": 0.8390375375747681, + "learning_rate": 1.6352398314901236e-07, + "loss": 0.424, + "step": 1704 + }, + { + "epoch": 5.328810020876826, + "grad_norm": 0.8126906156539917, + "learning_rate": 1.6198732077177203e-07, + "loss": 0.397, + "step": 1705 + }, + { + "epoch": 5.3319415448851775, + "grad_norm": 0.7679522633552551, + "learning_rate": 1.6045767087934012e-07, + "loss": 0.4669, + "step": 1706 + }, + { + "epoch": 5.335073068893529, + "grad_norm": 0.8407344818115234, + "learning_rate": 1.589350380596505e-07, + "loss": 0.4046, + "step": 1707 + }, + { + "epoch": 5.338204592901879, + "grad_norm": 0.7537122368812561, + "learning_rate": 1.5741942687959006e-07, + "loss": 0.4431, + "step": 1708 + }, + { + "epoch": 5.34133611691023, + "grad_norm": 0.8342767953872681, + "learning_rate": 1.559108418849853e-07, + "loss": 0.4134, + "step": 1709 + }, + { + "epoch": 5.34446764091858, + "grad_norm": 0.8210666179656982, + "learning_rate": 1.5440928760058842e-07, + "loss": 0.4029, + "step": 1710 + }, + { + "epoch": 5.347599164926931, + "grad_norm": 0.7971834540367126, + "learning_rate": 1.5291476853006527e-07, + "loss": 0.42, + "step": 1711 + }, + { + "epoch": 5.350730688935282, + "grad_norm": 0.7891254425048828, + "learning_rate": 1.5142728915598065e-07, + "loss": 0.4341, + "step": 1712 + }, + { + "epoch": 5.353862212943633, + "grad_norm": 0.8925033211708069, + "learning_rate": 1.499468539397836e-07, + "loss": 0.4396, + "step": 1713 + }, + { + "epoch": 5.356993736951983, + "grad_norm": 0.8453183174133301, + "learning_rate": 1.4847346732179717e-07, + "loss": 0.4176, + "step": 1714 + }, + { + "epoch": 5.360125260960334, + "grad_norm": 0.8093660473823547, + "learning_rate": 1.4700713372120172e-07, + "loss": 0.3894, + "step": 1715 + }, + { + "epoch": 5.3632567849686845, + "grad_norm": 0.7943024635314941, + "learning_rate": 1.455478575360239e-07, + "loss": 0.3693, + "step": 1716 + }, + { + "epoch": 5.366388308977036, + "grad_norm": 0.8238469958305359, + "learning_rate": 1.4409564314312313e-07, + "loss": 0.4362, + "step": 1717 + }, + { + "epoch": 5.369519832985386, + "grad_norm": 0.943553626537323, + "learning_rate": 1.4265049489817774e-07, + "loss": 0.4461, + "step": 1718 + }, + { + "epoch": 5.372651356993737, + "grad_norm": 1.0485329627990723, + "learning_rate": 1.412124171356727e-07, + "loss": 0.4278, + "step": 1719 + }, + { + "epoch": 5.375782881002087, + "grad_norm": 0.807830810546875, + "learning_rate": 1.3978141416888497e-07, + "loss": 0.406, + "step": 1720 + }, + { + "epoch": 5.3789144050104385, + "grad_norm": 0.8568108081817627, + "learning_rate": 1.3835749028987405e-07, + "loss": 0.449, + "step": 1721 + }, + { + "epoch": 5.382045929018789, + "grad_norm": 0.7883656024932861, + "learning_rate": 1.3694064976946448e-07, + "loss": 0.4199, + "step": 1722 + }, + { + "epoch": 5.38517745302714, + "grad_norm": 0.7353293299674988, + "learning_rate": 1.355308968572372e-07, + "loss": 0.4175, + "step": 1723 + }, + { + "epoch": 5.388308977035491, + "grad_norm": 0.7996500134468079, + "learning_rate": 1.3412823578151434e-07, + "loss": 0.4159, + "step": 1724 + }, + { + "epoch": 5.391440501043841, + "grad_norm": 0.7814679145812988, + "learning_rate": 1.327326707493473e-07, + "loss": 0.4024, + "step": 1725 + }, + { + "epoch": 5.3945720250521925, + "grad_norm": 0.7922082543373108, + "learning_rate": 1.3134420594650364e-07, + "loss": 0.4466, + "step": 1726 + }, + { + "epoch": 5.397703549060543, + "grad_norm": 1.122597098350525, + "learning_rate": 1.2996284553745575e-07, + "loss": 0.4318, + "step": 1727 + }, + { + "epoch": 5.400835073068894, + "grad_norm": 0.7850015163421631, + "learning_rate": 1.285885936653672e-07, + "loss": 0.4284, + "step": 1728 + }, + { + "epoch": 5.403966597077244, + "grad_norm": 0.8766868710517883, + "learning_rate": 1.2722145445208057e-07, + "loss": 0.4119, + "step": 1729 + }, + { + "epoch": 5.407098121085595, + "grad_norm": 0.8224661350250244, + "learning_rate": 1.2586143199810553e-07, + "loss": 0.438, + "step": 1730 + }, + { + "epoch": 5.4102296450939455, + "grad_norm": 0.9190678000450134, + "learning_rate": 1.245085303826049e-07, + "loss": 0.4196, + "step": 1731 + }, + { + "epoch": 5.413361169102297, + "grad_norm": 0.8007112145423889, + "learning_rate": 1.2316275366338586e-07, + "loss": 0.428, + "step": 1732 + }, + { + "epoch": 5.416492693110647, + "grad_norm": 0.8234992623329163, + "learning_rate": 1.2182410587688403e-07, + "loss": 0.4542, + "step": 1733 + }, + { + "epoch": 5.419624217118998, + "grad_norm": 0.8217283487319946, + "learning_rate": 1.2049259103815353e-07, + "loss": 0.4475, + "step": 1734 + }, + { + "epoch": 5.422755741127348, + "grad_norm": 0.8184598684310913, + "learning_rate": 1.1916821314085448e-07, + "loss": 0.4142, + "step": 1735 + }, + { + "epoch": 5.4258872651356995, + "grad_norm": 0.7703188061714172, + "learning_rate": 1.178509761572405e-07, + "loss": 0.4012, + "step": 1736 + }, + { + "epoch": 5.42901878914405, + "grad_norm": 0.7523705959320068, + "learning_rate": 1.1654088403814712e-07, + "loss": 0.3856, + "step": 1737 + }, + { + "epoch": 5.432150313152401, + "grad_norm": 0.8626607656478882, + "learning_rate": 1.1523794071298139e-07, + "loss": 0.3987, + "step": 1738 + }, + { + "epoch": 5.435281837160751, + "grad_norm": 0.7535544037818909, + "learning_rate": 1.1394215008970672e-07, + "loss": 0.4301, + "step": 1739 + }, + { + "epoch": 5.438413361169102, + "grad_norm": 0.7843238711357117, + "learning_rate": 1.1265351605483477e-07, + "loss": 0.3988, + "step": 1740 + }, + { + "epoch": 5.4415448851774535, + "grad_norm": 0.750119149684906, + "learning_rate": 1.1137204247341077e-07, + "loss": 0.394, + "step": 1741 + }, + { + "epoch": 5.444676409185804, + "grad_norm": 0.7726014852523804, + "learning_rate": 1.1009773318900518e-07, + "loss": 0.4411, + "step": 1742 + }, + { + "epoch": 5.447807933194154, + "grad_norm": 0.8338330984115601, + "learning_rate": 1.0883059202369816e-07, + "loss": 0.4115, + "step": 1743 + }, + { + "epoch": 5.450939457202505, + "grad_norm": 0.8153620958328247, + "learning_rate": 1.0757062277807206e-07, + "loss": 0.4479, + "step": 1744 + }, + { + "epoch": 5.454070981210856, + "grad_norm": 0.8071407079696655, + "learning_rate": 1.0631782923119782e-07, + "loss": 0.4279, + "step": 1745 + }, + { + "epoch": 5.457202505219207, + "grad_norm": 0.9648112654685974, + "learning_rate": 1.050722151406236e-07, + "loss": 0.4216, + "step": 1746 + }, + { + "epoch": 5.460334029227558, + "grad_norm": 0.8332387804985046, + "learning_rate": 1.038337842423634e-07, + "loss": 0.4021, + "step": 1747 + }, + { + "epoch": 5.463465553235908, + "grad_norm": 0.747900128364563, + "learning_rate": 1.0260254025088818e-07, + "loss": 0.4268, + "step": 1748 + }, + { + "epoch": 5.466597077244259, + "grad_norm": 0.7799843549728394, + "learning_rate": 1.013784868591114e-07, + "loss": 0.4241, + "step": 1749 + }, + { + "epoch": 5.469728601252609, + "grad_norm": 0.8218986988067627, + "learning_rate": 1.0016162773838045e-07, + "loss": 0.4348, + "step": 1750 + }, + { + "epoch": 5.4728601252609606, + "grad_norm": 0.8404920101165771, + "learning_rate": 9.895196653846417e-08, + "loss": 0.4649, + "step": 1751 + }, + { + "epoch": 5.475991649269311, + "grad_norm": 0.8962766528129578, + "learning_rate": 9.774950688754197e-08, + "loss": 0.4226, + "step": 1752 + }, + { + "epoch": 5.479123173277662, + "grad_norm": 0.7923446893692017, + "learning_rate": 9.655425239219446e-08, + "loss": 0.4602, + "step": 1753 + }, + { + "epoch": 5.482254697286012, + "grad_norm": 0.8076177835464478, + "learning_rate": 9.536620663739143e-08, + "loss": 0.4244, + "step": 1754 + }, + { + "epoch": 5.485386221294363, + "grad_norm": 0.8613052368164062, + "learning_rate": 9.418537318648057e-08, + "loss": 0.4146, + "step": 1755 + }, + { + "epoch": 5.488517745302714, + "grad_norm": 0.7825137376785278, + "learning_rate": 9.301175558117743e-08, + "loss": 0.4217, + "step": 1756 + }, + { + "epoch": 5.491649269311065, + "grad_norm": 0.9981064796447754, + "learning_rate": 9.184535734155509e-08, + "loss": 0.4339, + "step": 1757 + }, + { + "epoch": 5.494780793319415, + "grad_norm": 1.1833750009536743, + "learning_rate": 9.068618196603401e-08, + "loss": 0.4064, + "step": 1758 + }, + { + "epoch": 5.497912317327766, + "grad_norm": 0.7761838436126709, + "learning_rate": 8.953423293137004e-08, + "loss": 0.4289, + "step": 1759 + }, + { + "epoch": 5.5010438413361165, + "grad_norm": 0.8973391652107239, + "learning_rate": 8.838951369264492e-08, + "loss": 0.4289, + "step": 1760 + }, + { + "epoch": 5.504175365344468, + "grad_norm": 0.7715950608253479, + "learning_rate": 8.725202768325558e-08, + "loss": 0.4468, + "step": 1761 + }, + { + "epoch": 5.507306889352819, + "grad_norm": 0.7941539883613586, + "learning_rate": 8.612177831490459e-08, + "loss": 0.4296, + "step": 1762 + }, + { + "epoch": 5.510438413361169, + "grad_norm": 0.7699092030525208, + "learning_rate": 8.49987689775894e-08, + "loss": 0.4121, + "step": 1763 + }, + { + "epoch": 5.51356993736952, + "grad_norm": 1.026188850402832, + "learning_rate": 8.38830030395918e-08, + "loss": 0.4293, + "step": 1764 + }, + { + "epoch": 5.51670146137787, + "grad_norm": 0.7858873605728149, + "learning_rate": 8.277448384746839e-08, + "loss": 0.431, + "step": 1765 + }, + { + "epoch": 5.519832985386222, + "grad_norm": 0.7994810342788696, + "learning_rate": 8.167321472604017e-08, + "loss": 0.4263, + "step": 1766 + }, + { + "epoch": 5.522964509394572, + "grad_norm": 0.7905458807945251, + "learning_rate": 8.057919897838301e-08, + "loss": 0.4203, + "step": 1767 + }, + { + "epoch": 5.526096033402923, + "grad_norm": 0.7660766839981079, + "learning_rate": 7.949243988581767e-08, + "loss": 0.4495, + "step": 1768 + }, + { + "epoch": 5.529227557411273, + "grad_norm": 0.8373368382453918, + "learning_rate": 7.841294070789873e-08, + "loss": 0.4132, + "step": 1769 + }, + { + "epoch": 5.532359081419624, + "grad_norm": 0.8575682044029236, + "learning_rate": 7.734070468240735e-08, + "loss": 0.4139, + "step": 1770 + }, + { + "epoch": 5.535490605427975, + "grad_norm": 0.8791792392730713, + "learning_rate": 7.627573502533847e-08, + "loss": 0.4256, + "step": 1771 + }, + { + "epoch": 5.538622129436326, + "grad_norm": 0.7887570261955261, + "learning_rate": 7.521803493089453e-08, + "loss": 0.4227, + "step": 1772 + }, + { + "epoch": 5.541753653444676, + "grad_norm": 0.7709775567054749, + "learning_rate": 7.41676075714723e-08, + "loss": 0.4127, + "step": 1773 + }, + { + "epoch": 5.544885177453027, + "grad_norm": 0.9252105355262756, + "learning_rate": 7.312445609765628e-08, + "loss": 0.4128, + "step": 1774 + }, + { + "epoch": 5.5480167014613775, + "grad_norm": 0.8159840703010559, + "learning_rate": 7.20885836382082e-08, + "loss": 0.4344, + "step": 1775 + }, + { + "epoch": 5.551148225469729, + "grad_norm": 0.8233450651168823, + "learning_rate": 7.105999330005691e-08, + "loss": 0.4146, + "step": 1776 + }, + { + "epoch": 5.554279749478079, + "grad_norm": 0.8269393444061279, + "learning_rate": 7.00386881682899e-08, + "loss": 0.4379, + "step": 1777 + }, + { + "epoch": 5.55741127348643, + "grad_norm": 0.8312438726425171, + "learning_rate": 6.90246713061446e-08, + "loss": 0.441, + "step": 1778 + }, + { + "epoch": 5.560542797494781, + "grad_norm": 0.7896298766136169, + "learning_rate": 6.80179457549976e-08, + "loss": 0.4102, + "step": 1779 + }, + { + "epoch": 5.5636743215031315, + "grad_norm": 0.942116379737854, + "learning_rate": 6.701851453435715e-08, + "loss": 0.4141, + "step": 1780 + }, + { + "epoch": 5.566805845511483, + "grad_norm": 0.8050079941749573, + "learning_rate": 6.602638064185262e-08, + "loss": 0.4087, + "step": 1781 + }, + { + "epoch": 5.569937369519833, + "grad_norm": 0.7318202257156372, + "learning_rate": 6.504154705322646e-08, + "loss": 0.4179, + "step": 1782 + }, + { + "epoch": 5.573068893528184, + "grad_norm": 0.8148002624511719, + "learning_rate": 6.406401672232527e-08, + "loss": 0.4439, + "step": 1783 + }, + { + "epoch": 5.576200417536534, + "grad_norm": 0.8027870655059814, + "learning_rate": 6.309379258109093e-08, + "loss": 0.4291, + "step": 1784 + }, + { + "epoch": 5.579331941544885, + "grad_norm": 0.8059787154197693, + "learning_rate": 6.213087753955127e-08, + "loss": 0.4184, + "step": 1785 + }, + { + "epoch": 5.582463465553236, + "grad_norm": 0.8198132514953613, + "learning_rate": 6.117527448581101e-08, + "loss": 0.4197, + "step": 1786 + }, + { + "epoch": 5.585594989561587, + "grad_norm": 0.8053491115570068, + "learning_rate": 6.022698628604523e-08, + "loss": 0.4202, + "step": 1787 + }, + { + "epoch": 5.588726513569937, + "grad_norm": 0.7784984707832336, + "learning_rate": 5.928601578448823e-08, + "loss": 0.4169, + "step": 1788 + }, + { + "epoch": 5.591858037578288, + "grad_norm": 0.797520101070404, + "learning_rate": 5.835236580342657e-08, + "loss": 0.3941, + "step": 1789 + }, + { + "epoch": 5.5949895615866385, + "grad_norm": 0.8359535932540894, + "learning_rate": 5.742603914318995e-08, + "loss": 0.4393, + "step": 1790 + }, + { + "epoch": 5.59812108559499, + "grad_norm": 0.7943177223205566, + "learning_rate": 5.6507038582142574e-08, + "loss": 0.3962, + "step": 1791 + }, + { + "epoch": 5.60125260960334, + "grad_norm": 0.7825118899345398, + "learning_rate": 5.559536687667566e-08, + "loss": 0.4247, + "step": 1792 + }, + { + "epoch": 5.604384133611691, + "grad_norm": 0.771355390548706, + "learning_rate": 5.469102676119942e-08, + "loss": 0.4542, + "step": 1793 + }, + { + "epoch": 5.607515657620041, + "grad_norm": 0.8478569984436035, + "learning_rate": 5.379402094813246e-08, + "loss": 0.4225, + "step": 1794 + }, + { + "epoch": 5.6106471816283925, + "grad_norm": 1.0207805633544922, + "learning_rate": 5.2904352127896855e-08, + "loss": 0.433, + "step": 1795 + }, + { + "epoch": 5.613778705636744, + "grad_norm": 0.868712842464447, + "learning_rate": 5.2022022968907794e-08, + "loss": 0.418, + "step": 1796 + }, + { + "epoch": 5.616910229645094, + "grad_norm": 0.8502968549728394, + "learning_rate": 5.114703611756672e-08, + "loss": 0.3906, + "step": 1797 + }, + { + "epoch": 5.620041753653445, + "grad_norm": 0.8621317148208618, + "learning_rate": 5.027939419825295e-08, + "loss": 0.4502, + "step": 1798 + }, + { + "epoch": 5.623173277661795, + "grad_norm": 0.7986813187599182, + "learning_rate": 4.941909981331539e-08, + "loss": 0.4507, + "step": 1799 + }, + { + "epoch": 5.6263048016701465, + "grad_norm": 0.7599334716796875, + "learning_rate": 4.856615554306637e-08, + "loss": 0.4249, + "step": 1800 + }, + { + "epoch": 5.629436325678497, + "grad_norm": 0.854983389377594, + "learning_rate": 4.7720563945771456e-08, + "loss": 0.3809, + "step": 1801 + }, + { + "epoch": 5.632567849686848, + "grad_norm": 0.7154690027236938, + "learning_rate": 4.688232755764438e-08, + "loss": 0.4397, + "step": 1802 + }, + { + "epoch": 5.635699373695198, + "grad_norm": 0.7829597592353821, + "learning_rate": 4.605144889283652e-08, + "loss": 0.4104, + "step": 1803 + }, + { + "epoch": 5.638830897703549, + "grad_norm": 0.8672685623168945, + "learning_rate": 4.5227930443432485e-08, + "loss": 0.397, + "step": 1804 + }, + { + "epoch": 5.6419624217118995, + "grad_norm": 0.8553320169448853, + "learning_rate": 4.441177467944035e-08, + "loss": 0.447, + "step": 1805 + }, + { + "epoch": 5.645093945720251, + "grad_norm": 0.8556696176528931, + "learning_rate": 4.3602984048785324e-08, + "loss": 0.4379, + "step": 1806 + }, + { + "epoch": 5.648225469728601, + "grad_norm": 0.8407424688339233, + "learning_rate": 4.280156097730192e-08, + "loss": 0.4257, + "step": 1807 + }, + { + "epoch": 5.651356993736952, + "grad_norm": 0.8491048812866211, + "learning_rate": 4.200750786872682e-08, + "loss": 0.4181, + "step": 1808 + }, + { + "epoch": 5.654488517745302, + "grad_norm": 0.8149058222770691, + "learning_rate": 4.1220827104691826e-08, + "loss": 0.4265, + "step": 1809 + }, + { + "epoch": 5.6576200417536535, + "grad_norm": 0.7961820960044861, + "learning_rate": 4.044152104471705e-08, + "loss": 0.3905, + "step": 1810 + }, + { + "epoch": 5.660751565762004, + "grad_norm": 0.781095564365387, + "learning_rate": 3.966959202620274e-08, + "loss": 0.4311, + "step": 1811 + }, + { + "epoch": 5.663883089770355, + "grad_norm": 0.8125353455543518, + "learning_rate": 3.890504236442272e-08, + "loss": 0.4112, + "step": 1812 + }, + { + "epoch": 5.667014613778706, + "grad_norm": 0.7369498610496521, + "learning_rate": 3.814787435251821e-08, + "loss": 0.3749, + "step": 1813 + }, + { + "epoch": 5.670146137787056, + "grad_norm": 0.7556062936782837, + "learning_rate": 3.739809026149066e-08, + "loss": 0.4301, + "step": 1814 + }, + { + "epoch": 5.673277661795407, + "grad_norm": 0.7914175391197205, + "learning_rate": 3.665569234019367e-08, + "loss": 0.403, + "step": 1815 + }, + { + "epoch": 5.676409185803758, + "grad_norm": 0.8738980293273926, + "learning_rate": 3.592068281532857e-08, + "loss": 0.4194, + "step": 1816 + }, + { + "epoch": 5.679540709812109, + "grad_norm": 0.7967526912689209, + "learning_rate": 3.519306389143468e-08, + "loss": 0.4199, + "step": 1817 + }, + { + "epoch": 5.682672233820459, + "grad_norm": 0.8052618503570557, + "learning_rate": 3.4472837750885726e-08, + "loss": 0.4301, + "step": 1818 + }, + { + "epoch": 5.68580375782881, + "grad_norm": 1.0626697540283203, + "learning_rate": 3.376000655388234e-08, + "loss": 0.4073, + "step": 1819 + }, + { + "epoch": 5.688935281837161, + "grad_norm": 0.8473744988441467, + "learning_rate": 3.3054572438443454e-08, + "loss": 0.4323, + "step": 1820 + }, + { + "epoch": 5.692066805845512, + "grad_norm": 0.785251796245575, + "learning_rate": 3.23565375204038e-08, + "loss": 0.4214, + "step": 1821 + }, + { + "epoch": 5.695198329853862, + "grad_norm": 0.8004971742630005, + "learning_rate": 3.1665903893403626e-08, + "loss": 0.4119, + "step": 1822 + }, + { + "epoch": 5.698329853862213, + "grad_norm": 0.9184899926185608, + "learning_rate": 3.098267362888541e-08, + "loss": 0.4267, + "step": 1823 + }, + { + "epoch": 5.701461377870563, + "grad_norm": 0.807667076587677, + "learning_rate": 3.030684877608575e-08, + "loss": 0.4301, + "step": 1824 + }, + { + "epoch": 5.7045929018789145, + "grad_norm": 0.7712723612785339, + "learning_rate": 2.963843136203043e-08, + "loss": 0.4182, + "step": 1825 + }, + { + "epoch": 5.707724425887265, + "grad_norm": 0.734040379524231, + "learning_rate": 2.8977423391527703e-08, + "loss": 0.4443, + "step": 1826 + }, + { + "epoch": 5.710855949895616, + "grad_norm": 1.1767381429672241, + "learning_rate": 2.8323826847162507e-08, + "loss": 0.4125, + "step": 1827 + }, + { + "epoch": 5.713987473903966, + "grad_norm": 0.8537562489509583, + "learning_rate": 2.7677643689289492e-08, + "loss": 0.4221, + "step": 1828 + }, + { + "epoch": 5.717118997912317, + "grad_norm": 0.7683084011077881, + "learning_rate": 2.7038875856029712e-08, + "loss": 0.4167, + "step": 1829 + }, + { + "epoch": 5.7202505219206685, + "grad_norm": 0.8932932615280151, + "learning_rate": 2.6407525263261736e-08, + "loss": 0.4277, + "step": 1830 + }, + { + "epoch": 5.723382045929019, + "grad_norm": 0.7592349052429199, + "learning_rate": 2.5783593804618312e-08, + "loss": 0.3938, + "step": 1831 + }, + { + "epoch": 5.726513569937369, + "grad_norm": 0.7657082676887512, + "learning_rate": 2.516708335147916e-08, + "loss": 0.3895, + "step": 1832 + }, + { + "epoch": 5.72964509394572, + "grad_norm": 0.8797202110290527, + "learning_rate": 2.4557995752965967e-08, + "loss": 0.4119, + "step": 1833 + }, + { + "epoch": 5.732776617954071, + "grad_norm": 0.7941285967826843, + "learning_rate": 2.3956332835937123e-08, + "loss": 0.3737, + "step": 1834 + }, + { + "epoch": 5.735908141962422, + "grad_norm": 0.7556982040405273, + "learning_rate": 2.3362096404981605e-08, + "loss": 0.42, + "step": 1835 + }, + { + "epoch": 5.739039665970773, + "grad_norm": 0.8625621795654297, + "learning_rate": 2.2775288242413705e-08, + "loss": 0.4112, + "step": 1836 + }, + { + "epoch": 5.742171189979123, + "grad_norm": 0.8050491809844971, + "learning_rate": 2.219591010826805e-08, + "loss": 0.4222, + "step": 1837 + }, + { + "epoch": 5.745302713987474, + "grad_norm": 0.7006460428237915, + "learning_rate": 2.1623963740294574e-08, + "loss": 0.4187, + "step": 1838 + }, + { + "epoch": 5.748434237995824, + "grad_norm": 0.8389616012573242, + "learning_rate": 2.105945085395189e-08, + "loss": 0.4258, + "step": 1839 + }, + { + "epoch": 5.751565762004176, + "grad_norm": 0.8111906051635742, + "learning_rate": 2.050237314240422e-08, + "loss": 0.4078, + "step": 1840 + }, + { + "epoch": 5.754697286012526, + "grad_norm": 0.7622024416923523, + "learning_rate": 1.9952732276514174e-08, + "loss": 0.448, + "step": 1841 + }, + { + "epoch": 5.757828810020877, + "grad_norm": 0.8661556243896484, + "learning_rate": 1.941052990483916e-08, + "loss": 0.4058, + "step": 1842 + }, + { + "epoch": 5.760960334029227, + "grad_norm": 0.8311405181884766, + "learning_rate": 1.8875767653626366e-08, + "loss": 0.4126, + "step": 1843 + }, + { + "epoch": 5.764091858037578, + "grad_norm": 0.90278559923172, + "learning_rate": 1.834844712680722e-08, + "loss": 0.4541, + "step": 1844 + }, + { + "epoch": 5.767223382045929, + "grad_norm": 0.854500949382782, + "learning_rate": 1.782856990599269e-08, + "loss": 0.4233, + "step": 1845 + }, + { + "epoch": 5.77035490605428, + "grad_norm": 1.0461066961288452, + "learning_rate": 1.7316137550469625e-08, + "loss": 0.4011, + "step": 1846 + }, + { + "epoch": 5.773486430062631, + "grad_norm": 0.7584487199783325, + "learning_rate": 1.6811151597194153e-08, + "loss": 0.4156, + "step": 1847 + }, + { + "epoch": 5.776617954070981, + "grad_norm": 0.8103524446487427, + "learning_rate": 1.6313613560788865e-08, + "loss": 0.437, + "step": 1848 + }, + { + "epoch": 5.7797494780793315, + "grad_norm": 0.7988206744194031, + "learning_rate": 1.5823524933537283e-08, + "loss": 0.4259, + "step": 1849 + }, + { + "epoch": 5.782881002087683, + "grad_norm": 0.788215696811676, + "learning_rate": 1.5340887185379684e-08, + "loss": 0.4377, + "step": 1850 + }, + { + "epoch": 5.786012526096034, + "grad_norm": 0.835552990436554, + "learning_rate": 1.4865701763908947e-08, + "loss": 0.4378, + "step": 1851 + }, + { + "epoch": 5.789144050104384, + "grad_norm": 0.8064237236976624, + "learning_rate": 1.4397970094364999e-08, + "loss": 0.4396, + "step": 1852 + }, + { + "epoch": 5.792275574112735, + "grad_norm": 0.8313791155815125, + "learning_rate": 1.3937693579632872e-08, + "loss": 0.4411, + "step": 1853 + }, + { + "epoch": 5.7954070981210855, + "grad_norm": 0.8237503170967102, + "learning_rate": 1.3484873600236037e-08, + "loss": 0.4385, + "step": 1854 + }, + { + "epoch": 5.798538622129437, + "grad_norm": 0.8513125777244568, + "learning_rate": 1.3039511514333359e-08, + "loss": 0.4298, + "step": 1855 + }, + { + "epoch": 5.801670146137787, + "grad_norm": 0.7637052536010742, + "learning_rate": 1.2601608657715758e-08, + "loss": 0.4116, + "step": 1856 + }, + { + "epoch": 5.804801670146138, + "grad_norm": 0.7973178625106812, + "learning_rate": 1.2171166343800388e-08, + "loss": 0.4333, + "step": 1857 + }, + { + "epoch": 5.807933194154488, + "grad_norm": 0.7717419266700745, + "learning_rate": 1.1748185863628414e-08, + "loss": 0.4138, + "step": 1858 + }, + { + "epoch": 5.811064718162839, + "grad_norm": 0.773685097694397, + "learning_rate": 1.1332668485860287e-08, + "loss": 0.389, + "step": 1859 + }, + { + "epoch": 5.81419624217119, + "grad_norm": 0.8590340614318848, + "learning_rate": 1.092461545677187e-08, + "loss": 0.4393, + "step": 1860 + }, + { + "epoch": 5.817327766179541, + "grad_norm": 0.7766469717025757, + "learning_rate": 1.0524028000251651e-08, + "loss": 0.4546, + "step": 1861 + }, + { + "epoch": 5.820459290187891, + "grad_norm": 0.843730628490448, + "learning_rate": 1.0130907317795757e-08, + "loss": 0.4473, + "step": 1862 + }, + { + "epoch": 5.823590814196242, + "grad_norm": 0.7406162619590759, + "learning_rate": 9.745254588504615e-09, + "loss": 0.4064, + "step": 1863 + }, + { + "epoch": 5.826722338204593, + "grad_norm": 0.8189436793327332, + "learning_rate": 9.36707096908046e-09, + "loss": 0.4095, + "step": 1864 + }, + { + "epoch": 5.829853862212944, + "grad_norm": 0.8491302132606506, + "learning_rate": 8.996357593823445e-09, + "loss": 0.4425, + "step": 1865 + }, + { + "epoch": 5.832985386221294, + "grad_norm": 0.8554711937904358, + "learning_rate": 8.633115574626649e-09, + "loss": 0.43, + "step": 1866 + }, + { + "epoch": 5.836116910229645, + "grad_norm": 0.8352468013763428, + "learning_rate": 8.277346000974961e-09, + "loss": 0.4471, + "step": 1867 + }, + { + "epoch": 5.839248434237996, + "grad_norm": 0.7572605609893799, + "learning_rate": 7.929049939940924e-09, + "loss": 0.3961, + "step": 1868 + }, + { + "epoch": 5.8423799582463465, + "grad_norm": 0.7781088948249817, + "learning_rate": 7.5882284361814e-09, + "loss": 0.4198, + "step": 1869 + }, + { + "epoch": 5.845511482254698, + "grad_norm": 0.780062198638916, + "learning_rate": 7.254882511934236e-09, + "loss": 0.4275, + "step": 1870 + }, + { + "epoch": 5.848643006263048, + "grad_norm": 0.7632731795310974, + "learning_rate": 6.929013167016052e-09, + "loss": 0.448, + "step": 1871 + }, + { + "epoch": 5.851774530271399, + "grad_norm": 0.8311516642570496, + "learning_rate": 6.610621378818349e-09, + "loss": 0.4242, + "step": 1872 + }, + { + "epoch": 5.854906054279749, + "grad_norm": 0.790691077709198, + "learning_rate": 6.299708102304736e-09, + "loss": 0.4362, + "step": 1873 + }, + { + "epoch": 5.8580375782881005, + "grad_norm": 0.8050321340560913, + "learning_rate": 5.996274270008429e-09, + "loss": 0.4145, + "step": 1874 + }, + { + "epoch": 5.861169102296451, + "grad_norm": 0.8381393551826477, + "learning_rate": 5.700320792029757e-09, + "loss": 0.4229, + "step": 1875 + }, + { + "epoch": 5.864300626304802, + "grad_norm": 0.8540353775024414, + "learning_rate": 5.411848556032273e-09, + "loss": 0.4271, + "step": 1876 + }, + { + "epoch": 5.867432150313152, + "grad_norm": 0.7595441341400146, + "learning_rate": 5.1308584272410925e-09, + "loss": 0.4085, + "step": 1877 + }, + { + "epoch": 5.870563674321503, + "grad_norm": 0.8113922476768494, + "learning_rate": 4.857351248440112e-09, + "loss": 0.3848, + "step": 1878 + }, + { + "epoch": 5.8736951983298535, + "grad_norm": 0.8255279660224915, + "learning_rate": 4.591327839968685e-09, + "loss": 0.3989, + "step": 1879 + }, + { + "epoch": 5.876826722338205, + "grad_norm": 0.9661470651626587, + "learning_rate": 4.332788999720505e-09, + "loss": 0.449, + "step": 1880 + }, + { + "epoch": 5.879958246346555, + "grad_norm": 0.8207715153694153, + "learning_rate": 4.0817355031405606e-09, + "loss": 0.4236, + "step": 1881 + }, + { + "epoch": 5.883089770354906, + "grad_norm": 0.8630990982055664, + "learning_rate": 3.838168103222628e-09, + "loss": 0.4362, + "step": 1882 + }, + { + "epoch": 5.886221294363256, + "grad_norm": 0.729789137840271, + "learning_rate": 3.6020875305067816e-09, + "loss": 0.4224, + "step": 1883 + }, + { + "epoch": 5.8893528183716075, + "grad_norm": 0.8223000764846802, + "learning_rate": 3.373494493078e-09, + "loss": 0.4534, + "step": 1884 + }, + { + "epoch": 5.892484342379959, + "grad_norm": 0.8721533417701721, + "learning_rate": 3.1523896765633945e-09, + "loss": 0.4232, + "step": 1885 + }, + { + "epoch": 5.895615866388309, + "grad_norm": 0.9347804188728333, + "learning_rate": 2.9387737441308183e-09, + "loss": 0.4525, + "step": 1886 + }, + { + "epoch": 5.89874739039666, + "grad_norm": 0.8137363195419312, + "learning_rate": 2.732647336486094e-09, + "loss": 0.4049, + "step": 1887 + }, + { + "epoch": 5.90187891440501, + "grad_norm": 0.9206838607788086, + "learning_rate": 2.534011071871345e-09, + "loss": 0.4177, + "step": 1888 + }, + { + "epoch": 5.9050104384133615, + "grad_norm": 0.8816237449645996, + "learning_rate": 2.342865546063611e-09, + "loss": 0.4168, + "step": 1889 + }, + { + "epoch": 5.908141962421712, + "grad_norm": 0.7795571088790894, + "learning_rate": 2.1592113323720686e-09, + "loss": 0.4694, + "step": 1890 + }, + { + "epoch": 5.911273486430063, + "grad_norm": 0.8177521228790283, + "learning_rate": 1.983048981638036e-09, + "loss": 0.4035, + "step": 1891 + }, + { + "epoch": 5.914405010438413, + "grad_norm": 0.9472755789756775, + "learning_rate": 1.8143790222308055e-09, + "loss": 0.4372, + "step": 1892 + }, + { + "epoch": 5.917536534446764, + "grad_norm": 0.7970467805862427, + "learning_rate": 1.653201960048756e-09, + "loss": 0.4002, + "step": 1893 + }, + { + "epoch": 5.920668058455115, + "grad_norm": 0.811830461025238, + "learning_rate": 1.4995182785157437e-09, + "loss": 0.4343, + "step": 1894 + }, + { + "epoch": 5.923799582463466, + "grad_norm": 0.8110922574996948, + "learning_rate": 1.353328438580548e-09, + "loss": 0.4022, + "step": 1895 + }, + { + "epoch": 5.926931106471816, + "grad_norm": 0.7824969291687012, + "learning_rate": 1.2146328787160378e-09, + "loss": 0.4078, + "step": 1896 + }, + { + "epoch": 5.930062630480167, + "grad_norm": 0.8015698790550232, + "learning_rate": 1.0834320149163969e-09, + "loss": 0.4223, + "step": 1897 + }, + { + "epoch": 5.933194154488517, + "grad_norm": 1.0146117210388184, + "learning_rate": 9.59726240697123e-10, + "loss": 0.4295, + "step": 1898 + }, + { + "epoch": 5.9363256784968685, + "grad_norm": 0.8403705358505249, + "learning_rate": 8.435159270933635e-10, + "loss": 0.4097, + "step": 1899 + }, + { + "epoch": 5.939457202505219, + "grad_norm": 0.8461981415748596, + "learning_rate": 7.348014226588041e-10, + "loss": 0.4141, + "step": 1900 + }, + { + "epoch": 5.94258872651357, + "grad_norm": 0.750941812992096, + "learning_rate": 6.335830534648368e-10, + "loss": 0.4372, + "step": 1901 + }, + { + "epoch": 5.945720250521921, + "grad_norm": 0.7628970146179199, + "learning_rate": 5.3986112309945e-10, + "loss": 0.4195, + "step": 1902 + }, + { + "epoch": 5.948851774530271, + "grad_norm": 0.7590934038162231, + "learning_rate": 4.536359126658396e-10, + "loss": 0.4095, + "step": 1903 + }, + { + "epoch": 5.9519832985386225, + "grad_norm": 0.8931977152824402, + "learning_rate": 3.749076807826879e-10, + "loss": 0.4644, + "step": 1904 + }, + { + "epoch": 5.955114822546973, + "grad_norm": 1.068721055984497, + "learning_rate": 3.036766635822197e-10, + "loss": 0.4522, + "step": 1905 + }, + { + "epoch": 5.958246346555324, + "grad_norm": 0.7380474805831909, + "learning_rate": 2.3994307471020275e-10, + "loss": 0.4296, + "step": 1906 + }, + { + "epoch": 5.961377870563674, + "grad_norm": 0.765549898147583, + "learning_rate": 1.8370710532483738e-10, + "loss": 0.4462, + "step": 1907 + }, + { + "epoch": 5.964509394572025, + "grad_norm": 0.7848723530769348, + "learning_rate": 1.349689240970342e-10, + "loss": 0.4282, + "step": 1908 + }, + { + "epoch": 5.967640918580376, + "grad_norm": 0.8109093308448792, + "learning_rate": 9.372867720874867e-11, + "loss": 0.4064, + "step": 1909 + }, + { + "epoch": 5.970772442588727, + "grad_norm": 0.8530039191246033, + "learning_rate": 5.99864883532586e-11, + "loss": 0.437, + "step": 1910 + }, + { + "epoch": 5.973903966597077, + "grad_norm": 0.7433366179466248, + "learning_rate": 3.3742458734886684e-11, + "loss": 0.4156, + "step": 1911 + }, + { + "epoch": 5.977035490605428, + "grad_norm": 0.8591077923774719, + "learning_rate": 1.499666706844538e-11, + "loss": 0.4239, + "step": 1912 + }, + { + "epoch": 5.980167014613778, + "grad_norm": 0.8396080732345581, + "learning_rate": 3.7491695784042105e-12, + "loss": 0.4363, + "step": 1913 + }, + { + "epoch": 5.98329853862213, + "grad_norm": 0.775199294090271, + "learning_rate": 0.0, + "loss": 0.4328, + "step": 1914 + } + ], + "logging_steps": 1, + "max_steps": 1914, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 319, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 5.798644914657742e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-1914/training_args.bin b/checkpoint-1914/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8067ee9c1c0bc752bdfd00cfcaf1a6e717d2356b --- /dev/null +++ b/checkpoint-1914/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c650156a192cae0a2070c4323ee8a93e9b52fb76041d59ae0633b98389585727 +size 7928 diff --git a/checkpoint-1914/zero_to_fp32.py b/checkpoint-1914/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-1914/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-319/README.md b/checkpoint-319/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f4a3934800eeb082a0cb833d7b6af4f68eed3615 --- /dev/null +++ b/checkpoint-319/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3_3-Nemotron-Super-49B-v1 +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.15.0 \ No newline at end of file diff --git a/checkpoint-319/adapter_config.json b/checkpoint-319/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..1edb33780e2306c6b19fd727be8e9b8b35f237c4 --- /dev/null +++ b/checkpoint-319/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3_3-Nemotron-Super-49B-v1", + "bias": "none", + "corda_config": null, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "k_proj", + "down_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-319/adapter_model.safetensors b/checkpoint-319/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d8f396f36935f54e9ebf74f4fa741d56788b35f7 --- /dev/null +++ b/checkpoint-319/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c1d3e52970fd1c9220ff78f8dcfbe7d50076195a3997fc7cf94457e7972616fe +size 9016826528 diff --git a/checkpoint-319/global_step319/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-319/global_step319/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..92fe5b826731a4c210b62fe28df91b6ef75ef83f --- /dev/null +++ b/checkpoint-319/global_step319/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea7c861852a5aa99dba3052911a00ffe96b3172c1a2b2a5dad54f0aac4d02b5e +size 27050164444 diff --git a/checkpoint-319/global_step319/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-319/global_step319/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..3060e2b85be4c68e2bf431dc8059b0342ae145f4 --- /dev/null +++ b/checkpoint-319/global_step319/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b1c237d41ade0cfcb45da2bce873d64ad89d906aaa159c7f712adf225e230add +size 27050169884 diff --git a/checkpoint-319/global_step319/mp_rank_00_model_states.pt b/checkpoint-319/global_step319/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..77758dd3eb63a655fec29a2bf149fde2592398a8 --- /dev/null +++ b/checkpoint-319/global_step319/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a5b37167ceb09ad4ac9008da28670137f0be4d37889138803cb026f28e59d959 +size 9776788601 diff --git a/checkpoint-319/latest b/checkpoint-319/latest new file mode 100644 index 0000000000000000000000000000000000000000..ce0a9f612309feaeb884433bac375e628247dd48 --- /dev/null +++ b/checkpoint-319/latest @@ -0,0 +1 @@ +global_step319 \ No newline at end of file diff --git a/checkpoint-319/rng_state_0.pth b/checkpoint-319/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..b75feb823766df9e6ee9c6c1c5c1abc161824288 --- /dev/null +++ b/checkpoint-319/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:48d5fdf33d1910923036f72a78ece8e55f970c798de41ce0a723ac98cc6299e6 +size 14512 diff --git a/checkpoint-319/rng_state_1.pth b/checkpoint-319/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..8b6068a4d49244e0256e72ae95d75668111571be --- /dev/null +++ b/checkpoint-319/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:add034b30147f78680605f81076318e2111c9bfc3c9d831bfbcd6a7fd1ffed80 +size 14512 diff --git a/checkpoint-319/scheduler.pt b/checkpoint-319/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..4cf1ca1199c39d96498b722ffde6e486a7bc41cb --- /dev/null +++ b/checkpoint-319/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2eb3e60ca57802cfb48a712a9f58abd00a68f442d18b53579c9399f950b437d8 +size 1064 diff --git a/checkpoint-319/special_tokens_map.json b/checkpoint-319/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-319/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-319/tokenizer.json b/checkpoint-319/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-319/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-319/tokenizer_config.json b/checkpoint-319/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..edd01b980c1db496ea102a51c972ee8f5d1a2c74 --- /dev/null +++ b/checkpoint-319/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "chat_template": "{{- bos_token }}{%- if messages[0]['role'] == 'system' %}{%- set system_message = messages[0]['content']|trim %}{%- set messages = messages[1:] %}{%- else %}{%- set system_message = \"\" %}{%- endif %}{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}{{- system_message }}{{- \"<|eot_id|>\" }}{%- for message in messages %}{%- if message['role'] == 'assistant' and '' in message['content'] %}{%- set content = message['content'].split('')[-1].lstrip() %}{%- else %}{%- set content = message['content'] %}{%- endif %}{{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n' + content | trim + '<|eot_id|>' }}{%- endfor %}{%- if add_generation_prompt %}{{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}{%- endif %}", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-319/trainer_state.json b/checkpoint-319/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..a8028f671b0dffd593c979d107be62d39698bca4 --- /dev/null +++ b/checkpoint-319/trainer_state.json @@ -0,0 +1,2266 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.9989561586638831, + "eval_steps": 500, + "global_step": 319, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.003131524008350731, + "grad_norm": 13.917898178100586, + "learning_rate": 5.0000000000000004e-08, + "loss": 4.1051, + "step": 1 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 17.327869415283203, + "learning_rate": 1.0000000000000001e-07, + "loss": 4.1048, + "step": 2 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 14.063946723937988, + "learning_rate": 1.5000000000000002e-07, + "loss": 4.0741, + "step": 3 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 16.817699432373047, + "learning_rate": 2.0000000000000002e-07, + "loss": 4.2002, + "step": 4 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 14.47036361694336, + "learning_rate": 2.5000000000000004e-07, + "loss": 4.2652, + "step": 5 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 14.474193572998047, + "learning_rate": 3.0000000000000004e-07, + "loss": 4.0888, + "step": 6 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 14.865458488464355, + "learning_rate": 3.5000000000000004e-07, + "loss": 4.0014, + "step": 7 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 15.338888168334961, + "learning_rate": 4.0000000000000003e-07, + "loss": 4.13, + "step": 8 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 15.154336929321289, + "learning_rate": 4.5000000000000003e-07, + "loss": 4.2493, + "step": 9 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 15.919597625732422, + "learning_rate": 5.000000000000001e-07, + "loss": 4.0535, + "step": 10 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 14.981926918029785, + "learning_rate": 5.5e-07, + "loss": 3.9064, + "step": 11 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 13.36101245880127, + "learning_rate": 6.000000000000001e-07, + "loss": 4.1939, + "step": 12 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 15.58773422241211, + "learning_rate": 6.5e-07, + "loss": 4.18, + "step": 13 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 13.560139656066895, + "learning_rate": 7.000000000000001e-07, + "loss": 3.9414, + "step": 14 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 12.307971954345703, + "learning_rate": 7.5e-07, + "loss": 3.8836, + "step": 15 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 14.533182144165039, + "learning_rate": 8.000000000000001e-07, + "loss": 4.1551, + "step": 16 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 13.453729629516602, + "learning_rate": 8.500000000000001e-07, + "loss": 4.0048, + "step": 17 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 13.45992374420166, + "learning_rate": 9.000000000000001e-07, + "loss": 4.0745, + "step": 18 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 11.857145309448242, + "learning_rate": 9.500000000000001e-07, + "loss": 3.9871, + "step": 19 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 11.872294425964355, + "learning_rate": 1.0000000000000002e-06, + "loss": 3.8959, + "step": 20 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.969825744628906, + "learning_rate": 1.0500000000000001e-06, + "loss": 4.0308, + "step": 21 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 12.33769416809082, + "learning_rate": 1.1e-06, + "loss": 3.9341, + "step": 22 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 12.669405937194824, + "learning_rate": 1.1500000000000002e-06, + "loss": 3.8511, + "step": 23 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 10.677213668823242, + "learning_rate": 1.2000000000000002e-06, + "loss": 3.7764, + "step": 24 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 10.366402626037598, + "learning_rate": 1.25e-06, + "loss": 3.5291, + "step": 25 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 11.211421012878418, + "learning_rate": 1.3e-06, + "loss": 3.5765, + "step": 26 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 11.313716888427734, + "learning_rate": 1.3500000000000002e-06, + "loss": 3.4849, + "step": 27 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 10.41294002532959, + "learning_rate": 1.4000000000000001e-06, + "loss": 3.2653, + "step": 28 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 10.40064525604248, + "learning_rate": 1.45e-06, + "loss": 3.3384, + "step": 29 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 10.05427074432373, + "learning_rate": 1.5e-06, + "loss": 3.2257, + "step": 30 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 9.583163261413574, + "learning_rate": 1.5500000000000002e-06, + "loss": 3.1371, + "step": 31 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 10.09977912902832, + "learning_rate": 1.6000000000000001e-06, + "loss": 3.0658, + "step": 32 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 9.271486282348633, + "learning_rate": 1.6500000000000003e-06, + "loss": 2.9693, + "step": 33 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 10.687992095947266, + "learning_rate": 1.7000000000000002e-06, + "loss": 2.95, + "step": 34 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 8.762290000915527, + "learning_rate": 1.75e-06, + "loss": 2.8286, + "step": 35 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 10.13785171508789, + "learning_rate": 1.8000000000000001e-06, + "loss": 2.3664, + "step": 36 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 18.301353454589844, + "learning_rate": 1.85e-06, + "loss": 2.5533, + "step": 37 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 11.490377426147461, + "learning_rate": 1.9000000000000002e-06, + "loss": 2.6133, + "step": 38 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 15.614163398742676, + "learning_rate": 1.9500000000000004e-06, + "loss": 2.3596, + "step": 39 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 17.757442474365234, + "learning_rate": 2.0000000000000003e-06, + "loss": 2.3491, + "step": 40 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 17.18431854248047, + "learning_rate": 2.05e-06, + "loss": 2.2361, + "step": 41 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 16.149789810180664, + "learning_rate": 2.1000000000000002e-06, + "loss": 2.1457, + "step": 42 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 15.256914138793945, + "learning_rate": 2.15e-06, + "loss": 2.12, + "step": 43 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 15.537406921386719, + "learning_rate": 2.2e-06, + "loss": 2.1877, + "step": 44 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 7.947713851928711, + "learning_rate": 2.25e-06, + "loss": 2.1648, + "step": 45 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 8.818676948547363, + "learning_rate": 2.3000000000000004e-06, + "loss": 2.134, + "step": 46 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 5.175768852233887, + "learning_rate": 2.35e-06, + "loss": 2.0796, + "step": 47 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 6.750611305236816, + "learning_rate": 2.4000000000000003e-06, + "loss": 1.9174, + "step": 48 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 6.2147979736328125, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.8065, + "step": 49 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 13.291611671447754, + "learning_rate": 2.5e-06, + "loss": 1.7061, + "step": 50 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 7.251201629638672, + "learning_rate": 2.55e-06, + "loss": 1.7924, + "step": 51 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 5.2126054763793945, + "learning_rate": 2.6e-06, + "loss": 1.6735, + "step": 52 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 5.435528755187988, + "learning_rate": 2.6500000000000005e-06, + "loss": 1.6265, + "step": 53 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 4.505807399749756, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.4851, + "step": 54 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 5.128388404846191, + "learning_rate": 2.7500000000000004e-06, + "loss": 1.5832, + "step": 55 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 16.935827255249023, + "learning_rate": 2.8000000000000003e-06, + "loss": 1.6553, + "step": 56 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 3.664458990097046, + "learning_rate": 2.85e-06, + "loss": 1.5, + "step": 57 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 7.763802528381348, + "learning_rate": 2.9e-06, + "loss": 1.367, + "step": 58 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 3.2216155529022217, + "learning_rate": 2.95e-06, + "loss": 1.3863, + "step": 59 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 4.384445667266846, + "learning_rate": 3e-06, + "loss": 1.4247, + "step": 60 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 4.8080878257751465, + "learning_rate": 3.05e-06, + "loss": 1.3257, + "step": 61 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 4.154761791229248, + "learning_rate": 3.1000000000000004e-06, + "loss": 1.321, + "step": 62 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 6.4742112159729, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.2823, + "step": 63 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 2.583422899246216, + "learning_rate": 3.2000000000000003e-06, + "loss": 1.2136, + "step": 64 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 4.1933488845825195, + "learning_rate": 3.2500000000000002e-06, + "loss": 1.1855, + "step": 65 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 4.11049747467041, + "learning_rate": 3.3000000000000006e-06, + "loss": 1.2389, + "step": 66 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 2.264458417892456, + "learning_rate": 3.3500000000000005e-06, + "loss": 1.0651, + "step": 67 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 2.5408174991607666, + "learning_rate": 3.4000000000000005e-06, + "loss": 1.1389, + "step": 68 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 7.82421350479126, + "learning_rate": 3.45e-06, + "loss": 1.0956, + "step": 69 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 3.070939064025879, + "learning_rate": 3.5e-06, + "loss": 1.0451, + "step": 70 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 2.6310527324676514, + "learning_rate": 3.5500000000000003e-06, + "loss": 1.0538, + "step": 71 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 7.630155563354492, + "learning_rate": 3.6000000000000003e-06, + "loss": 1.0052, + "step": 72 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 6.950636863708496, + "learning_rate": 3.65e-06, + "loss": 1.0473, + "step": 73 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 2.2703945636749268, + "learning_rate": 3.7e-06, + "loss": 1.0576, + "step": 74 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 3.3817710876464844, + "learning_rate": 3.7500000000000005e-06, + "loss": 1.0177, + "step": 75 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 7.266414642333984, + "learning_rate": 3.8000000000000005e-06, + "loss": 1.0645, + "step": 76 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 5.782608509063721, + "learning_rate": 3.85e-06, + "loss": 1.0162, + "step": 77 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 2.7938575744628906, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9664, + "step": 78 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 6.681935787200928, + "learning_rate": 3.95e-06, + "loss": 0.953, + "step": 79 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 2.253279209136963, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9568, + "step": 80 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 1.4875826835632324, + "learning_rate": 4.05e-06, + "loss": 0.9448, + "step": 81 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 2.4987940788269043, + "learning_rate": 4.1e-06, + "loss": 0.9393, + "step": 82 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 4.712948322296143, + "learning_rate": 4.15e-06, + "loss": 0.9532, + "step": 83 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 6.9030632972717285, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.96, + "step": 84 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 3.4780967235565186, + "learning_rate": 4.25e-06, + "loss": 0.8993, + "step": 85 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.526064395904541, + "learning_rate": 4.3e-06, + "loss": 0.9021, + "step": 86 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 10.727686882019043, + "learning_rate": 4.350000000000001e-06, + "loss": 0.856, + "step": 87 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 12.483160972595215, + "learning_rate": 4.4e-06, + "loss": 0.9357, + "step": 88 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 6.544492244720459, + "learning_rate": 4.450000000000001e-06, + "loss": 0.9168, + "step": 89 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 1.178139567375183, + "learning_rate": 4.5e-06, + "loss": 0.8748, + "step": 90 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 1.711506962776184, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8425, + "step": 91 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 3.281747341156006, + "learning_rate": 4.600000000000001e-06, + "loss": 0.8491, + "step": 92 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 2.2964377403259277, + "learning_rate": 4.65e-06, + "loss": 0.8038, + "step": 93 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 1.959700345993042, + "learning_rate": 4.7e-06, + "loss": 0.8439, + "step": 94 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 3.979384183883667, + "learning_rate": 4.75e-06, + "loss": 0.8839, + "step": 95 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 1.4721262454986572, + "learning_rate": 4.800000000000001e-06, + "loss": 0.845, + "step": 96 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 2.862248659133911, + "learning_rate": 4.85e-06, + "loss": 0.7748, + "step": 97 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 3.7439088821411133, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8145, + "step": 98 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 1.6654618978500366, + "learning_rate": 4.95e-06, + "loss": 0.8326, + "step": 99 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 7.8437581062316895, + "learning_rate": 5e-06, + "loss": 0.8666, + "step": 100 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 6.429738521575928, + "learning_rate": 4.999996250830422e-06, + "loss": 0.836, + "step": 101 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 2.6017794609069824, + "learning_rate": 4.9999850033329326e-06, + "loss": 0.7785, + "step": 102 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 1.0575449466705322, + "learning_rate": 4.999966257541265e-06, + "loss": 0.7639, + "step": 103 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 2.6932010650634766, + "learning_rate": 4.999940013511647e-06, + "loss": 0.8214, + "step": 104 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 2.925288438796997, + "learning_rate": 4.999906271322792e-06, + "loss": 0.8797, + "step": 105 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.3570607900619507, + "learning_rate": 4.9998650310759035e-06, + "loss": 0.792, + "step": 106 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 5.126713752746582, + "learning_rate": 4.999816292894676e-06, + "loss": 0.8352, + "step": 107 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 1.8966432809829712, + "learning_rate": 4.99976005692529e-06, + "loss": 0.7663, + "step": 108 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 1.3100829124450684, + "learning_rate": 4.999696323336418e-06, + "loss": 0.771, + "step": 109 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 2.4025354385375977, + "learning_rate": 4.999625092319218e-06, + "loss": 0.7618, + "step": 110 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.130232810974121, + "learning_rate": 4.999546364087334e-06, + "loss": 0.7705, + "step": 111 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 3.430262327194214, + "learning_rate": 4.999460138876901e-06, + "loss": 0.77, + "step": 112 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 1.1272103786468506, + "learning_rate": 4.999366416946536e-06, + "loss": 0.7133, + "step": 113 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 1.1740471124649048, + "learning_rate": 4.999265198577342e-06, + "loss": 0.7684, + "step": 114 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.3138248920440674, + "learning_rate": 4.999156484072907e-06, + "loss": 0.7888, + "step": 115 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 1.061711311340332, + "learning_rate": 4.999040273759304e-06, + "loss": 0.7484, + "step": 116 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.4682390689849854, + "learning_rate": 4.998916567985083e-06, + "loss": 0.7296, + "step": 117 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 2.884068250656128, + "learning_rate": 4.998785367121284e-06, + "loss": 0.7662, + "step": 118 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9812761545181274, + "learning_rate": 4.9986466715614205e-06, + "loss": 0.7307, + "step": 119 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 2.2237496376037598, + "learning_rate": 4.998500481721484e-06, + "loss": 0.6761, + "step": 120 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 1.4004178047180176, + "learning_rate": 4.998346798039952e-06, + "loss": 0.7505, + "step": 121 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 5.54975700378418, + "learning_rate": 4.99818562097777e-06, + "loss": 0.7615, + "step": 122 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 6.17140531539917, + "learning_rate": 4.9980169510183624e-06, + "loss": 0.7002, + "step": 123 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 4.974380016326904, + "learning_rate": 4.997840788667628e-06, + "loss": 0.7449, + "step": 124 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 1.4133399724960327, + "learning_rate": 4.997657134453937e-06, + "loss": 0.7442, + "step": 125 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 1.868915319442749, + "learning_rate": 4.9974659889281295e-06, + "loss": 0.7104, + "step": 126 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 1.2599350214004517, + "learning_rate": 4.997267352663514e-06, + "loss": 0.7385, + "step": 127 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.4353641271591187, + "learning_rate": 4.997061226255869e-06, + "loss": 0.7081, + "step": 128 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 3.2492141723632812, + "learning_rate": 4.996847610323437e-06, + "loss": 0.7859, + "step": 129 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 9.599719047546387, + "learning_rate": 4.996626505506923e-06, + "loss": 0.7241, + "step": 130 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 10.053650856018066, + "learning_rate": 4.996397912469494e-06, + "loss": 0.6841, + "step": 131 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 1.323876976966858, + "learning_rate": 4.996161831896777e-06, + "loss": 0.7317, + "step": 132 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 1.4180598258972168, + "learning_rate": 4.9959182644968594e-06, + "loss": 0.692, + "step": 133 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 1.2194396257400513, + "learning_rate": 4.99566721100028e-06, + "loss": 0.7068, + "step": 134 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0984960794448853, + "learning_rate": 4.995408672160031e-06, + "loss": 0.6946, + "step": 135 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 1.9341071844100952, + "learning_rate": 4.995142648751561e-06, + "loss": 0.7467, + "step": 136 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 1.9960932731628418, + "learning_rate": 4.9948691415727594e-06, + "loss": 0.7379, + "step": 137 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.8743917942047119, + "learning_rate": 4.994588151443968e-06, + "loss": 0.66, + "step": 138 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8655261993408203, + "learning_rate": 4.99429967920797e-06, + "loss": 0.6646, + "step": 139 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 5.462070941925049, + "learning_rate": 4.994003725729992e-06, + "loss": 0.643, + "step": 140 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 2.1401469707489014, + "learning_rate": 4.993700291897695e-06, + "loss": 0.6639, + "step": 141 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 1.8219833374023438, + "learning_rate": 4.9933893786211815e-06, + "loss": 0.6673, + "step": 142 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 1.641079306602478, + "learning_rate": 4.993070986832984e-06, + "loss": 0.658, + "step": 143 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 1.1739819049835205, + "learning_rate": 4.992745117488066e-06, + "loss": 0.6826, + "step": 144 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 2.309185743331909, + "learning_rate": 4.9924117715638185e-06, + "loss": 0.6536, + "step": 145 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 1.09304940700531, + "learning_rate": 4.99207095006006e-06, + "loss": 0.721, + "step": 146 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.9056984186172485, + "learning_rate": 4.991722653999025e-06, + "loss": 0.7019, + "step": 147 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.8440625667572021, + "learning_rate": 4.991366884425374e-06, + "loss": 0.707, + "step": 148 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 1.2244676351547241, + "learning_rate": 4.991003642406177e-06, + "loss": 0.6407, + "step": 149 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9258589744567871, + "learning_rate": 4.99063292903092e-06, + "loss": 0.6954, + "step": 150 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 4.176390647888184, + "learning_rate": 4.990254745411496e-06, + "loss": 0.6812, + "step": 151 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 1.4322530031204224, + "learning_rate": 4.989869092682205e-06, + "loss": 0.6808, + "step": 152 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.8017717003822327, + "learning_rate": 4.989475971999748e-06, + "loss": 0.687, + "step": 153 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 1.5641374588012695, + "learning_rate": 4.989075384543228e-06, + "loss": 0.6599, + "step": 154 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 1.1522141695022583, + "learning_rate": 4.98866733151414e-06, + "loss": 0.6546, + "step": 155 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8593171238899231, + "learning_rate": 4.988251814136372e-06, + "loss": 0.6857, + "step": 156 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 2.668159246444702, + "learning_rate": 4.9878288336562e-06, + "loss": 0.661, + "step": 157 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9953671097755432, + "learning_rate": 4.987398391342285e-06, + "loss": 0.6512, + "step": 158 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.042872667312622, + "learning_rate": 4.986960488485667e-06, + "loss": 0.6311, + "step": 159 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.9070663452148438, + "learning_rate": 4.9865151263997645e-06, + "loss": 0.675, + "step": 160 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.8460433483123779, + "learning_rate": 4.986062306420367e-06, + "loss": 0.6635, + "step": 161 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 1.2639834880828857, + "learning_rate": 4.985602029905635e-06, + "loss": 0.6327, + "step": 162 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.8775074481964111, + "learning_rate": 4.985134298236091e-06, + "loss": 0.644, + "step": 163 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 1.2031961679458618, + "learning_rate": 4.98465911281462e-06, + "loss": 0.6254, + "step": 164 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.892494797706604, + "learning_rate": 4.984176475066463e-06, + "loss": 0.7122, + "step": 165 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 2.7122485637664795, + "learning_rate": 4.983686386439212e-06, + "loss": 0.6679, + "step": 166 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9344426989555359, + "learning_rate": 4.983188848402806e-06, + "loss": 0.6319, + "step": 167 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 1.4093577861785889, + "learning_rate": 4.982683862449531e-06, + "loss": 0.6425, + "step": 168 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 1.1285009384155273, + "learning_rate": 4.982171430094007e-06, + "loss": 0.6298, + "step": 169 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.952778935432434, + "learning_rate": 4.981651552873193e-06, + "loss": 0.7066, + "step": 170 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 5.133765697479248, + "learning_rate": 4.981124232346374e-06, + "loss": 0.6634, + "step": 171 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9770542979240417, + "learning_rate": 4.980589470095161e-06, + "loss": 0.7121, + "step": 172 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.8414323925971985, + "learning_rate": 4.980047267723487e-06, + "loss": 0.6397, + "step": 173 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 1.9173879623413086, + "learning_rate": 4.979497626857596e-06, + "loss": 0.6228, + "step": 174 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 1.0823363065719604, + "learning_rate": 4.978940549146048e-06, + "loss": 0.6475, + "step": 175 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 3.715353488922119, + "learning_rate": 4.978376036259706e-06, + "loss": 0.7127, + "step": 176 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.981584370136261, + "learning_rate": 4.9778040898917325e-06, + "loss": 0.6468, + "step": 177 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.70566987991333, + "learning_rate": 4.977224711757587e-06, + "loss": 0.6476, + "step": 178 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9217923283576965, + "learning_rate": 4.976637903595019e-06, + "loss": 0.6731, + "step": 179 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.8994677662849426, + "learning_rate": 4.976043667164063e-06, + "loss": 0.6562, + "step": 180 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 1.1613017320632935, + "learning_rate": 4.975442004247034e-06, + "loss": 0.6417, + "step": 181 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 1.6041977405548096, + "learning_rate": 4.974832916648521e-06, + "loss": 0.6029, + "step": 182 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.7978405952453613, + "learning_rate": 4.974216406195383e-06, + "loss": 0.6269, + "step": 183 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.6021920442581177, + "learning_rate": 4.973592474736739e-06, + "loss": 0.6149, + "step": 184 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8973568677902222, + "learning_rate": 4.972961124143971e-06, + "loss": 0.6648, + "step": 185 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 1.9432591199874878, + "learning_rate": 4.972322356310711e-06, + "loss": 0.6299, + "step": 186 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 4.457028388977051, + "learning_rate": 4.971676173152839e-06, + "loss": 0.656, + "step": 187 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 2.0989716053009033, + "learning_rate": 4.971022576608473e-06, + "loss": 0.6539, + "step": 188 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 1.0646967887878418, + "learning_rate": 4.97036156863797e-06, + "loss": 0.6727, + "step": 189 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.6522265672683716, + "learning_rate": 4.969693151223914e-06, + "loss": 0.6643, + "step": 190 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.7503505945205688, + "learning_rate": 4.969017326371115e-06, + "loss": 0.6402, + "step": 191 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 1.2341989278793335, + "learning_rate": 4.968334096106597e-06, + "loss": 0.6413, + "step": 192 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 3.089054584503174, + "learning_rate": 4.967643462479597e-06, + "loss": 0.6825, + "step": 193 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 2.711623430252075, + "learning_rate": 4.966945427561557e-06, + "loss": 0.65, + "step": 194 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 4.641184329986572, + "learning_rate": 4.966239993446118e-06, + "loss": 0.6229, + "step": 195 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.7984074354171753, + "learning_rate": 4.965527162249114e-06, + "loss": 0.6473, + "step": 196 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 1.1643115282058716, + "learning_rate": 4.964806936108566e-06, + "loss": 0.6404, + "step": 197 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 2.1877920627593994, + "learning_rate": 4.9640793171846725e-06, + "loss": 0.6185, + "step": 198 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.7970566749572754, + "learning_rate": 4.963344307659807e-06, + "loss": 0.634, + "step": 199 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 1.6014361381530762, + "learning_rate": 4.96260190973851e-06, + "loss": 0.6562, + "step": 200 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.8743320107460022, + "learning_rate": 4.961852125647482e-06, + "loss": 0.6133, + "step": 201 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 1.9526551961898804, + "learning_rate": 4.961094957635578e-06, + "loss": 0.6451, + "step": 202 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 3.6597347259521484, + "learning_rate": 4.960330407973798e-06, + "loss": 0.6386, + "step": 203 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 1.7180207967758179, + "learning_rate": 4.959558478955283e-06, + "loss": 0.6688, + "step": 204 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 0.9058470129966736, + "learning_rate": 4.958779172895308e-06, + "loss": 0.6161, + "step": 205 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 1.0031033754348755, + "learning_rate": 4.957992492131274e-06, + "loss": 0.6437, + "step": 206 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 1.5846725702285767, + "learning_rate": 4.9571984390226985e-06, + "loss": 0.6332, + "step": 207 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 1.9951609373092651, + "learning_rate": 4.956397015951215e-06, + "loss": 0.636, + "step": 208 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 1.4122583866119385, + "learning_rate": 4.95558822532056e-06, + "loss": 0.6586, + "step": 209 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.2243481874465942, + "learning_rate": 4.954772069556568e-06, + "loss": 0.6313, + "step": 210 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.8756356835365295, + "learning_rate": 4.953948551107164e-06, + "loss": 0.6406, + "step": 211 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 2.9979734420776367, + "learning_rate": 4.953117672442356e-06, + "loss": 0.5803, + "step": 212 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 2.1859359741210938, + "learning_rate": 4.952279436054229e-06, + "loss": 0.6607, + "step": 213 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.6929755806922913, + "learning_rate": 4.9514338444569346e-06, + "loss": 0.5989, + "step": 214 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 1.0361783504486084, + "learning_rate": 4.950580900186685e-06, + "loss": 0.6654, + "step": 215 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 1.210898518562317, + "learning_rate": 4.9497206058017475e-06, + "loss": 0.6213, + "step": 216 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 1.200990080833435, + "learning_rate": 4.948852963882434e-06, + "loss": 0.6654, + "step": 217 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 1.481831669807434, + "learning_rate": 4.947977977031093e-06, + "loss": 0.6474, + "step": 218 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.9883334636688232, + "learning_rate": 4.947095647872103e-06, + "loss": 0.6735, + "step": 219 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.7436536550521851, + "learning_rate": 4.946205979051868e-06, + "loss": 0.6456, + "step": 220 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9057570099830627, + "learning_rate": 4.945308973238802e-06, + "loss": 0.6228, + "step": 221 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 1.341081142425537, + "learning_rate": 4.944404633123324e-06, + "loss": 0.6417, + "step": 222 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.7958157062530518, + "learning_rate": 4.943492961417859e-06, + "loss": 0.6494, + "step": 223 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.216025471687317, + "learning_rate": 4.9425739608568106e-06, + "loss": 0.6566, + "step": 224 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.9774854779243469, + "learning_rate": 4.9416476341965735e-06, + "loss": 0.6171, + "step": 225 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 2.1562681198120117, + "learning_rate": 4.940713984215512e-06, + "loss": 0.629, + "step": 226 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 1.9521286487579346, + "learning_rate": 4.9397730137139556e-06, + "loss": 0.6475, + "step": 227 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 1.5749104022979736, + "learning_rate": 4.9388247255141895e-06, + "loss": 0.6053, + "step": 228 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 1.2008254528045654, + "learning_rate": 4.937869122460449e-06, + "loss": 0.6052, + "step": 229 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 1.0774102210998535, + "learning_rate": 4.93690620741891e-06, + "loss": 0.6099, + "step": 230 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 1.0929996967315674, + "learning_rate": 4.935935983277675e-06, + "loss": 0.6363, + "step": 231 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.8830653429031372, + "learning_rate": 4.934958452946774e-06, + "loss": 0.6136, + "step": 232 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 3.591218948364258, + "learning_rate": 4.933973619358147e-06, + "loss": 0.5962, + "step": 233 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 2.5797672271728516, + "learning_rate": 4.932981485465643e-06, + "loss": 0.6405, + "step": 234 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 1.0467664003372192, + "learning_rate": 4.9319820542450025e-06, + "loss": 0.6155, + "step": 235 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.8099795579910278, + "learning_rate": 4.930975328693856e-06, + "loss": 0.5615, + "step": 236 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.8906702995300293, + "learning_rate": 4.92996131183171e-06, + "loss": 0.6501, + "step": 237 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 1.0871416330337524, + "learning_rate": 4.928940006699944e-06, + "loss": 0.6282, + "step": 238 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 1.3209614753723145, + "learning_rate": 4.927911416361792e-06, + "loss": 0.598, + "step": 239 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.2252682447433472, + "learning_rate": 4.926875543902344e-06, + "loss": 0.6433, + "step": 240 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0569007396697998, + "learning_rate": 4.9258323924285285e-06, + "loss": 0.5927, + "step": 241 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9309014081954956, + "learning_rate": 4.924781965069106e-06, + "loss": 0.5927, + "step": 242 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0200378894805908, + "learning_rate": 4.923724264974662e-06, + "loss": 0.6064, + "step": 243 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 1.0533075332641602, + "learning_rate": 4.922659295317593e-06, + "loss": 0.6373, + "step": 244 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 0.7889382839202881, + "learning_rate": 4.921587059292102e-06, + "loss": 0.5887, + "step": 245 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.7943588495254517, + "learning_rate": 4.920507560114183e-06, + "loss": 0.593, + "step": 246 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.8247205018997192, + "learning_rate": 4.919420801021617e-06, + "loss": 0.6151, + "step": 247 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 0.9979158043861389, + "learning_rate": 4.91832678527396e-06, + "loss": 0.6019, + "step": 248 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9346868991851807, + "learning_rate": 4.917225516152532e-06, + "loss": 0.6098, + "step": 249 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.7487881183624268, + "learning_rate": 4.916116996960408e-06, + "loss": 0.5965, + "step": 250 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.821576714515686, + "learning_rate": 4.915001231022411e-06, + "loss": 0.6483, + "step": 251 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 1.0413196086883545, + "learning_rate": 4.913878221685096e-06, + "loss": 0.6108, + "step": 252 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9560331702232361, + "learning_rate": 4.912747972316745e-06, + "loss": 0.5758, + "step": 253 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.8964638113975525, + "learning_rate": 4.911610486307356e-06, + "loss": 0.6432, + "step": 254 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.8418346047401428, + "learning_rate": 4.910465767068631e-06, + "loss": 0.6027, + "step": 255 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 1.792371153831482, + "learning_rate": 4.909313818033966e-06, + "loss": 0.6198, + "step": 256 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 1.036665439605713, + "learning_rate": 4.908154642658446e-06, + "loss": 0.6255, + "step": 257 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.7592151165008545, + "learning_rate": 4.906988244418823e-06, + "loss": 0.6035, + "step": 258 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.8843073844909668, + "learning_rate": 4.90581462681352e-06, + "loss": 0.6299, + "step": 259 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9489964246749878, + "learning_rate": 4.9046337933626086e-06, + "loss": 0.5869, + "step": 260 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.851691722869873, + "learning_rate": 4.903445747607806e-06, + "loss": 0.603, + "step": 261 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 1.3722106218338013, + "learning_rate": 4.902250493112458e-06, + "loss": 0.5939, + "step": 262 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 1.1002827882766724, + "learning_rate": 4.901048033461537e-06, + "loss": 0.6452, + "step": 263 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 0.8428632020950317, + "learning_rate": 4.89983837226162e-06, + "loss": 0.5956, + "step": 264 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.7666584849357605, + "learning_rate": 4.898621513140889e-06, + "loss": 0.6067, + "step": 265 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.8413611054420471, + "learning_rate": 4.897397459749113e-06, + "loss": 0.5985, + "step": 266 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 2.3374335765838623, + "learning_rate": 4.896166215757638e-06, + "loss": 0.5885, + "step": 267 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 2.236640214920044, + "learning_rate": 4.894927784859377e-06, + "loss": 0.6408, + "step": 268 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.9715856313705444, + "learning_rate": 4.893682170768802e-06, + "loss": 0.5954, + "step": 269 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0249912738800049, + "learning_rate": 4.892429377221928e-06, + "loss": 0.6186, + "step": 270 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 1.255426049232483, + "learning_rate": 4.891169407976302e-06, + "loss": 0.6351, + "step": 271 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.9339559674263, + "learning_rate": 4.889902266810995e-06, + "loss": 0.5944, + "step": 272 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 1.2473429441452026, + "learning_rate": 4.888627957526589e-06, + "loss": 0.544, + "step": 273 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 1.0589442253112793, + "learning_rate": 4.887346483945166e-06, + "loss": 0.5543, + "step": 274 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9844024777412415, + "learning_rate": 4.886057849910294e-06, + "loss": 0.5941, + "step": 275 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 2.88578200340271, + "learning_rate": 4.8847620592870196e-06, + "loss": 0.6124, + "step": 276 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 0.7496054172515869, + "learning_rate": 4.8834591159618524e-06, + "loss": 0.6006, + "step": 277 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 0.7403052449226379, + "learning_rate": 4.88214902384276e-06, + "loss": 0.5911, + "step": 278 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9003771543502808, + "learning_rate": 4.880831786859146e-06, + "loss": 0.6347, + "step": 279 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0345501899719238, + "learning_rate": 4.879507408961847e-06, + "loss": 0.6111, + "step": 280 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.4385879039764404, + "learning_rate": 4.878175894123116e-06, + "loss": 0.6454, + "step": 281 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.8469482064247131, + "learning_rate": 4.8768372463366145e-06, + "loss": 0.6163, + "step": 282 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.8859589695930481, + "learning_rate": 4.875491469617395e-06, + "loss": 0.6144, + "step": 283 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 1.8436834812164307, + "learning_rate": 4.874138568001895e-06, + "loss": 0.6275, + "step": 284 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.6646101474761963, + "learning_rate": 4.87277854554792e-06, + "loss": 0.615, + "step": 285 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 1.0070925951004028, + "learning_rate": 4.871411406334633e-06, + "loss": 0.5898, + "step": 286 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9785194993019104, + "learning_rate": 4.870037154462545e-06, + "loss": 0.5992, + "step": 287 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.7244889736175537, + "learning_rate": 4.868655794053497e-06, + "loss": 0.6078, + "step": 288 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.4496444463729858, + "learning_rate": 4.8672673292506535e-06, + "loss": 0.5855, + "step": 289 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 1.8514957427978516, + "learning_rate": 4.865871764218486e-06, + "loss": 0.5707, + "step": 290 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.8439773321151733, + "learning_rate": 4.864469103142763e-06, + "loss": 0.5562, + "step": 291 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.8146086931228638, + "learning_rate": 4.8630593502305355e-06, + "loss": 0.6161, + "step": 292 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.8920315504074097, + "learning_rate": 4.861642509710126e-06, + "loss": 0.6139, + "step": 293 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.4980088472366333, + "learning_rate": 4.860218585831116e-06, + "loss": 0.6187, + "step": 294 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 0.9910127520561218, + "learning_rate": 4.8587875828643285e-06, + "loss": 0.5852, + "step": 295 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.819600522518158, + "learning_rate": 4.857349505101823e-06, + "loss": 0.6172, + "step": 296 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 1.1059772968292236, + "learning_rate": 4.855904356856878e-06, + "loss": 0.5868, + "step": 297 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 1.2362196445465088, + "learning_rate": 4.854452142463977e-06, + "loss": 0.625, + "step": 298 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.9956470727920532, + "learning_rate": 4.852992866278799e-06, + "loss": 0.5923, + "step": 299 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 0.864109218120575, + "learning_rate": 4.851526532678203e-06, + "loss": 0.6315, + "step": 300 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 0.8900614380836487, + "learning_rate": 4.850053146060217e-06, + "loss": 0.6128, + "step": 301 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.927254855632782, + "learning_rate": 4.84857271084402e-06, + "loss": 0.5955, + "step": 302 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0046517848968506, + "learning_rate": 4.847085231469935e-06, + "loss": 0.6134, + "step": 303 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.734597384929657, + "learning_rate": 4.8455907123994125e-06, + "loss": 0.5927, + "step": 304 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.7338348031044006, + "learning_rate": 4.844089158115016e-06, + "loss": 0.5897, + "step": 305 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9163988828659058, + "learning_rate": 4.8425805731204106e-06, + "loss": 0.6051, + "step": 306 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 1.050246238708496, + "learning_rate": 4.84106496194035e-06, + "loss": 0.5751, + "step": 307 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 0.7637603878974915, + "learning_rate": 4.83954232912066e-06, + "loss": 0.5677, + "step": 308 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.7110525965690613, + "learning_rate": 4.838012679228229e-06, + "loss": 0.6051, + "step": 309 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.7662068605422974, + "learning_rate": 4.836476016850988e-06, + "loss": 0.59, + "step": 310 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 0.8907375335693359, + "learning_rate": 4.834932346597906e-06, + "loss": 0.5792, + "step": 311 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.8939849138259888, + "learning_rate": 4.833381673098966e-06, + "loss": 0.6062, + "step": 312 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 0.8878788948059082, + "learning_rate": 4.8318240010051595e-06, + "loss": 0.5694, + "step": 313 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 1.2523870468139648, + "learning_rate": 4.830259334988468e-06, + "loss": 0.5809, + "step": 314 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 1.0836797952651978, + "learning_rate": 4.82868767974185e-06, + "loss": 0.5949, + "step": 315 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.7985473871231079, + "learning_rate": 4.827109039979226e-06, + "loss": 0.6057, + "step": 316 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 1.042951226234436, + "learning_rate": 4.825523420435469e-06, + "loss": 0.6004, + "step": 317 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.7845115661621094, + "learning_rate": 4.823930825866381e-06, + "loss": 0.6161, + "step": 318 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.931854784488678, + "learning_rate": 4.82233126104869e-06, + "loss": 0.5912, + "step": 319 + } + ], + "logging_steps": 1, + "max_steps": 1914, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 319, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 9.681902049591034e+18, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-319/training_args.bin b/checkpoint-319/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8067ee9c1c0bc752bdfd00cfcaf1a6e717d2356b --- /dev/null +++ b/checkpoint-319/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c650156a192cae0a2070c4323ee8a93e9b52fb76041d59ae0633b98389585727 +size 7928 diff --git a/checkpoint-319/zero_to_fp32.py b/checkpoint-319/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-319/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-638/README.md b/checkpoint-638/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f4a3934800eeb082a0cb833d7b6af4f68eed3615 --- /dev/null +++ b/checkpoint-638/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3_3-Nemotron-Super-49B-v1 +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.15.0 \ No newline at end of file diff --git a/checkpoint-638/adapter_config.json b/checkpoint-638/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..1edb33780e2306c6b19fd727be8e9b8b35f237c4 --- /dev/null +++ b/checkpoint-638/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3_3-Nemotron-Super-49B-v1", + "bias": "none", + "corda_config": null, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "k_proj", + "down_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-638/adapter_model.safetensors b/checkpoint-638/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..35842b9651b371450da9928a14475aeb7e3f269c --- /dev/null +++ b/checkpoint-638/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0812464ec0eceb2dda23eb97599723c8b0355ef227db40ef5b2f930dc4ef6186 +size 9016826528 diff --git a/checkpoint-638/global_step637/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-638/global_step637/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..dca4213777557e059dd0d043ae7d375bea6ea395 --- /dev/null +++ b/checkpoint-638/global_step637/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d9173edb4c9ebf323ced262faf58fd939e9af4f3604110f45aa3753c063f5caf +size 27050164444 diff --git a/checkpoint-638/global_step637/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-638/global_step637/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..da98948aa0c11dedf9fe16150a1506186fcc309d --- /dev/null +++ b/checkpoint-638/global_step637/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a7edd77adf20792561b7de39e823bcff5fdd6be758d10ad71579e00e7b00cb90 +size 27050169884 diff --git a/checkpoint-638/global_step637/mp_rank_00_model_states.pt b/checkpoint-638/global_step637/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..ecc121bdbbe2508b39af8bbafa231faaa36ea1e7 --- /dev/null +++ b/checkpoint-638/global_step637/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac91ce79a28628551a8979923f07d5118365f62a421e59c08c015b4bdae41290 +size 9776788601 diff --git a/checkpoint-638/latest b/checkpoint-638/latest new file mode 100644 index 0000000000000000000000000000000000000000..d012d0ce403671314e126e6402c39bcddcfe6480 --- /dev/null +++ b/checkpoint-638/latest @@ -0,0 +1 @@ +global_step637 \ No newline at end of file diff --git a/checkpoint-638/rng_state_0.pth b/checkpoint-638/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..d84c77535f8c7a4c2676f9274073d68c3007aeba --- /dev/null +++ b/checkpoint-638/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4c1befa2d454c450cd89379bc71c69f965ee33aca6ae6a7a757d8f3f6d6f8b70 +size 14512 diff --git a/checkpoint-638/rng_state_1.pth b/checkpoint-638/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..c0de96118a7a564558659b7142cfe50a9a694f2b --- /dev/null +++ b/checkpoint-638/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e2e65f53006556f6db1a94f122adf6f3b167ecff4df64a9d2980168de08b7e1 +size 14512 diff --git a/checkpoint-638/scheduler.pt b/checkpoint-638/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..580124aead66a410b0fed61fa136800732c3f226 --- /dev/null +++ b/checkpoint-638/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d8c0b9ca4dbe6069aa5d148004daaceb5003d77e6d5d93cc67757781a96bf7a7 +size 1064 diff --git a/checkpoint-638/special_tokens_map.json b/checkpoint-638/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-638/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-638/tokenizer.json b/checkpoint-638/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-638/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-638/tokenizer_config.json b/checkpoint-638/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..edd01b980c1db496ea102a51c972ee8f5d1a2c74 --- /dev/null +++ b/checkpoint-638/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "chat_template": "{{- bos_token }}{%- if messages[0]['role'] == 'system' %}{%- set system_message = messages[0]['content']|trim %}{%- set messages = messages[1:] %}{%- else %}{%- set system_message = \"\" %}{%- endif %}{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}{{- system_message }}{{- \"<|eot_id|>\" }}{%- for message in messages %}{%- if message['role'] == 'assistant' and '' in message['content'] %}{%- set content = message['content'].split('')[-1].lstrip() %}{%- else %}{%- set content = message['content'] %}{%- endif %}{{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n' + content | trim + '<|eot_id|>' }}{%- endfor %}{%- if add_generation_prompt %}{{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}{%- endif %}", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-638/trainer_state.json b/checkpoint-638/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..572435ec3e53693dcf53e3286d185caaee1b1c44 --- /dev/null +++ b/checkpoint-638/trainer_state.json @@ -0,0 +1,4499 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.9958246346555324, + "eval_steps": 500, + "global_step": 638, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.003131524008350731, + "grad_norm": 13.917898178100586, + "learning_rate": 5.0000000000000004e-08, + "loss": 4.1051, + "step": 1 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 17.327869415283203, + "learning_rate": 1.0000000000000001e-07, + "loss": 4.1048, + "step": 2 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 14.063946723937988, + "learning_rate": 1.5000000000000002e-07, + "loss": 4.0741, + "step": 3 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 16.817699432373047, + "learning_rate": 2.0000000000000002e-07, + "loss": 4.2002, + "step": 4 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 14.47036361694336, + "learning_rate": 2.5000000000000004e-07, + "loss": 4.2652, + "step": 5 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 14.474193572998047, + "learning_rate": 3.0000000000000004e-07, + "loss": 4.0888, + "step": 6 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 14.865458488464355, + "learning_rate": 3.5000000000000004e-07, + "loss": 4.0014, + "step": 7 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 15.338888168334961, + "learning_rate": 4.0000000000000003e-07, + "loss": 4.13, + "step": 8 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 15.154336929321289, + "learning_rate": 4.5000000000000003e-07, + "loss": 4.2493, + "step": 9 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 15.919597625732422, + "learning_rate": 5.000000000000001e-07, + "loss": 4.0535, + "step": 10 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 14.981926918029785, + "learning_rate": 5.5e-07, + "loss": 3.9064, + "step": 11 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 13.36101245880127, + "learning_rate": 6.000000000000001e-07, + "loss": 4.1939, + "step": 12 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 15.58773422241211, + "learning_rate": 6.5e-07, + "loss": 4.18, + "step": 13 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 13.560139656066895, + "learning_rate": 7.000000000000001e-07, + "loss": 3.9414, + "step": 14 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 12.307971954345703, + "learning_rate": 7.5e-07, + "loss": 3.8836, + "step": 15 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 14.533182144165039, + "learning_rate": 8.000000000000001e-07, + "loss": 4.1551, + "step": 16 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 13.453729629516602, + "learning_rate": 8.500000000000001e-07, + "loss": 4.0048, + "step": 17 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 13.45992374420166, + "learning_rate": 9.000000000000001e-07, + "loss": 4.0745, + "step": 18 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 11.857145309448242, + "learning_rate": 9.500000000000001e-07, + "loss": 3.9871, + "step": 19 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 11.872294425964355, + "learning_rate": 1.0000000000000002e-06, + "loss": 3.8959, + "step": 20 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.969825744628906, + "learning_rate": 1.0500000000000001e-06, + "loss": 4.0308, + "step": 21 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 12.33769416809082, + "learning_rate": 1.1e-06, + "loss": 3.9341, + "step": 22 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 12.669405937194824, + "learning_rate": 1.1500000000000002e-06, + "loss": 3.8511, + "step": 23 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 10.677213668823242, + "learning_rate": 1.2000000000000002e-06, + "loss": 3.7764, + "step": 24 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 10.366402626037598, + "learning_rate": 1.25e-06, + "loss": 3.5291, + "step": 25 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 11.211421012878418, + "learning_rate": 1.3e-06, + "loss": 3.5765, + "step": 26 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 11.313716888427734, + "learning_rate": 1.3500000000000002e-06, + "loss": 3.4849, + "step": 27 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 10.41294002532959, + "learning_rate": 1.4000000000000001e-06, + "loss": 3.2653, + "step": 28 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 10.40064525604248, + "learning_rate": 1.45e-06, + "loss": 3.3384, + "step": 29 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 10.05427074432373, + "learning_rate": 1.5e-06, + "loss": 3.2257, + "step": 30 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 9.583163261413574, + "learning_rate": 1.5500000000000002e-06, + "loss": 3.1371, + "step": 31 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 10.09977912902832, + "learning_rate": 1.6000000000000001e-06, + "loss": 3.0658, + "step": 32 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 9.271486282348633, + "learning_rate": 1.6500000000000003e-06, + "loss": 2.9693, + "step": 33 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 10.687992095947266, + "learning_rate": 1.7000000000000002e-06, + "loss": 2.95, + "step": 34 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 8.762290000915527, + "learning_rate": 1.75e-06, + "loss": 2.8286, + "step": 35 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 10.13785171508789, + "learning_rate": 1.8000000000000001e-06, + "loss": 2.3664, + "step": 36 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 18.301353454589844, + "learning_rate": 1.85e-06, + "loss": 2.5533, + "step": 37 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 11.490377426147461, + "learning_rate": 1.9000000000000002e-06, + "loss": 2.6133, + "step": 38 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 15.614163398742676, + "learning_rate": 1.9500000000000004e-06, + "loss": 2.3596, + "step": 39 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 17.757442474365234, + "learning_rate": 2.0000000000000003e-06, + "loss": 2.3491, + "step": 40 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 17.18431854248047, + "learning_rate": 2.05e-06, + "loss": 2.2361, + "step": 41 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 16.149789810180664, + "learning_rate": 2.1000000000000002e-06, + "loss": 2.1457, + "step": 42 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 15.256914138793945, + "learning_rate": 2.15e-06, + "loss": 2.12, + "step": 43 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 15.537406921386719, + "learning_rate": 2.2e-06, + "loss": 2.1877, + "step": 44 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 7.947713851928711, + "learning_rate": 2.25e-06, + "loss": 2.1648, + "step": 45 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 8.818676948547363, + "learning_rate": 2.3000000000000004e-06, + "loss": 2.134, + "step": 46 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 5.175768852233887, + "learning_rate": 2.35e-06, + "loss": 2.0796, + "step": 47 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 6.750611305236816, + "learning_rate": 2.4000000000000003e-06, + "loss": 1.9174, + "step": 48 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 6.2147979736328125, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.8065, + "step": 49 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 13.291611671447754, + "learning_rate": 2.5e-06, + "loss": 1.7061, + "step": 50 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 7.251201629638672, + "learning_rate": 2.55e-06, + "loss": 1.7924, + "step": 51 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 5.2126054763793945, + "learning_rate": 2.6e-06, + "loss": 1.6735, + "step": 52 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 5.435528755187988, + "learning_rate": 2.6500000000000005e-06, + "loss": 1.6265, + "step": 53 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 4.505807399749756, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.4851, + "step": 54 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 5.128388404846191, + "learning_rate": 2.7500000000000004e-06, + "loss": 1.5832, + "step": 55 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 16.935827255249023, + "learning_rate": 2.8000000000000003e-06, + "loss": 1.6553, + "step": 56 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 3.664458990097046, + "learning_rate": 2.85e-06, + "loss": 1.5, + "step": 57 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 7.763802528381348, + "learning_rate": 2.9e-06, + "loss": 1.367, + "step": 58 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 3.2216155529022217, + "learning_rate": 2.95e-06, + "loss": 1.3863, + "step": 59 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 4.384445667266846, + "learning_rate": 3e-06, + "loss": 1.4247, + "step": 60 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 4.8080878257751465, + "learning_rate": 3.05e-06, + "loss": 1.3257, + "step": 61 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 4.154761791229248, + "learning_rate": 3.1000000000000004e-06, + "loss": 1.321, + "step": 62 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 6.4742112159729, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.2823, + "step": 63 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 2.583422899246216, + "learning_rate": 3.2000000000000003e-06, + "loss": 1.2136, + "step": 64 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 4.1933488845825195, + "learning_rate": 3.2500000000000002e-06, + "loss": 1.1855, + "step": 65 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 4.11049747467041, + "learning_rate": 3.3000000000000006e-06, + "loss": 1.2389, + "step": 66 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 2.264458417892456, + "learning_rate": 3.3500000000000005e-06, + "loss": 1.0651, + "step": 67 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 2.5408174991607666, + "learning_rate": 3.4000000000000005e-06, + "loss": 1.1389, + "step": 68 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 7.82421350479126, + "learning_rate": 3.45e-06, + "loss": 1.0956, + "step": 69 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 3.070939064025879, + "learning_rate": 3.5e-06, + "loss": 1.0451, + "step": 70 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 2.6310527324676514, + "learning_rate": 3.5500000000000003e-06, + "loss": 1.0538, + "step": 71 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 7.630155563354492, + "learning_rate": 3.6000000000000003e-06, + "loss": 1.0052, + "step": 72 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 6.950636863708496, + "learning_rate": 3.65e-06, + "loss": 1.0473, + "step": 73 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 2.2703945636749268, + "learning_rate": 3.7e-06, + "loss": 1.0576, + "step": 74 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 3.3817710876464844, + "learning_rate": 3.7500000000000005e-06, + "loss": 1.0177, + "step": 75 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 7.266414642333984, + "learning_rate": 3.8000000000000005e-06, + "loss": 1.0645, + "step": 76 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 5.782608509063721, + "learning_rate": 3.85e-06, + "loss": 1.0162, + "step": 77 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 2.7938575744628906, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9664, + "step": 78 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 6.681935787200928, + "learning_rate": 3.95e-06, + "loss": 0.953, + "step": 79 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 2.253279209136963, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9568, + "step": 80 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 1.4875826835632324, + "learning_rate": 4.05e-06, + "loss": 0.9448, + "step": 81 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 2.4987940788269043, + "learning_rate": 4.1e-06, + "loss": 0.9393, + "step": 82 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 4.712948322296143, + "learning_rate": 4.15e-06, + "loss": 0.9532, + "step": 83 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 6.9030632972717285, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.96, + "step": 84 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 3.4780967235565186, + "learning_rate": 4.25e-06, + "loss": 0.8993, + "step": 85 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.526064395904541, + "learning_rate": 4.3e-06, + "loss": 0.9021, + "step": 86 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 10.727686882019043, + "learning_rate": 4.350000000000001e-06, + "loss": 0.856, + "step": 87 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 12.483160972595215, + "learning_rate": 4.4e-06, + "loss": 0.9357, + "step": 88 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 6.544492244720459, + "learning_rate": 4.450000000000001e-06, + "loss": 0.9168, + "step": 89 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 1.178139567375183, + "learning_rate": 4.5e-06, + "loss": 0.8748, + "step": 90 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 1.711506962776184, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8425, + "step": 91 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 3.281747341156006, + "learning_rate": 4.600000000000001e-06, + "loss": 0.8491, + "step": 92 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 2.2964377403259277, + "learning_rate": 4.65e-06, + "loss": 0.8038, + "step": 93 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 1.959700345993042, + "learning_rate": 4.7e-06, + "loss": 0.8439, + "step": 94 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 3.979384183883667, + "learning_rate": 4.75e-06, + "loss": 0.8839, + "step": 95 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 1.4721262454986572, + "learning_rate": 4.800000000000001e-06, + "loss": 0.845, + "step": 96 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 2.862248659133911, + "learning_rate": 4.85e-06, + "loss": 0.7748, + "step": 97 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 3.7439088821411133, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8145, + "step": 98 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 1.6654618978500366, + "learning_rate": 4.95e-06, + "loss": 0.8326, + "step": 99 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 7.8437581062316895, + "learning_rate": 5e-06, + "loss": 0.8666, + "step": 100 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 6.429738521575928, + "learning_rate": 4.999996250830422e-06, + "loss": 0.836, + "step": 101 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 2.6017794609069824, + "learning_rate": 4.9999850033329326e-06, + "loss": 0.7785, + "step": 102 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 1.0575449466705322, + "learning_rate": 4.999966257541265e-06, + "loss": 0.7639, + "step": 103 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 2.6932010650634766, + "learning_rate": 4.999940013511647e-06, + "loss": 0.8214, + "step": 104 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 2.925288438796997, + "learning_rate": 4.999906271322792e-06, + "loss": 0.8797, + "step": 105 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.3570607900619507, + "learning_rate": 4.9998650310759035e-06, + "loss": 0.792, + "step": 106 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 5.126713752746582, + "learning_rate": 4.999816292894676e-06, + "loss": 0.8352, + "step": 107 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 1.8966432809829712, + "learning_rate": 4.99976005692529e-06, + "loss": 0.7663, + "step": 108 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 1.3100829124450684, + "learning_rate": 4.999696323336418e-06, + "loss": 0.771, + "step": 109 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 2.4025354385375977, + "learning_rate": 4.999625092319218e-06, + "loss": 0.7618, + "step": 110 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.130232810974121, + "learning_rate": 4.999546364087334e-06, + "loss": 0.7705, + "step": 111 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 3.430262327194214, + "learning_rate": 4.999460138876901e-06, + "loss": 0.77, + "step": 112 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 1.1272103786468506, + "learning_rate": 4.999366416946536e-06, + "loss": 0.7133, + "step": 113 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 1.1740471124649048, + "learning_rate": 4.999265198577342e-06, + "loss": 0.7684, + "step": 114 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.3138248920440674, + "learning_rate": 4.999156484072907e-06, + "loss": 0.7888, + "step": 115 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 1.061711311340332, + "learning_rate": 4.999040273759304e-06, + "loss": 0.7484, + "step": 116 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.4682390689849854, + "learning_rate": 4.998916567985083e-06, + "loss": 0.7296, + "step": 117 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 2.884068250656128, + "learning_rate": 4.998785367121284e-06, + "loss": 0.7662, + "step": 118 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9812761545181274, + "learning_rate": 4.9986466715614205e-06, + "loss": 0.7307, + "step": 119 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 2.2237496376037598, + "learning_rate": 4.998500481721484e-06, + "loss": 0.6761, + "step": 120 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 1.4004178047180176, + "learning_rate": 4.998346798039952e-06, + "loss": 0.7505, + "step": 121 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 5.54975700378418, + "learning_rate": 4.99818562097777e-06, + "loss": 0.7615, + "step": 122 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 6.17140531539917, + "learning_rate": 4.9980169510183624e-06, + "loss": 0.7002, + "step": 123 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 4.974380016326904, + "learning_rate": 4.997840788667628e-06, + "loss": 0.7449, + "step": 124 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 1.4133399724960327, + "learning_rate": 4.997657134453937e-06, + "loss": 0.7442, + "step": 125 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 1.868915319442749, + "learning_rate": 4.9974659889281295e-06, + "loss": 0.7104, + "step": 126 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 1.2599350214004517, + "learning_rate": 4.997267352663514e-06, + "loss": 0.7385, + "step": 127 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.4353641271591187, + "learning_rate": 4.997061226255869e-06, + "loss": 0.7081, + "step": 128 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 3.2492141723632812, + "learning_rate": 4.996847610323437e-06, + "loss": 0.7859, + "step": 129 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 9.599719047546387, + "learning_rate": 4.996626505506923e-06, + "loss": 0.7241, + "step": 130 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 10.053650856018066, + "learning_rate": 4.996397912469494e-06, + "loss": 0.6841, + "step": 131 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 1.323876976966858, + "learning_rate": 4.996161831896777e-06, + "loss": 0.7317, + "step": 132 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 1.4180598258972168, + "learning_rate": 4.9959182644968594e-06, + "loss": 0.692, + "step": 133 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 1.2194396257400513, + "learning_rate": 4.99566721100028e-06, + "loss": 0.7068, + "step": 134 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0984960794448853, + "learning_rate": 4.995408672160031e-06, + "loss": 0.6946, + "step": 135 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 1.9341071844100952, + "learning_rate": 4.995142648751561e-06, + "loss": 0.7467, + "step": 136 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 1.9960932731628418, + "learning_rate": 4.9948691415727594e-06, + "loss": 0.7379, + "step": 137 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.8743917942047119, + "learning_rate": 4.994588151443968e-06, + "loss": 0.66, + "step": 138 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8655261993408203, + "learning_rate": 4.99429967920797e-06, + "loss": 0.6646, + "step": 139 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 5.462070941925049, + "learning_rate": 4.994003725729992e-06, + "loss": 0.643, + "step": 140 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 2.1401469707489014, + "learning_rate": 4.993700291897695e-06, + "loss": 0.6639, + "step": 141 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 1.8219833374023438, + "learning_rate": 4.9933893786211815e-06, + "loss": 0.6673, + "step": 142 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 1.641079306602478, + "learning_rate": 4.993070986832984e-06, + "loss": 0.658, + "step": 143 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 1.1739819049835205, + "learning_rate": 4.992745117488066e-06, + "loss": 0.6826, + "step": 144 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 2.309185743331909, + "learning_rate": 4.9924117715638185e-06, + "loss": 0.6536, + "step": 145 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 1.09304940700531, + "learning_rate": 4.99207095006006e-06, + "loss": 0.721, + "step": 146 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.9056984186172485, + "learning_rate": 4.991722653999025e-06, + "loss": 0.7019, + "step": 147 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.8440625667572021, + "learning_rate": 4.991366884425374e-06, + "loss": 0.707, + "step": 148 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 1.2244676351547241, + "learning_rate": 4.991003642406177e-06, + "loss": 0.6407, + "step": 149 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9258589744567871, + "learning_rate": 4.99063292903092e-06, + "loss": 0.6954, + "step": 150 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 4.176390647888184, + "learning_rate": 4.990254745411496e-06, + "loss": 0.6812, + "step": 151 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 1.4322530031204224, + "learning_rate": 4.989869092682205e-06, + "loss": 0.6808, + "step": 152 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.8017717003822327, + "learning_rate": 4.989475971999748e-06, + "loss": 0.687, + "step": 153 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 1.5641374588012695, + "learning_rate": 4.989075384543228e-06, + "loss": 0.6599, + "step": 154 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 1.1522141695022583, + "learning_rate": 4.98866733151414e-06, + "loss": 0.6546, + "step": 155 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8593171238899231, + "learning_rate": 4.988251814136372e-06, + "loss": 0.6857, + "step": 156 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 2.668159246444702, + "learning_rate": 4.9878288336562e-06, + "loss": 0.661, + "step": 157 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9953671097755432, + "learning_rate": 4.987398391342285e-06, + "loss": 0.6512, + "step": 158 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.042872667312622, + "learning_rate": 4.986960488485667e-06, + "loss": 0.6311, + "step": 159 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.9070663452148438, + "learning_rate": 4.9865151263997645e-06, + "loss": 0.675, + "step": 160 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.8460433483123779, + "learning_rate": 4.986062306420367e-06, + "loss": 0.6635, + "step": 161 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 1.2639834880828857, + "learning_rate": 4.985602029905635e-06, + "loss": 0.6327, + "step": 162 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.8775074481964111, + "learning_rate": 4.985134298236091e-06, + "loss": 0.644, + "step": 163 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 1.2031961679458618, + "learning_rate": 4.98465911281462e-06, + "loss": 0.6254, + "step": 164 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.892494797706604, + "learning_rate": 4.984176475066463e-06, + "loss": 0.7122, + "step": 165 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 2.7122485637664795, + "learning_rate": 4.983686386439212e-06, + "loss": 0.6679, + "step": 166 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9344426989555359, + "learning_rate": 4.983188848402806e-06, + "loss": 0.6319, + "step": 167 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 1.4093577861785889, + "learning_rate": 4.982683862449531e-06, + "loss": 0.6425, + "step": 168 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 1.1285009384155273, + "learning_rate": 4.982171430094007e-06, + "loss": 0.6298, + "step": 169 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.952778935432434, + "learning_rate": 4.981651552873193e-06, + "loss": 0.7066, + "step": 170 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 5.133765697479248, + "learning_rate": 4.981124232346374e-06, + "loss": 0.6634, + "step": 171 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9770542979240417, + "learning_rate": 4.980589470095161e-06, + "loss": 0.7121, + "step": 172 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.8414323925971985, + "learning_rate": 4.980047267723487e-06, + "loss": 0.6397, + "step": 173 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 1.9173879623413086, + "learning_rate": 4.979497626857596e-06, + "loss": 0.6228, + "step": 174 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 1.0823363065719604, + "learning_rate": 4.978940549146048e-06, + "loss": 0.6475, + "step": 175 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 3.715353488922119, + "learning_rate": 4.978376036259706e-06, + "loss": 0.7127, + "step": 176 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.981584370136261, + "learning_rate": 4.9778040898917325e-06, + "loss": 0.6468, + "step": 177 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.70566987991333, + "learning_rate": 4.977224711757587e-06, + "loss": 0.6476, + "step": 178 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9217923283576965, + "learning_rate": 4.976637903595019e-06, + "loss": 0.6731, + "step": 179 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.8994677662849426, + "learning_rate": 4.976043667164063e-06, + "loss": 0.6562, + "step": 180 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 1.1613017320632935, + "learning_rate": 4.975442004247034e-06, + "loss": 0.6417, + "step": 181 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 1.6041977405548096, + "learning_rate": 4.974832916648521e-06, + "loss": 0.6029, + "step": 182 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.7978405952453613, + "learning_rate": 4.974216406195383e-06, + "loss": 0.6269, + "step": 183 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.6021920442581177, + "learning_rate": 4.973592474736739e-06, + "loss": 0.6149, + "step": 184 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8973568677902222, + "learning_rate": 4.972961124143971e-06, + "loss": 0.6648, + "step": 185 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 1.9432591199874878, + "learning_rate": 4.972322356310711e-06, + "loss": 0.6299, + "step": 186 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 4.457028388977051, + "learning_rate": 4.971676173152839e-06, + "loss": 0.656, + "step": 187 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 2.0989716053009033, + "learning_rate": 4.971022576608473e-06, + "loss": 0.6539, + "step": 188 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 1.0646967887878418, + "learning_rate": 4.97036156863797e-06, + "loss": 0.6727, + "step": 189 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.6522265672683716, + "learning_rate": 4.969693151223914e-06, + "loss": 0.6643, + "step": 190 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.7503505945205688, + "learning_rate": 4.969017326371115e-06, + "loss": 0.6402, + "step": 191 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 1.2341989278793335, + "learning_rate": 4.968334096106597e-06, + "loss": 0.6413, + "step": 192 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 3.089054584503174, + "learning_rate": 4.967643462479597e-06, + "loss": 0.6825, + "step": 193 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 2.711623430252075, + "learning_rate": 4.966945427561557e-06, + "loss": 0.65, + "step": 194 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 4.641184329986572, + "learning_rate": 4.966239993446118e-06, + "loss": 0.6229, + "step": 195 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.7984074354171753, + "learning_rate": 4.965527162249114e-06, + "loss": 0.6473, + "step": 196 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 1.1643115282058716, + "learning_rate": 4.964806936108566e-06, + "loss": 0.6404, + "step": 197 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 2.1877920627593994, + "learning_rate": 4.9640793171846725e-06, + "loss": 0.6185, + "step": 198 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.7970566749572754, + "learning_rate": 4.963344307659807e-06, + "loss": 0.634, + "step": 199 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 1.6014361381530762, + "learning_rate": 4.96260190973851e-06, + "loss": 0.6562, + "step": 200 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.8743320107460022, + "learning_rate": 4.961852125647482e-06, + "loss": 0.6133, + "step": 201 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 1.9526551961898804, + "learning_rate": 4.961094957635578e-06, + "loss": 0.6451, + "step": 202 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 3.6597347259521484, + "learning_rate": 4.960330407973798e-06, + "loss": 0.6386, + "step": 203 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 1.7180207967758179, + "learning_rate": 4.959558478955283e-06, + "loss": 0.6688, + "step": 204 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 0.9058470129966736, + "learning_rate": 4.958779172895308e-06, + "loss": 0.6161, + "step": 205 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 1.0031033754348755, + "learning_rate": 4.957992492131274e-06, + "loss": 0.6437, + "step": 206 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 1.5846725702285767, + "learning_rate": 4.9571984390226985e-06, + "loss": 0.6332, + "step": 207 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 1.9951609373092651, + "learning_rate": 4.956397015951215e-06, + "loss": 0.636, + "step": 208 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 1.4122583866119385, + "learning_rate": 4.95558822532056e-06, + "loss": 0.6586, + "step": 209 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.2243481874465942, + "learning_rate": 4.954772069556568e-06, + "loss": 0.6313, + "step": 210 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.8756356835365295, + "learning_rate": 4.953948551107164e-06, + "loss": 0.6406, + "step": 211 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 2.9979734420776367, + "learning_rate": 4.953117672442356e-06, + "loss": 0.5803, + "step": 212 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 2.1859359741210938, + "learning_rate": 4.952279436054229e-06, + "loss": 0.6607, + "step": 213 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.6929755806922913, + "learning_rate": 4.9514338444569346e-06, + "loss": 0.5989, + "step": 214 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 1.0361783504486084, + "learning_rate": 4.950580900186685e-06, + "loss": 0.6654, + "step": 215 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 1.210898518562317, + "learning_rate": 4.9497206058017475e-06, + "loss": 0.6213, + "step": 216 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 1.200990080833435, + "learning_rate": 4.948852963882434e-06, + "loss": 0.6654, + "step": 217 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 1.481831669807434, + "learning_rate": 4.947977977031093e-06, + "loss": 0.6474, + "step": 218 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.9883334636688232, + "learning_rate": 4.947095647872103e-06, + "loss": 0.6735, + "step": 219 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.7436536550521851, + "learning_rate": 4.946205979051868e-06, + "loss": 0.6456, + "step": 220 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9057570099830627, + "learning_rate": 4.945308973238802e-06, + "loss": 0.6228, + "step": 221 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 1.341081142425537, + "learning_rate": 4.944404633123324e-06, + "loss": 0.6417, + "step": 222 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.7958157062530518, + "learning_rate": 4.943492961417859e-06, + "loss": 0.6494, + "step": 223 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.216025471687317, + "learning_rate": 4.9425739608568106e-06, + "loss": 0.6566, + "step": 224 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.9774854779243469, + "learning_rate": 4.9416476341965735e-06, + "loss": 0.6171, + "step": 225 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 2.1562681198120117, + "learning_rate": 4.940713984215512e-06, + "loss": 0.629, + "step": 226 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 1.9521286487579346, + "learning_rate": 4.9397730137139556e-06, + "loss": 0.6475, + "step": 227 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 1.5749104022979736, + "learning_rate": 4.9388247255141895e-06, + "loss": 0.6053, + "step": 228 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 1.2008254528045654, + "learning_rate": 4.937869122460449e-06, + "loss": 0.6052, + "step": 229 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 1.0774102210998535, + "learning_rate": 4.93690620741891e-06, + "loss": 0.6099, + "step": 230 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 1.0929996967315674, + "learning_rate": 4.935935983277675e-06, + "loss": 0.6363, + "step": 231 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.8830653429031372, + "learning_rate": 4.934958452946774e-06, + "loss": 0.6136, + "step": 232 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 3.591218948364258, + "learning_rate": 4.933973619358147e-06, + "loss": 0.5962, + "step": 233 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 2.5797672271728516, + "learning_rate": 4.932981485465643e-06, + "loss": 0.6405, + "step": 234 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 1.0467664003372192, + "learning_rate": 4.9319820542450025e-06, + "loss": 0.6155, + "step": 235 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.8099795579910278, + "learning_rate": 4.930975328693856e-06, + "loss": 0.5615, + "step": 236 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.8906702995300293, + "learning_rate": 4.92996131183171e-06, + "loss": 0.6501, + "step": 237 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 1.0871416330337524, + "learning_rate": 4.928940006699944e-06, + "loss": 0.6282, + "step": 238 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 1.3209614753723145, + "learning_rate": 4.927911416361792e-06, + "loss": 0.598, + "step": 239 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.2252682447433472, + "learning_rate": 4.926875543902344e-06, + "loss": 0.6433, + "step": 240 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0569007396697998, + "learning_rate": 4.9258323924285285e-06, + "loss": 0.5927, + "step": 241 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9309014081954956, + "learning_rate": 4.924781965069106e-06, + "loss": 0.5927, + "step": 242 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0200378894805908, + "learning_rate": 4.923724264974662e-06, + "loss": 0.6064, + "step": 243 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 1.0533075332641602, + "learning_rate": 4.922659295317593e-06, + "loss": 0.6373, + "step": 244 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 0.7889382839202881, + "learning_rate": 4.921587059292102e-06, + "loss": 0.5887, + "step": 245 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.7943588495254517, + "learning_rate": 4.920507560114183e-06, + "loss": 0.593, + "step": 246 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.8247205018997192, + "learning_rate": 4.919420801021617e-06, + "loss": 0.6151, + "step": 247 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 0.9979158043861389, + "learning_rate": 4.91832678527396e-06, + "loss": 0.6019, + "step": 248 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9346868991851807, + "learning_rate": 4.917225516152532e-06, + "loss": 0.6098, + "step": 249 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.7487881183624268, + "learning_rate": 4.916116996960408e-06, + "loss": 0.5965, + "step": 250 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.821576714515686, + "learning_rate": 4.915001231022411e-06, + "loss": 0.6483, + "step": 251 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 1.0413196086883545, + "learning_rate": 4.913878221685096e-06, + "loss": 0.6108, + "step": 252 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9560331702232361, + "learning_rate": 4.912747972316745e-06, + "loss": 0.5758, + "step": 253 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.8964638113975525, + "learning_rate": 4.911610486307356e-06, + "loss": 0.6432, + "step": 254 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.8418346047401428, + "learning_rate": 4.910465767068631e-06, + "loss": 0.6027, + "step": 255 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 1.792371153831482, + "learning_rate": 4.909313818033966e-06, + "loss": 0.6198, + "step": 256 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 1.036665439605713, + "learning_rate": 4.908154642658446e-06, + "loss": 0.6255, + "step": 257 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.7592151165008545, + "learning_rate": 4.906988244418823e-06, + "loss": 0.6035, + "step": 258 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.8843073844909668, + "learning_rate": 4.90581462681352e-06, + "loss": 0.6299, + "step": 259 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9489964246749878, + "learning_rate": 4.9046337933626086e-06, + "loss": 0.5869, + "step": 260 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.851691722869873, + "learning_rate": 4.903445747607806e-06, + "loss": 0.603, + "step": 261 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 1.3722106218338013, + "learning_rate": 4.902250493112458e-06, + "loss": 0.5939, + "step": 262 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 1.1002827882766724, + "learning_rate": 4.901048033461537e-06, + "loss": 0.6452, + "step": 263 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 0.8428632020950317, + "learning_rate": 4.89983837226162e-06, + "loss": 0.5956, + "step": 264 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.7666584849357605, + "learning_rate": 4.898621513140889e-06, + "loss": 0.6067, + "step": 265 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.8413611054420471, + "learning_rate": 4.897397459749113e-06, + "loss": 0.5985, + "step": 266 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 2.3374335765838623, + "learning_rate": 4.896166215757638e-06, + "loss": 0.5885, + "step": 267 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 2.236640214920044, + "learning_rate": 4.894927784859377e-06, + "loss": 0.6408, + "step": 268 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.9715856313705444, + "learning_rate": 4.893682170768802e-06, + "loss": 0.5954, + "step": 269 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0249912738800049, + "learning_rate": 4.892429377221928e-06, + "loss": 0.6186, + "step": 270 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 1.255426049232483, + "learning_rate": 4.891169407976302e-06, + "loss": 0.6351, + "step": 271 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.9339559674263, + "learning_rate": 4.889902266810995e-06, + "loss": 0.5944, + "step": 272 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 1.2473429441452026, + "learning_rate": 4.888627957526589e-06, + "loss": 0.544, + "step": 273 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 1.0589442253112793, + "learning_rate": 4.887346483945166e-06, + "loss": 0.5543, + "step": 274 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9844024777412415, + "learning_rate": 4.886057849910294e-06, + "loss": 0.5941, + "step": 275 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 2.88578200340271, + "learning_rate": 4.8847620592870196e-06, + "loss": 0.6124, + "step": 276 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 0.7496054172515869, + "learning_rate": 4.8834591159618524e-06, + "loss": 0.6006, + "step": 277 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 0.7403052449226379, + "learning_rate": 4.88214902384276e-06, + "loss": 0.5911, + "step": 278 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9003771543502808, + "learning_rate": 4.880831786859146e-06, + "loss": 0.6347, + "step": 279 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0345501899719238, + "learning_rate": 4.879507408961847e-06, + "loss": 0.6111, + "step": 280 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.4385879039764404, + "learning_rate": 4.878175894123116e-06, + "loss": 0.6454, + "step": 281 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.8469482064247131, + "learning_rate": 4.8768372463366145e-06, + "loss": 0.6163, + "step": 282 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.8859589695930481, + "learning_rate": 4.875491469617395e-06, + "loss": 0.6144, + "step": 283 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 1.8436834812164307, + "learning_rate": 4.874138568001895e-06, + "loss": 0.6275, + "step": 284 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.6646101474761963, + "learning_rate": 4.87277854554792e-06, + "loss": 0.615, + "step": 285 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 1.0070925951004028, + "learning_rate": 4.871411406334633e-06, + "loss": 0.5898, + "step": 286 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9785194993019104, + "learning_rate": 4.870037154462545e-06, + "loss": 0.5992, + "step": 287 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.7244889736175537, + "learning_rate": 4.868655794053497e-06, + "loss": 0.6078, + "step": 288 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.4496444463729858, + "learning_rate": 4.8672673292506535e-06, + "loss": 0.5855, + "step": 289 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 1.8514957427978516, + "learning_rate": 4.865871764218486e-06, + "loss": 0.5707, + "step": 290 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.8439773321151733, + "learning_rate": 4.864469103142763e-06, + "loss": 0.5562, + "step": 291 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.8146086931228638, + "learning_rate": 4.8630593502305355e-06, + "loss": 0.6161, + "step": 292 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.8920315504074097, + "learning_rate": 4.861642509710126e-06, + "loss": 0.6139, + "step": 293 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.4980088472366333, + "learning_rate": 4.860218585831116e-06, + "loss": 0.6187, + "step": 294 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 0.9910127520561218, + "learning_rate": 4.8587875828643285e-06, + "loss": 0.5852, + "step": 295 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.819600522518158, + "learning_rate": 4.857349505101823e-06, + "loss": 0.6172, + "step": 296 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 1.1059772968292236, + "learning_rate": 4.855904356856878e-06, + "loss": 0.5868, + "step": 297 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 1.2362196445465088, + "learning_rate": 4.854452142463977e-06, + "loss": 0.625, + "step": 298 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.9956470727920532, + "learning_rate": 4.852992866278799e-06, + "loss": 0.5923, + "step": 299 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 0.864109218120575, + "learning_rate": 4.851526532678203e-06, + "loss": 0.6315, + "step": 300 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 0.8900614380836487, + "learning_rate": 4.850053146060217e-06, + "loss": 0.6128, + "step": 301 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.927254855632782, + "learning_rate": 4.84857271084402e-06, + "loss": 0.5955, + "step": 302 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0046517848968506, + "learning_rate": 4.847085231469935e-06, + "loss": 0.6134, + "step": 303 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.734597384929657, + "learning_rate": 4.8455907123994125e-06, + "loss": 0.5927, + "step": 304 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.7338348031044006, + "learning_rate": 4.844089158115016e-06, + "loss": 0.5897, + "step": 305 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9163988828659058, + "learning_rate": 4.8425805731204106e-06, + "loss": 0.6051, + "step": 306 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 1.050246238708496, + "learning_rate": 4.84106496194035e-06, + "loss": 0.5751, + "step": 307 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 0.7637603878974915, + "learning_rate": 4.83954232912066e-06, + "loss": 0.5677, + "step": 308 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.7110525965690613, + "learning_rate": 4.838012679228229e-06, + "loss": 0.6051, + "step": 309 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.7662068605422974, + "learning_rate": 4.836476016850988e-06, + "loss": 0.59, + "step": 310 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 0.8907375335693359, + "learning_rate": 4.834932346597906e-06, + "loss": 0.5792, + "step": 311 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.8939849138259888, + "learning_rate": 4.833381673098966e-06, + "loss": 0.6062, + "step": 312 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 0.8878788948059082, + "learning_rate": 4.8318240010051595e-06, + "loss": 0.5694, + "step": 313 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 1.2523870468139648, + "learning_rate": 4.830259334988468e-06, + "loss": 0.5809, + "step": 314 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 1.0836797952651978, + "learning_rate": 4.82868767974185e-06, + "loss": 0.5949, + "step": 315 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.7985473871231079, + "learning_rate": 4.827109039979226e-06, + "loss": 0.6057, + "step": 316 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 1.042951226234436, + "learning_rate": 4.825523420435469e-06, + "loss": 0.6004, + "step": 317 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.7845115661621094, + "learning_rate": 4.823930825866381e-06, + "loss": 0.6161, + "step": 318 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.931854784488678, + "learning_rate": 4.82233126104869e-06, + "loss": 0.5912, + "step": 319 + }, + { + "epoch": 1.0, + "grad_norm": 0.931854784488678, + "learning_rate": 4.8207247307800275e-06, + "loss": 0.1914, + "step": 320 + }, + { + "epoch": 1.0031315240083507, + "grad_norm": 0.751028835773468, + "learning_rate": 4.819111239878916e-06, + "loss": 0.5802, + "step": 321 + }, + { + "epoch": 1.0062630480167014, + "grad_norm": 1.4943569898605347, + "learning_rate": 4.817490793184758e-06, + "loss": 0.613, + "step": 322 + }, + { + "epoch": 1.0093945720250521, + "grad_norm": 2.296318531036377, + "learning_rate": 4.815863395557816e-06, + "loss": 0.5453, + "step": 323 + }, + { + "epoch": 1.0125260960334028, + "grad_norm": 0.760101318359375, + "learning_rate": 4.814229051879202e-06, + "loss": 0.5302, + "step": 324 + }, + { + "epoch": 1.0156576200417538, + "grad_norm": 0.8145846128463745, + "learning_rate": 4.812587767050861e-06, + "loss": 0.5831, + "step": 325 + }, + { + "epoch": 1.0187891440501045, + "grad_norm": 0.9169796109199524, + "learning_rate": 4.8109395459955565e-06, + "loss": 0.5756, + "step": 326 + }, + { + "epoch": 1.0219206680584552, + "grad_norm": 0.8791524171829224, + "learning_rate": 4.809284393656858e-06, + "loss": 0.5988, + "step": 327 + }, + { + "epoch": 1.0250521920668059, + "grad_norm": 1.0184170007705688, + "learning_rate": 4.807622314999122e-06, + "loss": 0.5476, + "step": 328 + }, + { + "epoch": 1.0281837160751566, + "grad_norm": 0.8095184564590454, + "learning_rate": 4.8059533150074805e-06, + "loss": 0.5723, + "step": 329 + }, + { + "epoch": 1.0313152400835073, + "grad_norm": 0.7621930241584778, + "learning_rate": 4.804277398687826e-06, + "loss": 0.5841, + "step": 330 + }, + { + "epoch": 1.034446764091858, + "grad_norm": 3.729628324508667, + "learning_rate": 4.802594571066791e-06, + "loss": 0.5639, + "step": 331 + }, + { + "epoch": 1.0375782881002087, + "grad_norm": 1.6502974033355713, + "learning_rate": 4.800904837191743e-06, + "loss": 0.6024, + "step": 332 + }, + { + "epoch": 1.0407098121085594, + "grad_norm": 0.8031198978424072, + "learning_rate": 4.799208202130762e-06, + "loss": 0.5305, + "step": 333 + }, + { + "epoch": 1.0438413361169103, + "grad_norm": 0.939644992351532, + "learning_rate": 4.797504670972623e-06, + "loss": 0.5446, + "step": 334 + }, + { + "epoch": 1.046972860125261, + "grad_norm": 1.0589954853057861, + "learning_rate": 4.795794248826789e-06, + "loss": 0.5366, + "step": 335 + }, + { + "epoch": 1.0501043841336117, + "grad_norm": 0.9089614748954773, + "learning_rate": 4.794076940823391e-06, + "loss": 0.5795, + "step": 336 + }, + { + "epoch": 1.0532359081419624, + "grad_norm": 0.7732561230659485, + "learning_rate": 4.792352752113212e-06, + "loss": 0.5765, + "step": 337 + }, + { + "epoch": 1.0563674321503131, + "grad_norm": 1.811553955078125, + "learning_rate": 4.790621687867672e-06, + "loss": 0.561, + "step": 338 + }, + { + "epoch": 1.0594989561586639, + "grad_norm": 1.1930758953094482, + "learning_rate": 4.788883753278813e-06, + "loss": 0.5, + "step": 339 + }, + { + "epoch": 1.0626304801670146, + "grad_norm": 0.9551813006401062, + "learning_rate": 4.787138953559285e-06, + "loss": 0.5228, + "step": 340 + }, + { + "epoch": 1.0657620041753653, + "grad_norm": 0.9609586596488953, + "learning_rate": 4.785387293942329e-06, + "loss": 0.5827, + "step": 341 + }, + { + "epoch": 1.068893528183716, + "grad_norm": 0.8403449654579163, + "learning_rate": 4.783628779681759e-06, + "loss": 0.5585, + "step": 342 + }, + { + "epoch": 1.072025052192067, + "grad_norm": 0.9108251929283142, + "learning_rate": 4.7818634160519496e-06, + "loss": 0.6077, + "step": 343 + }, + { + "epoch": 1.0751565762004176, + "grad_norm": 0.9476898908615112, + "learning_rate": 4.780091208347819e-06, + "loss": 0.5493, + "step": 344 + }, + { + "epoch": 1.0782881002087683, + "grad_norm": 1.1943707466125488, + "learning_rate": 4.778312161884813e-06, + "loss": 0.5736, + "step": 345 + }, + { + "epoch": 1.081419624217119, + "grad_norm": 3.1342639923095703, + "learning_rate": 4.77652628199889e-06, + "loss": 0.5765, + "step": 346 + }, + { + "epoch": 1.0845511482254697, + "grad_norm": 2.7982125282287598, + "learning_rate": 4.7747335740465015e-06, + "loss": 0.6003, + "step": 347 + }, + { + "epoch": 1.0876826722338204, + "grad_norm": 1.5068914890289307, + "learning_rate": 4.7729340434045815e-06, + "loss": 0.5033, + "step": 348 + }, + { + "epoch": 1.0908141962421711, + "grad_norm": 0.8273429274559021, + "learning_rate": 4.771127695470527e-06, + "loss": 0.5309, + "step": 349 + }, + { + "epoch": 1.0939457202505218, + "grad_norm": 1.104974389076233, + "learning_rate": 4.76931453566218e-06, + "loss": 0.5244, + "step": 350 + }, + { + "epoch": 1.0970772442588728, + "grad_norm": 1.096509337425232, + "learning_rate": 4.7674945694178166e-06, + "loss": 0.5585, + "step": 351 + }, + { + "epoch": 1.1002087682672235, + "grad_norm": 1.0238200426101685, + "learning_rate": 4.765667802196127e-06, + "loss": 0.5589, + "step": 352 + }, + { + "epoch": 1.1033402922755742, + "grad_norm": 0.7515526413917542, + "learning_rate": 4.763834239476197e-06, + "loss": 0.5304, + "step": 353 + }, + { + "epoch": 1.1064718162839249, + "grad_norm": 1.0282566547393799, + "learning_rate": 4.761993886757499e-06, + "loss": 0.5476, + "step": 354 + }, + { + "epoch": 1.1096033402922756, + "grad_norm": 0.9962708950042725, + "learning_rate": 4.760146749559868e-06, + "loss": 0.5117, + "step": 355 + }, + { + "epoch": 1.1127348643006263, + "grad_norm": 0.7851671576499939, + "learning_rate": 4.758292833423488e-06, + "loss": 0.5542, + "step": 356 + }, + { + "epoch": 1.115866388308977, + "grad_norm": 0.8857759237289429, + "learning_rate": 4.756432143908876e-06, + "loss": 0.544, + "step": 357 + }, + { + "epoch": 1.1189979123173277, + "grad_norm": 0.9402740597724915, + "learning_rate": 4.7545646865968645e-06, + "loss": 0.5656, + "step": 358 + }, + { + "epoch": 1.1221294363256784, + "grad_norm": 0.8210407495498657, + "learning_rate": 4.752690467088584e-06, + "loss": 0.5733, + "step": 359 + }, + { + "epoch": 1.1252609603340291, + "grad_norm": 0.795684278011322, + "learning_rate": 4.750809491005449e-06, + "loss": 0.5678, + "step": 360 + }, + { + "epoch": 1.12839248434238, + "grad_norm": 0.8712463974952698, + "learning_rate": 4.748921763989139e-06, + "loss": 0.5777, + "step": 361 + }, + { + "epoch": 1.1315240083507307, + "grad_norm": 0.9810119867324829, + "learning_rate": 4.747027291701578e-06, + "loss": 0.5511, + "step": 362 + }, + { + "epoch": 1.1346555323590815, + "grad_norm": 0.81117844581604, + "learning_rate": 4.745126079824926e-06, + "loss": 0.5038, + "step": 363 + }, + { + "epoch": 1.1377870563674322, + "grad_norm": 0.7631494402885437, + "learning_rate": 4.743218134061556e-06, + "loss": 0.6272, + "step": 364 + }, + { + "epoch": 1.1409185803757829, + "grad_norm": 0.7601696252822876, + "learning_rate": 4.741303460134038e-06, + "loss": 0.571, + "step": 365 + }, + { + "epoch": 1.1440501043841336, + "grad_norm": 1.7977744340896606, + "learning_rate": 4.7393820637851205e-06, + "loss": 0.538, + "step": 366 + }, + { + "epoch": 1.1471816283924843, + "grad_norm": 2.022578001022339, + "learning_rate": 4.737453950777718e-06, + "loss": 0.5822, + "step": 367 + }, + { + "epoch": 1.150313152400835, + "grad_norm": 0.7586764693260193, + "learning_rate": 4.735519126894885e-06, + "loss": 0.5986, + "step": 368 + }, + { + "epoch": 1.153444676409186, + "grad_norm": 0.8970286846160889, + "learning_rate": 4.733577597939812e-06, + "loss": 0.542, + "step": 369 + }, + { + "epoch": 1.1565762004175366, + "grad_norm": 0.8546352982521057, + "learning_rate": 4.731629369735793e-06, + "loss": 0.5832, + "step": 370 + }, + { + "epoch": 1.1597077244258873, + "grad_norm": 0.9266164898872375, + "learning_rate": 4.72967444812622e-06, + "loss": 0.551, + "step": 371 + }, + { + "epoch": 1.162839248434238, + "grad_norm": 1.0413658618927002, + "learning_rate": 4.7277128389745595e-06, + "loss": 0.5866, + "step": 372 + }, + { + "epoch": 1.1659707724425887, + "grad_norm": 0.9312199950218201, + "learning_rate": 4.7257445481643334e-06, + "loss": 0.5723, + "step": 373 + }, + { + "epoch": 1.1691022964509394, + "grad_norm": 0.7389806509017944, + "learning_rate": 4.723769581599109e-06, + "loss": 0.5209, + "step": 374 + }, + { + "epoch": 1.1722338204592901, + "grad_norm": 3.053169012069702, + "learning_rate": 4.721787945202472e-06, + "loss": 0.6094, + "step": 375 + }, + { + "epoch": 1.1753653444676408, + "grad_norm": 1.288589596748352, + "learning_rate": 4.719799644918017e-06, + "loss": 0.5616, + "step": 376 + }, + { + "epoch": 1.1784968684759916, + "grad_norm": 0.7675042152404785, + "learning_rate": 4.717804686709323e-06, + "loss": 0.4963, + "step": 377 + }, + { + "epoch": 1.1816283924843423, + "grad_norm": 0.7246491312980652, + "learning_rate": 4.715803076559938e-06, + "loss": 0.5273, + "step": 378 + }, + { + "epoch": 1.1847599164926932, + "grad_norm": 0.8193361759185791, + "learning_rate": 4.713794820473366e-06, + "loss": 0.6107, + "step": 379 + }, + { + "epoch": 1.187891440501044, + "grad_norm": 0.9498510360717773, + "learning_rate": 4.711779924473037e-06, + "loss": 0.5421, + "step": 380 + }, + { + "epoch": 1.1910229645093946, + "grad_norm": 1.0479756593704224, + "learning_rate": 4.709758394602305e-06, + "loss": 0.5257, + "step": 381 + }, + { + "epoch": 1.1941544885177453, + "grad_norm": 0.907866895198822, + "learning_rate": 4.707730236924413e-06, + "loss": 0.5289, + "step": 382 + }, + { + "epoch": 1.197286012526096, + "grad_norm": 0.8861165642738342, + "learning_rate": 4.705695457522488e-06, + "loss": 0.5727, + "step": 383 + }, + { + "epoch": 1.2004175365344467, + "grad_norm": 0.7467761039733887, + "learning_rate": 4.703654062499516e-06, + "loss": 0.5602, + "step": 384 + }, + { + "epoch": 1.2035490605427974, + "grad_norm": 0.7456198334693909, + "learning_rate": 4.701606057978325e-06, + "loss": 0.5345, + "step": 385 + }, + { + "epoch": 1.2066805845511483, + "grad_norm": 1.9976060390472412, + "learning_rate": 4.699551450101571e-06, + "loss": 0.5504, + "step": 386 + }, + { + "epoch": 1.209812108559499, + "grad_norm": 1.5253807306289673, + "learning_rate": 4.697490245031709e-06, + "loss": 0.5568, + "step": 387 + }, + { + "epoch": 1.2129436325678498, + "grad_norm": 1.0786075592041016, + "learning_rate": 4.6954224489509885e-06, + "loss": 0.5564, + "step": 388 + }, + { + "epoch": 1.2160751565762005, + "grad_norm": 0.8385995030403137, + "learning_rate": 4.693348068061422e-06, + "loss": 0.5341, + "step": 389 + }, + { + "epoch": 1.2192066805845512, + "grad_norm": 0.8184949159622192, + "learning_rate": 4.691267108584774e-06, + "loss": 0.5614, + "step": 390 + }, + { + "epoch": 1.2223382045929019, + "grad_norm": 0.9964898824691772, + "learning_rate": 4.68917957676254e-06, + "loss": 0.5589, + "step": 391 + }, + { + "epoch": 1.2254697286012526, + "grad_norm": 1.0168914794921875, + "learning_rate": 4.687085478855931e-06, + "loss": 0.5892, + "step": 392 + }, + { + "epoch": 1.2286012526096033, + "grad_norm": 0.8841140866279602, + "learning_rate": 4.684984821145846e-06, + "loss": 0.5327, + "step": 393 + }, + { + "epoch": 1.231732776617954, + "grad_norm": 0.834431529045105, + "learning_rate": 4.682877609932866e-06, + "loss": 0.5594, + "step": 394 + }, + { + "epoch": 1.2348643006263047, + "grad_norm": 0.7256641983985901, + "learning_rate": 4.6807638515372234e-06, + "loss": 0.5443, + "step": 395 + }, + { + "epoch": 1.2379958246346556, + "grad_norm": 0.765096127986908, + "learning_rate": 4.678643552298788e-06, + "loss": 0.5439, + "step": 396 + }, + { + "epoch": 1.2411273486430063, + "grad_norm": 0.8760455846786499, + "learning_rate": 4.676516718577051e-06, + "loss": 0.5485, + "step": 397 + }, + { + "epoch": 1.244258872651357, + "grad_norm": 2.7111501693725586, + "learning_rate": 4.674383356751099e-06, + "loss": 0.5696, + "step": 398 + }, + { + "epoch": 1.2473903966597077, + "grad_norm": 1.0521738529205322, + "learning_rate": 4.672243473219601e-06, + "loss": 0.5503, + "step": 399 + }, + { + "epoch": 1.2505219206680585, + "grad_norm": 0.8909669518470764, + "learning_rate": 4.670097074400785e-06, + "loss": 0.5183, + "step": 400 + }, + { + "epoch": 1.2536534446764092, + "grad_norm": 0.7483847737312317, + "learning_rate": 4.667944166732424e-06, + "loss": 0.5669, + "step": 401 + }, + { + "epoch": 1.2567849686847599, + "grad_norm": 1.146997094154358, + "learning_rate": 4.665784756671808e-06, + "loss": 0.5464, + "step": 402 + }, + { + "epoch": 1.2599164926931106, + "grad_norm": 0.8998096585273743, + "learning_rate": 4.663618850695733e-06, + "loss": 0.5502, + "step": 403 + }, + { + "epoch": 1.2630480167014615, + "grad_norm": 0.8882688283920288, + "learning_rate": 4.6614464553004795e-06, + "loss": 0.5507, + "step": 404 + }, + { + "epoch": 1.2661795407098122, + "grad_norm": 0.8310684561729431, + "learning_rate": 4.659267577001789e-06, + "loss": 0.5164, + "step": 405 + }, + { + "epoch": 1.269311064718163, + "grad_norm": 0.9286114573478699, + "learning_rate": 4.657082222334851e-06, + "loss": 0.4813, + "step": 406 + }, + { + "epoch": 1.2724425887265136, + "grad_norm": 1.2394906282424927, + "learning_rate": 4.654890397854275e-06, + "loss": 0.5837, + "step": 407 + }, + { + "epoch": 1.2755741127348643, + "grad_norm": 4.00585412979126, + "learning_rate": 4.652692110134079e-06, + "loss": 0.5453, + "step": 408 + }, + { + "epoch": 1.278705636743215, + "grad_norm": 1.1667803525924683, + "learning_rate": 4.650487365767667e-06, + "loss": 0.5652, + "step": 409 + }, + { + "epoch": 1.2818371607515657, + "grad_norm": 0.9351289868354797, + "learning_rate": 4.648276171367807e-06, + "loss": 0.5576, + "step": 410 + }, + { + "epoch": 1.2849686847599164, + "grad_norm": 0.8107728958129883, + "learning_rate": 4.646058533566614e-06, + "loss": 0.5821, + "step": 411 + }, + { + "epoch": 1.2881002087682671, + "grad_norm": 0.7293011546134949, + "learning_rate": 4.643834459015525e-06, + "loss": 0.5363, + "step": 412 + }, + { + "epoch": 1.2912317327766178, + "grad_norm": 0.7550690770149231, + "learning_rate": 4.641603954385289e-06, + "loss": 0.53, + "step": 413 + }, + { + "epoch": 1.2943632567849686, + "grad_norm": 0.7626177072525024, + "learning_rate": 4.639367026365938e-06, + "loss": 0.5307, + "step": 414 + }, + { + "epoch": 1.2974947807933195, + "grad_norm": 1.0841096639633179, + "learning_rate": 4.637123681666769e-06, + "loss": 0.5162, + "step": 415 + }, + { + "epoch": 1.3006263048016702, + "grad_norm": 0.8814271092414856, + "learning_rate": 4.634873927016326e-06, + "loss": 0.5369, + "step": 416 + }, + { + "epoch": 1.303757828810021, + "grad_norm": 0.7402971982955933, + "learning_rate": 4.632617769162378e-06, + "loss": 0.5846, + "step": 417 + }, + { + "epoch": 1.3068893528183716, + "grad_norm": 0.8106061220169067, + "learning_rate": 4.6303552148719e-06, + "loss": 0.5289, + "step": 418 + }, + { + "epoch": 1.3100208768267223, + "grad_norm": 0.9241361618041992, + "learning_rate": 4.628086270931053e-06, + "loss": 0.5714, + "step": 419 + }, + { + "epoch": 1.313152400835073, + "grad_norm": 0.950332522392273, + "learning_rate": 4.625810944145159e-06, + "loss": 0.5817, + "step": 420 + }, + { + "epoch": 1.316283924843424, + "grad_norm": 0.9037718772888184, + "learning_rate": 4.623529241338689e-06, + "loss": 0.5902, + "step": 421 + }, + { + "epoch": 1.3194154488517746, + "grad_norm": 1.2110658884048462, + "learning_rate": 4.621241169355234e-06, + "loss": 0.561, + "step": 422 + }, + { + "epoch": 1.3225469728601253, + "grad_norm": 0.8582742214202881, + "learning_rate": 4.618946735057491e-06, + "loss": 0.5003, + "step": 423 + }, + { + "epoch": 1.325678496868476, + "grad_norm": 0.9203405976295471, + "learning_rate": 4.6166459453272386e-06, + "loss": 0.5639, + "step": 424 + }, + { + "epoch": 1.3288100208768268, + "grad_norm": 0.933721125125885, + "learning_rate": 4.614338807065317e-06, + "loss": 0.5766, + "step": 425 + }, + { + "epoch": 1.3319415448851775, + "grad_norm": 0.8435131311416626, + "learning_rate": 4.612025327191608e-06, + "loss": 0.5656, + "step": 426 + }, + { + "epoch": 1.3350730688935282, + "grad_norm": 0.795796811580658, + "learning_rate": 4.609705512645015e-06, + "loss": 0.4996, + "step": 427 + }, + { + "epoch": 1.3382045929018789, + "grad_norm": 0.8168228268623352, + "learning_rate": 4.6073793703834404e-06, + "loss": 0.5465, + "step": 428 + }, + { + "epoch": 1.3413361169102296, + "grad_norm": 0.8795569539070129, + "learning_rate": 4.605046907383765e-06, + "loss": 0.5407, + "step": 429 + }, + { + "epoch": 1.3444676409185803, + "grad_norm": 0.8504094481468201, + "learning_rate": 4.6027081306418295e-06, + "loss": 0.5589, + "step": 430 + }, + { + "epoch": 1.347599164926931, + "grad_norm": 1.485202431678772, + "learning_rate": 4.600363047172409e-06, + "loss": 0.5515, + "step": 431 + }, + { + "epoch": 1.350730688935282, + "grad_norm": 1.1156851053237915, + "learning_rate": 4.598011664009197e-06, + "loss": 0.5681, + "step": 432 + }, + { + "epoch": 1.3538622129436326, + "grad_norm": 0.8666876554489136, + "learning_rate": 4.595653988204779e-06, + "loss": 0.5451, + "step": 433 + }, + { + "epoch": 1.3569937369519833, + "grad_norm": 0.8192381858825684, + "learning_rate": 4.593290026830619e-06, + "loss": 0.5632, + "step": 434 + }, + { + "epoch": 1.360125260960334, + "grad_norm": 0.7994804978370667, + "learning_rate": 4.590919786977029e-06, + "loss": 0.5181, + "step": 435 + }, + { + "epoch": 1.3632567849686847, + "grad_norm": 0.8038607835769653, + "learning_rate": 4.5885432757531535e-06, + "loss": 0.5385, + "step": 436 + }, + { + "epoch": 1.3663883089770354, + "grad_norm": 0.7677503824234009, + "learning_rate": 4.586160500286948e-06, + "loss": 0.5455, + "step": 437 + }, + { + "epoch": 1.3695198329853862, + "grad_norm": 0.8293285369873047, + "learning_rate": 4.583771467725157e-06, + "loss": 0.5401, + "step": 438 + }, + { + "epoch": 1.372651356993737, + "grad_norm": 0.8607680797576904, + "learning_rate": 4.581376185233289e-06, + "loss": 0.5782, + "step": 439 + }, + { + "epoch": 1.3757828810020878, + "grad_norm": 0.8847081065177917, + "learning_rate": 4.578974659995601e-06, + "loss": 0.572, + "step": 440 + }, + { + "epoch": 1.3789144050104385, + "grad_norm": 0.7669641971588135, + "learning_rate": 4.576566899215075e-06, + "loss": 0.5655, + "step": 441 + }, + { + "epoch": 1.3820459290187892, + "grad_norm": 0.8514629006385803, + "learning_rate": 4.5741529101133904e-06, + "loss": 0.5218, + "step": 442 + }, + { + "epoch": 1.38517745302714, + "grad_norm": 0.8719842433929443, + "learning_rate": 4.5717326999309145e-06, + "loss": 0.5579, + "step": 443 + }, + { + "epoch": 1.3883089770354906, + "grad_norm": 1.1142809391021729, + "learning_rate": 4.569306275926667e-06, + "loss": 0.5535, + "step": 444 + }, + { + "epoch": 1.3914405010438413, + "grad_norm": 0.7392387986183167, + "learning_rate": 4.566873645378309e-06, + "loss": 0.5335, + "step": 445 + }, + { + "epoch": 1.394572025052192, + "grad_norm": 0.9066658616065979, + "learning_rate": 4.564434815582117e-06, + "loss": 0.5286, + "step": 446 + }, + { + "epoch": 1.3977035490605427, + "grad_norm": 0.8648932576179504, + "learning_rate": 4.561989793852959e-06, + "loss": 0.5008, + "step": 447 + }, + { + "epoch": 1.4008350730688934, + "grad_norm": 0.7768712043762207, + "learning_rate": 4.559538587524276e-06, + "loss": 0.5727, + "step": 448 + }, + { + "epoch": 1.4039665970772441, + "grad_norm": 0.7851182222366333, + "learning_rate": 4.557081203948059e-06, + "loss": 0.5731, + "step": 449 + }, + { + "epoch": 1.407098121085595, + "grad_norm": 0.8959861397743225, + "learning_rate": 4.5546176504948255e-06, + "loss": 0.5587, + "step": 450 + }, + { + "epoch": 1.4102296450939458, + "grad_norm": 1.0538026094436646, + "learning_rate": 4.552147934553601e-06, + "loss": 0.5808, + "step": 451 + }, + { + "epoch": 1.4133611691022965, + "grad_norm": 0.9887629151344299, + "learning_rate": 4.54967206353189e-06, + "loss": 0.5658, + "step": 452 + }, + { + "epoch": 1.4164926931106472, + "grad_norm": 0.9579302072525024, + "learning_rate": 4.547190044855663e-06, + "loss": 0.5092, + "step": 453 + }, + { + "epoch": 1.4196242171189979, + "grad_norm": 0.6993522047996521, + "learning_rate": 4.544701885969326e-06, + "loss": 0.5233, + "step": 454 + }, + { + "epoch": 1.4227557411273486, + "grad_norm": 0.8197568655014038, + "learning_rate": 4.542207594335703e-06, + "loss": 0.553, + "step": 455 + }, + { + "epoch": 1.4258872651356993, + "grad_norm": 2.921947717666626, + "learning_rate": 4.53970717743601e-06, + "loss": 0.4857, + "step": 456 + }, + { + "epoch": 1.4290187891440502, + "grad_norm": 1.3547242879867554, + "learning_rate": 4.53720064276984e-06, + "loss": 0.5676, + "step": 457 + }, + { + "epoch": 1.432150313152401, + "grad_norm": 1.4175567626953125, + "learning_rate": 4.534687997855131e-06, + "loss": 0.5164, + "step": 458 + }, + { + "epoch": 1.4352818371607516, + "grad_norm": 1.378146767616272, + "learning_rate": 4.532169250228145e-06, + "loss": 0.5429, + "step": 459 + }, + { + "epoch": 1.4384133611691023, + "grad_norm": 0.7811698317527771, + "learning_rate": 4.529644407443456e-06, + "loss": 0.524, + "step": 460 + }, + { + "epoch": 1.441544885177453, + "grad_norm": 1.1481678485870361, + "learning_rate": 4.527113477073914e-06, + "loss": 0.5513, + "step": 461 + }, + { + "epoch": 1.4446764091858038, + "grad_norm": 0.8450161218643188, + "learning_rate": 4.5245764667106266e-06, + "loss": 0.5632, + "step": 462 + }, + { + "epoch": 1.4478079331941545, + "grad_norm": 1.1582145690917969, + "learning_rate": 4.522033383962941e-06, + "loss": 0.5834, + "step": 463 + }, + { + "epoch": 1.4509394572025052, + "grad_norm": 1.0403447151184082, + "learning_rate": 4.519484236458416e-06, + "loss": 0.506, + "step": 464 + }, + { + "epoch": 1.4540709812108559, + "grad_norm": 0.7894920706748962, + "learning_rate": 4.516929031842799e-06, + "loss": 0.5526, + "step": 465 + }, + { + "epoch": 1.4572025052192066, + "grad_norm": 0.8092262744903564, + "learning_rate": 4.51436777778001e-06, + "loss": 0.5619, + "step": 466 + }, + { + "epoch": 1.4603340292275573, + "grad_norm": 0.9773806929588318, + "learning_rate": 4.511800481952106e-06, + "loss": 0.5179, + "step": 467 + }, + { + "epoch": 1.4634655532359082, + "grad_norm": 1.018676519393921, + "learning_rate": 4.509227152059271e-06, + "loss": 0.5415, + "step": 468 + }, + { + "epoch": 1.466597077244259, + "grad_norm": 0.7457838654518127, + "learning_rate": 4.506647795819784e-06, + "loss": 0.5473, + "step": 469 + }, + { + "epoch": 1.4697286012526096, + "grad_norm": 0.7826436161994934, + "learning_rate": 4.50406242097e-06, + "loss": 0.5526, + "step": 470 + }, + { + "epoch": 1.4728601252609603, + "grad_norm": 0.9492483139038086, + "learning_rate": 4.501471035264328e-06, + "loss": 0.5179, + "step": 471 + }, + { + "epoch": 1.475991649269311, + "grad_norm": 0.93398517370224, + "learning_rate": 4.4988736464752005e-06, + "loss": 0.5195, + "step": 472 + }, + { + "epoch": 1.4791231732776617, + "grad_norm": 0.8396487832069397, + "learning_rate": 4.496270262393061e-06, + "loss": 0.5447, + "step": 473 + }, + { + "epoch": 1.4822546972860124, + "grad_norm": 0.7450584173202515, + "learning_rate": 4.4936608908263315e-06, + "loss": 0.5207, + "step": 474 + }, + { + "epoch": 1.4853862212943634, + "grad_norm": 0.7887717485427856, + "learning_rate": 4.491045539601392e-06, + "loss": 0.523, + "step": 475 + }, + { + "epoch": 1.488517745302714, + "grad_norm": 1.2051388025283813, + "learning_rate": 4.48842421656256e-06, + "loss": 0.5402, + "step": 476 + }, + { + "epoch": 1.4916492693110648, + "grad_norm": 2.3103389739990234, + "learning_rate": 4.485796929572063e-06, + "loss": 0.5588, + "step": 477 + }, + { + "epoch": 1.4947807933194155, + "grad_norm": 0.7473112344741821, + "learning_rate": 4.483163686510016e-06, + "loss": 0.5731, + "step": 478 + }, + { + "epoch": 1.4979123173277662, + "grad_norm": 0.7545126676559448, + "learning_rate": 4.480524495274399e-06, + "loss": 0.5536, + "step": 479 + }, + { + "epoch": 1.501043841336117, + "grad_norm": 0.7801297903060913, + "learning_rate": 4.477879363781033e-06, + "loss": 0.5696, + "step": 480 + }, + { + "epoch": 1.5041753653444676, + "grad_norm": 0.7740563750267029, + "learning_rate": 4.475228299963554e-06, + "loss": 0.5526, + "step": 481 + }, + { + "epoch": 1.5073068893528183, + "grad_norm": 0.8600060343742371, + "learning_rate": 4.4725713117733936e-06, + "loss": 0.5051, + "step": 482 + }, + { + "epoch": 1.510438413361169, + "grad_norm": 0.6934283971786499, + "learning_rate": 4.46990840717975e-06, + "loss": 0.5564, + "step": 483 + }, + { + "epoch": 1.5135699373695197, + "grad_norm": 0.8927920460700989, + "learning_rate": 4.46723959416957e-06, + "loss": 0.5529, + "step": 484 + }, + { + "epoch": 1.5167014613778704, + "grad_norm": 0.9570988416671753, + "learning_rate": 4.464564880747517e-06, + "loss": 0.5661, + "step": 485 + }, + { + "epoch": 1.5198329853862211, + "grad_norm": 0.7229202389717102, + "learning_rate": 4.461884274935956e-06, + "loss": 0.5964, + "step": 486 + }, + { + "epoch": 1.522964509394572, + "grad_norm": 0.7367239594459534, + "learning_rate": 4.4591977847749225e-06, + "loss": 0.5455, + "step": 487 + }, + { + "epoch": 1.5260960334029228, + "grad_norm": 0.8062120676040649, + "learning_rate": 4.456505418322103e-06, + "loss": 0.5735, + "step": 488 + }, + { + "epoch": 1.5292275574112735, + "grad_norm": 0.8854482769966125, + "learning_rate": 4.453807183652808e-06, + "loss": 0.5421, + "step": 489 + }, + { + "epoch": 1.5323590814196242, + "grad_norm": 0.7518959045410156, + "learning_rate": 4.451103088859951e-06, + "loss": 0.5083, + "step": 490 + }, + { + "epoch": 1.535490605427975, + "grad_norm": 0.8621206879615784, + "learning_rate": 4.448393142054016e-06, + "loss": 0.4712, + "step": 491 + }, + { + "epoch": 1.5386221294363258, + "grad_norm": 1.0618741512298584, + "learning_rate": 4.445677351363046e-06, + "loss": 0.5808, + "step": 492 + }, + { + "epoch": 1.5417536534446765, + "grad_norm": 0.8261345028877258, + "learning_rate": 4.442955724932607e-06, + "loss": 0.5625, + "step": 493 + }, + { + "epoch": 1.5448851774530272, + "grad_norm": 0.7067139744758606, + "learning_rate": 4.440228270925772e-06, + "loss": 0.5661, + "step": 494 + }, + { + "epoch": 1.548016701461378, + "grad_norm": 0.9234416484832764, + "learning_rate": 4.437494997523091e-06, + "loss": 0.5428, + "step": 495 + }, + { + "epoch": 1.5511482254697286, + "grad_norm": 0.9273470044136047, + "learning_rate": 4.434755912922567e-06, + "loss": 0.5388, + "step": 496 + }, + { + "epoch": 1.5542797494780793, + "grad_norm": 1.0163263082504272, + "learning_rate": 4.4320110253396345e-06, + "loss": 0.5409, + "step": 497 + }, + { + "epoch": 1.55741127348643, + "grad_norm": 0.9542096853256226, + "learning_rate": 4.429260343007133e-06, + "loss": 0.5329, + "step": 498 + }, + { + "epoch": 1.5605427974947808, + "grad_norm": 0.8076801896095276, + "learning_rate": 4.426503874175283e-06, + "loss": 0.5616, + "step": 499 + }, + { + "epoch": 1.5636743215031315, + "grad_norm": 1.0063767433166504, + "learning_rate": 4.423741627111658e-06, + "loss": 0.5369, + "step": 500 + }, + { + "epoch": 1.5668058455114822, + "grad_norm": 1.040286898612976, + "learning_rate": 4.420973610101166e-06, + "loss": 0.5474, + "step": 501 + }, + { + "epoch": 1.5699373695198329, + "grad_norm": 0.7832860946655273, + "learning_rate": 4.4181998314460164e-06, + "loss": 0.5486, + "step": 502 + }, + { + "epoch": 1.5730688935281836, + "grad_norm": 0.8162257075309753, + "learning_rate": 4.415420299465706e-06, + "loss": 0.5054, + "step": 503 + }, + { + "epoch": 1.5762004175365343, + "grad_norm": 0.9108433127403259, + "learning_rate": 4.4126350224969814e-06, + "loss": 0.5399, + "step": 504 + }, + { + "epoch": 1.5793319415448852, + "grad_norm": 0.8002520799636841, + "learning_rate": 4.409844008893824e-06, + "loss": 0.5485, + "step": 505 + }, + { + "epoch": 1.582463465553236, + "grad_norm": 0.8543248772621155, + "learning_rate": 4.407047267027423e-06, + "loss": 0.4984, + "step": 506 + }, + { + "epoch": 1.5855949895615866, + "grad_norm": 0.7154155373573303, + "learning_rate": 4.404244805286141e-06, + "loss": 0.5392, + "step": 507 + }, + { + "epoch": 1.5887265135699373, + "grad_norm": 0.818553626537323, + "learning_rate": 4.401436632075504e-06, + "loss": 0.5178, + "step": 508 + }, + { + "epoch": 1.5918580375782883, + "grad_norm": 0.7535017728805542, + "learning_rate": 4.398622755818167e-06, + "loss": 0.5446, + "step": 509 + }, + { + "epoch": 1.594989561586639, + "grad_norm": 0.9328975677490234, + "learning_rate": 4.395803184953889e-06, + "loss": 0.5546, + "step": 510 + }, + { + "epoch": 1.5981210855949897, + "grad_norm": 0.7960026860237122, + "learning_rate": 4.392977927939508e-06, + "loss": 0.5451, + "step": 511 + }, + { + "epoch": 1.6012526096033404, + "grad_norm": 0.9686267971992493, + "learning_rate": 4.3901469932489195e-06, + "loss": 0.5198, + "step": 512 + }, + { + "epoch": 1.604384133611691, + "grad_norm": 0.903137743473053, + "learning_rate": 4.387310389373047e-06, + "loss": 0.5395, + "step": 513 + }, + { + "epoch": 1.6075156576200418, + "grad_norm": 1.0728516578674316, + "learning_rate": 4.384468124819816e-06, + "loss": 0.5843, + "step": 514 + }, + { + "epoch": 1.6106471816283925, + "grad_norm": 1.0245436429977417, + "learning_rate": 4.3816202081141345e-06, + "loss": 0.5672, + "step": 515 + }, + { + "epoch": 1.6137787056367432, + "grad_norm": 0.9672732353210449, + "learning_rate": 4.378766647797858e-06, + "loss": 0.5369, + "step": 516 + }, + { + "epoch": 1.616910229645094, + "grad_norm": 0.9149513840675354, + "learning_rate": 4.375907452429774e-06, + "loss": 0.4628, + "step": 517 + }, + { + "epoch": 1.6200417536534446, + "grad_norm": 0.7543843984603882, + "learning_rate": 4.373042630585567e-06, + "loss": 0.5344, + "step": 518 + }, + { + "epoch": 1.6231732776617953, + "grad_norm": 0.7589017152786255, + "learning_rate": 4.370172190857801e-06, + "loss": 0.5672, + "step": 519 + }, + { + "epoch": 1.626304801670146, + "grad_norm": 0.803040623664856, + "learning_rate": 4.367296141855887e-06, + "loss": 0.5313, + "step": 520 + }, + { + "epoch": 1.6294363256784967, + "grad_norm": 0.8305794596672058, + "learning_rate": 4.3644144922060625e-06, + "loss": 0.5754, + "step": 521 + }, + { + "epoch": 1.6325678496868476, + "grad_norm": 1.0086486339569092, + "learning_rate": 4.361527250551361e-06, + "loss": 0.5433, + "step": 522 + }, + { + "epoch": 1.6356993736951984, + "grad_norm": 0.7217550277709961, + "learning_rate": 4.35863442555159e-06, + "loss": 0.524, + "step": 523 + }, + { + "epoch": 1.638830897703549, + "grad_norm": 0.7788524627685547, + "learning_rate": 4.355736025883303e-06, + "loss": 0.536, + "step": 524 + }, + { + "epoch": 1.6419624217118998, + "grad_norm": 0.8460550904273987, + "learning_rate": 4.352832060239774e-06, + "loss": 0.5381, + "step": 525 + }, + { + "epoch": 1.6450939457202505, + "grad_norm": 0.7571215033531189, + "learning_rate": 4.3499225373309675e-06, + "loss": 0.541, + "step": 526 + }, + { + "epoch": 1.6482254697286014, + "grad_norm": 0.7343226671218872, + "learning_rate": 4.347007465883523e-06, + "loss": 0.5147, + "step": 527 + }, + { + "epoch": 1.651356993736952, + "grad_norm": 0.7271892428398132, + "learning_rate": 4.3440868546407165e-06, + "loss": 0.5311, + "step": 528 + }, + { + "epoch": 1.6544885177453028, + "grad_norm": 0.8166136741638184, + "learning_rate": 4.341160712362442e-06, + "loss": 0.5379, + "step": 529 + }, + { + "epoch": 1.6576200417536535, + "grad_norm": 1.5985233783721924, + "learning_rate": 4.338229047825182e-06, + "loss": 0.5782, + "step": 530 + }, + { + "epoch": 1.6607515657620042, + "grad_norm": 0.7835702896118164, + "learning_rate": 4.3352918698219835e-06, + "loss": 0.525, + "step": 531 + }, + { + "epoch": 1.663883089770355, + "grad_norm": 0.7278687953948975, + "learning_rate": 4.332349187162428e-06, + "loss": 0.5266, + "step": 532 + }, + { + "epoch": 1.6670146137787056, + "grad_norm": 0.8240190148353577, + "learning_rate": 4.329401008672608e-06, + "loss": 0.5515, + "step": 533 + }, + { + "epoch": 1.6701461377870563, + "grad_norm": 0.9447080492973328, + "learning_rate": 4.326447343195102e-06, + "loss": 0.5596, + "step": 534 + }, + { + "epoch": 1.673277661795407, + "grad_norm": 0.7827372550964355, + "learning_rate": 4.323488199588944e-06, + "loss": 0.5466, + "step": 535 + }, + { + "epoch": 1.6764091858037578, + "grad_norm": 0.9252517223358154, + "learning_rate": 4.320523586729599e-06, + "loss": 0.5433, + "step": 536 + }, + { + "epoch": 1.6795407098121085, + "grad_norm": 0.9437504410743713, + "learning_rate": 4.317553513508934e-06, + "loss": 0.5552, + "step": 537 + }, + { + "epoch": 1.6826722338204592, + "grad_norm": 0.8972746133804321, + "learning_rate": 4.3145779888351986e-06, + "loss": 0.5259, + "step": 538 + }, + { + "epoch": 1.6858037578288099, + "grad_norm": 0.8017446994781494, + "learning_rate": 4.311597021632988e-06, + "loss": 0.5263, + "step": 539 + }, + { + "epoch": 1.6889352818371608, + "grad_norm": 0.7875497341156006, + "learning_rate": 4.3086106208432235e-06, + "loss": 0.5316, + "step": 540 + }, + { + "epoch": 1.6920668058455115, + "grad_norm": 0.8204905986785889, + "learning_rate": 4.305618795423125e-06, + "loss": 0.5506, + "step": 541 + }, + { + "epoch": 1.6951983298538622, + "grad_norm": 0.888359785079956, + "learning_rate": 4.30262155434618e-06, + "loss": 0.4825, + "step": 542 + }, + { + "epoch": 1.698329853862213, + "grad_norm": 1.1026058197021484, + "learning_rate": 4.29961890660212e-06, + "loss": 0.5321, + "step": 543 + }, + { + "epoch": 1.7014613778705638, + "grad_norm": 0.7662535905838013, + "learning_rate": 4.2966108611968945e-06, + "loss": 0.5432, + "step": 544 + }, + { + "epoch": 1.7045929018789145, + "grad_norm": 1.1951749324798584, + "learning_rate": 4.293597427152641e-06, + "loss": 0.5123, + "step": 545 + }, + { + "epoch": 1.7077244258872653, + "grad_norm": 1.303183913230896, + "learning_rate": 4.290578613507661e-06, + "loss": 0.5346, + "step": 546 + }, + { + "epoch": 1.710855949895616, + "grad_norm": 0.7653357982635498, + "learning_rate": 4.287554429316387e-06, + "loss": 0.5397, + "step": 547 + }, + { + "epoch": 1.7139874739039667, + "grad_norm": 0.796215295791626, + "learning_rate": 4.284524883649366e-06, + "loss": 0.5421, + "step": 548 + }, + { + "epoch": 1.7171189979123174, + "grad_norm": 0.7599332332611084, + "learning_rate": 4.281489985593219e-06, + "loss": 0.5289, + "step": 549 + }, + { + "epoch": 1.720250521920668, + "grad_norm": 0.8029115796089172, + "learning_rate": 4.2784497442506265e-06, + "loss": 0.5409, + "step": 550 + }, + { + "epoch": 1.7233820459290188, + "grad_norm": 0.7194099426269531, + "learning_rate": 4.275404168740291e-06, + "loss": 0.5327, + "step": 551 + }, + { + "epoch": 1.7265135699373695, + "grad_norm": 0.7960740923881531, + "learning_rate": 4.272353268196917e-06, + "loss": 0.4896, + "step": 552 + }, + { + "epoch": 1.7296450939457202, + "grad_norm": 0.9572116732597351, + "learning_rate": 4.269297051771178e-06, + "loss": 0.5402, + "step": 553 + }, + { + "epoch": 1.732776617954071, + "grad_norm": 1.3604938983917236, + "learning_rate": 4.266235528629695e-06, + "loss": 0.5792, + "step": 554 + }, + { + "epoch": 1.7359081419624216, + "grad_norm": 2.067286729812622, + "learning_rate": 4.263168707955002e-06, + "loss": 0.5033, + "step": 555 + }, + { + "epoch": 1.7390396659707723, + "grad_norm": 0.8031097054481506, + "learning_rate": 4.260096598945523e-06, + "loss": 0.5117, + "step": 556 + }, + { + "epoch": 1.742171189979123, + "grad_norm": 1.0241729021072388, + "learning_rate": 4.257019210815546e-06, + "loss": 0.5359, + "step": 557 + }, + { + "epoch": 1.745302713987474, + "grad_norm": 0.7625218629837036, + "learning_rate": 4.25393655279519e-06, + "loss": 0.5625, + "step": 558 + }, + { + "epoch": 1.7484342379958246, + "grad_norm": 0.8603503704071045, + "learning_rate": 4.250848634130381e-06, + "loss": 0.5043, + "step": 559 + }, + { + "epoch": 1.7515657620041754, + "grad_norm": 0.9543750286102295, + "learning_rate": 4.247755464082824e-06, + "loss": 0.5364, + "step": 560 + }, + { + "epoch": 1.754697286012526, + "grad_norm": 0.9707463979721069, + "learning_rate": 4.244657051929973e-06, + "loss": 0.5184, + "step": 561 + }, + { + "epoch": 1.757828810020877, + "grad_norm": 0.7491432428359985, + "learning_rate": 4.241553406965008e-06, + "loss": 0.559, + "step": 562 + }, + { + "epoch": 1.7609603340292277, + "grad_norm": 0.7444972991943359, + "learning_rate": 4.238444538496801e-06, + "loss": 0.5327, + "step": 563 + }, + { + "epoch": 1.7640918580375784, + "grad_norm": 2.7108678817749023, + "learning_rate": 4.235330455849892e-06, + "loss": 0.55, + "step": 564 + }, + { + "epoch": 1.767223382045929, + "grad_norm": 1.6716049909591675, + "learning_rate": 4.232211168364459e-06, + "loss": 0.5093, + "step": 565 + }, + { + "epoch": 1.7703549060542798, + "grad_norm": 0.7023475170135498, + "learning_rate": 4.229086685396295e-06, + "loss": 0.569, + "step": 566 + }, + { + "epoch": 1.7734864300626305, + "grad_norm": 0.8596265316009521, + "learning_rate": 4.225957016316771e-06, + "loss": 0.5128, + "step": 567 + }, + { + "epoch": 1.7766179540709812, + "grad_norm": 0.8110849857330322, + "learning_rate": 4.222822170512816e-06, + "loss": 0.5142, + "step": 568 + }, + { + "epoch": 1.779749478079332, + "grad_norm": 0.7583725452423096, + "learning_rate": 4.219682157386884e-06, + "loss": 0.5584, + "step": 569 + }, + { + "epoch": 1.7828810020876826, + "grad_norm": 0.787811279296875, + "learning_rate": 4.21653698635693e-06, + "loss": 0.5068, + "step": 570 + }, + { + "epoch": 1.7860125260960333, + "grad_norm": 0.8298993110656738, + "learning_rate": 4.213386666856375e-06, + "loss": 0.5496, + "step": 571 + }, + { + "epoch": 1.789144050104384, + "grad_norm": 0.8999841213226318, + "learning_rate": 4.210231208334087e-06, + "loss": 0.5454, + "step": 572 + }, + { + "epoch": 1.7922755741127347, + "grad_norm": 4.264521598815918, + "learning_rate": 4.207070620254345e-06, + "loss": 0.5486, + "step": 573 + }, + { + "epoch": 1.7954070981210855, + "grad_norm": 0.8517448306083679, + "learning_rate": 4.203904912096812e-06, + "loss": 0.5566, + "step": 574 + }, + { + "epoch": 1.7985386221294362, + "grad_norm": 0.9230182766914368, + "learning_rate": 4.200734093356511e-06, + "loss": 0.4964, + "step": 575 + }, + { + "epoch": 1.801670146137787, + "grad_norm": 1.224039912223816, + "learning_rate": 4.197558173543791e-06, + "loss": 0.5356, + "step": 576 + }, + { + "epoch": 1.8048016701461378, + "grad_norm": 0.9998573660850525, + "learning_rate": 4.194377162184301e-06, + "loss": 0.5334, + "step": 577 + }, + { + "epoch": 1.8079331941544885, + "grad_norm": 0.865521252155304, + "learning_rate": 4.191191068818963e-06, + "loss": 0.5036, + "step": 578 + }, + { + "epoch": 1.8110647181628392, + "grad_norm": 0.8048138618469238, + "learning_rate": 4.18799990300394e-06, + "loss": 0.4979, + "step": 579 + }, + { + "epoch": 1.8141962421711901, + "grad_norm": 0.717815637588501, + "learning_rate": 4.184803674310609e-06, + "loss": 0.5623, + "step": 580 + }, + { + "epoch": 1.8173277661795408, + "grad_norm": 0.8403327465057373, + "learning_rate": 4.1816023923255335e-06, + "loss": 0.5055, + "step": 581 + }, + { + "epoch": 1.8204592901878915, + "grad_norm": 0.7298995852470398, + "learning_rate": 4.178396066650432e-06, + "loss": 0.5641, + "step": 582 + }, + { + "epoch": 1.8235908141962422, + "grad_norm": 0.9469727873802185, + "learning_rate": 4.1751847069021516e-06, + "loss": 0.5557, + "step": 583 + }, + { + "epoch": 1.826722338204593, + "grad_norm": 0.8641784191131592, + "learning_rate": 4.1719683227126386e-06, + "loss": 0.5153, + "step": 584 + }, + { + "epoch": 1.8298538622129437, + "grad_norm": 0.7316668629646301, + "learning_rate": 4.168746923728908e-06, + "loss": 0.4988, + "step": 585 + }, + { + "epoch": 1.8329853862212944, + "grad_norm": 0.8795468807220459, + "learning_rate": 4.165520519613017e-06, + "loss": 0.5483, + "step": 586 + }, + { + "epoch": 1.836116910229645, + "grad_norm": 0.7323560118675232, + "learning_rate": 4.162289120042034e-06, + "loss": 0.5194, + "step": 587 + }, + { + "epoch": 1.8392484342379958, + "grad_norm": 0.8217021822929382, + "learning_rate": 4.159052734708013e-06, + "loss": 0.532, + "step": 588 + }, + { + "epoch": 1.8423799582463465, + "grad_norm": 0.7669674754142761, + "learning_rate": 4.155811373317958e-06, + "loss": 0.541, + "step": 589 + }, + { + "epoch": 1.8455114822546972, + "grad_norm": 0.8312156200408936, + "learning_rate": 4.152565045593801e-06, + "loss": 0.5298, + "step": 590 + }, + { + "epoch": 1.848643006263048, + "grad_norm": 0.8967565298080444, + "learning_rate": 4.1493137612723665e-06, + "loss": 0.51, + "step": 591 + }, + { + "epoch": 1.8517745302713986, + "grad_norm": 0.8706664443016052, + "learning_rate": 4.14605753010535e-06, + "loss": 0.4941, + "step": 592 + }, + { + "epoch": 1.8549060542797495, + "grad_norm": 0.7585753798484802, + "learning_rate": 4.14279636185928e-06, + "loss": 0.5161, + "step": 593 + }, + { + "epoch": 1.8580375782881002, + "grad_norm": 0.7495241165161133, + "learning_rate": 4.1395302663154954e-06, + "loss": 0.5388, + "step": 594 + }, + { + "epoch": 1.861169102296451, + "grad_norm": 1.0746862888336182, + "learning_rate": 4.136259253270114e-06, + "loss": 0.4976, + "step": 595 + }, + { + "epoch": 1.8643006263048016, + "grad_norm": 0.872309684753418, + "learning_rate": 4.132983332534e-06, + "loss": 0.559, + "step": 596 + }, + { + "epoch": 1.8674321503131524, + "grad_norm": 0.8759891986846924, + "learning_rate": 4.1297025139327405e-06, + "loss": 0.5436, + "step": 597 + }, + { + "epoch": 1.8705636743215033, + "grad_norm": 1.1044493913650513, + "learning_rate": 4.126416807306611e-06, + "loss": 0.5476, + "step": 598 + }, + { + "epoch": 1.873695198329854, + "grad_norm": 0.8340442180633545, + "learning_rate": 4.123126222510549e-06, + "loss": 0.4592, + "step": 599 + }, + { + "epoch": 1.8768267223382047, + "grad_norm": 0.8331449031829834, + "learning_rate": 4.119830769414123e-06, + "loss": 0.5219, + "step": 600 + }, + { + "epoch": 1.8799582463465554, + "grad_norm": 1.0862973928451538, + "learning_rate": 4.116530457901503e-06, + "loss": 0.5159, + "step": 601 + }, + { + "epoch": 1.883089770354906, + "grad_norm": 0.8524414300918579, + "learning_rate": 4.113225297871431e-06, + "loss": 0.5502, + "step": 602 + }, + { + "epoch": 1.8862212943632568, + "grad_norm": 1.4945416450500488, + "learning_rate": 4.10991529923719e-06, + "loss": 0.5627, + "step": 603 + }, + { + "epoch": 1.8893528183716075, + "grad_norm": 1.5518157482147217, + "learning_rate": 4.10660047192658e-06, + "loss": 0.5517, + "step": 604 + }, + { + "epoch": 1.8924843423799582, + "grad_norm": 2.56638765335083, + "learning_rate": 4.103280825881878e-06, + "loss": 0.5422, + "step": 605 + }, + { + "epoch": 1.895615866388309, + "grad_norm": 0.867254912853241, + "learning_rate": 4.099956371059817e-06, + "loss": 0.4991, + "step": 606 + }, + { + "epoch": 1.8987473903966596, + "grad_norm": 0.9555892944335938, + "learning_rate": 4.096627117431554e-06, + "loss": 0.5339, + "step": 607 + }, + { + "epoch": 1.9018789144050103, + "grad_norm": 0.7905483245849609, + "learning_rate": 4.093293074982638e-06, + "loss": 0.5168, + "step": 608 + }, + { + "epoch": 1.905010438413361, + "grad_norm": 0.7500227093696594, + "learning_rate": 4.089954253712981e-06, + "loss": 0.5096, + "step": 609 + }, + { + "epoch": 1.9081419624217117, + "grad_norm": 0.8458324074745178, + "learning_rate": 4.086610663636828e-06, + "loss": 0.5296, + "step": 610 + }, + { + "epoch": 1.9112734864300627, + "grad_norm": 0.7392706871032715, + "learning_rate": 4.08326231478273e-06, + "loss": 0.5305, + "step": 611 + }, + { + "epoch": 1.9144050104384134, + "grad_norm": 0.8113343715667725, + "learning_rate": 4.079909217193508e-06, + "loss": 0.5044, + "step": 612 + }, + { + "epoch": 1.917536534446764, + "grad_norm": 0.7637801766395569, + "learning_rate": 4.076551380926226e-06, + "loss": 0.5298, + "step": 613 + }, + { + "epoch": 1.9206680584551148, + "grad_norm": 1.0523375272750854, + "learning_rate": 4.073188816052164e-06, + "loss": 0.5111, + "step": 614 + }, + { + "epoch": 1.9237995824634657, + "grad_norm": 0.8224868774414062, + "learning_rate": 4.069821532656781e-06, + "loss": 0.5178, + "step": 615 + }, + { + "epoch": 1.9269311064718164, + "grad_norm": 0.7270777821540833, + "learning_rate": 4.066449540839693e-06, + "loss": 0.5307, + "step": 616 + }, + { + "epoch": 1.9300626304801671, + "grad_norm": 0.7214602828025818, + "learning_rate": 4.063072850714631e-06, + "loss": 0.5171, + "step": 617 + }, + { + "epoch": 1.9331941544885178, + "grad_norm": 0.7333671450614929, + "learning_rate": 4.059691472409426e-06, + "loss": 0.56, + "step": 618 + }, + { + "epoch": 1.9363256784968685, + "grad_norm": 0.9166824221611023, + "learning_rate": 4.056305416065964e-06, + "loss": 0.5388, + "step": 619 + }, + { + "epoch": 1.9394572025052192, + "grad_norm": 0.7743303775787354, + "learning_rate": 4.052914691840167e-06, + "loss": 0.5134, + "step": 620 + }, + { + "epoch": 1.94258872651357, + "grad_norm": 0.704097330570221, + "learning_rate": 4.0495193099019524e-06, + "loss": 0.4926, + "step": 621 + }, + { + "epoch": 1.9457202505219207, + "grad_norm": 0.8508503437042236, + "learning_rate": 4.046119280435212e-06, + "loss": 0.5008, + "step": 622 + }, + { + "epoch": 1.9488517745302714, + "grad_norm": 0.725933313369751, + "learning_rate": 4.042714613637775e-06, + "loss": 0.5549, + "step": 623 + }, + { + "epoch": 1.951983298538622, + "grad_norm": 0.8919175863265991, + "learning_rate": 4.039305319721381e-06, + "loss": 0.5183, + "step": 624 + }, + { + "epoch": 1.9551148225469728, + "grad_norm": 0.827919065952301, + "learning_rate": 4.035891408911644e-06, + "loss": 0.5624, + "step": 625 + }, + { + "epoch": 1.9582463465553235, + "grad_norm": 0.7415187358856201, + "learning_rate": 4.032472891448032e-06, + "loss": 0.5454, + "step": 626 + }, + { + "epoch": 1.9613778705636742, + "grad_norm": 0.7675788998603821, + "learning_rate": 4.029049777583824e-06, + "loss": 0.5361, + "step": 627 + }, + { + "epoch": 1.964509394572025, + "grad_norm": 0.8464030623435974, + "learning_rate": 4.025622077586088e-06, + "loss": 0.5295, + "step": 628 + }, + { + "epoch": 1.9676409185803758, + "grad_norm": 0.7641633749008179, + "learning_rate": 4.022189801735646e-06, + "loss": 0.55, + "step": 629 + }, + { + "epoch": 1.9707724425887265, + "grad_norm": 0.7813227772712708, + "learning_rate": 4.018752960327048e-06, + "loss": 0.5587, + "step": 630 + }, + { + "epoch": 1.9739039665970772, + "grad_norm": 0.7576701641082764, + "learning_rate": 4.015311563668533e-06, + "loss": 0.5413, + "step": 631 + }, + { + "epoch": 1.977035490605428, + "grad_norm": 0.6949650049209595, + "learning_rate": 4.011865622082004e-06, + "loss": 0.5344, + "step": 632 + }, + { + "epoch": 1.9801670146137789, + "grad_norm": 0.9009145498275757, + "learning_rate": 4.008415145902997e-06, + "loss": 0.5233, + "step": 633 + }, + { + "epoch": 1.9832985386221296, + "grad_norm": 0.7635822892189026, + "learning_rate": 4.004960145480651e-06, + "loss": 0.4981, + "step": 634 + }, + { + "epoch": 1.9864300626304803, + "grad_norm": 0.8916334509849548, + "learning_rate": 4.0015006311776685e-06, + "loss": 0.5311, + "step": 635 + }, + { + "epoch": 1.989561586638831, + "grad_norm": 0.7197673320770264, + "learning_rate": 3.998036613370295e-06, + "loss": 0.5361, + "step": 636 + }, + { + "epoch": 1.9926931106471817, + "grad_norm": 0.8391228914260864, + "learning_rate": 3.994568102448284e-06, + "loss": 0.5473, + "step": 637 + }, + { + "epoch": 1.9958246346555324, + "grad_norm": 0.9371750950813293, + "learning_rate": 3.991095108814862e-06, + "loss": 0.5303, + "step": 638 + } + ], + "logging_steps": 1, + "max_steps": 1914, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 319, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.9343570238741414e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-638/training_args.bin b/checkpoint-638/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8067ee9c1c0bc752bdfd00cfcaf1a6e717d2356b --- /dev/null +++ b/checkpoint-638/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c650156a192cae0a2070c4323ee8a93e9b52fb76041d59ae0633b98389585727 +size 7928 diff --git a/checkpoint-638/zero_to_fp32.py b/checkpoint-638/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-638/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-957/README.md b/checkpoint-957/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f4a3934800eeb082a0cb833d7b6af4f68eed3615 --- /dev/null +++ b/checkpoint-957/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3_3-Nemotron-Super-49B-v1 +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.15.0 \ No newline at end of file diff --git a/checkpoint-957/adapter_config.json b/checkpoint-957/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..1edb33780e2306c6b19fd727be8e9b8b35f237c4 --- /dev/null +++ b/checkpoint-957/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3_3-Nemotron-Super-49B-v1", + "bias": "none", + "corda_config": null, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "k_proj", + "down_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-957/adapter_model.safetensors b/checkpoint-957/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a8c9d6b843d8111d855bbc5eb986462e822d368b --- /dev/null +++ b/checkpoint-957/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2c4fba160182b44f123fa5befe8bedc341afb7e8211a9cd9bc3fbca99ca305b5 +size 9016826528 diff --git a/checkpoint-957/global_step955/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-957/global_step955/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a724faedd79f5382b05276d9a359dcc1ede25443 --- /dev/null +++ b/checkpoint-957/global_step955/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:40076c054b3488882b5e837f16109e83c3315347b3ca4f298824bdcd2c9bacb4 +size 27050164444 diff --git a/checkpoint-957/global_step955/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-957/global_step955/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..874a416bdc64c3ad0c3c7031a834e25d15e4a672 --- /dev/null +++ b/checkpoint-957/global_step955/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eb9a32d60802d2a9b844ff380cf1876a06f7f0646db0c424c6579d3777bfdfdd +size 27050169884 diff --git a/checkpoint-957/global_step955/mp_rank_00_model_states.pt b/checkpoint-957/global_step955/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..0644f96cb6aab15fa0e1df62cedea128c2ceeadb --- /dev/null +++ b/checkpoint-957/global_step955/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2ecf4f5c9ba44f29a9e7ec10c139aad4b1f18fe5ef5681fa50628d9856f911a6 +size 9776788601 diff --git a/checkpoint-957/latest b/checkpoint-957/latest new file mode 100644 index 0000000000000000000000000000000000000000..a38f3310777331219ffe849f9792af39e25062aa --- /dev/null +++ b/checkpoint-957/latest @@ -0,0 +1 @@ +global_step955 \ No newline at end of file diff --git a/checkpoint-957/rng_state_0.pth b/checkpoint-957/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..cd433dc6ccf2ddc46e1878a11dadf9e5c078b35e --- /dev/null +++ b/checkpoint-957/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc4f5c99c78e3b43d8fbf8b1690558a76dbbc4bc1c01d9b02588cf55acbab3f5 +size 14512 diff --git a/checkpoint-957/rng_state_1.pth b/checkpoint-957/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..1566f5c64389939de8b378c9754ada5a631dc736 --- /dev/null +++ b/checkpoint-957/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3ea50dbf438d2f5ead49adeca5dc89fe039ae8b6ae03733eebb2d3e7f23a212c +size 14512 diff --git a/checkpoint-957/scheduler.pt b/checkpoint-957/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..3ab89c03232ef93703e3ba22df18f9336b987146 --- /dev/null +++ b/checkpoint-957/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6d456ff9f9125960678c7ceb13f24f7866014562349aeaab4aaf28f8cd5da86e +size 1064 diff --git a/checkpoint-957/special_tokens_map.json b/checkpoint-957/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-957/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-957/tokenizer.json b/checkpoint-957/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-957/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-957/tokenizer_config.json b/checkpoint-957/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..edd01b980c1db496ea102a51c972ee8f5d1a2c74 --- /dev/null +++ b/checkpoint-957/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "chat_template": "{{- bos_token }}{%- if messages[0]['role'] == 'system' %}{%- set system_message = messages[0]['content']|trim %}{%- set messages = messages[1:] %}{%- else %}{%- set system_message = \"\" %}{%- endif %}{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}{{- system_message }}{{- \"<|eot_id|>\" }}{%- for message in messages %}{%- if message['role'] == 'assistant' and '' in message['content'] %}{%- set content = message['content'].split('')[-1].lstrip() %}{%- else %}{%- set content = message['content'] %}{%- endif %}{{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n' + content | trim + '<|eot_id|>' }}{%- endfor %}{%- if add_generation_prompt %}{{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}{%- endif %}", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-957/trainer_state.json b/checkpoint-957/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..5e71b3f0db3ed1ba12f9b2adb765e62899ac6c5f --- /dev/null +++ b/checkpoint-957/trainer_state.json @@ -0,0 +1,6732 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.9926931106471817, + "eval_steps": 500, + "global_step": 957, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.003131524008350731, + "grad_norm": 13.917898178100586, + "learning_rate": 5.0000000000000004e-08, + "loss": 4.1051, + "step": 1 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 17.327869415283203, + "learning_rate": 1.0000000000000001e-07, + "loss": 4.1048, + "step": 2 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 14.063946723937988, + "learning_rate": 1.5000000000000002e-07, + "loss": 4.0741, + "step": 3 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 16.817699432373047, + "learning_rate": 2.0000000000000002e-07, + "loss": 4.2002, + "step": 4 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 14.47036361694336, + "learning_rate": 2.5000000000000004e-07, + "loss": 4.2652, + "step": 5 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 14.474193572998047, + "learning_rate": 3.0000000000000004e-07, + "loss": 4.0888, + "step": 6 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 14.865458488464355, + "learning_rate": 3.5000000000000004e-07, + "loss": 4.0014, + "step": 7 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 15.338888168334961, + "learning_rate": 4.0000000000000003e-07, + "loss": 4.13, + "step": 8 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 15.154336929321289, + "learning_rate": 4.5000000000000003e-07, + "loss": 4.2493, + "step": 9 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 15.919597625732422, + "learning_rate": 5.000000000000001e-07, + "loss": 4.0535, + "step": 10 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 14.981926918029785, + "learning_rate": 5.5e-07, + "loss": 3.9064, + "step": 11 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 13.36101245880127, + "learning_rate": 6.000000000000001e-07, + "loss": 4.1939, + "step": 12 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 15.58773422241211, + "learning_rate": 6.5e-07, + "loss": 4.18, + "step": 13 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 13.560139656066895, + "learning_rate": 7.000000000000001e-07, + "loss": 3.9414, + "step": 14 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 12.307971954345703, + "learning_rate": 7.5e-07, + "loss": 3.8836, + "step": 15 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 14.533182144165039, + "learning_rate": 8.000000000000001e-07, + "loss": 4.1551, + "step": 16 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 13.453729629516602, + "learning_rate": 8.500000000000001e-07, + "loss": 4.0048, + "step": 17 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 13.45992374420166, + "learning_rate": 9.000000000000001e-07, + "loss": 4.0745, + "step": 18 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 11.857145309448242, + "learning_rate": 9.500000000000001e-07, + "loss": 3.9871, + "step": 19 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 11.872294425964355, + "learning_rate": 1.0000000000000002e-06, + "loss": 3.8959, + "step": 20 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.969825744628906, + "learning_rate": 1.0500000000000001e-06, + "loss": 4.0308, + "step": 21 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 12.33769416809082, + "learning_rate": 1.1e-06, + "loss": 3.9341, + "step": 22 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 12.669405937194824, + "learning_rate": 1.1500000000000002e-06, + "loss": 3.8511, + "step": 23 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 10.677213668823242, + "learning_rate": 1.2000000000000002e-06, + "loss": 3.7764, + "step": 24 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 10.366402626037598, + "learning_rate": 1.25e-06, + "loss": 3.5291, + "step": 25 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 11.211421012878418, + "learning_rate": 1.3e-06, + "loss": 3.5765, + "step": 26 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 11.313716888427734, + "learning_rate": 1.3500000000000002e-06, + "loss": 3.4849, + "step": 27 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 10.41294002532959, + "learning_rate": 1.4000000000000001e-06, + "loss": 3.2653, + "step": 28 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 10.40064525604248, + "learning_rate": 1.45e-06, + "loss": 3.3384, + "step": 29 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 10.05427074432373, + "learning_rate": 1.5e-06, + "loss": 3.2257, + "step": 30 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 9.583163261413574, + "learning_rate": 1.5500000000000002e-06, + "loss": 3.1371, + "step": 31 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 10.09977912902832, + "learning_rate": 1.6000000000000001e-06, + "loss": 3.0658, + "step": 32 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 9.271486282348633, + "learning_rate": 1.6500000000000003e-06, + "loss": 2.9693, + "step": 33 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 10.687992095947266, + "learning_rate": 1.7000000000000002e-06, + "loss": 2.95, + "step": 34 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 8.762290000915527, + "learning_rate": 1.75e-06, + "loss": 2.8286, + "step": 35 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 10.13785171508789, + "learning_rate": 1.8000000000000001e-06, + "loss": 2.3664, + "step": 36 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 18.301353454589844, + "learning_rate": 1.85e-06, + "loss": 2.5533, + "step": 37 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 11.490377426147461, + "learning_rate": 1.9000000000000002e-06, + "loss": 2.6133, + "step": 38 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 15.614163398742676, + "learning_rate": 1.9500000000000004e-06, + "loss": 2.3596, + "step": 39 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 17.757442474365234, + "learning_rate": 2.0000000000000003e-06, + "loss": 2.3491, + "step": 40 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 17.18431854248047, + "learning_rate": 2.05e-06, + "loss": 2.2361, + "step": 41 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 16.149789810180664, + "learning_rate": 2.1000000000000002e-06, + "loss": 2.1457, + "step": 42 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 15.256914138793945, + "learning_rate": 2.15e-06, + "loss": 2.12, + "step": 43 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 15.537406921386719, + "learning_rate": 2.2e-06, + "loss": 2.1877, + "step": 44 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 7.947713851928711, + "learning_rate": 2.25e-06, + "loss": 2.1648, + "step": 45 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 8.818676948547363, + "learning_rate": 2.3000000000000004e-06, + "loss": 2.134, + "step": 46 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 5.175768852233887, + "learning_rate": 2.35e-06, + "loss": 2.0796, + "step": 47 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 6.750611305236816, + "learning_rate": 2.4000000000000003e-06, + "loss": 1.9174, + "step": 48 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 6.2147979736328125, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.8065, + "step": 49 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 13.291611671447754, + "learning_rate": 2.5e-06, + "loss": 1.7061, + "step": 50 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 7.251201629638672, + "learning_rate": 2.55e-06, + "loss": 1.7924, + "step": 51 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 5.2126054763793945, + "learning_rate": 2.6e-06, + "loss": 1.6735, + "step": 52 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 5.435528755187988, + "learning_rate": 2.6500000000000005e-06, + "loss": 1.6265, + "step": 53 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 4.505807399749756, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.4851, + "step": 54 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 5.128388404846191, + "learning_rate": 2.7500000000000004e-06, + "loss": 1.5832, + "step": 55 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 16.935827255249023, + "learning_rate": 2.8000000000000003e-06, + "loss": 1.6553, + "step": 56 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 3.664458990097046, + "learning_rate": 2.85e-06, + "loss": 1.5, + "step": 57 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 7.763802528381348, + "learning_rate": 2.9e-06, + "loss": 1.367, + "step": 58 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 3.2216155529022217, + "learning_rate": 2.95e-06, + "loss": 1.3863, + "step": 59 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 4.384445667266846, + "learning_rate": 3e-06, + "loss": 1.4247, + "step": 60 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 4.8080878257751465, + "learning_rate": 3.05e-06, + "loss": 1.3257, + "step": 61 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 4.154761791229248, + "learning_rate": 3.1000000000000004e-06, + "loss": 1.321, + "step": 62 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 6.4742112159729, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.2823, + "step": 63 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 2.583422899246216, + "learning_rate": 3.2000000000000003e-06, + "loss": 1.2136, + "step": 64 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 4.1933488845825195, + "learning_rate": 3.2500000000000002e-06, + "loss": 1.1855, + "step": 65 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 4.11049747467041, + "learning_rate": 3.3000000000000006e-06, + "loss": 1.2389, + "step": 66 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 2.264458417892456, + "learning_rate": 3.3500000000000005e-06, + "loss": 1.0651, + "step": 67 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 2.5408174991607666, + "learning_rate": 3.4000000000000005e-06, + "loss": 1.1389, + "step": 68 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 7.82421350479126, + "learning_rate": 3.45e-06, + "loss": 1.0956, + "step": 69 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 3.070939064025879, + "learning_rate": 3.5e-06, + "loss": 1.0451, + "step": 70 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 2.6310527324676514, + "learning_rate": 3.5500000000000003e-06, + "loss": 1.0538, + "step": 71 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 7.630155563354492, + "learning_rate": 3.6000000000000003e-06, + "loss": 1.0052, + "step": 72 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 6.950636863708496, + "learning_rate": 3.65e-06, + "loss": 1.0473, + "step": 73 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 2.2703945636749268, + "learning_rate": 3.7e-06, + "loss": 1.0576, + "step": 74 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 3.3817710876464844, + "learning_rate": 3.7500000000000005e-06, + "loss": 1.0177, + "step": 75 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 7.266414642333984, + "learning_rate": 3.8000000000000005e-06, + "loss": 1.0645, + "step": 76 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 5.782608509063721, + "learning_rate": 3.85e-06, + "loss": 1.0162, + "step": 77 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 2.7938575744628906, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9664, + "step": 78 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 6.681935787200928, + "learning_rate": 3.95e-06, + "loss": 0.953, + "step": 79 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 2.253279209136963, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9568, + "step": 80 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 1.4875826835632324, + "learning_rate": 4.05e-06, + "loss": 0.9448, + "step": 81 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 2.4987940788269043, + "learning_rate": 4.1e-06, + "loss": 0.9393, + "step": 82 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 4.712948322296143, + "learning_rate": 4.15e-06, + "loss": 0.9532, + "step": 83 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 6.9030632972717285, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.96, + "step": 84 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 3.4780967235565186, + "learning_rate": 4.25e-06, + "loss": 0.8993, + "step": 85 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.526064395904541, + "learning_rate": 4.3e-06, + "loss": 0.9021, + "step": 86 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 10.727686882019043, + "learning_rate": 4.350000000000001e-06, + "loss": 0.856, + "step": 87 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 12.483160972595215, + "learning_rate": 4.4e-06, + "loss": 0.9357, + "step": 88 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 6.544492244720459, + "learning_rate": 4.450000000000001e-06, + "loss": 0.9168, + "step": 89 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 1.178139567375183, + "learning_rate": 4.5e-06, + "loss": 0.8748, + "step": 90 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 1.711506962776184, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8425, + "step": 91 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 3.281747341156006, + "learning_rate": 4.600000000000001e-06, + "loss": 0.8491, + "step": 92 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 2.2964377403259277, + "learning_rate": 4.65e-06, + "loss": 0.8038, + "step": 93 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 1.959700345993042, + "learning_rate": 4.7e-06, + "loss": 0.8439, + "step": 94 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 3.979384183883667, + "learning_rate": 4.75e-06, + "loss": 0.8839, + "step": 95 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 1.4721262454986572, + "learning_rate": 4.800000000000001e-06, + "loss": 0.845, + "step": 96 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 2.862248659133911, + "learning_rate": 4.85e-06, + "loss": 0.7748, + "step": 97 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 3.7439088821411133, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8145, + "step": 98 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 1.6654618978500366, + "learning_rate": 4.95e-06, + "loss": 0.8326, + "step": 99 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 7.8437581062316895, + "learning_rate": 5e-06, + "loss": 0.8666, + "step": 100 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 6.429738521575928, + "learning_rate": 4.999996250830422e-06, + "loss": 0.836, + "step": 101 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 2.6017794609069824, + "learning_rate": 4.9999850033329326e-06, + "loss": 0.7785, + "step": 102 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 1.0575449466705322, + "learning_rate": 4.999966257541265e-06, + "loss": 0.7639, + "step": 103 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 2.6932010650634766, + "learning_rate": 4.999940013511647e-06, + "loss": 0.8214, + "step": 104 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 2.925288438796997, + "learning_rate": 4.999906271322792e-06, + "loss": 0.8797, + "step": 105 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.3570607900619507, + "learning_rate": 4.9998650310759035e-06, + "loss": 0.792, + "step": 106 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 5.126713752746582, + "learning_rate": 4.999816292894676e-06, + "loss": 0.8352, + "step": 107 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 1.8966432809829712, + "learning_rate": 4.99976005692529e-06, + "loss": 0.7663, + "step": 108 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 1.3100829124450684, + "learning_rate": 4.999696323336418e-06, + "loss": 0.771, + "step": 109 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 2.4025354385375977, + "learning_rate": 4.999625092319218e-06, + "loss": 0.7618, + "step": 110 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.130232810974121, + "learning_rate": 4.999546364087334e-06, + "loss": 0.7705, + "step": 111 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 3.430262327194214, + "learning_rate": 4.999460138876901e-06, + "loss": 0.77, + "step": 112 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 1.1272103786468506, + "learning_rate": 4.999366416946536e-06, + "loss": 0.7133, + "step": 113 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 1.1740471124649048, + "learning_rate": 4.999265198577342e-06, + "loss": 0.7684, + "step": 114 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.3138248920440674, + "learning_rate": 4.999156484072907e-06, + "loss": 0.7888, + "step": 115 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 1.061711311340332, + "learning_rate": 4.999040273759304e-06, + "loss": 0.7484, + "step": 116 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.4682390689849854, + "learning_rate": 4.998916567985083e-06, + "loss": 0.7296, + "step": 117 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 2.884068250656128, + "learning_rate": 4.998785367121284e-06, + "loss": 0.7662, + "step": 118 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9812761545181274, + "learning_rate": 4.9986466715614205e-06, + "loss": 0.7307, + "step": 119 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 2.2237496376037598, + "learning_rate": 4.998500481721484e-06, + "loss": 0.6761, + "step": 120 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 1.4004178047180176, + "learning_rate": 4.998346798039952e-06, + "loss": 0.7505, + "step": 121 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 5.54975700378418, + "learning_rate": 4.99818562097777e-06, + "loss": 0.7615, + "step": 122 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 6.17140531539917, + "learning_rate": 4.9980169510183624e-06, + "loss": 0.7002, + "step": 123 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 4.974380016326904, + "learning_rate": 4.997840788667628e-06, + "loss": 0.7449, + "step": 124 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 1.4133399724960327, + "learning_rate": 4.997657134453937e-06, + "loss": 0.7442, + "step": 125 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 1.868915319442749, + "learning_rate": 4.9974659889281295e-06, + "loss": 0.7104, + "step": 126 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 1.2599350214004517, + "learning_rate": 4.997267352663514e-06, + "loss": 0.7385, + "step": 127 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.4353641271591187, + "learning_rate": 4.997061226255869e-06, + "loss": 0.7081, + "step": 128 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 3.2492141723632812, + "learning_rate": 4.996847610323437e-06, + "loss": 0.7859, + "step": 129 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 9.599719047546387, + "learning_rate": 4.996626505506923e-06, + "loss": 0.7241, + "step": 130 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 10.053650856018066, + "learning_rate": 4.996397912469494e-06, + "loss": 0.6841, + "step": 131 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 1.323876976966858, + "learning_rate": 4.996161831896777e-06, + "loss": 0.7317, + "step": 132 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 1.4180598258972168, + "learning_rate": 4.9959182644968594e-06, + "loss": 0.692, + "step": 133 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 1.2194396257400513, + "learning_rate": 4.99566721100028e-06, + "loss": 0.7068, + "step": 134 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0984960794448853, + "learning_rate": 4.995408672160031e-06, + "loss": 0.6946, + "step": 135 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 1.9341071844100952, + "learning_rate": 4.995142648751561e-06, + "loss": 0.7467, + "step": 136 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 1.9960932731628418, + "learning_rate": 4.9948691415727594e-06, + "loss": 0.7379, + "step": 137 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.8743917942047119, + "learning_rate": 4.994588151443968e-06, + "loss": 0.66, + "step": 138 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8655261993408203, + "learning_rate": 4.99429967920797e-06, + "loss": 0.6646, + "step": 139 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 5.462070941925049, + "learning_rate": 4.994003725729992e-06, + "loss": 0.643, + "step": 140 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 2.1401469707489014, + "learning_rate": 4.993700291897695e-06, + "loss": 0.6639, + "step": 141 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 1.8219833374023438, + "learning_rate": 4.9933893786211815e-06, + "loss": 0.6673, + "step": 142 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 1.641079306602478, + "learning_rate": 4.993070986832984e-06, + "loss": 0.658, + "step": 143 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 1.1739819049835205, + "learning_rate": 4.992745117488066e-06, + "loss": 0.6826, + "step": 144 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 2.309185743331909, + "learning_rate": 4.9924117715638185e-06, + "loss": 0.6536, + "step": 145 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 1.09304940700531, + "learning_rate": 4.99207095006006e-06, + "loss": 0.721, + "step": 146 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.9056984186172485, + "learning_rate": 4.991722653999025e-06, + "loss": 0.7019, + "step": 147 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.8440625667572021, + "learning_rate": 4.991366884425374e-06, + "loss": 0.707, + "step": 148 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 1.2244676351547241, + "learning_rate": 4.991003642406177e-06, + "loss": 0.6407, + "step": 149 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9258589744567871, + "learning_rate": 4.99063292903092e-06, + "loss": 0.6954, + "step": 150 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 4.176390647888184, + "learning_rate": 4.990254745411496e-06, + "loss": 0.6812, + "step": 151 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 1.4322530031204224, + "learning_rate": 4.989869092682205e-06, + "loss": 0.6808, + "step": 152 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.8017717003822327, + "learning_rate": 4.989475971999748e-06, + "loss": 0.687, + "step": 153 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 1.5641374588012695, + "learning_rate": 4.989075384543228e-06, + "loss": 0.6599, + "step": 154 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 1.1522141695022583, + "learning_rate": 4.98866733151414e-06, + "loss": 0.6546, + "step": 155 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8593171238899231, + "learning_rate": 4.988251814136372e-06, + "loss": 0.6857, + "step": 156 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 2.668159246444702, + "learning_rate": 4.9878288336562e-06, + "loss": 0.661, + "step": 157 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9953671097755432, + "learning_rate": 4.987398391342285e-06, + "loss": 0.6512, + "step": 158 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.042872667312622, + "learning_rate": 4.986960488485667e-06, + "loss": 0.6311, + "step": 159 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.9070663452148438, + "learning_rate": 4.9865151263997645e-06, + "loss": 0.675, + "step": 160 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.8460433483123779, + "learning_rate": 4.986062306420367e-06, + "loss": 0.6635, + "step": 161 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 1.2639834880828857, + "learning_rate": 4.985602029905635e-06, + "loss": 0.6327, + "step": 162 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.8775074481964111, + "learning_rate": 4.985134298236091e-06, + "loss": 0.644, + "step": 163 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 1.2031961679458618, + "learning_rate": 4.98465911281462e-06, + "loss": 0.6254, + "step": 164 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.892494797706604, + "learning_rate": 4.984176475066463e-06, + "loss": 0.7122, + "step": 165 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 2.7122485637664795, + "learning_rate": 4.983686386439212e-06, + "loss": 0.6679, + "step": 166 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9344426989555359, + "learning_rate": 4.983188848402806e-06, + "loss": 0.6319, + "step": 167 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 1.4093577861785889, + "learning_rate": 4.982683862449531e-06, + "loss": 0.6425, + "step": 168 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 1.1285009384155273, + "learning_rate": 4.982171430094007e-06, + "loss": 0.6298, + "step": 169 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.952778935432434, + "learning_rate": 4.981651552873193e-06, + "loss": 0.7066, + "step": 170 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 5.133765697479248, + "learning_rate": 4.981124232346374e-06, + "loss": 0.6634, + "step": 171 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9770542979240417, + "learning_rate": 4.980589470095161e-06, + "loss": 0.7121, + "step": 172 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.8414323925971985, + "learning_rate": 4.980047267723487e-06, + "loss": 0.6397, + "step": 173 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 1.9173879623413086, + "learning_rate": 4.979497626857596e-06, + "loss": 0.6228, + "step": 174 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 1.0823363065719604, + "learning_rate": 4.978940549146048e-06, + "loss": 0.6475, + "step": 175 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 3.715353488922119, + "learning_rate": 4.978376036259706e-06, + "loss": 0.7127, + "step": 176 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.981584370136261, + "learning_rate": 4.9778040898917325e-06, + "loss": 0.6468, + "step": 177 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.70566987991333, + "learning_rate": 4.977224711757587e-06, + "loss": 0.6476, + "step": 178 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9217923283576965, + "learning_rate": 4.976637903595019e-06, + "loss": 0.6731, + "step": 179 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.8994677662849426, + "learning_rate": 4.976043667164063e-06, + "loss": 0.6562, + "step": 180 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 1.1613017320632935, + "learning_rate": 4.975442004247034e-06, + "loss": 0.6417, + "step": 181 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 1.6041977405548096, + "learning_rate": 4.974832916648521e-06, + "loss": 0.6029, + "step": 182 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.7978405952453613, + "learning_rate": 4.974216406195383e-06, + "loss": 0.6269, + "step": 183 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.6021920442581177, + "learning_rate": 4.973592474736739e-06, + "loss": 0.6149, + "step": 184 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8973568677902222, + "learning_rate": 4.972961124143971e-06, + "loss": 0.6648, + "step": 185 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 1.9432591199874878, + "learning_rate": 4.972322356310711e-06, + "loss": 0.6299, + "step": 186 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 4.457028388977051, + "learning_rate": 4.971676173152839e-06, + "loss": 0.656, + "step": 187 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 2.0989716053009033, + "learning_rate": 4.971022576608473e-06, + "loss": 0.6539, + "step": 188 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 1.0646967887878418, + "learning_rate": 4.97036156863797e-06, + "loss": 0.6727, + "step": 189 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.6522265672683716, + "learning_rate": 4.969693151223914e-06, + "loss": 0.6643, + "step": 190 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.7503505945205688, + "learning_rate": 4.969017326371115e-06, + "loss": 0.6402, + "step": 191 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 1.2341989278793335, + "learning_rate": 4.968334096106597e-06, + "loss": 0.6413, + "step": 192 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 3.089054584503174, + "learning_rate": 4.967643462479597e-06, + "loss": 0.6825, + "step": 193 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 2.711623430252075, + "learning_rate": 4.966945427561557e-06, + "loss": 0.65, + "step": 194 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 4.641184329986572, + "learning_rate": 4.966239993446118e-06, + "loss": 0.6229, + "step": 195 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.7984074354171753, + "learning_rate": 4.965527162249114e-06, + "loss": 0.6473, + "step": 196 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 1.1643115282058716, + "learning_rate": 4.964806936108566e-06, + "loss": 0.6404, + "step": 197 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 2.1877920627593994, + "learning_rate": 4.9640793171846725e-06, + "loss": 0.6185, + "step": 198 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.7970566749572754, + "learning_rate": 4.963344307659807e-06, + "loss": 0.634, + "step": 199 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 1.6014361381530762, + "learning_rate": 4.96260190973851e-06, + "loss": 0.6562, + "step": 200 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.8743320107460022, + "learning_rate": 4.961852125647482e-06, + "loss": 0.6133, + "step": 201 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 1.9526551961898804, + "learning_rate": 4.961094957635578e-06, + "loss": 0.6451, + "step": 202 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 3.6597347259521484, + "learning_rate": 4.960330407973798e-06, + "loss": 0.6386, + "step": 203 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 1.7180207967758179, + "learning_rate": 4.959558478955283e-06, + "loss": 0.6688, + "step": 204 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 0.9058470129966736, + "learning_rate": 4.958779172895308e-06, + "loss": 0.6161, + "step": 205 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 1.0031033754348755, + "learning_rate": 4.957992492131274e-06, + "loss": 0.6437, + "step": 206 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 1.5846725702285767, + "learning_rate": 4.9571984390226985e-06, + "loss": 0.6332, + "step": 207 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 1.9951609373092651, + "learning_rate": 4.956397015951215e-06, + "loss": 0.636, + "step": 208 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 1.4122583866119385, + "learning_rate": 4.95558822532056e-06, + "loss": 0.6586, + "step": 209 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.2243481874465942, + "learning_rate": 4.954772069556568e-06, + "loss": 0.6313, + "step": 210 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.8756356835365295, + "learning_rate": 4.953948551107164e-06, + "loss": 0.6406, + "step": 211 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 2.9979734420776367, + "learning_rate": 4.953117672442356e-06, + "loss": 0.5803, + "step": 212 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 2.1859359741210938, + "learning_rate": 4.952279436054229e-06, + "loss": 0.6607, + "step": 213 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.6929755806922913, + "learning_rate": 4.9514338444569346e-06, + "loss": 0.5989, + "step": 214 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 1.0361783504486084, + "learning_rate": 4.950580900186685e-06, + "loss": 0.6654, + "step": 215 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 1.210898518562317, + "learning_rate": 4.9497206058017475e-06, + "loss": 0.6213, + "step": 216 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 1.200990080833435, + "learning_rate": 4.948852963882434e-06, + "loss": 0.6654, + "step": 217 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 1.481831669807434, + "learning_rate": 4.947977977031093e-06, + "loss": 0.6474, + "step": 218 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.9883334636688232, + "learning_rate": 4.947095647872103e-06, + "loss": 0.6735, + "step": 219 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.7436536550521851, + "learning_rate": 4.946205979051868e-06, + "loss": 0.6456, + "step": 220 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9057570099830627, + "learning_rate": 4.945308973238802e-06, + "loss": 0.6228, + "step": 221 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 1.341081142425537, + "learning_rate": 4.944404633123324e-06, + "loss": 0.6417, + "step": 222 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.7958157062530518, + "learning_rate": 4.943492961417859e-06, + "loss": 0.6494, + "step": 223 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.216025471687317, + "learning_rate": 4.9425739608568106e-06, + "loss": 0.6566, + "step": 224 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.9774854779243469, + "learning_rate": 4.9416476341965735e-06, + "loss": 0.6171, + "step": 225 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 2.1562681198120117, + "learning_rate": 4.940713984215512e-06, + "loss": 0.629, + "step": 226 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 1.9521286487579346, + "learning_rate": 4.9397730137139556e-06, + "loss": 0.6475, + "step": 227 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 1.5749104022979736, + "learning_rate": 4.9388247255141895e-06, + "loss": 0.6053, + "step": 228 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 1.2008254528045654, + "learning_rate": 4.937869122460449e-06, + "loss": 0.6052, + "step": 229 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 1.0774102210998535, + "learning_rate": 4.93690620741891e-06, + "loss": 0.6099, + "step": 230 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 1.0929996967315674, + "learning_rate": 4.935935983277675e-06, + "loss": 0.6363, + "step": 231 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.8830653429031372, + "learning_rate": 4.934958452946774e-06, + "loss": 0.6136, + "step": 232 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 3.591218948364258, + "learning_rate": 4.933973619358147e-06, + "loss": 0.5962, + "step": 233 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 2.5797672271728516, + "learning_rate": 4.932981485465643e-06, + "loss": 0.6405, + "step": 234 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 1.0467664003372192, + "learning_rate": 4.9319820542450025e-06, + "loss": 0.6155, + "step": 235 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.8099795579910278, + "learning_rate": 4.930975328693856e-06, + "loss": 0.5615, + "step": 236 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.8906702995300293, + "learning_rate": 4.92996131183171e-06, + "loss": 0.6501, + "step": 237 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 1.0871416330337524, + "learning_rate": 4.928940006699944e-06, + "loss": 0.6282, + "step": 238 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 1.3209614753723145, + "learning_rate": 4.927911416361792e-06, + "loss": 0.598, + "step": 239 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.2252682447433472, + "learning_rate": 4.926875543902344e-06, + "loss": 0.6433, + "step": 240 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0569007396697998, + "learning_rate": 4.9258323924285285e-06, + "loss": 0.5927, + "step": 241 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9309014081954956, + "learning_rate": 4.924781965069106e-06, + "loss": 0.5927, + "step": 242 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0200378894805908, + "learning_rate": 4.923724264974662e-06, + "loss": 0.6064, + "step": 243 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 1.0533075332641602, + "learning_rate": 4.922659295317593e-06, + "loss": 0.6373, + "step": 244 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 0.7889382839202881, + "learning_rate": 4.921587059292102e-06, + "loss": 0.5887, + "step": 245 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.7943588495254517, + "learning_rate": 4.920507560114183e-06, + "loss": 0.593, + "step": 246 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.8247205018997192, + "learning_rate": 4.919420801021617e-06, + "loss": 0.6151, + "step": 247 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 0.9979158043861389, + "learning_rate": 4.91832678527396e-06, + "loss": 0.6019, + "step": 248 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9346868991851807, + "learning_rate": 4.917225516152532e-06, + "loss": 0.6098, + "step": 249 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.7487881183624268, + "learning_rate": 4.916116996960408e-06, + "loss": 0.5965, + "step": 250 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.821576714515686, + "learning_rate": 4.915001231022411e-06, + "loss": 0.6483, + "step": 251 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 1.0413196086883545, + "learning_rate": 4.913878221685096e-06, + "loss": 0.6108, + "step": 252 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9560331702232361, + "learning_rate": 4.912747972316745e-06, + "loss": 0.5758, + "step": 253 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.8964638113975525, + "learning_rate": 4.911610486307356e-06, + "loss": 0.6432, + "step": 254 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.8418346047401428, + "learning_rate": 4.910465767068631e-06, + "loss": 0.6027, + "step": 255 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 1.792371153831482, + "learning_rate": 4.909313818033966e-06, + "loss": 0.6198, + "step": 256 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 1.036665439605713, + "learning_rate": 4.908154642658446e-06, + "loss": 0.6255, + "step": 257 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.7592151165008545, + "learning_rate": 4.906988244418823e-06, + "loss": 0.6035, + "step": 258 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.8843073844909668, + "learning_rate": 4.90581462681352e-06, + "loss": 0.6299, + "step": 259 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9489964246749878, + "learning_rate": 4.9046337933626086e-06, + "loss": 0.5869, + "step": 260 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.851691722869873, + "learning_rate": 4.903445747607806e-06, + "loss": 0.603, + "step": 261 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 1.3722106218338013, + "learning_rate": 4.902250493112458e-06, + "loss": 0.5939, + "step": 262 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 1.1002827882766724, + "learning_rate": 4.901048033461537e-06, + "loss": 0.6452, + "step": 263 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 0.8428632020950317, + "learning_rate": 4.89983837226162e-06, + "loss": 0.5956, + "step": 264 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.7666584849357605, + "learning_rate": 4.898621513140889e-06, + "loss": 0.6067, + "step": 265 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.8413611054420471, + "learning_rate": 4.897397459749113e-06, + "loss": 0.5985, + "step": 266 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 2.3374335765838623, + "learning_rate": 4.896166215757638e-06, + "loss": 0.5885, + "step": 267 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 2.236640214920044, + "learning_rate": 4.894927784859377e-06, + "loss": 0.6408, + "step": 268 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.9715856313705444, + "learning_rate": 4.893682170768802e-06, + "loss": 0.5954, + "step": 269 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0249912738800049, + "learning_rate": 4.892429377221928e-06, + "loss": 0.6186, + "step": 270 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 1.255426049232483, + "learning_rate": 4.891169407976302e-06, + "loss": 0.6351, + "step": 271 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.9339559674263, + "learning_rate": 4.889902266810995e-06, + "loss": 0.5944, + "step": 272 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 1.2473429441452026, + "learning_rate": 4.888627957526589e-06, + "loss": 0.544, + "step": 273 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 1.0589442253112793, + "learning_rate": 4.887346483945166e-06, + "loss": 0.5543, + "step": 274 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9844024777412415, + "learning_rate": 4.886057849910294e-06, + "loss": 0.5941, + "step": 275 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 2.88578200340271, + "learning_rate": 4.8847620592870196e-06, + "loss": 0.6124, + "step": 276 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 0.7496054172515869, + "learning_rate": 4.8834591159618524e-06, + "loss": 0.6006, + "step": 277 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 0.7403052449226379, + "learning_rate": 4.88214902384276e-06, + "loss": 0.5911, + "step": 278 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9003771543502808, + "learning_rate": 4.880831786859146e-06, + "loss": 0.6347, + "step": 279 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0345501899719238, + "learning_rate": 4.879507408961847e-06, + "loss": 0.6111, + "step": 280 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.4385879039764404, + "learning_rate": 4.878175894123116e-06, + "loss": 0.6454, + "step": 281 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.8469482064247131, + "learning_rate": 4.8768372463366145e-06, + "loss": 0.6163, + "step": 282 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.8859589695930481, + "learning_rate": 4.875491469617395e-06, + "loss": 0.6144, + "step": 283 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 1.8436834812164307, + "learning_rate": 4.874138568001895e-06, + "loss": 0.6275, + "step": 284 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.6646101474761963, + "learning_rate": 4.87277854554792e-06, + "loss": 0.615, + "step": 285 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 1.0070925951004028, + "learning_rate": 4.871411406334633e-06, + "loss": 0.5898, + "step": 286 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9785194993019104, + "learning_rate": 4.870037154462545e-06, + "loss": 0.5992, + "step": 287 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.7244889736175537, + "learning_rate": 4.868655794053497e-06, + "loss": 0.6078, + "step": 288 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.4496444463729858, + "learning_rate": 4.8672673292506535e-06, + "loss": 0.5855, + "step": 289 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 1.8514957427978516, + "learning_rate": 4.865871764218486e-06, + "loss": 0.5707, + "step": 290 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.8439773321151733, + "learning_rate": 4.864469103142763e-06, + "loss": 0.5562, + "step": 291 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.8146086931228638, + "learning_rate": 4.8630593502305355e-06, + "loss": 0.6161, + "step": 292 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.8920315504074097, + "learning_rate": 4.861642509710126e-06, + "loss": 0.6139, + "step": 293 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.4980088472366333, + "learning_rate": 4.860218585831116e-06, + "loss": 0.6187, + "step": 294 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 0.9910127520561218, + "learning_rate": 4.8587875828643285e-06, + "loss": 0.5852, + "step": 295 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.819600522518158, + "learning_rate": 4.857349505101823e-06, + "loss": 0.6172, + "step": 296 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 1.1059772968292236, + "learning_rate": 4.855904356856878e-06, + "loss": 0.5868, + "step": 297 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 1.2362196445465088, + "learning_rate": 4.854452142463977e-06, + "loss": 0.625, + "step": 298 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.9956470727920532, + "learning_rate": 4.852992866278799e-06, + "loss": 0.5923, + "step": 299 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 0.864109218120575, + "learning_rate": 4.851526532678203e-06, + "loss": 0.6315, + "step": 300 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 0.8900614380836487, + "learning_rate": 4.850053146060217e-06, + "loss": 0.6128, + "step": 301 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.927254855632782, + "learning_rate": 4.84857271084402e-06, + "loss": 0.5955, + "step": 302 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0046517848968506, + "learning_rate": 4.847085231469935e-06, + "loss": 0.6134, + "step": 303 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.734597384929657, + "learning_rate": 4.8455907123994125e-06, + "loss": 0.5927, + "step": 304 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.7338348031044006, + "learning_rate": 4.844089158115016e-06, + "loss": 0.5897, + "step": 305 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9163988828659058, + "learning_rate": 4.8425805731204106e-06, + "loss": 0.6051, + "step": 306 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 1.050246238708496, + "learning_rate": 4.84106496194035e-06, + "loss": 0.5751, + "step": 307 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 0.7637603878974915, + "learning_rate": 4.83954232912066e-06, + "loss": 0.5677, + "step": 308 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.7110525965690613, + "learning_rate": 4.838012679228229e-06, + "loss": 0.6051, + "step": 309 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.7662068605422974, + "learning_rate": 4.836476016850988e-06, + "loss": 0.59, + "step": 310 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 0.8907375335693359, + "learning_rate": 4.834932346597906e-06, + "loss": 0.5792, + "step": 311 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.8939849138259888, + "learning_rate": 4.833381673098966e-06, + "loss": 0.6062, + "step": 312 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 0.8878788948059082, + "learning_rate": 4.8318240010051595e-06, + "loss": 0.5694, + "step": 313 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 1.2523870468139648, + "learning_rate": 4.830259334988468e-06, + "loss": 0.5809, + "step": 314 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 1.0836797952651978, + "learning_rate": 4.82868767974185e-06, + "loss": 0.5949, + "step": 315 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.7985473871231079, + "learning_rate": 4.827109039979226e-06, + "loss": 0.6057, + "step": 316 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 1.042951226234436, + "learning_rate": 4.825523420435469e-06, + "loss": 0.6004, + "step": 317 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.7845115661621094, + "learning_rate": 4.823930825866381e-06, + "loss": 0.6161, + "step": 318 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.931854784488678, + "learning_rate": 4.82233126104869e-06, + "loss": 0.5912, + "step": 319 + }, + { + "epoch": 1.0, + "grad_norm": 0.931854784488678, + "learning_rate": 4.8207247307800275e-06, + "loss": 0.1914, + "step": 320 + }, + { + "epoch": 1.0031315240083507, + "grad_norm": 0.751028835773468, + "learning_rate": 4.819111239878916e-06, + "loss": 0.5802, + "step": 321 + }, + { + "epoch": 1.0062630480167014, + "grad_norm": 1.4943569898605347, + "learning_rate": 4.817490793184758e-06, + "loss": 0.613, + "step": 322 + }, + { + "epoch": 1.0093945720250521, + "grad_norm": 2.296318531036377, + "learning_rate": 4.815863395557816e-06, + "loss": 0.5453, + "step": 323 + }, + { + "epoch": 1.0125260960334028, + "grad_norm": 0.760101318359375, + "learning_rate": 4.814229051879202e-06, + "loss": 0.5302, + "step": 324 + }, + { + "epoch": 1.0156576200417538, + "grad_norm": 0.8145846128463745, + "learning_rate": 4.812587767050861e-06, + "loss": 0.5831, + "step": 325 + }, + { + "epoch": 1.0187891440501045, + "grad_norm": 0.9169796109199524, + "learning_rate": 4.8109395459955565e-06, + "loss": 0.5756, + "step": 326 + }, + { + "epoch": 1.0219206680584552, + "grad_norm": 0.8791524171829224, + "learning_rate": 4.809284393656858e-06, + "loss": 0.5988, + "step": 327 + }, + { + "epoch": 1.0250521920668059, + "grad_norm": 1.0184170007705688, + "learning_rate": 4.807622314999122e-06, + "loss": 0.5476, + "step": 328 + }, + { + "epoch": 1.0281837160751566, + "grad_norm": 0.8095184564590454, + "learning_rate": 4.8059533150074805e-06, + "loss": 0.5723, + "step": 329 + }, + { + "epoch": 1.0313152400835073, + "grad_norm": 0.7621930241584778, + "learning_rate": 4.804277398687826e-06, + "loss": 0.5841, + "step": 330 + }, + { + "epoch": 1.034446764091858, + "grad_norm": 3.729628324508667, + "learning_rate": 4.802594571066791e-06, + "loss": 0.5639, + "step": 331 + }, + { + "epoch": 1.0375782881002087, + "grad_norm": 1.6502974033355713, + "learning_rate": 4.800904837191743e-06, + "loss": 0.6024, + "step": 332 + }, + { + "epoch": 1.0407098121085594, + "grad_norm": 0.8031198978424072, + "learning_rate": 4.799208202130762e-06, + "loss": 0.5305, + "step": 333 + }, + { + "epoch": 1.0438413361169103, + "grad_norm": 0.939644992351532, + "learning_rate": 4.797504670972623e-06, + "loss": 0.5446, + "step": 334 + }, + { + "epoch": 1.046972860125261, + "grad_norm": 1.0589954853057861, + "learning_rate": 4.795794248826789e-06, + "loss": 0.5366, + "step": 335 + }, + { + "epoch": 1.0501043841336117, + "grad_norm": 0.9089614748954773, + "learning_rate": 4.794076940823391e-06, + "loss": 0.5795, + "step": 336 + }, + { + "epoch": 1.0532359081419624, + "grad_norm": 0.7732561230659485, + "learning_rate": 4.792352752113212e-06, + "loss": 0.5765, + "step": 337 + }, + { + "epoch": 1.0563674321503131, + "grad_norm": 1.811553955078125, + "learning_rate": 4.790621687867672e-06, + "loss": 0.561, + "step": 338 + }, + { + "epoch": 1.0594989561586639, + "grad_norm": 1.1930758953094482, + "learning_rate": 4.788883753278813e-06, + "loss": 0.5, + "step": 339 + }, + { + "epoch": 1.0626304801670146, + "grad_norm": 0.9551813006401062, + "learning_rate": 4.787138953559285e-06, + "loss": 0.5228, + "step": 340 + }, + { + "epoch": 1.0657620041753653, + "grad_norm": 0.9609586596488953, + "learning_rate": 4.785387293942329e-06, + "loss": 0.5827, + "step": 341 + }, + { + "epoch": 1.068893528183716, + "grad_norm": 0.8403449654579163, + "learning_rate": 4.783628779681759e-06, + "loss": 0.5585, + "step": 342 + }, + { + "epoch": 1.072025052192067, + "grad_norm": 0.9108251929283142, + "learning_rate": 4.7818634160519496e-06, + "loss": 0.6077, + "step": 343 + }, + { + "epoch": 1.0751565762004176, + "grad_norm": 0.9476898908615112, + "learning_rate": 4.780091208347819e-06, + "loss": 0.5493, + "step": 344 + }, + { + "epoch": 1.0782881002087683, + "grad_norm": 1.1943707466125488, + "learning_rate": 4.778312161884813e-06, + "loss": 0.5736, + "step": 345 + }, + { + "epoch": 1.081419624217119, + "grad_norm": 3.1342639923095703, + "learning_rate": 4.77652628199889e-06, + "loss": 0.5765, + "step": 346 + }, + { + "epoch": 1.0845511482254697, + "grad_norm": 2.7982125282287598, + "learning_rate": 4.7747335740465015e-06, + "loss": 0.6003, + "step": 347 + }, + { + "epoch": 1.0876826722338204, + "grad_norm": 1.5068914890289307, + "learning_rate": 4.7729340434045815e-06, + "loss": 0.5033, + "step": 348 + }, + { + "epoch": 1.0908141962421711, + "grad_norm": 0.8273429274559021, + "learning_rate": 4.771127695470527e-06, + "loss": 0.5309, + "step": 349 + }, + { + "epoch": 1.0939457202505218, + "grad_norm": 1.104974389076233, + "learning_rate": 4.76931453566218e-06, + "loss": 0.5244, + "step": 350 + }, + { + "epoch": 1.0970772442588728, + "grad_norm": 1.096509337425232, + "learning_rate": 4.7674945694178166e-06, + "loss": 0.5585, + "step": 351 + }, + { + "epoch": 1.1002087682672235, + "grad_norm": 1.0238200426101685, + "learning_rate": 4.765667802196127e-06, + "loss": 0.5589, + "step": 352 + }, + { + "epoch": 1.1033402922755742, + "grad_norm": 0.7515526413917542, + "learning_rate": 4.763834239476197e-06, + "loss": 0.5304, + "step": 353 + }, + { + "epoch": 1.1064718162839249, + "grad_norm": 1.0282566547393799, + "learning_rate": 4.761993886757499e-06, + "loss": 0.5476, + "step": 354 + }, + { + "epoch": 1.1096033402922756, + "grad_norm": 0.9962708950042725, + "learning_rate": 4.760146749559868e-06, + "loss": 0.5117, + "step": 355 + }, + { + "epoch": 1.1127348643006263, + "grad_norm": 0.7851671576499939, + "learning_rate": 4.758292833423488e-06, + "loss": 0.5542, + "step": 356 + }, + { + "epoch": 1.115866388308977, + "grad_norm": 0.8857759237289429, + "learning_rate": 4.756432143908876e-06, + "loss": 0.544, + "step": 357 + }, + { + "epoch": 1.1189979123173277, + "grad_norm": 0.9402740597724915, + "learning_rate": 4.7545646865968645e-06, + "loss": 0.5656, + "step": 358 + }, + { + "epoch": 1.1221294363256784, + "grad_norm": 0.8210407495498657, + "learning_rate": 4.752690467088584e-06, + "loss": 0.5733, + "step": 359 + }, + { + "epoch": 1.1252609603340291, + "grad_norm": 0.795684278011322, + "learning_rate": 4.750809491005449e-06, + "loss": 0.5678, + "step": 360 + }, + { + "epoch": 1.12839248434238, + "grad_norm": 0.8712463974952698, + "learning_rate": 4.748921763989139e-06, + "loss": 0.5777, + "step": 361 + }, + { + "epoch": 1.1315240083507307, + "grad_norm": 0.9810119867324829, + "learning_rate": 4.747027291701578e-06, + "loss": 0.5511, + "step": 362 + }, + { + "epoch": 1.1346555323590815, + "grad_norm": 0.81117844581604, + "learning_rate": 4.745126079824926e-06, + "loss": 0.5038, + "step": 363 + }, + { + "epoch": 1.1377870563674322, + "grad_norm": 0.7631494402885437, + "learning_rate": 4.743218134061556e-06, + "loss": 0.6272, + "step": 364 + }, + { + "epoch": 1.1409185803757829, + "grad_norm": 0.7601696252822876, + "learning_rate": 4.741303460134038e-06, + "loss": 0.571, + "step": 365 + }, + { + "epoch": 1.1440501043841336, + "grad_norm": 1.7977744340896606, + "learning_rate": 4.7393820637851205e-06, + "loss": 0.538, + "step": 366 + }, + { + "epoch": 1.1471816283924843, + "grad_norm": 2.022578001022339, + "learning_rate": 4.737453950777718e-06, + "loss": 0.5822, + "step": 367 + }, + { + "epoch": 1.150313152400835, + "grad_norm": 0.7586764693260193, + "learning_rate": 4.735519126894885e-06, + "loss": 0.5986, + "step": 368 + }, + { + "epoch": 1.153444676409186, + "grad_norm": 0.8970286846160889, + "learning_rate": 4.733577597939812e-06, + "loss": 0.542, + "step": 369 + }, + { + "epoch": 1.1565762004175366, + "grad_norm": 0.8546352982521057, + "learning_rate": 4.731629369735793e-06, + "loss": 0.5832, + "step": 370 + }, + { + "epoch": 1.1597077244258873, + "grad_norm": 0.9266164898872375, + "learning_rate": 4.72967444812622e-06, + "loss": 0.551, + "step": 371 + }, + { + "epoch": 1.162839248434238, + "grad_norm": 1.0413658618927002, + "learning_rate": 4.7277128389745595e-06, + "loss": 0.5866, + "step": 372 + }, + { + "epoch": 1.1659707724425887, + "grad_norm": 0.9312199950218201, + "learning_rate": 4.7257445481643334e-06, + "loss": 0.5723, + "step": 373 + }, + { + "epoch": 1.1691022964509394, + "grad_norm": 0.7389806509017944, + "learning_rate": 4.723769581599109e-06, + "loss": 0.5209, + "step": 374 + }, + { + "epoch": 1.1722338204592901, + "grad_norm": 3.053169012069702, + "learning_rate": 4.721787945202472e-06, + "loss": 0.6094, + "step": 375 + }, + { + "epoch": 1.1753653444676408, + "grad_norm": 1.288589596748352, + "learning_rate": 4.719799644918017e-06, + "loss": 0.5616, + "step": 376 + }, + { + "epoch": 1.1784968684759916, + "grad_norm": 0.7675042152404785, + "learning_rate": 4.717804686709323e-06, + "loss": 0.4963, + "step": 377 + }, + { + "epoch": 1.1816283924843423, + "grad_norm": 0.7246491312980652, + "learning_rate": 4.715803076559938e-06, + "loss": 0.5273, + "step": 378 + }, + { + "epoch": 1.1847599164926932, + "grad_norm": 0.8193361759185791, + "learning_rate": 4.713794820473366e-06, + "loss": 0.6107, + "step": 379 + }, + { + "epoch": 1.187891440501044, + "grad_norm": 0.9498510360717773, + "learning_rate": 4.711779924473037e-06, + "loss": 0.5421, + "step": 380 + }, + { + "epoch": 1.1910229645093946, + "grad_norm": 1.0479756593704224, + "learning_rate": 4.709758394602305e-06, + "loss": 0.5257, + "step": 381 + }, + { + "epoch": 1.1941544885177453, + "grad_norm": 0.907866895198822, + "learning_rate": 4.707730236924413e-06, + "loss": 0.5289, + "step": 382 + }, + { + "epoch": 1.197286012526096, + "grad_norm": 0.8861165642738342, + "learning_rate": 4.705695457522488e-06, + "loss": 0.5727, + "step": 383 + }, + { + "epoch": 1.2004175365344467, + "grad_norm": 0.7467761039733887, + "learning_rate": 4.703654062499516e-06, + "loss": 0.5602, + "step": 384 + }, + { + "epoch": 1.2035490605427974, + "grad_norm": 0.7456198334693909, + "learning_rate": 4.701606057978325e-06, + "loss": 0.5345, + "step": 385 + }, + { + "epoch": 1.2066805845511483, + "grad_norm": 1.9976060390472412, + "learning_rate": 4.699551450101571e-06, + "loss": 0.5504, + "step": 386 + }, + { + "epoch": 1.209812108559499, + "grad_norm": 1.5253807306289673, + "learning_rate": 4.697490245031709e-06, + "loss": 0.5568, + "step": 387 + }, + { + "epoch": 1.2129436325678498, + "grad_norm": 1.0786075592041016, + "learning_rate": 4.6954224489509885e-06, + "loss": 0.5564, + "step": 388 + }, + { + "epoch": 1.2160751565762005, + "grad_norm": 0.8385995030403137, + "learning_rate": 4.693348068061422e-06, + "loss": 0.5341, + "step": 389 + }, + { + "epoch": 1.2192066805845512, + "grad_norm": 0.8184949159622192, + "learning_rate": 4.691267108584774e-06, + "loss": 0.5614, + "step": 390 + }, + { + "epoch": 1.2223382045929019, + "grad_norm": 0.9964898824691772, + "learning_rate": 4.68917957676254e-06, + "loss": 0.5589, + "step": 391 + }, + { + "epoch": 1.2254697286012526, + "grad_norm": 1.0168914794921875, + "learning_rate": 4.687085478855931e-06, + "loss": 0.5892, + "step": 392 + }, + { + "epoch": 1.2286012526096033, + "grad_norm": 0.8841140866279602, + "learning_rate": 4.684984821145846e-06, + "loss": 0.5327, + "step": 393 + }, + { + "epoch": 1.231732776617954, + "grad_norm": 0.834431529045105, + "learning_rate": 4.682877609932866e-06, + "loss": 0.5594, + "step": 394 + }, + { + "epoch": 1.2348643006263047, + "grad_norm": 0.7256641983985901, + "learning_rate": 4.6807638515372234e-06, + "loss": 0.5443, + "step": 395 + }, + { + "epoch": 1.2379958246346556, + "grad_norm": 0.765096127986908, + "learning_rate": 4.678643552298788e-06, + "loss": 0.5439, + "step": 396 + }, + { + "epoch": 1.2411273486430063, + "grad_norm": 0.8760455846786499, + "learning_rate": 4.676516718577051e-06, + "loss": 0.5485, + "step": 397 + }, + { + "epoch": 1.244258872651357, + "grad_norm": 2.7111501693725586, + "learning_rate": 4.674383356751099e-06, + "loss": 0.5696, + "step": 398 + }, + { + "epoch": 1.2473903966597077, + "grad_norm": 1.0521738529205322, + "learning_rate": 4.672243473219601e-06, + "loss": 0.5503, + "step": 399 + }, + { + "epoch": 1.2505219206680585, + "grad_norm": 0.8909669518470764, + "learning_rate": 4.670097074400785e-06, + "loss": 0.5183, + "step": 400 + }, + { + "epoch": 1.2536534446764092, + "grad_norm": 0.7483847737312317, + "learning_rate": 4.667944166732424e-06, + "loss": 0.5669, + "step": 401 + }, + { + "epoch": 1.2567849686847599, + "grad_norm": 1.146997094154358, + "learning_rate": 4.665784756671808e-06, + "loss": 0.5464, + "step": 402 + }, + { + "epoch": 1.2599164926931106, + "grad_norm": 0.8998096585273743, + "learning_rate": 4.663618850695733e-06, + "loss": 0.5502, + "step": 403 + }, + { + "epoch": 1.2630480167014615, + "grad_norm": 0.8882688283920288, + "learning_rate": 4.6614464553004795e-06, + "loss": 0.5507, + "step": 404 + }, + { + "epoch": 1.2661795407098122, + "grad_norm": 0.8310684561729431, + "learning_rate": 4.659267577001789e-06, + "loss": 0.5164, + "step": 405 + }, + { + "epoch": 1.269311064718163, + "grad_norm": 0.9286114573478699, + "learning_rate": 4.657082222334851e-06, + "loss": 0.4813, + "step": 406 + }, + { + "epoch": 1.2724425887265136, + "grad_norm": 1.2394906282424927, + "learning_rate": 4.654890397854275e-06, + "loss": 0.5837, + "step": 407 + }, + { + "epoch": 1.2755741127348643, + "grad_norm": 4.00585412979126, + "learning_rate": 4.652692110134079e-06, + "loss": 0.5453, + "step": 408 + }, + { + "epoch": 1.278705636743215, + "grad_norm": 1.1667803525924683, + "learning_rate": 4.650487365767667e-06, + "loss": 0.5652, + "step": 409 + }, + { + "epoch": 1.2818371607515657, + "grad_norm": 0.9351289868354797, + "learning_rate": 4.648276171367807e-06, + "loss": 0.5576, + "step": 410 + }, + { + "epoch": 1.2849686847599164, + "grad_norm": 0.8107728958129883, + "learning_rate": 4.646058533566614e-06, + "loss": 0.5821, + "step": 411 + }, + { + "epoch": 1.2881002087682671, + "grad_norm": 0.7293011546134949, + "learning_rate": 4.643834459015525e-06, + "loss": 0.5363, + "step": 412 + }, + { + "epoch": 1.2912317327766178, + "grad_norm": 0.7550690770149231, + "learning_rate": 4.641603954385289e-06, + "loss": 0.53, + "step": 413 + }, + { + "epoch": 1.2943632567849686, + "grad_norm": 0.7626177072525024, + "learning_rate": 4.639367026365938e-06, + "loss": 0.5307, + "step": 414 + }, + { + "epoch": 1.2974947807933195, + "grad_norm": 1.0841096639633179, + "learning_rate": 4.637123681666769e-06, + "loss": 0.5162, + "step": 415 + }, + { + "epoch": 1.3006263048016702, + "grad_norm": 0.8814271092414856, + "learning_rate": 4.634873927016326e-06, + "loss": 0.5369, + "step": 416 + }, + { + "epoch": 1.303757828810021, + "grad_norm": 0.7402971982955933, + "learning_rate": 4.632617769162378e-06, + "loss": 0.5846, + "step": 417 + }, + { + "epoch": 1.3068893528183716, + "grad_norm": 0.8106061220169067, + "learning_rate": 4.6303552148719e-06, + "loss": 0.5289, + "step": 418 + }, + { + "epoch": 1.3100208768267223, + "grad_norm": 0.9241361618041992, + "learning_rate": 4.628086270931053e-06, + "loss": 0.5714, + "step": 419 + }, + { + "epoch": 1.313152400835073, + "grad_norm": 0.950332522392273, + "learning_rate": 4.625810944145159e-06, + "loss": 0.5817, + "step": 420 + }, + { + "epoch": 1.316283924843424, + "grad_norm": 0.9037718772888184, + "learning_rate": 4.623529241338689e-06, + "loss": 0.5902, + "step": 421 + }, + { + "epoch": 1.3194154488517746, + "grad_norm": 1.2110658884048462, + "learning_rate": 4.621241169355234e-06, + "loss": 0.561, + "step": 422 + }, + { + "epoch": 1.3225469728601253, + "grad_norm": 0.8582742214202881, + "learning_rate": 4.618946735057491e-06, + "loss": 0.5003, + "step": 423 + }, + { + "epoch": 1.325678496868476, + "grad_norm": 0.9203405976295471, + "learning_rate": 4.6166459453272386e-06, + "loss": 0.5639, + "step": 424 + }, + { + "epoch": 1.3288100208768268, + "grad_norm": 0.933721125125885, + "learning_rate": 4.614338807065317e-06, + "loss": 0.5766, + "step": 425 + }, + { + "epoch": 1.3319415448851775, + "grad_norm": 0.8435131311416626, + "learning_rate": 4.612025327191608e-06, + "loss": 0.5656, + "step": 426 + }, + { + "epoch": 1.3350730688935282, + "grad_norm": 0.795796811580658, + "learning_rate": 4.609705512645015e-06, + "loss": 0.4996, + "step": 427 + }, + { + "epoch": 1.3382045929018789, + "grad_norm": 0.8168228268623352, + "learning_rate": 4.6073793703834404e-06, + "loss": 0.5465, + "step": 428 + }, + { + "epoch": 1.3413361169102296, + "grad_norm": 0.8795569539070129, + "learning_rate": 4.605046907383765e-06, + "loss": 0.5407, + "step": 429 + }, + { + "epoch": 1.3444676409185803, + "grad_norm": 0.8504094481468201, + "learning_rate": 4.6027081306418295e-06, + "loss": 0.5589, + "step": 430 + }, + { + "epoch": 1.347599164926931, + "grad_norm": 1.485202431678772, + "learning_rate": 4.600363047172409e-06, + "loss": 0.5515, + "step": 431 + }, + { + "epoch": 1.350730688935282, + "grad_norm": 1.1156851053237915, + "learning_rate": 4.598011664009197e-06, + "loss": 0.5681, + "step": 432 + }, + { + "epoch": 1.3538622129436326, + "grad_norm": 0.8666876554489136, + "learning_rate": 4.595653988204779e-06, + "loss": 0.5451, + "step": 433 + }, + { + "epoch": 1.3569937369519833, + "grad_norm": 0.8192381858825684, + "learning_rate": 4.593290026830619e-06, + "loss": 0.5632, + "step": 434 + }, + { + "epoch": 1.360125260960334, + "grad_norm": 0.7994804978370667, + "learning_rate": 4.590919786977029e-06, + "loss": 0.5181, + "step": 435 + }, + { + "epoch": 1.3632567849686847, + "grad_norm": 0.8038607835769653, + "learning_rate": 4.5885432757531535e-06, + "loss": 0.5385, + "step": 436 + }, + { + "epoch": 1.3663883089770354, + "grad_norm": 0.7677503824234009, + "learning_rate": 4.586160500286948e-06, + "loss": 0.5455, + "step": 437 + }, + { + "epoch": 1.3695198329853862, + "grad_norm": 0.8293285369873047, + "learning_rate": 4.583771467725157e-06, + "loss": 0.5401, + "step": 438 + }, + { + "epoch": 1.372651356993737, + "grad_norm": 0.8607680797576904, + "learning_rate": 4.581376185233289e-06, + "loss": 0.5782, + "step": 439 + }, + { + "epoch": 1.3757828810020878, + "grad_norm": 0.8847081065177917, + "learning_rate": 4.578974659995601e-06, + "loss": 0.572, + "step": 440 + }, + { + "epoch": 1.3789144050104385, + "grad_norm": 0.7669641971588135, + "learning_rate": 4.576566899215075e-06, + "loss": 0.5655, + "step": 441 + }, + { + "epoch": 1.3820459290187892, + "grad_norm": 0.8514629006385803, + "learning_rate": 4.5741529101133904e-06, + "loss": 0.5218, + "step": 442 + }, + { + "epoch": 1.38517745302714, + "grad_norm": 0.8719842433929443, + "learning_rate": 4.5717326999309145e-06, + "loss": 0.5579, + "step": 443 + }, + { + "epoch": 1.3883089770354906, + "grad_norm": 1.1142809391021729, + "learning_rate": 4.569306275926667e-06, + "loss": 0.5535, + "step": 444 + }, + { + "epoch": 1.3914405010438413, + "grad_norm": 0.7392387986183167, + "learning_rate": 4.566873645378309e-06, + "loss": 0.5335, + "step": 445 + }, + { + "epoch": 1.394572025052192, + "grad_norm": 0.9066658616065979, + "learning_rate": 4.564434815582117e-06, + "loss": 0.5286, + "step": 446 + }, + { + "epoch": 1.3977035490605427, + "grad_norm": 0.8648932576179504, + "learning_rate": 4.561989793852959e-06, + "loss": 0.5008, + "step": 447 + }, + { + "epoch": 1.4008350730688934, + "grad_norm": 0.7768712043762207, + "learning_rate": 4.559538587524276e-06, + "loss": 0.5727, + "step": 448 + }, + { + "epoch": 1.4039665970772441, + "grad_norm": 0.7851182222366333, + "learning_rate": 4.557081203948059e-06, + "loss": 0.5731, + "step": 449 + }, + { + "epoch": 1.407098121085595, + "grad_norm": 0.8959861397743225, + "learning_rate": 4.5546176504948255e-06, + "loss": 0.5587, + "step": 450 + }, + { + "epoch": 1.4102296450939458, + "grad_norm": 1.0538026094436646, + "learning_rate": 4.552147934553601e-06, + "loss": 0.5808, + "step": 451 + }, + { + "epoch": 1.4133611691022965, + "grad_norm": 0.9887629151344299, + "learning_rate": 4.54967206353189e-06, + "loss": 0.5658, + "step": 452 + }, + { + "epoch": 1.4164926931106472, + "grad_norm": 0.9579302072525024, + "learning_rate": 4.547190044855663e-06, + "loss": 0.5092, + "step": 453 + }, + { + "epoch": 1.4196242171189979, + "grad_norm": 0.6993522047996521, + "learning_rate": 4.544701885969326e-06, + "loss": 0.5233, + "step": 454 + }, + { + "epoch": 1.4227557411273486, + "grad_norm": 0.8197568655014038, + "learning_rate": 4.542207594335703e-06, + "loss": 0.553, + "step": 455 + }, + { + "epoch": 1.4258872651356993, + "grad_norm": 2.921947717666626, + "learning_rate": 4.53970717743601e-06, + "loss": 0.4857, + "step": 456 + }, + { + "epoch": 1.4290187891440502, + "grad_norm": 1.3547242879867554, + "learning_rate": 4.53720064276984e-06, + "loss": 0.5676, + "step": 457 + }, + { + "epoch": 1.432150313152401, + "grad_norm": 1.4175567626953125, + "learning_rate": 4.534687997855131e-06, + "loss": 0.5164, + "step": 458 + }, + { + "epoch": 1.4352818371607516, + "grad_norm": 1.378146767616272, + "learning_rate": 4.532169250228145e-06, + "loss": 0.5429, + "step": 459 + }, + { + "epoch": 1.4384133611691023, + "grad_norm": 0.7811698317527771, + "learning_rate": 4.529644407443456e-06, + "loss": 0.524, + "step": 460 + }, + { + "epoch": 1.441544885177453, + "grad_norm": 1.1481678485870361, + "learning_rate": 4.527113477073914e-06, + "loss": 0.5513, + "step": 461 + }, + { + "epoch": 1.4446764091858038, + "grad_norm": 0.8450161218643188, + "learning_rate": 4.5245764667106266e-06, + "loss": 0.5632, + "step": 462 + }, + { + "epoch": 1.4478079331941545, + "grad_norm": 1.1582145690917969, + "learning_rate": 4.522033383962941e-06, + "loss": 0.5834, + "step": 463 + }, + { + "epoch": 1.4509394572025052, + "grad_norm": 1.0403447151184082, + "learning_rate": 4.519484236458416e-06, + "loss": 0.506, + "step": 464 + }, + { + "epoch": 1.4540709812108559, + "grad_norm": 0.7894920706748962, + "learning_rate": 4.516929031842799e-06, + "loss": 0.5526, + "step": 465 + }, + { + "epoch": 1.4572025052192066, + "grad_norm": 0.8092262744903564, + "learning_rate": 4.51436777778001e-06, + "loss": 0.5619, + "step": 466 + }, + { + "epoch": 1.4603340292275573, + "grad_norm": 0.9773806929588318, + "learning_rate": 4.511800481952106e-06, + "loss": 0.5179, + "step": 467 + }, + { + "epoch": 1.4634655532359082, + "grad_norm": 1.018676519393921, + "learning_rate": 4.509227152059271e-06, + "loss": 0.5415, + "step": 468 + }, + { + "epoch": 1.466597077244259, + "grad_norm": 0.7457838654518127, + "learning_rate": 4.506647795819784e-06, + "loss": 0.5473, + "step": 469 + }, + { + "epoch": 1.4697286012526096, + "grad_norm": 0.7826436161994934, + "learning_rate": 4.50406242097e-06, + "loss": 0.5526, + "step": 470 + }, + { + "epoch": 1.4728601252609603, + "grad_norm": 0.9492483139038086, + "learning_rate": 4.501471035264328e-06, + "loss": 0.5179, + "step": 471 + }, + { + "epoch": 1.475991649269311, + "grad_norm": 0.93398517370224, + "learning_rate": 4.4988736464752005e-06, + "loss": 0.5195, + "step": 472 + }, + { + "epoch": 1.4791231732776617, + "grad_norm": 0.8396487832069397, + "learning_rate": 4.496270262393061e-06, + "loss": 0.5447, + "step": 473 + }, + { + "epoch": 1.4822546972860124, + "grad_norm": 0.7450584173202515, + "learning_rate": 4.4936608908263315e-06, + "loss": 0.5207, + "step": 474 + }, + { + "epoch": 1.4853862212943634, + "grad_norm": 0.7887717485427856, + "learning_rate": 4.491045539601392e-06, + "loss": 0.523, + "step": 475 + }, + { + "epoch": 1.488517745302714, + "grad_norm": 1.2051388025283813, + "learning_rate": 4.48842421656256e-06, + "loss": 0.5402, + "step": 476 + }, + { + "epoch": 1.4916492693110648, + "grad_norm": 2.3103389739990234, + "learning_rate": 4.485796929572063e-06, + "loss": 0.5588, + "step": 477 + }, + { + "epoch": 1.4947807933194155, + "grad_norm": 0.7473112344741821, + "learning_rate": 4.483163686510016e-06, + "loss": 0.5731, + "step": 478 + }, + { + "epoch": 1.4979123173277662, + "grad_norm": 0.7545126676559448, + "learning_rate": 4.480524495274399e-06, + "loss": 0.5536, + "step": 479 + }, + { + "epoch": 1.501043841336117, + "grad_norm": 0.7801297903060913, + "learning_rate": 4.477879363781033e-06, + "loss": 0.5696, + "step": 480 + }, + { + "epoch": 1.5041753653444676, + "grad_norm": 0.7740563750267029, + "learning_rate": 4.475228299963554e-06, + "loss": 0.5526, + "step": 481 + }, + { + "epoch": 1.5073068893528183, + "grad_norm": 0.8600060343742371, + "learning_rate": 4.4725713117733936e-06, + "loss": 0.5051, + "step": 482 + }, + { + "epoch": 1.510438413361169, + "grad_norm": 0.6934283971786499, + "learning_rate": 4.46990840717975e-06, + "loss": 0.5564, + "step": 483 + }, + { + "epoch": 1.5135699373695197, + "grad_norm": 0.8927920460700989, + "learning_rate": 4.46723959416957e-06, + "loss": 0.5529, + "step": 484 + }, + { + "epoch": 1.5167014613778704, + "grad_norm": 0.9570988416671753, + "learning_rate": 4.464564880747517e-06, + "loss": 0.5661, + "step": 485 + }, + { + "epoch": 1.5198329853862211, + "grad_norm": 0.7229202389717102, + "learning_rate": 4.461884274935956e-06, + "loss": 0.5964, + "step": 486 + }, + { + "epoch": 1.522964509394572, + "grad_norm": 0.7367239594459534, + "learning_rate": 4.4591977847749225e-06, + "loss": 0.5455, + "step": 487 + }, + { + "epoch": 1.5260960334029228, + "grad_norm": 0.8062120676040649, + "learning_rate": 4.456505418322103e-06, + "loss": 0.5735, + "step": 488 + }, + { + "epoch": 1.5292275574112735, + "grad_norm": 0.8854482769966125, + "learning_rate": 4.453807183652808e-06, + "loss": 0.5421, + "step": 489 + }, + { + "epoch": 1.5323590814196242, + "grad_norm": 0.7518959045410156, + "learning_rate": 4.451103088859951e-06, + "loss": 0.5083, + "step": 490 + }, + { + "epoch": 1.535490605427975, + "grad_norm": 0.8621206879615784, + "learning_rate": 4.448393142054016e-06, + "loss": 0.4712, + "step": 491 + }, + { + "epoch": 1.5386221294363258, + "grad_norm": 1.0618741512298584, + "learning_rate": 4.445677351363046e-06, + "loss": 0.5808, + "step": 492 + }, + { + "epoch": 1.5417536534446765, + "grad_norm": 0.8261345028877258, + "learning_rate": 4.442955724932607e-06, + "loss": 0.5625, + "step": 493 + }, + { + "epoch": 1.5448851774530272, + "grad_norm": 0.7067139744758606, + "learning_rate": 4.440228270925772e-06, + "loss": 0.5661, + "step": 494 + }, + { + "epoch": 1.548016701461378, + "grad_norm": 0.9234416484832764, + "learning_rate": 4.437494997523091e-06, + "loss": 0.5428, + "step": 495 + }, + { + "epoch": 1.5511482254697286, + "grad_norm": 0.9273470044136047, + "learning_rate": 4.434755912922567e-06, + "loss": 0.5388, + "step": 496 + }, + { + "epoch": 1.5542797494780793, + "grad_norm": 1.0163263082504272, + "learning_rate": 4.4320110253396345e-06, + "loss": 0.5409, + "step": 497 + }, + { + "epoch": 1.55741127348643, + "grad_norm": 0.9542096853256226, + "learning_rate": 4.429260343007133e-06, + "loss": 0.5329, + "step": 498 + }, + { + "epoch": 1.5605427974947808, + "grad_norm": 0.8076801896095276, + "learning_rate": 4.426503874175283e-06, + "loss": 0.5616, + "step": 499 + }, + { + "epoch": 1.5636743215031315, + "grad_norm": 1.0063767433166504, + "learning_rate": 4.423741627111658e-06, + "loss": 0.5369, + "step": 500 + }, + { + "epoch": 1.5668058455114822, + "grad_norm": 1.040286898612976, + "learning_rate": 4.420973610101166e-06, + "loss": 0.5474, + "step": 501 + }, + { + "epoch": 1.5699373695198329, + "grad_norm": 0.7832860946655273, + "learning_rate": 4.4181998314460164e-06, + "loss": 0.5486, + "step": 502 + }, + { + "epoch": 1.5730688935281836, + "grad_norm": 0.8162257075309753, + "learning_rate": 4.415420299465706e-06, + "loss": 0.5054, + "step": 503 + }, + { + "epoch": 1.5762004175365343, + "grad_norm": 0.9108433127403259, + "learning_rate": 4.4126350224969814e-06, + "loss": 0.5399, + "step": 504 + }, + { + "epoch": 1.5793319415448852, + "grad_norm": 0.8002520799636841, + "learning_rate": 4.409844008893824e-06, + "loss": 0.5485, + "step": 505 + }, + { + "epoch": 1.582463465553236, + "grad_norm": 0.8543248772621155, + "learning_rate": 4.407047267027423e-06, + "loss": 0.4984, + "step": 506 + }, + { + "epoch": 1.5855949895615866, + "grad_norm": 0.7154155373573303, + "learning_rate": 4.404244805286141e-06, + "loss": 0.5392, + "step": 507 + }, + { + "epoch": 1.5887265135699373, + "grad_norm": 0.818553626537323, + "learning_rate": 4.401436632075504e-06, + "loss": 0.5178, + "step": 508 + }, + { + "epoch": 1.5918580375782883, + "grad_norm": 0.7535017728805542, + "learning_rate": 4.398622755818167e-06, + "loss": 0.5446, + "step": 509 + }, + { + "epoch": 1.594989561586639, + "grad_norm": 0.9328975677490234, + "learning_rate": 4.395803184953889e-06, + "loss": 0.5546, + "step": 510 + }, + { + "epoch": 1.5981210855949897, + "grad_norm": 0.7960026860237122, + "learning_rate": 4.392977927939508e-06, + "loss": 0.5451, + "step": 511 + }, + { + "epoch": 1.6012526096033404, + "grad_norm": 0.9686267971992493, + "learning_rate": 4.3901469932489195e-06, + "loss": 0.5198, + "step": 512 + }, + { + "epoch": 1.604384133611691, + "grad_norm": 0.903137743473053, + "learning_rate": 4.387310389373047e-06, + "loss": 0.5395, + "step": 513 + }, + { + "epoch": 1.6075156576200418, + "grad_norm": 1.0728516578674316, + "learning_rate": 4.384468124819816e-06, + "loss": 0.5843, + "step": 514 + }, + { + "epoch": 1.6106471816283925, + "grad_norm": 1.0245436429977417, + "learning_rate": 4.3816202081141345e-06, + "loss": 0.5672, + "step": 515 + }, + { + "epoch": 1.6137787056367432, + "grad_norm": 0.9672732353210449, + "learning_rate": 4.378766647797858e-06, + "loss": 0.5369, + "step": 516 + }, + { + "epoch": 1.616910229645094, + "grad_norm": 0.9149513840675354, + "learning_rate": 4.375907452429774e-06, + "loss": 0.4628, + "step": 517 + }, + { + "epoch": 1.6200417536534446, + "grad_norm": 0.7543843984603882, + "learning_rate": 4.373042630585567e-06, + "loss": 0.5344, + "step": 518 + }, + { + "epoch": 1.6231732776617953, + "grad_norm": 0.7589017152786255, + "learning_rate": 4.370172190857801e-06, + "loss": 0.5672, + "step": 519 + }, + { + "epoch": 1.626304801670146, + "grad_norm": 0.803040623664856, + "learning_rate": 4.367296141855887e-06, + "loss": 0.5313, + "step": 520 + }, + { + "epoch": 1.6294363256784967, + "grad_norm": 0.8305794596672058, + "learning_rate": 4.3644144922060625e-06, + "loss": 0.5754, + "step": 521 + }, + { + "epoch": 1.6325678496868476, + "grad_norm": 1.0086486339569092, + "learning_rate": 4.361527250551361e-06, + "loss": 0.5433, + "step": 522 + }, + { + "epoch": 1.6356993736951984, + "grad_norm": 0.7217550277709961, + "learning_rate": 4.35863442555159e-06, + "loss": 0.524, + "step": 523 + }, + { + "epoch": 1.638830897703549, + "grad_norm": 0.7788524627685547, + "learning_rate": 4.355736025883303e-06, + "loss": 0.536, + "step": 524 + }, + { + "epoch": 1.6419624217118998, + "grad_norm": 0.8460550904273987, + "learning_rate": 4.352832060239774e-06, + "loss": 0.5381, + "step": 525 + }, + { + "epoch": 1.6450939457202505, + "grad_norm": 0.7571215033531189, + "learning_rate": 4.3499225373309675e-06, + "loss": 0.541, + "step": 526 + }, + { + "epoch": 1.6482254697286014, + "grad_norm": 0.7343226671218872, + "learning_rate": 4.347007465883523e-06, + "loss": 0.5147, + "step": 527 + }, + { + "epoch": 1.651356993736952, + "grad_norm": 0.7271892428398132, + "learning_rate": 4.3440868546407165e-06, + "loss": 0.5311, + "step": 528 + }, + { + "epoch": 1.6544885177453028, + "grad_norm": 0.8166136741638184, + "learning_rate": 4.341160712362442e-06, + "loss": 0.5379, + "step": 529 + }, + { + "epoch": 1.6576200417536535, + "grad_norm": 1.5985233783721924, + "learning_rate": 4.338229047825182e-06, + "loss": 0.5782, + "step": 530 + }, + { + "epoch": 1.6607515657620042, + "grad_norm": 0.7835702896118164, + "learning_rate": 4.3352918698219835e-06, + "loss": 0.525, + "step": 531 + }, + { + "epoch": 1.663883089770355, + "grad_norm": 0.7278687953948975, + "learning_rate": 4.332349187162428e-06, + "loss": 0.5266, + "step": 532 + }, + { + "epoch": 1.6670146137787056, + "grad_norm": 0.8240190148353577, + "learning_rate": 4.329401008672608e-06, + "loss": 0.5515, + "step": 533 + }, + { + "epoch": 1.6701461377870563, + "grad_norm": 0.9447080492973328, + "learning_rate": 4.326447343195102e-06, + "loss": 0.5596, + "step": 534 + }, + { + "epoch": 1.673277661795407, + "grad_norm": 0.7827372550964355, + "learning_rate": 4.323488199588944e-06, + "loss": 0.5466, + "step": 535 + }, + { + "epoch": 1.6764091858037578, + "grad_norm": 0.9252517223358154, + "learning_rate": 4.320523586729599e-06, + "loss": 0.5433, + "step": 536 + }, + { + "epoch": 1.6795407098121085, + "grad_norm": 0.9437504410743713, + "learning_rate": 4.317553513508934e-06, + "loss": 0.5552, + "step": 537 + }, + { + "epoch": 1.6826722338204592, + "grad_norm": 0.8972746133804321, + "learning_rate": 4.3145779888351986e-06, + "loss": 0.5259, + "step": 538 + }, + { + "epoch": 1.6858037578288099, + "grad_norm": 0.8017446994781494, + "learning_rate": 4.311597021632988e-06, + "loss": 0.5263, + "step": 539 + }, + { + "epoch": 1.6889352818371608, + "grad_norm": 0.7875497341156006, + "learning_rate": 4.3086106208432235e-06, + "loss": 0.5316, + "step": 540 + }, + { + "epoch": 1.6920668058455115, + "grad_norm": 0.8204905986785889, + "learning_rate": 4.305618795423125e-06, + "loss": 0.5506, + "step": 541 + }, + { + "epoch": 1.6951983298538622, + "grad_norm": 0.888359785079956, + "learning_rate": 4.30262155434618e-06, + "loss": 0.4825, + "step": 542 + }, + { + "epoch": 1.698329853862213, + "grad_norm": 1.1026058197021484, + "learning_rate": 4.29961890660212e-06, + "loss": 0.5321, + "step": 543 + }, + { + "epoch": 1.7014613778705638, + "grad_norm": 0.7662535905838013, + "learning_rate": 4.2966108611968945e-06, + "loss": 0.5432, + "step": 544 + }, + { + "epoch": 1.7045929018789145, + "grad_norm": 1.1951749324798584, + "learning_rate": 4.293597427152641e-06, + "loss": 0.5123, + "step": 545 + }, + { + "epoch": 1.7077244258872653, + "grad_norm": 1.303183913230896, + "learning_rate": 4.290578613507661e-06, + "loss": 0.5346, + "step": 546 + }, + { + "epoch": 1.710855949895616, + "grad_norm": 0.7653357982635498, + "learning_rate": 4.287554429316387e-06, + "loss": 0.5397, + "step": 547 + }, + { + "epoch": 1.7139874739039667, + "grad_norm": 0.796215295791626, + "learning_rate": 4.284524883649366e-06, + "loss": 0.5421, + "step": 548 + }, + { + "epoch": 1.7171189979123174, + "grad_norm": 0.7599332332611084, + "learning_rate": 4.281489985593219e-06, + "loss": 0.5289, + "step": 549 + }, + { + "epoch": 1.720250521920668, + "grad_norm": 0.8029115796089172, + "learning_rate": 4.2784497442506265e-06, + "loss": 0.5409, + "step": 550 + }, + { + "epoch": 1.7233820459290188, + "grad_norm": 0.7194099426269531, + "learning_rate": 4.275404168740291e-06, + "loss": 0.5327, + "step": 551 + }, + { + "epoch": 1.7265135699373695, + "grad_norm": 0.7960740923881531, + "learning_rate": 4.272353268196917e-06, + "loss": 0.4896, + "step": 552 + }, + { + "epoch": 1.7296450939457202, + "grad_norm": 0.9572116732597351, + "learning_rate": 4.269297051771178e-06, + "loss": 0.5402, + "step": 553 + }, + { + "epoch": 1.732776617954071, + "grad_norm": 1.3604938983917236, + "learning_rate": 4.266235528629695e-06, + "loss": 0.5792, + "step": 554 + }, + { + "epoch": 1.7359081419624216, + "grad_norm": 2.067286729812622, + "learning_rate": 4.263168707955002e-06, + "loss": 0.5033, + "step": 555 + }, + { + "epoch": 1.7390396659707723, + "grad_norm": 0.8031097054481506, + "learning_rate": 4.260096598945523e-06, + "loss": 0.5117, + "step": 556 + }, + { + "epoch": 1.742171189979123, + "grad_norm": 1.0241729021072388, + "learning_rate": 4.257019210815546e-06, + "loss": 0.5359, + "step": 557 + }, + { + "epoch": 1.745302713987474, + "grad_norm": 0.7625218629837036, + "learning_rate": 4.25393655279519e-06, + "loss": 0.5625, + "step": 558 + }, + { + "epoch": 1.7484342379958246, + "grad_norm": 0.8603503704071045, + "learning_rate": 4.250848634130381e-06, + "loss": 0.5043, + "step": 559 + }, + { + "epoch": 1.7515657620041754, + "grad_norm": 0.9543750286102295, + "learning_rate": 4.247755464082824e-06, + "loss": 0.5364, + "step": 560 + }, + { + "epoch": 1.754697286012526, + "grad_norm": 0.9707463979721069, + "learning_rate": 4.244657051929973e-06, + "loss": 0.5184, + "step": 561 + }, + { + "epoch": 1.757828810020877, + "grad_norm": 0.7491432428359985, + "learning_rate": 4.241553406965008e-06, + "loss": 0.559, + "step": 562 + }, + { + "epoch": 1.7609603340292277, + "grad_norm": 0.7444972991943359, + "learning_rate": 4.238444538496801e-06, + "loss": 0.5327, + "step": 563 + }, + { + "epoch": 1.7640918580375784, + "grad_norm": 2.7108678817749023, + "learning_rate": 4.235330455849892e-06, + "loss": 0.55, + "step": 564 + }, + { + "epoch": 1.767223382045929, + "grad_norm": 1.6716049909591675, + "learning_rate": 4.232211168364459e-06, + "loss": 0.5093, + "step": 565 + }, + { + "epoch": 1.7703549060542798, + "grad_norm": 0.7023475170135498, + "learning_rate": 4.229086685396295e-06, + "loss": 0.569, + "step": 566 + }, + { + "epoch": 1.7734864300626305, + "grad_norm": 0.8596265316009521, + "learning_rate": 4.225957016316771e-06, + "loss": 0.5128, + "step": 567 + }, + { + "epoch": 1.7766179540709812, + "grad_norm": 0.8110849857330322, + "learning_rate": 4.222822170512816e-06, + "loss": 0.5142, + "step": 568 + }, + { + "epoch": 1.779749478079332, + "grad_norm": 0.7583725452423096, + "learning_rate": 4.219682157386884e-06, + "loss": 0.5584, + "step": 569 + }, + { + "epoch": 1.7828810020876826, + "grad_norm": 0.787811279296875, + "learning_rate": 4.21653698635693e-06, + "loss": 0.5068, + "step": 570 + }, + { + "epoch": 1.7860125260960333, + "grad_norm": 0.8298993110656738, + "learning_rate": 4.213386666856375e-06, + "loss": 0.5496, + "step": 571 + }, + { + "epoch": 1.789144050104384, + "grad_norm": 0.8999841213226318, + "learning_rate": 4.210231208334087e-06, + "loss": 0.5454, + "step": 572 + }, + { + "epoch": 1.7922755741127347, + "grad_norm": 4.264521598815918, + "learning_rate": 4.207070620254345e-06, + "loss": 0.5486, + "step": 573 + }, + { + "epoch": 1.7954070981210855, + "grad_norm": 0.8517448306083679, + "learning_rate": 4.203904912096812e-06, + "loss": 0.5566, + "step": 574 + }, + { + "epoch": 1.7985386221294362, + "grad_norm": 0.9230182766914368, + "learning_rate": 4.200734093356511e-06, + "loss": 0.4964, + "step": 575 + }, + { + "epoch": 1.801670146137787, + "grad_norm": 1.224039912223816, + "learning_rate": 4.197558173543791e-06, + "loss": 0.5356, + "step": 576 + }, + { + "epoch": 1.8048016701461378, + "grad_norm": 0.9998573660850525, + "learning_rate": 4.194377162184301e-06, + "loss": 0.5334, + "step": 577 + }, + { + "epoch": 1.8079331941544885, + "grad_norm": 0.865521252155304, + "learning_rate": 4.191191068818963e-06, + "loss": 0.5036, + "step": 578 + }, + { + "epoch": 1.8110647181628392, + "grad_norm": 0.8048138618469238, + "learning_rate": 4.18799990300394e-06, + "loss": 0.4979, + "step": 579 + }, + { + "epoch": 1.8141962421711901, + "grad_norm": 0.717815637588501, + "learning_rate": 4.184803674310609e-06, + "loss": 0.5623, + "step": 580 + }, + { + "epoch": 1.8173277661795408, + "grad_norm": 0.8403327465057373, + "learning_rate": 4.1816023923255335e-06, + "loss": 0.5055, + "step": 581 + }, + { + "epoch": 1.8204592901878915, + "grad_norm": 0.7298995852470398, + "learning_rate": 4.178396066650432e-06, + "loss": 0.5641, + "step": 582 + }, + { + "epoch": 1.8235908141962422, + "grad_norm": 0.9469727873802185, + "learning_rate": 4.1751847069021516e-06, + "loss": 0.5557, + "step": 583 + }, + { + "epoch": 1.826722338204593, + "grad_norm": 0.8641784191131592, + "learning_rate": 4.1719683227126386e-06, + "loss": 0.5153, + "step": 584 + }, + { + "epoch": 1.8298538622129437, + "grad_norm": 0.7316668629646301, + "learning_rate": 4.168746923728908e-06, + "loss": 0.4988, + "step": 585 + }, + { + "epoch": 1.8329853862212944, + "grad_norm": 0.8795468807220459, + "learning_rate": 4.165520519613017e-06, + "loss": 0.5483, + "step": 586 + }, + { + "epoch": 1.836116910229645, + "grad_norm": 0.7323560118675232, + "learning_rate": 4.162289120042034e-06, + "loss": 0.5194, + "step": 587 + }, + { + "epoch": 1.8392484342379958, + "grad_norm": 0.8217021822929382, + "learning_rate": 4.159052734708013e-06, + "loss": 0.532, + "step": 588 + }, + { + "epoch": 1.8423799582463465, + "grad_norm": 0.7669674754142761, + "learning_rate": 4.155811373317958e-06, + "loss": 0.541, + "step": 589 + }, + { + "epoch": 1.8455114822546972, + "grad_norm": 0.8312156200408936, + "learning_rate": 4.152565045593801e-06, + "loss": 0.5298, + "step": 590 + }, + { + "epoch": 1.848643006263048, + "grad_norm": 0.8967565298080444, + "learning_rate": 4.1493137612723665e-06, + "loss": 0.51, + "step": 591 + }, + { + "epoch": 1.8517745302713986, + "grad_norm": 0.8706664443016052, + "learning_rate": 4.14605753010535e-06, + "loss": 0.4941, + "step": 592 + }, + { + "epoch": 1.8549060542797495, + "grad_norm": 0.7585753798484802, + "learning_rate": 4.14279636185928e-06, + "loss": 0.5161, + "step": 593 + }, + { + "epoch": 1.8580375782881002, + "grad_norm": 0.7495241165161133, + "learning_rate": 4.1395302663154954e-06, + "loss": 0.5388, + "step": 594 + }, + { + "epoch": 1.861169102296451, + "grad_norm": 1.0746862888336182, + "learning_rate": 4.136259253270114e-06, + "loss": 0.4976, + "step": 595 + }, + { + "epoch": 1.8643006263048016, + "grad_norm": 0.872309684753418, + "learning_rate": 4.132983332534e-06, + "loss": 0.559, + "step": 596 + }, + { + "epoch": 1.8674321503131524, + "grad_norm": 0.8759891986846924, + "learning_rate": 4.1297025139327405e-06, + "loss": 0.5436, + "step": 597 + }, + { + "epoch": 1.8705636743215033, + "grad_norm": 1.1044493913650513, + "learning_rate": 4.126416807306611e-06, + "loss": 0.5476, + "step": 598 + }, + { + "epoch": 1.873695198329854, + "grad_norm": 0.8340442180633545, + "learning_rate": 4.123126222510549e-06, + "loss": 0.4592, + "step": 599 + }, + { + "epoch": 1.8768267223382047, + "grad_norm": 0.8331449031829834, + "learning_rate": 4.119830769414123e-06, + "loss": 0.5219, + "step": 600 + }, + { + "epoch": 1.8799582463465554, + "grad_norm": 1.0862973928451538, + "learning_rate": 4.116530457901503e-06, + "loss": 0.5159, + "step": 601 + }, + { + "epoch": 1.883089770354906, + "grad_norm": 0.8524414300918579, + "learning_rate": 4.113225297871431e-06, + "loss": 0.5502, + "step": 602 + }, + { + "epoch": 1.8862212943632568, + "grad_norm": 1.4945416450500488, + "learning_rate": 4.10991529923719e-06, + "loss": 0.5627, + "step": 603 + }, + { + "epoch": 1.8893528183716075, + "grad_norm": 1.5518157482147217, + "learning_rate": 4.10660047192658e-06, + "loss": 0.5517, + "step": 604 + }, + { + "epoch": 1.8924843423799582, + "grad_norm": 2.56638765335083, + "learning_rate": 4.103280825881878e-06, + "loss": 0.5422, + "step": 605 + }, + { + "epoch": 1.895615866388309, + "grad_norm": 0.867254912853241, + "learning_rate": 4.099956371059817e-06, + "loss": 0.4991, + "step": 606 + }, + { + "epoch": 1.8987473903966596, + "grad_norm": 0.9555892944335938, + "learning_rate": 4.096627117431554e-06, + "loss": 0.5339, + "step": 607 + }, + { + "epoch": 1.9018789144050103, + "grad_norm": 0.7905483245849609, + "learning_rate": 4.093293074982638e-06, + "loss": 0.5168, + "step": 608 + }, + { + "epoch": 1.905010438413361, + "grad_norm": 0.7500227093696594, + "learning_rate": 4.089954253712981e-06, + "loss": 0.5096, + "step": 609 + }, + { + "epoch": 1.9081419624217117, + "grad_norm": 0.8458324074745178, + "learning_rate": 4.086610663636828e-06, + "loss": 0.5296, + "step": 610 + }, + { + "epoch": 1.9112734864300627, + "grad_norm": 0.7392706871032715, + "learning_rate": 4.08326231478273e-06, + "loss": 0.5305, + "step": 611 + }, + { + "epoch": 1.9144050104384134, + "grad_norm": 0.8113343715667725, + "learning_rate": 4.079909217193508e-06, + "loss": 0.5044, + "step": 612 + }, + { + "epoch": 1.917536534446764, + "grad_norm": 0.7637801766395569, + "learning_rate": 4.076551380926226e-06, + "loss": 0.5298, + "step": 613 + }, + { + "epoch": 1.9206680584551148, + "grad_norm": 1.0523375272750854, + "learning_rate": 4.073188816052164e-06, + "loss": 0.5111, + "step": 614 + }, + { + "epoch": 1.9237995824634657, + "grad_norm": 0.8224868774414062, + "learning_rate": 4.069821532656781e-06, + "loss": 0.5178, + "step": 615 + }, + { + "epoch": 1.9269311064718164, + "grad_norm": 0.7270777821540833, + "learning_rate": 4.066449540839693e-06, + "loss": 0.5307, + "step": 616 + }, + { + "epoch": 1.9300626304801671, + "grad_norm": 0.7214602828025818, + "learning_rate": 4.063072850714631e-06, + "loss": 0.5171, + "step": 617 + }, + { + "epoch": 1.9331941544885178, + "grad_norm": 0.7333671450614929, + "learning_rate": 4.059691472409426e-06, + "loss": 0.56, + "step": 618 + }, + { + "epoch": 1.9363256784968685, + "grad_norm": 0.9166824221611023, + "learning_rate": 4.056305416065964e-06, + "loss": 0.5388, + "step": 619 + }, + { + "epoch": 1.9394572025052192, + "grad_norm": 0.7743303775787354, + "learning_rate": 4.052914691840167e-06, + "loss": 0.5134, + "step": 620 + }, + { + "epoch": 1.94258872651357, + "grad_norm": 0.704097330570221, + "learning_rate": 4.0495193099019524e-06, + "loss": 0.4926, + "step": 621 + }, + { + "epoch": 1.9457202505219207, + "grad_norm": 0.8508503437042236, + "learning_rate": 4.046119280435212e-06, + "loss": 0.5008, + "step": 622 + }, + { + "epoch": 1.9488517745302714, + "grad_norm": 0.725933313369751, + "learning_rate": 4.042714613637775e-06, + "loss": 0.5549, + "step": 623 + }, + { + "epoch": 1.951983298538622, + "grad_norm": 0.8919175863265991, + "learning_rate": 4.039305319721381e-06, + "loss": 0.5183, + "step": 624 + }, + { + "epoch": 1.9551148225469728, + "grad_norm": 0.827919065952301, + "learning_rate": 4.035891408911644e-06, + "loss": 0.5624, + "step": 625 + }, + { + "epoch": 1.9582463465553235, + "grad_norm": 0.7415187358856201, + "learning_rate": 4.032472891448032e-06, + "loss": 0.5454, + "step": 626 + }, + { + "epoch": 1.9613778705636742, + "grad_norm": 0.7675788998603821, + "learning_rate": 4.029049777583824e-06, + "loss": 0.5361, + "step": 627 + }, + { + "epoch": 1.964509394572025, + "grad_norm": 0.8464030623435974, + "learning_rate": 4.025622077586088e-06, + "loss": 0.5295, + "step": 628 + }, + { + "epoch": 1.9676409185803758, + "grad_norm": 0.7641633749008179, + "learning_rate": 4.022189801735646e-06, + "loss": 0.55, + "step": 629 + }, + { + "epoch": 1.9707724425887265, + "grad_norm": 0.7813227772712708, + "learning_rate": 4.018752960327048e-06, + "loss": 0.5587, + "step": 630 + }, + { + "epoch": 1.9739039665970772, + "grad_norm": 0.7576701641082764, + "learning_rate": 4.015311563668533e-06, + "loss": 0.5413, + "step": 631 + }, + { + "epoch": 1.977035490605428, + "grad_norm": 0.6949650049209595, + "learning_rate": 4.011865622082004e-06, + "loss": 0.5344, + "step": 632 + }, + { + "epoch": 1.9801670146137789, + "grad_norm": 0.9009145498275757, + "learning_rate": 4.008415145902997e-06, + "loss": 0.5233, + "step": 633 + }, + { + "epoch": 1.9832985386221296, + "grad_norm": 0.7635822892189026, + "learning_rate": 4.004960145480651e-06, + "loss": 0.4981, + "step": 634 + }, + { + "epoch": 1.9864300626304803, + "grad_norm": 0.8916334509849548, + "learning_rate": 4.0015006311776685e-06, + "loss": 0.5311, + "step": 635 + }, + { + "epoch": 1.989561586638831, + "grad_norm": 0.7197673320770264, + "learning_rate": 3.998036613370295e-06, + "loss": 0.5361, + "step": 636 + }, + { + "epoch": 1.9926931106471817, + "grad_norm": 0.8391228914260864, + "learning_rate": 3.994568102448284e-06, + "loss": 0.5473, + "step": 637 + }, + { + "epoch": 1.9958246346555324, + "grad_norm": 0.9371750950813293, + "learning_rate": 3.991095108814862e-06, + "loss": 0.5303, + "step": 638 + }, + { + "epoch": 1.998956158663883, + "grad_norm": 0.8929619789123535, + "learning_rate": 3.9876176428867046e-06, + "loss": 0.533, + "step": 639 + }, + { + "epoch": 2.0, + "grad_norm": 0.8929619789123535, + "learning_rate": 3.9841357150938984e-06, + "loss": 0.1831, + "step": 640 + }, + { + "epoch": 2.0031315240083507, + "grad_norm": 0.8802503347396851, + "learning_rate": 3.9806493358799135e-06, + "loss": 0.493, + "step": 641 + }, + { + "epoch": 2.0062630480167014, + "grad_norm": 0.802759051322937, + "learning_rate": 3.977158515701571e-06, + "loss": 0.498, + "step": 642 + }, + { + "epoch": 2.009394572025052, + "grad_norm": 1.0235401391983032, + "learning_rate": 3.973663265029013e-06, + "loss": 0.4887, + "step": 643 + }, + { + "epoch": 2.012526096033403, + "grad_norm": 0.7219089865684509, + "learning_rate": 3.97016359434567e-06, + "loss": 0.4628, + "step": 644 + }, + { + "epoch": 2.0156576200417535, + "grad_norm": 0.7887073755264282, + "learning_rate": 3.966659514148229e-06, + "loss": 0.525, + "step": 645 + }, + { + "epoch": 2.0187891440501042, + "grad_norm": 0.7960914969444275, + "learning_rate": 3.963151034946602e-06, + "loss": 0.4643, + "step": 646 + }, + { + "epoch": 2.021920668058455, + "grad_norm": 0.7902271151542664, + "learning_rate": 3.959638167263895e-06, + "loss": 0.4922, + "step": 647 + }, + { + "epoch": 2.0250521920668056, + "grad_norm": 0.9501478672027588, + "learning_rate": 3.956120921636379e-06, + "loss": 0.5285, + "step": 648 + }, + { + "epoch": 2.028183716075157, + "grad_norm": 0.9510527849197388, + "learning_rate": 3.952599308613454e-06, + "loss": 0.4909, + "step": 649 + }, + { + "epoch": 2.0313152400835075, + "grad_norm": 0.9408219456672668, + "learning_rate": 3.949073338757619e-06, + "loss": 0.4912, + "step": 650 + }, + { + "epoch": 2.034446764091858, + "grad_norm": 0.7148041725158691, + "learning_rate": 3.945543022644441e-06, + "loss": 0.4792, + "step": 651 + }, + { + "epoch": 2.037578288100209, + "grad_norm": 0.7737464904785156, + "learning_rate": 3.942008370862522e-06, + "loss": 0.4694, + "step": 652 + }, + { + "epoch": 2.0407098121085596, + "grad_norm": 0.8405889868736267, + "learning_rate": 3.938469394013472e-06, + "loss": 0.5048, + "step": 653 + }, + { + "epoch": 2.0438413361169103, + "grad_norm": 0.7896456718444824, + "learning_rate": 3.934926102711869e-06, + "loss": 0.4882, + "step": 654 + }, + { + "epoch": 2.046972860125261, + "grad_norm": 0.9290387034416199, + "learning_rate": 3.931378507585231e-06, + "loss": 0.503, + "step": 655 + }, + { + "epoch": 2.0501043841336117, + "grad_norm": 0.7386118769645691, + "learning_rate": 3.927826619273991e-06, + "loss": 0.4918, + "step": 656 + }, + { + "epoch": 2.0532359081419624, + "grad_norm": 0.9878676533699036, + "learning_rate": 3.92427044843145e-06, + "loss": 0.4958, + "step": 657 + }, + { + "epoch": 2.056367432150313, + "grad_norm": 1.0111151933670044, + "learning_rate": 3.92071000572376e-06, + "loss": 0.4886, + "step": 658 + }, + { + "epoch": 2.059498956158664, + "grad_norm": 0.8612061738967896, + "learning_rate": 3.917145301829884e-06, + "loss": 0.5216, + "step": 659 + }, + { + "epoch": 2.0626304801670146, + "grad_norm": 0.7458518743515015, + "learning_rate": 3.913576347441564e-06, + "loss": 0.4807, + "step": 660 + }, + { + "epoch": 2.0657620041753653, + "grad_norm": 0.7775886058807373, + "learning_rate": 3.910003153263294e-06, + "loss": 0.4837, + "step": 661 + }, + { + "epoch": 2.068893528183716, + "grad_norm": 0.7144196629524231, + "learning_rate": 3.906425730012282e-06, + "loss": 0.5081, + "step": 662 + }, + { + "epoch": 2.0720250521920667, + "grad_norm": 0.844971776008606, + "learning_rate": 3.9028440884184215e-06, + "loss": 0.474, + "step": 663 + }, + { + "epoch": 2.0751565762004174, + "grad_norm": 0.9709283113479614, + "learning_rate": 3.899258239224257e-06, + "loss": 0.503, + "step": 664 + }, + { + "epoch": 2.078288100208768, + "grad_norm": 1.1325515508651733, + "learning_rate": 3.895668193184954e-06, + "loss": 0.5058, + "step": 665 + }, + { + "epoch": 2.081419624217119, + "grad_norm": 0.7211254239082336, + "learning_rate": 3.892073961068266e-06, + "loss": 0.4982, + "step": 666 + }, + { + "epoch": 2.08455114822547, + "grad_norm": 0.8975517153739929, + "learning_rate": 3.888475553654502e-06, + "loss": 0.4699, + "step": 667 + }, + { + "epoch": 2.0876826722338206, + "grad_norm": 0.8270771503448486, + "learning_rate": 3.884872981736493e-06, + "loss": 0.4586, + "step": 668 + }, + { + "epoch": 2.0908141962421714, + "grad_norm": 0.8606625199317932, + "learning_rate": 3.881266256119561e-06, + "loss": 0.5299, + "step": 669 + }, + { + "epoch": 2.093945720250522, + "grad_norm": 0.9013976454734802, + "learning_rate": 3.877655387621488e-06, + "loss": 0.4887, + "step": 670 + }, + { + "epoch": 2.0970772442588728, + "grad_norm": 0.7603903412818909, + "learning_rate": 3.8740403870724795e-06, + "loss": 0.4992, + "step": 671 + }, + { + "epoch": 2.1002087682672235, + "grad_norm": 1.0432350635528564, + "learning_rate": 3.870421265315137e-06, + "loss": 0.5035, + "step": 672 + }, + { + "epoch": 2.103340292275574, + "grad_norm": 0.7727136611938477, + "learning_rate": 3.8667980332044195e-06, + "loss": 0.5006, + "step": 673 + }, + { + "epoch": 2.106471816283925, + "grad_norm": 0.9764307141304016, + "learning_rate": 3.863170701607618e-06, + "loss": 0.5061, + "step": 674 + }, + { + "epoch": 2.1096033402922756, + "grad_norm": 0.747818648815155, + "learning_rate": 3.859539281404317e-06, + "loss": 0.4761, + "step": 675 + }, + { + "epoch": 2.1127348643006263, + "grad_norm": 0.7254915237426758, + "learning_rate": 3.855903783486364e-06, + "loss": 0.5166, + "step": 676 + }, + { + "epoch": 2.115866388308977, + "grad_norm": 0.7678592801094055, + "learning_rate": 3.852264218757839e-06, + "loss": 0.5122, + "step": 677 + }, + { + "epoch": 2.1189979123173277, + "grad_norm": 0.8140144348144531, + "learning_rate": 3.8486205981350165e-06, + "loss": 0.4551, + "step": 678 + }, + { + "epoch": 2.1221294363256784, + "grad_norm": 0.9417359232902527, + "learning_rate": 3.844972932546338e-06, + "loss": 0.4748, + "step": 679 + }, + { + "epoch": 2.125260960334029, + "grad_norm": 0.8035290241241455, + "learning_rate": 3.841321232932378e-06, + "loss": 0.5079, + "step": 680 + }, + { + "epoch": 2.12839248434238, + "grad_norm": 0.8300641775131226, + "learning_rate": 3.837665510245809e-06, + "loss": 0.5018, + "step": 681 + }, + { + "epoch": 2.1315240083507305, + "grad_norm": 0.8293547034263611, + "learning_rate": 3.8340057754513715e-06, + "loss": 0.5042, + "step": 682 + }, + { + "epoch": 2.1346555323590812, + "grad_norm": 0.7780007719993591, + "learning_rate": 3.8303420395258365e-06, + "loss": 0.5048, + "step": 683 + }, + { + "epoch": 2.137787056367432, + "grad_norm": 0.7519420385360718, + "learning_rate": 3.8266743134579785e-06, + "loss": 0.5108, + "step": 684 + }, + { + "epoch": 2.140918580375783, + "grad_norm": 0.7872384190559387, + "learning_rate": 3.8230026082485404e-06, + "loss": 0.4924, + "step": 685 + }, + { + "epoch": 2.144050104384134, + "grad_norm": 0.7479491829872131, + "learning_rate": 3.819326934910197e-06, + "loss": 0.5184, + "step": 686 + }, + { + "epoch": 2.1471816283924845, + "grad_norm": 0.8438299298286438, + "learning_rate": 3.815647304467527e-06, + "loss": 0.4791, + "step": 687 + }, + { + "epoch": 2.150313152400835, + "grad_norm": 0.7923721671104431, + "learning_rate": 3.8119637279569773e-06, + "loss": 0.5305, + "step": 688 + }, + { + "epoch": 2.153444676409186, + "grad_norm": 0.7856534719467163, + "learning_rate": 3.80827621642683e-06, + "loss": 0.5063, + "step": 689 + }, + { + "epoch": 2.1565762004175366, + "grad_norm": 0.8544500470161438, + "learning_rate": 3.8045847809371706e-06, + "loss": 0.4989, + "step": 690 + }, + { + "epoch": 2.1597077244258873, + "grad_norm": 0.865390956401825, + "learning_rate": 3.800889432559852e-06, + "loss": 0.4931, + "step": 691 + }, + { + "epoch": 2.162839248434238, + "grad_norm": 0.9809399247169495, + "learning_rate": 3.797190182378466e-06, + "loss": 0.4785, + "step": 692 + }, + { + "epoch": 2.1659707724425887, + "grad_norm": 0.7954536080360413, + "learning_rate": 3.793487041488304e-06, + "loss": 0.4847, + "step": 693 + }, + { + "epoch": 2.1691022964509394, + "grad_norm": 0.754704475402832, + "learning_rate": 3.7897800209963298e-06, + "loss": 0.5125, + "step": 694 + }, + { + "epoch": 2.17223382045929, + "grad_norm": 0.7319822311401367, + "learning_rate": 3.7860691320211414e-06, + "loss": 0.477, + "step": 695 + }, + { + "epoch": 2.175365344467641, + "grad_norm": 0.8198635578155518, + "learning_rate": 3.7823543856929403e-06, + "loss": 0.4764, + "step": 696 + }, + { + "epoch": 2.1784968684759916, + "grad_norm": 0.708933413028717, + "learning_rate": 3.7786357931534987e-06, + "loss": 0.4948, + "step": 697 + }, + { + "epoch": 2.1816283924843423, + "grad_norm": 0.8493193984031677, + "learning_rate": 3.774913365556123e-06, + "loss": 0.5271, + "step": 698 + }, + { + "epoch": 2.184759916492693, + "grad_norm": 0.7999475002288818, + "learning_rate": 3.771187114065622e-06, + "loss": 0.4804, + "step": 699 + }, + { + "epoch": 2.1878914405010437, + "grad_norm": 0.8366796970367432, + "learning_rate": 3.7674570498582776e-06, + "loss": 0.457, + "step": 700 + }, + { + "epoch": 2.1910229645093944, + "grad_norm": 0.7935530543327332, + "learning_rate": 3.7637231841218015e-06, + "loss": 0.5001, + "step": 701 + }, + { + "epoch": 2.1941544885177455, + "grad_norm": 0.7700153589248657, + "learning_rate": 3.7599855280553125e-06, + "loss": 0.5091, + "step": 702 + }, + { + "epoch": 2.1972860125260962, + "grad_norm": 0.7991652488708496, + "learning_rate": 3.756244092869294e-06, + "loss": 0.4955, + "step": 703 + }, + { + "epoch": 2.200417536534447, + "grad_norm": 0.720051646232605, + "learning_rate": 3.752498889785567e-06, + "loss": 0.4902, + "step": 704 + }, + { + "epoch": 2.2035490605427976, + "grad_norm": 0.7312369346618652, + "learning_rate": 3.748749930037252e-06, + "loss": 0.4935, + "step": 705 + }, + { + "epoch": 2.2066805845511483, + "grad_norm": 0.8418563008308411, + "learning_rate": 3.744997224868739e-06, + "loss": 0.5186, + "step": 706 + }, + { + "epoch": 2.209812108559499, + "grad_norm": 0.8324081301689148, + "learning_rate": 3.741240785535649e-06, + "loss": 0.482, + "step": 707 + }, + { + "epoch": 2.2129436325678498, + "grad_norm": 0.8051855564117432, + "learning_rate": 3.737480623304805e-06, + "loss": 0.4663, + "step": 708 + }, + { + "epoch": 2.2160751565762005, + "grad_norm": 0.9464184641838074, + "learning_rate": 3.7337167494541948e-06, + "loss": 0.451, + "step": 709 + }, + { + "epoch": 2.219206680584551, + "grad_norm": 1.0227075815200806, + "learning_rate": 3.729949175272942e-06, + "loss": 0.4428, + "step": 710 + }, + { + "epoch": 2.222338204592902, + "grad_norm": 0.7930364012718201, + "learning_rate": 3.7261779120612633e-06, + "loss": 0.5132, + "step": 711 + }, + { + "epoch": 2.2254697286012526, + "grad_norm": 0.9033688306808472, + "learning_rate": 3.7224029711304444e-06, + "loss": 0.476, + "step": 712 + }, + { + "epoch": 2.2286012526096033, + "grad_norm": 0.8026887774467468, + "learning_rate": 3.7186243638028007e-06, + "loss": 0.4959, + "step": 713 + }, + { + "epoch": 2.231732776617954, + "grad_norm": 0.9391745328903198, + "learning_rate": 3.714842101411642e-06, + "loss": 0.4962, + "step": 714 + }, + { + "epoch": 2.2348643006263047, + "grad_norm": 0.7774361371994019, + "learning_rate": 3.711056195301245e-06, + "loss": 0.4748, + "step": 715 + }, + { + "epoch": 2.2379958246346554, + "grad_norm": 0.9278722405433655, + "learning_rate": 3.7072666568268115e-06, + "loss": 0.5074, + "step": 716 + }, + { + "epoch": 2.241127348643006, + "grad_norm": 0.771423876285553, + "learning_rate": 3.7034734973544406e-06, + "loss": 0.5072, + "step": 717 + }, + { + "epoch": 2.244258872651357, + "grad_norm": 0.8707448244094849, + "learning_rate": 3.6996767282610892e-06, + "loss": 0.4851, + "step": 718 + }, + { + "epoch": 2.2473903966597075, + "grad_norm": 0.7641019821166992, + "learning_rate": 3.695876360934543e-06, + "loss": 0.4941, + "step": 719 + }, + { + "epoch": 2.2505219206680582, + "grad_norm": 0.7647167444229126, + "learning_rate": 3.69207240677338e-06, + "loss": 0.5225, + "step": 720 + }, + { + "epoch": 2.2536534446764094, + "grad_norm": 0.9108865261077881, + "learning_rate": 3.6882648771869345e-06, + "loss": 0.454, + "step": 721 + }, + { + "epoch": 2.25678496868476, + "grad_norm": 0.86728835105896, + "learning_rate": 3.6844537835952666e-06, + "loss": 0.4461, + "step": 722 + }, + { + "epoch": 2.259916492693111, + "grad_norm": 1.1055282354354858, + "learning_rate": 3.6806391374291238e-06, + "loss": 0.4618, + "step": 723 + }, + { + "epoch": 2.2630480167014615, + "grad_norm": 0.7591858506202698, + "learning_rate": 3.6768209501299116e-06, + "loss": 0.4901, + "step": 724 + }, + { + "epoch": 2.266179540709812, + "grad_norm": 0.7966359257698059, + "learning_rate": 3.6729992331496554e-06, + "loss": 0.5171, + "step": 725 + }, + { + "epoch": 2.269311064718163, + "grad_norm": 0.983161211013794, + "learning_rate": 3.6691739979509672e-06, + "loss": 0.4949, + "step": 726 + }, + { + "epoch": 2.2724425887265136, + "grad_norm": 0.9200037121772766, + "learning_rate": 3.6653452560070106e-06, + "loss": 0.5234, + "step": 727 + }, + { + "epoch": 2.2755741127348643, + "grad_norm": 1.0288461446762085, + "learning_rate": 3.6615130188014685e-06, + "loss": 0.4713, + "step": 728 + }, + { + "epoch": 2.278705636743215, + "grad_norm": 0.7325463891029358, + "learning_rate": 3.6576772978285065e-06, + "loss": 0.527, + "step": 729 + }, + { + "epoch": 2.2818371607515657, + "grad_norm": 1.0045446157455444, + "learning_rate": 3.6538381045927395e-06, + "loss": 0.5139, + "step": 730 + }, + { + "epoch": 2.2849686847599164, + "grad_norm": 0.7391849756240845, + "learning_rate": 3.6499954506091963e-06, + "loss": 0.4829, + "step": 731 + }, + { + "epoch": 2.288100208768267, + "grad_norm": 0.7808229923248291, + "learning_rate": 3.646149347403286e-06, + "loss": 0.4831, + "step": 732 + }, + { + "epoch": 2.291231732776618, + "grad_norm": 0.7056961059570312, + "learning_rate": 3.6422998065107628e-06, + "loss": 0.5066, + "step": 733 + }, + { + "epoch": 2.2943632567849686, + "grad_norm": 0.7498443126678467, + "learning_rate": 3.6384468394776935e-06, + "loss": 0.4724, + "step": 734 + }, + { + "epoch": 2.2974947807933193, + "grad_norm": 0.8511576056480408, + "learning_rate": 3.634590457860418e-06, + "loss": 0.5286, + "step": 735 + }, + { + "epoch": 2.30062630480167, + "grad_norm": 0.873635470867157, + "learning_rate": 3.63073067322552e-06, + "loss": 0.4751, + "step": 736 + }, + { + "epoch": 2.3037578288100207, + "grad_norm": 0.7427377104759216, + "learning_rate": 3.626867497149788e-06, + "loss": 0.475, + "step": 737 + }, + { + "epoch": 2.306889352818372, + "grad_norm": 1.0591017007827759, + "learning_rate": 3.623000941220186e-06, + "loss": 0.4591, + "step": 738 + }, + { + "epoch": 2.3100208768267225, + "grad_norm": 0.8767879009246826, + "learning_rate": 3.6191310170338114e-06, + "loss": 0.4673, + "step": 739 + }, + { + "epoch": 2.3131524008350732, + "grad_norm": 0.9156234860420227, + "learning_rate": 3.615257736197866e-06, + "loss": 0.4622, + "step": 740 + }, + { + "epoch": 2.316283924843424, + "grad_norm": 0.6743756532669067, + "learning_rate": 3.611381110329619e-06, + "loss": 0.4723, + "step": 741 + }, + { + "epoch": 2.3194154488517746, + "grad_norm": 0.8655558228492737, + "learning_rate": 3.6075011510563732e-06, + "loss": 0.471, + "step": 742 + }, + { + "epoch": 2.3225469728601253, + "grad_norm": 0.7652033567428589, + "learning_rate": 3.603617870015429e-06, + "loss": 0.5155, + "step": 743 + }, + { + "epoch": 2.325678496868476, + "grad_norm": 0.7970699667930603, + "learning_rate": 3.599731278854049e-06, + "loss": 0.4507, + "step": 744 + }, + { + "epoch": 2.3288100208768268, + "grad_norm": 0.7538278698921204, + "learning_rate": 3.5958413892294253e-06, + "loss": 0.5093, + "step": 745 + }, + { + "epoch": 2.3319415448851775, + "grad_norm": 0.735996663570404, + "learning_rate": 3.5919482128086414e-06, + "loss": 0.5008, + "step": 746 + }, + { + "epoch": 2.335073068893528, + "grad_norm": 0.7643904685974121, + "learning_rate": 3.588051761268642e-06, + "loss": 0.5072, + "step": 747 + }, + { + "epoch": 2.338204592901879, + "grad_norm": 0.7646260857582092, + "learning_rate": 3.584152046296191e-06, + "loss": 0.4578, + "step": 748 + }, + { + "epoch": 2.3413361169102296, + "grad_norm": 0.7873825430870056, + "learning_rate": 3.5802490795878446e-06, + "loss": 0.5249, + "step": 749 + }, + { + "epoch": 2.3444676409185803, + "grad_norm": 1.095333218574524, + "learning_rate": 3.5763428728499095e-06, + "loss": 0.4913, + "step": 750 + }, + { + "epoch": 2.347599164926931, + "grad_norm": 1.3425395488739014, + "learning_rate": 3.5724334377984107e-06, + "loss": 0.5317, + "step": 751 + }, + { + "epoch": 2.3507306889352817, + "grad_norm": 0.7151113748550415, + "learning_rate": 3.568520786159055e-06, + "loss": 0.5135, + "step": 752 + }, + { + "epoch": 2.3538622129436324, + "grad_norm": 0.8072878122329712, + "learning_rate": 3.5646049296672004e-06, + "loss": 0.4863, + "step": 753 + }, + { + "epoch": 2.356993736951983, + "grad_norm": 0.8040189743041992, + "learning_rate": 3.5606858800678123e-06, + "loss": 0.4668, + "step": 754 + }, + { + "epoch": 2.3601252609603343, + "grad_norm": 0.7749765515327454, + "learning_rate": 3.5567636491154385e-06, + "loss": 0.4681, + "step": 755 + }, + { + "epoch": 2.3632567849686845, + "grad_norm": 0.773013710975647, + "learning_rate": 3.5528382485741638e-06, + "loss": 0.5012, + "step": 756 + }, + { + "epoch": 2.3663883089770357, + "grad_norm": 0.7017714381217957, + "learning_rate": 3.5489096902175835e-06, + "loss": 0.5019, + "step": 757 + }, + { + "epoch": 2.3695198329853864, + "grad_norm": 1.132458209991455, + "learning_rate": 3.5449779858287625e-06, + "loss": 0.5131, + "step": 758 + }, + { + "epoch": 2.372651356993737, + "grad_norm": 0.7624574899673462, + "learning_rate": 3.541043147200202e-06, + "loss": 0.4856, + "step": 759 + }, + { + "epoch": 2.375782881002088, + "grad_norm": 0.9078478217124939, + "learning_rate": 3.5371051861338036e-06, + "loss": 0.4337, + "step": 760 + }, + { + "epoch": 2.3789144050104385, + "grad_norm": 0.8608354330062866, + "learning_rate": 3.5331641144408344e-06, + "loss": 0.5053, + "step": 761 + }, + { + "epoch": 2.382045929018789, + "grad_norm": 0.775047779083252, + "learning_rate": 3.529219943941892e-06, + "loss": 0.4779, + "step": 762 + }, + { + "epoch": 2.38517745302714, + "grad_norm": 0.7775866389274597, + "learning_rate": 3.525272686466866e-06, + "loss": 0.4979, + "step": 763 + }, + { + "epoch": 2.3883089770354906, + "grad_norm": 0.9386464357376099, + "learning_rate": 3.521322353854908e-06, + "loss": 0.5222, + "step": 764 + }, + { + "epoch": 2.3914405010438413, + "grad_norm": 0.874109148979187, + "learning_rate": 3.517368957954391e-06, + "loss": 0.4681, + "step": 765 + }, + { + "epoch": 2.394572025052192, + "grad_norm": 0.824588418006897, + "learning_rate": 3.5134125106228766e-06, + "loss": 0.4955, + "step": 766 + }, + { + "epoch": 2.3977035490605427, + "grad_norm": 0.8790764808654785, + "learning_rate": 3.5094530237270774e-06, + "loss": 0.4722, + "step": 767 + }, + { + "epoch": 2.4008350730688934, + "grad_norm": 1.1399786472320557, + "learning_rate": 3.5054905091428253e-06, + "loss": 0.4771, + "step": 768 + }, + { + "epoch": 2.403966597077244, + "grad_norm": 1.2586532831192017, + "learning_rate": 3.50152497875503e-06, + "loss": 0.4849, + "step": 769 + }, + { + "epoch": 2.407098121085595, + "grad_norm": 0.7706464529037476, + "learning_rate": 3.4975564444576487e-06, + "loss": 0.477, + "step": 770 + }, + { + "epoch": 2.4102296450939455, + "grad_norm": 0.7695909142494202, + "learning_rate": 3.4935849181536484e-06, + "loss": 0.4695, + "step": 771 + }, + { + "epoch": 2.4133611691022967, + "grad_norm": 0.7744433283805847, + "learning_rate": 3.489610411754969e-06, + "loss": 0.499, + "step": 772 + }, + { + "epoch": 2.416492693110647, + "grad_norm": 0.9265744686126709, + "learning_rate": 3.48563293718249e-06, + "loss": 0.481, + "step": 773 + }, + { + "epoch": 2.419624217118998, + "grad_norm": 1.0680506229400635, + "learning_rate": 3.481652506365992e-06, + "loss": 0.4898, + "step": 774 + }, + { + "epoch": 2.422755741127349, + "grad_norm": 0.721493661403656, + "learning_rate": 3.477669131244122e-06, + "loss": 0.4813, + "step": 775 + }, + { + "epoch": 2.4258872651356995, + "grad_norm": 0.7993559837341309, + "learning_rate": 3.4736828237643616e-06, + "loss": 0.5179, + "step": 776 + }, + { + "epoch": 2.4290187891440502, + "grad_norm": 0.8148090839385986, + "learning_rate": 3.4696935958829837e-06, + "loss": 0.4753, + "step": 777 + }, + { + "epoch": 2.432150313152401, + "grad_norm": 0.8006406426429749, + "learning_rate": 3.465701459565022e-06, + "loss": 0.501, + "step": 778 + }, + { + "epoch": 2.4352818371607516, + "grad_norm": 0.9307970404624939, + "learning_rate": 3.4617064267842327e-06, + "loss": 0.487, + "step": 779 + }, + { + "epoch": 2.4384133611691023, + "grad_norm": 0.7192814946174622, + "learning_rate": 3.45770850952306e-06, + "loss": 0.4769, + "step": 780 + }, + { + "epoch": 2.441544885177453, + "grad_norm": 0.7386271953582764, + "learning_rate": 3.4537077197726023e-06, + "loss": 0.4726, + "step": 781 + }, + { + "epoch": 2.4446764091858038, + "grad_norm": 0.8006314039230347, + "learning_rate": 3.449704069532567e-06, + "loss": 0.494, + "step": 782 + }, + { + "epoch": 2.4478079331941545, + "grad_norm": 0.7466752529144287, + "learning_rate": 3.4456975708112477e-06, + "loss": 0.4778, + "step": 783 + }, + { + "epoch": 2.450939457202505, + "grad_norm": 0.8348856568336487, + "learning_rate": 3.4416882356254777e-06, + "loss": 0.4766, + "step": 784 + }, + { + "epoch": 2.454070981210856, + "grad_norm": 0.754851758480072, + "learning_rate": 3.4376760760005994e-06, + "loss": 0.4673, + "step": 785 + }, + { + "epoch": 2.4572025052192066, + "grad_norm": 0.7854018807411194, + "learning_rate": 3.433661103970427e-06, + "loss": 0.4954, + "step": 786 + }, + { + "epoch": 2.4603340292275573, + "grad_norm": 0.7238256931304932, + "learning_rate": 3.4296433315772084e-06, + "loss": 0.496, + "step": 787 + }, + { + "epoch": 2.463465553235908, + "grad_norm": 0.7007659673690796, + "learning_rate": 3.4256227708715915e-06, + "loss": 0.4793, + "step": 788 + }, + { + "epoch": 2.4665970772442587, + "grad_norm": 0.7234371900558472, + "learning_rate": 3.421599433912588e-06, + "loss": 0.4935, + "step": 789 + }, + { + "epoch": 2.4697286012526094, + "grad_norm": 0.7537544965744019, + "learning_rate": 3.4175733327675355e-06, + "loss": 0.5194, + "step": 790 + }, + { + "epoch": 2.4728601252609606, + "grad_norm": 0.7608047127723694, + "learning_rate": 3.4135444795120633e-06, + "loss": 0.4793, + "step": 791 + }, + { + "epoch": 2.4759916492693113, + "grad_norm": 0.7847898006439209, + "learning_rate": 3.4095128862300542e-06, + "loss": 0.4877, + "step": 792 + }, + { + "epoch": 2.479123173277662, + "grad_norm": 0.8002011179924011, + "learning_rate": 3.405478565013609e-06, + "loss": 0.4927, + "step": 793 + }, + { + "epoch": 2.4822546972860127, + "grad_norm": 0.8200219869613647, + "learning_rate": 3.401441527963013e-06, + "loss": 0.4997, + "step": 794 + }, + { + "epoch": 2.4853862212943634, + "grad_norm": 0.7220162749290466, + "learning_rate": 3.3974017871866938e-06, + "loss": 0.4668, + "step": 795 + }, + { + "epoch": 2.488517745302714, + "grad_norm": 0.8022251129150391, + "learning_rate": 3.3933593548011912e-06, + "loss": 0.5179, + "step": 796 + }, + { + "epoch": 2.491649269311065, + "grad_norm": 0.7914465069770813, + "learning_rate": 3.389314242931115e-06, + "loss": 0.4943, + "step": 797 + }, + { + "epoch": 2.4947807933194155, + "grad_norm": 1.1399403810501099, + "learning_rate": 3.385266463709116e-06, + "loss": 0.4896, + "step": 798 + }, + { + "epoch": 2.497912317327766, + "grad_norm": 0.8098909854888916, + "learning_rate": 3.38121602927584e-06, + "loss": 0.4904, + "step": 799 + }, + { + "epoch": 2.501043841336117, + "grad_norm": 0.7434052228927612, + "learning_rate": 3.377162951779902e-06, + "loss": 0.4864, + "step": 800 + }, + { + "epoch": 2.5041753653444676, + "grad_norm": 0.7397809624671936, + "learning_rate": 3.3731072433778407e-06, + "loss": 0.486, + "step": 801 + }, + { + "epoch": 2.5073068893528183, + "grad_norm": 0.99027019739151, + "learning_rate": 3.3690489162340867e-06, + "loss": 0.5011, + "step": 802 + }, + { + "epoch": 2.510438413361169, + "grad_norm": 0.8443610668182373, + "learning_rate": 3.3649879825209246e-06, + "loss": 0.455, + "step": 803 + }, + { + "epoch": 2.5135699373695197, + "grad_norm": 0.755649983882904, + "learning_rate": 3.3609244544184604e-06, + "loss": 0.4563, + "step": 804 + }, + { + "epoch": 2.5167014613778704, + "grad_norm": 0.728018045425415, + "learning_rate": 3.3568583441145765e-06, + "loss": 0.471, + "step": 805 + }, + { + "epoch": 2.519832985386221, + "grad_norm": 0.7777130603790283, + "learning_rate": 3.352789663804904e-06, + "loss": 0.4667, + "step": 806 + }, + { + "epoch": 2.522964509394572, + "grad_norm": 0.7545619606971741, + "learning_rate": 3.3487184256927785e-06, + "loss": 0.4915, + "step": 807 + }, + { + "epoch": 2.526096033402923, + "grad_norm": 0.8374579548835754, + "learning_rate": 3.3446446419892127e-06, + "loss": 0.485, + "step": 808 + }, + { + "epoch": 2.5292275574112733, + "grad_norm": 0.7354666590690613, + "learning_rate": 3.340568324912849e-06, + "loss": 0.5254, + "step": 809 + }, + { + "epoch": 2.5323590814196244, + "grad_norm": 0.7581545114517212, + "learning_rate": 3.3364894866899324e-06, + "loss": 0.4483, + "step": 810 + }, + { + "epoch": 2.535490605427975, + "grad_norm": 0.8077559471130371, + "learning_rate": 3.3324081395542662e-06, + "loss": 0.5022, + "step": 811 + }, + { + "epoch": 2.538622129436326, + "grad_norm": 0.8827865719795227, + "learning_rate": 3.3283242957471806e-06, + "loss": 0.4909, + "step": 812 + }, + { + "epoch": 2.5417536534446765, + "grad_norm": 0.9139482378959656, + "learning_rate": 3.3242379675174953e-06, + "loss": 0.5205, + "step": 813 + }, + { + "epoch": 2.5448851774530272, + "grad_norm": 0.7616812586784363, + "learning_rate": 3.3201491671214797e-06, + "loss": 0.4744, + "step": 814 + }, + { + "epoch": 2.548016701461378, + "grad_norm": 0.987173318862915, + "learning_rate": 3.3160579068228183e-06, + "loss": 0.4876, + "step": 815 + }, + { + "epoch": 2.5511482254697286, + "grad_norm": 1.259137749671936, + "learning_rate": 3.311964198892574e-06, + "loss": 0.454, + "step": 816 + }, + { + "epoch": 2.5542797494780793, + "grad_norm": 0.7866336703300476, + "learning_rate": 3.3078680556091513e-06, + "loss": 0.5107, + "step": 817 + }, + { + "epoch": 2.55741127348643, + "grad_norm": 0.9311352372169495, + "learning_rate": 3.303769489258258e-06, + "loss": 0.4843, + "step": 818 + }, + { + "epoch": 2.5605427974947808, + "grad_norm": 0.8556346893310547, + "learning_rate": 3.299668512132872e-06, + "loss": 0.5017, + "step": 819 + }, + { + "epoch": 2.5636743215031315, + "grad_norm": 2.810598373413086, + "learning_rate": 3.2955651365331988e-06, + "loss": 0.5223, + "step": 820 + }, + { + "epoch": 2.566805845511482, + "grad_norm": 1.0120766162872314, + "learning_rate": 3.29145937476664e-06, + "loss": 0.4959, + "step": 821 + }, + { + "epoch": 2.569937369519833, + "grad_norm": 0.751412034034729, + "learning_rate": 3.287351239147752e-06, + "loss": 0.4941, + "step": 822 + }, + { + "epoch": 2.5730688935281836, + "grad_norm": 1.9308148622512817, + "learning_rate": 3.2832407419982136e-06, + "loss": 0.4965, + "step": 823 + }, + { + "epoch": 2.5762004175365343, + "grad_norm": 0.9215649962425232, + "learning_rate": 3.279127895646786e-06, + "loss": 0.5071, + "step": 824 + }, + { + "epoch": 2.5793319415448854, + "grad_norm": 0.7599574327468872, + "learning_rate": 3.2750127124292754e-06, + "loss": 0.5191, + "step": 825 + }, + { + "epoch": 2.5824634655532357, + "grad_norm": 0.8234940767288208, + "learning_rate": 3.270895204688496e-06, + "loss": 0.4947, + "step": 826 + }, + { + "epoch": 2.585594989561587, + "grad_norm": 0.8401572704315186, + "learning_rate": 3.266775384774238e-06, + "loss": 0.4547, + "step": 827 + }, + { + "epoch": 2.588726513569937, + "grad_norm": 0.8927991986274719, + "learning_rate": 3.262653265043223e-06, + "loss": 0.4296, + "step": 828 + }, + { + "epoch": 2.5918580375782883, + "grad_norm": 0.8009241223335266, + "learning_rate": 3.2585288578590716e-06, + "loss": 0.4578, + "step": 829 + }, + { + "epoch": 2.594989561586639, + "grad_norm": 0.7982021570205688, + "learning_rate": 3.2544021755922663e-06, + "loss": 0.4961, + "step": 830 + }, + { + "epoch": 2.5981210855949897, + "grad_norm": 0.7096095681190491, + "learning_rate": 3.2502732306201112e-06, + "loss": 0.4975, + "step": 831 + }, + { + "epoch": 2.6012526096033404, + "grad_norm": 1.1092045307159424, + "learning_rate": 3.246142035326699e-06, + "loss": 0.4705, + "step": 832 + }, + { + "epoch": 2.604384133611691, + "grad_norm": 0.785799503326416, + "learning_rate": 3.24200860210287e-06, + "loss": 0.479, + "step": 833 + }, + { + "epoch": 2.607515657620042, + "grad_norm": 0.7315773367881775, + "learning_rate": 3.2378729433461804e-06, + "loss": 0.5036, + "step": 834 + }, + { + "epoch": 2.6106471816283925, + "grad_norm": 0.7840189337730408, + "learning_rate": 3.233735071460856e-06, + "loss": 0.4967, + "step": 835 + }, + { + "epoch": 2.613778705636743, + "grad_norm": 0.7186565399169922, + "learning_rate": 3.2295949988577655e-06, + "loss": 0.4889, + "step": 836 + }, + { + "epoch": 2.616910229645094, + "grad_norm": 0.766054093837738, + "learning_rate": 3.2254527379543747e-06, + "loss": 0.539, + "step": 837 + }, + { + "epoch": 2.6200417536534446, + "grad_norm": 0.7705381512641907, + "learning_rate": 3.2213083011747165e-06, + "loss": 0.4968, + "step": 838 + }, + { + "epoch": 2.6231732776617953, + "grad_norm": 1.3530604839324951, + "learning_rate": 3.217161700949346e-06, + "loss": 0.52, + "step": 839 + }, + { + "epoch": 2.626304801670146, + "grad_norm": 0.737389862537384, + "learning_rate": 3.2130129497153107e-06, + "loss": 0.4823, + "step": 840 + }, + { + "epoch": 2.6294363256784967, + "grad_norm": 0.9121193885803223, + "learning_rate": 3.2088620599161064e-06, + "loss": 0.4592, + "step": 841 + }, + { + "epoch": 2.632567849686848, + "grad_norm": 0.8869616389274597, + "learning_rate": 3.2047090440016464e-06, + "loss": 0.5001, + "step": 842 + }, + { + "epoch": 2.635699373695198, + "grad_norm": 0.8447219133377075, + "learning_rate": 3.200553914428219e-06, + "loss": 0.4969, + "step": 843 + }, + { + "epoch": 2.6388308977035493, + "grad_norm": 0.8877657055854797, + "learning_rate": 3.1963966836584524e-06, + "loss": 0.4718, + "step": 844 + }, + { + "epoch": 2.6419624217118995, + "grad_norm": 1.045272946357727, + "learning_rate": 3.192237364161277e-06, + "loss": 0.4864, + "step": 845 + }, + { + "epoch": 2.6450939457202507, + "grad_norm": 0.8485913276672363, + "learning_rate": 3.1880759684118876e-06, + "loss": 0.4688, + "step": 846 + }, + { + "epoch": 2.6482254697286014, + "grad_norm": 0.7328930497169495, + "learning_rate": 3.183912508891709e-06, + "loss": 0.4728, + "step": 847 + }, + { + "epoch": 2.651356993736952, + "grad_norm": 0.7377315759658813, + "learning_rate": 3.179746998088351e-06, + "loss": 0.4672, + "step": 848 + }, + { + "epoch": 2.654488517745303, + "grad_norm": 0.8017002940177917, + "learning_rate": 3.1755794484955817e-06, + "loss": 0.4884, + "step": 849 + }, + { + "epoch": 2.6576200417536535, + "grad_norm": 1.045470952987671, + "learning_rate": 3.171409872613278e-06, + "loss": 0.4789, + "step": 850 + }, + { + "epoch": 2.6607515657620042, + "grad_norm": 0.8823987245559692, + "learning_rate": 3.1672382829473997e-06, + "loss": 0.5117, + "step": 851 + }, + { + "epoch": 2.663883089770355, + "grad_norm": 0.7395204901695251, + "learning_rate": 3.163064692009944e-06, + "loss": 0.5476, + "step": 852 + }, + { + "epoch": 2.6670146137787056, + "grad_norm": 0.7778941988945007, + "learning_rate": 3.1588891123189103e-06, + "loss": 0.5092, + "step": 853 + }, + { + "epoch": 2.6701461377870563, + "grad_norm": 0.8072531819343567, + "learning_rate": 3.1547115563982643e-06, + "loss": 0.4961, + "step": 854 + }, + { + "epoch": 2.673277661795407, + "grad_norm": 0.9018139243125916, + "learning_rate": 3.1505320367778993e-06, + "loss": 0.4624, + "step": 855 + }, + { + "epoch": 2.6764091858037578, + "grad_norm": 0.8554450869560242, + "learning_rate": 3.1463505659935957e-06, + "loss": 0.4971, + "step": 856 + }, + { + "epoch": 2.6795407098121085, + "grad_norm": 0.7727259397506714, + "learning_rate": 3.14216715658699e-06, + "loss": 0.4544, + "step": 857 + }, + { + "epoch": 2.682672233820459, + "grad_norm": 0.9253409504890442, + "learning_rate": 3.137981821105529e-06, + "loss": 0.4893, + "step": 858 + }, + { + "epoch": 2.68580375782881, + "grad_norm": 0.8809456825256348, + "learning_rate": 3.1337945721024403e-06, + "loss": 0.5242, + "step": 859 + }, + { + "epoch": 2.6889352818371606, + "grad_norm": 0.981755256652832, + "learning_rate": 3.129605422136689e-06, + "loss": 0.4686, + "step": 860 + }, + { + "epoch": 2.6920668058455117, + "grad_norm": 1.1278467178344727, + "learning_rate": 3.1254143837729412e-06, + "loss": 0.4813, + "step": 861 + }, + { + "epoch": 2.695198329853862, + "grad_norm": 0.8529123663902283, + "learning_rate": 3.1212214695815285e-06, + "loss": 0.4723, + "step": 862 + }, + { + "epoch": 2.698329853862213, + "grad_norm": 0.7764189839363098, + "learning_rate": 3.1170266921384075e-06, + "loss": 0.4777, + "step": 863 + }, + { + "epoch": 2.701461377870564, + "grad_norm": 0.7364740967750549, + "learning_rate": 3.112830064025124e-06, + "loss": 0.4975, + "step": 864 + }, + { + "epoch": 2.7045929018789145, + "grad_norm": 0.7594549059867859, + "learning_rate": 3.108631597828774e-06, + "loss": 0.5083, + "step": 865 + }, + { + "epoch": 2.7077244258872653, + "grad_norm": 0.7337073683738708, + "learning_rate": 3.104431306141968e-06, + "loss": 0.4778, + "step": 866 + }, + { + "epoch": 2.710855949895616, + "grad_norm": 0.7709932327270508, + "learning_rate": 3.1002292015627894e-06, + "loss": 0.4754, + "step": 867 + }, + { + "epoch": 2.7139874739039667, + "grad_norm": 0.8001313209533691, + "learning_rate": 3.0960252966947605e-06, + "loss": 0.4489, + "step": 868 + }, + { + "epoch": 2.7171189979123174, + "grad_norm": 0.8280592560768127, + "learning_rate": 3.091819604146804e-06, + "loss": 0.4606, + "step": 869 + }, + { + "epoch": 2.720250521920668, + "grad_norm": 0.7463534474372864, + "learning_rate": 3.0876121365332024e-06, + "loss": 0.5168, + "step": 870 + }, + { + "epoch": 2.723382045929019, + "grad_norm": 0.9011222124099731, + "learning_rate": 3.0834029064735636e-06, + "loss": 0.5163, + "step": 871 + }, + { + "epoch": 2.7265135699373695, + "grad_norm": 0.7811456322669983, + "learning_rate": 3.0791919265927827e-06, + "loss": 0.5004, + "step": 872 + }, + { + "epoch": 2.72964509394572, + "grad_norm": 0.9251837134361267, + "learning_rate": 3.0749792095210003e-06, + "loss": 0.5081, + "step": 873 + }, + { + "epoch": 2.732776617954071, + "grad_norm": 0.8347085118293762, + "learning_rate": 3.0707647678935695e-06, + "loss": 0.4793, + "step": 874 + }, + { + "epoch": 2.7359081419624216, + "grad_norm": 0.9766442179679871, + "learning_rate": 3.0665486143510153e-06, + "loss": 0.493, + "step": 875 + }, + { + "epoch": 2.7390396659707723, + "grad_norm": 0.7692548036575317, + "learning_rate": 3.0623307615389975e-06, + "loss": 0.4874, + "step": 876 + }, + { + "epoch": 2.742171189979123, + "grad_norm": 0.7714599370956421, + "learning_rate": 3.0581112221082727e-06, + "loss": 0.4929, + "step": 877 + }, + { + "epoch": 2.745302713987474, + "grad_norm": 0.7797786593437195, + "learning_rate": 3.053890008714655e-06, + "loss": 0.4359, + "step": 878 + }, + { + "epoch": 2.7484342379958244, + "grad_norm": 5.118397235870361, + "learning_rate": 3.049667134018981e-06, + "loss": 0.4634, + "step": 879 + }, + { + "epoch": 2.7515657620041756, + "grad_norm": 0.7684539556503296, + "learning_rate": 3.04544261068707e-06, + "loss": 0.4688, + "step": 880 + }, + { + "epoch": 2.754697286012526, + "grad_norm": 0.8678610920906067, + "learning_rate": 3.0412164513896846e-06, + "loss": 0.5213, + "step": 881 + }, + { + "epoch": 2.757828810020877, + "grad_norm": 0.80293869972229, + "learning_rate": 3.0369886688024954e-06, + "loss": 0.4392, + "step": 882 + }, + { + "epoch": 2.7609603340292277, + "grad_norm": 0.7438644766807556, + "learning_rate": 3.0327592756060412e-06, + "loss": 0.528, + "step": 883 + }, + { + "epoch": 2.7640918580375784, + "grad_norm": 0.7701645493507385, + "learning_rate": 3.0285282844856917e-06, + "loss": 0.504, + "step": 884 + }, + { + "epoch": 2.767223382045929, + "grad_norm": 0.7113856673240662, + "learning_rate": 3.024295708131611e-06, + "loss": 0.4819, + "step": 885 + }, + { + "epoch": 2.77035490605428, + "grad_norm": 1.2697532176971436, + "learning_rate": 3.020061559238714e-06, + "loss": 0.5009, + "step": 886 + }, + { + "epoch": 2.7734864300626305, + "grad_norm": 1.0299439430236816, + "learning_rate": 3.015825850506636e-06, + "loss": 0.4707, + "step": 887 + }, + { + "epoch": 2.776617954070981, + "grad_norm": 0.9703660607337952, + "learning_rate": 3.011588594639688e-06, + "loss": 0.4102, + "step": 888 + }, + { + "epoch": 2.779749478079332, + "grad_norm": 0.7357314825057983, + "learning_rate": 3.0073498043468247e-06, + "loss": 0.4649, + "step": 889 + }, + { + "epoch": 2.7828810020876826, + "grad_norm": 0.7815471291542053, + "learning_rate": 3.0031094923415993e-06, + "loss": 0.469, + "step": 890 + }, + { + "epoch": 2.7860125260960333, + "grad_norm": 0.7856019139289856, + "learning_rate": 2.9988676713421318e-06, + "loss": 0.4241, + "step": 891 + }, + { + "epoch": 2.789144050104384, + "grad_norm": 0.7668167352676392, + "learning_rate": 2.994624354071066e-06, + "loss": 0.5309, + "step": 892 + }, + { + "epoch": 2.7922755741127347, + "grad_norm": 0.7485945820808411, + "learning_rate": 2.990379553255535e-06, + "loss": 0.5173, + "step": 893 + }, + { + "epoch": 2.7954070981210855, + "grad_norm": 0.8065824508666992, + "learning_rate": 2.986133281627123e-06, + "loss": 0.4995, + "step": 894 + }, + { + "epoch": 2.798538622129436, + "grad_norm": 0.7156995534896851, + "learning_rate": 2.9818855519218217e-06, + "loss": 0.4642, + "step": 895 + }, + { + "epoch": 2.801670146137787, + "grad_norm": 0.9115403890609741, + "learning_rate": 2.97763637688e-06, + "loss": 0.4799, + "step": 896 + }, + { + "epoch": 2.804801670146138, + "grad_norm": 0.7466689944267273, + "learning_rate": 2.9733857692463584e-06, + "loss": 0.4942, + "step": 897 + }, + { + "epoch": 2.8079331941544883, + "grad_norm": 0.7484914064407349, + "learning_rate": 2.9691337417698974e-06, + "loss": 0.4618, + "step": 898 + }, + { + "epoch": 2.8110647181628394, + "grad_norm": 0.816704511642456, + "learning_rate": 2.9648803072038736e-06, + "loss": 0.4748, + "step": 899 + }, + { + "epoch": 2.81419624217119, + "grad_norm": 0.7627584934234619, + "learning_rate": 2.9606254783057666e-06, + "loss": 0.4667, + "step": 900 + }, + { + "epoch": 2.817327766179541, + "grad_norm": 0.7341011166572571, + "learning_rate": 2.9563692678372342e-06, + "loss": 0.4802, + "step": 901 + }, + { + "epoch": 2.8204592901878915, + "grad_norm": 1.2541382312774658, + "learning_rate": 2.952111688564082e-06, + "loss": 0.5231, + "step": 902 + }, + { + "epoch": 2.8235908141962422, + "grad_norm": 0.7172819375991821, + "learning_rate": 2.9478527532562184e-06, + "loss": 0.4488, + "step": 903 + }, + { + "epoch": 2.826722338204593, + "grad_norm": 0.774529218673706, + "learning_rate": 2.943592474687621e-06, + "loss": 0.4964, + "step": 904 + }, + { + "epoch": 2.8298538622129437, + "grad_norm": 0.7315672636032104, + "learning_rate": 2.939330865636294e-06, + "loss": 0.4817, + "step": 905 + }, + { + "epoch": 2.8329853862212944, + "grad_norm": 0.7698234915733337, + "learning_rate": 2.9350679388842347e-06, + "loss": 0.5075, + "step": 906 + }, + { + "epoch": 2.836116910229645, + "grad_norm": 0.7717766761779785, + "learning_rate": 2.93080370721739e-06, + "loss": 0.4789, + "step": 907 + }, + { + "epoch": 2.8392484342379958, + "grad_norm": 0.7383570075035095, + "learning_rate": 2.926538183425622e-06, + "loss": 0.4992, + "step": 908 + }, + { + "epoch": 2.8423799582463465, + "grad_norm": 0.7858864068984985, + "learning_rate": 2.92227138030267e-06, + "loss": 0.4993, + "step": 909 + }, + { + "epoch": 2.845511482254697, + "grad_norm": 0.8220369219779968, + "learning_rate": 2.9180033106461076e-06, + "loss": 0.4929, + "step": 910 + }, + { + "epoch": 2.848643006263048, + "grad_norm": 0.7507152557373047, + "learning_rate": 2.9137339872573086e-06, + "loss": 0.4394, + "step": 911 + }, + { + "epoch": 2.8517745302713986, + "grad_norm": 0.7935269474983215, + "learning_rate": 2.9094634229414063e-06, + "loss": 0.4656, + "step": 912 + }, + { + "epoch": 2.8549060542797493, + "grad_norm": 0.9187721610069275, + "learning_rate": 2.9051916305072576e-06, + "loss": 0.4918, + "step": 913 + }, + { + "epoch": 2.8580375782881005, + "grad_norm": 0.8699706792831421, + "learning_rate": 2.9009186227674e-06, + "loss": 0.5106, + "step": 914 + }, + { + "epoch": 2.8611691022964507, + "grad_norm": 0.7175673246383667, + "learning_rate": 2.896644412538021e-06, + "loss": 0.5105, + "step": 915 + }, + { + "epoch": 2.864300626304802, + "grad_norm": 0.8563990592956543, + "learning_rate": 2.892369012638909e-06, + "loss": 0.4993, + "step": 916 + }, + { + "epoch": 2.867432150313152, + "grad_norm": 0.7891882658004761, + "learning_rate": 2.8880924358934246e-06, + "loss": 0.4983, + "step": 917 + }, + { + "epoch": 2.8705636743215033, + "grad_norm": 0.9247110486030579, + "learning_rate": 2.8838146951284575e-06, + "loss": 0.4789, + "step": 918 + }, + { + "epoch": 2.873695198329854, + "grad_norm": 0.7523055672645569, + "learning_rate": 2.879535803174387e-06, + "loss": 0.4982, + "step": 919 + }, + { + "epoch": 2.8768267223382047, + "grad_norm": 0.8096909523010254, + "learning_rate": 2.8752557728650467e-06, + "loss": 0.4958, + "step": 920 + }, + { + "epoch": 2.8799582463465554, + "grad_norm": 2.3476874828338623, + "learning_rate": 2.870974617037684e-06, + "loss": 0.491, + "step": 921 + }, + { + "epoch": 2.883089770354906, + "grad_norm": 0.8388578295707703, + "learning_rate": 2.8666923485329224e-06, + "loss": 0.5275, + "step": 922 + }, + { + "epoch": 2.886221294363257, + "grad_norm": 0.8162729144096375, + "learning_rate": 2.8624089801947234e-06, + "loss": 0.4776, + "step": 923 + }, + { + "epoch": 2.8893528183716075, + "grad_norm": 0.7306103110313416, + "learning_rate": 2.858124524870345e-06, + "loss": 0.4814, + "step": 924 + }, + { + "epoch": 2.892484342379958, + "grad_norm": 0.8736817836761475, + "learning_rate": 2.853838995410307e-06, + "loss": 0.5097, + "step": 925 + }, + { + "epoch": 2.895615866388309, + "grad_norm": 0.7771823406219482, + "learning_rate": 2.8495524046683525e-06, + "loss": 0.4806, + "step": 926 + }, + { + "epoch": 2.8987473903966596, + "grad_norm": 0.9421334862709045, + "learning_rate": 2.845264765501404e-06, + "loss": 0.5055, + "step": 927 + }, + { + "epoch": 2.9018789144050103, + "grad_norm": 0.8403921127319336, + "learning_rate": 2.8409760907695314e-06, + "loss": 0.4775, + "step": 928 + }, + { + "epoch": 2.905010438413361, + "grad_norm": 0.8095362186431885, + "learning_rate": 2.836686393335909e-06, + "loss": 0.4532, + "step": 929 + }, + { + "epoch": 2.9081419624217117, + "grad_norm": 0.7340645790100098, + "learning_rate": 2.8323956860667813e-06, + "loss": 0.4835, + "step": 930 + }, + { + "epoch": 2.911273486430063, + "grad_norm": 0.6970911026000977, + "learning_rate": 2.828103981831417e-06, + "loss": 0.4999, + "step": 931 + }, + { + "epoch": 2.914405010438413, + "grad_norm": 0.8136418461799622, + "learning_rate": 2.8238112935020794e-06, + "loss": 0.5038, + "step": 932 + }, + { + "epoch": 2.9175365344467643, + "grad_norm": 0.9045608043670654, + "learning_rate": 2.8195176339539816e-06, + "loss": 0.486, + "step": 933 + }, + { + "epoch": 2.9206680584551146, + "grad_norm": 1.14940345287323, + "learning_rate": 2.815223016065249e-06, + "loss": 0.5079, + "step": 934 + }, + { + "epoch": 2.9237995824634657, + "grad_norm": 0.7411190867424011, + "learning_rate": 2.8109274527168826e-06, + "loss": 0.4564, + "step": 935 + }, + { + "epoch": 2.9269311064718164, + "grad_norm": 0.8903455138206482, + "learning_rate": 2.806630956792719e-06, + "loss": 0.451, + "step": 936 + }, + { + "epoch": 2.930062630480167, + "grad_norm": 0.7865445017814636, + "learning_rate": 2.8023335411793904e-06, + "loss": 0.4658, + "step": 937 + }, + { + "epoch": 2.933194154488518, + "grad_norm": 0.8185790777206421, + "learning_rate": 2.798035218766292e-06, + "loss": 0.4776, + "step": 938 + }, + { + "epoch": 2.9363256784968685, + "grad_norm": 0.7516276836395264, + "learning_rate": 2.793736002445531e-06, + "loss": 0.4447, + "step": 939 + }, + { + "epoch": 2.9394572025052192, + "grad_norm": 0.738080620765686, + "learning_rate": 2.789435905111903e-06, + "loss": 0.4832, + "step": 940 + }, + { + "epoch": 2.94258872651357, + "grad_norm": 0.7971507906913757, + "learning_rate": 2.785134939662843e-06, + "loss": 0.4835, + "step": 941 + }, + { + "epoch": 2.9457202505219207, + "grad_norm": 0.7529093623161316, + "learning_rate": 2.78083311899839e-06, + "loss": 0.4759, + "step": 942 + }, + { + "epoch": 2.9488517745302714, + "grad_norm": 0.8222358226776123, + "learning_rate": 2.7765304560211482e-06, + "loss": 0.4365, + "step": 943 + }, + { + "epoch": 2.951983298538622, + "grad_norm": 0.729945182800293, + "learning_rate": 2.7722269636362462e-06, + "loss": 0.5026, + "step": 944 + }, + { + "epoch": 2.9551148225469728, + "grad_norm": 0.7287900447845459, + "learning_rate": 2.767922654751306e-06, + "loss": 0.4916, + "step": 945 + }, + { + "epoch": 2.9582463465553235, + "grad_norm": 0.869637131690979, + "learning_rate": 2.763617542276391e-06, + "loss": 0.5018, + "step": 946 + }, + { + "epoch": 2.961377870563674, + "grad_norm": 1.004909634590149, + "learning_rate": 2.7593116391239806e-06, + "loss": 0.5152, + "step": 947 + }, + { + "epoch": 2.964509394572025, + "grad_norm": 0.8263046145439148, + "learning_rate": 2.7550049582089235e-06, + "loss": 0.5249, + "step": 948 + }, + { + "epoch": 2.9676409185803756, + "grad_norm": 0.7963895797729492, + "learning_rate": 2.750697512448401e-06, + "loss": 0.5084, + "step": 949 + }, + { + "epoch": 2.9707724425887267, + "grad_norm": 0.7211249470710754, + "learning_rate": 2.7463893147618893e-06, + "loss": 0.4691, + "step": 950 + }, + { + "epoch": 2.973903966597077, + "grad_norm": 0.8010216951370239, + "learning_rate": 2.742080378071118e-06, + "loss": 0.5026, + "step": 951 + }, + { + "epoch": 2.977035490605428, + "grad_norm": 0.780078649520874, + "learning_rate": 2.7377707153000356e-06, + "loss": 0.4758, + "step": 952 + }, + { + "epoch": 2.980167014613779, + "grad_norm": 0.7728193998336792, + "learning_rate": 2.7334603393747684e-06, + "loss": 0.488, + "step": 953 + }, + { + "epoch": 2.9832985386221296, + "grad_norm": 0.836329996585846, + "learning_rate": 2.7291492632235777e-06, + "loss": 0.456, + "step": 954 + }, + { + "epoch": 2.9864300626304803, + "grad_norm": 0.7241990566253662, + "learning_rate": 2.724837499776831e-06, + "loss": 0.4953, + "step": 955 + }, + { + "epoch": 2.989561586638831, + "grad_norm": 0.7595076560974121, + "learning_rate": 2.7205250619669527e-06, + "loss": 0.446, + "step": 956 + }, + { + "epoch": 2.9926931106471817, + "grad_norm": 0.8177686333656311, + "learning_rate": 2.716211962728392e-06, + "loss": 0.5057, + "step": 957 + } + ], + "logging_steps": 1, + "max_steps": 1914, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 319, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 2.9005238427891794e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-957/training_args.bin b/checkpoint-957/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8067ee9c1c0bc752bdfd00cfcaf1a6e717d2356b --- /dev/null +++ b/checkpoint-957/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c650156a192cae0a2070c4323ee8a93e9b52fb76041d59ae0633b98389585727 +size 7928 diff --git a/checkpoint-957/zero_to_fp32.py b/checkpoint-957/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-957/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/config.json b/config.json new file mode 100644 index 0000000000000000000000000000000000000000..b7324bbcd3035a34c2ac96f0e2a46dd94a5db25c --- /dev/null +++ b/config.json @@ -0,0 +1,1497 @@ +{ + "_attn_implementation_autoset": true, + "_name_or_path": "nvidia/Llama-3_3-Nemotron-Super-49B-v1", + "architectures": [ + "DeciLMForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "auto_map": { + "AutoConfig": "nvidia/Llama-3_3-Nemotron-Super-49B-v1--configuration_decilm.DeciLMConfig", + "AutoModelForCausalLM": "nvidia/Llama-3_3-Nemotron-Super-49B-v1--modeling_decilm.DeciLMForCausalLM" + }, + "block_configs": [ + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 2.625, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 2.625, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 2.625, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 3.28125, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.3125, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 2.625, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 2.625, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.3125, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.3125, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 2.625, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.3125, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.3125, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.3125, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.3125, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.0, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.0, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.3125, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.0, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.0, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.0, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.3125, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.3125, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 0.5, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 0.5, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.0, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.0, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 0.5, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 0.5, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 1.0, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 0.5, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": null, + "no_op": true, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 0.5, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + }, + { + "attention": { + "n_heads_in_group": 8, + "no_op": false, + "num_sink_tokens": null, + "replace_with_linear": false, + "sparsify": null, + "unshifted_sink": false, + "use_prefill_window_in_sink_attention": false, + "window_length": null + }, + "ffn": { + "ffn_mult": 5.25, + "no_op": false, + "replace_with_linear": false, + "sparsify": null + } + } + ], + "bos_token_id": 128000, + "eos_token_id": 128009, + "hidden_act": "silu", + "hidden_size": 8192, + "initializer_range": 0.02, + "intermediate_size": null, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "nemotron-nas", + "num_attention_heads": 64, + "num_hidden_layers": 80, + "num_key_value_heads": null, + "pretraining_tp": 1, + "quantization_config": { + "_load_in_4bit": true, + "_load_in_8bit": false, + "bnb_4bit_compute_dtype": "bfloat16", + "bnb_4bit_quant_storage": "bfloat16", + "bnb_4bit_quant_type": "nf4", + "bnb_4bit_use_double_quant": true, + "llm_int8_enable_fp32_cpu_offload": false, + "llm_int8_has_fp16_weight": false, + "llm_int8_skip_modules": null, + "llm_int8_threshold": 6.0, + "load_in_4bit": true, + "load_in_8bit": false, + "quant_method": "bitsandbytes" + }, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.49.0", + "use_cache": false, + "vocab_size": 128256 +} diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..edd01b980c1db496ea102a51c972ee8f5d1a2c74 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "chat_template": "{{- bos_token }}{%- if messages[0]['role'] == 'system' %}{%- set system_message = messages[0]['content']|trim %}{%- set messages = messages[1:] %}{%- else %}{%- set system_message = \"\" %}{%- endif %}{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}{{- system_message }}{{- \"<|eot_id|>\" }}{%- for message in messages %}{%- if message['role'] == 'assistant' and '' in message['content'] %}{%- set content = message['content'].split('')[-1].lstrip() %}{%- else %}{%- set content = message['content'] %}{%- endif %}{{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n' + content | trim + '<|eot_id|>' }}{%- endfor %}{%- if add_generation_prompt %}{{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}{%- endif %}", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +}