diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..f53664a60e94522124557690f3227b8d5ca5eb5a 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-1914/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-2552/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-3190/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-3828/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-638/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..7d1fda29137ff9bc7e661b3ab4be401685f35668 --- /dev/null +++ b/README.md @@ -0,0 +1,143 @@ +--- +library_name: peft +license: other +base_model: nvidia/Llama-3.1-Nemotron-Nano-8B-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-3.1-Nemotron-Nano-8B-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_llama31nvidia +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-3.1-Nemotron-Nano-8B-v1](https://huggingface.co/nvidia/Llama-3.1-Nemotron-Nano-8B-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 +- gradient_accumulation_steps: 3 +- total_train_batch_size: 12 +- 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 0.14.0 +- Transformers 4.50.0 +- 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..3db75f3ecbeb5f4990760e56c7957796d83680c5 --- /dev/null +++ b/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1", + "bias": "none", + "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": [ + "q_proj", + "down_proj", + "o_proj", + "v_proj", + "k_proj", + "gate_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "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..b7828c64e3e693a901fefa43b86429b0bf262f8d --- /dev/null +++ b/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c045b55e37980f1987824870ab9c9912e9bbf700733b076ba36f6b536e1393f +size 3443586272 diff --git a/checkpoint-1276/README.md b/checkpoint-1276/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f07164552af92bf3072e4e4326b5e572dea01ae8 --- /dev/null +++ b/checkpoint-1276/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-Nano-8B-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 0.14.0 \ 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..3db75f3ecbeb5f4990760e56c7957796d83680c5 --- /dev/null +++ b/checkpoint-1276/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1", + "bias": "none", + "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": [ + "q_proj", + "down_proj", + "o_proj", + "v_proj", + "k_proj", + "gate_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "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..e0ea2a48f0adbf38de1de5640d3521ae3642662e --- /dev/null +++ b/checkpoint-1276/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a990d40e1e96b695c539d344baac17ee14404323bf7f516131c72316f327144e +size 3443586272 diff --git a/checkpoint-1276/global_step1276/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-1276/global_step1276/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..ba2c1f9b44ac110be7119b67159b8bb687e0e9ed --- /dev/null +++ b/checkpoint-1276/global_step1276/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:16d5c3b2400d2869c0dc66292d3bacbe0f47436cf3b7321a1c0b3e47bf926382 +size 20661195036 diff --git a/checkpoint-1276/global_step1276/mp_rank_00_model_states.pt b/checkpoint-1276/global_step1276/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..2451269d9241a95c1092335c6d59ba2654308baa --- /dev/null +++ b/checkpoint-1276/global_step1276/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:12a12d6ed647c7a01e80b2c92241b28c31bba8d316a3fb0cea56d6d3ac0a6ca8 +size 3555326777 diff --git a/checkpoint-1276/latest b/checkpoint-1276/latest new file mode 100644 index 0000000000000000000000000000000000000000..bcecca951331714bd5998e9c34bedf350f44c403 --- /dev/null +++ b/checkpoint-1276/latest @@ -0,0 +1 @@ +global_step1276 \ No newline at end of file diff --git a/checkpoint-1276/rng_state.pth b/checkpoint-1276/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..2a33202b70d96a9a6889bdae7020bda9cbf709aa --- /dev/null +++ b/checkpoint-1276/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3185e5cfa8dad71d397b3dc46365a90ca0fc0588abd008e1ffbd9677fd8827e1 +size 14244 diff --git a/checkpoint-1276/scheduler.pt b/checkpoint-1276/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..16fe629bd1fcd3666a6e0916fada7d4fe923ca50 --- /dev/null +++ b/checkpoint-1276/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:041a823d8ac60d4220ba6794112b14838f396a6963c37e259a33d857bf43abb3 +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..b480d149dc274c0a23db079bfd38032d90082e49 --- /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": "{%- if messages[0]['role'] == 'system' -%}{%- set system_message = messages[0]['content'] | trim -%}{%- set messages = messages[1:] -%}{%- else -%}{%- set system_message = '' -%}{%- endif -%}{%- if tools is not none -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message -}} {{- '\n\n' if system_message else '' -}} {{- '[' -}} {% for t in tools %}{{- (t.function if t.function is defined else t) | tojson() -}}{{- ', ' if not loop.last else '' -}}{%- endfor -%} {{- ']' -}} {{- '<|eot_id|>' -}}{%- else -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message + '<|eot_id|>' -}}{%- endif -%}{%- for message in messages -%}{%- if (message['role'] in ['user', 'tool']) != (loop.index0 % 2 == 0) -%}{{- raise_exception('Conversation roles must alternate between user/tool and assistant') -}}{%- elif message['role'] == 'user' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- elif message['role'] == 'tool' -%}{%- set tool_response = '[' + message['content'] | trim + ']' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + tool_response + '<|eot_id|>' -}}{%- elif message['role'] == 'assistant' and message.get('tool_calls') is not none -%}{%- set tool_calls = message['tool_calls'] -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + '[' -}}{%- for tool_call in tool_calls -%}{{ '{' + '\"name\": \"' + tool_call.function.name + '\", \"arguments\": ' + tool_call.function.arguments | tojson + '}' }}{%- if not loop.last -%}{{ ', ' }}{%- else -%}{{ ']' + '<|eot_id|>' }}{%- endif -%}{%- endfor -%}{%- elif message['role'] == 'assistant' -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- endif -%}{%- 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..a0c8acc996e8ee32716583d75890d5aea43c3198 --- /dev/null +++ b/checkpoint-1276/trainer_state.json @@ -0,0 +1,8966 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.998956158663883, + "eval_steps": 500, + "global_step": 1276, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0015657620041753654, + "grad_norm": 658.1201782226562, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.3039, + "step": 1 + }, + { + "epoch": 0.003131524008350731, + "grad_norm": 749.5796508789062, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.4164, + "step": 2 + }, + { + "epoch": 0.004697286012526096, + "grad_norm": 394.2350769042969, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.3367, + "step": 3 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 384.08319091796875, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.3192, + "step": 4 + }, + { + "epoch": 0.007828810020876827, + "grad_norm": 376.9790954589844, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.2756, + "step": 5 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 175.25262451171875, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.3992, + "step": 6 + }, + { + "epoch": 0.010960334029227558, + "grad_norm": 100.1114730834961, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.2348, + "step": 7 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 131.44432067871094, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.2685, + "step": 8 + }, + { + "epoch": 0.014091858037578288, + "grad_norm": 67.03341674804688, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.3272, + "step": 9 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 32.0869140625, + "learning_rate": 5.000000000000001e-07, + "loss": 1.3458, + "step": 10 + }, + { + "epoch": 0.01722338204592902, + "grad_norm": 17.701913833618164, + "learning_rate": 5.5e-07, + "loss": 1.3608, + "step": 11 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 9.164765357971191, + "learning_rate": 6.000000000000001e-07, + "loss": 1.2961, + "step": 12 + }, + { + "epoch": 0.02035490605427975, + "grad_norm": 4.315349578857422, + "learning_rate": 6.5e-07, + "loss": 1.2069, + "step": 13 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 6.652385234832764, + "learning_rate": 7.000000000000001e-07, + "loss": 1.2588, + "step": 14 + }, + { + "epoch": 0.02348643006263048, + "grad_norm": 3.3322548866271973, + "learning_rate": 7.5e-07, + "loss": 1.2505, + "step": 15 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 9.599331855773926, + "learning_rate": 8.000000000000001e-07, + "loss": 1.1343, + "step": 16 + }, + { + "epoch": 0.02661795407098121, + "grad_norm": 10.33918285369873, + "learning_rate": 8.500000000000001e-07, + "loss": 1.3334, + "step": 17 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 9.453723907470703, + "learning_rate": 9.000000000000001e-07, + "loss": 1.3099, + "step": 18 + }, + { + "epoch": 0.029749478079331943, + "grad_norm": 7.075769901275635, + "learning_rate": 9.500000000000001e-07, + "loss": 1.1121, + "step": 19 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 7.770864009857178, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.1836, + "step": 20 + }, + { + "epoch": 0.03288100208768267, + "grad_norm": 9.582944869995117, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.1939, + "step": 21 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 9.496583938598633, + "learning_rate": 1.1e-06, + "loss": 1.2395, + "step": 22 + }, + { + "epoch": 0.0360125260960334, + "grad_norm": 5.912876129150391, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.1769, + "step": 23 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 2.6761083602905273, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.204, + "step": 24 + }, + { + "epoch": 0.03914405010438413, + "grad_norm": 11.136463165283203, + "learning_rate": 1.25e-06, + "loss": 1.2271, + "step": 25 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 23.912893295288086, + "learning_rate": 1.3e-06, + "loss": 1.1771, + "step": 26 + }, + { + "epoch": 0.042275574112734866, + "grad_norm": 57.08906555175781, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.2253, + "step": 27 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 36.11408615112305, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.1014, + "step": 28 + }, + { + "epoch": 0.04540709812108559, + "grad_norm": 51.39445495605469, + "learning_rate": 1.45e-06, + "loss": 1.1978, + "step": 29 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 27.394500732421875, + "learning_rate": 1.5e-06, + "loss": 1.1155, + "step": 30 + }, + { + "epoch": 0.048538622129436326, + "grad_norm": 6.844716548919678, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.1798, + "step": 31 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 2.397536277770996, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.1846, + "step": 32 + }, + { + "epoch": 0.05167014613778706, + "grad_norm": 7.821156024932861, + "learning_rate": 1.6500000000000003e-06, + "loss": 1.1312, + "step": 33 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 5.455922603607178, + "learning_rate": 1.7000000000000002e-06, + "loss": 1.2153, + "step": 34 + }, + { + "epoch": 0.054801670146137786, + "grad_norm": 5.377617835998535, + "learning_rate": 1.75e-06, + "loss": 1.1292, + "step": 35 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 7.172031879425049, + "learning_rate": 1.8000000000000001e-06, + "loss": 1.1459, + "step": 36 + }, + { + "epoch": 0.05793319415448852, + "grad_norm": 2.1464905738830566, + "learning_rate": 1.85e-06, + "loss": 1.1649, + "step": 37 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 2.7782726287841797, + "learning_rate": 1.9000000000000002e-06, + "loss": 1.159, + "step": 38 + }, + { + "epoch": 0.061064718162839246, + "grad_norm": 20.818374633789062, + "learning_rate": 1.9500000000000004e-06, + "loss": 1.0174, + "step": 39 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 22.776456832885742, + "learning_rate": 2.0000000000000003e-06, + "loss": 1.1425, + "step": 40 + }, + { + "epoch": 0.06419624217118998, + "grad_norm": 33.601287841796875, + "learning_rate": 2.05e-06, + "loss": 1.0498, + "step": 41 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.884684562683105, + "learning_rate": 2.1000000000000002e-06, + "loss": 1.1231, + "step": 42 + }, + { + "epoch": 0.06732776617954071, + "grad_norm": 1.951470971107483, + "learning_rate": 2.15e-06, + "loss": 1.1256, + "step": 43 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 2.584127426147461, + "learning_rate": 2.2e-06, + "loss": 1.1242, + "step": 44 + }, + { + "epoch": 0.07045929018789145, + "grad_norm": 6.982508659362793, + "learning_rate": 2.25e-06, + "loss": 1.0769, + "step": 45 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 7.940657138824463, + "learning_rate": 2.3000000000000004e-06, + "loss": 1.0516, + "step": 46 + }, + { + "epoch": 0.07359081419624217, + "grad_norm": 5.456567287445068, + "learning_rate": 2.35e-06, + "loss": 0.9894, + "step": 47 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 4.623636722564697, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9912, + "step": 48 + }, + { + "epoch": 0.0767223382045929, + "grad_norm": 2.182708978652954, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.1303, + "step": 49 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 2.388091802597046, + "learning_rate": 2.5e-06, + "loss": 1.0725, + "step": 50 + }, + { + "epoch": 0.07985386221294363, + "grad_norm": 2.2816522121429443, + "learning_rate": 2.55e-06, + "loss": 1.0895, + "step": 51 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 3.8391623497009277, + "learning_rate": 2.6e-06, + "loss": 1.0587, + "step": 52 + }, + { + "epoch": 0.08298538622129437, + "grad_norm": 3.700389862060547, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9795, + "step": 53 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 3.180649995803833, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.0434, + "step": 54 + }, + { + "epoch": 0.0861169102296451, + "grad_norm": 3.790774345397949, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9652, + "step": 55 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 1.8411438465118408, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.947, + "step": 56 + }, + { + "epoch": 0.08924843423799582, + "grad_norm": 2.100008010864258, + "learning_rate": 2.85e-06, + "loss": 1.0129, + "step": 57 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 1.103857159614563, + "learning_rate": 2.9e-06, + "loss": 0.9993, + "step": 58 + }, + { + "epoch": 0.09237995824634655, + "grad_norm": 1.232283353805542, + "learning_rate": 2.95e-06, + "loss": 0.9988, + "step": 59 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 2.4731297492980957, + "learning_rate": 3e-06, + "loss": 0.9873, + "step": 60 + }, + { + "epoch": 0.09551148225469729, + "grad_norm": 1.1080410480499268, + "learning_rate": 3.05e-06, + "loss": 1.0657, + "step": 61 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 1.1155121326446533, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.974, + "step": 62 + }, + { + "epoch": 0.09864300626304802, + "grad_norm": 1.763231873512268, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.0215, + "step": 63 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 1.1961321830749512, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.9955, + "step": 64 + }, + { + "epoch": 0.10177453027139875, + "grad_norm": 1.377040982246399, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.9834, + "step": 65 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 1.373260498046875, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.984, + "step": 66 + }, + { + "epoch": 0.10490605427974947, + "grad_norm": 2.034064292907715, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.973, + "step": 67 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 1.41798734664917, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.925, + "step": 68 + }, + { + "epoch": 0.1080375782881002, + "grad_norm": 1.7425439357757568, + "learning_rate": 3.45e-06, + "loss": 0.9884, + "step": 69 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 2.369657278060913, + "learning_rate": 3.5e-06, + "loss": 1.0048, + "step": 70 + }, + { + "epoch": 0.11116910229645094, + "grad_norm": 1.4699137210845947, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.9155, + "step": 71 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 1.1473263502120972, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8445, + "step": 72 + }, + { + "epoch": 0.11430062630480167, + "grad_norm": 2.315568685531616, + "learning_rate": 3.65e-06, + "loss": 0.977, + "step": 73 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 1.8554916381835938, + "learning_rate": 3.7e-06, + "loss": 0.841, + "step": 74 + }, + { + "epoch": 0.1174321503131524, + "grad_norm": 1.1517397165298462, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.9478, + "step": 75 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 2.799316883087158, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.9873, + "step": 76 + }, + { + "epoch": 0.12056367432150313, + "grad_norm": 1.11930251121521, + "learning_rate": 3.85e-06, + "loss": 0.9214, + "step": 77 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 1.0082950592041016, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9115, + "step": 78 + }, + { + "epoch": 0.12369519832985386, + "grad_norm": 1.5534541606903076, + "learning_rate": 3.95e-06, + "loss": 1.04, + "step": 79 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 1.3469066619873047, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9091, + "step": 80 + }, + { + "epoch": 0.1268267223382046, + "grad_norm": 1.1149636507034302, + "learning_rate": 4.05e-06, + "loss": 0.8829, + "step": 81 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 2.5933806896209717, + "learning_rate": 4.1e-06, + "loss": 0.9343, + "step": 82 + }, + { + "epoch": 0.1299582463465553, + "grad_norm": 1.980901837348938, + "learning_rate": 4.15e-06, + "loss": 0.9028, + "step": 83 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 0.8476768732070923, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.9038, + "step": 84 + }, + { + "epoch": 0.13308977035490605, + "grad_norm": 1.1239815950393677, + "learning_rate": 4.25e-06, + "loss": 1.0359, + "step": 85 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 1.0159156322479248, + "learning_rate": 4.3e-06, + "loss": 0.8705, + "step": 86 + }, + { + "epoch": 0.13622129436325678, + "grad_norm": 2.180856466293335, + "learning_rate": 4.350000000000001e-06, + "loss": 0.9834, + "step": 87 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 1.5725276470184326, + "learning_rate": 4.4e-06, + "loss": 0.9449, + "step": 88 + }, + { + "epoch": 0.1393528183716075, + "grad_norm": 1.5560917854309082, + "learning_rate": 4.450000000000001e-06, + "loss": 1.0109, + "step": 89 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 1.1096960306167603, + "learning_rate": 4.5e-06, + "loss": 0.9081, + "step": 90 + }, + { + "epoch": 0.14248434237995825, + "grad_norm": 1.2525160312652588, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.9907, + "step": 91 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 0.9198770523071289, + "learning_rate": 4.600000000000001e-06, + "loss": 0.9266, + "step": 92 + }, + { + "epoch": 0.14561586638830898, + "grad_norm": 0.992057740688324, + "learning_rate": 4.65e-06, + "loss": 0.9556, + "step": 93 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 1.00990891456604, + "learning_rate": 4.7e-06, + "loss": 0.9712, + "step": 94 + }, + { + "epoch": 0.1487473903966597, + "grad_norm": 1.1176848411560059, + "learning_rate": 4.75e-06, + "loss": 0.9604, + "step": 95 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 0.9323613047599792, + "learning_rate": 4.800000000000001e-06, + "loss": 0.9708, + "step": 96 + }, + { + "epoch": 0.15187891440501045, + "grad_norm": 0.9800642728805542, + "learning_rate": 4.85e-06, + "loss": 0.9868, + "step": 97 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 1.487841010093689, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8679, + "step": 98 + }, + { + "epoch": 0.15501043841336118, + "grad_norm": 0.95517897605896, + "learning_rate": 4.95e-06, + "loss": 0.7949, + "step": 99 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 0.9686336517333984, + "learning_rate": 5e-06, + "loss": 0.8655, + "step": 100 + }, + { + "epoch": 0.1581419624217119, + "grad_norm": 1.9481035470962524, + "learning_rate": 4.999999112316954e-06, + "loss": 0.9328, + "step": 101 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 1.272631287574768, + "learning_rate": 4.9999964492684465e-06, + "loss": 0.7869, + "step": 102 + }, + { + "epoch": 0.16127348643006262, + "grad_norm": 1.1117815971374512, + "learning_rate": 4.999992010856368e-06, + "loss": 0.949, + "step": 103 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 0.9681039452552795, + "learning_rate": 4.999985797083871e-06, + "loss": 1.0396, + "step": 104 + }, + { + "epoch": 0.16440501043841335, + "grad_norm": 0.9797878265380859, + "learning_rate": 4.999977807955368e-06, + "loss": 0.9393, + "step": 105 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 1.0239158868789673, + "learning_rate": 4.999968043476532e-06, + "loss": 0.8766, + "step": 106 + }, + { + "epoch": 0.16753653444676408, + "grad_norm": 0.9314751029014587, + "learning_rate": 4.999956503654299e-06, + "loss": 0.8545, + "step": 107 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 1.1627179384231567, + "learning_rate": 4.999943188496862e-06, + "loss": 0.9241, + "step": 108 + }, + { + "epoch": 0.17066805845511482, + "grad_norm": 1.022752285003662, + "learning_rate": 4.9999280980136765e-06, + "loss": 0.9662, + "step": 109 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 1.0513653755187988, + "learning_rate": 4.99991123221546e-06, + "loss": 0.9054, + "step": 110 + }, + { + "epoch": 0.17379958246346555, + "grad_norm": 0.9488444328308105, + "learning_rate": 4.99989259111419e-06, + "loss": 0.8921, + "step": 111 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 0.9934154748916626, + "learning_rate": 4.999872174723104e-06, + "loss": 0.9848, + "step": 112 + }, + { + "epoch": 0.17693110647181629, + "grad_norm": 1.2035515308380127, + "learning_rate": 4.9998499830567e-06, + "loss": 0.9833, + "step": 113 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 1.027538537979126, + "learning_rate": 4.999826016130739e-06, + "loss": 1.0089, + "step": 114 + }, + { + "epoch": 0.18006263048016702, + "grad_norm": 1.0635300874710083, + "learning_rate": 4.999800273962238e-06, + "loss": 0.8863, + "step": 115 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 0.9978796243667603, + "learning_rate": 4.999772756569482e-06, + "loss": 0.899, + "step": 116 + }, + { + "epoch": 0.18319415448851775, + "grad_norm": 0.9971838593482971, + "learning_rate": 4.999743463972008e-06, + "loss": 0.9515, + "step": 117 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 1.2436699867248535, + "learning_rate": 4.999712396190622e-06, + "loss": 0.9101, + "step": 118 + }, + { + "epoch": 0.18632567849686849, + "grad_norm": 0.9573557376861572, + "learning_rate": 4.999679553247383e-06, + "loss": 0.8817, + "step": 119 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 0.9105668067932129, + "learning_rate": 4.9996449351656165e-06, + "loss": 1.0077, + "step": 120 + }, + { + "epoch": 0.18945720250521922, + "grad_norm": 0.8741400241851807, + "learning_rate": 4.999608541969905e-06, + "loss": 0.912, + "step": 121 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 1.1172688007354736, + "learning_rate": 4.999570373686095e-06, + "loss": 0.8062, + "step": 122 + }, + { + "epoch": 0.19258872651356992, + "grad_norm": 0.940539538860321, + "learning_rate": 4.9995304303412905e-06, + "loss": 0.8944, + "step": 123 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 2.1082208156585693, + "learning_rate": 4.999488711963857e-06, + "loss": 0.9668, + "step": 124 + }, + { + "epoch": 0.19572025052192066, + "grad_norm": 1.0358761548995972, + "learning_rate": 4.99944521858342e-06, + "loss": 0.9032, + "step": 125 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 2.0804238319396973, + "learning_rate": 4.999399950230867e-06, + "loss": 0.9819, + "step": 126 + }, + { + "epoch": 0.1988517745302714, + "grad_norm": 1.0986278057098389, + "learning_rate": 4.9993529069383465e-06, + "loss": 0.9034, + "step": 127 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 1.2374722957611084, + "learning_rate": 4.999304088739263e-06, + "loss": 0.9255, + "step": 128 + }, + { + "epoch": 0.20198329853862212, + "grad_norm": 1.0443146228790283, + "learning_rate": 4.999253495668287e-06, + "loss": 0.9352, + "step": 129 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 1.009800910949707, + "learning_rate": 4.999201127761346e-06, + "loss": 0.8324, + "step": 130 + }, + { + "epoch": 0.20511482254697286, + "grad_norm": 0.892861008644104, + "learning_rate": 4.9991469850556295e-06, + "loss": 0.886, + "step": 131 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 1.1675922870635986, + "learning_rate": 4.999091067589587e-06, + "loss": 0.967, + "step": 132 + }, + { + "epoch": 0.2082463465553236, + "grad_norm": 1.0169954299926758, + "learning_rate": 4.999033375402926e-06, + "loss": 0.8498, + "step": 133 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 1.0006227493286133, + "learning_rate": 4.99897390853662e-06, + "loss": 0.8481, + "step": 134 + }, + { + "epoch": 0.21137787056367432, + "grad_norm": 0.9049347639083862, + "learning_rate": 4.9989126670328945e-06, + "loss": 0.9186, + "step": 135 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 1.2151603698730469, + "learning_rate": 4.9988496509352444e-06, + "loss": 0.9067, + "step": 136 + }, + { + "epoch": 0.21450939457202506, + "grad_norm": 1.0112006664276123, + "learning_rate": 4.998784860288417e-06, + "loss": 0.8843, + "step": 137 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 0.9316662549972534, + "learning_rate": 4.9987182951384264e-06, + "loss": 0.8635, + "step": 138 + }, + { + "epoch": 0.2176409185803758, + "grad_norm": 0.942743718624115, + "learning_rate": 4.9986499555325406e-06, + "loss": 0.8481, + "step": 139 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 1.0298817157745361, + "learning_rate": 4.998579841519292e-06, + "loss": 0.865, + "step": 140 + }, + { + "epoch": 0.22077244258872653, + "grad_norm": 1.1556729078292847, + "learning_rate": 4.998507953148472e-06, + "loss": 0.8731, + "step": 141 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 0.997958779335022, + "learning_rate": 4.9984342904711315e-06, + "loss": 0.7949, + "step": 142 + }, + { + "epoch": 0.22390396659707723, + "grad_norm": 0.9565998315811157, + "learning_rate": 4.998358853539582e-06, + "loss": 0.8875, + "step": 143 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 0.9667196273803711, + "learning_rate": 4.998281642407394e-06, + "loss": 0.8537, + "step": 144 + }, + { + "epoch": 0.22703549060542796, + "grad_norm": 0.9701303839683533, + "learning_rate": 4.998202657129401e-06, + "loss": 0.873, + "step": 145 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 0.9936159253120422, + "learning_rate": 4.998121897761692e-06, + "loss": 0.8915, + "step": 146 + }, + { + "epoch": 0.2301670146137787, + "grad_norm": 1.5054311752319336, + "learning_rate": 4.998039364361617e-06, + "loss": 0.9944, + "step": 147 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 1.5895891189575195, + "learning_rate": 4.99795505698779e-06, + "loss": 0.9282, + "step": 148 + }, + { + "epoch": 0.23329853862212943, + "grad_norm": 1.0141469240188599, + "learning_rate": 4.9978689757000785e-06, + "loss": 0.9166, + "step": 149 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 1.1720311641693115, + "learning_rate": 4.997781120559615e-06, + "loss": 0.9404, + "step": 150 + }, + { + "epoch": 0.23643006263048016, + "grad_norm": 0.9536433815956116, + "learning_rate": 4.997691491628789e-06, + "loss": 0.9339, + "step": 151 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 0.9407872557640076, + "learning_rate": 4.997600088971249e-06, + "loss": 0.9019, + "step": 152 + }, + { + "epoch": 0.2395615866388309, + "grad_norm": 0.9202889204025269, + "learning_rate": 4.997506912651906e-06, + "loss": 0.9598, + "step": 153 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 1.0630513429641724, + "learning_rate": 4.997411962736927e-06, + "loss": 0.8554, + "step": 154 + }, + { + "epoch": 0.24269311064718163, + "grad_norm": 0.8378801941871643, + "learning_rate": 4.997315239293742e-06, + "loss": 0.8583, + "step": 155 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 0.9442815184593201, + "learning_rate": 4.997216742391038e-06, + "loss": 0.8518, + "step": 156 + }, + { + "epoch": 0.24582463465553236, + "grad_norm": 0.9273308515548706, + "learning_rate": 4.997116472098763e-06, + "loss": 0.8942, + "step": 157 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 0.9028997421264648, + "learning_rate": 4.997014428488121e-06, + "loss": 0.8592, + "step": 158 + }, + { + "epoch": 0.2489561586638831, + "grad_norm": 0.9877567887306213, + "learning_rate": 4.996910611631582e-06, + "loss": 0.9316, + "step": 159 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 0.961592435836792, + "learning_rate": 4.996805021602868e-06, + "loss": 0.8875, + "step": 160 + }, + { + "epoch": 0.2520876826722338, + "grad_norm": 0.9923007488250732, + "learning_rate": 4.996697658476965e-06, + "loss": 0.8554, + "step": 161 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 0.956763744354248, + "learning_rate": 4.996588522330114e-06, + "loss": 0.9381, + "step": 162 + }, + { + "epoch": 0.25521920668058456, + "grad_norm": 0.919992208480835, + "learning_rate": 4.996477613239822e-06, + "loss": 0.9016, + "step": 163 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 0.9655581712722778, + "learning_rate": 4.996364931284847e-06, + "loss": 0.896, + "step": 164 + }, + { + "epoch": 0.25835073068893527, + "grad_norm": 0.9159606099128723, + "learning_rate": 4.996250476545212e-06, + "loss": 0.9475, + "step": 165 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 1.1147626638412476, + "learning_rate": 4.996134249102195e-06, + "loss": 0.8793, + "step": 166 + }, + { + "epoch": 0.26148225469728603, + "grad_norm": 1.112069010734558, + "learning_rate": 4.996016249038334e-06, + "loss": 0.8765, + "step": 167 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 1.0366944074630737, + "learning_rate": 4.995896476437428e-06, + "loss": 0.9356, + "step": 168 + }, + { + "epoch": 0.26461377870563674, + "grad_norm": 0.9494239091873169, + "learning_rate": 4.995774931384534e-06, + "loss": 0.8246, + "step": 169 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 0.9819200038909912, + "learning_rate": 4.995651613965964e-06, + "loss": 0.9403, + "step": 170 + }, + { + "epoch": 0.2677453027139875, + "grad_norm": 1.7329354286193848, + "learning_rate": 4.995526524269293e-06, + "loss": 0.9091, + "step": 171 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.1415139436721802, + "learning_rate": 4.995399662383352e-06, + "loss": 0.9106, + "step": 172 + }, + { + "epoch": 0.2708768267223382, + "grad_norm": 0.9471906423568726, + "learning_rate": 4.995271028398233e-06, + "loss": 0.8858, + "step": 173 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 1.096053123474121, + "learning_rate": 4.995140622405284e-06, + "loss": 0.7502, + "step": 174 + }, + { + "epoch": 0.2740083507306889, + "grad_norm": 1.285410761833191, + "learning_rate": 4.995008444497113e-06, + "loss": 0.9288, + "step": 175 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 0.9714595079421997, + "learning_rate": 4.994874494767585e-06, + "loss": 0.8824, + "step": 176 + }, + { + "epoch": 0.27713987473903967, + "grad_norm": 0.8399011492729187, + "learning_rate": 4.994738773311824e-06, + "loss": 0.8703, + "step": 177 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 0.9454224109649658, + "learning_rate": 4.994601280226212e-06, + "loss": 0.9627, + "step": 178 + }, + { + "epoch": 0.2802713987473904, + "grad_norm": 0.8983661532402039, + "learning_rate": 4.99446201560839e-06, + "loss": 0.9086, + "step": 179 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 0.9814153909683228, + "learning_rate": 4.994320979557256e-06, + "loss": 0.908, + "step": 180 + }, + { + "epoch": 0.28340292275574114, + "grad_norm": 0.8405355215072632, + "learning_rate": 4.9941781721729645e-06, + "loss": 0.8414, + "step": 181 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 0.950950026512146, + "learning_rate": 4.994033593556933e-06, + "loss": 0.9091, + "step": 182 + }, + { + "epoch": 0.28653444676409184, + "grad_norm": 1.1019747257232666, + "learning_rate": 4.99388724381183e-06, + "loss": 0.8305, + "step": 183 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 0.9339331984519958, + "learning_rate": 4.9937391230415875e-06, + "loss": 0.8542, + "step": 184 + }, + { + "epoch": 0.2896659707724426, + "grad_norm": 0.8813616633415222, + "learning_rate": 4.993589231351394e-06, + "loss": 0.848, + "step": 185 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 0.8916418552398682, + "learning_rate": 4.9934375688476905e-06, + "loss": 0.8032, + "step": 186 + }, + { + "epoch": 0.2927974947807933, + "grad_norm": 0.9891005754470825, + "learning_rate": 4.993284135638182e-06, + "loss": 0.7937, + "step": 187 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 0.9842082262039185, + "learning_rate": 4.99312893183183e-06, + "loss": 0.9532, + "step": 188 + }, + { + "epoch": 0.29592901878914407, + "grad_norm": 0.993478536605835, + "learning_rate": 4.99297195753885e-06, + "loss": 0.9406, + "step": 189 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 1.0654981136322021, + "learning_rate": 4.992813212870718e-06, + "loss": 0.907, + "step": 190 + }, + { + "epoch": 0.2990605427974948, + "grad_norm": 0.8937190771102905, + "learning_rate": 4.992652697940164e-06, + "loss": 0.9307, + "step": 191 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 0.9006052613258362, + "learning_rate": 4.992490412861178e-06, + "loss": 0.8808, + "step": 192 + }, + { + "epoch": 0.30219206680584554, + "grad_norm": 0.9601522088050842, + "learning_rate": 4.992326357749007e-06, + "loss": 0.8495, + "step": 193 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 0.9814243316650391, + "learning_rate": 4.992160532720153e-06, + "loss": 0.8204, + "step": 194 + }, + { + "epoch": 0.30532359081419624, + "grad_norm": 0.9660456776618958, + "learning_rate": 4.991992937892377e-06, + "loss": 0.8209, + "step": 195 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 0.8504270315170288, + "learning_rate": 4.991823573384695e-06, + "loss": 0.8415, + "step": 196 + }, + { + "epoch": 0.30845511482254695, + "grad_norm": 0.8733190894126892, + "learning_rate": 4.991652439317382e-06, + "loss": 0.8753, + "step": 197 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 0.9761794805526733, + "learning_rate": 4.991479535811967e-06, + "loss": 0.8923, + "step": 198 + }, + { + "epoch": 0.3115866388308977, + "grad_norm": 1.0400285720825195, + "learning_rate": 4.991304862991238e-06, + "loss": 0.899, + "step": 199 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 0.8561192750930786, + "learning_rate": 4.991128420979237e-06, + "loss": 0.9117, + "step": 200 + }, + { + "epoch": 0.3147181628392484, + "grad_norm": 0.8839634656906128, + "learning_rate": 4.990950209901263e-06, + "loss": 0.9022, + "step": 201 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 0.9569464921951294, + "learning_rate": 4.990770229883873e-06, + "loss": 0.9013, + "step": 202 + }, + { + "epoch": 0.3178496868475992, + "grad_norm": 1.0600861310958862, + "learning_rate": 4.990588481054881e-06, + "loss": 0.8387, + "step": 203 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 0.9492761492729187, + "learning_rate": 4.990404963543352e-06, + "loss": 0.8521, + "step": 204 + }, + { + "epoch": 0.3209812108559499, + "grad_norm": 0.9534817934036255, + "learning_rate": 4.990219677479612e-06, + "loss": 0.8009, + "step": 205 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 0.9918960928916931, + "learning_rate": 4.990032622995242e-06, + "loss": 0.8215, + "step": 206 + }, + { + "epoch": 0.32411273486430064, + "grad_norm": 0.8751630783081055, + "learning_rate": 4.989843800223077e-06, + "loss": 0.8744, + "step": 207 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 1.0034785270690918, + "learning_rate": 4.9896532092972085e-06, + "loss": 0.884, + "step": 208 + }, + { + "epoch": 0.32724425887265135, + "grad_norm": 1.2693432569503784, + "learning_rate": 4.989460850352984e-06, + "loss": 0.9689, + "step": 209 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 0.9923811554908752, + "learning_rate": 4.989266723527007e-06, + "loss": 0.9622, + "step": 210 + }, + { + "epoch": 0.3303757828810021, + "grad_norm": 0.9439074397087097, + "learning_rate": 4.989070828957137e-06, + "loss": 0.9155, + "step": 211 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.2564488649368286, + "learning_rate": 4.988873166782485e-06, + "loss": 0.8712, + "step": 212 + }, + { + "epoch": 0.3335073068893528, + "grad_norm": 1.1441881656646729, + "learning_rate": 4.988673737143423e-06, + "loss": 0.8895, + "step": 213 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 0.972155749797821, + "learning_rate": 4.988472540181574e-06, + "loss": 0.9158, + "step": 214 + }, + { + "epoch": 0.3366388308977035, + "grad_norm": 0.9536120295524597, + "learning_rate": 4.988269576039817e-06, + "loss": 0.7963, + "step": 215 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 0.8956594467163086, + "learning_rate": 4.988064844862288e-06, + "loss": 0.8947, + "step": 216 + }, + { + "epoch": 0.3397703549060543, + "grad_norm": 0.9503691792488098, + "learning_rate": 4.987858346794373e-06, + "loss": 0.7857, + "step": 217 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 0.9471456408500671, + "learning_rate": 4.987650081982719e-06, + "loss": 0.9118, + "step": 218 + }, + { + "epoch": 0.342901878914405, + "grad_norm": 0.9485713839530945, + "learning_rate": 4.987440050575222e-06, + "loss": 0.8188, + "step": 219 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 1.0775927305221558, + "learning_rate": 4.987228252721037e-06, + "loss": 0.9469, + "step": 220 + }, + { + "epoch": 0.34603340292275575, + "grad_norm": 0.9173988103866577, + "learning_rate": 4.9870146885705705e-06, + "loss": 0.8566, + "step": 221 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.008735179901123, + "learning_rate": 4.986799358275485e-06, + "loss": 0.8637, + "step": 222 + }, + { + "epoch": 0.34916492693110646, + "grad_norm": 0.8183198571205139, + "learning_rate": 4.986582261988695e-06, + "loss": 0.9321, + "step": 223 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 0.8951883316040039, + "learning_rate": 4.986363399864372e-06, + "loss": 0.8015, + "step": 224 + }, + { + "epoch": 0.3522964509394572, + "grad_norm": 0.9275704622268677, + "learning_rate": 4.986142772057941e-06, + "loss": 0.8242, + "step": 225 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 0.9548516273498535, + "learning_rate": 4.985920378726077e-06, + "loss": 0.7889, + "step": 226 + }, + { + "epoch": 0.3554279749478079, + "grad_norm": 1.1203585863113403, + "learning_rate": 4.985696220026714e-06, + "loss": 0.835, + "step": 227 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 0.9525771141052246, + "learning_rate": 4.985470296119038e-06, + "loss": 0.9162, + "step": 228 + }, + { + "epoch": 0.3585594989561587, + "grad_norm": 1.150404930114746, + "learning_rate": 4.985242607163488e-06, + "loss": 0.9456, + "step": 229 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.01652991771698, + "learning_rate": 4.985013153321755e-06, + "loss": 0.87, + "step": 230 + }, + { + "epoch": 0.3616910229645094, + "grad_norm": 0.9455218315124512, + "learning_rate": 4.984781934756786e-06, + "loss": 0.8466, + "step": 231 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 0.9125218391418457, + "learning_rate": 4.984548951632779e-06, + "loss": 0.8764, + "step": 232 + }, + { + "epoch": 0.36482254697286015, + "grad_norm": 2.0192980766296387, + "learning_rate": 4.984314204115187e-06, + "loss": 0.7866, + "step": 233 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.8118515014648438, + "learning_rate": 4.984077692370716e-06, + "loss": 0.8703, + "step": 234 + }, + { + "epoch": 0.36795407098121086, + "grad_norm": 1.2995024919509888, + "learning_rate": 4.983839416567322e-06, + "loss": 0.9459, + "step": 235 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 1.0611528158187866, + "learning_rate": 4.983599376874216e-06, + "loss": 0.8456, + "step": 236 + }, + { + "epoch": 0.37108559498956156, + "grad_norm": 1.2018043994903564, + "learning_rate": 4.983357573461863e-06, + "loss": 0.914, + "step": 237 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9091911315917969, + "learning_rate": 4.983114006501978e-06, + "loss": 0.791, + "step": 238 + }, + { + "epoch": 0.3742171189979123, + "grad_norm": 0.980873167514801, + "learning_rate": 4.982868676167528e-06, + "loss": 0.77, + "step": 239 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 0.8945695161819458, + "learning_rate": 4.982621582632735e-06, + "loss": 0.8038, + "step": 240 + }, + { + "epoch": 0.37734864300626303, + "grad_norm": 0.9325360655784607, + "learning_rate": 4.9823727260730705e-06, + "loss": 0.8908, + "step": 241 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 0.987630307674408, + "learning_rate": 4.98212210666526e-06, + "loss": 0.8642, + "step": 242 + }, + { + "epoch": 0.3804801670146138, + "grad_norm": 0.9151357412338257, + "learning_rate": 4.98186972458728e-06, + "loss": 0.8437, + "step": 243 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 0.9084850549697876, + "learning_rate": 4.981615580018358e-06, + "loss": 0.8775, + "step": 244 + }, + { + "epoch": 0.3836116910229645, + "grad_norm": 0.9786385297775269, + "learning_rate": 4.981359673138974e-06, + "loss": 0.8251, + "step": 245 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 0.8947877883911133, + "learning_rate": 4.98110200413086e-06, + "loss": 0.7709, + "step": 246 + }, + { + "epoch": 0.38674321503131526, + "grad_norm": 0.967019259929657, + "learning_rate": 4.9808425731769985e-06, + "loss": 0.8954, + "step": 247 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 1.1066704988479614, + "learning_rate": 4.980581380461622e-06, + "loss": 0.8544, + "step": 248 + }, + { + "epoch": 0.38987473903966596, + "grad_norm": 0.9750981330871582, + "learning_rate": 4.980318426170218e-06, + "loss": 0.8354, + "step": 249 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 0.8727232217788696, + "learning_rate": 4.980053710489521e-06, + "loss": 0.8911, + "step": 250 + }, + { + "epoch": 0.3930062630480167, + "grad_norm": 0.9337188601493835, + "learning_rate": 4.979787233607518e-06, + "loss": 0.8567, + "step": 251 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 0.890701174736023, + "learning_rate": 4.979518995713448e-06, + "loss": 0.8327, + "step": 252 + }, + { + "epoch": 0.39613778705636743, + "grad_norm": 0.9973195195198059, + "learning_rate": 4.979248996997797e-06, + "loss": 0.9065, + "step": 253 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 0.9894055128097534, + "learning_rate": 4.978977237652306e-06, + "loss": 0.8729, + "step": 254 + }, + { + "epoch": 0.39926931106471814, + "grad_norm": 1.0398536920547485, + "learning_rate": 4.978703717869961e-06, + "loss": 0.861, + "step": 255 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.1713942289352417, + "learning_rate": 4.978428437845003e-06, + "loss": 0.8183, + "step": 256 + }, + { + "epoch": 0.4024008350730689, + "grad_norm": 0.9868776202201843, + "learning_rate": 4.9781513977729225e-06, + "loss": 0.941, + "step": 257 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 0.9260863065719604, + "learning_rate": 4.977872597850455e-06, + "loss": 0.8963, + "step": 258 + }, + { + "epoch": 0.4055323590814196, + "grad_norm": 0.9553239941596985, + "learning_rate": 4.977592038275592e-06, + "loss": 0.8102, + "step": 259 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 0.9426907300949097, + "learning_rate": 4.977309719247571e-06, + "loss": 0.89, + "step": 260 + }, + { + "epoch": 0.40866388308977036, + "grad_norm": 0.9691527485847473, + "learning_rate": 4.97702564096688e-06, + "loss": 0.7967, + "step": 261 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 0.9917487502098083, + "learning_rate": 4.976739803635256e-06, + "loss": 0.8327, + "step": 262 + }, + { + "epoch": 0.41179540709812107, + "grad_norm": 0.992340087890625, + "learning_rate": 4.976452207455686e-06, + "loss": 0.9462, + "step": 263 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 0.9554483890533447, + "learning_rate": 4.9761628526324035e-06, + "loss": 0.8294, + "step": 264 + }, + { + "epoch": 0.41492693110647183, + "grad_norm": 0.9318789839744568, + "learning_rate": 4.975871739370895e-06, + "loss": 0.8079, + "step": 265 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 0.9158215522766113, + "learning_rate": 4.975578867877893e-06, + "loss": 0.8632, + "step": 266 + }, + { + "epoch": 0.41805845511482254, + "grad_norm": 0.9444370865821838, + "learning_rate": 4.975284238361378e-06, + "loss": 0.8897, + "step": 267 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 0.9949231743812561, + "learning_rate": 4.974987851030581e-06, + "loss": 0.7905, + "step": 268 + }, + { + "epoch": 0.4211899791231733, + "grad_norm": 0.9819440245628357, + "learning_rate": 4.974689706095981e-06, + "loss": 0.9183, + "step": 269 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0787206888198853, + "learning_rate": 4.974389803769304e-06, + "loss": 0.7739, + "step": 270 + }, + { + "epoch": 0.424321503131524, + "grad_norm": 0.8871361017227173, + "learning_rate": 4.9740881442635235e-06, + "loss": 0.9454, + "step": 271 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 0.9125893115997314, + "learning_rate": 4.973784727792863e-06, + "loss": 0.8017, + "step": 272 + }, + { + "epoch": 0.42745302713987476, + "grad_norm": 0.9102838635444641, + "learning_rate": 4.973479554572793e-06, + "loss": 0.8307, + "step": 273 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 0.8780474662780762, + "learning_rate": 4.973172624820031e-06, + "loss": 0.9402, + "step": 274 + }, + { + "epoch": 0.43058455114822547, + "grad_norm": 1.0059378147125244, + "learning_rate": 4.972863938752541e-06, + "loss": 0.7438, + "step": 275 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.9684615135192871, + "learning_rate": 4.972553496589537e-06, + "loss": 0.8897, + "step": 276 + }, + { + "epoch": 0.4337160751565762, + "grad_norm": 0.9615057110786438, + "learning_rate": 4.972241298551477e-06, + "loss": 0.8157, + "step": 277 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8589702844619751, + "learning_rate": 4.971927344860068e-06, + "loss": 0.7813, + "step": 278 + }, + { + "epoch": 0.43684759916492694, + "grad_norm": 0.900324285030365, + "learning_rate": 4.971611635738263e-06, + "loss": 0.7534, + "step": 279 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 0.8881833553314209, + "learning_rate": 4.971294171410261e-06, + "loss": 0.8255, + "step": 280 + }, + { + "epoch": 0.43997912317327764, + "grad_norm": 0.9723751544952393, + "learning_rate": 4.9709749521015085e-06, + "loss": 0.8796, + "step": 281 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 0.9341713190078735, + "learning_rate": 4.9706539780386995e-06, + "loss": 0.7429, + "step": 282 + }, + { + "epoch": 0.4431106471816284, + "grad_norm": 0.9354205131530762, + "learning_rate": 4.97033124944977e-06, + "loss": 0.8545, + "step": 283 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 0.9294624328613281, + "learning_rate": 4.970006766563906e-06, + "loss": 0.7783, + "step": 284 + }, + { + "epoch": 0.4462421711899791, + "grad_norm": 0.9584887027740479, + "learning_rate": 4.969680529611538e-06, + "loss": 0.8068, + "step": 285 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 0.9525735378265381, + "learning_rate": 4.9693525388243424e-06, + "loss": 0.8055, + "step": 286 + }, + { + "epoch": 0.44937369519832987, + "grad_norm": 0.8845317959785461, + "learning_rate": 4.96902279443524e-06, + "loss": 0.8318, + "step": 287 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 0.8932276964187622, + "learning_rate": 4.9686912966783965e-06, + "loss": 0.8185, + "step": 288 + }, + { + "epoch": 0.4525052192066806, + "grad_norm": 0.9881951808929443, + "learning_rate": 4.968358045789225e-06, + "loss": 0.9001, + "step": 289 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 0.995212197303772, + "learning_rate": 4.968023042004384e-06, + "loss": 0.75, + "step": 290 + }, + { + "epoch": 0.45563674321503134, + "grad_norm": 0.9997846484184265, + "learning_rate": 4.967686285561773e-06, + "loss": 0.8993, + "step": 291 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 0.9438545107841492, + "learning_rate": 4.967347776700538e-06, + "loss": 0.8964, + "step": 292 + }, + { + "epoch": 0.45876826722338204, + "grad_norm": 0.9164395332336426, + "learning_rate": 4.967007515661072e-06, + "loss": 0.8671, + "step": 293 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.895302951335907, + "learning_rate": 4.966665502685009e-06, + "loss": 0.8785, + "step": 294 + }, + { + "epoch": 0.4618997912317328, + "grad_norm": 0.9122863411903381, + "learning_rate": 4.966321738015227e-06, + "loss": 0.8499, + "step": 295 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.0097336769104004, + "learning_rate": 4.965976221895852e-06, + "loss": 0.83, + "step": 296 + }, + { + "epoch": 0.4650313152400835, + "grad_norm": 1.0664499998092651, + "learning_rate": 4.965628954572249e-06, + "loss": 0.7799, + "step": 297 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 0.9602341055870056, + "learning_rate": 4.96527993629103e-06, + "loss": 0.8061, + "step": 298 + }, + { + "epoch": 0.4681628392484342, + "grad_norm": 1.0133620500564575, + "learning_rate": 4.964929167300048e-06, + "loss": 0.9148, + "step": 299 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9351868033409119, + "learning_rate": 4.964576647848401e-06, + "loss": 0.83, + "step": 300 + }, + { + "epoch": 0.471294363256785, + "grad_norm": 1.0078860521316528, + "learning_rate": 4.964222378186429e-06, + "loss": 0.8391, + "step": 301 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 0.8966493606567383, + "learning_rate": 4.963866358565716e-06, + "loss": 0.8677, + "step": 302 + }, + { + "epoch": 0.4744258872651357, + "grad_norm": 0.9283657670021057, + "learning_rate": 4.963508589239087e-06, + "loss": 0.8787, + "step": 303 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 0.9540023803710938, + "learning_rate": 4.963149070460611e-06, + "loss": 0.8667, + "step": 304 + }, + { + "epoch": 0.47755741127348644, + "grad_norm": 0.9580283164978027, + "learning_rate": 4.9627878024855995e-06, + "loss": 0.8442, + "step": 305 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.9008278846740723, + "learning_rate": 4.962424785570605e-06, + "loss": 0.8605, + "step": 306 + }, + { + "epoch": 0.48068893528183715, + "grad_norm": 0.9681883454322815, + "learning_rate": 4.962060019973423e-06, + "loss": 0.7865, + "step": 307 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 0.8492597341537476, + "learning_rate": 4.9616935059530915e-06, + "loss": 0.8578, + "step": 308 + }, + { + "epoch": 0.4838204592901879, + "grad_norm": 0.9323164820671082, + "learning_rate": 4.9613252437698865e-06, + "loss": 0.8491, + "step": 309 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 0.9417833685874939, + "learning_rate": 4.960955233685331e-06, + "loss": 0.8108, + "step": 310 + }, + { + "epoch": 0.4869519832985386, + "grad_norm": 0.901364803314209, + "learning_rate": 4.960583475962185e-06, + "loss": 0.8958, + "step": 311 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8929979205131531, + "learning_rate": 4.96020997086445e-06, + "loss": 0.8237, + "step": 312 + }, + { + "epoch": 0.4900835073068894, + "grad_norm": 0.9422337412834167, + "learning_rate": 4.9598347186573716e-06, + "loss": 0.8024, + "step": 313 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 0.9165504574775696, + "learning_rate": 4.959457719607432e-06, + "loss": 0.8124, + "step": 314 + }, + { + "epoch": 0.4932150313152401, + "grad_norm": 1.0233370065689087, + "learning_rate": 4.959078973982356e-06, + "loss": 0.788, + "step": 315 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9375972151756287, + "learning_rate": 4.958698482051109e-06, + "loss": 0.839, + "step": 316 + }, + { + "epoch": 0.4963465553235908, + "grad_norm": 0.8805437684059143, + "learning_rate": 4.958316244083895e-06, + "loss": 0.7445, + "step": 317 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.0231893062591553, + "learning_rate": 4.95793226035216e-06, + "loss": 0.8257, + "step": 318 + }, + { + "epoch": 0.49947807933194155, + "grad_norm": 0.9125928282737732, + "learning_rate": 4.957546531128589e-06, + "loss": 0.7699, + "step": 319 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.8844016790390015, + "learning_rate": 4.957159056687104e-06, + "loss": 0.8784, + "step": 320 + }, + { + "epoch": 0.5026096033402923, + "grad_norm": 0.9770572781562805, + "learning_rate": 4.956769837302871e-06, + "loss": 0.8611, + "step": 321 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.9413725733757019, + "learning_rate": 4.956378873252293e-06, + "loss": 0.8056, + "step": 322 + }, + { + "epoch": 0.505741127348643, + "grad_norm": 0.9058850407600403, + "learning_rate": 4.955986164813008e-06, + "loss": 0.8162, + "step": 323 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 0.893100917339325, + "learning_rate": 4.955591712263901e-06, + "loss": 0.7554, + "step": 324 + }, + { + "epoch": 0.5088726513569938, + "grad_norm": 0.8967075943946838, + "learning_rate": 4.955195515885089e-06, + "loss": 0.8143, + "step": 325 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.9542785286903381, + "learning_rate": 4.954797575957929e-06, + "loss": 0.8102, + "step": 326 + }, + { + "epoch": 0.5120041753653445, + "grad_norm": 1.006451964378357, + "learning_rate": 4.954397892765018e-06, + "loss": 0.86, + "step": 327 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 0.9964715838432312, + "learning_rate": 4.953996466590189e-06, + "loss": 0.7166, + "step": 328 + }, + { + "epoch": 0.5151356993736952, + "grad_norm": 0.9568808674812317, + "learning_rate": 4.953593297718512e-06, + "loss": 0.8753, + "step": 329 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.9435825347900391, + "learning_rate": 4.953188386436298e-06, + "loss": 0.8961, + "step": 330 + }, + { + "epoch": 0.5182672233820459, + "grad_norm": 0.9386457800865173, + "learning_rate": 4.952781733031092e-06, + "loss": 0.8468, + "step": 331 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 0.9916554689407349, + "learning_rate": 4.952373337791678e-06, + "loss": 0.8274, + "step": 332 + }, + { + "epoch": 0.5213987473903967, + "grad_norm": 0.9351572394371033, + "learning_rate": 4.9519632010080765e-06, + "loss": 0.7964, + "step": 333 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9936269521713257, + "learning_rate": 4.951551322971545e-06, + "loss": 0.8354, + "step": 334 + }, + { + "epoch": 0.5245302713987474, + "grad_norm": 0.9678221940994263, + "learning_rate": 4.951137703974576e-06, + "loss": 0.9115, + "step": 335 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 0.9453133940696716, + "learning_rate": 4.9507223443109005e-06, + "loss": 0.7725, + "step": 336 + }, + { + "epoch": 0.5276617954070981, + "grad_norm": 0.9577226638793945, + "learning_rate": 4.950305244275484e-06, + "loss": 0.8679, + "step": 337 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 0.998489260673523, + "learning_rate": 4.94988640416453e-06, + "loss": 0.7505, + "step": 338 + }, + { + "epoch": 0.5307933194154488, + "grad_norm": 0.976026713848114, + "learning_rate": 4.949465824275475e-06, + "loss": 0.9083, + "step": 339 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.0041502714157104, + "learning_rate": 4.9490435049069925e-06, + "loss": 0.8815, + "step": 340 + }, + { + "epoch": 0.5339248434237995, + "grad_norm": 0.9839801788330078, + "learning_rate": 4.948619446358991e-06, + "loss": 0.8517, + "step": 341 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 0.9995140433311462, + "learning_rate": 4.948193648932616e-06, + "loss": 0.8157, + "step": 342 + }, + { + "epoch": 0.5370563674321504, + "grad_norm": 1.0782225131988525, + "learning_rate": 4.947766112930243e-06, + "loss": 0.9206, + "step": 343 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9981802105903625, + "learning_rate": 4.947336838655487e-06, + "loss": 0.8389, + "step": 344 + }, + { + "epoch": 0.5401878914405011, + "grad_norm": 0.9261236190795898, + "learning_rate": 4.946905826413195e-06, + "loss": 0.8035, + "step": 345 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.9195284247398376, + "learning_rate": 4.946473076509449e-06, + "loss": 0.8265, + "step": 346 + }, + { + "epoch": 0.5433194154488518, + "grad_norm": 0.9535123109817505, + "learning_rate": 4.946038589251566e-06, + "loss": 0.7528, + "step": 347 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 0.9969532489776611, + "learning_rate": 4.9456023649480935e-06, + "loss": 0.7787, + "step": 348 + }, + { + "epoch": 0.5464509394572025, + "grad_norm": 0.9612399935722351, + "learning_rate": 4.945164403908816e-06, + "loss": 0.8298, + "step": 349 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 0.9541627764701843, + "learning_rate": 4.944724706444749e-06, + "loss": 0.8039, + "step": 350 + }, + { + "epoch": 0.5495824634655533, + "grad_norm": 0.9916557669639587, + "learning_rate": 4.944283272868143e-06, + "loss": 0.9012, + "step": 351 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 0.8807661533355713, + "learning_rate": 4.94384010349248e-06, + "loss": 0.9005, + "step": 352 + }, + { + "epoch": 0.552713987473904, + "grad_norm": 1.0179708003997803, + "learning_rate": 4.943395198632476e-06, + "loss": 0.8845, + "step": 353 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.923014223575592, + "learning_rate": 4.942948558604077e-06, + "loss": 0.8173, + "step": 354 + }, + { + "epoch": 0.5558455114822547, + "grad_norm": 0.9318749308586121, + "learning_rate": 4.942500183724463e-06, + "loss": 0.8828, + "step": 355 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.0907000303268433, + "learning_rate": 4.942050074312048e-06, + "loss": 0.7951, + "step": 356 + }, + { + "epoch": 0.5589770354906054, + "grad_norm": 0.9811620712280273, + "learning_rate": 4.9415982306864744e-06, + "loss": 0.8826, + "step": 357 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9888150691986084, + "learning_rate": 4.9411446531686156e-06, + "loss": 0.8338, + "step": 358 + }, + { + "epoch": 0.5621085594989561, + "grad_norm": 0.9389257431030273, + "learning_rate": 4.9406893420805804e-06, + "loss": 0.8316, + "step": 359 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.9500603675842285, + "learning_rate": 4.940232297745705e-06, + "loss": 0.8764, + "step": 360 + }, + { + "epoch": 0.5652400835073069, + "grad_norm": 0.9748769998550415, + "learning_rate": 4.939773520488559e-06, + "loss": 0.8308, + "step": 361 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 0.9179681539535522, + "learning_rate": 4.93931301063494e-06, + "loss": 0.789, + "step": 362 + }, + { + "epoch": 0.5683716075156576, + "grad_norm": 0.8609325289726257, + "learning_rate": 4.938850768511878e-06, + "loss": 0.7368, + "step": 363 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 0.9375531673431396, + "learning_rate": 4.9383867944476325e-06, + "loss": 0.8164, + "step": 364 + }, + { + "epoch": 0.5715031315240083, + "grad_norm": 0.9386453032493591, + "learning_rate": 4.937921088771694e-06, + "loss": 0.7898, + "step": 365 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.025040864944458, + "learning_rate": 4.937453651814781e-06, + "loss": 0.8465, + "step": 366 + }, + { + "epoch": 0.574634655532359, + "grad_norm": 1.1037245988845825, + "learning_rate": 4.936984483908842e-06, + "loss": 0.7935, + "step": 367 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.230821132659912, + "learning_rate": 4.936513585387055e-06, + "loss": 0.8043, + "step": 368 + }, + { + "epoch": 0.5777661795407099, + "grad_norm": 1.090235948562622, + "learning_rate": 4.9360409565838265e-06, + "loss": 0.8726, + "step": 369 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8977810144424438, + "learning_rate": 4.935566597834793e-06, + "loss": 0.8667, + "step": 370 + }, + { + "epoch": 0.5808977035490606, + "grad_norm": 0.96401447057724, + "learning_rate": 4.935090509476818e-06, + "loss": 0.8494, + "step": 371 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 0.9290149211883545, + "learning_rate": 4.934612691847995e-06, + "loss": 0.7678, + "step": 372 + }, + { + "epoch": 0.5840292275574113, + "grad_norm": 0.9765584468841553, + "learning_rate": 4.934133145287644e-06, + "loss": 0.7947, + "step": 373 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 0.8939430117607117, + "learning_rate": 4.933651870136313e-06, + "loss": 0.885, + "step": 374 + }, + { + "epoch": 0.587160751565762, + "grad_norm": 0.9512760043144226, + "learning_rate": 4.933168866735777e-06, + "loss": 0.8863, + "step": 375 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 1.0334441661834717, + "learning_rate": 4.932684135429041e-06, + "loss": 0.8362, + "step": 376 + }, + { + "epoch": 0.5902922755741128, + "grad_norm": 0.9402525424957275, + "learning_rate": 4.932197676560334e-06, + "loss": 0.8758, + "step": 377 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 0.9615777730941772, + "learning_rate": 4.931709490475113e-06, + "loss": 0.8324, + "step": 378 + }, + { + "epoch": 0.5934237995824635, + "grad_norm": 0.9407138228416443, + "learning_rate": 4.931219577520061e-06, + "loss": 0.8454, + "step": 379 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.0415815114974976, + "learning_rate": 4.930727938043091e-06, + "loss": 0.8649, + "step": 380 + }, + { + "epoch": 0.5965553235908142, + "grad_norm": 1.0946165323257446, + "learning_rate": 4.9302345723933344e-06, + "loss": 0.7794, + "step": 381 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.0532523393630981, + "learning_rate": 4.929739480921157e-06, + "loss": 0.8555, + "step": 382 + }, + { + "epoch": 0.5996868475991649, + "grad_norm": 0.9441399574279785, + "learning_rate": 4.929242663978144e-06, + "loss": 0.7613, + "step": 383 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 0.9970793128013611, + "learning_rate": 4.92874412191711e-06, + "loss": 0.9037, + "step": 384 + }, + { + "epoch": 0.6028183716075156, + "grad_norm": 0.9777354001998901, + "learning_rate": 4.928243855092091e-06, + "loss": 0.9044, + "step": 385 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 0.9507855772972107, + "learning_rate": 4.92774186385835e-06, + "loss": 0.8359, + "step": 386 + }, + { + "epoch": 0.6059498956158664, + "grad_norm": 0.9827553033828735, + "learning_rate": 4.9272381485723765e-06, + "loss": 0.7938, + "step": 387 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 0.8896475434303284, + "learning_rate": 4.926732709591879e-06, + "loss": 0.8545, + "step": 388 + }, + { + "epoch": 0.6090814196242171, + "grad_norm": 0.9817705750465393, + "learning_rate": 4.926225547275795e-06, + "loss": 0.8456, + "step": 389 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 0.9646008610725403, + "learning_rate": 4.925716661984283e-06, + "loss": 0.7906, + "step": 390 + }, + { + "epoch": 0.6122129436325678, + "grad_norm": 0.8940444588661194, + "learning_rate": 4.9252060540787275e-06, + "loss": 0.8371, + "step": 391 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.0812394618988037, + "learning_rate": 4.924693723921734e-06, + "loss": 0.8056, + "step": 392 + }, + { + "epoch": 0.6153444676409185, + "grad_norm": 0.9953856468200684, + "learning_rate": 4.924179671877132e-06, + "loss": 0.8651, + "step": 393 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 0.9760703444480896, + "learning_rate": 4.9236638983099735e-06, + "loss": 0.845, + "step": 394 + }, + { + "epoch": 0.6184759916492694, + "grad_norm": 1.0127615928649902, + "learning_rate": 4.923146403586533e-06, + "loss": 0.7987, + "step": 395 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 0.8771396279335022, + "learning_rate": 4.9226271880743086e-06, + "loss": 0.8092, + "step": 396 + }, + { + "epoch": 0.6216075156576201, + "grad_norm": 0.8906831741333008, + "learning_rate": 4.9221062521420196e-06, + "loss": 0.8039, + "step": 397 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.0218392610549927, + "learning_rate": 4.9215835961596045e-06, + "loss": 0.8354, + "step": 398 + }, + { + "epoch": 0.6247390396659708, + "grad_norm": 1.0382038354873657, + "learning_rate": 4.921059220498227e-06, + "loss": 0.8021, + "step": 399 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 0.8997588157653809, + "learning_rate": 4.920533125530272e-06, + "loss": 0.8716, + "step": 400 + }, + { + "epoch": 0.6278705636743215, + "grad_norm": 0.8725724220275879, + "learning_rate": 4.920005311629342e-06, + "loss": 0.7934, + "step": 401 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.9289800524711609, + "learning_rate": 4.9194757791702625e-06, + "loss": 0.812, + "step": 402 + }, + { + "epoch": 0.6310020876826722, + "grad_norm": 0.9969078898429871, + "learning_rate": 4.918944528529079e-06, + "loss": 0.866, + "step": 403 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 0.9544941186904907, + "learning_rate": 4.918411560083058e-06, + "loss": 0.8035, + "step": 404 + }, + { + "epoch": 0.634133611691023, + "grad_norm": 0.9678053855895996, + "learning_rate": 4.917876874210686e-06, + "loss": 0.8254, + "step": 405 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 0.9782965779304504, + "learning_rate": 4.917340471291666e-06, + "loss": 0.7954, + "step": 406 + }, + { + "epoch": 0.6372651356993737, + "grad_norm": 0.9380600452423096, + "learning_rate": 4.916802351706923e-06, + "loss": 0.8287, + "step": 407 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 0.9455891847610474, + "learning_rate": 4.916262515838603e-06, + "loss": 0.8714, + "step": 408 + }, + { + "epoch": 0.6403966597077244, + "grad_norm": 0.8937395811080933, + "learning_rate": 4.915720964070065e-06, + "loss": 0.7618, + "step": 409 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 1.0304689407348633, + "learning_rate": 4.915177696785894e-06, + "loss": 0.8436, + "step": 410 + }, + { + "epoch": 0.6435281837160751, + "grad_norm": 0.9623110294342041, + "learning_rate": 4.914632714371885e-06, + "loss": 0.865, + "step": 411 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 0.9212867021560669, + "learning_rate": 4.914086017215059e-06, + "loss": 0.8232, + "step": 412 + }, + { + "epoch": 0.6466597077244259, + "grad_norm": 0.9817205667495728, + "learning_rate": 4.9135376057036485e-06, + "loss": 0.7926, + "step": 413 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 0.9576719403266907, + "learning_rate": 4.912987480227108e-06, + "loss": 0.8812, + "step": 414 + }, + { + "epoch": 0.6497912317327766, + "grad_norm": 0.9415407180786133, + "learning_rate": 4.912435641176106e-06, + "loss": 0.8985, + "step": 415 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 0.9084275364875793, + "learning_rate": 4.91188208894253e-06, + "loss": 0.8085, + "step": 416 + }, + { + "epoch": 0.6529227557411273, + "grad_norm": 0.8890969753265381, + "learning_rate": 4.911326823919482e-06, + "loss": 0.8127, + "step": 417 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 0.8727678656578064, + "learning_rate": 4.9107698465012815e-06, + "loss": 0.8656, + "step": 418 + }, + { + "epoch": 0.656054279749478, + "grad_norm": 1.0174667835235596, + "learning_rate": 4.910211157083465e-06, + "loss": 0.8358, + "step": 419 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.045121431350708, + "learning_rate": 4.909650756062782e-06, + "loss": 0.8239, + "step": 420 + }, + { + "epoch": 0.6591858037578288, + "grad_norm": 1.0688676834106445, + "learning_rate": 4.909088643837202e-06, + "loss": 0.9246, + "step": 421 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.9371877908706665, + "learning_rate": 4.908524820805905e-06, + "loss": 0.7862, + "step": 422 + }, + { + "epoch": 0.6623173277661796, + "grad_norm": 0.9567264318466187, + "learning_rate": 4.907959287369288e-06, + "loss": 0.8164, + "step": 423 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 0.9109166860580444, + "learning_rate": 4.907392043928964e-06, + "loss": 0.7016, + "step": 424 + }, + { + "epoch": 0.6654488517745303, + "grad_norm": 1.0378203392028809, + "learning_rate": 4.906823090887757e-06, + "loss": 0.8986, + "step": 425 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 0.9871951341629028, + "learning_rate": 4.906252428649708e-06, + "loss": 0.8366, + "step": 426 + }, + { + "epoch": 0.668580375782881, + "grad_norm": 0.8682780265808105, + "learning_rate": 4.905680057620072e-06, + "loss": 0.738, + "step": 427 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.8834647536277771, + "learning_rate": 4.9051059782053125e-06, + "loss": 0.8304, + "step": 428 + }, + { + "epoch": 0.6717118997912317, + "grad_norm": 1.0133378505706787, + "learning_rate": 4.904530190813113e-06, + "loss": 0.9369, + "step": 429 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 0.9967373013496399, + "learning_rate": 4.9039526958523676e-06, + "loss": 0.782, + "step": 430 + }, + { + "epoch": 0.6748434237995825, + "grad_norm": 1.0258032083511353, + "learning_rate": 4.90337349373318e-06, + "loss": 0.8214, + "step": 431 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 0.9814366102218628, + "learning_rate": 4.90279258486687e-06, + "loss": 0.8173, + "step": 432 + }, + { + "epoch": 0.6779749478079332, + "grad_norm": 0.9351440668106079, + "learning_rate": 4.902209969665966e-06, + "loss": 0.8815, + "step": 433 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 0.9490398168563843, + "learning_rate": 4.901625648544212e-06, + "loss": 0.8295, + "step": 434 + }, + { + "epoch": 0.6811064718162839, + "grad_norm": 1.0183534622192383, + "learning_rate": 4.9010396219165614e-06, + "loss": 0.8538, + "step": 435 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 0.9546058773994446, + "learning_rate": 4.900451890199179e-06, + "loss": 0.8385, + "step": 436 + }, + { + "epoch": 0.6842379958246346, + "grad_norm": 1.076120138168335, + "learning_rate": 4.8998624538094394e-06, + "loss": 0.8958, + "step": 437 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.8935732841491699, + "learning_rate": 4.899271313165929e-06, + "loss": 0.8674, + "step": 438 + }, + { + "epoch": 0.6873695198329853, + "grad_norm": 0.9574604630470276, + "learning_rate": 4.898678468688445e-06, + "loss": 0.8245, + "step": 439 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.931751549243927, + "learning_rate": 4.898083920797993e-06, + "loss": 0.8409, + "step": 440 + }, + { + "epoch": 0.6905010438413361, + "grad_norm": 0.937874436378479, + "learning_rate": 4.89748766991679e-06, + "loss": 0.8222, + "step": 441 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9606375694274902, + "learning_rate": 4.896889716468259e-06, + "loss": 0.8123, + "step": 442 + }, + { + "epoch": 0.6936325678496869, + "grad_norm": 0.8692249059677124, + "learning_rate": 4.896290060877038e-06, + "loss": 0.8532, + "step": 443 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 0.9597892165184021, + "learning_rate": 4.895688703568968e-06, + "loss": 0.8365, + "step": 444 + }, + { + "epoch": 0.6967640918580376, + "grad_norm": 0.958342432975769, + "learning_rate": 4.895085644971102e-06, + "loss": 0.8535, + "step": 445 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.9603601098060608, + "learning_rate": 4.894480885511699e-06, + "loss": 0.8227, + "step": 446 + }, + { + "epoch": 0.6998956158663883, + "grad_norm": 1.0010526180267334, + "learning_rate": 4.893874425620226e-06, + "loss": 0.8785, + "step": 447 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.0061219930648804, + "learning_rate": 4.893266265727361e-06, + "loss": 0.8336, + "step": 448 + }, + { + "epoch": 0.7030271398747391, + "grad_norm": 0.8967173099517822, + "learning_rate": 4.892656406264984e-06, + "loss": 0.8305, + "step": 449 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.962729275226593, + "learning_rate": 4.892044847666186e-06, + "loss": 0.7843, + "step": 450 + }, + { + "epoch": 0.7061586638830898, + "grad_norm": 1.0067342519760132, + "learning_rate": 4.891431590365262e-06, + "loss": 0.8111, + "step": 451 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 1.0252268314361572, + "learning_rate": 4.890816634797716e-06, + "loss": 0.8253, + "step": 452 + }, + { + "epoch": 0.7092901878914405, + "grad_norm": 0.9098564386367798, + "learning_rate": 4.890199981400255e-06, + "loss": 0.8526, + "step": 453 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 0.893135130405426, + "learning_rate": 4.889581630610793e-06, + "loss": 0.8217, + "step": 454 + }, + { + "epoch": 0.7124217118997912, + "grad_norm": 0.9225817322731018, + "learning_rate": 4.888961582868453e-06, + "loss": 0.8296, + "step": 455 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 0.973386824131012, + "learning_rate": 4.888339838613555e-06, + "loss": 0.7429, + "step": 456 + }, + { + "epoch": 0.715553235908142, + "grad_norm": 0.9829993844032288, + "learning_rate": 4.8877163982876306e-06, + "loss": 0.7405, + "step": 457 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 0.927728533744812, + "learning_rate": 4.887091262333414e-06, + "loss": 0.8257, + "step": 458 + }, + { + "epoch": 0.7186847599164927, + "grad_norm": 0.9419897794723511, + "learning_rate": 4.886464431194844e-06, + "loss": 0.7762, + "step": 459 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 0.9358823299407959, + "learning_rate": 4.885835905317061e-06, + "loss": 0.7794, + "step": 460 + }, + { + "epoch": 0.7218162839248434, + "grad_norm": 1.0290783643722534, + "learning_rate": 4.88520568514641e-06, + "loss": 0.8723, + "step": 461 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 0.9909068942070007, + "learning_rate": 4.884573771130442e-06, + "loss": 0.8302, + "step": 462 + }, + { + "epoch": 0.7249478079331941, + "grad_norm": 0.9972653985023499, + "learning_rate": 4.883940163717906e-06, + "loss": 0.7961, + "step": 463 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.9245638251304626, + "learning_rate": 4.883304863358757e-06, + "loss": 0.8057, + "step": 464 + }, + { + "epoch": 0.7280793319415448, + "grad_norm": 1.0097148418426514, + "learning_rate": 4.882667870504153e-06, + "loss": 0.7296, + "step": 465 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 0.9792670607566833, + "learning_rate": 4.882029185606448e-06, + "loss": 0.8006, + "step": 466 + }, + { + "epoch": 0.7312108559498957, + "grad_norm": 0.9308120608329773, + "learning_rate": 4.881388809119206e-06, + "loss": 0.7957, + "step": 467 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 0.9472301602363586, + "learning_rate": 4.880746741497187e-06, + "loss": 0.8984, + "step": 468 + }, + { + "epoch": 0.7343423799582464, + "grad_norm": 0.9898937344551086, + "learning_rate": 4.880102983196351e-06, + "loss": 0.8365, + "step": 469 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 0.9376612901687622, + "learning_rate": 4.879457534673863e-06, + "loss": 0.7838, + "step": 470 + }, + { + "epoch": 0.7374739039665971, + "grad_norm": 1.030616283416748, + "learning_rate": 4.878810396388085e-06, + "loss": 0.7698, + "step": 471 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.9583387970924377, + "learning_rate": 4.87816156879858e-06, + "loss": 0.7381, + "step": 472 + }, + { + "epoch": 0.7406054279749478, + "grad_norm": 1.00180184841156, + "learning_rate": 4.87751105236611e-06, + "loss": 0.8497, + "step": 473 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.9382562637329102, + "learning_rate": 4.876858847552638e-06, + "loss": 0.8698, + "step": 474 + }, + { + "epoch": 0.7437369519832986, + "grad_norm": 1.0104063749313354, + "learning_rate": 4.8762049548213255e-06, + "loss": 0.8444, + "step": 475 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 0.9321765303611755, + "learning_rate": 4.87554937463653e-06, + "loss": 0.8027, + "step": 476 + }, + { + "epoch": 0.7468684759916493, + "grad_norm": 0.9527666568756104, + "learning_rate": 4.874892107463811e-06, + "loss": 0.8069, + "step": 477 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 0.9576907157897949, + "learning_rate": 4.874233153769924e-06, + "loss": 0.774, + "step": 478 + }, + { + "epoch": 0.75, + "grad_norm": 1.0516161918640137, + "learning_rate": 4.873572514022822e-06, + "loss": 0.8176, + "step": 479 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.026928186416626, + "learning_rate": 4.872910188691657e-06, + "loss": 0.8797, + "step": 480 + }, + { + "epoch": 0.7531315240083507, + "grad_norm": 0.9312179088592529, + "learning_rate": 4.872246178246776e-06, + "loss": 0.7747, + "step": 481 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0676255226135254, + "learning_rate": 4.871580483159725e-06, + "loss": 0.7801, + "step": 482 + }, + { + "epoch": 0.7562630480167014, + "grad_norm": 1.066056489944458, + "learning_rate": 4.870913103903245e-06, + "loss": 0.7837, + "step": 483 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9388124346733093, + "learning_rate": 4.87024404095127e-06, + "loss": 0.7801, + "step": 484 + }, + { + "epoch": 0.7593945720250522, + "grad_norm": 0.9879791140556335, + "learning_rate": 4.8695732947789356e-06, + "loss": 0.7849, + "step": 485 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0002318620681763, + "learning_rate": 4.868900865862569e-06, + "loss": 0.8219, + "step": 486 + }, + { + "epoch": 0.7625260960334029, + "grad_norm": 0.9599125385284424, + "learning_rate": 4.8682267546796945e-06, + "loss": 0.8888, + "step": 487 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 0.9099973440170288, + "learning_rate": 4.867550961709027e-06, + "loss": 0.7585, + "step": 488 + }, + { + "epoch": 0.7656576200417536, + "grad_norm": 0.9715955257415771, + "learning_rate": 4.866873487430481e-06, + "loss": 0.8237, + "step": 489 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 1.0194706916809082, + "learning_rate": 4.86619433232516e-06, + "loss": 0.7708, + "step": 490 + }, + { + "epoch": 0.7687891440501043, + "grad_norm": 1.044005036354065, + "learning_rate": 4.865513496875367e-06, + "loss": 0.8, + "step": 491 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.9776361584663391, + "learning_rate": 4.8648309815645915e-06, + "loss": 0.7696, + "step": 492 + }, + { + "epoch": 0.7719206680584552, + "grad_norm": 0.9213030934333801, + "learning_rate": 4.864146786877521e-06, + "loss": 0.8558, + "step": 493 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.9361440539360046, + "learning_rate": 4.8634609133000336e-06, + "loss": 0.7851, + "step": 494 + }, + { + "epoch": 0.7750521920668059, + "grad_norm": 1.0199540853500366, + "learning_rate": 4.8627733613192e-06, + "loss": 0.8229, + "step": 495 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 1.0244604349136353, + "learning_rate": 4.862084131423283e-06, + "loss": 0.772, + "step": 496 + }, + { + "epoch": 0.7781837160751566, + "grad_norm": 0.9414368271827698, + "learning_rate": 4.861393224101736e-06, + "loss": 0.7579, + "step": 497 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9198965430259705, + "learning_rate": 4.860700639845205e-06, + "loss": 0.855, + "step": 498 + }, + { + "epoch": 0.7813152400835073, + "grad_norm": 0.9398712515830994, + "learning_rate": 4.8600063791455275e-06, + "loss": 0.7824, + "step": 499 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.9544969201087952, + "learning_rate": 4.8593104424957275e-06, + "loss": 0.8104, + "step": 500 + }, + { + "epoch": 0.784446764091858, + "grad_norm": 0.9522370100021362, + "learning_rate": 4.858612830390024e-06, + "loss": 0.813, + "step": 501 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.9830234050750732, + "learning_rate": 4.857913543323823e-06, + "loss": 0.8656, + "step": 502 + }, + { + "epoch": 0.7875782881002088, + "grad_norm": 1.0527008771896362, + "learning_rate": 4.8572125817937195e-06, + "loss": 0.8417, + "step": 503 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 0.9772782921791077, + "learning_rate": 4.856509946297501e-06, + "loss": 0.7983, + "step": 504 + }, + { + "epoch": 0.7907098121085595, + "grad_norm": 0.9250578880310059, + "learning_rate": 4.855805637334141e-06, + "loss": 0.7391, + "step": 505 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9792859554290771, + "learning_rate": 4.855099655403802e-06, + "loss": 0.7753, + "step": 506 + }, + { + "epoch": 0.7938413361169102, + "grad_norm": 0.8876479864120483, + "learning_rate": 4.854392001007834e-06, + "loss": 0.8249, + "step": 507 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.9336771368980408, + "learning_rate": 4.853682674648775e-06, + "loss": 0.8462, + "step": 508 + }, + { + "epoch": 0.7969728601252609, + "grad_norm": 0.9732388854026794, + "learning_rate": 4.852971676830352e-06, + "loss": 0.8791, + "step": 509 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.9135870337486267, + "learning_rate": 4.852259008057476e-06, + "loss": 0.7279, + "step": 510 + }, + { + "epoch": 0.8001043841336117, + "grad_norm": 0.9782464504241943, + "learning_rate": 4.8515446688362476e-06, + "loss": 0.7797, + "step": 511 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 0.9571824073791504, + "learning_rate": 4.85082865967395e-06, + "loss": 0.8349, + "step": 512 + }, + { + "epoch": 0.8032359081419624, + "grad_norm": 1.011786937713623, + "learning_rate": 4.850110981079057e-06, + "loss": 0.8057, + "step": 513 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 0.9418787956237793, + "learning_rate": 4.8493916335612255e-06, + "loss": 0.8592, + "step": 514 + }, + { + "epoch": 0.8063674321503131, + "grad_norm": 0.9292604923248291, + "learning_rate": 4.848670617631296e-06, + "loss": 0.7471, + "step": 515 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.9303188323974609, + "learning_rate": 4.847947933801296e-06, + "loss": 0.8768, + "step": 516 + }, + { + "epoch": 0.8094989561586639, + "grad_norm": 0.9897359609603882, + "learning_rate": 4.847223582584437e-06, + "loss": 0.8343, + "step": 517 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.9695891737937927, + "learning_rate": 4.846497564495114e-06, + "loss": 0.8318, + "step": 518 + }, + { + "epoch": 0.8126304801670147, + "grad_norm": 1.0175273418426514, + "learning_rate": 4.845769880048906e-06, + "loss": 0.9137, + "step": 519 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9497922658920288, + "learning_rate": 4.845040529762577e-06, + "loss": 0.6549, + "step": 520 + }, + { + "epoch": 0.8157620041753654, + "grad_norm": 1.0647574663162231, + "learning_rate": 4.844309514154071e-06, + "loss": 0.8243, + "step": 521 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.9710965156555176, + "learning_rate": 4.843576833742517e-06, + "loss": 0.7864, + "step": 522 + }, + { + "epoch": 0.8188935281837161, + "grad_norm": 0.9881441593170166, + "learning_rate": 4.842842489048225e-06, + "loss": 0.7799, + "step": 523 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 0.9455597400665283, + "learning_rate": 4.842106480592687e-06, + "loss": 0.7943, + "step": 524 + }, + { + "epoch": 0.8220250521920668, + "grad_norm": 0.9948442578315735, + "learning_rate": 4.841368808898577e-06, + "loss": 0.8587, + "step": 525 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 0.9959186315536499, + "learning_rate": 4.84062947448975e-06, + "loss": 0.8477, + "step": 526 + }, + { + "epoch": 0.8251565762004175, + "grad_norm": 0.9791848659515381, + "learning_rate": 4.839888477891243e-06, + "loss": 0.7568, + "step": 527 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 1.1499426364898682, + "learning_rate": 4.839145819629269e-06, + "loss": 0.8296, + "step": 528 + }, + { + "epoch": 0.8282881002087683, + "grad_norm": 0.9269506335258484, + "learning_rate": 4.8384015002312265e-06, + "loss": 0.7935, + "step": 529 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.970565140247345, + "learning_rate": 4.837655520225692e-06, + "loss": 0.8138, + "step": 530 + }, + { + "epoch": 0.831419624217119, + "grad_norm": 0.9852119088172913, + "learning_rate": 4.836907880142418e-06, + "loss": 0.7909, + "step": 531 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.927129328250885, + "learning_rate": 4.836158580512339e-06, + "loss": 0.8105, + "step": 532 + }, + { + "epoch": 0.8345511482254697, + "grad_norm": 1.0711781978607178, + "learning_rate": 4.835407621867569e-06, + "loss": 0.8241, + "step": 533 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 1.0472421646118164, + "learning_rate": 4.834655004741397e-06, + "loss": 0.7923, + "step": 534 + }, + { + "epoch": 0.8376826722338204, + "grad_norm": 0.9761579632759094, + "learning_rate": 4.833900729668293e-06, + "loss": 0.8203, + "step": 535 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 1.0559637546539307, + "learning_rate": 4.833144797183902e-06, + "loss": 0.8915, + "step": 536 + }, + { + "epoch": 0.8408141962421712, + "grad_norm": 0.9951052069664001, + "learning_rate": 4.832387207825045e-06, + "loss": 0.8253, + "step": 537 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.894079864025116, + "learning_rate": 4.831627962129725e-06, + "loss": 0.7479, + "step": 538 + }, + { + "epoch": 0.843945720250522, + "grad_norm": 0.9832531213760376, + "learning_rate": 4.830867060637115e-06, + "loss": 0.8484, + "step": 539 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0477591753005981, + "learning_rate": 4.8301045038875665e-06, + "loss": 0.8052, + "step": 540 + }, + { + "epoch": 0.8470772442588727, + "grad_norm": 0.9388875365257263, + "learning_rate": 4.829340292422607e-06, + "loss": 0.7998, + "step": 541 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 0.8975532054901123, + "learning_rate": 4.8285744267849395e-06, + "loss": 0.848, + "step": 542 + }, + { + "epoch": 0.8502087682672234, + "grad_norm": 0.9211076498031616, + "learning_rate": 4.8278069075184396e-06, + "loss": 0.8337, + "step": 543 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.970134437084198, + "learning_rate": 4.827037735168159e-06, + "loss": 0.7843, + "step": 544 + }, + { + "epoch": 0.8533402922755741, + "grad_norm": 0.9714353084564209, + "learning_rate": 4.826266910280322e-06, + "loss": 0.6991, + "step": 545 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 0.8607689142227173, + "learning_rate": 4.8254944334023265e-06, + "loss": 0.7314, + "step": 546 + }, + { + "epoch": 0.8564718162839249, + "grad_norm": 0.9843030571937561, + "learning_rate": 4.824720305082747e-06, + "loss": 0.7497, + "step": 547 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 0.9638474583625793, + "learning_rate": 4.823944525871324e-06, + "loss": 0.776, + "step": 548 + }, + { + "epoch": 0.8596033402922756, + "grad_norm": 0.9342365860939026, + "learning_rate": 4.823167096318979e-06, + "loss": 0.77, + "step": 549 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9608579874038696, + "learning_rate": 4.822388016977797e-06, + "loss": 0.8281, + "step": 550 + }, + { + "epoch": 0.8627348643006263, + "grad_norm": 0.8996383547782898, + "learning_rate": 4.8216072884010395e-06, + "loss": 0.7519, + "step": 551 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 0.9922810196876526, + "learning_rate": 4.820824911143139e-06, + "loss": 0.8714, + "step": 552 + }, + { + "epoch": 0.865866388308977, + "grad_norm": 0.9861842393875122, + "learning_rate": 4.820040885759697e-06, + "loss": 0.7998, + "step": 553 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 1.013168215751648, + "learning_rate": 4.819255212807486e-06, + "loss": 0.8291, + "step": 554 + }, + { + "epoch": 0.8689979123173278, + "grad_norm": 0.9695484638214111, + "learning_rate": 4.818467892844451e-06, + "loss": 0.8257, + "step": 555 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 1.0018948316574097, + "learning_rate": 4.817678926429702e-06, + "loss": 0.7551, + "step": 556 + }, + { + "epoch": 0.8721294363256785, + "grad_norm": 1.0259422063827515, + "learning_rate": 4.816888314123521e-06, + "loss": 0.8582, + "step": 557 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9045076966285706, + "learning_rate": 4.81609605648736e-06, + "loss": 0.811, + "step": 558 + }, + { + "epoch": 0.8752609603340292, + "grad_norm": 0.9112316370010376, + "learning_rate": 4.815302154083837e-06, + "loss": 0.818, + "step": 559 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0060032606124878, + "learning_rate": 4.814506607476739e-06, + "loss": 0.766, + "step": 560 + }, + { + "epoch": 0.8783924843423799, + "grad_norm": 0.9126970171928406, + "learning_rate": 4.813709417231021e-06, + "loss": 0.8903, + "step": 561 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.0825906991958618, + "learning_rate": 4.812910583912804e-06, + "loss": 0.8466, + "step": 562 + }, + { + "epoch": 0.8815240083507306, + "grad_norm": 0.9530975818634033, + "learning_rate": 4.812110108089377e-06, + "loss": 0.8163, + "step": 563 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.9566405415534973, + "learning_rate": 4.8113079903291955e-06, + "loss": 0.8244, + "step": 564 + }, + { + "epoch": 0.8846555323590815, + "grad_norm": 0.919980525970459, + "learning_rate": 4.81050423120188e-06, + "loss": 0.8335, + "step": 565 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.9123868346214294, + "learning_rate": 4.809698831278217e-06, + "loss": 0.7712, + "step": 566 + }, + { + "epoch": 0.8877870563674322, + "grad_norm": 0.9589663743972778, + "learning_rate": 4.8088917911301595e-06, + "loss": 0.8467, + "step": 567 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 0.9052798748016357, + "learning_rate": 4.808083111330823e-06, + "loss": 0.8078, + "step": 568 + }, + { + "epoch": 0.8909185803757829, + "grad_norm": 0.9250494241714478, + "learning_rate": 4.807272792454489e-06, + "loss": 0.8536, + "step": 569 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.8451298475265503, + "learning_rate": 4.806460835076603e-06, + "loss": 0.7752, + "step": 570 + }, + { + "epoch": 0.8940501043841336, + "grad_norm": 1.0324240922927856, + "learning_rate": 4.805647239773773e-06, + "loss": 0.7871, + "step": 571 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 0.9338528513908386, + "learning_rate": 4.804832007123771e-06, + "loss": 0.8082, + "step": 572 + }, + { + "epoch": 0.8971816283924844, + "grad_norm": 1.0080739259719849, + "learning_rate": 4.804015137705531e-06, + "loss": 0.7636, + "step": 573 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9817036986351013, + "learning_rate": 4.803196632099152e-06, + "loss": 0.8599, + "step": 574 + }, + { + "epoch": 0.9003131524008351, + "grad_norm": 0.9284219145774841, + "learning_rate": 4.80237649088589e-06, + "loss": 0.8152, + "step": 575 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.8922317028045654, + "learning_rate": 4.801554714648166e-06, + "loss": 0.8146, + "step": 576 + }, + { + "epoch": 0.9034446764091858, + "grad_norm": 0.9465014338493347, + "learning_rate": 4.800731303969565e-06, + "loss": 0.7663, + "step": 577 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.0325525999069214, + "learning_rate": 4.799906259434824e-06, + "loss": 0.8458, + "step": 578 + }, + { + "epoch": 0.9065762004175365, + "grad_norm": 0.9366307258605957, + "learning_rate": 4.7990795816298485e-06, + "loss": 0.7992, + "step": 579 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 0.9923847317695618, + "learning_rate": 4.7982512711416995e-06, + "loss": 0.7256, + "step": 580 + }, + { + "epoch": 0.9097077244258872, + "grad_norm": 1.0387133359909058, + "learning_rate": 4.7974213285586e-06, + "loss": 0.7766, + "step": 581 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.9617043733596802, + "learning_rate": 4.79658975446993e-06, + "loss": 0.7602, + "step": 582 + }, + { + "epoch": 0.912839248434238, + "grad_norm": 0.9530353546142578, + "learning_rate": 4.795756549466229e-06, + "loss": 0.7922, + "step": 583 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.9305918216705322, + "learning_rate": 4.794921714139194e-06, + "loss": 0.8377, + "step": 584 + }, + { + "epoch": 0.9159707724425887, + "grad_norm": 0.899860143661499, + "learning_rate": 4.794085249081682e-06, + "loss": 0.8343, + "step": 585 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.9024940729141235, + "learning_rate": 4.793247154887703e-06, + "loss": 0.8271, + "step": 586 + }, + { + "epoch": 0.9191022964509394, + "grad_norm": 0.980869472026825, + "learning_rate": 4.7924074321524286e-06, + "loss": 0.7468, + "step": 587 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.0506541728973389, + "learning_rate": 4.791566081472185e-06, + "loss": 0.9207, + "step": 588 + }, + { + "epoch": 0.9222338204592901, + "grad_norm": 1.0723917484283447, + "learning_rate": 4.790723103444454e-06, + "loss": 0.7615, + "step": 589 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 1.0455081462860107, + "learning_rate": 4.789878498667873e-06, + "loss": 0.8115, + "step": 590 + }, + { + "epoch": 0.925365344467641, + "grad_norm": 0.9729639291763306, + "learning_rate": 4.789032267742234e-06, + "loss": 0.8622, + "step": 591 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.9695239067077637, + "learning_rate": 4.788184411268488e-06, + "loss": 0.8165, + "step": 592 + }, + { + "epoch": 0.9284968684759917, + "grad_norm": 1.0099278688430786, + "learning_rate": 4.7873349298487345e-06, + "loss": 0.759, + "step": 593 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 0.9377149939537048, + "learning_rate": 4.786483824086231e-06, + "loss": 0.8323, + "step": 594 + }, + { + "epoch": 0.9316283924843424, + "grad_norm": 0.8885690569877625, + "learning_rate": 4.785631094585387e-06, + "loss": 0.8448, + "step": 595 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 0.9133430123329163, + "learning_rate": 4.784776741951766e-06, + "loss": 0.8135, + "step": 596 + }, + { + "epoch": 0.9347599164926931, + "grad_norm": 1.015851378440857, + "learning_rate": 4.783920766792082e-06, + "loss": 0.8464, + "step": 597 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.8883825540542603, + "learning_rate": 4.783063169714203e-06, + "loss": 0.7445, + "step": 598 + }, + { + "epoch": 0.9378914405010439, + "grad_norm": 0.9872002005577087, + "learning_rate": 4.782203951327149e-06, + "loss": 0.8293, + "step": 599 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 1.0305813550949097, + "learning_rate": 4.781343112241091e-06, + "loss": 0.8606, + "step": 600 + }, + { + "epoch": 0.9410229645093946, + "grad_norm": 0.9667104482650757, + "learning_rate": 4.780480653067351e-06, + "loss": 0.8057, + "step": 601 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 1.067185640335083, + "learning_rate": 4.7796165744184e-06, + "loss": 0.849, + "step": 602 + }, + { + "epoch": 0.9441544885177453, + "grad_norm": 0.9017406105995178, + "learning_rate": 4.778750876907862e-06, + "loss": 0.7927, + "step": 603 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.9827666878700256, + "learning_rate": 4.777883561150508e-06, + "loss": 0.8087, + "step": 604 + }, + { + "epoch": 0.947286012526096, + "grad_norm": 0.9627811908721924, + "learning_rate": 4.77701462776226e-06, + "loss": 0.8488, + "step": 605 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0321338176727295, + "learning_rate": 4.776144077360186e-06, + "loss": 0.8289, + "step": 606 + }, + { + "epoch": 0.9504175365344467, + "grad_norm": 1.0012633800506592, + "learning_rate": 4.775271910562508e-06, + "loss": 0.7184, + "step": 607 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.9186056852340698, + "learning_rate": 4.7743981279885875e-06, + "loss": 0.8727, + "step": 608 + }, + { + "epoch": 0.9535490605427975, + "grad_norm": 0.910044252872467, + "learning_rate": 4.773522730258942e-06, + "loss": 0.8455, + "step": 609 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.9342944025993347, + "learning_rate": 4.77264571799523e-06, + "loss": 0.7548, + "step": 610 + }, + { + "epoch": 0.9566805845511482, + "grad_norm": 0.9609628319740295, + "learning_rate": 4.771767091820259e-06, + "loss": 0.8294, + "step": 611 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9708508253097534, + "learning_rate": 4.770886852357983e-06, + "loss": 0.7857, + "step": 612 + }, + { + "epoch": 0.9598121085594989, + "grad_norm": 1.0910807847976685, + "learning_rate": 4.7700050002334995e-06, + "loss": 0.8162, + "step": 613 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 0.9703771471977234, + "learning_rate": 4.769121536073054e-06, + "loss": 0.7205, + "step": 614 + }, + { + "epoch": 0.9629436325678496, + "grad_norm": 1.027938961982727, + "learning_rate": 4.768236460504035e-06, + "loss": 0.7641, + "step": 615 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 1.0003989934921265, + "learning_rate": 4.767349774154974e-06, + "loss": 0.7588, + "step": 616 + }, + { + "epoch": 0.9660751565762005, + "grad_norm": 0.9572570323944092, + "learning_rate": 4.766461477655552e-06, + "loss": 0.8274, + "step": 617 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.977287769317627, + "learning_rate": 4.765571571636586e-06, + "loss": 0.7856, + "step": 618 + }, + { + "epoch": 0.9692066805845512, + "grad_norm": 0.9582562446594238, + "learning_rate": 4.764680056730042e-06, + "loss": 0.8102, + "step": 619 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.9842986464500427, + "learning_rate": 4.763786933569025e-06, + "loss": 0.7515, + "step": 620 + }, + { + "epoch": 0.9723382045929019, + "grad_norm": 0.9955743551254272, + "learning_rate": 4.762892202787782e-06, + "loss": 0.7398, + "step": 621 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 1.0011107921600342, + "learning_rate": 4.761995865021706e-06, + "loss": 0.8182, + "step": 622 + }, + { + "epoch": 0.9754697286012526, + "grad_norm": 0.9586448669433594, + "learning_rate": 4.761097920907325e-06, + "loss": 0.8434, + "step": 623 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.979123055934906, + "learning_rate": 4.760198371082312e-06, + "loss": 0.7954, + "step": 624 + }, + { + "epoch": 0.9786012526096033, + "grad_norm": 1.0081250667572021, + "learning_rate": 4.75929721618548e-06, + "loss": 0.7931, + "step": 625 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 1.027185320854187, + "learning_rate": 4.75839445685678e-06, + "loss": 0.7339, + "step": 626 + }, + { + "epoch": 0.9817327766179541, + "grad_norm": 1.0553215742111206, + "learning_rate": 4.757490093737303e-06, + "loss": 0.7639, + "step": 627 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 0.8578981757164001, + "learning_rate": 4.75658412746928e-06, + "loss": 0.8128, + "step": 628 + }, + { + "epoch": 0.9848643006263048, + "grad_norm": 1.0143071413040161, + "learning_rate": 4.75567655869608e-06, + "loss": 0.8169, + "step": 629 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 0.9743554592132568, + "learning_rate": 4.754767388062208e-06, + "loss": 0.8179, + "step": 630 + }, + { + "epoch": 0.9879958246346555, + "grad_norm": 1.0276602506637573, + "learning_rate": 4.7538566162133105e-06, + "loss": 0.8009, + "step": 631 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.9296634793281555, + "learning_rate": 4.752944243796167e-06, + "loss": 0.8262, + "step": 632 + }, + { + "epoch": 0.9911273486430062, + "grad_norm": 0.9216791987419128, + "learning_rate": 4.752030271458696e-06, + "loss": 0.8929, + "step": 633 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 0.9272832274436951, + "learning_rate": 4.751114699849954e-06, + "loss": 0.7396, + "step": 634 + }, + { + "epoch": 0.994258872651357, + "grad_norm": 0.9121203422546387, + "learning_rate": 4.750197529620127e-06, + "loss": 0.8311, + "step": 635 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.999600887298584, + "learning_rate": 4.7492787614205425e-06, + "loss": 0.8125, + "step": 636 + }, + { + "epoch": 0.9973903966597077, + "grad_norm": 1.012978196144104, + "learning_rate": 4.748358395903661e-06, + "loss": 0.7966, + "step": 637 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.9767918586730957, + "learning_rate": 4.747436433723075e-06, + "loss": 0.8024, + "step": 638 + }, + { + "epoch": 1.0015657620041754, + "grad_norm": 1.8302738666534424, + "learning_rate": 4.7465128755335135e-06, + "loss": 1.6427, + "step": 639 + }, + { + "epoch": 1.0031315240083507, + "grad_norm": 0.8704890012741089, + "learning_rate": 4.7455877219908386e-06, + "loss": 0.7766, + "step": 640 + }, + { + "epoch": 1.004697286012526, + "grad_norm": 0.9848434925079346, + "learning_rate": 4.744660973752044e-06, + "loss": 0.8314, + "step": 641 + }, + { + "epoch": 1.0062630480167014, + "grad_norm": 1.054824709892273, + "learning_rate": 4.743732631475258e-06, + "loss": 0.797, + "step": 642 + }, + { + "epoch": 1.0078288100208768, + "grad_norm": 0.9286114573478699, + "learning_rate": 4.742802695819737e-06, + "loss": 0.764, + "step": 643 + }, + { + "epoch": 1.0093945720250521, + "grad_norm": 0.8625781536102295, + "learning_rate": 4.7418711674458735e-06, + "loss": 0.6691, + "step": 644 + }, + { + "epoch": 1.0109603340292275, + "grad_norm": 0.8710452914237976, + "learning_rate": 4.740938047015188e-06, + "loss": 0.7277, + "step": 645 + }, + { + "epoch": 1.0125260960334028, + "grad_norm": 0.8873130679130554, + "learning_rate": 4.740003335190334e-06, + "loss": 0.6531, + "step": 646 + }, + { + "epoch": 1.0140918580375784, + "grad_norm": 0.9777164459228516, + "learning_rate": 4.7390670326350925e-06, + "loss": 0.8047, + "step": 647 + }, + { + "epoch": 1.0156576200417538, + "grad_norm": 0.886025071144104, + "learning_rate": 4.738129140014377e-06, + "loss": 0.715, + "step": 648 + }, + { + "epoch": 1.017223382045929, + "grad_norm": 0.9825332164764404, + "learning_rate": 4.737189657994226e-06, + "loss": 0.7783, + "step": 649 + }, + { + "epoch": 1.0187891440501045, + "grad_norm": 0.9331428408622742, + "learning_rate": 4.736248587241812e-06, + "loss": 0.7598, + "step": 650 + }, + { + "epoch": 1.0203549060542798, + "grad_norm": 0.9406535625457764, + "learning_rate": 4.735305928425431e-06, + "loss": 0.7989, + "step": 651 + }, + { + "epoch": 1.0219206680584552, + "grad_norm": 1.0644745826721191, + "learning_rate": 4.734361682214511e-06, + "loss": 0.8135, + "step": 652 + }, + { + "epoch": 1.0234864300626305, + "grad_norm": 0.9037052392959595, + "learning_rate": 4.733415849279603e-06, + "loss": 0.7051, + "step": 653 + }, + { + "epoch": 1.0250521920668059, + "grad_norm": 1.000563144683838, + "learning_rate": 4.7324684302923865e-06, + "loss": 0.7427, + "step": 654 + }, + { + "epoch": 1.0266179540709812, + "grad_norm": 0.9181354641914368, + "learning_rate": 4.731519425925671e-06, + "loss": 0.7236, + "step": 655 + }, + { + "epoch": 1.0281837160751566, + "grad_norm": 1.0283046960830688, + "learning_rate": 4.730568836853384e-06, + "loss": 0.7998, + "step": 656 + }, + { + "epoch": 1.029749478079332, + "grad_norm": 0.8692315220832825, + "learning_rate": 4.729616663750587e-06, + "loss": 0.7974, + "step": 657 + }, + { + "epoch": 1.0313152400835073, + "grad_norm": 0.928920567035675, + "learning_rate": 4.72866290729346e-06, + "loss": 0.7199, + "step": 658 + }, + { + "epoch": 1.0328810020876826, + "grad_norm": 0.9066957235336304, + "learning_rate": 4.727707568159311e-06, + "loss": 0.791, + "step": 659 + }, + { + "epoch": 1.034446764091858, + "grad_norm": 1.0086729526519775, + "learning_rate": 4.726750647026569e-06, + "loss": 0.7113, + "step": 660 + }, + { + "epoch": 1.0360125260960333, + "grad_norm": 0.931856632232666, + "learning_rate": 4.72579214457479e-06, + "loss": 0.7977, + "step": 661 + }, + { + "epoch": 1.0375782881002087, + "grad_norm": 0.9191203713417053, + "learning_rate": 4.724832061484651e-06, + "loss": 0.7979, + "step": 662 + }, + { + "epoch": 1.039144050104384, + "grad_norm": 0.9614707827568054, + "learning_rate": 4.72387039843795e-06, + "loss": 0.6994, + "step": 663 + }, + { + "epoch": 1.0407098121085594, + "grad_norm": 1.014823317527771, + "learning_rate": 4.72290715611761e-06, + "loss": 0.7539, + "step": 664 + }, + { + "epoch": 1.042275574112735, + "grad_norm": 0.9177424311637878, + "learning_rate": 4.721942335207673e-06, + "loss": 0.6625, + "step": 665 + }, + { + "epoch": 1.0438413361169103, + "grad_norm": 1.0253031253814697, + "learning_rate": 4.720975936393305e-06, + "loss": 0.7747, + "step": 666 + }, + { + "epoch": 1.0454070981210857, + "grad_norm": 1.0004568099975586, + "learning_rate": 4.720007960360788e-06, + "loss": 0.7008, + "step": 667 + }, + { + "epoch": 1.046972860125261, + "grad_norm": 1.1344349384307861, + "learning_rate": 4.719038407797529e-06, + "loss": 0.7493, + "step": 668 + }, + { + "epoch": 1.0485386221294364, + "grad_norm": 0.9944799542427063, + "learning_rate": 4.718067279392052e-06, + "loss": 0.7823, + "step": 669 + }, + { + "epoch": 1.0501043841336117, + "grad_norm": 0.9469346404075623, + "learning_rate": 4.717094575833999e-06, + "loss": 0.7419, + "step": 670 + }, + { + "epoch": 1.051670146137787, + "grad_norm": 1.015339732170105, + "learning_rate": 4.716120297814133e-06, + "loss": 0.7168, + "step": 671 + }, + { + "epoch": 1.0532359081419624, + "grad_norm": 0.9544686675071716, + "learning_rate": 4.715144446024333e-06, + "loss": 0.791, + "step": 672 + }, + { + "epoch": 1.0548016701461378, + "grad_norm": 1.0285652875900269, + "learning_rate": 4.714167021157599e-06, + "loss": 0.7649, + "step": 673 + }, + { + "epoch": 1.0563674321503131, + "grad_norm": 0.9366872906684875, + "learning_rate": 4.713188023908044e-06, + "loss": 0.7524, + "step": 674 + }, + { + "epoch": 1.0579331941544885, + "grad_norm": 1.0348883867263794, + "learning_rate": 4.712207454970899e-06, + "loss": 0.6965, + "step": 675 + }, + { + "epoch": 1.0594989561586639, + "grad_norm": 0.8716043829917908, + "learning_rate": 4.711225315042513e-06, + "loss": 0.6264, + "step": 676 + }, + { + "epoch": 1.0610647181628392, + "grad_norm": 1.0421313047409058, + "learning_rate": 4.710241604820348e-06, + "loss": 0.7548, + "step": 677 + }, + { + "epoch": 1.0626304801670146, + "grad_norm": 1.0886270999908447, + "learning_rate": 4.709256325002983e-06, + "loss": 0.6352, + "step": 678 + }, + { + "epoch": 1.06419624217119, + "grad_norm": 0.9752539396286011, + "learning_rate": 4.708269476290111e-06, + "loss": 0.7764, + "step": 679 + }, + { + "epoch": 1.0657620041753653, + "grad_norm": 1.0658339262008667, + "learning_rate": 4.7072810593825385e-06, + "loss": 0.766, + "step": 680 + }, + { + "epoch": 1.0673277661795406, + "grad_norm": 1.0299617052078247, + "learning_rate": 4.706291074982187e-06, + "loss": 0.7558, + "step": 681 + }, + { + "epoch": 1.068893528183716, + "grad_norm": 0.9171326160430908, + "learning_rate": 4.705299523792089e-06, + "loss": 0.7084, + "step": 682 + }, + { + "epoch": 1.0704592901878915, + "grad_norm": 0.9924243688583374, + "learning_rate": 4.7043064065163925e-06, + "loss": 0.8393, + "step": 683 + }, + { + "epoch": 1.072025052192067, + "grad_norm": 1.0243877172470093, + "learning_rate": 4.703311723860356e-06, + "loss": 0.7755, + "step": 684 + }, + { + "epoch": 1.0735908141962422, + "grad_norm": 1.0019137859344482, + "learning_rate": 4.702315476530349e-06, + "loss": 0.7254, + "step": 685 + }, + { + "epoch": 1.0751565762004176, + "grad_norm": 0.8541996479034424, + "learning_rate": 4.701317665233853e-06, + "loss": 0.7389, + "step": 686 + }, + { + "epoch": 1.076722338204593, + "grad_norm": 1.1119922399520874, + "learning_rate": 4.700318290679462e-06, + "loss": 0.7803, + "step": 687 + }, + { + "epoch": 1.0782881002087683, + "grad_norm": 1.078452706336975, + "learning_rate": 4.699317353576876e-06, + "loss": 0.756, + "step": 688 + }, + { + "epoch": 1.0798538622129437, + "grad_norm": 1.0079457759857178, + "learning_rate": 4.698314854636908e-06, + "loss": 0.7365, + "step": 689 + }, + { + "epoch": 1.081419624217119, + "grad_norm": 0.931423544883728, + "learning_rate": 4.6973107945714795e-06, + "loss": 0.7881, + "step": 690 + }, + { + "epoch": 1.0829853862212944, + "grad_norm": 1.0135142803192139, + "learning_rate": 4.696305174093619e-06, + "loss": 0.7884, + "step": 691 + }, + { + "epoch": 1.0845511482254697, + "grad_norm": 1.1194610595703125, + "learning_rate": 4.695297993917465e-06, + "loss": 0.7969, + "step": 692 + }, + { + "epoch": 1.086116910229645, + "grad_norm": 0.9972780346870422, + "learning_rate": 4.6942892547582625e-06, + "loss": 0.5873, + "step": 693 + }, + { + "epoch": 1.0876826722338204, + "grad_norm": 1.0711641311645508, + "learning_rate": 4.693278957332364e-06, + "loss": 0.7273, + "step": 694 + }, + { + "epoch": 1.0892484342379958, + "grad_norm": 0.9708715081214905, + "learning_rate": 4.6922671023572305e-06, + "loss": 0.6786, + "step": 695 + }, + { + "epoch": 1.0908141962421711, + "grad_norm": 0.9773644804954529, + "learning_rate": 4.691253690551424e-06, + "loss": 0.7839, + "step": 696 + }, + { + "epoch": 1.0923799582463465, + "grad_norm": 1.039009690284729, + "learning_rate": 4.690238722634618e-06, + "loss": 0.6807, + "step": 697 + }, + { + "epoch": 1.0939457202505218, + "grad_norm": 1.140777349472046, + "learning_rate": 4.689222199327586e-06, + "loss": 0.7506, + "step": 698 + }, + { + "epoch": 1.0955114822546972, + "grad_norm": 0.9779882431030273, + "learning_rate": 4.688204121352211e-06, + "loss": 0.7732, + "step": 699 + }, + { + "epoch": 1.0970772442588728, + "grad_norm": 1.0412248373031616, + "learning_rate": 4.687184489431476e-06, + "loss": 0.7034, + "step": 700 + }, + { + "epoch": 1.0986430062630481, + "grad_norm": 0.9802400469779968, + "learning_rate": 4.686163304289469e-06, + "loss": 0.729, + "step": 701 + }, + { + "epoch": 1.1002087682672235, + "grad_norm": 1.0531584024429321, + "learning_rate": 4.6851405666513804e-06, + "loss": 0.7635, + "step": 702 + }, + { + "epoch": 1.1017745302713988, + "grad_norm": 0.983923077583313, + "learning_rate": 4.684116277243505e-06, + "loss": 0.7062, + "step": 703 + }, + { + "epoch": 1.1033402922755742, + "grad_norm": 0.9886301159858704, + "learning_rate": 4.683090436793238e-06, + "loss": 0.7299, + "step": 704 + }, + { + "epoch": 1.1049060542797495, + "grad_norm": 1.0247817039489746, + "learning_rate": 4.682063046029075e-06, + "loss": 0.729, + "step": 705 + }, + { + "epoch": 1.1064718162839249, + "grad_norm": 0.9860950708389282, + "learning_rate": 4.681034105680616e-06, + "loss": 0.7392, + "step": 706 + }, + { + "epoch": 1.1080375782881002, + "grad_norm": 0.9543380737304688, + "learning_rate": 4.680003616478556e-06, + "loss": 0.7061, + "step": 707 + }, + { + "epoch": 1.1096033402922756, + "grad_norm": 1.0253225564956665, + "learning_rate": 4.678971579154698e-06, + "loss": 0.6577, + "step": 708 + }, + { + "epoch": 1.111169102296451, + "grad_norm": 0.9013616442680359, + "learning_rate": 4.677937994441935e-06, + "loss": 0.7422, + "step": 709 + }, + { + "epoch": 1.1127348643006263, + "grad_norm": 0.9716471433639526, + "learning_rate": 4.676902863074268e-06, + "loss": 0.7353, + "step": 710 + }, + { + "epoch": 1.1143006263048016, + "grad_norm": 1.040898084640503, + "learning_rate": 4.675866185786787e-06, + "loss": 0.7396, + "step": 711 + }, + { + "epoch": 1.115866388308977, + "grad_norm": 0.9931808114051819, + "learning_rate": 4.674827963315688e-06, + "loss": 0.7354, + "step": 712 + }, + { + "epoch": 1.1174321503131524, + "grad_norm": 1.020512580871582, + "learning_rate": 4.6737881963982605e-06, + "loss": 0.7366, + "step": 713 + }, + { + "epoch": 1.1189979123173277, + "grad_norm": 1.0201045274734497, + "learning_rate": 4.67274688577289e-06, + "loss": 0.7907, + "step": 714 + }, + { + "epoch": 1.120563674321503, + "grad_norm": 1.0163429975509644, + "learning_rate": 4.671704032179061e-06, + "loss": 0.7452, + "step": 715 + }, + { + "epoch": 1.1221294363256784, + "grad_norm": 0.9667377471923828, + "learning_rate": 4.670659636357352e-06, + "loss": 0.7983, + "step": 716 + }, + { + "epoch": 1.1236951983298538, + "grad_norm": 1.0788301229476929, + "learning_rate": 4.669613699049436e-06, + "loss": 0.7861, + "step": 717 + }, + { + "epoch": 1.1252609603340291, + "grad_norm": 1.0512142181396484, + "learning_rate": 4.668566220998082e-06, + "loss": 0.7582, + "step": 718 + }, + { + "epoch": 1.1268267223382047, + "grad_norm": 0.9225794076919556, + "learning_rate": 4.667517202947154e-06, + "loss": 0.7801, + "step": 719 + }, + { + "epoch": 1.12839248434238, + "grad_norm": 1.1193747520446777, + "learning_rate": 4.666466645641607e-06, + "loss": 0.8073, + "step": 720 + }, + { + "epoch": 1.1299582463465554, + "grad_norm": 0.934438943862915, + "learning_rate": 4.6654145498274915e-06, + "loss": 0.774, + "step": 721 + }, + { + "epoch": 1.1315240083507307, + "grad_norm": 1.0614968538284302, + "learning_rate": 4.664360916251949e-06, + "loss": 0.7216, + "step": 722 + }, + { + "epoch": 1.133089770354906, + "grad_norm": 1.0807424783706665, + "learning_rate": 4.663305745663212e-06, + "loss": 0.6434, + "step": 723 + }, + { + "epoch": 1.1346555323590815, + "grad_norm": 0.9515354633331299, + "learning_rate": 4.66224903881061e-06, + "loss": 0.7336, + "step": 724 + }, + { + "epoch": 1.1362212943632568, + "grad_norm": 0.9852489233016968, + "learning_rate": 4.661190796444557e-06, + "loss": 0.8156, + "step": 725 + }, + { + "epoch": 1.1377870563674322, + "grad_norm": 1.0071501731872559, + "learning_rate": 4.66013101931656e-06, + "loss": 0.8836, + "step": 726 + }, + { + "epoch": 1.1393528183716075, + "grad_norm": 0.9622960686683655, + "learning_rate": 4.6590697081792176e-06, + "loss": 0.7631, + "step": 727 + }, + { + "epoch": 1.1409185803757829, + "grad_norm": 1.0588105916976929, + "learning_rate": 4.6580068637862144e-06, + "loss": 0.7874, + "step": 728 + }, + { + "epoch": 1.1424843423799582, + "grad_norm": 0.9920447468757629, + "learning_rate": 4.656942486892326e-06, + "loss": 0.7161, + "step": 729 + }, + { + "epoch": 1.1440501043841336, + "grad_norm": 1.0036561489105225, + "learning_rate": 4.655876578253416e-06, + "loss": 0.7116, + "step": 730 + }, + { + "epoch": 1.145615866388309, + "grad_norm": 1.0208396911621094, + "learning_rate": 4.654809138626436e-06, + "loss": 0.818, + "step": 731 + }, + { + "epoch": 1.1471816283924843, + "grad_norm": 1.1095491647720337, + "learning_rate": 4.653740168769424e-06, + "loss": 0.7319, + "step": 732 + }, + { + "epoch": 1.1487473903966596, + "grad_norm": 0.8151328563690186, + "learning_rate": 4.652669669441505e-06, + "loss": 0.774, + "step": 733 + }, + { + "epoch": 1.150313152400835, + "grad_norm": 1.0747915506362915, + "learning_rate": 4.651597641402891e-06, + "loss": 0.8275, + "step": 734 + }, + { + "epoch": 1.1518789144050103, + "grad_norm": 0.9265272617340088, + "learning_rate": 4.650524085414878e-06, + "loss": 0.6946, + "step": 735 + }, + { + "epoch": 1.153444676409186, + "grad_norm": 0.9531672596931458, + "learning_rate": 4.649449002239849e-06, + "loss": 0.7899, + "step": 736 + }, + { + "epoch": 1.1550104384133613, + "grad_norm": 1.0496262311935425, + "learning_rate": 4.648372392641269e-06, + "loss": 0.8237, + "step": 737 + }, + { + "epoch": 1.1565762004175366, + "grad_norm": 0.8967193961143494, + "learning_rate": 4.64729425738369e-06, + "loss": 0.736, + "step": 738 + }, + { + "epoch": 1.158141962421712, + "grad_norm": 1.061038851737976, + "learning_rate": 4.646214597232745e-06, + "loss": 0.7308, + "step": 739 + }, + { + "epoch": 1.1597077244258873, + "grad_norm": 0.9556939601898193, + "learning_rate": 4.64513341295515e-06, + "loss": 0.7526, + "step": 740 + }, + { + "epoch": 1.1612734864300627, + "grad_norm": 1.000093936920166, + "learning_rate": 4.6440507053187075e-06, + "loss": 0.7488, + "step": 741 + }, + { + "epoch": 1.162839248434238, + "grad_norm": 0.9552261233329773, + "learning_rate": 4.642966475092294e-06, + "loss": 0.8291, + "step": 742 + }, + { + "epoch": 1.1644050104384134, + "grad_norm": 1.0215847492218018, + "learning_rate": 4.641880723045875e-06, + "loss": 0.756, + "step": 743 + }, + { + "epoch": 1.1659707724425887, + "grad_norm": 1.0543124675750732, + "learning_rate": 4.640793449950493e-06, + "loss": 0.7668, + "step": 744 + }, + { + "epoch": 1.167536534446764, + "grad_norm": 0.9648609161376953, + "learning_rate": 4.639704656578269e-06, + "loss": 0.6531, + "step": 745 + }, + { + "epoch": 1.1691022964509394, + "grad_norm": 1.0117427110671997, + "learning_rate": 4.638614343702408e-06, + "loss": 0.7084, + "step": 746 + }, + { + "epoch": 1.1706680584551148, + "grad_norm": 0.9475041627883911, + "learning_rate": 4.6375225120971915e-06, + "loss": 0.8578, + "step": 747 + }, + { + "epoch": 1.1722338204592901, + "grad_norm": 1.0122299194335938, + "learning_rate": 4.6364291625379785e-06, + "loss": 0.7541, + "step": 748 + }, + { + "epoch": 1.1737995824634655, + "grad_norm": 0.9487535357475281, + "learning_rate": 4.635334295801208e-06, + "loss": 0.7203, + "step": 749 + }, + { + "epoch": 1.1753653444676408, + "grad_norm": 1.0628844499588013, + "learning_rate": 4.6342379126643965e-06, + "loss": 0.7532, + "step": 750 + }, + { + "epoch": 1.1769311064718162, + "grad_norm": 0.9647781848907471, + "learning_rate": 4.633140013906136e-06, + "loss": 0.5918, + "step": 751 + }, + { + "epoch": 1.1784968684759916, + "grad_norm": 0.9011932611465454, + "learning_rate": 4.632040600306095e-06, + "loss": 0.7395, + "step": 752 + }, + { + "epoch": 1.1800626304801671, + "grad_norm": 1.0524142980575562, + "learning_rate": 4.630939672645018e-06, + "loss": 0.7487, + "step": 753 + }, + { + "epoch": 1.1816283924843423, + "grad_norm": 1.165077567100525, + "learning_rate": 4.6298372317047265e-06, + "loss": 0.6817, + "step": 754 + }, + { + "epoch": 1.1831941544885178, + "grad_norm": 0.9301269054412842, + "learning_rate": 4.628733278268113e-06, + "loss": 0.8446, + "step": 755 + }, + { + "epoch": 1.1847599164926932, + "grad_norm": 0.9762471318244934, + "learning_rate": 4.627627813119147e-06, + "loss": 0.7527, + "step": 756 + }, + { + "epoch": 1.1863256784968685, + "grad_norm": 0.8634734153747559, + "learning_rate": 4.626520837042871e-06, + "loss": 0.7177, + "step": 757 + }, + { + "epoch": 1.187891440501044, + "grad_norm": 1.0592188835144043, + "learning_rate": 4.625412350825399e-06, + "loss": 0.7293, + "step": 758 + }, + { + "epoch": 1.1894572025052192, + "grad_norm": 1.1318559646606445, + "learning_rate": 4.62430235525392e-06, + "loss": 0.6592, + "step": 759 + }, + { + "epoch": 1.1910229645093946, + "grad_norm": 0.9689339399337769, + "learning_rate": 4.623190851116692e-06, + "loss": 0.7667, + "step": 760 + }, + { + "epoch": 1.19258872651357, + "grad_norm": 0.9965885281562805, + "learning_rate": 4.6220778392030464e-06, + "loss": 0.6982, + "step": 761 + }, + { + "epoch": 1.1941544885177453, + "grad_norm": 1.1428629159927368, + "learning_rate": 4.6209633203033846e-06, + "loss": 0.7342, + "step": 762 + }, + { + "epoch": 1.1957202505219207, + "grad_norm": 0.901360809803009, + "learning_rate": 4.6198472952091785e-06, + "loss": 0.8371, + "step": 763 + }, + { + "epoch": 1.197286012526096, + "grad_norm": 0.9882062077522278, + "learning_rate": 4.618729764712969e-06, + "loss": 0.6882, + "step": 764 + }, + { + "epoch": 1.1988517745302714, + "grad_norm": 1.0692517757415771, + "learning_rate": 4.617610729608366e-06, + "loss": 0.7327, + "step": 765 + }, + { + "epoch": 1.2004175365344467, + "grad_norm": 1.1056455373764038, + "learning_rate": 4.616490190690049e-06, + "loss": 0.781, + "step": 766 + }, + { + "epoch": 1.201983298538622, + "grad_norm": 0.9928885698318481, + "learning_rate": 4.615368148753765e-06, + "loss": 0.7463, + "step": 767 + }, + { + "epoch": 1.2035490605427974, + "grad_norm": 1.0006698369979858, + "learning_rate": 4.614244604596327e-06, + "loss": 0.6994, + "step": 768 + }, + { + "epoch": 1.2051148225469728, + "grad_norm": 0.907508909702301, + "learning_rate": 4.613119559015616e-06, + "loss": 0.7117, + "step": 769 + }, + { + "epoch": 1.2066805845511483, + "grad_norm": 1.1037627458572388, + "learning_rate": 4.61199301281058e-06, + "loss": 0.772, + "step": 770 + }, + { + "epoch": 1.2082463465553235, + "grad_norm": 0.9474150538444519, + "learning_rate": 4.610864966781231e-06, + "loss": 0.7601, + "step": 771 + }, + { + "epoch": 1.209812108559499, + "grad_norm": 1.003667950630188, + "learning_rate": 4.609735421728647e-06, + "loss": 0.7141, + "step": 772 + }, + { + "epoch": 1.2113778705636744, + "grad_norm": 0.9724787473678589, + "learning_rate": 4.608604378454971e-06, + "loss": 0.7841, + "step": 773 + }, + { + "epoch": 1.2129436325678498, + "grad_norm": 1.0460760593414307, + "learning_rate": 4.607471837763409e-06, + "loss": 0.7563, + "step": 774 + }, + { + "epoch": 1.2145093945720251, + "grad_norm": 0.9798458218574524, + "learning_rate": 4.606337800458231e-06, + "loss": 0.6812, + "step": 775 + }, + { + "epoch": 1.2160751565762005, + "grad_norm": 1.1637914180755615, + "learning_rate": 4.605202267344768e-06, + "loss": 0.7571, + "step": 776 + }, + { + "epoch": 1.2176409185803758, + "grad_norm": 0.9384608864784241, + "learning_rate": 4.6040652392294175e-06, + "loss": 0.7842, + "step": 777 + }, + { + "epoch": 1.2192066805845512, + "grad_norm": 1.0490702390670776, + "learning_rate": 4.602926716919633e-06, + "loss": 0.7157, + "step": 778 + }, + { + "epoch": 1.2207724425887265, + "grad_norm": 1.131402611732483, + "learning_rate": 4.601786701223934e-06, + "loss": 0.7724, + "step": 779 + }, + { + "epoch": 1.2223382045929019, + "grad_norm": 1.0159722566604614, + "learning_rate": 4.600645192951898e-06, + "loss": 0.7666, + "step": 780 + }, + { + "epoch": 1.2239039665970772, + "grad_norm": 1.039806604385376, + "learning_rate": 4.599502192914164e-06, + "loss": 0.7845, + "step": 781 + }, + { + "epoch": 1.2254697286012526, + "grad_norm": 1.0722614526748657, + "learning_rate": 4.598357701922428e-06, + "loss": 0.782, + "step": 782 + }, + { + "epoch": 1.227035490605428, + "grad_norm": 0.9443033933639526, + "learning_rate": 4.5972117207894465e-06, + "loss": 0.6573, + "step": 783 + }, + { + "epoch": 1.2286012526096033, + "grad_norm": 0.9738026857376099, + "learning_rate": 4.596064250329035e-06, + "loss": 0.7475, + "step": 784 + }, + { + "epoch": 1.2301670146137786, + "grad_norm": 0.9932008385658264, + "learning_rate": 4.594915291356063e-06, + "loss": 0.757, + "step": 785 + }, + { + "epoch": 1.231732776617954, + "grad_norm": 1.0612564086914062, + "learning_rate": 4.5937648446864625e-06, + "loss": 0.7824, + "step": 786 + }, + { + "epoch": 1.2332985386221293, + "grad_norm": 1.0500893592834473, + "learning_rate": 4.592612911137218e-06, + "loss": 0.7646, + "step": 787 + }, + { + "epoch": 1.2348643006263047, + "grad_norm": 0.9607855081558228, + "learning_rate": 4.591459491526371e-06, + "loss": 0.754, + "step": 788 + }, + { + "epoch": 1.2364300626304803, + "grad_norm": 1.0871951580047607, + "learning_rate": 4.590304586673018e-06, + "loss": 0.7226, + "step": 789 + }, + { + "epoch": 1.2379958246346556, + "grad_norm": 1.0030092000961304, + "learning_rate": 4.589148197397311e-06, + "loss": 0.7357, + "step": 790 + }, + { + "epoch": 1.239561586638831, + "grad_norm": 1.0297991037368774, + "learning_rate": 4.587990324520455e-06, + "loss": 0.7724, + "step": 791 + }, + { + "epoch": 1.2411273486430063, + "grad_norm": 1.0192303657531738, + "learning_rate": 4.586830968864711e-06, + "loss": 0.6921, + "step": 792 + }, + { + "epoch": 1.2426931106471817, + "grad_norm": 0.9617307186126709, + "learning_rate": 4.585670131253389e-06, + "loss": 0.7443, + "step": 793 + }, + { + "epoch": 1.244258872651357, + "grad_norm": 1.0355238914489746, + "learning_rate": 4.584507812510856e-06, + "loss": 0.7741, + "step": 794 + }, + { + "epoch": 1.2458246346555324, + "grad_norm": 0.9938539862632751, + "learning_rate": 4.583344013462526e-06, + "loss": 0.7372, + "step": 795 + }, + { + "epoch": 1.2473903966597077, + "grad_norm": 1.0573335886001587, + "learning_rate": 4.582178734934869e-06, + "loss": 0.7191, + "step": 796 + }, + { + "epoch": 1.248956158663883, + "grad_norm": 0.963642954826355, + "learning_rate": 4.581011977755402e-06, + "loss": 0.6283, + "step": 797 + }, + { + "epoch": 1.2505219206680585, + "grad_norm": 1.0119287967681885, + "learning_rate": 4.579843742752694e-06, + "loss": 0.7599, + "step": 798 + }, + { + "epoch": 1.2520876826722338, + "grad_norm": 1.0628082752227783, + "learning_rate": 4.578674030756364e-06, + "loss": 0.7741, + "step": 799 + }, + { + "epoch": 1.2536534446764092, + "grad_norm": 1.0510241985321045, + "learning_rate": 4.577502842597077e-06, + "loss": 0.7615, + "step": 800 + }, + { + "epoch": 1.2552192066805845, + "grad_norm": 1.1180559396743774, + "learning_rate": 4.5763301791065474e-06, + "loss": 0.7119, + "step": 801 + }, + { + "epoch": 1.2567849686847599, + "grad_norm": 1.0198049545288086, + "learning_rate": 4.575156041117541e-06, + "loss": 0.7487, + "step": 802 + }, + { + "epoch": 1.2583507306889352, + "grad_norm": 1.018902063369751, + "learning_rate": 4.5739804294638655e-06, + "loss": 0.7461, + "step": 803 + }, + { + "epoch": 1.2599164926931106, + "grad_norm": 0.9230141639709473, + "learning_rate": 4.572803344980378e-06, + "loss": 0.7295, + "step": 804 + }, + { + "epoch": 1.261482254697286, + "grad_norm": 1.0011619329452515, + "learning_rate": 4.571624788502981e-06, + "loss": 0.7325, + "step": 805 + }, + { + "epoch": 1.2630480167014615, + "grad_norm": 0.9569756388664246, + "learning_rate": 4.570444760868622e-06, + "loss": 0.7494, + "step": 806 + }, + { + "epoch": 1.2646137787056366, + "grad_norm": 0.9860491156578064, + "learning_rate": 4.569263262915293e-06, + "loss": 0.6536, + "step": 807 + }, + { + "epoch": 1.2661795407098122, + "grad_norm": 1.0986074209213257, + "learning_rate": 4.568080295482032e-06, + "loss": 0.7337, + "step": 808 + }, + { + "epoch": 1.2677453027139876, + "grad_norm": 1.0782687664031982, + "learning_rate": 4.566895859408916e-06, + "loss": 0.7193, + "step": 809 + }, + { + "epoch": 1.269311064718163, + "grad_norm": 1.1424410343170166, + "learning_rate": 4.565709955537071e-06, + "loss": 0.6153, + "step": 810 + }, + { + "epoch": 1.2708768267223383, + "grad_norm": 0.9178900718688965, + "learning_rate": 4.564522584708662e-06, + "loss": 0.7802, + "step": 811 + }, + { + "epoch": 1.2724425887265136, + "grad_norm": 1.0006638765335083, + "learning_rate": 4.563333747766896e-06, + "loss": 0.7882, + "step": 812 + }, + { + "epoch": 1.274008350730689, + "grad_norm": 1.0139516592025757, + "learning_rate": 4.56214344555602e-06, + "loss": 0.6562, + "step": 813 + }, + { + "epoch": 1.2755741127348643, + "grad_norm": 1.0360203981399536, + "learning_rate": 4.560951678921324e-06, + "loss": 0.7951, + "step": 814 + }, + { + "epoch": 1.2771398747390397, + "grad_norm": 1.042457103729248, + "learning_rate": 4.559758448709138e-06, + "loss": 0.7572, + "step": 815 + }, + { + "epoch": 1.278705636743215, + "grad_norm": 1.0056653022766113, + "learning_rate": 4.558563755766827e-06, + "loss": 0.7679, + "step": 816 + }, + { + "epoch": 1.2802713987473904, + "grad_norm": 0.9838577508926392, + "learning_rate": 4.5573676009428e-06, + "loss": 0.7245, + "step": 817 + }, + { + "epoch": 1.2818371607515657, + "grad_norm": 1.0474631786346436, + "learning_rate": 4.556169985086503e-06, + "loss": 0.7823, + "step": 818 + }, + { + "epoch": 1.283402922755741, + "grad_norm": 1.0177569389343262, + "learning_rate": 4.554970909048417e-06, + "loss": 0.7599, + "step": 819 + }, + { + "epoch": 1.2849686847599164, + "grad_norm": 1.0120493173599243, + "learning_rate": 4.553770373680062e-06, + "loss": 0.7842, + "step": 820 + }, + { + "epoch": 1.2865344467640918, + "grad_norm": 1.1045918464660645, + "learning_rate": 4.5525683798339946e-06, + "loss": 0.7075, + "step": 821 + }, + { + "epoch": 1.2881002087682671, + "grad_norm": 1.097714900970459, + "learning_rate": 4.551364928363805e-06, + "loss": 0.7571, + "step": 822 + }, + { + "epoch": 1.2896659707724427, + "grad_norm": 0.9125490188598633, + "learning_rate": 4.550160020124121e-06, + "loss": 0.7626, + "step": 823 + }, + { + "epoch": 1.2912317327766178, + "grad_norm": 1.1127262115478516, + "learning_rate": 4.548953655970604e-06, + "loss": 0.6618, + "step": 824 + }, + { + "epoch": 1.2927974947807934, + "grad_norm": 1.0372540950775146, + "learning_rate": 4.547745836759949e-06, + "loss": 0.7467, + "step": 825 + }, + { + "epoch": 1.2943632567849686, + "grad_norm": 1.0158213376998901, + "learning_rate": 4.546536563349885e-06, + "loss": 0.6421, + "step": 826 + }, + { + "epoch": 1.2959290187891441, + "grad_norm": 1.0057018995285034, + "learning_rate": 4.545325836599171e-06, + "loss": 0.7252, + "step": 827 + }, + { + "epoch": 1.2974947807933195, + "grad_norm": 1.0552566051483154, + "learning_rate": 4.544113657367604e-06, + "loss": 0.6645, + "step": 828 + }, + { + "epoch": 1.2990605427974948, + "grad_norm": 1.0798696279525757, + "learning_rate": 4.542900026516006e-06, + "loss": 0.7627, + "step": 829 + }, + { + "epoch": 1.3006263048016702, + "grad_norm": 1.0450047254562378, + "learning_rate": 4.541684944906233e-06, + "loss": 0.705, + "step": 830 + }, + { + "epoch": 1.3021920668058455, + "grad_norm": 0.9975817203521729, + "learning_rate": 4.540468413401171e-06, + "loss": 0.7639, + "step": 831 + }, + { + "epoch": 1.303757828810021, + "grad_norm": 0.9081487059593201, + "learning_rate": 4.539250432864736e-06, + "loss": 0.8129, + "step": 832 + }, + { + "epoch": 1.3053235908141962, + "grad_norm": 1.228684902191162, + "learning_rate": 4.538031004161872e-06, + "loss": 0.6984, + "step": 833 + }, + { + "epoch": 1.3068893528183716, + "grad_norm": 1.0577794313430786, + "learning_rate": 4.536810128158552e-06, + "loss": 0.7436, + "step": 834 + }, + { + "epoch": 1.308455114822547, + "grad_norm": 1.1270350217819214, + "learning_rate": 4.535587805721777e-06, + "loss": 0.7589, + "step": 835 + }, + { + "epoch": 1.3100208768267223, + "grad_norm": 1.1473027467727661, + "learning_rate": 4.5343640377195766e-06, + "loss": 0.7988, + "step": 836 + }, + { + "epoch": 1.3115866388308977, + "grad_norm": 0.9700750112533569, + "learning_rate": 4.533138825021001e-06, + "loss": 0.7388, + "step": 837 + }, + { + "epoch": 1.313152400835073, + "grad_norm": 0.9596260786056519, + "learning_rate": 4.531912168496135e-06, + "loss": 0.8177, + "step": 838 + }, + { + "epoch": 1.3147181628392484, + "grad_norm": 0.9888957738876343, + "learning_rate": 4.530684069016081e-06, + "loss": 0.809, + "step": 839 + }, + { + "epoch": 1.316283924843424, + "grad_norm": 1.219224214553833, + "learning_rate": 4.529454527452972e-06, + "loss": 0.7761, + "step": 840 + }, + { + "epoch": 1.317849686847599, + "grad_norm": 0.9730296730995178, + "learning_rate": 4.5282235446799624e-06, + "loss": 0.8044, + "step": 841 + }, + { + "epoch": 1.3194154488517746, + "grad_norm": 1.1046907901763916, + "learning_rate": 4.526991121571228e-06, + "loss": 0.7452, + "step": 842 + }, + { + "epoch": 1.3209812108559498, + "grad_norm": 0.9707220196723938, + "learning_rate": 4.525757259001972e-06, + "loss": 0.7112, + "step": 843 + }, + { + "epoch": 1.3225469728601253, + "grad_norm": 1.0085593461990356, + "learning_rate": 4.524521957848416e-06, + "loss": 0.6266, + "step": 844 + }, + { + "epoch": 1.3241127348643007, + "grad_norm": 1.131439447402954, + "learning_rate": 4.5232852189878055e-06, + "loss": 0.7393, + "step": 845 + }, + { + "epoch": 1.325678496868476, + "grad_norm": 1.1238782405853271, + "learning_rate": 4.522047043298406e-06, + "loss": 0.7541, + "step": 846 + }, + { + "epoch": 1.3272442588726514, + "grad_norm": 0.9149152636528015, + "learning_rate": 4.520807431659504e-06, + "loss": 0.8162, + "step": 847 + }, + { + "epoch": 1.3288100208768268, + "grad_norm": 1.0428732633590698, + "learning_rate": 4.519566384951403e-06, + "loss": 0.7511, + "step": 848 + }, + { + "epoch": 1.3303757828810021, + "grad_norm": 1.0263090133666992, + "learning_rate": 4.518323904055432e-06, + "loss": 0.7482, + "step": 849 + }, + { + "epoch": 1.3319415448851775, + "grad_norm": 1.113810420036316, + "learning_rate": 4.517079989853931e-06, + "loss": 0.797, + "step": 850 + }, + { + "epoch": 1.3335073068893528, + "grad_norm": 1.0090736150741577, + "learning_rate": 4.515834643230263e-06, + "loss": 0.688, + "step": 851 + }, + { + "epoch": 1.3350730688935282, + "grad_norm": 1.012824296951294, + "learning_rate": 4.514587865068806e-06, + "loss": 0.6683, + "step": 852 + }, + { + "epoch": 1.3366388308977035, + "grad_norm": 0.9442676305770874, + "learning_rate": 4.513339656254954e-06, + "loss": 0.7812, + "step": 853 + }, + { + "epoch": 1.3382045929018789, + "grad_norm": 1.0870879888534546, + "learning_rate": 4.512090017675119e-06, + "loss": 0.6943, + "step": 854 + }, + { + "epoch": 1.3397703549060542, + "grad_norm": 0.9352210164070129, + "learning_rate": 4.510838950216727e-06, + "loss": 0.7571, + "step": 855 + }, + { + "epoch": 1.3413361169102296, + "grad_norm": 1.173621654510498, + "learning_rate": 4.509586454768221e-06, + "loss": 0.6985, + "step": 856 + }, + { + "epoch": 1.342901878914405, + "grad_norm": 0.9790180921554565, + "learning_rate": 4.508332532219053e-06, + "loss": 0.7132, + "step": 857 + }, + { + "epoch": 1.3444676409185803, + "grad_norm": 1.0999163389205933, + "learning_rate": 4.507077183459694e-06, + "loss": 0.7858, + "step": 858 + }, + { + "epoch": 1.3460334029227559, + "grad_norm": 1.018182635307312, + "learning_rate": 4.505820409381625e-06, + "loss": 0.712, + "step": 859 + }, + { + "epoch": 1.347599164926931, + "grad_norm": 1.1015701293945312, + "learning_rate": 4.504562210877338e-06, + "loss": 0.8004, + "step": 860 + }, + { + "epoch": 1.3491649269311066, + "grad_norm": 1.0872273445129395, + "learning_rate": 4.50330258884034e-06, + "loss": 0.7478, + "step": 861 + }, + { + "epoch": 1.350730688935282, + "grad_norm": 1.1098262071609497, + "learning_rate": 4.502041544165148e-06, + "loss": 0.7894, + "step": 862 + }, + { + "epoch": 1.3522964509394573, + "grad_norm": 1.0535573959350586, + "learning_rate": 4.500779077747285e-06, + "loss": 0.7185, + "step": 863 + }, + { + "epoch": 1.3538622129436326, + "grad_norm": 1.0424474477767944, + "learning_rate": 4.499515190483291e-06, + "loss": 0.748, + "step": 864 + }, + { + "epoch": 1.355427974947808, + "grad_norm": 1.005065679550171, + "learning_rate": 4.498249883270708e-06, + "loss": 0.7948, + "step": 865 + }, + { + "epoch": 1.3569937369519833, + "grad_norm": 1.0052721500396729, + "learning_rate": 4.49698315700809e-06, + "loss": 0.7619, + "step": 866 + }, + { + "epoch": 1.3585594989561587, + "grad_norm": 1.0953692197799683, + "learning_rate": 4.495715012594999e-06, + "loss": 0.7363, + "step": 867 + }, + { + "epoch": 1.360125260960334, + "grad_norm": 1.1238093376159668, + "learning_rate": 4.494445450932003e-06, + "loss": 0.6576, + "step": 868 + }, + { + "epoch": 1.3616910229645094, + "grad_norm": 0.8546008467674255, + "learning_rate": 4.493174472920676e-06, + "loss": 0.7513, + "step": 869 + }, + { + "epoch": 1.3632567849686847, + "grad_norm": 1.0863126516342163, + "learning_rate": 4.4919020794635995e-06, + "loss": 0.6886, + "step": 870 + }, + { + "epoch": 1.36482254697286, + "grad_norm": 0.9943820834159851, + "learning_rate": 4.490628271464359e-06, + "loss": 0.7253, + "step": 871 + }, + { + "epoch": 1.3663883089770354, + "grad_norm": 0.9377979636192322, + "learning_rate": 4.4893530498275436e-06, + "loss": 0.7026, + "step": 872 + }, + { + "epoch": 1.3679540709812108, + "grad_norm": 1.100799798965454, + "learning_rate": 4.488076415458749e-06, + "loss": 0.719, + "step": 873 + }, + { + "epoch": 1.3695198329853862, + "grad_norm": 1.015074610710144, + "learning_rate": 4.486798369264572e-06, + "loss": 0.7513, + "step": 874 + }, + { + "epoch": 1.3710855949895615, + "grad_norm": 1.047530174255371, + "learning_rate": 4.485518912152611e-06, + "loss": 0.7757, + "step": 875 + }, + { + "epoch": 1.372651356993737, + "grad_norm": 1.0264610052108765, + "learning_rate": 4.484238045031471e-06, + "loss": 0.7712, + "step": 876 + }, + { + "epoch": 1.3742171189979122, + "grad_norm": 0.9848430156707764, + "learning_rate": 4.482955768810752e-06, + "loss": 0.7569, + "step": 877 + }, + { + "epoch": 1.3757828810020878, + "grad_norm": 1.070816159248352, + "learning_rate": 4.48167208440106e-06, + "loss": 0.7857, + "step": 878 + }, + { + "epoch": 1.377348643006263, + "grad_norm": 1.0134745836257935, + "learning_rate": 4.480386992713998e-06, + "loss": 0.7372, + "step": 879 + }, + { + "epoch": 1.3789144050104385, + "grad_norm": 1.0736140012741089, + "learning_rate": 4.47910049466217e-06, + "loss": 0.7794, + "step": 880 + }, + { + "epoch": 1.3804801670146138, + "grad_norm": 1.0163054466247559, + "learning_rate": 4.477812591159176e-06, + "loss": 0.7341, + "step": 881 + }, + { + "epoch": 1.3820459290187892, + "grad_norm": 1.074578881263733, + "learning_rate": 4.476523283119619e-06, + "loss": 0.6924, + "step": 882 + }, + { + "epoch": 1.3836116910229646, + "grad_norm": 1.183667540550232, + "learning_rate": 4.475232571459095e-06, + "loss": 0.7866, + "step": 883 + }, + { + "epoch": 1.38517745302714, + "grad_norm": 1.030827522277832, + "learning_rate": 4.473940457094199e-06, + "loss": 0.7845, + "step": 884 + }, + { + "epoch": 1.3867432150313153, + "grad_norm": 1.0043658018112183, + "learning_rate": 4.47264694094252e-06, + "loss": 0.7399, + "step": 885 + }, + { + "epoch": 1.3883089770354906, + "grad_norm": 1.1135472059249878, + "learning_rate": 4.471352023922645e-06, + "loss": 0.771, + "step": 886 + }, + { + "epoch": 1.389874739039666, + "grad_norm": 1.0338670015335083, + "learning_rate": 4.470055706954154e-06, + "loss": 0.7282, + "step": 887 + }, + { + "epoch": 1.3914405010438413, + "grad_norm": 0.918855607509613, + "learning_rate": 4.468757990957623e-06, + "loss": 0.728, + "step": 888 + }, + { + "epoch": 1.3930062630480167, + "grad_norm": 1.06833016872406, + "learning_rate": 4.467458876854619e-06, + "loss": 0.7012, + "step": 889 + }, + { + "epoch": 1.394572025052192, + "grad_norm": 1.0103005170822144, + "learning_rate": 4.4661583655677045e-06, + "loss": 0.7074, + "step": 890 + }, + { + "epoch": 1.3961377870563674, + "grad_norm": 1.0418543815612793, + "learning_rate": 4.464856458020433e-06, + "loss": 0.6858, + "step": 891 + }, + { + "epoch": 1.3977035490605427, + "grad_norm": 0.9720668196678162, + "learning_rate": 4.463553155137348e-06, + "loss": 0.7024, + "step": 892 + }, + { + "epoch": 1.399269311064718, + "grad_norm": 1.1870561838150024, + "learning_rate": 4.462248457843986e-06, + "loss": 0.8069, + "step": 893 + }, + { + "epoch": 1.4008350730688934, + "grad_norm": 0.9888094663619995, + "learning_rate": 4.460942367066874e-06, + "loss": 0.7281, + "step": 894 + }, + { + "epoch": 1.402400835073069, + "grad_norm": 0.9120670557022095, + "learning_rate": 4.459634883733527e-06, + "loss": 0.7992, + "step": 895 + }, + { + "epoch": 1.4039665970772441, + "grad_norm": 1.0239481925964355, + "learning_rate": 4.45832600877245e-06, + "loss": 0.7312, + "step": 896 + }, + { + "epoch": 1.4055323590814197, + "grad_norm": 0.9379130005836487, + "learning_rate": 4.4570157431131355e-06, + "loss": 0.7019, + "step": 897 + }, + { + "epoch": 1.407098121085595, + "grad_norm": 1.0789109468460083, + "learning_rate": 4.455704087686063e-06, + "loss": 0.8149, + "step": 898 + }, + { + "epoch": 1.4086638830897704, + "grad_norm": 1.0521917343139648, + "learning_rate": 4.454391043422703e-06, + "loss": 0.7873, + "step": 899 + }, + { + "epoch": 1.4102296450939458, + "grad_norm": 0.9582643508911133, + "learning_rate": 4.453076611255507e-06, + "loss": 0.7815, + "step": 900 + }, + { + "epoch": 1.4117954070981211, + "grad_norm": 1.1096739768981934, + "learning_rate": 4.451760792117914e-06, + "loss": 0.7636, + "step": 901 + }, + { + "epoch": 1.4133611691022965, + "grad_norm": 1.043574333190918, + "learning_rate": 4.450443586944349e-06, + "loss": 0.7411, + "step": 902 + }, + { + "epoch": 1.4149269311064718, + "grad_norm": 1.0280033349990845, + "learning_rate": 4.449124996670221e-06, + "loss": 0.6525, + "step": 903 + }, + { + "epoch": 1.4164926931106472, + "grad_norm": 1.0329992771148682, + "learning_rate": 4.44780502223192e-06, + "loss": 0.7133, + "step": 904 + }, + { + "epoch": 1.4180584551148225, + "grad_norm": 1.016168475151062, + "learning_rate": 4.446483664566825e-06, + "loss": 0.7098, + "step": 905 + }, + { + "epoch": 1.4196242171189979, + "grad_norm": 1.2022786140441895, + "learning_rate": 4.4451609246132895e-06, + "loss": 0.7303, + "step": 906 + }, + { + "epoch": 1.4211899791231732, + "grad_norm": 1.0311683416366577, + "learning_rate": 4.443836803310655e-06, + "loss": 0.7475, + "step": 907 + }, + { + "epoch": 1.4227557411273486, + "grad_norm": 1.0073823928833008, + "learning_rate": 4.442511301599241e-06, + "loss": 0.7513, + "step": 908 + }, + { + "epoch": 1.424321503131524, + "grad_norm": 1.0702322721481323, + "learning_rate": 4.441184420420347e-06, + "loss": 0.6918, + "step": 909 + }, + { + "epoch": 1.4258872651356993, + "grad_norm": 1.1416171789169312, + "learning_rate": 4.439856160716254e-06, + "loss": 0.6324, + "step": 910 + }, + { + "epoch": 1.4274530271398747, + "grad_norm": 1.0560253858566284, + "learning_rate": 4.438526523430221e-06, + "loss": 0.749, + "step": 911 + }, + { + "epoch": 1.4290187891440502, + "grad_norm": 1.021008849143982, + "learning_rate": 4.4371955095064835e-06, + "loss": 0.7666, + "step": 912 + }, + { + "epoch": 1.4305845511482254, + "grad_norm": 1.151904821395874, + "learning_rate": 4.435863119890258e-06, + "loss": 0.6904, + "step": 913 + }, + { + "epoch": 1.432150313152401, + "grad_norm": 1.1216702461242676, + "learning_rate": 4.434529355527736e-06, + "loss": 0.7068, + "step": 914 + }, + { + "epoch": 1.433716075156576, + "grad_norm": 1.1130588054656982, + "learning_rate": 4.433194217366085e-06, + "loss": 0.6524, + "step": 915 + }, + { + "epoch": 1.4352818371607516, + "grad_norm": 1.07612144947052, + "learning_rate": 4.431857706353449e-06, + "loss": 0.8078, + "step": 916 + }, + { + "epoch": 1.436847599164927, + "grad_norm": 1.0220447778701782, + "learning_rate": 4.430519823438946e-06, + "loss": 0.6736, + "step": 917 + }, + { + "epoch": 1.4384133611691023, + "grad_norm": 1.0657141208648682, + "learning_rate": 4.429180569572669e-06, + "loss": 0.7457, + "step": 918 + }, + { + "epoch": 1.4399791231732777, + "grad_norm": 1.0646854639053345, + "learning_rate": 4.427839945705684e-06, + "loss": 0.7925, + "step": 919 + }, + { + "epoch": 1.441544885177453, + "grad_norm": 1.0619502067565918, + "learning_rate": 4.426497952790031e-06, + "loss": 0.6968, + "step": 920 + }, + { + "epoch": 1.4431106471816284, + "grad_norm": 0.9684338569641113, + "learning_rate": 4.425154591778722e-06, + "loss": 0.7411, + "step": 921 + }, + { + "epoch": 1.4446764091858038, + "grad_norm": 1.1225167512893677, + "learning_rate": 4.423809863625739e-06, + "loss": 0.7623, + "step": 922 + }, + { + "epoch": 1.446242171189979, + "grad_norm": 0.9310746788978577, + "learning_rate": 4.422463769286036e-06, + "loss": 0.8219, + "step": 923 + }, + { + "epoch": 1.4478079331941545, + "grad_norm": 1.1107957363128662, + "learning_rate": 4.4211163097155375e-06, + "loss": 0.7755, + "step": 924 + }, + { + "epoch": 1.4493736951983298, + "grad_norm": 0.9827596545219421, + "learning_rate": 4.419767485871136e-06, + "loss": 0.7102, + "step": 925 + }, + { + "epoch": 1.4509394572025052, + "grad_norm": 1.1075087785720825, + "learning_rate": 4.418417298710695e-06, + "loss": 0.6399, + "step": 926 + }, + { + "epoch": 1.4525052192066805, + "grad_norm": 0.9575802087783813, + "learning_rate": 4.417065749193045e-06, + "loss": 0.721, + "step": 927 + }, + { + "epoch": 1.4540709812108559, + "grad_norm": 1.0016006231307983, + "learning_rate": 4.415712838277985e-06, + "loss": 0.762, + "step": 928 + }, + { + "epoch": 1.4556367432150314, + "grad_norm": 1.1907106637954712, + "learning_rate": 4.414358566926279e-06, + "loss": 0.7904, + "step": 929 + }, + { + "epoch": 1.4572025052192066, + "grad_norm": 1.0011543035507202, + "learning_rate": 4.413002936099656e-06, + "loss": 0.7179, + "step": 930 + }, + { + "epoch": 1.4587682672233822, + "grad_norm": 0.980684757232666, + "learning_rate": 4.411645946760817e-06, + "loss": 0.6845, + "step": 931 + }, + { + "epoch": 1.4603340292275573, + "grad_norm": 0.9458135366439819, + "learning_rate": 4.4102875998734176e-06, + "loss": 0.6911, + "step": 932 + }, + { + "epoch": 1.4618997912317329, + "grad_norm": 1.020966649055481, + "learning_rate": 4.408927896402087e-06, + "loss": 0.7284, + "step": 933 + }, + { + "epoch": 1.4634655532359082, + "grad_norm": 1.041833519935608, + "learning_rate": 4.407566837312413e-06, + "loss": 0.7451, + "step": 934 + }, + { + "epoch": 1.4650313152400836, + "grad_norm": 0.9076465964317322, + "learning_rate": 4.406204423570946e-06, + "loss": 0.7553, + "step": 935 + }, + { + "epoch": 1.466597077244259, + "grad_norm": 1.1963952779769897, + "learning_rate": 4.404840656145199e-06, + "loss": 0.7437, + "step": 936 + }, + { + "epoch": 1.4681628392484343, + "grad_norm": 1.0181536674499512, + "learning_rate": 4.403475536003648e-06, + "loss": 0.7283, + "step": 937 + }, + { + "epoch": 1.4697286012526096, + "grad_norm": 0.9563003778457642, + "learning_rate": 4.402109064115727e-06, + "loss": 0.7375, + "step": 938 + }, + { + "epoch": 1.471294363256785, + "grad_norm": 1.05129873752594, + "learning_rate": 4.400741241451832e-06, + "loss": 0.6805, + "step": 939 + }, + { + "epoch": 1.4728601252609603, + "grad_norm": 1.0912166833877563, + "learning_rate": 4.399372068983317e-06, + "loss": 0.7076, + "step": 940 + }, + { + "epoch": 1.4744258872651357, + "grad_norm": 1.108500599861145, + "learning_rate": 4.398001547682494e-06, + "loss": 0.6836, + "step": 941 + }, + { + "epoch": 1.475991649269311, + "grad_norm": 1.0741651058197021, + "learning_rate": 4.396629678522636e-06, + "loss": 0.7405, + "step": 942 + }, + { + "epoch": 1.4775574112734864, + "grad_norm": 1.0603930950164795, + "learning_rate": 4.39525646247797e-06, + "loss": 0.7602, + "step": 943 + }, + { + "epoch": 1.4791231732776617, + "grad_norm": 0.9814990162849426, + "learning_rate": 4.39388190052368e-06, + "loss": 0.7324, + "step": 944 + }, + { + "epoch": 1.480688935281837, + "grad_norm": 0.9643695950508118, + "learning_rate": 4.392505993635906e-06, + "loss": 0.7034, + "step": 945 + }, + { + "epoch": 1.4822546972860124, + "grad_norm": 1.0490994453430176, + "learning_rate": 4.391128742791743e-06, + "loss": 0.7358, + "step": 946 + }, + { + "epoch": 1.4838204592901878, + "grad_norm": 0.9223467111587524, + "learning_rate": 4.389750148969244e-06, + "loss": 0.698, + "step": 947 + }, + { + "epoch": 1.4853862212943634, + "grad_norm": 1.1107758283615112, + "learning_rate": 4.388370213147409e-06, + "loss": 0.7264, + "step": 948 + }, + { + "epoch": 1.4869519832985385, + "grad_norm": 1.05939781665802, + "learning_rate": 4.386988936306196e-06, + "loss": 0.7209, + "step": 949 + }, + { + "epoch": 1.488517745302714, + "grad_norm": 1.024686574935913, + "learning_rate": 4.385606319426512e-06, + "loss": 0.7798, + "step": 950 + }, + { + "epoch": 1.4900835073068894, + "grad_norm": 0.9920977354049683, + "learning_rate": 4.3842223634902205e-06, + "loss": 0.8045, + "step": 951 + }, + { + "epoch": 1.4916492693110648, + "grad_norm": 0.950927197933197, + "learning_rate": 4.382837069480131e-06, + "loss": 0.6777, + "step": 952 + }, + { + "epoch": 1.4932150313152401, + "grad_norm": 1.0072081089019775, + "learning_rate": 4.3814504383800056e-06, + "loss": 0.7458, + "step": 953 + }, + { + "epoch": 1.4947807933194155, + "grad_norm": 0.9982030987739563, + "learning_rate": 4.3800624711745546e-06, + "loss": 0.8215, + "step": 954 + }, + { + "epoch": 1.4963465553235908, + "grad_norm": 0.9682126641273499, + "learning_rate": 4.3786731688494396e-06, + "loss": 0.7157, + "step": 955 + }, + { + "epoch": 1.4979123173277662, + "grad_norm": 1.1599830389022827, + "learning_rate": 4.377282532391267e-06, + "loss": 0.8091, + "step": 956 + }, + { + "epoch": 1.4994780793319415, + "grad_norm": 1.0624948740005493, + "learning_rate": 4.3758905627875934e-06, + "loss": 0.7428, + "step": 957 + }, + { + "epoch": 1.501043841336117, + "grad_norm": 1.14694082736969, + "learning_rate": 4.374497261026921e-06, + "loss": 0.7752, + "step": 958 + }, + { + "epoch": 1.5026096033402923, + "grad_norm": 1.0355825424194336, + "learning_rate": 4.373102628098697e-06, + "loss": 0.7348, + "step": 959 + }, + { + "epoch": 1.5041753653444676, + "grad_norm": 1.036026954650879, + "learning_rate": 4.371706664993316e-06, + "loss": 0.7536, + "step": 960 + }, + { + "epoch": 1.505741127348643, + "grad_norm": 1.2319790124893188, + "learning_rate": 4.370309372702116e-06, + "loss": 0.6985, + "step": 961 + }, + { + "epoch": 1.5073068893528183, + "grad_norm": 0.908687949180603, + "learning_rate": 4.36891075221738e-06, + "loss": 0.7173, + "step": 962 + }, + { + "epoch": 1.5088726513569939, + "grad_norm": 1.0482099056243896, + "learning_rate": 4.367510804532331e-06, + "loss": 0.7514, + "step": 963 + }, + { + "epoch": 1.510438413361169, + "grad_norm": 1.0434144735336304, + "learning_rate": 4.36610953064114e-06, + "loss": 0.754, + "step": 964 + }, + { + "epoch": 1.5120041753653446, + "grad_norm": 1.0122450590133667, + "learning_rate": 4.364706931538914e-06, + "loss": 0.7767, + "step": 965 + }, + { + "epoch": 1.5135699373695197, + "grad_norm": 1.0471900701522827, + "learning_rate": 4.363303008221705e-06, + "loss": 0.7337, + "step": 966 + }, + { + "epoch": 1.5151356993736953, + "grad_norm": 1.0691189765930176, + "learning_rate": 4.3618977616865045e-06, + "loss": 0.8047, + "step": 967 + }, + { + "epoch": 1.5167014613778704, + "grad_norm": 1.015114188194275, + "learning_rate": 4.360491192931242e-06, + "loss": 0.726, + "step": 968 + }, + { + "epoch": 1.518267223382046, + "grad_norm": 0.9770560264587402, + "learning_rate": 4.359083302954788e-06, + "loss": 0.7692, + "step": 969 + }, + { + "epoch": 1.5198329853862211, + "grad_norm": 1.079991340637207, + "learning_rate": 4.357674092756951e-06, + "loss": 0.8424, + "step": 970 + }, + { + "epoch": 1.5213987473903967, + "grad_norm": 1.068708896636963, + "learning_rate": 4.356263563338476e-06, + "loss": 0.758, + "step": 971 + }, + { + "epoch": 1.522964509394572, + "grad_norm": 1.1005477905273438, + "learning_rate": 4.354851715701046e-06, + "loss": 0.7209, + "step": 972 + }, + { + "epoch": 1.5245302713987474, + "grad_norm": 0.9870902299880981, + "learning_rate": 4.3534385508472786e-06, + "loss": 0.7272, + "step": 973 + }, + { + "epoch": 1.5260960334029228, + "grad_norm": 0.9616238474845886, + "learning_rate": 4.352024069780729e-06, + "loss": 0.8124, + "step": 974 + }, + { + "epoch": 1.5276617954070981, + "grad_norm": 1.0341622829437256, + "learning_rate": 4.3506082735058855e-06, + "loss": 0.7434, + "step": 975 + }, + { + "epoch": 1.5292275574112735, + "grad_norm": 1.0157302618026733, + "learning_rate": 4.349191163028171e-06, + "loss": 0.7534, + "step": 976 + }, + { + "epoch": 1.5307933194154488, + "grad_norm": 1.041379690170288, + "learning_rate": 4.347772739353939e-06, + "loss": 0.6942, + "step": 977 + }, + { + "epoch": 1.5323590814196242, + "grad_norm": 1.016205072402954, + "learning_rate": 4.346353003490481e-06, + "loss": 0.6707, + "step": 978 + }, + { + "epoch": 1.5339248434237995, + "grad_norm": 1.0034714937210083, + "learning_rate": 4.344931956446017e-06, + "loss": 0.6358, + "step": 979 + }, + { + "epoch": 1.535490605427975, + "grad_norm": 0.9671767354011536, + "learning_rate": 4.343509599229697e-06, + "loss": 0.6333, + "step": 980 + }, + { + "epoch": 1.5370563674321502, + "grad_norm": 1.1384506225585938, + "learning_rate": 4.342085932851605e-06, + "loss": 0.803, + "step": 981 + }, + { + "epoch": 1.5386221294363258, + "grad_norm": 1.023031234741211, + "learning_rate": 4.34066095832275e-06, + "loss": 0.7801, + "step": 982 + }, + { + "epoch": 1.540187891440501, + "grad_norm": 1.0675241947174072, + "learning_rate": 4.339234676655075e-06, + "loss": 0.7335, + "step": 983 + }, + { + "epoch": 1.5417536534446765, + "grad_norm": 1.0397543907165527, + "learning_rate": 4.337807088861447e-06, + "loss": 0.7722, + "step": 984 + }, + { + "epoch": 1.5433194154488517, + "grad_norm": 1.0892809629440308, + "learning_rate": 4.3363781959556635e-06, + "loss": 0.7603, + "step": 985 + }, + { + "epoch": 1.5448851774530272, + "grad_norm": 0.9335037469863892, + "learning_rate": 4.334947998952448e-06, + "loss": 0.7678, + "step": 986 + }, + { + "epoch": 1.5464509394572024, + "grad_norm": 1.0991342067718506, + "learning_rate": 4.333516498867449e-06, + "loss": 0.7461, + "step": 987 + }, + { + "epoch": 1.548016701461378, + "grad_norm": 1.112265944480896, + "learning_rate": 4.332083696717242e-06, + "loss": 0.7593, + "step": 988 + }, + { + "epoch": 1.5495824634655533, + "grad_norm": 0.9525822997093201, + "learning_rate": 4.330649593519325e-06, + "loss": 0.7629, + "step": 989 + }, + { + "epoch": 1.5511482254697286, + "grad_norm": 1.0624464750289917, + "learning_rate": 4.329214190292122e-06, + "loss": 0.7261, + "step": 990 + }, + { + "epoch": 1.552713987473904, + "grad_norm": 1.1969516277313232, + "learning_rate": 4.327777488054979e-06, + "loss": 0.7483, + "step": 991 + }, + { + "epoch": 1.5542797494780793, + "grad_norm": 1.0239863395690918, + "learning_rate": 4.326339487828167e-06, + "loss": 0.7265, + "step": 992 + }, + { + "epoch": 1.5558455114822547, + "grad_norm": 1.0881083011627197, + "learning_rate": 4.324900190632874e-06, + "loss": 0.724, + "step": 993 + }, + { + "epoch": 1.55741127348643, + "grad_norm": 1.098827838897705, + "learning_rate": 4.323459597491214e-06, + "loss": 0.7222, + "step": 994 + }, + { + "epoch": 1.5589770354906054, + "grad_norm": 1.105057716369629, + "learning_rate": 4.322017709426217e-06, + "loss": 0.7827, + "step": 995 + }, + { + "epoch": 1.5605427974947808, + "grad_norm": 0.9830679297447205, + "learning_rate": 4.3205745274618365e-06, + "loss": 0.7316, + "step": 996 + }, + { + "epoch": 1.562108559498956, + "grad_norm": 0.9951156973838806, + "learning_rate": 4.319130052622942e-06, + "loss": 0.7181, + "step": 997 + }, + { + "epoch": 1.5636743215031315, + "grad_norm": 1.0374284982681274, + "learning_rate": 4.317684285935323e-06, + "loss": 0.7355, + "step": 998 + }, + { + "epoch": 1.565240083507307, + "grad_norm": 1.0139906406402588, + "learning_rate": 4.316237228425683e-06, + "loss": 0.7249, + "step": 999 + }, + { + "epoch": 1.5668058455114822, + "grad_norm": 0.9555441737174988, + "learning_rate": 4.3147888811216496e-06, + "loss": 0.7387, + "step": 1000 + }, + { + "epoch": 1.5683716075156577, + "grad_norm": 1.0600265264511108, + "learning_rate": 4.313339245051758e-06, + "loss": 0.7445, + "step": 1001 + }, + { + "epoch": 1.5699373695198329, + "grad_norm": 1.0070478916168213, + "learning_rate": 4.311888321245461e-06, + "loss": 0.7605, + "step": 1002 + }, + { + "epoch": 1.5715031315240084, + "grad_norm": 1.019710898399353, + "learning_rate": 4.31043611073313e-06, + "loss": 0.7262, + "step": 1003 + }, + { + "epoch": 1.5730688935281836, + "grad_norm": 1.0476088523864746, + "learning_rate": 4.308982614546045e-06, + "loss": 0.6766, + "step": 1004 + }, + { + "epoch": 1.5746346555323592, + "grad_norm": 1.0670830011367798, + "learning_rate": 4.307527833716403e-06, + "loss": 0.7465, + "step": 1005 + }, + { + "epoch": 1.5762004175365343, + "grad_norm": 1.1807290315628052, + "learning_rate": 4.30607176927731e-06, + "loss": 0.7032, + "step": 1006 + }, + { + "epoch": 1.5777661795407099, + "grad_norm": 1.0482438802719116, + "learning_rate": 4.3046144222627855e-06, + "loss": 0.7156, + "step": 1007 + }, + { + "epoch": 1.5793319415448852, + "grad_norm": 0.9434391856193542, + "learning_rate": 4.303155793707759e-06, + "loss": 0.7816, + "step": 1008 + }, + { + "epoch": 1.5808977035490606, + "grad_norm": 1.0497010946273804, + "learning_rate": 4.3016958846480706e-06, + "loss": 0.735, + "step": 1009 + }, + { + "epoch": 1.582463465553236, + "grad_norm": 0.955949068069458, + "learning_rate": 4.300234696120469e-06, + "loss": 0.6414, + "step": 1010 + }, + { + "epoch": 1.5840292275574113, + "grad_norm": 1.0462027788162231, + "learning_rate": 4.298772229162613e-06, + "loss": 0.6949, + "step": 1011 + }, + { + "epoch": 1.5855949895615866, + "grad_norm": 1.0634411573410034, + "learning_rate": 4.297308484813067e-06, + "loss": 0.7736, + "step": 1012 + }, + { + "epoch": 1.587160751565762, + "grad_norm": 1.0463027954101562, + "learning_rate": 4.295843464111306e-06, + "loss": 0.6812, + "step": 1013 + }, + { + "epoch": 1.5887265135699373, + "grad_norm": 1.0409252643585205, + "learning_rate": 4.294377168097706e-06, + "loss": 0.7319, + "step": 1014 + }, + { + "epoch": 1.5902922755741127, + "grad_norm": 0.9714667201042175, + "learning_rate": 4.292909597813555e-06, + "loss": 0.7473, + "step": 1015 + }, + { + "epoch": 1.5918580375782883, + "grad_norm": 1.03997802734375, + "learning_rate": 4.291440754301041e-06, + "loss": 0.7261, + "step": 1016 + }, + { + "epoch": 1.5934237995824634, + "grad_norm": 1.063293218612671, + "learning_rate": 4.289970638603258e-06, + "loss": 0.7909, + "step": 1017 + }, + { + "epoch": 1.594989561586639, + "grad_norm": 1.110007643699646, + "learning_rate": 4.288499251764205e-06, + "loss": 0.7, + "step": 1018 + }, + { + "epoch": 1.596555323590814, + "grad_norm": 0.981261134147644, + "learning_rate": 4.2870265948287806e-06, + "loss": 0.7516, + "step": 1019 + }, + { + "epoch": 1.5981210855949897, + "grad_norm": 1.0653973817825317, + "learning_rate": 4.2855526688427875e-06, + "loss": 0.7278, + "step": 1020 + }, + { + "epoch": 1.5996868475991648, + "grad_norm": 0.9212517738342285, + "learning_rate": 4.284077474852928e-06, + "loss": 0.7251, + "step": 1021 + }, + { + "epoch": 1.6012526096033404, + "grad_norm": 1.0887659788131714, + "learning_rate": 4.282601013906808e-06, + "loss": 0.7271, + "step": 1022 + }, + { + "epoch": 1.6028183716075155, + "grad_norm": 1.0753779411315918, + "learning_rate": 4.281123287052928e-06, + "loss": 0.7679, + "step": 1023 + }, + { + "epoch": 1.604384133611691, + "grad_norm": 1.0282115936279297, + "learning_rate": 4.2796442953406935e-06, + "loss": 0.695, + "step": 1024 + }, + { + "epoch": 1.6059498956158664, + "grad_norm": 0.9782281517982483, + "learning_rate": 4.2781640398204036e-06, + "loss": 0.7717, + "step": 1025 + }, + { + "epoch": 1.6075156576200418, + "grad_norm": 1.06228768825531, + "learning_rate": 4.276682521543256e-06, + "loss": 0.8069, + "step": 1026 + }, + { + "epoch": 1.6090814196242171, + "grad_norm": 0.8853397965431213, + "learning_rate": 4.2751997415613465e-06, + "loss": 0.8174, + "step": 1027 + }, + { + "epoch": 1.6106471816283925, + "grad_norm": 1.1525455713272095, + "learning_rate": 4.273715700927666e-06, + "loss": 0.7149, + "step": 1028 + }, + { + "epoch": 1.6122129436325678, + "grad_norm": 0.8743720650672913, + "learning_rate": 4.272230400696101e-06, + "loss": 0.7296, + "step": 1029 + }, + { + "epoch": 1.6137787056367432, + "grad_norm": 1.1136301755905151, + "learning_rate": 4.270743841921431e-06, + "loss": 0.7539, + "step": 1030 + }, + { + "epoch": 1.6153444676409185, + "grad_norm": 0.9830031991004944, + "learning_rate": 4.269256025659332e-06, + "loss": 0.6545, + "step": 1031 + }, + { + "epoch": 1.616910229645094, + "grad_norm": 1.1221113204956055, + "learning_rate": 4.267766952966369e-06, + "loss": 0.6144, + "step": 1032 + }, + { + "epoch": 1.6184759916492695, + "grad_norm": 0.975713849067688, + "learning_rate": 4.266276624900004e-06, + "loss": 0.746, + "step": 1033 + }, + { + "epoch": 1.6200417536534446, + "grad_norm": 1.1131149530410767, + "learning_rate": 4.264785042518587e-06, + "loss": 0.705, + "step": 1034 + }, + { + "epoch": 1.6216075156576202, + "grad_norm": 1.0574990510940552, + "learning_rate": 4.263292206881361e-06, + "loss": 0.7681, + "step": 1035 + }, + { + "epoch": 1.6231732776617953, + "grad_norm": 0.9931074976921082, + "learning_rate": 4.261798119048456e-06, + "loss": 0.7854, + "step": 1036 + }, + { + "epoch": 1.6247390396659709, + "grad_norm": 1.1598386764526367, + "learning_rate": 4.260302780080896e-06, + "loss": 0.7406, + "step": 1037 + }, + { + "epoch": 1.626304801670146, + "grad_norm": 0.9910677671432495, + "learning_rate": 4.258806191040587e-06, + "loss": 0.7107, + "step": 1038 + }, + { + "epoch": 1.6278705636743216, + "grad_norm": 1.1023072004318237, + "learning_rate": 4.2573083529903294e-06, + "loss": 0.7964, + "step": 1039 + }, + { + "epoch": 1.6294363256784967, + "grad_norm": 0.991082489490509, + "learning_rate": 4.255809266993806e-06, + "loss": 0.762, + "step": 1040 + }, + { + "epoch": 1.6310020876826723, + "grad_norm": 0.9281193017959595, + "learning_rate": 4.254308934115589e-06, + "loss": 0.7272, + "step": 1041 + }, + { + "epoch": 1.6325678496868476, + "grad_norm": 1.20292067527771, + "learning_rate": 4.252807355421133e-06, + "loss": 0.7535, + "step": 1042 + }, + { + "epoch": 1.634133611691023, + "grad_norm": 1.0362178087234497, + "learning_rate": 4.251304531976778e-06, + "loss": 0.7524, + "step": 1043 + }, + { + "epoch": 1.6356993736951984, + "grad_norm": 1.0267268419265747, + "learning_rate": 4.249800464849751e-06, + "loss": 0.6588, + "step": 1044 + }, + { + "epoch": 1.6372651356993737, + "grad_norm": 1.0531187057495117, + "learning_rate": 4.248295155108158e-06, + "loss": 0.7328, + "step": 1045 + }, + { + "epoch": 1.638830897703549, + "grad_norm": 1.0008069276809692, + "learning_rate": 4.246788603820989e-06, + "loss": 0.719, + "step": 1046 + }, + { + "epoch": 1.6403966597077244, + "grad_norm": 1.064780831336975, + "learning_rate": 4.245280812058119e-06, + "loss": 0.7388, + "step": 1047 + }, + { + "epoch": 1.6419624217118998, + "grad_norm": 1.06492018699646, + "learning_rate": 4.243771780890298e-06, + "loss": 0.7495, + "step": 1048 + }, + { + "epoch": 1.6435281837160751, + "grad_norm": 0.9472903609275818, + "learning_rate": 4.24226151138916e-06, + "loss": 0.7409, + "step": 1049 + }, + { + "epoch": 1.6450939457202505, + "grad_norm": 1.0612895488739014, + "learning_rate": 4.240750004627217e-06, + "loss": 0.7605, + "step": 1050 + }, + { + "epoch": 1.6466597077244258, + "grad_norm": 0.9634467959403992, + "learning_rate": 4.239237261677863e-06, + "loss": 0.7109, + "step": 1051 + }, + { + "epoch": 1.6482254697286014, + "grad_norm": 1.1347044706344604, + "learning_rate": 4.2377232836153635e-06, + "loss": 0.693, + "step": 1052 + }, + { + "epoch": 1.6497912317327765, + "grad_norm": 1.0814287662506104, + "learning_rate": 4.236208071514865e-06, + "loss": 0.7159, + "step": 1053 + }, + { + "epoch": 1.651356993736952, + "grad_norm": 0.9427600502967834, + "learning_rate": 4.234691626452392e-06, + "loss": 0.7014, + "step": 1054 + }, + { + "epoch": 1.6529227557411272, + "grad_norm": 1.0177332162857056, + "learning_rate": 4.233173949504842e-06, + "loss": 0.7717, + "step": 1055 + }, + { + "epoch": 1.6544885177453028, + "grad_norm": 1.0621635913848877, + "learning_rate": 4.231655041749985e-06, + "loss": 0.7078, + "step": 1056 + }, + { + "epoch": 1.656054279749478, + "grad_norm": 1.0474066734313965, + "learning_rate": 4.230134904266472e-06, + "loss": 0.7869, + "step": 1057 + }, + { + "epoch": 1.6576200417536535, + "grad_norm": 0.9170332551002502, + "learning_rate": 4.228613538133821e-06, + "loss": 0.7806, + "step": 1058 + }, + { + "epoch": 1.6591858037578286, + "grad_norm": 1.010465383529663, + "learning_rate": 4.227090944432426e-06, + "loss": 0.7352, + "step": 1059 + }, + { + "epoch": 1.6607515657620042, + "grad_norm": 0.988068163394928, + "learning_rate": 4.22556712424355e-06, + "loss": 0.7173, + "step": 1060 + }, + { + "epoch": 1.6623173277661796, + "grad_norm": 1.0753575563430786, + "learning_rate": 4.22404207864933e-06, + "loss": 0.6981, + "step": 1061 + }, + { + "epoch": 1.663883089770355, + "grad_norm": 1.0075691938400269, + "learning_rate": 4.222515808732769e-06, + "loss": 0.7279, + "step": 1062 + }, + { + "epoch": 1.6654488517745303, + "grad_norm": 1.0954983234405518, + "learning_rate": 4.220988315577745e-06, + "loss": 0.7675, + "step": 1063 + }, + { + "epoch": 1.6670146137787056, + "grad_norm": 1.0368746519088745, + "learning_rate": 4.2194596002690016e-06, + "loss": 0.7466, + "step": 1064 + }, + { + "epoch": 1.668580375782881, + "grad_norm": 1.083906650543213, + "learning_rate": 4.217929663892149e-06, + "loss": 0.7517, + "step": 1065 + }, + { + "epoch": 1.6701461377870563, + "grad_norm": 1.1356595754623413, + "learning_rate": 4.216398507533667e-06, + "loss": 0.7601, + "step": 1066 + }, + { + "epoch": 1.6717118997912317, + "grad_norm": 1.1610008478164673, + "learning_rate": 4.2148661322809e-06, + "loss": 0.7356, + "step": 1067 + }, + { + "epoch": 1.673277661795407, + "grad_norm": 1.019673466682434, + "learning_rate": 4.21333253922206e-06, + "loss": 0.7618, + "step": 1068 + }, + { + "epoch": 1.6748434237995826, + "grad_norm": 1.1470829248428345, + "learning_rate": 4.2117977294462235e-06, + "loss": 0.8001, + "step": 1069 + }, + { + "epoch": 1.6764091858037578, + "grad_norm": 0.9750358462333679, + "learning_rate": 4.210261704043327e-06, + "loss": 0.6736, + "step": 1070 + }, + { + "epoch": 1.6779749478079333, + "grad_norm": 1.061862826347351, + "learning_rate": 4.208724464104176e-06, + "loss": 0.7751, + "step": 1071 + }, + { + "epoch": 1.6795407098121085, + "grad_norm": 1.0311503410339355, + "learning_rate": 4.2071860107204346e-06, + "loss": 0.7629, + "step": 1072 + }, + { + "epoch": 1.681106471816284, + "grad_norm": 1.0023622512817383, + "learning_rate": 4.2056463449846305e-06, + "loss": 0.6935, + "step": 1073 + }, + { + "epoch": 1.6826722338204592, + "grad_norm": 1.1624165773391724, + "learning_rate": 4.204105467990153e-06, + "loss": 0.7451, + "step": 1074 + }, + { + "epoch": 1.6842379958246347, + "grad_norm": 0.9852432608604431, + "learning_rate": 4.202563380831249e-06, + "loss": 0.7446, + "step": 1075 + }, + { + "epoch": 1.6858037578288099, + "grad_norm": 1.0009260177612305, + "learning_rate": 4.201020084603027e-06, + "loss": 0.726, + "step": 1076 + }, + { + "epoch": 1.6873695198329854, + "grad_norm": 1.0374701023101807, + "learning_rate": 4.199475580401452e-06, + "loss": 0.6687, + "step": 1077 + }, + { + "epoch": 1.6889352818371608, + "grad_norm": 1.1437046527862549, + "learning_rate": 4.19792986932335e-06, + "loss": 0.7802, + "step": 1078 + }, + { + "epoch": 1.6905010438413361, + "grad_norm": 0.9866487383842468, + "learning_rate": 4.1963829524664e-06, + "loss": 0.7117, + "step": 1079 + }, + { + "epoch": 1.6920668058455115, + "grad_norm": 0.9720652103424072, + "learning_rate": 4.194834830929142e-06, + "loss": 0.778, + "step": 1080 + }, + { + "epoch": 1.6936325678496869, + "grad_norm": 1.009422779083252, + "learning_rate": 4.1932855058109675e-06, + "loss": 0.6783, + "step": 1081 + }, + { + "epoch": 1.6951983298538622, + "grad_norm": 1.024781346321106, + "learning_rate": 4.191734978212124e-06, + "loss": 0.6697, + "step": 1082 + }, + { + "epoch": 1.6967640918580376, + "grad_norm": 1.060354471206665, + "learning_rate": 4.190183249233715e-06, + "loss": 0.7151, + "step": 1083 + }, + { + "epoch": 1.698329853862213, + "grad_norm": 1.118486762046814, + "learning_rate": 4.1886303199776924e-06, + "loss": 0.7408, + "step": 1084 + }, + { + "epoch": 1.6998956158663883, + "grad_norm": 0.9334209561347961, + "learning_rate": 4.187076191546865e-06, + "loss": 0.7646, + "step": 1085 + }, + { + "epoch": 1.7014613778705638, + "grad_norm": 1.0506161451339722, + "learning_rate": 4.185520865044892e-06, + "loss": 0.7299, + "step": 1086 + }, + { + "epoch": 1.703027139874739, + "grad_norm": 1.1261264085769653, + "learning_rate": 4.1839643415762825e-06, + "loss": 0.6813, + "step": 1087 + }, + { + "epoch": 1.7045929018789145, + "grad_norm": 0.9690954685211182, + "learning_rate": 4.182406622246396e-06, + "loss": 0.7417, + "step": 1088 + }, + { + "epoch": 1.7061586638830897, + "grad_norm": 1.0872795581817627, + "learning_rate": 4.180847708161442e-06, + "loss": 0.676, + "step": 1089 + }, + { + "epoch": 1.7077244258872653, + "grad_norm": 1.054091453552246, + "learning_rate": 4.179287600428476e-06, + "loss": 0.785, + "step": 1090 + }, + { + "epoch": 1.7092901878914404, + "grad_norm": 1.109984278678894, + "learning_rate": 4.177726300155404e-06, + "loss": 0.8033, + "step": 1091 + }, + { + "epoch": 1.710855949895616, + "grad_norm": 1.0984348058700562, + "learning_rate": 4.176163808450978e-06, + "loss": 0.6964, + "step": 1092 + }, + { + "epoch": 1.712421711899791, + "grad_norm": 1.0788499116897583, + "learning_rate": 4.174600126424797e-06, + "loss": 0.7596, + "step": 1093 + }, + { + "epoch": 1.7139874739039667, + "grad_norm": 1.0846409797668457, + "learning_rate": 4.173035255187302e-06, + "loss": 0.7322, + "step": 1094 + }, + { + "epoch": 1.715553235908142, + "grad_norm": 1.1376627683639526, + "learning_rate": 4.171469195849782e-06, + "loss": 0.6934, + "step": 1095 + }, + { + "epoch": 1.7171189979123174, + "grad_norm": 1.003432035446167, + "learning_rate": 4.169901949524368e-06, + "loss": 0.7663, + "step": 1096 + }, + { + "epoch": 1.7186847599164927, + "grad_norm": 0.9824592471122742, + "learning_rate": 4.168333517324036e-06, + "loss": 0.8253, + "step": 1097 + }, + { + "epoch": 1.720250521920668, + "grad_norm": 1.0256636142730713, + "learning_rate": 4.1667639003626e-06, + "loss": 0.6614, + "step": 1098 + }, + { + "epoch": 1.7218162839248434, + "grad_norm": 1.0290439128875732, + "learning_rate": 4.16519309975472e-06, + "loss": 0.7403, + "step": 1099 + }, + { + "epoch": 1.7233820459290188, + "grad_norm": 1.0358202457427979, + "learning_rate": 4.163621116615892e-06, + "loss": 0.7332, + "step": 1100 + }, + { + "epoch": 1.7249478079331941, + "grad_norm": 1.1046626567840576, + "learning_rate": 4.162047952062456e-06, + "loss": 0.6672, + "step": 1101 + }, + { + "epoch": 1.7265135699373695, + "grad_norm": 1.116408109664917, + "learning_rate": 4.16047360721159e-06, + "loss": 0.6997, + "step": 1102 + }, + { + "epoch": 1.7280793319415448, + "grad_norm": 0.98244309425354, + "learning_rate": 4.158898083181308e-06, + "loss": 0.754, + "step": 1103 + }, + { + "epoch": 1.7296450939457202, + "grad_norm": 1.033761978149414, + "learning_rate": 4.1573213810904635e-06, + "loss": 0.7317, + "step": 1104 + }, + { + "epoch": 1.7312108559498958, + "grad_norm": 1.1415889263153076, + "learning_rate": 4.155743502058745e-06, + "loss": 0.8361, + "step": 1105 + }, + { + "epoch": 1.732776617954071, + "grad_norm": 0.9897739291191101, + "learning_rate": 4.154164447206679e-06, + "loss": 0.7361, + "step": 1106 + }, + { + "epoch": 1.7343423799582465, + "grad_norm": 1.0837332010269165, + "learning_rate": 4.1525842176556235e-06, + "loss": 0.7007, + "step": 1107 + }, + { + "epoch": 1.7359081419624216, + "grad_norm": 1.021332025527954, + "learning_rate": 4.151002814527774e-06, + "loss": 0.723, + "step": 1108 + }, + { + "epoch": 1.7374739039665972, + "grad_norm": 0.9808626770973206, + "learning_rate": 4.14942023894616e-06, + "loss": 0.6121, + "step": 1109 + }, + { + "epoch": 1.7390396659707723, + "grad_norm": 1.0266727209091187, + "learning_rate": 4.14783649203464e-06, + "loss": 0.8221, + "step": 1110 + }, + { + "epoch": 1.7406054279749479, + "grad_norm": 1.0880744457244873, + "learning_rate": 4.146251574917907e-06, + "loss": 0.7159, + "step": 1111 + }, + { + "epoch": 1.742171189979123, + "grad_norm": 1.0125631093978882, + "learning_rate": 4.144665488721483e-06, + "loss": 0.7228, + "step": 1112 + }, + { + "epoch": 1.7437369519832986, + "grad_norm": 0.9339308142662048, + "learning_rate": 4.143078234571723e-06, + "loss": 0.7472, + "step": 1113 + }, + { + "epoch": 1.745302713987474, + "grad_norm": 1.1234889030456543, + "learning_rate": 4.141489813595809e-06, + "loss": 0.7875, + "step": 1114 + }, + { + "epoch": 1.7468684759916493, + "grad_norm": 1.0862795114517212, + "learning_rate": 4.139900226921753e-06, + "loss": 0.7219, + "step": 1115 + }, + { + "epoch": 1.7484342379958246, + "grad_norm": 1.027912974357605, + "learning_rate": 4.138309475678393e-06, + "loss": 0.6616, + "step": 1116 + }, + { + "epoch": 1.75, + "grad_norm": 1.0271416902542114, + "learning_rate": 4.136717560995398e-06, + "loss": 0.7245, + "step": 1117 + }, + { + "epoch": 1.7515657620041754, + "grad_norm": 0.9430286288261414, + "learning_rate": 4.135124484003258e-06, + "loss": 0.7347, + "step": 1118 + }, + { + "epoch": 1.7531315240083507, + "grad_norm": 1.1191871166229248, + "learning_rate": 4.133530245833292e-06, + "loss": 0.6713, + "step": 1119 + }, + { + "epoch": 1.754697286012526, + "grad_norm": 1.0734976530075073, + "learning_rate": 4.131934847617642e-06, + "loss": 0.7268, + "step": 1120 + }, + { + "epoch": 1.7562630480167014, + "grad_norm": 1.0727618932724, + "learning_rate": 4.130338290489276e-06, + "loss": 0.7855, + "step": 1121 + }, + { + "epoch": 1.757828810020877, + "grad_norm": 1.0351287126541138, + "learning_rate": 4.1287405755819805e-06, + "loss": 0.7297, + "step": 1122 + }, + { + "epoch": 1.7593945720250521, + "grad_norm": 1.046382188796997, + "learning_rate": 4.127141704030371e-06, + "loss": 0.6508, + "step": 1123 + }, + { + "epoch": 1.7609603340292277, + "grad_norm": 1.0040491819381714, + "learning_rate": 4.125541676969876e-06, + "loss": 0.7707, + "step": 1124 + }, + { + "epoch": 1.7625260960334028, + "grad_norm": 1.1626756191253662, + "learning_rate": 4.123940495536752e-06, + "loss": 0.7366, + "step": 1125 + }, + { + "epoch": 1.7640918580375784, + "grad_norm": 0.9720292091369629, + "learning_rate": 4.122338160868071e-06, + "loss": 0.7641, + "step": 1126 + }, + { + "epoch": 1.7656576200417535, + "grad_norm": 0.9957835078239441, + "learning_rate": 4.1207346741017265e-06, + "loss": 0.7398, + "step": 1127 + }, + { + "epoch": 1.767223382045929, + "grad_norm": 0.972172737121582, + "learning_rate": 4.1191300363764276e-06, + "loss": 0.6565, + "step": 1128 + }, + { + "epoch": 1.7687891440501042, + "grad_norm": 1.0128676891326904, + "learning_rate": 4.117524248831702e-06, + "loss": 0.7433, + "step": 1129 + }, + { + "epoch": 1.7703549060542798, + "grad_norm": 1.1161550283432007, + "learning_rate": 4.115917312607895e-06, + "loss": 0.8241, + "step": 1130 + }, + { + "epoch": 1.7719206680584552, + "grad_norm": 1.009810209274292, + "learning_rate": 4.114309228846167e-06, + "loss": 0.6939, + "step": 1131 + }, + { + "epoch": 1.7734864300626305, + "grad_norm": 1.0082894563674927, + "learning_rate": 4.112699998688492e-06, + "loss": 0.6772, + "step": 1132 + }, + { + "epoch": 1.7750521920668059, + "grad_norm": 0.9453067183494568, + "learning_rate": 4.111089623277658e-06, + "loss": 0.6982, + "step": 1133 + }, + { + "epoch": 1.7766179540709812, + "grad_norm": 1.074747920036316, + "learning_rate": 4.1094781037572696e-06, + "loss": 0.7296, + "step": 1134 + }, + { + "epoch": 1.7781837160751566, + "grad_norm": 1.0477592945098877, + "learning_rate": 4.1078654412717404e-06, + "loss": 0.8162, + "step": 1135 + }, + { + "epoch": 1.779749478079332, + "grad_norm": 1.1362730264663696, + "learning_rate": 4.106251636966297e-06, + "loss": 0.6942, + "step": 1136 + }, + { + "epoch": 1.7813152400835073, + "grad_norm": 0.9564191102981567, + "learning_rate": 4.104636691986977e-06, + "loss": 0.7613, + "step": 1137 + }, + { + "epoch": 1.7828810020876826, + "grad_norm": 1.0384364128112793, + "learning_rate": 4.103020607480628e-06, + "loss": 0.6482, + "step": 1138 + }, + { + "epoch": 1.784446764091858, + "grad_norm": 1.1258740425109863, + "learning_rate": 4.101403384594905e-06, + "loss": 0.7736, + "step": 1139 + }, + { + "epoch": 1.7860125260960333, + "grad_norm": 1.0740283727645874, + "learning_rate": 4.099785024478276e-06, + "loss": 0.7219, + "step": 1140 + }, + { + "epoch": 1.787578288100209, + "grad_norm": 1.0377696752548218, + "learning_rate": 4.098165528280012e-06, + "loss": 0.7132, + "step": 1141 + }, + { + "epoch": 1.789144050104384, + "grad_norm": 1.0551276206970215, + "learning_rate": 4.096544897150192e-06, + "loss": 0.7879, + "step": 1142 + }, + { + "epoch": 1.7907098121085596, + "grad_norm": 1.096389651298523, + "learning_rate": 4.0949231322397036e-06, + "loss": 0.7516, + "step": 1143 + }, + { + "epoch": 1.7922755741127347, + "grad_norm": 0.9602596163749695, + "learning_rate": 4.093300234700234e-06, + "loss": 0.7197, + "step": 1144 + }, + { + "epoch": 1.7938413361169103, + "grad_norm": 1.1918690204620361, + "learning_rate": 4.09167620568428e-06, + "loss": 0.7634, + "step": 1145 + }, + { + "epoch": 1.7954070981210855, + "grad_norm": 1.1249059438705444, + "learning_rate": 4.09005104634514e-06, + "loss": 0.7718, + "step": 1146 + }, + { + "epoch": 1.796972860125261, + "grad_norm": 1.0596246719360352, + "learning_rate": 4.0884247578369164e-06, + "loss": 0.7258, + "step": 1147 + }, + { + "epoch": 1.7985386221294362, + "grad_norm": 1.026201844215393, + "learning_rate": 4.086797341314509e-06, + "loss": 0.6426, + "step": 1148 + }, + { + "epoch": 1.8001043841336117, + "grad_norm": 1.0791343450546265, + "learning_rate": 4.085168797933625e-06, + "loss": 0.7147, + "step": 1149 + }, + { + "epoch": 1.801670146137787, + "grad_norm": 0.9660344123840332, + "learning_rate": 4.083539128850766e-06, + "loss": 0.7335, + "step": 1150 + }, + { + "epoch": 1.8032359081419624, + "grad_norm": 1.0022811889648438, + "learning_rate": 4.081908335223238e-06, + "loss": 0.6824, + "step": 1151 + }, + { + "epoch": 1.8048016701461378, + "grad_norm": 0.9261506199836731, + "learning_rate": 4.080276418209143e-06, + "loss": 0.7538, + "step": 1152 + }, + { + "epoch": 1.8063674321503131, + "grad_norm": 1.088552713394165, + "learning_rate": 4.078643378967379e-06, + "loss": 0.6988, + "step": 1153 + }, + { + "epoch": 1.8079331941544885, + "grad_norm": 1.0581461191177368, + "learning_rate": 4.077009218657645e-06, + "loss": 0.7075, + "step": 1154 + }, + { + "epoch": 1.8094989561586639, + "grad_norm": 1.0722301006317139, + "learning_rate": 4.0753739384404335e-06, + "loss": 0.6974, + "step": 1155 + }, + { + "epoch": 1.8110647181628392, + "grad_norm": 0.9519115090370178, + "learning_rate": 4.073737539477033e-06, + "loss": 0.6898, + "step": 1156 + }, + { + "epoch": 1.8126304801670146, + "grad_norm": 1.1201752424240112, + "learning_rate": 4.072100022929526e-06, + "loss": 0.8572, + "step": 1157 + }, + { + "epoch": 1.8141962421711901, + "grad_norm": 1.24104642868042, + "learning_rate": 4.070461389960789e-06, + "loss": 0.7407, + "step": 1158 + }, + { + "epoch": 1.8157620041753653, + "grad_norm": 1.0625057220458984, + "learning_rate": 4.068821641734492e-06, + "loss": 0.655, + "step": 1159 + }, + { + "epoch": 1.8173277661795408, + "grad_norm": 1.1174049377441406, + "learning_rate": 4.067180779415097e-06, + "loss": 0.7214, + "step": 1160 + }, + { + "epoch": 1.818893528183716, + "grad_norm": 1.0683097839355469, + "learning_rate": 4.065538804167855e-06, + "loss": 0.7863, + "step": 1161 + }, + { + "epoch": 1.8204592901878915, + "grad_norm": 1.031198263168335, + "learning_rate": 4.063895717158809e-06, + "loss": 0.7663, + "step": 1162 + }, + { + "epoch": 1.8220250521920667, + "grad_norm": 1.1091554164886475, + "learning_rate": 4.062251519554791e-06, + "loss": 0.7234, + "step": 1163 + }, + { + "epoch": 1.8235908141962422, + "grad_norm": 1.0818198919296265, + "learning_rate": 4.060606212523425e-06, + "loss": 0.7709, + "step": 1164 + }, + { + "epoch": 1.8251565762004174, + "grad_norm": 1.0104292631149292, + "learning_rate": 4.058959797233116e-06, + "loss": 0.7225, + "step": 1165 + }, + { + "epoch": 1.826722338204593, + "grad_norm": 1.1726491451263428, + "learning_rate": 4.057312274853063e-06, + "loss": 0.6831, + "step": 1166 + }, + { + "epoch": 1.8282881002087683, + "grad_norm": 0.9735565185546875, + "learning_rate": 4.055663646553248e-06, + "loss": 0.7123, + "step": 1167 + }, + { + "epoch": 1.8298538622129437, + "grad_norm": 1.011709213256836, + "learning_rate": 4.054013913504436e-06, + "loss": 0.6885, + "step": 1168 + }, + { + "epoch": 1.831419624217119, + "grad_norm": 1.0374364852905273, + "learning_rate": 4.05236307687818e-06, + "loss": 0.782, + "step": 1169 + }, + { + "epoch": 1.8329853862212944, + "grad_norm": 1.011202096939087, + "learning_rate": 4.050711137846817e-06, + "loss": 0.7381, + "step": 1170 + }, + { + "epoch": 1.8345511482254697, + "grad_norm": 1.0984755754470825, + "learning_rate": 4.049058097583464e-06, + "loss": 0.7472, + "step": 1171 + }, + { + "epoch": 1.836116910229645, + "grad_norm": 1.090078353881836, + "learning_rate": 4.047403957262024e-06, + "loss": 0.6773, + "step": 1172 + }, + { + "epoch": 1.8376826722338204, + "grad_norm": 0.9866096377372742, + "learning_rate": 4.045748718057176e-06, + "loss": 0.6812, + "step": 1173 + }, + { + "epoch": 1.8392484342379958, + "grad_norm": 1.1006139516830444, + "learning_rate": 4.0440923811443846e-06, + "loss": 0.7953, + "step": 1174 + }, + { + "epoch": 1.8408141962421714, + "grad_norm": 0.9621697664260864, + "learning_rate": 4.04243494769989e-06, + "loss": 0.7557, + "step": 1175 + }, + { + "epoch": 1.8423799582463465, + "grad_norm": 1.1520715951919556, + "learning_rate": 4.040776418900714e-06, + "loss": 0.7422, + "step": 1176 + }, + { + "epoch": 1.843945720250522, + "grad_norm": 1.0679237842559814, + "learning_rate": 4.039116795924653e-06, + "loss": 0.6738, + "step": 1177 + }, + { + "epoch": 1.8455114822546972, + "grad_norm": 1.026510238647461, + "learning_rate": 4.037456079950284e-06, + "loss": 0.7741, + "step": 1178 + }, + { + "epoch": 1.8470772442588728, + "grad_norm": 1.0111522674560547, + "learning_rate": 4.0357942721569585e-06, + "loss": 0.6968, + "step": 1179 + }, + { + "epoch": 1.848643006263048, + "grad_norm": 1.0290493965148926, + "learning_rate": 4.034131373724802e-06, + "loss": 0.7131, + "step": 1180 + }, + { + "epoch": 1.8502087682672235, + "grad_norm": 0.9878649711608887, + "learning_rate": 4.032467385834718e-06, + "loss": 0.7477, + "step": 1181 + }, + { + "epoch": 1.8517745302713986, + "grad_norm": 1.0317823886871338, + "learning_rate": 4.030802309668379e-06, + "loss": 0.6318, + "step": 1182 + }, + { + "epoch": 1.8533402922755742, + "grad_norm": 1.0163817405700684, + "learning_rate": 4.029136146408235e-06, + "loss": 0.7027, + "step": 1183 + }, + { + "epoch": 1.8549060542797495, + "grad_norm": 0.9590600728988647, + "learning_rate": 4.027468897237505e-06, + "loss": 0.6988, + "step": 1184 + }, + { + "epoch": 1.8564718162839249, + "grad_norm": 1.0645931959152222, + "learning_rate": 4.0258005633401795e-06, + "loss": 0.7588, + "step": 1185 + }, + { + "epoch": 1.8580375782881002, + "grad_norm": 1.0380210876464844, + "learning_rate": 4.024131145901022e-06, + "loss": 0.6984, + "step": 1186 + }, + { + "epoch": 1.8596033402922756, + "grad_norm": 1.0558087825775146, + "learning_rate": 4.022460646105561e-06, + "loss": 0.6729, + "step": 1187 + }, + { + "epoch": 1.861169102296451, + "grad_norm": 1.0865014791488647, + "learning_rate": 4.020789065140097e-06, + "loss": 0.7208, + "step": 1188 + }, + { + "epoch": 1.8627348643006263, + "grad_norm": 1.0456374883651733, + "learning_rate": 4.019116404191697e-06, + "loss": 0.7759, + "step": 1189 + }, + { + "epoch": 1.8643006263048016, + "grad_norm": 0.9724026918411255, + "learning_rate": 4.017442664448197e-06, + "loss": 0.7475, + "step": 1190 + }, + { + "epoch": 1.865866388308977, + "grad_norm": 1.1170289516448975, + "learning_rate": 4.015767847098194e-06, + "loss": 0.6993, + "step": 1191 + }, + { + "epoch": 1.8674321503131524, + "grad_norm": 1.1052978038787842, + "learning_rate": 4.014091953331056e-06, + "loss": 0.748, + "step": 1192 + }, + { + "epoch": 1.8689979123173277, + "grad_norm": 1.1223454475402832, + "learning_rate": 4.012414984336912e-06, + "loss": 0.6921, + "step": 1193 + }, + { + "epoch": 1.8705636743215033, + "grad_norm": 1.0611138343811035, + "learning_rate": 4.010736941306655e-06, + "loss": 0.7859, + "step": 1194 + }, + { + "epoch": 1.8721294363256784, + "grad_norm": 1.169189691543579, + "learning_rate": 4.009057825431944e-06, + "loss": 0.5874, + "step": 1195 + }, + { + "epoch": 1.873695198329854, + "grad_norm": 1.036626935005188, + "learning_rate": 4.0073776379051945e-06, + "loss": 0.6801, + "step": 1196 + }, + { + "epoch": 1.8752609603340291, + "grad_norm": 0.9431620240211487, + "learning_rate": 4.005696379919586e-06, + "loss": 0.7476, + "step": 1197 + }, + { + "epoch": 1.8768267223382047, + "grad_norm": 1.0381059646606445, + "learning_rate": 4.0040140526690585e-06, + "loss": 0.6691, + "step": 1198 + }, + { + "epoch": 1.8783924843423798, + "grad_norm": 1.120033860206604, + "learning_rate": 4.0023306573483105e-06, + "loss": 0.6669, + "step": 1199 + }, + { + "epoch": 1.8799582463465554, + "grad_norm": 1.0298967361450195, + "learning_rate": 4.000646195152798e-06, + "loss": 0.7355, + "step": 1200 + }, + { + "epoch": 1.8815240083507305, + "grad_norm": 1.0036594867706299, + "learning_rate": 3.99896066727874e-06, + "loss": 0.7545, + "step": 1201 + }, + { + "epoch": 1.883089770354906, + "grad_norm": 1.1166342496871948, + "learning_rate": 3.997274074923103e-06, + "loss": 0.7468, + "step": 1202 + }, + { + "epoch": 1.8846555323590815, + "grad_norm": 1.0512094497680664, + "learning_rate": 3.995586419283617e-06, + "loss": 0.7647, + "step": 1203 + }, + { + "epoch": 1.8862212943632568, + "grad_norm": 1.0115973949432373, + "learning_rate": 3.993897701558764e-06, + "loss": 0.7893, + "step": 1204 + }, + { + "epoch": 1.8877870563674322, + "grad_norm": 1.0839310884475708, + "learning_rate": 3.992207922947783e-06, + "loss": 0.6824, + "step": 1205 + }, + { + "epoch": 1.8893528183716075, + "grad_norm": 0.9987578988075256, + "learning_rate": 3.99051708465066e-06, + "loss": 0.8045, + "step": 1206 + }, + { + "epoch": 1.8909185803757829, + "grad_norm": 1.0759059190750122, + "learning_rate": 3.9888251878681425e-06, + "loss": 0.7756, + "step": 1207 + }, + { + "epoch": 1.8924843423799582, + "grad_norm": 1.1055301427841187, + "learning_rate": 3.987132233801722e-06, + "loss": 0.6947, + "step": 1208 + }, + { + "epoch": 1.8940501043841336, + "grad_norm": 1.042575478553772, + "learning_rate": 3.985438223653644e-06, + "loss": 0.6669, + "step": 1209 + }, + { + "epoch": 1.895615866388309, + "grad_norm": 1.0595242977142334, + "learning_rate": 3.983743158626904e-06, + "loss": 0.6937, + "step": 1210 + }, + { + "epoch": 1.8971816283924845, + "grad_norm": 1.0498789548873901, + "learning_rate": 3.982047039925247e-06, + "loss": 0.7244, + "step": 1211 + }, + { + "epoch": 1.8987473903966596, + "grad_norm": 1.0960590839385986, + "learning_rate": 3.980349868753166e-06, + "loss": 0.7661, + "step": 1212 + }, + { + "epoch": 1.9003131524008352, + "grad_norm": 1.108823299407959, + "learning_rate": 3.9786516463159e-06, + "loss": 0.6937, + "step": 1213 + }, + { + "epoch": 1.9018789144050103, + "grad_norm": 1.051416039466858, + "learning_rate": 3.976952373819436e-06, + "loss": 0.7263, + "step": 1214 + }, + { + "epoch": 1.903444676409186, + "grad_norm": 0.9884061217308044, + "learning_rate": 3.9752520524705065e-06, + "loss": 0.6577, + "step": 1215 + }, + { + "epoch": 1.905010438413361, + "grad_norm": 1.0950369834899902, + "learning_rate": 3.9735506834765885e-06, + "loss": 0.7376, + "step": 1216 + }, + { + "epoch": 1.9065762004175366, + "grad_norm": 1.1140044927597046, + "learning_rate": 3.971848268045902e-06, + "loss": 0.7125, + "step": 1217 + }, + { + "epoch": 1.9081419624217117, + "grad_norm": 1.0908716917037964, + "learning_rate": 3.970144807387414e-06, + "loss": 0.7796, + "step": 1218 + }, + { + "epoch": 1.9097077244258873, + "grad_norm": 1.1068549156188965, + "learning_rate": 3.968440302710829e-06, + "loss": 0.7454, + "step": 1219 + }, + { + "epoch": 1.9112734864300627, + "grad_norm": 1.102594017982483, + "learning_rate": 3.9667347552265945e-06, + "loss": 0.757, + "step": 1220 + }, + { + "epoch": 1.912839248434238, + "grad_norm": 1.1588326692581177, + "learning_rate": 3.9650281661459015e-06, + "loss": 0.6491, + "step": 1221 + }, + { + "epoch": 1.9144050104384134, + "grad_norm": 0.9896461367607117, + "learning_rate": 3.963320536680675e-06, + "loss": 0.773, + "step": 1222 + }, + { + "epoch": 1.9159707724425887, + "grad_norm": 1.0855952501296997, + "learning_rate": 3.961611868043583e-06, + "loss": 0.6906, + "step": 1223 + }, + { + "epoch": 1.917536534446764, + "grad_norm": 1.0556972026824951, + "learning_rate": 3.959902161448032e-06, + "loss": 0.7577, + "step": 1224 + }, + { + "epoch": 1.9191022964509394, + "grad_norm": 1.0284528732299805, + "learning_rate": 3.958191418108162e-06, + "loss": 0.6844, + "step": 1225 + }, + { + "epoch": 1.9206680584551148, + "grad_norm": 0.9578842520713806, + "learning_rate": 3.956479639238852e-06, + "loss": 0.7216, + "step": 1226 + }, + { + "epoch": 1.9222338204592901, + "grad_norm": 1.0366833209991455, + "learning_rate": 3.954766826055717e-06, + "loss": 0.7365, + "step": 1227 + }, + { + "epoch": 1.9237995824634657, + "grad_norm": 1.0498946905136108, + "learning_rate": 3.953052979775103e-06, + "loss": 0.6788, + "step": 1228 + }, + { + "epoch": 1.9253653444676408, + "grad_norm": 1.010860562324524, + "learning_rate": 3.951338101614094e-06, + "loss": 0.7181, + "step": 1229 + }, + { + "epoch": 1.9269311064718164, + "grad_norm": 1.0203677415847778, + "learning_rate": 3.949622192790502e-06, + "loss": 0.7277, + "step": 1230 + }, + { + "epoch": 1.9284968684759916, + "grad_norm": 0.9907971024513245, + "learning_rate": 3.947905254522876e-06, + "loss": 0.6858, + "step": 1231 + }, + { + "epoch": 1.9300626304801671, + "grad_norm": 0.9824337363243103, + "learning_rate": 3.946187288030493e-06, + "loss": 0.7517, + "step": 1232 + }, + { + "epoch": 1.9316283924843423, + "grad_norm": 1.0206327438354492, + "learning_rate": 3.944468294533359e-06, + "loss": 0.7225, + "step": 1233 + }, + { + "epoch": 1.9331941544885178, + "grad_norm": 1.0529900789260864, + "learning_rate": 3.942748275252214e-06, + "loss": 0.7965, + "step": 1234 + }, + { + "epoch": 1.934759916492693, + "grad_norm": 1.0501983165740967, + "learning_rate": 3.941027231408521e-06, + "loss": 0.7646, + "step": 1235 + }, + { + "epoch": 1.9363256784968685, + "grad_norm": 0.9482662081718445, + "learning_rate": 3.939305164224474e-06, + "loss": 0.7449, + "step": 1236 + }, + { + "epoch": 1.937891440501044, + "grad_norm": 1.096723198890686, + "learning_rate": 3.937582074922993e-06, + "loss": 0.7464, + "step": 1237 + }, + { + "epoch": 1.9394572025052192, + "grad_norm": 1.0417059659957886, + "learning_rate": 3.935857964727725e-06, + "loss": 0.6586, + "step": 1238 + }, + { + "epoch": 1.9410229645093946, + "grad_norm": 1.1008849143981934, + "learning_rate": 3.934132834863039e-06, + "loss": 0.6236, + "step": 1239 + }, + { + "epoch": 1.94258872651357, + "grad_norm": 0.9107227921485901, + "learning_rate": 3.93240668655403e-06, + "loss": 0.7069, + "step": 1240 + }, + { + "epoch": 1.9441544885177453, + "grad_norm": 1.0516945123672485, + "learning_rate": 3.930679521026517e-06, + "loss": 0.773, + "step": 1241 + }, + { + "epoch": 1.9457202505219207, + "grad_norm": 0.9941951632499695, + "learning_rate": 3.92895133950704e-06, + "loss": 0.6002, + "step": 1242 + }, + { + "epoch": 1.947286012526096, + "grad_norm": 1.031784176826477, + "learning_rate": 3.927222143222859e-06, + "loss": 0.7737, + "step": 1243 + }, + { + "epoch": 1.9488517745302714, + "grad_norm": 0.986473798751831, + "learning_rate": 3.925491933401961e-06, + "loss": 0.7455, + "step": 1244 + }, + { + "epoch": 1.9504175365344467, + "grad_norm": 1.109964370727539, + "learning_rate": 3.923760711273044e-06, + "loss": 0.7048, + "step": 1245 + }, + { + "epoch": 1.951983298538622, + "grad_norm": 1.1303372383117676, + "learning_rate": 3.922028478065532e-06, + "loss": 0.749, + "step": 1246 + }, + { + "epoch": 1.9535490605427976, + "grad_norm": 0.999824583530426, + "learning_rate": 3.920295235009562e-06, + "loss": 0.7596, + "step": 1247 + }, + { + "epoch": 1.9551148225469728, + "grad_norm": 1.0440820455551147, + "learning_rate": 3.918560983335993e-06, + "loss": 0.7841, + "step": 1248 + }, + { + "epoch": 1.9566805845511483, + "grad_norm": 1.0420323610305786, + "learning_rate": 3.916825724276395e-06, + "loss": 0.7837, + "step": 1249 + }, + { + "epoch": 1.9582463465553235, + "grad_norm": 1.001997947692871, + "learning_rate": 3.915089459063058e-06, + "loss": 0.6956, + "step": 1250 + }, + { + "epoch": 1.959812108559499, + "grad_norm": 1.030626893043518, + "learning_rate": 3.913352188928984e-06, + "loss": 0.74, + "step": 1251 + }, + { + "epoch": 1.9613778705636742, + "grad_norm": 1.0981760025024414, + "learning_rate": 3.911613915107888e-06, + "loss": 0.713, + "step": 1252 + }, + { + "epoch": 1.9629436325678498, + "grad_norm": 1.020772933959961, + "learning_rate": 3.9098746388342006e-06, + "loss": 0.7605, + "step": 1253 + }, + { + "epoch": 1.964509394572025, + "grad_norm": 1.0206503868103027, + "learning_rate": 3.908134361343062e-06, + "loss": 0.7176, + "step": 1254 + }, + { + "epoch": 1.9660751565762005, + "grad_norm": 1.0471689701080322, + "learning_rate": 3.906393083870325e-06, + "loss": 0.7698, + "step": 1255 + }, + { + "epoch": 1.9676409185803758, + "grad_norm": 1.0658202171325684, + "learning_rate": 3.904650807652549e-06, + "loss": 0.7179, + "step": 1256 + }, + { + "epoch": 1.9692066805845512, + "grad_norm": 1.196192741394043, + "learning_rate": 3.902907533927007e-06, + "loss": 0.7801, + "step": 1257 + }, + { + "epoch": 1.9707724425887265, + "grad_norm": 1.0874967575073242, + "learning_rate": 3.90116326393168e-06, + "loss": 0.7672, + "step": 1258 + }, + { + "epoch": 1.9723382045929019, + "grad_norm": 1.153617024421692, + "learning_rate": 3.899417998905253e-06, + "loss": 0.752, + "step": 1259 + }, + { + "epoch": 1.9739039665970772, + "grad_norm": 1.0750453472137451, + "learning_rate": 3.89767174008712e-06, + "loss": 0.7471, + "step": 1260 + }, + { + "epoch": 1.9754697286012526, + "grad_norm": 1.0693799257278442, + "learning_rate": 3.895924488717381e-06, + "loss": 0.6704, + "step": 1261 + }, + { + "epoch": 1.977035490605428, + "grad_norm": 0.9552210569381714, + "learning_rate": 3.89417624603684e-06, + "loss": 0.7793, + "step": 1262 + }, + { + "epoch": 1.9786012526096033, + "grad_norm": 0.96712726354599, + "learning_rate": 3.892427013287007e-06, + "loss": 0.6843, + "step": 1263 + }, + { + "epoch": 1.9801670146137789, + "grad_norm": 1.1453518867492676, + "learning_rate": 3.890676791710091e-06, + "loss": 0.7532, + "step": 1264 + }, + { + "epoch": 1.981732776617954, + "grad_norm": 1.0294159650802612, + "learning_rate": 3.888925582549006e-06, + "loss": 0.703, + "step": 1265 + }, + { + "epoch": 1.9832985386221296, + "grad_norm": 1.0790715217590332, + "learning_rate": 3.8871733870473686e-06, + "loss": 0.6731, + "step": 1266 + }, + { + "epoch": 1.9848643006263047, + "grad_norm": 1.052460789680481, + "learning_rate": 3.8854202064494915e-06, + "loss": 0.748, + "step": 1267 + }, + { + "epoch": 1.9864300626304803, + "grad_norm": 1.112462043762207, + "learning_rate": 3.883666042000392e-06, + "loss": 0.7408, + "step": 1268 + }, + { + "epoch": 1.9879958246346554, + "grad_norm": 1.0738476514816284, + "learning_rate": 3.881910894945783e-06, + "loss": 0.759, + "step": 1269 + }, + { + "epoch": 1.989561586638831, + "grad_norm": 1.136472463607788, + "learning_rate": 3.8801547665320775e-06, + "loss": 0.754, + "step": 1270 + }, + { + "epoch": 1.9911273486430061, + "grad_norm": 1.0640734434127808, + "learning_rate": 3.87839765800638e-06, + "loss": 0.719, + "step": 1271 + }, + { + "epoch": 1.9926931106471817, + "grad_norm": 0.9155470132827759, + "learning_rate": 3.876639570616498e-06, + "loss": 0.762, + "step": 1272 + }, + { + "epoch": 1.994258872651357, + "grad_norm": 1.015743374824524, + "learning_rate": 3.87488050561093e-06, + "loss": 0.6926, + "step": 1273 + }, + { + "epoch": 1.9958246346555324, + "grad_norm": 1.0829862356185913, + "learning_rate": 3.87312046423887e-06, + "loss": 0.7644, + "step": 1274 + }, + { + "epoch": 1.9973903966597077, + "grad_norm": 1.0962239503860474, + "learning_rate": 3.871359447750205e-06, + "loss": 0.7038, + "step": 1275 + }, + { + "epoch": 1.998956158663883, + "grad_norm": 1.0470675230026245, + "learning_rate": 3.869597457395514e-06, + "loss": 0.7629, + "step": 1276 + } + ], + "logging_steps": 1, + "max_steps": 3828, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 638, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3.224064837298422e+18, + "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..07c8c9aae2a66d9fb6a8b65759b1cd86d2f7d08b --- /dev/null +++ b/checkpoint-1276/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81aeca374ba1e1a5384aca380c5dd9da735cc6d6164e0b5b0681e227cf61f41c +size 7992 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-1914/README.md b/checkpoint-1914/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f07164552af92bf3072e4e4326b5e572dea01ae8 --- /dev/null +++ b/checkpoint-1914/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-Nano-8B-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 0.14.0 \ 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..3db75f3ecbeb5f4990760e56c7957796d83680c5 --- /dev/null +++ b/checkpoint-1914/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1", + "bias": "none", + "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": [ + "q_proj", + "down_proj", + "o_proj", + "v_proj", + "k_proj", + "gate_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "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..15cd23ea62bb456df292d2b42c3639e7da011a4c --- /dev/null +++ b/checkpoint-1914/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31547923706c7daeef14d37d011841f77bbbbf083412859850d7d3f6e62c75d8 +size 3443586272 diff --git a/checkpoint-1914/global_step1914/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-1914/global_step1914/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..d3b531748d0208d6fe01afb1fdaf73fa01fac57b --- /dev/null +++ b/checkpoint-1914/global_step1914/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:d6bcc52ee2860f10c283c0b859330b58a82aab5c73f418b6d43fb1aef3a60e36 +size 20661195036 diff --git a/checkpoint-1914/global_step1914/mp_rank_00_model_states.pt b/checkpoint-1914/global_step1914/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..ce17955b203896712252622d445c0d75841ad147 --- /dev/null +++ b/checkpoint-1914/global_step1914/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1e184cfed735e30ea9bfeb5de9a812f8dc752c77fcb292e81a4c9e044371460c +size 3555326777 diff --git a/checkpoint-1914/latest b/checkpoint-1914/latest new file mode 100644 index 0000000000000000000000000000000000000000..c6bbbb48dd57813bdb7ac0c25d9e48c1ca8f7cea --- /dev/null +++ b/checkpoint-1914/latest @@ -0,0 +1 @@ +global_step1914 \ No newline at end of file diff --git a/checkpoint-1914/rng_state.pth b/checkpoint-1914/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..86b0813c4a0c7a17891d89bcfa128ba50d72f478 --- /dev/null +++ b/checkpoint-1914/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37d9830ebeea0ce3961a1488310aac4e7a48f4cb122dbcfef3b23b8beec02dae +size 14244 diff --git a/checkpoint-1914/scheduler.pt b/checkpoint-1914/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..c097111f2793c7624662bd06cfea9f4dbf090740 --- /dev/null +++ b/checkpoint-1914/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f12a8ff5eecfe5f70909219aae0aa75bb1e904574aacd44db5a940d952a6cc5d +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..b480d149dc274c0a23db079bfd38032d90082e49 --- /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": "{%- if messages[0]['role'] == 'system' -%}{%- set system_message = messages[0]['content'] | trim -%}{%- set messages = messages[1:] -%}{%- else -%}{%- set system_message = '' -%}{%- endif -%}{%- if tools is not none -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message -}} {{- '\n\n' if system_message else '' -}} {{- '[' -}} {% for t in tools %}{{- (t.function if t.function is defined else t) | tojson() -}}{{- ', ' if not loop.last else '' -}}{%- endfor -%} {{- ']' -}} {{- '<|eot_id|>' -}}{%- else -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message + '<|eot_id|>' -}}{%- endif -%}{%- for message in messages -%}{%- if (message['role'] in ['user', 'tool']) != (loop.index0 % 2 == 0) -%}{{- raise_exception('Conversation roles must alternate between user/tool and assistant') -}}{%- elif message['role'] == 'user' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- elif message['role'] == 'tool' -%}{%- set tool_response = '[' + message['content'] | trim + ']' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + tool_response + '<|eot_id|>' -}}{%- elif message['role'] == 'assistant' and message.get('tool_calls') is not none -%}{%- set tool_calls = message['tool_calls'] -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + '[' -}}{%- for tool_call in tool_calls -%}{{ '{' + '\"name\": \"' + tool_call.function.name + '\", \"arguments\": ' + tool_call.function.arguments | tojson + '}' }}{%- if not loop.last -%}{{ ', ' }}{%- else -%}{{ ']' + '<|eot_id|>' }}{%- endif -%}{%- endfor -%}{%- elif message['role'] == 'assistant' -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- endif -%}{%- 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..55c6f7fa9198bf9a3ddd341137dc9617a7976034 --- /dev/null +++ b/checkpoint-1914/trainer_state.json @@ -0,0 +1,13432 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.998956158663883, + "eval_steps": 500, + "global_step": 1914, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0015657620041753654, + "grad_norm": 658.1201782226562, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.3039, + "step": 1 + }, + { + "epoch": 0.003131524008350731, + "grad_norm": 749.5796508789062, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.4164, + "step": 2 + }, + { + "epoch": 0.004697286012526096, + "grad_norm": 394.2350769042969, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.3367, + "step": 3 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 384.08319091796875, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.3192, + "step": 4 + }, + { + "epoch": 0.007828810020876827, + "grad_norm": 376.9790954589844, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.2756, + "step": 5 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 175.25262451171875, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.3992, + "step": 6 + }, + { + "epoch": 0.010960334029227558, + "grad_norm": 100.1114730834961, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.2348, + "step": 7 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 131.44432067871094, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.2685, + "step": 8 + }, + { + "epoch": 0.014091858037578288, + "grad_norm": 67.03341674804688, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.3272, + "step": 9 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 32.0869140625, + "learning_rate": 5.000000000000001e-07, + "loss": 1.3458, + "step": 10 + }, + { + "epoch": 0.01722338204592902, + "grad_norm": 17.701913833618164, + "learning_rate": 5.5e-07, + "loss": 1.3608, + "step": 11 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 9.164765357971191, + "learning_rate": 6.000000000000001e-07, + "loss": 1.2961, + "step": 12 + }, + { + "epoch": 0.02035490605427975, + "grad_norm": 4.315349578857422, + "learning_rate": 6.5e-07, + "loss": 1.2069, + "step": 13 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 6.652385234832764, + "learning_rate": 7.000000000000001e-07, + "loss": 1.2588, + "step": 14 + }, + { + "epoch": 0.02348643006263048, + "grad_norm": 3.3322548866271973, + "learning_rate": 7.5e-07, + "loss": 1.2505, + "step": 15 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 9.599331855773926, + "learning_rate": 8.000000000000001e-07, + "loss": 1.1343, + "step": 16 + }, + { + "epoch": 0.02661795407098121, + "grad_norm": 10.33918285369873, + "learning_rate": 8.500000000000001e-07, + "loss": 1.3334, + "step": 17 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 9.453723907470703, + "learning_rate": 9.000000000000001e-07, + "loss": 1.3099, + "step": 18 + }, + { + "epoch": 0.029749478079331943, + "grad_norm": 7.075769901275635, + "learning_rate": 9.500000000000001e-07, + "loss": 1.1121, + "step": 19 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 7.770864009857178, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.1836, + "step": 20 + }, + { + "epoch": 0.03288100208768267, + "grad_norm": 9.582944869995117, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.1939, + "step": 21 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 9.496583938598633, + "learning_rate": 1.1e-06, + "loss": 1.2395, + "step": 22 + }, + { + "epoch": 0.0360125260960334, + "grad_norm": 5.912876129150391, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.1769, + "step": 23 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 2.6761083602905273, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.204, + "step": 24 + }, + { + "epoch": 0.03914405010438413, + "grad_norm": 11.136463165283203, + "learning_rate": 1.25e-06, + "loss": 1.2271, + "step": 25 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 23.912893295288086, + "learning_rate": 1.3e-06, + "loss": 1.1771, + "step": 26 + }, + { + "epoch": 0.042275574112734866, + "grad_norm": 57.08906555175781, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.2253, + "step": 27 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 36.11408615112305, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.1014, + "step": 28 + }, + { + "epoch": 0.04540709812108559, + "grad_norm": 51.39445495605469, + "learning_rate": 1.45e-06, + "loss": 1.1978, + "step": 29 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 27.394500732421875, + "learning_rate": 1.5e-06, + "loss": 1.1155, + "step": 30 + }, + { + "epoch": 0.048538622129436326, + "grad_norm": 6.844716548919678, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.1798, + "step": 31 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 2.397536277770996, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.1846, + "step": 32 + }, + { + "epoch": 0.05167014613778706, + "grad_norm": 7.821156024932861, + "learning_rate": 1.6500000000000003e-06, + "loss": 1.1312, + "step": 33 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 5.455922603607178, + "learning_rate": 1.7000000000000002e-06, + "loss": 1.2153, + "step": 34 + }, + { + "epoch": 0.054801670146137786, + "grad_norm": 5.377617835998535, + "learning_rate": 1.75e-06, + "loss": 1.1292, + "step": 35 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 7.172031879425049, + "learning_rate": 1.8000000000000001e-06, + "loss": 1.1459, + "step": 36 + }, + { + "epoch": 0.05793319415448852, + "grad_norm": 2.1464905738830566, + "learning_rate": 1.85e-06, + "loss": 1.1649, + "step": 37 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 2.7782726287841797, + "learning_rate": 1.9000000000000002e-06, + "loss": 1.159, + "step": 38 + }, + { + "epoch": 0.061064718162839246, + "grad_norm": 20.818374633789062, + "learning_rate": 1.9500000000000004e-06, + "loss": 1.0174, + "step": 39 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 22.776456832885742, + "learning_rate": 2.0000000000000003e-06, + "loss": 1.1425, + "step": 40 + }, + { + "epoch": 0.06419624217118998, + "grad_norm": 33.601287841796875, + "learning_rate": 2.05e-06, + "loss": 1.0498, + "step": 41 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.884684562683105, + "learning_rate": 2.1000000000000002e-06, + "loss": 1.1231, + "step": 42 + }, + { + "epoch": 0.06732776617954071, + "grad_norm": 1.951470971107483, + "learning_rate": 2.15e-06, + "loss": 1.1256, + "step": 43 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 2.584127426147461, + "learning_rate": 2.2e-06, + "loss": 1.1242, + "step": 44 + }, + { + "epoch": 0.07045929018789145, + "grad_norm": 6.982508659362793, + "learning_rate": 2.25e-06, + "loss": 1.0769, + "step": 45 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 7.940657138824463, + "learning_rate": 2.3000000000000004e-06, + "loss": 1.0516, + "step": 46 + }, + { + "epoch": 0.07359081419624217, + "grad_norm": 5.456567287445068, + "learning_rate": 2.35e-06, + "loss": 0.9894, + "step": 47 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 4.623636722564697, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9912, + "step": 48 + }, + { + "epoch": 0.0767223382045929, + "grad_norm": 2.182708978652954, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.1303, + "step": 49 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 2.388091802597046, + "learning_rate": 2.5e-06, + "loss": 1.0725, + "step": 50 + }, + { + "epoch": 0.07985386221294363, + "grad_norm": 2.2816522121429443, + "learning_rate": 2.55e-06, + "loss": 1.0895, + "step": 51 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 3.8391623497009277, + "learning_rate": 2.6e-06, + "loss": 1.0587, + "step": 52 + }, + { + "epoch": 0.08298538622129437, + "grad_norm": 3.700389862060547, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9795, + "step": 53 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 3.180649995803833, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.0434, + "step": 54 + }, + { + "epoch": 0.0861169102296451, + "grad_norm": 3.790774345397949, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9652, + "step": 55 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 1.8411438465118408, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.947, + "step": 56 + }, + { + "epoch": 0.08924843423799582, + "grad_norm": 2.100008010864258, + "learning_rate": 2.85e-06, + "loss": 1.0129, + "step": 57 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 1.103857159614563, + "learning_rate": 2.9e-06, + "loss": 0.9993, + "step": 58 + }, + { + "epoch": 0.09237995824634655, + "grad_norm": 1.232283353805542, + "learning_rate": 2.95e-06, + "loss": 0.9988, + "step": 59 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 2.4731297492980957, + "learning_rate": 3e-06, + "loss": 0.9873, + "step": 60 + }, + { + "epoch": 0.09551148225469729, + "grad_norm": 1.1080410480499268, + "learning_rate": 3.05e-06, + "loss": 1.0657, + "step": 61 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 1.1155121326446533, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.974, + "step": 62 + }, + { + "epoch": 0.09864300626304802, + "grad_norm": 1.763231873512268, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.0215, + "step": 63 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 1.1961321830749512, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.9955, + "step": 64 + }, + { + "epoch": 0.10177453027139875, + "grad_norm": 1.377040982246399, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.9834, + "step": 65 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 1.373260498046875, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.984, + "step": 66 + }, + { + "epoch": 0.10490605427974947, + "grad_norm": 2.034064292907715, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.973, + "step": 67 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 1.41798734664917, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.925, + "step": 68 + }, + { + "epoch": 0.1080375782881002, + "grad_norm": 1.7425439357757568, + "learning_rate": 3.45e-06, + "loss": 0.9884, + "step": 69 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 2.369657278060913, + "learning_rate": 3.5e-06, + "loss": 1.0048, + "step": 70 + }, + { + "epoch": 0.11116910229645094, + "grad_norm": 1.4699137210845947, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.9155, + "step": 71 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 1.1473263502120972, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8445, + "step": 72 + }, + { + "epoch": 0.11430062630480167, + "grad_norm": 2.315568685531616, + "learning_rate": 3.65e-06, + "loss": 0.977, + "step": 73 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 1.8554916381835938, + "learning_rate": 3.7e-06, + "loss": 0.841, + "step": 74 + }, + { + "epoch": 0.1174321503131524, + "grad_norm": 1.1517397165298462, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.9478, + "step": 75 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 2.799316883087158, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.9873, + "step": 76 + }, + { + "epoch": 0.12056367432150313, + "grad_norm": 1.11930251121521, + "learning_rate": 3.85e-06, + "loss": 0.9214, + "step": 77 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 1.0082950592041016, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9115, + "step": 78 + }, + { + "epoch": 0.12369519832985386, + "grad_norm": 1.5534541606903076, + "learning_rate": 3.95e-06, + "loss": 1.04, + "step": 79 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 1.3469066619873047, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9091, + "step": 80 + }, + { + "epoch": 0.1268267223382046, + "grad_norm": 1.1149636507034302, + "learning_rate": 4.05e-06, + "loss": 0.8829, + "step": 81 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 2.5933806896209717, + "learning_rate": 4.1e-06, + "loss": 0.9343, + "step": 82 + }, + { + "epoch": 0.1299582463465553, + "grad_norm": 1.980901837348938, + "learning_rate": 4.15e-06, + "loss": 0.9028, + "step": 83 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 0.8476768732070923, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.9038, + "step": 84 + }, + { + "epoch": 0.13308977035490605, + "grad_norm": 1.1239815950393677, + "learning_rate": 4.25e-06, + "loss": 1.0359, + "step": 85 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 1.0159156322479248, + "learning_rate": 4.3e-06, + "loss": 0.8705, + "step": 86 + }, + { + "epoch": 0.13622129436325678, + "grad_norm": 2.180856466293335, + "learning_rate": 4.350000000000001e-06, + "loss": 0.9834, + "step": 87 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 1.5725276470184326, + "learning_rate": 4.4e-06, + "loss": 0.9449, + "step": 88 + }, + { + "epoch": 0.1393528183716075, + "grad_norm": 1.5560917854309082, + "learning_rate": 4.450000000000001e-06, + "loss": 1.0109, + "step": 89 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 1.1096960306167603, + "learning_rate": 4.5e-06, + "loss": 0.9081, + "step": 90 + }, + { + "epoch": 0.14248434237995825, + "grad_norm": 1.2525160312652588, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.9907, + "step": 91 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 0.9198770523071289, + "learning_rate": 4.600000000000001e-06, + "loss": 0.9266, + "step": 92 + }, + { + "epoch": 0.14561586638830898, + "grad_norm": 0.992057740688324, + "learning_rate": 4.65e-06, + "loss": 0.9556, + "step": 93 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 1.00990891456604, + "learning_rate": 4.7e-06, + "loss": 0.9712, + "step": 94 + }, + { + "epoch": 0.1487473903966597, + "grad_norm": 1.1176848411560059, + "learning_rate": 4.75e-06, + "loss": 0.9604, + "step": 95 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 0.9323613047599792, + "learning_rate": 4.800000000000001e-06, + "loss": 0.9708, + "step": 96 + }, + { + "epoch": 0.15187891440501045, + "grad_norm": 0.9800642728805542, + "learning_rate": 4.85e-06, + "loss": 0.9868, + "step": 97 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 1.487841010093689, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8679, + "step": 98 + }, + { + "epoch": 0.15501043841336118, + "grad_norm": 0.95517897605896, + "learning_rate": 4.95e-06, + "loss": 0.7949, + "step": 99 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 0.9686336517333984, + "learning_rate": 5e-06, + "loss": 0.8655, + "step": 100 + }, + { + "epoch": 0.1581419624217119, + "grad_norm": 1.9481035470962524, + "learning_rate": 4.999999112316954e-06, + "loss": 0.9328, + "step": 101 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 1.272631287574768, + "learning_rate": 4.9999964492684465e-06, + "loss": 0.7869, + "step": 102 + }, + { + "epoch": 0.16127348643006262, + "grad_norm": 1.1117815971374512, + "learning_rate": 4.999992010856368e-06, + "loss": 0.949, + "step": 103 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 0.9681039452552795, + "learning_rate": 4.999985797083871e-06, + "loss": 1.0396, + "step": 104 + }, + { + "epoch": 0.16440501043841335, + "grad_norm": 0.9797878265380859, + "learning_rate": 4.999977807955368e-06, + "loss": 0.9393, + "step": 105 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 1.0239158868789673, + "learning_rate": 4.999968043476532e-06, + "loss": 0.8766, + "step": 106 + }, + { + "epoch": 0.16753653444676408, + "grad_norm": 0.9314751029014587, + "learning_rate": 4.999956503654299e-06, + "loss": 0.8545, + "step": 107 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 1.1627179384231567, + "learning_rate": 4.999943188496862e-06, + "loss": 0.9241, + "step": 108 + }, + { + "epoch": 0.17066805845511482, + "grad_norm": 1.022752285003662, + "learning_rate": 4.9999280980136765e-06, + "loss": 0.9662, + "step": 109 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 1.0513653755187988, + "learning_rate": 4.99991123221546e-06, + "loss": 0.9054, + "step": 110 + }, + { + "epoch": 0.17379958246346555, + "grad_norm": 0.9488444328308105, + "learning_rate": 4.99989259111419e-06, + "loss": 0.8921, + "step": 111 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 0.9934154748916626, + "learning_rate": 4.999872174723104e-06, + "loss": 0.9848, + "step": 112 + }, + { + "epoch": 0.17693110647181629, + "grad_norm": 1.2035515308380127, + "learning_rate": 4.9998499830567e-06, + "loss": 0.9833, + "step": 113 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 1.027538537979126, + "learning_rate": 4.999826016130739e-06, + "loss": 1.0089, + "step": 114 + }, + { + "epoch": 0.18006263048016702, + "grad_norm": 1.0635300874710083, + "learning_rate": 4.999800273962238e-06, + "loss": 0.8863, + "step": 115 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 0.9978796243667603, + "learning_rate": 4.999772756569482e-06, + "loss": 0.899, + "step": 116 + }, + { + "epoch": 0.18319415448851775, + "grad_norm": 0.9971838593482971, + "learning_rate": 4.999743463972008e-06, + "loss": 0.9515, + "step": 117 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 1.2436699867248535, + "learning_rate": 4.999712396190622e-06, + "loss": 0.9101, + "step": 118 + }, + { + "epoch": 0.18632567849686849, + "grad_norm": 0.9573557376861572, + "learning_rate": 4.999679553247383e-06, + "loss": 0.8817, + "step": 119 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 0.9105668067932129, + "learning_rate": 4.9996449351656165e-06, + "loss": 1.0077, + "step": 120 + }, + { + "epoch": 0.18945720250521922, + "grad_norm": 0.8741400241851807, + "learning_rate": 4.999608541969905e-06, + "loss": 0.912, + "step": 121 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 1.1172688007354736, + "learning_rate": 4.999570373686095e-06, + "loss": 0.8062, + "step": 122 + }, + { + "epoch": 0.19258872651356992, + "grad_norm": 0.940539538860321, + "learning_rate": 4.9995304303412905e-06, + "loss": 0.8944, + "step": 123 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 2.1082208156585693, + "learning_rate": 4.999488711963857e-06, + "loss": 0.9668, + "step": 124 + }, + { + "epoch": 0.19572025052192066, + "grad_norm": 1.0358761548995972, + "learning_rate": 4.99944521858342e-06, + "loss": 0.9032, + "step": 125 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 2.0804238319396973, + "learning_rate": 4.999399950230867e-06, + "loss": 0.9819, + "step": 126 + }, + { + "epoch": 0.1988517745302714, + "grad_norm": 1.0986278057098389, + "learning_rate": 4.9993529069383465e-06, + "loss": 0.9034, + "step": 127 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 1.2374722957611084, + "learning_rate": 4.999304088739263e-06, + "loss": 0.9255, + "step": 128 + }, + { + "epoch": 0.20198329853862212, + "grad_norm": 1.0443146228790283, + "learning_rate": 4.999253495668287e-06, + "loss": 0.9352, + "step": 129 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 1.009800910949707, + "learning_rate": 4.999201127761346e-06, + "loss": 0.8324, + "step": 130 + }, + { + "epoch": 0.20511482254697286, + "grad_norm": 0.892861008644104, + "learning_rate": 4.9991469850556295e-06, + "loss": 0.886, + "step": 131 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 1.1675922870635986, + "learning_rate": 4.999091067589587e-06, + "loss": 0.967, + "step": 132 + }, + { + "epoch": 0.2082463465553236, + "grad_norm": 1.0169954299926758, + "learning_rate": 4.999033375402926e-06, + "loss": 0.8498, + "step": 133 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 1.0006227493286133, + "learning_rate": 4.99897390853662e-06, + "loss": 0.8481, + "step": 134 + }, + { + "epoch": 0.21137787056367432, + "grad_norm": 0.9049347639083862, + "learning_rate": 4.9989126670328945e-06, + "loss": 0.9186, + "step": 135 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 1.2151603698730469, + "learning_rate": 4.9988496509352444e-06, + "loss": 0.9067, + "step": 136 + }, + { + "epoch": 0.21450939457202506, + "grad_norm": 1.0112006664276123, + "learning_rate": 4.998784860288417e-06, + "loss": 0.8843, + "step": 137 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 0.9316662549972534, + "learning_rate": 4.9987182951384264e-06, + "loss": 0.8635, + "step": 138 + }, + { + "epoch": 0.2176409185803758, + "grad_norm": 0.942743718624115, + "learning_rate": 4.9986499555325406e-06, + "loss": 0.8481, + "step": 139 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 1.0298817157745361, + "learning_rate": 4.998579841519292e-06, + "loss": 0.865, + "step": 140 + }, + { + "epoch": 0.22077244258872653, + "grad_norm": 1.1556729078292847, + "learning_rate": 4.998507953148472e-06, + "loss": 0.8731, + "step": 141 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 0.997958779335022, + "learning_rate": 4.9984342904711315e-06, + "loss": 0.7949, + "step": 142 + }, + { + "epoch": 0.22390396659707723, + "grad_norm": 0.9565998315811157, + "learning_rate": 4.998358853539582e-06, + "loss": 0.8875, + "step": 143 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 0.9667196273803711, + "learning_rate": 4.998281642407394e-06, + "loss": 0.8537, + "step": 144 + }, + { + "epoch": 0.22703549060542796, + "grad_norm": 0.9701303839683533, + "learning_rate": 4.998202657129401e-06, + "loss": 0.873, + "step": 145 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 0.9936159253120422, + "learning_rate": 4.998121897761692e-06, + "loss": 0.8915, + "step": 146 + }, + { + "epoch": 0.2301670146137787, + "grad_norm": 1.5054311752319336, + "learning_rate": 4.998039364361617e-06, + "loss": 0.9944, + "step": 147 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 1.5895891189575195, + "learning_rate": 4.99795505698779e-06, + "loss": 0.9282, + "step": 148 + }, + { + "epoch": 0.23329853862212943, + "grad_norm": 1.0141469240188599, + "learning_rate": 4.9978689757000785e-06, + "loss": 0.9166, + "step": 149 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 1.1720311641693115, + "learning_rate": 4.997781120559615e-06, + "loss": 0.9404, + "step": 150 + }, + { + "epoch": 0.23643006263048016, + "grad_norm": 0.9536433815956116, + "learning_rate": 4.997691491628789e-06, + "loss": 0.9339, + "step": 151 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 0.9407872557640076, + "learning_rate": 4.997600088971249e-06, + "loss": 0.9019, + "step": 152 + }, + { + "epoch": 0.2395615866388309, + "grad_norm": 0.9202889204025269, + "learning_rate": 4.997506912651906e-06, + "loss": 0.9598, + "step": 153 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 1.0630513429641724, + "learning_rate": 4.997411962736927e-06, + "loss": 0.8554, + "step": 154 + }, + { + "epoch": 0.24269311064718163, + "grad_norm": 0.8378801941871643, + "learning_rate": 4.997315239293742e-06, + "loss": 0.8583, + "step": 155 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 0.9442815184593201, + "learning_rate": 4.997216742391038e-06, + "loss": 0.8518, + "step": 156 + }, + { + "epoch": 0.24582463465553236, + "grad_norm": 0.9273308515548706, + "learning_rate": 4.997116472098763e-06, + "loss": 0.8942, + "step": 157 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 0.9028997421264648, + "learning_rate": 4.997014428488121e-06, + "loss": 0.8592, + "step": 158 + }, + { + "epoch": 0.2489561586638831, + "grad_norm": 0.9877567887306213, + "learning_rate": 4.996910611631582e-06, + "loss": 0.9316, + "step": 159 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 0.961592435836792, + "learning_rate": 4.996805021602868e-06, + "loss": 0.8875, + "step": 160 + }, + { + "epoch": 0.2520876826722338, + "grad_norm": 0.9923007488250732, + "learning_rate": 4.996697658476965e-06, + "loss": 0.8554, + "step": 161 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 0.956763744354248, + "learning_rate": 4.996588522330114e-06, + "loss": 0.9381, + "step": 162 + }, + { + "epoch": 0.25521920668058456, + "grad_norm": 0.919992208480835, + "learning_rate": 4.996477613239822e-06, + "loss": 0.9016, + "step": 163 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 0.9655581712722778, + "learning_rate": 4.996364931284847e-06, + "loss": 0.896, + "step": 164 + }, + { + "epoch": 0.25835073068893527, + "grad_norm": 0.9159606099128723, + "learning_rate": 4.996250476545212e-06, + "loss": 0.9475, + "step": 165 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 1.1147626638412476, + "learning_rate": 4.996134249102195e-06, + "loss": 0.8793, + "step": 166 + }, + { + "epoch": 0.26148225469728603, + "grad_norm": 1.112069010734558, + "learning_rate": 4.996016249038334e-06, + "loss": 0.8765, + "step": 167 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 1.0366944074630737, + "learning_rate": 4.995896476437428e-06, + "loss": 0.9356, + "step": 168 + }, + { + "epoch": 0.26461377870563674, + "grad_norm": 0.9494239091873169, + "learning_rate": 4.995774931384534e-06, + "loss": 0.8246, + "step": 169 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 0.9819200038909912, + "learning_rate": 4.995651613965964e-06, + "loss": 0.9403, + "step": 170 + }, + { + "epoch": 0.2677453027139875, + "grad_norm": 1.7329354286193848, + "learning_rate": 4.995526524269293e-06, + "loss": 0.9091, + "step": 171 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.1415139436721802, + "learning_rate": 4.995399662383352e-06, + "loss": 0.9106, + "step": 172 + }, + { + "epoch": 0.2708768267223382, + "grad_norm": 0.9471906423568726, + "learning_rate": 4.995271028398233e-06, + "loss": 0.8858, + "step": 173 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 1.096053123474121, + "learning_rate": 4.995140622405284e-06, + "loss": 0.7502, + "step": 174 + }, + { + "epoch": 0.2740083507306889, + "grad_norm": 1.285410761833191, + "learning_rate": 4.995008444497113e-06, + "loss": 0.9288, + "step": 175 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 0.9714595079421997, + "learning_rate": 4.994874494767585e-06, + "loss": 0.8824, + "step": 176 + }, + { + "epoch": 0.27713987473903967, + "grad_norm": 0.8399011492729187, + "learning_rate": 4.994738773311824e-06, + "loss": 0.8703, + "step": 177 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 0.9454224109649658, + "learning_rate": 4.994601280226212e-06, + "loss": 0.9627, + "step": 178 + }, + { + "epoch": 0.2802713987473904, + "grad_norm": 0.8983661532402039, + "learning_rate": 4.99446201560839e-06, + "loss": 0.9086, + "step": 179 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 0.9814153909683228, + "learning_rate": 4.994320979557256e-06, + "loss": 0.908, + "step": 180 + }, + { + "epoch": 0.28340292275574114, + "grad_norm": 0.8405355215072632, + "learning_rate": 4.9941781721729645e-06, + "loss": 0.8414, + "step": 181 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 0.950950026512146, + "learning_rate": 4.994033593556933e-06, + "loss": 0.9091, + "step": 182 + }, + { + "epoch": 0.28653444676409184, + "grad_norm": 1.1019747257232666, + "learning_rate": 4.99388724381183e-06, + "loss": 0.8305, + "step": 183 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 0.9339331984519958, + "learning_rate": 4.9937391230415875e-06, + "loss": 0.8542, + "step": 184 + }, + { + "epoch": 0.2896659707724426, + "grad_norm": 0.8813616633415222, + "learning_rate": 4.993589231351394e-06, + "loss": 0.848, + "step": 185 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 0.8916418552398682, + "learning_rate": 4.9934375688476905e-06, + "loss": 0.8032, + "step": 186 + }, + { + "epoch": 0.2927974947807933, + "grad_norm": 0.9891005754470825, + "learning_rate": 4.993284135638182e-06, + "loss": 0.7937, + "step": 187 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 0.9842082262039185, + "learning_rate": 4.99312893183183e-06, + "loss": 0.9532, + "step": 188 + }, + { + "epoch": 0.29592901878914407, + "grad_norm": 0.993478536605835, + "learning_rate": 4.99297195753885e-06, + "loss": 0.9406, + "step": 189 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 1.0654981136322021, + "learning_rate": 4.992813212870718e-06, + "loss": 0.907, + "step": 190 + }, + { + "epoch": 0.2990605427974948, + "grad_norm": 0.8937190771102905, + "learning_rate": 4.992652697940164e-06, + "loss": 0.9307, + "step": 191 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 0.9006052613258362, + "learning_rate": 4.992490412861178e-06, + "loss": 0.8808, + "step": 192 + }, + { + "epoch": 0.30219206680584554, + "grad_norm": 0.9601522088050842, + "learning_rate": 4.992326357749007e-06, + "loss": 0.8495, + "step": 193 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 0.9814243316650391, + "learning_rate": 4.992160532720153e-06, + "loss": 0.8204, + "step": 194 + }, + { + "epoch": 0.30532359081419624, + "grad_norm": 0.9660456776618958, + "learning_rate": 4.991992937892377e-06, + "loss": 0.8209, + "step": 195 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 0.8504270315170288, + "learning_rate": 4.991823573384695e-06, + "loss": 0.8415, + "step": 196 + }, + { + "epoch": 0.30845511482254695, + "grad_norm": 0.8733190894126892, + "learning_rate": 4.991652439317382e-06, + "loss": 0.8753, + "step": 197 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 0.9761794805526733, + "learning_rate": 4.991479535811967e-06, + "loss": 0.8923, + "step": 198 + }, + { + "epoch": 0.3115866388308977, + "grad_norm": 1.0400285720825195, + "learning_rate": 4.991304862991238e-06, + "loss": 0.899, + "step": 199 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 0.8561192750930786, + "learning_rate": 4.991128420979237e-06, + "loss": 0.9117, + "step": 200 + }, + { + "epoch": 0.3147181628392484, + "grad_norm": 0.8839634656906128, + "learning_rate": 4.990950209901263e-06, + "loss": 0.9022, + "step": 201 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 0.9569464921951294, + "learning_rate": 4.990770229883873e-06, + "loss": 0.9013, + "step": 202 + }, + { + "epoch": 0.3178496868475992, + "grad_norm": 1.0600861310958862, + "learning_rate": 4.990588481054881e-06, + "loss": 0.8387, + "step": 203 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 0.9492761492729187, + "learning_rate": 4.990404963543352e-06, + "loss": 0.8521, + "step": 204 + }, + { + "epoch": 0.3209812108559499, + "grad_norm": 0.9534817934036255, + "learning_rate": 4.990219677479612e-06, + "loss": 0.8009, + "step": 205 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 0.9918960928916931, + "learning_rate": 4.990032622995242e-06, + "loss": 0.8215, + "step": 206 + }, + { + "epoch": 0.32411273486430064, + "grad_norm": 0.8751630783081055, + "learning_rate": 4.989843800223077e-06, + "loss": 0.8744, + "step": 207 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 1.0034785270690918, + "learning_rate": 4.9896532092972085e-06, + "loss": 0.884, + "step": 208 + }, + { + "epoch": 0.32724425887265135, + "grad_norm": 1.2693432569503784, + "learning_rate": 4.989460850352984e-06, + "loss": 0.9689, + "step": 209 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 0.9923811554908752, + "learning_rate": 4.989266723527007e-06, + "loss": 0.9622, + "step": 210 + }, + { + "epoch": 0.3303757828810021, + "grad_norm": 0.9439074397087097, + "learning_rate": 4.989070828957137e-06, + "loss": 0.9155, + "step": 211 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.2564488649368286, + "learning_rate": 4.988873166782485e-06, + "loss": 0.8712, + "step": 212 + }, + { + "epoch": 0.3335073068893528, + "grad_norm": 1.1441881656646729, + "learning_rate": 4.988673737143423e-06, + "loss": 0.8895, + "step": 213 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 0.972155749797821, + "learning_rate": 4.988472540181574e-06, + "loss": 0.9158, + "step": 214 + }, + { + "epoch": 0.3366388308977035, + "grad_norm": 0.9536120295524597, + "learning_rate": 4.988269576039817e-06, + "loss": 0.7963, + "step": 215 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 0.8956594467163086, + "learning_rate": 4.988064844862288e-06, + "loss": 0.8947, + "step": 216 + }, + { + "epoch": 0.3397703549060543, + "grad_norm": 0.9503691792488098, + "learning_rate": 4.987858346794373e-06, + "loss": 0.7857, + "step": 217 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 0.9471456408500671, + "learning_rate": 4.987650081982719e-06, + "loss": 0.9118, + "step": 218 + }, + { + "epoch": 0.342901878914405, + "grad_norm": 0.9485713839530945, + "learning_rate": 4.987440050575222e-06, + "loss": 0.8188, + "step": 219 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 1.0775927305221558, + "learning_rate": 4.987228252721037e-06, + "loss": 0.9469, + "step": 220 + }, + { + "epoch": 0.34603340292275575, + "grad_norm": 0.9173988103866577, + "learning_rate": 4.9870146885705705e-06, + "loss": 0.8566, + "step": 221 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.008735179901123, + "learning_rate": 4.986799358275485e-06, + "loss": 0.8637, + "step": 222 + }, + { + "epoch": 0.34916492693110646, + "grad_norm": 0.8183198571205139, + "learning_rate": 4.986582261988695e-06, + "loss": 0.9321, + "step": 223 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 0.8951883316040039, + "learning_rate": 4.986363399864372e-06, + "loss": 0.8015, + "step": 224 + }, + { + "epoch": 0.3522964509394572, + "grad_norm": 0.9275704622268677, + "learning_rate": 4.986142772057941e-06, + "loss": 0.8242, + "step": 225 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 0.9548516273498535, + "learning_rate": 4.985920378726077e-06, + "loss": 0.7889, + "step": 226 + }, + { + "epoch": 0.3554279749478079, + "grad_norm": 1.1203585863113403, + "learning_rate": 4.985696220026714e-06, + "loss": 0.835, + "step": 227 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 0.9525771141052246, + "learning_rate": 4.985470296119038e-06, + "loss": 0.9162, + "step": 228 + }, + { + "epoch": 0.3585594989561587, + "grad_norm": 1.150404930114746, + "learning_rate": 4.985242607163488e-06, + "loss": 0.9456, + "step": 229 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.01652991771698, + "learning_rate": 4.985013153321755e-06, + "loss": 0.87, + "step": 230 + }, + { + "epoch": 0.3616910229645094, + "grad_norm": 0.9455218315124512, + "learning_rate": 4.984781934756786e-06, + "loss": 0.8466, + "step": 231 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 0.9125218391418457, + "learning_rate": 4.984548951632779e-06, + "loss": 0.8764, + "step": 232 + }, + { + "epoch": 0.36482254697286015, + "grad_norm": 2.0192980766296387, + "learning_rate": 4.984314204115187e-06, + "loss": 0.7866, + "step": 233 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.8118515014648438, + "learning_rate": 4.984077692370716e-06, + "loss": 0.8703, + "step": 234 + }, + { + "epoch": 0.36795407098121086, + "grad_norm": 1.2995024919509888, + "learning_rate": 4.983839416567322e-06, + "loss": 0.9459, + "step": 235 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 1.0611528158187866, + "learning_rate": 4.983599376874216e-06, + "loss": 0.8456, + "step": 236 + }, + { + "epoch": 0.37108559498956156, + "grad_norm": 1.2018043994903564, + "learning_rate": 4.983357573461863e-06, + "loss": 0.914, + "step": 237 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9091911315917969, + "learning_rate": 4.983114006501978e-06, + "loss": 0.791, + "step": 238 + }, + { + "epoch": 0.3742171189979123, + "grad_norm": 0.980873167514801, + "learning_rate": 4.982868676167528e-06, + "loss": 0.77, + "step": 239 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 0.8945695161819458, + "learning_rate": 4.982621582632735e-06, + "loss": 0.8038, + "step": 240 + }, + { + "epoch": 0.37734864300626303, + "grad_norm": 0.9325360655784607, + "learning_rate": 4.9823727260730705e-06, + "loss": 0.8908, + "step": 241 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 0.987630307674408, + "learning_rate": 4.98212210666526e-06, + "loss": 0.8642, + "step": 242 + }, + { + "epoch": 0.3804801670146138, + "grad_norm": 0.9151357412338257, + "learning_rate": 4.98186972458728e-06, + "loss": 0.8437, + "step": 243 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 0.9084850549697876, + "learning_rate": 4.981615580018358e-06, + "loss": 0.8775, + "step": 244 + }, + { + "epoch": 0.3836116910229645, + "grad_norm": 0.9786385297775269, + "learning_rate": 4.981359673138974e-06, + "loss": 0.8251, + "step": 245 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 0.8947877883911133, + "learning_rate": 4.98110200413086e-06, + "loss": 0.7709, + "step": 246 + }, + { + "epoch": 0.38674321503131526, + "grad_norm": 0.967019259929657, + "learning_rate": 4.9808425731769985e-06, + "loss": 0.8954, + "step": 247 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 1.1066704988479614, + "learning_rate": 4.980581380461622e-06, + "loss": 0.8544, + "step": 248 + }, + { + "epoch": 0.38987473903966596, + "grad_norm": 0.9750981330871582, + "learning_rate": 4.980318426170218e-06, + "loss": 0.8354, + "step": 249 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 0.8727232217788696, + "learning_rate": 4.980053710489521e-06, + "loss": 0.8911, + "step": 250 + }, + { + "epoch": 0.3930062630480167, + "grad_norm": 0.9337188601493835, + "learning_rate": 4.979787233607518e-06, + "loss": 0.8567, + "step": 251 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 0.890701174736023, + "learning_rate": 4.979518995713448e-06, + "loss": 0.8327, + "step": 252 + }, + { + "epoch": 0.39613778705636743, + "grad_norm": 0.9973195195198059, + "learning_rate": 4.979248996997797e-06, + "loss": 0.9065, + "step": 253 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 0.9894055128097534, + "learning_rate": 4.978977237652306e-06, + "loss": 0.8729, + "step": 254 + }, + { + "epoch": 0.39926931106471814, + "grad_norm": 1.0398536920547485, + "learning_rate": 4.978703717869961e-06, + "loss": 0.861, + "step": 255 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.1713942289352417, + "learning_rate": 4.978428437845003e-06, + "loss": 0.8183, + "step": 256 + }, + { + "epoch": 0.4024008350730689, + "grad_norm": 0.9868776202201843, + "learning_rate": 4.9781513977729225e-06, + "loss": 0.941, + "step": 257 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 0.9260863065719604, + "learning_rate": 4.977872597850455e-06, + "loss": 0.8963, + "step": 258 + }, + { + "epoch": 0.4055323590814196, + "grad_norm": 0.9553239941596985, + "learning_rate": 4.977592038275592e-06, + "loss": 0.8102, + "step": 259 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 0.9426907300949097, + "learning_rate": 4.977309719247571e-06, + "loss": 0.89, + "step": 260 + }, + { + "epoch": 0.40866388308977036, + "grad_norm": 0.9691527485847473, + "learning_rate": 4.97702564096688e-06, + "loss": 0.7967, + "step": 261 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 0.9917487502098083, + "learning_rate": 4.976739803635256e-06, + "loss": 0.8327, + "step": 262 + }, + { + "epoch": 0.41179540709812107, + "grad_norm": 0.992340087890625, + "learning_rate": 4.976452207455686e-06, + "loss": 0.9462, + "step": 263 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 0.9554483890533447, + "learning_rate": 4.9761628526324035e-06, + "loss": 0.8294, + "step": 264 + }, + { + "epoch": 0.41492693110647183, + "grad_norm": 0.9318789839744568, + "learning_rate": 4.975871739370895e-06, + "loss": 0.8079, + "step": 265 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 0.9158215522766113, + "learning_rate": 4.975578867877893e-06, + "loss": 0.8632, + "step": 266 + }, + { + "epoch": 0.41805845511482254, + "grad_norm": 0.9444370865821838, + "learning_rate": 4.975284238361378e-06, + "loss": 0.8897, + "step": 267 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 0.9949231743812561, + "learning_rate": 4.974987851030581e-06, + "loss": 0.7905, + "step": 268 + }, + { + "epoch": 0.4211899791231733, + "grad_norm": 0.9819440245628357, + "learning_rate": 4.974689706095981e-06, + "loss": 0.9183, + "step": 269 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0787206888198853, + "learning_rate": 4.974389803769304e-06, + "loss": 0.7739, + "step": 270 + }, + { + "epoch": 0.424321503131524, + "grad_norm": 0.8871361017227173, + "learning_rate": 4.9740881442635235e-06, + "loss": 0.9454, + "step": 271 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 0.9125893115997314, + "learning_rate": 4.973784727792863e-06, + "loss": 0.8017, + "step": 272 + }, + { + "epoch": 0.42745302713987476, + "grad_norm": 0.9102838635444641, + "learning_rate": 4.973479554572793e-06, + "loss": 0.8307, + "step": 273 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 0.8780474662780762, + "learning_rate": 4.973172624820031e-06, + "loss": 0.9402, + "step": 274 + }, + { + "epoch": 0.43058455114822547, + "grad_norm": 1.0059378147125244, + "learning_rate": 4.972863938752541e-06, + "loss": 0.7438, + "step": 275 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.9684615135192871, + "learning_rate": 4.972553496589537e-06, + "loss": 0.8897, + "step": 276 + }, + { + "epoch": 0.4337160751565762, + "grad_norm": 0.9615057110786438, + "learning_rate": 4.972241298551477e-06, + "loss": 0.8157, + "step": 277 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8589702844619751, + "learning_rate": 4.971927344860068e-06, + "loss": 0.7813, + "step": 278 + }, + { + "epoch": 0.43684759916492694, + "grad_norm": 0.900324285030365, + "learning_rate": 4.971611635738263e-06, + "loss": 0.7534, + "step": 279 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 0.8881833553314209, + "learning_rate": 4.971294171410261e-06, + "loss": 0.8255, + "step": 280 + }, + { + "epoch": 0.43997912317327764, + "grad_norm": 0.9723751544952393, + "learning_rate": 4.9709749521015085e-06, + "loss": 0.8796, + "step": 281 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 0.9341713190078735, + "learning_rate": 4.9706539780386995e-06, + "loss": 0.7429, + "step": 282 + }, + { + "epoch": 0.4431106471816284, + "grad_norm": 0.9354205131530762, + "learning_rate": 4.97033124944977e-06, + "loss": 0.8545, + "step": 283 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 0.9294624328613281, + "learning_rate": 4.970006766563906e-06, + "loss": 0.7783, + "step": 284 + }, + { + "epoch": 0.4462421711899791, + "grad_norm": 0.9584887027740479, + "learning_rate": 4.969680529611538e-06, + "loss": 0.8068, + "step": 285 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 0.9525735378265381, + "learning_rate": 4.9693525388243424e-06, + "loss": 0.8055, + "step": 286 + }, + { + "epoch": 0.44937369519832987, + "grad_norm": 0.8845317959785461, + "learning_rate": 4.96902279443524e-06, + "loss": 0.8318, + "step": 287 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 0.8932276964187622, + "learning_rate": 4.9686912966783965e-06, + "loss": 0.8185, + "step": 288 + }, + { + "epoch": 0.4525052192066806, + "grad_norm": 0.9881951808929443, + "learning_rate": 4.968358045789225e-06, + "loss": 0.9001, + "step": 289 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 0.995212197303772, + "learning_rate": 4.968023042004384e-06, + "loss": 0.75, + "step": 290 + }, + { + "epoch": 0.45563674321503134, + "grad_norm": 0.9997846484184265, + "learning_rate": 4.967686285561773e-06, + "loss": 0.8993, + "step": 291 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 0.9438545107841492, + "learning_rate": 4.967347776700538e-06, + "loss": 0.8964, + "step": 292 + }, + { + "epoch": 0.45876826722338204, + "grad_norm": 0.9164395332336426, + "learning_rate": 4.967007515661072e-06, + "loss": 0.8671, + "step": 293 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.895302951335907, + "learning_rate": 4.966665502685009e-06, + "loss": 0.8785, + "step": 294 + }, + { + "epoch": 0.4618997912317328, + "grad_norm": 0.9122863411903381, + "learning_rate": 4.966321738015227e-06, + "loss": 0.8499, + "step": 295 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.0097336769104004, + "learning_rate": 4.965976221895852e-06, + "loss": 0.83, + "step": 296 + }, + { + "epoch": 0.4650313152400835, + "grad_norm": 1.0664499998092651, + "learning_rate": 4.965628954572249e-06, + "loss": 0.7799, + "step": 297 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 0.9602341055870056, + "learning_rate": 4.96527993629103e-06, + "loss": 0.8061, + "step": 298 + }, + { + "epoch": 0.4681628392484342, + "grad_norm": 1.0133620500564575, + "learning_rate": 4.964929167300048e-06, + "loss": 0.9148, + "step": 299 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9351868033409119, + "learning_rate": 4.964576647848401e-06, + "loss": 0.83, + "step": 300 + }, + { + "epoch": 0.471294363256785, + "grad_norm": 1.0078860521316528, + "learning_rate": 4.964222378186429e-06, + "loss": 0.8391, + "step": 301 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 0.8966493606567383, + "learning_rate": 4.963866358565716e-06, + "loss": 0.8677, + "step": 302 + }, + { + "epoch": 0.4744258872651357, + "grad_norm": 0.9283657670021057, + "learning_rate": 4.963508589239087e-06, + "loss": 0.8787, + "step": 303 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 0.9540023803710938, + "learning_rate": 4.963149070460611e-06, + "loss": 0.8667, + "step": 304 + }, + { + "epoch": 0.47755741127348644, + "grad_norm": 0.9580283164978027, + "learning_rate": 4.9627878024855995e-06, + "loss": 0.8442, + "step": 305 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.9008278846740723, + "learning_rate": 4.962424785570605e-06, + "loss": 0.8605, + "step": 306 + }, + { + "epoch": 0.48068893528183715, + "grad_norm": 0.9681883454322815, + "learning_rate": 4.962060019973423e-06, + "loss": 0.7865, + "step": 307 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 0.8492597341537476, + "learning_rate": 4.9616935059530915e-06, + "loss": 0.8578, + "step": 308 + }, + { + "epoch": 0.4838204592901879, + "grad_norm": 0.9323164820671082, + "learning_rate": 4.9613252437698865e-06, + "loss": 0.8491, + "step": 309 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 0.9417833685874939, + "learning_rate": 4.960955233685331e-06, + "loss": 0.8108, + "step": 310 + }, + { + "epoch": 0.4869519832985386, + "grad_norm": 0.901364803314209, + "learning_rate": 4.960583475962185e-06, + "loss": 0.8958, + "step": 311 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8929979205131531, + "learning_rate": 4.96020997086445e-06, + "loss": 0.8237, + "step": 312 + }, + { + "epoch": 0.4900835073068894, + "grad_norm": 0.9422337412834167, + "learning_rate": 4.9598347186573716e-06, + "loss": 0.8024, + "step": 313 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 0.9165504574775696, + "learning_rate": 4.959457719607432e-06, + "loss": 0.8124, + "step": 314 + }, + { + "epoch": 0.4932150313152401, + "grad_norm": 1.0233370065689087, + "learning_rate": 4.959078973982356e-06, + "loss": 0.788, + "step": 315 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9375972151756287, + "learning_rate": 4.958698482051109e-06, + "loss": 0.839, + "step": 316 + }, + { + "epoch": 0.4963465553235908, + "grad_norm": 0.8805437684059143, + "learning_rate": 4.958316244083895e-06, + "loss": 0.7445, + "step": 317 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.0231893062591553, + "learning_rate": 4.95793226035216e-06, + "loss": 0.8257, + "step": 318 + }, + { + "epoch": 0.49947807933194155, + "grad_norm": 0.9125928282737732, + "learning_rate": 4.957546531128589e-06, + "loss": 0.7699, + "step": 319 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.8844016790390015, + "learning_rate": 4.957159056687104e-06, + "loss": 0.8784, + "step": 320 + }, + { + "epoch": 0.5026096033402923, + "grad_norm": 0.9770572781562805, + "learning_rate": 4.956769837302871e-06, + "loss": 0.8611, + "step": 321 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.9413725733757019, + "learning_rate": 4.956378873252293e-06, + "loss": 0.8056, + "step": 322 + }, + { + "epoch": 0.505741127348643, + "grad_norm": 0.9058850407600403, + "learning_rate": 4.955986164813008e-06, + "loss": 0.8162, + "step": 323 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 0.893100917339325, + "learning_rate": 4.955591712263901e-06, + "loss": 0.7554, + "step": 324 + }, + { + "epoch": 0.5088726513569938, + "grad_norm": 0.8967075943946838, + "learning_rate": 4.955195515885089e-06, + "loss": 0.8143, + "step": 325 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.9542785286903381, + "learning_rate": 4.954797575957929e-06, + "loss": 0.8102, + "step": 326 + }, + { + "epoch": 0.5120041753653445, + "grad_norm": 1.006451964378357, + "learning_rate": 4.954397892765018e-06, + "loss": 0.86, + "step": 327 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 0.9964715838432312, + "learning_rate": 4.953996466590189e-06, + "loss": 0.7166, + "step": 328 + }, + { + "epoch": 0.5151356993736952, + "grad_norm": 0.9568808674812317, + "learning_rate": 4.953593297718512e-06, + "loss": 0.8753, + "step": 329 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.9435825347900391, + "learning_rate": 4.953188386436298e-06, + "loss": 0.8961, + "step": 330 + }, + { + "epoch": 0.5182672233820459, + "grad_norm": 0.9386457800865173, + "learning_rate": 4.952781733031092e-06, + "loss": 0.8468, + "step": 331 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 0.9916554689407349, + "learning_rate": 4.952373337791678e-06, + "loss": 0.8274, + "step": 332 + }, + { + "epoch": 0.5213987473903967, + "grad_norm": 0.9351572394371033, + "learning_rate": 4.9519632010080765e-06, + "loss": 0.7964, + "step": 333 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9936269521713257, + "learning_rate": 4.951551322971545e-06, + "loss": 0.8354, + "step": 334 + }, + { + "epoch": 0.5245302713987474, + "grad_norm": 0.9678221940994263, + "learning_rate": 4.951137703974576e-06, + "loss": 0.9115, + "step": 335 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 0.9453133940696716, + "learning_rate": 4.9507223443109005e-06, + "loss": 0.7725, + "step": 336 + }, + { + "epoch": 0.5276617954070981, + "grad_norm": 0.9577226638793945, + "learning_rate": 4.950305244275484e-06, + "loss": 0.8679, + "step": 337 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 0.998489260673523, + "learning_rate": 4.94988640416453e-06, + "loss": 0.7505, + "step": 338 + }, + { + "epoch": 0.5307933194154488, + "grad_norm": 0.976026713848114, + "learning_rate": 4.949465824275475e-06, + "loss": 0.9083, + "step": 339 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.0041502714157104, + "learning_rate": 4.9490435049069925e-06, + "loss": 0.8815, + "step": 340 + }, + { + "epoch": 0.5339248434237995, + "grad_norm": 0.9839801788330078, + "learning_rate": 4.948619446358991e-06, + "loss": 0.8517, + "step": 341 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 0.9995140433311462, + "learning_rate": 4.948193648932616e-06, + "loss": 0.8157, + "step": 342 + }, + { + "epoch": 0.5370563674321504, + "grad_norm": 1.0782225131988525, + "learning_rate": 4.947766112930243e-06, + "loss": 0.9206, + "step": 343 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9981802105903625, + "learning_rate": 4.947336838655487e-06, + "loss": 0.8389, + "step": 344 + }, + { + "epoch": 0.5401878914405011, + "grad_norm": 0.9261236190795898, + "learning_rate": 4.946905826413195e-06, + "loss": 0.8035, + "step": 345 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.9195284247398376, + "learning_rate": 4.946473076509449e-06, + "loss": 0.8265, + "step": 346 + }, + { + "epoch": 0.5433194154488518, + "grad_norm": 0.9535123109817505, + "learning_rate": 4.946038589251566e-06, + "loss": 0.7528, + "step": 347 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 0.9969532489776611, + "learning_rate": 4.9456023649480935e-06, + "loss": 0.7787, + "step": 348 + }, + { + "epoch": 0.5464509394572025, + "grad_norm": 0.9612399935722351, + "learning_rate": 4.945164403908816e-06, + "loss": 0.8298, + "step": 349 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 0.9541627764701843, + "learning_rate": 4.944724706444749e-06, + "loss": 0.8039, + "step": 350 + }, + { + "epoch": 0.5495824634655533, + "grad_norm": 0.9916557669639587, + "learning_rate": 4.944283272868143e-06, + "loss": 0.9012, + "step": 351 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 0.8807661533355713, + "learning_rate": 4.94384010349248e-06, + "loss": 0.9005, + "step": 352 + }, + { + "epoch": 0.552713987473904, + "grad_norm": 1.0179708003997803, + "learning_rate": 4.943395198632476e-06, + "loss": 0.8845, + "step": 353 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.923014223575592, + "learning_rate": 4.942948558604077e-06, + "loss": 0.8173, + "step": 354 + }, + { + "epoch": 0.5558455114822547, + "grad_norm": 0.9318749308586121, + "learning_rate": 4.942500183724463e-06, + "loss": 0.8828, + "step": 355 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.0907000303268433, + "learning_rate": 4.942050074312048e-06, + "loss": 0.7951, + "step": 356 + }, + { + "epoch": 0.5589770354906054, + "grad_norm": 0.9811620712280273, + "learning_rate": 4.9415982306864744e-06, + "loss": 0.8826, + "step": 357 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9888150691986084, + "learning_rate": 4.9411446531686156e-06, + "loss": 0.8338, + "step": 358 + }, + { + "epoch": 0.5621085594989561, + "grad_norm": 0.9389257431030273, + "learning_rate": 4.9406893420805804e-06, + "loss": 0.8316, + "step": 359 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.9500603675842285, + "learning_rate": 4.940232297745705e-06, + "loss": 0.8764, + "step": 360 + }, + { + "epoch": 0.5652400835073069, + "grad_norm": 0.9748769998550415, + "learning_rate": 4.939773520488559e-06, + "loss": 0.8308, + "step": 361 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 0.9179681539535522, + "learning_rate": 4.93931301063494e-06, + "loss": 0.789, + "step": 362 + }, + { + "epoch": 0.5683716075156576, + "grad_norm": 0.8609325289726257, + "learning_rate": 4.938850768511878e-06, + "loss": 0.7368, + "step": 363 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 0.9375531673431396, + "learning_rate": 4.9383867944476325e-06, + "loss": 0.8164, + "step": 364 + }, + { + "epoch": 0.5715031315240083, + "grad_norm": 0.9386453032493591, + "learning_rate": 4.937921088771694e-06, + "loss": 0.7898, + "step": 365 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.025040864944458, + "learning_rate": 4.937453651814781e-06, + "loss": 0.8465, + "step": 366 + }, + { + "epoch": 0.574634655532359, + "grad_norm": 1.1037245988845825, + "learning_rate": 4.936984483908842e-06, + "loss": 0.7935, + "step": 367 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.230821132659912, + "learning_rate": 4.936513585387055e-06, + "loss": 0.8043, + "step": 368 + }, + { + "epoch": 0.5777661795407099, + "grad_norm": 1.090235948562622, + "learning_rate": 4.9360409565838265e-06, + "loss": 0.8726, + "step": 369 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8977810144424438, + "learning_rate": 4.935566597834793e-06, + "loss": 0.8667, + "step": 370 + }, + { + "epoch": 0.5808977035490606, + "grad_norm": 0.96401447057724, + "learning_rate": 4.935090509476818e-06, + "loss": 0.8494, + "step": 371 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 0.9290149211883545, + "learning_rate": 4.934612691847995e-06, + "loss": 0.7678, + "step": 372 + }, + { + "epoch": 0.5840292275574113, + "grad_norm": 0.9765584468841553, + "learning_rate": 4.934133145287644e-06, + "loss": 0.7947, + "step": 373 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 0.8939430117607117, + "learning_rate": 4.933651870136313e-06, + "loss": 0.885, + "step": 374 + }, + { + "epoch": 0.587160751565762, + "grad_norm": 0.9512760043144226, + "learning_rate": 4.933168866735777e-06, + "loss": 0.8863, + "step": 375 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 1.0334441661834717, + "learning_rate": 4.932684135429041e-06, + "loss": 0.8362, + "step": 376 + }, + { + "epoch": 0.5902922755741128, + "grad_norm": 0.9402525424957275, + "learning_rate": 4.932197676560334e-06, + "loss": 0.8758, + "step": 377 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 0.9615777730941772, + "learning_rate": 4.931709490475113e-06, + "loss": 0.8324, + "step": 378 + }, + { + "epoch": 0.5934237995824635, + "grad_norm": 0.9407138228416443, + "learning_rate": 4.931219577520061e-06, + "loss": 0.8454, + "step": 379 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.0415815114974976, + "learning_rate": 4.930727938043091e-06, + "loss": 0.8649, + "step": 380 + }, + { + "epoch": 0.5965553235908142, + "grad_norm": 1.0946165323257446, + "learning_rate": 4.9302345723933344e-06, + "loss": 0.7794, + "step": 381 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.0532523393630981, + "learning_rate": 4.929739480921157e-06, + "loss": 0.8555, + "step": 382 + }, + { + "epoch": 0.5996868475991649, + "grad_norm": 0.9441399574279785, + "learning_rate": 4.929242663978144e-06, + "loss": 0.7613, + "step": 383 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 0.9970793128013611, + "learning_rate": 4.92874412191711e-06, + "loss": 0.9037, + "step": 384 + }, + { + "epoch": 0.6028183716075156, + "grad_norm": 0.9777354001998901, + "learning_rate": 4.928243855092091e-06, + "loss": 0.9044, + "step": 385 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 0.9507855772972107, + "learning_rate": 4.92774186385835e-06, + "loss": 0.8359, + "step": 386 + }, + { + "epoch": 0.6059498956158664, + "grad_norm": 0.9827553033828735, + "learning_rate": 4.9272381485723765e-06, + "loss": 0.7938, + "step": 387 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 0.8896475434303284, + "learning_rate": 4.926732709591879e-06, + "loss": 0.8545, + "step": 388 + }, + { + "epoch": 0.6090814196242171, + "grad_norm": 0.9817705750465393, + "learning_rate": 4.926225547275795e-06, + "loss": 0.8456, + "step": 389 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 0.9646008610725403, + "learning_rate": 4.925716661984283e-06, + "loss": 0.7906, + "step": 390 + }, + { + "epoch": 0.6122129436325678, + "grad_norm": 0.8940444588661194, + "learning_rate": 4.9252060540787275e-06, + "loss": 0.8371, + "step": 391 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.0812394618988037, + "learning_rate": 4.924693723921734e-06, + "loss": 0.8056, + "step": 392 + }, + { + "epoch": 0.6153444676409185, + "grad_norm": 0.9953856468200684, + "learning_rate": 4.924179671877132e-06, + "loss": 0.8651, + "step": 393 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 0.9760703444480896, + "learning_rate": 4.9236638983099735e-06, + "loss": 0.845, + "step": 394 + }, + { + "epoch": 0.6184759916492694, + "grad_norm": 1.0127615928649902, + "learning_rate": 4.923146403586533e-06, + "loss": 0.7987, + "step": 395 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 0.8771396279335022, + "learning_rate": 4.9226271880743086e-06, + "loss": 0.8092, + "step": 396 + }, + { + "epoch": 0.6216075156576201, + "grad_norm": 0.8906831741333008, + "learning_rate": 4.9221062521420196e-06, + "loss": 0.8039, + "step": 397 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.0218392610549927, + "learning_rate": 4.9215835961596045e-06, + "loss": 0.8354, + "step": 398 + }, + { + "epoch": 0.6247390396659708, + "grad_norm": 1.0382038354873657, + "learning_rate": 4.921059220498227e-06, + "loss": 0.8021, + "step": 399 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 0.8997588157653809, + "learning_rate": 4.920533125530272e-06, + "loss": 0.8716, + "step": 400 + }, + { + "epoch": 0.6278705636743215, + "grad_norm": 0.8725724220275879, + "learning_rate": 4.920005311629342e-06, + "loss": 0.7934, + "step": 401 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.9289800524711609, + "learning_rate": 4.9194757791702625e-06, + "loss": 0.812, + "step": 402 + }, + { + "epoch": 0.6310020876826722, + "grad_norm": 0.9969078898429871, + "learning_rate": 4.918944528529079e-06, + "loss": 0.866, + "step": 403 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 0.9544941186904907, + "learning_rate": 4.918411560083058e-06, + "loss": 0.8035, + "step": 404 + }, + { + "epoch": 0.634133611691023, + "grad_norm": 0.9678053855895996, + "learning_rate": 4.917876874210686e-06, + "loss": 0.8254, + "step": 405 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 0.9782965779304504, + "learning_rate": 4.917340471291666e-06, + "loss": 0.7954, + "step": 406 + }, + { + "epoch": 0.6372651356993737, + "grad_norm": 0.9380600452423096, + "learning_rate": 4.916802351706923e-06, + "loss": 0.8287, + "step": 407 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 0.9455891847610474, + "learning_rate": 4.916262515838603e-06, + "loss": 0.8714, + "step": 408 + }, + { + "epoch": 0.6403966597077244, + "grad_norm": 0.8937395811080933, + "learning_rate": 4.915720964070065e-06, + "loss": 0.7618, + "step": 409 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 1.0304689407348633, + "learning_rate": 4.915177696785894e-06, + "loss": 0.8436, + "step": 410 + }, + { + "epoch": 0.6435281837160751, + "grad_norm": 0.9623110294342041, + "learning_rate": 4.914632714371885e-06, + "loss": 0.865, + "step": 411 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 0.9212867021560669, + "learning_rate": 4.914086017215059e-06, + "loss": 0.8232, + "step": 412 + }, + { + "epoch": 0.6466597077244259, + "grad_norm": 0.9817205667495728, + "learning_rate": 4.9135376057036485e-06, + "loss": 0.7926, + "step": 413 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 0.9576719403266907, + "learning_rate": 4.912987480227108e-06, + "loss": 0.8812, + "step": 414 + }, + { + "epoch": 0.6497912317327766, + "grad_norm": 0.9415407180786133, + "learning_rate": 4.912435641176106e-06, + "loss": 0.8985, + "step": 415 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 0.9084275364875793, + "learning_rate": 4.91188208894253e-06, + "loss": 0.8085, + "step": 416 + }, + { + "epoch": 0.6529227557411273, + "grad_norm": 0.8890969753265381, + "learning_rate": 4.911326823919482e-06, + "loss": 0.8127, + "step": 417 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 0.8727678656578064, + "learning_rate": 4.9107698465012815e-06, + "loss": 0.8656, + "step": 418 + }, + { + "epoch": 0.656054279749478, + "grad_norm": 1.0174667835235596, + "learning_rate": 4.910211157083465e-06, + "loss": 0.8358, + "step": 419 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.045121431350708, + "learning_rate": 4.909650756062782e-06, + "loss": 0.8239, + "step": 420 + }, + { + "epoch": 0.6591858037578288, + "grad_norm": 1.0688676834106445, + "learning_rate": 4.909088643837202e-06, + "loss": 0.9246, + "step": 421 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.9371877908706665, + "learning_rate": 4.908524820805905e-06, + "loss": 0.7862, + "step": 422 + }, + { + "epoch": 0.6623173277661796, + "grad_norm": 0.9567264318466187, + "learning_rate": 4.907959287369288e-06, + "loss": 0.8164, + "step": 423 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 0.9109166860580444, + "learning_rate": 4.907392043928964e-06, + "loss": 0.7016, + "step": 424 + }, + { + "epoch": 0.6654488517745303, + "grad_norm": 1.0378203392028809, + "learning_rate": 4.906823090887757e-06, + "loss": 0.8986, + "step": 425 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 0.9871951341629028, + "learning_rate": 4.906252428649708e-06, + "loss": 0.8366, + "step": 426 + }, + { + "epoch": 0.668580375782881, + "grad_norm": 0.8682780265808105, + "learning_rate": 4.905680057620072e-06, + "loss": 0.738, + "step": 427 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.8834647536277771, + "learning_rate": 4.9051059782053125e-06, + "loss": 0.8304, + "step": 428 + }, + { + "epoch": 0.6717118997912317, + "grad_norm": 1.0133378505706787, + "learning_rate": 4.904530190813113e-06, + "loss": 0.9369, + "step": 429 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 0.9967373013496399, + "learning_rate": 4.9039526958523676e-06, + "loss": 0.782, + "step": 430 + }, + { + "epoch": 0.6748434237995825, + "grad_norm": 1.0258032083511353, + "learning_rate": 4.90337349373318e-06, + "loss": 0.8214, + "step": 431 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 0.9814366102218628, + "learning_rate": 4.90279258486687e-06, + "loss": 0.8173, + "step": 432 + }, + { + "epoch": 0.6779749478079332, + "grad_norm": 0.9351440668106079, + "learning_rate": 4.902209969665966e-06, + "loss": 0.8815, + "step": 433 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 0.9490398168563843, + "learning_rate": 4.901625648544212e-06, + "loss": 0.8295, + "step": 434 + }, + { + "epoch": 0.6811064718162839, + "grad_norm": 1.0183534622192383, + "learning_rate": 4.9010396219165614e-06, + "loss": 0.8538, + "step": 435 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 0.9546058773994446, + "learning_rate": 4.900451890199179e-06, + "loss": 0.8385, + "step": 436 + }, + { + "epoch": 0.6842379958246346, + "grad_norm": 1.076120138168335, + "learning_rate": 4.8998624538094394e-06, + "loss": 0.8958, + "step": 437 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.8935732841491699, + "learning_rate": 4.899271313165929e-06, + "loss": 0.8674, + "step": 438 + }, + { + "epoch": 0.6873695198329853, + "grad_norm": 0.9574604630470276, + "learning_rate": 4.898678468688445e-06, + "loss": 0.8245, + "step": 439 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.931751549243927, + "learning_rate": 4.898083920797993e-06, + "loss": 0.8409, + "step": 440 + }, + { + "epoch": 0.6905010438413361, + "grad_norm": 0.937874436378479, + "learning_rate": 4.89748766991679e-06, + "loss": 0.8222, + "step": 441 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9606375694274902, + "learning_rate": 4.896889716468259e-06, + "loss": 0.8123, + "step": 442 + }, + { + "epoch": 0.6936325678496869, + "grad_norm": 0.8692249059677124, + "learning_rate": 4.896290060877038e-06, + "loss": 0.8532, + "step": 443 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 0.9597892165184021, + "learning_rate": 4.895688703568968e-06, + "loss": 0.8365, + "step": 444 + }, + { + "epoch": 0.6967640918580376, + "grad_norm": 0.958342432975769, + "learning_rate": 4.895085644971102e-06, + "loss": 0.8535, + "step": 445 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.9603601098060608, + "learning_rate": 4.894480885511699e-06, + "loss": 0.8227, + "step": 446 + }, + { + "epoch": 0.6998956158663883, + "grad_norm": 1.0010526180267334, + "learning_rate": 4.893874425620226e-06, + "loss": 0.8785, + "step": 447 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.0061219930648804, + "learning_rate": 4.893266265727361e-06, + "loss": 0.8336, + "step": 448 + }, + { + "epoch": 0.7030271398747391, + "grad_norm": 0.8967173099517822, + "learning_rate": 4.892656406264984e-06, + "loss": 0.8305, + "step": 449 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.962729275226593, + "learning_rate": 4.892044847666186e-06, + "loss": 0.7843, + "step": 450 + }, + { + "epoch": 0.7061586638830898, + "grad_norm": 1.0067342519760132, + "learning_rate": 4.891431590365262e-06, + "loss": 0.8111, + "step": 451 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 1.0252268314361572, + "learning_rate": 4.890816634797716e-06, + "loss": 0.8253, + "step": 452 + }, + { + "epoch": 0.7092901878914405, + "grad_norm": 0.9098564386367798, + "learning_rate": 4.890199981400255e-06, + "loss": 0.8526, + "step": 453 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 0.893135130405426, + "learning_rate": 4.889581630610793e-06, + "loss": 0.8217, + "step": 454 + }, + { + "epoch": 0.7124217118997912, + "grad_norm": 0.9225817322731018, + "learning_rate": 4.888961582868453e-06, + "loss": 0.8296, + "step": 455 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 0.973386824131012, + "learning_rate": 4.888339838613555e-06, + "loss": 0.7429, + "step": 456 + }, + { + "epoch": 0.715553235908142, + "grad_norm": 0.9829993844032288, + "learning_rate": 4.8877163982876306e-06, + "loss": 0.7405, + "step": 457 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 0.927728533744812, + "learning_rate": 4.887091262333414e-06, + "loss": 0.8257, + "step": 458 + }, + { + "epoch": 0.7186847599164927, + "grad_norm": 0.9419897794723511, + "learning_rate": 4.886464431194844e-06, + "loss": 0.7762, + "step": 459 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 0.9358823299407959, + "learning_rate": 4.885835905317061e-06, + "loss": 0.7794, + "step": 460 + }, + { + "epoch": 0.7218162839248434, + "grad_norm": 1.0290783643722534, + "learning_rate": 4.88520568514641e-06, + "loss": 0.8723, + "step": 461 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 0.9909068942070007, + "learning_rate": 4.884573771130442e-06, + "loss": 0.8302, + "step": 462 + }, + { + "epoch": 0.7249478079331941, + "grad_norm": 0.9972653985023499, + "learning_rate": 4.883940163717906e-06, + "loss": 0.7961, + "step": 463 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.9245638251304626, + "learning_rate": 4.883304863358757e-06, + "loss": 0.8057, + "step": 464 + }, + { + "epoch": 0.7280793319415448, + "grad_norm": 1.0097148418426514, + "learning_rate": 4.882667870504153e-06, + "loss": 0.7296, + "step": 465 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 0.9792670607566833, + "learning_rate": 4.882029185606448e-06, + "loss": 0.8006, + "step": 466 + }, + { + "epoch": 0.7312108559498957, + "grad_norm": 0.9308120608329773, + "learning_rate": 4.881388809119206e-06, + "loss": 0.7957, + "step": 467 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 0.9472301602363586, + "learning_rate": 4.880746741497187e-06, + "loss": 0.8984, + "step": 468 + }, + { + "epoch": 0.7343423799582464, + "grad_norm": 0.9898937344551086, + "learning_rate": 4.880102983196351e-06, + "loss": 0.8365, + "step": 469 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 0.9376612901687622, + "learning_rate": 4.879457534673863e-06, + "loss": 0.7838, + "step": 470 + }, + { + "epoch": 0.7374739039665971, + "grad_norm": 1.030616283416748, + "learning_rate": 4.878810396388085e-06, + "loss": 0.7698, + "step": 471 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.9583387970924377, + "learning_rate": 4.87816156879858e-06, + "loss": 0.7381, + "step": 472 + }, + { + "epoch": 0.7406054279749478, + "grad_norm": 1.00180184841156, + "learning_rate": 4.87751105236611e-06, + "loss": 0.8497, + "step": 473 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.9382562637329102, + "learning_rate": 4.876858847552638e-06, + "loss": 0.8698, + "step": 474 + }, + { + "epoch": 0.7437369519832986, + "grad_norm": 1.0104063749313354, + "learning_rate": 4.8762049548213255e-06, + "loss": 0.8444, + "step": 475 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 0.9321765303611755, + "learning_rate": 4.87554937463653e-06, + "loss": 0.8027, + "step": 476 + }, + { + "epoch": 0.7468684759916493, + "grad_norm": 0.9527666568756104, + "learning_rate": 4.874892107463811e-06, + "loss": 0.8069, + "step": 477 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 0.9576907157897949, + "learning_rate": 4.874233153769924e-06, + "loss": 0.774, + "step": 478 + }, + { + "epoch": 0.75, + "grad_norm": 1.0516161918640137, + "learning_rate": 4.873572514022822e-06, + "loss": 0.8176, + "step": 479 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.026928186416626, + "learning_rate": 4.872910188691657e-06, + "loss": 0.8797, + "step": 480 + }, + { + "epoch": 0.7531315240083507, + "grad_norm": 0.9312179088592529, + "learning_rate": 4.872246178246776e-06, + "loss": 0.7747, + "step": 481 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0676255226135254, + "learning_rate": 4.871580483159725e-06, + "loss": 0.7801, + "step": 482 + }, + { + "epoch": 0.7562630480167014, + "grad_norm": 1.066056489944458, + "learning_rate": 4.870913103903245e-06, + "loss": 0.7837, + "step": 483 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9388124346733093, + "learning_rate": 4.87024404095127e-06, + "loss": 0.7801, + "step": 484 + }, + { + "epoch": 0.7593945720250522, + "grad_norm": 0.9879791140556335, + "learning_rate": 4.8695732947789356e-06, + "loss": 0.7849, + "step": 485 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0002318620681763, + "learning_rate": 4.868900865862569e-06, + "loss": 0.8219, + "step": 486 + }, + { + "epoch": 0.7625260960334029, + "grad_norm": 0.9599125385284424, + "learning_rate": 4.8682267546796945e-06, + "loss": 0.8888, + "step": 487 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 0.9099973440170288, + "learning_rate": 4.867550961709027e-06, + "loss": 0.7585, + "step": 488 + }, + { + "epoch": 0.7656576200417536, + "grad_norm": 0.9715955257415771, + "learning_rate": 4.866873487430481e-06, + "loss": 0.8237, + "step": 489 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 1.0194706916809082, + "learning_rate": 4.86619433232516e-06, + "loss": 0.7708, + "step": 490 + }, + { + "epoch": 0.7687891440501043, + "grad_norm": 1.044005036354065, + "learning_rate": 4.865513496875367e-06, + "loss": 0.8, + "step": 491 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.9776361584663391, + "learning_rate": 4.8648309815645915e-06, + "loss": 0.7696, + "step": 492 + }, + { + "epoch": 0.7719206680584552, + "grad_norm": 0.9213030934333801, + "learning_rate": 4.864146786877521e-06, + "loss": 0.8558, + "step": 493 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.9361440539360046, + "learning_rate": 4.8634609133000336e-06, + "loss": 0.7851, + "step": 494 + }, + { + "epoch": 0.7750521920668059, + "grad_norm": 1.0199540853500366, + "learning_rate": 4.8627733613192e-06, + "loss": 0.8229, + "step": 495 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 1.0244604349136353, + "learning_rate": 4.862084131423283e-06, + "loss": 0.772, + "step": 496 + }, + { + "epoch": 0.7781837160751566, + "grad_norm": 0.9414368271827698, + "learning_rate": 4.861393224101736e-06, + "loss": 0.7579, + "step": 497 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9198965430259705, + "learning_rate": 4.860700639845205e-06, + "loss": 0.855, + "step": 498 + }, + { + "epoch": 0.7813152400835073, + "grad_norm": 0.9398712515830994, + "learning_rate": 4.8600063791455275e-06, + "loss": 0.7824, + "step": 499 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.9544969201087952, + "learning_rate": 4.8593104424957275e-06, + "loss": 0.8104, + "step": 500 + }, + { + "epoch": 0.784446764091858, + "grad_norm": 0.9522370100021362, + "learning_rate": 4.858612830390024e-06, + "loss": 0.813, + "step": 501 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.9830234050750732, + "learning_rate": 4.857913543323823e-06, + "loss": 0.8656, + "step": 502 + }, + { + "epoch": 0.7875782881002088, + "grad_norm": 1.0527008771896362, + "learning_rate": 4.8572125817937195e-06, + "loss": 0.8417, + "step": 503 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 0.9772782921791077, + "learning_rate": 4.856509946297501e-06, + "loss": 0.7983, + "step": 504 + }, + { + "epoch": 0.7907098121085595, + "grad_norm": 0.9250578880310059, + "learning_rate": 4.855805637334141e-06, + "loss": 0.7391, + "step": 505 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9792859554290771, + "learning_rate": 4.855099655403802e-06, + "loss": 0.7753, + "step": 506 + }, + { + "epoch": 0.7938413361169102, + "grad_norm": 0.8876479864120483, + "learning_rate": 4.854392001007834e-06, + "loss": 0.8249, + "step": 507 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.9336771368980408, + "learning_rate": 4.853682674648775e-06, + "loss": 0.8462, + "step": 508 + }, + { + "epoch": 0.7969728601252609, + "grad_norm": 0.9732388854026794, + "learning_rate": 4.852971676830352e-06, + "loss": 0.8791, + "step": 509 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.9135870337486267, + "learning_rate": 4.852259008057476e-06, + "loss": 0.7279, + "step": 510 + }, + { + "epoch": 0.8001043841336117, + "grad_norm": 0.9782464504241943, + "learning_rate": 4.8515446688362476e-06, + "loss": 0.7797, + "step": 511 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 0.9571824073791504, + "learning_rate": 4.85082865967395e-06, + "loss": 0.8349, + "step": 512 + }, + { + "epoch": 0.8032359081419624, + "grad_norm": 1.011786937713623, + "learning_rate": 4.850110981079057e-06, + "loss": 0.8057, + "step": 513 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 0.9418787956237793, + "learning_rate": 4.8493916335612255e-06, + "loss": 0.8592, + "step": 514 + }, + { + "epoch": 0.8063674321503131, + "grad_norm": 0.9292604923248291, + "learning_rate": 4.848670617631296e-06, + "loss": 0.7471, + "step": 515 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.9303188323974609, + "learning_rate": 4.847947933801296e-06, + "loss": 0.8768, + "step": 516 + }, + { + "epoch": 0.8094989561586639, + "grad_norm": 0.9897359609603882, + "learning_rate": 4.847223582584437e-06, + "loss": 0.8343, + "step": 517 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.9695891737937927, + "learning_rate": 4.846497564495114e-06, + "loss": 0.8318, + "step": 518 + }, + { + "epoch": 0.8126304801670147, + "grad_norm": 1.0175273418426514, + "learning_rate": 4.845769880048906e-06, + "loss": 0.9137, + "step": 519 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9497922658920288, + "learning_rate": 4.845040529762577e-06, + "loss": 0.6549, + "step": 520 + }, + { + "epoch": 0.8157620041753654, + "grad_norm": 1.0647574663162231, + "learning_rate": 4.844309514154071e-06, + "loss": 0.8243, + "step": 521 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.9710965156555176, + "learning_rate": 4.843576833742517e-06, + "loss": 0.7864, + "step": 522 + }, + { + "epoch": 0.8188935281837161, + "grad_norm": 0.9881441593170166, + "learning_rate": 4.842842489048225e-06, + "loss": 0.7799, + "step": 523 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 0.9455597400665283, + "learning_rate": 4.842106480592687e-06, + "loss": 0.7943, + "step": 524 + }, + { + "epoch": 0.8220250521920668, + "grad_norm": 0.9948442578315735, + "learning_rate": 4.841368808898577e-06, + "loss": 0.8587, + "step": 525 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 0.9959186315536499, + "learning_rate": 4.84062947448975e-06, + "loss": 0.8477, + "step": 526 + }, + { + "epoch": 0.8251565762004175, + "grad_norm": 0.9791848659515381, + "learning_rate": 4.839888477891243e-06, + "loss": 0.7568, + "step": 527 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 1.1499426364898682, + "learning_rate": 4.839145819629269e-06, + "loss": 0.8296, + "step": 528 + }, + { + "epoch": 0.8282881002087683, + "grad_norm": 0.9269506335258484, + "learning_rate": 4.8384015002312265e-06, + "loss": 0.7935, + "step": 529 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.970565140247345, + "learning_rate": 4.837655520225692e-06, + "loss": 0.8138, + "step": 530 + }, + { + "epoch": 0.831419624217119, + "grad_norm": 0.9852119088172913, + "learning_rate": 4.836907880142418e-06, + "loss": 0.7909, + "step": 531 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.927129328250885, + "learning_rate": 4.836158580512339e-06, + "loss": 0.8105, + "step": 532 + }, + { + "epoch": 0.8345511482254697, + "grad_norm": 1.0711781978607178, + "learning_rate": 4.835407621867569e-06, + "loss": 0.8241, + "step": 533 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 1.0472421646118164, + "learning_rate": 4.834655004741397e-06, + "loss": 0.7923, + "step": 534 + }, + { + "epoch": 0.8376826722338204, + "grad_norm": 0.9761579632759094, + "learning_rate": 4.833900729668293e-06, + "loss": 0.8203, + "step": 535 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 1.0559637546539307, + "learning_rate": 4.833144797183902e-06, + "loss": 0.8915, + "step": 536 + }, + { + "epoch": 0.8408141962421712, + "grad_norm": 0.9951052069664001, + "learning_rate": 4.832387207825045e-06, + "loss": 0.8253, + "step": 537 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.894079864025116, + "learning_rate": 4.831627962129725e-06, + "loss": 0.7479, + "step": 538 + }, + { + "epoch": 0.843945720250522, + "grad_norm": 0.9832531213760376, + "learning_rate": 4.830867060637115e-06, + "loss": 0.8484, + "step": 539 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0477591753005981, + "learning_rate": 4.8301045038875665e-06, + "loss": 0.8052, + "step": 540 + }, + { + "epoch": 0.8470772442588727, + "grad_norm": 0.9388875365257263, + "learning_rate": 4.829340292422607e-06, + "loss": 0.7998, + "step": 541 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 0.8975532054901123, + "learning_rate": 4.8285744267849395e-06, + "loss": 0.848, + "step": 542 + }, + { + "epoch": 0.8502087682672234, + "grad_norm": 0.9211076498031616, + "learning_rate": 4.8278069075184396e-06, + "loss": 0.8337, + "step": 543 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.970134437084198, + "learning_rate": 4.827037735168159e-06, + "loss": 0.7843, + "step": 544 + }, + { + "epoch": 0.8533402922755741, + "grad_norm": 0.9714353084564209, + "learning_rate": 4.826266910280322e-06, + "loss": 0.6991, + "step": 545 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 0.8607689142227173, + "learning_rate": 4.8254944334023265e-06, + "loss": 0.7314, + "step": 546 + }, + { + "epoch": 0.8564718162839249, + "grad_norm": 0.9843030571937561, + "learning_rate": 4.824720305082747e-06, + "loss": 0.7497, + "step": 547 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 0.9638474583625793, + "learning_rate": 4.823944525871324e-06, + "loss": 0.776, + "step": 548 + }, + { + "epoch": 0.8596033402922756, + "grad_norm": 0.9342365860939026, + "learning_rate": 4.823167096318979e-06, + "loss": 0.77, + "step": 549 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9608579874038696, + "learning_rate": 4.822388016977797e-06, + "loss": 0.8281, + "step": 550 + }, + { + "epoch": 0.8627348643006263, + "grad_norm": 0.8996383547782898, + "learning_rate": 4.8216072884010395e-06, + "loss": 0.7519, + "step": 551 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 0.9922810196876526, + "learning_rate": 4.820824911143139e-06, + "loss": 0.8714, + "step": 552 + }, + { + "epoch": 0.865866388308977, + "grad_norm": 0.9861842393875122, + "learning_rate": 4.820040885759697e-06, + "loss": 0.7998, + "step": 553 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 1.013168215751648, + "learning_rate": 4.819255212807486e-06, + "loss": 0.8291, + "step": 554 + }, + { + "epoch": 0.8689979123173278, + "grad_norm": 0.9695484638214111, + "learning_rate": 4.818467892844451e-06, + "loss": 0.8257, + "step": 555 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 1.0018948316574097, + "learning_rate": 4.817678926429702e-06, + "loss": 0.7551, + "step": 556 + }, + { + "epoch": 0.8721294363256785, + "grad_norm": 1.0259422063827515, + "learning_rate": 4.816888314123521e-06, + "loss": 0.8582, + "step": 557 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9045076966285706, + "learning_rate": 4.81609605648736e-06, + "loss": 0.811, + "step": 558 + }, + { + "epoch": 0.8752609603340292, + "grad_norm": 0.9112316370010376, + "learning_rate": 4.815302154083837e-06, + "loss": 0.818, + "step": 559 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0060032606124878, + "learning_rate": 4.814506607476739e-06, + "loss": 0.766, + "step": 560 + }, + { + "epoch": 0.8783924843423799, + "grad_norm": 0.9126970171928406, + "learning_rate": 4.813709417231021e-06, + "loss": 0.8903, + "step": 561 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.0825906991958618, + "learning_rate": 4.812910583912804e-06, + "loss": 0.8466, + "step": 562 + }, + { + "epoch": 0.8815240083507306, + "grad_norm": 0.9530975818634033, + "learning_rate": 4.812110108089377e-06, + "loss": 0.8163, + "step": 563 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.9566405415534973, + "learning_rate": 4.8113079903291955e-06, + "loss": 0.8244, + "step": 564 + }, + { + "epoch": 0.8846555323590815, + "grad_norm": 0.919980525970459, + "learning_rate": 4.81050423120188e-06, + "loss": 0.8335, + "step": 565 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.9123868346214294, + "learning_rate": 4.809698831278217e-06, + "loss": 0.7712, + "step": 566 + }, + { + "epoch": 0.8877870563674322, + "grad_norm": 0.9589663743972778, + "learning_rate": 4.8088917911301595e-06, + "loss": 0.8467, + "step": 567 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 0.9052798748016357, + "learning_rate": 4.808083111330823e-06, + "loss": 0.8078, + "step": 568 + }, + { + "epoch": 0.8909185803757829, + "grad_norm": 0.9250494241714478, + "learning_rate": 4.807272792454489e-06, + "loss": 0.8536, + "step": 569 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.8451298475265503, + "learning_rate": 4.806460835076603e-06, + "loss": 0.7752, + "step": 570 + }, + { + "epoch": 0.8940501043841336, + "grad_norm": 1.0324240922927856, + "learning_rate": 4.805647239773773e-06, + "loss": 0.7871, + "step": 571 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 0.9338528513908386, + "learning_rate": 4.804832007123771e-06, + "loss": 0.8082, + "step": 572 + }, + { + "epoch": 0.8971816283924844, + "grad_norm": 1.0080739259719849, + "learning_rate": 4.804015137705531e-06, + "loss": 0.7636, + "step": 573 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9817036986351013, + "learning_rate": 4.803196632099152e-06, + "loss": 0.8599, + "step": 574 + }, + { + "epoch": 0.9003131524008351, + "grad_norm": 0.9284219145774841, + "learning_rate": 4.80237649088589e-06, + "loss": 0.8152, + "step": 575 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.8922317028045654, + "learning_rate": 4.801554714648166e-06, + "loss": 0.8146, + "step": 576 + }, + { + "epoch": 0.9034446764091858, + "grad_norm": 0.9465014338493347, + "learning_rate": 4.800731303969565e-06, + "loss": 0.7663, + "step": 577 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.0325525999069214, + "learning_rate": 4.799906259434824e-06, + "loss": 0.8458, + "step": 578 + }, + { + "epoch": 0.9065762004175365, + "grad_norm": 0.9366307258605957, + "learning_rate": 4.7990795816298485e-06, + "loss": 0.7992, + "step": 579 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 0.9923847317695618, + "learning_rate": 4.7982512711416995e-06, + "loss": 0.7256, + "step": 580 + }, + { + "epoch": 0.9097077244258872, + "grad_norm": 1.0387133359909058, + "learning_rate": 4.7974213285586e-06, + "loss": 0.7766, + "step": 581 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.9617043733596802, + "learning_rate": 4.79658975446993e-06, + "loss": 0.7602, + "step": 582 + }, + { + "epoch": 0.912839248434238, + "grad_norm": 0.9530353546142578, + "learning_rate": 4.795756549466229e-06, + "loss": 0.7922, + "step": 583 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.9305918216705322, + "learning_rate": 4.794921714139194e-06, + "loss": 0.8377, + "step": 584 + }, + { + "epoch": 0.9159707724425887, + "grad_norm": 0.899860143661499, + "learning_rate": 4.794085249081682e-06, + "loss": 0.8343, + "step": 585 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.9024940729141235, + "learning_rate": 4.793247154887703e-06, + "loss": 0.8271, + "step": 586 + }, + { + "epoch": 0.9191022964509394, + "grad_norm": 0.980869472026825, + "learning_rate": 4.7924074321524286e-06, + "loss": 0.7468, + "step": 587 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.0506541728973389, + "learning_rate": 4.791566081472185e-06, + "loss": 0.9207, + "step": 588 + }, + { + "epoch": 0.9222338204592901, + "grad_norm": 1.0723917484283447, + "learning_rate": 4.790723103444454e-06, + "loss": 0.7615, + "step": 589 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 1.0455081462860107, + "learning_rate": 4.789878498667873e-06, + "loss": 0.8115, + "step": 590 + }, + { + "epoch": 0.925365344467641, + "grad_norm": 0.9729639291763306, + "learning_rate": 4.789032267742234e-06, + "loss": 0.8622, + "step": 591 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.9695239067077637, + "learning_rate": 4.788184411268488e-06, + "loss": 0.8165, + "step": 592 + }, + { + "epoch": 0.9284968684759917, + "grad_norm": 1.0099278688430786, + "learning_rate": 4.7873349298487345e-06, + "loss": 0.759, + "step": 593 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 0.9377149939537048, + "learning_rate": 4.786483824086231e-06, + "loss": 0.8323, + "step": 594 + }, + { + "epoch": 0.9316283924843424, + "grad_norm": 0.8885690569877625, + "learning_rate": 4.785631094585387e-06, + "loss": 0.8448, + "step": 595 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 0.9133430123329163, + "learning_rate": 4.784776741951766e-06, + "loss": 0.8135, + "step": 596 + }, + { + "epoch": 0.9347599164926931, + "grad_norm": 1.015851378440857, + "learning_rate": 4.783920766792082e-06, + "loss": 0.8464, + "step": 597 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.8883825540542603, + "learning_rate": 4.783063169714203e-06, + "loss": 0.7445, + "step": 598 + }, + { + "epoch": 0.9378914405010439, + "grad_norm": 0.9872002005577087, + "learning_rate": 4.782203951327149e-06, + "loss": 0.8293, + "step": 599 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 1.0305813550949097, + "learning_rate": 4.781343112241091e-06, + "loss": 0.8606, + "step": 600 + }, + { + "epoch": 0.9410229645093946, + "grad_norm": 0.9667104482650757, + "learning_rate": 4.780480653067351e-06, + "loss": 0.8057, + "step": 601 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 1.067185640335083, + "learning_rate": 4.7796165744184e-06, + "loss": 0.849, + "step": 602 + }, + { + "epoch": 0.9441544885177453, + "grad_norm": 0.9017406105995178, + "learning_rate": 4.778750876907862e-06, + "loss": 0.7927, + "step": 603 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.9827666878700256, + "learning_rate": 4.777883561150508e-06, + "loss": 0.8087, + "step": 604 + }, + { + "epoch": 0.947286012526096, + "grad_norm": 0.9627811908721924, + "learning_rate": 4.77701462776226e-06, + "loss": 0.8488, + "step": 605 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0321338176727295, + "learning_rate": 4.776144077360186e-06, + "loss": 0.8289, + "step": 606 + }, + { + "epoch": 0.9504175365344467, + "grad_norm": 1.0012633800506592, + "learning_rate": 4.775271910562508e-06, + "loss": 0.7184, + "step": 607 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.9186056852340698, + "learning_rate": 4.7743981279885875e-06, + "loss": 0.8727, + "step": 608 + }, + { + "epoch": 0.9535490605427975, + "grad_norm": 0.910044252872467, + "learning_rate": 4.773522730258942e-06, + "loss": 0.8455, + "step": 609 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.9342944025993347, + "learning_rate": 4.77264571799523e-06, + "loss": 0.7548, + "step": 610 + }, + { + "epoch": 0.9566805845511482, + "grad_norm": 0.9609628319740295, + "learning_rate": 4.771767091820259e-06, + "loss": 0.8294, + "step": 611 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9708508253097534, + "learning_rate": 4.770886852357983e-06, + "loss": 0.7857, + "step": 612 + }, + { + "epoch": 0.9598121085594989, + "grad_norm": 1.0910807847976685, + "learning_rate": 4.7700050002334995e-06, + "loss": 0.8162, + "step": 613 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 0.9703771471977234, + "learning_rate": 4.769121536073054e-06, + "loss": 0.7205, + "step": 614 + }, + { + "epoch": 0.9629436325678496, + "grad_norm": 1.027938961982727, + "learning_rate": 4.768236460504035e-06, + "loss": 0.7641, + "step": 615 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 1.0003989934921265, + "learning_rate": 4.767349774154974e-06, + "loss": 0.7588, + "step": 616 + }, + { + "epoch": 0.9660751565762005, + "grad_norm": 0.9572570323944092, + "learning_rate": 4.766461477655552e-06, + "loss": 0.8274, + "step": 617 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.977287769317627, + "learning_rate": 4.765571571636586e-06, + "loss": 0.7856, + "step": 618 + }, + { + "epoch": 0.9692066805845512, + "grad_norm": 0.9582562446594238, + "learning_rate": 4.764680056730042e-06, + "loss": 0.8102, + "step": 619 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.9842986464500427, + "learning_rate": 4.763786933569025e-06, + "loss": 0.7515, + "step": 620 + }, + { + "epoch": 0.9723382045929019, + "grad_norm": 0.9955743551254272, + "learning_rate": 4.762892202787782e-06, + "loss": 0.7398, + "step": 621 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 1.0011107921600342, + "learning_rate": 4.761995865021706e-06, + "loss": 0.8182, + "step": 622 + }, + { + "epoch": 0.9754697286012526, + "grad_norm": 0.9586448669433594, + "learning_rate": 4.761097920907325e-06, + "loss": 0.8434, + "step": 623 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.979123055934906, + "learning_rate": 4.760198371082312e-06, + "loss": 0.7954, + "step": 624 + }, + { + "epoch": 0.9786012526096033, + "grad_norm": 1.0081250667572021, + "learning_rate": 4.75929721618548e-06, + "loss": 0.7931, + "step": 625 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 1.027185320854187, + "learning_rate": 4.75839445685678e-06, + "loss": 0.7339, + "step": 626 + }, + { + "epoch": 0.9817327766179541, + "grad_norm": 1.0553215742111206, + "learning_rate": 4.757490093737303e-06, + "loss": 0.7639, + "step": 627 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 0.8578981757164001, + "learning_rate": 4.75658412746928e-06, + "loss": 0.8128, + "step": 628 + }, + { + "epoch": 0.9848643006263048, + "grad_norm": 1.0143071413040161, + "learning_rate": 4.75567655869608e-06, + "loss": 0.8169, + "step": 629 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 0.9743554592132568, + "learning_rate": 4.754767388062208e-06, + "loss": 0.8179, + "step": 630 + }, + { + "epoch": 0.9879958246346555, + "grad_norm": 1.0276602506637573, + "learning_rate": 4.7538566162133105e-06, + "loss": 0.8009, + "step": 631 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.9296634793281555, + "learning_rate": 4.752944243796167e-06, + "loss": 0.8262, + "step": 632 + }, + { + "epoch": 0.9911273486430062, + "grad_norm": 0.9216791987419128, + "learning_rate": 4.752030271458696e-06, + "loss": 0.8929, + "step": 633 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 0.9272832274436951, + "learning_rate": 4.751114699849954e-06, + "loss": 0.7396, + "step": 634 + }, + { + "epoch": 0.994258872651357, + "grad_norm": 0.9121203422546387, + "learning_rate": 4.750197529620127e-06, + "loss": 0.8311, + "step": 635 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.999600887298584, + "learning_rate": 4.7492787614205425e-06, + "loss": 0.8125, + "step": 636 + }, + { + "epoch": 0.9973903966597077, + "grad_norm": 1.012978196144104, + "learning_rate": 4.748358395903661e-06, + "loss": 0.7966, + "step": 637 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.9767918586730957, + "learning_rate": 4.747436433723075e-06, + "loss": 0.8024, + "step": 638 + }, + { + "epoch": 1.0015657620041754, + "grad_norm": 1.8302738666534424, + "learning_rate": 4.7465128755335135e-06, + "loss": 1.6427, + "step": 639 + }, + { + "epoch": 1.0031315240083507, + "grad_norm": 0.8704890012741089, + "learning_rate": 4.7455877219908386e-06, + "loss": 0.7766, + "step": 640 + }, + { + "epoch": 1.004697286012526, + "grad_norm": 0.9848434925079346, + "learning_rate": 4.744660973752044e-06, + "loss": 0.8314, + "step": 641 + }, + { + "epoch": 1.0062630480167014, + "grad_norm": 1.054824709892273, + "learning_rate": 4.743732631475258e-06, + "loss": 0.797, + "step": 642 + }, + { + "epoch": 1.0078288100208768, + "grad_norm": 0.9286114573478699, + "learning_rate": 4.742802695819737e-06, + "loss": 0.764, + "step": 643 + }, + { + "epoch": 1.0093945720250521, + "grad_norm": 0.8625781536102295, + "learning_rate": 4.7418711674458735e-06, + "loss": 0.6691, + "step": 644 + }, + { + "epoch": 1.0109603340292275, + "grad_norm": 0.8710452914237976, + "learning_rate": 4.740938047015188e-06, + "loss": 0.7277, + "step": 645 + }, + { + "epoch": 1.0125260960334028, + "grad_norm": 0.8873130679130554, + "learning_rate": 4.740003335190334e-06, + "loss": 0.6531, + "step": 646 + }, + { + "epoch": 1.0140918580375784, + "grad_norm": 0.9777164459228516, + "learning_rate": 4.7390670326350925e-06, + "loss": 0.8047, + "step": 647 + }, + { + "epoch": 1.0156576200417538, + "grad_norm": 0.886025071144104, + "learning_rate": 4.738129140014377e-06, + "loss": 0.715, + "step": 648 + }, + { + "epoch": 1.017223382045929, + "grad_norm": 0.9825332164764404, + "learning_rate": 4.737189657994226e-06, + "loss": 0.7783, + "step": 649 + }, + { + "epoch": 1.0187891440501045, + "grad_norm": 0.9331428408622742, + "learning_rate": 4.736248587241812e-06, + "loss": 0.7598, + "step": 650 + }, + { + "epoch": 1.0203549060542798, + "grad_norm": 0.9406535625457764, + "learning_rate": 4.735305928425431e-06, + "loss": 0.7989, + "step": 651 + }, + { + "epoch": 1.0219206680584552, + "grad_norm": 1.0644745826721191, + "learning_rate": 4.734361682214511e-06, + "loss": 0.8135, + "step": 652 + }, + { + "epoch": 1.0234864300626305, + "grad_norm": 0.9037052392959595, + "learning_rate": 4.733415849279603e-06, + "loss": 0.7051, + "step": 653 + }, + { + "epoch": 1.0250521920668059, + "grad_norm": 1.000563144683838, + "learning_rate": 4.7324684302923865e-06, + "loss": 0.7427, + "step": 654 + }, + { + "epoch": 1.0266179540709812, + "grad_norm": 0.9181354641914368, + "learning_rate": 4.731519425925671e-06, + "loss": 0.7236, + "step": 655 + }, + { + "epoch": 1.0281837160751566, + "grad_norm": 1.0283046960830688, + "learning_rate": 4.730568836853384e-06, + "loss": 0.7998, + "step": 656 + }, + { + "epoch": 1.029749478079332, + "grad_norm": 0.8692315220832825, + "learning_rate": 4.729616663750587e-06, + "loss": 0.7974, + "step": 657 + }, + { + "epoch": 1.0313152400835073, + "grad_norm": 0.928920567035675, + "learning_rate": 4.72866290729346e-06, + "loss": 0.7199, + "step": 658 + }, + { + "epoch": 1.0328810020876826, + "grad_norm": 0.9066957235336304, + "learning_rate": 4.727707568159311e-06, + "loss": 0.791, + "step": 659 + }, + { + "epoch": 1.034446764091858, + "grad_norm": 1.0086729526519775, + "learning_rate": 4.726750647026569e-06, + "loss": 0.7113, + "step": 660 + }, + { + "epoch": 1.0360125260960333, + "grad_norm": 0.931856632232666, + "learning_rate": 4.72579214457479e-06, + "loss": 0.7977, + "step": 661 + }, + { + "epoch": 1.0375782881002087, + "grad_norm": 0.9191203713417053, + "learning_rate": 4.724832061484651e-06, + "loss": 0.7979, + "step": 662 + }, + { + "epoch": 1.039144050104384, + "grad_norm": 0.9614707827568054, + "learning_rate": 4.72387039843795e-06, + "loss": 0.6994, + "step": 663 + }, + { + "epoch": 1.0407098121085594, + "grad_norm": 1.014823317527771, + "learning_rate": 4.72290715611761e-06, + "loss": 0.7539, + "step": 664 + }, + { + "epoch": 1.042275574112735, + "grad_norm": 0.9177424311637878, + "learning_rate": 4.721942335207673e-06, + "loss": 0.6625, + "step": 665 + }, + { + "epoch": 1.0438413361169103, + "grad_norm": 1.0253031253814697, + "learning_rate": 4.720975936393305e-06, + "loss": 0.7747, + "step": 666 + }, + { + "epoch": 1.0454070981210857, + "grad_norm": 1.0004568099975586, + "learning_rate": 4.720007960360788e-06, + "loss": 0.7008, + "step": 667 + }, + { + "epoch": 1.046972860125261, + "grad_norm": 1.1344349384307861, + "learning_rate": 4.719038407797529e-06, + "loss": 0.7493, + "step": 668 + }, + { + "epoch": 1.0485386221294364, + "grad_norm": 0.9944799542427063, + "learning_rate": 4.718067279392052e-06, + "loss": 0.7823, + "step": 669 + }, + { + "epoch": 1.0501043841336117, + "grad_norm": 0.9469346404075623, + "learning_rate": 4.717094575833999e-06, + "loss": 0.7419, + "step": 670 + }, + { + "epoch": 1.051670146137787, + "grad_norm": 1.015339732170105, + "learning_rate": 4.716120297814133e-06, + "loss": 0.7168, + "step": 671 + }, + { + "epoch": 1.0532359081419624, + "grad_norm": 0.9544686675071716, + "learning_rate": 4.715144446024333e-06, + "loss": 0.791, + "step": 672 + }, + { + "epoch": 1.0548016701461378, + "grad_norm": 1.0285652875900269, + "learning_rate": 4.714167021157599e-06, + "loss": 0.7649, + "step": 673 + }, + { + "epoch": 1.0563674321503131, + "grad_norm": 0.9366872906684875, + "learning_rate": 4.713188023908044e-06, + "loss": 0.7524, + "step": 674 + }, + { + "epoch": 1.0579331941544885, + "grad_norm": 1.0348883867263794, + "learning_rate": 4.712207454970899e-06, + "loss": 0.6965, + "step": 675 + }, + { + "epoch": 1.0594989561586639, + "grad_norm": 0.8716043829917908, + "learning_rate": 4.711225315042513e-06, + "loss": 0.6264, + "step": 676 + }, + { + "epoch": 1.0610647181628392, + "grad_norm": 1.0421313047409058, + "learning_rate": 4.710241604820348e-06, + "loss": 0.7548, + "step": 677 + }, + { + "epoch": 1.0626304801670146, + "grad_norm": 1.0886270999908447, + "learning_rate": 4.709256325002983e-06, + "loss": 0.6352, + "step": 678 + }, + { + "epoch": 1.06419624217119, + "grad_norm": 0.9752539396286011, + "learning_rate": 4.708269476290111e-06, + "loss": 0.7764, + "step": 679 + }, + { + "epoch": 1.0657620041753653, + "grad_norm": 1.0658339262008667, + "learning_rate": 4.7072810593825385e-06, + "loss": 0.766, + "step": 680 + }, + { + "epoch": 1.0673277661795406, + "grad_norm": 1.0299617052078247, + "learning_rate": 4.706291074982187e-06, + "loss": 0.7558, + "step": 681 + }, + { + "epoch": 1.068893528183716, + "grad_norm": 0.9171326160430908, + "learning_rate": 4.705299523792089e-06, + "loss": 0.7084, + "step": 682 + }, + { + "epoch": 1.0704592901878915, + "grad_norm": 0.9924243688583374, + "learning_rate": 4.7043064065163925e-06, + "loss": 0.8393, + "step": 683 + }, + { + "epoch": 1.072025052192067, + "grad_norm": 1.0243877172470093, + "learning_rate": 4.703311723860356e-06, + "loss": 0.7755, + "step": 684 + }, + { + "epoch": 1.0735908141962422, + "grad_norm": 1.0019137859344482, + "learning_rate": 4.702315476530349e-06, + "loss": 0.7254, + "step": 685 + }, + { + "epoch": 1.0751565762004176, + "grad_norm": 0.8541996479034424, + "learning_rate": 4.701317665233853e-06, + "loss": 0.7389, + "step": 686 + }, + { + "epoch": 1.076722338204593, + "grad_norm": 1.1119922399520874, + "learning_rate": 4.700318290679462e-06, + "loss": 0.7803, + "step": 687 + }, + { + "epoch": 1.0782881002087683, + "grad_norm": 1.078452706336975, + "learning_rate": 4.699317353576876e-06, + "loss": 0.756, + "step": 688 + }, + { + "epoch": 1.0798538622129437, + "grad_norm": 1.0079457759857178, + "learning_rate": 4.698314854636908e-06, + "loss": 0.7365, + "step": 689 + }, + { + "epoch": 1.081419624217119, + "grad_norm": 0.931423544883728, + "learning_rate": 4.6973107945714795e-06, + "loss": 0.7881, + "step": 690 + }, + { + "epoch": 1.0829853862212944, + "grad_norm": 1.0135142803192139, + "learning_rate": 4.696305174093619e-06, + "loss": 0.7884, + "step": 691 + }, + { + "epoch": 1.0845511482254697, + "grad_norm": 1.1194610595703125, + "learning_rate": 4.695297993917465e-06, + "loss": 0.7969, + "step": 692 + }, + { + "epoch": 1.086116910229645, + "grad_norm": 0.9972780346870422, + "learning_rate": 4.6942892547582625e-06, + "loss": 0.5873, + "step": 693 + }, + { + "epoch": 1.0876826722338204, + "grad_norm": 1.0711641311645508, + "learning_rate": 4.693278957332364e-06, + "loss": 0.7273, + "step": 694 + }, + { + "epoch": 1.0892484342379958, + "grad_norm": 0.9708715081214905, + "learning_rate": 4.6922671023572305e-06, + "loss": 0.6786, + "step": 695 + }, + { + "epoch": 1.0908141962421711, + "grad_norm": 0.9773644804954529, + "learning_rate": 4.691253690551424e-06, + "loss": 0.7839, + "step": 696 + }, + { + "epoch": 1.0923799582463465, + "grad_norm": 1.039009690284729, + "learning_rate": 4.690238722634618e-06, + "loss": 0.6807, + "step": 697 + }, + { + "epoch": 1.0939457202505218, + "grad_norm": 1.140777349472046, + "learning_rate": 4.689222199327586e-06, + "loss": 0.7506, + "step": 698 + }, + { + "epoch": 1.0955114822546972, + "grad_norm": 0.9779882431030273, + "learning_rate": 4.688204121352211e-06, + "loss": 0.7732, + "step": 699 + }, + { + "epoch": 1.0970772442588728, + "grad_norm": 1.0412248373031616, + "learning_rate": 4.687184489431476e-06, + "loss": 0.7034, + "step": 700 + }, + { + "epoch": 1.0986430062630481, + "grad_norm": 0.9802400469779968, + "learning_rate": 4.686163304289469e-06, + "loss": 0.729, + "step": 701 + }, + { + "epoch": 1.1002087682672235, + "grad_norm": 1.0531584024429321, + "learning_rate": 4.6851405666513804e-06, + "loss": 0.7635, + "step": 702 + }, + { + "epoch": 1.1017745302713988, + "grad_norm": 0.983923077583313, + "learning_rate": 4.684116277243505e-06, + "loss": 0.7062, + "step": 703 + }, + { + "epoch": 1.1033402922755742, + "grad_norm": 0.9886301159858704, + "learning_rate": 4.683090436793238e-06, + "loss": 0.7299, + "step": 704 + }, + { + "epoch": 1.1049060542797495, + "grad_norm": 1.0247817039489746, + "learning_rate": 4.682063046029075e-06, + "loss": 0.729, + "step": 705 + }, + { + "epoch": 1.1064718162839249, + "grad_norm": 0.9860950708389282, + "learning_rate": 4.681034105680616e-06, + "loss": 0.7392, + "step": 706 + }, + { + "epoch": 1.1080375782881002, + "grad_norm": 0.9543380737304688, + "learning_rate": 4.680003616478556e-06, + "loss": 0.7061, + "step": 707 + }, + { + "epoch": 1.1096033402922756, + "grad_norm": 1.0253225564956665, + "learning_rate": 4.678971579154698e-06, + "loss": 0.6577, + "step": 708 + }, + { + "epoch": 1.111169102296451, + "grad_norm": 0.9013616442680359, + "learning_rate": 4.677937994441935e-06, + "loss": 0.7422, + "step": 709 + }, + { + "epoch": 1.1127348643006263, + "grad_norm": 0.9716471433639526, + "learning_rate": 4.676902863074268e-06, + "loss": 0.7353, + "step": 710 + }, + { + "epoch": 1.1143006263048016, + "grad_norm": 1.040898084640503, + "learning_rate": 4.675866185786787e-06, + "loss": 0.7396, + "step": 711 + }, + { + "epoch": 1.115866388308977, + "grad_norm": 0.9931808114051819, + "learning_rate": 4.674827963315688e-06, + "loss": 0.7354, + "step": 712 + }, + { + "epoch": 1.1174321503131524, + "grad_norm": 1.020512580871582, + "learning_rate": 4.6737881963982605e-06, + "loss": 0.7366, + "step": 713 + }, + { + "epoch": 1.1189979123173277, + "grad_norm": 1.0201045274734497, + "learning_rate": 4.67274688577289e-06, + "loss": 0.7907, + "step": 714 + }, + { + "epoch": 1.120563674321503, + "grad_norm": 1.0163429975509644, + "learning_rate": 4.671704032179061e-06, + "loss": 0.7452, + "step": 715 + }, + { + "epoch": 1.1221294363256784, + "grad_norm": 0.9667377471923828, + "learning_rate": 4.670659636357352e-06, + "loss": 0.7983, + "step": 716 + }, + { + "epoch": 1.1236951983298538, + "grad_norm": 1.0788301229476929, + "learning_rate": 4.669613699049436e-06, + "loss": 0.7861, + "step": 717 + }, + { + "epoch": 1.1252609603340291, + "grad_norm": 1.0512142181396484, + "learning_rate": 4.668566220998082e-06, + "loss": 0.7582, + "step": 718 + }, + { + "epoch": 1.1268267223382047, + "grad_norm": 0.9225794076919556, + "learning_rate": 4.667517202947154e-06, + "loss": 0.7801, + "step": 719 + }, + { + "epoch": 1.12839248434238, + "grad_norm": 1.1193747520446777, + "learning_rate": 4.666466645641607e-06, + "loss": 0.8073, + "step": 720 + }, + { + "epoch": 1.1299582463465554, + "grad_norm": 0.934438943862915, + "learning_rate": 4.6654145498274915e-06, + "loss": 0.774, + "step": 721 + }, + { + "epoch": 1.1315240083507307, + "grad_norm": 1.0614968538284302, + "learning_rate": 4.664360916251949e-06, + "loss": 0.7216, + "step": 722 + }, + { + "epoch": 1.133089770354906, + "grad_norm": 1.0807424783706665, + "learning_rate": 4.663305745663212e-06, + "loss": 0.6434, + "step": 723 + }, + { + "epoch": 1.1346555323590815, + "grad_norm": 0.9515354633331299, + "learning_rate": 4.66224903881061e-06, + "loss": 0.7336, + "step": 724 + }, + { + "epoch": 1.1362212943632568, + "grad_norm": 0.9852489233016968, + "learning_rate": 4.661190796444557e-06, + "loss": 0.8156, + "step": 725 + }, + { + "epoch": 1.1377870563674322, + "grad_norm": 1.0071501731872559, + "learning_rate": 4.66013101931656e-06, + "loss": 0.8836, + "step": 726 + }, + { + "epoch": 1.1393528183716075, + "grad_norm": 0.9622960686683655, + "learning_rate": 4.6590697081792176e-06, + "loss": 0.7631, + "step": 727 + }, + { + "epoch": 1.1409185803757829, + "grad_norm": 1.0588105916976929, + "learning_rate": 4.6580068637862144e-06, + "loss": 0.7874, + "step": 728 + }, + { + "epoch": 1.1424843423799582, + "grad_norm": 0.9920447468757629, + "learning_rate": 4.656942486892326e-06, + "loss": 0.7161, + "step": 729 + }, + { + "epoch": 1.1440501043841336, + "grad_norm": 1.0036561489105225, + "learning_rate": 4.655876578253416e-06, + "loss": 0.7116, + "step": 730 + }, + { + "epoch": 1.145615866388309, + "grad_norm": 1.0208396911621094, + "learning_rate": 4.654809138626436e-06, + "loss": 0.818, + "step": 731 + }, + { + "epoch": 1.1471816283924843, + "grad_norm": 1.1095491647720337, + "learning_rate": 4.653740168769424e-06, + "loss": 0.7319, + "step": 732 + }, + { + "epoch": 1.1487473903966596, + "grad_norm": 0.8151328563690186, + "learning_rate": 4.652669669441505e-06, + "loss": 0.774, + "step": 733 + }, + { + "epoch": 1.150313152400835, + "grad_norm": 1.0747915506362915, + "learning_rate": 4.651597641402891e-06, + "loss": 0.8275, + "step": 734 + }, + { + "epoch": 1.1518789144050103, + "grad_norm": 0.9265272617340088, + "learning_rate": 4.650524085414878e-06, + "loss": 0.6946, + "step": 735 + }, + { + "epoch": 1.153444676409186, + "grad_norm": 0.9531672596931458, + "learning_rate": 4.649449002239849e-06, + "loss": 0.7899, + "step": 736 + }, + { + "epoch": 1.1550104384133613, + "grad_norm": 1.0496262311935425, + "learning_rate": 4.648372392641269e-06, + "loss": 0.8237, + "step": 737 + }, + { + "epoch": 1.1565762004175366, + "grad_norm": 0.8967193961143494, + "learning_rate": 4.64729425738369e-06, + "loss": 0.736, + "step": 738 + }, + { + "epoch": 1.158141962421712, + "grad_norm": 1.061038851737976, + "learning_rate": 4.646214597232745e-06, + "loss": 0.7308, + "step": 739 + }, + { + "epoch": 1.1597077244258873, + "grad_norm": 0.9556939601898193, + "learning_rate": 4.64513341295515e-06, + "loss": 0.7526, + "step": 740 + }, + { + "epoch": 1.1612734864300627, + "grad_norm": 1.000093936920166, + "learning_rate": 4.6440507053187075e-06, + "loss": 0.7488, + "step": 741 + }, + { + "epoch": 1.162839248434238, + "grad_norm": 0.9552261233329773, + "learning_rate": 4.642966475092294e-06, + "loss": 0.8291, + "step": 742 + }, + { + "epoch": 1.1644050104384134, + "grad_norm": 1.0215847492218018, + "learning_rate": 4.641880723045875e-06, + "loss": 0.756, + "step": 743 + }, + { + "epoch": 1.1659707724425887, + "grad_norm": 1.0543124675750732, + "learning_rate": 4.640793449950493e-06, + "loss": 0.7668, + "step": 744 + }, + { + "epoch": 1.167536534446764, + "grad_norm": 0.9648609161376953, + "learning_rate": 4.639704656578269e-06, + "loss": 0.6531, + "step": 745 + }, + { + "epoch": 1.1691022964509394, + "grad_norm": 1.0117427110671997, + "learning_rate": 4.638614343702408e-06, + "loss": 0.7084, + "step": 746 + }, + { + "epoch": 1.1706680584551148, + "grad_norm": 0.9475041627883911, + "learning_rate": 4.6375225120971915e-06, + "loss": 0.8578, + "step": 747 + }, + { + "epoch": 1.1722338204592901, + "grad_norm": 1.0122299194335938, + "learning_rate": 4.6364291625379785e-06, + "loss": 0.7541, + "step": 748 + }, + { + "epoch": 1.1737995824634655, + "grad_norm": 0.9487535357475281, + "learning_rate": 4.635334295801208e-06, + "loss": 0.7203, + "step": 749 + }, + { + "epoch": 1.1753653444676408, + "grad_norm": 1.0628844499588013, + "learning_rate": 4.6342379126643965e-06, + "loss": 0.7532, + "step": 750 + }, + { + "epoch": 1.1769311064718162, + "grad_norm": 0.9647781848907471, + "learning_rate": 4.633140013906136e-06, + "loss": 0.5918, + "step": 751 + }, + { + "epoch": 1.1784968684759916, + "grad_norm": 0.9011932611465454, + "learning_rate": 4.632040600306095e-06, + "loss": 0.7395, + "step": 752 + }, + { + "epoch": 1.1800626304801671, + "grad_norm": 1.0524142980575562, + "learning_rate": 4.630939672645018e-06, + "loss": 0.7487, + "step": 753 + }, + { + "epoch": 1.1816283924843423, + "grad_norm": 1.165077567100525, + "learning_rate": 4.6298372317047265e-06, + "loss": 0.6817, + "step": 754 + }, + { + "epoch": 1.1831941544885178, + "grad_norm": 0.9301269054412842, + "learning_rate": 4.628733278268113e-06, + "loss": 0.8446, + "step": 755 + }, + { + "epoch": 1.1847599164926932, + "grad_norm": 0.9762471318244934, + "learning_rate": 4.627627813119147e-06, + "loss": 0.7527, + "step": 756 + }, + { + "epoch": 1.1863256784968685, + "grad_norm": 0.8634734153747559, + "learning_rate": 4.626520837042871e-06, + "loss": 0.7177, + "step": 757 + }, + { + "epoch": 1.187891440501044, + "grad_norm": 1.0592188835144043, + "learning_rate": 4.625412350825399e-06, + "loss": 0.7293, + "step": 758 + }, + { + "epoch": 1.1894572025052192, + "grad_norm": 1.1318559646606445, + "learning_rate": 4.62430235525392e-06, + "loss": 0.6592, + "step": 759 + }, + { + "epoch": 1.1910229645093946, + "grad_norm": 0.9689339399337769, + "learning_rate": 4.623190851116692e-06, + "loss": 0.7667, + "step": 760 + }, + { + "epoch": 1.19258872651357, + "grad_norm": 0.9965885281562805, + "learning_rate": 4.6220778392030464e-06, + "loss": 0.6982, + "step": 761 + }, + { + "epoch": 1.1941544885177453, + "grad_norm": 1.1428629159927368, + "learning_rate": 4.6209633203033846e-06, + "loss": 0.7342, + "step": 762 + }, + { + "epoch": 1.1957202505219207, + "grad_norm": 0.901360809803009, + "learning_rate": 4.6198472952091785e-06, + "loss": 0.8371, + "step": 763 + }, + { + "epoch": 1.197286012526096, + "grad_norm": 0.9882062077522278, + "learning_rate": 4.618729764712969e-06, + "loss": 0.6882, + "step": 764 + }, + { + "epoch": 1.1988517745302714, + "grad_norm": 1.0692517757415771, + "learning_rate": 4.617610729608366e-06, + "loss": 0.7327, + "step": 765 + }, + { + "epoch": 1.2004175365344467, + "grad_norm": 1.1056455373764038, + "learning_rate": 4.616490190690049e-06, + "loss": 0.781, + "step": 766 + }, + { + "epoch": 1.201983298538622, + "grad_norm": 0.9928885698318481, + "learning_rate": 4.615368148753765e-06, + "loss": 0.7463, + "step": 767 + }, + { + "epoch": 1.2035490605427974, + "grad_norm": 1.0006698369979858, + "learning_rate": 4.614244604596327e-06, + "loss": 0.6994, + "step": 768 + }, + { + "epoch": 1.2051148225469728, + "grad_norm": 0.907508909702301, + "learning_rate": 4.613119559015616e-06, + "loss": 0.7117, + "step": 769 + }, + { + "epoch": 1.2066805845511483, + "grad_norm": 1.1037627458572388, + "learning_rate": 4.61199301281058e-06, + "loss": 0.772, + "step": 770 + }, + { + "epoch": 1.2082463465553235, + "grad_norm": 0.9474150538444519, + "learning_rate": 4.610864966781231e-06, + "loss": 0.7601, + "step": 771 + }, + { + "epoch": 1.209812108559499, + "grad_norm": 1.003667950630188, + "learning_rate": 4.609735421728647e-06, + "loss": 0.7141, + "step": 772 + }, + { + "epoch": 1.2113778705636744, + "grad_norm": 0.9724787473678589, + "learning_rate": 4.608604378454971e-06, + "loss": 0.7841, + "step": 773 + }, + { + "epoch": 1.2129436325678498, + "grad_norm": 1.0460760593414307, + "learning_rate": 4.607471837763409e-06, + "loss": 0.7563, + "step": 774 + }, + { + "epoch": 1.2145093945720251, + "grad_norm": 0.9798458218574524, + "learning_rate": 4.606337800458231e-06, + "loss": 0.6812, + "step": 775 + }, + { + "epoch": 1.2160751565762005, + "grad_norm": 1.1637914180755615, + "learning_rate": 4.605202267344768e-06, + "loss": 0.7571, + "step": 776 + }, + { + "epoch": 1.2176409185803758, + "grad_norm": 0.9384608864784241, + "learning_rate": 4.6040652392294175e-06, + "loss": 0.7842, + "step": 777 + }, + { + "epoch": 1.2192066805845512, + "grad_norm": 1.0490702390670776, + "learning_rate": 4.602926716919633e-06, + "loss": 0.7157, + "step": 778 + }, + { + "epoch": 1.2207724425887265, + "grad_norm": 1.131402611732483, + "learning_rate": 4.601786701223934e-06, + "loss": 0.7724, + "step": 779 + }, + { + "epoch": 1.2223382045929019, + "grad_norm": 1.0159722566604614, + "learning_rate": 4.600645192951898e-06, + "loss": 0.7666, + "step": 780 + }, + { + "epoch": 1.2239039665970772, + "grad_norm": 1.039806604385376, + "learning_rate": 4.599502192914164e-06, + "loss": 0.7845, + "step": 781 + }, + { + "epoch": 1.2254697286012526, + "grad_norm": 1.0722614526748657, + "learning_rate": 4.598357701922428e-06, + "loss": 0.782, + "step": 782 + }, + { + "epoch": 1.227035490605428, + "grad_norm": 0.9443033933639526, + "learning_rate": 4.5972117207894465e-06, + "loss": 0.6573, + "step": 783 + }, + { + "epoch": 1.2286012526096033, + "grad_norm": 0.9738026857376099, + "learning_rate": 4.596064250329035e-06, + "loss": 0.7475, + "step": 784 + }, + { + "epoch": 1.2301670146137786, + "grad_norm": 0.9932008385658264, + "learning_rate": 4.594915291356063e-06, + "loss": 0.757, + "step": 785 + }, + { + "epoch": 1.231732776617954, + "grad_norm": 1.0612564086914062, + "learning_rate": 4.5937648446864625e-06, + "loss": 0.7824, + "step": 786 + }, + { + "epoch": 1.2332985386221293, + "grad_norm": 1.0500893592834473, + "learning_rate": 4.592612911137218e-06, + "loss": 0.7646, + "step": 787 + }, + { + "epoch": 1.2348643006263047, + "grad_norm": 0.9607855081558228, + "learning_rate": 4.591459491526371e-06, + "loss": 0.754, + "step": 788 + }, + { + "epoch": 1.2364300626304803, + "grad_norm": 1.0871951580047607, + "learning_rate": 4.590304586673018e-06, + "loss": 0.7226, + "step": 789 + }, + { + "epoch": 1.2379958246346556, + "grad_norm": 1.0030092000961304, + "learning_rate": 4.589148197397311e-06, + "loss": 0.7357, + "step": 790 + }, + { + "epoch": 1.239561586638831, + "grad_norm": 1.0297991037368774, + "learning_rate": 4.587990324520455e-06, + "loss": 0.7724, + "step": 791 + }, + { + "epoch": 1.2411273486430063, + "grad_norm": 1.0192303657531738, + "learning_rate": 4.586830968864711e-06, + "loss": 0.6921, + "step": 792 + }, + { + "epoch": 1.2426931106471817, + "grad_norm": 0.9617307186126709, + "learning_rate": 4.585670131253389e-06, + "loss": 0.7443, + "step": 793 + }, + { + "epoch": 1.244258872651357, + "grad_norm": 1.0355238914489746, + "learning_rate": 4.584507812510856e-06, + "loss": 0.7741, + "step": 794 + }, + { + "epoch": 1.2458246346555324, + "grad_norm": 0.9938539862632751, + "learning_rate": 4.583344013462526e-06, + "loss": 0.7372, + "step": 795 + }, + { + "epoch": 1.2473903966597077, + "grad_norm": 1.0573335886001587, + "learning_rate": 4.582178734934869e-06, + "loss": 0.7191, + "step": 796 + }, + { + "epoch": 1.248956158663883, + "grad_norm": 0.963642954826355, + "learning_rate": 4.581011977755402e-06, + "loss": 0.6283, + "step": 797 + }, + { + "epoch": 1.2505219206680585, + "grad_norm": 1.0119287967681885, + "learning_rate": 4.579843742752694e-06, + "loss": 0.7599, + "step": 798 + }, + { + "epoch": 1.2520876826722338, + "grad_norm": 1.0628082752227783, + "learning_rate": 4.578674030756364e-06, + "loss": 0.7741, + "step": 799 + }, + { + "epoch": 1.2536534446764092, + "grad_norm": 1.0510241985321045, + "learning_rate": 4.577502842597077e-06, + "loss": 0.7615, + "step": 800 + }, + { + "epoch": 1.2552192066805845, + "grad_norm": 1.1180559396743774, + "learning_rate": 4.5763301791065474e-06, + "loss": 0.7119, + "step": 801 + }, + { + "epoch": 1.2567849686847599, + "grad_norm": 1.0198049545288086, + "learning_rate": 4.575156041117541e-06, + "loss": 0.7487, + "step": 802 + }, + { + "epoch": 1.2583507306889352, + "grad_norm": 1.018902063369751, + "learning_rate": 4.5739804294638655e-06, + "loss": 0.7461, + "step": 803 + }, + { + "epoch": 1.2599164926931106, + "grad_norm": 0.9230141639709473, + "learning_rate": 4.572803344980378e-06, + "loss": 0.7295, + "step": 804 + }, + { + "epoch": 1.261482254697286, + "grad_norm": 1.0011619329452515, + "learning_rate": 4.571624788502981e-06, + "loss": 0.7325, + "step": 805 + }, + { + "epoch": 1.2630480167014615, + "grad_norm": 0.9569756388664246, + "learning_rate": 4.570444760868622e-06, + "loss": 0.7494, + "step": 806 + }, + { + "epoch": 1.2646137787056366, + "grad_norm": 0.9860491156578064, + "learning_rate": 4.569263262915293e-06, + "loss": 0.6536, + "step": 807 + }, + { + "epoch": 1.2661795407098122, + "grad_norm": 1.0986074209213257, + "learning_rate": 4.568080295482032e-06, + "loss": 0.7337, + "step": 808 + }, + { + "epoch": 1.2677453027139876, + "grad_norm": 1.0782687664031982, + "learning_rate": 4.566895859408916e-06, + "loss": 0.7193, + "step": 809 + }, + { + "epoch": 1.269311064718163, + "grad_norm": 1.1424410343170166, + "learning_rate": 4.565709955537071e-06, + "loss": 0.6153, + "step": 810 + }, + { + "epoch": 1.2708768267223383, + "grad_norm": 0.9178900718688965, + "learning_rate": 4.564522584708662e-06, + "loss": 0.7802, + "step": 811 + }, + { + "epoch": 1.2724425887265136, + "grad_norm": 1.0006638765335083, + "learning_rate": 4.563333747766896e-06, + "loss": 0.7882, + "step": 812 + }, + { + "epoch": 1.274008350730689, + "grad_norm": 1.0139516592025757, + "learning_rate": 4.56214344555602e-06, + "loss": 0.6562, + "step": 813 + }, + { + "epoch": 1.2755741127348643, + "grad_norm": 1.0360203981399536, + "learning_rate": 4.560951678921324e-06, + "loss": 0.7951, + "step": 814 + }, + { + "epoch": 1.2771398747390397, + "grad_norm": 1.042457103729248, + "learning_rate": 4.559758448709138e-06, + "loss": 0.7572, + "step": 815 + }, + { + "epoch": 1.278705636743215, + "grad_norm": 1.0056653022766113, + "learning_rate": 4.558563755766827e-06, + "loss": 0.7679, + "step": 816 + }, + { + "epoch": 1.2802713987473904, + "grad_norm": 0.9838577508926392, + "learning_rate": 4.5573676009428e-06, + "loss": 0.7245, + "step": 817 + }, + { + "epoch": 1.2818371607515657, + "grad_norm": 1.0474631786346436, + "learning_rate": 4.556169985086503e-06, + "loss": 0.7823, + "step": 818 + }, + { + "epoch": 1.283402922755741, + "grad_norm": 1.0177569389343262, + "learning_rate": 4.554970909048417e-06, + "loss": 0.7599, + "step": 819 + }, + { + "epoch": 1.2849686847599164, + "grad_norm": 1.0120493173599243, + "learning_rate": 4.553770373680062e-06, + "loss": 0.7842, + "step": 820 + }, + { + "epoch": 1.2865344467640918, + "grad_norm": 1.1045918464660645, + "learning_rate": 4.5525683798339946e-06, + "loss": 0.7075, + "step": 821 + }, + { + "epoch": 1.2881002087682671, + "grad_norm": 1.097714900970459, + "learning_rate": 4.551364928363805e-06, + "loss": 0.7571, + "step": 822 + }, + { + "epoch": 1.2896659707724427, + "grad_norm": 0.9125490188598633, + "learning_rate": 4.550160020124121e-06, + "loss": 0.7626, + "step": 823 + }, + { + "epoch": 1.2912317327766178, + "grad_norm": 1.1127262115478516, + "learning_rate": 4.548953655970604e-06, + "loss": 0.6618, + "step": 824 + }, + { + "epoch": 1.2927974947807934, + "grad_norm": 1.0372540950775146, + "learning_rate": 4.547745836759949e-06, + "loss": 0.7467, + "step": 825 + }, + { + "epoch": 1.2943632567849686, + "grad_norm": 1.0158213376998901, + "learning_rate": 4.546536563349885e-06, + "loss": 0.6421, + "step": 826 + }, + { + "epoch": 1.2959290187891441, + "grad_norm": 1.0057018995285034, + "learning_rate": 4.545325836599171e-06, + "loss": 0.7252, + "step": 827 + }, + { + "epoch": 1.2974947807933195, + "grad_norm": 1.0552566051483154, + "learning_rate": 4.544113657367604e-06, + "loss": 0.6645, + "step": 828 + }, + { + "epoch": 1.2990605427974948, + "grad_norm": 1.0798696279525757, + "learning_rate": 4.542900026516006e-06, + "loss": 0.7627, + "step": 829 + }, + { + "epoch": 1.3006263048016702, + "grad_norm": 1.0450047254562378, + "learning_rate": 4.541684944906233e-06, + "loss": 0.705, + "step": 830 + }, + { + "epoch": 1.3021920668058455, + "grad_norm": 0.9975817203521729, + "learning_rate": 4.540468413401171e-06, + "loss": 0.7639, + "step": 831 + }, + { + "epoch": 1.303757828810021, + "grad_norm": 0.9081487059593201, + "learning_rate": 4.539250432864736e-06, + "loss": 0.8129, + "step": 832 + }, + { + "epoch": 1.3053235908141962, + "grad_norm": 1.228684902191162, + "learning_rate": 4.538031004161872e-06, + "loss": 0.6984, + "step": 833 + }, + { + "epoch": 1.3068893528183716, + "grad_norm": 1.0577794313430786, + "learning_rate": 4.536810128158552e-06, + "loss": 0.7436, + "step": 834 + }, + { + "epoch": 1.308455114822547, + "grad_norm": 1.1270350217819214, + "learning_rate": 4.535587805721777e-06, + "loss": 0.7589, + "step": 835 + }, + { + "epoch": 1.3100208768267223, + "grad_norm": 1.1473027467727661, + "learning_rate": 4.5343640377195766e-06, + "loss": 0.7988, + "step": 836 + }, + { + "epoch": 1.3115866388308977, + "grad_norm": 0.9700750112533569, + "learning_rate": 4.533138825021001e-06, + "loss": 0.7388, + "step": 837 + }, + { + "epoch": 1.313152400835073, + "grad_norm": 0.9596260786056519, + "learning_rate": 4.531912168496135e-06, + "loss": 0.8177, + "step": 838 + }, + { + "epoch": 1.3147181628392484, + "grad_norm": 0.9888957738876343, + "learning_rate": 4.530684069016081e-06, + "loss": 0.809, + "step": 839 + }, + { + "epoch": 1.316283924843424, + "grad_norm": 1.219224214553833, + "learning_rate": 4.529454527452972e-06, + "loss": 0.7761, + "step": 840 + }, + { + "epoch": 1.317849686847599, + "grad_norm": 0.9730296730995178, + "learning_rate": 4.5282235446799624e-06, + "loss": 0.8044, + "step": 841 + }, + { + "epoch": 1.3194154488517746, + "grad_norm": 1.1046907901763916, + "learning_rate": 4.526991121571228e-06, + "loss": 0.7452, + "step": 842 + }, + { + "epoch": 1.3209812108559498, + "grad_norm": 0.9707220196723938, + "learning_rate": 4.525757259001972e-06, + "loss": 0.7112, + "step": 843 + }, + { + "epoch": 1.3225469728601253, + "grad_norm": 1.0085593461990356, + "learning_rate": 4.524521957848416e-06, + "loss": 0.6266, + "step": 844 + }, + { + "epoch": 1.3241127348643007, + "grad_norm": 1.131439447402954, + "learning_rate": 4.5232852189878055e-06, + "loss": 0.7393, + "step": 845 + }, + { + "epoch": 1.325678496868476, + "grad_norm": 1.1238782405853271, + "learning_rate": 4.522047043298406e-06, + "loss": 0.7541, + "step": 846 + }, + { + "epoch": 1.3272442588726514, + "grad_norm": 0.9149152636528015, + "learning_rate": 4.520807431659504e-06, + "loss": 0.8162, + "step": 847 + }, + { + "epoch": 1.3288100208768268, + "grad_norm": 1.0428732633590698, + "learning_rate": 4.519566384951403e-06, + "loss": 0.7511, + "step": 848 + }, + { + "epoch": 1.3303757828810021, + "grad_norm": 1.0263090133666992, + "learning_rate": 4.518323904055432e-06, + "loss": 0.7482, + "step": 849 + }, + { + "epoch": 1.3319415448851775, + "grad_norm": 1.113810420036316, + "learning_rate": 4.517079989853931e-06, + "loss": 0.797, + "step": 850 + }, + { + "epoch": 1.3335073068893528, + "grad_norm": 1.0090736150741577, + "learning_rate": 4.515834643230263e-06, + "loss": 0.688, + "step": 851 + }, + { + "epoch": 1.3350730688935282, + "grad_norm": 1.012824296951294, + "learning_rate": 4.514587865068806e-06, + "loss": 0.6683, + "step": 852 + }, + { + "epoch": 1.3366388308977035, + "grad_norm": 0.9442676305770874, + "learning_rate": 4.513339656254954e-06, + "loss": 0.7812, + "step": 853 + }, + { + "epoch": 1.3382045929018789, + "grad_norm": 1.0870879888534546, + "learning_rate": 4.512090017675119e-06, + "loss": 0.6943, + "step": 854 + }, + { + "epoch": 1.3397703549060542, + "grad_norm": 0.9352210164070129, + "learning_rate": 4.510838950216727e-06, + "loss": 0.7571, + "step": 855 + }, + { + "epoch": 1.3413361169102296, + "grad_norm": 1.173621654510498, + "learning_rate": 4.509586454768221e-06, + "loss": 0.6985, + "step": 856 + }, + { + "epoch": 1.342901878914405, + "grad_norm": 0.9790180921554565, + "learning_rate": 4.508332532219053e-06, + "loss": 0.7132, + "step": 857 + }, + { + "epoch": 1.3444676409185803, + "grad_norm": 1.0999163389205933, + "learning_rate": 4.507077183459694e-06, + "loss": 0.7858, + "step": 858 + }, + { + "epoch": 1.3460334029227559, + "grad_norm": 1.018182635307312, + "learning_rate": 4.505820409381625e-06, + "loss": 0.712, + "step": 859 + }, + { + "epoch": 1.347599164926931, + "grad_norm": 1.1015701293945312, + "learning_rate": 4.504562210877338e-06, + "loss": 0.8004, + "step": 860 + }, + { + "epoch": 1.3491649269311066, + "grad_norm": 1.0872273445129395, + "learning_rate": 4.50330258884034e-06, + "loss": 0.7478, + "step": 861 + }, + { + "epoch": 1.350730688935282, + "grad_norm": 1.1098262071609497, + "learning_rate": 4.502041544165148e-06, + "loss": 0.7894, + "step": 862 + }, + { + "epoch": 1.3522964509394573, + "grad_norm": 1.0535573959350586, + "learning_rate": 4.500779077747285e-06, + "loss": 0.7185, + "step": 863 + }, + { + "epoch": 1.3538622129436326, + "grad_norm": 1.0424474477767944, + "learning_rate": 4.499515190483291e-06, + "loss": 0.748, + "step": 864 + }, + { + "epoch": 1.355427974947808, + "grad_norm": 1.005065679550171, + "learning_rate": 4.498249883270708e-06, + "loss": 0.7948, + "step": 865 + }, + { + "epoch": 1.3569937369519833, + "grad_norm": 1.0052721500396729, + "learning_rate": 4.49698315700809e-06, + "loss": 0.7619, + "step": 866 + }, + { + "epoch": 1.3585594989561587, + "grad_norm": 1.0953692197799683, + "learning_rate": 4.495715012594999e-06, + "loss": 0.7363, + "step": 867 + }, + { + "epoch": 1.360125260960334, + "grad_norm": 1.1238093376159668, + "learning_rate": 4.494445450932003e-06, + "loss": 0.6576, + "step": 868 + }, + { + "epoch": 1.3616910229645094, + "grad_norm": 0.8546008467674255, + "learning_rate": 4.493174472920676e-06, + "loss": 0.7513, + "step": 869 + }, + { + "epoch": 1.3632567849686847, + "grad_norm": 1.0863126516342163, + "learning_rate": 4.4919020794635995e-06, + "loss": 0.6886, + "step": 870 + }, + { + "epoch": 1.36482254697286, + "grad_norm": 0.9943820834159851, + "learning_rate": 4.490628271464359e-06, + "loss": 0.7253, + "step": 871 + }, + { + "epoch": 1.3663883089770354, + "grad_norm": 0.9377979636192322, + "learning_rate": 4.4893530498275436e-06, + "loss": 0.7026, + "step": 872 + }, + { + "epoch": 1.3679540709812108, + "grad_norm": 1.100799798965454, + "learning_rate": 4.488076415458749e-06, + "loss": 0.719, + "step": 873 + }, + { + "epoch": 1.3695198329853862, + "grad_norm": 1.015074610710144, + "learning_rate": 4.486798369264572e-06, + "loss": 0.7513, + "step": 874 + }, + { + "epoch": 1.3710855949895615, + "grad_norm": 1.047530174255371, + "learning_rate": 4.485518912152611e-06, + "loss": 0.7757, + "step": 875 + }, + { + "epoch": 1.372651356993737, + "grad_norm": 1.0264610052108765, + "learning_rate": 4.484238045031471e-06, + "loss": 0.7712, + "step": 876 + }, + { + "epoch": 1.3742171189979122, + "grad_norm": 0.9848430156707764, + "learning_rate": 4.482955768810752e-06, + "loss": 0.7569, + "step": 877 + }, + { + "epoch": 1.3757828810020878, + "grad_norm": 1.070816159248352, + "learning_rate": 4.48167208440106e-06, + "loss": 0.7857, + "step": 878 + }, + { + "epoch": 1.377348643006263, + "grad_norm": 1.0134745836257935, + "learning_rate": 4.480386992713998e-06, + "loss": 0.7372, + "step": 879 + }, + { + "epoch": 1.3789144050104385, + "grad_norm": 1.0736140012741089, + "learning_rate": 4.47910049466217e-06, + "loss": 0.7794, + "step": 880 + }, + { + "epoch": 1.3804801670146138, + "grad_norm": 1.0163054466247559, + "learning_rate": 4.477812591159176e-06, + "loss": 0.7341, + "step": 881 + }, + { + "epoch": 1.3820459290187892, + "grad_norm": 1.074578881263733, + "learning_rate": 4.476523283119619e-06, + "loss": 0.6924, + "step": 882 + }, + { + "epoch": 1.3836116910229646, + "grad_norm": 1.183667540550232, + "learning_rate": 4.475232571459095e-06, + "loss": 0.7866, + "step": 883 + }, + { + "epoch": 1.38517745302714, + "grad_norm": 1.030827522277832, + "learning_rate": 4.473940457094199e-06, + "loss": 0.7845, + "step": 884 + }, + { + "epoch": 1.3867432150313153, + "grad_norm": 1.0043658018112183, + "learning_rate": 4.47264694094252e-06, + "loss": 0.7399, + "step": 885 + }, + { + "epoch": 1.3883089770354906, + "grad_norm": 1.1135472059249878, + "learning_rate": 4.471352023922645e-06, + "loss": 0.771, + "step": 886 + }, + { + "epoch": 1.389874739039666, + "grad_norm": 1.0338670015335083, + "learning_rate": 4.470055706954154e-06, + "loss": 0.7282, + "step": 887 + }, + { + "epoch": 1.3914405010438413, + "grad_norm": 0.918855607509613, + "learning_rate": 4.468757990957623e-06, + "loss": 0.728, + "step": 888 + }, + { + "epoch": 1.3930062630480167, + "grad_norm": 1.06833016872406, + "learning_rate": 4.467458876854619e-06, + "loss": 0.7012, + "step": 889 + }, + { + "epoch": 1.394572025052192, + "grad_norm": 1.0103005170822144, + "learning_rate": 4.4661583655677045e-06, + "loss": 0.7074, + "step": 890 + }, + { + "epoch": 1.3961377870563674, + "grad_norm": 1.0418543815612793, + "learning_rate": 4.464856458020433e-06, + "loss": 0.6858, + "step": 891 + }, + { + "epoch": 1.3977035490605427, + "grad_norm": 0.9720668196678162, + "learning_rate": 4.463553155137348e-06, + "loss": 0.7024, + "step": 892 + }, + { + "epoch": 1.399269311064718, + "grad_norm": 1.1870561838150024, + "learning_rate": 4.462248457843986e-06, + "loss": 0.8069, + "step": 893 + }, + { + "epoch": 1.4008350730688934, + "grad_norm": 0.9888094663619995, + "learning_rate": 4.460942367066874e-06, + "loss": 0.7281, + "step": 894 + }, + { + "epoch": 1.402400835073069, + "grad_norm": 0.9120670557022095, + "learning_rate": 4.459634883733527e-06, + "loss": 0.7992, + "step": 895 + }, + { + "epoch": 1.4039665970772441, + "grad_norm": 1.0239481925964355, + "learning_rate": 4.45832600877245e-06, + "loss": 0.7312, + "step": 896 + }, + { + "epoch": 1.4055323590814197, + "grad_norm": 0.9379130005836487, + "learning_rate": 4.4570157431131355e-06, + "loss": 0.7019, + "step": 897 + }, + { + "epoch": 1.407098121085595, + "grad_norm": 1.0789109468460083, + "learning_rate": 4.455704087686063e-06, + "loss": 0.8149, + "step": 898 + }, + { + "epoch": 1.4086638830897704, + "grad_norm": 1.0521917343139648, + "learning_rate": 4.454391043422703e-06, + "loss": 0.7873, + "step": 899 + }, + { + "epoch": 1.4102296450939458, + "grad_norm": 0.9582643508911133, + "learning_rate": 4.453076611255507e-06, + "loss": 0.7815, + "step": 900 + }, + { + "epoch": 1.4117954070981211, + "grad_norm": 1.1096739768981934, + "learning_rate": 4.451760792117914e-06, + "loss": 0.7636, + "step": 901 + }, + { + "epoch": 1.4133611691022965, + "grad_norm": 1.043574333190918, + "learning_rate": 4.450443586944349e-06, + "loss": 0.7411, + "step": 902 + }, + { + "epoch": 1.4149269311064718, + "grad_norm": 1.0280033349990845, + "learning_rate": 4.449124996670221e-06, + "loss": 0.6525, + "step": 903 + }, + { + "epoch": 1.4164926931106472, + "grad_norm": 1.0329992771148682, + "learning_rate": 4.44780502223192e-06, + "loss": 0.7133, + "step": 904 + }, + { + "epoch": 1.4180584551148225, + "grad_norm": 1.016168475151062, + "learning_rate": 4.446483664566825e-06, + "loss": 0.7098, + "step": 905 + }, + { + "epoch": 1.4196242171189979, + "grad_norm": 1.2022786140441895, + "learning_rate": 4.4451609246132895e-06, + "loss": 0.7303, + "step": 906 + }, + { + "epoch": 1.4211899791231732, + "grad_norm": 1.0311683416366577, + "learning_rate": 4.443836803310655e-06, + "loss": 0.7475, + "step": 907 + }, + { + "epoch": 1.4227557411273486, + "grad_norm": 1.0073823928833008, + "learning_rate": 4.442511301599241e-06, + "loss": 0.7513, + "step": 908 + }, + { + "epoch": 1.424321503131524, + "grad_norm": 1.0702322721481323, + "learning_rate": 4.441184420420347e-06, + "loss": 0.6918, + "step": 909 + }, + { + "epoch": 1.4258872651356993, + "grad_norm": 1.1416171789169312, + "learning_rate": 4.439856160716254e-06, + "loss": 0.6324, + "step": 910 + }, + { + "epoch": 1.4274530271398747, + "grad_norm": 1.0560253858566284, + "learning_rate": 4.438526523430221e-06, + "loss": 0.749, + "step": 911 + }, + { + "epoch": 1.4290187891440502, + "grad_norm": 1.021008849143982, + "learning_rate": 4.4371955095064835e-06, + "loss": 0.7666, + "step": 912 + }, + { + "epoch": 1.4305845511482254, + "grad_norm": 1.151904821395874, + "learning_rate": 4.435863119890258e-06, + "loss": 0.6904, + "step": 913 + }, + { + "epoch": 1.432150313152401, + "grad_norm": 1.1216702461242676, + "learning_rate": 4.434529355527736e-06, + "loss": 0.7068, + "step": 914 + }, + { + "epoch": 1.433716075156576, + "grad_norm": 1.1130588054656982, + "learning_rate": 4.433194217366085e-06, + "loss": 0.6524, + "step": 915 + }, + { + "epoch": 1.4352818371607516, + "grad_norm": 1.07612144947052, + "learning_rate": 4.431857706353449e-06, + "loss": 0.8078, + "step": 916 + }, + { + "epoch": 1.436847599164927, + "grad_norm": 1.0220447778701782, + "learning_rate": 4.430519823438946e-06, + "loss": 0.6736, + "step": 917 + }, + { + "epoch": 1.4384133611691023, + "grad_norm": 1.0657141208648682, + "learning_rate": 4.429180569572669e-06, + "loss": 0.7457, + "step": 918 + }, + { + "epoch": 1.4399791231732777, + "grad_norm": 1.0646854639053345, + "learning_rate": 4.427839945705684e-06, + "loss": 0.7925, + "step": 919 + }, + { + "epoch": 1.441544885177453, + "grad_norm": 1.0619502067565918, + "learning_rate": 4.426497952790031e-06, + "loss": 0.6968, + "step": 920 + }, + { + "epoch": 1.4431106471816284, + "grad_norm": 0.9684338569641113, + "learning_rate": 4.425154591778722e-06, + "loss": 0.7411, + "step": 921 + }, + { + "epoch": 1.4446764091858038, + "grad_norm": 1.1225167512893677, + "learning_rate": 4.423809863625739e-06, + "loss": 0.7623, + "step": 922 + }, + { + "epoch": 1.446242171189979, + "grad_norm": 0.9310746788978577, + "learning_rate": 4.422463769286036e-06, + "loss": 0.8219, + "step": 923 + }, + { + "epoch": 1.4478079331941545, + "grad_norm": 1.1107957363128662, + "learning_rate": 4.4211163097155375e-06, + "loss": 0.7755, + "step": 924 + }, + { + "epoch": 1.4493736951983298, + "grad_norm": 0.9827596545219421, + "learning_rate": 4.419767485871136e-06, + "loss": 0.7102, + "step": 925 + }, + { + "epoch": 1.4509394572025052, + "grad_norm": 1.1075087785720825, + "learning_rate": 4.418417298710695e-06, + "loss": 0.6399, + "step": 926 + }, + { + "epoch": 1.4525052192066805, + "grad_norm": 0.9575802087783813, + "learning_rate": 4.417065749193045e-06, + "loss": 0.721, + "step": 927 + }, + { + "epoch": 1.4540709812108559, + "grad_norm": 1.0016006231307983, + "learning_rate": 4.415712838277985e-06, + "loss": 0.762, + "step": 928 + }, + { + "epoch": 1.4556367432150314, + "grad_norm": 1.1907106637954712, + "learning_rate": 4.414358566926279e-06, + "loss": 0.7904, + "step": 929 + }, + { + "epoch": 1.4572025052192066, + "grad_norm": 1.0011543035507202, + "learning_rate": 4.413002936099656e-06, + "loss": 0.7179, + "step": 930 + }, + { + "epoch": 1.4587682672233822, + "grad_norm": 0.980684757232666, + "learning_rate": 4.411645946760817e-06, + "loss": 0.6845, + "step": 931 + }, + { + "epoch": 1.4603340292275573, + "grad_norm": 0.9458135366439819, + "learning_rate": 4.4102875998734176e-06, + "loss": 0.6911, + "step": 932 + }, + { + "epoch": 1.4618997912317329, + "grad_norm": 1.020966649055481, + "learning_rate": 4.408927896402087e-06, + "loss": 0.7284, + "step": 933 + }, + { + "epoch": 1.4634655532359082, + "grad_norm": 1.041833519935608, + "learning_rate": 4.407566837312413e-06, + "loss": 0.7451, + "step": 934 + }, + { + "epoch": 1.4650313152400836, + "grad_norm": 0.9076465964317322, + "learning_rate": 4.406204423570946e-06, + "loss": 0.7553, + "step": 935 + }, + { + "epoch": 1.466597077244259, + "grad_norm": 1.1963952779769897, + "learning_rate": 4.404840656145199e-06, + "loss": 0.7437, + "step": 936 + }, + { + "epoch": 1.4681628392484343, + "grad_norm": 1.0181536674499512, + "learning_rate": 4.403475536003648e-06, + "loss": 0.7283, + "step": 937 + }, + { + "epoch": 1.4697286012526096, + "grad_norm": 0.9563003778457642, + "learning_rate": 4.402109064115727e-06, + "loss": 0.7375, + "step": 938 + }, + { + "epoch": 1.471294363256785, + "grad_norm": 1.05129873752594, + "learning_rate": 4.400741241451832e-06, + "loss": 0.6805, + "step": 939 + }, + { + "epoch": 1.4728601252609603, + "grad_norm": 1.0912166833877563, + "learning_rate": 4.399372068983317e-06, + "loss": 0.7076, + "step": 940 + }, + { + "epoch": 1.4744258872651357, + "grad_norm": 1.108500599861145, + "learning_rate": 4.398001547682494e-06, + "loss": 0.6836, + "step": 941 + }, + { + "epoch": 1.475991649269311, + "grad_norm": 1.0741651058197021, + "learning_rate": 4.396629678522636e-06, + "loss": 0.7405, + "step": 942 + }, + { + "epoch": 1.4775574112734864, + "grad_norm": 1.0603930950164795, + "learning_rate": 4.39525646247797e-06, + "loss": 0.7602, + "step": 943 + }, + { + "epoch": 1.4791231732776617, + "grad_norm": 0.9814990162849426, + "learning_rate": 4.39388190052368e-06, + "loss": 0.7324, + "step": 944 + }, + { + "epoch": 1.480688935281837, + "grad_norm": 0.9643695950508118, + "learning_rate": 4.392505993635906e-06, + "loss": 0.7034, + "step": 945 + }, + { + "epoch": 1.4822546972860124, + "grad_norm": 1.0490994453430176, + "learning_rate": 4.391128742791743e-06, + "loss": 0.7358, + "step": 946 + }, + { + "epoch": 1.4838204592901878, + "grad_norm": 0.9223467111587524, + "learning_rate": 4.389750148969244e-06, + "loss": 0.698, + "step": 947 + }, + { + "epoch": 1.4853862212943634, + "grad_norm": 1.1107758283615112, + "learning_rate": 4.388370213147409e-06, + "loss": 0.7264, + "step": 948 + }, + { + "epoch": 1.4869519832985385, + "grad_norm": 1.05939781665802, + "learning_rate": 4.386988936306196e-06, + "loss": 0.7209, + "step": 949 + }, + { + "epoch": 1.488517745302714, + "grad_norm": 1.024686574935913, + "learning_rate": 4.385606319426512e-06, + "loss": 0.7798, + "step": 950 + }, + { + "epoch": 1.4900835073068894, + "grad_norm": 0.9920977354049683, + "learning_rate": 4.3842223634902205e-06, + "loss": 0.8045, + "step": 951 + }, + { + "epoch": 1.4916492693110648, + "grad_norm": 0.950927197933197, + "learning_rate": 4.382837069480131e-06, + "loss": 0.6777, + "step": 952 + }, + { + "epoch": 1.4932150313152401, + "grad_norm": 1.0072081089019775, + "learning_rate": 4.3814504383800056e-06, + "loss": 0.7458, + "step": 953 + }, + { + "epoch": 1.4947807933194155, + "grad_norm": 0.9982030987739563, + "learning_rate": 4.3800624711745546e-06, + "loss": 0.8215, + "step": 954 + }, + { + "epoch": 1.4963465553235908, + "grad_norm": 0.9682126641273499, + "learning_rate": 4.3786731688494396e-06, + "loss": 0.7157, + "step": 955 + }, + { + "epoch": 1.4979123173277662, + "grad_norm": 1.1599830389022827, + "learning_rate": 4.377282532391267e-06, + "loss": 0.8091, + "step": 956 + }, + { + "epoch": 1.4994780793319415, + "grad_norm": 1.0624948740005493, + "learning_rate": 4.3758905627875934e-06, + "loss": 0.7428, + "step": 957 + }, + { + "epoch": 1.501043841336117, + "grad_norm": 1.14694082736969, + "learning_rate": 4.374497261026921e-06, + "loss": 0.7752, + "step": 958 + }, + { + "epoch": 1.5026096033402923, + "grad_norm": 1.0355825424194336, + "learning_rate": 4.373102628098697e-06, + "loss": 0.7348, + "step": 959 + }, + { + "epoch": 1.5041753653444676, + "grad_norm": 1.036026954650879, + "learning_rate": 4.371706664993316e-06, + "loss": 0.7536, + "step": 960 + }, + { + "epoch": 1.505741127348643, + "grad_norm": 1.2319790124893188, + "learning_rate": 4.370309372702116e-06, + "loss": 0.6985, + "step": 961 + }, + { + "epoch": 1.5073068893528183, + "grad_norm": 0.908687949180603, + "learning_rate": 4.36891075221738e-06, + "loss": 0.7173, + "step": 962 + }, + { + "epoch": 1.5088726513569939, + "grad_norm": 1.0482099056243896, + "learning_rate": 4.367510804532331e-06, + "loss": 0.7514, + "step": 963 + }, + { + "epoch": 1.510438413361169, + "grad_norm": 1.0434144735336304, + "learning_rate": 4.36610953064114e-06, + "loss": 0.754, + "step": 964 + }, + { + "epoch": 1.5120041753653446, + "grad_norm": 1.0122450590133667, + "learning_rate": 4.364706931538914e-06, + "loss": 0.7767, + "step": 965 + }, + { + "epoch": 1.5135699373695197, + "grad_norm": 1.0471900701522827, + "learning_rate": 4.363303008221705e-06, + "loss": 0.7337, + "step": 966 + }, + { + "epoch": 1.5151356993736953, + "grad_norm": 1.0691189765930176, + "learning_rate": 4.3618977616865045e-06, + "loss": 0.8047, + "step": 967 + }, + { + "epoch": 1.5167014613778704, + "grad_norm": 1.015114188194275, + "learning_rate": 4.360491192931242e-06, + "loss": 0.726, + "step": 968 + }, + { + "epoch": 1.518267223382046, + "grad_norm": 0.9770560264587402, + "learning_rate": 4.359083302954788e-06, + "loss": 0.7692, + "step": 969 + }, + { + "epoch": 1.5198329853862211, + "grad_norm": 1.079991340637207, + "learning_rate": 4.357674092756951e-06, + "loss": 0.8424, + "step": 970 + }, + { + "epoch": 1.5213987473903967, + "grad_norm": 1.068708896636963, + "learning_rate": 4.356263563338476e-06, + "loss": 0.758, + "step": 971 + }, + { + "epoch": 1.522964509394572, + "grad_norm": 1.1005477905273438, + "learning_rate": 4.354851715701046e-06, + "loss": 0.7209, + "step": 972 + }, + { + "epoch": 1.5245302713987474, + "grad_norm": 0.9870902299880981, + "learning_rate": 4.3534385508472786e-06, + "loss": 0.7272, + "step": 973 + }, + { + "epoch": 1.5260960334029228, + "grad_norm": 0.9616238474845886, + "learning_rate": 4.352024069780729e-06, + "loss": 0.8124, + "step": 974 + }, + { + "epoch": 1.5276617954070981, + "grad_norm": 1.0341622829437256, + "learning_rate": 4.3506082735058855e-06, + "loss": 0.7434, + "step": 975 + }, + { + "epoch": 1.5292275574112735, + "grad_norm": 1.0157302618026733, + "learning_rate": 4.349191163028171e-06, + "loss": 0.7534, + "step": 976 + }, + { + "epoch": 1.5307933194154488, + "grad_norm": 1.041379690170288, + "learning_rate": 4.347772739353939e-06, + "loss": 0.6942, + "step": 977 + }, + { + "epoch": 1.5323590814196242, + "grad_norm": 1.016205072402954, + "learning_rate": 4.346353003490481e-06, + "loss": 0.6707, + "step": 978 + }, + { + "epoch": 1.5339248434237995, + "grad_norm": 1.0034714937210083, + "learning_rate": 4.344931956446017e-06, + "loss": 0.6358, + "step": 979 + }, + { + "epoch": 1.535490605427975, + "grad_norm": 0.9671767354011536, + "learning_rate": 4.343509599229697e-06, + "loss": 0.6333, + "step": 980 + }, + { + "epoch": 1.5370563674321502, + "grad_norm": 1.1384506225585938, + "learning_rate": 4.342085932851605e-06, + "loss": 0.803, + "step": 981 + }, + { + "epoch": 1.5386221294363258, + "grad_norm": 1.023031234741211, + "learning_rate": 4.34066095832275e-06, + "loss": 0.7801, + "step": 982 + }, + { + "epoch": 1.540187891440501, + "grad_norm": 1.0675241947174072, + "learning_rate": 4.339234676655075e-06, + "loss": 0.7335, + "step": 983 + }, + { + "epoch": 1.5417536534446765, + "grad_norm": 1.0397543907165527, + "learning_rate": 4.337807088861447e-06, + "loss": 0.7722, + "step": 984 + }, + { + "epoch": 1.5433194154488517, + "grad_norm": 1.0892809629440308, + "learning_rate": 4.3363781959556635e-06, + "loss": 0.7603, + "step": 985 + }, + { + "epoch": 1.5448851774530272, + "grad_norm": 0.9335037469863892, + "learning_rate": 4.334947998952448e-06, + "loss": 0.7678, + "step": 986 + }, + { + "epoch": 1.5464509394572024, + "grad_norm": 1.0991342067718506, + "learning_rate": 4.333516498867449e-06, + "loss": 0.7461, + "step": 987 + }, + { + "epoch": 1.548016701461378, + "grad_norm": 1.112265944480896, + "learning_rate": 4.332083696717242e-06, + "loss": 0.7593, + "step": 988 + }, + { + "epoch": 1.5495824634655533, + "grad_norm": 0.9525822997093201, + "learning_rate": 4.330649593519325e-06, + "loss": 0.7629, + "step": 989 + }, + { + "epoch": 1.5511482254697286, + "grad_norm": 1.0624464750289917, + "learning_rate": 4.329214190292122e-06, + "loss": 0.7261, + "step": 990 + }, + { + "epoch": 1.552713987473904, + "grad_norm": 1.1969516277313232, + "learning_rate": 4.327777488054979e-06, + "loss": 0.7483, + "step": 991 + }, + { + "epoch": 1.5542797494780793, + "grad_norm": 1.0239863395690918, + "learning_rate": 4.326339487828167e-06, + "loss": 0.7265, + "step": 992 + }, + { + "epoch": 1.5558455114822547, + "grad_norm": 1.0881083011627197, + "learning_rate": 4.324900190632874e-06, + "loss": 0.724, + "step": 993 + }, + { + "epoch": 1.55741127348643, + "grad_norm": 1.098827838897705, + "learning_rate": 4.323459597491214e-06, + "loss": 0.7222, + "step": 994 + }, + { + "epoch": 1.5589770354906054, + "grad_norm": 1.105057716369629, + "learning_rate": 4.322017709426217e-06, + "loss": 0.7827, + "step": 995 + }, + { + "epoch": 1.5605427974947808, + "grad_norm": 0.9830679297447205, + "learning_rate": 4.3205745274618365e-06, + "loss": 0.7316, + "step": 996 + }, + { + "epoch": 1.562108559498956, + "grad_norm": 0.9951156973838806, + "learning_rate": 4.319130052622942e-06, + "loss": 0.7181, + "step": 997 + }, + { + "epoch": 1.5636743215031315, + "grad_norm": 1.0374284982681274, + "learning_rate": 4.317684285935323e-06, + "loss": 0.7355, + "step": 998 + }, + { + "epoch": 1.565240083507307, + "grad_norm": 1.0139906406402588, + "learning_rate": 4.316237228425683e-06, + "loss": 0.7249, + "step": 999 + }, + { + "epoch": 1.5668058455114822, + "grad_norm": 0.9555441737174988, + "learning_rate": 4.3147888811216496e-06, + "loss": 0.7387, + "step": 1000 + }, + { + "epoch": 1.5683716075156577, + "grad_norm": 1.0600265264511108, + "learning_rate": 4.313339245051758e-06, + "loss": 0.7445, + "step": 1001 + }, + { + "epoch": 1.5699373695198329, + "grad_norm": 1.0070478916168213, + "learning_rate": 4.311888321245461e-06, + "loss": 0.7605, + "step": 1002 + }, + { + "epoch": 1.5715031315240084, + "grad_norm": 1.019710898399353, + "learning_rate": 4.31043611073313e-06, + "loss": 0.7262, + "step": 1003 + }, + { + "epoch": 1.5730688935281836, + "grad_norm": 1.0476088523864746, + "learning_rate": 4.308982614546045e-06, + "loss": 0.6766, + "step": 1004 + }, + { + "epoch": 1.5746346555323592, + "grad_norm": 1.0670830011367798, + "learning_rate": 4.307527833716403e-06, + "loss": 0.7465, + "step": 1005 + }, + { + "epoch": 1.5762004175365343, + "grad_norm": 1.1807290315628052, + "learning_rate": 4.30607176927731e-06, + "loss": 0.7032, + "step": 1006 + }, + { + "epoch": 1.5777661795407099, + "grad_norm": 1.0482438802719116, + "learning_rate": 4.3046144222627855e-06, + "loss": 0.7156, + "step": 1007 + }, + { + "epoch": 1.5793319415448852, + "grad_norm": 0.9434391856193542, + "learning_rate": 4.303155793707759e-06, + "loss": 0.7816, + "step": 1008 + }, + { + "epoch": 1.5808977035490606, + "grad_norm": 1.0497010946273804, + "learning_rate": 4.3016958846480706e-06, + "loss": 0.735, + "step": 1009 + }, + { + "epoch": 1.582463465553236, + "grad_norm": 0.955949068069458, + "learning_rate": 4.300234696120469e-06, + "loss": 0.6414, + "step": 1010 + }, + { + "epoch": 1.5840292275574113, + "grad_norm": 1.0462027788162231, + "learning_rate": 4.298772229162613e-06, + "loss": 0.6949, + "step": 1011 + }, + { + "epoch": 1.5855949895615866, + "grad_norm": 1.0634411573410034, + "learning_rate": 4.297308484813067e-06, + "loss": 0.7736, + "step": 1012 + }, + { + "epoch": 1.587160751565762, + "grad_norm": 1.0463027954101562, + "learning_rate": 4.295843464111306e-06, + "loss": 0.6812, + "step": 1013 + }, + { + "epoch": 1.5887265135699373, + "grad_norm": 1.0409252643585205, + "learning_rate": 4.294377168097706e-06, + "loss": 0.7319, + "step": 1014 + }, + { + "epoch": 1.5902922755741127, + "grad_norm": 0.9714667201042175, + "learning_rate": 4.292909597813555e-06, + "loss": 0.7473, + "step": 1015 + }, + { + "epoch": 1.5918580375782883, + "grad_norm": 1.03997802734375, + "learning_rate": 4.291440754301041e-06, + "loss": 0.7261, + "step": 1016 + }, + { + "epoch": 1.5934237995824634, + "grad_norm": 1.063293218612671, + "learning_rate": 4.289970638603258e-06, + "loss": 0.7909, + "step": 1017 + }, + { + "epoch": 1.594989561586639, + "grad_norm": 1.110007643699646, + "learning_rate": 4.288499251764205e-06, + "loss": 0.7, + "step": 1018 + }, + { + "epoch": 1.596555323590814, + "grad_norm": 0.981261134147644, + "learning_rate": 4.2870265948287806e-06, + "loss": 0.7516, + "step": 1019 + }, + { + "epoch": 1.5981210855949897, + "grad_norm": 1.0653973817825317, + "learning_rate": 4.2855526688427875e-06, + "loss": 0.7278, + "step": 1020 + }, + { + "epoch": 1.5996868475991648, + "grad_norm": 0.9212517738342285, + "learning_rate": 4.284077474852928e-06, + "loss": 0.7251, + "step": 1021 + }, + { + "epoch": 1.6012526096033404, + "grad_norm": 1.0887659788131714, + "learning_rate": 4.282601013906808e-06, + "loss": 0.7271, + "step": 1022 + }, + { + "epoch": 1.6028183716075155, + "grad_norm": 1.0753779411315918, + "learning_rate": 4.281123287052928e-06, + "loss": 0.7679, + "step": 1023 + }, + { + "epoch": 1.604384133611691, + "grad_norm": 1.0282115936279297, + "learning_rate": 4.2796442953406935e-06, + "loss": 0.695, + "step": 1024 + }, + { + "epoch": 1.6059498956158664, + "grad_norm": 0.9782281517982483, + "learning_rate": 4.2781640398204036e-06, + "loss": 0.7717, + "step": 1025 + }, + { + "epoch": 1.6075156576200418, + "grad_norm": 1.06228768825531, + "learning_rate": 4.276682521543256e-06, + "loss": 0.8069, + "step": 1026 + }, + { + "epoch": 1.6090814196242171, + "grad_norm": 0.8853397965431213, + "learning_rate": 4.2751997415613465e-06, + "loss": 0.8174, + "step": 1027 + }, + { + "epoch": 1.6106471816283925, + "grad_norm": 1.1525455713272095, + "learning_rate": 4.273715700927666e-06, + "loss": 0.7149, + "step": 1028 + }, + { + "epoch": 1.6122129436325678, + "grad_norm": 0.8743720650672913, + "learning_rate": 4.272230400696101e-06, + "loss": 0.7296, + "step": 1029 + }, + { + "epoch": 1.6137787056367432, + "grad_norm": 1.1136301755905151, + "learning_rate": 4.270743841921431e-06, + "loss": 0.7539, + "step": 1030 + }, + { + "epoch": 1.6153444676409185, + "grad_norm": 0.9830031991004944, + "learning_rate": 4.269256025659332e-06, + "loss": 0.6545, + "step": 1031 + }, + { + "epoch": 1.616910229645094, + "grad_norm": 1.1221113204956055, + "learning_rate": 4.267766952966369e-06, + "loss": 0.6144, + "step": 1032 + }, + { + "epoch": 1.6184759916492695, + "grad_norm": 0.975713849067688, + "learning_rate": 4.266276624900004e-06, + "loss": 0.746, + "step": 1033 + }, + { + "epoch": 1.6200417536534446, + "grad_norm": 1.1131149530410767, + "learning_rate": 4.264785042518587e-06, + "loss": 0.705, + "step": 1034 + }, + { + "epoch": 1.6216075156576202, + "grad_norm": 1.0574990510940552, + "learning_rate": 4.263292206881361e-06, + "loss": 0.7681, + "step": 1035 + }, + { + "epoch": 1.6231732776617953, + "grad_norm": 0.9931074976921082, + "learning_rate": 4.261798119048456e-06, + "loss": 0.7854, + "step": 1036 + }, + { + "epoch": 1.6247390396659709, + "grad_norm": 1.1598386764526367, + "learning_rate": 4.260302780080896e-06, + "loss": 0.7406, + "step": 1037 + }, + { + "epoch": 1.626304801670146, + "grad_norm": 0.9910677671432495, + "learning_rate": 4.258806191040587e-06, + "loss": 0.7107, + "step": 1038 + }, + { + "epoch": 1.6278705636743216, + "grad_norm": 1.1023072004318237, + "learning_rate": 4.2573083529903294e-06, + "loss": 0.7964, + "step": 1039 + }, + { + "epoch": 1.6294363256784967, + "grad_norm": 0.991082489490509, + "learning_rate": 4.255809266993806e-06, + "loss": 0.762, + "step": 1040 + }, + { + "epoch": 1.6310020876826723, + "grad_norm": 0.9281193017959595, + "learning_rate": 4.254308934115589e-06, + "loss": 0.7272, + "step": 1041 + }, + { + "epoch": 1.6325678496868476, + "grad_norm": 1.20292067527771, + "learning_rate": 4.252807355421133e-06, + "loss": 0.7535, + "step": 1042 + }, + { + "epoch": 1.634133611691023, + "grad_norm": 1.0362178087234497, + "learning_rate": 4.251304531976778e-06, + "loss": 0.7524, + "step": 1043 + }, + { + "epoch": 1.6356993736951984, + "grad_norm": 1.0267268419265747, + "learning_rate": 4.249800464849751e-06, + "loss": 0.6588, + "step": 1044 + }, + { + "epoch": 1.6372651356993737, + "grad_norm": 1.0531187057495117, + "learning_rate": 4.248295155108158e-06, + "loss": 0.7328, + "step": 1045 + }, + { + "epoch": 1.638830897703549, + "grad_norm": 1.0008069276809692, + "learning_rate": 4.246788603820989e-06, + "loss": 0.719, + "step": 1046 + }, + { + "epoch": 1.6403966597077244, + "grad_norm": 1.064780831336975, + "learning_rate": 4.245280812058119e-06, + "loss": 0.7388, + "step": 1047 + }, + { + "epoch": 1.6419624217118998, + "grad_norm": 1.06492018699646, + "learning_rate": 4.243771780890298e-06, + "loss": 0.7495, + "step": 1048 + }, + { + "epoch": 1.6435281837160751, + "grad_norm": 0.9472903609275818, + "learning_rate": 4.24226151138916e-06, + "loss": 0.7409, + "step": 1049 + }, + { + "epoch": 1.6450939457202505, + "grad_norm": 1.0612895488739014, + "learning_rate": 4.240750004627217e-06, + "loss": 0.7605, + "step": 1050 + }, + { + "epoch": 1.6466597077244258, + "grad_norm": 0.9634467959403992, + "learning_rate": 4.239237261677863e-06, + "loss": 0.7109, + "step": 1051 + }, + { + "epoch": 1.6482254697286014, + "grad_norm": 1.1347044706344604, + "learning_rate": 4.2377232836153635e-06, + "loss": 0.693, + "step": 1052 + }, + { + "epoch": 1.6497912317327765, + "grad_norm": 1.0814287662506104, + "learning_rate": 4.236208071514865e-06, + "loss": 0.7159, + "step": 1053 + }, + { + "epoch": 1.651356993736952, + "grad_norm": 0.9427600502967834, + "learning_rate": 4.234691626452392e-06, + "loss": 0.7014, + "step": 1054 + }, + { + "epoch": 1.6529227557411272, + "grad_norm": 1.0177332162857056, + "learning_rate": 4.233173949504842e-06, + "loss": 0.7717, + "step": 1055 + }, + { + "epoch": 1.6544885177453028, + "grad_norm": 1.0621635913848877, + "learning_rate": 4.231655041749985e-06, + "loss": 0.7078, + "step": 1056 + }, + { + "epoch": 1.656054279749478, + "grad_norm": 1.0474066734313965, + "learning_rate": 4.230134904266472e-06, + "loss": 0.7869, + "step": 1057 + }, + { + "epoch": 1.6576200417536535, + "grad_norm": 0.9170332551002502, + "learning_rate": 4.228613538133821e-06, + "loss": 0.7806, + "step": 1058 + }, + { + "epoch": 1.6591858037578286, + "grad_norm": 1.010465383529663, + "learning_rate": 4.227090944432426e-06, + "loss": 0.7352, + "step": 1059 + }, + { + "epoch": 1.6607515657620042, + "grad_norm": 0.988068163394928, + "learning_rate": 4.22556712424355e-06, + "loss": 0.7173, + "step": 1060 + }, + { + "epoch": 1.6623173277661796, + "grad_norm": 1.0753575563430786, + "learning_rate": 4.22404207864933e-06, + "loss": 0.6981, + "step": 1061 + }, + { + "epoch": 1.663883089770355, + "grad_norm": 1.0075691938400269, + "learning_rate": 4.222515808732769e-06, + "loss": 0.7279, + "step": 1062 + }, + { + "epoch": 1.6654488517745303, + "grad_norm": 1.0954983234405518, + "learning_rate": 4.220988315577745e-06, + "loss": 0.7675, + "step": 1063 + }, + { + "epoch": 1.6670146137787056, + "grad_norm": 1.0368746519088745, + "learning_rate": 4.2194596002690016e-06, + "loss": 0.7466, + "step": 1064 + }, + { + "epoch": 1.668580375782881, + "grad_norm": 1.083906650543213, + "learning_rate": 4.217929663892149e-06, + "loss": 0.7517, + "step": 1065 + }, + { + "epoch": 1.6701461377870563, + "grad_norm": 1.1356595754623413, + "learning_rate": 4.216398507533667e-06, + "loss": 0.7601, + "step": 1066 + }, + { + "epoch": 1.6717118997912317, + "grad_norm": 1.1610008478164673, + "learning_rate": 4.2148661322809e-06, + "loss": 0.7356, + "step": 1067 + }, + { + "epoch": 1.673277661795407, + "grad_norm": 1.019673466682434, + "learning_rate": 4.21333253922206e-06, + "loss": 0.7618, + "step": 1068 + }, + { + "epoch": 1.6748434237995826, + "grad_norm": 1.1470829248428345, + "learning_rate": 4.2117977294462235e-06, + "loss": 0.8001, + "step": 1069 + }, + { + "epoch": 1.6764091858037578, + "grad_norm": 0.9750358462333679, + "learning_rate": 4.210261704043327e-06, + "loss": 0.6736, + "step": 1070 + }, + { + "epoch": 1.6779749478079333, + "grad_norm": 1.061862826347351, + "learning_rate": 4.208724464104176e-06, + "loss": 0.7751, + "step": 1071 + }, + { + "epoch": 1.6795407098121085, + "grad_norm": 1.0311503410339355, + "learning_rate": 4.2071860107204346e-06, + "loss": 0.7629, + "step": 1072 + }, + { + "epoch": 1.681106471816284, + "grad_norm": 1.0023622512817383, + "learning_rate": 4.2056463449846305e-06, + "loss": 0.6935, + "step": 1073 + }, + { + "epoch": 1.6826722338204592, + "grad_norm": 1.1624165773391724, + "learning_rate": 4.204105467990153e-06, + "loss": 0.7451, + "step": 1074 + }, + { + "epoch": 1.6842379958246347, + "grad_norm": 0.9852432608604431, + "learning_rate": 4.202563380831249e-06, + "loss": 0.7446, + "step": 1075 + }, + { + "epoch": 1.6858037578288099, + "grad_norm": 1.0009260177612305, + "learning_rate": 4.201020084603027e-06, + "loss": 0.726, + "step": 1076 + }, + { + "epoch": 1.6873695198329854, + "grad_norm": 1.0374701023101807, + "learning_rate": 4.199475580401452e-06, + "loss": 0.6687, + "step": 1077 + }, + { + "epoch": 1.6889352818371608, + "grad_norm": 1.1437046527862549, + "learning_rate": 4.19792986932335e-06, + "loss": 0.7802, + "step": 1078 + }, + { + "epoch": 1.6905010438413361, + "grad_norm": 0.9866487383842468, + "learning_rate": 4.1963829524664e-06, + "loss": 0.7117, + "step": 1079 + }, + { + "epoch": 1.6920668058455115, + "grad_norm": 0.9720652103424072, + "learning_rate": 4.194834830929142e-06, + "loss": 0.778, + "step": 1080 + }, + { + "epoch": 1.6936325678496869, + "grad_norm": 1.009422779083252, + "learning_rate": 4.1932855058109675e-06, + "loss": 0.6783, + "step": 1081 + }, + { + "epoch": 1.6951983298538622, + "grad_norm": 1.024781346321106, + "learning_rate": 4.191734978212124e-06, + "loss": 0.6697, + "step": 1082 + }, + { + "epoch": 1.6967640918580376, + "grad_norm": 1.060354471206665, + "learning_rate": 4.190183249233715e-06, + "loss": 0.7151, + "step": 1083 + }, + { + "epoch": 1.698329853862213, + "grad_norm": 1.118486762046814, + "learning_rate": 4.1886303199776924e-06, + "loss": 0.7408, + "step": 1084 + }, + { + "epoch": 1.6998956158663883, + "grad_norm": 0.9334209561347961, + "learning_rate": 4.187076191546865e-06, + "loss": 0.7646, + "step": 1085 + }, + { + "epoch": 1.7014613778705638, + "grad_norm": 1.0506161451339722, + "learning_rate": 4.185520865044892e-06, + "loss": 0.7299, + "step": 1086 + }, + { + "epoch": 1.703027139874739, + "grad_norm": 1.1261264085769653, + "learning_rate": 4.1839643415762825e-06, + "loss": 0.6813, + "step": 1087 + }, + { + "epoch": 1.7045929018789145, + "grad_norm": 0.9690954685211182, + "learning_rate": 4.182406622246396e-06, + "loss": 0.7417, + "step": 1088 + }, + { + "epoch": 1.7061586638830897, + "grad_norm": 1.0872795581817627, + "learning_rate": 4.180847708161442e-06, + "loss": 0.676, + "step": 1089 + }, + { + "epoch": 1.7077244258872653, + "grad_norm": 1.054091453552246, + "learning_rate": 4.179287600428476e-06, + "loss": 0.785, + "step": 1090 + }, + { + "epoch": 1.7092901878914404, + "grad_norm": 1.109984278678894, + "learning_rate": 4.177726300155404e-06, + "loss": 0.8033, + "step": 1091 + }, + { + "epoch": 1.710855949895616, + "grad_norm": 1.0984348058700562, + "learning_rate": 4.176163808450978e-06, + "loss": 0.6964, + "step": 1092 + }, + { + "epoch": 1.712421711899791, + "grad_norm": 1.0788499116897583, + "learning_rate": 4.174600126424797e-06, + "loss": 0.7596, + "step": 1093 + }, + { + "epoch": 1.7139874739039667, + "grad_norm": 1.0846409797668457, + "learning_rate": 4.173035255187302e-06, + "loss": 0.7322, + "step": 1094 + }, + { + "epoch": 1.715553235908142, + "grad_norm": 1.1376627683639526, + "learning_rate": 4.171469195849782e-06, + "loss": 0.6934, + "step": 1095 + }, + { + "epoch": 1.7171189979123174, + "grad_norm": 1.003432035446167, + "learning_rate": 4.169901949524368e-06, + "loss": 0.7663, + "step": 1096 + }, + { + "epoch": 1.7186847599164927, + "grad_norm": 0.9824592471122742, + "learning_rate": 4.168333517324036e-06, + "loss": 0.8253, + "step": 1097 + }, + { + "epoch": 1.720250521920668, + "grad_norm": 1.0256636142730713, + "learning_rate": 4.1667639003626e-06, + "loss": 0.6614, + "step": 1098 + }, + { + "epoch": 1.7218162839248434, + "grad_norm": 1.0290439128875732, + "learning_rate": 4.16519309975472e-06, + "loss": 0.7403, + "step": 1099 + }, + { + "epoch": 1.7233820459290188, + "grad_norm": 1.0358202457427979, + "learning_rate": 4.163621116615892e-06, + "loss": 0.7332, + "step": 1100 + }, + { + "epoch": 1.7249478079331941, + "grad_norm": 1.1046626567840576, + "learning_rate": 4.162047952062456e-06, + "loss": 0.6672, + "step": 1101 + }, + { + "epoch": 1.7265135699373695, + "grad_norm": 1.116408109664917, + "learning_rate": 4.16047360721159e-06, + "loss": 0.6997, + "step": 1102 + }, + { + "epoch": 1.7280793319415448, + "grad_norm": 0.98244309425354, + "learning_rate": 4.158898083181308e-06, + "loss": 0.754, + "step": 1103 + }, + { + "epoch": 1.7296450939457202, + "grad_norm": 1.033761978149414, + "learning_rate": 4.1573213810904635e-06, + "loss": 0.7317, + "step": 1104 + }, + { + "epoch": 1.7312108559498958, + "grad_norm": 1.1415889263153076, + "learning_rate": 4.155743502058745e-06, + "loss": 0.8361, + "step": 1105 + }, + { + "epoch": 1.732776617954071, + "grad_norm": 0.9897739291191101, + "learning_rate": 4.154164447206679e-06, + "loss": 0.7361, + "step": 1106 + }, + { + "epoch": 1.7343423799582465, + "grad_norm": 1.0837332010269165, + "learning_rate": 4.1525842176556235e-06, + "loss": 0.7007, + "step": 1107 + }, + { + "epoch": 1.7359081419624216, + "grad_norm": 1.021332025527954, + "learning_rate": 4.151002814527774e-06, + "loss": 0.723, + "step": 1108 + }, + { + "epoch": 1.7374739039665972, + "grad_norm": 0.9808626770973206, + "learning_rate": 4.14942023894616e-06, + "loss": 0.6121, + "step": 1109 + }, + { + "epoch": 1.7390396659707723, + "grad_norm": 1.0266727209091187, + "learning_rate": 4.14783649203464e-06, + "loss": 0.8221, + "step": 1110 + }, + { + "epoch": 1.7406054279749479, + "grad_norm": 1.0880744457244873, + "learning_rate": 4.146251574917907e-06, + "loss": 0.7159, + "step": 1111 + }, + { + "epoch": 1.742171189979123, + "grad_norm": 1.0125631093978882, + "learning_rate": 4.144665488721483e-06, + "loss": 0.7228, + "step": 1112 + }, + { + "epoch": 1.7437369519832986, + "grad_norm": 0.9339308142662048, + "learning_rate": 4.143078234571723e-06, + "loss": 0.7472, + "step": 1113 + }, + { + "epoch": 1.745302713987474, + "grad_norm": 1.1234889030456543, + "learning_rate": 4.141489813595809e-06, + "loss": 0.7875, + "step": 1114 + }, + { + "epoch": 1.7468684759916493, + "grad_norm": 1.0862795114517212, + "learning_rate": 4.139900226921753e-06, + "loss": 0.7219, + "step": 1115 + }, + { + "epoch": 1.7484342379958246, + "grad_norm": 1.027912974357605, + "learning_rate": 4.138309475678393e-06, + "loss": 0.6616, + "step": 1116 + }, + { + "epoch": 1.75, + "grad_norm": 1.0271416902542114, + "learning_rate": 4.136717560995398e-06, + "loss": 0.7245, + "step": 1117 + }, + { + "epoch": 1.7515657620041754, + "grad_norm": 0.9430286288261414, + "learning_rate": 4.135124484003258e-06, + "loss": 0.7347, + "step": 1118 + }, + { + "epoch": 1.7531315240083507, + "grad_norm": 1.1191871166229248, + "learning_rate": 4.133530245833292e-06, + "loss": 0.6713, + "step": 1119 + }, + { + "epoch": 1.754697286012526, + "grad_norm": 1.0734976530075073, + "learning_rate": 4.131934847617642e-06, + "loss": 0.7268, + "step": 1120 + }, + { + "epoch": 1.7562630480167014, + "grad_norm": 1.0727618932724, + "learning_rate": 4.130338290489276e-06, + "loss": 0.7855, + "step": 1121 + }, + { + "epoch": 1.757828810020877, + "grad_norm": 1.0351287126541138, + "learning_rate": 4.1287405755819805e-06, + "loss": 0.7297, + "step": 1122 + }, + { + "epoch": 1.7593945720250521, + "grad_norm": 1.046382188796997, + "learning_rate": 4.127141704030371e-06, + "loss": 0.6508, + "step": 1123 + }, + { + "epoch": 1.7609603340292277, + "grad_norm": 1.0040491819381714, + "learning_rate": 4.125541676969876e-06, + "loss": 0.7707, + "step": 1124 + }, + { + "epoch": 1.7625260960334028, + "grad_norm": 1.1626756191253662, + "learning_rate": 4.123940495536752e-06, + "loss": 0.7366, + "step": 1125 + }, + { + "epoch": 1.7640918580375784, + "grad_norm": 0.9720292091369629, + "learning_rate": 4.122338160868071e-06, + "loss": 0.7641, + "step": 1126 + }, + { + "epoch": 1.7656576200417535, + "grad_norm": 0.9957835078239441, + "learning_rate": 4.1207346741017265e-06, + "loss": 0.7398, + "step": 1127 + }, + { + "epoch": 1.767223382045929, + "grad_norm": 0.972172737121582, + "learning_rate": 4.1191300363764276e-06, + "loss": 0.6565, + "step": 1128 + }, + { + "epoch": 1.7687891440501042, + "grad_norm": 1.0128676891326904, + "learning_rate": 4.117524248831702e-06, + "loss": 0.7433, + "step": 1129 + }, + { + "epoch": 1.7703549060542798, + "grad_norm": 1.1161550283432007, + "learning_rate": 4.115917312607895e-06, + "loss": 0.8241, + "step": 1130 + }, + { + "epoch": 1.7719206680584552, + "grad_norm": 1.009810209274292, + "learning_rate": 4.114309228846167e-06, + "loss": 0.6939, + "step": 1131 + }, + { + "epoch": 1.7734864300626305, + "grad_norm": 1.0082894563674927, + "learning_rate": 4.112699998688492e-06, + "loss": 0.6772, + "step": 1132 + }, + { + "epoch": 1.7750521920668059, + "grad_norm": 0.9453067183494568, + "learning_rate": 4.111089623277658e-06, + "loss": 0.6982, + "step": 1133 + }, + { + "epoch": 1.7766179540709812, + "grad_norm": 1.074747920036316, + "learning_rate": 4.1094781037572696e-06, + "loss": 0.7296, + "step": 1134 + }, + { + "epoch": 1.7781837160751566, + "grad_norm": 1.0477592945098877, + "learning_rate": 4.1078654412717404e-06, + "loss": 0.8162, + "step": 1135 + }, + { + "epoch": 1.779749478079332, + "grad_norm": 1.1362730264663696, + "learning_rate": 4.106251636966297e-06, + "loss": 0.6942, + "step": 1136 + }, + { + "epoch": 1.7813152400835073, + "grad_norm": 0.9564191102981567, + "learning_rate": 4.104636691986977e-06, + "loss": 0.7613, + "step": 1137 + }, + { + "epoch": 1.7828810020876826, + "grad_norm": 1.0384364128112793, + "learning_rate": 4.103020607480628e-06, + "loss": 0.6482, + "step": 1138 + }, + { + "epoch": 1.784446764091858, + "grad_norm": 1.1258740425109863, + "learning_rate": 4.101403384594905e-06, + "loss": 0.7736, + "step": 1139 + }, + { + "epoch": 1.7860125260960333, + "grad_norm": 1.0740283727645874, + "learning_rate": 4.099785024478276e-06, + "loss": 0.7219, + "step": 1140 + }, + { + "epoch": 1.787578288100209, + "grad_norm": 1.0377696752548218, + "learning_rate": 4.098165528280012e-06, + "loss": 0.7132, + "step": 1141 + }, + { + "epoch": 1.789144050104384, + "grad_norm": 1.0551276206970215, + "learning_rate": 4.096544897150192e-06, + "loss": 0.7879, + "step": 1142 + }, + { + "epoch": 1.7907098121085596, + "grad_norm": 1.096389651298523, + "learning_rate": 4.0949231322397036e-06, + "loss": 0.7516, + "step": 1143 + }, + { + "epoch": 1.7922755741127347, + "grad_norm": 0.9602596163749695, + "learning_rate": 4.093300234700234e-06, + "loss": 0.7197, + "step": 1144 + }, + { + "epoch": 1.7938413361169103, + "grad_norm": 1.1918690204620361, + "learning_rate": 4.09167620568428e-06, + "loss": 0.7634, + "step": 1145 + }, + { + "epoch": 1.7954070981210855, + "grad_norm": 1.1249059438705444, + "learning_rate": 4.09005104634514e-06, + "loss": 0.7718, + "step": 1146 + }, + { + "epoch": 1.796972860125261, + "grad_norm": 1.0596246719360352, + "learning_rate": 4.0884247578369164e-06, + "loss": 0.7258, + "step": 1147 + }, + { + "epoch": 1.7985386221294362, + "grad_norm": 1.026201844215393, + "learning_rate": 4.086797341314509e-06, + "loss": 0.6426, + "step": 1148 + }, + { + "epoch": 1.8001043841336117, + "grad_norm": 1.0791343450546265, + "learning_rate": 4.085168797933625e-06, + "loss": 0.7147, + "step": 1149 + }, + { + "epoch": 1.801670146137787, + "grad_norm": 0.9660344123840332, + "learning_rate": 4.083539128850766e-06, + "loss": 0.7335, + "step": 1150 + }, + { + "epoch": 1.8032359081419624, + "grad_norm": 1.0022811889648438, + "learning_rate": 4.081908335223238e-06, + "loss": 0.6824, + "step": 1151 + }, + { + "epoch": 1.8048016701461378, + "grad_norm": 0.9261506199836731, + "learning_rate": 4.080276418209143e-06, + "loss": 0.7538, + "step": 1152 + }, + { + "epoch": 1.8063674321503131, + "grad_norm": 1.088552713394165, + "learning_rate": 4.078643378967379e-06, + "loss": 0.6988, + "step": 1153 + }, + { + "epoch": 1.8079331941544885, + "grad_norm": 1.0581461191177368, + "learning_rate": 4.077009218657645e-06, + "loss": 0.7075, + "step": 1154 + }, + { + "epoch": 1.8094989561586639, + "grad_norm": 1.0722301006317139, + "learning_rate": 4.0753739384404335e-06, + "loss": 0.6974, + "step": 1155 + }, + { + "epoch": 1.8110647181628392, + "grad_norm": 0.9519115090370178, + "learning_rate": 4.073737539477033e-06, + "loss": 0.6898, + "step": 1156 + }, + { + "epoch": 1.8126304801670146, + "grad_norm": 1.1201752424240112, + "learning_rate": 4.072100022929526e-06, + "loss": 0.8572, + "step": 1157 + }, + { + "epoch": 1.8141962421711901, + "grad_norm": 1.24104642868042, + "learning_rate": 4.070461389960789e-06, + "loss": 0.7407, + "step": 1158 + }, + { + "epoch": 1.8157620041753653, + "grad_norm": 1.0625057220458984, + "learning_rate": 4.068821641734492e-06, + "loss": 0.655, + "step": 1159 + }, + { + "epoch": 1.8173277661795408, + "grad_norm": 1.1174049377441406, + "learning_rate": 4.067180779415097e-06, + "loss": 0.7214, + "step": 1160 + }, + { + "epoch": 1.818893528183716, + "grad_norm": 1.0683097839355469, + "learning_rate": 4.065538804167855e-06, + "loss": 0.7863, + "step": 1161 + }, + { + "epoch": 1.8204592901878915, + "grad_norm": 1.031198263168335, + "learning_rate": 4.063895717158809e-06, + "loss": 0.7663, + "step": 1162 + }, + { + "epoch": 1.8220250521920667, + "grad_norm": 1.1091554164886475, + "learning_rate": 4.062251519554791e-06, + "loss": 0.7234, + "step": 1163 + }, + { + "epoch": 1.8235908141962422, + "grad_norm": 1.0818198919296265, + "learning_rate": 4.060606212523425e-06, + "loss": 0.7709, + "step": 1164 + }, + { + "epoch": 1.8251565762004174, + "grad_norm": 1.0104292631149292, + "learning_rate": 4.058959797233116e-06, + "loss": 0.7225, + "step": 1165 + }, + { + "epoch": 1.826722338204593, + "grad_norm": 1.1726491451263428, + "learning_rate": 4.057312274853063e-06, + "loss": 0.6831, + "step": 1166 + }, + { + "epoch": 1.8282881002087683, + "grad_norm": 0.9735565185546875, + "learning_rate": 4.055663646553248e-06, + "loss": 0.7123, + "step": 1167 + }, + { + "epoch": 1.8298538622129437, + "grad_norm": 1.011709213256836, + "learning_rate": 4.054013913504436e-06, + "loss": 0.6885, + "step": 1168 + }, + { + "epoch": 1.831419624217119, + "grad_norm": 1.0374364852905273, + "learning_rate": 4.05236307687818e-06, + "loss": 0.782, + "step": 1169 + }, + { + "epoch": 1.8329853862212944, + "grad_norm": 1.011202096939087, + "learning_rate": 4.050711137846817e-06, + "loss": 0.7381, + "step": 1170 + }, + { + "epoch": 1.8345511482254697, + "grad_norm": 1.0984755754470825, + "learning_rate": 4.049058097583464e-06, + "loss": 0.7472, + "step": 1171 + }, + { + "epoch": 1.836116910229645, + "grad_norm": 1.090078353881836, + "learning_rate": 4.047403957262024e-06, + "loss": 0.6773, + "step": 1172 + }, + { + "epoch": 1.8376826722338204, + "grad_norm": 0.9866096377372742, + "learning_rate": 4.045748718057176e-06, + "loss": 0.6812, + "step": 1173 + }, + { + "epoch": 1.8392484342379958, + "grad_norm": 1.1006139516830444, + "learning_rate": 4.0440923811443846e-06, + "loss": 0.7953, + "step": 1174 + }, + { + "epoch": 1.8408141962421714, + "grad_norm": 0.9621697664260864, + "learning_rate": 4.04243494769989e-06, + "loss": 0.7557, + "step": 1175 + }, + { + "epoch": 1.8423799582463465, + "grad_norm": 1.1520715951919556, + "learning_rate": 4.040776418900714e-06, + "loss": 0.7422, + "step": 1176 + }, + { + "epoch": 1.843945720250522, + "grad_norm": 1.0679237842559814, + "learning_rate": 4.039116795924653e-06, + "loss": 0.6738, + "step": 1177 + }, + { + "epoch": 1.8455114822546972, + "grad_norm": 1.026510238647461, + "learning_rate": 4.037456079950284e-06, + "loss": 0.7741, + "step": 1178 + }, + { + "epoch": 1.8470772442588728, + "grad_norm": 1.0111522674560547, + "learning_rate": 4.0357942721569585e-06, + "loss": 0.6968, + "step": 1179 + }, + { + "epoch": 1.848643006263048, + "grad_norm": 1.0290493965148926, + "learning_rate": 4.034131373724802e-06, + "loss": 0.7131, + "step": 1180 + }, + { + "epoch": 1.8502087682672235, + "grad_norm": 0.9878649711608887, + "learning_rate": 4.032467385834718e-06, + "loss": 0.7477, + "step": 1181 + }, + { + "epoch": 1.8517745302713986, + "grad_norm": 1.0317823886871338, + "learning_rate": 4.030802309668379e-06, + "loss": 0.6318, + "step": 1182 + }, + { + "epoch": 1.8533402922755742, + "grad_norm": 1.0163817405700684, + "learning_rate": 4.029136146408235e-06, + "loss": 0.7027, + "step": 1183 + }, + { + "epoch": 1.8549060542797495, + "grad_norm": 0.9590600728988647, + "learning_rate": 4.027468897237505e-06, + "loss": 0.6988, + "step": 1184 + }, + { + "epoch": 1.8564718162839249, + "grad_norm": 1.0645931959152222, + "learning_rate": 4.0258005633401795e-06, + "loss": 0.7588, + "step": 1185 + }, + { + "epoch": 1.8580375782881002, + "grad_norm": 1.0380210876464844, + "learning_rate": 4.024131145901022e-06, + "loss": 0.6984, + "step": 1186 + }, + { + "epoch": 1.8596033402922756, + "grad_norm": 1.0558087825775146, + "learning_rate": 4.022460646105561e-06, + "loss": 0.6729, + "step": 1187 + }, + { + "epoch": 1.861169102296451, + "grad_norm": 1.0865014791488647, + "learning_rate": 4.020789065140097e-06, + "loss": 0.7208, + "step": 1188 + }, + { + "epoch": 1.8627348643006263, + "grad_norm": 1.0456374883651733, + "learning_rate": 4.019116404191697e-06, + "loss": 0.7759, + "step": 1189 + }, + { + "epoch": 1.8643006263048016, + "grad_norm": 0.9724026918411255, + "learning_rate": 4.017442664448197e-06, + "loss": 0.7475, + "step": 1190 + }, + { + "epoch": 1.865866388308977, + "grad_norm": 1.1170289516448975, + "learning_rate": 4.015767847098194e-06, + "loss": 0.6993, + "step": 1191 + }, + { + "epoch": 1.8674321503131524, + "grad_norm": 1.1052978038787842, + "learning_rate": 4.014091953331056e-06, + "loss": 0.748, + "step": 1192 + }, + { + "epoch": 1.8689979123173277, + "grad_norm": 1.1223454475402832, + "learning_rate": 4.012414984336912e-06, + "loss": 0.6921, + "step": 1193 + }, + { + "epoch": 1.8705636743215033, + "grad_norm": 1.0611138343811035, + "learning_rate": 4.010736941306655e-06, + "loss": 0.7859, + "step": 1194 + }, + { + "epoch": 1.8721294363256784, + "grad_norm": 1.169189691543579, + "learning_rate": 4.009057825431944e-06, + "loss": 0.5874, + "step": 1195 + }, + { + "epoch": 1.873695198329854, + "grad_norm": 1.036626935005188, + "learning_rate": 4.0073776379051945e-06, + "loss": 0.6801, + "step": 1196 + }, + { + "epoch": 1.8752609603340291, + "grad_norm": 0.9431620240211487, + "learning_rate": 4.005696379919586e-06, + "loss": 0.7476, + "step": 1197 + }, + { + "epoch": 1.8768267223382047, + "grad_norm": 1.0381059646606445, + "learning_rate": 4.0040140526690585e-06, + "loss": 0.6691, + "step": 1198 + }, + { + "epoch": 1.8783924843423798, + "grad_norm": 1.120033860206604, + "learning_rate": 4.0023306573483105e-06, + "loss": 0.6669, + "step": 1199 + }, + { + "epoch": 1.8799582463465554, + "grad_norm": 1.0298967361450195, + "learning_rate": 4.000646195152798e-06, + "loss": 0.7355, + "step": 1200 + }, + { + "epoch": 1.8815240083507305, + "grad_norm": 1.0036594867706299, + "learning_rate": 3.99896066727874e-06, + "loss": 0.7545, + "step": 1201 + }, + { + "epoch": 1.883089770354906, + "grad_norm": 1.1166342496871948, + "learning_rate": 3.997274074923103e-06, + "loss": 0.7468, + "step": 1202 + }, + { + "epoch": 1.8846555323590815, + "grad_norm": 1.0512094497680664, + "learning_rate": 3.995586419283617e-06, + "loss": 0.7647, + "step": 1203 + }, + { + "epoch": 1.8862212943632568, + "grad_norm": 1.0115973949432373, + "learning_rate": 3.993897701558764e-06, + "loss": 0.7893, + "step": 1204 + }, + { + "epoch": 1.8877870563674322, + "grad_norm": 1.0839310884475708, + "learning_rate": 3.992207922947783e-06, + "loss": 0.6824, + "step": 1205 + }, + { + "epoch": 1.8893528183716075, + "grad_norm": 0.9987578988075256, + "learning_rate": 3.99051708465066e-06, + "loss": 0.8045, + "step": 1206 + }, + { + "epoch": 1.8909185803757829, + "grad_norm": 1.0759059190750122, + "learning_rate": 3.9888251878681425e-06, + "loss": 0.7756, + "step": 1207 + }, + { + "epoch": 1.8924843423799582, + "grad_norm": 1.1055301427841187, + "learning_rate": 3.987132233801722e-06, + "loss": 0.6947, + "step": 1208 + }, + { + "epoch": 1.8940501043841336, + "grad_norm": 1.042575478553772, + "learning_rate": 3.985438223653644e-06, + "loss": 0.6669, + "step": 1209 + }, + { + "epoch": 1.895615866388309, + "grad_norm": 1.0595242977142334, + "learning_rate": 3.983743158626904e-06, + "loss": 0.6937, + "step": 1210 + }, + { + "epoch": 1.8971816283924845, + "grad_norm": 1.0498789548873901, + "learning_rate": 3.982047039925247e-06, + "loss": 0.7244, + "step": 1211 + }, + { + "epoch": 1.8987473903966596, + "grad_norm": 1.0960590839385986, + "learning_rate": 3.980349868753166e-06, + "loss": 0.7661, + "step": 1212 + }, + { + "epoch": 1.9003131524008352, + "grad_norm": 1.108823299407959, + "learning_rate": 3.9786516463159e-06, + "loss": 0.6937, + "step": 1213 + }, + { + "epoch": 1.9018789144050103, + "grad_norm": 1.051416039466858, + "learning_rate": 3.976952373819436e-06, + "loss": 0.7263, + "step": 1214 + }, + { + "epoch": 1.903444676409186, + "grad_norm": 0.9884061217308044, + "learning_rate": 3.9752520524705065e-06, + "loss": 0.6577, + "step": 1215 + }, + { + "epoch": 1.905010438413361, + "grad_norm": 1.0950369834899902, + "learning_rate": 3.9735506834765885e-06, + "loss": 0.7376, + "step": 1216 + }, + { + "epoch": 1.9065762004175366, + "grad_norm": 1.1140044927597046, + "learning_rate": 3.971848268045902e-06, + "loss": 0.7125, + "step": 1217 + }, + { + "epoch": 1.9081419624217117, + "grad_norm": 1.0908716917037964, + "learning_rate": 3.970144807387414e-06, + "loss": 0.7796, + "step": 1218 + }, + { + "epoch": 1.9097077244258873, + "grad_norm": 1.1068549156188965, + "learning_rate": 3.968440302710829e-06, + "loss": 0.7454, + "step": 1219 + }, + { + "epoch": 1.9112734864300627, + "grad_norm": 1.102594017982483, + "learning_rate": 3.9667347552265945e-06, + "loss": 0.757, + "step": 1220 + }, + { + "epoch": 1.912839248434238, + "grad_norm": 1.1588326692581177, + "learning_rate": 3.9650281661459015e-06, + "loss": 0.6491, + "step": 1221 + }, + { + "epoch": 1.9144050104384134, + "grad_norm": 0.9896461367607117, + "learning_rate": 3.963320536680675e-06, + "loss": 0.773, + "step": 1222 + }, + { + "epoch": 1.9159707724425887, + "grad_norm": 1.0855952501296997, + "learning_rate": 3.961611868043583e-06, + "loss": 0.6906, + "step": 1223 + }, + { + "epoch": 1.917536534446764, + "grad_norm": 1.0556972026824951, + "learning_rate": 3.959902161448032e-06, + "loss": 0.7577, + "step": 1224 + }, + { + "epoch": 1.9191022964509394, + "grad_norm": 1.0284528732299805, + "learning_rate": 3.958191418108162e-06, + "loss": 0.6844, + "step": 1225 + }, + { + "epoch": 1.9206680584551148, + "grad_norm": 0.9578842520713806, + "learning_rate": 3.956479639238852e-06, + "loss": 0.7216, + "step": 1226 + }, + { + "epoch": 1.9222338204592901, + "grad_norm": 1.0366833209991455, + "learning_rate": 3.954766826055717e-06, + "loss": 0.7365, + "step": 1227 + }, + { + "epoch": 1.9237995824634657, + "grad_norm": 1.0498946905136108, + "learning_rate": 3.953052979775103e-06, + "loss": 0.6788, + "step": 1228 + }, + { + "epoch": 1.9253653444676408, + "grad_norm": 1.010860562324524, + "learning_rate": 3.951338101614094e-06, + "loss": 0.7181, + "step": 1229 + }, + { + "epoch": 1.9269311064718164, + "grad_norm": 1.0203677415847778, + "learning_rate": 3.949622192790502e-06, + "loss": 0.7277, + "step": 1230 + }, + { + "epoch": 1.9284968684759916, + "grad_norm": 0.9907971024513245, + "learning_rate": 3.947905254522876e-06, + "loss": 0.6858, + "step": 1231 + }, + { + "epoch": 1.9300626304801671, + "grad_norm": 0.9824337363243103, + "learning_rate": 3.946187288030493e-06, + "loss": 0.7517, + "step": 1232 + }, + { + "epoch": 1.9316283924843423, + "grad_norm": 1.0206327438354492, + "learning_rate": 3.944468294533359e-06, + "loss": 0.7225, + "step": 1233 + }, + { + "epoch": 1.9331941544885178, + "grad_norm": 1.0529900789260864, + "learning_rate": 3.942748275252214e-06, + "loss": 0.7965, + "step": 1234 + }, + { + "epoch": 1.934759916492693, + "grad_norm": 1.0501983165740967, + "learning_rate": 3.941027231408521e-06, + "loss": 0.7646, + "step": 1235 + }, + { + "epoch": 1.9363256784968685, + "grad_norm": 0.9482662081718445, + "learning_rate": 3.939305164224474e-06, + "loss": 0.7449, + "step": 1236 + }, + { + "epoch": 1.937891440501044, + "grad_norm": 1.096723198890686, + "learning_rate": 3.937582074922993e-06, + "loss": 0.7464, + "step": 1237 + }, + { + "epoch": 1.9394572025052192, + "grad_norm": 1.0417059659957886, + "learning_rate": 3.935857964727725e-06, + "loss": 0.6586, + "step": 1238 + }, + { + "epoch": 1.9410229645093946, + "grad_norm": 1.1008849143981934, + "learning_rate": 3.934132834863039e-06, + "loss": 0.6236, + "step": 1239 + }, + { + "epoch": 1.94258872651357, + "grad_norm": 0.9107227921485901, + "learning_rate": 3.93240668655403e-06, + "loss": 0.7069, + "step": 1240 + }, + { + "epoch": 1.9441544885177453, + "grad_norm": 1.0516945123672485, + "learning_rate": 3.930679521026517e-06, + "loss": 0.773, + "step": 1241 + }, + { + "epoch": 1.9457202505219207, + "grad_norm": 0.9941951632499695, + "learning_rate": 3.92895133950704e-06, + "loss": 0.6002, + "step": 1242 + }, + { + "epoch": 1.947286012526096, + "grad_norm": 1.031784176826477, + "learning_rate": 3.927222143222859e-06, + "loss": 0.7737, + "step": 1243 + }, + { + "epoch": 1.9488517745302714, + "grad_norm": 0.986473798751831, + "learning_rate": 3.925491933401961e-06, + "loss": 0.7455, + "step": 1244 + }, + { + "epoch": 1.9504175365344467, + "grad_norm": 1.109964370727539, + "learning_rate": 3.923760711273044e-06, + "loss": 0.7048, + "step": 1245 + }, + { + "epoch": 1.951983298538622, + "grad_norm": 1.1303372383117676, + "learning_rate": 3.922028478065532e-06, + "loss": 0.749, + "step": 1246 + }, + { + "epoch": 1.9535490605427976, + "grad_norm": 0.999824583530426, + "learning_rate": 3.920295235009562e-06, + "loss": 0.7596, + "step": 1247 + }, + { + "epoch": 1.9551148225469728, + "grad_norm": 1.0440820455551147, + "learning_rate": 3.918560983335993e-06, + "loss": 0.7841, + "step": 1248 + }, + { + "epoch": 1.9566805845511483, + "grad_norm": 1.0420323610305786, + "learning_rate": 3.916825724276395e-06, + "loss": 0.7837, + "step": 1249 + }, + { + "epoch": 1.9582463465553235, + "grad_norm": 1.001997947692871, + "learning_rate": 3.915089459063058e-06, + "loss": 0.6956, + "step": 1250 + }, + { + "epoch": 1.959812108559499, + "grad_norm": 1.030626893043518, + "learning_rate": 3.913352188928984e-06, + "loss": 0.74, + "step": 1251 + }, + { + "epoch": 1.9613778705636742, + "grad_norm": 1.0981760025024414, + "learning_rate": 3.911613915107888e-06, + "loss": 0.713, + "step": 1252 + }, + { + "epoch": 1.9629436325678498, + "grad_norm": 1.020772933959961, + "learning_rate": 3.9098746388342006e-06, + "loss": 0.7605, + "step": 1253 + }, + { + "epoch": 1.964509394572025, + "grad_norm": 1.0206503868103027, + "learning_rate": 3.908134361343062e-06, + "loss": 0.7176, + "step": 1254 + }, + { + "epoch": 1.9660751565762005, + "grad_norm": 1.0471689701080322, + "learning_rate": 3.906393083870325e-06, + "loss": 0.7698, + "step": 1255 + }, + { + "epoch": 1.9676409185803758, + "grad_norm": 1.0658202171325684, + "learning_rate": 3.904650807652549e-06, + "loss": 0.7179, + "step": 1256 + }, + { + "epoch": 1.9692066805845512, + "grad_norm": 1.196192741394043, + "learning_rate": 3.902907533927007e-06, + "loss": 0.7801, + "step": 1257 + }, + { + "epoch": 1.9707724425887265, + "grad_norm": 1.0874967575073242, + "learning_rate": 3.90116326393168e-06, + "loss": 0.7672, + "step": 1258 + }, + { + "epoch": 1.9723382045929019, + "grad_norm": 1.153617024421692, + "learning_rate": 3.899417998905253e-06, + "loss": 0.752, + "step": 1259 + }, + { + "epoch": 1.9739039665970772, + "grad_norm": 1.0750453472137451, + "learning_rate": 3.89767174008712e-06, + "loss": 0.7471, + "step": 1260 + }, + { + "epoch": 1.9754697286012526, + "grad_norm": 1.0693799257278442, + "learning_rate": 3.895924488717381e-06, + "loss": 0.6704, + "step": 1261 + }, + { + "epoch": 1.977035490605428, + "grad_norm": 0.9552210569381714, + "learning_rate": 3.89417624603684e-06, + "loss": 0.7793, + "step": 1262 + }, + { + "epoch": 1.9786012526096033, + "grad_norm": 0.96712726354599, + "learning_rate": 3.892427013287007e-06, + "loss": 0.6843, + "step": 1263 + }, + { + "epoch": 1.9801670146137789, + "grad_norm": 1.1453518867492676, + "learning_rate": 3.890676791710091e-06, + "loss": 0.7532, + "step": 1264 + }, + { + "epoch": 1.981732776617954, + "grad_norm": 1.0294159650802612, + "learning_rate": 3.888925582549006e-06, + "loss": 0.703, + "step": 1265 + }, + { + "epoch": 1.9832985386221296, + "grad_norm": 1.0790715217590332, + "learning_rate": 3.8871733870473686e-06, + "loss": 0.6731, + "step": 1266 + }, + { + "epoch": 1.9848643006263047, + "grad_norm": 1.052460789680481, + "learning_rate": 3.8854202064494915e-06, + "loss": 0.748, + "step": 1267 + }, + { + "epoch": 1.9864300626304803, + "grad_norm": 1.112462043762207, + "learning_rate": 3.883666042000392e-06, + "loss": 0.7408, + "step": 1268 + }, + { + "epoch": 1.9879958246346554, + "grad_norm": 1.0738476514816284, + "learning_rate": 3.881910894945783e-06, + "loss": 0.759, + "step": 1269 + }, + { + "epoch": 1.989561586638831, + "grad_norm": 1.136472463607788, + "learning_rate": 3.8801547665320775e-06, + "loss": 0.754, + "step": 1270 + }, + { + "epoch": 1.9911273486430061, + "grad_norm": 1.0640734434127808, + "learning_rate": 3.87839765800638e-06, + "loss": 0.719, + "step": 1271 + }, + { + "epoch": 1.9926931106471817, + "grad_norm": 0.9155470132827759, + "learning_rate": 3.876639570616498e-06, + "loss": 0.762, + "step": 1272 + }, + { + "epoch": 1.994258872651357, + "grad_norm": 1.015743374824524, + "learning_rate": 3.87488050561093e-06, + "loss": 0.6926, + "step": 1273 + }, + { + "epoch": 1.9958246346555324, + "grad_norm": 1.0829862356185913, + "learning_rate": 3.87312046423887e-06, + "loss": 0.7644, + "step": 1274 + }, + { + "epoch": 1.9973903966597077, + "grad_norm": 1.0962239503860474, + "learning_rate": 3.871359447750205e-06, + "loss": 0.7038, + "step": 1275 + }, + { + "epoch": 1.998956158663883, + "grad_norm": 1.0470675230026245, + "learning_rate": 3.869597457395514e-06, + "loss": 0.7629, + "step": 1276 + }, + { + "epoch": 2.0015657620041756, + "grad_norm": 1.962398648262024, + "learning_rate": 3.867834494426069e-06, + "loss": 1.4251, + "step": 1277 + }, + { + "epoch": 2.0031315240083507, + "grad_norm": 1.0987200736999512, + "learning_rate": 3.866070560093832e-06, + "loss": 0.6873, + "step": 1278 + }, + { + "epoch": 2.0046972860125263, + "grad_norm": 1.0096511840820312, + "learning_rate": 3.864305655651453e-06, + "loss": 0.7113, + "step": 1279 + }, + { + "epoch": 2.0062630480167014, + "grad_norm": 0.9988889098167419, + "learning_rate": 3.8625397823522755e-06, + "loss": 0.6555, + "step": 1280 + }, + { + "epoch": 2.007828810020877, + "grad_norm": 0.9960927963256836, + "learning_rate": 3.860772941450326e-06, + "loss": 0.6458, + "step": 1281 + }, + { + "epoch": 2.009394572025052, + "grad_norm": 0.971942126750946, + "learning_rate": 3.85900513420032e-06, + "loss": 0.6821, + "step": 1282 + }, + { + "epoch": 2.0109603340292277, + "grad_norm": 1.0935462713241577, + "learning_rate": 3.857236361857662e-06, + "loss": 0.5862, + "step": 1283 + }, + { + "epoch": 2.012526096033403, + "grad_norm": 1.0295177698135376, + "learning_rate": 3.855466625678435e-06, + "loss": 0.723, + "step": 1284 + }, + { + "epoch": 2.0140918580375784, + "grad_norm": 0.9750491976737976, + "learning_rate": 3.853695926919415e-06, + "loss": 0.7105, + "step": 1285 + }, + { + "epoch": 2.0156576200417535, + "grad_norm": 1.1649892330169678, + "learning_rate": 3.851924266838056e-06, + "loss": 0.7106, + "step": 1286 + }, + { + "epoch": 2.017223382045929, + "grad_norm": 1.068339467048645, + "learning_rate": 3.850151646692496e-06, + "loss": 0.668, + "step": 1287 + }, + { + "epoch": 2.0187891440501042, + "grad_norm": 0.9659110903739929, + "learning_rate": 3.848378067741555e-06, + "loss": 0.6201, + "step": 1288 + }, + { + "epoch": 2.02035490605428, + "grad_norm": 1.0711618661880493, + "learning_rate": 3.846603531244734e-06, + "loss": 0.6379, + "step": 1289 + }, + { + "epoch": 2.021920668058455, + "grad_norm": 1.092109203338623, + "learning_rate": 3.844828038462213e-06, + "loss": 0.7105, + "step": 1290 + }, + { + "epoch": 2.0234864300626305, + "grad_norm": 1.1485183238983154, + "learning_rate": 3.843051590654852e-06, + "loss": 0.724, + "step": 1291 + }, + { + "epoch": 2.0250521920668056, + "grad_norm": 1.1114039421081543, + "learning_rate": 3.841274189084189e-06, + "loss": 0.6928, + "step": 1292 + }, + { + "epoch": 2.026617954070981, + "grad_norm": 1.1090906858444214, + "learning_rate": 3.839495835012441e-06, + "loss": 0.6988, + "step": 1293 + }, + { + "epoch": 2.028183716075157, + "grad_norm": 1.0808806419372559, + "learning_rate": 3.837716529702498e-06, + "loss": 0.6374, + "step": 1294 + }, + { + "epoch": 2.029749478079332, + "grad_norm": 1.0822157859802246, + "learning_rate": 3.835936274417927e-06, + "loss": 0.6249, + "step": 1295 + }, + { + "epoch": 2.0313152400835075, + "grad_norm": 1.0366182327270508, + "learning_rate": 3.834155070422972e-06, + "loss": 0.702, + "step": 1296 + }, + { + "epoch": 2.0328810020876826, + "grad_norm": 1.0268456935882568, + "learning_rate": 3.832372918982547e-06, + "loss": 0.6809, + "step": 1297 + }, + { + "epoch": 2.034446764091858, + "grad_norm": 1.1484824419021606, + "learning_rate": 3.830589821362242e-06, + "loss": 0.6442, + "step": 1298 + }, + { + "epoch": 2.0360125260960333, + "grad_norm": 1.0607216358184814, + "learning_rate": 3.828805778828315e-06, + "loss": 0.6469, + "step": 1299 + }, + { + "epoch": 2.037578288100209, + "grad_norm": 1.0950021743774414, + "learning_rate": 3.8270207926477e-06, + "loss": 0.6756, + "step": 1300 + }, + { + "epoch": 2.039144050104384, + "grad_norm": 0.9932938814163208, + "learning_rate": 3.825234864087998e-06, + "loss": 0.706, + "step": 1301 + }, + { + "epoch": 2.0407098121085596, + "grad_norm": 1.0706391334533691, + "learning_rate": 3.823447994417477e-06, + "loss": 0.6799, + "step": 1302 + }, + { + "epoch": 2.0422755741127347, + "grad_norm": 1.0093334913253784, + "learning_rate": 3.821660184905081e-06, + "loss": 0.676, + "step": 1303 + }, + { + "epoch": 2.0438413361169103, + "grad_norm": 1.0283772945404053, + "learning_rate": 3.819871436820413e-06, + "loss": 0.6738, + "step": 1304 + }, + { + "epoch": 2.0454070981210855, + "grad_norm": 1.0533452033996582, + "learning_rate": 3.818081751433746e-06, + "loss": 0.7053, + "step": 1305 + }, + { + "epoch": 2.046972860125261, + "grad_norm": 1.1219292879104614, + "learning_rate": 3.816291130016021e-06, + "loss": 0.6563, + "step": 1306 + }, + { + "epoch": 2.048538622129436, + "grad_norm": 1.025702953338623, + "learning_rate": 3.8144995738388397e-06, + "loss": 0.7264, + "step": 1307 + }, + { + "epoch": 2.0501043841336117, + "grad_norm": 1.0799325704574585, + "learning_rate": 3.8127070841744695e-06, + "loss": 0.6541, + "step": 1308 + }, + { + "epoch": 2.051670146137787, + "grad_norm": 1.043544888496399, + "learning_rate": 3.810913662295841e-06, + "loss": 0.627, + "step": 1309 + }, + { + "epoch": 2.0532359081419624, + "grad_norm": 0.9250067472457886, + "learning_rate": 3.809119309476547e-06, + "loss": 0.7447, + "step": 1310 + }, + { + "epoch": 2.0548016701461376, + "grad_norm": 1.0885530710220337, + "learning_rate": 3.8073240269908385e-06, + "loss": 0.6509, + "step": 1311 + }, + { + "epoch": 2.056367432150313, + "grad_norm": 1.2108066082000732, + "learning_rate": 3.805527816113631e-06, + "loss": 0.6844, + "step": 1312 + }, + { + "epoch": 2.0579331941544887, + "grad_norm": 1.1016451120376587, + "learning_rate": 3.8037306781204974e-06, + "loss": 0.6942, + "step": 1313 + }, + { + "epoch": 2.059498956158664, + "grad_norm": 1.080930233001709, + "learning_rate": 3.801932614287668e-06, + "loss": 0.7379, + "step": 1314 + }, + { + "epoch": 2.0610647181628394, + "grad_norm": 1.1148338317871094, + "learning_rate": 3.800133625892031e-06, + "loss": 0.6817, + "step": 1315 + }, + { + "epoch": 2.0626304801670146, + "grad_norm": 1.0792765617370605, + "learning_rate": 3.798333714211132e-06, + "loss": 0.6787, + "step": 1316 + }, + { + "epoch": 2.06419624217119, + "grad_norm": 1.219314455986023, + "learning_rate": 3.7965328805231727e-06, + "loss": 0.6451, + "step": 1317 + }, + { + "epoch": 2.0657620041753653, + "grad_norm": 1.0637946128845215, + "learning_rate": 3.7947311261070073e-06, + "loss": 0.6869, + "step": 1318 + }, + { + "epoch": 2.067327766179541, + "grad_norm": 1.112642765045166, + "learning_rate": 3.792928452242146e-06, + "loss": 0.6785, + "step": 1319 + }, + { + "epoch": 2.068893528183716, + "grad_norm": 1.006885051727295, + "learning_rate": 3.791124860208752e-06, + "loss": 0.7289, + "step": 1320 + }, + { + "epoch": 2.0704592901878915, + "grad_norm": 1.1578105688095093, + "learning_rate": 3.7893203512876385e-06, + "loss": 0.6472, + "step": 1321 + }, + { + "epoch": 2.0720250521920667, + "grad_norm": 1.0113657712936401, + "learning_rate": 3.7875149267602725e-06, + "loss": 0.6314, + "step": 1322 + }, + { + "epoch": 2.0735908141962422, + "grad_norm": 1.0363519191741943, + "learning_rate": 3.785708587908767e-06, + "loss": 0.6949, + "step": 1323 + }, + { + "epoch": 2.0751565762004174, + "grad_norm": 1.0536452531814575, + "learning_rate": 3.7839013360158904e-06, + "loss": 0.6817, + "step": 1324 + }, + { + "epoch": 2.076722338204593, + "grad_norm": 1.2700945138931274, + "learning_rate": 3.7820931723650547e-06, + "loss": 0.7364, + "step": 1325 + }, + { + "epoch": 2.078288100208768, + "grad_norm": 0.979896068572998, + "learning_rate": 3.7802840982403204e-06, + "loss": 0.6583, + "step": 1326 + }, + { + "epoch": 2.0798538622129437, + "grad_norm": 1.1849523782730103, + "learning_rate": 3.778474114926396e-06, + "loss": 0.6623, + "step": 1327 + }, + { + "epoch": 2.081419624217119, + "grad_norm": 1.0124396085739136, + "learning_rate": 3.7766632237086336e-06, + "loss": 0.7018, + "step": 1328 + }, + { + "epoch": 2.0829853862212944, + "grad_norm": 1.0735864639282227, + "learning_rate": 3.774851425873032e-06, + "loss": 0.6178, + "step": 1329 + }, + { + "epoch": 2.08455114822547, + "grad_norm": 1.1188064813613892, + "learning_rate": 3.773038722706234e-06, + "loss": 0.6603, + "step": 1330 + }, + { + "epoch": 2.086116910229645, + "grad_norm": 1.1021878719329834, + "learning_rate": 3.771225115495522e-06, + "loss": 0.7027, + "step": 1331 + }, + { + "epoch": 2.0876826722338206, + "grad_norm": 1.067917823791504, + "learning_rate": 3.769410605528824e-06, + "loss": 0.5733, + "step": 1332 + }, + { + "epoch": 2.0892484342379958, + "grad_norm": 1.199495553970337, + "learning_rate": 3.7675951940947075e-06, + "loss": 0.7376, + "step": 1333 + }, + { + "epoch": 2.0908141962421714, + "grad_norm": 1.104844093322754, + "learning_rate": 3.7657788824823805e-06, + "loss": 0.7099, + "step": 1334 + }, + { + "epoch": 2.0923799582463465, + "grad_norm": 1.1797598600387573, + "learning_rate": 3.7639616719816907e-06, + "loss": 0.6264, + "step": 1335 + }, + { + "epoch": 2.093945720250522, + "grad_norm": 1.0947678089141846, + "learning_rate": 3.7621435638831226e-06, + "loss": 0.7066, + "step": 1336 + }, + { + "epoch": 2.095511482254697, + "grad_norm": 1.1054636240005493, + "learning_rate": 3.7603245594778e-06, + "loss": 0.7439, + "step": 1337 + }, + { + "epoch": 2.0970772442588728, + "grad_norm": 1.106891393661499, + "learning_rate": 3.758504660057482e-06, + "loss": 0.6446, + "step": 1338 + }, + { + "epoch": 2.098643006263048, + "grad_norm": 1.043440818786621, + "learning_rate": 3.7566838669145644e-06, + "loss": 0.6885, + "step": 1339 + }, + { + "epoch": 2.1002087682672235, + "grad_norm": 1.0602154731750488, + "learning_rate": 3.7548621813420765e-06, + "loss": 0.6808, + "step": 1340 + }, + { + "epoch": 2.1017745302713986, + "grad_norm": 0.9517614841461182, + "learning_rate": 3.7530396046336814e-06, + "loss": 0.6369, + "step": 1341 + }, + { + "epoch": 2.103340292275574, + "grad_norm": 1.1436609029769897, + "learning_rate": 3.7512161380836763e-06, + "loss": 0.7427, + "step": 1342 + }, + { + "epoch": 2.1049060542797493, + "grad_norm": 1.00765860080719, + "learning_rate": 3.7493917829869887e-06, + "loss": 0.7012, + "step": 1343 + }, + { + "epoch": 2.106471816283925, + "grad_norm": 1.1718883514404297, + "learning_rate": 3.747566540639178e-06, + "loss": 0.6689, + "step": 1344 + }, + { + "epoch": 2.1080375782881, + "grad_norm": 1.0489915609359741, + "learning_rate": 3.745740412336434e-06, + "loss": 0.6735, + "step": 1345 + }, + { + "epoch": 2.1096033402922756, + "grad_norm": 1.1287360191345215, + "learning_rate": 3.743913399375575e-06, + "loss": 0.6211, + "step": 1346 + }, + { + "epoch": 2.1111691022964507, + "grad_norm": 1.0830544233322144, + "learning_rate": 3.7420855030540476e-06, + "loss": 0.7219, + "step": 1347 + }, + { + "epoch": 2.1127348643006263, + "grad_norm": 1.1535146236419678, + "learning_rate": 3.7402567246699257e-06, + "loss": 0.6995, + "step": 1348 + }, + { + "epoch": 2.114300626304802, + "grad_norm": 1.1392090320587158, + "learning_rate": 3.7384270655219102e-06, + "loss": 0.7118, + "step": 1349 + }, + { + "epoch": 2.115866388308977, + "grad_norm": 1.009692668914795, + "learning_rate": 3.736596526909327e-06, + "loss": 0.7211, + "step": 1350 + }, + { + "epoch": 2.1174321503131526, + "grad_norm": 1.2041988372802734, + "learning_rate": 3.7347651101321258e-06, + "loss": 0.5969, + "step": 1351 + }, + { + "epoch": 2.1189979123173277, + "grad_norm": 1.0518646240234375, + "learning_rate": 3.7329328164908818e-06, + "loss": 0.6624, + "step": 1352 + }, + { + "epoch": 2.1205636743215033, + "grad_norm": 1.1757079362869263, + "learning_rate": 3.7310996472867916e-06, + "loss": 0.7201, + "step": 1353 + }, + { + "epoch": 2.1221294363256784, + "grad_norm": 1.0831660032272339, + "learning_rate": 3.7292656038216722e-06, + "loss": 0.6302, + "step": 1354 + }, + { + "epoch": 2.123695198329854, + "grad_norm": 1.1849291324615479, + "learning_rate": 3.7274306873979654e-06, + "loss": 0.7365, + "step": 1355 + }, + { + "epoch": 2.125260960334029, + "grad_norm": 1.1130744218826294, + "learning_rate": 3.72559489931873e-06, + "loss": 0.6759, + "step": 1356 + }, + { + "epoch": 2.1268267223382047, + "grad_norm": 1.0014575719833374, + "learning_rate": 3.7237582408876426e-06, + "loss": 0.7179, + "step": 1357 + }, + { + "epoch": 2.12839248434238, + "grad_norm": 1.1398662328720093, + "learning_rate": 3.7219207134090013e-06, + "loss": 0.6678, + "step": 1358 + }, + { + "epoch": 2.1299582463465554, + "grad_norm": 1.1094143390655518, + "learning_rate": 3.720082318187719e-06, + "loss": 0.6962, + "step": 1359 + }, + { + "epoch": 2.1315240083507305, + "grad_norm": 1.0060160160064697, + "learning_rate": 3.718243056529326e-06, + "loss": 0.688, + "step": 1360 + }, + { + "epoch": 2.133089770354906, + "grad_norm": 1.0628772974014282, + "learning_rate": 3.716402929739967e-06, + "loss": 0.7124, + "step": 1361 + }, + { + "epoch": 2.1346555323590812, + "grad_norm": 1.1098920106887817, + "learning_rate": 3.714561939126402e-06, + "loss": 0.7016, + "step": 1362 + }, + { + "epoch": 2.136221294363257, + "grad_norm": 0.9880667328834534, + "learning_rate": 3.712720085996003e-06, + "loss": 0.7333, + "step": 1363 + }, + { + "epoch": 2.137787056367432, + "grad_norm": 1.1795120239257812, + "learning_rate": 3.710877371656757e-06, + "loss": 0.6729, + "step": 1364 + }, + { + "epoch": 2.1393528183716075, + "grad_norm": 1.1155129671096802, + "learning_rate": 3.7090337974172595e-06, + "loss": 0.6676, + "step": 1365 + }, + { + "epoch": 2.140918580375783, + "grad_norm": 1.1053357124328613, + "learning_rate": 3.707189364586719e-06, + "loss": 0.7043, + "step": 1366 + }, + { + "epoch": 2.142484342379958, + "grad_norm": 1.1484512090682983, + "learning_rate": 3.705344074474952e-06, + "loss": 0.7166, + "step": 1367 + }, + { + "epoch": 2.144050104384134, + "grad_norm": 1.056976318359375, + "learning_rate": 3.703497928392386e-06, + "loss": 0.7034, + "step": 1368 + }, + { + "epoch": 2.145615866388309, + "grad_norm": 1.093001365661621, + "learning_rate": 3.7016509276500546e-06, + "loss": 0.5952, + "step": 1369 + }, + { + "epoch": 2.1471816283924845, + "grad_norm": 1.094588041305542, + "learning_rate": 3.699803073559598e-06, + "loss": 0.7126, + "step": 1370 + }, + { + "epoch": 2.1487473903966596, + "grad_norm": 1.126368522644043, + "learning_rate": 3.6979543674332634e-06, + "loss": 0.714, + "step": 1371 + }, + { + "epoch": 2.150313152400835, + "grad_norm": 1.0853040218353271, + "learning_rate": 3.696104810583904e-06, + "loss": 0.721, + "step": 1372 + }, + { + "epoch": 2.1518789144050103, + "grad_norm": 1.091913104057312, + "learning_rate": 3.694254404324975e-06, + "loss": 0.6752, + "step": 1373 + }, + { + "epoch": 2.153444676409186, + "grad_norm": 1.1095316410064697, + "learning_rate": 3.6924031499705366e-06, + "loss": 0.6988, + "step": 1374 + }, + { + "epoch": 2.155010438413361, + "grad_norm": 1.1327425241470337, + "learning_rate": 3.690551048835249e-06, + "loss": 0.6563, + "step": 1375 + }, + { + "epoch": 2.1565762004175366, + "grad_norm": 1.1065635681152344, + "learning_rate": 3.688698102234377e-06, + "loss": 0.683, + "step": 1376 + }, + { + "epoch": 2.1581419624217117, + "grad_norm": 1.050032138824463, + "learning_rate": 3.6868443114837828e-06, + "loss": 0.6471, + "step": 1377 + }, + { + "epoch": 2.1597077244258873, + "grad_norm": 1.2008466720581055, + "learning_rate": 3.6849896778999293e-06, + "loss": 0.7257, + "step": 1378 + }, + { + "epoch": 2.1612734864300625, + "grad_norm": 1.101426601409912, + "learning_rate": 3.6831342027998796e-06, + "loss": 0.6542, + "step": 1379 + }, + { + "epoch": 2.162839248434238, + "grad_norm": 1.0931123495101929, + "learning_rate": 3.68127788750129e-06, + "loss": 0.665, + "step": 1380 + }, + { + "epoch": 2.164405010438413, + "grad_norm": 1.2418708801269531, + "learning_rate": 3.6794207333224184e-06, + "loss": 0.6399, + "step": 1381 + }, + { + "epoch": 2.1659707724425887, + "grad_norm": 1.1205544471740723, + "learning_rate": 3.6775627415821158e-06, + "loss": 0.7029, + "step": 1382 + }, + { + "epoch": 2.1675365344467643, + "grad_norm": 1.0413703918457031, + "learning_rate": 3.6757039135998276e-06, + "loss": 0.7521, + "step": 1383 + }, + { + "epoch": 2.1691022964509394, + "grad_norm": 1.0559121370315552, + "learning_rate": 3.673844250695594e-06, + "loss": 0.6664, + "step": 1384 + }, + { + "epoch": 2.170668058455115, + "grad_norm": 1.1533021926879883, + "learning_rate": 3.6719837541900495e-06, + "loss": 0.6425, + "step": 1385 + }, + { + "epoch": 2.17223382045929, + "grad_norm": 1.246163010597229, + "learning_rate": 3.6701224254044173e-06, + "loss": 0.6683, + "step": 1386 + }, + { + "epoch": 2.1737995824634657, + "grad_norm": 1.1539034843444824, + "learning_rate": 3.6682602656605136e-06, + "loss": 0.6699, + "step": 1387 + }, + { + "epoch": 2.175365344467641, + "grad_norm": 1.021262526512146, + "learning_rate": 3.6663972762807453e-06, + "loss": 0.6409, + "step": 1388 + }, + { + "epoch": 2.1769311064718164, + "grad_norm": 1.09273099899292, + "learning_rate": 3.664533458588107e-06, + "loss": 0.7315, + "step": 1389 + }, + { + "epoch": 2.1784968684759916, + "grad_norm": 1.169747233390808, + "learning_rate": 3.662668813906183e-06, + "loss": 0.6675, + "step": 1390 + }, + { + "epoch": 2.180062630480167, + "grad_norm": 1.189447045326233, + "learning_rate": 3.6608033435591435e-06, + "loss": 0.7016, + "step": 1391 + }, + { + "epoch": 2.1816283924843423, + "grad_norm": 1.0325034856796265, + "learning_rate": 3.658937048871746e-06, + "loss": 0.74, + "step": 1392 + }, + { + "epoch": 2.183194154488518, + "grad_norm": 1.1599889993667603, + "learning_rate": 3.6570699311693316e-06, + "loss": 0.6525, + "step": 1393 + }, + { + "epoch": 2.184759916492693, + "grad_norm": 1.018955945968628, + "learning_rate": 3.6552019917778294e-06, + "loss": 0.6779, + "step": 1394 + }, + { + "epoch": 2.1863256784968685, + "grad_norm": 1.1707937717437744, + "learning_rate": 3.653333232023748e-06, + "loss": 0.6321, + "step": 1395 + }, + { + "epoch": 2.1878914405010437, + "grad_norm": 1.1663832664489746, + "learning_rate": 3.6514636532341825e-06, + "loss": 0.645, + "step": 1396 + }, + { + "epoch": 2.1894572025052192, + "grad_norm": 1.0475002527236938, + "learning_rate": 3.6495932567368054e-06, + "loss": 0.6209, + "step": 1397 + }, + { + "epoch": 2.1910229645093944, + "grad_norm": 1.052640438079834, + "learning_rate": 3.647722043859874e-06, + "loss": 0.7383, + "step": 1398 + }, + { + "epoch": 2.19258872651357, + "grad_norm": 1.1010091304779053, + "learning_rate": 3.6458500159322226e-06, + "loss": 0.669, + "step": 1399 + }, + { + "epoch": 2.1941544885177455, + "grad_norm": 1.0842148065567017, + "learning_rate": 3.6439771742832645e-06, + "loss": 0.7248, + "step": 1400 + }, + { + "epoch": 2.1957202505219207, + "grad_norm": 1.1516163349151611, + "learning_rate": 3.642103520242993e-06, + "loss": 0.6501, + "step": 1401 + }, + { + "epoch": 2.1972860125260962, + "grad_norm": 1.1424976587295532, + "learning_rate": 3.6402290551419754e-06, + "loss": 0.7191, + "step": 1402 + }, + { + "epoch": 2.1988517745302714, + "grad_norm": 1.132794737815857, + "learning_rate": 3.6383537803113577e-06, + "loss": 0.676, + "step": 1403 + }, + { + "epoch": 2.200417536534447, + "grad_norm": 1.0671226978302002, + "learning_rate": 3.6364776970828586e-06, + "loss": 0.6878, + "step": 1404 + }, + { + "epoch": 2.201983298538622, + "grad_norm": 1.0326734781265259, + "learning_rate": 3.6346008067887724e-06, + "loss": 0.6559, + "step": 1405 + }, + { + "epoch": 2.2035490605427976, + "grad_norm": 1.0672800540924072, + "learning_rate": 3.6327231107619664e-06, + "loss": 0.6826, + "step": 1406 + }, + { + "epoch": 2.2051148225469728, + "grad_norm": 1.0678563117980957, + "learning_rate": 3.6308446103358785e-06, + "loss": 0.6772, + "step": 1407 + }, + { + "epoch": 2.2066805845511483, + "grad_norm": 1.1288729906082153, + "learning_rate": 3.628965306844521e-06, + "loss": 0.7248, + "step": 1408 + }, + { + "epoch": 2.2082463465553235, + "grad_norm": 1.0807738304138184, + "learning_rate": 3.6270852016224735e-06, + "loss": 0.6994, + "step": 1409 + }, + { + "epoch": 2.209812108559499, + "grad_norm": 1.2594057321548462, + "learning_rate": 3.625204296004885e-06, + "loss": 0.6429, + "step": 1410 + }, + { + "epoch": 2.211377870563674, + "grad_norm": 1.159574031829834, + "learning_rate": 3.623322591327476e-06, + "loss": 0.6292, + "step": 1411 + }, + { + "epoch": 2.2129436325678498, + "grad_norm": 1.2090522050857544, + "learning_rate": 3.621440088926531e-06, + "loss": 0.663, + "step": 1412 + }, + { + "epoch": 2.214509394572025, + "grad_norm": 1.176761507987976, + "learning_rate": 3.619556790138903e-06, + "loss": 0.5766, + "step": 1413 + }, + { + "epoch": 2.2160751565762005, + "grad_norm": 1.0924803018569946, + "learning_rate": 3.6176726963020085e-06, + "loss": 0.6838, + "step": 1414 + }, + { + "epoch": 2.2176409185803756, + "grad_norm": 1.0821013450622559, + "learning_rate": 3.6157878087538322e-06, + "loss": 0.6464, + "step": 1415 + }, + { + "epoch": 2.219206680584551, + "grad_norm": 1.0731130838394165, + "learning_rate": 3.613902128832919e-06, + "loss": 0.5345, + "step": 1416 + }, + { + "epoch": 2.2207724425887267, + "grad_norm": 1.0433294773101807, + "learning_rate": 3.612015657878378e-06, + "loss": 0.6799, + "step": 1417 + }, + { + "epoch": 2.222338204592902, + "grad_norm": 1.2107945680618286, + "learning_rate": 3.61012839722988e-06, + "loss": 0.746, + "step": 1418 + }, + { + "epoch": 2.223903966597077, + "grad_norm": 1.2256337404251099, + "learning_rate": 3.608240348227656e-06, + "loss": 0.6438, + "step": 1419 + }, + { + "epoch": 2.2254697286012526, + "grad_norm": 1.0490953922271729, + "learning_rate": 3.6063515122124975e-06, + "loss": 0.6618, + "step": 1420 + }, + { + "epoch": 2.227035490605428, + "grad_norm": 1.134993553161621, + "learning_rate": 3.6044618905257546e-06, + "loss": 0.6669, + "step": 1421 + }, + { + "epoch": 2.2286012526096033, + "grad_norm": 1.120219349861145, + "learning_rate": 3.602571484509336e-06, + "loss": 0.7164, + "step": 1422 + }, + { + "epoch": 2.230167014613779, + "grad_norm": 1.2099931240081787, + "learning_rate": 3.6006802955057057e-06, + "loss": 0.6499, + "step": 1423 + }, + { + "epoch": 2.231732776617954, + "grad_norm": 1.276110291481018, + "learning_rate": 3.5987883248578857e-06, + "loss": 0.7182, + "step": 1424 + }, + { + "epoch": 2.2332985386221296, + "grad_norm": 1.12351655960083, + "learning_rate": 3.5968955739094522e-06, + "loss": 0.6672, + "step": 1425 + }, + { + "epoch": 2.2348643006263047, + "grad_norm": 1.1756179332733154, + "learning_rate": 3.595002044004535e-06, + "loss": 0.6417, + "step": 1426 + }, + { + "epoch": 2.2364300626304803, + "grad_norm": 1.1440640687942505, + "learning_rate": 3.593107736487818e-06, + "loss": 0.6694, + "step": 1427 + }, + { + "epoch": 2.2379958246346554, + "grad_norm": 1.160673975944519, + "learning_rate": 3.5912126527045368e-06, + "loss": 0.7067, + "step": 1428 + }, + { + "epoch": 2.239561586638831, + "grad_norm": 1.1037743091583252, + "learning_rate": 3.589316794000478e-06, + "loss": 0.7258, + "step": 1429 + }, + { + "epoch": 2.241127348643006, + "grad_norm": 1.1246981620788574, + "learning_rate": 3.58742016172198e-06, + "loss": 0.6715, + "step": 1430 + }, + { + "epoch": 2.2426931106471817, + "grad_norm": 1.185558795928955, + "learning_rate": 3.5855227572159283e-06, + "loss": 0.6932, + "step": 1431 + }, + { + "epoch": 2.244258872651357, + "grad_norm": 1.0668050050735474, + "learning_rate": 3.583624581829759e-06, + "loss": 0.6836, + "step": 1432 + }, + { + "epoch": 2.2458246346555324, + "grad_norm": 1.1782646179199219, + "learning_rate": 3.581725636911454e-06, + "loss": 0.6787, + "step": 1433 + }, + { + "epoch": 2.2473903966597075, + "grad_norm": 1.2447540760040283, + "learning_rate": 3.579825923809542e-06, + "loss": 0.6702, + "step": 1434 + }, + { + "epoch": 2.248956158663883, + "grad_norm": 1.029355525970459, + "learning_rate": 3.577925443873098e-06, + "loss": 0.637, + "step": 1435 + }, + { + "epoch": 2.2505219206680582, + "grad_norm": 1.1154545545578003, + "learning_rate": 3.5760241984517397e-06, + "loss": 0.7274, + "step": 1436 + }, + { + "epoch": 2.252087682672234, + "grad_norm": 1.116469144821167, + "learning_rate": 3.5741221888956325e-06, + "loss": 0.6321, + "step": 1437 + }, + { + "epoch": 2.2536534446764094, + "grad_norm": 1.2985416650772095, + "learning_rate": 3.5722194165554795e-06, + "loss": 0.648, + "step": 1438 + }, + { + "epoch": 2.2552192066805845, + "grad_norm": 1.2045787572860718, + "learning_rate": 3.5703158827825277e-06, + "loss": 0.6138, + "step": 1439 + }, + { + "epoch": 2.25678496868476, + "grad_norm": 1.0173790454864502, + "learning_rate": 3.5684115889285663e-06, + "loss": 0.6144, + "step": 1440 + }, + { + "epoch": 2.258350730688935, + "grad_norm": 1.1876413822174072, + "learning_rate": 3.5665065363459215e-06, + "loss": 0.5951, + "step": 1441 + }, + { + "epoch": 2.259916492693111, + "grad_norm": 1.0441265106201172, + "learning_rate": 3.564600726387461e-06, + "loss": 0.6681, + "step": 1442 + }, + { + "epoch": 2.261482254697286, + "grad_norm": 1.2096240520477295, + "learning_rate": 3.5626941604065866e-06, + "loss": 0.7213, + "step": 1443 + }, + { + "epoch": 2.2630480167014615, + "grad_norm": 1.0928987264633179, + "learning_rate": 3.560786839757242e-06, + "loss": 0.6129, + "step": 1444 + }, + { + "epoch": 2.2646137787056366, + "grad_norm": 1.169466495513916, + "learning_rate": 3.558878765793902e-06, + "loss": 0.7212, + "step": 1445 + }, + { + "epoch": 2.266179540709812, + "grad_norm": 1.0599427223205566, + "learning_rate": 3.5569699398715794e-06, + "loss": 0.683, + "step": 1446 + }, + { + "epoch": 2.2677453027139873, + "grad_norm": 1.126380443572998, + "learning_rate": 3.5550603633458202e-06, + "loss": 0.6886, + "step": 1447 + }, + { + "epoch": 2.269311064718163, + "grad_norm": 1.2540115118026733, + "learning_rate": 3.553150037572704e-06, + "loss": 0.6515, + "step": 1448 + }, + { + "epoch": 2.270876826722338, + "grad_norm": 1.1289552450180054, + "learning_rate": 3.55123896390884e-06, + "loss": 0.7067, + "step": 1449 + }, + { + "epoch": 2.2724425887265136, + "grad_norm": 1.0645952224731445, + "learning_rate": 3.549327143711372e-06, + "loss": 0.7059, + "step": 1450 + }, + { + "epoch": 2.2740083507306887, + "grad_norm": 0.9427555799484253, + "learning_rate": 3.5474145783379728e-06, + "loss": 0.6472, + "step": 1451 + }, + { + "epoch": 2.2755741127348643, + "grad_norm": 1.1604477167129517, + "learning_rate": 3.5455012691468417e-06, + "loss": 0.6602, + "step": 1452 + }, + { + "epoch": 2.2771398747390394, + "grad_norm": 1.1867820024490356, + "learning_rate": 3.5435872174967102e-06, + "loss": 0.7162, + "step": 1453 + }, + { + "epoch": 2.278705636743215, + "grad_norm": 1.0727450847625732, + "learning_rate": 3.541672424746834e-06, + "loss": 0.7171, + "step": 1454 + }, + { + "epoch": 2.2802713987473906, + "grad_norm": 1.1541563272476196, + "learning_rate": 3.539756892256997e-06, + "loss": 0.7214, + "step": 1455 + }, + { + "epoch": 2.2818371607515657, + "grad_norm": 1.0292634963989258, + "learning_rate": 3.5378406213875084e-06, + "loss": 0.702, + "step": 1456 + }, + { + "epoch": 2.2834029227557413, + "grad_norm": 1.1958640813827515, + "learning_rate": 3.5359236134991996e-06, + "loss": 0.6456, + "step": 1457 + }, + { + "epoch": 2.2849686847599164, + "grad_norm": 1.191214680671692, + "learning_rate": 3.5340058699534287e-06, + "loss": 0.7149, + "step": 1458 + }, + { + "epoch": 2.286534446764092, + "grad_norm": 1.0933841466903687, + "learning_rate": 3.5320873921120735e-06, + "loss": 0.657, + "step": 1459 + }, + { + "epoch": 2.288100208768267, + "grad_norm": 1.0500751733779907, + "learning_rate": 3.5301681813375343e-06, + "loss": 0.6708, + "step": 1460 + }, + { + "epoch": 2.2896659707724427, + "grad_norm": 1.0765804052352905, + "learning_rate": 3.5282482389927313e-06, + "loss": 0.6528, + "step": 1461 + }, + { + "epoch": 2.291231732776618, + "grad_norm": 1.0991063117980957, + "learning_rate": 3.526327566441105e-06, + "loss": 0.7167, + "step": 1462 + }, + { + "epoch": 2.2927974947807934, + "grad_norm": 1.096111536026001, + "learning_rate": 3.5244061650466155e-06, + "loss": 0.6983, + "step": 1463 + }, + { + "epoch": 2.2943632567849686, + "grad_norm": 1.0495761632919312, + "learning_rate": 3.522484036173737e-06, + "loss": 0.6019, + "step": 1464 + }, + { + "epoch": 2.295929018789144, + "grad_norm": 1.204898476600647, + "learning_rate": 3.520561181187464e-06, + "loss": 0.7198, + "step": 1465 + }, + { + "epoch": 2.2974947807933193, + "grad_norm": 1.0273269414901733, + "learning_rate": 3.518637601453304e-06, + "loss": 0.7077, + "step": 1466 + }, + { + "epoch": 2.299060542797495, + "grad_norm": 1.0621248483657837, + "learning_rate": 3.5167132983372808e-06, + "loss": 0.703, + "step": 1467 + }, + { + "epoch": 2.30062630480167, + "grad_norm": 1.216728687286377, + "learning_rate": 3.5147882732059323e-06, + "loss": 0.5898, + "step": 1468 + }, + { + "epoch": 2.3021920668058455, + "grad_norm": 1.065699815750122, + "learning_rate": 3.5128625274263074e-06, + "loss": 0.611, + "step": 1469 + }, + { + "epoch": 2.3037578288100207, + "grad_norm": 1.1529189348220825, + "learning_rate": 3.5109360623659672e-06, + "loss": 0.6778, + "step": 1470 + }, + { + "epoch": 2.3053235908141962, + "grad_norm": 1.1462815999984741, + "learning_rate": 3.5090088793929842e-06, + "loss": 0.6366, + "step": 1471 + }, + { + "epoch": 2.306889352818372, + "grad_norm": 1.0135819911956787, + "learning_rate": 3.507080979875941e-06, + "loss": 0.6393, + "step": 1472 + }, + { + "epoch": 2.308455114822547, + "grad_norm": 1.112345814704895, + "learning_rate": 3.5051523651839288e-06, + "loss": 0.595, + "step": 1473 + }, + { + "epoch": 2.3100208768267225, + "grad_norm": 1.0799155235290527, + "learning_rate": 3.503223036686545e-06, + "loss": 0.7024, + "step": 1474 + }, + { + "epoch": 2.3115866388308977, + "grad_norm": 1.2584905624389648, + "learning_rate": 3.501292995753897e-06, + "loss": 0.6308, + "step": 1475 + }, + { + "epoch": 2.3131524008350732, + "grad_norm": 1.0637624263763428, + "learning_rate": 3.4993622437565955e-06, + "loss": 0.6502, + "step": 1476 + }, + { + "epoch": 2.3147181628392484, + "grad_norm": 1.2071688175201416, + "learning_rate": 3.4974307820657584e-06, + "loss": 0.6389, + "step": 1477 + }, + { + "epoch": 2.316283924843424, + "grad_norm": 1.2328526973724365, + "learning_rate": 3.4954986120530042e-06, + "loss": 0.6748, + "step": 1478 + }, + { + "epoch": 2.317849686847599, + "grad_norm": 1.1051819324493408, + "learning_rate": 3.493565735090458e-06, + "loss": 0.6675, + "step": 1479 + }, + { + "epoch": 2.3194154488517746, + "grad_norm": 1.1707861423492432, + "learning_rate": 3.491632152550746e-06, + "loss": 0.6305, + "step": 1480 + }, + { + "epoch": 2.3209812108559498, + "grad_norm": 1.0759583711624146, + "learning_rate": 3.4896978658069937e-06, + "loss": 0.683, + "step": 1481 + }, + { + "epoch": 2.3225469728601253, + "grad_norm": 1.1699646711349487, + "learning_rate": 3.487762876232828e-06, + "loss": 0.7231, + "step": 1482 + }, + { + "epoch": 2.3241127348643005, + "grad_norm": 1.0097750425338745, + "learning_rate": 3.4858271852023756e-06, + "loss": 0.5897, + "step": 1483 + }, + { + "epoch": 2.325678496868476, + "grad_norm": 1.0517746210098267, + "learning_rate": 3.4838907940902607e-06, + "loss": 0.6386, + "step": 1484 + }, + { + "epoch": 2.327244258872651, + "grad_norm": 1.2001336812973022, + "learning_rate": 3.4819537042716033e-06, + "loss": 0.6854, + "step": 1485 + }, + { + "epoch": 2.3288100208768268, + "grad_norm": 1.1421329975128174, + "learning_rate": 3.480015917122023e-06, + "loss": 0.7186, + "step": 1486 + }, + { + "epoch": 2.330375782881002, + "grad_norm": 1.0638641119003296, + "learning_rate": 3.47807743401763e-06, + "loss": 0.6848, + "step": 1487 + }, + { + "epoch": 2.3319415448851775, + "grad_norm": 1.019140601158142, + "learning_rate": 3.476138256335033e-06, + "loss": 0.6904, + "step": 1488 + }, + { + "epoch": 2.333507306889353, + "grad_norm": 1.2467342615127563, + "learning_rate": 3.4741983854513306e-06, + "loss": 0.753, + "step": 1489 + }, + { + "epoch": 2.335073068893528, + "grad_norm": 1.1455196142196655, + "learning_rate": 3.472257822744116e-06, + "loss": 0.6544, + "step": 1490 + }, + { + "epoch": 2.3366388308977033, + "grad_norm": 1.1508904695510864, + "learning_rate": 3.4703165695914737e-06, + "loss": 0.6216, + "step": 1491 + }, + { + "epoch": 2.338204592901879, + "grad_norm": 1.1695144176483154, + "learning_rate": 3.4683746273719754e-06, + "loss": 0.6478, + "step": 1492 + }, + { + "epoch": 2.3397703549060545, + "grad_norm": 1.041002631187439, + "learning_rate": 3.4664319974646872e-06, + "loss": 0.7142, + "step": 1493 + }, + { + "epoch": 2.3413361169102296, + "grad_norm": 1.0497469902038574, + "learning_rate": 3.4644886812491598e-06, + "loss": 0.7028, + "step": 1494 + }, + { + "epoch": 2.342901878914405, + "grad_norm": 1.0564930438995361, + "learning_rate": 3.4625446801054312e-06, + "loss": 0.6796, + "step": 1495 + }, + { + "epoch": 2.3444676409185803, + "grad_norm": 1.2151904106140137, + "learning_rate": 3.460599995414028e-06, + "loss": 0.6933, + "step": 1496 + }, + { + "epoch": 2.346033402922756, + "grad_norm": 1.197263240814209, + "learning_rate": 3.458654628555961e-06, + "loss": 0.7615, + "step": 1497 + }, + { + "epoch": 2.347599164926931, + "grad_norm": 1.037292718887329, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.7014, + "step": 1498 + }, + { + "epoch": 2.3491649269311066, + "grad_norm": 1.0831308364868164, + "learning_rate": 3.454761853866299e-06, + "loss": 0.6682, + "step": 1499 + }, + { + "epoch": 2.3507306889352817, + "grad_norm": 1.0729385614395142, + "learning_rate": 3.4528144487991448e-06, + "loss": 0.725, + "step": 1500 + }, + { + "epoch": 2.3522964509394573, + "grad_norm": 1.010629415512085, + "learning_rate": 3.4508663670942044e-06, + "loss": 0.6579, + "step": 1501 + }, + { + "epoch": 2.3538622129436324, + "grad_norm": 1.0924932956695557, + "learning_rate": 3.448917610134902e-06, + "loss": 0.6892, + "step": 1502 + }, + { + "epoch": 2.355427974947808, + "grad_norm": 1.1776976585388184, + "learning_rate": 3.4469681793051403e-06, + "loss": 0.6366, + "step": 1503 + }, + { + "epoch": 2.356993736951983, + "grad_norm": 1.062983751296997, + "learning_rate": 3.4450180759892994e-06, + "loss": 0.6518, + "step": 1504 + }, + { + "epoch": 2.3585594989561587, + "grad_norm": 1.1387819051742554, + "learning_rate": 3.4430673015722393e-06, + "loss": 0.623, + "step": 1505 + }, + { + "epoch": 2.3601252609603343, + "grad_norm": 1.035445213317871, + "learning_rate": 3.4411158574392956e-06, + "loss": 0.6876, + "step": 1506 + }, + { + "epoch": 2.3616910229645094, + "grad_norm": 1.0623654127120972, + "learning_rate": 3.439163744976279e-06, + "loss": 0.6795, + "step": 1507 + }, + { + "epoch": 2.3632567849686845, + "grad_norm": 1.0046356916427612, + "learning_rate": 3.437210965569475e-06, + "loss": 0.7043, + "step": 1508 + }, + { + "epoch": 2.36482254697286, + "grad_norm": 1.183174729347229, + "learning_rate": 3.4352575206056426e-06, + "loss": 0.7059, + "step": 1509 + }, + { + "epoch": 2.3663883089770357, + "grad_norm": 1.0844060182571411, + "learning_rate": 3.4333034114720155e-06, + "loss": 0.6811, + "step": 1510 + }, + { + "epoch": 2.367954070981211, + "grad_norm": 1.058577299118042, + "learning_rate": 3.431348639556296e-06, + "loss": 0.7258, + "step": 1511 + }, + { + "epoch": 2.3695198329853864, + "grad_norm": 1.0441113710403442, + "learning_rate": 3.429393206246659e-06, + "loss": 0.6659, + "step": 1512 + }, + { + "epoch": 2.3710855949895615, + "grad_norm": 1.111039400100708, + "learning_rate": 3.4274371129317476e-06, + "loss": 0.6113, + "step": 1513 + }, + { + "epoch": 2.372651356993737, + "grad_norm": 1.1454105377197266, + "learning_rate": 3.425480361000675e-06, + "loss": 0.727, + "step": 1514 + }, + { + "epoch": 2.374217118997912, + "grad_norm": 1.1241672039031982, + "learning_rate": 3.4235229518430212e-06, + "loss": 0.5266, + "step": 1515 + }, + { + "epoch": 2.375782881002088, + "grad_norm": 1.0866069793701172, + "learning_rate": 3.421564886848835e-06, + "loss": 0.6871, + "step": 1516 + }, + { + "epoch": 2.377348643006263, + "grad_norm": 1.211893081665039, + "learning_rate": 3.4196061674086276e-06, + "loss": 0.7064, + "step": 1517 + }, + { + "epoch": 2.3789144050104385, + "grad_norm": 1.1039608716964722, + "learning_rate": 3.417646794913376e-06, + "loss": 0.6821, + "step": 1518 + }, + { + "epoch": 2.3804801670146136, + "grad_norm": 1.0984389781951904, + "learning_rate": 3.4156867707545237e-06, + "loss": 0.6443, + "step": 1519 + }, + { + "epoch": 2.382045929018789, + "grad_norm": 1.11539888381958, + "learning_rate": 3.413726096323975e-06, + "loss": 0.6896, + "step": 1520 + }, + { + "epoch": 2.3836116910229643, + "grad_norm": 1.1566188335418701, + "learning_rate": 3.4117647730140935e-06, + "loss": 0.699, + "step": 1521 + }, + { + "epoch": 2.38517745302714, + "grad_norm": 1.0620112419128418, + "learning_rate": 3.409802802217707e-06, + "loss": 0.6741, + "step": 1522 + }, + { + "epoch": 2.3867432150313155, + "grad_norm": 1.117667317390442, + "learning_rate": 3.4078401853281022e-06, + "loss": 0.7027, + "step": 1523 + }, + { + "epoch": 2.3883089770354906, + "grad_norm": 1.0997360944747925, + "learning_rate": 3.4058769237390254e-06, + "loss": 0.7249, + "step": 1524 + }, + { + "epoch": 2.3898747390396657, + "grad_norm": 1.1101152896881104, + "learning_rate": 3.403913018844678e-06, + "loss": 0.6794, + "step": 1525 + }, + { + "epoch": 2.3914405010438413, + "grad_norm": 1.1798306703567505, + "learning_rate": 3.4019484720397217e-06, + "loss": 0.6219, + "step": 1526 + }, + { + "epoch": 2.393006263048017, + "grad_norm": 1.0730222463607788, + "learning_rate": 3.3999832847192716e-06, + "loss": 0.6622, + "step": 1527 + }, + { + "epoch": 2.394572025052192, + "grad_norm": 1.0943461656570435, + "learning_rate": 3.3980174582788982e-06, + "loss": 0.6999, + "step": 1528 + }, + { + "epoch": 2.3961377870563676, + "grad_norm": 1.1592621803283691, + "learning_rate": 3.396050994114628e-06, + "loss": 0.6419, + "step": 1529 + }, + { + "epoch": 2.3977035490605427, + "grad_norm": 1.1652766466140747, + "learning_rate": 3.3940838936229357e-06, + "loss": 0.6952, + "step": 1530 + }, + { + "epoch": 2.3992693110647183, + "grad_norm": 1.1424206495285034, + "learning_rate": 3.3921161582007524e-06, + "loss": 0.6521, + "step": 1531 + }, + { + "epoch": 2.4008350730688934, + "grad_norm": 1.0935698747634888, + "learning_rate": 3.3901477892454583e-06, + "loss": 0.6707, + "step": 1532 + }, + { + "epoch": 2.402400835073069, + "grad_norm": 1.2164827585220337, + "learning_rate": 3.3881787881548835e-06, + "loss": 0.6485, + "step": 1533 + }, + { + "epoch": 2.403966597077244, + "grad_norm": 1.0536198616027832, + "learning_rate": 3.386209156327307e-06, + "loss": 0.6997, + "step": 1534 + }, + { + "epoch": 2.4055323590814197, + "grad_norm": 1.1083351373672485, + "learning_rate": 3.3842388951614545e-06, + "loss": 0.6264, + "step": 1535 + }, + { + "epoch": 2.407098121085595, + "grad_norm": 1.133180856704712, + "learning_rate": 3.3822680060565026e-06, + "loss": 0.6905, + "step": 1536 + }, + { + "epoch": 2.4086638830897704, + "grad_norm": 1.0531203746795654, + "learning_rate": 3.3802964904120703e-06, + "loss": 0.6492, + "step": 1537 + }, + { + "epoch": 2.4102296450939455, + "grad_norm": 1.0389595031738281, + "learning_rate": 3.3783243496282215e-06, + "loss": 0.6506, + "step": 1538 + }, + { + "epoch": 2.411795407098121, + "grad_norm": 1.1839019060134888, + "learning_rate": 3.376351585105465e-06, + "loss": 0.6933, + "step": 1539 + }, + { + "epoch": 2.4133611691022967, + "grad_norm": 1.0715898275375366, + "learning_rate": 3.3743781982447533e-06, + "loss": 0.7043, + "step": 1540 + }, + { + "epoch": 2.414926931106472, + "grad_norm": 1.1898013353347778, + "learning_rate": 3.3724041904474793e-06, + "loss": 0.7054, + "step": 1541 + }, + { + "epoch": 2.416492693110647, + "grad_norm": 1.026489019393921, + "learning_rate": 3.370429563115478e-06, + "loss": 0.6458, + "step": 1542 + }, + { + "epoch": 2.4180584551148225, + "grad_norm": 1.2772443294525146, + "learning_rate": 3.368454317651024e-06, + "loss": 0.6647, + "step": 1543 + }, + { + "epoch": 2.419624217118998, + "grad_norm": 1.0573022365570068, + "learning_rate": 3.366478455456831e-06, + "loss": 0.6723, + "step": 1544 + }, + { + "epoch": 2.4211899791231732, + "grad_norm": 1.2983055114746094, + "learning_rate": 3.3645019779360493e-06, + "loss": 0.6922, + "step": 1545 + }, + { + "epoch": 2.422755741127349, + "grad_norm": 1.0773214101791382, + "learning_rate": 3.36252488649227e-06, + "loss": 0.6538, + "step": 1546 + }, + { + "epoch": 2.424321503131524, + "grad_norm": 1.1467341184616089, + "learning_rate": 3.3605471825295143e-06, + "loss": 0.7274, + "step": 1547 + }, + { + "epoch": 2.4258872651356995, + "grad_norm": 1.1962897777557373, + "learning_rate": 3.3585688674522438e-06, + "loss": 0.7113, + "step": 1548 + }, + { + "epoch": 2.4274530271398747, + "grad_norm": 0.997977614402771, + "learning_rate": 3.3565899426653515e-06, + "loss": 0.6731, + "step": 1549 + }, + { + "epoch": 2.4290187891440502, + "grad_norm": 1.095198392868042, + "learning_rate": 3.3546104095741634e-06, + "loss": 0.6353, + "step": 1550 + }, + { + "epoch": 2.4305845511482254, + "grad_norm": 1.0401874780654907, + "learning_rate": 3.3526302695844382e-06, + "loss": 0.7459, + "step": 1551 + }, + { + "epoch": 2.432150313152401, + "grad_norm": 1.174526333808899, + "learning_rate": 3.3506495241023646e-06, + "loss": 0.6435, + "step": 1552 + }, + { + "epoch": 2.433716075156576, + "grad_norm": 1.0898199081420898, + "learning_rate": 3.3486681745345627e-06, + "loss": 0.6991, + "step": 1553 + }, + { + "epoch": 2.4352818371607516, + "grad_norm": 1.0757240056991577, + "learning_rate": 3.346686222288081e-06, + "loss": 0.6447, + "step": 1554 + }, + { + "epoch": 2.4368475991649268, + "grad_norm": 1.0817614793777466, + "learning_rate": 3.3447036687703956e-06, + "loss": 0.6546, + "step": 1555 + }, + { + "epoch": 2.4384133611691023, + "grad_norm": 1.181427240371704, + "learning_rate": 3.3427205153894088e-06, + "loss": 0.6592, + "step": 1556 + }, + { + "epoch": 2.4399791231732775, + "grad_norm": 1.0263524055480957, + "learning_rate": 3.340736763553451e-06, + "loss": 0.6766, + "step": 1557 + }, + { + "epoch": 2.441544885177453, + "grad_norm": 1.0632904767990112, + "learning_rate": 3.3387524146712762e-06, + "loss": 0.6368, + "step": 1558 + }, + { + "epoch": 2.443110647181628, + "grad_norm": 1.165933609008789, + "learning_rate": 3.3367674701520627e-06, + "loss": 0.692, + "step": 1559 + }, + { + "epoch": 2.4446764091858038, + "grad_norm": 1.109712839126587, + "learning_rate": 3.3347819314054124e-06, + "loss": 0.6721, + "step": 1560 + }, + { + "epoch": 2.4462421711899793, + "grad_norm": 1.0221387147903442, + "learning_rate": 3.332795799841347e-06, + "loss": 0.6547, + "step": 1561 + }, + { + "epoch": 2.4478079331941545, + "grad_norm": 1.1391754150390625, + "learning_rate": 3.3308090768703118e-06, + "loss": 0.68, + "step": 1562 + }, + { + "epoch": 2.44937369519833, + "grad_norm": 1.0794706344604492, + "learning_rate": 3.328821763903172e-06, + "loss": 0.6531, + "step": 1563 + }, + { + "epoch": 2.450939457202505, + "grad_norm": 1.0358234643936157, + "learning_rate": 3.3268338623512094e-06, + "loss": 0.6638, + "step": 1564 + }, + { + "epoch": 2.4525052192066807, + "grad_norm": 1.1351442337036133, + "learning_rate": 3.3248453736261247e-06, + "loss": 0.6356, + "step": 1565 + }, + { + "epoch": 2.454070981210856, + "grad_norm": 1.010600209236145, + "learning_rate": 3.322856299140038e-06, + "loss": 0.6722, + "step": 1566 + }, + { + "epoch": 2.4556367432150314, + "grad_norm": 1.216776728630066, + "learning_rate": 3.320866640305482e-06, + "loss": 0.658, + "step": 1567 + }, + { + "epoch": 2.4572025052192066, + "grad_norm": 1.0781611204147339, + "learning_rate": 3.3188763985354055e-06, + "loss": 0.711, + "step": 1568 + }, + { + "epoch": 2.458768267223382, + "grad_norm": 1.1265203952789307, + "learning_rate": 3.3168855752431732e-06, + "loss": 0.6746, + "step": 1569 + }, + { + "epoch": 2.4603340292275573, + "grad_norm": 1.1726053953170776, + "learning_rate": 3.3148941718425606e-06, + "loss": 0.6836, + "step": 1570 + }, + { + "epoch": 2.461899791231733, + "grad_norm": 1.1113516092300415, + "learning_rate": 3.3129021897477546e-06, + "loss": 0.6322, + "step": 1571 + }, + { + "epoch": 2.463465553235908, + "grad_norm": 1.0234968662261963, + "learning_rate": 3.3109096303733564e-06, + "loss": 0.7179, + "step": 1572 + }, + { + "epoch": 2.4650313152400836, + "grad_norm": 1.1797139644622803, + "learning_rate": 3.308916495134372e-06, + "loss": 0.6996, + "step": 1573 + }, + { + "epoch": 2.4665970772442587, + "grad_norm": 0.9623079895973206, + "learning_rate": 3.3069227854462216e-06, + "loss": 0.6583, + "step": 1574 + }, + { + "epoch": 2.4681628392484343, + "grad_norm": 1.1546630859375, + "learning_rate": 3.30492850272473e-06, + "loss": 0.7457, + "step": 1575 + }, + { + "epoch": 2.4697286012526094, + "grad_norm": 1.0559042692184448, + "learning_rate": 3.302933648386131e-06, + "loss": 0.692, + "step": 1576 + }, + { + "epoch": 2.471294363256785, + "grad_norm": 1.0812047719955444, + "learning_rate": 3.300938223847062e-06, + "loss": 0.6795, + "step": 1577 + }, + { + "epoch": 2.4728601252609606, + "grad_norm": 1.0247067213058472, + "learning_rate": 3.298942230524567e-06, + "loss": 0.695, + "step": 1578 + }, + { + "epoch": 2.4744258872651357, + "grad_norm": 1.354367971420288, + "learning_rate": 3.296945669836094e-06, + "loss": 0.7014, + "step": 1579 + }, + { + "epoch": 2.4759916492693113, + "grad_norm": 1.0897941589355469, + "learning_rate": 3.2949485431994945e-06, + "loss": 0.6338, + "step": 1580 + }, + { + "epoch": 2.4775574112734864, + "grad_norm": 0.9883112907409668, + "learning_rate": 3.292950852033018e-06, + "loss": 0.7102, + "step": 1581 + }, + { + "epoch": 2.479123173277662, + "grad_norm": 1.1789237260818481, + "learning_rate": 3.290952597755319e-06, + "loss": 0.6802, + "step": 1582 + }, + { + "epoch": 2.480688935281837, + "grad_norm": 1.1877659559249878, + "learning_rate": 3.288953781785452e-06, + "loss": 0.6846, + "step": 1583 + }, + { + "epoch": 2.4822546972860127, + "grad_norm": 1.0334861278533936, + "learning_rate": 3.2869544055428678e-06, + "loss": 0.7138, + "step": 1584 + }, + { + "epoch": 2.483820459290188, + "grad_norm": 1.2335622310638428, + "learning_rate": 3.284954470447416e-06, + "loss": 0.6125, + "step": 1585 + }, + { + "epoch": 2.4853862212943634, + "grad_norm": 1.1405894756317139, + "learning_rate": 3.282953977919345e-06, + "loss": 0.6801, + "step": 1586 + }, + { + "epoch": 2.4869519832985385, + "grad_norm": 1.1099573373794556, + "learning_rate": 3.2809529293792953e-06, + "loss": 0.7071, + "step": 1587 + }, + { + "epoch": 2.488517745302714, + "grad_norm": 1.1282474994659424, + "learning_rate": 3.2789513262483053e-06, + "loss": 0.7082, + "step": 1588 + }, + { + "epoch": 2.490083507306889, + "grad_norm": 1.1693743467330933, + "learning_rate": 3.2769491699478077e-06, + "loss": 0.7137, + "step": 1589 + }, + { + "epoch": 2.491649269311065, + "grad_norm": 1.1324049234390259, + "learning_rate": 3.2749464618996244e-06, + "loss": 0.6402, + "step": 1590 + }, + { + "epoch": 2.49321503131524, + "grad_norm": 1.1645187139511108, + "learning_rate": 3.2729432035259733e-06, + "loss": 0.6941, + "step": 1591 + }, + { + "epoch": 2.4947807933194155, + "grad_norm": 1.1647381782531738, + "learning_rate": 3.2709393962494597e-06, + "loss": 0.6583, + "step": 1592 + }, + { + "epoch": 2.4963465553235906, + "grad_norm": 1.1749935150146484, + "learning_rate": 3.268935041493081e-06, + "loss": 0.7099, + "step": 1593 + }, + { + "epoch": 2.497912317327766, + "grad_norm": 1.1859298944473267, + "learning_rate": 3.266930140680222e-06, + "loss": 0.6551, + "step": 1594 + }, + { + "epoch": 2.4994780793319418, + "grad_norm": 1.1642869710922241, + "learning_rate": 3.264924695234657e-06, + "loss": 0.6712, + "step": 1595 + }, + { + "epoch": 2.501043841336117, + "grad_norm": 1.1522237062454224, + "learning_rate": 3.2629187065805445e-06, + "loss": 0.6773, + "step": 1596 + }, + { + "epoch": 2.502609603340292, + "grad_norm": 1.142119288444519, + "learning_rate": 3.2609121761424312e-06, + "loss": 0.6077, + "step": 1597 + }, + { + "epoch": 2.5041753653444676, + "grad_norm": 1.137640357017517, + "learning_rate": 3.258905105345248e-06, + "loss": 0.7086, + "step": 1598 + }, + { + "epoch": 2.505741127348643, + "grad_norm": 1.1106284856796265, + "learning_rate": 3.2568974956143073e-06, + "loss": 0.6795, + "step": 1599 + }, + { + "epoch": 2.5073068893528183, + "grad_norm": 1.0668182373046875, + "learning_rate": 3.254889348375307e-06, + "loss": 0.7163, + "step": 1600 + }, + { + "epoch": 2.508872651356994, + "grad_norm": 1.2222528457641602, + "learning_rate": 3.2528806650543264e-06, + "loss": 0.6589, + "step": 1601 + }, + { + "epoch": 2.510438413361169, + "grad_norm": 1.0996103286743164, + "learning_rate": 3.250871447077824e-06, + "loss": 0.6097, + "step": 1602 + }, + { + "epoch": 2.5120041753653446, + "grad_norm": 1.2791520357131958, + "learning_rate": 3.248861695872639e-06, + "loss": 0.6367, + "step": 1603 + }, + { + "epoch": 2.5135699373695197, + "grad_norm": 1.1636749505996704, + "learning_rate": 3.2468514128659884e-06, + "loss": 0.6271, + "step": 1604 + }, + { + "epoch": 2.5151356993736953, + "grad_norm": 1.1000317335128784, + "learning_rate": 3.2448405994854683e-06, + "loss": 0.6746, + "step": 1605 + }, + { + "epoch": 2.5167014613778704, + "grad_norm": 1.163297414779663, + "learning_rate": 3.2428292571590514e-06, + "loss": 0.6248, + "step": 1606 + }, + { + "epoch": 2.518267223382046, + "grad_norm": 1.0616698265075684, + "learning_rate": 3.240817387315083e-06, + "loss": 0.6249, + "step": 1607 + }, + { + "epoch": 2.519832985386221, + "grad_norm": 1.2272953987121582, + "learning_rate": 3.2388049913822874e-06, + "loss": 0.6856, + "step": 1608 + }, + { + "epoch": 2.5213987473903967, + "grad_norm": 1.0715241432189941, + "learning_rate": 3.2367920707897594e-06, + "loss": 0.6348, + "step": 1609 + }, + { + "epoch": 2.522964509394572, + "grad_norm": 1.069440245628357, + "learning_rate": 3.234778626966968e-06, + "loss": 0.698, + "step": 1610 + }, + { + "epoch": 2.5245302713987474, + "grad_norm": 1.0824602842330933, + "learning_rate": 3.2327646613437525e-06, + "loss": 0.6809, + "step": 1611 + }, + { + "epoch": 2.526096033402923, + "grad_norm": 1.1153802871704102, + "learning_rate": 3.230750175350324e-06, + "loss": 0.6562, + "step": 1612 + }, + { + "epoch": 2.527661795407098, + "grad_norm": 1.1219229698181152, + "learning_rate": 3.228735170417261e-06, + "loss": 0.6819, + "step": 1613 + }, + { + "epoch": 2.5292275574112733, + "grad_norm": 0.9705930352210999, + "learning_rate": 3.226719647975514e-06, + "loss": 0.7457, + "step": 1614 + }, + { + "epoch": 2.530793319415449, + "grad_norm": 1.0846235752105713, + "learning_rate": 3.2247036094563992e-06, + "loss": 0.6122, + "step": 1615 + }, + { + "epoch": 2.5323590814196244, + "grad_norm": 1.0858566761016846, + "learning_rate": 3.222687056291598e-06, + "loss": 0.6141, + "step": 1616 + }, + { + "epoch": 2.5339248434237995, + "grad_norm": 1.0649619102478027, + "learning_rate": 3.2206699899131576e-06, + "loss": 0.6813, + "step": 1617 + }, + { + "epoch": 2.535490605427975, + "grad_norm": 1.2037075757980347, + "learning_rate": 3.2186524117534935e-06, + "loss": 0.7131, + "step": 1618 + }, + { + "epoch": 2.5370563674321502, + "grad_norm": 1.217882513999939, + "learning_rate": 3.216634323245379e-06, + "loss": 0.6977, + "step": 1619 + }, + { + "epoch": 2.538622129436326, + "grad_norm": 1.1430108547210693, + "learning_rate": 3.2146157258219534e-06, + "loss": 0.6799, + "step": 1620 + }, + { + "epoch": 2.540187891440501, + "grad_norm": 1.1914441585540771, + "learning_rate": 3.2125966209167166e-06, + "loss": 0.7284, + "step": 1621 + }, + { + "epoch": 2.5417536534446765, + "grad_norm": 1.1642671823501587, + "learning_rate": 3.210577009963528e-06, + "loss": 0.7203, + "step": 1622 + }, + { + "epoch": 2.5433194154488517, + "grad_norm": 1.0995980501174927, + "learning_rate": 3.208556894396608e-06, + "loss": 0.6635, + "step": 1623 + }, + { + "epoch": 2.5448851774530272, + "grad_norm": 1.1318360567092896, + "learning_rate": 3.2065362756505345e-06, + "loss": 0.6537, + "step": 1624 + }, + { + "epoch": 2.5464509394572024, + "grad_norm": 1.2233246564865112, + "learning_rate": 3.204515155160242e-06, + "loss": 0.7006, + "step": 1625 + }, + { + "epoch": 2.548016701461378, + "grad_norm": 0.9861915707588196, + "learning_rate": 3.202493534361022e-06, + "loss": 0.6647, + "step": 1626 + }, + { + "epoch": 2.549582463465553, + "grad_norm": 1.2189120054244995, + "learning_rate": 3.2004714146885225e-06, + "loss": 0.6374, + "step": 1627 + }, + { + "epoch": 2.5511482254697286, + "grad_norm": 1.165047526359558, + "learning_rate": 3.1984487975787433e-06, + "loss": 0.6287, + "step": 1628 + }, + { + "epoch": 2.552713987473904, + "grad_norm": 1.1922754049301147, + "learning_rate": 3.196425684468039e-06, + "loss": 0.7103, + "step": 1629 + }, + { + "epoch": 2.5542797494780793, + "grad_norm": 1.1970739364624023, + "learning_rate": 3.194402076793116e-06, + "loss": 0.6951, + "step": 1630 + }, + { + "epoch": 2.5558455114822545, + "grad_norm": 1.1461929082870483, + "learning_rate": 3.1923779759910327e-06, + "loss": 0.6989, + "step": 1631 + }, + { + "epoch": 2.55741127348643, + "grad_norm": 1.1536473035812378, + "learning_rate": 3.190353383499198e-06, + "loss": 0.6433, + "step": 1632 + }, + { + "epoch": 2.5589770354906056, + "grad_norm": 1.3109079599380493, + "learning_rate": 3.188328300755367e-06, + "loss": 0.6873, + "step": 1633 + }, + { + "epoch": 2.5605427974947808, + "grad_norm": 1.1532987356185913, + "learning_rate": 3.186302729197645e-06, + "loss": 0.6818, + "step": 1634 + }, + { + "epoch": 2.562108559498956, + "grad_norm": 1.1802146434783936, + "learning_rate": 3.184276670264486e-06, + "loss": 0.7233, + "step": 1635 + }, + { + "epoch": 2.5636743215031315, + "grad_norm": 1.1282404661178589, + "learning_rate": 3.1822501253946875e-06, + "loss": 0.7203, + "step": 1636 + }, + { + "epoch": 2.565240083507307, + "grad_norm": 1.0870387554168701, + "learning_rate": 3.180223096027394e-06, + "loss": 0.6472, + "step": 1637 + }, + { + "epoch": 2.566805845511482, + "grad_norm": 1.038499355316162, + "learning_rate": 3.178195583602092e-06, + "loss": 0.7197, + "step": 1638 + }, + { + "epoch": 2.5683716075156577, + "grad_norm": 1.0605285167694092, + "learning_rate": 3.176167589558613e-06, + "loss": 0.7222, + "step": 1639 + }, + { + "epoch": 2.569937369519833, + "grad_norm": 1.1737507581710815, + "learning_rate": 3.174139115337129e-06, + "loss": 0.6215, + "step": 1640 + }, + { + "epoch": 2.5715031315240084, + "grad_norm": 1.0304044485092163, + "learning_rate": 3.1721101623781556e-06, + "loss": 0.7113, + "step": 1641 + }, + { + "epoch": 2.5730688935281836, + "grad_norm": 1.1509263515472412, + "learning_rate": 3.170080732122545e-06, + "loss": 0.6521, + "step": 1642 + }, + { + "epoch": 2.574634655532359, + "grad_norm": 1.2465925216674805, + "learning_rate": 3.1680508260114882e-06, + "loss": 0.6921, + "step": 1643 + }, + { + "epoch": 2.5762004175365343, + "grad_norm": 1.0584062337875366, + "learning_rate": 3.1660204454865203e-06, + "loss": 0.7041, + "step": 1644 + }, + { + "epoch": 2.57776617954071, + "grad_norm": 1.1217284202575684, + "learning_rate": 3.163989591989505e-06, + "loss": 0.7177, + "step": 1645 + }, + { + "epoch": 2.5793319415448854, + "grad_norm": 1.1175196170806885, + "learning_rate": 3.1619582669626457e-06, + "loss": 0.7192, + "step": 1646 + }, + { + "epoch": 2.5808977035490606, + "grad_norm": 1.1421573162078857, + "learning_rate": 3.1599264718484835e-06, + "loss": 0.7199, + "step": 1647 + }, + { + "epoch": 2.5824634655532357, + "grad_norm": 1.0428645610809326, + "learning_rate": 3.1578942080898876e-06, + "loss": 0.6496, + "step": 1648 + }, + { + "epoch": 2.5840292275574113, + "grad_norm": 0.9507089257240295, + "learning_rate": 3.1558614771300645e-06, + "loss": 0.6271, + "step": 1649 + }, + { + "epoch": 2.585594989561587, + "grad_norm": 1.2682905197143555, + "learning_rate": 3.1538282804125507e-06, + "loss": 0.6635, + "step": 1650 + }, + { + "epoch": 2.587160751565762, + "grad_norm": 1.098144769668579, + "learning_rate": 3.1517946193812132e-06, + "loss": 0.6631, + "step": 1651 + }, + { + "epoch": 2.588726513569937, + "grad_norm": 1.1256821155548096, + "learning_rate": 3.1497604954802485e-06, + "loss": 0.5353, + "step": 1652 + }, + { + "epoch": 2.5902922755741127, + "grad_norm": 1.1058018207550049, + "learning_rate": 3.1477259101541834e-06, + "loss": 0.6324, + "step": 1653 + }, + { + "epoch": 2.5918580375782883, + "grad_norm": 1.200184941291809, + "learning_rate": 3.145690864847871e-06, + "loss": 0.6484, + "step": 1654 + }, + { + "epoch": 2.5934237995824634, + "grad_norm": 1.0515025854110718, + "learning_rate": 3.1436553610064923e-06, + "loss": 0.6774, + "step": 1655 + }, + { + "epoch": 2.594989561586639, + "grad_norm": 1.0765128135681152, + "learning_rate": 3.1416194000755517e-06, + "loss": 0.6838, + "step": 1656 + }, + { + "epoch": 2.596555323590814, + "grad_norm": 1.157872200012207, + "learning_rate": 3.139582983500881e-06, + "loss": 0.7306, + "step": 1657 + }, + { + "epoch": 2.5981210855949897, + "grad_norm": 1.056947946548462, + "learning_rate": 3.1375461127286334e-06, + "loss": 0.628, + "step": 1658 + }, + { + "epoch": 2.599686847599165, + "grad_norm": 1.1467186212539673, + "learning_rate": 3.135508789205285e-06, + "loss": 0.6829, + "step": 1659 + }, + { + "epoch": 2.6012526096033404, + "grad_norm": 1.0062966346740723, + "learning_rate": 3.1334710143776346e-06, + "loss": 0.654, + "step": 1660 + }, + { + "epoch": 2.6028183716075155, + "grad_norm": 1.1658058166503906, + "learning_rate": 3.1314327896928006e-06, + "loss": 0.7067, + "step": 1661 + }, + { + "epoch": 2.604384133611691, + "grad_norm": 1.2473435401916504, + "learning_rate": 3.12939411659822e-06, + "loss": 0.668, + "step": 1662 + }, + { + "epoch": 2.6059498956158667, + "grad_norm": 1.171295404434204, + "learning_rate": 3.127354996541651e-06, + "loss": 0.6847, + "step": 1663 + }, + { + "epoch": 2.607515657620042, + "grad_norm": 1.118043303489685, + "learning_rate": 3.1253154309711663e-06, + "loss": 0.6948, + "step": 1664 + }, + { + "epoch": 2.609081419624217, + "grad_norm": 1.0427628755569458, + "learning_rate": 3.123275421335157e-06, + "loss": 0.7186, + "step": 1665 + }, + { + "epoch": 2.6106471816283925, + "grad_norm": 1.231054425239563, + "learning_rate": 3.1212349690823273e-06, + "loss": 0.6574, + "step": 1666 + }, + { + "epoch": 2.612212943632568, + "grad_norm": 1.0482161045074463, + "learning_rate": 3.1191940756616983e-06, + "loss": 0.7376, + "step": 1667 + }, + { + "epoch": 2.613778705636743, + "grad_norm": 1.2325738668441772, + "learning_rate": 3.1171527425226027e-06, + "loss": 0.6741, + "step": 1668 + }, + { + "epoch": 2.6153444676409183, + "grad_norm": 1.183379888534546, + "learning_rate": 3.1151109711146854e-06, + "loss": 0.7424, + "step": 1669 + }, + { + "epoch": 2.616910229645094, + "grad_norm": 1.1731956005096436, + "learning_rate": 3.1130687628879047e-06, + "loss": 0.7913, + "step": 1670 + }, + { + "epoch": 2.6184759916492695, + "grad_norm": 1.155126690864563, + "learning_rate": 3.111026119292526e-06, + "loss": 0.6724, + "step": 1671 + }, + { + "epoch": 2.6200417536534446, + "grad_norm": 1.0947262048721313, + "learning_rate": 3.1089830417791255e-06, + "loss": 0.6959, + "step": 1672 + }, + { + "epoch": 2.62160751565762, + "grad_norm": 1.073900580406189, + "learning_rate": 3.1069395317985884e-06, + "loss": 0.7274, + "step": 1673 + }, + { + "epoch": 2.6231732776617953, + "grad_norm": 1.1424334049224854, + "learning_rate": 3.104895590802105e-06, + "loss": 0.657, + "step": 1674 + }, + { + "epoch": 2.624739039665971, + "grad_norm": 1.100588321685791, + "learning_rate": 3.1028512202411736e-06, + "loss": 0.6451, + "step": 1675 + }, + { + "epoch": 2.626304801670146, + "grad_norm": 1.0387758016586304, + "learning_rate": 3.100806421567596e-06, + "loss": 0.7331, + "step": 1676 + }, + { + "epoch": 2.6278705636743216, + "grad_norm": 1.2019600868225098, + "learning_rate": 3.0987611962334784e-06, + "loss": 0.6224, + "step": 1677 + }, + { + "epoch": 2.6294363256784967, + "grad_norm": 1.0992989540100098, + "learning_rate": 3.0967155456912313e-06, + "loss": 0.6557, + "step": 1678 + }, + { + "epoch": 2.6310020876826723, + "grad_norm": 1.259507417678833, + "learning_rate": 3.0946694713935655e-06, + "loss": 0.6668, + "step": 1679 + }, + { + "epoch": 2.632567849686848, + "grad_norm": 1.1520506143569946, + "learning_rate": 3.0926229747934932e-06, + "loss": 0.7199, + "step": 1680 + }, + { + "epoch": 2.634133611691023, + "grad_norm": 1.0905909538269043, + "learning_rate": 3.0905760573443278e-06, + "loss": 0.6927, + "step": 1681 + }, + { + "epoch": 2.635699373695198, + "grad_norm": 1.1551228761672974, + "learning_rate": 3.0885287204996784e-06, + "loss": 0.6872, + "step": 1682 + }, + { + "epoch": 2.6372651356993737, + "grad_norm": 1.177915096282959, + "learning_rate": 3.0864809657134553e-06, + "loss": 0.6102, + "step": 1683 + }, + { + "epoch": 2.6388308977035493, + "grad_norm": 1.2892553806304932, + "learning_rate": 3.084432794439865e-06, + "loss": 0.6576, + "step": 1684 + }, + { + "epoch": 2.6403966597077244, + "grad_norm": 1.1799864768981934, + "learning_rate": 3.0823842081334075e-06, + "loss": 0.7214, + "step": 1685 + }, + { + "epoch": 2.6419624217118995, + "grad_norm": 1.11050283908844, + "learning_rate": 3.08033520824888e-06, + "loss": 0.6156, + "step": 1686 + }, + { + "epoch": 2.643528183716075, + "grad_norm": 1.15585196018219, + "learning_rate": 3.078285796241372e-06, + "loss": 0.6311, + "step": 1687 + }, + { + "epoch": 2.6450939457202507, + "grad_norm": 1.1588464975357056, + "learning_rate": 3.0762359735662663e-06, + "loss": 0.6804, + "step": 1688 + }, + { + "epoch": 2.646659707724426, + "grad_norm": 1.1857856512069702, + "learning_rate": 3.0741857416792374e-06, + "loss": 0.6737, + "step": 1689 + }, + { + "epoch": 2.6482254697286014, + "grad_norm": 1.1420296430587769, + "learning_rate": 3.0721351020362495e-06, + "loss": 0.6241, + "step": 1690 + }, + { + "epoch": 2.6497912317327765, + "grad_norm": 1.2236177921295166, + "learning_rate": 3.070084056093558e-06, + "loss": 0.7142, + "step": 1691 + }, + { + "epoch": 2.651356993736952, + "grad_norm": 1.1257219314575195, + "learning_rate": 3.0680326053077047e-06, + "loss": 0.5786, + "step": 1692 + }, + { + "epoch": 2.6529227557411272, + "grad_norm": 1.1258952617645264, + "learning_rate": 3.065980751135522e-06, + "loss": 0.7306, + "step": 1693 + }, + { + "epoch": 2.654488517745303, + "grad_norm": 1.1650607585906982, + "learning_rate": 3.0639284950341246e-06, + "loss": 0.6776, + "step": 1694 + }, + { + "epoch": 2.656054279749478, + "grad_norm": 1.1687644720077515, + "learning_rate": 3.0618758384609156e-06, + "loss": 0.6602, + "step": 1695 + }, + { + "epoch": 2.6576200417536535, + "grad_norm": 1.160516381263733, + "learning_rate": 3.0598227828735825e-06, + "loss": 0.6676, + "step": 1696 + }, + { + "epoch": 2.6591858037578286, + "grad_norm": 1.0864505767822266, + "learning_rate": 3.0577693297300947e-06, + "loss": 0.6843, + "step": 1697 + }, + { + "epoch": 2.6607515657620042, + "grad_norm": 1.0814275741577148, + "learning_rate": 3.055715480488706e-06, + "loss": 0.7214, + "step": 1698 + }, + { + "epoch": 2.6623173277661794, + "grad_norm": 1.1549509763717651, + "learning_rate": 3.0536612366079465e-06, + "loss": 0.7756, + "step": 1699 + }, + { + "epoch": 2.663883089770355, + "grad_norm": 1.125248908996582, + "learning_rate": 3.0516065995466336e-06, + "loss": 0.7133, + "step": 1700 + }, + { + "epoch": 2.6654488517745305, + "grad_norm": 1.0562750101089478, + "learning_rate": 3.0495515707638605e-06, + "loss": 0.6655, + "step": 1701 + }, + { + "epoch": 2.6670146137787056, + "grad_norm": 1.1991479396820068, + "learning_rate": 3.0474961517189965e-06, + "loss": 0.7382, + "step": 1702 + }, + { + "epoch": 2.6685803757828808, + "grad_norm": 1.0789530277252197, + "learning_rate": 3.0454403438716918e-06, + "loss": 0.6608, + "step": 1703 + }, + { + "epoch": 2.6701461377870563, + "grad_norm": 1.1776909828186035, + "learning_rate": 3.04338414868187e-06, + "loss": 0.7493, + "step": 1704 + }, + { + "epoch": 2.671711899791232, + "grad_norm": 1.1587637662887573, + "learning_rate": 3.041327567609731e-06, + "loss": 0.7042, + "step": 1705 + }, + { + "epoch": 2.673277661795407, + "grad_norm": 1.151100516319275, + "learning_rate": 3.0392706021157485e-06, + "loss": 0.5642, + "step": 1706 + }, + { + "epoch": 2.6748434237995826, + "grad_norm": 1.098358392715454, + "learning_rate": 3.03721325366067e-06, + "loss": 0.6177, + "step": 1707 + }, + { + "epoch": 2.6764091858037578, + "grad_norm": 1.1176345348358154, + "learning_rate": 3.0351555237055135e-06, + "loss": 0.7474, + "step": 1708 + }, + { + "epoch": 2.6779749478079333, + "grad_norm": 1.1730945110321045, + "learning_rate": 3.0330974137115683e-06, + "loss": 0.5899, + "step": 1709 + }, + { + "epoch": 2.6795407098121085, + "grad_norm": 1.1680086851119995, + "learning_rate": 3.031038925140395e-06, + "loss": 0.6576, + "step": 1710 + }, + { + "epoch": 2.681106471816284, + "grad_norm": 1.1488760709762573, + "learning_rate": 3.028980059453821e-06, + "loss": 0.6541, + "step": 1711 + }, + { + "epoch": 2.682672233820459, + "grad_norm": 1.1394031047821045, + "learning_rate": 3.0269208181139413e-06, + "loss": 0.6725, + "step": 1712 + }, + { + "epoch": 2.6842379958246347, + "grad_norm": 1.105613112449646, + "learning_rate": 3.024861202583122e-06, + "loss": 0.7088, + "step": 1713 + }, + { + "epoch": 2.68580375782881, + "grad_norm": 1.1166354417800903, + "learning_rate": 3.022801214323989e-06, + "loss": 0.7273, + "step": 1714 + }, + { + "epoch": 2.6873695198329854, + "grad_norm": 1.05818772315979, + "learning_rate": 3.020740854799436e-06, + "loss": 0.591, + "step": 1715 + }, + { + "epoch": 2.6889352818371606, + "grad_norm": 1.019797682762146, + "learning_rate": 3.0186801254726213e-06, + "loss": 0.6676, + "step": 1716 + }, + { + "epoch": 2.690501043841336, + "grad_norm": 1.1530799865722656, + "learning_rate": 3.016619027806963e-06, + "loss": 0.6181, + "step": 1717 + }, + { + "epoch": 2.6920668058455117, + "grad_norm": 1.132246971130371, + "learning_rate": 3.0145575632661426e-06, + "loss": 0.6932, + "step": 1718 + }, + { + "epoch": 2.693632567849687, + "grad_norm": 1.2640838623046875, + "learning_rate": 3.0124957333141036e-06, + "loss": 0.7029, + "step": 1719 + }, + { + "epoch": 2.695198329853862, + "grad_norm": 1.1203837394714355, + "learning_rate": 3.0104335394150447e-06, + "loss": 0.6522, + "step": 1720 + }, + { + "epoch": 2.6967640918580376, + "grad_norm": 1.1490654945373535, + "learning_rate": 3.008370983033426e-06, + "loss": 0.6837, + "step": 1721 + }, + { + "epoch": 2.698329853862213, + "grad_norm": 1.0226919651031494, + "learning_rate": 3.006308065633966e-06, + "loss": 0.6344, + "step": 1722 + }, + { + "epoch": 2.6998956158663883, + "grad_norm": 1.2632393836975098, + "learning_rate": 3.004244788681637e-06, + "loss": 0.7157, + "step": 1723 + }, + { + "epoch": 2.701461377870564, + "grad_norm": 1.0831620693206787, + "learning_rate": 3.0021811536416676e-06, + "loss": 0.6671, + "step": 1724 + }, + { + "epoch": 2.703027139874739, + "grad_norm": 1.1284756660461426, + "learning_rate": 3.0001171619795417e-06, + "loss": 0.6669, + "step": 1725 + }, + { + "epoch": 2.7045929018789145, + "grad_norm": 1.0045720338821411, + "learning_rate": 2.9980528151609945e-06, + "loss": 0.7276, + "step": 1726 + }, + { + "epoch": 2.7061586638830897, + "grad_norm": 1.1337859630584717, + "learning_rate": 2.9959881146520164e-06, + "loss": 0.6734, + "step": 1727 + }, + { + "epoch": 2.7077244258872653, + "grad_norm": 1.0121667385101318, + "learning_rate": 2.993923061918845e-06, + "loss": 0.6201, + "step": 1728 + }, + { + "epoch": 2.7092901878914404, + "grad_norm": 1.2311614751815796, + "learning_rate": 2.991857658427972e-06, + "loss": 0.6392, + "step": 1729 + }, + { + "epoch": 2.710855949895616, + "grad_norm": 1.0462316274642944, + "learning_rate": 2.989791905646135e-06, + "loss": 0.6736, + "step": 1730 + }, + { + "epoch": 2.712421711899791, + "grad_norm": 1.1915360689163208, + "learning_rate": 2.9877258050403214e-06, + "loss": 0.5977, + "step": 1731 + }, + { + "epoch": 2.7139874739039667, + "grad_norm": 1.0216567516326904, + "learning_rate": 2.985659358077765e-06, + "loss": 0.6393, + "step": 1732 + }, + { + "epoch": 2.715553235908142, + "grad_norm": 1.114071011543274, + "learning_rate": 2.9835925662259457e-06, + "loss": 0.6854, + "step": 1733 + }, + { + "epoch": 2.7171189979123174, + "grad_norm": 1.2007335424423218, + "learning_rate": 2.981525430952589e-06, + "loss": 0.584, + "step": 1734 + }, + { + "epoch": 2.718684759916493, + "grad_norm": 1.221422791481018, + "learning_rate": 2.979457953725663e-06, + "loss": 0.742, + "step": 1735 + }, + { + "epoch": 2.720250521920668, + "grad_norm": 1.0720511674880981, + "learning_rate": 2.9773901360133804e-06, + "loss": 0.6722, + "step": 1736 + }, + { + "epoch": 2.721816283924843, + "grad_norm": 1.1078705787658691, + "learning_rate": 2.975321979284193e-06, + "loss": 0.6837, + "step": 1737 + }, + { + "epoch": 2.723382045929019, + "grad_norm": 1.1647889614105225, + "learning_rate": 2.973253485006795e-06, + "loss": 0.7406, + "step": 1738 + }, + { + "epoch": 2.7249478079331944, + "grad_norm": 1.1482031345367432, + "learning_rate": 2.971184654650122e-06, + "loss": 0.7109, + "step": 1739 + }, + { + "epoch": 2.7265135699373695, + "grad_norm": 1.095131278038025, + "learning_rate": 2.9691154896833454e-06, + "loss": 0.6795, + "step": 1740 + }, + { + "epoch": 2.7280793319415446, + "grad_norm": 1.0917677879333496, + "learning_rate": 2.9670459915758746e-06, + "loss": 0.7142, + "step": 1741 + }, + { + "epoch": 2.72964509394572, + "grad_norm": 1.0911260843276978, + "learning_rate": 2.9649761617973572e-06, + "loss": 0.6917, + "step": 1742 + }, + { + "epoch": 2.7312108559498958, + "grad_norm": 1.2418321371078491, + "learning_rate": 2.9629060018176757e-06, + "loss": 0.6562, + "step": 1743 + }, + { + "epoch": 2.732776617954071, + "grad_norm": 1.114390254020691, + "learning_rate": 2.9608355131069464e-06, + "loss": 0.6812, + "step": 1744 + }, + { + "epoch": 2.7343423799582465, + "grad_norm": 1.0684012174606323, + "learning_rate": 2.9587646971355194e-06, + "loss": 0.688, + "step": 1745 + }, + { + "epoch": 2.7359081419624216, + "grad_norm": 1.1974529027938843, + "learning_rate": 2.9566935553739773e-06, + "loss": 0.6697, + "step": 1746 + }, + { + "epoch": 2.737473903966597, + "grad_norm": 1.0794286727905273, + "learning_rate": 2.9546220892931336e-06, + "loss": 0.6587, + "step": 1747 + }, + { + "epoch": 2.7390396659707723, + "grad_norm": 1.1143841743469238, + "learning_rate": 2.9525503003640336e-06, + "loss": 0.6814, + "step": 1748 + }, + { + "epoch": 2.740605427974948, + "grad_norm": 1.1970027685165405, + "learning_rate": 2.9504781900579504e-06, + "loss": 0.6714, + "step": 1749 + }, + { + "epoch": 2.742171189979123, + "grad_norm": 1.0299553871154785, + "learning_rate": 2.948405759846386e-06, + "loss": 0.6707, + "step": 1750 + }, + { + "epoch": 2.7437369519832986, + "grad_norm": 1.103028416633606, + "learning_rate": 2.946333011201068e-06, + "loss": 0.6163, + "step": 1751 + }, + { + "epoch": 2.745302713987474, + "grad_norm": 1.1230628490447998, + "learning_rate": 2.944259945593953e-06, + "loss": 0.5917, + "step": 1752 + }, + { + "epoch": 2.7468684759916493, + "grad_norm": 1.1009881496429443, + "learning_rate": 2.942186564497221e-06, + "loss": 0.7048, + "step": 1753 + }, + { + "epoch": 2.7484342379958244, + "grad_norm": 1.1123842000961304, + "learning_rate": 2.9401128693832752e-06, + "loss": 0.6063, + "step": 1754 + }, + { + "epoch": 2.75, + "grad_norm": 1.0923182964324951, + "learning_rate": 2.9380388617247433e-06, + "loss": 0.663, + "step": 1755 + }, + { + "epoch": 2.7515657620041756, + "grad_norm": 1.1761754751205444, + "learning_rate": 2.935964542994475e-06, + "loss": 0.6494, + "step": 1756 + }, + { + "epoch": 2.7531315240083507, + "grad_norm": 1.0580326318740845, + "learning_rate": 2.9338899146655396e-06, + "loss": 0.7391, + "step": 1757 + }, + { + "epoch": 2.754697286012526, + "grad_norm": 1.1655254364013672, + "learning_rate": 2.931814978211227e-06, + "loss": 0.6595, + "step": 1758 + }, + { + "epoch": 2.7562630480167014, + "grad_norm": 1.1162359714508057, + "learning_rate": 2.929739735105047e-06, + "loss": 0.6438, + "step": 1759 + }, + { + "epoch": 2.757828810020877, + "grad_norm": 1.173933744430542, + "learning_rate": 2.9276641868207244e-06, + "loss": 0.5591, + "step": 1760 + }, + { + "epoch": 2.759394572025052, + "grad_norm": 1.2183250188827515, + "learning_rate": 2.9255883348322038e-06, + "loss": 0.7317, + "step": 1761 + }, + { + "epoch": 2.7609603340292277, + "grad_norm": 1.0836644172668457, + "learning_rate": 2.9235121806136446e-06, + "loss": 0.7539, + "step": 1762 + }, + { + "epoch": 2.762526096033403, + "grad_norm": 1.1108825206756592, + "learning_rate": 2.921435725639419e-06, + "loss": 0.6598, + "step": 1763 + }, + { + "epoch": 2.7640918580375784, + "grad_norm": 1.090196967124939, + "learning_rate": 2.9193589713841132e-06, + "loss": 0.7412, + "step": 1764 + }, + { + "epoch": 2.7656576200417535, + "grad_norm": 1.1506152153015137, + "learning_rate": 2.9172819193225296e-06, + "loss": 0.6857, + "step": 1765 + }, + { + "epoch": 2.767223382045929, + "grad_norm": 1.127022385597229, + "learning_rate": 2.9152045709296766e-06, + "loss": 0.6383, + "step": 1766 + }, + { + "epoch": 2.7687891440501042, + "grad_norm": 1.1096911430358887, + "learning_rate": 2.9131269276807782e-06, + "loss": 0.6363, + "step": 1767 + }, + { + "epoch": 2.77035490605428, + "grad_norm": 1.1879534721374512, + "learning_rate": 2.911048991051263e-06, + "loss": 0.7313, + "step": 1768 + }, + { + "epoch": 2.7719206680584554, + "grad_norm": 1.0864057540893555, + "learning_rate": 2.908970762516772e-06, + "loss": 0.6602, + "step": 1769 + }, + { + "epoch": 2.7734864300626305, + "grad_norm": 1.174880027770996, + "learning_rate": 2.9068922435531515e-06, + "loss": 0.6731, + "step": 1770 + }, + { + "epoch": 2.7750521920668056, + "grad_norm": 1.1670523881912231, + "learning_rate": 2.904813435636453e-06, + "loss": 0.5982, + "step": 1771 + }, + { + "epoch": 2.776617954070981, + "grad_norm": 1.016558289527893, + "learning_rate": 2.902734340242937e-06, + "loss": 0.5549, + "step": 1772 + }, + { + "epoch": 2.778183716075157, + "grad_norm": 1.2644622325897217, + "learning_rate": 2.9006549588490636e-06, + "loss": 0.6209, + "step": 1773 + }, + { + "epoch": 2.779749478079332, + "grad_norm": 1.1689661741256714, + "learning_rate": 2.8985752929314985e-06, + "loss": 0.6629, + "step": 1774 + }, + { + "epoch": 2.781315240083507, + "grad_norm": 1.123618721961975, + "learning_rate": 2.8964953439671095e-06, + "loss": 0.6986, + "step": 1775 + }, + { + "epoch": 2.7828810020876826, + "grad_norm": 1.1276949644088745, + "learning_rate": 2.8944151134329646e-06, + "loss": 0.6125, + "step": 1776 + }, + { + "epoch": 2.784446764091858, + "grad_norm": 1.1804574728012085, + "learning_rate": 2.8923346028063328e-06, + "loss": 0.5912, + "step": 1777 + }, + { + "epoch": 2.7860125260960333, + "grad_norm": 1.1222927570343018, + "learning_rate": 2.89025381356468e-06, + "loss": 0.5837, + "step": 1778 + }, + { + "epoch": 2.787578288100209, + "grad_norm": 1.063511848449707, + "learning_rate": 2.888172747185673e-06, + "loss": 0.7243, + "step": 1779 + }, + { + "epoch": 2.789144050104384, + "grad_norm": 1.1183874607086182, + "learning_rate": 2.8860914051471722e-06, + "loss": 0.7204, + "step": 1780 + }, + { + "epoch": 2.7907098121085596, + "grad_norm": 1.135906457901001, + "learning_rate": 2.884009788927235e-06, + "loss": 0.7308, + "step": 1781 + }, + { + "epoch": 2.7922755741127347, + "grad_norm": 1.2013477087020874, + "learning_rate": 2.8819279000041156e-06, + "loss": 0.7004, + "step": 1782 + }, + { + "epoch": 2.7938413361169103, + "grad_norm": 1.074418306350708, + "learning_rate": 2.8798457398562584e-06, + "loss": 0.6404, + "step": 1783 + }, + { + "epoch": 2.7954070981210855, + "grad_norm": 1.0827600955963135, + "learning_rate": 2.8777633099623027e-06, + "loss": 0.7488, + "step": 1784 + }, + { + "epoch": 2.796972860125261, + "grad_norm": 1.1087040901184082, + "learning_rate": 2.8756806118010784e-06, + "loss": 0.6554, + "step": 1785 + }, + { + "epoch": 2.798538622129436, + "grad_norm": 1.0850439071655273, + "learning_rate": 2.8735976468516057e-06, + "loss": 0.6501, + "step": 1786 + }, + { + "epoch": 2.8001043841336117, + "grad_norm": 1.129045844078064, + "learning_rate": 2.871514416593096e-06, + "loss": 0.6697, + "step": 1787 + }, + { + "epoch": 2.801670146137787, + "grad_norm": 1.0854123830795288, + "learning_rate": 2.869430922504947e-06, + "loss": 0.6787, + "step": 1788 + }, + { + "epoch": 2.8032359081419624, + "grad_norm": 1.2950549125671387, + "learning_rate": 2.8673471660667448e-06, + "loss": 0.6682, + "step": 1789 + }, + { + "epoch": 2.804801670146138, + "grad_norm": 1.0760269165039062, + "learning_rate": 2.865263148758261e-06, + "loss": 0.6846, + "step": 1790 + }, + { + "epoch": 2.806367432150313, + "grad_norm": 1.1861540079116821, + "learning_rate": 2.863178872059454e-06, + "loss": 0.6029, + "step": 1791 + }, + { + "epoch": 2.8079331941544883, + "grad_norm": 1.081518292427063, + "learning_rate": 2.8610943374504645e-06, + "loss": 0.6801, + "step": 1792 + }, + { + "epoch": 2.809498956158664, + "grad_norm": 1.1814477443695068, + "learning_rate": 2.8590095464116192e-06, + "loss": 0.68, + "step": 1793 + }, + { + "epoch": 2.8110647181628394, + "grad_norm": 1.154828667640686, + "learning_rate": 2.856924500423422e-06, + "loss": 0.6164, + "step": 1794 + }, + { + "epoch": 2.8126304801670146, + "grad_norm": 1.0605709552764893, + "learning_rate": 2.8548392009665638e-06, + "loss": 0.6146, + "step": 1795 + }, + { + "epoch": 2.81419624217119, + "grad_norm": 1.1769119501113892, + "learning_rate": 2.852753649521911e-06, + "loss": 0.6789, + "step": 1796 + }, + { + "epoch": 2.8157620041753653, + "grad_norm": 1.006847858428955, + "learning_rate": 2.8506678475705114e-06, + "loss": 0.6719, + "step": 1797 + }, + { + "epoch": 2.817327766179541, + "grad_norm": 1.1305062770843506, + "learning_rate": 2.8485817965935886e-06, + "loss": 0.6548, + "step": 1798 + }, + { + "epoch": 2.818893528183716, + "grad_norm": 1.0815331935882568, + "learning_rate": 2.846495498072545e-06, + "loss": 0.7306, + "step": 1799 + }, + { + "epoch": 2.8204592901878915, + "grad_norm": 1.1019850969314575, + "learning_rate": 2.8444089534889587e-06, + "loss": 0.7188, + "step": 1800 + }, + { + "epoch": 2.8220250521920667, + "grad_norm": 1.1405457258224487, + "learning_rate": 2.8423221643245807e-06, + "loss": 0.6482, + "step": 1801 + }, + { + "epoch": 2.8235908141962422, + "grad_norm": 1.1295759677886963, + "learning_rate": 2.8402351320613374e-06, + "loss": 0.5882, + "step": 1802 + }, + { + "epoch": 2.8251565762004174, + "grad_norm": 1.1165974140167236, + "learning_rate": 2.8381478581813275e-06, + "loss": 0.6714, + "step": 1803 + }, + { + "epoch": 2.826722338204593, + "grad_norm": 1.0910004377365112, + "learning_rate": 2.836060344166821e-06, + "loss": 0.6971, + "step": 1804 + }, + { + "epoch": 2.828288100208768, + "grad_norm": 1.093596339225769, + "learning_rate": 2.833972591500259e-06, + "loss": 0.7041, + "step": 1805 + }, + { + "epoch": 2.8298538622129437, + "grad_norm": 1.1507813930511475, + "learning_rate": 2.8318846016642507e-06, + "loss": 0.6455, + "step": 1806 + }, + { + "epoch": 2.8314196242171192, + "grad_norm": 1.055422306060791, + "learning_rate": 2.8297963761415748e-06, + "loss": 0.7499, + "step": 1807 + }, + { + "epoch": 2.8329853862212944, + "grad_norm": 1.0835453271865845, + "learning_rate": 2.8277079164151782e-06, + "loss": 0.6739, + "step": 1808 + }, + { + "epoch": 2.8345511482254695, + "grad_norm": 1.247808575630188, + "learning_rate": 2.8256192239681723e-06, + "loss": 0.6774, + "step": 1809 + }, + { + "epoch": 2.836116910229645, + "grad_norm": 1.0836130380630493, + "learning_rate": 2.8235303002838343e-06, + "loss": 0.6592, + "step": 1810 + }, + { + "epoch": 2.8376826722338206, + "grad_norm": 1.199852466583252, + "learning_rate": 2.8214411468456067e-06, + "loss": 0.6728, + "step": 1811 + }, + { + "epoch": 2.8392484342379958, + "grad_norm": 1.0706546306610107, + "learning_rate": 2.8193517651370934e-06, + "loss": 0.7438, + "step": 1812 + }, + { + "epoch": 2.8408141962421714, + "grad_norm": 1.301465630531311, + "learning_rate": 2.8172621566420633e-06, + "loss": 0.6736, + "step": 1813 + }, + { + "epoch": 2.8423799582463465, + "grad_norm": 1.0916422605514526, + "learning_rate": 2.815172322844442e-06, + "loss": 0.7258, + "step": 1814 + }, + { + "epoch": 2.843945720250522, + "grad_norm": 1.1184017658233643, + "learning_rate": 2.8130822652283183e-06, + "loss": 0.6831, + "step": 1815 + }, + { + "epoch": 2.845511482254697, + "grad_norm": 1.1714346408843994, + "learning_rate": 2.8109919852779406e-06, + "loss": 0.6866, + "step": 1816 + }, + { + "epoch": 2.8470772442588728, + "grad_norm": 1.0131067037582397, + "learning_rate": 2.808901484477712e-06, + "loss": 0.6574, + "step": 1817 + }, + { + "epoch": 2.848643006263048, + "grad_norm": 1.2358931303024292, + "learning_rate": 2.8068107643121952e-06, + "loss": 0.5596, + "step": 1818 + }, + { + "epoch": 2.8502087682672235, + "grad_norm": 0.9713279604911804, + "learning_rate": 2.8047198262661074e-06, + "loss": 0.6502, + "step": 1819 + }, + { + "epoch": 2.8517745302713986, + "grad_norm": 1.0890235900878906, + "learning_rate": 2.80262867182432e-06, + "loss": 0.6566, + "step": 1820 + }, + { + "epoch": 2.853340292275574, + "grad_norm": 1.084876298904419, + "learning_rate": 2.8005373024718597e-06, + "loss": 0.6356, + "step": 1821 + }, + { + "epoch": 2.8549060542797493, + "grad_norm": 1.1338250637054443, + "learning_rate": 2.7984457196939056e-06, + "loss": 0.7226, + "step": 1822 + }, + { + "epoch": 2.856471816283925, + "grad_norm": 1.1295956373214722, + "learning_rate": 2.7963539249757872e-06, + "loss": 0.724, + "step": 1823 + }, + { + "epoch": 2.8580375782881005, + "grad_norm": 1.0954629182815552, + "learning_rate": 2.794261919802984e-06, + "loss": 0.685, + "step": 1824 + }, + { + "epoch": 2.8596033402922756, + "grad_norm": 1.0435383319854736, + "learning_rate": 2.792169705661128e-06, + "loss": 0.7532, + "step": 1825 + }, + { + "epoch": 2.8611691022964507, + "grad_norm": 1.1580495834350586, + "learning_rate": 2.790077284035996e-06, + "loss": 0.6745, + "step": 1826 + }, + { + "epoch": 2.8627348643006263, + "grad_norm": 1.0154657363891602, + "learning_rate": 2.787984656413514e-06, + "loss": 0.7312, + "step": 1827 + }, + { + "epoch": 2.864300626304802, + "grad_norm": 1.1673710346221924, + "learning_rate": 2.785891824279755e-06, + "loss": 0.6769, + "step": 1828 + }, + { + "epoch": 2.865866388308977, + "grad_norm": 1.1436103582382202, + "learning_rate": 2.7837987891209352e-06, + "loss": 0.6987, + "step": 1829 + }, + { + "epoch": 2.867432150313152, + "grad_norm": 1.2176343202590942, + "learning_rate": 2.7817055524234167e-06, + "loss": 0.6768, + "step": 1830 + }, + { + "epoch": 2.8689979123173277, + "grad_norm": 1.1192113161087036, + "learning_rate": 2.7796121156737042e-06, + "loss": 0.6468, + "step": 1831 + }, + { + "epoch": 2.8705636743215033, + "grad_norm": 1.085511326789856, + "learning_rate": 2.777518480358444e-06, + "loss": 0.6923, + "step": 1832 + }, + { + "epoch": 2.8721294363256784, + "grad_norm": 1.1534816026687622, + "learning_rate": 2.7754246479644227e-06, + "loss": 0.7897, + "step": 1833 + }, + { + "epoch": 2.873695198329854, + "grad_norm": 1.2280727624893188, + "learning_rate": 2.7733306199785704e-06, + "loss": 0.5913, + "step": 1834 + }, + { + "epoch": 2.875260960334029, + "grad_norm": 1.1843805313110352, + "learning_rate": 2.771236397887952e-06, + "loss": 0.6554, + "step": 1835 + }, + { + "epoch": 2.8768267223382047, + "grad_norm": 1.1090878248214722, + "learning_rate": 2.7691419831797724e-06, + "loss": 0.7155, + "step": 1836 + }, + { + "epoch": 2.87839248434238, + "grad_norm": 1.3048434257507324, + "learning_rate": 2.767047377341371e-06, + "loss": 0.689, + "step": 1837 + }, + { + "epoch": 2.8799582463465554, + "grad_norm": 1.0179991722106934, + "learning_rate": 2.764952581860227e-06, + "loss": 0.6919, + "step": 1838 + }, + { + "epoch": 2.8815240083507305, + "grad_norm": 1.0855897665023804, + "learning_rate": 2.7628575982239513e-06, + "loss": 0.7377, + "step": 1839 + }, + { + "epoch": 2.883089770354906, + "grad_norm": 1.0885690450668335, + "learning_rate": 2.7607624279202887e-06, + "loss": 0.7436, + "step": 1840 + }, + { + "epoch": 2.8846555323590817, + "grad_norm": 0.9842485189437866, + "learning_rate": 2.758667072437117e-06, + "loss": 0.6818, + "step": 1841 + }, + { + "epoch": 2.886221294363257, + "grad_norm": 1.2647449970245361, + "learning_rate": 2.7565715332624455e-06, + "loss": 0.6892, + "step": 1842 + }, + { + "epoch": 2.887787056367432, + "grad_norm": 1.0584148168563843, + "learning_rate": 2.7544758118844134e-06, + "loss": 0.6945, + "step": 1843 + }, + { + "epoch": 2.8893528183716075, + "grad_norm": 1.2626157999038696, + "learning_rate": 2.7523799097912905e-06, + "loss": 0.6617, + "step": 1844 + }, + { + "epoch": 2.890918580375783, + "grad_norm": 1.1559371948242188, + "learning_rate": 2.7502838284714746e-06, + "loss": 0.679, + "step": 1845 + }, + { + "epoch": 2.892484342379958, + "grad_norm": 1.1553645133972168, + "learning_rate": 2.7481875694134884e-06, + "loss": 0.7216, + "step": 1846 + }, + { + "epoch": 2.8940501043841333, + "grad_norm": 1.1300830841064453, + "learning_rate": 2.7460911341059847e-06, + "loss": 0.6761, + "step": 1847 + }, + { + "epoch": 2.895615866388309, + "grad_norm": 1.1680023670196533, + "learning_rate": 2.7439945240377395e-06, + "loss": 0.675, + "step": 1848 + }, + { + "epoch": 2.8971816283924845, + "grad_norm": 1.0944335460662842, + "learning_rate": 2.7418977406976526e-06, + "loss": 0.6959, + "step": 1849 + }, + { + "epoch": 2.8987473903966596, + "grad_norm": 1.1949477195739746, + "learning_rate": 2.7398007855747455e-06, + "loss": 0.6822, + "step": 1850 + }, + { + "epoch": 2.900313152400835, + "grad_norm": 1.2108348608016968, + "learning_rate": 2.737703660158167e-06, + "loss": 0.6568, + "step": 1851 + }, + { + "epoch": 2.9018789144050103, + "grad_norm": 1.119547963142395, + "learning_rate": 2.73560636593718e-06, + "loss": 0.6508, + "step": 1852 + }, + { + "epoch": 2.903444676409186, + "grad_norm": 1.1591404676437378, + "learning_rate": 2.733508904401172e-06, + "loss": 0.6462, + "step": 1853 + }, + { + "epoch": 2.905010438413361, + "grad_norm": 1.1378757953643799, + "learning_rate": 2.7314112770396474e-06, + "loss": 0.6202, + "step": 1854 + }, + { + "epoch": 2.9065762004175366, + "grad_norm": 1.1146254539489746, + "learning_rate": 2.729313485342229e-06, + "loss": 0.6618, + "step": 1855 + }, + { + "epoch": 2.9081419624217117, + "grad_norm": 1.1829131841659546, + "learning_rate": 2.727215530798656e-06, + "loss": 0.6708, + "step": 1856 + }, + { + "epoch": 2.9097077244258873, + "grad_norm": 1.0372153520584106, + "learning_rate": 2.725117414898784e-06, + "loss": 0.6666, + "step": 1857 + }, + { + "epoch": 2.911273486430063, + "grad_norm": 1.1896759271621704, + "learning_rate": 2.723019139132581e-06, + "loss": 0.7169, + "step": 1858 + }, + { + "epoch": 2.912839248434238, + "grad_norm": 1.137988805770874, + "learning_rate": 2.720920704990131e-06, + "loss": 0.6872, + "step": 1859 + }, + { + "epoch": 2.914405010438413, + "grad_norm": 1.1891788244247437, + "learning_rate": 2.7188221139616303e-06, + "loss": 0.7105, + "step": 1860 + }, + { + "epoch": 2.9159707724425887, + "grad_norm": 1.0640588998794556, + "learning_rate": 2.716723367537385e-06, + "loss": 0.6905, + "step": 1861 + }, + { + "epoch": 2.9175365344467643, + "grad_norm": 1.2714836597442627, + "learning_rate": 2.7146244672078125e-06, + "loss": 0.6713, + "step": 1862 + }, + { + "epoch": 2.9191022964509394, + "grad_norm": 1.222401738166809, + "learning_rate": 2.7125254144634388e-06, + "loss": 0.6638, + "step": 1863 + }, + { + "epoch": 2.9206680584551146, + "grad_norm": 1.0254578590393066, + "learning_rate": 2.7104262107948993e-06, + "loss": 0.7374, + "step": 1864 + }, + { + "epoch": 2.92223382045929, + "grad_norm": 1.089826226234436, + "learning_rate": 2.7083268576929368e-06, + "loss": 0.6742, + "step": 1865 + }, + { + "epoch": 2.9237995824634657, + "grad_norm": 1.1508773565292358, + "learning_rate": 2.7062273566483984e-06, + "loss": 0.5628, + "step": 1866 + }, + { + "epoch": 2.925365344467641, + "grad_norm": 1.1214534044265747, + "learning_rate": 2.7041277091522365e-06, + "loss": 0.5819, + "step": 1867 + }, + { + "epoch": 2.9269311064718164, + "grad_norm": 1.2162528038024902, + "learning_rate": 2.70202791669551e-06, + "loss": 0.6774, + "step": 1868 + }, + { + "epoch": 2.9284968684759916, + "grad_norm": 1.130321741104126, + "learning_rate": 2.6999279807693783e-06, + "loss": 0.6617, + "step": 1869 + }, + { + "epoch": 2.930062630480167, + "grad_norm": 1.1052041053771973, + "learning_rate": 2.697827902865103e-06, + "loss": 0.6348, + "step": 1870 + }, + { + "epoch": 2.9316283924843423, + "grad_norm": 1.243033766746521, + "learning_rate": 2.695727684474048e-06, + "loss": 0.671, + "step": 1871 + }, + { + "epoch": 2.933194154488518, + "grad_norm": 1.163845419883728, + "learning_rate": 2.693627327087675e-06, + "loss": 0.6605, + "step": 1872 + }, + { + "epoch": 2.934759916492693, + "grad_norm": 1.0230295658111572, + "learning_rate": 2.6915268321975454e-06, + "loss": 0.6511, + "step": 1873 + }, + { + "epoch": 2.9363256784968685, + "grad_norm": 1.0699726343154907, + "learning_rate": 2.6894262012953194e-06, + "loss": 0.5641, + "step": 1874 + }, + { + "epoch": 2.937891440501044, + "grad_norm": 1.139090657234192, + "learning_rate": 2.6873254358727503e-06, + "loss": 0.7327, + "step": 1875 + }, + { + "epoch": 2.9394572025052192, + "grad_norm": 1.0480340719223022, + "learning_rate": 2.68522453742169e-06, + "loss": 0.6559, + "step": 1876 + }, + { + "epoch": 2.9410229645093944, + "grad_norm": 1.1894673109054565, + "learning_rate": 2.6831235074340854e-06, + "loss": 0.6235, + "step": 1877 + }, + { + "epoch": 2.94258872651357, + "grad_norm": 1.1843831539154053, + "learning_rate": 2.681022347401975e-06, + "loss": 0.7161, + "step": 1878 + }, + { + "epoch": 2.9441544885177455, + "grad_norm": 1.278631329536438, + "learning_rate": 2.6789210588174887e-06, + "loss": 0.6651, + "step": 1879 + }, + { + "epoch": 2.9457202505219207, + "grad_norm": 1.1330087184906006, + "learning_rate": 2.67681964317285e-06, + "loss": 0.682, + "step": 1880 + }, + { + "epoch": 2.947286012526096, + "grad_norm": 1.1718090772628784, + "learning_rate": 2.6747181019603718e-06, + "loss": 0.6656, + "step": 1881 + }, + { + "epoch": 2.9488517745302714, + "grad_norm": 1.0476222038269043, + "learning_rate": 2.672616436672456e-06, + "loss": 0.543, + "step": 1882 + }, + { + "epoch": 2.950417536534447, + "grad_norm": 1.1602777242660522, + "learning_rate": 2.6705146488015933e-06, + "loss": 0.6887, + "step": 1883 + }, + { + "epoch": 2.951983298538622, + "grad_norm": 1.1091994047164917, + "learning_rate": 2.66841273984036e-06, + "loss": 0.7068, + "step": 1884 + }, + { + "epoch": 2.9535490605427976, + "grad_norm": 1.1577891111373901, + "learning_rate": 2.666310711281419e-06, + "loss": 0.7285, + "step": 1885 + }, + { + "epoch": 2.9551148225469728, + "grad_norm": 1.1392951011657715, + "learning_rate": 2.6642085646175196e-06, + "loss": 0.6308, + "step": 1886 + }, + { + "epoch": 2.9566805845511483, + "grad_norm": 1.3026043176651, + "learning_rate": 2.6621063013414934e-06, + "loss": 0.7092, + "step": 1887 + }, + { + "epoch": 2.9582463465553235, + "grad_norm": 1.0123710632324219, + "learning_rate": 2.660003922946255e-06, + "loss": 0.7045, + "step": 1888 + }, + { + "epoch": 2.959812108559499, + "grad_norm": 1.0754269361495972, + "learning_rate": 2.6579014309247998e-06, + "loss": 0.7301, + "step": 1889 + }, + { + "epoch": 2.961377870563674, + "grad_norm": 1.19624662399292, + "learning_rate": 2.655798826770206e-06, + "loss": 0.699, + "step": 1890 + }, + { + "epoch": 2.9629436325678498, + "grad_norm": 1.0728397369384766, + "learning_rate": 2.6536961119756317e-06, + "loss": 0.7188, + "step": 1891 + }, + { + "epoch": 2.964509394572025, + "grad_norm": 1.231490135192871, + "learning_rate": 2.6515932880343103e-06, + "loss": 0.7309, + "step": 1892 + }, + { + "epoch": 2.9660751565762005, + "grad_norm": 1.2304288148880005, + "learning_rate": 2.6494903564395563e-06, + "loss": 0.7267, + "step": 1893 + }, + { + "epoch": 2.9676409185803756, + "grad_norm": 1.0533643960952759, + "learning_rate": 2.6473873186847577e-06, + "loss": 0.7011, + "step": 1894 + }, + { + "epoch": 2.969206680584551, + "grad_norm": 1.1855214834213257, + "learning_rate": 2.6452841762633795e-06, + "loss": 0.707, + "step": 1895 + }, + { + "epoch": 2.9707724425887267, + "grad_norm": 1.1279842853546143, + "learning_rate": 2.643180930668962e-06, + "loss": 0.6033, + "step": 1896 + }, + { + "epoch": 2.972338204592902, + "grad_norm": 1.102318525314331, + "learning_rate": 2.6410775833951164e-06, + "loss": 0.6696, + "step": 1897 + }, + { + "epoch": 2.973903966597077, + "grad_norm": 1.1372147798538208, + "learning_rate": 2.6389741359355275e-06, + "loss": 0.7164, + "step": 1898 + }, + { + "epoch": 2.9754697286012526, + "grad_norm": 1.2800220251083374, + "learning_rate": 2.6368705897839515e-06, + "loss": 0.733, + "step": 1899 + }, + { + "epoch": 2.977035490605428, + "grad_norm": 1.0855633020401, + "learning_rate": 2.634766946434214e-06, + "loss": 0.5894, + "step": 1900 + }, + { + "epoch": 2.9786012526096033, + "grad_norm": 1.1212704181671143, + "learning_rate": 2.6326632073802092e-06, + "loss": 0.6753, + "step": 1901 + }, + { + "epoch": 2.980167014613779, + "grad_norm": 1.0885487794876099, + "learning_rate": 2.6305593741159004e-06, + "loss": 0.6802, + "step": 1902 + }, + { + "epoch": 2.981732776617954, + "grad_norm": 1.1605111360549927, + "learning_rate": 2.6284554481353182e-06, + "loss": 0.5837, + "step": 1903 + }, + { + "epoch": 2.9832985386221296, + "grad_norm": 1.1119557619094849, + "learning_rate": 2.626351430932556e-06, + "loss": 0.6767, + "step": 1904 + }, + { + "epoch": 2.9848643006263047, + "grad_norm": 1.1678719520568848, + "learning_rate": 2.6242473240017757e-06, + "loss": 0.7052, + "step": 1905 + }, + { + "epoch": 2.9864300626304803, + "grad_norm": 1.1222507953643799, + "learning_rate": 2.622143128837201e-06, + "loss": 0.6751, + "step": 1906 + }, + { + "epoch": 2.9879958246346554, + "grad_norm": 1.1091363430023193, + "learning_rate": 2.620038846933119e-06, + "loss": 0.6427, + "step": 1907 + }, + { + "epoch": 2.989561586638831, + "grad_norm": 1.0949740409851074, + "learning_rate": 2.6179344797838775e-06, + "loss": 0.6353, + "step": 1908 + }, + { + "epoch": 2.991127348643006, + "grad_norm": 1.2391382455825806, + "learning_rate": 2.615830028883885e-06, + "loss": 0.7133, + "step": 1909 + }, + { + "epoch": 2.9926931106471817, + "grad_norm": 1.1227985620498657, + "learning_rate": 2.6137254957276105e-06, + "loss": 0.6942, + "step": 1910 + }, + { + "epoch": 2.994258872651357, + "grad_norm": 1.2233500480651855, + "learning_rate": 2.61162088180958e-06, + "loss": 0.7006, + "step": 1911 + }, + { + "epoch": 2.9958246346555324, + "grad_norm": 1.1352185010910034, + "learning_rate": 2.609516188624378e-06, + "loss": 0.6304, + "step": 1912 + }, + { + "epoch": 2.997390396659708, + "grad_norm": 1.2024694681167603, + "learning_rate": 2.6074114176666455e-06, + "loss": 0.6911, + "step": 1913 + }, + { + "epoch": 2.998956158663883, + "grad_norm": 1.1703287363052368, + "learning_rate": 2.605306570431077e-06, + "loss": 0.6528, + "step": 1914 + } + ], + "logging_steps": 1, + "max_steps": 3828, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 638, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 4.836518262063759e+18, + "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..07c8c9aae2a66d9fb6a8b65759b1cd86d2f7d08b --- /dev/null +++ b/checkpoint-1914/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81aeca374ba1e1a5384aca380c5dd9da735cc6d6164e0b5b0681e227cf61f41c +size 7992 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-2552/README.md b/checkpoint-2552/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f07164552af92bf3072e4e4326b5e572dea01ae8 --- /dev/null +++ b/checkpoint-2552/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-Nano-8B-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 0.14.0 \ No newline at end of file diff --git a/checkpoint-2552/adapter_config.json b/checkpoint-2552/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3db75f3ecbeb5f4990760e56c7957796d83680c5 --- /dev/null +++ b/checkpoint-2552/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1", + "bias": "none", + "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": [ + "q_proj", + "down_proj", + "o_proj", + "v_proj", + "k_proj", + "gate_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-2552/adapter_model.safetensors b/checkpoint-2552/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..26d03711defe5998cc94091509278b2801f58c3b --- /dev/null +++ b/checkpoint-2552/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:547e38ea8faf6d21fc7c1a875c311e9bedc35091a2505980b9c2970ded19d0d3 +size 3443586272 diff --git a/checkpoint-2552/global_step2553/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-2552/global_step2553/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..ada7bed7f41fb2759931cd654e30aa6d69b3f1f7 --- /dev/null +++ b/checkpoint-2552/global_step2553/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:98662b90c78a75896fd7e24e07a080368630bea2b63febcefb6ef2657c93d2ee +size 20661195036 diff --git a/checkpoint-2552/global_step2553/mp_rank_00_model_states.pt b/checkpoint-2552/global_step2553/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..35247d29c8fd252cb3b3d744761f7504948069de --- /dev/null +++ b/checkpoint-2552/global_step2553/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1064c179cef74d8b87b412ae8d48db0414caaf69c43444bcd1a317f4afea3793 +size 3555326777 diff --git a/checkpoint-2552/latest b/checkpoint-2552/latest new file mode 100644 index 0000000000000000000000000000000000000000..44bf527a63224167555f3c0574e0919486ed2ebb --- /dev/null +++ b/checkpoint-2552/latest @@ -0,0 +1 @@ +global_step2553 \ No newline at end of file diff --git a/checkpoint-2552/rng_state.pth b/checkpoint-2552/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..96048c8e9bb369dceb2a89c9a5a9e7614eb1eaef --- /dev/null +++ b/checkpoint-2552/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dd18c51e587e9f54f414a5031eeb17d6d52945d6e1b95a09e82f751e35e16968 +size 14244 diff --git a/checkpoint-2552/scheduler.pt b/checkpoint-2552/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..0adc54f6cc85ad8e0ac70a6681e8be79a6b24e59 --- /dev/null +++ b/checkpoint-2552/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a7c821777236c94d8cf7be5d1ae669fa4ac91c7b0c50c7bcd7c6f70f723ea74 +size 1064 diff --git a/checkpoint-2552/special_tokens_map.json b/checkpoint-2552/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-2552/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-2552/tokenizer.json b/checkpoint-2552/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-2552/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-2552/tokenizer_config.json b/checkpoint-2552/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b480d149dc274c0a23db079bfd38032d90082e49 --- /dev/null +++ b/checkpoint-2552/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": "{%- if messages[0]['role'] == 'system' -%}{%- set system_message = messages[0]['content'] | trim -%}{%- set messages = messages[1:] -%}{%- else -%}{%- set system_message = '' -%}{%- endif -%}{%- if tools is not none -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message -}} {{- '\n\n' if system_message else '' -}} {{- '[' -}} {% for t in tools %}{{- (t.function if t.function is defined else t) | tojson() -}}{{- ', ' if not loop.last else '' -}}{%- endfor -%} {{- ']' -}} {{- '<|eot_id|>' -}}{%- else -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message + '<|eot_id|>' -}}{%- endif -%}{%- for message in messages -%}{%- if (message['role'] in ['user', 'tool']) != (loop.index0 % 2 == 0) -%}{{- raise_exception('Conversation roles must alternate between user/tool and assistant') -}}{%- elif message['role'] == 'user' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- elif message['role'] == 'tool' -%}{%- set tool_response = '[' + message['content'] | trim + ']' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + tool_response + '<|eot_id|>' -}}{%- elif message['role'] == 'assistant' and message.get('tool_calls') is not none -%}{%- set tool_calls = message['tool_calls'] -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + '[' -}}{%- for tool_call in tool_calls -%}{{ '{' + '\"name\": \"' + tool_call.function.name + '\", \"arguments\": ' + tool_call.function.arguments | tojson + '}' }}{%- if not loop.last -%}{{ ', ' }}{%- else -%}{{ ']' + '<|eot_id|>' }}{%- endif -%}{%- endfor -%}{%- elif message['role'] == 'assistant' -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- endif -%}{%- 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-2552/trainer_state.json b/checkpoint-2552/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..e0d517155b51cde509992970d73bea2f0d114a3f --- /dev/null +++ b/checkpoint-2552/trainer_state.json @@ -0,0 +1,17898 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 3.998956158663883, + "eval_steps": 500, + "global_step": 2552, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0015657620041753654, + "grad_norm": 658.1201782226562, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.3039, + "step": 1 + }, + { + "epoch": 0.003131524008350731, + "grad_norm": 749.5796508789062, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.4164, + "step": 2 + }, + { + "epoch": 0.004697286012526096, + "grad_norm": 394.2350769042969, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.3367, + "step": 3 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 384.08319091796875, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.3192, + "step": 4 + }, + { + "epoch": 0.007828810020876827, + "grad_norm": 376.9790954589844, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.2756, + "step": 5 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 175.25262451171875, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.3992, + "step": 6 + }, + { + "epoch": 0.010960334029227558, + "grad_norm": 100.1114730834961, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.2348, + "step": 7 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 131.44432067871094, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.2685, + "step": 8 + }, + { + "epoch": 0.014091858037578288, + "grad_norm": 67.03341674804688, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.3272, + "step": 9 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 32.0869140625, + "learning_rate": 5.000000000000001e-07, + "loss": 1.3458, + "step": 10 + }, + { + "epoch": 0.01722338204592902, + "grad_norm": 17.701913833618164, + "learning_rate": 5.5e-07, + "loss": 1.3608, + "step": 11 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 9.164765357971191, + "learning_rate": 6.000000000000001e-07, + "loss": 1.2961, + "step": 12 + }, + { + "epoch": 0.02035490605427975, + "grad_norm": 4.315349578857422, + "learning_rate": 6.5e-07, + "loss": 1.2069, + "step": 13 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 6.652385234832764, + "learning_rate": 7.000000000000001e-07, + "loss": 1.2588, + "step": 14 + }, + { + "epoch": 0.02348643006263048, + "grad_norm": 3.3322548866271973, + "learning_rate": 7.5e-07, + "loss": 1.2505, + "step": 15 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 9.599331855773926, + "learning_rate": 8.000000000000001e-07, + "loss": 1.1343, + "step": 16 + }, + { + "epoch": 0.02661795407098121, + "grad_norm": 10.33918285369873, + "learning_rate": 8.500000000000001e-07, + "loss": 1.3334, + "step": 17 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 9.453723907470703, + "learning_rate": 9.000000000000001e-07, + "loss": 1.3099, + "step": 18 + }, + { + "epoch": 0.029749478079331943, + "grad_norm": 7.075769901275635, + "learning_rate": 9.500000000000001e-07, + "loss": 1.1121, + "step": 19 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 7.770864009857178, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.1836, + "step": 20 + }, + { + "epoch": 0.03288100208768267, + "grad_norm": 9.582944869995117, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.1939, + "step": 21 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 9.496583938598633, + "learning_rate": 1.1e-06, + "loss": 1.2395, + "step": 22 + }, + { + "epoch": 0.0360125260960334, + "grad_norm": 5.912876129150391, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.1769, + "step": 23 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 2.6761083602905273, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.204, + "step": 24 + }, + { + "epoch": 0.03914405010438413, + "grad_norm": 11.136463165283203, + "learning_rate": 1.25e-06, + "loss": 1.2271, + "step": 25 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 23.912893295288086, + "learning_rate": 1.3e-06, + "loss": 1.1771, + "step": 26 + }, + { + "epoch": 0.042275574112734866, + "grad_norm": 57.08906555175781, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.2253, + "step": 27 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 36.11408615112305, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.1014, + "step": 28 + }, + { + "epoch": 0.04540709812108559, + "grad_norm": 51.39445495605469, + "learning_rate": 1.45e-06, + "loss": 1.1978, + "step": 29 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 27.394500732421875, + "learning_rate": 1.5e-06, + "loss": 1.1155, + "step": 30 + }, + { + "epoch": 0.048538622129436326, + "grad_norm": 6.844716548919678, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.1798, + "step": 31 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 2.397536277770996, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.1846, + "step": 32 + }, + { + "epoch": 0.05167014613778706, + "grad_norm": 7.821156024932861, + "learning_rate": 1.6500000000000003e-06, + "loss": 1.1312, + "step": 33 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 5.455922603607178, + "learning_rate": 1.7000000000000002e-06, + "loss": 1.2153, + "step": 34 + }, + { + "epoch": 0.054801670146137786, + "grad_norm": 5.377617835998535, + "learning_rate": 1.75e-06, + "loss": 1.1292, + "step": 35 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 7.172031879425049, + "learning_rate": 1.8000000000000001e-06, + "loss": 1.1459, + "step": 36 + }, + { + "epoch": 0.05793319415448852, + "grad_norm": 2.1464905738830566, + "learning_rate": 1.85e-06, + "loss": 1.1649, + "step": 37 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 2.7782726287841797, + "learning_rate": 1.9000000000000002e-06, + "loss": 1.159, + "step": 38 + }, + { + "epoch": 0.061064718162839246, + "grad_norm": 20.818374633789062, + "learning_rate": 1.9500000000000004e-06, + "loss": 1.0174, + "step": 39 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 22.776456832885742, + "learning_rate": 2.0000000000000003e-06, + "loss": 1.1425, + "step": 40 + }, + { + "epoch": 0.06419624217118998, + "grad_norm": 33.601287841796875, + "learning_rate": 2.05e-06, + "loss": 1.0498, + "step": 41 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.884684562683105, + "learning_rate": 2.1000000000000002e-06, + "loss": 1.1231, + "step": 42 + }, + { + "epoch": 0.06732776617954071, + "grad_norm": 1.951470971107483, + "learning_rate": 2.15e-06, + "loss": 1.1256, + "step": 43 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 2.584127426147461, + "learning_rate": 2.2e-06, + "loss": 1.1242, + "step": 44 + }, + { + "epoch": 0.07045929018789145, + "grad_norm": 6.982508659362793, + "learning_rate": 2.25e-06, + "loss": 1.0769, + "step": 45 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 7.940657138824463, + "learning_rate": 2.3000000000000004e-06, + "loss": 1.0516, + "step": 46 + }, + { + "epoch": 0.07359081419624217, + "grad_norm": 5.456567287445068, + "learning_rate": 2.35e-06, + "loss": 0.9894, + "step": 47 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 4.623636722564697, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9912, + "step": 48 + }, + { + "epoch": 0.0767223382045929, + "grad_norm": 2.182708978652954, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.1303, + "step": 49 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 2.388091802597046, + "learning_rate": 2.5e-06, + "loss": 1.0725, + "step": 50 + }, + { + "epoch": 0.07985386221294363, + "grad_norm": 2.2816522121429443, + "learning_rate": 2.55e-06, + "loss": 1.0895, + "step": 51 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 3.8391623497009277, + "learning_rate": 2.6e-06, + "loss": 1.0587, + "step": 52 + }, + { + "epoch": 0.08298538622129437, + "grad_norm": 3.700389862060547, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9795, + "step": 53 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 3.180649995803833, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.0434, + "step": 54 + }, + { + "epoch": 0.0861169102296451, + "grad_norm": 3.790774345397949, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9652, + "step": 55 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 1.8411438465118408, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.947, + "step": 56 + }, + { + "epoch": 0.08924843423799582, + "grad_norm": 2.100008010864258, + "learning_rate": 2.85e-06, + "loss": 1.0129, + "step": 57 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 1.103857159614563, + "learning_rate": 2.9e-06, + "loss": 0.9993, + "step": 58 + }, + { + "epoch": 0.09237995824634655, + "grad_norm": 1.232283353805542, + "learning_rate": 2.95e-06, + "loss": 0.9988, + "step": 59 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 2.4731297492980957, + "learning_rate": 3e-06, + "loss": 0.9873, + "step": 60 + }, + { + "epoch": 0.09551148225469729, + "grad_norm": 1.1080410480499268, + "learning_rate": 3.05e-06, + "loss": 1.0657, + "step": 61 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 1.1155121326446533, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.974, + "step": 62 + }, + { + "epoch": 0.09864300626304802, + "grad_norm": 1.763231873512268, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.0215, + "step": 63 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 1.1961321830749512, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.9955, + "step": 64 + }, + { + "epoch": 0.10177453027139875, + "grad_norm": 1.377040982246399, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.9834, + "step": 65 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 1.373260498046875, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.984, + "step": 66 + }, + { + "epoch": 0.10490605427974947, + "grad_norm": 2.034064292907715, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.973, + "step": 67 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 1.41798734664917, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.925, + "step": 68 + }, + { + "epoch": 0.1080375782881002, + "grad_norm": 1.7425439357757568, + "learning_rate": 3.45e-06, + "loss": 0.9884, + "step": 69 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 2.369657278060913, + "learning_rate": 3.5e-06, + "loss": 1.0048, + "step": 70 + }, + { + "epoch": 0.11116910229645094, + "grad_norm": 1.4699137210845947, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.9155, + "step": 71 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 1.1473263502120972, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8445, + "step": 72 + }, + { + "epoch": 0.11430062630480167, + "grad_norm": 2.315568685531616, + "learning_rate": 3.65e-06, + "loss": 0.977, + "step": 73 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 1.8554916381835938, + "learning_rate": 3.7e-06, + "loss": 0.841, + "step": 74 + }, + { + "epoch": 0.1174321503131524, + "grad_norm": 1.1517397165298462, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.9478, + "step": 75 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 2.799316883087158, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.9873, + "step": 76 + }, + { + "epoch": 0.12056367432150313, + "grad_norm": 1.11930251121521, + "learning_rate": 3.85e-06, + "loss": 0.9214, + "step": 77 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 1.0082950592041016, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9115, + "step": 78 + }, + { + "epoch": 0.12369519832985386, + "grad_norm": 1.5534541606903076, + "learning_rate": 3.95e-06, + "loss": 1.04, + "step": 79 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 1.3469066619873047, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9091, + "step": 80 + }, + { + "epoch": 0.1268267223382046, + "grad_norm": 1.1149636507034302, + "learning_rate": 4.05e-06, + "loss": 0.8829, + "step": 81 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 2.5933806896209717, + "learning_rate": 4.1e-06, + "loss": 0.9343, + "step": 82 + }, + { + "epoch": 0.1299582463465553, + "grad_norm": 1.980901837348938, + "learning_rate": 4.15e-06, + "loss": 0.9028, + "step": 83 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 0.8476768732070923, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.9038, + "step": 84 + }, + { + "epoch": 0.13308977035490605, + "grad_norm": 1.1239815950393677, + "learning_rate": 4.25e-06, + "loss": 1.0359, + "step": 85 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 1.0159156322479248, + "learning_rate": 4.3e-06, + "loss": 0.8705, + "step": 86 + }, + { + "epoch": 0.13622129436325678, + "grad_norm": 2.180856466293335, + "learning_rate": 4.350000000000001e-06, + "loss": 0.9834, + "step": 87 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 1.5725276470184326, + "learning_rate": 4.4e-06, + "loss": 0.9449, + "step": 88 + }, + { + "epoch": 0.1393528183716075, + "grad_norm": 1.5560917854309082, + "learning_rate": 4.450000000000001e-06, + "loss": 1.0109, + "step": 89 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 1.1096960306167603, + "learning_rate": 4.5e-06, + "loss": 0.9081, + "step": 90 + }, + { + "epoch": 0.14248434237995825, + "grad_norm": 1.2525160312652588, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.9907, + "step": 91 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 0.9198770523071289, + "learning_rate": 4.600000000000001e-06, + "loss": 0.9266, + "step": 92 + }, + { + "epoch": 0.14561586638830898, + "grad_norm": 0.992057740688324, + "learning_rate": 4.65e-06, + "loss": 0.9556, + "step": 93 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 1.00990891456604, + "learning_rate": 4.7e-06, + "loss": 0.9712, + "step": 94 + }, + { + "epoch": 0.1487473903966597, + "grad_norm": 1.1176848411560059, + "learning_rate": 4.75e-06, + "loss": 0.9604, + "step": 95 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 0.9323613047599792, + "learning_rate": 4.800000000000001e-06, + "loss": 0.9708, + "step": 96 + }, + { + "epoch": 0.15187891440501045, + "grad_norm": 0.9800642728805542, + "learning_rate": 4.85e-06, + "loss": 0.9868, + "step": 97 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 1.487841010093689, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8679, + "step": 98 + }, + { + "epoch": 0.15501043841336118, + "grad_norm": 0.95517897605896, + "learning_rate": 4.95e-06, + "loss": 0.7949, + "step": 99 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 0.9686336517333984, + "learning_rate": 5e-06, + "loss": 0.8655, + "step": 100 + }, + { + "epoch": 0.1581419624217119, + "grad_norm": 1.9481035470962524, + "learning_rate": 4.999999112316954e-06, + "loss": 0.9328, + "step": 101 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 1.272631287574768, + "learning_rate": 4.9999964492684465e-06, + "loss": 0.7869, + "step": 102 + }, + { + "epoch": 0.16127348643006262, + "grad_norm": 1.1117815971374512, + "learning_rate": 4.999992010856368e-06, + "loss": 0.949, + "step": 103 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 0.9681039452552795, + "learning_rate": 4.999985797083871e-06, + "loss": 1.0396, + "step": 104 + }, + { + "epoch": 0.16440501043841335, + "grad_norm": 0.9797878265380859, + "learning_rate": 4.999977807955368e-06, + "loss": 0.9393, + "step": 105 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 1.0239158868789673, + "learning_rate": 4.999968043476532e-06, + "loss": 0.8766, + "step": 106 + }, + { + "epoch": 0.16753653444676408, + "grad_norm": 0.9314751029014587, + "learning_rate": 4.999956503654299e-06, + "loss": 0.8545, + "step": 107 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 1.1627179384231567, + "learning_rate": 4.999943188496862e-06, + "loss": 0.9241, + "step": 108 + }, + { + "epoch": 0.17066805845511482, + "grad_norm": 1.022752285003662, + "learning_rate": 4.9999280980136765e-06, + "loss": 0.9662, + "step": 109 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 1.0513653755187988, + "learning_rate": 4.99991123221546e-06, + "loss": 0.9054, + "step": 110 + }, + { + "epoch": 0.17379958246346555, + "grad_norm": 0.9488444328308105, + "learning_rate": 4.99989259111419e-06, + "loss": 0.8921, + "step": 111 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 0.9934154748916626, + "learning_rate": 4.999872174723104e-06, + "loss": 0.9848, + "step": 112 + }, + { + "epoch": 0.17693110647181629, + "grad_norm": 1.2035515308380127, + "learning_rate": 4.9998499830567e-06, + "loss": 0.9833, + "step": 113 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 1.027538537979126, + "learning_rate": 4.999826016130739e-06, + "loss": 1.0089, + "step": 114 + }, + { + "epoch": 0.18006263048016702, + "grad_norm": 1.0635300874710083, + "learning_rate": 4.999800273962238e-06, + "loss": 0.8863, + "step": 115 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 0.9978796243667603, + "learning_rate": 4.999772756569482e-06, + "loss": 0.899, + "step": 116 + }, + { + "epoch": 0.18319415448851775, + "grad_norm": 0.9971838593482971, + "learning_rate": 4.999743463972008e-06, + "loss": 0.9515, + "step": 117 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 1.2436699867248535, + "learning_rate": 4.999712396190622e-06, + "loss": 0.9101, + "step": 118 + }, + { + "epoch": 0.18632567849686849, + "grad_norm": 0.9573557376861572, + "learning_rate": 4.999679553247383e-06, + "loss": 0.8817, + "step": 119 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 0.9105668067932129, + "learning_rate": 4.9996449351656165e-06, + "loss": 1.0077, + "step": 120 + }, + { + "epoch": 0.18945720250521922, + "grad_norm": 0.8741400241851807, + "learning_rate": 4.999608541969905e-06, + "loss": 0.912, + "step": 121 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 1.1172688007354736, + "learning_rate": 4.999570373686095e-06, + "loss": 0.8062, + "step": 122 + }, + { + "epoch": 0.19258872651356992, + "grad_norm": 0.940539538860321, + "learning_rate": 4.9995304303412905e-06, + "loss": 0.8944, + "step": 123 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 2.1082208156585693, + "learning_rate": 4.999488711963857e-06, + "loss": 0.9668, + "step": 124 + }, + { + "epoch": 0.19572025052192066, + "grad_norm": 1.0358761548995972, + "learning_rate": 4.99944521858342e-06, + "loss": 0.9032, + "step": 125 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 2.0804238319396973, + "learning_rate": 4.999399950230867e-06, + "loss": 0.9819, + "step": 126 + }, + { + "epoch": 0.1988517745302714, + "grad_norm": 1.0986278057098389, + "learning_rate": 4.9993529069383465e-06, + "loss": 0.9034, + "step": 127 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 1.2374722957611084, + "learning_rate": 4.999304088739263e-06, + "loss": 0.9255, + "step": 128 + }, + { + "epoch": 0.20198329853862212, + "grad_norm": 1.0443146228790283, + "learning_rate": 4.999253495668287e-06, + "loss": 0.9352, + "step": 129 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 1.009800910949707, + "learning_rate": 4.999201127761346e-06, + "loss": 0.8324, + "step": 130 + }, + { + "epoch": 0.20511482254697286, + "grad_norm": 0.892861008644104, + "learning_rate": 4.9991469850556295e-06, + "loss": 0.886, + "step": 131 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 1.1675922870635986, + "learning_rate": 4.999091067589587e-06, + "loss": 0.967, + "step": 132 + }, + { + "epoch": 0.2082463465553236, + "grad_norm": 1.0169954299926758, + "learning_rate": 4.999033375402926e-06, + "loss": 0.8498, + "step": 133 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 1.0006227493286133, + "learning_rate": 4.99897390853662e-06, + "loss": 0.8481, + "step": 134 + }, + { + "epoch": 0.21137787056367432, + "grad_norm": 0.9049347639083862, + "learning_rate": 4.9989126670328945e-06, + "loss": 0.9186, + "step": 135 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 1.2151603698730469, + "learning_rate": 4.9988496509352444e-06, + "loss": 0.9067, + "step": 136 + }, + { + "epoch": 0.21450939457202506, + "grad_norm": 1.0112006664276123, + "learning_rate": 4.998784860288417e-06, + "loss": 0.8843, + "step": 137 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 0.9316662549972534, + "learning_rate": 4.9987182951384264e-06, + "loss": 0.8635, + "step": 138 + }, + { + "epoch": 0.2176409185803758, + "grad_norm": 0.942743718624115, + "learning_rate": 4.9986499555325406e-06, + "loss": 0.8481, + "step": 139 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 1.0298817157745361, + "learning_rate": 4.998579841519292e-06, + "loss": 0.865, + "step": 140 + }, + { + "epoch": 0.22077244258872653, + "grad_norm": 1.1556729078292847, + "learning_rate": 4.998507953148472e-06, + "loss": 0.8731, + "step": 141 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 0.997958779335022, + "learning_rate": 4.9984342904711315e-06, + "loss": 0.7949, + "step": 142 + }, + { + "epoch": 0.22390396659707723, + "grad_norm": 0.9565998315811157, + "learning_rate": 4.998358853539582e-06, + "loss": 0.8875, + "step": 143 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 0.9667196273803711, + "learning_rate": 4.998281642407394e-06, + "loss": 0.8537, + "step": 144 + }, + { + "epoch": 0.22703549060542796, + "grad_norm": 0.9701303839683533, + "learning_rate": 4.998202657129401e-06, + "loss": 0.873, + "step": 145 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 0.9936159253120422, + "learning_rate": 4.998121897761692e-06, + "loss": 0.8915, + "step": 146 + }, + { + "epoch": 0.2301670146137787, + "grad_norm": 1.5054311752319336, + "learning_rate": 4.998039364361617e-06, + "loss": 0.9944, + "step": 147 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 1.5895891189575195, + "learning_rate": 4.99795505698779e-06, + "loss": 0.9282, + "step": 148 + }, + { + "epoch": 0.23329853862212943, + "grad_norm": 1.0141469240188599, + "learning_rate": 4.9978689757000785e-06, + "loss": 0.9166, + "step": 149 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 1.1720311641693115, + "learning_rate": 4.997781120559615e-06, + "loss": 0.9404, + "step": 150 + }, + { + "epoch": 0.23643006263048016, + "grad_norm": 0.9536433815956116, + "learning_rate": 4.997691491628789e-06, + "loss": 0.9339, + "step": 151 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 0.9407872557640076, + "learning_rate": 4.997600088971249e-06, + "loss": 0.9019, + "step": 152 + }, + { + "epoch": 0.2395615866388309, + "grad_norm": 0.9202889204025269, + "learning_rate": 4.997506912651906e-06, + "loss": 0.9598, + "step": 153 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 1.0630513429641724, + "learning_rate": 4.997411962736927e-06, + "loss": 0.8554, + "step": 154 + }, + { + "epoch": 0.24269311064718163, + "grad_norm": 0.8378801941871643, + "learning_rate": 4.997315239293742e-06, + "loss": 0.8583, + "step": 155 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 0.9442815184593201, + "learning_rate": 4.997216742391038e-06, + "loss": 0.8518, + "step": 156 + }, + { + "epoch": 0.24582463465553236, + "grad_norm": 0.9273308515548706, + "learning_rate": 4.997116472098763e-06, + "loss": 0.8942, + "step": 157 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 0.9028997421264648, + "learning_rate": 4.997014428488121e-06, + "loss": 0.8592, + "step": 158 + }, + { + "epoch": 0.2489561586638831, + "grad_norm": 0.9877567887306213, + "learning_rate": 4.996910611631582e-06, + "loss": 0.9316, + "step": 159 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 0.961592435836792, + "learning_rate": 4.996805021602868e-06, + "loss": 0.8875, + "step": 160 + }, + { + "epoch": 0.2520876826722338, + "grad_norm": 0.9923007488250732, + "learning_rate": 4.996697658476965e-06, + "loss": 0.8554, + "step": 161 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 0.956763744354248, + "learning_rate": 4.996588522330114e-06, + "loss": 0.9381, + "step": 162 + }, + { + "epoch": 0.25521920668058456, + "grad_norm": 0.919992208480835, + "learning_rate": 4.996477613239822e-06, + "loss": 0.9016, + "step": 163 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 0.9655581712722778, + "learning_rate": 4.996364931284847e-06, + "loss": 0.896, + "step": 164 + }, + { + "epoch": 0.25835073068893527, + "grad_norm": 0.9159606099128723, + "learning_rate": 4.996250476545212e-06, + "loss": 0.9475, + "step": 165 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 1.1147626638412476, + "learning_rate": 4.996134249102195e-06, + "loss": 0.8793, + "step": 166 + }, + { + "epoch": 0.26148225469728603, + "grad_norm": 1.112069010734558, + "learning_rate": 4.996016249038334e-06, + "loss": 0.8765, + "step": 167 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 1.0366944074630737, + "learning_rate": 4.995896476437428e-06, + "loss": 0.9356, + "step": 168 + }, + { + "epoch": 0.26461377870563674, + "grad_norm": 0.9494239091873169, + "learning_rate": 4.995774931384534e-06, + "loss": 0.8246, + "step": 169 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 0.9819200038909912, + "learning_rate": 4.995651613965964e-06, + "loss": 0.9403, + "step": 170 + }, + { + "epoch": 0.2677453027139875, + "grad_norm": 1.7329354286193848, + "learning_rate": 4.995526524269293e-06, + "loss": 0.9091, + "step": 171 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.1415139436721802, + "learning_rate": 4.995399662383352e-06, + "loss": 0.9106, + "step": 172 + }, + { + "epoch": 0.2708768267223382, + "grad_norm": 0.9471906423568726, + "learning_rate": 4.995271028398233e-06, + "loss": 0.8858, + "step": 173 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 1.096053123474121, + "learning_rate": 4.995140622405284e-06, + "loss": 0.7502, + "step": 174 + }, + { + "epoch": 0.2740083507306889, + "grad_norm": 1.285410761833191, + "learning_rate": 4.995008444497113e-06, + "loss": 0.9288, + "step": 175 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 0.9714595079421997, + "learning_rate": 4.994874494767585e-06, + "loss": 0.8824, + "step": 176 + }, + { + "epoch": 0.27713987473903967, + "grad_norm": 0.8399011492729187, + "learning_rate": 4.994738773311824e-06, + "loss": 0.8703, + "step": 177 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 0.9454224109649658, + "learning_rate": 4.994601280226212e-06, + "loss": 0.9627, + "step": 178 + }, + { + "epoch": 0.2802713987473904, + "grad_norm": 0.8983661532402039, + "learning_rate": 4.99446201560839e-06, + "loss": 0.9086, + "step": 179 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 0.9814153909683228, + "learning_rate": 4.994320979557256e-06, + "loss": 0.908, + "step": 180 + }, + { + "epoch": 0.28340292275574114, + "grad_norm": 0.8405355215072632, + "learning_rate": 4.9941781721729645e-06, + "loss": 0.8414, + "step": 181 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 0.950950026512146, + "learning_rate": 4.994033593556933e-06, + "loss": 0.9091, + "step": 182 + }, + { + "epoch": 0.28653444676409184, + "grad_norm": 1.1019747257232666, + "learning_rate": 4.99388724381183e-06, + "loss": 0.8305, + "step": 183 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 0.9339331984519958, + "learning_rate": 4.9937391230415875e-06, + "loss": 0.8542, + "step": 184 + }, + { + "epoch": 0.2896659707724426, + "grad_norm": 0.8813616633415222, + "learning_rate": 4.993589231351394e-06, + "loss": 0.848, + "step": 185 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 0.8916418552398682, + "learning_rate": 4.9934375688476905e-06, + "loss": 0.8032, + "step": 186 + }, + { + "epoch": 0.2927974947807933, + "grad_norm": 0.9891005754470825, + "learning_rate": 4.993284135638182e-06, + "loss": 0.7937, + "step": 187 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 0.9842082262039185, + "learning_rate": 4.99312893183183e-06, + "loss": 0.9532, + "step": 188 + }, + { + "epoch": 0.29592901878914407, + "grad_norm": 0.993478536605835, + "learning_rate": 4.99297195753885e-06, + "loss": 0.9406, + "step": 189 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 1.0654981136322021, + "learning_rate": 4.992813212870718e-06, + "loss": 0.907, + "step": 190 + }, + { + "epoch": 0.2990605427974948, + "grad_norm": 0.8937190771102905, + "learning_rate": 4.992652697940164e-06, + "loss": 0.9307, + "step": 191 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 0.9006052613258362, + "learning_rate": 4.992490412861178e-06, + "loss": 0.8808, + "step": 192 + }, + { + "epoch": 0.30219206680584554, + "grad_norm": 0.9601522088050842, + "learning_rate": 4.992326357749007e-06, + "loss": 0.8495, + "step": 193 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 0.9814243316650391, + "learning_rate": 4.992160532720153e-06, + "loss": 0.8204, + "step": 194 + }, + { + "epoch": 0.30532359081419624, + "grad_norm": 0.9660456776618958, + "learning_rate": 4.991992937892377e-06, + "loss": 0.8209, + "step": 195 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 0.8504270315170288, + "learning_rate": 4.991823573384695e-06, + "loss": 0.8415, + "step": 196 + }, + { + "epoch": 0.30845511482254695, + "grad_norm": 0.8733190894126892, + "learning_rate": 4.991652439317382e-06, + "loss": 0.8753, + "step": 197 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 0.9761794805526733, + "learning_rate": 4.991479535811967e-06, + "loss": 0.8923, + "step": 198 + }, + { + "epoch": 0.3115866388308977, + "grad_norm": 1.0400285720825195, + "learning_rate": 4.991304862991238e-06, + "loss": 0.899, + "step": 199 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 0.8561192750930786, + "learning_rate": 4.991128420979237e-06, + "loss": 0.9117, + "step": 200 + }, + { + "epoch": 0.3147181628392484, + "grad_norm": 0.8839634656906128, + "learning_rate": 4.990950209901263e-06, + "loss": 0.9022, + "step": 201 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 0.9569464921951294, + "learning_rate": 4.990770229883873e-06, + "loss": 0.9013, + "step": 202 + }, + { + "epoch": 0.3178496868475992, + "grad_norm": 1.0600861310958862, + "learning_rate": 4.990588481054881e-06, + "loss": 0.8387, + "step": 203 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 0.9492761492729187, + "learning_rate": 4.990404963543352e-06, + "loss": 0.8521, + "step": 204 + }, + { + "epoch": 0.3209812108559499, + "grad_norm": 0.9534817934036255, + "learning_rate": 4.990219677479612e-06, + "loss": 0.8009, + "step": 205 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 0.9918960928916931, + "learning_rate": 4.990032622995242e-06, + "loss": 0.8215, + "step": 206 + }, + { + "epoch": 0.32411273486430064, + "grad_norm": 0.8751630783081055, + "learning_rate": 4.989843800223077e-06, + "loss": 0.8744, + "step": 207 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 1.0034785270690918, + "learning_rate": 4.9896532092972085e-06, + "loss": 0.884, + "step": 208 + }, + { + "epoch": 0.32724425887265135, + "grad_norm": 1.2693432569503784, + "learning_rate": 4.989460850352984e-06, + "loss": 0.9689, + "step": 209 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 0.9923811554908752, + "learning_rate": 4.989266723527007e-06, + "loss": 0.9622, + "step": 210 + }, + { + "epoch": 0.3303757828810021, + "grad_norm": 0.9439074397087097, + "learning_rate": 4.989070828957137e-06, + "loss": 0.9155, + "step": 211 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.2564488649368286, + "learning_rate": 4.988873166782485e-06, + "loss": 0.8712, + "step": 212 + }, + { + "epoch": 0.3335073068893528, + "grad_norm": 1.1441881656646729, + "learning_rate": 4.988673737143423e-06, + "loss": 0.8895, + "step": 213 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 0.972155749797821, + "learning_rate": 4.988472540181574e-06, + "loss": 0.9158, + "step": 214 + }, + { + "epoch": 0.3366388308977035, + "grad_norm": 0.9536120295524597, + "learning_rate": 4.988269576039817e-06, + "loss": 0.7963, + "step": 215 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 0.8956594467163086, + "learning_rate": 4.988064844862288e-06, + "loss": 0.8947, + "step": 216 + }, + { + "epoch": 0.3397703549060543, + "grad_norm": 0.9503691792488098, + "learning_rate": 4.987858346794373e-06, + "loss": 0.7857, + "step": 217 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 0.9471456408500671, + "learning_rate": 4.987650081982719e-06, + "loss": 0.9118, + "step": 218 + }, + { + "epoch": 0.342901878914405, + "grad_norm": 0.9485713839530945, + "learning_rate": 4.987440050575222e-06, + "loss": 0.8188, + "step": 219 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 1.0775927305221558, + "learning_rate": 4.987228252721037e-06, + "loss": 0.9469, + "step": 220 + }, + { + "epoch": 0.34603340292275575, + "grad_norm": 0.9173988103866577, + "learning_rate": 4.9870146885705705e-06, + "loss": 0.8566, + "step": 221 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.008735179901123, + "learning_rate": 4.986799358275485e-06, + "loss": 0.8637, + "step": 222 + }, + { + "epoch": 0.34916492693110646, + "grad_norm": 0.8183198571205139, + "learning_rate": 4.986582261988695e-06, + "loss": 0.9321, + "step": 223 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 0.8951883316040039, + "learning_rate": 4.986363399864372e-06, + "loss": 0.8015, + "step": 224 + }, + { + "epoch": 0.3522964509394572, + "grad_norm": 0.9275704622268677, + "learning_rate": 4.986142772057941e-06, + "loss": 0.8242, + "step": 225 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 0.9548516273498535, + "learning_rate": 4.985920378726077e-06, + "loss": 0.7889, + "step": 226 + }, + { + "epoch": 0.3554279749478079, + "grad_norm": 1.1203585863113403, + "learning_rate": 4.985696220026714e-06, + "loss": 0.835, + "step": 227 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 0.9525771141052246, + "learning_rate": 4.985470296119038e-06, + "loss": 0.9162, + "step": 228 + }, + { + "epoch": 0.3585594989561587, + "grad_norm": 1.150404930114746, + "learning_rate": 4.985242607163488e-06, + "loss": 0.9456, + "step": 229 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.01652991771698, + "learning_rate": 4.985013153321755e-06, + "loss": 0.87, + "step": 230 + }, + { + "epoch": 0.3616910229645094, + "grad_norm": 0.9455218315124512, + "learning_rate": 4.984781934756786e-06, + "loss": 0.8466, + "step": 231 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 0.9125218391418457, + "learning_rate": 4.984548951632779e-06, + "loss": 0.8764, + "step": 232 + }, + { + "epoch": 0.36482254697286015, + "grad_norm": 2.0192980766296387, + "learning_rate": 4.984314204115187e-06, + "loss": 0.7866, + "step": 233 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.8118515014648438, + "learning_rate": 4.984077692370716e-06, + "loss": 0.8703, + "step": 234 + }, + { + "epoch": 0.36795407098121086, + "grad_norm": 1.2995024919509888, + "learning_rate": 4.983839416567322e-06, + "loss": 0.9459, + "step": 235 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 1.0611528158187866, + "learning_rate": 4.983599376874216e-06, + "loss": 0.8456, + "step": 236 + }, + { + "epoch": 0.37108559498956156, + "grad_norm": 1.2018043994903564, + "learning_rate": 4.983357573461863e-06, + "loss": 0.914, + "step": 237 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9091911315917969, + "learning_rate": 4.983114006501978e-06, + "loss": 0.791, + "step": 238 + }, + { + "epoch": 0.3742171189979123, + "grad_norm": 0.980873167514801, + "learning_rate": 4.982868676167528e-06, + "loss": 0.77, + "step": 239 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 0.8945695161819458, + "learning_rate": 4.982621582632735e-06, + "loss": 0.8038, + "step": 240 + }, + { + "epoch": 0.37734864300626303, + "grad_norm": 0.9325360655784607, + "learning_rate": 4.9823727260730705e-06, + "loss": 0.8908, + "step": 241 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 0.987630307674408, + "learning_rate": 4.98212210666526e-06, + "loss": 0.8642, + "step": 242 + }, + { + "epoch": 0.3804801670146138, + "grad_norm": 0.9151357412338257, + "learning_rate": 4.98186972458728e-06, + "loss": 0.8437, + "step": 243 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 0.9084850549697876, + "learning_rate": 4.981615580018358e-06, + "loss": 0.8775, + "step": 244 + }, + { + "epoch": 0.3836116910229645, + "grad_norm": 0.9786385297775269, + "learning_rate": 4.981359673138974e-06, + "loss": 0.8251, + "step": 245 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 0.8947877883911133, + "learning_rate": 4.98110200413086e-06, + "loss": 0.7709, + "step": 246 + }, + { + "epoch": 0.38674321503131526, + "grad_norm": 0.967019259929657, + "learning_rate": 4.9808425731769985e-06, + "loss": 0.8954, + "step": 247 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 1.1066704988479614, + "learning_rate": 4.980581380461622e-06, + "loss": 0.8544, + "step": 248 + }, + { + "epoch": 0.38987473903966596, + "grad_norm": 0.9750981330871582, + "learning_rate": 4.980318426170218e-06, + "loss": 0.8354, + "step": 249 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 0.8727232217788696, + "learning_rate": 4.980053710489521e-06, + "loss": 0.8911, + "step": 250 + }, + { + "epoch": 0.3930062630480167, + "grad_norm": 0.9337188601493835, + "learning_rate": 4.979787233607518e-06, + "loss": 0.8567, + "step": 251 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 0.890701174736023, + "learning_rate": 4.979518995713448e-06, + "loss": 0.8327, + "step": 252 + }, + { + "epoch": 0.39613778705636743, + "grad_norm": 0.9973195195198059, + "learning_rate": 4.979248996997797e-06, + "loss": 0.9065, + "step": 253 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 0.9894055128097534, + "learning_rate": 4.978977237652306e-06, + "loss": 0.8729, + "step": 254 + }, + { + "epoch": 0.39926931106471814, + "grad_norm": 1.0398536920547485, + "learning_rate": 4.978703717869961e-06, + "loss": 0.861, + "step": 255 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.1713942289352417, + "learning_rate": 4.978428437845003e-06, + "loss": 0.8183, + "step": 256 + }, + { + "epoch": 0.4024008350730689, + "grad_norm": 0.9868776202201843, + "learning_rate": 4.9781513977729225e-06, + "loss": 0.941, + "step": 257 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 0.9260863065719604, + "learning_rate": 4.977872597850455e-06, + "loss": 0.8963, + "step": 258 + }, + { + "epoch": 0.4055323590814196, + "grad_norm": 0.9553239941596985, + "learning_rate": 4.977592038275592e-06, + "loss": 0.8102, + "step": 259 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 0.9426907300949097, + "learning_rate": 4.977309719247571e-06, + "loss": 0.89, + "step": 260 + }, + { + "epoch": 0.40866388308977036, + "grad_norm": 0.9691527485847473, + "learning_rate": 4.97702564096688e-06, + "loss": 0.7967, + "step": 261 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 0.9917487502098083, + "learning_rate": 4.976739803635256e-06, + "loss": 0.8327, + "step": 262 + }, + { + "epoch": 0.41179540709812107, + "grad_norm": 0.992340087890625, + "learning_rate": 4.976452207455686e-06, + "loss": 0.9462, + "step": 263 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 0.9554483890533447, + "learning_rate": 4.9761628526324035e-06, + "loss": 0.8294, + "step": 264 + }, + { + "epoch": 0.41492693110647183, + "grad_norm": 0.9318789839744568, + "learning_rate": 4.975871739370895e-06, + "loss": 0.8079, + "step": 265 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 0.9158215522766113, + "learning_rate": 4.975578867877893e-06, + "loss": 0.8632, + "step": 266 + }, + { + "epoch": 0.41805845511482254, + "grad_norm": 0.9444370865821838, + "learning_rate": 4.975284238361378e-06, + "loss": 0.8897, + "step": 267 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 0.9949231743812561, + "learning_rate": 4.974987851030581e-06, + "loss": 0.7905, + "step": 268 + }, + { + "epoch": 0.4211899791231733, + "grad_norm": 0.9819440245628357, + "learning_rate": 4.974689706095981e-06, + "loss": 0.9183, + "step": 269 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0787206888198853, + "learning_rate": 4.974389803769304e-06, + "loss": 0.7739, + "step": 270 + }, + { + "epoch": 0.424321503131524, + "grad_norm": 0.8871361017227173, + "learning_rate": 4.9740881442635235e-06, + "loss": 0.9454, + "step": 271 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 0.9125893115997314, + "learning_rate": 4.973784727792863e-06, + "loss": 0.8017, + "step": 272 + }, + { + "epoch": 0.42745302713987476, + "grad_norm": 0.9102838635444641, + "learning_rate": 4.973479554572793e-06, + "loss": 0.8307, + "step": 273 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 0.8780474662780762, + "learning_rate": 4.973172624820031e-06, + "loss": 0.9402, + "step": 274 + }, + { + "epoch": 0.43058455114822547, + "grad_norm": 1.0059378147125244, + "learning_rate": 4.972863938752541e-06, + "loss": 0.7438, + "step": 275 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.9684615135192871, + "learning_rate": 4.972553496589537e-06, + "loss": 0.8897, + "step": 276 + }, + { + "epoch": 0.4337160751565762, + "grad_norm": 0.9615057110786438, + "learning_rate": 4.972241298551477e-06, + "loss": 0.8157, + "step": 277 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8589702844619751, + "learning_rate": 4.971927344860068e-06, + "loss": 0.7813, + "step": 278 + }, + { + "epoch": 0.43684759916492694, + "grad_norm": 0.900324285030365, + "learning_rate": 4.971611635738263e-06, + "loss": 0.7534, + "step": 279 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 0.8881833553314209, + "learning_rate": 4.971294171410261e-06, + "loss": 0.8255, + "step": 280 + }, + { + "epoch": 0.43997912317327764, + "grad_norm": 0.9723751544952393, + "learning_rate": 4.9709749521015085e-06, + "loss": 0.8796, + "step": 281 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 0.9341713190078735, + "learning_rate": 4.9706539780386995e-06, + "loss": 0.7429, + "step": 282 + }, + { + "epoch": 0.4431106471816284, + "grad_norm": 0.9354205131530762, + "learning_rate": 4.97033124944977e-06, + "loss": 0.8545, + "step": 283 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 0.9294624328613281, + "learning_rate": 4.970006766563906e-06, + "loss": 0.7783, + "step": 284 + }, + { + "epoch": 0.4462421711899791, + "grad_norm": 0.9584887027740479, + "learning_rate": 4.969680529611538e-06, + "loss": 0.8068, + "step": 285 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 0.9525735378265381, + "learning_rate": 4.9693525388243424e-06, + "loss": 0.8055, + "step": 286 + }, + { + "epoch": 0.44937369519832987, + "grad_norm": 0.8845317959785461, + "learning_rate": 4.96902279443524e-06, + "loss": 0.8318, + "step": 287 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 0.8932276964187622, + "learning_rate": 4.9686912966783965e-06, + "loss": 0.8185, + "step": 288 + }, + { + "epoch": 0.4525052192066806, + "grad_norm": 0.9881951808929443, + "learning_rate": 4.968358045789225e-06, + "loss": 0.9001, + "step": 289 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 0.995212197303772, + "learning_rate": 4.968023042004384e-06, + "loss": 0.75, + "step": 290 + }, + { + "epoch": 0.45563674321503134, + "grad_norm": 0.9997846484184265, + "learning_rate": 4.967686285561773e-06, + "loss": 0.8993, + "step": 291 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 0.9438545107841492, + "learning_rate": 4.967347776700538e-06, + "loss": 0.8964, + "step": 292 + }, + { + "epoch": 0.45876826722338204, + "grad_norm": 0.9164395332336426, + "learning_rate": 4.967007515661072e-06, + "loss": 0.8671, + "step": 293 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.895302951335907, + "learning_rate": 4.966665502685009e-06, + "loss": 0.8785, + "step": 294 + }, + { + "epoch": 0.4618997912317328, + "grad_norm": 0.9122863411903381, + "learning_rate": 4.966321738015227e-06, + "loss": 0.8499, + "step": 295 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.0097336769104004, + "learning_rate": 4.965976221895852e-06, + "loss": 0.83, + "step": 296 + }, + { + "epoch": 0.4650313152400835, + "grad_norm": 1.0664499998092651, + "learning_rate": 4.965628954572249e-06, + "loss": 0.7799, + "step": 297 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 0.9602341055870056, + "learning_rate": 4.96527993629103e-06, + "loss": 0.8061, + "step": 298 + }, + { + "epoch": 0.4681628392484342, + "grad_norm": 1.0133620500564575, + "learning_rate": 4.964929167300048e-06, + "loss": 0.9148, + "step": 299 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9351868033409119, + "learning_rate": 4.964576647848401e-06, + "loss": 0.83, + "step": 300 + }, + { + "epoch": 0.471294363256785, + "grad_norm": 1.0078860521316528, + "learning_rate": 4.964222378186429e-06, + "loss": 0.8391, + "step": 301 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 0.8966493606567383, + "learning_rate": 4.963866358565716e-06, + "loss": 0.8677, + "step": 302 + }, + { + "epoch": 0.4744258872651357, + "grad_norm": 0.9283657670021057, + "learning_rate": 4.963508589239087e-06, + "loss": 0.8787, + "step": 303 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 0.9540023803710938, + "learning_rate": 4.963149070460611e-06, + "loss": 0.8667, + "step": 304 + }, + { + "epoch": 0.47755741127348644, + "grad_norm": 0.9580283164978027, + "learning_rate": 4.9627878024855995e-06, + "loss": 0.8442, + "step": 305 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.9008278846740723, + "learning_rate": 4.962424785570605e-06, + "loss": 0.8605, + "step": 306 + }, + { + "epoch": 0.48068893528183715, + "grad_norm": 0.9681883454322815, + "learning_rate": 4.962060019973423e-06, + "loss": 0.7865, + "step": 307 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 0.8492597341537476, + "learning_rate": 4.9616935059530915e-06, + "loss": 0.8578, + "step": 308 + }, + { + "epoch": 0.4838204592901879, + "grad_norm": 0.9323164820671082, + "learning_rate": 4.9613252437698865e-06, + "loss": 0.8491, + "step": 309 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 0.9417833685874939, + "learning_rate": 4.960955233685331e-06, + "loss": 0.8108, + "step": 310 + }, + { + "epoch": 0.4869519832985386, + "grad_norm": 0.901364803314209, + "learning_rate": 4.960583475962185e-06, + "loss": 0.8958, + "step": 311 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8929979205131531, + "learning_rate": 4.96020997086445e-06, + "loss": 0.8237, + "step": 312 + }, + { + "epoch": 0.4900835073068894, + "grad_norm": 0.9422337412834167, + "learning_rate": 4.9598347186573716e-06, + "loss": 0.8024, + "step": 313 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 0.9165504574775696, + "learning_rate": 4.959457719607432e-06, + "loss": 0.8124, + "step": 314 + }, + { + "epoch": 0.4932150313152401, + "grad_norm": 1.0233370065689087, + "learning_rate": 4.959078973982356e-06, + "loss": 0.788, + "step": 315 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9375972151756287, + "learning_rate": 4.958698482051109e-06, + "loss": 0.839, + "step": 316 + }, + { + "epoch": 0.4963465553235908, + "grad_norm": 0.8805437684059143, + "learning_rate": 4.958316244083895e-06, + "loss": 0.7445, + "step": 317 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.0231893062591553, + "learning_rate": 4.95793226035216e-06, + "loss": 0.8257, + "step": 318 + }, + { + "epoch": 0.49947807933194155, + "grad_norm": 0.9125928282737732, + "learning_rate": 4.957546531128589e-06, + "loss": 0.7699, + "step": 319 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.8844016790390015, + "learning_rate": 4.957159056687104e-06, + "loss": 0.8784, + "step": 320 + }, + { + "epoch": 0.5026096033402923, + "grad_norm": 0.9770572781562805, + "learning_rate": 4.956769837302871e-06, + "loss": 0.8611, + "step": 321 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.9413725733757019, + "learning_rate": 4.956378873252293e-06, + "loss": 0.8056, + "step": 322 + }, + { + "epoch": 0.505741127348643, + "grad_norm": 0.9058850407600403, + "learning_rate": 4.955986164813008e-06, + "loss": 0.8162, + "step": 323 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 0.893100917339325, + "learning_rate": 4.955591712263901e-06, + "loss": 0.7554, + "step": 324 + }, + { + "epoch": 0.5088726513569938, + "grad_norm": 0.8967075943946838, + "learning_rate": 4.955195515885089e-06, + "loss": 0.8143, + "step": 325 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.9542785286903381, + "learning_rate": 4.954797575957929e-06, + "loss": 0.8102, + "step": 326 + }, + { + "epoch": 0.5120041753653445, + "grad_norm": 1.006451964378357, + "learning_rate": 4.954397892765018e-06, + "loss": 0.86, + "step": 327 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 0.9964715838432312, + "learning_rate": 4.953996466590189e-06, + "loss": 0.7166, + "step": 328 + }, + { + "epoch": 0.5151356993736952, + "grad_norm": 0.9568808674812317, + "learning_rate": 4.953593297718512e-06, + "loss": 0.8753, + "step": 329 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.9435825347900391, + "learning_rate": 4.953188386436298e-06, + "loss": 0.8961, + "step": 330 + }, + { + "epoch": 0.5182672233820459, + "grad_norm": 0.9386457800865173, + "learning_rate": 4.952781733031092e-06, + "loss": 0.8468, + "step": 331 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 0.9916554689407349, + "learning_rate": 4.952373337791678e-06, + "loss": 0.8274, + "step": 332 + }, + { + "epoch": 0.5213987473903967, + "grad_norm": 0.9351572394371033, + "learning_rate": 4.9519632010080765e-06, + "loss": 0.7964, + "step": 333 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9936269521713257, + "learning_rate": 4.951551322971545e-06, + "loss": 0.8354, + "step": 334 + }, + { + "epoch": 0.5245302713987474, + "grad_norm": 0.9678221940994263, + "learning_rate": 4.951137703974576e-06, + "loss": 0.9115, + "step": 335 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 0.9453133940696716, + "learning_rate": 4.9507223443109005e-06, + "loss": 0.7725, + "step": 336 + }, + { + "epoch": 0.5276617954070981, + "grad_norm": 0.9577226638793945, + "learning_rate": 4.950305244275484e-06, + "loss": 0.8679, + "step": 337 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 0.998489260673523, + "learning_rate": 4.94988640416453e-06, + "loss": 0.7505, + "step": 338 + }, + { + "epoch": 0.5307933194154488, + "grad_norm": 0.976026713848114, + "learning_rate": 4.949465824275475e-06, + "loss": 0.9083, + "step": 339 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.0041502714157104, + "learning_rate": 4.9490435049069925e-06, + "loss": 0.8815, + "step": 340 + }, + { + "epoch": 0.5339248434237995, + "grad_norm": 0.9839801788330078, + "learning_rate": 4.948619446358991e-06, + "loss": 0.8517, + "step": 341 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 0.9995140433311462, + "learning_rate": 4.948193648932616e-06, + "loss": 0.8157, + "step": 342 + }, + { + "epoch": 0.5370563674321504, + "grad_norm": 1.0782225131988525, + "learning_rate": 4.947766112930243e-06, + "loss": 0.9206, + "step": 343 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9981802105903625, + "learning_rate": 4.947336838655487e-06, + "loss": 0.8389, + "step": 344 + }, + { + "epoch": 0.5401878914405011, + "grad_norm": 0.9261236190795898, + "learning_rate": 4.946905826413195e-06, + "loss": 0.8035, + "step": 345 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.9195284247398376, + "learning_rate": 4.946473076509449e-06, + "loss": 0.8265, + "step": 346 + }, + { + "epoch": 0.5433194154488518, + "grad_norm": 0.9535123109817505, + "learning_rate": 4.946038589251566e-06, + "loss": 0.7528, + "step": 347 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 0.9969532489776611, + "learning_rate": 4.9456023649480935e-06, + "loss": 0.7787, + "step": 348 + }, + { + "epoch": 0.5464509394572025, + "grad_norm": 0.9612399935722351, + "learning_rate": 4.945164403908816e-06, + "loss": 0.8298, + "step": 349 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 0.9541627764701843, + "learning_rate": 4.944724706444749e-06, + "loss": 0.8039, + "step": 350 + }, + { + "epoch": 0.5495824634655533, + "grad_norm": 0.9916557669639587, + "learning_rate": 4.944283272868143e-06, + "loss": 0.9012, + "step": 351 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 0.8807661533355713, + "learning_rate": 4.94384010349248e-06, + "loss": 0.9005, + "step": 352 + }, + { + "epoch": 0.552713987473904, + "grad_norm": 1.0179708003997803, + "learning_rate": 4.943395198632476e-06, + "loss": 0.8845, + "step": 353 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.923014223575592, + "learning_rate": 4.942948558604077e-06, + "loss": 0.8173, + "step": 354 + }, + { + "epoch": 0.5558455114822547, + "grad_norm": 0.9318749308586121, + "learning_rate": 4.942500183724463e-06, + "loss": 0.8828, + "step": 355 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.0907000303268433, + "learning_rate": 4.942050074312048e-06, + "loss": 0.7951, + "step": 356 + }, + { + "epoch": 0.5589770354906054, + "grad_norm": 0.9811620712280273, + "learning_rate": 4.9415982306864744e-06, + "loss": 0.8826, + "step": 357 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9888150691986084, + "learning_rate": 4.9411446531686156e-06, + "loss": 0.8338, + "step": 358 + }, + { + "epoch": 0.5621085594989561, + "grad_norm": 0.9389257431030273, + "learning_rate": 4.9406893420805804e-06, + "loss": 0.8316, + "step": 359 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.9500603675842285, + "learning_rate": 4.940232297745705e-06, + "loss": 0.8764, + "step": 360 + }, + { + "epoch": 0.5652400835073069, + "grad_norm": 0.9748769998550415, + "learning_rate": 4.939773520488559e-06, + "loss": 0.8308, + "step": 361 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 0.9179681539535522, + "learning_rate": 4.93931301063494e-06, + "loss": 0.789, + "step": 362 + }, + { + "epoch": 0.5683716075156576, + "grad_norm": 0.8609325289726257, + "learning_rate": 4.938850768511878e-06, + "loss": 0.7368, + "step": 363 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 0.9375531673431396, + "learning_rate": 4.9383867944476325e-06, + "loss": 0.8164, + "step": 364 + }, + { + "epoch": 0.5715031315240083, + "grad_norm": 0.9386453032493591, + "learning_rate": 4.937921088771694e-06, + "loss": 0.7898, + "step": 365 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.025040864944458, + "learning_rate": 4.937453651814781e-06, + "loss": 0.8465, + "step": 366 + }, + { + "epoch": 0.574634655532359, + "grad_norm": 1.1037245988845825, + "learning_rate": 4.936984483908842e-06, + "loss": 0.7935, + "step": 367 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.230821132659912, + "learning_rate": 4.936513585387055e-06, + "loss": 0.8043, + "step": 368 + }, + { + "epoch": 0.5777661795407099, + "grad_norm": 1.090235948562622, + "learning_rate": 4.9360409565838265e-06, + "loss": 0.8726, + "step": 369 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8977810144424438, + "learning_rate": 4.935566597834793e-06, + "loss": 0.8667, + "step": 370 + }, + { + "epoch": 0.5808977035490606, + "grad_norm": 0.96401447057724, + "learning_rate": 4.935090509476818e-06, + "loss": 0.8494, + "step": 371 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 0.9290149211883545, + "learning_rate": 4.934612691847995e-06, + "loss": 0.7678, + "step": 372 + }, + { + "epoch": 0.5840292275574113, + "grad_norm": 0.9765584468841553, + "learning_rate": 4.934133145287644e-06, + "loss": 0.7947, + "step": 373 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 0.8939430117607117, + "learning_rate": 4.933651870136313e-06, + "loss": 0.885, + "step": 374 + }, + { + "epoch": 0.587160751565762, + "grad_norm": 0.9512760043144226, + "learning_rate": 4.933168866735777e-06, + "loss": 0.8863, + "step": 375 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 1.0334441661834717, + "learning_rate": 4.932684135429041e-06, + "loss": 0.8362, + "step": 376 + }, + { + "epoch": 0.5902922755741128, + "grad_norm": 0.9402525424957275, + "learning_rate": 4.932197676560334e-06, + "loss": 0.8758, + "step": 377 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 0.9615777730941772, + "learning_rate": 4.931709490475113e-06, + "loss": 0.8324, + "step": 378 + }, + { + "epoch": 0.5934237995824635, + "grad_norm": 0.9407138228416443, + "learning_rate": 4.931219577520061e-06, + "loss": 0.8454, + "step": 379 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.0415815114974976, + "learning_rate": 4.930727938043091e-06, + "loss": 0.8649, + "step": 380 + }, + { + "epoch": 0.5965553235908142, + "grad_norm": 1.0946165323257446, + "learning_rate": 4.9302345723933344e-06, + "loss": 0.7794, + "step": 381 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.0532523393630981, + "learning_rate": 4.929739480921157e-06, + "loss": 0.8555, + "step": 382 + }, + { + "epoch": 0.5996868475991649, + "grad_norm": 0.9441399574279785, + "learning_rate": 4.929242663978144e-06, + "loss": 0.7613, + "step": 383 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 0.9970793128013611, + "learning_rate": 4.92874412191711e-06, + "loss": 0.9037, + "step": 384 + }, + { + "epoch": 0.6028183716075156, + "grad_norm": 0.9777354001998901, + "learning_rate": 4.928243855092091e-06, + "loss": 0.9044, + "step": 385 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 0.9507855772972107, + "learning_rate": 4.92774186385835e-06, + "loss": 0.8359, + "step": 386 + }, + { + "epoch": 0.6059498956158664, + "grad_norm": 0.9827553033828735, + "learning_rate": 4.9272381485723765e-06, + "loss": 0.7938, + "step": 387 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 0.8896475434303284, + "learning_rate": 4.926732709591879e-06, + "loss": 0.8545, + "step": 388 + }, + { + "epoch": 0.6090814196242171, + "grad_norm": 0.9817705750465393, + "learning_rate": 4.926225547275795e-06, + "loss": 0.8456, + "step": 389 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 0.9646008610725403, + "learning_rate": 4.925716661984283e-06, + "loss": 0.7906, + "step": 390 + }, + { + "epoch": 0.6122129436325678, + "grad_norm": 0.8940444588661194, + "learning_rate": 4.9252060540787275e-06, + "loss": 0.8371, + "step": 391 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.0812394618988037, + "learning_rate": 4.924693723921734e-06, + "loss": 0.8056, + "step": 392 + }, + { + "epoch": 0.6153444676409185, + "grad_norm": 0.9953856468200684, + "learning_rate": 4.924179671877132e-06, + "loss": 0.8651, + "step": 393 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 0.9760703444480896, + "learning_rate": 4.9236638983099735e-06, + "loss": 0.845, + "step": 394 + }, + { + "epoch": 0.6184759916492694, + "grad_norm": 1.0127615928649902, + "learning_rate": 4.923146403586533e-06, + "loss": 0.7987, + "step": 395 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 0.8771396279335022, + "learning_rate": 4.9226271880743086e-06, + "loss": 0.8092, + "step": 396 + }, + { + "epoch": 0.6216075156576201, + "grad_norm": 0.8906831741333008, + "learning_rate": 4.9221062521420196e-06, + "loss": 0.8039, + "step": 397 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.0218392610549927, + "learning_rate": 4.9215835961596045e-06, + "loss": 0.8354, + "step": 398 + }, + { + "epoch": 0.6247390396659708, + "grad_norm": 1.0382038354873657, + "learning_rate": 4.921059220498227e-06, + "loss": 0.8021, + "step": 399 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 0.8997588157653809, + "learning_rate": 4.920533125530272e-06, + "loss": 0.8716, + "step": 400 + }, + { + "epoch": 0.6278705636743215, + "grad_norm": 0.8725724220275879, + "learning_rate": 4.920005311629342e-06, + "loss": 0.7934, + "step": 401 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.9289800524711609, + "learning_rate": 4.9194757791702625e-06, + "loss": 0.812, + "step": 402 + }, + { + "epoch": 0.6310020876826722, + "grad_norm": 0.9969078898429871, + "learning_rate": 4.918944528529079e-06, + "loss": 0.866, + "step": 403 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 0.9544941186904907, + "learning_rate": 4.918411560083058e-06, + "loss": 0.8035, + "step": 404 + }, + { + "epoch": 0.634133611691023, + "grad_norm": 0.9678053855895996, + "learning_rate": 4.917876874210686e-06, + "loss": 0.8254, + "step": 405 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 0.9782965779304504, + "learning_rate": 4.917340471291666e-06, + "loss": 0.7954, + "step": 406 + }, + { + "epoch": 0.6372651356993737, + "grad_norm": 0.9380600452423096, + "learning_rate": 4.916802351706923e-06, + "loss": 0.8287, + "step": 407 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 0.9455891847610474, + "learning_rate": 4.916262515838603e-06, + "loss": 0.8714, + "step": 408 + }, + { + "epoch": 0.6403966597077244, + "grad_norm": 0.8937395811080933, + "learning_rate": 4.915720964070065e-06, + "loss": 0.7618, + "step": 409 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 1.0304689407348633, + "learning_rate": 4.915177696785894e-06, + "loss": 0.8436, + "step": 410 + }, + { + "epoch": 0.6435281837160751, + "grad_norm": 0.9623110294342041, + "learning_rate": 4.914632714371885e-06, + "loss": 0.865, + "step": 411 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 0.9212867021560669, + "learning_rate": 4.914086017215059e-06, + "loss": 0.8232, + "step": 412 + }, + { + "epoch": 0.6466597077244259, + "grad_norm": 0.9817205667495728, + "learning_rate": 4.9135376057036485e-06, + "loss": 0.7926, + "step": 413 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 0.9576719403266907, + "learning_rate": 4.912987480227108e-06, + "loss": 0.8812, + "step": 414 + }, + { + "epoch": 0.6497912317327766, + "grad_norm": 0.9415407180786133, + "learning_rate": 4.912435641176106e-06, + "loss": 0.8985, + "step": 415 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 0.9084275364875793, + "learning_rate": 4.91188208894253e-06, + "loss": 0.8085, + "step": 416 + }, + { + "epoch": 0.6529227557411273, + "grad_norm": 0.8890969753265381, + "learning_rate": 4.911326823919482e-06, + "loss": 0.8127, + "step": 417 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 0.8727678656578064, + "learning_rate": 4.9107698465012815e-06, + "loss": 0.8656, + "step": 418 + }, + { + "epoch": 0.656054279749478, + "grad_norm": 1.0174667835235596, + "learning_rate": 4.910211157083465e-06, + "loss": 0.8358, + "step": 419 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.045121431350708, + "learning_rate": 4.909650756062782e-06, + "loss": 0.8239, + "step": 420 + }, + { + "epoch": 0.6591858037578288, + "grad_norm": 1.0688676834106445, + "learning_rate": 4.909088643837202e-06, + "loss": 0.9246, + "step": 421 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.9371877908706665, + "learning_rate": 4.908524820805905e-06, + "loss": 0.7862, + "step": 422 + }, + { + "epoch": 0.6623173277661796, + "grad_norm": 0.9567264318466187, + "learning_rate": 4.907959287369288e-06, + "loss": 0.8164, + "step": 423 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 0.9109166860580444, + "learning_rate": 4.907392043928964e-06, + "loss": 0.7016, + "step": 424 + }, + { + "epoch": 0.6654488517745303, + "grad_norm": 1.0378203392028809, + "learning_rate": 4.906823090887757e-06, + "loss": 0.8986, + "step": 425 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 0.9871951341629028, + "learning_rate": 4.906252428649708e-06, + "loss": 0.8366, + "step": 426 + }, + { + "epoch": 0.668580375782881, + "grad_norm": 0.8682780265808105, + "learning_rate": 4.905680057620072e-06, + "loss": 0.738, + "step": 427 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.8834647536277771, + "learning_rate": 4.9051059782053125e-06, + "loss": 0.8304, + "step": 428 + }, + { + "epoch": 0.6717118997912317, + "grad_norm": 1.0133378505706787, + "learning_rate": 4.904530190813113e-06, + "loss": 0.9369, + "step": 429 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 0.9967373013496399, + "learning_rate": 4.9039526958523676e-06, + "loss": 0.782, + "step": 430 + }, + { + "epoch": 0.6748434237995825, + "grad_norm": 1.0258032083511353, + "learning_rate": 4.90337349373318e-06, + "loss": 0.8214, + "step": 431 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 0.9814366102218628, + "learning_rate": 4.90279258486687e-06, + "loss": 0.8173, + "step": 432 + }, + { + "epoch": 0.6779749478079332, + "grad_norm": 0.9351440668106079, + "learning_rate": 4.902209969665966e-06, + "loss": 0.8815, + "step": 433 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 0.9490398168563843, + "learning_rate": 4.901625648544212e-06, + "loss": 0.8295, + "step": 434 + }, + { + "epoch": 0.6811064718162839, + "grad_norm": 1.0183534622192383, + "learning_rate": 4.9010396219165614e-06, + "loss": 0.8538, + "step": 435 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 0.9546058773994446, + "learning_rate": 4.900451890199179e-06, + "loss": 0.8385, + "step": 436 + }, + { + "epoch": 0.6842379958246346, + "grad_norm": 1.076120138168335, + "learning_rate": 4.8998624538094394e-06, + "loss": 0.8958, + "step": 437 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.8935732841491699, + "learning_rate": 4.899271313165929e-06, + "loss": 0.8674, + "step": 438 + }, + { + "epoch": 0.6873695198329853, + "grad_norm": 0.9574604630470276, + "learning_rate": 4.898678468688445e-06, + "loss": 0.8245, + "step": 439 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.931751549243927, + "learning_rate": 4.898083920797993e-06, + "loss": 0.8409, + "step": 440 + }, + { + "epoch": 0.6905010438413361, + "grad_norm": 0.937874436378479, + "learning_rate": 4.89748766991679e-06, + "loss": 0.8222, + "step": 441 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9606375694274902, + "learning_rate": 4.896889716468259e-06, + "loss": 0.8123, + "step": 442 + }, + { + "epoch": 0.6936325678496869, + "grad_norm": 0.8692249059677124, + "learning_rate": 4.896290060877038e-06, + "loss": 0.8532, + "step": 443 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 0.9597892165184021, + "learning_rate": 4.895688703568968e-06, + "loss": 0.8365, + "step": 444 + }, + { + "epoch": 0.6967640918580376, + "grad_norm": 0.958342432975769, + "learning_rate": 4.895085644971102e-06, + "loss": 0.8535, + "step": 445 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.9603601098060608, + "learning_rate": 4.894480885511699e-06, + "loss": 0.8227, + "step": 446 + }, + { + "epoch": 0.6998956158663883, + "grad_norm": 1.0010526180267334, + "learning_rate": 4.893874425620226e-06, + "loss": 0.8785, + "step": 447 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.0061219930648804, + "learning_rate": 4.893266265727361e-06, + "loss": 0.8336, + "step": 448 + }, + { + "epoch": 0.7030271398747391, + "grad_norm": 0.8967173099517822, + "learning_rate": 4.892656406264984e-06, + "loss": 0.8305, + "step": 449 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.962729275226593, + "learning_rate": 4.892044847666186e-06, + "loss": 0.7843, + "step": 450 + }, + { + "epoch": 0.7061586638830898, + "grad_norm": 1.0067342519760132, + "learning_rate": 4.891431590365262e-06, + "loss": 0.8111, + "step": 451 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 1.0252268314361572, + "learning_rate": 4.890816634797716e-06, + "loss": 0.8253, + "step": 452 + }, + { + "epoch": 0.7092901878914405, + "grad_norm": 0.9098564386367798, + "learning_rate": 4.890199981400255e-06, + "loss": 0.8526, + "step": 453 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 0.893135130405426, + "learning_rate": 4.889581630610793e-06, + "loss": 0.8217, + "step": 454 + }, + { + "epoch": 0.7124217118997912, + "grad_norm": 0.9225817322731018, + "learning_rate": 4.888961582868453e-06, + "loss": 0.8296, + "step": 455 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 0.973386824131012, + "learning_rate": 4.888339838613555e-06, + "loss": 0.7429, + "step": 456 + }, + { + "epoch": 0.715553235908142, + "grad_norm": 0.9829993844032288, + "learning_rate": 4.8877163982876306e-06, + "loss": 0.7405, + "step": 457 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 0.927728533744812, + "learning_rate": 4.887091262333414e-06, + "loss": 0.8257, + "step": 458 + }, + { + "epoch": 0.7186847599164927, + "grad_norm": 0.9419897794723511, + "learning_rate": 4.886464431194844e-06, + "loss": 0.7762, + "step": 459 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 0.9358823299407959, + "learning_rate": 4.885835905317061e-06, + "loss": 0.7794, + "step": 460 + }, + { + "epoch": 0.7218162839248434, + "grad_norm": 1.0290783643722534, + "learning_rate": 4.88520568514641e-06, + "loss": 0.8723, + "step": 461 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 0.9909068942070007, + "learning_rate": 4.884573771130442e-06, + "loss": 0.8302, + "step": 462 + }, + { + "epoch": 0.7249478079331941, + "grad_norm": 0.9972653985023499, + "learning_rate": 4.883940163717906e-06, + "loss": 0.7961, + "step": 463 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.9245638251304626, + "learning_rate": 4.883304863358757e-06, + "loss": 0.8057, + "step": 464 + }, + { + "epoch": 0.7280793319415448, + "grad_norm": 1.0097148418426514, + "learning_rate": 4.882667870504153e-06, + "loss": 0.7296, + "step": 465 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 0.9792670607566833, + "learning_rate": 4.882029185606448e-06, + "loss": 0.8006, + "step": 466 + }, + { + "epoch": 0.7312108559498957, + "grad_norm": 0.9308120608329773, + "learning_rate": 4.881388809119206e-06, + "loss": 0.7957, + "step": 467 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 0.9472301602363586, + "learning_rate": 4.880746741497187e-06, + "loss": 0.8984, + "step": 468 + }, + { + "epoch": 0.7343423799582464, + "grad_norm": 0.9898937344551086, + "learning_rate": 4.880102983196351e-06, + "loss": 0.8365, + "step": 469 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 0.9376612901687622, + "learning_rate": 4.879457534673863e-06, + "loss": 0.7838, + "step": 470 + }, + { + "epoch": 0.7374739039665971, + "grad_norm": 1.030616283416748, + "learning_rate": 4.878810396388085e-06, + "loss": 0.7698, + "step": 471 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.9583387970924377, + "learning_rate": 4.87816156879858e-06, + "loss": 0.7381, + "step": 472 + }, + { + "epoch": 0.7406054279749478, + "grad_norm": 1.00180184841156, + "learning_rate": 4.87751105236611e-06, + "loss": 0.8497, + "step": 473 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.9382562637329102, + "learning_rate": 4.876858847552638e-06, + "loss": 0.8698, + "step": 474 + }, + { + "epoch": 0.7437369519832986, + "grad_norm": 1.0104063749313354, + "learning_rate": 4.8762049548213255e-06, + "loss": 0.8444, + "step": 475 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 0.9321765303611755, + "learning_rate": 4.87554937463653e-06, + "loss": 0.8027, + "step": 476 + }, + { + "epoch": 0.7468684759916493, + "grad_norm": 0.9527666568756104, + "learning_rate": 4.874892107463811e-06, + "loss": 0.8069, + "step": 477 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 0.9576907157897949, + "learning_rate": 4.874233153769924e-06, + "loss": 0.774, + "step": 478 + }, + { + "epoch": 0.75, + "grad_norm": 1.0516161918640137, + "learning_rate": 4.873572514022822e-06, + "loss": 0.8176, + "step": 479 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.026928186416626, + "learning_rate": 4.872910188691657e-06, + "loss": 0.8797, + "step": 480 + }, + { + "epoch": 0.7531315240083507, + "grad_norm": 0.9312179088592529, + "learning_rate": 4.872246178246776e-06, + "loss": 0.7747, + "step": 481 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0676255226135254, + "learning_rate": 4.871580483159725e-06, + "loss": 0.7801, + "step": 482 + }, + { + "epoch": 0.7562630480167014, + "grad_norm": 1.066056489944458, + "learning_rate": 4.870913103903245e-06, + "loss": 0.7837, + "step": 483 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9388124346733093, + "learning_rate": 4.87024404095127e-06, + "loss": 0.7801, + "step": 484 + }, + { + "epoch": 0.7593945720250522, + "grad_norm": 0.9879791140556335, + "learning_rate": 4.8695732947789356e-06, + "loss": 0.7849, + "step": 485 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0002318620681763, + "learning_rate": 4.868900865862569e-06, + "loss": 0.8219, + "step": 486 + }, + { + "epoch": 0.7625260960334029, + "grad_norm": 0.9599125385284424, + "learning_rate": 4.8682267546796945e-06, + "loss": 0.8888, + "step": 487 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 0.9099973440170288, + "learning_rate": 4.867550961709027e-06, + "loss": 0.7585, + "step": 488 + }, + { + "epoch": 0.7656576200417536, + "grad_norm": 0.9715955257415771, + "learning_rate": 4.866873487430481e-06, + "loss": 0.8237, + "step": 489 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 1.0194706916809082, + "learning_rate": 4.86619433232516e-06, + "loss": 0.7708, + "step": 490 + }, + { + "epoch": 0.7687891440501043, + "grad_norm": 1.044005036354065, + "learning_rate": 4.865513496875367e-06, + "loss": 0.8, + "step": 491 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.9776361584663391, + "learning_rate": 4.8648309815645915e-06, + "loss": 0.7696, + "step": 492 + }, + { + "epoch": 0.7719206680584552, + "grad_norm": 0.9213030934333801, + "learning_rate": 4.864146786877521e-06, + "loss": 0.8558, + "step": 493 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.9361440539360046, + "learning_rate": 4.8634609133000336e-06, + "loss": 0.7851, + "step": 494 + }, + { + "epoch": 0.7750521920668059, + "grad_norm": 1.0199540853500366, + "learning_rate": 4.8627733613192e-06, + "loss": 0.8229, + "step": 495 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 1.0244604349136353, + "learning_rate": 4.862084131423283e-06, + "loss": 0.772, + "step": 496 + }, + { + "epoch": 0.7781837160751566, + "grad_norm": 0.9414368271827698, + "learning_rate": 4.861393224101736e-06, + "loss": 0.7579, + "step": 497 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9198965430259705, + "learning_rate": 4.860700639845205e-06, + "loss": 0.855, + "step": 498 + }, + { + "epoch": 0.7813152400835073, + "grad_norm": 0.9398712515830994, + "learning_rate": 4.8600063791455275e-06, + "loss": 0.7824, + "step": 499 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.9544969201087952, + "learning_rate": 4.8593104424957275e-06, + "loss": 0.8104, + "step": 500 + }, + { + "epoch": 0.784446764091858, + "grad_norm": 0.9522370100021362, + "learning_rate": 4.858612830390024e-06, + "loss": 0.813, + "step": 501 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.9830234050750732, + "learning_rate": 4.857913543323823e-06, + "loss": 0.8656, + "step": 502 + }, + { + "epoch": 0.7875782881002088, + "grad_norm": 1.0527008771896362, + "learning_rate": 4.8572125817937195e-06, + "loss": 0.8417, + "step": 503 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 0.9772782921791077, + "learning_rate": 4.856509946297501e-06, + "loss": 0.7983, + "step": 504 + }, + { + "epoch": 0.7907098121085595, + "grad_norm": 0.9250578880310059, + "learning_rate": 4.855805637334141e-06, + "loss": 0.7391, + "step": 505 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9792859554290771, + "learning_rate": 4.855099655403802e-06, + "loss": 0.7753, + "step": 506 + }, + { + "epoch": 0.7938413361169102, + "grad_norm": 0.8876479864120483, + "learning_rate": 4.854392001007834e-06, + "loss": 0.8249, + "step": 507 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.9336771368980408, + "learning_rate": 4.853682674648775e-06, + "loss": 0.8462, + "step": 508 + }, + { + "epoch": 0.7969728601252609, + "grad_norm": 0.9732388854026794, + "learning_rate": 4.852971676830352e-06, + "loss": 0.8791, + "step": 509 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.9135870337486267, + "learning_rate": 4.852259008057476e-06, + "loss": 0.7279, + "step": 510 + }, + { + "epoch": 0.8001043841336117, + "grad_norm": 0.9782464504241943, + "learning_rate": 4.8515446688362476e-06, + "loss": 0.7797, + "step": 511 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 0.9571824073791504, + "learning_rate": 4.85082865967395e-06, + "loss": 0.8349, + "step": 512 + }, + { + "epoch": 0.8032359081419624, + "grad_norm": 1.011786937713623, + "learning_rate": 4.850110981079057e-06, + "loss": 0.8057, + "step": 513 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 0.9418787956237793, + "learning_rate": 4.8493916335612255e-06, + "loss": 0.8592, + "step": 514 + }, + { + "epoch": 0.8063674321503131, + "grad_norm": 0.9292604923248291, + "learning_rate": 4.848670617631296e-06, + "loss": 0.7471, + "step": 515 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.9303188323974609, + "learning_rate": 4.847947933801296e-06, + "loss": 0.8768, + "step": 516 + }, + { + "epoch": 0.8094989561586639, + "grad_norm": 0.9897359609603882, + "learning_rate": 4.847223582584437e-06, + "loss": 0.8343, + "step": 517 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.9695891737937927, + "learning_rate": 4.846497564495114e-06, + "loss": 0.8318, + "step": 518 + }, + { + "epoch": 0.8126304801670147, + "grad_norm": 1.0175273418426514, + "learning_rate": 4.845769880048906e-06, + "loss": 0.9137, + "step": 519 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9497922658920288, + "learning_rate": 4.845040529762577e-06, + "loss": 0.6549, + "step": 520 + }, + { + "epoch": 0.8157620041753654, + "grad_norm": 1.0647574663162231, + "learning_rate": 4.844309514154071e-06, + "loss": 0.8243, + "step": 521 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.9710965156555176, + "learning_rate": 4.843576833742517e-06, + "loss": 0.7864, + "step": 522 + }, + { + "epoch": 0.8188935281837161, + "grad_norm": 0.9881441593170166, + "learning_rate": 4.842842489048225e-06, + "loss": 0.7799, + "step": 523 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 0.9455597400665283, + "learning_rate": 4.842106480592687e-06, + "loss": 0.7943, + "step": 524 + }, + { + "epoch": 0.8220250521920668, + "grad_norm": 0.9948442578315735, + "learning_rate": 4.841368808898577e-06, + "loss": 0.8587, + "step": 525 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 0.9959186315536499, + "learning_rate": 4.84062947448975e-06, + "loss": 0.8477, + "step": 526 + }, + { + "epoch": 0.8251565762004175, + "grad_norm": 0.9791848659515381, + "learning_rate": 4.839888477891243e-06, + "loss": 0.7568, + "step": 527 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 1.1499426364898682, + "learning_rate": 4.839145819629269e-06, + "loss": 0.8296, + "step": 528 + }, + { + "epoch": 0.8282881002087683, + "grad_norm": 0.9269506335258484, + "learning_rate": 4.8384015002312265e-06, + "loss": 0.7935, + "step": 529 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.970565140247345, + "learning_rate": 4.837655520225692e-06, + "loss": 0.8138, + "step": 530 + }, + { + "epoch": 0.831419624217119, + "grad_norm": 0.9852119088172913, + "learning_rate": 4.836907880142418e-06, + "loss": 0.7909, + "step": 531 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.927129328250885, + "learning_rate": 4.836158580512339e-06, + "loss": 0.8105, + "step": 532 + }, + { + "epoch": 0.8345511482254697, + "grad_norm": 1.0711781978607178, + "learning_rate": 4.835407621867569e-06, + "loss": 0.8241, + "step": 533 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 1.0472421646118164, + "learning_rate": 4.834655004741397e-06, + "loss": 0.7923, + "step": 534 + }, + { + "epoch": 0.8376826722338204, + "grad_norm": 0.9761579632759094, + "learning_rate": 4.833900729668293e-06, + "loss": 0.8203, + "step": 535 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 1.0559637546539307, + "learning_rate": 4.833144797183902e-06, + "loss": 0.8915, + "step": 536 + }, + { + "epoch": 0.8408141962421712, + "grad_norm": 0.9951052069664001, + "learning_rate": 4.832387207825045e-06, + "loss": 0.8253, + "step": 537 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.894079864025116, + "learning_rate": 4.831627962129725e-06, + "loss": 0.7479, + "step": 538 + }, + { + "epoch": 0.843945720250522, + "grad_norm": 0.9832531213760376, + "learning_rate": 4.830867060637115e-06, + "loss": 0.8484, + "step": 539 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0477591753005981, + "learning_rate": 4.8301045038875665e-06, + "loss": 0.8052, + "step": 540 + }, + { + "epoch": 0.8470772442588727, + "grad_norm": 0.9388875365257263, + "learning_rate": 4.829340292422607e-06, + "loss": 0.7998, + "step": 541 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 0.8975532054901123, + "learning_rate": 4.8285744267849395e-06, + "loss": 0.848, + "step": 542 + }, + { + "epoch": 0.8502087682672234, + "grad_norm": 0.9211076498031616, + "learning_rate": 4.8278069075184396e-06, + "loss": 0.8337, + "step": 543 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.970134437084198, + "learning_rate": 4.827037735168159e-06, + "loss": 0.7843, + "step": 544 + }, + { + "epoch": 0.8533402922755741, + "grad_norm": 0.9714353084564209, + "learning_rate": 4.826266910280322e-06, + "loss": 0.6991, + "step": 545 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 0.8607689142227173, + "learning_rate": 4.8254944334023265e-06, + "loss": 0.7314, + "step": 546 + }, + { + "epoch": 0.8564718162839249, + "grad_norm": 0.9843030571937561, + "learning_rate": 4.824720305082747e-06, + "loss": 0.7497, + "step": 547 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 0.9638474583625793, + "learning_rate": 4.823944525871324e-06, + "loss": 0.776, + "step": 548 + }, + { + "epoch": 0.8596033402922756, + "grad_norm": 0.9342365860939026, + "learning_rate": 4.823167096318979e-06, + "loss": 0.77, + "step": 549 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9608579874038696, + "learning_rate": 4.822388016977797e-06, + "loss": 0.8281, + "step": 550 + }, + { + "epoch": 0.8627348643006263, + "grad_norm": 0.8996383547782898, + "learning_rate": 4.8216072884010395e-06, + "loss": 0.7519, + "step": 551 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 0.9922810196876526, + "learning_rate": 4.820824911143139e-06, + "loss": 0.8714, + "step": 552 + }, + { + "epoch": 0.865866388308977, + "grad_norm": 0.9861842393875122, + "learning_rate": 4.820040885759697e-06, + "loss": 0.7998, + "step": 553 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 1.013168215751648, + "learning_rate": 4.819255212807486e-06, + "loss": 0.8291, + "step": 554 + }, + { + "epoch": 0.8689979123173278, + "grad_norm": 0.9695484638214111, + "learning_rate": 4.818467892844451e-06, + "loss": 0.8257, + "step": 555 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 1.0018948316574097, + "learning_rate": 4.817678926429702e-06, + "loss": 0.7551, + "step": 556 + }, + { + "epoch": 0.8721294363256785, + "grad_norm": 1.0259422063827515, + "learning_rate": 4.816888314123521e-06, + "loss": 0.8582, + "step": 557 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9045076966285706, + "learning_rate": 4.81609605648736e-06, + "loss": 0.811, + "step": 558 + }, + { + "epoch": 0.8752609603340292, + "grad_norm": 0.9112316370010376, + "learning_rate": 4.815302154083837e-06, + "loss": 0.818, + "step": 559 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0060032606124878, + "learning_rate": 4.814506607476739e-06, + "loss": 0.766, + "step": 560 + }, + { + "epoch": 0.8783924843423799, + "grad_norm": 0.9126970171928406, + "learning_rate": 4.813709417231021e-06, + "loss": 0.8903, + "step": 561 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.0825906991958618, + "learning_rate": 4.812910583912804e-06, + "loss": 0.8466, + "step": 562 + }, + { + "epoch": 0.8815240083507306, + "grad_norm": 0.9530975818634033, + "learning_rate": 4.812110108089377e-06, + "loss": 0.8163, + "step": 563 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.9566405415534973, + "learning_rate": 4.8113079903291955e-06, + "loss": 0.8244, + "step": 564 + }, + { + "epoch": 0.8846555323590815, + "grad_norm": 0.919980525970459, + "learning_rate": 4.81050423120188e-06, + "loss": 0.8335, + "step": 565 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.9123868346214294, + "learning_rate": 4.809698831278217e-06, + "loss": 0.7712, + "step": 566 + }, + { + "epoch": 0.8877870563674322, + "grad_norm": 0.9589663743972778, + "learning_rate": 4.8088917911301595e-06, + "loss": 0.8467, + "step": 567 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 0.9052798748016357, + "learning_rate": 4.808083111330823e-06, + "loss": 0.8078, + "step": 568 + }, + { + "epoch": 0.8909185803757829, + "grad_norm": 0.9250494241714478, + "learning_rate": 4.807272792454489e-06, + "loss": 0.8536, + "step": 569 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.8451298475265503, + "learning_rate": 4.806460835076603e-06, + "loss": 0.7752, + "step": 570 + }, + { + "epoch": 0.8940501043841336, + "grad_norm": 1.0324240922927856, + "learning_rate": 4.805647239773773e-06, + "loss": 0.7871, + "step": 571 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 0.9338528513908386, + "learning_rate": 4.804832007123771e-06, + "loss": 0.8082, + "step": 572 + }, + { + "epoch": 0.8971816283924844, + "grad_norm": 1.0080739259719849, + "learning_rate": 4.804015137705531e-06, + "loss": 0.7636, + "step": 573 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9817036986351013, + "learning_rate": 4.803196632099152e-06, + "loss": 0.8599, + "step": 574 + }, + { + "epoch": 0.9003131524008351, + "grad_norm": 0.9284219145774841, + "learning_rate": 4.80237649088589e-06, + "loss": 0.8152, + "step": 575 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.8922317028045654, + "learning_rate": 4.801554714648166e-06, + "loss": 0.8146, + "step": 576 + }, + { + "epoch": 0.9034446764091858, + "grad_norm": 0.9465014338493347, + "learning_rate": 4.800731303969565e-06, + "loss": 0.7663, + "step": 577 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.0325525999069214, + "learning_rate": 4.799906259434824e-06, + "loss": 0.8458, + "step": 578 + }, + { + "epoch": 0.9065762004175365, + "grad_norm": 0.9366307258605957, + "learning_rate": 4.7990795816298485e-06, + "loss": 0.7992, + "step": 579 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 0.9923847317695618, + "learning_rate": 4.7982512711416995e-06, + "loss": 0.7256, + "step": 580 + }, + { + "epoch": 0.9097077244258872, + "grad_norm": 1.0387133359909058, + "learning_rate": 4.7974213285586e-06, + "loss": 0.7766, + "step": 581 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.9617043733596802, + "learning_rate": 4.79658975446993e-06, + "loss": 0.7602, + "step": 582 + }, + { + "epoch": 0.912839248434238, + "grad_norm": 0.9530353546142578, + "learning_rate": 4.795756549466229e-06, + "loss": 0.7922, + "step": 583 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.9305918216705322, + "learning_rate": 4.794921714139194e-06, + "loss": 0.8377, + "step": 584 + }, + { + "epoch": 0.9159707724425887, + "grad_norm": 0.899860143661499, + "learning_rate": 4.794085249081682e-06, + "loss": 0.8343, + "step": 585 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.9024940729141235, + "learning_rate": 4.793247154887703e-06, + "loss": 0.8271, + "step": 586 + }, + { + "epoch": 0.9191022964509394, + "grad_norm": 0.980869472026825, + "learning_rate": 4.7924074321524286e-06, + "loss": 0.7468, + "step": 587 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.0506541728973389, + "learning_rate": 4.791566081472185e-06, + "loss": 0.9207, + "step": 588 + }, + { + "epoch": 0.9222338204592901, + "grad_norm": 1.0723917484283447, + "learning_rate": 4.790723103444454e-06, + "loss": 0.7615, + "step": 589 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 1.0455081462860107, + "learning_rate": 4.789878498667873e-06, + "loss": 0.8115, + "step": 590 + }, + { + "epoch": 0.925365344467641, + "grad_norm": 0.9729639291763306, + "learning_rate": 4.789032267742234e-06, + "loss": 0.8622, + "step": 591 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.9695239067077637, + "learning_rate": 4.788184411268488e-06, + "loss": 0.8165, + "step": 592 + }, + { + "epoch": 0.9284968684759917, + "grad_norm": 1.0099278688430786, + "learning_rate": 4.7873349298487345e-06, + "loss": 0.759, + "step": 593 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 0.9377149939537048, + "learning_rate": 4.786483824086231e-06, + "loss": 0.8323, + "step": 594 + }, + { + "epoch": 0.9316283924843424, + "grad_norm": 0.8885690569877625, + "learning_rate": 4.785631094585387e-06, + "loss": 0.8448, + "step": 595 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 0.9133430123329163, + "learning_rate": 4.784776741951766e-06, + "loss": 0.8135, + "step": 596 + }, + { + "epoch": 0.9347599164926931, + "grad_norm": 1.015851378440857, + "learning_rate": 4.783920766792082e-06, + "loss": 0.8464, + "step": 597 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.8883825540542603, + "learning_rate": 4.783063169714203e-06, + "loss": 0.7445, + "step": 598 + }, + { + "epoch": 0.9378914405010439, + "grad_norm": 0.9872002005577087, + "learning_rate": 4.782203951327149e-06, + "loss": 0.8293, + "step": 599 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 1.0305813550949097, + "learning_rate": 4.781343112241091e-06, + "loss": 0.8606, + "step": 600 + }, + { + "epoch": 0.9410229645093946, + "grad_norm": 0.9667104482650757, + "learning_rate": 4.780480653067351e-06, + "loss": 0.8057, + "step": 601 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 1.067185640335083, + "learning_rate": 4.7796165744184e-06, + "loss": 0.849, + "step": 602 + }, + { + "epoch": 0.9441544885177453, + "grad_norm": 0.9017406105995178, + "learning_rate": 4.778750876907862e-06, + "loss": 0.7927, + "step": 603 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.9827666878700256, + "learning_rate": 4.777883561150508e-06, + "loss": 0.8087, + "step": 604 + }, + { + "epoch": 0.947286012526096, + "grad_norm": 0.9627811908721924, + "learning_rate": 4.77701462776226e-06, + "loss": 0.8488, + "step": 605 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0321338176727295, + "learning_rate": 4.776144077360186e-06, + "loss": 0.8289, + "step": 606 + }, + { + "epoch": 0.9504175365344467, + "grad_norm": 1.0012633800506592, + "learning_rate": 4.775271910562508e-06, + "loss": 0.7184, + "step": 607 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.9186056852340698, + "learning_rate": 4.7743981279885875e-06, + "loss": 0.8727, + "step": 608 + }, + { + "epoch": 0.9535490605427975, + "grad_norm": 0.910044252872467, + "learning_rate": 4.773522730258942e-06, + "loss": 0.8455, + "step": 609 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.9342944025993347, + "learning_rate": 4.77264571799523e-06, + "loss": 0.7548, + "step": 610 + }, + { + "epoch": 0.9566805845511482, + "grad_norm": 0.9609628319740295, + "learning_rate": 4.771767091820259e-06, + "loss": 0.8294, + "step": 611 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9708508253097534, + "learning_rate": 4.770886852357983e-06, + "loss": 0.7857, + "step": 612 + }, + { + "epoch": 0.9598121085594989, + "grad_norm": 1.0910807847976685, + "learning_rate": 4.7700050002334995e-06, + "loss": 0.8162, + "step": 613 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 0.9703771471977234, + "learning_rate": 4.769121536073054e-06, + "loss": 0.7205, + "step": 614 + }, + { + "epoch": 0.9629436325678496, + "grad_norm": 1.027938961982727, + "learning_rate": 4.768236460504035e-06, + "loss": 0.7641, + "step": 615 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 1.0003989934921265, + "learning_rate": 4.767349774154974e-06, + "loss": 0.7588, + "step": 616 + }, + { + "epoch": 0.9660751565762005, + "grad_norm": 0.9572570323944092, + "learning_rate": 4.766461477655552e-06, + "loss": 0.8274, + "step": 617 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.977287769317627, + "learning_rate": 4.765571571636586e-06, + "loss": 0.7856, + "step": 618 + }, + { + "epoch": 0.9692066805845512, + "grad_norm": 0.9582562446594238, + "learning_rate": 4.764680056730042e-06, + "loss": 0.8102, + "step": 619 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.9842986464500427, + "learning_rate": 4.763786933569025e-06, + "loss": 0.7515, + "step": 620 + }, + { + "epoch": 0.9723382045929019, + "grad_norm": 0.9955743551254272, + "learning_rate": 4.762892202787782e-06, + "loss": 0.7398, + "step": 621 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 1.0011107921600342, + "learning_rate": 4.761995865021706e-06, + "loss": 0.8182, + "step": 622 + }, + { + "epoch": 0.9754697286012526, + "grad_norm": 0.9586448669433594, + "learning_rate": 4.761097920907325e-06, + "loss": 0.8434, + "step": 623 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.979123055934906, + "learning_rate": 4.760198371082312e-06, + "loss": 0.7954, + "step": 624 + }, + { + "epoch": 0.9786012526096033, + "grad_norm": 1.0081250667572021, + "learning_rate": 4.75929721618548e-06, + "loss": 0.7931, + "step": 625 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 1.027185320854187, + "learning_rate": 4.75839445685678e-06, + "loss": 0.7339, + "step": 626 + }, + { + "epoch": 0.9817327766179541, + "grad_norm": 1.0553215742111206, + "learning_rate": 4.757490093737303e-06, + "loss": 0.7639, + "step": 627 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 0.8578981757164001, + "learning_rate": 4.75658412746928e-06, + "loss": 0.8128, + "step": 628 + }, + { + "epoch": 0.9848643006263048, + "grad_norm": 1.0143071413040161, + "learning_rate": 4.75567655869608e-06, + "loss": 0.8169, + "step": 629 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 0.9743554592132568, + "learning_rate": 4.754767388062208e-06, + "loss": 0.8179, + "step": 630 + }, + { + "epoch": 0.9879958246346555, + "grad_norm": 1.0276602506637573, + "learning_rate": 4.7538566162133105e-06, + "loss": 0.8009, + "step": 631 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.9296634793281555, + "learning_rate": 4.752944243796167e-06, + "loss": 0.8262, + "step": 632 + }, + { + "epoch": 0.9911273486430062, + "grad_norm": 0.9216791987419128, + "learning_rate": 4.752030271458696e-06, + "loss": 0.8929, + "step": 633 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 0.9272832274436951, + "learning_rate": 4.751114699849954e-06, + "loss": 0.7396, + "step": 634 + }, + { + "epoch": 0.994258872651357, + "grad_norm": 0.9121203422546387, + "learning_rate": 4.750197529620127e-06, + "loss": 0.8311, + "step": 635 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.999600887298584, + "learning_rate": 4.7492787614205425e-06, + "loss": 0.8125, + "step": 636 + }, + { + "epoch": 0.9973903966597077, + "grad_norm": 1.012978196144104, + "learning_rate": 4.748358395903661e-06, + "loss": 0.7966, + "step": 637 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.9767918586730957, + "learning_rate": 4.747436433723075e-06, + "loss": 0.8024, + "step": 638 + }, + { + "epoch": 1.0015657620041754, + "grad_norm": 1.8302738666534424, + "learning_rate": 4.7465128755335135e-06, + "loss": 1.6427, + "step": 639 + }, + { + "epoch": 1.0031315240083507, + "grad_norm": 0.8704890012741089, + "learning_rate": 4.7455877219908386e-06, + "loss": 0.7766, + "step": 640 + }, + { + "epoch": 1.004697286012526, + "grad_norm": 0.9848434925079346, + "learning_rate": 4.744660973752044e-06, + "loss": 0.8314, + "step": 641 + }, + { + "epoch": 1.0062630480167014, + "grad_norm": 1.054824709892273, + "learning_rate": 4.743732631475258e-06, + "loss": 0.797, + "step": 642 + }, + { + "epoch": 1.0078288100208768, + "grad_norm": 0.9286114573478699, + "learning_rate": 4.742802695819737e-06, + "loss": 0.764, + "step": 643 + }, + { + "epoch": 1.0093945720250521, + "grad_norm": 0.8625781536102295, + "learning_rate": 4.7418711674458735e-06, + "loss": 0.6691, + "step": 644 + }, + { + "epoch": 1.0109603340292275, + "grad_norm": 0.8710452914237976, + "learning_rate": 4.740938047015188e-06, + "loss": 0.7277, + "step": 645 + }, + { + "epoch": 1.0125260960334028, + "grad_norm": 0.8873130679130554, + "learning_rate": 4.740003335190334e-06, + "loss": 0.6531, + "step": 646 + }, + { + "epoch": 1.0140918580375784, + "grad_norm": 0.9777164459228516, + "learning_rate": 4.7390670326350925e-06, + "loss": 0.8047, + "step": 647 + }, + { + "epoch": 1.0156576200417538, + "grad_norm": 0.886025071144104, + "learning_rate": 4.738129140014377e-06, + "loss": 0.715, + "step": 648 + }, + { + "epoch": 1.017223382045929, + "grad_norm": 0.9825332164764404, + "learning_rate": 4.737189657994226e-06, + "loss": 0.7783, + "step": 649 + }, + { + "epoch": 1.0187891440501045, + "grad_norm": 0.9331428408622742, + "learning_rate": 4.736248587241812e-06, + "loss": 0.7598, + "step": 650 + }, + { + "epoch": 1.0203549060542798, + "grad_norm": 0.9406535625457764, + "learning_rate": 4.735305928425431e-06, + "loss": 0.7989, + "step": 651 + }, + { + "epoch": 1.0219206680584552, + "grad_norm": 1.0644745826721191, + "learning_rate": 4.734361682214511e-06, + "loss": 0.8135, + "step": 652 + }, + { + "epoch": 1.0234864300626305, + "grad_norm": 0.9037052392959595, + "learning_rate": 4.733415849279603e-06, + "loss": 0.7051, + "step": 653 + }, + { + "epoch": 1.0250521920668059, + "grad_norm": 1.000563144683838, + "learning_rate": 4.7324684302923865e-06, + "loss": 0.7427, + "step": 654 + }, + { + "epoch": 1.0266179540709812, + "grad_norm": 0.9181354641914368, + "learning_rate": 4.731519425925671e-06, + "loss": 0.7236, + "step": 655 + }, + { + "epoch": 1.0281837160751566, + "grad_norm": 1.0283046960830688, + "learning_rate": 4.730568836853384e-06, + "loss": 0.7998, + "step": 656 + }, + { + "epoch": 1.029749478079332, + "grad_norm": 0.8692315220832825, + "learning_rate": 4.729616663750587e-06, + "loss": 0.7974, + "step": 657 + }, + { + "epoch": 1.0313152400835073, + "grad_norm": 0.928920567035675, + "learning_rate": 4.72866290729346e-06, + "loss": 0.7199, + "step": 658 + }, + { + "epoch": 1.0328810020876826, + "grad_norm": 0.9066957235336304, + "learning_rate": 4.727707568159311e-06, + "loss": 0.791, + "step": 659 + }, + { + "epoch": 1.034446764091858, + "grad_norm": 1.0086729526519775, + "learning_rate": 4.726750647026569e-06, + "loss": 0.7113, + "step": 660 + }, + { + "epoch": 1.0360125260960333, + "grad_norm": 0.931856632232666, + "learning_rate": 4.72579214457479e-06, + "loss": 0.7977, + "step": 661 + }, + { + "epoch": 1.0375782881002087, + "grad_norm": 0.9191203713417053, + "learning_rate": 4.724832061484651e-06, + "loss": 0.7979, + "step": 662 + }, + { + "epoch": 1.039144050104384, + "grad_norm": 0.9614707827568054, + "learning_rate": 4.72387039843795e-06, + "loss": 0.6994, + "step": 663 + }, + { + "epoch": 1.0407098121085594, + "grad_norm": 1.014823317527771, + "learning_rate": 4.72290715611761e-06, + "loss": 0.7539, + "step": 664 + }, + { + "epoch": 1.042275574112735, + "grad_norm": 0.9177424311637878, + "learning_rate": 4.721942335207673e-06, + "loss": 0.6625, + "step": 665 + }, + { + "epoch": 1.0438413361169103, + "grad_norm": 1.0253031253814697, + "learning_rate": 4.720975936393305e-06, + "loss": 0.7747, + "step": 666 + }, + { + "epoch": 1.0454070981210857, + "grad_norm": 1.0004568099975586, + "learning_rate": 4.720007960360788e-06, + "loss": 0.7008, + "step": 667 + }, + { + "epoch": 1.046972860125261, + "grad_norm": 1.1344349384307861, + "learning_rate": 4.719038407797529e-06, + "loss": 0.7493, + "step": 668 + }, + { + "epoch": 1.0485386221294364, + "grad_norm": 0.9944799542427063, + "learning_rate": 4.718067279392052e-06, + "loss": 0.7823, + "step": 669 + }, + { + "epoch": 1.0501043841336117, + "grad_norm": 0.9469346404075623, + "learning_rate": 4.717094575833999e-06, + "loss": 0.7419, + "step": 670 + }, + { + "epoch": 1.051670146137787, + "grad_norm": 1.015339732170105, + "learning_rate": 4.716120297814133e-06, + "loss": 0.7168, + "step": 671 + }, + { + "epoch": 1.0532359081419624, + "grad_norm": 0.9544686675071716, + "learning_rate": 4.715144446024333e-06, + "loss": 0.791, + "step": 672 + }, + { + "epoch": 1.0548016701461378, + "grad_norm": 1.0285652875900269, + "learning_rate": 4.714167021157599e-06, + "loss": 0.7649, + "step": 673 + }, + { + "epoch": 1.0563674321503131, + "grad_norm": 0.9366872906684875, + "learning_rate": 4.713188023908044e-06, + "loss": 0.7524, + "step": 674 + }, + { + "epoch": 1.0579331941544885, + "grad_norm": 1.0348883867263794, + "learning_rate": 4.712207454970899e-06, + "loss": 0.6965, + "step": 675 + }, + { + "epoch": 1.0594989561586639, + "grad_norm": 0.8716043829917908, + "learning_rate": 4.711225315042513e-06, + "loss": 0.6264, + "step": 676 + }, + { + "epoch": 1.0610647181628392, + "grad_norm": 1.0421313047409058, + "learning_rate": 4.710241604820348e-06, + "loss": 0.7548, + "step": 677 + }, + { + "epoch": 1.0626304801670146, + "grad_norm": 1.0886270999908447, + "learning_rate": 4.709256325002983e-06, + "loss": 0.6352, + "step": 678 + }, + { + "epoch": 1.06419624217119, + "grad_norm": 0.9752539396286011, + "learning_rate": 4.708269476290111e-06, + "loss": 0.7764, + "step": 679 + }, + { + "epoch": 1.0657620041753653, + "grad_norm": 1.0658339262008667, + "learning_rate": 4.7072810593825385e-06, + "loss": 0.766, + "step": 680 + }, + { + "epoch": 1.0673277661795406, + "grad_norm": 1.0299617052078247, + "learning_rate": 4.706291074982187e-06, + "loss": 0.7558, + "step": 681 + }, + { + "epoch": 1.068893528183716, + "grad_norm": 0.9171326160430908, + "learning_rate": 4.705299523792089e-06, + "loss": 0.7084, + "step": 682 + }, + { + "epoch": 1.0704592901878915, + "grad_norm": 0.9924243688583374, + "learning_rate": 4.7043064065163925e-06, + "loss": 0.8393, + "step": 683 + }, + { + "epoch": 1.072025052192067, + "grad_norm": 1.0243877172470093, + "learning_rate": 4.703311723860356e-06, + "loss": 0.7755, + "step": 684 + }, + { + "epoch": 1.0735908141962422, + "grad_norm": 1.0019137859344482, + "learning_rate": 4.702315476530349e-06, + "loss": 0.7254, + "step": 685 + }, + { + "epoch": 1.0751565762004176, + "grad_norm": 0.8541996479034424, + "learning_rate": 4.701317665233853e-06, + "loss": 0.7389, + "step": 686 + }, + { + "epoch": 1.076722338204593, + "grad_norm": 1.1119922399520874, + "learning_rate": 4.700318290679462e-06, + "loss": 0.7803, + "step": 687 + }, + { + "epoch": 1.0782881002087683, + "grad_norm": 1.078452706336975, + "learning_rate": 4.699317353576876e-06, + "loss": 0.756, + "step": 688 + }, + { + "epoch": 1.0798538622129437, + "grad_norm": 1.0079457759857178, + "learning_rate": 4.698314854636908e-06, + "loss": 0.7365, + "step": 689 + }, + { + "epoch": 1.081419624217119, + "grad_norm": 0.931423544883728, + "learning_rate": 4.6973107945714795e-06, + "loss": 0.7881, + "step": 690 + }, + { + "epoch": 1.0829853862212944, + "grad_norm": 1.0135142803192139, + "learning_rate": 4.696305174093619e-06, + "loss": 0.7884, + "step": 691 + }, + { + "epoch": 1.0845511482254697, + "grad_norm": 1.1194610595703125, + "learning_rate": 4.695297993917465e-06, + "loss": 0.7969, + "step": 692 + }, + { + "epoch": 1.086116910229645, + "grad_norm": 0.9972780346870422, + "learning_rate": 4.6942892547582625e-06, + "loss": 0.5873, + "step": 693 + }, + { + "epoch": 1.0876826722338204, + "grad_norm": 1.0711641311645508, + "learning_rate": 4.693278957332364e-06, + "loss": 0.7273, + "step": 694 + }, + { + "epoch": 1.0892484342379958, + "grad_norm": 0.9708715081214905, + "learning_rate": 4.6922671023572305e-06, + "loss": 0.6786, + "step": 695 + }, + { + "epoch": 1.0908141962421711, + "grad_norm": 0.9773644804954529, + "learning_rate": 4.691253690551424e-06, + "loss": 0.7839, + "step": 696 + }, + { + "epoch": 1.0923799582463465, + "grad_norm": 1.039009690284729, + "learning_rate": 4.690238722634618e-06, + "loss": 0.6807, + "step": 697 + }, + { + "epoch": 1.0939457202505218, + "grad_norm": 1.140777349472046, + "learning_rate": 4.689222199327586e-06, + "loss": 0.7506, + "step": 698 + }, + { + "epoch": 1.0955114822546972, + "grad_norm": 0.9779882431030273, + "learning_rate": 4.688204121352211e-06, + "loss": 0.7732, + "step": 699 + }, + { + "epoch": 1.0970772442588728, + "grad_norm": 1.0412248373031616, + "learning_rate": 4.687184489431476e-06, + "loss": 0.7034, + "step": 700 + }, + { + "epoch": 1.0986430062630481, + "grad_norm": 0.9802400469779968, + "learning_rate": 4.686163304289469e-06, + "loss": 0.729, + "step": 701 + }, + { + "epoch": 1.1002087682672235, + "grad_norm": 1.0531584024429321, + "learning_rate": 4.6851405666513804e-06, + "loss": 0.7635, + "step": 702 + }, + { + "epoch": 1.1017745302713988, + "grad_norm": 0.983923077583313, + "learning_rate": 4.684116277243505e-06, + "loss": 0.7062, + "step": 703 + }, + { + "epoch": 1.1033402922755742, + "grad_norm": 0.9886301159858704, + "learning_rate": 4.683090436793238e-06, + "loss": 0.7299, + "step": 704 + }, + { + "epoch": 1.1049060542797495, + "grad_norm": 1.0247817039489746, + "learning_rate": 4.682063046029075e-06, + "loss": 0.729, + "step": 705 + }, + { + "epoch": 1.1064718162839249, + "grad_norm": 0.9860950708389282, + "learning_rate": 4.681034105680616e-06, + "loss": 0.7392, + "step": 706 + }, + { + "epoch": 1.1080375782881002, + "grad_norm": 0.9543380737304688, + "learning_rate": 4.680003616478556e-06, + "loss": 0.7061, + "step": 707 + }, + { + "epoch": 1.1096033402922756, + "grad_norm": 1.0253225564956665, + "learning_rate": 4.678971579154698e-06, + "loss": 0.6577, + "step": 708 + }, + { + "epoch": 1.111169102296451, + "grad_norm": 0.9013616442680359, + "learning_rate": 4.677937994441935e-06, + "loss": 0.7422, + "step": 709 + }, + { + "epoch": 1.1127348643006263, + "grad_norm": 0.9716471433639526, + "learning_rate": 4.676902863074268e-06, + "loss": 0.7353, + "step": 710 + }, + { + "epoch": 1.1143006263048016, + "grad_norm": 1.040898084640503, + "learning_rate": 4.675866185786787e-06, + "loss": 0.7396, + "step": 711 + }, + { + "epoch": 1.115866388308977, + "grad_norm": 0.9931808114051819, + "learning_rate": 4.674827963315688e-06, + "loss": 0.7354, + "step": 712 + }, + { + "epoch": 1.1174321503131524, + "grad_norm": 1.020512580871582, + "learning_rate": 4.6737881963982605e-06, + "loss": 0.7366, + "step": 713 + }, + { + "epoch": 1.1189979123173277, + "grad_norm": 1.0201045274734497, + "learning_rate": 4.67274688577289e-06, + "loss": 0.7907, + "step": 714 + }, + { + "epoch": 1.120563674321503, + "grad_norm": 1.0163429975509644, + "learning_rate": 4.671704032179061e-06, + "loss": 0.7452, + "step": 715 + }, + { + "epoch": 1.1221294363256784, + "grad_norm": 0.9667377471923828, + "learning_rate": 4.670659636357352e-06, + "loss": 0.7983, + "step": 716 + }, + { + "epoch": 1.1236951983298538, + "grad_norm": 1.0788301229476929, + "learning_rate": 4.669613699049436e-06, + "loss": 0.7861, + "step": 717 + }, + { + "epoch": 1.1252609603340291, + "grad_norm": 1.0512142181396484, + "learning_rate": 4.668566220998082e-06, + "loss": 0.7582, + "step": 718 + }, + { + "epoch": 1.1268267223382047, + "grad_norm": 0.9225794076919556, + "learning_rate": 4.667517202947154e-06, + "loss": 0.7801, + "step": 719 + }, + { + "epoch": 1.12839248434238, + "grad_norm": 1.1193747520446777, + "learning_rate": 4.666466645641607e-06, + "loss": 0.8073, + "step": 720 + }, + { + "epoch": 1.1299582463465554, + "grad_norm": 0.934438943862915, + "learning_rate": 4.6654145498274915e-06, + "loss": 0.774, + "step": 721 + }, + { + "epoch": 1.1315240083507307, + "grad_norm": 1.0614968538284302, + "learning_rate": 4.664360916251949e-06, + "loss": 0.7216, + "step": 722 + }, + { + "epoch": 1.133089770354906, + "grad_norm": 1.0807424783706665, + "learning_rate": 4.663305745663212e-06, + "loss": 0.6434, + "step": 723 + }, + { + "epoch": 1.1346555323590815, + "grad_norm": 0.9515354633331299, + "learning_rate": 4.66224903881061e-06, + "loss": 0.7336, + "step": 724 + }, + { + "epoch": 1.1362212943632568, + "grad_norm": 0.9852489233016968, + "learning_rate": 4.661190796444557e-06, + "loss": 0.8156, + "step": 725 + }, + { + "epoch": 1.1377870563674322, + "grad_norm": 1.0071501731872559, + "learning_rate": 4.66013101931656e-06, + "loss": 0.8836, + "step": 726 + }, + { + "epoch": 1.1393528183716075, + "grad_norm": 0.9622960686683655, + "learning_rate": 4.6590697081792176e-06, + "loss": 0.7631, + "step": 727 + }, + { + "epoch": 1.1409185803757829, + "grad_norm": 1.0588105916976929, + "learning_rate": 4.6580068637862144e-06, + "loss": 0.7874, + "step": 728 + }, + { + "epoch": 1.1424843423799582, + "grad_norm": 0.9920447468757629, + "learning_rate": 4.656942486892326e-06, + "loss": 0.7161, + "step": 729 + }, + { + "epoch": 1.1440501043841336, + "grad_norm": 1.0036561489105225, + "learning_rate": 4.655876578253416e-06, + "loss": 0.7116, + "step": 730 + }, + { + "epoch": 1.145615866388309, + "grad_norm": 1.0208396911621094, + "learning_rate": 4.654809138626436e-06, + "loss": 0.818, + "step": 731 + }, + { + "epoch": 1.1471816283924843, + "grad_norm": 1.1095491647720337, + "learning_rate": 4.653740168769424e-06, + "loss": 0.7319, + "step": 732 + }, + { + "epoch": 1.1487473903966596, + "grad_norm": 0.8151328563690186, + "learning_rate": 4.652669669441505e-06, + "loss": 0.774, + "step": 733 + }, + { + "epoch": 1.150313152400835, + "grad_norm": 1.0747915506362915, + "learning_rate": 4.651597641402891e-06, + "loss": 0.8275, + "step": 734 + }, + { + "epoch": 1.1518789144050103, + "grad_norm": 0.9265272617340088, + "learning_rate": 4.650524085414878e-06, + "loss": 0.6946, + "step": 735 + }, + { + "epoch": 1.153444676409186, + "grad_norm": 0.9531672596931458, + "learning_rate": 4.649449002239849e-06, + "loss": 0.7899, + "step": 736 + }, + { + "epoch": 1.1550104384133613, + "grad_norm": 1.0496262311935425, + "learning_rate": 4.648372392641269e-06, + "loss": 0.8237, + "step": 737 + }, + { + "epoch": 1.1565762004175366, + "grad_norm": 0.8967193961143494, + "learning_rate": 4.64729425738369e-06, + "loss": 0.736, + "step": 738 + }, + { + "epoch": 1.158141962421712, + "grad_norm": 1.061038851737976, + "learning_rate": 4.646214597232745e-06, + "loss": 0.7308, + "step": 739 + }, + { + "epoch": 1.1597077244258873, + "grad_norm": 0.9556939601898193, + "learning_rate": 4.64513341295515e-06, + "loss": 0.7526, + "step": 740 + }, + { + "epoch": 1.1612734864300627, + "grad_norm": 1.000093936920166, + "learning_rate": 4.6440507053187075e-06, + "loss": 0.7488, + "step": 741 + }, + { + "epoch": 1.162839248434238, + "grad_norm": 0.9552261233329773, + "learning_rate": 4.642966475092294e-06, + "loss": 0.8291, + "step": 742 + }, + { + "epoch": 1.1644050104384134, + "grad_norm": 1.0215847492218018, + "learning_rate": 4.641880723045875e-06, + "loss": 0.756, + "step": 743 + }, + { + "epoch": 1.1659707724425887, + "grad_norm": 1.0543124675750732, + "learning_rate": 4.640793449950493e-06, + "loss": 0.7668, + "step": 744 + }, + { + "epoch": 1.167536534446764, + "grad_norm": 0.9648609161376953, + "learning_rate": 4.639704656578269e-06, + "loss": 0.6531, + "step": 745 + }, + { + "epoch": 1.1691022964509394, + "grad_norm": 1.0117427110671997, + "learning_rate": 4.638614343702408e-06, + "loss": 0.7084, + "step": 746 + }, + { + "epoch": 1.1706680584551148, + "grad_norm": 0.9475041627883911, + "learning_rate": 4.6375225120971915e-06, + "loss": 0.8578, + "step": 747 + }, + { + "epoch": 1.1722338204592901, + "grad_norm": 1.0122299194335938, + "learning_rate": 4.6364291625379785e-06, + "loss": 0.7541, + "step": 748 + }, + { + "epoch": 1.1737995824634655, + "grad_norm": 0.9487535357475281, + "learning_rate": 4.635334295801208e-06, + "loss": 0.7203, + "step": 749 + }, + { + "epoch": 1.1753653444676408, + "grad_norm": 1.0628844499588013, + "learning_rate": 4.6342379126643965e-06, + "loss": 0.7532, + "step": 750 + }, + { + "epoch": 1.1769311064718162, + "grad_norm": 0.9647781848907471, + "learning_rate": 4.633140013906136e-06, + "loss": 0.5918, + "step": 751 + }, + { + "epoch": 1.1784968684759916, + "grad_norm": 0.9011932611465454, + "learning_rate": 4.632040600306095e-06, + "loss": 0.7395, + "step": 752 + }, + { + "epoch": 1.1800626304801671, + "grad_norm": 1.0524142980575562, + "learning_rate": 4.630939672645018e-06, + "loss": 0.7487, + "step": 753 + }, + { + "epoch": 1.1816283924843423, + "grad_norm": 1.165077567100525, + "learning_rate": 4.6298372317047265e-06, + "loss": 0.6817, + "step": 754 + }, + { + "epoch": 1.1831941544885178, + "grad_norm": 0.9301269054412842, + "learning_rate": 4.628733278268113e-06, + "loss": 0.8446, + "step": 755 + }, + { + "epoch": 1.1847599164926932, + "grad_norm": 0.9762471318244934, + "learning_rate": 4.627627813119147e-06, + "loss": 0.7527, + "step": 756 + }, + { + "epoch": 1.1863256784968685, + "grad_norm": 0.8634734153747559, + "learning_rate": 4.626520837042871e-06, + "loss": 0.7177, + "step": 757 + }, + { + "epoch": 1.187891440501044, + "grad_norm": 1.0592188835144043, + "learning_rate": 4.625412350825399e-06, + "loss": 0.7293, + "step": 758 + }, + { + "epoch": 1.1894572025052192, + "grad_norm": 1.1318559646606445, + "learning_rate": 4.62430235525392e-06, + "loss": 0.6592, + "step": 759 + }, + { + "epoch": 1.1910229645093946, + "grad_norm": 0.9689339399337769, + "learning_rate": 4.623190851116692e-06, + "loss": 0.7667, + "step": 760 + }, + { + "epoch": 1.19258872651357, + "grad_norm": 0.9965885281562805, + "learning_rate": 4.6220778392030464e-06, + "loss": 0.6982, + "step": 761 + }, + { + "epoch": 1.1941544885177453, + "grad_norm": 1.1428629159927368, + "learning_rate": 4.6209633203033846e-06, + "loss": 0.7342, + "step": 762 + }, + { + "epoch": 1.1957202505219207, + "grad_norm": 0.901360809803009, + "learning_rate": 4.6198472952091785e-06, + "loss": 0.8371, + "step": 763 + }, + { + "epoch": 1.197286012526096, + "grad_norm": 0.9882062077522278, + "learning_rate": 4.618729764712969e-06, + "loss": 0.6882, + "step": 764 + }, + { + "epoch": 1.1988517745302714, + "grad_norm": 1.0692517757415771, + "learning_rate": 4.617610729608366e-06, + "loss": 0.7327, + "step": 765 + }, + { + "epoch": 1.2004175365344467, + "grad_norm": 1.1056455373764038, + "learning_rate": 4.616490190690049e-06, + "loss": 0.781, + "step": 766 + }, + { + "epoch": 1.201983298538622, + "grad_norm": 0.9928885698318481, + "learning_rate": 4.615368148753765e-06, + "loss": 0.7463, + "step": 767 + }, + { + "epoch": 1.2035490605427974, + "grad_norm": 1.0006698369979858, + "learning_rate": 4.614244604596327e-06, + "loss": 0.6994, + "step": 768 + }, + { + "epoch": 1.2051148225469728, + "grad_norm": 0.907508909702301, + "learning_rate": 4.613119559015616e-06, + "loss": 0.7117, + "step": 769 + }, + { + "epoch": 1.2066805845511483, + "grad_norm": 1.1037627458572388, + "learning_rate": 4.61199301281058e-06, + "loss": 0.772, + "step": 770 + }, + { + "epoch": 1.2082463465553235, + "grad_norm": 0.9474150538444519, + "learning_rate": 4.610864966781231e-06, + "loss": 0.7601, + "step": 771 + }, + { + "epoch": 1.209812108559499, + "grad_norm": 1.003667950630188, + "learning_rate": 4.609735421728647e-06, + "loss": 0.7141, + "step": 772 + }, + { + "epoch": 1.2113778705636744, + "grad_norm": 0.9724787473678589, + "learning_rate": 4.608604378454971e-06, + "loss": 0.7841, + "step": 773 + }, + { + "epoch": 1.2129436325678498, + "grad_norm": 1.0460760593414307, + "learning_rate": 4.607471837763409e-06, + "loss": 0.7563, + "step": 774 + }, + { + "epoch": 1.2145093945720251, + "grad_norm": 0.9798458218574524, + "learning_rate": 4.606337800458231e-06, + "loss": 0.6812, + "step": 775 + }, + { + "epoch": 1.2160751565762005, + "grad_norm": 1.1637914180755615, + "learning_rate": 4.605202267344768e-06, + "loss": 0.7571, + "step": 776 + }, + { + "epoch": 1.2176409185803758, + "grad_norm": 0.9384608864784241, + "learning_rate": 4.6040652392294175e-06, + "loss": 0.7842, + "step": 777 + }, + { + "epoch": 1.2192066805845512, + "grad_norm": 1.0490702390670776, + "learning_rate": 4.602926716919633e-06, + "loss": 0.7157, + "step": 778 + }, + { + "epoch": 1.2207724425887265, + "grad_norm": 1.131402611732483, + "learning_rate": 4.601786701223934e-06, + "loss": 0.7724, + "step": 779 + }, + { + "epoch": 1.2223382045929019, + "grad_norm": 1.0159722566604614, + "learning_rate": 4.600645192951898e-06, + "loss": 0.7666, + "step": 780 + }, + { + "epoch": 1.2239039665970772, + "grad_norm": 1.039806604385376, + "learning_rate": 4.599502192914164e-06, + "loss": 0.7845, + "step": 781 + }, + { + "epoch": 1.2254697286012526, + "grad_norm": 1.0722614526748657, + "learning_rate": 4.598357701922428e-06, + "loss": 0.782, + "step": 782 + }, + { + "epoch": 1.227035490605428, + "grad_norm": 0.9443033933639526, + "learning_rate": 4.5972117207894465e-06, + "loss": 0.6573, + "step": 783 + }, + { + "epoch": 1.2286012526096033, + "grad_norm": 0.9738026857376099, + "learning_rate": 4.596064250329035e-06, + "loss": 0.7475, + "step": 784 + }, + { + "epoch": 1.2301670146137786, + "grad_norm": 0.9932008385658264, + "learning_rate": 4.594915291356063e-06, + "loss": 0.757, + "step": 785 + }, + { + "epoch": 1.231732776617954, + "grad_norm": 1.0612564086914062, + "learning_rate": 4.5937648446864625e-06, + "loss": 0.7824, + "step": 786 + }, + { + "epoch": 1.2332985386221293, + "grad_norm": 1.0500893592834473, + "learning_rate": 4.592612911137218e-06, + "loss": 0.7646, + "step": 787 + }, + { + "epoch": 1.2348643006263047, + "grad_norm": 0.9607855081558228, + "learning_rate": 4.591459491526371e-06, + "loss": 0.754, + "step": 788 + }, + { + "epoch": 1.2364300626304803, + "grad_norm": 1.0871951580047607, + "learning_rate": 4.590304586673018e-06, + "loss": 0.7226, + "step": 789 + }, + { + "epoch": 1.2379958246346556, + "grad_norm": 1.0030092000961304, + "learning_rate": 4.589148197397311e-06, + "loss": 0.7357, + "step": 790 + }, + { + "epoch": 1.239561586638831, + "grad_norm": 1.0297991037368774, + "learning_rate": 4.587990324520455e-06, + "loss": 0.7724, + "step": 791 + }, + { + "epoch": 1.2411273486430063, + "grad_norm": 1.0192303657531738, + "learning_rate": 4.586830968864711e-06, + "loss": 0.6921, + "step": 792 + }, + { + "epoch": 1.2426931106471817, + "grad_norm": 0.9617307186126709, + "learning_rate": 4.585670131253389e-06, + "loss": 0.7443, + "step": 793 + }, + { + "epoch": 1.244258872651357, + "grad_norm": 1.0355238914489746, + "learning_rate": 4.584507812510856e-06, + "loss": 0.7741, + "step": 794 + }, + { + "epoch": 1.2458246346555324, + "grad_norm": 0.9938539862632751, + "learning_rate": 4.583344013462526e-06, + "loss": 0.7372, + "step": 795 + }, + { + "epoch": 1.2473903966597077, + "grad_norm": 1.0573335886001587, + "learning_rate": 4.582178734934869e-06, + "loss": 0.7191, + "step": 796 + }, + { + "epoch": 1.248956158663883, + "grad_norm": 0.963642954826355, + "learning_rate": 4.581011977755402e-06, + "loss": 0.6283, + "step": 797 + }, + { + "epoch": 1.2505219206680585, + "grad_norm": 1.0119287967681885, + "learning_rate": 4.579843742752694e-06, + "loss": 0.7599, + "step": 798 + }, + { + "epoch": 1.2520876826722338, + "grad_norm": 1.0628082752227783, + "learning_rate": 4.578674030756364e-06, + "loss": 0.7741, + "step": 799 + }, + { + "epoch": 1.2536534446764092, + "grad_norm": 1.0510241985321045, + "learning_rate": 4.577502842597077e-06, + "loss": 0.7615, + "step": 800 + }, + { + "epoch": 1.2552192066805845, + "grad_norm": 1.1180559396743774, + "learning_rate": 4.5763301791065474e-06, + "loss": 0.7119, + "step": 801 + }, + { + "epoch": 1.2567849686847599, + "grad_norm": 1.0198049545288086, + "learning_rate": 4.575156041117541e-06, + "loss": 0.7487, + "step": 802 + }, + { + "epoch": 1.2583507306889352, + "grad_norm": 1.018902063369751, + "learning_rate": 4.5739804294638655e-06, + "loss": 0.7461, + "step": 803 + }, + { + "epoch": 1.2599164926931106, + "grad_norm": 0.9230141639709473, + "learning_rate": 4.572803344980378e-06, + "loss": 0.7295, + "step": 804 + }, + { + "epoch": 1.261482254697286, + "grad_norm": 1.0011619329452515, + "learning_rate": 4.571624788502981e-06, + "loss": 0.7325, + "step": 805 + }, + { + "epoch": 1.2630480167014615, + "grad_norm": 0.9569756388664246, + "learning_rate": 4.570444760868622e-06, + "loss": 0.7494, + "step": 806 + }, + { + "epoch": 1.2646137787056366, + "grad_norm": 0.9860491156578064, + "learning_rate": 4.569263262915293e-06, + "loss": 0.6536, + "step": 807 + }, + { + "epoch": 1.2661795407098122, + "grad_norm": 1.0986074209213257, + "learning_rate": 4.568080295482032e-06, + "loss": 0.7337, + "step": 808 + }, + { + "epoch": 1.2677453027139876, + "grad_norm": 1.0782687664031982, + "learning_rate": 4.566895859408916e-06, + "loss": 0.7193, + "step": 809 + }, + { + "epoch": 1.269311064718163, + "grad_norm": 1.1424410343170166, + "learning_rate": 4.565709955537071e-06, + "loss": 0.6153, + "step": 810 + }, + { + "epoch": 1.2708768267223383, + "grad_norm": 0.9178900718688965, + "learning_rate": 4.564522584708662e-06, + "loss": 0.7802, + "step": 811 + }, + { + "epoch": 1.2724425887265136, + "grad_norm": 1.0006638765335083, + "learning_rate": 4.563333747766896e-06, + "loss": 0.7882, + "step": 812 + }, + { + "epoch": 1.274008350730689, + "grad_norm": 1.0139516592025757, + "learning_rate": 4.56214344555602e-06, + "loss": 0.6562, + "step": 813 + }, + { + "epoch": 1.2755741127348643, + "grad_norm": 1.0360203981399536, + "learning_rate": 4.560951678921324e-06, + "loss": 0.7951, + "step": 814 + }, + { + "epoch": 1.2771398747390397, + "grad_norm": 1.042457103729248, + "learning_rate": 4.559758448709138e-06, + "loss": 0.7572, + "step": 815 + }, + { + "epoch": 1.278705636743215, + "grad_norm": 1.0056653022766113, + "learning_rate": 4.558563755766827e-06, + "loss": 0.7679, + "step": 816 + }, + { + "epoch": 1.2802713987473904, + "grad_norm": 0.9838577508926392, + "learning_rate": 4.5573676009428e-06, + "loss": 0.7245, + "step": 817 + }, + { + "epoch": 1.2818371607515657, + "grad_norm": 1.0474631786346436, + "learning_rate": 4.556169985086503e-06, + "loss": 0.7823, + "step": 818 + }, + { + "epoch": 1.283402922755741, + "grad_norm": 1.0177569389343262, + "learning_rate": 4.554970909048417e-06, + "loss": 0.7599, + "step": 819 + }, + { + "epoch": 1.2849686847599164, + "grad_norm": 1.0120493173599243, + "learning_rate": 4.553770373680062e-06, + "loss": 0.7842, + "step": 820 + }, + { + "epoch": 1.2865344467640918, + "grad_norm": 1.1045918464660645, + "learning_rate": 4.5525683798339946e-06, + "loss": 0.7075, + "step": 821 + }, + { + "epoch": 1.2881002087682671, + "grad_norm": 1.097714900970459, + "learning_rate": 4.551364928363805e-06, + "loss": 0.7571, + "step": 822 + }, + { + "epoch": 1.2896659707724427, + "grad_norm": 0.9125490188598633, + "learning_rate": 4.550160020124121e-06, + "loss": 0.7626, + "step": 823 + }, + { + "epoch": 1.2912317327766178, + "grad_norm": 1.1127262115478516, + "learning_rate": 4.548953655970604e-06, + "loss": 0.6618, + "step": 824 + }, + { + "epoch": 1.2927974947807934, + "grad_norm": 1.0372540950775146, + "learning_rate": 4.547745836759949e-06, + "loss": 0.7467, + "step": 825 + }, + { + "epoch": 1.2943632567849686, + "grad_norm": 1.0158213376998901, + "learning_rate": 4.546536563349885e-06, + "loss": 0.6421, + "step": 826 + }, + { + "epoch": 1.2959290187891441, + "grad_norm": 1.0057018995285034, + "learning_rate": 4.545325836599171e-06, + "loss": 0.7252, + "step": 827 + }, + { + "epoch": 1.2974947807933195, + "grad_norm": 1.0552566051483154, + "learning_rate": 4.544113657367604e-06, + "loss": 0.6645, + "step": 828 + }, + { + "epoch": 1.2990605427974948, + "grad_norm": 1.0798696279525757, + "learning_rate": 4.542900026516006e-06, + "loss": 0.7627, + "step": 829 + }, + { + "epoch": 1.3006263048016702, + "grad_norm": 1.0450047254562378, + "learning_rate": 4.541684944906233e-06, + "loss": 0.705, + "step": 830 + }, + { + "epoch": 1.3021920668058455, + "grad_norm": 0.9975817203521729, + "learning_rate": 4.540468413401171e-06, + "loss": 0.7639, + "step": 831 + }, + { + "epoch": 1.303757828810021, + "grad_norm": 0.9081487059593201, + "learning_rate": 4.539250432864736e-06, + "loss": 0.8129, + "step": 832 + }, + { + "epoch": 1.3053235908141962, + "grad_norm": 1.228684902191162, + "learning_rate": 4.538031004161872e-06, + "loss": 0.6984, + "step": 833 + }, + { + "epoch": 1.3068893528183716, + "grad_norm": 1.0577794313430786, + "learning_rate": 4.536810128158552e-06, + "loss": 0.7436, + "step": 834 + }, + { + "epoch": 1.308455114822547, + "grad_norm": 1.1270350217819214, + "learning_rate": 4.535587805721777e-06, + "loss": 0.7589, + "step": 835 + }, + { + "epoch": 1.3100208768267223, + "grad_norm": 1.1473027467727661, + "learning_rate": 4.5343640377195766e-06, + "loss": 0.7988, + "step": 836 + }, + { + "epoch": 1.3115866388308977, + "grad_norm": 0.9700750112533569, + "learning_rate": 4.533138825021001e-06, + "loss": 0.7388, + "step": 837 + }, + { + "epoch": 1.313152400835073, + "grad_norm": 0.9596260786056519, + "learning_rate": 4.531912168496135e-06, + "loss": 0.8177, + "step": 838 + }, + { + "epoch": 1.3147181628392484, + "grad_norm": 0.9888957738876343, + "learning_rate": 4.530684069016081e-06, + "loss": 0.809, + "step": 839 + }, + { + "epoch": 1.316283924843424, + "grad_norm": 1.219224214553833, + "learning_rate": 4.529454527452972e-06, + "loss": 0.7761, + "step": 840 + }, + { + "epoch": 1.317849686847599, + "grad_norm": 0.9730296730995178, + "learning_rate": 4.5282235446799624e-06, + "loss": 0.8044, + "step": 841 + }, + { + "epoch": 1.3194154488517746, + "grad_norm": 1.1046907901763916, + "learning_rate": 4.526991121571228e-06, + "loss": 0.7452, + "step": 842 + }, + { + "epoch": 1.3209812108559498, + "grad_norm": 0.9707220196723938, + "learning_rate": 4.525757259001972e-06, + "loss": 0.7112, + "step": 843 + }, + { + "epoch": 1.3225469728601253, + "grad_norm": 1.0085593461990356, + "learning_rate": 4.524521957848416e-06, + "loss": 0.6266, + "step": 844 + }, + { + "epoch": 1.3241127348643007, + "grad_norm": 1.131439447402954, + "learning_rate": 4.5232852189878055e-06, + "loss": 0.7393, + "step": 845 + }, + { + "epoch": 1.325678496868476, + "grad_norm": 1.1238782405853271, + "learning_rate": 4.522047043298406e-06, + "loss": 0.7541, + "step": 846 + }, + { + "epoch": 1.3272442588726514, + "grad_norm": 0.9149152636528015, + "learning_rate": 4.520807431659504e-06, + "loss": 0.8162, + "step": 847 + }, + { + "epoch": 1.3288100208768268, + "grad_norm": 1.0428732633590698, + "learning_rate": 4.519566384951403e-06, + "loss": 0.7511, + "step": 848 + }, + { + "epoch": 1.3303757828810021, + "grad_norm": 1.0263090133666992, + "learning_rate": 4.518323904055432e-06, + "loss": 0.7482, + "step": 849 + }, + { + "epoch": 1.3319415448851775, + "grad_norm": 1.113810420036316, + "learning_rate": 4.517079989853931e-06, + "loss": 0.797, + "step": 850 + }, + { + "epoch": 1.3335073068893528, + "grad_norm": 1.0090736150741577, + "learning_rate": 4.515834643230263e-06, + "loss": 0.688, + "step": 851 + }, + { + "epoch": 1.3350730688935282, + "grad_norm": 1.012824296951294, + "learning_rate": 4.514587865068806e-06, + "loss": 0.6683, + "step": 852 + }, + { + "epoch": 1.3366388308977035, + "grad_norm": 0.9442676305770874, + "learning_rate": 4.513339656254954e-06, + "loss": 0.7812, + "step": 853 + }, + { + "epoch": 1.3382045929018789, + "grad_norm": 1.0870879888534546, + "learning_rate": 4.512090017675119e-06, + "loss": 0.6943, + "step": 854 + }, + { + "epoch": 1.3397703549060542, + "grad_norm": 0.9352210164070129, + "learning_rate": 4.510838950216727e-06, + "loss": 0.7571, + "step": 855 + }, + { + "epoch": 1.3413361169102296, + "grad_norm": 1.173621654510498, + "learning_rate": 4.509586454768221e-06, + "loss": 0.6985, + "step": 856 + }, + { + "epoch": 1.342901878914405, + "grad_norm": 0.9790180921554565, + "learning_rate": 4.508332532219053e-06, + "loss": 0.7132, + "step": 857 + }, + { + "epoch": 1.3444676409185803, + "grad_norm": 1.0999163389205933, + "learning_rate": 4.507077183459694e-06, + "loss": 0.7858, + "step": 858 + }, + { + "epoch": 1.3460334029227559, + "grad_norm": 1.018182635307312, + "learning_rate": 4.505820409381625e-06, + "loss": 0.712, + "step": 859 + }, + { + "epoch": 1.347599164926931, + "grad_norm": 1.1015701293945312, + "learning_rate": 4.504562210877338e-06, + "loss": 0.8004, + "step": 860 + }, + { + "epoch": 1.3491649269311066, + "grad_norm": 1.0872273445129395, + "learning_rate": 4.50330258884034e-06, + "loss": 0.7478, + "step": 861 + }, + { + "epoch": 1.350730688935282, + "grad_norm": 1.1098262071609497, + "learning_rate": 4.502041544165148e-06, + "loss": 0.7894, + "step": 862 + }, + { + "epoch": 1.3522964509394573, + "grad_norm": 1.0535573959350586, + "learning_rate": 4.500779077747285e-06, + "loss": 0.7185, + "step": 863 + }, + { + "epoch": 1.3538622129436326, + "grad_norm": 1.0424474477767944, + "learning_rate": 4.499515190483291e-06, + "loss": 0.748, + "step": 864 + }, + { + "epoch": 1.355427974947808, + "grad_norm": 1.005065679550171, + "learning_rate": 4.498249883270708e-06, + "loss": 0.7948, + "step": 865 + }, + { + "epoch": 1.3569937369519833, + "grad_norm": 1.0052721500396729, + "learning_rate": 4.49698315700809e-06, + "loss": 0.7619, + "step": 866 + }, + { + "epoch": 1.3585594989561587, + "grad_norm": 1.0953692197799683, + "learning_rate": 4.495715012594999e-06, + "loss": 0.7363, + "step": 867 + }, + { + "epoch": 1.360125260960334, + "grad_norm": 1.1238093376159668, + "learning_rate": 4.494445450932003e-06, + "loss": 0.6576, + "step": 868 + }, + { + "epoch": 1.3616910229645094, + "grad_norm": 0.8546008467674255, + "learning_rate": 4.493174472920676e-06, + "loss": 0.7513, + "step": 869 + }, + { + "epoch": 1.3632567849686847, + "grad_norm": 1.0863126516342163, + "learning_rate": 4.4919020794635995e-06, + "loss": 0.6886, + "step": 870 + }, + { + "epoch": 1.36482254697286, + "grad_norm": 0.9943820834159851, + "learning_rate": 4.490628271464359e-06, + "loss": 0.7253, + "step": 871 + }, + { + "epoch": 1.3663883089770354, + "grad_norm": 0.9377979636192322, + "learning_rate": 4.4893530498275436e-06, + "loss": 0.7026, + "step": 872 + }, + { + "epoch": 1.3679540709812108, + "grad_norm": 1.100799798965454, + "learning_rate": 4.488076415458749e-06, + "loss": 0.719, + "step": 873 + }, + { + "epoch": 1.3695198329853862, + "grad_norm": 1.015074610710144, + "learning_rate": 4.486798369264572e-06, + "loss": 0.7513, + "step": 874 + }, + { + "epoch": 1.3710855949895615, + "grad_norm": 1.047530174255371, + "learning_rate": 4.485518912152611e-06, + "loss": 0.7757, + "step": 875 + }, + { + "epoch": 1.372651356993737, + "grad_norm": 1.0264610052108765, + "learning_rate": 4.484238045031471e-06, + "loss": 0.7712, + "step": 876 + }, + { + "epoch": 1.3742171189979122, + "grad_norm": 0.9848430156707764, + "learning_rate": 4.482955768810752e-06, + "loss": 0.7569, + "step": 877 + }, + { + "epoch": 1.3757828810020878, + "grad_norm": 1.070816159248352, + "learning_rate": 4.48167208440106e-06, + "loss": 0.7857, + "step": 878 + }, + { + "epoch": 1.377348643006263, + "grad_norm": 1.0134745836257935, + "learning_rate": 4.480386992713998e-06, + "loss": 0.7372, + "step": 879 + }, + { + "epoch": 1.3789144050104385, + "grad_norm": 1.0736140012741089, + "learning_rate": 4.47910049466217e-06, + "loss": 0.7794, + "step": 880 + }, + { + "epoch": 1.3804801670146138, + "grad_norm": 1.0163054466247559, + "learning_rate": 4.477812591159176e-06, + "loss": 0.7341, + "step": 881 + }, + { + "epoch": 1.3820459290187892, + "grad_norm": 1.074578881263733, + "learning_rate": 4.476523283119619e-06, + "loss": 0.6924, + "step": 882 + }, + { + "epoch": 1.3836116910229646, + "grad_norm": 1.183667540550232, + "learning_rate": 4.475232571459095e-06, + "loss": 0.7866, + "step": 883 + }, + { + "epoch": 1.38517745302714, + "grad_norm": 1.030827522277832, + "learning_rate": 4.473940457094199e-06, + "loss": 0.7845, + "step": 884 + }, + { + "epoch": 1.3867432150313153, + "grad_norm": 1.0043658018112183, + "learning_rate": 4.47264694094252e-06, + "loss": 0.7399, + "step": 885 + }, + { + "epoch": 1.3883089770354906, + "grad_norm": 1.1135472059249878, + "learning_rate": 4.471352023922645e-06, + "loss": 0.771, + "step": 886 + }, + { + "epoch": 1.389874739039666, + "grad_norm": 1.0338670015335083, + "learning_rate": 4.470055706954154e-06, + "loss": 0.7282, + "step": 887 + }, + { + "epoch": 1.3914405010438413, + "grad_norm": 0.918855607509613, + "learning_rate": 4.468757990957623e-06, + "loss": 0.728, + "step": 888 + }, + { + "epoch": 1.3930062630480167, + "grad_norm": 1.06833016872406, + "learning_rate": 4.467458876854619e-06, + "loss": 0.7012, + "step": 889 + }, + { + "epoch": 1.394572025052192, + "grad_norm": 1.0103005170822144, + "learning_rate": 4.4661583655677045e-06, + "loss": 0.7074, + "step": 890 + }, + { + "epoch": 1.3961377870563674, + "grad_norm": 1.0418543815612793, + "learning_rate": 4.464856458020433e-06, + "loss": 0.6858, + "step": 891 + }, + { + "epoch": 1.3977035490605427, + "grad_norm": 0.9720668196678162, + "learning_rate": 4.463553155137348e-06, + "loss": 0.7024, + "step": 892 + }, + { + "epoch": 1.399269311064718, + "grad_norm": 1.1870561838150024, + "learning_rate": 4.462248457843986e-06, + "loss": 0.8069, + "step": 893 + }, + { + "epoch": 1.4008350730688934, + "grad_norm": 0.9888094663619995, + "learning_rate": 4.460942367066874e-06, + "loss": 0.7281, + "step": 894 + }, + { + "epoch": 1.402400835073069, + "grad_norm": 0.9120670557022095, + "learning_rate": 4.459634883733527e-06, + "loss": 0.7992, + "step": 895 + }, + { + "epoch": 1.4039665970772441, + "grad_norm": 1.0239481925964355, + "learning_rate": 4.45832600877245e-06, + "loss": 0.7312, + "step": 896 + }, + { + "epoch": 1.4055323590814197, + "grad_norm": 0.9379130005836487, + "learning_rate": 4.4570157431131355e-06, + "loss": 0.7019, + "step": 897 + }, + { + "epoch": 1.407098121085595, + "grad_norm": 1.0789109468460083, + "learning_rate": 4.455704087686063e-06, + "loss": 0.8149, + "step": 898 + }, + { + "epoch": 1.4086638830897704, + "grad_norm": 1.0521917343139648, + "learning_rate": 4.454391043422703e-06, + "loss": 0.7873, + "step": 899 + }, + { + "epoch": 1.4102296450939458, + "grad_norm": 0.9582643508911133, + "learning_rate": 4.453076611255507e-06, + "loss": 0.7815, + "step": 900 + }, + { + "epoch": 1.4117954070981211, + "grad_norm": 1.1096739768981934, + "learning_rate": 4.451760792117914e-06, + "loss": 0.7636, + "step": 901 + }, + { + "epoch": 1.4133611691022965, + "grad_norm": 1.043574333190918, + "learning_rate": 4.450443586944349e-06, + "loss": 0.7411, + "step": 902 + }, + { + "epoch": 1.4149269311064718, + "grad_norm": 1.0280033349990845, + "learning_rate": 4.449124996670221e-06, + "loss": 0.6525, + "step": 903 + }, + { + "epoch": 1.4164926931106472, + "grad_norm": 1.0329992771148682, + "learning_rate": 4.44780502223192e-06, + "loss": 0.7133, + "step": 904 + }, + { + "epoch": 1.4180584551148225, + "grad_norm": 1.016168475151062, + "learning_rate": 4.446483664566825e-06, + "loss": 0.7098, + "step": 905 + }, + { + "epoch": 1.4196242171189979, + "grad_norm": 1.2022786140441895, + "learning_rate": 4.4451609246132895e-06, + "loss": 0.7303, + "step": 906 + }, + { + "epoch": 1.4211899791231732, + "grad_norm": 1.0311683416366577, + "learning_rate": 4.443836803310655e-06, + "loss": 0.7475, + "step": 907 + }, + { + "epoch": 1.4227557411273486, + "grad_norm": 1.0073823928833008, + "learning_rate": 4.442511301599241e-06, + "loss": 0.7513, + "step": 908 + }, + { + "epoch": 1.424321503131524, + "grad_norm": 1.0702322721481323, + "learning_rate": 4.441184420420347e-06, + "loss": 0.6918, + "step": 909 + }, + { + "epoch": 1.4258872651356993, + "grad_norm": 1.1416171789169312, + "learning_rate": 4.439856160716254e-06, + "loss": 0.6324, + "step": 910 + }, + { + "epoch": 1.4274530271398747, + "grad_norm": 1.0560253858566284, + "learning_rate": 4.438526523430221e-06, + "loss": 0.749, + "step": 911 + }, + { + "epoch": 1.4290187891440502, + "grad_norm": 1.021008849143982, + "learning_rate": 4.4371955095064835e-06, + "loss": 0.7666, + "step": 912 + }, + { + "epoch": 1.4305845511482254, + "grad_norm": 1.151904821395874, + "learning_rate": 4.435863119890258e-06, + "loss": 0.6904, + "step": 913 + }, + { + "epoch": 1.432150313152401, + "grad_norm": 1.1216702461242676, + "learning_rate": 4.434529355527736e-06, + "loss": 0.7068, + "step": 914 + }, + { + "epoch": 1.433716075156576, + "grad_norm": 1.1130588054656982, + "learning_rate": 4.433194217366085e-06, + "loss": 0.6524, + "step": 915 + }, + { + "epoch": 1.4352818371607516, + "grad_norm": 1.07612144947052, + "learning_rate": 4.431857706353449e-06, + "loss": 0.8078, + "step": 916 + }, + { + "epoch": 1.436847599164927, + "grad_norm": 1.0220447778701782, + "learning_rate": 4.430519823438946e-06, + "loss": 0.6736, + "step": 917 + }, + { + "epoch": 1.4384133611691023, + "grad_norm": 1.0657141208648682, + "learning_rate": 4.429180569572669e-06, + "loss": 0.7457, + "step": 918 + }, + { + "epoch": 1.4399791231732777, + "grad_norm": 1.0646854639053345, + "learning_rate": 4.427839945705684e-06, + "loss": 0.7925, + "step": 919 + }, + { + "epoch": 1.441544885177453, + "grad_norm": 1.0619502067565918, + "learning_rate": 4.426497952790031e-06, + "loss": 0.6968, + "step": 920 + }, + { + "epoch": 1.4431106471816284, + "grad_norm": 0.9684338569641113, + "learning_rate": 4.425154591778722e-06, + "loss": 0.7411, + "step": 921 + }, + { + "epoch": 1.4446764091858038, + "grad_norm": 1.1225167512893677, + "learning_rate": 4.423809863625739e-06, + "loss": 0.7623, + "step": 922 + }, + { + "epoch": 1.446242171189979, + "grad_norm": 0.9310746788978577, + "learning_rate": 4.422463769286036e-06, + "loss": 0.8219, + "step": 923 + }, + { + "epoch": 1.4478079331941545, + "grad_norm": 1.1107957363128662, + "learning_rate": 4.4211163097155375e-06, + "loss": 0.7755, + "step": 924 + }, + { + "epoch": 1.4493736951983298, + "grad_norm": 0.9827596545219421, + "learning_rate": 4.419767485871136e-06, + "loss": 0.7102, + "step": 925 + }, + { + "epoch": 1.4509394572025052, + "grad_norm": 1.1075087785720825, + "learning_rate": 4.418417298710695e-06, + "loss": 0.6399, + "step": 926 + }, + { + "epoch": 1.4525052192066805, + "grad_norm": 0.9575802087783813, + "learning_rate": 4.417065749193045e-06, + "loss": 0.721, + "step": 927 + }, + { + "epoch": 1.4540709812108559, + "grad_norm": 1.0016006231307983, + "learning_rate": 4.415712838277985e-06, + "loss": 0.762, + "step": 928 + }, + { + "epoch": 1.4556367432150314, + "grad_norm": 1.1907106637954712, + "learning_rate": 4.414358566926279e-06, + "loss": 0.7904, + "step": 929 + }, + { + "epoch": 1.4572025052192066, + "grad_norm": 1.0011543035507202, + "learning_rate": 4.413002936099656e-06, + "loss": 0.7179, + "step": 930 + }, + { + "epoch": 1.4587682672233822, + "grad_norm": 0.980684757232666, + "learning_rate": 4.411645946760817e-06, + "loss": 0.6845, + "step": 931 + }, + { + "epoch": 1.4603340292275573, + "grad_norm": 0.9458135366439819, + "learning_rate": 4.4102875998734176e-06, + "loss": 0.6911, + "step": 932 + }, + { + "epoch": 1.4618997912317329, + "grad_norm": 1.020966649055481, + "learning_rate": 4.408927896402087e-06, + "loss": 0.7284, + "step": 933 + }, + { + "epoch": 1.4634655532359082, + "grad_norm": 1.041833519935608, + "learning_rate": 4.407566837312413e-06, + "loss": 0.7451, + "step": 934 + }, + { + "epoch": 1.4650313152400836, + "grad_norm": 0.9076465964317322, + "learning_rate": 4.406204423570946e-06, + "loss": 0.7553, + "step": 935 + }, + { + "epoch": 1.466597077244259, + "grad_norm": 1.1963952779769897, + "learning_rate": 4.404840656145199e-06, + "loss": 0.7437, + "step": 936 + }, + { + "epoch": 1.4681628392484343, + "grad_norm": 1.0181536674499512, + "learning_rate": 4.403475536003648e-06, + "loss": 0.7283, + "step": 937 + }, + { + "epoch": 1.4697286012526096, + "grad_norm": 0.9563003778457642, + "learning_rate": 4.402109064115727e-06, + "loss": 0.7375, + "step": 938 + }, + { + "epoch": 1.471294363256785, + "grad_norm": 1.05129873752594, + "learning_rate": 4.400741241451832e-06, + "loss": 0.6805, + "step": 939 + }, + { + "epoch": 1.4728601252609603, + "grad_norm": 1.0912166833877563, + "learning_rate": 4.399372068983317e-06, + "loss": 0.7076, + "step": 940 + }, + { + "epoch": 1.4744258872651357, + "grad_norm": 1.108500599861145, + "learning_rate": 4.398001547682494e-06, + "loss": 0.6836, + "step": 941 + }, + { + "epoch": 1.475991649269311, + "grad_norm": 1.0741651058197021, + "learning_rate": 4.396629678522636e-06, + "loss": 0.7405, + "step": 942 + }, + { + "epoch": 1.4775574112734864, + "grad_norm": 1.0603930950164795, + "learning_rate": 4.39525646247797e-06, + "loss": 0.7602, + "step": 943 + }, + { + "epoch": 1.4791231732776617, + "grad_norm": 0.9814990162849426, + "learning_rate": 4.39388190052368e-06, + "loss": 0.7324, + "step": 944 + }, + { + "epoch": 1.480688935281837, + "grad_norm": 0.9643695950508118, + "learning_rate": 4.392505993635906e-06, + "loss": 0.7034, + "step": 945 + }, + { + "epoch": 1.4822546972860124, + "grad_norm": 1.0490994453430176, + "learning_rate": 4.391128742791743e-06, + "loss": 0.7358, + "step": 946 + }, + { + "epoch": 1.4838204592901878, + "grad_norm": 0.9223467111587524, + "learning_rate": 4.389750148969244e-06, + "loss": 0.698, + "step": 947 + }, + { + "epoch": 1.4853862212943634, + "grad_norm": 1.1107758283615112, + "learning_rate": 4.388370213147409e-06, + "loss": 0.7264, + "step": 948 + }, + { + "epoch": 1.4869519832985385, + "grad_norm": 1.05939781665802, + "learning_rate": 4.386988936306196e-06, + "loss": 0.7209, + "step": 949 + }, + { + "epoch": 1.488517745302714, + "grad_norm": 1.024686574935913, + "learning_rate": 4.385606319426512e-06, + "loss": 0.7798, + "step": 950 + }, + { + "epoch": 1.4900835073068894, + "grad_norm": 0.9920977354049683, + "learning_rate": 4.3842223634902205e-06, + "loss": 0.8045, + "step": 951 + }, + { + "epoch": 1.4916492693110648, + "grad_norm": 0.950927197933197, + "learning_rate": 4.382837069480131e-06, + "loss": 0.6777, + "step": 952 + }, + { + "epoch": 1.4932150313152401, + "grad_norm": 1.0072081089019775, + "learning_rate": 4.3814504383800056e-06, + "loss": 0.7458, + "step": 953 + }, + { + "epoch": 1.4947807933194155, + "grad_norm": 0.9982030987739563, + "learning_rate": 4.3800624711745546e-06, + "loss": 0.8215, + "step": 954 + }, + { + "epoch": 1.4963465553235908, + "grad_norm": 0.9682126641273499, + "learning_rate": 4.3786731688494396e-06, + "loss": 0.7157, + "step": 955 + }, + { + "epoch": 1.4979123173277662, + "grad_norm": 1.1599830389022827, + "learning_rate": 4.377282532391267e-06, + "loss": 0.8091, + "step": 956 + }, + { + "epoch": 1.4994780793319415, + "grad_norm": 1.0624948740005493, + "learning_rate": 4.3758905627875934e-06, + "loss": 0.7428, + "step": 957 + }, + { + "epoch": 1.501043841336117, + "grad_norm": 1.14694082736969, + "learning_rate": 4.374497261026921e-06, + "loss": 0.7752, + "step": 958 + }, + { + "epoch": 1.5026096033402923, + "grad_norm": 1.0355825424194336, + "learning_rate": 4.373102628098697e-06, + "loss": 0.7348, + "step": 959 + }, + { + "epoch": 1.5041753653444676, + "grad_norm": 1.036026954650879, + "learning_rate": 4.371706664993316e-06, + "loss": 0.7536, + "step": 960 + }, + { + "epoch": 1.505741127348643, + "grad_norm": 1.2319790124893188, + "learning_rate": 4.370309372702116e-06, + "loss": 0.6985, + "step": 961 + }, + { + "epoch": 1.5073068893528183, + "grad_norm": 0.908687949180603, + "learning_rate": 4.36891075221738e-06, + "loss": 0.7173, + "step": 962 + }, + { + "epoch": 1.5088726513569939, + "grad_norm": 1.0482099056243896, + "learning_rate": 4.367510804532331e-06, + "loss": 0.7514, + "step": 963 + }, + { + "epoch": 1.510438413361169, + "grad_norm": 1.0434144735336304, + "learning_rate": 4.36610953064114e-06, + "loss": 0.754, + "step": 964 + }, + { + "epoch": 1.5120041753653446, + "grad_norm": 1.0122450590133667, + "learning_rate": 4.364706931538914e-06, + "loss": 0.7767, + "step": 965 + }, + { + "epoch": 1.5135699373695197, + "grad_norm": 1.0471900701522827, + "learning_rate": 4.363303008221705e-06, + "loss": 0.7337, + "step": 966 + }, + { + "epoch": 1.5151356993736953, + "grad_norm": 1.0691189765930176, + "learning_rate": 4.3618977616865045e-06, + "loss": 0.8047, + "step": 967 + }, + { + "epoch": 1.5167014613778704, + "grad_norm": 1.015114188194275, + "learning_rate": 4.360491192931242e-06, + "loss": 0.726, + "step": 968 + }, + { + "epoch": 1.518267223382046, + "grad_norm": 0.9770560264587402, + "learning_rate": 4.359083302954788e-06, + "loss": 0.7692, + "step": 969 + }, + { + "epoch": 1.5198329853862211, + "grad_norm": 1.079991340637207, + "learning_rate": 4.357674092756951e-06, + "loss": 0.8424, + "step": 970 + }, + { + "epoch": 1.5213987473903967, + "grad_norm": 1.068708896636963, + "learning_rate": 4.356263563338476e-06, + "loss": 0.758, + "step": 971 + }, + { + "epoch": 1.522964509394572, + "grad_norm": 1.1005477905273438, + "learning_rate": 4.354851715701046e-06, + "loss": 0.7209, + "step": 972 + }, + { + "epoch": 1.5245302713987474, + "grad_norm": 0.9870902299880981, + "learning_rate": 4.3534385508472786e-06, + "loss": 0.7272, + "step": 973 + }, + { + "epoch": 1.5260960334029228, + "grad_norm": 0.9616238474845886, + "learning_rate": 4.352024069780729e-06, + "loss": 0.8124, + "step": 974 + }, + { + "epoch": 1.5276617954070981, + "grad_norm": 1.0341622829437256, + "learning_rate": 4.3506082735058855e-06, + "loss": 0.7434, + "step": 975 + }, + { + "epoch": 1.5292275574112735, + "grad_norm": 1.0157302618026733, + "learning_rate": 4.349191163028171e-06, + "loss": 0.7534, + "step": 976 + }, + { + "epoch": 1.5307933194154488, + "grad_norm": 1.041379690170288, + "learning_rate": 4.347772739353939e-06, + "loss": 0.6942, + "step": 977 + }, + { + "epoch": 1.5323590814196242, + "grad_norm": 1.016205072402954, + "learning_rate": 4.346353003490481e-06, + "loss": 0.6707, + "step": 978 + }, + { + "epoch": 1.5339248434237995, + "grad_norm": 1.0034714937210083, + "learning_rate": 4.344931956446017e-06, + "loss": 0.6358, + "step": 979 + }, + { + "epoch": 1.535490605427975, + "grad_norm": 0.9671767354011536, + "learning_rate": 4.343509599229697e-06, + "loss": 0.6333, + "step": 980 + }, + { + "epoch": 1.5370563674321502, + "grad_norm": 1.1384506225585938, + "learning_rate": 4.342085932851605e-06, + "loss": 0.803, + "step": 981 + }, + { + "epoch": 1.5386221294363258, + "grad_norm": 1.023031234741211, + "learning_rate": 4.34066095832275e-06, + "loss": 0.7801, + "step": 982 + }, + { + "epoch": 1.540187891440501, + "grad_norm": 1.0675241947174072, + "learning_rate": 4.339234676655075e-06, + "loss": 0.7335, + "step": 983 + }, + { + "epoch": 1.5417536534446765, + "grad_norm": 1.0397543907165527, + "learning_rate": 4.337807088861447e-06, + "loss": 0.7722, + "step": 984 + }, + { + "epoch": 1.5433194154488517, + "grad_norm": 1.0892809629440308, + "learning_rate": 4.3363781959556635e-06, + "loss": 0.7603, + "step": 985 + }, + { + "epoch": 1.5448851774530272, + "grad_norm": 0.9335037469863892, + "learning_rate": 4.334947998952448e-06, + "loss": 0.7678, + "step": 986 + }, + { + "epoch": 1.5464509394572024, + "grad_norm": 1.0991342067718506, + "learning_rate": 4.333516498867449e-06, + "loss": 0.7461, + "step": 987 + }, + { + "epoch": 1.548016701461378, + "grad_norm": 1.112265944480896, + "learning_rate": 4.332083696717242e-06, + "loss": 0.7593, + "step": 988 + }, + { + "epoch": 1.5495824634655533, + "grad_norm": 0.9525822997093201, + "learning_rate": 4.330649593519325e-06, + "loss": 0.7629, + "step": 989 + }, + { + "epoch": 1.5511482254697286, + "grad_norm": 1.0624464750289917, + "learning_rate": 4.329214190292122e-06, + "loss": 0.7261, + "step": 990 + }, + { + "epoch": 1.552713987473904, + "grad_norm": 1.1969516277313232, + "learning_rate": 4.327777488054979e-06, + "loss": 0.7483, + "step": 991 + }, + { + "epoch": 1.5542797494780793, + "grad_norm": 1.0239863395690918, + "learning_rate": 4.326339487828167e-06, + "loss": 0.7265, + "step": 992 + }, + { + "epoch": 1.5558455114822547, + "grad_norm": 1.0881083011627197, + "learning_rate": 4.324900190632874e-06, + "loss": 0.724, + "step": 993 + }, + { + "epoch": 1.55741127348643, + "grad_norm": 1.098827838897705, + "learning_rate": 4.323459597491214e-06, + "loss": 0.7222, + "step": 994 + }, + { + "epoch": 1.5589770354906054, + "grad_norm": 1.105057716369629, + "learning_rate": 4.322017709426217e-06, + "loss": 0.7827, + "step": 995 + }, + { + "epoch": 1.5605427974947808, + "grad_norm": 0.9830679297447205, + "learning_rate": 4.3205745274618365e-06, + "loss": 0.7316, + "step": 996 + }, + { + "epoch": 1.562108559498956, + "grad_norm": 0.9951156973838806, + "learning_rate": 4.319130052622942e-06, + "loss": 0.7181, + "step": 997 + }, + { + "epoch": 1.5636743215031315, + "grad_norm": 1.0374284982681274, + "learning_rate": 4.317684285935323e-06, + "loss": 0.7355, + "step": 998 + }, + { + "epoch": 1.565240083507307, + "grad_norm": 1.0139906406402588, + "learning_rate": 4.316237228425683e-06, + "loss": 0.7249, + "step": 999 + }, + { + "epoch": 1.5668058455114822, + "grad_norm": 0.9555441737174988, + "learning_rate": 4.3147888811216496e-06, + "loss": 0.7387, + "step": 1000 + }, + { + "epoch": 1.5683716075156577, + "grad_norm": 1.0600265264511108, + "learning_rate": 4.313339245051758e-06, + "loss": 0.7445, + "step": 1001 + }, + { + "epoch": 1.5699373695198329, + "grad_norm": 1.0070478916168213, + "learning_rate": 4.311888321245461e-06, + "loss": 0.7605, + "step": 1002 + }, + { + "epoch": 1.5715031315240084, + "grad_norm": 1.019710898399353, + "learning_rate": 4.31043611073313e-06, + "loss": 0.7262, + "step": 1003 + }, + { + "epoch": 1.5730688935281836, + "grad_norm": 1.0476088523864746, + "learning_rate": 4.308982614546045e-06, + "loss": 0.6766, + "step": 1004 + }, + { + "epoch": 1.5746346555323592, + "grad_norm": 1.0670830011367798, + "learning_rate": 4.307527833716403e-06, + "loss": 0.7465, + "step": 1005 + }, + { + "epoch": 1.5762004175365343, + "grad_norm": 1.1807290315628052, + "learning_rate": 4.30607176927731e-06, + "loss": 0.7032, + "step": 1006 + }, + { + "epoch": 1.5777661795407099, + "grad_norm": 1.0482438802719116, + "learning_rate": 4.3046144222627855e-06, + "loss": 0.7156, + "step": 1007 + }, + { + "epoch": 1.5793319415448852, + "grad_norm": 0.9434391856193542, + "learning_rate": 4.303155793707759e-06, + "loss": 0.7816, + "step": 1008 + }, + { + "epoch": 1.5808977035490606, + "grad_norm": 1.0497010946273804, + "learning_rate": 4.3016958846480706e-06, + "loss": 0.735, + "step": 1009 + }, + { + "epoch": 1.582463465553236, + "grad_norm": 0.955949068069458, + "learning_rate": 4.300234696120469e-06, + "loss": 0.6414, + "step": 1010 + }, + { + "epoch": 1.5840292275574113, + "grad_norm": 1.0462027788162231, + "learning_rate": 4.298772229162613e-06, + "loss": 0.6949, + "step": 1011 + }, + { + "epoch": 1.5855949895615866, + "grad_norm": 1.0634411573410034, + "learning_rate": 4.297308484813067e-06, + "loss": 0.7736, + "step": 1012 + }, + { + "epoch": 1.587160751565762, + "grad_norm": 1.0463027954101562, + "learning_rate": 4.295843464111306e-06, + "loss": 0.6812, + "step": 1013 + }, + { + "epoch": 1.5887265135699373, + "grad_norm": 1.0409252643585205, + "learning_rate": 4.294377168097706e-06, + "loss": 0.7319, + "step": 1014 + }, + { + "epoch": 1.5902922755741127, + "grad_norm": 0.9714667201042175, + "learning_rate": 4.292909597813555e-06, + "loss": 0.7473, + "step": 1015 + }, + { + "epoch": 1.5918580375782883, + "grad_norm": 1.03997802734375, + "learning_rate": 4.291440754301041e-06, + "loss": 0.7261, + "step": 1016 + }, + { + "epoch": 1.5934237995824634, + "grad_norm": 1.063293218612671, + "learning_rate": 4.289970638603258e-06, + "loss": 0.7909, + "step": 1017 + }, + { + "epoch": 1.594989561586639, + "grad_norm": 1.110007643699646, + "learning_rate": 4.288499251764205e-06, + "loss": 0.7, + "step": 1018 + }, + { + "epoch": 1.596555323590814, + "grad_norm": 0.981261134147644, + "learning_rate": 4.2870265948287806e-06, + "loss": 0.7516, + "step": 1019 + }, + { + "epoch": 1.5981210855949897, + "grad_norm": 1.0653973817825317, + "learning_rate": 4.2855526688427875e-06, + "loss": 0.7278, + "step": 1020 + }, + { + "epoch": 1.5996868475991648, + "grad_norm": 0.9212517738342285, + "learning_rate": 4.284077474852928e-06, + "loss": 0.7251, + "step": 1021 + }, + { + "epoch": 1.6012526096033404, + "grad_norm": 1.0887659788131714, + "learning_rate": 4.282601013906808e-06, + "loss": 0.7271, + "step": 1022 + }, + { + "epoch": 1.6028183716075155, + "grad_norm": 1.0753779411315918, + "learning_rate": 4.281123287052928e-06, + "loss": 0.7679, + "step": 1023 + }, + { + "epoch": 1.604384133611691, + "grad_norm": 1.0282115936279297, + "learning_rate": 4.2796442953406935e-06, + "loss": 0.695, + "step": 1024 + }, + { + "epoch": 1.6059498956158664, + "grad_norm": 0.9782281517982483, + "learning_rate": 4.2781640398204036e-06, + "loss": 0.7717, + "step": 1025 + }, + { + "epoch": 1.6075156576200418, + "grad_norm": 1.06228768825531, + "learning_rate": 4.276682521543256e-06, + "loss": 0.8069, + "step": 1026 + }, + { + "epoch": 1.6090814196242171, + "grad_norm": 0.8853397965431213, + "learning_rate": 4.2751997415613465e-06, + "loss": 0.8174, + "step": 1027 + }, + { + "epoch": 1.6106471816283925, + "grad_norm": 1.1525455713272095, + "learning_rate": 4.273715700927666e-06, + "loss": 0.7149, + "step": 1028 + }, + { + "epoch": 1.6122129436325678, + "grad_norm": 0.8743720650672913, + "learning_rate": 4.272230400696101e-06, + "loss": 0.7296, + "step": 1029 + }, + { + "epoch": 1.6137787056367432, + "grad_norm": 1.1136301755905151, + "learning_rate": 4.270743841921431e-06, + "loss": 0.7539, + "step": 1030 + }, + { + "epoch": 1.6153444676409185, + "grad_norm": 0.9830031991004944, + "learning_rate": 4.269256025659332e-06, + "loss": 0.6545, + "step": 1031 + }, + { + "epoch": 1.616910229645094, + "grad_norm": 1.1221113204956055, + "learning_rate": 4.267766952966369e-06, + "loss": 0.6144, + "step": 1032 + }, + { + "epoch": 1.6184759916492695, + "grad_norm": 0.975713849067688, + "learning_rate": 4.266276624900004e-06, + "loss": 0.746, + "step": 1033 + }, + { + "epoch": 1.6200417536534446, + "grad_norm": 1.1131149530410767, + "learning_rate": 4.264785042518587e-06, + "loss": 0.705, + "step": 1034 + }, + { + "epoch": 1.6216075156576202, + "grad_norm": 1.0574990510940552, + "learning_rate": 4.263292206881361e-06, + "loss": 0.7681, + "step": 1035 + }, + { + "epoch": 1.6231732776617953, + "grad_norm": 0.9931074976921082, + "learning_rate": 4.261798119048456e-06, + "loss": 0.7854, + "step": 1036 + }, + { + "epoch": 1.6247390396659709, + "grad_norm": 1.1598386764526367, + "learning_rate": 4.260302780080896e-06, + "loss": 0.7406, + "step": 1037 + }, + { + "epoch": 1.626304801670146, + "grad_norm": 0.9910677671432495, + "learning_rate": 4.258806191040587e-06, + "loss": 0.7107, + "step": 1038 + }, + { + "epoch": 1.6278705636743216, + "grad_norm": 1.1023072004318237, + "learning_rate": 4.2573083529903294e-06, + "loss": 0.7964, + "step": 1039 + }, + { + "epoch": 1.6294363256784967, + "grad_norm": 0.991082489490509, + "learning_rate": 4.255809266993806e-06, + "loss": 0.762, + "step": 1040 + }, + { + "epoch": 1.6310020876826723, + "grad_norm": 0.9281193017959595, + "learning_rate": 4.254308934115589e-06, + "loss": 0.7272, + "step": 1041 + }, + { + "epoch": 1.6325678496868476, + "grad_norm": 1.20292067527771, + "learning_rate": 4.252807355421133e-06, + "loss": 0.7535, + "step": 1042 + }, + { + "epoch": 1.634133611691023, + "grad_norm": 1.0362178087234497, + "learning_rate": 4.251304531976778e-06, + "loss": 0.7524, + "step": 1043 + }, + { + "epoch": 1.6356993736951984, + "grad_norm": 1.0267268419265747, + "learning_rate": 4.249800464849751e-06, + "loss": 0.6588, + "step": 1044 + }, + { + "epoch": 1.6372651356993737, + "grad_norm": 1.0531187057495117, + "learning_rate": 4.248295155108158e-06, + "loss": 0.7328, + "step": 1045 + }, + { + "epoch": 1.638830897703549, + "grad_norm": 1.0008069276809692, + "learning_rate": 4.246788603820989e-06, + "loss": 0.719, + "step": 1046 + }, + { + "epoch": 1.6403966597077244, + "grad_norm": 1.064780831336975, + "learning_rate": 4.245280812058119e-06, + "loss": 0.7388, + "step": 1047 + }, + { + "epoch": 1.6419624217118998, + "grad_norm": 1.06492018699646, + "learning_rate": 4.243771780890298e-06, + "loss": 0.7495, + "step": 1048 + }, + { + "epoch": 1.6435281837160751, + "grad_norm": 0.9472903609275818, + "learning_rate": 4.24226151138916e-06, + "loss": 0.7409, + "step": 1049 + }, + { + "epoch": 1.6450939457202505, + "grad_norm": 1.0612895488739014, + "learning_rate": 4.240750004627217e-06, + "loss": 0.7605, + "step": 1050 + }, + { + "epoch": 1.6466597077244258, + "grad_norm": 0.9634467959403992, + "learning_rate": 4.239237261677863e-06, + "loss": 0.7109, + "step": 1051 + }, + { + "epoch": 1.6482254697286014, + "grad_norm": 1.1347044706344604, + "learning_rate": 4.2377232836153635e-06, + "loss": 0.693, + "step": 1052 + }, + { + "epoch": 1.6497912317327765, + "grad_norm": 1.0814287662506104, + "learning_rate": 4.236208071514865e-06, + "loss": 0.7159, + "step": 1053 + }, + { + "epoch": 1.651356993736952, + "grad_norm": 0.9427600502967834, + "learning_rate": 4.234691626452392e-06, + "loss": 0.7014, + "step": 1054 + }, + { + "epoch": 1.6529227557411272, + "grad_norm": 1.0177332162857056, + "learning_rate": 4.233173949504842e-06, + "loss": 0.7717, + "step": 1055 + }, + { + "epoch": 1.6544885177453028, + "grad_norm": 1.0621635913848877, + "learning_rate": 4.231655041749985e-06, + "loss": 0.7078, + "step": 1056 + }, + { + "epoch": 1.656054279749478, + "grad_norm": 1.0474066734313965, + "learning_rate": 4.230134904266472e-06, + "loss": 0.7869, + "step": 1057 + }, + { + "epoch": 1.6576200417536535, + "grad_norm": 0.9170332551002502, + "learning_rate": 4.228613538133821e-06, + "loss": 0.7806, + "step": 1058 + }, + { + "epoch": 1.6591858037578286, + "grad_norm": 1.010465383529663, + "learning_rate": 4.227090944432426e-06, + "loss": 0.7352, + "step": 1059 + }, + { + "epoch": 1.6607515657620042, + "grad_norm": 0.988068163394928, + "learning_rate": 4.22556712424355e-06, + "loss": 0.7173, + "step": 1060 + }, + { + "epoch": 1.6623173277661796, + "grad_norm": 1.0753575563430786, + "learning_rate": 4.22404207864933e-06, + "loss": 0.6981, + "step": 1061 + }, + { + "epoch": 1.663883089770355, + "grad_norm": 1.0075691938400269, + "learning_rate": 4.222515808732769e-06, + "loss": 0.7279, + "step": 1062 + }, + { + "epoch": 1.6654488517745303, + "grad_norm": 1.0954983234405518, + "learning_rate": 4.220988315577745e-06, + "loss": 0.7675, + "step": 1063 + }, + { + "epoch": 1.6670146137787056, + "grad_norm": 1.0368746519088745, + "learning_rate": 4.2194596002690016e-06, + "loss": 0.7466, + "step": 1064 + }, + { + "epoch": 1.668580375782881, + "grad_norm": 1.083906650543213, + "learning_rate": 4.217929663892149e-06, + "loss": 0.7517, + "step": 1065 + }, + { + "epoch": 1.6701461377870563, + "grad_norm": 1.1356595754623413, + "learning_rate": 4.216398507533667e-06, + "loss": 0.7601, + "step": 1066 + }, + { + "epoch": 1.6717118997912317, + "grad_norm": 1.1610008478164673, + "learning_rate": 4.2148661322809e-06, + "loss": 0.7356, + "step": 1067 + }, + { + "epoch": 1.673277661795407, + "grad_norm": 1.019673466682434, + "learning_rate": 4.21333253922206e-06, + "loss": 0.7618, + "step": 1068 + }, + { + "epoch": 1.6748434237995826, + "grad_norm": 1.1470829248428345, + "learning_rate": 4.2117977294462235e-06, + "loss": 0.8001, + "step": 1069 + }, + { + "epoch": 1.6764091858037578, + "grad_norm": 0.9750358462333679, + "learning_rate": 4.210261704043327e-06, + "loss": 0.6736, + "step": 1070 + }, + { + "epoch": 1.6779749478079333, + "grad_norm": 1.061862826347351, + "learning_rate": 4.208724464104176e-06, + "loss": 0.7751, + "step": 1071 + }, + { + "epoch": 1.6795407098121085, + "grad_norm": 1.0311503410339355, + "learning_rate": 4.2071860107204346e-06, + "loss": 0.7629, + "step": 1072 + }, + { + "epoch": 1.681106471816284, + "grad_norm": 1.0023622512817383, + "learning_rate": 4.2056463449846305e-06, + "loss": 0.6935, + "step": 1073 + }, + { + "epoch": 1.6826722338204592, + "grad_norm": 1.1624165773391724, + "learning_rate": 4.204105467990153e-06, + "loss": 0.7451, + "step": 1074 + }, + { + "epoch": 1.6842379958246347, + "grad_norm": 0.9852432608604431, + "learning_rate": 4.202563380831249e-06, + "loss": 0.7446, + "step": 1075 + }, + { + "epoch": 1.6858037578288099, + "grad_norm": 1.0009260177612305, + "learning_rate": 4.201020084603027e-06, + "loss": 0.726, + "step": 1076 + }, + { + "epoch": 1.6873695198329854, + "grad_norm": 1.0374701023101807, + "learning_rate": 4.199475580401452e-06, + "loss": 0.6687, + "step": 1077 + }, + { + "epoch": 1.6889352818371608, + "grad_norm": 1.1437046527862549, + "learning_rate": 4.19792986932335e-06, + "loss": 0.7802, + "step": 1078 + }, + { + "epoch": 1.6905010438413361, + "grad_norm": 0.9866487383842468, + "learning_rate": 4.1963829524664e-06, + "loss": 0.7117, + "step": 1079 + }, + { + "epoch": 1.6920668058455115, + "grad_norm": 0.9720652103424072, + "learning_rate": 4.194834830929142e-06, + "loss": 0.778, + "step": 1080 + }, + { + "epoch": 1.6936325678496869, + "grad_norm": 1.009422779083252, + "learning_rate": 4.1932855058109675e-06, + "loss": 0.6783, + "step": 1081 + }, + { + "epoch": 1.6951983298538622, + "grad_norm": 1.024781346321106, + "learning_rate": 4.191734978212124e-06, + "loss": 0.6697, + "step": 1082 + }, + { + "epoch": 1.6967640918580376, + "grad_norm": 1.060354471206665, + "learning_rate": 4.190183249233715e-06, + "loss": 0.7151, + "step": 1083 + }, + { + "epoch": 1.698329853862213, + "grad_norm": 1.118486762046814, + "learning_rate": 4.1886303199776924e-06, + "loss": 0.7408, + "step": 1084 + }, + { + "epoch": 1.6998956158663883, + "grad_norm": 0.9334209561347961, + "learning_rate": 4.187076191546865e-06, + "loss": 0.7646, + "step": 1085 + }, + { + "epoch": 1.7014613778705638, + "grad_norm": 1.0506161451339722, + "learning_rate": 4.185520865044892e-06, + "loss": 0.7299, + "step": 1086 + }, + { + "epoch": 1.703027139874739, + "grad_norm": 1.1261264085769653, + "learning_rate": 4.1839643415762825e-06, + "loss": 0.6813, + "step": 1087 + }, + { + "epoch": 1.7045929018789145, + "grad_norm": 0.9690954685211182, + "learning_rate": 4.182406622246396e-06, + "loss": 0.7417, + "step": 1088 + }, + { + "epoch": 1.7061586638830897, + "grad_norm": 1.0872795581817627, + "learning_rate": 4.180847708161442e-06, + "loss": 0.676, + "step": 1089 + }, + { + "epoch": 1.7077244258872653, + "grad_norm": 1.054091453552246, + "learning_rate": 4.179287600428476e-06, + "loss": 0.785, + "step": 1090 + }, + { + "epoch": 1.7092901878914404, + "grad_norm": 1.109984278678894, + "learning_rate": 4.177726300155404e-06, + "loss": 0.8033, + "step": 1091 + }, + { + "epoch": 1.710855949895616, + "grad_norm": 1.0984348058700562, + "learning_rate": 4.176163808450978e-06, + "loss": 0.6964, + "step": 1092 + }, + { + "epoch": 1.712421711899791, + "grad_norm": 1.0788499116897583, + "learning_rate": 4.174600126424797e-06, + "loss": 0.7596, + "step": 1093 + }, + { + "epoch": 1.7139874739039667, + "grad_norm": 1.0846409797668457, + "learning_rate": 4.173035255187302e-06, + "loss": 0.7322, + "step": 1094 + }, + { + "epoch": 1.715553235908142, + "grad_norm": 1.1376627683639526, + "learning_rate": 4.171469195849782e-06, + "loss": 0.6934, + "step": 1095 + }, + { + "epoch": 1.7171189979123174, + "grad_norm": 1.003432035446167, + "learning_rate": 4.169901949524368e-06, + "loss": 0.7663, + "step": 1096 + }, + { + "epoch": 1.7186847599164927, + "grad_norm": 0.9824592471122742, + "learning_rate": 4.168333517324036e-06, + "loss": 0.8253, + "step": 1097 + }, + { + "epoch": 1.720250521920668, + "grad_norm": 1.0256636142730713, + "learning_rate": 4.1667639003626e-06, + "loss": 0.6614, + "step": 1098 + }, + { + "epoch": 1.7218162839248434, + "grad_norm": 1.0290439128875732, + "learning_rate": 4.16519309975472e-06, + "loss": 0.7403, + "step": 1099 + }, + { + "epoch": 1.7233820459290188, + "grad_norm": 1.0358202457427979, + "learning_rate": 4.163621116615892e-06, + "loss": 0.7332, + "step": 1100 + }, + { + "epoch": 1.7249478079331941, + "grad_norm": 1.1046626567840576, + "learning_rate": 4.162047952062456e-06, + "loss": 0.6672, + "step": 1101 + }, + { + "epoch": 1.7265135699373695, + "grad_norm": 1.116408109664917, + "learning_rate": 4.16047360721159e-06, + "loss": 0.6997, + "step": 1102 + }, + { + "epoch": 1.7280793319415448, + "grad_norm": 0.98244309425354, + "learning_rate": 4.158898083181308e-06, + "loss": 0.754, + "step": 1103 + }, + { + "epoch": 1.7296450939457202, + "grad_norm": 1.033761978149414, + "learning_rate": 4.1573213810904635e-06, + "loss": 0.7317, + "step": 1104 + }, + { + "epoch": 1.7312108559498958, + "grad_norm": 1.1415889263153076, + "learning_rate": 4.155743502058745e-06, + "loss": 0.8361, + "step": 1105 + }, + { + "epoch": 1.732776617954071, + "grad_norm": 0.9897739291191101, + "learning_rate": 4.154164447206679e-06, + "loss": 0.7361, + "step": 1106 + }, + { + "epoch": 1.7343423799582465, + "grad_norm": 1.0837332010269165, + "learning_rate": 4.1525842176556235e-06, + "loss": 0.7007, + "step": 1107 + }, + { + "epoch": 1.7359081419624216, + "grad_norm": 1.021332025527954, + "learning_rate": 4.151002814527774e-06, + "loss": 0.723, + "step": 1108 + }, + { + "epoch": 1.7374739039665972, + "grad_norm": 0.9808626770973206, + "learning_rate": 4.14942023894616e-06, + "loss": 0.6121, + "step": 1109 + }, + { + "epoch": 1.7390396659707723, + "grad_norm": 1.0266727209091187, + "learning_rate": 4.14783649203464e-06, + "loss": 0.8221, + "step": 1110 + }, + { + "epoch": 1.7406054279749479, + "grad_norm": 1.0880744457244873, + "learning_rate": 4.146251574917907e-06, + "loss": 0.7159, + "step": 1111 + }, + { + "epoch": 1.742171189979123, + "grad_norm": 1.0125631093978882, + "learning_rate": 4.144665488721483e-06, + "loss": 0.7228, + "step": 1112 + }, + { + "epoch": 1.7437369519832986, + "grad_norm": 0.9339308142662048, + "learning_rate": 4.143078234571723e-06, + "loss": 0.7472, + "step": 1113 + }, + { + "epoch": 1.745302713987474, + "grad_norm": 1.1234889030456543, + "learning_rate": 4.141489813595809e-06, + "loss": 0.7875, + "step": 1114 + }, + { + "epoch": 1.7468684759916493, + "grad_norm": 1.0862795114517212, + "learning_rate": 4.139900226921753e-06, + "loss": 0.7219, + "step": 1115 + }, + { + "epoch": 1.7484342379958246, + "grad_norm": 1.027912974357605, + "learning_rate": 4.138309475678393e-06, + "loss": 0.6616, + "step": 1116 + }, + { + "epoch": 1.75, + "grad_norm": 1.0271416902542114, + "learning_rate": 4.136717560995398e-06, + "loss": 0.7245, + "step": 1117 + }, + { + "epoch": 1.7515657620041754, + "grad_norm": 0.9430286288261414, + "learning_rate": 4.135124484003258e-06, + "loss": 0.7347, + "step": 1118 + }, + { + "epoch": 1.7531315240083507, + "grad_norm": 1.1191871166229248, + "learning_rate": 4.133530245833292e-06, + "loss": 0.6713, + "step": 1119 + }, + { + "epoch": 1.754697286012526, + "grad_norm": 1.0734976530075073, + "learning_rate": 4.131934847617642e-06, + "loss": 0.7268, + "step": 1120 + }, + { + "epoch": 1.7562630480167014, + "grad_norm": 1.0727618932724, + "learning_rate": 4.130338290489276e-06, + "loss": 0.7855, + "step": 1121 + }, + { + "epoch": 1.757828810020877, + "grad_norm": 1.0351287126541138, + "learning_rate": 4.1287405755819805e-06, + "loss": 0.7297, + "step": 1122 + }, + { + "epoch": 1.7593945720250521, + "grad_norm": 1.046382188796997, + "learning_rate": 4.127141704030371e-06, + "loss": 0.6508, + "step": 1123 + }, + { + "epoch": 1.7609603340292277, + "grad_norm": 1.0040491819381714, + "learning_rate": 4.125541676969876e-06, + "loss": 0.7707, + "step": 1124 + }, + { + "epoch": 1.7625260960334028, + "grad_norm": 1.1626756191253662, + "learning_rate": 4.123940495536752e-06, + "loss": 0.7366, + "step": 1125 + }, + { + "epoch": 1.7640918580375784, + "grad_norm": 0.9720292091369629, + "learning_rate": 4.122338160868071e-06, + "loss": 0.7641, + "step": 1126 + }, + { + "epoch": 1.7656576200417535, + "grad_norm": 0.9957835078239441, + "learning_rate": 4.1207346741017265e-06, + "loss": 0.7398, + "step": 1127 + }, + { + "epoch": 1.767223382045929, + "grad_norm": 0.972172737121582, + "learning_rate": 4.1191300363764276e-06, + "loss": 0.6565, + "step": 1128 + }, + { + "epoch": 1.7687891440501042, + "grad_norm": 1.0128676891326904, + "learning_rate": 4.117524248831702e-06, + "loss": 0.7433, + "step": 1129 + }, + { + "epoch": 1.7703549060542798, + "grad_norm": 1.1161550283432007, + "learning_rate": 4.115917312607895e-06, + "loss": 0.8241, + "step": 1130 + }, + { + "epoch": 1.7719206680584552, + "grad_norm": 1.009810209274292, + "learning_rate": 4.114309228846167e-06, + "loss": 0.6939, + "step": 1131 + }, + { + "epoch": 1.7734864300626305, + "grad_norm": 1.0082894563674927, + "learning_rate": 4.112699998688492e-06, + "loss": 0.6772, + "step": 1132 + }, + { + "epoch": 1.7750521920668059, + "grad_norm": 0.9453067183494568, + "learning_rate": 4.111089623277658e-06, + "loss": 0.6982, + "step": 1133 + }, + { + "epoch": 1.7766179540709812, + "grad_norm": 1.074747920036316, + "learning_rate": 4.1094781037572696e-06, + "loss": 0.7296, + "step": 1134 + }, + { + "epoch": 1.7781837160751566, + "grad_norm": 1.0477592945098877, + "learning_rate": 4.1078654412717404e-06, + "loss": 0.8162, + "step": 1135 + }, + { + "epoch": 1.779749478079332, + "grad_norm": 1.1362730264663696, + "learning_rate": 4.106251636966297e-06, + "loss": 0.6942, + "step": 1136 + }, + { + "epoch": 1.7813152400835073, + "grad_norm": 0.9564191102981567, + "learning_rate": 4.104636691986977e-06, + "loss": 0.7613, + "step": 1137 + }, + { + "epoch": 1.7828810020876826, + "grad_norm": 1.0384364128112793, + "learning_rate": 4.103020607480628e-06, + "loss": 0.6482, + "step": 1138 + }, + { + "epoch": 1.784446764091858, + "grad_norm": 1.1258740425109863, + "learning_rate": 4.101403384594905e-06, + "loss": 0.7736, + "step": 1139 + }, + { + "epoch": 1.7860125260960333, + "grad_norm": 1.0740283727645874, + "learning_rate": 4.099785024478276e-06, + "loss": 0.7219, + "step": 1140 + }, + { + "epoch": 1.787578288100209, + "grad_norm": 1.0377696752548218, + "learning_rate": 4.098165528280012e-06, + "loss": 0.7132, + "step": 1141 + }, + { + "epoch": 1.789144050104384, + "grad_norm": 1.0551276206970215, + "learning_rate": 4.096544897150192e-06, + "loss": 0.7879, + "step": 1142 + }, + { + "epoch": 1.7907098121085596, + "grad_norm": 1.096389651298523, + "learning_rate": 4.0949231322397036e-06, + "loss": 0.7516, + "step": 1143 + }, + { + "epoch": 1.7922755741127347, + "grad_norm": 0.9602596163749695, + "learning_rate": 4.093300234700234e-06, + "loss": 0.7197, + "step": 1144 + }, + { + "epoch": 1.7938413361169103, + "grad_norm": 1.1918690204620361, + "learning_rate": 4.09167620568428e-06, + "loss": 0.7634, + "step": 1145 + }, + { + "epoch": 1.7954070981210855, + "grad_norm": 1.1249059438705444, + "learning_rate": 4.09005104634514e-06, + "loss": 0.7718, + "step": 1146 + }, + { + "epoch": 1.796972860125261, + "grad_norm": 1.0596246719360352, + "learning_rate": 4.0884247578369164e-06, + "loss": 0.7258, + "step": 1147 + }, + { + "epoch": 1.7985386221294362, + "grad_norm": 1.026201844215393, + "learning_rate": 4.086797341314509e-06, + "loss": 0.6426, + "step": 1148 + }, + { + "epoch": 1.8001043841336117, + "grad_norm": 1.0791343450546265, + "learning_rate": 4.085168797933625e-06, + "loss": 0.7147, + "step": 1149 + }, + { + "epoch": 1.801670146137787, + "grad_norm": 0.9660344123840332, + "learning_rate": 4.083539128850766e-06, + "loss": 0.7335, + "step": 1150 + }, + { + "epoch": 1.8032359081419624, + "grad_norm": 1.0022811889648438, + "learning_rate": 4.081908335223238e-06, + "loss": 0.6824, + "step": 1151 + }, + { + "epoch": 1.8048016701461378, + "grad_norm": 0.9261506199836731, + "learning_rate": 4.080276418209143e-06, + "loss": 0.7538, + "step": 1152 + }, + { + "epoch": 1.8063674321503131, + "grad_norm": 1.088552713394165, + "learning_rate": 4.078643378967379e-06, + "loss": 0.6988, + "step": 1153 + }, + { + "epoch": 1.8079331941544885, + "grad_norm": 1.0581461191177368, + "learning_rate": 4.077009218657645e-06, + "loss": 0.7075, + "step": 1154 + }, + { + "epoch": 1.8094989561586639, + "grad_norm": 1.0722301006317139, + "learning_rate": 4.0753739384404335e-06, + "loss": 0.6974, + "step": 1155 + }, + { + "epoch": 1.8110647181628392, + "grad_norm": 0.9519115090370178, + "learning_rate": 4.073737539477033e-06, + "loss": 0.6898, + "step": 1156 + }, + { + "epoch": 1.8126304801670146, + "grad_norm": 1.1201752424240112, + "learning_rate": 4.072100022929526e-06, + "loss": 0.8572, + "step": 1157 + }, + { + "epoch": 1.8141962421711901, + "grad_norm": 1.24104642868042, + "learning_rate": 4.070461389960789e-06, + "loss": 0.7407, + "step": 1158 + }, + { + "epoch": 1.8157620041753653, + "grad_norm": 1.0625057220458984, + "learning_rate": 4.068821641734492e-06, + "loss": 0.655, + "step": 1159 + }, + { + "epoch": 1.8173277661795408, + "grad_norm": 1.1174049377441406, + "learning_rate": 4.067180779415097e-06, + "loss": 0.7214, + "step": 1160 + }, + { + "epoch": 1.818893528183716, + "grad_norm": 1.0683097839355469, + "learning_rate": 4.065538804167855e-06, + "loss": 0.7863, + "step": 1161 + }, + { + "epoch": 1.8204592901878915, + "grad_norm": 1.031198263168335, + "learning_rate": 4.063895717158809e-06, + "loss": 0.7663, + "step": 1162 + }, + { + "epoch": 1.8220250521920667, + "grad_norm": 1.1091554164886475, + "learning_rate": 4.062251519554791e-06, + "loss": 0.7234, + "step": 1163 + }, + { + "epoch": 1.8235908141962422, + "grad_norm": 1.0818198919296265, + "learning_rate": 4.060606212523425e-06, + "loss": 0.7709, + "step": 1164 + }, + { + "epoch": 1.8251565762004174, + "grad_norm": 1.0104292631149292, + "learning_rate": 4.058959797233116e-06, + "loss": 0.7225, + "step": 1165 + }, + { + "epoch": 1.826722338204593, + "grad_norm": 1.1726491451263428, + "learning_rate": 4.057312274853063e-06, + "loss": 0.6831, + "step": 1166 + }, + { + "epoch": 1.8282881002087683, + "grad_norm": 0.9735565185546875, + "learning_rate": 4.055663646553248e-06, + "loss": 0.7123, + "step": 1167 + }, + { + "epoch": 1.8298538622129437, + "grad_norm": 1.011709213256836, + "learning_rate": 4.054013913504436e-06, + "loss": 0.6885, + "step": 1168 + }, + { + "epoch": 1.831419624217119, + "grad_norm": 1.0374364852905273, + "learning_rate": 4.05236307687818e-06, + "loss": 0.782, + "step": 1169 + }, + { + "epoch": 1.8329853862212944, + "grad_norm": 1.011202096939087, + "learning_rate": 4.050711137846817e-06, + "loss": 0.7381, + "step": 1170 + }, + { + "epoch": 1.8345511482254697, + "grad_norm": 1.0984755754470825, + "learning_rate": 4.049058097583464e-06, + "loss": 0.7472, + "step": 1171 + }, + { + "epoch": 1.836116910229645, + "grad_norm": 1.090078353881836, + "learning_rate": 4.047403957262024e-06, + "loss": 0.6773, + "step": 1172 + }, + { + "epoch": 1.8376826722338204, + "grad_norm": 0.9866096377372742, + "learning_rate": 4.045748718057176e-06, + "loss": 0.6812, + "step": 1173 + }, + { + "epoch": 1.8392484342379958, + "grad_norm": 1.1006139516830444, + "learning_rate": 4.0440923811443846e-06, + "loss": 0.7953, + "step": 1174 + }, + { + "epoch": 1.8408141962421714, + "grad_norm": 0.9621697664260864, + "learning_rate": 4.04243494769989e-06, + "loss": 0.7557, + "step": 1175 + }, + { + "epoch": 1.8423799582463465, + "grad_norm": 1.1520715951919556, + "learning_rate": 4.040776418900714e-06, + "loss": 0.7422, + "step": 1176 + }, + { + "epoch": 1.843945720250522, + "grad_norm": 1.0679237842559814, + "learning_rate": 4.039116795924653e-06, + "loss": 0.6738, + "step": 1177 + }, + { + "epoch": 1.8455114822546972, + "grad_norm": 1.026510238647461, + "learning_rate": 4.037456079950284e-06, + "loss": 0.7741, + "step": 1178 + }, + { + "epoch": 1.8470772442588728, + "grad_norm": 1.0111522674560547, + "learning_rate": 4.0357942721569585e-06, + "loss": 0.6968, + "step": 1179 + }, + { + "epoch": 1.848643006263048, + "grad_norm": 1.0290493965148926, + "learning_rate": 4.034131373724802e-06, + "loss": 0.7131, + "step": 1180 + }, + { + "epoch": 1.8502087682672235, + "grad_norm": 0.9878649711608887, + "learning_rate": 4.032467385834718e-06, + "loss": 0.7477, + "step": 1181 + }, + { + "epoch": 1.8517745302713986, + "grad_norm": 1.0317823886871338, + "learning_rate": 4.030802309668379e-06, + "loss": 0.6318, + "step": 1182 + }, + { + "epoch": 1.8533402922755742, + "grad_norm": 1.0163817405700684, + "learning_rate": 4.029136146408235e-06, + "loss": 0.7027, + "step": 1183 + }, + { + "epoch": 1.8549060542797495, + "grad_norm": 0.9590600728988647, + "learning_rate": 4.027468897237505e-06, + "loss": 0.6988, + "step": 1184 + }, + { + "epoch": 1.8564718162839249, + "grad_norm": 1.0645931959152222, + "learning_rate": 4.0258005633401795e-06, + "loss": 0.7588, + "step": 1185 + }, + { + "epoch": 1.8580375782881002, + "grad_norm": 1.0380210876464844, + "learning_rate": 4.024131145901022e-06, + "loss": 0.6984, + "step": 1186 + }, + { + "epoch": 1.8596033402922756, + "grad_norm": 1.0558087825775146, + "learning_rate": 4.022460646105561e-06, + "loss": 0.6729, + "step": 1187 + }, + { + "epoch": 1.861169102296451, + "grad_norm": 1.0865014791488647, + "learning_rate": 4.020789065140097e-06, + "loss": 0.7208, + "step": 1188 + }, + { + "epoch": 1.8627348643006263, + "grad_norm": 1.0456374883651733, + "learning_rate": 4.019116404191697e-06, + "loss": 0.7759, + "step": 1189 + }, + { + "epoch": 1.8643006263048016, + "grad_norm": 0.9724026918411255, + "learning_rate": 4.017442664448197e-06, + "loss": 0.7475, + "step": 1190 + }, + { + "epoch": 1.865866388308977, + "grad_norm": 1.1170289516448975, + "learning_rate": 4.015767847098194e-06, + "loss": 0.6993, + "step": 1191 + }, + { + "epoch": 1.8674321503131524, + "grad_norm": 1.1052978038787842, + "learning_rate": 4.014091953331056e-06, + "loss": 0.748, + "step": 1192 + }, + { + "epoch": 1.8689979123173277, + "grad_norm": 1.1223454475402832, + "learning_rate": 4.012414984336912e-06, + "loss": 0.6921, + "step": 1193 + }, + { + "epoch": 1.8705636743215033, + "grad_norm": 1.0611138343811035, + "learning_rate": 4.010736941306655e-06, + "loss": 0.7859, + "step": 1194 + }, + { + "epoch": 1.8721294363256784, + "grad_norm": 1.169189691543579, + "learning_rate": 4.009057825431944e-06, + "loss": 0.5874, + "step": 1195 + }, + { + "epoch": 1.873695198329854, + "grad_norm": 1.036626935005188, + "learning_rate": 4.0073776379051945e-06, + "loss": 0.6801, + "step": 1196 + }, + { + "epoch": 1.8752609603340291, + "grad_norm": 0.9431620240211487, + "learning_rate": 4.005696379919586e-06, + "loss": 0.7476, + "step": 1197 + }, + { + "epoch": 1.8768267223382047, + "grad_norm": 1.0381059646606445, + "learning_rate": 4.0040140526690585e-06, + "loss": 0.6691, + "step": 1198 + }, + { + "epoch": 1.8783924843423798, + "grad_norm": 1.120033860206604, + "learning_rate": 4.0023306573483105e-06, + "loss": 0.6669, + "step": 1199 + }, + { + "epoch": 1.8799582463465554, + "grad_norm": 1.0298967361450195, + "learning_rate": 4.000646195152798e-06, + "loss": 0.7355, + "step": 1200 + }, + { + "epoch": 1.8815240083507305, + "grad_norm": 1.0036594867706299, + "learning_rate": 3.99896066727874e-06, + "loss": 0.7545, + "step": 1201 + }, + { + "epoch": 1.883089770354906, + "grad_norm": 1.1166342496871948, + "learning_rate": 3.997274074923103e-06, + "loss": 0.7468, + "step": 1202 + }, + { + "epoch": 1.8846555323590815, + "grad_norm": 1.0512094497680664, + "learning_rate": 3.995586419283617e-06, + "loss": 0.7647, + "step": 1203 + }, + { + "epoch": 1.8862212943632568, + "grad_norm": 1.0115973949432373, + "learning_rate": 3.993897701558764e-06, + "loss": 0.7893, + "step": 1204 + }, + { + "epoch": 1.8877870563674322, + "grad_norm": 1.0839310884475708, + "learning_rate": 3.992207922947783e-06, + "loss": 0.6824, + "step": 1205 + }, + { + "epoch": 1.8893528183716075, + "grad_norm": 0.9987578988075256, + "learning_rate": 3.99051708465066e-06, + "loss": 0.8045, + "step": 1206 + }, + { + "epoch": 1.8909185803757829, + "grad_norm": 1.0759059190750122, + "learning_rate": 3.9888251878681425e-06, + "loss": 0.7756, + "step": 1207 + }, + { + "epoch": 1.8924843423799582, + "grad_norm": 1.1055301427841187, + "learning_rate": 3.987132233801722e-06, + "loss": 0.6947, + "step": 1208 + }, + { + "epoch": 1.8940501043841336, + "grad_norm": 1.042575478553772, + "learning_rate": 3.985438223653644e-06, + "loss": 0.6669, + "step": 1209 + }, + { + "epoch": 1.895615866388309, + "grad_norm": 1.0595242977142334, + "learning_rate": 3.983743158626904e-06, + "loss": 0.6937, + "step": 1210 + }, + { + "epoch": 1.8971816283924845, + "grad_norm": 1.0498789548873901, + "learning_rate": 3.982047039925247e-06, + "loss": 0.7244, + "step": 1211 + }, + { + "epoch": 1.8987473903966596, + "grad_norm": 1.0960590839385986, + "learning_rate": 3.980349868753166e-06, + "loss": 0.7661, + "step": 1212 + }, + { + "epoch": 1.9003131524008352, + "grad_norm": 1.108823299407959, + "learning_rate": 3.9786516463159e-06, + "loss": 0.6937, + "step": 1213 + }, + { + "epoch": 1.9018789144050103, + "grad_norm": 1.051416039466858, + "learning_rate": 3.976952373819436e-06, + "loss": 0.7263, + "step": 1214 + }, + { + "epoch": 1.903444676409186, + "grad_norm": 0.9884061217308044, + "learning_rate": 3.9752520524705065e-06, + "loss": 0.6577, + "step": 1215 + }, + { + "epoch": 1.905010438413361, + "grad_norm": 1.0950369834899902, + "learning_rate": 3.9735506834765885e-06, + "loss": 0.7376, + "step": 1216 + }, + { + "epoch": 1.9065762004175366, + "grad_norm": 1.1140044927597046, + "learning_rate": 3.971848268045902e-06, + "loss": 0.7125, + "step": 1217 + }, + { + "epoch": 1.9081419624217117, + "grad_norm": 1.0908716917037964, + "learning_rate": 3.970144807387414e-06, + "loss": 0.7796, + "step": 1218 + }, + { + "epoch": 1.9097077244258873, + "grad_norm": 1.1068549156188965, + "learning_rate": 3.968440302710829e-06, + "loss": 0.7454, + "step": 1219 + }, + { + "epoch": 1.9112734864300627, + "grad_norm": 1.102594017982483, + "learning_rate": 3.9667347552265945e-06, + "loss": 0.757, + "step": 1220 + }, + { + "epoch": 1.912839248434238, + "grad_norm": 1.1588326692581177, + "learning_rate": 3.9650281661459015e-06, + "loss": 0.6491, + "step": 1221 + }, + { + "epoch": 1.9144050104384134, + "grad_norm": 0.9896461367607117, + "learning_rate": 3.963320536680675e-06, + "loss": 0.773, + "step": 1222 + }, + { + "epoch": 1.9159707724425887, + "grad_norm": 1.0855952501296997, + "learning_rate": 3.961611868043583e-06, + "loss": 0.6906, + "step": 1223 + }, + { + "epoch": 1.917536534446764, + "grad_norm": 1.0556972026824951, + "learning_rate": 3.959902161448032e-06, + "loss": 0.7577, + "step": 1224 + }, + { + "epoch": 1.9191022964509394, + "grad_norm": 1.0284528732299805, + "learning_rate": 3.958191418108162e-06, + "loss": 0.6844, + "step": 1225 + }, + { + "epoch": 1.9206680584551148, + "grad_norm": 0.9578842520713806, + "learning_rate": 3.956479639238852e-06, + "loss": 0.7216, + "step": 1226 + }, + { + "epoch": 1.9222338204592901, + "grad_norm": 1.0366833209991455, + "learning_rate": 3.954766826055717e-06, + "loss": 0.7365, + "step": 1227 + }, + { + "epoch": 1.9237995824634657, + "grad_norm": 1.0498946905136108, + "learning_rate": 3.953052979775103e-06, + "loss": 0.6788, + "step": 1228 + }, + { + "epoch": 1.9253653444676408, + "grad_norm": 1.010860562324524, + "learning_rate": 3.951338101614094e-06, + "loss": 0.7181, + "step": 1229 + }, + { + "epoch": 1.9269311064718164, + "grad_norm": 1.0203677415847778, + "learning_rate": 3.949622192790502e-06, + "loss": 0.7277, + "step": 1230 + }, + { + "epoch": 1.9284968684759916, + "grad_norm": 0.9907971024513245, + "learning_rate": 3.947905254522876e-06, + "loss": 0.6858, + "step": 1231 + }, + { + "epoch": 1.9300626304801671, + "grad_norm": 0.9824337363243103, + "learning_rate": 3.946187288030493e-06, + "loss": 0.7517, + "step": 1232 + }, + { + "epoch": 1.9316283924843423, + "grad_norm": 1.0206327438354492, + "learning_rate": 3.944468294533359e-06, + "loss": 0.7225, + "step": 1233 + }, + { + "epoch": 1.9331941544885178, + "grad_norm": 1.0529900789260864, + "learning_rate": 3.942748275252214e-06, + "loss": 0.7965, + "step": 1234 + }, + { + "epoch": 1.934759916492693, + "grad_norm": 1.0501983165740967, + "learning_rate": 3.941027231408521e-06, + "loss": 0.7646, + "step": 1235 + }, + { + "epoch": 1.9363256784968685, + "grad_norm": 0.9482662081718445, + "learning_rate": 3.939305164224474e-06, + "loss": 0.7449, + "step": 1236 + }, + { + "epoch": 1.937891440501044, + "grad_norm": 1.096723198890686, + "learning_rate": 3.937582074922993e-06, + "loss": 0.7464, + "step": 1237 + }, + { + "epoch": 1.9394572025052192, + "grad_norm": 1.0417059659957886, + "learning_rate": 3.935857964727725e-06, + "loss": 0.6586, + "step": 1238 + }, + { + "epoch": 1.9410229645093946, + "grad_norm": 1.1008849143981934, + "learning_rate": 3.934132834863039e-06, + "loss": 0.6236, + "step": 1239 + }, + { + "epoch": 1.94258872651357, + "grad_norm": 0.9107227921485901, + "learning_rate": 3.93240668655403e-06, + "loss": 0.7069, + "step": 1240 + }, + { + "epoch": 1.9441544885177453, + "grad_norm": 1.0516945123672485, + "learning_rate": 3.930679521026517e-06, + "loss": 0.773, + "step": 1241 + }, + { + "epoch": 1.9457202505219207, + "grad_norm": 0.9941951632499695, + "learning_rate": 3.92895133950704e-06, + "loss": 0.6002, + "step": 1242 + }, + { + "epoch": 1.947286012526096, + "grad_norm": 1.031784176826477, + "learning_rate": 3.927222143222859e-06, + "loss": 0.7737, + "step": 1243 + }, + { + "epoch": 1.9488517745302714, + "grad_norm": 0.986473798751831, + "learning_rate": 3.925491933401961e-06, + "loss": 0.7455, + "step": 1244 + }, + { + "epoch": 1.9504175365344467, + "grad_norm": 1.109964370727539, + "learning_rate": 3.923760711273044e-06, + "loss": 0.7048, + "step": 1245 + }, + { + "epoch": 1.951983298538622, + "grad_norm": 1.1303372383117676, + "learning_rate": 3.922028478065532e-06, + "loss": 0.749, + "step": 1246 + }, + { + "epoch": 1.9535490605427976, + "grad_norm": 0.999824583530426, + "learning_rate": 3.920295235009562e-06, + "loss": 0.7596, + "step": 1247 + }, + { + "epoch": 1.9551148225469728, + "grad_norm": 1.0440820455551147, + "learning_rate": 3.918560983335993e-06, + "loss": 0.7841, + "step": 1248 + }, + { + "epoch": 1.9566805845511483, + "grad_norm": 1.0420323610305786, + "learning_rate": 3.916825724276395e-06, + "loss": 0.7837, + "step": 1249 + }, + { + "epoch": 1.9582463465553235, + "grad_norm": 1.001997947692871, + "learning_rate": 3.915089459063058e-06, + "loss": 0.6956, + "step": 1250 + }, + { + "epoch": 1.959812108559499, + "grad_norm": 1.030626893043518, + "learning_rate": 3.913352188928984e-06, + "loss": 0.74, + "step": 1251 + }, + { + "epoch": 1.9613778705636742, + "grad_norm": 1.0981760025024414, + "learning_rate": 3.911613915107888e-06, + "loss": 0.713, + "step": 1252 + }, + { + "epoch": 1.9629436325678498, + "grad_norm": 1.020772933959961, + "learning_rate": 3.9098746388342006e-06, + "loss": 0.7605, + "step": 1253 + }, + { + "epoch": 1.964509394572025, + "grad_norm": 1.0206503868103027, + "learning_rate": 3.908134361343062e-06, + "loss": 0.7176, + "step": 1254 + }, + { + "epoch": 1.9660751565762005, + "grad_norm": 1.0471689701080322, + "learning_rate": 3.906393083870325e-06, + "loss": 0.7698, + "step": 1255 + }, + { + "epoch": 1.9676409185803758, + "grad_norm": 1.0658202171325684, + "learning_rate": 3.904650807652549e-06, + "loss": 0.7179, + "step": 1256 + }, + { + "epoch": 1.9692066805845512, + "grad_norm": 1.196192741394043, + "learning_rate": 3.902907533927007e-06, + "loss": 0.7801, + "step": 1257 + }, + { + "epoch": 1.9707724425887265, + "grad_norm": 1.0874967575073242, + "learning_rate": 3.90116326393168e-06, + "loss": 0.7672, + "step": 1258 + }, + { + "epoch": 1.9723382045929019, + "grad_norm": 1.153617024421692, + "learning_rate": 3.899417998905253e-06, + "loss": 0.752, + "step": 1259 + }, + { + "epoch": 1.9739039665970772, + "grad_norm": 1.0750453472137451, + "learning_rate": 3.89767174008712e-06, + "loss": 0.7471, + "step": 1260 + }, + { + "epoch": 1.9754697286012526, + "grad_norm": 1.0693799257278442, + "learning_rate": 3.895924488717381e-06, + "loss": 0.6704, + "step": 1261 + }, + { + "epoch": 1.977035490605428, + "grad_norm": 0.9552210569381714, + "learning_rate": 3.89417624603684e-06, + "loss": 0.7793, + "step": 1262 + }, + { + "epoch": 1.9786012526096033, + "grad_norm": 0.96712726354599, + "learning_rate": 3.892427013287007e-06, + "loss": 0.6843, + "step": 1263 + }, + { + "epoch": 1.9801670146137789, + "grad_norm": 1.1453518867492676, + "learning_rate": 3.890676791710091e-06, + "loss": 0.7532, + "step": 1264 + }, + { + "epoch": 1.981732776617954, + "grad_norm": 1.0294159650802612, + "learning_rate": 3.888925582549006e-06, + "loss": 0.703, + "step": 1265 + }, + { + "epoch": 1.9832985386221296, + "grad_norm": 1.0790715217590332, + "learning_rate": 3.8871733870473686e-06, + "loss": 0.6731, + "step": 1266 + }, + { + "epoch": 1.9848643006263047, + "grad_norm": 1.052460789680481, + "learning_rate": 3.8854202064494915e-06, + "loss": 0.748, + "step": 1267 + }, + { + "epoch": 1.9864300626304803, + "grad_norm": 1.112462043762207, + "learning_rate": 3.883666042000392e-06, + "loss": 0.7408, + "step": 1268 + }, + { + "epoch": 1.9879958246346554, + "grad_norm": 1.0738476514816284, + "learning_rate": 3.881910894945783e-06, + "loss": 0.759, + "step": 1269 + }, + { + "epoch": 1.989561586638831, + "grad_norm": 1.136472463607788, + "learning_rate": 3.8801547665320775e-06, + "loss": 0.754, + "step": 1270 + }, + { + "epoch": 1.9911273486430061, + "grad_norm": 1.0640734434127808, + "learning_rate": 3.87839765800638e-06, + "loss": 0.719, + "step": 1271 + }, + { + "epoch": 1.9926931106471817, + "grad_norm": 0.9155470132827759, + "learning_rate": 3.876639570616498e-06, + "loss": 0.762, + "step": 1272 + }, + { + "epoch": 1.994258872651357, + "grad_norm": 1.015743374824524, + "learning_rate": 3.87488050561093e-06, + "loss": 0.6926, + "step": 1273 + }, + { + "epoch": 1.9958246346555324, + "grad_norm": 1.0829862356185913, + "learning_rate": 3.87312046423887e-06, + "loss": 0.7644, + "step": 1274 + }, + { + "epoch": 1.9973903966597077, + "grad_norm": 1.0962239503860474, + "learning_rate": 3.871359447750205e-06, + "loss": 0.7038, + "step": 1275 + }, + { + "epoch": 1.998956158663883, + "grad_norm": 1.0470675230026245, + "learning_rate": 3.869597457395514e-06, + "loss": 0.7629, + "step": 1276 + }, + { + "epoch": 2.0015657620041756, + "grad_norm": 1.962398648262024, + "learning_rate": 3.867834494426069e-06, + "loss": 1.4251, + "step": 1277 + }, + { + "epoch": 2.0031315240083507, + "grad_norm": 1.0987200736999512, + "learning_rate": 3.866070560093832e-06, + "loss": 0.6873, + "step": 1278 + }, + { + "epoch": 2.0046972860125263, + "grad_norm": 1.0096511840820312, + "learning_rate": 3.864305655651453e-06, + "loss": 0.7113, + "step": 1279 + }, + { + "epoch": 2.0062630480167014, + "grad_norm": 0.9988889098167419, + "learning_rate": 3.8625397823522755e-06, + "loss": 0.6555, + "step": 1280 + }, + { + "epoch": 2.007828810020877, + "grad_norm": 0.9960927963256836, + "learning_rate": 3.860772941450326e-06, + "loss": 0.6458, + "step": 1281 + }, + { + "epoch": 2.009394572025052, + "grad_norm": 0.971942126750946, + "learning_rate": 3.85900513420032e-06, + "loss": 0.6821, + "step": 1282 + }, + { + "epoch": 2.0109603340292277, + "grad_norm": 1.0935462713241577, + "learning_rate": 3.857236361857662e-06, + "loss": 0.5862, + "step": 1283 + }, + { + "epoch": 2.012526096033403, + "grad_norm": 1.0295177698135376, + "learning_rate": 3.855466625678435e-06, + "loss": 0.723, + "step": 1284 + }, + { + "epoch": 2.0140918580375784, + "grad_norm": 0.9750491976737976, + "learning_rate": 3.853695926919415e-06, + "loss": 0.7105, + "step": 1285 + }, + { + "epoch": 2.0156576200417535, + "grad_norm": 1.1649892330169678, + "learning_rate": 3.851924266838056e-06, + "loss": 0.7106, + "step": 1286 + }, + { + "epoch": 2.017223382045929, + "grad_norm": 1.068339467048645, + "learning_rate": 3.850151646692496e-06, + "loss": 0.668, + "step": 1287 + }, + { + "epoch": 2.0187891440501042, + "grad_norm": 0.9659110903739929, + "learning_rate": 3.848378067741555e-06, + "loss": 0.6201, + "step": 1288 + }, + { + "epoch": 2.02035490605428, + "grad_norm": 1.0711618661880493, + "learning_rate": 3.846603531244734e-06, + "loss": 0.6379, + "step": 1289 + }, + { + "epoch": 2.021920668058455, + "grad_norm": 1.092109203338623, + "learning_rate": 3.844828038462213e-06, + "loss": 0.7105, + "step": 1290 + }, + { + "epoch": 2.0234864300626305, + "grad_norm": 1.1485183238983154, + "learning_rate": 3.843051590654852e-06, + "loss": 0.724, + "step": 1291 + }, + { + "epoch": 2.0250521920668056, + "grad_norm": 1.1114039421081543, + "learning_rate": 3.841274189084189e-06, + "loss": 0.6928, + "step": 1292 + }, + { + "epoch": 2.026617954070981, + "grad_norm": 1.1090906858444214, + "learning_rate": 3.839495835012441e-06, + "loss": 0.6988, + "step": 1293 + }, + { + "epoch": 2.028183716075157, + "grad_norm": 1.0808806419372559, + "learning_rate": 3.837716529702498e-06, + "loss": 0.6374, + "step": 1294 + }, + { + "epoch": 2.029749478079332, + "grad_norm": 1.0822157859802246, + "learning_rate": 3.835936274417927e-06, + "loss": 0.6249, + "step": 1295 + }, + { + "epoch": 2.0313152400835075, + "grad_norm": 1.0366182327270508, + "learning_rate": 3.834155070422972e-06, + "loss": 0.702, + "step": 1296 + }, + { + "epoch": 2.0328810020876826, + "grad_norm": 1.0268456935882568, + "learning_rate": 3.832372918982547e-06, + "loss": 0.6809, + "step": 1297 + }, + { + "epoch": 2.034446764091858, + "grad_norm": 1.1484824419021606, + "learning_rate": 3.830589821362242e-06, + "loss": 0.6442, + "step": 1298 + }, + { + "epoch": 2.0360125260960333, + "grad_norm": 1.0607216358184814, + "learning_rate": 3.828805778828315e-06, + "loss": 0.6469, + "step": 1299 + }, + { + "epoch": 2.037578288100209, + "grad_norm": 1.0950021743774414, + "learning_rate": 3.8270207926477e-06, + "loss": 0.6756, + "step": 1300 + }, + { + "epoch": 2.039144050104384, + "grad_norm": 0.9932938814163208, + "learning_rate": 3.825234864087998e-06, + "loss": 0.706, + "step": 1301 + }, + { + "epoch": 2.0407098121085596, + "grad_norm": 1.0706391334533691, + "learning_rate": 3.823447994417477e-06, + "loss": 0.6799, + "step": 1302 + }, + { + "epoch": 2.0422755741127347, + "grad_norm": 1.0093334913253784, + "learning_rate": 3.821660184905081e-06, + "loss": 0.676, + "step": 1303 + }, + { + "epoch": 2.0438413361169103, + "grad_norm": 1.0283772945404053, + "learning_rate": 3.819871436820413e-06, + "loss": 0.6738, + "step": 1304 + }, + { + "epoch": 2.0454070981210855, + "grad_norm": 1.0533452033996582, + "learning_rate": 3.818081751433746e-06, + "loss": 0.7053, + "step": 1305 + }, + { + "epoch": 2.046972860125261, + "grad_norm": 1.1219292879104614, + "learning_rate": 3.816291130016021e-06, + "loss": 0.6563, + "step": 1306 + }, + { + "epoch": 2.048538622129436, + "grad_norm": 1.025702953338623, + "learning_rate": 3.8144995738388397e-06, + "loss": 0.7264, + "step": 1307 + }, + { + "epoch": 2.0501043841336117, + "grad_norm": 1.0799325704574585, + "learning_rate": 3.8127070841744695e-06, + "loss": 0.6541, + "step": 1308 + }, + { + "epoch": 2.051670146137787, + "grad_norm": 1.043544888496399, + "learning_rate": 3.810913662295841e-06, + "loss": 0.627, + "step": 1309 + }, + { + "epoch": 2.0532359081419624, + "grad_norm": 0.9250067472457886, + "learning_rate": 3.809119309476547e-06, + "loss": 0.7447, + "step": 1310 + }, + { + "epoch": 2.0548016701461376, + "grad_norm": 1.0885530710220337, + "learning_rate": 3.8073240269908385e-06, + "loss": 0.6509, + "step": 1311 + }, + { + "epoch": 2.056367432150313, + "grad_norm": 1.2108066082000732, + "learning_rate": 3.805527816113631e-06, + "loss": 0.6844, + "step": 1312 + }, + { + "epoch": 2.0579331941544887, + "grad_norm": 1.1016451120376587, + "learning_rate": 3.8037306781204974e-06, + "loss": 0.6942, + "step": 1313 + }, + { + "epoch": 2.059498956158664, + "grad_norm": 1.080930233001709, + "learning_rate": 3.801932614287668e-06, + "loss": 0.7379, + "step": 1314 + }, + { + "epoch": 2.0610647181628394, + "grad_norm": 1.1148338317871094, + "learning_rate": 3.800133625892031e-06, + "loss": 0.6817, + "step": 1315 + }, + { + "epoch": 2.0626304801670146, + "grad_norm": 1.0792765617370605, + "learning_rate": 3.798333714211132e-06, + "loss": 0.6787, + "step": 1316 + }, + { + "epoch": 2.06419624217119, + "grad_norm": 1.219314455986023, + "learning_rate": 3.7965328805231727e-06, + "loss": 0.6451, + "step": 1317 + }, + { + "epoch": 2.0657620041753653, + "grad_norm": 1.0637946128845215, + "learning_rate": 3.7947311261070073e-06, + "loss": 0.6869, + "step": 1318 + }, + { + "epoch": 2.067327766179541, + "grad_norm": 1.112642765045166, + "learning_rate": 3.792928452242146e-06, + "loss": 0.6785, + "step": 1319 + }, + { + "epoch": 2.068893528183716, + "grad_norm": 1.006885051727295, + "learning_rate": 3.791124860208752e-06, + "loss": 0.7289, + "step": 1320 + }, + { + "epoch": 2.0704592901878915, + "grad_norm": 1.1578105688095093, + "learning_rate": 3.7893203512876385e-06, + "loss": 0.6472, + "step": 1321 + }, + { + "epoch": 2.0720250521920667, + "grad_norm": 1.0113657712936401, + "learning_rate": 3.7875149267602725e-06, + "loss": 0.6314, + "step": 1322 + }, + { + "epoch": 2.0735908141962422, + "grad_norm": 1.0363519191741943, + "learning_rate": 3.785708587908767e-06, + "loss": 0.6949, + "step": 1323 + }, + { + "epoch": 2.0751565762004174, + "grad_norm": 1.0536452531814575, + "learning_rate": 3.7839013360158904e-06, + "loss": 0.6817, + "step": 1324 + }, + { + "epoch": 2.076722338204593, + "grad_norm": 1.2700945138931274, + "learning_rate": 3.7820931723650547e-06, + "loss": 0.7364, + "step": 1325 + }, + { + "epoch": 2.078288100208768, + "grad_norm": 0.979896068572998, + "learning_rate": 3.7802840982403204e-06, + "loss": 0.6583, + "step": 1326 + }, + { + "epoch": 2.0798538622129437, + "grad_norm": 1.1849523782730103, + "learning_rate": 3.778474114926396e-06, + "loss": 0.6623, + "step": 1327 + }, + { + "epoch": 2.081419624217119, + "grad_norm": 1.0124396085739136, + "learning_rate": 3.7766632237086336e-06, + "loss": 0.7018, + "step": 1328 + }, + { + "epoch": 2.0829853862212944, + "grad_norm": 1.0735864639282227, + "learning_rate": 3.774851425873032e-06, + "loss": 0.6178, + "step": 1329 + }, + { + "epoch": 2.08455114822547, + "grad_norm": 1.1188064813613892, + "learning_rate": 3.773038722706234e-06, + "loss": 0.6603, + "step": 1330 + }, + { + "epoch": 2.086116910229645, + "grad_norm": 1.1021878719329834, + "learning_rate": 3.771225115495522e-06, + "loss": 0.7027, + "step": 1331 + }, + { + "epoch": 2.0876826722338206, + "grad_norm": 1.067917823791504, + "learning_rate": 3.769410605528824e-06, + "loss": 0.5733, + "step": 1332 + }, + { + "epoch": 2.0892484342379958, + "grad_norm": 1.199495553970337, + "learning_rate": 3.7675951940947075e-06, + "loss": 0.7376, + "step": 1333 + }, + { + "epoch": 2.0908141962421714, + "grad_norm": 1.104844093322754, + "learning_rate": 3.7657788824823805e-06, + "loss": 0.7099, + "step": 1334 + }, + { + "epoch": 2.0923799582463465, + "grad_norm": 1.1797598600387573, + "learning_rate": 3.7639616719816907e-06, + "loss": 0.6264, + "step": 1335 + }, + { + "epoch": 2.093945720250522, + "grad_norm": 1.0947678089141846, + "learning_rate": 3.7621435638831226e-06, + "loss": 0.7066, + "step": 1336 + }, + { + "epoch": 2.095511482254697, + "grad_norm": 1.1054636240005493, + "learning_rate": 3.7603245594778e-06, + "loss": 0.7439, + "step": 1337 + }, + { + "epoch": 2.0970772442588728, + "grad_norm": 1.106891393661499, + "learning_rate": 3.758504660057482e-06, + "loss": 0.6446, + "step": 1338 + }, + { + "epoch": 2.098643006263048, + "grad_norm": 1.043440818786621, + "learning_rate": 3.7566838669145644e-06, + "loss": 0.6885, + "step": 1339 + }, + { + "epoch": 2.1002087682672235, + "grad_norm": 1.0602154731750488, + "learning_rate": 3.7548621813420765e-06, + "loss": 0.6808, + "step": 1340 + }, + { + "epoch": 2.1017745302713986, + "grad_norm": 0.9517614841461182, + "learning_rate": 3.7530396046336814e-06, + "loss": 0.6369, + "step": 1341 + }, + { + "epoch": 2.103340292275574, + "grad_norm": 1.1436609029769897, + "learning_rate": 3.7512161380836763e-06, + "loss": 0.7427, + "step": 1342 + }, + { + "epoch": 2.1049060542797493, + "grad_norm": 1.00765860080719, + "learning_rate": 3.7493917829869887e-06, + "loss": 0.7012, + "step": 1343 + }, + { + "epoch": 2.106471816283925, + "grad_norm": 1.1718883514404297, + "learning_rate": 3.747566540639178e-06, + "loss": 0.6689, + "step": 1344 + }, + { + "epoch": 2.1080375782881, + "grad_norm": 1.0489915609359741, + "learning_rate": 3.745740412336434e-06, + "loss": 0.6735, + "step": 1345 + }, + { + "epoch": 2.1096033402922756, + "grad_norm": 1.1287360191345215, + "learning_rate": 3.743913399375575e-06, + "loss": 0.6211, + "step": 1346 + }, + { + "epoch": 2.1111691022964507, + "grad_norm": 1.0830544233322144, + "learning_rate": 3.7420855030540476e-06, + "loss": 0.7219, + "step": 1347 + }, + { + "epoch": 2.1127348643006263, + "grad_norm": 1.1535146236419678, + "learning_rate": 3.7402567246699257e-06, + "loss": 0.6995, + "step": 1348 + }, + { + "epoch": 2.114300626304802, + "grad_norm": 1.1392090320587158, + "learning_rate": 3.7384270655219102e-06, + "loss": 0.7118, + "step": 1349 + }, + { + "epoch": 2.115866388308977, + "grad_norm": 1.009692668914795, + "learning_rate": 3.736596526909327e-06, + "loss": 0.7211, + "step": 1350 + }, + { + "epoch": 2.1174321503131526, + "grad_norm": 1.2041988372802734, + "learning_rate": 3.7347651101321258e-06, + "loss": 0.5969, + "step": 1351 + }, + { + "epoch": 2.1189979123173277, + "grad_norm": 1.0518646240234375, + "learning_rate": 3.7329328164908818e-06, + "loss": 0.6624, + "step": 1352 + }, + { + "epoch": 2.1205636743215033, + "grad_norm": 1.1757079362869263, + "learning_rate": 3.7310996472867916e-06, + "loss": 0.7201, + "step": 1353 + }, + { + "epoch": 2.1221294363256784, + "grad_norm": 1.0831660032272339, + "learning_rate": 3.7292656038216722e-06, + "loss": 0.6302, + "step": 1354 + }, + { + "epoch": 2.123695198329854, + "grad_norm": 1.1849291324615479, + "learning_rate": 3.7274306873979654e-06, + "loss": 0.7365, + "step": 1355 + }, + { + "epoch": 2.125260960334029, + "grad_norm": 1.1130744218826294, + "learning_rate": 3.72559489931873e-06, + "loss": 0.6759, + "step": 1356 + }, + { + "epoch": 2.1268267223382047, + "grad_norm": 1.0014575719833374, + "learning_rate": 3.7237582408876426e-06, + "loss": 0.7179, + "step": 1357 + }, + { + "epoch": 2.12839248434238, + "grad_norm": 1.1398662328720093, + "learning_rate": 3.7219207134090013e-06, + "loss": 0.6678, + "step": 1358 + }, + { + "epoch": 2.1299582463465554, + "grad_norm": 1.1094143390655518, + "learning_rate": 3.720082318187719e-06, + "loss": 0.6962, + "step": 1359 + }, + { + "epoch": 2.1315240083507305, + "grad_norm": 1.0060160160064697, + "learning_rate": 3.718243056529326e-06, + "loss": 0.688, + "step": 1360 + }, + { + "epoch": 2.133089770354906, + "grad_norm": 1.0628772974014282, + "learning_rate": 3.716402929739967e-06, + "loss": 0.7124, + "step": 1361 + }, + { + "epoch": 2.1346555323590812, + "grad_norm": 1.1098920106887817, + "learning_rate": 3.714561939126402e-06, + "loss": 0.7016, + "step": 1362 + }, + { + "epoch": 2.136221294363257, + "grad_norm": 0.9880667328834534, + "learning_rate": 3.712720085996003e-06, + "loss": 0.7333, + "step": 1363 + }, + { + "epoch": 2.137787056367432, + "grad_norm": 1.1795120239257812, + "learning_rate": 3.710877371656757e-06, + "loss": 0.6729, + "step": 1364 + }, + { + "epoch": 2.1393528183716075, + "grad_norm": 1.1155129671096802, + "learning_rate": 3.7090337974172595e-06, + "loss": 0.6676, + "step": 1365 + }, + { + "epoch": 2.140918580375783, + "grad_norm": 1.1053357124328613, + "learning_rate": 3.707189364586719e-06, + "loss": 0.7043, + "step": 1366 + }, + { + "epoch": 2.142484342379958, + "grad_norm": 1.1484512090682983, + "learning_rate": 3.705344074474952e-06, + "loss": 0.7166, + "step": 1367 + }, + { + "epoch": 2.144050104384134, + "grad_norm": 1.056976318359375, + "learning_rate": 3.703497928392386e-06, + "loss": 0.7034, + "step": 1368 + }, + { + "epoch": 2.145615866388309, + "grad_norm": 1.093001365661621, + "learning_rate": 3.7016509276500546e-06, + "loss": 0.5952, + "step": 1369 + }, + { + "epoch": 2.1471816283924845, + "grad_norm": 1.094588041305542, + "learning_rate": 3.699803073559598e-06, + "loss": 0.7126, + "step": 1370 + }, + { + "epoch": 2.1487473903966596, + "grad_norm": 1.126368522644043, + "learning_rate": 3.6979543674332634e-06, + "loss": 0.714, + "step": 1371 + }, + { + "epoch": 2.150313152400835, + "grad_norm": 1.0853040218353271, + "learning_rate": 3.696104810583904e-06, + "loss": 0.721, + "step": 1372 + }, + { + "epoch": 2.1518789144050103, + "grad_norm": 1.091913104057312, + "learning_rate": 3.694254404324975e-06, + "loss": 0.6752, + "step": 1373 + }, + { + "epoch": 2.153444676409186, + "grad_norm": 1.1095316410064697, + "learning_rate": 3.6924031499705366e-06, + "loss": 0.6988, + "step": 1374 + }, + { + "epoch": 2.155010438413361, + "grad_norm": 1.1327425241470337, + "learning_rate": 3.690551048835249e-06, + "loss": 0.6563, + "step": 1375 + }, + { + "epoch": 2.1565762004175366, + "grad_norm": 1.1065635681152344, + "learning_rate": 3.688698102234377e-06, + "loss": 0.683, + "step": 1376 + }, + { + "epoch": 2.1581419624217117, + "grad_norm": 1.050032138824463, + "learning_rate": 3.6868443114837828e-06, + "loss": 0.6471, + "step": 1377 + }, + { + "epoch": 2.1597077244258873, + "grad_norm": 1.2008466720581055, + "learning_rate": 3.6849896778999293e-06, + "loss": 0.7257, + "step": 1378 + }, + { + "epoch": 2.1612734864300625, + "grad_norm": 1.101426601409912, + "learning_rate": 3.6831342027998796e-06, + "loss": 0.6542, + "step": 1379 + }, + { + "epoch": 2.162839248434238, + "grad_norm": 1.0931123495101929, + "learning_rate": 3.68127788750129e-06, + "loss": 0.665, + "step": 1380 + }, + { + "epoch": 2.164405010438413, + "grad_norm": 1.2418708801269531, + "learning_rate": 3.6794207333224184e-06, + "loss": 0.6399, + "step": 1381 + }, + { + "epoch": 2.1659707724425887, + "grad_norm": 1.1205544471740723, + "learning_rate": 3.6775627415821158e-06, + "loss": 0.7029, + "step": 1382 + }, + { + "epoch": 2.1675365344467643, + "grad_norm": 1.0413703918457031, + "learning_rate": 3.6757039135998276e-06, + "loss": 0.7521, + "step": 1383 + }, + { + "epoch": 2.1691022964509394, + "grad_norm": 1.0559121370315552, + "learning_rate": 3.673844250695594e-06, + "loss": 0.6664, + "step": 1384 + }, + { + "epoch": 2.170668058455115, + "grad_norm": 1.1533021926879883, + "learning_rate": 3.6719837541900495e-06, + "loss": 0.6425, + "step": 1385 + }, + { + "epoch": 2.17223382045929, + "grad_norm": 1.246163010597229, + "learning_rate": 3.6701224254044173e-06, + "loss": 0.6683, + "step": 1386 + }, + { + "epoch": 2.1737995824634657, + "grad_norm": 1.1539034843444824, + "learning_rate": 3.6682602656605136e-06, + "loss": 0.6699, + "step": 1387 + }, + { + "epoch": 2.175365344467641, + "grad_norm": 1.021262526512146, + "learning_rate": 3.6663972762807453e-06, + "loss": 0.6409, + "step": 1388 + }, + { + "epoch": 2.1769311064718164, + "grad_norm": 1.09273099899292, + "learning_rate": 3.664533458588107e-06, + "loss": 0.7315, + "step": 1389 + }, + { + "epoch": 2.1784968684759916, + "grad_norm": 1.169747233390808, + "learning_rate": 3.662668813906183e-06, + "loss": 0.6675, + "step": 1390 + }, + { + "epoch": 2.180062630480167, + "grad_norm": 1.189447045326233, + "learning_rate": 3.6608033435591435e-06, + "loss": 0.7016, + "step": 1391 + }, + { + "epoch": 2.1816283924843423, + "grad_norm": 1.0325034856796265, + "learning_rate": 3.658937048871746e-06, + "loss": 0.74, + "step": 1392 + }, + { + "epoch": 2.183194154488518, + "grad_norm": 1.1599889993667603, + "learning_rate": 3.6570699311693316e-06, + "loss": 0.6525, + "step": 1393 + }, + { + "epoch": 2.184759916492693, + "grad_norm": 1.018955945968628, + "learning_rate": 3.6552019917778294e-06, + "loss": 0.6779, + "step": 1394 + }, + { + "epoch": 2.1863256784968685, + "grad_norm": 1.1707937717437744, + "learning_rate": 3.653333232023748e-06, + "loss": 0.6321, + "step": 1395 + }, + { + "epoch": 2.1878914405010437, + "grad_norm": 1.1663832664489746, + "learning_rate": 3.6514636532341825e-06, + "loss": 0.645, + "step": 1396 + }, + { + "epoch": 2.1894572025052192, + "grad_norm": 1.0475002527236938, + "learning_rate": 3.6495932567368054e-06, + "loss": 0.6209, + "step": 1397 + }, + { + "epoch": 2.1910229645093944, + "grad_norm": 1.052640438079834, + "learning_rate": 3.647722043859874e-06, + "loss": 0.7383, + "step": 1398 + }, + { + "epoch": 2.19258872651357, + "grad_norm": 1.1010091304779053, + "learning_rate": 3.6458500159322226e-06, + "loss": 0.669, + "step": 1399 + }, + { + "epoch": 2.1941544885177455, + "grad_norm": 1.0842148065567017, + "learning_rate": 3.6439771742832645e-06, + "loss": 0.7248, + "step": 1400 + }, + { + "epoch": 2.1957202505219207, + "grad_norm": 1.1516163349151611, + "learning_rate": 3.642103520242993e-06, + "loss": 0.6501, + "step": 1401 + }, + { + "epoch": 2.1972860125260962, + "grad_norm": 1.1424976587295532, + "learning_rate": 3.6402290551419754e-06, + "loss": 0.7191, + "step": 1402 + }, + { + "epoch": 2.1988517745302714, + "grad_norm": 1.132794737815857, + "learning_rate": 3.6383537803113577e-06, + "loss": 0.676, + "step": 1403 + }, + { + "epoch": 2.200417536534447, + "grad_norm": 1.0671226978302002, + "learning_rate": 3.6364776970828586e-06, + "loss": 0.6878, + "step": 1404 + }, + { + "epoch": 2.201983298538622, + "grad_norm": 1.0326734781265259, + "learning_rate": 3.6346008067887724e-06, + "loss": 0.6559, + "step": 1405 + }, + { + "epoch": 2.2035490605427976, + "grad_norm": 1.0672800540924072, + "learning_rate": 3.6327231107619664e-06, + "loss": 0.6826, + "step": 1406 + }, + { + "epoch": 2.2051148225469728, + "grad_norm": 1.0678563117980957, + "learning_rate": 3.6308446103358785e-06, + "loss": 0.6772, + "step": 1407 + }, + { + "epoch": 2.2066805845511483, + "grad_norm": 1.1288729906082153, + "learning_rate": 3.628965306844521e-06, + "loss": 0.7248, + "step": 1408 + }, + { + "epoch": 2.2082463465553235, + "grad_norm": 1.0807738304138184, + "learning_rate": 3.6270852016224735e-06, + "loss": 0.6994, + "step": 1409 + }, + { + "epoch": 2.209812108559499, + "grad_norm": 1.2594057321548462, + "learning_rate": 3.625204296004885e-06, + "loss": 0.6429, + "step": 1410 + }, + { + "epoch": 2.211377870563674, + "grad_norm": 1.159574031829834, + "learning_rate": 3.623322591327476e-06, + "loss": 0.6292, + "step": 1411 + }, + { + "epoch": 2.2129436325678498, + "grad_norm": 1.2090522050857544, + "learning_rate": 3.621440088926531e-06, + "loss": 0.663, + "step": 1412 + }, + { + "epoch": 2.214509394572025, + "grad_norm": 1.176761507987976, + "learning_rate": 3.619556790138903e-06, + "loss": 0.5766, + "step": 1413 + }, + { + "epoch": 2.2160751565762005, + "grad_norm": 1.0924803018569946, + "learning_rate": 3.6176726963020085e-06, + "loss": 0.6838, + "step": 1414 + }, + { + "epoch": 2.2176409185803756, + "grad_norm": 1.0821013450622559, + "learning_rate": 3.6157878087538322e-06, + "loss": 0.6464, + "step": 1415 + }, + { + "epoch": 2.219206680584551, + "grad_norm": 1.0731130838394165, + "learning_rate": 3.613902128832919e-06, + "loss": 0.5345, + "step": 1416 + }, + { + "epoch": 2.2207724425887267, + "grad_norm": 1.0433294773101807, + "learning_rate": 3.612015657878378e-06, + "loss": 0.6799, + "step": 1417 + }, + { + "epoch": 2.222338204592902, + "grad_norm": 1.2107945680618286, + "learning_rate": 3.61012839722988e-06, + "loss": 0.746, + "step": 1418 + }, + { + "epoch": 2.223903966597077, + "grad_norm": 1.2256337404251099, + "learning_rate": 3.608240348227656e-06, + "loss": 0.6438, + "step": 1419 + }, + { + "epoch": 2.2254697286012526, + "grad_norm": 1.0490953922271729, + "learning_rate": 3.6063515122124975e-06, + "loss": 0.6618, + "step": 1420 + }, + { + "epoch": 2.227035490605428, + "grad_norm": 1.134993553161621, + "learning_rate": 3.6044618905257546e-06, + "loss": 0.6669, + "step": 1421 + }, + { + "epoch": 2.2286012526096033, + "grad_norm": 1.120219349861145, + "learning_rate": 3.602571484509336e-06, + "loss": 0.7164, + "step": 1422 + }, + { + "epoch": 2.230167014613779, + "grad_norm": 1.2099931240081787, + "learning_rate": 3.6006802955057057e-06, + "loss": 0.6499, + "step": 1423 + }, + { + "epoch": 2.231732776617954, + "grad_norm": 1.276110291481018, + "learning_rate": 3.5987883248578857e-06, + "loss": 0.7182, + "step": 1424 + }, + { + "epoch": 2.2332985386221296, + "grad_norm": 1.12351655960083, + "learning_rate": 3.5968955739094522e-06, + "loss": 0.6672, + "step": 1425 + }, + { + "epoch": 2.2348643006263047, + "grad_norm": 1.1756179332733154, + "learning_rate": 3.595002044004535e-06, + "loss": 0.6417, + "step": 1426 + }, + { + "epoch": 2.2364300626304803, + "grad_norm": 1.1440640687942505, + "learning_rate": 3.593107736487818e-06, + "loss": 0.6694, + "step": 1427 + }, + { + "epoch": 2.2379958246346554, + "grad_norm": 1.160673975944519, + "learning_rate": 3.5912126527045368e-06, + "loss": 0.7067, + "step": 1428 + }, + { + "epoch": 2.239561586638831, + "grad_norm": 1.1037743091583252, + "learning_rate": 3.589316794000478e-06, + "loss": 0.7258, + "step": 1429 + }, + { + "epoch": 2.241127348643006, + "grad_norm": 1.1246981620788574, + "learning_rate": 3.58742016172198e-06, + "loss": 0.6715, + "step": 1430 + }, + { + "epoch": 2.2426931106471817, + "grad_norm": 1.185558795928955, + "learning_rate": 3.5855227572159283e-06, + "loss": 0.6932, + "step": 1431 + }, + { + "epoch": 2.244258872651357, + "grad_norm": 1.0668050050735474, + "learning_rate": 3.583624581829759e-06, + "loss": 0.6836, + "step": 1432 + }, + { + "epoch": 2.2458246346555324, + "grad_norm": 1.1782646179199219, + "learning_rate": 3.581725636911454e-06, + "loss": 0.6787, + "step": 1433 + }, + { + "epoch": 2.2473903966597075, + "grad_norm": 1.2447540760040283, + "learning_rate": 3.579825923809542e-06, + "loss": 0.6702, + "step": 1434 + }, + { + "epoch": 2.248956158663883, + "grad_norm": 1.029355525970459, + "learning_rate": 3.577925443873098e-06, + "loss": 0.637, + "step": 1435 + }, + { + "epoch": 2.2505219206680582, + "grad_norm": 1.1154545545578003, + "learning_rate": 3.5760241984517397e-06, + "loss": 0.7274, + "step": 1436 + }, + { + "epoch": 2.252087682672234, + "grad_norm": 1.116469144821167, + "learning_rate": 3.5741221888956325e-06, + "loss": 0.6321, + "step": 1437 + }, + { + "epoch": 2.2536534446764094, + "grad_norm": 1.2985416650772095, + "learning_rate": 3.5722194165554795e-06, + "loss": 0.648, + "step": 1438 + }, + { + "epoch": 2.2552192066805845, + "grad_norm": 1.2045787572860718, + "learning_rate": 3.5703158827825277e-06, + "loss": 0.6138, + "step": 1439 + }, + { + "epoch": 2.25678496868476, + "grad_norm": 1.0173790454864502, + "learning_rate": 3.5684115889285663e-06, + "loss": 0.6144, + "step": 1440 + }, + { + "epoch": 2.258350730688935, + "grad_norm": 1.1876413822174072, + "learning_rate": 3.5665065363459215e-06, + "loss": 0.5951, + "step": 1441 + }, + { + "epoch": 2.259916492693111, + "grad_norm": 1.0441265106201172, + "learning_rate": 3.564600726387461e-06, + "loss": 0.6681, + "step": 1442 + }, + { + "epoch": 2.261482254697286, + "grad_norm": 1.2096240520477295, + "learning_rate": 3.5626941604065866e-06, + "loss": 0.7213, + "step": 1443 + }, + { + "epoch": 2.2630480167014615, + "grad_norm": 1.0928987264633179, + "learning_rate": 3.560786839757242e-06, + "loss": 0.6129, + "step": 1444 + }, + { + "epoch": 2.2646137787056366, + "grad_norm": 1.169466495513916, + "learning_rate": 3.558878765793902e-06, + "loss": 0.7212, + "step": 1445 + }, + { + "epoch": 2.266179540709812, + "grad_norm": 1.0599427223205566, + "learning_rate": 3.5569699398715794e-06, + "loss": 0.683, + "step": 1446 + }, + { + "epoch": 2.2677453027139873, + "grad_norm": 1.126380443572998, + "learning_rate": 3.5550603633458202e-06, + "loss": 0.6886, + "step": 1447 + }, + { + "epoch": 2.269311064718163, + "grad_norm": 1.2540115118026733, + "learning_rate": 3.553150037572704e-06, + "loss": 0.6515, + "step": 1448 + }, + { + "epoch": 2.270876826722338, + "grad_norm": 1.1289552450180054, + "learning_rate": 3.55123896390884e-06, + "loss": 0.7067, + "step": 1449 + }, + { + "epoch": 2.2724425887265136, + "grad_norm": 1.0645952224731445, + "learning_rate": 3.549327143711372e-06, + "loss": 0.7059, + "step": 1450 + }, + { + "epoch": 2.2740083507306887, + "grad_norm": 0.9427555799484253, + "learning_rate": 3.5474145783379728e-06, + "loss": 0.6472, + "step": 1451 + }, + { + "epoch": 2.2755741127348643, + "grad_norm": 1.1604477167129517, + "learning_rate": 3.5455012691468417e-06, + "loss": 0.6602, + "step": 1452 + }, + { + "epoch": 2.2771398747390394, + "grad_norm": 1.1867820024490356, + "learning_rate": 3.5435872174967102e-06, + "loss": 0.7162, + "step": 1453 + }, + { + "epoch": 2.278705636743215, + "grad_norm": 1.0727450847625732, + "learning_rate": 3.541672424746834e-06, + "loss": 0.7171, + "step": 1454 + }, + { + "epoch": 2.2802713987473906, + "grad_norm": 1.1541563272476196, + "learning_rate": 3.539756892256997e-06, + "loss": 0.7214, + "step": 1455 + }, + { + "epoch": 2.2818371607515657, + "grad_norm": 1.0292634963989258, + "learning_rate": 3.5378406213875084e-06, + "loss": 0.702, + "step": 1456 + }, + { + "epoch": 2.2834029227557413, + "grad_norm": 1.1958640813827515, + "learning_rate": 3.5359236134991996e-06, + "loss": 0.6456, + "step": 1457 + }, + { + "epoch": 2.2849686847599164, + "grad_norm": 1.191214680671692, + "learning_rate": 3.5340058699534287e-06, + "loss": 0.7149, + "step": 1458 + }, + { + "epoch": 2.286534446764092, + "grad_norm": 1.0933841466903687, + "learning_rate": 3.5320873921120735e-06, + "loss": 0.657, + "step": 1459 + }, + { + "epoch": 2.288100208768267, + "grad_norm": 1.0500751733779907, + "learning_rate": 3.5301681813375343e-06, + "loss": 0.6708, + "step": 1460 + }, + { + "epoch": 2.2896659707724427, + "grad_norm": 1.0765804052352905, + "learning_rate": 3.5282482389927313e-06, + "loss": 0.6528, + "step": 1461 + }, + { + "epoch": 2.291231732776618, + "grad_norm": 1.0991063117980957, + "learning_rate": 3.526327566441105e-06, + "loss": 0.7167, + "step": 1462 + }, + { + "epoch": 2.2927974947807934, + "grad_norm": 1.096111536026001, + "learning_rate": 3.5244061650466155e-06, + "loss": 0.6983, + "step": 1463 + }, + { + "epoch": 2.2943632567849686, + "grad_norm": 1.0495761632919312, + "learning_rate": 3.522484036173737e-06, + "loss": 0.6019, + "step": 1464 + }, + { + "epoch": 2.295929018789144, + "grad_norm": 1.204898476600647, + "learning_rate": 3.520561181187464e-06, + "loss": 0.7198, + "step": 1465 + }, + { + "epoch": 2.2974947807933193, + "grad_norm": 1.0273269414901733, + "learning_rate": 3.518637601453304e-06, + "loss": 0.7077, + "step": 1466 + }, + { + "epoch": 2.299060542797495, + "grad_norm": 1.0621248483657837, + "learning_rate": 3.5167132983372808e-06, + "loss": 0.703, + "step": 1467 + }, + { + "epoch": 2.30062630480167, + "grad_norm": 1.216728687286377, + "learning_rate": 3.5147882732059323e-06, + "loss": 0.5898, + "step": 1468 + }, + { + "epoch": 2.3021920668058455, + "grad_norm": 1.065699815750122, + "learning_rate": 3.5128625274263074e-06, + "loss": 0.611, + "step": 1469 + }, + { + "epoch": 2.3037578288100207, + "grad_norm": 1.1529189348220825, + "learning_rate": 3.5109360623659672e-06, + "loss": 0.6778, + "step": 1470 + }, + { + "epoch": 2.3053235908141962, + "grad_norm": 1.1462815999984741, + "learning_rate": 3.5090088793929842e-06, + "loss": 0.6366, + "step": 1471 + }, + { + "epoch": 2.306889352818372, + "grad_norm": 1.0135819911956787, + "learning_rate": 3.507080979875941e-06, + "loss": 0.6393, + "step": 1472 + }, + { + "epoch": 2.308455114822547, + "grad_norm": 1.112345814704895, + "learning_rate": 3.5051523651839288e-06, + "loss": 0.595, + "step": 1473 + }, + { + "epoch": 2.3100208768267225, + "grad_norm": 1.0799155235290527, + "learning_rate": 3.503223036686545e-06, + "loss": 0.7024, + "step": 1474 + }, + { + "epoch": 2.3115866388308977, + "grad_norm": 1.2584905624389648, + "learning_rate": 3.501292995753897e-06, + "loss": 0.6308, + "step": 1475 + }, + { + "epoch": 2.3131524008350732, + "grad_norm": 1.0637624263763428, + "learning_rate": 3.4993622437565955e-06, + "loss": 0.6502, + "step": 1476 + }, + { + "epoch": 2.3147181628392484, + "grad_norm": 1.2071688175201416, + "learning_rate": 3.4974307820657584e-06, + "loss": 0.6389, + "step": 1477 + }, + { + "epoch": 2.316283924843424, + "grad_norm": 1.2328526973724365, + "learning_rate": 3.4954986120530042e-06, + "loss": 0.6748, + "step": 1478 + }, + { + "epoch": 2.317849686847599, + "grad_norm": 1.1051819324493408, + "learning_rate": 3.493565735090458e-06, + "loss": 0.6675, + "step": 1479 + }, + { + "epoch": 2.3194154488517746, + "grad_norm": 1.1707861423492432, + "learning_rate": 3.491632152550746e-06, + "loss": 0.6305, + "step": 1480 + }, + { + "epoch": 2.3209812108559498, + "grad_norm": 1.0759583711624146, + "learning_rate": 3.4896978658069937e-06, + "loss": 0.683, + "step": 1481 + }, + { + "epoch": 2.3225469728601253, + "grad_norm": 1.1699646711349487, + "learning_rate": 3.487762876232828e-06, + "loss": 0.7231, + "step": 1482 + }, + { + "epoch": 2.3241127348643005, + "grad_norm": 1.0097750425338745, + "learning_rate": 3.4858271852023756e-06, + "loss": 0.5897, + "step": 1483 + }, + { + "epoch": 2.325678496868476, + "grad_norm": 1.0517746210098267, + "learning_rate": 3.4838907940902607e-06, + "loss": 0.6386, + "step": 1484 + }, + { + "epoch": 2.327244258872651, + "grad_norm": 1.2001336812973022, + "learning_rate": 3.4819537042716033e-06, + "loss": 0.6854, + "step": 1485 + }, + { + "epoch": 2.3288100208768268, + "grad_norm": 1.1421329975128174, + "learning_rate": 3.480015917122023e-06, + "loss": 0.7186, + "step": 1486 + }, + { + "epoch": 2.330375782881002, + "grad_norm": 1.0638641119003296, + "learning_rate": 3.47807743401763e-06, + "loss": 0.6848, + "step": 1487 + }, + { + "epoch": 2.3319415448851775, + "grad_norm": 1.019140601158142, + "learning_rate": 3.476138256335033e-06, + "loss": 0.6904, + "step": 1488 + }, + { + "epoch": 2.333507306889353, + "grad_norm": 1.2467342615127563, + "learning_rate": 3.4741983854513306e-06, + "loss": 0.753, + "step": 1489 + }, + { + "epoch": 2.335073068893528, + "grad_norm": 1.1455196142196655, + "learning_rate": 3.472257822744116e-06, + "loss": 0.6544, + "step": 1490 + }, + { + "epoch": 2.3366388308977033, + "grad_norm": 1.1508904695510864, + "learning_rate": 3.4703165695914737e-06, + "loss": 0.6216, + "step": 1491 + }, + { + "epoch": 2.338204592901879, + "grad_norm": 1.1695144176483154, + "learning_rate": 3.4683746273719754e-06, + "loss": 0.6478, + "step": 1492 + }, + { + "epoch": 2.3397703549060545, + "grad_norm": 1.041002631187439, + "learning_rate": 3.4664319974646872e-06, + "loss": 0.7142, + "step": 1493 + }, + { + "epoch": 2.3413361169102296, + "grad_norm": 1.0497469902038574, + "learning_rate": 3.4644886812491598e-06, + "loss": 0.7028, + "step": 1494 + }, + { + "epoch": 2.342901878914405, + "grad_norm": 1.0564930438995361, + "learning_rate": 3.4625446801054312e-06, + "loss": 0.6796, + "step": 1495 + }, + { + "epoch": 2.3444676409185803, + "grad_norm": 1.2151904106140137, + "learning_rate": 3.460599995414028e-06, + "loss": 0.6933, + "step": 1496 + }, + { + "epoch": 2.346033402922756, + "grad_norm": 1.197263240814209, + "learning_rate": 3.458654628555961e-06, + "loss": 0.7615, + "step": 1497 + }, + { + "epoch": 2.347599164926931, + "grad_norm": 1.037292718887329, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.7014, + "step": 1498 + }, + { + "epoch": 2.3491649269311066, + "grad_norm": 1.0831308364868164, + "learning_rate": 3.454761853866299e-06, + "loss": 0.6682, + "step": 1499 + }, + { + "epoch": 2.3507306889352817, + "grad_norm": 1.0729385614395142, + "learning_rate": 3.4528144487991448e-06, + "loss": 0.725, + "step": 1500 + }, + { + "epoch": 2.3522964509394573, + "grad_norm": 1.010629415512085, + "learning_rate": 3.4508663670942044e-06, + "loss": 0.6579, + "step": 1501 + }, + { + "epoch": 2.3538622129436324, + "grad_norm": 1.0924932956695557, + "learning_rate": 3.448917610134902e-06, + "loss": 0.6892, + "step": 1502 + }, + { + "epoch": 2.355427974947808, + "grad_norm": 1.1776976585388184, + "learning_rate": 3.4469681793051403e-06, + "loss": 0.6366, + "step": 1503 + }, + { + "epoch": 2.356993736951983, + "grad_norm": 1.062983751296997, + "learning_rate": 3.4450180759892994e-06, + "loss": 0.6518, + "step": 1504 + }, + { + "epoch": 2.3585594989561587, + "grad_norm": 1.1387819051742554, + "learning_rate": 3.4430673015722393e-06, + "loss": 0.623, + "step": 1505 + }, + { + "epoch": 2.3601252609603343, + "grad_norm": 1.035445213317871, + "learning_rate": 3.4411158574392956e-06, + "loss": 0.6876, + "step": 1506 + }, + { + "epoch": 2.3616910229645094, + "grad_norm": 1.0623654127120972, + "learning_rate": 3.439163744976279e-06, + "loss": 0.6795, + "step": 1507 + }, + { + "epoch": 2.3632567849686845, + "grad_norm": 1.0046356916427612, + "learning_rate": 3.437210965569475e-06, + "loss": 0.7043, + "step": 1508 + }, + { + "epoch": 2.36482254697286, + "grad_norm": 1.183174729347229, + "learning_rate": 3.4352575206056426e-06, + "loss": 0.7059, + "step": 1509 + }, + { + "epoch": 2.3663883089770357, + "grad_norm": 1.0844060182571411, + "learning_rate": 3.4333034114720155e-06, + "loss": 0.6811, + "step": 1510 + }, + { + "epoch": 2.367954070981211, + "grad_norm": 1.058577299118042, + "learning_rate": 3.431348639556296e-06, + "loss": 0.7258, + "step": 1511 + }, + { + "epoch": 2.3695198329853864, + "grad_norm": 1.0441113710403442, + "learning_rate": 3.429393206246659e-06, + "loss": 0.6659, + "step": 1512 + }, + { + "epoch": 2.3710855949895615, + "grad_norm": 1.111039400100708, + "learning_rate": 3.4274371129317476e-06, + "loss": 0.6113, + "step": 1513 + }, + { + "epoch": 2.372651356993737, + "grad_norm": 1.1454105377197266, + "learning_rate": 3.425480361000675e-06, + "loss": 0.727, + "step": 1514 + }, + { + "epoch": 2.374217118997912, + "grad_norm": 1.1241672039031982, + "learning_rate": 3.4235229518430212e-06, + "loss": 0.5266, + "step": 1515 + }, + { + "epoch": 2.375782881002088, + "grad_norm": 1.0866069793701172, + "learning_rate": 3.421564886848835e-06, + "loss": 0.6871, + "step": 1516 + }, + { + "epoch": 2.377348643006263, + "grad_norm": 1.211893081665039, + "learning_rate": 3.4196061674086276e-06, + "loss": 0.7064, + "step": 1517 + }, + { + "epoch": 2.3789144050104385, + "grad_norm": 1.1039608716964722, + "learning_rate": 3.417646794913376e-06, + "loss": 0.6821, + "step": 1518 + }, + { + "epoch": 2.3804801670146136, + "grad_norm": 1.0984389781951904, + "learning_rate": 3.4156867707545237e-06, + "loss": 0.6443, + "step": 1519 + }, + { + "epoch": 2.382045929018789, + "grad_norm": 1.11539888381958, + "learning_rate": 3.413726096323975e-06, + "loss": 0.6896, + "step": 1520 + }, + { + "epoch": 2.3836116910229643, + "grad_norm": 1.1566188335418701, + "learning_rate": 3.4117647730140935e-06, + "loss": 0.699, + "step": 1521 + }, + { + "epoch": 2.38517745302714, + "grad_norm": 1.0620112419128418, + "learning_rate": 3.409802802217707e-06, + "loss": 0.6741, + "step": 1522 + }, + { + "epoch": 2.3867432150313155, + "grad_norm": 1.117667317390442, + "learning_rate": 3.4078401853281022e-06, + "loss": 0.7027, + "step": 1523 + }, + { + "epoch": 2.3883089770354906, + "grad_norm": 1.0997360944747925, + "learning_rate": 3.4058769237390254e-06, + "loss": 0.7249, + "step": 1524 + }, + { + "epoch": 2.3898747390396657, + "grad_norm": 1.1101152896881104, + "learning_rate": 3.403913018844678e-06, + "loss": 0.6794, + "step": 1525 + }, + { + "epoch": 2.3914405010438413, + "grad_norm": 1.1798306703567505, + "learning_rate": 3.4019484720397217e-06, + "loss": 0.6219, + "step": 1526 + }, + { + "epoch": 2.393006263048017, + "grad_norm": 1.0730222463607788, + "learning_rate": 3.3999832847192716e-06, + "loss": 0.6622, + "step": 1527 + }, + { + "epoch": 2.394572025052192, + "grad_norm": 1.0943461656570435, + "learning_rate": 3.3980174582788982e-06, + "loss": 0.6999, + "step": 1528 + }, + { + "epoch": 2.3961377870563676, + "grad_norm": 1.1592621803283691, + "learning_rate": 3.396050994114628e-06, + "loss": 0.6419, + "step": 1529 + }, + { + "epoch": 2.3977035490605427, + "grad_norm": 1.1652766466140747, + "learning_rate": 3.3940838936229357e-06, + "loss": 0.6952, + "step": 1530 + }, + { + "epoch": 2.3992693110647183, + "grad_norm": 1.1424206495285034, + "learning_rate": 3.3921161582007524e-06, + "loss": 0.6521, + "step": 1531 + }, + { + "epoch": 2.4008350730688934, + "grad_norm": 1.0935698747634888, + "learning_rate": 3.3901477892454583e-06, + "loss": 0.6707, + "step": 1532 + }, + { + "epoch": 2.402400835073069, + "grad_norm": 1.2164827585220337, + "learning_rate": 3.3881787881548835e-06, + "loss": 0.6485, + "step": 1533 + }, + { + "epoch": 2.403966597077244, + "grad_norm": 1.0536198616027832, + "learning_rate": 3.386209156327307e-06, + "loss": 0.6997, + "step": 1534 + }, + { + "epoch": 2.4055323590814197, + "grad_norm": 1.1083351373672485, + "learning_rate": 3.3842388951614545e-06, + "loss": 0.6264, + "step": 1535 + }, + { + "epoch": 2.407098121085595, + "grad_norm": 1.133180856704712, + "learning_rate": 3.3822680060565026e-06, + "loss": 0.6905, + "step": 1536 + }, + { + "epoch": 2.4086638830897704, + "grad_norm": 1.0531203746795654, + "learning_rate": 3.3802964904120703e-06, + "loss": 0.6492, + "step": 1537 + }, + { + "epoch": 2.4102296450939455, + "grad_norm": 1.0389595031738281, + "learning_rate": 3.3783243496282215e-06, + "loss": 0.6506, + "step": 1538 + }, + { + "epoch": 2.411795407098121, + "grad_norm": 1.1839019060134888, + "learning_rate": 3.376351585105465e-06, + "loss": 0.6933, + "step": 1539 + }, + { + "epoch": 2.4133611691022967, + "grad_norm": 1.0715898275375366, + "learning_rate": 3.3743781982447533e-06, + "loss": 0.7043, + "step": 1540 + }, + { + "epoch": 2.414926931106472, + "grad_norm": 1.1898013353347778, + "learning_rate": 3.3724041904474793e-06, + "loss": 0.7054, + "step": 1541 + }, + { + "epoch": 2.416492693110647, + "grad_norm": 1.026489019393921, + "learning_rate": 3.370429563115478e-06, + "loss": 0.6458, + "step": 1542 + }, + { + "epoch": 2.4180584551148225, + "grad_norm": 1.2772443294525146, + "learning_rate": 3.368454317651024e-06, + "loss": 0.6647, + "step": 1543 + }, + { + "epoch": 2.419624217118998, + "grad_norm": 1.0573022365570068, + "learning_rate": 3.366478455456831e-06, + "loss": 0.6723, + "step": 1544 + }, + { + "epoch": 2.4211899791231732, + "grad_norm": 1.2983055114746094, + "learning_rate": 3.3645019779360493e-06, + "loss": 0.6922, + "step": 1545 + }, + { + "epoch": 2.422755741127349, + "grad_norm": 1.0773214101791382, + "learning_rate": 3.36252488649227e-06, + "loss": 0.6538, + "step": 1546 + }, + { + "epoch": 2.424321503131524, + "grad_norm": 1.1467341184616089, + "learning_rate": 3.3605471825295143e-06, + "loss": 0.7274, + "step": 1547 + }, + { + "epoch": 2.4258872651356995, + "grad_norm": 1.1962897777557373, + "learning_rate": 3.3585688674522438e-06, + "loss": 0.7113, + "step": 1548 + }, + { + "epoch": 2.4274530271398747, + "grad_norm": 0.997977614402771, + "learning_rate": 3.3565899426653515e-06, + "loss": 0.6731, + "step": 1549 + }, + { + "epoch": 2.4290187891440502, + "grad_norm": 1.095198392868042, + "learning_rate": 3.3546104095741634e-06, + "loss": 0.6353, + "step": 1550 + }, + { + "epoch": 2.4305845511482254, + "grad_norm": 1.0401874780654907, + "learning_rate": 3.3526302695844382e-06, + "loss": 0.7459, + "step": 1551 + }, + { + "epoch": 2.432150313152401, + "grad_norm": 1.174526333808899, + "learning_rate": 3.3506495241023646e-06, + "loss": 0.6435, + "step": 1552 + }, + { + "epoch": 2.433716075156576, + "grad_norm": 1.0898199081420898, + "learning_rate": 3.3486681745345627e-06, + "loss": 0.6991, + "step": 1553 + }, + { + "epoch": 2.4352818371607516, + "grad_norm": 1.0757240056991577, + "learning_rate": 3.346686222288081e-06, + "loss": 0.6447, + "step": 1554 + }, + { + "epoch": 2.4368475991649268, + "grad_norm": 1.0817614793777466, + "learning_rate": 3.3447036687703956e-06, + "loss": 0.6546, + "step": 1555 + }, + { + "epoch": 2.4384133611691023, + "grad_norm": 1.181427240371704, + "learning_rate": 3.3427205153894088e-06, + "loss": 0.6592, + "step": 1556 + }, + { + "epoch": 2.4399791231732775, + "grad_norm": 1.0263524055480957, + "learning_rate": 3.340736763553451e-06, + "loss": 0.6766, + "step": 1557 + }, + { + "epoch": 2.441544885177453, + "grad_norm": 1.0632904767990112, + "learning_rate": 3.3387524146712762e-06, + "loss": 0.6368, + "step": 1558 + }, + { + "epoch": 2.443110647181628, + "grad_norm": 1.165933609008789, + "learning_rate": 3.3367674701520627e-06, + "loss": 0.692, + "step": 1559 + }, + { + "epoch": 2.4446764091858038, + "grad_norm": 1.109712839126587, + "learning_rate": 3.3347819314054124e-06, + "loss": 0.6721, + "step": 1560 + }, + { + "epoch": 2.4462421711899793, + "grad_norm": 1.0221387147903442, + "learning_rate": 3.332795799841347e-06, + "loss": 0.6547, + "step": 1561 + }, + { + "epoch": 2.4478079331941545, + "grad_norm": 1.1391754150390625, + "learning_rate": 3.3308090768703118e-06, + "loss": 0.68, + "step": 1562 + }, + { + "epoch": 2.44937369519833, + "grad_norm": 1.0794706344604492, + "learning_rate": 3.328821763903172e-06, + "loss": 0.6531, + "step": 1563 + }, + { + "epoch": 2.450939457202505, + "grad_norm": 1.0358234643936157, + "learning_rate": 3.3268338623512094e-06, + "loss": 0.6638, + "step": 1564 + }, + { + "epoch": 2.4525052192066807, + "grad_norm": 1.1351442337036133, + "learning_rate": 3.3248453736261247e-06, + "loss": 0.6356, + "step": 1565 + }, + { + "epoch": 2.454070981210856, + "grad_norm": 1.010600209236145, + "learning_rate": 3.322856299140038e-06, + "loss": 0.6722, + "step": 1566 + }, + { + "epoch": 2.4556367432150314, + "grad_norm": 1.216776728630066, + "learning_rate": 3.320866640305482e-06, + "loss": 0.658, + "step": 1567 + }, + { + "epoch": 2.4572025052192066, + "grad_norm": 1.0781611204147339, + "learning_rate": 3.3188763985354055e-06, + "loss": 0.711, + "step": 1568 + }, + { + "epoch": 2.458768267223382, + "grad_norm": 1.1265203952789307, + "learning_rate": 3.3168855752431732e-06, + "loss": 0.6746, + "step": 1569 + }, + { + "epoch": 2.4603340292275573, + "grad_norm": 1.1726053953170776, + "learning_rate": 3.3148941718425606e-06, + "loss": 0.6836, + "step": 1570 + }, + { + "epoch": 2.461899791231733, + "grad_norm": 1.1113516092300415, + "learning_rate": 3.3129021897477546e-06, + "loss": 0.6322, + "step": 1571 + }, + { + "epoch": 2.463465553235908, + "grad_norm": 1.0234968662261963, + "learning_rate": 3.3109096303733564e-06, + "loss": 0.7179, + "step": 1572 + }, + { + "epoch": 2.4650313152400836, + "grad_norm": 1.1797139644622803, + "learning_rate": 3.308916495134372e-06, + "loss": 0.6996, + "step": 1573 + }, + { + "epoch": 2.4665970772442587, + "grad_norm": 0.9623079895973206, + "learning_rate": 3.3069227854462216e-06, + "loss": 0.6583, + "step": 1574 + }, + { + "epoch": 2.4681628392484343, + "grad_norm": 1.1546630859375, + "learning_rate": 3.30492850272473e-06, + "loss": 0.7457, + "step": 1575 + }, + { + "epoch": 2.4697286012526094, + "grad_norm": 1.0559042692184448, + "learning_rate": 3.302933648386131e-06, + "loss": 0.692, + "step": 1576 + }, + { + "epoch": 2.471294363256785, + "grad_norm": 1.0812047719955444, + "learning_rate": 3.300938223847062e-06, + "loss": 0.6795, + "step": 1577 + }, + { + "epoch": 2.4728601252609606, + "grad_norm": 1.0247067213058472, + "learning_rate": 3.298942230524567e-06, + "loss": 0.695, + "step": 1578 + }, + { + "epoch": 2.4744258872651357, + "grad_norm": 1.354367971420288, + "learning_rate": 3.296945669836094e-06, + "loss": 0.7014, + "step": 1579 + }, + { + "epoch": 2.4759916492693113, + "grad_norm": 1.0897941589355469, + "learning_rate": 3.2949485431994945e-06, + "loss": 0.6338, + "step": 1580 + }, + { + "epoch": 2.4775574112734864, + "grad_norm": 0.9883112907409668, + "learning_rate": 3.292950852033018e-06, + "loss": 0.7102, + "step": 1581 + }, + { + "epoch": 2.479123173277662, + "grad_norm": 1.1789237260818481, + "learning_rate": 3.290952597755319e-06, + "loss": 0.6802, + "step": 1582 + }, + { + "epoch": 2.480688935281837, + "grad_norm": 1.1877659559249878, + "learning_rate": 3.288953781785452e-06, + "loss": 0.6846, + "step": 1583 + }, + { + "epoch": 2.4822546972860127, + "grad_norm": 1.0334861278533936, + "learning_rate": 3.2869544055428678e-06, + "loss": 0.7138, + "step": 1584 + }, + { + "epoch": 2.483820459290188, + "grad_norm": 1.2335622310638428, + "learning_rate": 3.284954470447416e-06, + "loss": 0.6125, + "step": 1585 + }, + { + "epoch": 2.4853862212943634, + "grad_norm": 1.1405894756317139, + "learning_rate": 3.282953977919345e-06, + "loss": 0.6801, + "step": 1586 + }, + { + "epoch": 2.4869519832985385, + "grad_norm": 1.1099573373794556, + "learning_rate": 3.2809529293792953e-06, + "loss": 0.7071, + "step": 1587 + }, + { + "epoch": 2.488517745302714, + "grad_norm": 1.1282474994659424, + "learning_rate": 3.2789513262483053e-06, + "loss": 0.7082, + "step": 1588 + }, + { + "epoch": 2.490083507306889, + "grad_norm": 1.1693743467330933, + "learning_rate": 3.2769491699478077e-06, + "loss": 0.7137, + "step": 1589 + }, + { + "epoch": 2.491649269311065, + "grad_norm": 1.1324049234390259, + "learning_rate": 3.2749464618996244e-06, + "loss": 0.6402, + "step": 1590 + }, + { + "epoch": 2.49321503131524, + "grad_norm": 1.1645187139511108, + "learning_rate": 3.2729432035259733e-06, + "loss": 0.6941, + "step": 1591 + }, + { + "epoch": 2.4947807933194155, + "grad_norm": 1.1647381782531738, + "learning_rate": 3.2709393962494597e-06, + "loss": 0.6583, + "step": 1592 + }, + { + "epoch": 2.4963465553235906, + "grad_norm": 1.1749935150146484, + "learning_rate": 3.268935041493081e-06, + "loss": 0.7099, + "step": 1593 + }, + { + "epoch": 2.497912317327766, + "grad_norm": 1.1859298944473267, + "learning_rate": 3.266930140680222e-06, + "loss": 0.6551, + "step": 1594 + }, + { + "epoch": 2.4994780793319418, + "grad_norm": 1.1642869710922241, + "learning_rate": 3.264924695234657e-06, + "loss": 0.6712, + "step": 1595 + }, + { + "epoch": 2.501043841336117, + "grad_norm": 1.1522237062454224, + "learning_rate": 3.2629187065805445e-06, + "loss": 0.6773, + "step": 1596 + }, + { + "epoch": 2.502609603340292, + "grad_norm": 1.142119288444519, + "learning_rate": 3.2609121761424312e-06, + "loss": 0.6077, + "step": 1597 + }, + { + "epoch": 2.5041753653444676, + "grad_norm": 1.137640357017517, + "learning_rate": 3.258905105345248e-06, + "loss": 0.7086, + "step": 1598 + }, + { + "epoch": 2.505741127348643, + "grad_norm": 1.1106284856796265, + "learning_rate": 3.2568974956143073e-06, + "loss": 0.6795, + "step": 1599 + }, + { + "epoch": 2.5073068893528183, + "grad_norm": 1.0668182373046875, + "learning_rate": 3.254889348375307e-06, + "loss": 0.7163, + "step": 1600 + }, + { + "epoch": 2.508872651356994, + "grad_norm": 1.2222528457641602, + "learning_rate": 3.2528806650543264e-06, + "loss": 0.6589, + "step": 1601 + }, + { + "epoch": 2.510438413361169, + "grad_norm": 1.0996103286743164, + "learning_rate": 3.250871447077824e-06, + "loss": 0.6097, + "step": 1602 + }, + { + "epoch": 2.5120041753653446, + "grad_norm": 1.2791520357131958, + "learning_rate": 3.248861695872639e-06, + "loss": 0.6367, + "step": 1603 + }, + { + "epoch": 2.5135699373695197, + "grad_norm": 1.1636749505996704, + "learning_rate": 3.2468514128659884e-06, + "loss": 0.6271, + "step": 1604 + }, + { + "epoch": 2.5151356993736953, + "grad_norm": 1.1000317335128784, + "learning_rate": 3.2448405994854683e-06, + "loss": 0.6746, + "step": 1605 + }, + { + "epoch": 2.5167014613778704, + "grad_norm": 1.163297414779663, + "learning_rate": 3.2428292571590514e-06, + "loss": 0.6248, + "step": 1606 + }, + { + "epoch": 2.518267223382046, + "grad_norm": 1.0616698265075684, + "learning_rate": 3.240817387315083e-06, + "loss": 0.6249, + "step": 1607 + }, + { + "epoch": 2.519832985386221, + "grad_norm": 1.2272953987121582, + "learning_rate": 3.2388049913822874e-06, + "loss": 0.6856, + "step": 1608 + }, + { + "epoch": 2.5213987473903967, + "grad_norm": 1.0715241432189941, + "learning_rate": 3.2367920707897594e-06, + "loss": 0.6348, + "step": 1609 + }, + { + "epoch": 2.522964509394572, + "grad_norm": 1.069440245628357, + "learning_rate": 3.234778626966968e-06, + "loss": 0.698, + "step": 1610 + }, + { + "epoch": 2.5245302713987474, + "grad_norm": 1.0824602842330933, + "learning_rate": 3.2327646613437525e-06, + "loss": 0.6809, + "step": 1611 + }, + { + "epoch": 2.526096033402923, + "grad_norm": 1.1153802871704102, + "learning_rate": 3.230750175350324e-06, + "loss": 0.6562, + "step": 1612 + }, + { + "epoch": 2.527661795407098, + "grad_norm": 1.1219229698181152, + "learning_rate": 3.228735170417261e-06, + "loss": 0.6819, + "step": 1613 + }, + { + "epoch": 2.5292275574112733, + "grad_norm": 0.9705930352210999, + "learning_rate": 3.226719647975514e-06, + "loss": 0.7457, + "step": 1614 + }, + { + "epoch": 2.530793319415449, + "grad_norm": 1.0846235752105713, + "learning_rate": 3.2247036094563992e-06, + "loss": 0.6122, + "step": 1615 + }, + { + "epoch": 2.5323590814196244, + "grad_norm": 1.0858566761016846, + "learning_rate": 3.222687056291598e-06, + "loss": 0.6141, + "step": 1616 + }, + { + "epoch": 2.5339248434237995, + "grad_norm": 1.0649619102478027, + "learning_rate": 3.2206699899131576e-06, + "loss": 0.6813, + "step": 1617 + }, + { + "epoch": 2.535490605427975, + "grad_norm": 1.2037075757980347, + "learning_rate": 3.2186524117534935e-06, + "loss": 0.7131, + "step": 1618 + }, + { + "epoch": 2.5370563674321502, + "grad_norm": 1.217882513999939, + "learning_rate": 3.216634323245379e-06, + "loss": 0.6977, + "step": 1619 + }, + { + "epoch": 2.538622129436326, + "grad_norm": 1.1430108547210693, + "learning_rate": 3.2146157258219534e-06, + "loss": 0.6799, + "step": 1620 + }, + { + "epoch": 2.540187891440501, + "grad_norm": 1.1914441585540771, + "learning_rate": 3.2125966209167166e-06, + "loss": 0.7284, + "step": 1621 + }, + { + "epoch": 2.5417536534446765, + "grad_norm": 1.1642671823501587, + "learning_rate": 3.210577009963528e-06, + "loss": 0.7203, + "step": 1622 + }, + { + "epoch": 2.5433194154488517, + "grad_norm": 1.0995980501174927, + "learning_rate": 3.208556894396608e-06, + "loss": 0.6635, + "step": 1623 + }, + { + "epoch": 2.5448851774530272, + "grad_norm": 1.1318360567092896, + "learning_rate": 3.2065362756505345e-06, + "loss": 0.6537, + "step": 1624 + }, + { + "epoch": 2.5464509394572024, + "grad_norm": 1.2233246564865112, + "learning_rate": 3.204515155160242e-06, + "loss": 0.7006, + "step": 1625 + }, + { + "epoch": 2.548016701461378, + "grad_norm": 0.9861915707588196, + "learning_rate": 3.202493534361022e-06, + "loss": 0.6647, + "step": 1626 + }, + { + "epoch": 2.549582463465553, + "grad_norm": 1.2189120054244995, + "learning_rate": 3.2004714146885225e-06, + "loss": 0.6374, + "step": 1627 + }, + { + "epoch": 2.5511482254697286, + "grad_norm": 1.165047526359558, + "learning_rate": 3.1984487975787433e-06, + "loss": 0.6287, + "step": 1628 + }, + { + "epoch": 2.552713987473904, + "grad_norm": 1.1922754049301147, + "learning_rate": 3.196425684468039e-06, + "loss": 0.7103, + "step": 1629 + }, + { + "epoch": 2.5542797494780793, + "grad_norm": 1.1970739364624023, + "learning_rate": 3.194402076793116e-06, + "loss": 0.6951, + "step": 1630 + }, + { + "epoch": 2.5558455114822545, + "grad_norm": 1.1461929082870483, + "learning_rate": 3.1923779759910327e-06, + "loss": 0.6989, + "step": 1631 + }, + { + "epoch": 2.55741127348643, + "grad_norm": 1.1536473035812378, + "learning_rate": 3.190353383499198e-06, + "loss": 0.6433, + "step": 1632 + }, + { + "epoch": 2.5589770354906056, + "grad_norm": 1.3109079599380493, + "learning_rate": 3.188328300755367e-06, + "loss": 0.6873, + "step": 1633 + }, + { + "epoch": 2.5605427974947808, + "grad_norm": 1.1532987356185913, + "learning_rate": 3.186302729197645e-06, + "loss": 0.6818, + "step": 1634 + }, + { + "epoch": 2.562108559498956, + "grad_norm": 1.1802146434783936, + "learning_rate": 3.184276670264486e-06, + "loss": 0.7233, + "step": 1635 + }, + { + "epoch": 2.5636743215031315, + "grad_norm": 1.1282404661178589, + "learning_rate": 3.1822501253946875e-06, + "loss": 0.7203, + "step": 1636 + }, + { + "epoch": 2.565240083507307, + "grad_norm": 1.0870387554168701, + "learning_rate": 3.180223096027394e-06, + "loss": 0.6472, + "step": 1637 + }, + { + "epoch": 2.566805845511482, + "grad_norm": 1.038499355316162, + "learning_rate": 3.178195583602092e-06, + "loss": 0.7197, + "step": 1638 + }, + { + "epoch": 2.5683716075156577, + "grad_norm": 1.0605285167694092, + "learning_rate": 3.176167589558613e-06, + "loss": 0.7222, + "step": 1639 + }, + { + "epoch": 2.569937369519833, + "grad_norm": 1.1737507581710815, + "learning_rate": 3.174139115337129e-06, + "loss": 0.6215, + "step": 1640 + }, + { + "epoch": 2.5715031315240084, + "grad_norm": 1.0304044485092163, + "learning_rate": 3.1721101623781556e-06, + "loss": 0.7113, + "step": 1641 + }, + { + "epoch": 2.5730688935281836, + "grad_norm": 1.1509263515472412, + "learning_rate": 3.170080732122545e-06, + "loss": 0.6521, + "step": 1642 + }, + { + "epoch": 2.574634655532359, + "grad_norm": 1.2465925216674805, + "learning_rate": 3.1680508260114882e-06, + "loss": 0.6921, + "step": 1643 + }, + { + "epoch": 2.5762004175365343, + "grad_norm": 1.0584062337875366, + "learning_rate": 3.1660204454865203e-06, + "loss": 0.7041, + "step": 1644 + }, + { + "epoch": 2.57776617954071, + "grad_norm": 1.1217284202575684, + "learning_rate": 3.163989591989505e-06, + "loss": 0.7177, + "step": 1645 + }, + { + "epoch": 2.5793319415448854, + "grad_norm": 1.1175196170806885, + "learning_rate": 3.1619582669626457e-06, + "loss": 0.7192, + "step": 1646 + }, + { + "epoch": 2.5808977035490606, + "grad_norm": 1.1421573162078857, + "learning_rate": 3.1599264718484835e-06, + "loss": 0.7199, + "step": 1647 + }, + { + "epoch": 2.5824634655532357, + "grad_norm": 1.0428645610809326, + "learning_rate": 3.1578942080898876e-06, + "loss": 0.6496, + "step": 1648 + }, + { + "epoch": 2.5840292275574113, + "grad_norm": 0.9507089257240295, + "learning_rate": 3.1558614771300645e-06, + "loss": 0.6271, + "step": 1649 + }, + { + "epoch": 2.585594989561587, + "grad_norm": 1.2682905197143555, + "learning_rate": 3.1538282804125507e-06, + "loss": 0.6635, + "step": 1650 + }, + { + "epoch": 2.587160751565762, + "grad_norm": 1.098144769668579, + "learning_rate": 3.1517946193812132e-06, + "loss": 0.6631, + "step": 1651 + }, + { + "epoch": 2.588726513569937, + "grad_norm": 1.1256821155548096, + "learning_rate": 3.1497604954802485e-06, + "loss": 0.5353, + "step": 1652 + }, + { + "epoch": 2.5902922755741127, + "grad_norm": 1.1058018207550049, + "learning_rate": 3.1477259101541834e-06, + "loss": 0.6324, + "step": 1653 + }, + { + "epoch": 2.5918580375782883, + "grad_norm": 1.200184941291809, + "learning_rate": 3.145690864847871e-06, + "loss": 0.6484, + "step": 1654 + }, + { + "epoch": 2.5934237995824634, + "grad_norm": 1.0515025854110718, + "learning_rate": 3.1436553610064923e-06, + "loss": 0.6774, + "step": 1655 + }, + { + "epoch": 2.594989561586639, + "grad_norm": 1.0765128135681152, + "learning_rate": 3.1416194000755517e-06, + "loss": 0.6838, + "step": 1656 + }, + { + "epoch": 2.596555323590814, + "grad_norm": 1.157872200012207, + "learning_rate": 3.139582983500881e-06, + "loss": 0.7306, + "step": 1657 + }, + { + "epoch": 2.5981210855949897, + "grad_norm": 1.056947946548462, + "learning_rate": 3.1375461127286334e-06, + "loss": 0.628, + "step": 1658 + }, + { + "epoch": 2.599686847599165, + "grad_norm": 1.1467186212539673, + "learning_rate": 3.135508789205285e-06, + "loss": 0.6829, + "step": 1659 + }, + { + "epoch": 2.6012526096033404, + "grad_norm": 1.0062966346740723, + "learning_rate": 3.1334710143776346e-06, + "loss": 0.654, + "step": 1660 + }, + { + "epoch": 2.6028183716075155, + "grad_norm": 1.1658058166503906, + "learning_rate": 3.1314327896928006e-06, + "loss": 0.7067, + "step": 1661 + }, + { + "epoch": 2.604384133611691, + "grad_norm": 1.2473435401916504, + "learning_rate": 3.12939411659822e-06, + "loss": 0.668, + "step": 1662 + }, + { + "epoch": 2.6059498956158667, + "grad_norm": 1.171295404434204, + "learning_rate": 3.127354996541651e-06, + "loss": 0.6847, + "step": 1663 + }, + { + "epoch": 2.607515657620042, + "grad_norm": 1.118043303489685, + "learning_rate": 3.1253154309711663e-06, + "loss": 0.6948, + "step": 1664 + }, + { + "epoch": 2.609081419624217, + "grad_norm": 1.0427628755569458, + "learning_rate": 3.123275421335157e-06, + "loss": 0.7186, + "step": 1665 + }, + { + "epoch": 2.6106471816283925, + "grad_norm": 1.231054425239563, + "learning_rate": 3.1212349690823273e-06, + "loss": 0.6574, + "step": 1666 + }, + { + "epoch": 2.612212943632568, + "grad_norm": 1.0482161045074463, + "learning_rate": 3.1191940756616983e-06, + "loss": 0.7376, + "step": 1667 + }, + { + "epoch": 2.613778705636743, + "grad_norm": 1.2325738668441772, + "learning_rate": 3.1171527425226027e-06, + "loss": 0.6741, + "step": 1668 + }, + { + "epoch": 2.6153444676409183, + "grad_norm": 1.183379888534546, + "learning_rate": 3.1151109711146854e-06, + "loss": 0.7424, + "step": 1669 + }, + { + "epoch": 2.616910229645094, + "grad_norm": 1.1731956005096436, + "learning_rate": 3.1130687628879047e-06, + "loss": 0.7913, + "step": 1670 + }, + { + "epoch": 2.6184759916492695, + "grad_norm": 1.155126690864563, + "learning_rate": 3.111026119292526e-06, + "loss": 0.6724, + "step": 1671 + }, + { + "epoch": 2.6200417536534446, + "grad_norm": 1.0947262048721313, + "learning_rate": 3.1089830417791255e-06, + "loss": 0.6959, + "step": 1672 + }, + { + "epoch": 2.62160751565762, + "grad_norm": 1.073900580406189, + "learning_rate": 3.1069395317985884e-06, + "loss": 0.7274, + "step": 1673 + }, + { + "epoch": 2.6231732776617953, + "grad_norm": 1.1424334049224854, + "learning_rate": 3.104895590802105e-06, + "loss": 0.657, + "step": 1674 + }, + { + "epoch": 2.624739039665971, + "grad_norm": 1.100588321685791, + "learning_rate": 3.1028512202411736e-06, + "loss": 0.6451, + "step": 1675 + }, + { + "epoch": 2.626304801670146, + "grad_norm": 1.0387758016586304, + "learning_rate": 3.100806421567596e-06, + "loss": 0.7331, + "step": 1676 + }, + { + "epoch": 2.6278705636743216, + "grad_norm": 1.2019600868225098, + "learning_rate": 3.0987611962334784e-06, + "loss": 0.6224, + "step": 1677 + }, + { + "epoch": 2.6294363256784967, + "grad_norm": 1.0992989540100098, + "learning_rate": 3.0967155456912313e-06, + "loss": 0.6557, + "step": 1678 + }, + { + "epoch": 2.6310020876826723, + "grad_norm": 1.259507417678833, + "learning_rate": 3.0946694713935655e-06, + "loss": 0.6668, + "step": 1679 + }, + { + "epoch": 2.632567849686848, + "grad_norm": 1.1520506143569946, + "learning_rate": 3.0926229747934932e-06, + "loss": 0.7199, + "step": 1680 + }, + { + "epoch": 2.634133611691023, + "grad_norm": 1.0905909538269043, + "learning_rate": 3.0905760573443278e-06, + "loss": 0.6927, + "step": 1681 + }, + { + "epoch": 2.635699373695198, + "grad_norm": 1.1551228761672974, + "learning_rate": 3.0885287204996784e-06, + "loss": 0.6872, + "step": 1682 + }, + { + "epoch": 2.6372651356993737, + "grad_norm": 1.177915096282959, + "learning_rate": 3.0864809657134553e-06, + "loss": 0.6102, + "step": 1683 + }, + { + "epoch": 2.6388308977035493, + "grad_norm": 1.2892553806304932, + "learning_rate": 3.084432794439865e-06, + "loss": 0.6576, + "step": 1684 + }, + { + "epoch": 2.6403966597077244, + "grad_norm": 1.1799864768981934, + "learning_rate": 3.0823842081334075e-06, + "loss": 0.7214, + "step": 1685 + }, + { + "epoch": 2.6419624217118995, + "grad_norm": 1.11050283908844, + "learning_rate": 3.08033520824888e-06, + "loss": 0.6156, + "step": 1686 + }, + { + "epoch": 2.643528183716075, + "grad_norm": 1.15585196018219, + "learning_rate": 3.078285796241372e-06, + "loss": 0.6311, + "step": 1687 + }, + { + "epoch": 2.6450939457202507, + "grad_norm": 1.1588464975357056, + "learning_rate": 3.0762359735662663e-06, + "loss": 0.6804, + "step": 1688 + }, + { + "epoch": 2.646659707724426, + "grad_norm": 1.1857856512069702, + "learning_rate": 3.0741857416792374e-06, + "loss": 0.6737, + "step": 1689 + }, + { + "epoch": 2.6482254697286014, + "grad_norm": 1.1420296430587769, + "learning_rate": 3.0721351020362495e-06, + "loss": 0.6241, + "step": 1690 + }, + { + "epoch": 2.6497912317327765, + "grad_norm": 1.2236177921295166, + "learning_rate": 3.070084056093558e-06, + "loss": 0.7142, + "step": 1691 + }, + { + "epoch": 2.651356993736952, + "grad_norm": 1.1257219314575195, + "learning_rate": 3.0680326053077047e-06, + "loss": 0.5786, + "step": 1692 + }, + { + "epoch": 2.6529227557411272, + "grad_norm": 1.1258952617645264, + "learning_rate": 3.065980751135522e-06, + "loss": 0.7306, + "step": 1693 + }, + { + "epoch": 2.654488517745303, + "grad_norm": 1.1650607585906982, + "learning_rate": 3.0639284950341246e-06, + "loss": 0.6776, + "step": 1694 + }, + { + "epoch": 2.656054279749478, + "grad_norm": 1.1687644720077515, + "learning_rate": 3.0618758384609156e-06, + "loss": 0.6602, + "step": 1695 + }, + { + "epoch": 2.6576200417536535, + "grad_norm": 1.160516381263733, + "learning_rate": 3.0598227828735825e-06, + "loss": 0.6676, + "step": 1696 + }, + { + "epoch": 2.6591858037578286, + "grad_norm": 1.0864505767822266, + "learning_rate": 3.0577693297300947e-06, + "loss": 0.6843, + "step": 1697 + }, + { + "epoch": 2.6607515657620042, + "grad_norm": 1.0814275741577148, + "learning_rate": 3.055715480488706e-06, + "loss": 0.7214, + "step": 1698 + }, + { + "epoch": 2.6623173277661794, + "grad_norm": 1.1549509763717651, + "learning_rate": 3.0536612366079465e-06, + "loss": 0.7756, + "step": 1699 + }, + { + "epoch": 2.663883089770355, + "grad_norm": 1.125248908996582, + "learning_rate": 3.0516065995466336e-06, + "loss": 0.7133, + "step": 1700 + }, + { + "epoch": 2.6654488517745305, + "grad_norm": 1.0562750101089478, + "learning_rate": 3.0495515707638605e-06, + "loss": 0.6655, + "step": 1701 + }, + { + "epoch": 2.6670146137787056, + "grad_norm": 1.1991479396820068, + "learning_rate": 3.0474961517189965e-06, + "loss": 0.7382, + "step": 1702 + }, + { + "epoch": 2.6685803757828808, + "grad_norm": 1.0789530277252197, + "learning_rate": 3.0454403438716918e-06, + "loss": 0.6608, + "step": 1703 + }, + { + "epoch": 2.6701461377870563, + "grad_norm": 1.1776909828186035, + "learning_rate": 3.04338414868187e-06, + "loss": 0.7493, + "step": 1704 + }, + { + "epoch": 2.671711899791232, + "grad_norm": 1.1587637662887573, + "learning_rate": 3.041327567609731e-06, + "loss": 0.7042, + "step": 1705 + }, + { + "epoch": 2.673277661795407, + "grad_norm": 1.151100516319275, + "learning_rate": 3.0392706021157485e-06, + "loss": 0.5642, + "step": 1706 + }, + { + "epoch": 2.6748434237995826, + "grad_norm": 1.098358392715454, + "learning_rate": 3.03721325366067e-06, + "loss": 0.6177, + "step": 1707 + }, + { + "epoch": 2.6764091858037578, + "grad_norm": 1.1176345348358154, + "learning_rate": 3.0351555237055135e-06, + "loss": 0.7474, + "step": 1708 + }, + { + "epoch": 2.6779749478079333, + "grad_norm": 1.1730945110321045, + "learning_rate": 3.0330974137115683e-06, + "loss": 0.5899, + "step": 1709 + }, + { + "epoch": 2.6795407098121085, + "grad_norm": 1.1680086851119995, + "learning_rate": 3.031038925140395e-06, + "loss": 0.6576, + "step": 1710 + }, + { + "epoch": 2.681106471816284, + "grad_norm": 1.1488760709762573, + "learning_rate": 3.028980059453821e-06, + "loss": 0.6541, + "step": 1711 + }, + { + "epoch": 2.682672233820459, + "grad_norm": 1.1394031047821045, + "learning_rate": 3.0269208181139413e-06, + "loss": 0.6725, + "step": 1712 + }, + { + "epoch": 2.6842379958246347, + "grad_norm": 1.105613112449646, + "learning_rate": 3.024861202583122e-06, + "loss": 0.7088, + "step": 1713 + }, + { + "epoch": 2.68580375782881, + "grad_norm": 1.1166354417800903, + "learning_rate": 3.022801214323989e-06, + "loss": 0.7273, + "step": 1714 + }, + { + "epoch": 2.6873695198329854, + "grad_norm": 1.05818772315979, + "learning_rate": 3.020740854799436e-06, + "loss": 0.591, + "step": 1715 + }, + { + "epoch": 2.6889352818371606, + "grad_norm": 1.019797682762146, + "learning_rate": 3.0186801254726213e-06, + "loss": 0.6676, + "step": 1716 + }, + { + "epoch": 2.690501043841336, + "grad_norm": 1.1530799865722656, + "learning_rate": 3.016619027806963e-06, + "loss": 0.6181, + "step": 1717 + }, + { + "epoch": 2.6920668058455117, + "grad_norm": 1.132246971130371, + "learning_rate": 3.0145575632661426e-06, + "loss": 0.6932, + "step": 1718 + }, + { + "epoch": 2.693632567849687, + "grad_norm": 1.2640838623046875, + "learning_rate": 3.0124957333141036e-06, + "loss": 0.7029, + "step": 1719 + }, + { + "epoch": 2.695198329853862, + "grad_norm": 1.1203837394714355, + "learning_rate": 3.0104335394150447e-06, + "loss": 0.6522, + "step": 1720 + }, + { + "epoch": 2.6967640918580376, + "grad_norm": 1.1490654945373535, + "learning_rate": 3.008370983033426e-06, + "loss": 0.6837, + "step": 1721 + }, + { + "epoch": 2.698329853862213, + "grad_norm": 1.0226919651031494, + "learning_rate": 3.006308065633966e-06, + "loss": 0.6344, + "step": 1722 + }, + { + "epoch": 2.6998956158663883, + "grad_norm": 1.2632393836975098, + "learning_rate": 3.004244788681637e-06, + "loss": 0.7157, + "step": 1723 + }, + { + "epoch": 2.701461377870564, + "grad_norm": 1.0831620693206787, + "learning_rate": 3.0021811536416676e-06, + "loss": 0.6671, + "step": 1724 + }, + { + "epoch": 2.703027139874739, + "grad_norm": 1.1284756660461426, + "learning_rate": 3.0001171619795417e-06, + "loss": 0.6669, + "step": 1725 + }, + { + "epoch": 2.7045929018789145, + "grad_norm": 1.0045720338821411, + "learning_rate": 2.9980528151609945e-06, + "loss": 0.7276, + "step": 1726 + }, + { + "epoch": 2.7061586638830897, + "grad_norm": 1.1337859630584717, + "learning_rate": 2.9959881146520164e-06, + "loss": 0.6734, + "step": 1727 + }, + { + "epoch": 2.7077244258872653, + "grad_norm": 1.0121667385101318, + "learning_rate": 2.993923061918845e-06, + "loss": 0.6201, + "step": 1728 + }, + { + "epoch": 2.7092901878914404, + "grad_norm": 1.2311614751815796, + "learning_rate": 2.991857658427972e-06, + "loss": 0.6392, + "step": 1729 + }, + { + "epoch": 2.710855949895616, + "grad_norm": 1.0462316274642944, + "learning_rate": 2.989791905646135e-06, + "loss": 0.6736, + "step": 1730 + }, + { + "epoch": 2.712421711899791, + "grad_norm": 1.1915360689163208, + "learning_rate": 2.9877258050403214e-06, + "loss": 0.5977, + "step": 1731 + }, + { + "epoch": 2.7139874739039667, + "grad_norm": 1.0216567516326904, + "learning_rate": 2.985659358077765e-06, + "loss": 0.6393, + "step": 1732 + }, + { + "epoch": 2.715553235908142, + "grad_norm": 1.114071011543274, + "learning_rate": 2.9835925662259457e-06, + "loss": 0.6854, + "step": 1733 + }, + { + "epoch": 2.7171189979123174, + "grad_norm": 1.2007335424423218, + "learning_rate": 2.981525430952589e-06, + "loss": 0.584, + "step": 1734 + }, + { + "epoch": 2.718684759916493, + "grad_norm": 1.221422791481018, + "learning_rate": 2.979457953725663e-06, + "loss": 0.742, + "step": 1735 + }, + { + "epoch": 2.720250521920668, + "grad_norm": 1.0720511674880981, + "learning_rate": 2.9773901360133804e-06, + "loss": 0.6722, + "step": 1736 + }, + { + "epoch": 2.721816283924843, + "grad_norm": 1.1078705787658691, + "learning_rate": 2.975321979284193e-06, + "loss": 0.6837, + "step": 1737 + }, + { + "epoch": 2.723382045929019, + "grad_norm": 1.1647889614105225, + "learning_rate": 2.973253485006795e-06, + "loss": 0.7406, + "step": 1738 + }, + { + "epoch": 2.7249478079331944, + "grad_norm": 1.1482031345367432, + "learning_rate": 2.971184654650122e-06, + "loss": 0.7109, + "step": 1739 + }, + { + "epoch": 2.7265135699373695, + "grad_norm": 1.095131278038025, + "learning_rate": 2.9691154896833454e-06, + "loss": 0.6795, + "step": 1740 + }, + { + "epoch": 2.7280793319415446, + "grad_norm": 1.0917677879333496, + "learning_rate": 2.9670459915758746e-06, + "loss": 0.7142, + "step": 1741 + }, + { + "epoch": 2.72964509394572, + "grad_norm": 1.0911260843276978, + "learning_rate": 2.9649761617973572e-06, + "loss": 0.6917, + "step": 1742 + }, + { + "epoch": 2.7312108559498958, + "grad_norm": 1.2418321371078491, + "learning_rate": 2.9629060018176757e-06, + "loss": 0.6562, + "step": 1743 + }, + { + "epoch": 2.732776617954071, + "grad_norm": 1.114390254020691, + "learning_rate": 2.9608355131069464e-06, + "loss": 0.6812, + "step": 1744 + }, + { + "epoch": 2.7343423799582465, + "grad_norm": 1.0684012174606323, + "learning_rate": 2.9587646971355194e-06, + "loss": 0.688, + "step": 1745 + }, + { + "epoch": 2.7359081419624216, + "grad_norm": 1.1974529027938843, + "learning_rate": 2.9566935553739773e-06, + "loss": 0.6697, + "step": 1746 + }, + { + "epoch": 2.737473903966597, + "grad_norm": 1.0794286727905273, + "learning_rate": 2.9546220892931336e-06, + "loss": 0.6587, + "step": 1747 + }, + { + "epoch": 2.7390396659707723, + "grad_norm": 1.1143841743469238, + "learning_rate": 2.9525503003640336e-06, + "loss": 0.6814, + "step": 1748 + }, + { + "epoch": 2.740605427974948, + "grad_norm": 1.1970027685165405, + "learning_rate": 2.9504781900579504e-06, + "loss": 0.6714, + "step": 1749 + }, + { + "epoch": 2.742171189979123, + "grad_norm": 1.0299553871154785, + "learning_rate": 2.948405759846386e-06, + "loss": 0.6707, + "step": 1750 + }, + { + "epoch": 2.7437369519832986, + "grad_norm": 1.103028416633606, + "learning_rate": 2.946333011201068e-06, + "loss": 0.6163, + "step": 1751 + }, + { + "epoch": 2.745302713987474, + "grad_norm": 1.1230628490447998, + "learning_rate": 2.944259945593953e-06, + "loss": 0.5917, + "step": 1752 + }, + { + "epoch": 2.7468684759916493, + "grad_norm": 1.1009881496429443, + "learning_rate": 2.942186564497221e-06, + "loss": 0.7048, + "step": 1753 + }, + { + "epoch": 2.7484342379958244, + "grad_norm": 1.1123842000961304, + "learning_rate": 2.9401128693832752e-06, + "loss": 0.6063, + "step": 1754 + }, + { + "epoch": 2.75, + "grad_norm": 1.0923182964324951, + "learning_rate": 2.9380388617247433e-06, + "loss": 0.663, + "step": 1755 + }, + { + "epoch": 2.7515657620041756, + "grad_norm": 1.1761754751205444, + "learning_rate": 2.935964542994475e-06, + "loss": 0.6494, + "step": 1756 + }, + { + "epoch": 2.7531315240083507, + "grad_norm": 1.0580326318740845, + "learning_rate": 2.9338899146655396e-06, + "loss": 0.7391, + "step": 1757 + }, + { + "epoch": 2.754697286012526, + "grad_norm": 1.1655254364013672, + "learning_rate": 2.931814978211227e-06, + "loss": 0.6595, + "step": 1758 + }, + { + "epoch": 2.7562630480167014, + "grad_norm": 1.1162359714508057, + "learning_rate": 2.929739735105047e-06, + "loss": 0.6438, + "step": 1759 + }, + { + "epoch": 2.757828810020877, + "grad_norm": 1.173933744430542, + "learning_rate": 2.9276641868207244e-06, + "loss": 0.5591, + "step": 1760 + }, + { + "epoch": 2.759394572025052, + "grad_norm": 1.2183250188827515, + "learning_rate": 2.9255883348322038e-06, + "loss": 0.7317, + "step": 1761 + }, + { + "epoch": 2.7609603340292277, + "grad_norm": 1.0836644172668457, + "learning_rate": 2.9235121806136446e-06, + "loss": 0.7539, + "step": 1762 + }, + { + "epoch": 2.762526096033403, + "grad_norm": 1.1108825206756592, + "learning_rate": 2.921435725639419e-06, + "loss": 0.6598, + "step": 1763 + }, + { + "epoch": 2.7640918580375784, + "grad_norm": 1.090196967124939, + "learning_rate": 2.9193589713841132e-06, + "loss": 0.7412, + "step": 1764 + }, + { + "epoch": 2.7656576200417535, + "grad_norm": 1.1506152153015137, + "learning_rate": 2.9172819193225296e-06, + "loss": 0.6857, + "step": 1765 + }, + { + "epoch": 2.767223382045929, + "grad_norm": 1.127022385597229, + "learning_rate": 2.9152045709296766e-06, + "loss": 0.6383, + "step": 1766 + }, + { + "epoch": 2.7687891440501042, + "grad_norm": 1.1096911430358887, + "learning_rate": 2.9131269276807782e-06, + "loss": 0.6363, + "step": 1767 + }, + { + "epoch": 2.77035490605428, + "grad_norm": 1.1879534721374512, + "learning_rate": 2.911048991051263e-06, + "loss": 0.7313, + "step": 1768 + }, + { + "epoch": 2.7719206680584554, + "grad_norm": 1.0864057540893555, + "learning_rate": 2.908970762516772e-06, + "loss": 0.6602, + "step": 1769 + }, + { + "epoch": 2.7734864300626305, + "grad_norm": 1.174880027770996, + "learning_rate": 2.9068922435531515e-06, + "loss": 0.6731, + "step": 1770 + }, + { + "epoch": 2.7750521920668056, + "grad_norm": 1.1670523881912231, + "learning_rate": 2.904813435636453e-06, + "loss": 0.5982, + "step": 1771 + }, + { + "epoch": 2.776617954070981, + "grad_norm": 1.016558289527893, + "learning_rate": 2.902734340242937e-06, + "loss": 0.5549, + "step": 1772 + }, + { + "epoch": 2.778183716075157, + "grad_norm": 1.2644622325897217, + "learning_rate": 2.9006549588490636e-06, + "loss": 0.6209, + "step": 1773 + }, + { + "epoch": 2.779749478079332, + "grad_norm": 1.1689661741256714, + "learning_rate": 2.8985752929314985e-06, + "loss": 0.6629, + "step": 1774 + }, + { + "epoch": 2.781315240083507, + "grad_norm": 1.123618721961975, + "learning_rate": 2.8964953439671095e-06, + "loss": 0.6986, + "step": 1775 + }, + { + "epoch": 2.7828810020876826, + "grad_norm": 1.1276949644088745, + "learning_rate": 2.8944151134329646e-06, + "loss": 0.6125, + "step": 1776 + }, + { + "epoch": 2.784446764091858, + "grad_norm": 1.1804574728012085, + "learning_rate": 2.8923346028063328e-06, + "loss": 0.5912, + "step": 1777 + }, + { + "epoch": 2.7860125260960333, + "grad_norm": 1.1222927570343018, + "learning_rate": 2.89025381356468e-06, + "loss": 0.5837, + "step": 1778 + }, + { + "epoch": 2.787578288100209, + "grad_norm": 1.063511848449707, + "learning_rate": 2.888172747185673e-06, + "loss": 0.7243, + "step": 1779 + }, + { + "epoch": 2.789144050104384, + "grad_norm": 1.1183874607086182, + "learning_rate": 2.8860914051471722e-06, + "loss": 0.7204, + "step": 1780 + }, + { + "epoch": 2.7907098121085596, + "grad_norm": 1.135906457901001, + "learning_rate": 2.884009788927235e-06, + "loss": 0.7308, + "step": 1781 + }, + { + "epoch": 2.7922755741127347, + "grad_norm": 1.2013477087020874, + "learning_rate": 2.8819279000041156e-06, + "loss": 0.7004, + "step": 1782 + }, + { + "epoch": 2.7938413361169103, + "grad_norm": 1.074418306350708, + "learning_rate": 2.8798457398562584e-06, + "loss": 0.6404, + "step": 1783 + }, + { + "epoch": 2.7954070981210855, + "grad_norm": 1.0827600955963135, + "learning_rate": 2.8777633099623027e-06, + "loss": 0.7488, + "step": 1784 + }, + { + "epoch": 2.796972860125261, + "grad_norm": 1.1087040901184082, + "learning_rate": 2.8756806118010784e-06, + "loss": 0.6554, + "step": 1785 + }, + { + "epoch": 2.798538622129436, + "grad_norm": 1.0850439071655273, + "learning_rate": 2.8735976468516057e-06, + "loss": 0.6501, + "step": 1786 + }, + { + "epoch": 2.8001043841336117, + "grad_norm": 1.129045844078064, + "learning_rate": 2.871514416593096e-06, + "loss": 0.6697, + "step": 1787 + }, + { + "epoch": 2.801670146137787, + "grad_norm": 1.0854123830795288, + "learning_rate": 2.869430922504947e-06, + "loss": 0.6787, + "step": 1788 + }, + { + "epoch": 2.8032359081419624, + "grad_norm": 1.2950549125671387, + "learning_rate": 2.8673471660667448e-06, + "loss": 0.6682, + "step": 1789 + }, + { + "epoch": 2.804801670146138, + "grad_norm": 1.0760269165039062, + "learning_rate": 2.865263148758261e-06, + "loss": 0.6846, + "step": 1790 + }, + { + "epoch": 2.806367432150313, + "grad_norm": 1.1861540079116821, + "learning_rate": 2.863178872059454e-06, + "loss": 0.6029, + "step": 1791 + }, + { + "epoch": 2.8079331941544883, + "grad_norm": 1.081518292427063, + "learning_rate": 2.8610943374504645e-06, + "loss": 0.6801, + "step": 1792 + }, + { + "epoch": 2.809498956158664, + "grad_norm": 1.1814477443695068, + "learning_rate": 2.8590095464116192e-06, + "loss": 0.68, + "step": 1793 + }, + { + "epoch": 2.8110647181628394, + "grad_norm": 1.154828667640686, + "learning_rate": 2.856924500423422e-06, + "loss": 0.6164, + "step": 1794 + }, + { + "epoch": 2.8126304801670146, + "grad_norm": 1.0605709552764893, + "learning_rate": 2.8548392009665638e-06, + "loss": 0.6146, + "step": 1795 + }, + { + "epoch": 2.81419624217119, + "grad_norm": 1.1769119501113892, + "learning_rate": 2.852753649521911e-06, + "loss": 0.6789, + "step": 1796 + }, + { + "epoch": 2.8157620041753653, + "grad_norm": 1.006847858428955, + "learning_rate": 2.8506678475705114e-06, + "loss": 0.6719, + "step": 1797 + }, + { + "epoch": 2.817327766179541, + "grad_norm": 1.1305062770843506, + "learning_rate": 2.8485817965935886e-06, + "loss": 0.6548, + "step": 1798 + }, + { + "epoch": 2.818893528183716, + "grad_norm": 1.0815331935882568, + "learning_rate": 2.846495498072545e-06, + "loss": 0.7306, + "step": 1799 + }, + { + "epoch": 2.8204592901878915, + "grad_norm": 1.1019850969314575, + "learning_rate": 2.8444089534889587e-06, + "loss": 0.7188, + "step": 1800 + }, + { + "epoch": 2.8220250521920667, + "grad_norm": 1.1405457258224487, + "learning_rate": 2.8423221643245807e-06, + "loss": 0.6482, + "step": 1801 + }, + { + "epoch": 2.8235908141962422, + "grad_norm": 1.1295759677886963, + "learning_rate": 2.8402351320613374e-06, + "loss": 0.5882, + "step": 1802 + }, + { + "epoch": 2.8251565762004174, + "grad_norm": 1.1165974140167236, + "learning_rate": 2.8381478581813275e-06, + "loss": 0.6714, + "step": 1803 + }, + { + "epoch": 2.826722338204593, + "grad_norm": 1.0910004377365112, + "learning_rate": 2.836060344166821e-06, + "loss": 0.6971, + "step": 1804 + }, + { + "epoch": 2.828288100208768, + "grad_norm": 1.093596339225769, + "learning_rate": 2.833972591500259e-06, + "loss": 0.7041, + "step": 1805 + }, + { + "epoch": 2.8298538622129437, + "grad_norm": 1.1507813930511475, + "learning_rate": 2.8318846016642507e-06, + "loss": 0.6455, + "step": 1806 + }, + { + "epoch": 2.8314196242171192, + "grad_norm": 1.055422306060791, + "learning_rate": 2.8297963761415748e-06, + "loss": 0.7499, + "step": 1807 + }, + { + "epoch": 2.8329853862212944, + "grad_norm": 1.0835453271865845, + "learning_rate": 2.8277079164151782e-06, + "loss": 0.6739, + "step": 1808 + }, + { + "epoch": 2.8345511482254695, + "grad_norm": 1.247808575630188, + "learning_rate": 2.8256192239681723e-06, + "loss": 0.6774, + "step": 1809 + }, + { + "epoch": 2.836116910229645, + "grad_norm": 1.0836130380630493, + "learning_rate": 2.8235303002838343e-06, + "loss": 0.6592, + "step": 1810 + }, + { + "epoch": 2.8376826722338206, + "grad_norm": 1.199852466583252, + "learning_rate": 2.8214411468456067e-06, + "loss": 0.6728, + "step": 1811 + }, + { + "epoch": 2.8392484342379958, + "grad_norm": 1.0706546306610107, + "learning_rate": 2.8193517651370934e-06, + "loss": 0.7438, + "step": 1812 + }, + { + "epoch": 2.8408141962421714, + "grad_norm": 1.301465630531311, + "learning_rate": 2.8172621566420633e-06, + "loss": 0.6736, + "step": 1813 + }, + { + "epoch": 2.8423799582463465, + "grad_norm": 1.0916422605514526, + "learning_rate": 2.815172322844442e-06, + "loss": 0.7258, + "step": 1814 + }, + { + "epoch": 2.843945720250522, + "grad_norm": 1.1184017658233643, + "learning_rate": 2.8130822652283183e-06, + "loss": 0.6831, + "step": 1815 + }, + { + "epoch": 2.845511482254697, + "grad_norm": 1.1714346408843994, + "learning_rate": 2.8109919852779406e-06, + "loss": 0.6866, + "step": 1816 + }, + { + "epoch": 2.8470772442588728, + "grad_norm": 1.0131067037582397, + "learning_rate": 2.808901484477712e-06, + "loss": 0.6574, + "step": 1817 + }, + { + "epoch": 2.848643006263048, + "grad_norm": 1.2358931303024292, + "learning_rate": 2.8068107643121952e-06, + "loss": 0.5596, + "step": 1818 + }, + { + "epoch": 2.8502087682672235, + "grad_norm": 0.9713279604911804, + "learning_rate": 2.8047198262661074e-06, + "loss": 0.6502, + "step": 1819 + }, + { + "epoch": 2.8517745302713986, + "grad_norm": 1.0890235900878906, + "learning_rate": 2.80262867182432e-06, + "loss": 0.6566, + "step": 1820 + }, + { + "epoch": 2.853340292275574, + "grad_norm": 1.084876298904419, + "learning_rate": 2.8005373024718597e-06, + "loss": 0.6356, + "step": 1821 + }, + { + "epoch": 2.8549060542797493, + "grad_norm": 1.1338250637054443, + "learning_rate": 2.7984457196939056e-06, + "loss": 0.7226, + "step": 1822 + }, + { + "epoch": 2.856471816283925, + "grad_norm": 1.1295956373214722, + "learning_rate": 2.7963539249757872e-06, + "loss": 0.724, + "step": 1823 + }, + { + "epoch": 2.8580375782881005, + "grad_norm": 1.0954629182815552, + "learning_rate": 2.794261919802984e-06, + "loss": 0.685, + "step": 1824 + }, + { + "epoch": 2.8596033402922756, + "grad_norm": 1.0435383319854736, + "learning_rate": 2.792169705661128e-06, + "loss": 0.7532, + "step": 1825 + }, + { + "epoch": 2.8611691022964507, + "grad_norm": 1.1580495834350586, + "learning_rate": 2.790077284035996e-06, + "loss": 0.6745, + "step": 1826 + }, + { + "epoch": 2.8627348643006263, + "grad_norm": 1.0154657363891602, + "learning_rate": 2.787984656413514e-06, + "loss": 0.7312, + "step": 1827 + }, + { + "epoch": 2.864300626304802, + "grad_norm": 1.1673710346221924, + "learning_rate": 2.785891824279755e-06, + "loss": 0.6769, + "step": 1828 + }, + { + "epoch": 2.865866388308977, + "grad_norm": 1.1436103582382202, + "learning_rate": 2.7837987891209352e-06, + "loss": 0.6987, + "step": 1829 + }, + { + "epoch": 2.867432150313152, + "grad_norm": 1.2176343202590942, + "learning_rate": 2.7817055524234167e-06, + "loss": 0.6768, + "step": 1830 + }, + { + "epoch": 2.8689979123173277, + "grad_norm": 1.1192113161087036, + "learning_rate": 2.7796121156737042e-06, + "loss": 0.6468, + "step": 1831 + }, + { + "epoch": 2.8705636743215033, + "grad_norm": 1.085511326789856, + "learning_rate": 2.777518480358444e-06, + "loss": 0.6923, + "step": 1832 + }, + { + "epoch": 2.8721294363256784, + "grad_norm": 1.1534816026687622, + "learning_rate": 2.7754246479644227e-06, + "loss": 0.7897, + "step": 1833 + }, + { + "epoch": 2.873695198329854, + "grad_norm": 1.2280727624893188, + "learning_rate": 2.7733306199785704e-06, + "loss": 0.5913, + "step": 1834 + }, + { + "epoch": 2.875260960334029, + "grad_norm": 1.1843805313110352, + "learning_rate": 2.771236397887952e-06, + "loss": 0.6554, + "step": 1835 + }, + { + "epoch": 2.8768267223382047, + "grad_norm": 1.1090878248214722, + "learning_rate": 2.7691419831797724e-06, + "loss": 0.7155, + "step": 1836 + }, + { + "epoch": 2.87839248434238, + "grad_norm": 1.3048434257507324, + "learning_rate": 2.767047377341371e-06, + "loss": 0.689, + "step": 1837 + }, + { + "epoch": 2.8799582463465554, + "grad_norm": 1.0179991722106934, + "learning_rate": 2.764952581860227e-06, + "loss": 0.6919, + "step": 1838 + }, + { + "epoch": 2.8815240083507305, + "grad_norm": 1.0855897665023804, + "learning_rate": 2.7628575982239513e-06, + "loss": 0.7377, + "step": 1839 + }, + { + "epoch": 2.883089770354906, + "grad_norm": 1.0885690450668335, + "learning_rate": 2.7607624279202887e-06, + "loss": 0.7436, + "step": 1840 + }, + { + "epoch": 2.8846555323590817, + "grad_norm": 0.9842485189437866, + "learning_rate": 2.758667072437117e-06, + "loss": 0.6818, + "step": 1841 + }, + { + "epoch": 2.886221294363257, + "grad_norm": 1.2647449970245361, + "learning_rate": 2.7565715332624455e-06, + "loss": 0.6892, + "step": 1842 + }, + { + "epoch": 2.887787056367432, + "grad_norm": 1.0584148168563843, + "learning_rate": 2.7544758118844134e-06, + "loss": 0.6945, + "step": 1843 + }, + { + "epoch": 2.8893528183716075, + "grad_norm": 1.2626157999038696, + "learning_rate": 2.7523799097912905e-06, + "loss": 0.6617, + "step": 1844 + }, + { + "epoch": 2.890918580375783, + "grad_norm": 1.1559371948242188, + "learning_rate": 2.7502838284714746e-06, + "loss": 0.679, + "step": 1845 + }, + { + "epoch": 2.892484342379958, + "grad_norm": 1.1553645133972168, + "learning_rate": 2.7481875694134884e-06, + "loss": 0.7216, + "step": 1846 + }, + { + "epoch": 2.8940501043841333, + "grad_norm": 1.1300830841064453, + "learning_rate": 2.7460911341059847e-06, + "loss": 0.6761, + "step": 1847 + }, + { + "epoch": 2.895615866388309, + "grad_norm": 1.1680023670196533, + "learning_rate": 2.7439945240377395e-06, + "loss": 0.675, + "step": 1848 + }, + { + "epoch": 2.8971816283924845, + "grad_norm": 1.0944335460662842, + "learning_rate": 2.7418977406976526e-06, + "loss": 0.6959, + "step": 1849 + }, + { + "epoch": 2.8987473903966596, + "grad_norm": 1.1949477195739746, + "learning_rate": 2.7398007855747455e-06, + "loss": 0.6822, + "step": 1850 + }, + { + "epoch": 2.900313152400835, + "grad_norm": 1.2108348608016968, + "learning_rate": 2.737703660158167e-06, + "loss": 0.6568, + "step": 1851 + }, + { + "epoch": 2.9018789144050103, + "grad_norm": 1.119547963142395, + "learning_rate": 2.73560636593718e-06, + "loss": 0.6508, + "step": 1852 + }, + { + "epoch": 2.903444676409186, + "grad_norm": 1.1591404676437378, + "learning_rate": 2.733508904401172e-06, + "loss": 0.6462, + "step": 1853 + }, + { + "epoch": 2.905010438413361, + "grad_norm": 1.1378757953643799, + "learning_rate": 2.7314112770396474e-06, + "loss": 0.6202, + "step": 1854 + }, + { + "epoch": 2.9065762004175366, + "grad_norm": 1.1146254539489746, + "learning_rate": 2.729313485342229e-06, + "loss": 0.6618, + "step": 1855 + }, + { + "epoch": 2.9081419624217117, + "grad_norm": 1.1829131841659546, + "learning_rate": 2.727215530798656e-06, + "loss": 0.6708, + "step": 1856 + }, + { + "epoch": 2.9097077244258873, + "grad_norm": 1.0372153520584106, + "learning_rate": 2.725117414898784e-06, + "loss": 0.6666, + "step": 1857 + }, + { + "epoch": 2.911273486430063, + "grad_norm": 1.1896759271621704, + "learning_rate": 2.723019139132581e-06, + "loss": 0.7169, + "step": 1858 + }, + { + "epoch": 2.912839248434238, + "grad_norm": 1.137988805770874, + "learning_rate": 2.720920704990131e-06, + "loss": 0.6872, + "step": 1859 + }, + { + "epoch": 2.914405010438413, + "grad_norm": 1.1891788244247437, + "learning_rate": 2.7188221139616303e-06, + "loss": 0.7105, + "step": 1860 + }, + { + "epoch": 2.9159707724425887, + "grad_norm": 1.0640588998794556, + "learning_rate": 2.716723367537385e-06, + "loss": 0.6905, + "step": 1861 + }, + { + "epoch": 2.9175365344467643, + "grad_norm": 1.2714836597442627, + "learning_rate": 2.7146244672078125e-06, + "loss": 0.6713, + "step": 1862 + }, + { + "epoch": 2.9191022964509394, + "grad_norm": 1.222401738166809, + "learning_rate": 2.7125254144634388e-06, + "loss": 0.6638, + "step": 1863 + }, + { + "epoch": 2.9206680584551146, + "grad_norm": 1.0254578590393066, + "learning_rate": 2.7104262107948993e-06, + "loss": 0.7374, + "step": 1864 + }, + { + "epoch": 2.92223382045929, + "grad_norm": 1.089826226234436, + "learning_rate": 2.7083268576929368e-06, + "loss": 0.6742, + "step": 1865 + }, + { + "epoch": 2.9237995824634657, + "grad_norm": 1.1508773565292358, + "learning_rate": 2.7062273566483984e-06, + "loss": 0.5628, + "step": 1866 + }, + { + "epoch": 2.925365344467641, + "grad_norm": 1.1214534044265747, + "learning_rate": 2.7041277091522365e-06, + "loss": 0.5819, + "step": 1867 + }, + { + "epoch": 2.9269311064718164, + "grad_norm": 1.2162528038024902, + "learning_rate": 2.70202791669551e-06, + "loss": 0.6774, + "step": 1868 + }, + { + "epoch": 2.9284968684759916, + "grad_norm": 1.130321741104126, + "learning_rate": 2.6999279807693783e-06, + "loss": 0.6617, + "step": 1869 + }, + { + "epoch": 2.930062630480167, + "grad_norm": 1.1052041053771973, + "learning_rate": 2.697827902865103e-06, + "loss": 0.6348, + "step": 1870 + }, + { + "epoch": 2.9316283924843423, + "grad_norm": 1.243033766746521, + "learning_rate": 2.695727684474048e-06, + "loss": 0.671, + "step": 1871 + }, + { + "epoch": 2.933194154488518, + "grad_norm": 1.163845419883728, + "learning_rate": 2.693627327087675e-06, + "loss": 0.6605, + "step": 1872 + }, + { + "epoch": 2.934759916492693, + "grad_norm": 1.0230295658111572, + "learning_rate": 2.6915268321975454e-06, + "loss": 0.6511, + "step": 1873 + }, + { + "epoch": 2.9363256784968685, + "grad_norm": 1.0699726343154907, + "learning_rate": 2.6894262012953194e-06, + "loss": 0.5641, + "step": 1874 + }, + { + "epoch": 2.937891440501044, + "grad_norm": 1.139090657234192, + "learning_rate": 2.6873254358727503e-06, + "loss": 0.7327, + "step": 1875 + }, + { + "epoch": 2.9394572025052192, + "grad_norm": 1.0480340719223022, + "learning_rate": 2.68522453742169e-06, + "loss": 0.6559, + "step": 1876 + }, + { + "epoch": 2.9410229645093944, + "grad_norm": 1.1894673109054565, + "learning_rate": 2.6831235074340854e-06, + "loss": 0.6235, + "step": 1877 + }, + { + "epoch": 2.94258872651357, + "grad_norm": 1.1843831539154053, + "learning_rate": 2.681022347401975e-06, + "loss": 0.7161, + "step": 1878 + }, + { + "epoch": 2.9441544885177455, + "grad_norm": 1.278631329536438, + "learning_rate": 2.6789210588174887e-06, + "loss": 0.6651, + "step": 1879 + }, + { + "epoch": 2.9457202505219207, + "grad_norm": 1.1330087184906006, + "learning_rate": 2.67681964317285e-06, + "loss": 0.682, + "step": 1880 + }, + { + "epoch": 2.947286012526096, + "grad_norm": 1.1718090772628784, + "learning_rate": 2.6747181019603718e-06, + "loss": 0.6656, + "step": 1881 + }, + { + "epoch": 2.9488517745302714, + "grad_norm": 1.0476222038269043, + "learning_rate": 2.672616436672456e-06, + "loss": 0.543, + "step": 1882 + }, + { + "epoch": 2.950417536534447, + "grad_norm": 1.1602777242660522, + "learning_rate": 2.6705146488015933e-06, + "loss": 0.6887, + "step": 1883 + }, + { + "epoch": 2.951983298538622, + "grad_norm": 1.1091994047164917, + "learning_rate": 2.66841273984036e-06, + "loss": 0.7068, + "step": 1884 + }, + { + "epoch": 2.9535490605427976, + "grad_norm": 1.1577891111373901, + "learning_rate": 2.666310711281419e-06, + "loss": 0.7285, + "step": 1885 + }, + { + "epoch": 2.9551148225469728, + "grad_norm": 1.1392951011657715, + "learning_rate": 2.6642085646175196e-06, + "loss": 0.6308, + "step": 1886 + }, + { + "epoch": 2.9566805845511483, + "grad_norm": 1.3026043176651, + "learning_rate": 2.6621063013414934e-06, + "loss": 0.7092, + "step": 1887 + }, + { + "epoch": 2.9582463465553235, + "grad_norm": 1.0123710632324219, + "learning_rate": 2.660003922946255e-06, + "loss": 0.7045, + "step": 1888 + }, + { + "epoch": 2.959812108559499, + "grad_norm": 1.0754269361495972, + "learning_rate": 2.6579014309247998e-06, + "loss": 0.7301, + "step": 1889 + }, + { + "epoch": 2.961377870563674, + "grad_norm": 1.19624662399292, + "learning_rate": 2.655798826770206e-06, + "loss": 0.699, + "step": 1890 + }, + { + "epoch": 2.9629436325678498, + "grad_norm": 1.0728397369384766, + "learning_rate": 2.6536961119756317e-06, + "loss": 0.7188, + "step": 1891 + }, + { + "epoch": 2.964509394572025, + "grad_norm": 1.231490135192871, + "learning_rate": 2.6515932880343103e-06, + "loss": 0.7309, + "step": 1892 + }, + { + "epoch": 2.9660751565762005, + "grad_norm": 1.2304288148880005, + "learning_rate": 2.6494903564395563e-06, + "loss": 0.7267, + "step": 1893 + }, + { + "epoch": 2.9676409185803756, + "grad_norm": 1.0533643960952759, + "learning_rate": 2.6473873186847577e-06, + "loss": 0.7011, + "step": 1894 + }, + { + "epoch": 2.969206680584551, + "grad_norm": 1.1855214834213257, + "learning_rate": 2.6452841762633795e-06, + "loss": 0.707, + "step": 1895 + }, + { + "epoch": 2.9707724425887267, + "grad_norm": 1.1279842853546143, + "learning_rate": 2.643180930668962e-06, + "loss": 0.6033, + "step": 1896 + }, + { + "epoch": 2.972338204592902, + "grad_norm": 1.102318525314331, + "learning_rate": 2.6410775833951164e-06, + "loss": 0.6696, + "step": 1897 + }, + { + "epoch": 2.973903966597077, + "grad_norm": 1.1372147798538208, + "learning_rate": 2.6389741359355275e-06, + "loss": 0.7164, + "step": 1898 + }, + { + "epoch": 2.9754697286012526, + "grad_norm": 1.2800220251083374, + "learning_rate": 2.6368705897839515e-06, + "loss": 0.733, + "step": 1899 + }, + { + "epoch": 2.977035490605428, + "grad_norm": 1.0855633020401, + "learning_rate": 2.634766946434214e-06, + "loss": 0.5894, + "step": 1900 + }, + { + "epoch": 2.9786012526096033, + "grad_norm": 1.1212704181671143, + "learning_rate": 2.6326632073802092e-06, + "loss": 0.6753, + "step": 1901 + }, + { + "epoch": 2.980167014613779, + "grad_norm": 1.0885487794876099, + "learning_rate": 2.6305593741159004e-06, + "loss": 0.6802, + "step": 1902 + }, + { + "epoch": 2.981732776617954, + "grad_norm": 1.1605111360549927, + "learning_rate": 2.6284554481353182e-06, + "loss": 0.5837, + "step": 1903 + }, + { + "epoch": 2.9832985386221296, + "grad_norm": 1.1119557619094849, + "learning_rate": 2.626351430932556e-06, + "loss": 0.6767, + "step": 1904 + }, + { + "epoch": 2.9848643006263047, + "grad_norm": 1.1678719520568848, + "learning_rate": 2.6242473240017757e-06, + "loss": 0.7052, + "step": 1905 + }, + { + "epoch": 2.9864300626304803, + "grad_norm": 1.1222507953643799, + "learning_rate": 2.622143128837201e-06, + "loss": 0.6751, + "step": 1906 + }, + { + "epoch": 2.9879958246346554, + "grad_norm": 1.1091363430023193, + "learning_rate": 2.620038846933119e-06, + "loss": 0.6427, + "step": 1907 + }, + { + "epoch": 2.989561586638831, + "grad_norm": 1.0949740409851074, + "learning_rate": 2.6179344797838775e-06, + "loss": 0.6353, + "step": 1908 + }, + { + "epoch": 2.991127348643006, + "grad_norm": 1.2391382455825806, + "learning_rate": 2.615830028883885e-06, + "loss": 0.7133, + "step": 1909 + }, + { + "epoch": 2.9926931106471817, + "grad_norm": 1.1227985620498657, + "learning_rate": 2.6137254957276105e-06, + "loss": 0.6942, + "step": 1910 + }, + { + "epoch": 2.994258872651357, + "grad_norm": 1.2233500480651855, + "learning_rate": 2.61162088180958e-06, + "loss": 0.7006, + "step": 1911 + }, + { + "epoch": 2.9958246346555324, + "grad_norm": 1.1352185010910034, + "learning_rate": 2.609516188624378e-06, + "loss": 0.6304, + "step": 1912 + }, + { + "epoch": 2.997390396659708, + "grad_norm": 1.2024694681167603, + "learning_rate": 2.6074114176666455e-06, + "loss": 0.6911, + "step": 1913 + }, + { + "epoch": 2.998956158663883, + "grad_norm": 1.1703287363052368, + "learning_rate": 2.605306570431077e-06, + "loss": 0.6528, + "step": 1914 + }, + { + "epoch": 3.0015657620041756, + "grad_norm": 1.1240348815917969, + "learning_rate": 2.6032016484124224e-06, + "loss": 1.3204, + "step": 1915 + }, + { + "epoch": 3.0031315240083507, + "grad_norm": 1.0691215991973877, + "learning_rate": 2.6010966531054852e-06, + "loss": 0.643, + "step": 1916 + }, + { + "epoch": 3.0046972860125263, + "grad_norm": 1.1520895957946777, + "learning_rate": 2.5989915860051205e-06, + "loss": 0.6326, + "step": 1917 + }, + { + "epoch": 3.0062630480167014, + "grad_norm": 1.1557644605636597, + "learning_rate": 2.5968864486062334e-06, + "loss": 0.6654, + "step": 1918 + }, + { + "epoch": 3.007828810020877, + "grad_norm": 1.216338872909546, + "learning_rate": 2.59478124240378e-06, + "loss": 0.6109, + "step": 1919 + }, + { + "epoch": 3.009394572025052, + "grad_norm": 1.168593406677246, + "learning_rate": 2.592675968892765e-06, + "loss": 0.6167, + "step": 1920 + }, + { + "epoch": 3.0109603340292277, + "grad_norm": 1.103684902191162, + "learning_rate": 2.590570629568241e-06, + "loss": 0.5883, + "step": 1921 + }, + { + "epoch": 3.012526096033403, + "grad_norm": 1.1893324851989746, + "learning_rate": 2.5884652259253067e-06, + "loss": 0.6295, + "step": 1922 + }, + { + "epoch": 3.0140918580375784, + "grad_norm": 1.046500325202942, + "learning_rate": 2.5863597594591077e-06, + "loss": 0.6338, + "step": 1923 + }, + { + "epoch": 3.0156576200417535, + "grad_norm": 1.1588988304138184, + "learning_rate": 2.5842542316648333e-06, + "loss": 0.6208, + "step": 1924 + }, + { + "epoch": 3.017223382045929, + "grad_norm": 1.0734959840774536, + "learning_rate": 2.5821486440377168e-06, + "loss": 0.5612, + "step": 1925 + }, + { + "epoch": 3.0187891440501042, + "grad_norm": 1.128631353378296, + "learning_rate": 2.5800429980730335e-06, + "loss": 0.6208, + "step": 1926 + }, + { + "epoch": 3.02035490605428, + "grad_norm": 1.1834310293197632, + "learning_rate": 2.5779372952661003e-06, + "loss": 0.6418, + "step": 1927 + }, + { + "epoch": 3.021920668058455, + "grad_norm": 1.217034101486206, + "learning_rate": 2.5758315371122733e-06, + "loss": 0.7117, + "step": 1928 + }, + { + "epoch": 3.0234864300626305, + "grad_norm": 1.060900330543518, + "learning_rate": 2.573725725106952e-06, + "loss": 0.6104, + "step": 1929 + }, + { + "epoch": 3.0250521920668056, + "grad_norm": 1.1937845945358276, + "learning_rate": 2.571619860745569e-06, + "loss": 0.672, + "step": 1930 + }, + { + "epoch": 3.026617954070981, + "grad_norm": 1.0636308193206787, + "learning_rate": 2.569513945523598e-06, + "loss": 0.6525, + "step": 1931 + }, + { + "epoch": 3.028183716075157, + "grad_norm": 1.2644332647323608, + "learning_rate": 2.5674079809365443e-06, + "loss": 0.6109, + "step": 1932 + }, + { + "epoch": 3.029749478079332, + "grad_norm": 1.1538089513778687, + "learning_rate": 2.5653019684799545e-06, + "loss": 0.6813, + "step": 1933 + }, + { + "epoch": 3.0313152400835075, + "grad_norm": 1.1390219926834106, + "learning_rate": 2.5631959096494047e-06, + "loss": 0.6725, + "step": 1934 + }, + { + "epoch": 3.0328810020876826, + "grad_norm": 1.17123281955719, + "learning_rate": 2.5610898059405044e-06, + "loss": 0.5332, + "step": 1935 + }, + { + "epoch": 3.034446764091858, + "grad_norm": 1.1410160064697266, + "learning_rate": 2.5589836588488958e-06, + "loss": 0.6445, + "step": 1936 + }, + { + "epoch": 3.0360125260960333, + "grad_norm": 1.134597897529602, + "learning_rate": 2.5568774698702526e-06, + "loss": 0.6513, + "step": 1937 + }, + { + "epoch": 3.037578288100209, + "grad_norm": 1.1042146682739258, + "learning_rate": 2.5547712405002762e-06, + "loss": 0.6622, + "step": 1938 + }, + { + "epoch": 3.039144050104384, + "grad_norm": 1.129697561264038, + "learning_rate": 2.5526649722346986e-06, + "loss": 0.674, + "step": 1939 + }, + { + "epoch": 3.0407098121085596, + "grad_norm": 1.1788016557693481, + "learning_rate": 2.550558666569279e-06, + "loss": 0.6768, + "step": 1940 + }, + { + "epoch": 3.0422755741127347, + "grad_norm": 1.0913262367248535, + "learning_rate": 2.5484523249998022e-06, + "loss": 0.6817, + "step": 1941 + }, + { + "epoch": 3.0438413361169103, + "grad_norm": 1.1731491088867188, + "learning_rate": 2.5463459490220794e-06, + "loss": 0.652, + "step": 1942 + }, + { + "epoch": 3.0454070981210855, + "grad_norm": 1.139214038848877, + "learning_rate": 2.544239540131947e-06, + "loss": 0.6462, + "step": 1943 + }, + { + "epoch": 3.046972860125261, + "grad_norm": 1.163565754890442, + "learning_rate": 2.5421330998252617e-06, + "loss": 0.611, + "step": 1944 + }, + { + "epoch": 3.048538622129436, + "grad_norm": 1.136677861213684, + "learning_rate": 2.5400266295979055e-06, + "loss": 0.6502, + "step": 1945 + }, + { + "epoch": 3.0501043841336117, + "grad_norm": 1.1309559345245361, + "learning_rate": 2.537920130945782e-06, + "loss": 0.5777, + "step": 1946 + }, + { + "epoch": 3.051670146137787, + "grad_norm": 1.0693000555038452, + "learning_rate": 2.535813605364812e-06, + "loss": 0.6596, + "step": 1947 + }, + { + "epoch": 3.0532359081419624, + "grad_norm": 1.2112295627593994, + "learning_rate": 2.533707054350938e-06, + "loss": 0.6033, + "step": 1948 + }, + { + "epoch": 3.0548016701461376, + "grad_norm": 1.186867594718933, + "learning_rate": 2.531600479400119e-06, + "loss": 0.6399, + "step": 1949 + }, + { + "epoch": 3.056367432150313, + "grad_norm": 1.146904468536377, + "learning_rate": 2.5294938820083324e-06, + "loss": 0.6431, + "step": 1950 + }, + { + "epoch": 3.0579331941544887, + "grad_norm": 1.0728389024734497, + "learning_rate": 2.5273872636715707e-06, + "loss": 0.6754, + "step": 1951 + }, + { + "epoch": 3.059498956158664, + "grad_norm": 1.219117522239685, + "learning_rate": 2.5252806258858416e-06, + "loss": 0.5961, + "step": 1952 + }, + { + "epoch": 3.0610647181628394, + "grad_norm": 1.1819669008255005, + "learning_rate": 2.5231739701471657e-06, + "loss": 0.6809, + "step": 1953 + }, + { + "epoch": 3.0626304801670146, + "grad_norm": 1.181559443473816, + "learning_rate": 2.5210672979515773e-06, + "loss": 0.5754, + "step": 1954 + }, + { + "epoch": 3.06419624217119, + "grad_norm": 1.1187459230422974, + "learning_rate": 2.5189606107951225e-06, + "loss": 0.6324, + "step": 1955 + }, + { + "epoch": 3.0657620041753653, + "grad_norm": 1.20028555393219, + "learning_rate": 2.5168539101738576e-06, + "loss": 0.6178, + "step": 1956 + }, + { + "epoch": 3.067327766179541, + "grad_norm": 1.1661895513534546, + "learning_rate": 2.514747197583849e-06, + "loss": 0.5646, + "step": 1957 + }, + { + "epoch": 3.068893528183716, + "grad_norm": 1.1893212795257568, + "learning_rate": 2.512640474521169e-06, + "loss": 0.6387, + "step": 1958 + }, + { + "epoch": 3.0704592901878915, + "grad_norm": 1.1453262567520142, + "learning_rate": 2.510533742481902e-06, + "loss": 0.6416, + "step": 1959 + }, + { + "epoch": 3.0720250521920667, + "grad_norm": 1.2452423572540283, + "learning_rate": 2.5084270029621353e-06, + "loss": 0.6435, + "step": 1960 + }, + { + "epoch": 3.0735908141962422, + "grad_norm": 1.1707086563110352, + "learning_rate": 2.5063202574579625e-06, + "loss": 0.6328, + "step": 1961 + }, + { + "epoch": 3.0751565762004174, + "grad_norm": 1.2181516885757446, + "learning_rate": 2.5042135074654815e-06, + "loss": 0.6084, + "step": 1962 + }, + { + "epoch": 3.076722338204593, + "grad_norm": 1.1872752904891968, + "learning_rate": 2.502106754480793e-06, + "loss": 0.5128, + "step": 1963 + }, + { + "epoch": 3.078288100208768, + "grad_norm": 1.1264151334762573, + "learning_rate": 2.5e-06, + "loss": 0.6391, + "step": 1964 + }, + { + "epoch": 3.0798538622129437, + "grad_norm": 1.1406108140945435, + "learning_rate": 2.497893245519208e-06, + "loss": 0.6686, + "step": 1965 + }, + { + "epoch": 3.081419624217119, + "grad_norm": 1.1387429237365723, + "learning_rate": 2.495786492534519e-06, + "loss": 0.6449, + "step": 1966 + }, + { + "epoch": 3.0829853862212944, + "grad_norm": 1.1874876022338867, + "learning_rate": 2.493679742542038e-06, + "loss": 0.6471, + "step": 1967 + }, + { + "epoch": 3.08455114822547, + "grad_norm": 1.2808914184570312, + "learning_rate": 2.4915729970378655e-06, + "loss": 0.5295, + "step": 1968 + }, + { + "epoch": 3.086116910229645, + "grad_norm": 1.203076720237732, + "learning_rate": 2.4894662575180982e-06, + "loss": 0.6161, + "step": 1969 + }, + { + "epoch": 3.0876826722338206, + "grad_norm": 1.0870691537857056, + "learning_rate": 2.4873595254788317e-06, + "loss": 0.6344, + "step": 1970 + }, + { + "epoch": 3.0892484342379958, + "grad_norm": 1.174511194229126, + "learning_rate": 2.4852528024161526e-06, + "loss": 0.5958, + "step": 1971 + }, + { + "epoch": 3.0908141962421714, + "grad_norm": 1.2232667207717896, + "learning_rate": 2.4831460898261428e-06, + "loss": 0.5745, + "step": 1972 + }, + { + "epoch": 3.0923799582463465, + "grad_norm": 1.1356587409973145, + "learning_rate": 2.4810393892048783e-06, + "loss": 0.645, + "step": 1973 + }, + { + "epoch": 3.093945720250522, + "grad_norm": 1.2031471729278564, + "learning_rate": 2.478932702048424e-06, + "loss": 0.6246, + "step": 1974 + }, + { + "epoch": 3.095511482254697, + "grad_norm": 1.1597861051559448, + "learning_rate": 2.476826029852835e-06, + "loss": 0.6527, + "step": 1975 + }, + { + "epoch": 3.0970772442588728, + "grad_norm": 1.106753945350647, + "learning_rate": 2.4747193741141593e-06, + "loss": 0.7164, + "step": 1976 + }, + { + "epoch": 3.098643006263048, + "grad_norm": 1.147586703300476, + "learning_rate": 2.47261273632843e-06, + "loss": 0.6295, + "step": 1977 + }, + { + "epoch": 3.1002087682672235, + "grad_norm": 1.2318609952926636, + "learning_rate": 2.470506117991668e-06, + "loss": 0.5468, + "step": 1978 + }, + { + "epoch": 3.1017745302713986, + "grad_norm": 1.1728826761245728, + "learning_rate": 2.4683995205998818e-06, + "loss": 0.6144, + "step": 1979 + }, + { + "epoch": 3.103340292275574, + "grad_norm": 1.1983925104141235, + "learning_rate": 2.4662929456490633e-06, + "loss": 0.6214, + "step": 1980 + }, + { + "epoch": 3.1049060542797493, + "grad_norm": 1.134922981262207, + "learning_rate": 2.4641863946351884e-06, + "loss": 0.631, + "step": 1981 + }, + { + "epoch": 3.106471816283925, + "grad_norm": 1.2184958457946777, + "learning_rate": 2.4620798690542184e-06, + "loss": 0.6642, + "step": 1982 + }, + { + "epoch": 3.1080375782881, + "grad_norm": 1.1365389823913574, + "learning_rate": 2.459973370402095e-06, + "loss": 0.631, + "step": 1983 + }, + { + "epoch": 3.1096033402922756, + "grad_norm": 1.107224464416504, + "learning_rate": 2.4578669001747387e-06, + "loss": 0.6455, + "step": 1984 + }, + { + "epoch": 3.1111691022964507, + "grad_norm": 1.09954833984375, + "learning_rate": 2.455760459868054e-06, + "loss": 0.6674, + "step": 1985 + }, + { + "epoch": 3.1127348643006263, + "grad_norm": 1.1493922472000122, + "learning_rate": 2.453654050977921e-06, + "loss": 0.6144, + "step": 1986 + }, + { + "epoch": 3.114300626304802, + "grad_norm": 1.1649893522262573, + "learning_rate": 2.4515476750001978e-06, + "loss": 0.6621, + "step": 1987 + }, + { + "epoch": 3.115866388308977, + "grad_norm": 1.16617751121521, + "learning_rate": 2.449441333430722e-06, + "loss": 0.7146, + "step": 1988 + }, + { + "epoch": 3.1174321503131526, + "grad_norm": 1.137216567993164, + "learning_rate": 2.447335027765302e-06, + "loss": 0.6294, + "step": 1989 + }, + { + "epoch": 3.1189979123173277, + "grad_norm": 1.1343148946762085, + "learning_rate": 2.445228759499725e-06, + "loss": 0.6289, + "step": 1990 + }, + { + "epoch": 3.1205636743215033, + "grad_norm": 1.113688588142395, + "learning_rate": 2.4431225301297482e-06, + "loss": 0.6331, + "step": 1991 + }, + { + "epoch": 3.1221294363256784, + "grad_norm": 1.1988941431045532, + "learning_rate": 2.4410163411511046e-06, + "loss": 0.5989, + "step": 1992 + }, + { + "epoch": 3.123695198329854, + "grad_norm": 1.1929535865783691, + "learning_rate": 2.438910194059497e-06, + "loss": 0.6789, + "step": 1993 + }, + { + "epoch": 3.125260960334029, + "grad_norm": 1.2655651569366455, + "learning_rate": 2.436804090350596e-06, + "loss": 0.5709, + "step": 1994 + }, + { + "epoch": 3.1268267223382047, + "grad_norm": 1.187930941581726, + "learning_rate": 2.4346980315200464e-06, + "loss": 0.6324, + "step": 1995 + }, + { + "epoch": 3.12839248434238, + "grad_norm": 1.1696043014526367, + "learning_rate": 2.432592019063456e-06, + "loss": 0.5987, + "step": 1996 + }, + { + "epoch": 3.1299582463465554, + "grad_norm": 1.228115439414978, + "learning_rate": 2.430486054476403e-06, + "loss": 0.6581, + "step": 1997 + }, + { + "epoch": 3.1315240083507305, + "grad_norm": 1.1672264337539673, + "learning_rate": 2.4283801392544317e-06, + "loss": 0.6412, + "step": 1998 + }, + { + "epoch": 3.133089770354906, + "grad_norm": 1.2138696908950806, + "learning_rate": 2.4262742748930493e-06, + "loss": 0.6761, + "step": 1999 + }, + { + "epoch": 3.1346555323590812, + "grad_norm": 1.215649962425232, + "learning_rate": 2.4241684628877267e-06, + "loss": 0.6074, + "step": 2000 + }, + { + "epoch": 3.136221294363257, + "grad_norm": 1.2324877977371216, + "learning_rate": 2.4220627047339006e-06, + "loss": 0.6402, + "step": 2001 + }, + { + "epoch": 3.137787056367432, + "grad_norm": 1.230507254600525, + "learning_rate": 2.4199570019269678e-06, + "loss": 0.6784, + "step": 2002 + }, + { + "epoch": 3.1393528183716075, + "grad_norm": 1.056394100189209, + "learning_rate": 2.4178513559622836e-06, + "loss": 0.6209, + "step": 2003 + }, + { + "epoch": 3.140918580375783, + "grad_norm": 1.178450107574463, + "learning_rate": 2.415745768335167e-06, + "loss": 0.6463, + "step": 2004 + }, + { + "epoch": 3.142484342379958, + "grad_norm": 1.2509713172912598, + "learning_rate": 2.4136402405408927e-06, + "loss": 0.5886, + "step": 2005 + }, + { + "epoch": 3.144050104384134, + "grad_norm": 1.1313499212265015, + "learning_rate": 2.411534774074693e-06, + "loss": 0.6083, + "step": 2006 + }, + { + "epoch": 3.145615866388309, + "grad_norm": 1.208187460899353, + "learning_rate": 2.4094293704317596e-06, + "loss": 0.6488, + "step": 2007 + }, + { + "epoch": 3.1471816283924845, + "grad_norm": 1.1712708473205566, + "learning_rate": 2.4073240311072363e-06, + "loss": 0.6386, + "step": 2008 + }, + { + "epoch": 3.1487473903966596, + "grad_norm": 1.2195419073104858, + "learning_rate": 2.4052187575962206e-06, + "loss": 0.6016, + "step": 2009 + }, + { + "epoch": 3.150313152400835, + "grad_norm": 1.2239069938659668, + "learning_rate": 2.4031135513937675e-06, + "loss": 0.6095, + "step": 2010 + }, + { + "epoch": 3.1518789144050103, + "grad_norm": 1.274491786956787, + "learning_rate": 2.4010084139948807e-06, + "loss": 0.5551, + "step": 2011 + }, + { + "epoch": 3.153444676409186, + "grad_norm": 1.104736089706421, + "learning_rate": 2.398903346894515e-06, + "loss": 0.6555, + "step": 2012 + }, + { + "epoch": 3.155010438413361, + "grad_norm": 1.159359097480774, + "learning_rate": 2.396798351587578e-06, + "loss": 0.7065, + "step": 2013 + }, + { + "epoch": 3.1565762004175366, + "grad_norm": 1.2481755018234253, + "learning_rate": 2.394693429568924e-06, + "loss": 0.6085, + "step": 2014 + }, + { + "epoch": 3.1581419624217117, + "grad_norm": 1.2071157693862915, + "learning_rate": 2.392588582333355e-06, + "loss": 0.6871, + "step": 2015 + }, + { + "epoch": 3.1597077244258873, + "grad_norm": 1.1674847602844238, + "learning_rate": 2.3904838113756228e-06, + "loss": 0.5756, + "step": 2016 + }, + { + "epoch": 3.1612734864300625, + "grad_norm": 1.1972459554672241, + "learning_rate": 2.3883791181904216e-06, + "loss": 0.6307, + "step": 2017 + }, + { + "epoch": 3.162839248434238, + "grad_norm": 1.2457903623580933, + "learning_rate": 2.3862745042723907e-06, + "loss": 0.5552, + "step": 2018 + }, + { + "epoch": 3.164405010438413, + "grad_norm": 1.2475699186325073, + "learning_rate": 2.3841699711161155e-06, + "loss": 0.6054, + "step": 2019 + }, + { + "epoch": 3.1659707724425887, + "grad_norm": 1.2934919595718384, + "learning_rate": 2.3820655202161237e-06, + "loss": 0.6986, + "step": 2020 + }, + { + "epoch": 3.1675365344467643, + "grad_norm": 1.1634780168533325, + "learning_rate": 2.3799611530668814e-06, + "loss": 0.6697, + "step": 2021 + }, + { + "epoch": 3.1691022964509394, + "grad_norm": 1.1987229585647583, + "learning_rate": 2.3778568711628e-06, + "loss": 0.6376, + "step": 2022 + }, + { + "epoch": 3.170668058455115, + "grad_norm": 1.0842245817184448, + "learning_rate": 2.375752675998225e-06, + "loss": 0.629, + "step": 2023 + }, + { + "epoch": 3.17223382045929, + "grad_norm": 1.1573731899261475, + "learning_rate": 2.373648569067444e-06, + "loss": 0.6231, + "step": 2024 + }, + { + "epoch": 3.1737995824634657, + "grad_norm": 1.1786472797393799, + "learning_rate": 2.371544551864683e-06, + "loss": 0.7082, + "step": 2025 + }, + { + "epoch": 3.175365344467641, + "grad_norm": 1.2212613821029663, + "learning_rate": 2.3694406258841004e-06, + "loss": 0.6382, + "step": 2026 + }, + { + "epoch": 3.1769311064718164, + "grad_norm": 1.0942133665084839, + "learning_rate": 2.367336792619791e-06, + "loss": 0.6157, + "step": 2027 + }, + { + "epoch": 3.1784968684759916, + "grad_norm": 1.198861837387085, + "learning_rate": 2.365233053565787e-06, + "loss": 0.6172, + "step": 2028 + }, + { + "epoch": 3.180062630480167, + "grad_norm": 1.2345949411392212, + "learning_rate": 2.3631294102160493e-06, + "loss": 0.6561, + "step": 2029 + }, + { + "epoch": 3.1816283924843423, + "grad_norm": 1.1807947158813477, + "learning_rate": 2.3610258640644725e-06, + "loss": 0.6323, + "step": 2030 + }, + { + "epoch": 3.183194154488518, + "grad_norm": 1.272806167602539, + "learning_rate": 2.3589224166048844e-06, + "loss": 0.6286, + "step": 2031 + }, + { + "epoch": 3.184759916492693, + "grad_norm": 1.1519666910171509, + "learning_rate": 2.3568190693310388e-06, + "loss": 0.6351, + "step": 2032 + }, + { + "epoch": 3.1863256784968685, + "grad_norm": 1.3199100494384766, + "learning_rate": 2.35471582373662e-06, + "loss": 0.6282, + "step": 2033 + }, + { + "epoch": 3.1878914405010437, + "grad_norm": 1.1525390148162842, + "learning_rate": 2.352612681315243e-06, + "loss": 0.5453, + "step": 2034 + }, + { + "epoch": 3.1894572025052192, + "grad_norm": 1.1851332187652588, + "learning_rate": 2.3505096435604446e-06, + "loss": 0.5935, + "step": 2035 + }, + { + "epoch": 3.1910229645093944, + "grad_norm": 1.1144623756408691, + "learning_rate": 2.3484067119656905e-06, + "loss": 0.6756, + "step": 2036 + }, + { + "epoch": 3.19258872651357, + "grad_norm": 1.2141358852386475, + "learning_rate": 2.3463038880243687e-06, + "loss": 0.649, + "step": 2037 + }, + { + "epoch": 3.1941544885177455, + "grad_norm": 1.206795573234558, + "learning_rate": 2.3442011732297944e-06, + "loss": 0.7051, + "step": 2038 + }, + { + "epoch": 3.1957202505219207, + "grad_norm": 1.1212406158447266, + "learning_rate": 2.3420985690752015e-06, + "loss": 0.6279, + "step": 2039 + }, + { + "epoch": 3.1972860125260962, + "grad_norm": 1.1145480871200562, + "learning_rate": 2.339996077053746e-06, + "loss": 0.6885, + "step": 2040 + }, + { + "epoch": 3.1988517745302714, + "grad_norm": 1.235465168952942, + "learning_rate": 2.337893698658508e-06, + "loss": 0.6146, + "step": 2041 + }, + { + "epoch": 3.200417536534447, + "grad_norm": 1.2589614391326904, + "learning_rate": 2.3357914353824816e-06, + "loss": 0.6892, + "step": 2042 + }, + { + "epoch": 3.201983298538622, + "grad_norm": 1.3488636016845703, + "learning_rate": 2.3336892887185816e-06, + "loss": 0.5532, + "step": 2043 + }, + { + "epoch": 3.2035490605427976, + "grad_norm": 1.218481183052063, + "learning_rate": 2.331587260159641e-06, + "loss": 0.6554, + "step": 2044 + }, + { + "epoch": 3.2051148225469728, + "grad_norm": 1.207727074623108, + "learning_rate": 2.329485351198408e-06, + "loss": 0.5789, + "step": 2045 + }, + { + "epoch": 3.2066805845511483, + "grad_norm": 1.143080711364746, + "learning_rate": 2.3273835633275444e-06, + "loss": 0.6171, + "step": 2046 + }, + { + "epoch": 3.2082463465553235, + "grad_norm": 1.1894160509109497, + "learning_rate": 2.325281898039629e-06, + "loss": 0.6514, + "step": 2047 + }, + { + "epoch": 3.209812108559499, + "grad_norm": 1.1722674369812012, + "learning_rate": 2.323180356827151e-06, + "loss": 0.6305, + "step": 2048 + }, + { + "epoch": 3.211377870563674, + "grad_norm": 1.131166934967041, + "learning_rate": 2.3210789411825117e-06, + "loss": 0.651, + "step": 2049 + }, + { + "epoch": 3.2129436325678498, + "grad_norm": 1.0506892204284668, + "learning_rate": 2.318977652598026e-06, + "loss": 0.6041, + "step": 2050 + }, + { + "epoch": 3.214509394572025, + "grad_norm": 1.2478179931640625, + "learning_rate": 2.3168764925659155e-06, + "loss": 0.6418, + "step": 2051 + }, + { + "epoch": 3.2160751565762005, + "grad_norm": 1.1366816759109497, + "learning_rate": 2.31477546257831e-06, + "loss": 0.6269, + "step": 2052 + }, + { + "epoch": 3.2176409185803756, + "grad_norm": 1.1247450113296509, + "learning_rate": 2.3126745641272506e-06, + "loss": 0.6706, + "step": 2053 + }, + { + "epoch": 3.219206680584551, + "grad_norm": 1.166385531425476, + "learning_rate": 2.3105737987046823e-06, + "loss": 0.6267, + "step": 2054 + }, + { + "epoch": 3.2207724425887267, + "grad_norm": 1.134945273399353, + "learning_rate": 2.3084731678024546e-06, + "loss": 0.6367, + "step": 2055 + }, + { + "epoch": 3.222338204592902, + "grad_norm": 1.129788875579834, + "learning_rate": 2.306372672912326e-06, + "loss": 0.6327, + "step": 2056 + }, + { + "epoch": 3.223903966597077, + "grad_norm": 1.0779662132263184, + "learning_rate": 2.3042723155259526e-06, + "loss": 0.6261, + "step": 2057 + }, + { + "epoch": 3.2254697286012526, + "grad_norm": 1.2631971836090088, + "learning_rate": 2.302172097134897e-06, + "loss": 0.5617, + "step": 2058 + }, + { + "epoch": 3.227035490605428, + "grad_norm": 1.180594801902771, + "learning_rate": 2.3000720192306226e-06, + "loss": 0.585, + "step": 2059 + }, + { + "epoch": 3.2286012526096033, + "grad_norm": 1.1890621185302734, + "learning_rate": 2.297972083304491e-06, + "loss": 0.6525, + "step": 2060 + }, + { + "epoch": 3.230167014613779, + "grad_norm": 1.199937343597412, + "learning_rate": 2.295872290847764e-06, + "loss": 0.6024, + "step": 2061 + }, + { + "epoch": 3.231732776617954, + "grad_norm": 1.2697222232818604, + "learning_rate": 2.293772643351603e-06, + "loss": 0.6007, + "step": 2062 + }, + { + "epoch": 3.2332985386221296, + "grad_norm": 1.173552393913269, + "learning_rate": 2.2916731423070645e-06, + "loss": 0.6576, + "step": 2063 + }, + { + "epoch": 3.2348643006263047, + "grad_norm": 1.2051188945770264, + "learning_rate": 2.2895737892051007e-06, + "loss": 0.6862, + "step": 2064 + }, + { + "epoch": 3.2364300626304803, + "grad_norm": 1.1974767446517944, + "learning_rate": 2.287474585536562e-06, + "loss": 0.6201, + "step": 2065 + }, + { + "epoch": 3.2379958246346554, + "grad_norm": 1.2346913814544678, + "learning_rate": 2.2853755327921883e-06, + "loss": 0.5762, + "step": 2066 + }, + { + "epoch": 3.239561586638831, + "grad_norm": 1.1864664554595947, + "learning_rate": 2.2832766324626153e-06, + "loss": 0.6389, + "step": 2067 + }, + { + "epoch": 3.241127348643006, + "grad_norm": 1.300667405128479, + "learning_rate": 2.28117788603837e-06, + "loss": 0.6645, + "step": 2068 + }, + { + "epoch": 3.2426931106471817, + "grad_norm": 1.2889467477798462, + "learning_rate": 2.2790792950098695e-06, + "loss": 0.6399, + "step": 2069 + }, + { + "epoch": 3.244258872651357, + "grad_norm": 1.1807681322097778, + "learning_rate": 2.2769808608674193e-06, + "loss": 0.5974, + "step": 2070 + }, + { + "epoch": 3.2458246346555324, + "grad_norm": 1.2693206071853638, + "learning_rate": 2.274882585101217e-06, + "loss": 0.7067, + "step": 2071 + }, + { + "epoch": 3.2473903966597075, + "grad_norm": 1.1668689250946045, + "learning_rate": 2.2727844692013448e-06, + "loss": 0.5987, + "step": 2072 + }, + { + "epoch": 3.248956158663883, + "grad_norm": 1.2399280071258545, + "learning_rate": 2.2706865146577716e-06, + "loss": 0.6821, + "step": 2073 + }, + { + "epoch": 3.2505219206680582, + "grad_norm": 1.2630313634872437, + "learning_rate": 2.2685887229603534e-06, + "loss": 0.5846, + "step": 2074 + }, + { + "epoch": 3.252087682672234, + "grad_norm": 1.1724474430084229, + "learning_rate": 2.266491095598829e-06, + "loss": 0.6272, + "step": 2075 + }, + { + "epoch": 3.2536534446764094, + "grad_norm": 1.3041969537734985, + "learning_rate": 2.2643936340628205e-06, + "loss": 0.6246, + "step": 2076 + }, + { + "epoch": 3.2552192066805845, + "grad_norm": 1.152488350868225, + "learning_rate": 2.262296339841834e-06, + "loss": 0.6827, + "step": 2077 + }, + { + "epoch": 3.25678496868476, + "grad_norm": 1.1879392862319946, + "learning_rate": 2.260199214425255e-06, + "loss": 0.681, + "step": 2078 + }, + { + "epoch": 3.258350730688935, + "grad_norm": 1.1999655961990356, + "learning_rate": 2.2581022593023483e-06, + "loss": 0.7044, + "step": 2079 + }, + { + "epoch": 3.259916492693111, + "grad_norm": 1.2020212411880493, + "learning_rate": 2.256005475962261e-06, + "loss": 0.6206, + "step": 2080 + }, + { + "epoch": 3.261482254697286, + "grad_norm": 1.312864065170288, + "learning_rate": 2.2539088658940158e-06, + "loss": 0.6385, + "step": 2081 + }, + { + "epoch": 3.2630480167014615, + "grad_norm": 1.1875925064086914, + "learning_rate": 2.251812430586512e-06, + "loss": 0.5999, + "step": 2082 + }, + { + "epoch": 3.2646137787056366, + "grad_norm": 1.0950852632522583, + "learning_rate": 2.249716171528526e-06, + "loss": 0.7346, + "step": 2083 + }, + { + "epoch": 3.266179540709812, + "grad_norm": 1.2490246295928955, + "learning_rate": 2.24762009020871e-06, + "loss": 0.5889, + "step": 2084 + }, + { + "epoch": 3.2677453027139873, + "grad_norm": 1.1770070791244507, + "learning_rate": 2.2455241881155874e-06, + "loss": 0.681, + "step": 2085 + }, + { + "epoch": 3.269311064718163, + "grad_norm": 1.1888784170150757, + "learning_rate": 2.2434284667375553e-06, + "loss": 0.566, + "step": 2086 + }, + { + "epoch": 3.270876826722338, + "grad_norm": 1.290913701057434, + "learning_rate": 2.2413329275628836e-06, + "loss": 0.6737, + "step": 2087 + }, + { + "epoch": 3.2724425887265136, + "grad_norm": 1.1826350688934326, + "learning_rate": 2.2392375720797126e-06, + "loss": 0.6149, + "step": 2088 + }, + { + "epoch": 3.2740083507306887, + "grad_norm": 1.1560386419296265, + "learning_rate": 2.237142401776049e-06, + "loss": 0.6214, + "step": 2089 + }, + { + "epoch": 3.2755741127348643, + "grad_norm": 1.1863820552825928, + "learning_rate": 2.2350474181397736e-06, + "loss": 0.592, + "step": 2090 + }, + { + "epoch": 3.2771398747390394, + "grad_norm": 1.2404558658599854, + "learning_rate": 2.2329526226586294e-06, + "loss": 0.6648, + "step": 2091 + }, + { + "epoch": 3.278705636743215, + "grad_norm": 1.1192048788070679, + "learning_rate": 2.2308580168202284e-06, + "loss": 0.6657, + "step": 2092 + }, + { + "epoch": 3.2802713987473906, + "grad_norm": 1.2412128448486328, + "learning_rate": 2.228763602112049e-06, + "loss": 0.5999, + "step": 2093 + }, + { + "epoch": 3.2818371607515657, + "grad_norm": 1.2265843152999878, + "learning_rate": 2.2266693800214305e-06, + "loss": 0.6335, + "step": 2094 + }, + { + "epoch": 3.2834029227557413, + "grad_norm": 1.1712675094604492, + "learning_rate": 2.2245753520355773e-06, + "loss": 0.6533, + "step": 2095 + }, + { + "epoch": 3.2849686847599164, + "grad_norm": 1.2727960348129272, + "learning_rate": 2.222481519641557e-06, + "loss": 0.6233, + "step": 2096 + }, + { + "epoch": 3.286534446764092, + "grad_norm": 1.0983766317367554, + "learning_rate": 2.220387884326297e-06, + "loss": 0.6268, + "step": 2097 + }, + { + "epoch": 3.288100208768267, + "grad_norm": 1.2702240943908691, + "learning_rate": 2.2182944475765837e-06, + "loss": 0.6453, + "step": 2098 + }, + { + "epoch": 3.2896659707724427, + "grad_norm": 1.2952048778533936, + "learning_rate": 2.2162012108790656e-06, + "loss": 0.6177, + "step": 2099 + }, + { + "epoch": 3.291231732776618, + "grad_norm": 1.1496988534927368, + "learning_rate": 2.214108175720246e-06, + "loss": 0.6123, + "step": 2100 + }, + { + "epoch": 3.2927974947807934, + "grad_norm": 1.1527506113052368, + "learning_rate": 2.212015343586486e-06, + "loss": 0.5732, + "step": 2101 + }, + { + "epoch": 3.2943632567849686, + "grad_norm": 1.1944189071655273, + "learning_rate": 2.209922715964005e-06, + "loss": 0.6502, + "step": 2102 + }, + { + "epoch": 3.295929018789144, + "grad_norm": 1.2035129070281982, + "learning_rate": 2.2078302943388736e-06, + "loss": 0.6233, + "step": 2103 + }, + { + "epoch": 3.2974947807933193, + "grad_norm": 1.111652135848999, + "learning_rate": 2.2057380801970164e-06, + "loss": 0.6082, + "step": 2104 + }, + { + "epoch": 3.299060542797495, + "grad_norm": 1.223018765449524, + "learning_rate": 2.2036460750242136e-06, + "loss": 0.6537, + "step": 2105 + }, + { + "epoch": 3.30062630480167, + "grad_norm": 1.2612967491149902, + "learning_rate": 2.2015542803060952e-06, + "loss": 0.6857, + "step": 2106 + }, + { + "epoch": 3.3021920668058455, + "grad_norm": 1.0949403047561646, + "learning_rate": 2.1994626975281403e-06, + "loss": 0.591, + "step": 2107 + }, + { + "epoch": 3.3037578288100207, + "grad_norm": 1.1419187784194946, + "learning_rate": 2.197371328175681e-06, + "loss": 0.6729, + "step": 2108 + }, + { + "epoch": 3.3053235908141962, + "grad_norm": 1.1853833198547363, + "learning_rate": 2.195280173733894e-06, + "loss": 0.6015, + "step": 2109 + }, + { + "epoch": 3.306889352818372, + "grad_norm": 1.2181181907653809, + "learning_rate": 2.193189235687805e-06, + "loss": 0.6378, + "step": 2110 + }, + { + "epoch": 3.308455114822547, + "grad_norm": 1.291874885559082, + "learning_rate": 2.191098515522289e-06, + "loss": 0.6883, + "step": 2111 + }, + { + "epoch": 3.3100208768267225, + "grad_norm": 1.1752232313156128, + "learning_rate": 2.1890080147220606e-06, + "loss": 0.5764, + "step": 2112 + }, + { + "epoch": 3.3115866388308977, + "grad_norm": 1.2201036214828491, + "learning_rate": 2.186917734771682e-06, + "loss": 0.6593, + "step": 2113 + }, + { + "epoch": 3.3131524008350732, + "grad_norm": 1.2628161907196045, + "learning_rate": 2.1848276771555586e-06, + "loss": 0.5571, + "step": 2114 + }, + { + "epoch": 3.3147181628392484, + "grad_norm": 1.2265757322311401, + "learning_rate": 2.1827378433579384e-06, + "loss": 0.5375, + "step": 2115 + }, + { + "epoch": 3.316283924843424, + "grad_norm": 1.2193812131881714, + "learning_rate": 2.1806482348629065e-06, + "loss": 0.6333, + "step": 2116 + }, + { + "epoch": 3.317849686847599, + "grad_norm": 1.1655728816986084, + "learning_rate": 2.1785588531543945e-06, + "loss": 0.6113, + "step": 2117 + }, + { + "epoch": 3.3194154488517746, + "grad_norm": 1.2318220138549805, + "learning_rate": 2.1764696997161665e-06, + "loss": 0.659, + "step": 2118 + }, + { + "epoch": 3.3209812108559498, + "grad_norm": 1.1320031881332397, + "learning_rate": 2.174380776031828e-06, + "loss": 0.656, + "step": 2119 + }, + { + "epoch": 3.3225469728601253, + "grad_norm": 1.1628968715667725, + "learning_rate": 2.1722920835848226e-06, + "loss": 0.6614, + "step": 2120 + }, + { + "epoch": 3.3241127348643005, + "grad_norm": 1.2760236263275146, + "learning_rate": 2.170203623858426e-06, + "loss": 0.5455, + "step": 2121 + }, + { + "epoch": 3.325678496868476, + "grad_norm": 1.1655547618865967, + "learning_rate": 2.1681153983357497e-06, + "loss": 0.6167, + "step": 2122 + }, + { + "epoch": 3.327244258872651, + "grad_norm": 1.1703981161117554, + "learning_rate": 2.1660274084997414e-06, + "loss": 0.6874, + "step": 2123 + }, + { + "epoch": 3.3288100208768268, + "grad_norm": 1.2664302587509155, + "learning_rate": 2.1639396558331794e-06, + "loss": 0.6355, + "step": 2124 + }, + { + "epoch": 3.330375782881002, + "grad_norm": 1.1683710813522339, + "learning_rate": 2.1618521418186725e-06, + "loss": 0.6236, + "step": 2125 + }, + { + "epoch": 3.3319415448851775, + "grad_norm": 1.219686508178711, + "learning_rate": 2.159764867938663e-06, + "loss": 0.622, + "step": 2126 + }, + { + "epoch": 3.333507306889353, + "grad_norm": 1.1496042013168335, + "learning_rate": 2.15767783567542e-06, + "loss": 0.5684, + "step": 2127 + }, + { + "epoch": 3.335073068893528, + "grad_norm": 1.1844868659973145, + "learning_rate": 2.155591046511043e-06, + "loss": 0.5401, + "step": 2128 + }, + { + "epoch": 3.3366388308977033, + "grad_norm": 1.0850353240966797, + "learning_rate": 2.1535045019274557e-06, + "loss": 0.7124, + "step": 2129 + }, + { + "epoch": 3.338204592901879, + "grad_norm": 1.1659971475601196, + "learning_rate": 2.1514182034064123e-06, + "loss": 0.6502, + "step": 2130 + }, + { + "epoch": 3.3397703549060545, + "grad_norm": 1.1342421770095825, + "learning_rate": 2.1493321524294902e-06, + "loss": 0.6352, + "step": 2131 + }, + { + "epoch": 3.3413361169102296, + "grad_norm": 1.1351344585418701, + "learning_rate": 2.1472463504780893e-06, + "loss": 0.7151, + "step": 2132 + }, + { + "epoch": 3.342901878914405, + "grad_norm": 1.2182797193527222, + "learning_rate": 2.145160799033437e-06, + "loss": 0.6671, + "step": 2133 + }, + { + "epoch": 3.3444676409185803, + "grad_norm": 1.0785852670669556, + "learning_rate": 2.1430754995765783e-06, + "loss": 0.6595, + "step": 2134 + }, + { + "epoch": 3.346033402922756, + "grad_norm": 1.2832294702529907, + "learning_rate": 2.1409904535883816e-06, + "loss": 0.6094, + "step": 2135 + }, + { + "epoch": 3.347599164926931, + "grad_norm": 1.1902724504470825, + "learning_rate": 2.138905662549536e-06, + "loss": 0.6205, + "step": 2136 + }, + { + "epoch": 3.3491649269311066, + "grad_norm": 1.1487525701522827, + "learning_rate": 2.136821127940547e-06, + "loss": 0.6812, + "step": 2137 + }, + { + "epoch": 3.3507306889352817, + "grad_norm": 1.214062213897705, + "learning_rate": 2.1347368512417394e-06, + "loss": 0.6414, + "step": 2138 + }, + { + "epoch": 3.3522964509394573, + "grad_norm": 1.1467952728271484, + "learning_rate": 2.132652833933256e-06, + "loss": 0.6825, + "step": 2139 + }, + { + "epoch": 3.3538622129436324, + "grad_norm": 1.1381312608718872, + "learning_rate": 2.1305690774950543e-06, + "loss": 0.5342, + "step": 2140 + }, + { + "epoch": 3.355427974947808, + "grad_norm": 1.2287116050720215, + "learning_rate": 2.1284855834069044e-06, + "loss": 0.6098, + "step": 2141 + }, + { + "epoch": 3.356993736951983, + "grad_norm": 1.1297441720962524, + "learning_rate": 2.1264023531483947e-06, + "loss": 0.6527, + "step": 2142 + }, + { + "epoch": 3.3585594989561587, + "grad_norm": 1.1599788665771484, + "learning_rate": 2.124319388198923e-06, + "loss": 0.6767, + "step": 2143 + }, + { + "epoch": 3.3601252609603343, + "grad_norm": 1.2442164421081543, + "learning_rate": 2.1222366900376977e-06, + "loss": 0.618, + "step": 2144 + }, + { + "epoch": 3.3616910229645094, + "grad_norm": 1.2075343132019043, + "learning_rate": 2.120154260143742e-06, + "loss": 0.6413, + "step": 2145 + }, + { + "epoch": 3.3632567849686845, + "grad_norm": 1.173141360282898, + "learning_rate": 2.1180720999958857e-06, + "loss": 0.6272, + "step": 2146 + }, + { + "epoch": 3.36482254697286, + "grad_norm": 1.2015706300735474, + "learning_rate": 2.115990211072765e-06, + "loss": 0.6148, + "step": 2147 + }, + { + "epoch": 3.3663883089770357, + "grad_norm": 1.2163746356964111, + "learning_rate": 2.1139085948528286e-06, + "loss": 0.697, + "step": 2148 + }, + { + "epoch": 3.367954070981211, + "grad_norm": 1.2009632587432861, + "learning_rate": 2.111827252814328e-06, + "loss": 0.6202, + "step": 2149 + }, + { + "epoch": 3.3695198329853864, + "grad_norm": 1.1929175853729248, + "learning_rate": 2.1097461864353202e-06, + "loss": 0.5118, + "step": 2150 + }, + { + "epoch": 3.3710855949895615, + "grad_norm": 1.1137278079986572, + "learning_rate": 2.1076653971936685e-06, + "loss": 0.6489, + "step": 2151 + }, + { + "epoch": 3.372651356993737, + "grad_norm": 1.248371958732605, + "learning_rate": 2.1055848865670363e-06, + "loss": 0.6153, + "step": 2152 + }, + { + "epoch": 3.374217118997912, + "grad_norm": 1.1446887254714966, + "learning_rate": 2.103504656032891e-06, + "loss": 0.586, + "step": 2153 + }, + { + "epoch": 3.375782881002088, + "grad_norm": 1.3197392225265503, + "learning_rate": 2.1014247070685024e-06, + "loss": 0.6469, + "step": 2154 + }, + { + "epoch": 3.377348643006263, + "grad_norm": 1.1482443809509277, + "learning_rate": 2.099345041150938e-06, + "loss": 0.5673, + "step": 2155 + }, + { + "epoch": 3.3789144050104385, + "grad_norm": 1.3151997327804565, + "learning_rate": 2.097265659757064e-06, + "loss": 0.6684, + "step": 2156 + }, + { + "epoch": 3.3804801670146136, + "grad_norm": 1.1619030237197876, + "learning_rate": 2.0951865643635472e-06, + "loss": 0.6521, + "step": 2157 + }, + { + "epoch": 3.382045929018789, + "grad_norm": 1.2650394439697266, + "learning_rate": 2.09310775644685e-06, + "loss": 0.5612, + "step": 2158 + }, + { + "epoch": 3.3836116910229643, + "grad_norm": 1.1590896844863892, + "learning_rate": 2.0910292374832285e-06, + "loss": 0.6129, + "step": 2159 + }, + { + "epoch": 3.38517745302714, + "grad_norm": 1.1127238273620605, + "learning_rate": 2.088951008948738e-06, + "loss": 0.5928, + "step": 2160 + }, + { + "epoch": 3.3867432150313155, + "grad_norm": 1.203165888786316, + "learning_rate": 2.086873072319223e-06, + "loss": 0.6821, + "step": 2161 + }, + { + "epoch": 3.3883089770354906, + "grad_norm": 1.186804175376892, + "learning_rate": 2.0847954290703234e-06, + "loss": 0.6913, + "step": 2162 + }, + { + "epoch": 3.3898747390396657, + "grad_norm": 1.2237906455993652, + "learning_rate": 2.0827180806774717e-06, + "loss": 0.635, + "step": 2163 + }, + { + "epoch": 3.3914405010438413, + "grad_norm": 1.1457163095474243, + "learning_rate": 2.080641028615888e-06, + "loss": 0.6628, + "step": 2164 + }, + { + "epoch": 3.393006263048017, + "grad_norm": 1.2211480140686035, + "learning_rate": 2.078564274360582e-06, + "loss": 0.6259, + "step": 2165 + }, + { + "epoch": 3.394572025052192, + "grad_norm": 1.1769284009933472, + "learning_rate": 2.076487819386356e-06, + "loss": 0.6458, + "step": 2166 + }, + { + "epoch": 3.3961377870563676, + "grad_norm": 1.1229456663131714, + "learning_rate": 2.0744116651677967e-06, + "loss": 0.6518, + "step": 2167 + }, + { + "epoch": 3.3977035490605427, + "grad_norm": 1.2193604707717896, + "learning_rate": 2.072335813179275e-06, + "loss": 0.636, + "step": 2168 + }, + { + "epoch": 3.3992693110647183, + "grad_norm": 1.2844295501708984, + "learning_rate": 2.0702602648949536e-06, + "loss": 0.6275, + "step": 2169 + }, + { + "epoch": 3.4008350730688934, + "grad_norm": 1.2357110977172852, + "learning_rate": 2.0681850217887737e-06, + "loss": 0.6535, + "step": 2170 + }, + { + "epoch": 3.402400835073069, + "grad_norm": 1.191166877746582, + "learning_rate": 2.0661100853344608e-06, + "loss": 0.63, + "step": 2171 + }, + { + "epoch": 3.403966597077244, + "grad_norm": 1.1162155866622925, + "learning_rate": 2.064035457005526e-06, + "loss": 0.6438, + "step": 2172 + }, + { + "epoch": 3.4055323590814197, + "grad_norm": 1.1535841226577759, + "learning_rate": 2.0619611382752575e-06, + "loss": 0.6106, + "step": 2173 + }, + { + "epoch": 3.407098121085595, + "grad_norm": 1.2141941785812378, + "learning_rate": 2.059887130616726e-06, + "loss": 0.6009, + "step": 2174 + }, + { + "epoch": 3.4086638830897704, + "grad_norm": 1.1389793157577515, + "learning_rate": 2.05781343550278e-06, + "loss": 0.6691, + "step": 2175 + }, + { + "epoch": 3.4102296450939455, + "grad_norm": 1.2746304273605347, + "learning_rate": 2.055740054406048e-06, + "loss": 0.5752, + "step": 2176 + }, + { + "epoch": 3.411795407098121, + "grad_norm": 1.250404953956604, + "learning_rate": 2.053666988798933e-06, + "loss": 0.6839, + "step": 2177 + }, + { + "epoch": 3.4133611691022967, + "grad_norm": 1.2025327682495117, + "learning_rate": 2.0515942401536148e-06, + "loss": 0.6806, + "step": 2178 + }, + { + "epoch": 3.414926931106472, + "grad_norm": 1.1835360527038574, + "learning_rate": 2.0495218099420504e-06, + "loss": 0.6353, + "step": 2179 + }, + { + "epoch": 3.416492693110647, + "grad_norm": 1.3114181756973267, + "learning_rate": 2.0474496996359676e-06, + "loss": 0.6561, + "step": 2180 + }, + { + "epoch": 3.4180584551148225, + "grad_norm": 1.0880002975463867, + "learning_rate": 2.045377910706867e-06, + "loss": 0.6646, + "step": 2181 + }, + { + "epoch": 3.419624217118998, + "grad_norm": 1.1725163459777832, + "learning_rate": 2.0433064446260235e-06, + "loss": 0.5227, + "step": 2182 + }, + { + "epoch": 3.4211899791231732, + "grad_norm": 1.1769042015075684, + "learning_rate": 2.041235302864482e-06, + "loss": 0.6465, + "step": 2183 + }, + { + "epoch": 3.422755741127349, + "grad_norm": 1.2654465436935425, + "learning_rate": 2.039164486893054e-06, + "loss": 0.6407, + "step": 2184 + }, + { + "epoch": 3.424321503131524, + "grad_norm": 1.2088981866836548, + "learning_rate": 2.037093998182325e-06, + "loss": 0.6607, + "step": 2185 + }, + { + "epoch": 3.4258872651356995, + "grad_norm": 1.2232211828231812, + "learning_rate": 2.035023838202643e-06, + "loss": 0.6702, + "step": 2186 + }, + { + "epoch": 3.4274530271398747, + "grad_norm": 1.214909553527832, + "learning_rate": 2.032954008424126e-06, + "loss": 0.6322, + "step": 2187 + }, + { + "epoch": 3.4290187891440502, + "grad_norm": 1.1763339042663574, + "learning_rate": 2.0308845103166555e-06, + "loss": 0.6287, + "step": 2188 + }, + { + "epoch": 3.4305845511482254, + "grad_norm": 1.3962702751159668, + "learning_rate": 2.028815345349879e-06, + "loss": 0.547, + "step": 2189 + }, + { + "epoch": 3.432150313152401, + "grad_norm": 1.2137348651885986, + "learning_rate": 2.026746514993205e-06, + "loss": 0.6401, + "step": 2190 + }, + { + "epoch": 3.433716075156576, + "grad_norm": 1.143447756767273, + "learning_rate": 2.024678020715808e-06, + "loss": 0.6563, + "step": 2191 + }, + { + "epoch": 3.4352818371607516, + "grad_norm": 1.1660840511322021, + "learning_rate": 2.022609863986621e-06, + "loss": 0.6908, + "step": 2192 + }, + { + "epoch": 3.4368475991649268, + "grad_norm": 1.2545430660247803, + "learning_rate": 2.020542046274337e-06, + "loss": 0.6209, + "step": 2193 + }, + { + "epoch": 3.4384133611691023, + "grad_norm": 1.3274216651916504, + "learning_rate": 2.0184745690474116e-06, + "loss": 0.6243, + "step": 2194 + }, + { + "epoch": 3.4399791231732775, + "grad_norm": 1.2278743982315063, + "learning_rate": 2.0164074337740548e-06, + "loss": 0.6487, + "step": 2195 + }, + { + "epoch": 3.441544885177453, + "grad_norm": 1.1612262725830078, + "learning_rate": 2.0143406419222354e-06, + "loss": 0.5982, + "step": 2196 + }, + { + "epoch": 3.443110647181628, + "grad_norm": 1.2417254447937012, + "learning_rate": 2.01227419495968e-06, + "loss": 0.6292, + "step": 2197 + }, + { + "epoch": 3.4446764091858038, + "grad_norm": 1.1197260618209839, + "learning_rate": 2.0102080943538664e-06, + "loss": 0.6478, + "step": 2198 + }, + { + "epoch": 3.4462421711899793, + "grad_norm": 1.113791584968567, + "learning_rate": 2.008142341572029e-06, + "loss": 0.5747, + "step": 2199 + }, + { + "epoch": 3.4478079331941545, + "grad_norm": 1.1108101606369019, + "learning_rate": 2.0060769380811552e-06, + "loss": 0.6325, + "step": 2200 + }, + { + "epoch": 3.44937369519833, + "grad_norm": 1.2102993726730347, + "learning_rate": 2.004011885347985e-06, + "loss": 0.6066, + "step": 2201 + }, + { + "epoch": 3.450939457202505, + "grad_norm": 1.304426908493042, + "learning_rate": 2.0019471848390055e-06, + "loss": 0.6503, + "step": 2202 + }, + { + "epoch": 3.4525052192066807, + "grad_norm": 1.244708776473999, + "learning_rate": 1.9998828380204596e-06, + "loss": 0.6723, + "step": 2203 + }, + { + "epoch": 3.454070981210856, + "grad_norm": 1.1601401567459106, + "learning_rate": 1.997818846358333e-06, + "loss": 0.684, + "step": 2204 + }, + { + "epoch": 3.4556367432150314, + "grad_norm": 1.1886024475097656, + "learning_rate": 1.9957552113183635e-06, + "loss": 0.6277, + "step": 2205 + }, + { + "epoch": 3.4572025052192066, + "grad_norm": 1.2624013423919678, + "learning_rate": 1.993691934366035e-06, + "loss": 0.6569, + "step": 2206 + }, + { + "epoch": 3.458768267223382, + "grad_norm": 1.272737979888916, + "learning_rate": 1.991629016966575e-06, + "loss": 0.6481, + "step": 2207 + }, + { + "epoch": 3.4603340292275573, + "grad_norm": 1.2370678186416626, + "learning_rate": 1.989566460584956e-06, + "loss": 0.6598, + "step": 2208 + }, + { + "epoch": 3.461899791231733, + "grad_norm": 1.1626964807510376, + "learning_rate": 1.9875042666858973e-06, + "loss": 0.6582, + "step": 2209 + }, + { + "epoch": 3.463465553235908, + "grad_norm": 1.1608844995498657, + "learning_rate": 1.9854424367338578e-06, + "loss": 0.5931, + "step": 2210 + }, + { + "epoch": 3.4650313152400836, + "grad_norm": 1.2470782995224, + "learning_rate": 1.983380972193037e-06, + "loss": 0.6186, + "step": 2211 + }, + { + "epoch": 3.4665970772442587, + "grad_norm": 1.1697843074798584, + "learning_rate": 1.98131987452738e-06, + "loss": 0.6373, + "step": 2212 + }, + { + "epoch": 3.4681628392484343, + "grad_norm": 1.1374456882476807, + "learning_rate": 1.9792591452005645e-06, + "loss": 0.6518, + "step": 2213 + }, + { + "epoch": 3.4697286012526094, + "grad_norm": 1.1979857683181763, + "learning_rate": 1.9771987856760115e-06, + "loss": 0.6174, + "step": 2214 + }, + { + "epoch": 3.471294363256785, + "grad_norm": 1.2085691690444946, + "learning_rate": 1.975138797416879e-06, + "loss": 0.6296, + "step": 2215 + }, + { + "epoch": 3.4728601252609606, + "grad_norm": 1.1843719482421875, + "learning_rate": 1.973079181886059e-06, + "loss": 0.6796, + "step": 2216 + }, + { + "epoch": 3.4744258872651357, + "grad_norm": 1.230175256729126, + "learning_rate": 1.97101994054618e-06, + "loss": 0.6626, + "step": 2217 + }, + { + "epoch": 3.4759916492693113, + "grad_norm": 1.14729642868042, + "learning_rate": 1.9689610748596056e-06, + "loss": 0.5719, + "step": 2218 + }, + { + "epoch": 3.4775574112734864, + "grad_norm": 1.3141698837280273, + "learning_rate": 1.966902586288432e-06, + "loss": 0.6455, + "step": 2219 + }, + { + "epoch": 3.479123173277662, + "grad_norm": 1.15304434299469, + "learning_rate": 1.964844476294487e-06, + "loss": 0.6873, + "step": 2220 + }, + { + "epoch": 3.480688935281837, + "grad_norm": 1.193190336227417, + "learning_rate": 1.9627867463393308e-06, + "loss": 0.6656, + "step": 2221 + }, + { + "epoch": 3.4822546972860127, + "grad_norm": 1.2585062980651855, + "learning_rate": 1.960729397884252e-06, + "loss": 0.673, + "step": 2222 + }, + { + "epoch": 3.483820459290188, + "grad_norm": 1.2261855602264404, + "learning_rate": 1.9586724323902704e-06, + "loss": 0.6563, + "step": 2223 + }, + { + "epoch": 3.4853862212943634, + "grad_norm": 1.1301530599594116, + "learning_rate": 1.9566158513181312e-06, + "loss": 0.6392, + "step": 2224 + }, + { + "epoch": 3.4869519832985385, + "grad_norm": 1.1394745111465454, + "learning_rate": 1.954559656128309e-06, + "loss": 0.5913, + "step": 2225 + }, + { + "epoch": 3.488517745302714, + "grad_norm": 1.2290774583816528, + "learning_rate": 1.9525038482810044e-06, + "loss": 0.6117, + "step": 2226 + }, + { + "epoch": 3.490083507306889, + "grad_norm": 1.1881879568099976, + "learning_rate": 1.9504484292361403e-06, + "loss": 0.6634, + "step": 2227 + }, + { + "epoch": 3.491649269311065, + "grad_norm": 1.157094955444336, + "learning_rate": 1.948393400453367e-06, + "loss": 0.608, + "step": 2228 + }, + { + "epoch": 3.49321503131524, + "grad_norm": 1.1708686351776123, + "learning_rate": 1.9463387633920543e-06, + "loss": 0.6324, + "step": 2229 + }, + { + "epoch": 3.4947807933194155, + "grad_norm": 1.1796098947525024, + "learning_rate": 1.9442845195112954e-06, + "loss": 0.6601, + "step": 2230 + }, + { + "epoch": 3.4963465553235906, + "grad_norm": 1.2861493825912476, + "learning_rate": 1.942230670269906e-06, + "loss": 0.615, + "step": 2231 + }, + { + "epoch": 3.497912317327766, + "grad_norm": 1.2033058404922485, + "learning_rate": 1.9401772171264187e-06, + "loss": 0.6148, + "step": 2232 + }, + { + "epoch": 3.4994780793319418, + "grad_norm": 1.2392219305038452, + "learning_rate": 1.938124161539085e-06, + "loss": 0.5715, + "step": 2233 + }, + { + "epoch": 3.501043841336117, + "grad_norm": 1.1681913137435913, + "learning_rate": 1.936071504965876e-06, + "loss": 0.5809, + "step": 2234 + }, + { + "epoch": 3.502609603340292, + "grad_norm": 1.2426320314407349, + "learning_rate": 1.9340192488644793e-06, + "loss": 0.6475, + "step": 2235 + }, + { + "epoch": 3.5041753653444676, + "grad_norm": 1.2066962718963623, + "learning_rate": 1.9319673946922953e-06, + "loss": 0.5845, + "step": 2236 + }, + { + "epoch": 3.505741127348643, + "grad_norm": 1.3178675174713135, + "learning_rate": 1.929915943906443e-06, + "loss": 0.6422, + "step": 2237 + }, + { + "epoch": 3.5073068893528183, + "grad_norm": 1.1881810426712036, + "learning_rate": 1.927864897963751e-06, + "loss": 0.6726, + "step": 2238 + }, + { + "epoch": 3.508872651356994, + "grad_norm": 1.130273699760437, + "learning_rate": 1.925814258320763e-06, + "loss": 0.6027, + "step": 2239 + }, + { + "epoch": 3.510438413361169, + "grad_norm": 1.1589324474334717, + "learning_rate": 1.923764026433734e-06, + "loss": 0.5668, + "step": 2240 + }, + { + "epoch": 3.5120041753653446, + "grad_norm": 1.2391080856323242, + "learning_rate": 1.921714203758629e-06, + "loss": 0.6494, + "step": 2241 + }, + { + "epoch": 3.5135699373695197, + "grad_norm": 1.2799794673919678, + "learning_rate": 1.9196647917511207e-06, + "loss": 0.6579, + "step": 2242 + }, + { + "epoch": 3.5151356993736953, + "grad_norm": 1.1656668186187744, + "learning_rate": 1.917615791866593e-06, + "loss": 0.6021, + "step": 2243 + }, + { + "epoch": 3.5167014613778704, + "grad_norm": 1.142409086227417, + "learning_rate": 1.9155672055601364e-06, + "loss": 0.6258, + "step": 2244 + }, + { + "epoch": 3.518267223382046, + "grad_norm": 1.1905341148376465, + "learning_rate": 1.9135190342865447e-06, + "loss": 0.6112, + "step": 2245 + }, + { + "epoch": 3.519832985386221, + "grad_norm": 1.143678903579712, + "learning_rate": 1.911471279500323e-06, + "loss": 0.6136, + "step": 2246 + }, + { + "epoch": 3.5213987473903967, + "grad_norm": 1.240808367729187, + "learning_rate": 1.909423942655674e-06, + "loss": 0.6403, + "step": 2247 + }, + { + "epoch": 3.522964509394572, + "grad_norm": 1.2007147073745728, + "learning_rate": 1.907377025206507e-06, + "loss": 0.6564, + "step": 2248 + }, + { + "epoch": 3.5245302713987474, + "grad_norm": 1.1846612691879272, + "learning_rate": 1.9053305286064356e-06, + "loss": 0.6839, + "step": 2249 + }, + { + "epoch": 3.526096033402923, + "grad_norm": 1.1430362462997437, + "learning_rate": 1.90328445430877e-06, + "loss": 0.6447, + "step": 2250 + }, + { + "epoch": 3.527661795407098, + "grad_norm": 1.1141353845596313, + "learning_rate": 1.9012388037665222e-06, + "loss": 0.637, + "step": 2251 + }, + { + "epoch": 3.5292275574112733, + "grad_norm": 1.329208254814148, + "learning_rate": 1.8991935784324048e-06, + "loss": 0.7307, + "step": 2252 + }, + { + "epoch": 3.530793319415449, + "grad_norm": 1.3457560539245605, + "learning_rate": 1.8971487797588276e-06, + "loss": 0.6345, + "step": 2253 + }, + { + "epoch": 3.5323590814196244, + "grad_norm": 1.1213788986206055, + "learning_rate": 1.8951044091978951e-06, + "loss": 0.5832, + "step": 2254 + }, + { + "epoch": 3.5339248434237995, + "grad_norm": 1.2542991638183594, + "learning_rate": 1.8930604682014125e-06, + "loss": 0.7679, + "step": 2255 + }, + { + "epoch": 3.535490605427975, + "grad_norm": 1.20833158493042, + "learning_rate": 1.891016958220875e-06, + "loss": 0.6436, + "step": 2256 + }, + { + "epoch": 3.5370563674321502, + "grad_norm": 1.2129275798797607, + "learning_rate": 1.8889738807074743e-06, + "loss": 0.6603, + "step": 2257 + }, + { + "epoch": 3.538622129436326, + "grad_norm": 1.2521908283233643, + "learning_rate": 1.886931237112096e-06, + "loss": 0.635, + "step": 2258 + }, + { + "epoch": 3.540187891440501, + "grad_norm": 1.248688817024231, + "learning_rate": 1.8848890288853152e-06, + "loss": 0.6358, + "step": 2259 + }, + { + "epoch": 3.5417536534446765, + "grad_norm": 1.1389580965042114, + "learning_rate": 1.882847257477398e-06, + "loss": 0.621, + "step": 2260 + }, + { + "epoch": 3.5433194154488517, + "grad_norm": 1.2111111879348755, + "learning_rate": 1.880805924338302e-06, + "loss": 0.6486, + "step": 2261 + }, + { + "epoch": 3.5448851774530272, + "grad_norm": 1.1926789283752441, + "learning_rate": 1.8787650309176735e-06, + "loss": 0.6613, + "step": 2262 + }, + { + "epoch": 3.5464509394572024, + "grad_norm": 1.144853949546814, + "learning_rate": 1.8767245786648435e-06, + "loss": 0.6353, + "step": 2263 + }, + { + "epoch": 3.548016701461378, + "grad_norm": 1.2768452167510986, + "learning_rate": 1.874684569028834e-06, + "loss": 0.6479, + "step": 2264 + }, + { + "epoch": 3.549582463465553, + "grad_norm": 1.2346423864364624, + "learning_rate": 1.8726450034583496e-06, + "loss": 0.6315, + "step": 2265 + }, + { + "epoch": 3.5511482254697286, + "grad_norm": 1.1488447189331055, + "learning_rate": 1.87060588340178e-06, + "loss": 0.6493, + "step": 2266 + }, + { + "epoch": 3.552713987473904, + "grad_norm": 1.1464084386825562, + "learning_rate": 1.8685672103072005e-06, + "loss": 0.6181, + "step": 2267 + }, + { + "epoch": 3.5542797494780793, + "grad_norm": 1.289974570274353, + "learning_rate": 1.8665289856223662e-06, + "loss": 0.6522, + "step": 2268 + }, + { + "epoch": 3.5558455114822545, + "grad_norm": 1.2859828472137451, + "learning_rate": 1.8644912107947164e-06, + "loss": 0.57, + "step": 2269 + }, + { + "epoch": 3.55741127348643, + "grad_norm": 1.176608920097351, + "learning_rate": 1.8624538872713676e-06, + "loss": 0.6546, + "step": 2270 + }, + { + "epoch": 3.5589770354906056, + "grad_norm": 1.335280418395996, + "learning_rate": 1.86041701649912e-06, + "loss": 0.6659, + "step": 2271 + }, + { + "epoch": 3.5605427974947808, + "grad_norm": 1.2693639993667603, + "learning_rate": 1.858380599924449e-06, + "loss": 0.6348, + "step": 2272 + }, + { + "epoch": 3.562108559498956, + "grad_norm": 1.1791412830352783, + "learning_rate": 1.856344638993508e-06, + "loss": 0.6755, + "step": 2273 + }, + { + "epoch": 3.5636743215031315, + "grad_norm": 1.1750035285949707, + "learning_rate": 1.8543091351521292e-06, + "loss": 0.6508, + "step": 2274 + }, + { + "epoch": 3.565240083507307, + "grad_norm": 1.1812372207641602, + "learning_rate": 1.8522740898458178e-06, + "loss": 0.5811, + "step": 2275 + }, + { + "epoch": 3.566805845511482, + "grad_norm": 1.1619631052017212, + "learning_rate": 1.8502395045197522e-06, + "loss": 0.6227, + "step": 2276 + }, + { + "epoch": 3.5683716075156577, + "grad_norm": 1.2588560581207275, + "learning_rate": 1.8482053806187878e-06, + "loss": 0.6806, + "step": 2277 + }, + { + "epoch": 3.569937369519833, + "grad_norm": 1.0842543840408325, + "learning_rate": 1.8461717195874504e-06, + "loss": 0.6315, + "step": 2278 + }, + { + "epoch": 3.5715031315240084, + "grad_norm": 1.1269354820251465, + "learning_rate": 1.8441385228699355e-06, + "loss": 0.6387, + "step": 2279 + }, + { + "epoch": 3.5730688935281836, + "grad_norm": 1.196655511856079, + "learning_rate": 1.8421057919101128e-06, + "loss": 0.6872, + "step": 2280 + }, + { + "epoch": 3.574634655532359, + "grad_norm": 1.3561420440673828, + "learning_rate": 1.8400735281515175e-06, + "loss": 0.6538, + "step": 2281 + }, + { + "epoch": 3.5762004175365343, + "grad_norm": 1.1209763288497925, + "learning_rate": 1.838041733037354e-06, + "loss": 0.6934, + "step": 2282 + }, + { + "epoch": 3.57776617954071, + "grad_norm": 1.0936529636383057, + "learning_rate": 1.8360104080104962e-06, + "loss": 0.7136, + "step": 2283 + }, + { + "epoch": 3.5793319415448854, + "grad_norm": 1.2066773176193237, + "learning_rate": 1.8339795545134814e-06, + "loss": 0.6259, + "step": 2284 + }, + { + "epoch": 3.5808977035490606, + "grad_norm": 1.0662962198257446, + "learning_rate": 1.8319491739885115e-06, + "loss": 0.6195, + "step": 2285 + }, + { + "epoch": 3.5824634655532357, + "grad_norm": 1.284588098526001, + "learning_rate": 1.8299192678774564e-06, + "loss": 0.6154, + "step": 2286 + }, + { + "epoch": 3.5840292275574113, + "grad_norm": 1.1617132425308228, + "learning_rate": 1.8278898376218456e-06, + "loss": 0.6574, + "step": 2287 + }, + { + "epoch": 3.585594989561587, + "grad_norm": 1.1630313396453857, + "learning_rate": 1.8258608846628708e-06, + "loss": 0.577, + "step": 2288 + }, + { + "epoch": 3.587160751565762, + "grad_norm": 1.1374728679656982, + "learning_rate": 1.8238324104413879e-06, + "loss": 0.6908, + "step": 2289 + }, + { + "epoch": 3.588726513569937, + "grad_norm": 1.1832219362258911, + "learning_rate": 1.8218044163979088e-06, + "loss": 0.6409, + "step": 2290 + }, + { + "epoch": 3.5902922755741127, + "grad_norm": 1.4350699186325073, + "learning_rate": 1.8197769039726065e-06, + "loss": 0.6043, + "step": 2291 + }, + { + "epoch": 3.5918580375782883, + "grad_norm": 1.0792481899261475, + "learning_rate": 1.8177498746053129e-06, + "loss": 0.5953, + "step": 2292 + }, + { + "epoch": 3.5934237995824634, + "grad_norm": 1.2297285795211792, + "learning_rate": 1.8157233297355154e-06, + "loss": 0.646, + "step": 2293 + }, + { + "epoch": 3.594989561586639, + "grad_norm": 1.1795411109924316, + "learning_rate": 1.8136972708023557e-06, + "loss": 0.6187, + "step": 2294 + }, + { + "epoch": 3.596555323590814, + "grad_norm": 1.2370169162750244, + "learning_rate": 1.8116716992446342e-06, + "loss": 0.6899, + "step": 2295 + }, + { + "epoch": 3.5981210855949897, + "grad_norm": 1.1697649955749512, + "learning_rate": 1.8096466165008037e-06, + "loss": 0.6495, + "step": 2296 + }, + { + "epoch": 3.599686847599165, + "grad_norm": 1.1952897310256958, + "learning_rate": 1.807622024008967e-06, + "loss": 0.6007, + "step": 2297 + }, + { + "epoch": 3.6012526096033404, + "grad_norm": 1.2342556715011597, + "learning_rate": 1.8055979232068845e-06, + "loss": 0.5842, + "step": 2298 + }, + { + "epoch": 3.6028183716075155, + "grad_norm": 1.2582154273986816, + "learning_rate": 1.8035743155319618e-06, + "loss": 0.6939, + "step": 2299 + }, + { + "epoch": 3.604384133611691, + "grad_norm": 1.1596347093582153, + "learning_rate": 1.8015512024212573e-06, + "loss": 0.6609, + "step": 2300 + }, + { + "epoch": 3.6059498956158667, + "grad_norm": 1.2669062614440918, + "learning_rate": 1.7995285853114785e-06, + "loss": 0.6747, + "step": 2301 + }, + { + "epoch": 3.607515657620042, + "grad_norm": 1.2295914888381958, + "learning_rate": 1.797506465638979e-06, + "loss": 0.6374, + "step": 2302 + }, + { + "epoch": 3.609081419624217, + "grad_norm": 1.2566529512405396, + "learning_rate": 1.795484844839759e-06, + "loss": 0.6034, + "step": 2303 + }, + { + "epoch": 3.6106471816283925, + "grad_norm": 1.2655516862869263, + "learning_rate": 1.7934637243494663e-06, + "loss": 0.6394, + "step": 2304 + }, + { + "epoch": 3.612212943632568, + "grad_norm": 1.2168526649475098, + "learning_rate": 1.7914431056033927e-06, + "loss": 0.639, + "step": 2305 + }, + { + "epoch": 3.613778705636743, + "grad_norm": 1.309919834136963, + "learning_rate": 1.7894229900364721e-06, + "loss": 0.648, + "step": 2306 + }, + { + "epoch": 3.6153444676409183, + "grad_norm": 1.3890948295593262, + "learning_rate": 1.7874033790832846e-06, + "loss": 0.5951, + "step": 2307 + }, + { + "epoch": 3.616910229645094, + "grad_norm": 1.1613552570343018, + "learning_rate": 1.7853842741780474e-06, + "loss": 0.6083, + "step": 2308 + }, + { + "epoch": 3.6184759916492695, + "grad_norm": 1.3169668912887573, + "learning_rate": 1.783365676754621e-06, + "loss": 0.6439, + "step": 2309 + }, + { + "epoch": 3.6200417536534446, + "grad_norm": 1.21388578414917, + "learning_rate": 1.781347588246507e-06, + "loss": 0.6702, + "step": 2310 + }, + { + "epoch": 3.62160751565762, + "grad_norm": 1.202735185623169, + "learning_rate": 1.7793300100868428e-06, + "loss": 0.6417, + "step": 2311 + }, + { + "epoch": 3.6231732776617953, + "grad_norm": 1.2919602394104004, + "learning_rate": 1.7773129437084028e-06, + "loss": 0.6139, + "step": 2312 + }, + { + "epoch": 3.624739039665971, + "grad_norm": 1.1894437074661255, + "learning_rate": 1.7752963905436014e-06, + "loss": 0.6689, + "step": 2313 + }, + { + "epoch": 3.626304801670146, + "grad_norm": 1.1630206108093262, + "learning_rate": 1.7732803520244866e-06, + "loss": 0.6381, + "step": 2314 + }, + { + "epoch": 3.6278705636743216, + "grad_norm": 1.2039670944213867, + "learning_rate": 1.7712648295827395e-06, + "loss": 0.6972, + "step": 2315 + }, + { + "epoch": 3.6294363256784967, + "grad_norm": 1.2051432132720947, + "learning_rate": 1.769249824649677e-06, + "loss": 0.6404, + "step": 2316 + }, + { + "epoch": 3.6310020876826723, + "grad_norm": 1.2452560663223267, + "learning_rate": 1.7672353386562486e-06, + "loss": 0.6057, + "step": 2317 + }, + { + "epoch": 3.632567849686848, + "grad_norm": 1.1576342582702637, + "learning_rate": 1.7652213730330336e-06, + "loss": 0.6235, + "step": 2318 + }, + { + "epoch": 3.634133611691023, + "grad_norm": 1.3302104473114014, + "learning_rate": 1.7632079292102412e-06, + "loss": 0.6464, + "step": 2319 + }, + { + "epoch": 3.635699373695198, + "grad_norm": 1.1526870727539062, + "learning_rate": 1.7611950086177132e-06, + "loss": 0.6619, + "step": 2320 + }, + { + "epoch": 3.6372651356993737, + "grad_norm": 1.1325047016143799, + "learning_rate": 1.7591826126849179e-06, + "loss": 0.6815, + "step": 2321 + }, + { + "epoch": 3.6388308977035493, + "grad_norm": 1.2510312795639038, + "learning_rate": 1.7571707428409496e-06, + "loss": 0.6344, + "step": 2322 + }, + { + "epoch": 3.6403966597077244, + "grad_norm": 1.3015072345733643, + "learning_rate": 1.7551594005145323e-06, + "loss": 0.6082, + "step": 2323 + }, + { + "epoch": 3.6419624217118995, + "grad_norm": 1.1332790851593018, + "learning_rate": 1.7531485871340122e-06, + "loss": 0.6854, + "step": 2324 + }, + { + "epoch": 3.643528183716075, + "grad_norm": 1.1714104413986206, + "learning_rate": 1.7511383041273617e-06, + "loss": 0.6497, + "step": 2325 + }, + { + "epoch": 3.6450939457202507, + "grad_norm": 1.1853827238082886, + "learning_rate": 1.7491285529221768e-06, + "loss": 0.5655, + "step": 2326 + }, + { + "epoch": 3.646659707724426, + "grad_norm": 1.2981195449829102, + "learning_rate": 1.7471193349456746e-06, + "loss": 0.6327, + "step": 2327 + }, + { + "epoch": 3.6482254697286014, + "grad_norm": 1.2444530725479126, + "learning_rate": 1.7451106516246932e-06, + "loss": 0.643, + "step": 2328 + }, + { + "epoch": 3.6497912317327765, + "grad_norm": 1.1355667114257812, + "learning_rate": 1.7431025043856931e-06, + "loss": 0.6439, + "step": 2329 + }, + { + "epoch": 3.651356993736952, + "grad_norm": 1.1224238872528076, + "learning_rate": 1.7410948946547536e-06, + "loss": 0.6606, + "step": 2330 + }, + { + "epoch": 3.6529227557411272, + "grad_norm": 1.163265347480774, + "learning_rate": 1.739087823857569e-06, + "loss": 0.5989, + "step": 2331 + }, + { + "epoch": 3.654488517745303, + "grad_norm": 1.3157579898834229, + "learning_rate": 1.7370812934194565e-06, + "loss": 0.6818, + "step": 2332 + }, + { + "epoch": 3.656054279749478, + "grad_norm": 1.1727818250656128, + "learning_rate": 1.735075304765344e-06, + "loss": 0.6271, + "step": 2333 + }, + { + "epoch": 3.6576200417536535, + "grad_norm": 1.2371114492416382, + "learning_rate": 1.733069859319778e-06, + "loss": 0.667, + "step": 2334 + }, + { + "epoch": 3.6591858037578286, + "grad_norm": 1.210724115371704, + "learning_rate": 1.73106495850692e-06, + "loss": 0.725, + "step": 2335 + }, + { + "epoch": 3.6607515657620042, + "grad_norm": 1.1199427843093872, + "learning_rate": 1.7290606037505417e-06, + "loss": 0.6634, + "step": 2336 + }, + { + "epoch": 3.6623173277661794, + "grad_norm": 1.2661418914794922, + "learning_rate": 1.7270567964740275e-06, + "loss": 0.6259, + "step": 2337 + }, + { + "epoch": 3.663883089770355, + "grad_norm": 1.2222251892089844, + "learning_rate": 1.7250535381003758e-06, + "loss": 0.6082, + "step": 2338 + }, + { + "epoch": 3.6654488517745305, + "grad_norm": 1.1748756170272827, + "learning_rate": 1.7230508300521935e-06, + "loss": 0.6017, + "step": 2339 + }, + { + "epoch": 3.6670146137787056, + "grad_norm": 1.2061327695846558, + "learning_rate": 1.7210486737516947e-06, + "loss": 0.6196, + "step": 2340 + }, + { + "epoch": 3.6685803757828808, + "grad_norm": 1.168503999710083, + "learning_rate": 1.7190470706207058e-06, + "loss": 0.6709, + "step": 2341 + }, + { + "epoch": 3.6701461377870563, + "grad_norm": 1.1397614479064941, + "learning_rate": 1.7170460220806563e-06, + "loss": 0.6043, + "step": 2342 + }, + { + "epoch": 3.671711899791232, + "grad_norm": 1.1928285360336304, + "learning_rate": 1.7150455295525842e-06, + "loss": 0.5985, + "step": 2343 + }, + { + "epoch": 3.673277661795407, + "grad_norm": 1.205994725227356, + "learning_rate": 1.713045594457133e-06, + "loss": 0.6461, + "step": 2344 + }, + { + "epoch": 3.6748434237995826, + "grad_norm": 1.192114233970642, + "learning_rate": 1.7110462182145494e-06, + "loss": 0.6461, + "step": 2345 + }, + { + "epoch": 3.6764091858037578, + "grad_norm": 1.2672173976898193, + "learning_rate": 1.7090474022446812e-06, + "loss": 0.6623, + "step": 2346 + }, + { + "epoch": 3.6779749478079333, + "grad_norm": 1.1894220113754272, + "learning_rate": 1.707049147966983e-06, + "loss": 0.6687, + "step": 2347 + }, + { + "epoch": 3.6795407098121085, + "grad_norm": 1.170541763305664, + "learning_rate": 1.7050514568005072e-06, + "loss": 0.6258, + "step": 2348 + }, + { + "epoch": 3.681106471816284, + "grad_norm": 1.2120347023010254, + "learning_rate": 1.7030543301639057e-06, + "loss": 0.5854, + "step": 2349 + }, + { + "epoch": 3.682672233820459, + "grad_norm": 1.1434928178787231, + "learning_rate": 1.7010577694754332e-06, + "loss": 0.6177, + "step": 2350 + }, + { + "epoch": 3.6842379958246347, + "grad_norm": 1.2849304676055908, + "learning_rate": 1.6990617761529388e-06, + "loss": 0.5642, + "step": 2351 + }, + { + "epoch": 3.68580375782881, + "grad_norm": 1.1992167234420776, + "learning_rate": 1.6970663516138692e-06, + "loss": 0.6594, + "step": 2352 + }, + { + "epoch": 3.6873695198329854, + "grad_norm": 1.2290526628494263, + "learning_rate": 1.6950714972752702e-06, + "loss": 0.6506, + "step": 2353 + }, + { + "epoch": 3.6889352818371606, + "grad_norm": 1.1701496839523315, + "learning_rate": 1.6930772145537794e-06, + "loss": 0.5947, + "step": 2354 + }, + { + "epoch": 3.690501043841336, + "grad_norm": 1.2298767566680908, + "learning_rate": 1.6910835048656282e-06, + "loss": 0.6824, + "step": 2355 + }, + { + "epoch": 3.6920668058455117, + "grad_norm": 1.1752413511276245, + "learning_rate": 1.6890903696266447e-06, + "loss": 0.5763, + "step": 2356 + }, + { + "epoch": 3.693632567849687, + "grad_norm": 1.174170732498169, + "learning_rate": 1.6870978102522462e-06, + "loss": 0.5934, + "step": 2357 + }, + { + "epoch": 3.695198329853862, + "grad_norm": 1.1316381692886353, + "learning_rate": 1.68510582815744e-06, + "loss": 0.5857, + "step": 2358 + }, + { + "epoch": 3.6967640918580376, + "grad_norm": 1.2187336683273315, + "learning_rate": 1.6831144247568274e-06, + "loss": 0.6832, + "step": 2359 + }, + { + "epoch": 3.698329853862213, + "grad_norm": 1.2257745265960693, + "learning_rate": 1.6811236014645949e-06, + "loss": 0.6001, + "step": 2360 + }, + { + "epoch": 3.6998956158663883, + "grad_norm": 1.2153297662734985, + "learning_rate": 1.6791333596945197e-06, + "loss": 0.6452, + "step": 2361 + }, + { + "epoch": 3.701461377870564, + "grad_norm": 1.203791618347168, + "learning_rate": 1.677143700859963e-06, + "loss": 0.6032, + "step": 2362 + }, + { + "epoch": 3.703027139874739, + "grad_norm": 1.215394139289856, + "learning_rate": 1.6751546263738761e-06, + "loss": 0.694, + "step": 2363 + }, + { + "epoch": 3.7045929018789145, + "grad_norm": 1.1475002765655518, + "learning_rate": 1.6731661376487923e-06, + "loss": 0.6627, + "step": 2364 + }, + { + "epoch": 3.7061586638830897, + "grad_norm": 1.2240904569625854, + "learning_rate": 1.6711782360968288e-06, + "loss": 0.6605, + "step": 2365 + }, + { + "epoch": 3.7077244258872653, + "grad_norm": 1.1701643466949463, + "learning_rate": 1.6691909231296888e-06, + "loss": 0.5705, + "step": 2366 + }, + { + "epoch": 3.7092901878914404, + "grad_norm": 1.1641205549240112, + "learning_rate": 1.6672042001586538e-06, + "loss": 0.6604, + "step": 2367 + }, + { + "epoch": 3.710855949895616, + "grad_norm": 1.120191216468811, + "learning_rate": 1.6652180685945884e-06, + "loss": 0.567, + "step": 2368 + }, + { + "epoch": 3.712421711899791, + "grad_norm": 1.1677566766738892, + "learning_rate": 1.663232529847938e-06, + "loss": 0.6484, + "step": 2369 + }, + { + "epoch": 3.7139874739039667, + "grad_norm": 1.1286911964416504, + "learning_rate": 1.6612475853287252e-06, + "loss": 0.5604, + "step": 2370 + }, + { + "epoch": 3.715553235908142, + "grad_norm": 1.2938463687896729, + "learning_rate": 1.65926323644655e-06, + "loss": 0.683, + "step": 2371 + }, + { + "epoch": 3.7171189979123174, + "grad_norm": 1.160536527633667, + "learning_rate": 1.6572794846105919e-06, + "loss": 0.6467, + "step": 2372 + }, + { + "epoch": 3.718684759916493, + "grad_norm": 1.1587703227996826, + "learning_rate": 1.655296331229606e-06, + "loss": 0.6386, + "step": 2373 + }, + { + "epoch": 3.720250521920668, + "grad_norm": 1.2075384855270386, + "learning_rate": 1.6533137777119196e-06, + "loss": 0.6331, + "step": 2374 + }, + { + "epoch": 3.721816283924843, + "grad_norm": 1.2803996801376343, + "learning_rate": 1.651331825465438e-06, + "loss": 0.6421, + "step": 2375 + }, + { + "epoch": 3.723382045929019, + "grad_norm": 1.1343461275100708, + "learning_rate": 1.6493504758976358e-06, + "loss": 0.6807, + "step": 2376 + }, + { + "epoch": 3.7249478079331944, + "grad_norm": 1.211895227432251, + "learning_rate": 1.6473697304155622e-06, + "loss": 0.6533, + "step": 2377 + }, + { + "epoch": 3.7265135699373695, + "grad_norm": 1.191689372062683, + "learning_rate": 1.6453895904258372e-06, + "loss": 0.6559, + "step": 2378 + }, + { + "epoch": 3.7280793319415446, + "grad_norm": 1.2990779876708984, + "learning_rate": 1.6434100573346496e-06, + "loss": 0.6354, + "step": 2379 + }, + { + "epoch": 3.72964509394572, + "grad_norm": 1.2150678634643555, + "learning_rate": 1.6414311325477567e-06, + "loss": 0.6438, + "step": 2380 + }, + { + "epoch": 3.7312108559498958, + "grad_norm": 1.1220753192901611, + "learning_rate": 1.639452817470486e-06, + "loss": 0.6315, + "step": 2381 + }, + { + "epoch": 3.732776617954071, + "grad_norm": 1.096580147743225, + "learning_rate": 1.6374751135077315e-06, + "loss": 0.6525, + "step": 2382 + }, + { + "epoch": 3.7343423799582465, + "grad_norm": 1.2590006589889526, + "learning_rate": 1.6354980220639504e-06, + "loss": 0.6609, + "step": 2383 + }, + { + "epoch": 3.7359081419624216, + "grad_norm": 1.2841553688049316, + "learning_rate": 1.6335215445431702e-06, + "loss": 0.6861, + "step": 2384 + }, + { + "epoch": 3.737473903966597, + "grad_norm": 1.1383644342422485, + "learning_rate": 1.6315456823489766e-06, + "loss": 0.6707, + "step": 2385 + }, + { + "epoch": 3.7390396659707723, + "grad_norm": 1.3325951099395752, + "learning_rate": 1.6295704368845221e-06, + "loss": 0.5672, + "step": 2386 + }, + { + "epoch": 3.740605427974948, + "grad_norm": 1.2384133338928223, + "learning_rate": 1.6275958095525217e-06, + "loss": 0.6274, + "step": 2387 + }, + { + "epoch": 3.742171189979123, + "grad_norm": 1.173836588859558, + "learning_rate": 1.6256218017552484e-06, + "loss": 0.5975, + "step": 2388 + }, + { + "epoch": 3.7437369519832986, + "grad_norm": 1.1901642084121704, + "learning_rate": 1.623648414894536e-06, + "loss": 0.6735, + "step": 2389 + }, + { + "epoch": 3.745302713987474, + "grad_norm": 1.2956198453903198, + "learning_rate": 1.6216756503717795e-06, + "loss": 0.6171, + "step": 2390 + }, + { + "epoch": 3.7468684759916493, + "grad_norm": 1.173884391784668, + "learning_rate": 1.619703509587931e-06, + "loss": 0.6375, + "step": 2391 + }, + { + "epoch": 3.7484342379958244, + "grad_norm": 1.1478065252304077, + "learning_rate": 1.6177319939434974e-06, + "loss": 0.6806, + "step": 2392 + }, + { + "epoch": 3.75, + "grad_norm": 1.1560224294662476, + "learning_rate": 1.6157611048385457e-06, + "loss": 0.5867, + "step": 2393 + }, + { + "epoch": 3.7515657620041756, + "grad_norm": 1.1529253721237183, + "learning_rate": 1.6137908436726943e-06, + "loss": 0.686, + "step": 2394 + }, + { + "epoch": 3.7531315240083507, + "grad_norm": 1.2027194499969482, + "learning_rate": 1.6118212118451171e-06, + "loss": 0.6579, + "step": 2395 + }, + { + "epoch": 3.754697286012526, + "grad_norm": 1.1857298612594604, + "learning_rate": 1.6098522107545426e-06, + "loss": 0.6274, + "step": 2396 + }, + { + "epoch": 3.7562630480167014, + "grad_norm": 1.2853041887283325, + "learning_rate": 1.6078838417992487e-06, + "loss": 0.6866, + "step": 2397 + }, + { + "epoch": 3.757828810020877, + "grad_norm": 1.1234532594680786, + "learning_rate": 1.6059161063770651e-06, + "loss": 0.5872, + "step": 2398 + }, + { + "epoch": 3.759394572025052, + "grad_norm": 1.2463502883911133, + "learning_rate": 1.603949005885373e-06, + "loss": 0.6566, + "step": 2399 + }, + { + "epoch": 3.7609603340292277, + "grad_norm": 1.1617919206619263, + "learning_rate": 1.6019825417211022e-06, + "loss": 0.6428, + "step": 2400 + }, + { + "epoch": 3.762526096033403, + "grad_norm": 1.1768699884414673, + "learning_rate": 1.6000167152807288e-06, + "loss": 0.6147, + "step": 2401 + }, + { + "epoch": 3.7640918580375784, + "grad_norm": 1.2775728702545166, + "learning_rate": 1.598051527960279e-06, + "loss": 0.6594, + "step": 2402 + }, + { + "epoch": 3.7656576200417535, + "grad_norm": 1.2250229120254517, + "learning_rate": 1.5960869811553225e-06, + "loss": 0.6844, + "step": 2403 + }, + { + "epoch": 3.767223382045929, + "grad_norm": 1.3038227558135986, + "learning_rate": 1.594123076260975e-06, + "loss": 0.6515, + "step": 2404 + }, + { + "epoch": 3.7687891440501042, + "grad_norm": 1.2010571956634521, + "learning_rate": 1.592159814671898e-06, + "loss": 0.6293, + "step": 2405 + }, + { + "epoch": 3.77035490605428, + "grad_norm": 1.2710856199264526, + "learning_rate": 1.590197197782294e-06, + "loss": 0.6096, + "step": 2406 + }, + { + "epoch": 3.7719206680584554, + "grad_norm": 1.2024561166763306, + "learning_rate": 1.588235226985908e-06, + "loss": 0.6227, + "step": 2407 + }, + { + "epoch": 3.7734864300626305, + "grad_norm": 1.1803232431411743, + "learning_rate": 1.5862739036760262e-06, + "loss": 0.6251, + "step": 2408 + }, + { + "epoch": 3.7750521920668056, + "grad_norm": 1.2758967876434326, + "learning_rate": 1.5843132292454765e-06, + "loss": 0.6948, + "step": 2409 + }, + { + "epoch": 3.776617954070981, + "grad_norm": 1.2238482236862183, + "learning_rate": 1.5823532050866242e-06, + "loss": 0.6186, + "step": 2410 + }, + { + "epoch": 3.778183716075157, + "grad_norm": 1.24042546749115, + "learning_rate": 1.5803938325913732e-06, + "loss": 0.6261, + "step": 2411 + }, + { + "epoch": 3.779749478079332, + "grad_norm": 1.137797474861145, + "learning_rate": 1.578435113151166e-06, + "loss": 0.5794, + "step": 2412 + }, + { + "epoch": 3.781315240083507, + "grad_norm": 1.2196292877197266, + "learning_rate": 1.5764770481569796e-06, + "loss": 0.64, + "step": 2413 + }, + { + "epoch": 3.7828810020876826, + "grad_norm": 1.1684080362319946, + "learning_rate": 1.5745196389993259e-06, + "loss": 0.5861, + "step": 2414 + }, + { + "epoch": 3.784446764091858, + "grad_norm": 1.2441654205322266, + "learning_rate": 1.5725628870682532e-06, + "loss": 0.6084, + "step": 2415 + }, + { + "epoch": 3.7860125260960333, + "grad_norm": 1.187728762626648, + "learning_rate": 1.5706067937533427e-06, + "loss": 0.6436, + "step": 2416 + }, + { + "epoch": 3.787578288100209, + "grad_norm": 1.2837245464324951, + "learning_rate": 1.5686513604437043e-06, + "loss": 0.5967, + "step": 2417 + }, + { + "epoch": 3.789144050104384, + "grad_norm": 1.1892480850219727, + "learning_rate": 1.566696588527985e-06, + "loss": 0.6547, + "step": 2418 + }, + { + "epoch": 3.7907098121085596, + "grad_norm": 1.2308993339538574, + "learning_rate": 1.5647424793943579e-06, + "loss": 0.6512, + "step": 2419 + }, + { + "epoch": 3.7922755741127347, + "grad_norm": 1.3206483125686646, + "learning_rate": 1.5627890344305256e-06, + "loss": 0.6845, + "step": 2420 + }, + { + "epoch": 3.7938413361169103, + "grad_norm": 1.151696801185608, + "learning_rate": 1.560836255023722e-06, + "loss": 0.597, + "step": 2421 + }, + { + "epoch": 3.7954070981210855, + "grad_norm": 1.2277007102966309, + "learning_rate": 1.5588841425607056e-06, + "loss": 0.6035, + "step": 2422 + }, + { + "epoch": 3.796972860125261, + "grad_norm": 1.2150050401687622, + "learning_rate": 1.556932698427761e-06, + "loss": 0.6679, + "step": 2423 + }, + { + "epoch": 3.798538622129436, + "grad_norm": 1.075766921043396, + "learning_rate": 1.5549819240107012e-06, + "loss": 0.6112, + "step": 2424 + }, + { + "epoch": 3.8001043841336117, + "grad_norm": 1.2026962041854858, + "learning_rate": 1.5530318206948607e-06, + "loss": 0.6104, + "step": 2425 + }, + { + "epoch": 3.801670146137787, + "grad_norm": 1.2209258079528809, + "learning_rate": 1.551082389865098e-06, + "loss": 0.6391, + "step": 2426 + }, + { + "epoch": 3.8032359081419624, + "grad_norm": 1.1685758829116821, + "learning_rate": 1.549133632905796e-06, + "loss": 0.6352, + "step": 2427 + }, + { + "epoch": 3.804801670146138, + "grad_norm": 1.1990925073623657, + "learning_rate": 1.547185551200856e-06, + "loss": 0.631, + "step": 2428 + }, + { + "epoch": 3.806367432150313, + "grad_norm": 1.1937427520751953, + "learning_rate": 1.5452381461337013e-06, + "loss": 0.6393, + "step": 2429 + }, + { + "epoch": 3.8079331941544883, + "grad_norm": 1.1711888313293457, + "learning_rate": 1.5432914190872757e-06, + "loss": 0.6415, + "step": 2430 + }, + { + "epoch": 3.809498956158664, + "grad_norm": 1.3688654899597168, + "learning_rate": 1.5413453714440404e-06, + "loss": 0.6711, + "step": 2431 + }, + { + "epoch": 3.8110647181628394, + "grad_norm": 1.1071898937225342, + "learning_rate": 1.5394000045859728e-06, + "loss": 0.5966, + "step": 2432 + }, + { + "epoch": 3.8126304801670146, + "grad_norm": 1.1659204959869385, + "learning_rate": 1.5374553198945696e-06, + "loss": 0.6672, + "step": 2433 + }, + { + "epoch": 3.81419624217119, + "grad_norm": 1.1961894035339355, + "learning_rate": 1.5355113187508413e-06, + "loss": 0.6229, + "step": 2434 + }, + { + "epoch": 3.8157620041753653, + "grad_norm": 1.2803820371627808, + "learning_rate": 1.5335680025353132e-06, + "loss": 0.657, + "step": 2435 + }, + { + "epoch": 3.817327766179541, + "grad_norm": 1.1563304662704468, + "learning_rate": 1.531625372628025e-06, + "loss": 0.6386, + "step": 2436 + }, + { + "epoch": 3.818893528183716, + "grad_norm": 1.2539798021316528, + "learning_rate": 1.5296834304085276e-06, + "loss": 0.5733, + "step": 2437 + }, + { + "epoch": 3.8204592901878915, + "grad_norm": 1.2615854740142822, + "learning_rate": 1.5277421772558842e-06, + "loss": 0.6073, + "step": 2438 + }, + { + "epoch": 3.8220250521920667, + "grad_norm": 1.2182807922363281, + "learning_rate": 1.5258016145486705e-06, + "loss": 0.6719, + "step": 2439 + }, + { + "epoch": 3.8235908141962422, + "grad_norm": 1.2267918586730957, + "learning_rate": 1.5238617436649689e-06, + "loss": 0.616, + "step": 2440 + }, + { + "epoch": 3.8251565762004174, + "grad_norm": 1.2033759355545044, + "learning_rate": 1.5219225659823706e-06, + "loss": 0.6759, + "step": 2441 + }, + { + "epoch": 3.826722338204593, + "grad_norm": 1.3071197271347046, + "learning_rate": 1.5199840828779777e-06, + "loss": 0.6672, + "step": 2442 + }, + { + "epoch": 3.828288100208768, + "grad_norm": 1.2482445240020752, + "learning_rate": 1.5180462957283971e-06, + "loss": 0.6363, + "step": 2443 + }, + { + "epoch": 3.8298538622129437, + "grad_norm": 1.2446298599243164, + "learning_rate": 1.5161092059097399e-06, + "loss": 0.6506, + "step": 2444 + }, + { + "epoch": 3.8314196242171192, + "grad_norm": 1.217212438583374, + "learning_rate": 1.5141728147976248e-06, + "loss": 0.7131, + "step": 2445 + }, + { + "epoch": 3.8329853862212944, + "grad_norm": 1.2086701393127441, + "learning_rate": 1.5122371237671726e-06, + "loss": 0.6107, + "step": 2446 + }, + { + "epoch": 3.8345511482254695, + "grad_norm": 1.244437336921692, + "learning_rate": 1.510302134193007e-06, + "loss": 0.5585, + "step": 2447 + }, + { + "epoch": 3.836116910229645, + "grad_norm": 1.1509884595870972, + "learning_rate": 1.5083678474492546e-06, + "loss": 0.5905, + "step": 2448 + }, + { + "epoch": 3.8376826722338206, + "grad_norm": 1.188522458076477, + "learning_rate": 1.5064342649095428e-06, + "loss": 0.6584, + "step": 2449 + }, + { + "epoch": 3.8392484342379958, + "grad_norm": 1.2903960943222046, + "learning_rate": 1.504501387946996e-06, + "loss": 0.6286, + "step": 2450 + }, + { + "epoch": 3.8408141962421714, + "grad_norm": 1.135545253753662, + "learning_rate": 1.5025692179342425e-06, + "loss": 0.6769, + "step": 2451 + }, + { + "epoch": 3.8423799582463465, + "grad_norm": 1.2375710010528564, + "learning_rate": 1.500637756243405e-06, + "loss": 0.6506, + "step": 2452 + }, + { + "epoch": 3.843945720250522, + "grad_norm": 1.7769895792007446, + "learning_rate": 1.4987070042461032e-06, + "loss": 0.6302, + "step": 2453 + }, + { + "epoch": 3.845511482254697, + "grad_norm": 1.0754798650741577, + "learning_rate": 1.4967769633134555e-06, + "loss": 0.6019, + "step": 2454 + }, + { + "epoch": 3.8470772442588728, + "grad_norm": 1.2362276315689087, + "learning_rate": 1.4948476348160723e-06, + "loss": 0.578, + "step": 2455 + }, + { + "epoch": 3.848643006263048, + "grad_norm": 1.1476119756698608, + "learning_rate": 1.4929190201240601e-06, + "loss": 0.5866, + "step": 2456 + }, + { + "epoch": 3.8502087682672235, + "grad_norm": 1.14879310131073, + "learning_rate": 1.4909911206070162e-06, + "loss": 0.6431, + "step": 2457 + }, + { + "epoch": 3.8517745302713986, + "grad_norm": 1.1493771076202393, + "learning_rate": 1.4890639376340336e-06, + "loss": 0.6297, + "step": 2458 + }, + { + "epoch": 3.853340292275574, + "grad_norm": 1.2373193502426147, + "learning_rate": 1.4871374725736936e-06, + "loss": 0.6387, + "step": 2459 + }, + { + "epoch": 3.8549060542797493, + "grad_norm": 1.2413831949234009, + "learning_rate": 1.485211726794068e-06, + "loss": 0.6569, + "step": 2460 + }, + { + "epoch": 3.856471816283925, + "grad_norm": 1.2246074676513672, + "learning_rate": 1.4832867016627194e-06, + "loss": 0.5908, + "step": 2461 + }, + { + "epoch": 3.8580375782881005, + "grad_norm": 1.1326402425765991, + "learning_rate": 1.4813623985466968e-06, + "loss": 0.6488, + "step": 2462 + }, + { + "epoch": 3.8596033402922756, + "grad_norm": 1.2424553632736206, + "learning_rate": 1.4794388188125364e-06, + "loss": 0.6299, + "step": 2463 + }, + { + "epoch": 3.8611691022964507, + "grad_norm": 1.2312356233596802, + "learning_rate": 1.4775159638262637e-06, + "loss": 0.6359, + "step": 2464 + }, + { + "epoch": 3.8627348643006263, + "grad_norm": 1.049404263496399, + "learning_rate": 1.475593834953386e-06, + "loss": 0.5736, + "step": 2465 + }, + { + "epoch": 3.864300626304802, + "grad_norm": 1.2294747829437256, + "learning_rate": 1.4736724335588952e-06, + "loss": 0.6342, + "step": 2466 + }, + { + "epoch": 3.865866388308977, + "grad_norm": 1.2269598245620728, + "learning_rate": 1.471751761007269e-06, + "loss": 0.6316, + "step": 2467 + }, + { + "epoch": 3.867432150313152, + "grad_norm": 1.138282299041748, + "learning_rate": 1.469831818662467e-06, + "loss": 0.6402, + "step": 2468 + }, + { + "epoch": 3.8689979123173277, + "grad_norm": 1.2380733489990234, + "learning_rate": 1.467912607887927e-06, + "loss": 0.639, + "step": 2469 + }, + { + "epoch": 3.8705636743215033, + "grad_norm": 1.109584093093872, + "learning_rate": 1.465994130046572e-06, + "loss": 0.6178, + "step": 2470 + }, + { + "epoch": 3.8721294363256784, + "grad_norm": 1.3927913904190063, + "learning_rate": 1.4640763865008006e-06, + "loss": 0.5844, + "step": 2471 + }, + { + "epoch": 3.873695198329854, + "grad_norm": 1.1995818614959717, + "learning_rate": 1.462159378612492e-06, + "loss": 0.6532, + "step": 2472 + }, + { + "epoch": 3.875260960334029, + "grad_norm": 1.244868278503418, + "learning_rate": 1.4602431077430035e-06, + "loss": 0.5823, + "step": 2473 + }, + { + "epoch": 3.8768267223382047, + "grad_norm": 1.2030248641967773, + "learning_rate": 1.4583275752531672e-06, + "loss": 0.6096, + "step": 2474 + }, + { + "epoch": 3.87839248434238, + "grad_norm": 1.1157633066177368, + "learning_rate": 1.4564127825032904e-06, + "loss": 0.6433, + "step": 2475 + }, + { + "epoch": 3.8799582463465554, + "grad_norm": 1.1652717590332031, + "learning_rate": 1.4544987308531594e-06, + "loss": 0.5917, + "step": 2476 + }, + { + "epoch": 3.8815240083507305, + "grad_norm": 1.2521084547042847, + "learning_rate": 1.452585421662029e-06, + "loss": 0.6254, + "step": 2477 + }, + { + "epoch": 3.883089770354906, + "grad_norm": 1.2177410125732422, + "learning_rate": 1.4506728562886285e-06, + "loss": 0.6565, + "step": 2478 + }, + { + "epoch": 3.8846555323590817, + "grad_norm": 1.2197600603103638, + "learning_rate": 1.4487610360911606e-06, + "loss": 0.6001, + "step": 2479 + }, + { + "epoch": 3.886221294363257, + "grad_norm": 1.186809778213501, + "learning_rate": 1.4468499624272972e-06, + "loss": 0.611, + "step": 2480 + }, + { + "epoch": 3.887787056367432, + "grad_norm": 1.2346223592758179, + "learning_rate": 1.4449396366541802e-06, + "loss": 0.7188, + "step": 2481 + }, + { + "epoch": 3.8893528183716075, + "grad_norm": 1.148040533065796, + "learning_rate": 1.4430300601284214e-06, + "loss": 0.6711, + "step": 2482 + }, + { + "epoch": 3.890918580375783, + "grad_norm": 1.2020211219787598, + "learning_rate": 1.4411212342060988e-06, + "loss": 0.6765, + "step": 2483 + }, + { + "epoch": 3.892484342379958, + "grad_norm": 1.2378863096237183, + "learning_rate": 1.439213160242759e-06, + "loss": 0.6814, + "step": 2484 + }, + { + "epoch": 3.8940501043841333, + "grad_norm": 1.2072421312332153, + "learning_rate": 1.4373058395934136e-06, + "loss": 0.6178, + "step": 2485 + }, + { + "epoch": 3.895615866388309, + "grad_norm": 1.3510164022445679, + "learning_rate": 1.4353992736125396e-06, + "loss": 0.6775, + "step": 2486 + }, + { + "epoch": 3.8971816283924845, + "grad_norm": 1.2395039796829224, + "learning_rate": 1.4334934636540782e-06, + "loss": 0.5773, + "step": 2487 + }, + { + "epoch": 3.8987473903966596, + "grad_norm": 1.208178162574768, + "learning_rate": 1.4315884110714337e-06, + "loss": 0.5234, + "step": 2488 + }, + { + "epoch": 3.900313152400835, + "grad_norm": 1.2231117486953735, + "learning_rate": 1.429684117217473e-06, + "loss": 0.6405, + "step": 2489 + }, + { + "epoch": 3.9018789144050103, + "grad_norm": 1.1081833839416504, + "learning_rate": 1.4277805834445207e-06, + "loss": 0.6349, + "step": 2490 + }, + { + "epoch": 3.903444676409186, + "grad_norm": 1.3395150899887085, + "learning_rate": 1.4258778111043674e-06, + "loss": 0.6236, + "step": 2491 + }, + { + "epoch": 3.905010438413361, + "grad_norm": 1.2174855470657349, + "learning_rate": 1.4239758015482607e-06, + "loss": 0.667, + "step": 2492 + }, + { + "epoch": 3.9065762004175366, + "grad_norm": 1.3547643423080444, + "learning_rate": 1.4220745561269022e-06, + "loss": 0.6654, + "step": 2493 + }, + { + "epoch": 3.9081419624217117, + "grad_norm": 1.1408005952835083, + "learning_rate": 1.4201740761904591e-06, + "loss": 0.638, + "step": 2494 + }, + { + "epoch": 3.9097077244258873, + "grad_norm": 1.299989938735962, + "learning_rate": 1.418274363088547e-06, + "loss": 0.5875, + "step": 2495 + }, + { + "epoch": 3.911273486430063, + "grad_norm": 1.1767139434814453, + "learning_rate": 1.4163754181702408e-06, + "loss": 0.6723, + "step": 2496 + }, + { + "epoch": 3.912839248434238, + "grad_norm": 1.2104045152664185, + "learning_rate": 1.414477242784072e-06, + "loss": 0.5944, + "step": 2497 + }, + { + "epoch": 3.914405010438413, + "grad_norm": 1.2391446828842163, + "learning_rate": 1.4125798382780207e-06, + "loss": 0.6747, + "step": 2498 + }, + { + "epoch": 3.9159707724425887, + "grad_norm": 1.1312291622161865, + "learning_rate": 1.4106832059995224e-06, + "loss": 0.6705, + "step": 2499 + }, + { + "epoch": 3.9175365344467643, + "grad_norm": 1.2023922204971313, + "learning_rate": 1.4087873472954638e-06, + "loss": 0.6417, + "step": 2500 + }, + { + "epoch": 3.9191022964509394, + "grad_norm": 1.2477304935455322, + "learning_rate": 1.4068922635121828e-06, + "loss": 0.6471, + "step": 2501 + }, + { + "epoch": 3.9206680584551146, + "grad_norm": 1.140338659286499, + "learning_rate": 1.4049979559954666e-06, + "loss": 0.6042, + "step": 2502 + }, + { + "epoch": 3.92223382045929, + "grad_norm": 1.1643264293670654, + "learning_rate": 1.4031044260905486e-06, + "loss": 0.7038, + "step": 2503 + }, + { + "epoch": 3.9237995824634657, + "grad_norm": 1.2577917575836182, + "learning_rate": 1.4012116751421145e-06, + "loss": 0.6826, + "step": 2504 + }, + { + "epoch": 3.925365344467641, + "grad_norm": 1.1324503421783447, + "learning_rate": 1.3993197044942958e-06, + "loss": 0.6337, + "step": 2505 + }, + { + "epoch": 3.9269311064718164, + "grad_norm": 1.3015128374099731, + "learning_rate": 1.3974285154906647e-06, + "loss": 0.6443, + "step": 2506 + }, + { + "epoch": 3.9284968684759916, + "grad_norm": 1.139436960220337, + "learning_rate": 1.3955381094742464e-06, + "loss": 0.6569, + "step": 2507 + }, + { + "epoch": 3.930062630480167, + "grad_norm": 1.2854009866714478, + "learning_rate": 1.393648487787504e-06, + "loss": 0.6191, + "step": 2508 + }, + { + "epoch": 3.9316283924843423, + "grad_norm": 1.2040328979492188, + "learning_rate": 1.3917596517723441e-06, + "loss": 0.6355, + "step": 2509 + }, + { + "epoch": 3.933194154488518, + "grad_norm": 1.2754182815551758, + "learning_rate": 1.3898716027701214e-06, + "loss": 0.658, + "step": 2510 + }, + { + "epoch": 3.934759916492693, + "grad_norm": 1.2027299404144287, + "learning_rate": 1.387984342121623e-06, + "loss": 0.5478, + "step": 2511 + }, + { + "epoch": 3.9363256784968685, + "grad_norm": 1.1748281717300415, + "learning_rate": 1.3860978711670819e-06, + "loss": 0.6427, + "step": 2512 + }, + { + "epoch": 3.937891440501044, + "grad_norm": 1.209522008895874, + "learning_rate": 1.3842121912461686e-06, + "loss": 0.6234, + "step": 2513 + }, + { + "epoch": 3.9394572025052192, + "grad_norm": 1.1207284927368164, + "learning_rate": 1.382327303697992e-06, + "loss": 0.6092, + "step": 2514 + }, + { + "epoch": 3.9410229645093944, + "grad_norm": 1.2785614728927612, + "learning_rate": 1.3804432098610982e-06, + "loss": 0.6507, + "step": 2515 + }, + { + "epoch": 3.94258872651357, + "grad_norm": 1.2241697311401367, + "learning_rate": 1.37855991107347e-06, + "loss": 0.5863, + "step": 2516 + }, + { + "epoch": 3.9441544885177455, + "grad_norm": 1.1810356378555298, + "learning_rate": 1.3766774086725246e-06, + "loss": 0.6718, + "step": 2517 + }, + { + "epoch": 3.9457202505219207, + "grad_norm": 1.1314932107925415, + "learning_rate": 1.3747957039951152e-06, + "loss": 0.6726, + "step": 2518 + }, + { + "epoch": 3.947286012526096, + "grad_norm": 1.2174315452575684, + "learning_rate": 1.3729147983775272e-06, + "loss": 0.6201, + "step": 2519 + }, + { + "epoch": 3.9488517745302714, + "grad_norm": 1.2153561115264893, + "learning_rate": 1.371034693155479e-06, + "loss": 0.6499, + "step": 2520 + }, + { + "epoch": 3.950417536534447, + "grad_norm": 1.219773292541504, + "learning_rate": 1.3691553896641213e-06, + "loss": 0.6046, + "step": 2521 + }, + { + "epoch": 3.951983298538622, + "grad_norm": 1.284940481185913, + "learning_rate": 1.367276889238034e-06, + "loss": 0.6899, + "step": 2522 + }, + { + "epoch": 3.9535490605427976, + "grad_norm": 1.356799602508545, + "learning_rate": 1.3653991932112287e-06, + "loss": 0.6861, + "step": 2523 + }, + { + "epoch": 3.9551148225469728, + "grad_norm": 1.166818618774414, + "learning_rate": 1.3635223029171418e-06, + "loss": 0.5631, + "step": 2524 + }, + { + "epoch": 3.9566805845511483, + "grad_norm": 1.2792567014694214, + "learning_rate": 1.3616462196886427e-06, + "loss": 0.6471, + "step": 2525 + }, + { + "epoch": 3.9582463465553235, + "grad_norm": 1.2961795330047607, + "learning_rate": 1.3597709448580257e-06, + "loss": 0.6074, + "step": 2526 + }, + { + "epoch": 3.959812108559499, + "grad_norm": 1.2090468406677246, + "learning_rate": 1.3578964797570077e-06, + "loss": 0.6225, + "step": 2527 + }, + { + "epoch": 3.961377870563674, + "grad_norm": 1.1482014656066895, + "learning_rate": 1.3560228257167365e-06, + "loss": 0.5968, + "step": 2528 + }, + { + "epoch": 3.9629436325678498, + "grad_norm": 1.2505629062652588, + "learning_rate": 1.354149984067779e-06, + "loss": 0.6198, + "step": 2529 + }, + { + "epoch": 3.964509394572025, + "grad_norm": 1.1473637819290161, + "learning_rate": 1.3522779561401262e-06, + "loss": 0.5933, + "step": 2530 + }, + { + "epoch": 3.9660751565762005, + "grad_norm": 1.2111499309539795, + "learning_rate": 1.3504067432631954e-06, + "loss": 0.6173, + "step": 2531 + }, + { + "epoch": 3.9676409185803756, + "grad_norm": 1.2065558433532715, + "learning_rate": 1.3485363467658186e-06, + "loss": 0.6819, + "step": 2532 + }, + { + "epoch": 3.969206680584551, + "grad_norm": 1.3822040557861328, + "learning_rate": 1.3466667679762525e-06, + "loss": 0.6258, + "step": 2533 + }, + { + "epoch": 3.9707724425887267, + "grad_norm": 1.2183210849761963, + "learning_rate": 1.3447980082221719e-06, + "loss": 0.6501, + "step": 2534 + }, + { + "epoch": 3.972338204592902, + "grad_norm": 1.1406798362731934, + "learning_rate": 1.342930068830669e-06, + "loss": 0.5535, + "step": 2535 + }, + { + "epoch": 3.973903966597077, + "grad_norm": 1.2553014755249023, + "learning_rate": 1.3410629511282553e-06, + "loss": 0.6809, + "step": 2536 + }, + { + "epoch": 3.9754697286012526, + "grad_norm": 1.2471221685409546, + "learning_rate": 1.3391966564408571e-06, + "loss": 0.635, + "step": 2537 + }, + { + "epoch": 3.977035490605428, + "grad_norm": 1.318433165550232, + "learning_rate": 1.3373311860938177e-06, + "loss": 0.6571, + "step": 2538 + }, + { + "epoch": 3.9786012526096033, + "grad_norm": 1.2130142450332642, + "learning_rate": 1.335466541411893e-06, + "loss": 0.6643, + "step": 2539 + }, + { + "epoch": 3.980167014613779, + "grad_norm": 1.1460458040237427, + "learning_rate": 1.3336027237192551e-06, + "loss": 0.5837, + "step": 2540 + }, + { + "epoch": 3.981732776617954, + "grad_norm": 1.1360803842544556, + "learning_rate": 1.3317397343394879e-06, + "loss": 0.6432, + "step": 2541 + }, + { + "epoch": 3.9832985386221296, + "grad_norm": 1.1411229372024536, + "learning_rate": 1.3298775745955833e-06, + "loss": 0.5568, + "step": 2542 + }, + { + "epoch": 3.9848643006263047, + "grad_norm": 1.3322495222091675, + "learning_rate": 1.328016245809951e-06, + "loss": 0.7113, + "step": 2543 + }, + { + "epoch": 3.9864300626304803, + "grad_norm": 1.1256479024887085, + "learning_rate": 1.3261557493044067e-06, + "loss": 0.5875, + "step": 2544 + }, + { + "epoch": 3.9879958246346554, + "grad_norm": 1.1804993152618408, + "learning_rate": 1.3242960864001724e-06, + "loss": 0.6372, + "step": 2545 + }, + { + "epoch": 3.989561586638831, + "grad_norm": 1.1208686828613281, + "learning_rate": 1.3224372584178855e-06, + "loss": 0.4952, + "step": 2546 + }, + { + "epoch": 3.991127348643006, + "grad_norm": 1.2875077724456787, + "learning_rate": 1.3205792666775824e-06, + "loss": 0.6465, + "step": 2547 + }, + { + "epoch": 3.9926931106471817, + "grad_norm": 1.2553578615188599, + "learning_rate": 1.3187221124987107e-06, + "loss": 0.6053, + "step": 2548 + }, + { + "epoch": 3.994258872651357, + "grad_norm": 1.1742136478424072, + "learning_rate": 1.3168657972001214e-06, + "loss": 0.6711, + "step": 2549 + }, + { + "epoch": 3.9958246346555324, + "grad_norm": 1.184242844581604, + "learning_rate": 1.3150103221000709e-06, + "loss": 0.6405, + "step": 2550 + }, + { + "epoch": 3.997390396659708, + "grad_norm": 1.1263542175292969, + "learning_rate": 1.313155688516218e-06, + "loss": 0.5631, + "step": 2551 + }, + { + "epoch": 3.998956158663883, + "grad_norm": 1.2463663816452026, + "learning_rate": 1.3113018977656238e-06, + "loss": 0.5974, + "step": 2552 + } + ], + "logging_steps": 1, + "max_steps": 3828, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 638, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 6.448971686829097e+18, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-2552/training_args.bin b/checkpoint-2552/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..07c8c9aae2a66d9fb6a8b65759b1cd86d2f7d08b --- /dev/null +++ b/checkpoint-2552/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81aeca374ba1e1a5384aca380c5dd9da735cc6d6164e0b5b0681e227cf61f41c +size 7992 diff --git a/checkpoint-2552/zero_to_fp32.py b/checkpoint-2552/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-2552/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-3190/README.md b/checkpoint-3190/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f07164552af92bf3072e4e4326b5e572dea01ae8 --- /dev/null +++ b/checkpoint-3190/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-Nano-8B-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 0.14.0 \ No newline at end of file diff --git a/checkpoint-3190/adapter_config.json b/checkpoint-3190/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3db75f3ecbeb5f4990760e56c7957796d83680c5 --- /dev/null +++ b/checkpoint-3190/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1", + "bias": "none", + "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": [ + "q_proj", + "down_proj", + "o_proj", + "v_proj", + "k_proj", + "gate_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-3190/adapter_model.safetensors b/checkpoint-3190/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1e514658834dd796174255a5ef942358bd569ef0 --- /dev/null +++ b/checkpoint-3190/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3bf2a31625cc9ec30596a80b5e9847d8248c3041e8e1f9fcba6dca57b2777bf9 +size 3443586272 diff --git a/checkpoint-3190/global_step3191/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-3190/global_step3191/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..abffc54ca7c0bd257048e4c679355baab127dd68 --- /dev/null +++ b/checkpoint-3190/global_step3191/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e7065306a35b68d8431598195ce714c602f68073c8b4dadcd31e11890b9b47a +size 20661195036 diff --git a/checkpoint-3190/global_step3191/mp_rank_00_model_states.pt b/checkpoint-3190/global_step3191/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..c8f8b4f37e1fa353495e72ac51f9a47d7a3e8cf7 --- /dev/null +++ b/checkpoint-3190/global_step3191/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8210864b3cc01cb979e11b3172b621e163e9f726aa4db0a137aef0fe2274222c +size 3555326777 diff --git a/checkpoint-3190/latest b/checkpoint-3190/latest new file mode 100644 index 0000000000000000000000000000000000000000..84195709a61927cab0aeadad35ca79343dbb2d7d --- /dev/null +++ b/checkpoint-3190/latest @@ -0,0 +1 @@ +global_step3191 \ No newline at end of file diff --git a/checkpoint-3190/rng_state.pth b/checkpoint-3190/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..4449972ad8cdce622fdc0332d0258a23e59fc34b --- /dev/null +++ b/checkpoint-3190/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8fd8e6188aa63711eadde587a8c9a996106de0c8617ca553131f9b71727d3c74 +size 14244 diff --git a/checkpoint-3190/scheduler.pt b/checkpoint-3190/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..424aa871addd75d6c871ece610718efd57089472 --- /dev/null +++ b/checkpoint-3190/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4eb7eebbdf158333a141f17fe04b9cabbfa396055322564e842ed01d5eee7ae6 +size 1064 diff --git a/checkpoint-3190/special_tokens_map.json b/checkpoint-3190/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-3190/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-3190/tokenizer.json b/checkpoint-3190/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-3190/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-3190/tokenizer_config.json b/checkpoint-3190/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b480d149dc274c0a23db079bfd38032d90082e49 --- /dev/null +++ b/checkpoint-3190/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": "{%- if messages[0]['role'] == 'system' -%}{%- set system_message = messages[0]['content'] | trim -%}{%- set messages = messages[1:] -%}{%- else -%}{%- set system_message = '' -%}{%- endif -%}{%- if tools is not none -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message -}} {{- '\n\n' if system_message else '' -}} {{- '[' -}} {% for t in tools %}{{- (t.function if t.function is defined else t) | tojson() -}}{{- ', ' if not loop.last else '' -}}{%- endfor -%} {{- ']' -}} {{- '<|eot_id|>' -}}{%- else -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message + '<|eot_id|>' -}}{%- endif -%}{%- for message in messages -%}{%- if (message['role'] in ['user', 'tool']) != (loop.index0 % 2 == 0) -%}{{- raise_exception('Conversation roles must alternate between user/tool and assistant') -}}{%- elif message['role'] == 'user' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- elif message['role'] == 'tool' -%}{%- set tool_response = '[' + message['content'] | trim + ']' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + tool_response + '<|eot_id|>' -}}{%- elif message['role'] == 'assistant' and message.get('tool_calls') is not none -%}{%- set tool_calls = message['tool_calls'] -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + '[' -}}{%- for tool_call in tool_calls -%}{{ '{' + '\"name\": \"' + tool_call.function.name + '\", \"arguments\": ' + tool_call.function.arguments | tojson + '}' }}{%- if not loop.last -%}{{ ', ' }}{%- else -%}{{ ']' + '<|eot_id|>' }}{%- endif -%}{%- endfor -%}{%- elif message['role'] == 'assistant' -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- endif -%}{%- 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-3190/trainer_state.json b/checkpoint-3190/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..5afe0967a569591d4d83ebd5962dee1fedb36e50 --- /dev/null +++ b/checkpoint-3190/trainer_state.json @@ -0,0 +1,22364 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 4.9989561586638835, + "eval_steps": 500, + "global_step": 3190, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0015657620041753654, + "grad_norm": 658.1201782226562, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.3039, + "step": 1 + }, + { + "epoch": 0.003131524008350731, + "grad_norm": 749.5796508789062, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.4164, + "step": 2 + }, + { + "epoch": 0.004697286012526096, + "grad_norm": 394.2350769042969, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.3367, + "step": 3 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 384.08319091796875, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.3192, + "step": 4 + }, + { + "epoch": 0.007828810020876827, + "grad_norm": 376.9790954589844, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.2756, + "step": 5 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 175.25262451171875, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.3992, + "step": 6 + }, + { + "epoch": 0.010960334029227558, + "grad_norm": 100.1114730834961, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.2348, + "step": 7 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 131.44432067871094, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.2685, + "step": 8 + }, + { + "epoch": 0.014091858037578288, + "grad_norm": 67.03341674804688, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.3272, + "step": 9 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 32.0869140625, + "learning_rate": 5.000000000000001e-07, + "loss": 1.3458, + "step": 10 + }, + { + "epoch": 0.01722338204592902, + "grad_norm": 17.701913833618164, + "learning_rate": 5.5e-07, + "loss": 1.3608, + "step": 11 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 9.164765357971191, + "learning_rate": 6.000000000000001e-07, + "loss": 1.2961, + "step": 12 + }, + { + "epoch": 0.02035490605427975, + "grad_norm": 4.315349578857422, + "learning_rate": 6.5e-07, + "loss": 1.2069, + "step": 13 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 6.652385234832764, + "learning_rate": 7.000000000000001e-07, + "loss": 1.2588, + "step": 14 + }, + { + "epoch": 0.02348643006263048, + "grad_norm": 3.3322548866271973, + "learning_rate": 7.5e-07, + "loss": 1.2505, + "step": 15 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 9.599331855773926, + "learning_rate": 8.000000000000001e-07, + "loss": 1.1343, + "step": 16 + }, + { + "epoch": 0.02661795407098121, + "grad_norm": 10.33918285369873, + "learning_rate": 8.500000000000001e-07, + "loss": 1.3334, + "step": 17 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 9.453723907470703, + "learning_rate": 9.000000000000001e-07, + "loss": 1.3099, + "step": 18 + }, + { + "epoch": 0.029749478079331943, + "grad_norm": 7.075769901275635, + "learning_rate": 9.500000000000001e-07, + "loss": 1.1121, + "step": 19 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 7.770864009857178, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.1836, + "step": 20 + }, + { + "epoch": 0.03288100208768267, + "grad_norm": 9.582944869995117, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.1939, + "step": 21 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 9.496583938598633, + "learning_rate": 1.1e-06, + "loss": 1.2395, + "step": 22 + }, + { + "epoch": 0.0360125260960334, + "grad_norm": 5.912876129150391, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.1769, + "step": 23 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 2.6761083602905273, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.204, + "step": 24 + }, + { + "epoch": 0.03914405010438413, + "grad_norm": 11.136463165283203, + "learning_rate": 1.25e-06, + "loss": 1.2271, + "step": 25 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 23.912893295288086, + "learning_rate": 1.3e-06, + "loss": 1.1771, + "step": 26 + }, + { + "epoch": 0.042275574112734866, + "grad_norm": 57.08906555175781, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.2253, + "step": 27 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 36.11408615112305, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.1014, + "step": 28 + }, + { + "epoch": 0.04540709812108559, + "grad_norm": 51.39445495605469, + "learning_rate": 1.45e-06, + "loss": 1.1978, + "step": 29 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 27.394500732421875, + "learning_rate": 1.5e-06, + "loss": 1.1155, + "step": 30 + }, + { + "epoch": 0.048538622129436326, + "grad_norm": 6.844716548919678, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.1798, + "step": 31 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 2.397536277770996, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.1846, + "step": 32 + }, + { + "epoch": 0.05167014613778706, + "grad_norm": 7.821156024932861, + "learning_rate": 1.6500000000000003e-06, + "loss": 1.1312, + "step": 33 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 5.455922603607178, + "learning_rate": 1.7000000000000002e-06, + "loss": 1.2153, + "step": 34 + }, + { + "epoch": 0.054801670146137786, + "grad_norm": 5.377617835998535, + "learning_rate": 1.75e-06, + "loss": 1.1292, + "step": 35 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 7.172031879425049, + "learning_rate": 1.8000000000000001e-06, + "loss": 1.1459, + "step": 36 + }, + { + "epoch": 0.05793319415448852, + "grad_norm": 2.1464905738830566, + "learning_rate": 1.85e-06, + "loss": 1.1649, + "step": 37 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 2.7782726287841797, + "learning_rate": 1.9000000000000002e-06, + "loss": 1.159, + "step": 38 + }, + { + "epoch": 0.061064718162839246, + "grad_norm": 20.818374633789062, + "learning_rate": 1.9500000000000004e-06, + "loss": 1.0174, + "step": 39 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 22.776456832885742, + "learning_rate": 2.0000000000000003e-06, + "loss": 1.1425, + "step": 40 + }, + { + "epoch": 0.06419624217118998, + "grad_norm": 33.601287841796875, + "learning_rate": 2.05e-06, + "loss": 1.0498, + "step": 41 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.884684562683105, + "learning_rate": 2.1000000000000002e-06, + "loss": 1.1231, + "step": 42 + }, + { + "epoch": 0.06732776617954071, + "grad_norm": 1.951470971107483, + "learning_rate": 2.15e-06, + "loss": 1.1256, + "step": 43 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 2.584127426147461, + "learning_rate": 2.2e-06, + "loss": 1.1242, + "step": 44 + }, + { + "epoch": 0.07045929018789145, + "grad_norm": 6.982508659362793, + "learning_rate": 2.25e-06, + "loss": 1.0769, + "step": 45 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 7.940657138824463, + "learning_rate": 2.3000000000000004e-06, + "loss": 1.0516, + "step": 46 + }, + { + "epoch": 0.07359081419624217, + "grad_norm": 5.456567287445068, + "learning_rate": 2.35e-06, + "loss": 0.9894, + "step": 47 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 4.623636722564697, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9912, + "step": 48 + }, + { + "epoch": 0.0767223382045929, + "grad_norm": 2.182708978652954, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.1303, + "step": 49 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 2.388091802597046, + "learning_rate": 2.5e-06, + "loss": 1.0725, + "step": 50 + }, + { + "epoch": 0.07985386221294363, + "grad_norm": 2.2816522121429443, + "learning_rate": 2.55e-06, + "loss": 1.0895, + "step": 51 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 3.8391623497009277, + "learning_rate": 2.6e-06, + "loss": 1.0587, + "step": 52 + }, + { + "epoch": 0.08298538622129437, + "grad_norm": 3.700389862060547, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9795, + "step": 53 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 3.180649995803833, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.0434, + "step": 54 + }, + { + "epoch": 0.0861169102296451, + "grad_norm": 3.790774345397949, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9652, + "step": 55 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 1.8411438465118408, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.947, + "step": 56 + }, + { + "epoch": 0.08924843423799582, + "grad_norm": 2.100008010864258, + "learning_rate": 2.85e-06, + "loss": 1.0129, + "step": 57 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 1.103857159614563, + "learning_rate": 2.9e-06, + "loss": 0.9993, + "step": 58 + }, + { + "epoch": 0.09237995824634655, + "grad_norm": 1.232283353805542, + "learning_rate": 2.95e-06, + "loss": 0.9988, + "step": 59 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 2.4731297492980957, + "learning_rate": 3e-06, + "loss": 0.9873, + "step": 60 + }, + { + "epoch": 0.09551148225469729, + "grad_norm": 1.1080410480499268, + "learning_rate": 3.05e-06, + "loss": 1.0657, + "step": 61 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 1.1155121326446533, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.974, + "step": 62 + }, + { + "epoch": 0.09864300626304802, + "grad_norm": 1.763231873512268, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.0215, + "step": 63 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 1.1961321830749512, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.9955, + "step": 64 + }, + { + "epoch": 0.10177453027139875, + "grad_norm": 1.377040982246399, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.9834, + "step": 65 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 1.373260498046875, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.984, + "step": 66 + }, + { + "epoch": 0.10490605427974947, + "grad_norm": 2.034064292907715, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.973, + "step": 67 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 1.41798734664917, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.925, + "step": 68 + }, + { + "epoch": 0.1080375782881002, + "grad_norm": 1.7425439357757568, + "learning_rate": 3.45e-06, + "loss": 0.9884, + "step": 69 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 2.369657278060913, + "learning_rate": 3.5e-06, + "loss": 1.0048, + "step": 70 + }, + { + "epoch": 0.11116910229645094, + "grad_norm": 1.4699137210845947, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.9155, + "step": 71 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 1.1473263502120972, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8445, + "step": 72 + }, + { + "epoch": 0.11430062630480167, + "grad_norm": 2.315568685531616, + "learning_rate": 3.65e-06, + "loss": 0.977, + "step": 73 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 1.8554916381835938, + "learning_rate": 3.7e-06, + "loss": 0.841, + "step": 74 + }, + { + "epoch": 0.1174321503131524, + "grad_norm": 1.1517397165298462, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.9478, + "step": 75 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 2.799316883087158, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.9873, + "step": 76 + }, + { + "epoch": 0.12056367432150313, + "grad_norm": 1.11930251121521, + "learning_rate": 3.85e-06, + "loss": 0.9214, + "step": 77 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 1.0082950592041016, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9115, + "step": 78 + }, + { + "epoch": 0.12369519832985386, + "grad_norm": 1.5534541606903076, + "learning_rate": 3.95e-06, + "loss": 1.04, + "step": 79 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 1.3469066619873047, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9091, + "step": 80 + }, + { + "epoch": 0.1268267223382046, + "grad_norm": 1.1149636507034302, + "learning_rate": 4.05e-06, + "loss": 0.8829, + "step": 81 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 2.5933806896209717, + "learning_rate": 4.1e-06, + "loss": 0.9343, + "step": 82 + }, + { + "epoch": 0.1299582463465553, + "grad_norm": 1.980901837348938, + "learning_rate": 4.15e-06, + "loss": 0.9028, + "step": 83 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 0.8476768732070923, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.9038, + "step": 84 + }, + { + "epoch": 0.13308977035490605, + "grad_norm": 1.1239815950393677, + "learning_rate": 4.25e-06, + "loss": 1.0359, + "step": 85 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 1.0159156322479248, + "learning_rate": 4.3e-06, + "loss": 0.8705, + "step": 86 + }, + { + "epoch": 0.13622129436325678, + "grad_norm": 2.180856466293335, + "learning_rate": 4.350000000000001e-06, + "loss": 0.9834, + "step": 87 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 1.5725276470184326, + "learning_rate": 4.4e-06, + "loss": 0.9449, + "step": 88 + }, + { + "epoch": 0.1393528183716075, + "grad_norm": 1.5560917854309082, + "learning_rate": 4.450000000000001e-06, + "loss": 1.0109, + "step": 89 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 1.1096960306167603, + "learning_rate": 4.5e-06, + "loss": 0.9081, + "step": 90 + }, + { + "epoch": 0.14248434237995825, + "grad_norm": 1.2525160312652588, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.9907, + "step": 91 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 0.9198770523071289, + "learning_rate": 4.600000000000001e-06, + "loss": 0.9266, + "step": 92 + }, + { + "epoch": 0.14561586638830898, + "grad_norm": 0.992057740688324, + "learning_rate": 4.65e-06, + "loss": 0.9556, + "step": 93 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 1.00990891456604, + "learning_rate": 4.7e-06, + "loss": 0.9712, + "step": 94 + }, + { + "epoch": 0.1487473903966597, + "grad_norm": 1.1176848411560059, + "learning_rate": 4.75e-06, + "loss": 0.9604, + "step": 95 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 0.9323613047599792, + "learning_rate": 4.800000000000001e-06, + "loss": 0.9708, + "step": 96 + }, + { + "epoch": 0.15187891440501045, + "grad_norm": 0.9800642728805542, + "learning_rate": 4.85e-06, + "loss": 0.9868, + "step": 97 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 1.487841010093689, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8679, + "step": 98 + }, + { + "epoch": 0.15501043841336118, + "grad_norm": 0.95517897605896, + "learning_rate": 4.95e-06, + "loss": 0.7949, + "step": 99 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 0.9686336517333984, + "learning_rate": 5e-06, + "loss": 0.8655, + "step": 100 + }, + { + "epoch": 0.1581419624217119, + "grad_norm": 1.9481035470962524, + "learning_rate": 4.999999112316954e-06, + "loss": 0.9328, + "step": 101 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 1.272631287574768, + "learning_rate": 4.9999964492684465e-06, + "loss": 0.7869, + "step": 102 + }, + { + "epoch": 0.16127348643006262, + "grad_norm": 1.1117815971374512, + "learning_rate": 4.999992010856368e-06, + "loss": 0.949, + "step": 103 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 0.9681039452552795, + "learning_rate": 4.999985797083871e-06, + "loss": 1.0396, + "step": 104 + }, + { + "epoch": 0.16440501043841335, + "grad_norm": 0.9797878265380859, + "learning_rate": 4.999977807955368e-06, + "loss": 0.9393, + "step": 105 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 1.0239158868789673, + "learning_rate": 4.999968043476532e-06, + "loss": 0.8766, + "step": 106 + }, + { + "epoch": 0.16753653444676408, + "grad_norm": 0.9314751029014587, + "learning_rate": 4.999956503654299e-06, + "loss": 0.8545, + "step": 107 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 1.1627179384231567, + "learning_rate": 4.999943188496862e-06, + "loss": 0.9241, + "step": 108 + }, + { + "epoch": 0.17066805845511482, + "grad_norm": 1.022752285003662, + "learning_rate": 4.9999280980136765e-06, + "loss": 0.9662, + "step": 109 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 1.0513653755187988, + "learning_rate": 4.99991123221546e-06, + "loss": 0.9054, + "step": 110 + }, + { + "epoch": 0.17379958246346555, + "grad_norm": 0.9488444328308105, + "learning_rate": 4.99989259111419e-06, + "loss": 0.8921, + "step": 111 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 0.9934154748916626, + "learning_rate": 4.999872174723104e-06, + "loss": 0.9848, + "step": 112 + }, + { + "epoch": 0.17693110647181629, + "grad_norm": 1.2035515308380127, + "learning_rate": 4.9998499830567e-06, + "loss": 0.9833, + "step": 113 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 1.027538537979126, + "learning_rate": 4.999826016130739e-06, + "loss": 1.0089, + "step": 114 + }, + { + "epoch": 0.18006263048016702, + "grad_norm": 1.0635300874710083, + "learning_rate": 4.999800273962238e-06, + "loss": 0.8863, + "step": 115 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 0.9978796243667603, + "learning_rate": 4.999772756569482e-06, + "loss": 0.899, + "step": 116 + }, + { + "epoch": 0.18319415448851775, + "grad_norm": 0.9971838593482971, + "learning_rate": 4.999743463972008e-06, + "loss": 0.9515, + "step": 117 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 1.2436699867248535, + "learning_rate": 4.999712396190622e-06, + "loss": 0.9101, + "step": 118 + }, + { + "epoch": 0.18632567849686849, + "grad_norm": 0.9573557376861572, + "learning_rate": 4.999679553247383e-06, + "loss": 0.8817, + "step": 119 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 0.9105668067932129, + "learning_rate": 4.9996449351656165e-06, + "loss": 1.0077, + "step": 120 + }, + { + "epoch": 0.18945720250521922, + "grad_norm": 0.8741400241851807, + "learning_rate": 4.999608541969905e-06, + "loss": 0.912, + "step": 121 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 1.1172688007354736, + "learning_rate": 4.999570373686095e-06, + "loss": 0.8062, + "step": 122 + }, + { + "epoch": 0.19258872651356992, + "grad_norm": 0.940539538860321, + "learning_rate": 4.9995304303412905e-06, + "loss": 0.8944, + "step": 123 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 2.1082208156585693, + "learning_rate": 4.999488711963857e-06, + "loss": 0.9668, + "step": 124 + }, + { + "epoch": 0.19572025052192066, + "grad_norm": 1.0358761548995972, + "learning_rate": 4.99944521858342e-06, + "loss": 0.9032, + "step": 125 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 2.0804238319396973, + "learning_rate": 4.999399950230867e-06, + "loss": 0.9819, + "step": 126 + }, + { + "epoch": 0.1988517745302714, + "grad_norm": 1.0986278057098389, + "learning_rate": 4.9993529069383465e-06, + "loss": 0.9034, + "step": 127 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 1.2374722957611084, + "learning_rate": 4.999304088739263e-06, + "loss": 0.9255, + "step": 128 + }, + { + "epoch": 0.20198329853862212, + "grad_norm": 1.0443146228790283, + "learning_rate": 4.999253495668287e-06, + "loss": 0.9352, + "step": 129 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 1.009800910949707, + "learning_rate": 4.999201127761346e-06, + "loss": 0.8324, + "step": 130 + }, + { + "epoch": 0.20511482254697286, + "grad_norm": 0.892861008644104, + "learning_rate": 4.9991469850556295e-06, + "loss": 0.886, + "step": 131 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 1.1675922870635986, + "learning_rate": 4.999091067589587e-06, + "loss": 0.967, + "step": 132 + }, + { + "epoch": 0.2082463465553236, + "grad_norm": 1.0169954299926758, + "learning_rate": 4.999033375402926e-06, + "loss": 0.8498, + "step": 133 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 1.0006227493286133, + "learning_rate": 4.99897390853662e-06, + "loss": 0.8481, + "step": 134 + }, + { + "epoch": 0.21137787056367432, + "grad_norm": 0.9049347639083862, + "learning_rate": 4.9989126670328945e-06, + "loss": 0.9186, + "step": 135 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 1.2151603698730469, + "learning_rate": 4.9988496509352444e-06, + "loss": 0.9067, + "step": 136 + }, + { + "epoch": 0.21450939457202506, + "grad_norm": 1.0112006664276123, + "learning_rate": 4.998784860288417e-06, + "loss": 0.8843, + "step": 137 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 0.9316662549972534, + "learning_rate": 4.9987182951384264e-06, + "loss": 0.8635, + "step": 138 + }, + { + "epoch": 0.2176409185803758, + "grad_norm": 0.942743718624115, + "learning_rate": 4.9986499555325406e-06, + "loss": 0.8481, + "step": 139 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 1.0298817157745361, + "learning_rate": 4.998579841519292e-06, + "loss": 0.865, + "step": 140 + }, + { + "epoch": 0.22077244258872653, + "grad_norm": 1.1556729078292847, + "learning_rate": 4.998507953148472e-06, + "loss": 0.8731, + "step": 141 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 0.997958779335022, + "learning_rate": 4.9984342904711315e-06, + "loss": 0.7949, + "step": 142 + }, + { + "epoch": 0.22390396659707723, + "grad_norm": 0.9565998315811157, + "learning_rate": 4.998358853539582e-06, + "loss": 0.8875, + "step": 143 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 0.9667196273803711, + "learning_rate": 4.998281642407394e-06, + "loss": 0.8537, + "step": 144 + }, + { + "epoch": 0.22703549060542796, + "grad_norm": 0.9701303839683533, + "learning_rate": 4.998202657129401e-06, + "loss": 0.873, + "step": 145 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 0.9936159253120422, + "learning_rate": 4.998121897761692e-06, + "loss": 0.8915, + "step": 146 + }, + { + "epoch": 0.2301670146137787, + "grad_norm": 1.5054311752319336, + "learning_rate": 4.998039364361617e-06, + "loss": 0.9944, + "step": 147 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 1.5895891189575195, + "learning_rate": 4.99795505698779e-06, + "loss": 0.9282, + "step": 148 + }, + { + "epoch": 0.23329853862212943, + "grad_norm": 1.0141469240188599, + "learning_rate": 4.9978689757000785e-06, + "loss": 0.9166, + "step": 149 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 1.1720311641693115, + "learning_rate": 4.997781120559615e-06, + "loss": 0.9404, + "step": 150 + }, + { + "epoch": 0.23643006263048016, + "grad_norm": 0.9536433815956116, + "learning_rate": 4.997691491628789e-06, + "loss": 0.9339, + "step": 151 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 0.9407872557640076, + "learning_rate": 4.997600088971249e-06, + "loss": 0.9019, + "step": 152 + }, + { + "epoch": 0.2395615866388309, + "grad_norm": 0.9202889204025269, + "learning_rate": 4.997506912651906e-06, + "loss": 0.9598, + "step": 153 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 1.0630513429641724, + "learning_rate": 4.997411962736927e-06, + "loss": 0.8554, + "step": 154 + }, + { + "epoch": 0.24269311064718163, + "grad_norm": 0.8378801941871643, + "learning_rate": 4.997315239293742e-06, + "loss": 0.8583, + "step": 155 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 0.9442815184593201, + "learning_rate": 4.997216742391038e-06, + "loss": 0.8518, + "step": 156 + }, + { + "epoch": 0.24582463465553236, + "grad_norm": 0.9273308515548706, + "learning_rate": 4.997116472098763e-06, + "loss": 0.8942, + "step": 157 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 0.9028997421264648, + "learning_rate": 4.997014428488121e-06, + "loss": 0.8592, + "step": 158 + }, + { + "epoch": 0.2489561586638831, + "grad_norm": 0.9877567887306213, + "learning_rate": 4.996910611631582e-06, + "loss": 0.9316, + "step": 159 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 0.961592435836792, + "learning_rate": 4.996805021602868e-06, + "loss": 0.8875, + "step": 160 + }, + { + "epoch": 0.2520876826722338, + "grad_norm": 0.9923007488250732, + "learning_rate": 4.996697658476965e-06, + "loss": 0.8554, + "step": 161 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 0.956763744354248, + "learning_rate": 4.996588522330114e-06, + "loss": 0.9381, + "step": 162 + }, + { + "epoch": 0.25521920668058456, + "grad_norm": 0.919992208480835, + "learning_rate": 4.996477613239822e-06, + "loss": 0.9016, + "step": 163 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 0.9655581712722778, + "learning_rate": 4.996364931284847e-06, + "loss": 0.896, + "step": 164 + }, + { + "epoch": 0.25835073068893527, + "grad_norm": 0.9159606099128723, + "learning_rate": 4.996250476545212e-06, + "loss": 0.9475, + "step": 165 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 1.1147626638412476, + "learning_rate": 4.996134249102195e-06, + "loss": 0.8793, + "step": 166 + }, + { + "epoch": 0.26148225469728603, + "grad_norm": 1.112069010734558, + "learning_rate": 4.996016249038334e-06, + "loss": 0.8765, + "step": 167 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 1.0366944074630737, + "learning_rate": 4.995896476437428e-06, + "loss": 0.9356, + "step": 168 + }, + { + "epoch": 0.26461377870563674, + "grad_norm": 0.9494239091873169, + "learning_rate": 4.995774931384534e-06, + "loss": 0.8246, + "step": 169 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 0.9819200038909912, + "learning_rate": 4.995651613965964e-06, + "loss": 0.9403, + "step": 170 + }, + { + "epoch": 0.2677453027139875, + "grad_norm": 1.7329354286193848, + "learning_rate": 4.995526524269293e-06, + "loss": 0.9091, + "step": 171 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.1415139436721802, + "learning_rate": 4.995399662383352e-06, + "loss": 0.9106, + "step": 172 + }, + { + "epoch": 0.2708768267223382, + "grad_norm": 0.9471906423568726, + "learning_rate": 4.995271028398233e-06, + "loss": 0.8858, + "step": 173 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 1.096053123474121, + "learning_rate": 4.995140622405284e-06, + "loss": 0.7502, + "step": 174 + }, + { + "epoch": 0.2740083507306889, + "grad_norm": 1.285410761833191, + "learning_rate": 4.995008444497113e-06, + "loss": 0.9288, + "step": 175 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 0.9714595079421997, + "learning_rate": 4.994874494767585e-06, + "loss": 0.8824, + "step": 176 + }, + { + "epoch": 0.27713987473903967, + "grad_norm": 0.8399011492729187, + "learning_rate": 4.994738773311824e-06, + "loss": 0.8703, + "step": 177 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 0.9454224109649658, + "learning_rate": 4.994601280226212e-06, + "loss": 0.9627, + "step": 178 + }, + { + "epoch": 0.2802713987473904, + "grad_norm": 0.8983661532402039, + "learning_rate": 4.99446201560839e-06, + "loss": 0.9086, + "step": 179 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 0.9814153909683228, + "learning_rate": 4.994320979557256e-06, + "loss": 0.908, + "step": 180 + }, + { + "epoch": 0.28340292275574114, + "grad_norm": 0.8405355215072632, + "learning_rate": 4.9941781721729645e-06, + "loss": 0.8414, + "step": 181 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 0.950950026512146, + "learning_rate": 4.994033593556933e-06, + "loss": 0.9091, + "step": 182 + }, + { + "epoch": 0.28653444676409184, + "grad_norm": 1.1019747257232666, + "learning_rate": 4.99388724381183e-06, + "loss": 0.8305, + "step": 183 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 0.9339331984519958, + "learning_rate": 4.9937391230415875e-06, + "loss": 0.8542, + "step": 184 + }, + { + "epoch": 0.2896659707724426, + "grad_norm": 0.8813616633415222, + "learning_rate": 4.993589231351394e-06, + "loss": 0.848, + "step": 185 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 0.8916418552398682, + "learning_rate": 4.9934375688476905e-06, + "loss": 0.8032, + "step": 186 + }, + { + "epoch": 0.2927974947807933, + "grad_norm": 0.9891005754470825, + "learning_rate": 4.993284135638182e-06, + "loss": 0.7937, + "step": 187 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 0.9842082262039185, + "learning_rate": 4.99312893183183e-06, + "loss": 0.9532, + "step": 188 + }, + { + "epoch": 0.29592901878914407, + "grad_norm": 0.993478536605835, + "learning_rate": 4.99297195753885e-06, + "loss": 0.9406, + "step": 189 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 1.0654981136322021, + "learning_rate": 4.992813212870718e-06, + "loss": 0.907, + "step": 190 + }, + { + "epoch": 0.2990605427974948, + "grad_norm": 0.8937190771102905, + "learning_rate": 4.992652697940164e-06, + "loss": 0.9307, + "step": 191 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 0.9006052613258362, + "learning_rate": 4.992490412861178e-06, + "loss": 0.8808, + "step": 192 + }, + { + "epoch": 0.30219206680584554, + "grad_norm": 0.9601522088050842, + "learning_rate": 4.992326357749007e-06, + "loss": 0.8495, + "step": 193 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 0.9814243316650391, + "learning_rate": 4.992160532720153e-06, + "loss": 0.8204, + "step": 194 + }, + { + "epoch": 0.30532359081419624, + "grad_norm": 0.9660456776618958, + "learning_rate": 4.991992937892377e-06, + "loss": 0.8209, + "step": 195 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 0.8504270315170288, + "learning_rate": 4.991823573384695e-06, + "loss": 0.8415, + "step": 196 + }, + { + "epoch": 0.30845511482254695, + "grad_norm": 0.8733190894126892, + "learning_rate": 4.991652439317382e-06, + "loss": 0.8753, + "step": 197 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 0.9761794805526733, + "learning_rate": 4.991479535811967e-06, + "loss": 0.8923, + "step": 198 + }, + { + "epoch": 0.3115866388308977, + "grad_norm": 1.0400285720825195, + "learning_rate": 4.991304862991238e-06, + "loss": 0.899, + "step": 199 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 0.8561192750930786, + "learning_rate": 4.991128420979237e-06, + "loss": 0.9117, + "step": 200 + }, + { + "epoch": 0.3147181628392484, + "grad_norm": 0.8839634656906128, + "learning_rate": 4.990950209901263e-06, + "loss": 0.9022, + "step": 201 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 0.9569464921951294, + "learning_rate": 4.990770229883873e-06, + "loss": 0.9013, + "step": 202 + }, + { + "epoch": 0.3178496868475992, + "grad_norm": 1.0600861310958862, + "learning_rate": 4.990588481054881e-06, + "loss": 0.8387, + "step": 203 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 0.9492761492729187, + "learning_rate": 4.990404963543352e-06, + "loss": 0.8521, + "step": 204 + }, + { + "epoch": 0.3209812108559499, + "grad_norm": 0.9534817934036255, + "learning_rate": 4.990219677479612e-06, + "loss": 0.8009, + "step": 205 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 0.9918960928916931, + "learning_rate": 4.990032622995242e-06, + "loss": 0.8215, + "step": 206 + }, + { + "epoch": 0.32411273486430064, + "grad_norm": 0.8751630783081055, + "learning_rate": 4.989843800223077e-06, + "loss": 0.8744, + "step": 207 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 1.0034785270690918, + "learning_rate": 4.9896532092972085e-06, + "loss": 0.884, + "step": 208 + }, + { + "epoch": 0.32724425887265135, + "grad_norm": 1.2693432569503784, + "learning_rate": 4.989460850352984e-06, + "loss": 0.9689, + "step": 209 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 0.9923811554908752, + "learning_rate": 4.989266723527007e-06, + "loss": 0.9622, + "step": 210 + }, + { + "epoch": 0.3303757828810021, + "grad_norm": 0.9439074397087097, + "learning_rate": 4.989070828957137e-06, + "loss": 0.9155, + "step": 211 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.2564488649368286, + "learning_rate": 4.988873166782485e-06, + "loss": 0.8712, + "step": 212 + }, + { + "epoch": 0.3335073068893528, + "grad_norm": 1.1441881656646729, + "learning_rate": 4.988673737143423e-06, + "loss": 0.8895, + "step": 213 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 0.972155749797821, + "learning_rate": 4.988472540181574e-06, + "loss": 0.9158, + "step": 214 + }, + { + "epoch": 0.3366388308977035, + "grad_norm": 0.9536120295524597, + "learning_rate": 4.988269576039817e-06, + "loss": 0.7963, + "step": 215 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 0.8956594467163086, + "learning_rate": 4.988064844862288e-06, + "loss": 0.8947, + "step": 216 + }, + { + "epoch": 0.3397703549060543, + "grad_norm": 0.9503691792488098, + "learning_rate": 4.987858346794373e-06, + "loss": 0.7857, + "step": 217 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 0.9471456408500671, + "learning_rate": 4.987650081982719e-06, + "loss": 0.9118, + "step": 218 + }, + { + "epoch": 0.342901878914405, + "grad_norm": 0.9485713839530945, + "learning_rate": 4.987440050575222e-06, + "loss": 0.8188, + "step": 219 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 1.0775927305221558, + "learning_rate": 4.987228252721037e-06, + "loss": 0.9469, + "step": 220 + }, + { + "epoch": 0.34603340292275575, + "grad_norm": 0.9173988103866577, + "learning_rate": 4.9870146885705705e-06, + "loss": 0.8566, + "step": 221 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.008735179901123, + "learning_rate": 4.986799358275485e-06, + "loss": 0.8637, + "step": 222 + }, + { + "epoch": 0.34916492693110646, + "grad_norm": 0.8183198571205139, + "learning_rate": 4.986582261988695e-06, + "loss": 0.9321, + "step": 223 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 0.8951883316040039, + "learning_rate": 4.986363399864372e-06, + "loss": 0.8015, + "step": 224 + }, + { + "epoch": 0.3522964509394572, + "grad_norm": 0.9275704622268677, + "learning_rate": 4.986142772057941e-06, + "loss": 0.8242, + "step": 225 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 0.9548516273498535, + "learning_rate": 4.985920378726077e-06, + "loss": 0.7889, + "step": 226 + }, + { + "epoch": 0.3554279749478079, + "grad_norm": 1.1203585863113403, + "learning_rate": 4.985696220026714e-06, + "loss": 0.835, + "step": 227 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 0.9525771141052246, + "learning_rate": 4.985470296119038e-06, + "loss": 0.9162, + "step": 228 + }, + { + "epoch": 0.3585594989561587, + "grad_norm": 1.150404930114746, + "learning_rate": 4.985242607163488e-06, + "loss": 0.9456, + "step": 229 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.01652991771698, + "learning_rate": 4.985013153321755e-06, + "loss": 0.87, + "step": 230 + }, + { + "epoch": 0.3616910229645094, + "grad_norm": 0.9455218315124512, + "learning_rate": 4.984781934756786e-06, + "loss": 0.8466, + "step": 231 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 0.9125218391418457, + "learning_rate": 4.984548951632779e-06, + "loss": 0.8764, + "step": 232 + }, + { + "epoch": 0.36482254697286015, + "grad_norm": 2.0192980766296387, + "learning_rate": 4.984314204115187e-06, + "loss": 0.7866, + "step": 233 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.8118515014648438, + "learning_rate": 4.984077692370716e-06, + "loss": 0.8703, + "step": 234 + }, + { + "epoch": 0.36795407098121086, + "grad_norm": 1.2995024919509888, + "learning_rate": 4.983839416567322e-06, + "loss": 0.9459, + "step": 235 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 1.0611528158187866, + "learning_rate": 4.983599376874216e-06, + "loss": 0.8456, + "step": 236 + }, + { + "epoch": 0.37108559498956156, + "grad_norm": 1.2018043994903564, + "learning_rate": 4.983357573461863e-06, + "loss": 0.914, + "step": 237 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9091911315917969, + "learning_rate": 4.983114006501978e-06, + "loss": 0.791, + "step": 238 + }, + { + "epoch": 0.3742171189979123, + "grad_norm": 0.980873167514801, + "learning_rate": 4.982868676167528e-06, + "loss": 0.77, + "step": 239 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 0.8945695161819458, + "learning_rate": 4.982621582632735e-06, + "loss": 0.8038, + "step": 240 + }, + { + "epoch": 0.37734864300626303, + "grad_norm": 0.9325360655784607, + "learning_rate": 4.9823727260730705e-06, + "loss": 0.8908, + "step": 241 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 0.987630307674408, + "learning_rate": 4.98212210666526e-06, + "loss": 0.8642, + "step": 242 + }, + { + "epoch": 0.3804801670146138, + "grad_norm": 0.9151357412338257, + "learning_rate": 4.98186972458728e-06, + "loss": 0.8437, + "step": 243 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 0.9084850549697876, + "learning_rate": 4.981615580018358e-06, + "loss": 0.8775, + "step": 244 + }, + { + "epoch": 0.3836116910229645, + "grad_norm": 0.9786385297775269, + "learning_rate": 4.981359673138974e-06, + "loss": 0.8251, + "step": 245 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 0.8947877883911133, + "learning_rate": 4.98110200413086e-06, + "loss": 0.7709, + "step": 246 + }, + { + "epoch": 0.38674321503131526, + "grad_norm": 0.967019259929657, + "learning_rate": 4.9808425731769985e-06, + "loss": 0.8954, + "step": 247 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 1.1066704988479614, + "learning_rate": 4.980581380461622e-06, + "loss": 0.8544, + "step": 248 + }, + { + "epoch": 0.38987473903966596, + "grad_norm": 0.9750981330871582, + "learning_rate": 4.980318426170218e-06, + "loss": 0.8354, + "step": 249 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 0.8727232217788696, + "learning_rate": 4.980053710489521e-06, + "loss": 0.8911, + "step": 250 + }, + { + "epoch": 0.3930062630480167, + "grad_norm": 0.9337188601493835, + "learning_rate": 4.979787233607518e-06, + "loss": 0.8567, + "step": 251 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 0.890701174736023, + "learning_rate": 4.979518995713448e-06, + "loss": 0.8327, + "step": 252 + }, + { + "epoch": 0.39613778705636743, + "grad_norm": 0.9973195195198059, + "learning_rate": 4.979248996997797e-06, + "loss": 0.9065, + "step": 253 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 0.9894055128097534, + "learning_rate": 4.978977237652306e-06, + "loss": 0.8729, + "step": 254 + }, + { + "epoch": 0.39926931106471814, + "grad_norm": 1.0398536920547485, + "learning_rate": 4.978703717869961e-06, + "loss": 0.861, + "step": 255 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.1713942289352417, + "learning_rate": 4.978428437845003e-06, + "loss": 0.8183, + "step": 256 + }, + { + "epoch": 0.4024008350730689, + "grad_norm": 0.9868776202201843, + "learning_rate": 4.9781513977729225e-06, + "loss": 0.941, + "step": 257 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 0.9260863065719604, + "learning_rate": 4.977872597850455e-06, + "loss": 0.8963, + "step": 258 + }, + { + "epoch": 0.4055323590814196, + "grad_norm": 0.9553239941596985, + "learning_rate": 4.977592038275592e-06, + "loss": 0.8102, + "step": 259 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 0.9426907300949097, + "learning_rate": 4.977309719247571e-06, + "loss": 0.89, + "step": 260 + }, + { + "epoch": 0.40866388308977036, + "grad_norm": 0.9691527485847473, + "learning_rate": 4.97702564096688e-06, + "loss": 0.7967, + "step": 261 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 0.9917487502098083, + "learning_rate": 4.976739803635256e-06, + "loss": 0.8327, + "step": 262 + }, + { + "epoch": 0.41179540709812107, + "grad_norm": 0.992340087890625, + "learning_rate": 4.976452207455686e-06, + "loss": 0.9462, + "step": 263 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 0.9554483890533447, + "learning_rate": 4.9761628526324035e-06, + "loss": 0.8294, + "step": 264 + }, + { + "epoch": 0.41492693110647183, + "grad_norm": 0.9318789839744568, + "learning_rate": 4.975871739370895e-06, + "loss": 0.8079, + "step": 265 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 0.9158215522766113, + "learning_rate": 4.975578867877893e-06, + "loss": 0.8632, + "step": 266 + }, + { + "epoch": 0.41805845511482254, + "grad_norm": 0.9444370865821838, + "learning_rate": 4.975284238361378e-06, + "loss": 0.8897, + "step": 267 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 0.9949231743812561, + "learning_rate": 4.974987851030581e-06, + "loss": 0.7905, + "step": 268 + }, + { + "epoch": 0.4211899791231733, + "grad_norm": 0.9819440245628357, + "learning_rate": 4.974689706095981e-06, + "loss": 0.9183, + "step": 269 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0787206888198853, + "learning_rate": 4.974389803769304e-06, + "loss": 0.7739, + "step": 270 + }, + { + "epoch": 0.424321503131524, + "grad_norm": 0.8871361017227173, + "learning_rate": 4.9740881442635235e-06, + "loss": 0.9454, + "step": 271 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 0.9125893115997314, + "learning_rate": 4.973784727792863e-06, + "loss": 0.8017, + "step": 272 + }, + { + "epoch": 0.42745302713987476, + "grad_norm": 0.9102838635444641, + "learning_rate": 4.973479554572793e-06, + "loss": 0.8307, + "step": 273 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 0.8780474662780762, + "learning_rate": 4.973172624820031e-06, + "loss": 0.9402, + "step": 274 + }, + { + "epoch": 0.43058455114822547, + "grad_norm": 1.0059378147125244, + "learning_rate": 4.972863938752541e-06, + "loss": 0.7438, + "step": 275 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.9684615135192871, + "learning_rate": 4.972553496589537e-06, + "loss": 0.8897, + "step": 276 + }, + { + "epoch": 0.4337160751565762, + "grad_norm": 0.9615057110786438, + "learning_rate": 4.972241298551477e-06, + "loss": 0.8157, + "step": 277 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8589702844619751, + "learning_rate": 4.971927344860068e-06, + "loss": 0.7813, + "step": 278 + }, + { + "epoch": 0.43684759916492694, + "grad_norm": 0.900324285030365, + "learning_rate": 4.971611635738263e-06, + "loss": 0.7534, + "step": 279 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 0.8881833553314209, + "learning_rate": 4.971294171410261e-06, + "loss": 0.8255, + "step": 280 + }, + { + "epoch": 0.43997912317327764, + "grad_norm": 0.9723751544952393, + "learning_rate": 4.9709749521015085e-06, + "loss": 0.8796, + "step": 281 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 0.9341713190078735, + "learning_rate": 4.9706539780386995e-06, + "loss": 0.7429, + "step": 282 + }, + { + "epoch": 0.4431106471816284, + "grad_norm": 0.9354205131530762, + "learning_rate": 4.97033124944977e-06, + "loss": 0.8545, + "step": 283 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 0.9294624328613281, + "learning_rate": 4.970006766563906e-06, + "loss": 0.7783, + "step": 284 + }, + { + "epoch": 0.4462421711899791, + "grad_norm": 0.9584887027740479, + "learning_rate": 4.969680529611538e-06, + "loss": 0.8068, + "step": 285 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 0.9525735378265381, + "learning_rate": 4.9693525388243424e-06, + "loss": 0.8055, + "step": 286 + }, + { + "epoch": 0.44937369519832987, + "grad_norm": 0.8845317959785461, + "learning_rate": 4.96902279443524e-06, + "loss": 0.8318, + "step": 287 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 0.8932276964187622, + "learning_rate": 4.9686912966783965e-06, + "loss": 0.8185, + "step": 288 + }, + { + "epoch": 0.4525052192066806, + "grad_norm": 0.9881951808929443, + "learning_rate": 4.968358045789225e-06, + "loss": 0.9001, + "step": 289 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 0.995212197303772, + "learning_rate": 4.968023042004384e-06, + "loss": 0.75, + "step": 290 + }, + { + "epoch": 0.45563674321503134, + "grad_norm": 0.9997846484184265, + "learning_rate": 4.967686285561773e-06, + "loss": 0.8993, + "step": 291 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 0.9438545107841492, + "learning_rate": 4.967347776700538e-06, + "loss": 0.8964, + "step": 292 + }, + { + "epoch": 0.45876826722338204, + "grad_norm": 0.9164395332336426, + "learning_rate": 4.967007515661072e-06, + "loss": 0.8671, + "step": 293 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.895302951335907, + "learning_rate": 4.966665502685009e-06, + "loss": 0.8785, + "step": 294 + }, + { + "epoch": 0.4618997912317328, + "grad_norm": 0.9122863411903381, + "learning_rate": 4.966321738015227e-06, + "loss": 0.8499, + "step": 295 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.0097336769104004, + "learning_rate": 4.965976221895852e-06, + "loss": 0.83, + "step": 296 + }, + { + "epoch": 0.4650313152400835, + "grad_norm": 1.0664499998092651, + "learning_rate": 4.965628954572249e-06, + "loss": 0.7799, + "step": 297 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 0.9602341055870056, + "learning_rate": 4.96527993629103e-06, + "loss": 0.8061, + "step": 298 + }, + { + "epoch": 0.4681628392484342, + "grad_norm": 1.0133620500564575, + "learning_rate": 4.964929167300048e-06, + "loss": 0.9148, + "step": 299 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9351868033409119, + "learning_rate": 4.964576647848401e-06, + "loss": 0.83, + "step": 300 + }, + { + "epoch": 0.471294363256785, + "grad_norm": 1.0078860521316528, + "learning_rate": 4.964222378186429e-06, + "loss": 0.8391, + "step": 301 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 0.8966493606567383, + "learning_rate": 4.963866358565716e-06, + "loss": 0.8677, + "step": 302 + }, + { + "epoch": 0.4744258872651357, + "grad_norm": 0.9283657670021057, + "learning_rate": 4.963508589239087e-06, + "loss": 0.8787, + "step": 303 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 0.9540023803710938, + "learning_rate": 4.963149070460611e-06, + "loss": 0.8667, + "step": 304 + }, + { + "epoch": 0.47755741127348644, + "grad_norm": 0.9580283164978027, + "learning_rate": 4.9627878024855995e-06, + "loss": 0.8442, + "step": 305 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.9008278846740723, + "learning_rate": 4.962424785570605e-06, + "loss": 0.8605, + "step": 306 + }, + { + "epoch": 0.48068893528183715, + "grad_norm": 0.9681883454322815, + "learning_rate": 4.962060019973423e-06, + "loss": 0.7865, + "step": 307 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 0.8492597341537476, + "learning_rate": 4.9616935059530915e-06, + "loss": 0.8578, + "step": 308 + }, + { + "epoch": 0.4838204592901879, + "grad_norm": 0.9323164820671082, + "learning_rate": 4.9613252437698865e-06, + "loss": 0.8491, + "step": 309 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 0.9417833685874939, + "learning_rate": 4.960955233685331e-06, + "loss": 0.8108, + "step": 310 + }, + { + "epoch": 0.4869519832985386, + "grad_norm": 0.901364803314209, + "learning_rate": 4.960583475962185e-06, + "loss": 0.8958, + "step": 311 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8929979205131531, + "learning_rate": 4.96020997086445e-06, + "loss": 0.8237, + "step": 312 + }, + { + "epoch": 0.4900835073068894, + "grad_norm": 0.9422337412834167, + "learning_rate": 4.9598347186573716e-06, + "loss": 0.8024, + "step": 313 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 0.9165504574775696, + "learning_rate": 4.959457719607432e-06, + "loss": 0.8124, + "step": 314 + }, + { + "epoch": 0.4932150313152401, + "grad_norm": 1.0233370065689087, + "learning_rate": 4.959078973982356e-06, + "loss": 0.788, + "step": 315 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9375972151756287, + "learning_rate": 4.958698482051109e-06, + "loss": 0.839, + "step": 316 + }, + { + "epoch": 0.4963465553235908, + "grad_norm": 0.8805437684059143, + "learning_rate": 4.958316244083895e-06, + "loss": 0.7445, + "step": 317 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.0231893062591553, + "learning_rate": 4.95793226035216e-06, + "loss": 0.8257, + "step": 318 + }, + { + "epoch": 0.49947807933194155, + "grad_norm": 0.9125928282737732, + "learning_rate": 4.957546531128589e-06, + "loss": 0.7699, + "step": 319 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.8844016790390015, + "learning_rate": 4.957159056687104e-06, + "loss": 0.8784, + "step": 320 + }, + { + "epoch": 0.5026096033402923, + "grad_norm": 0.9770572781562805, + "learning_rate": 4.956769837302871e-06, + "loss": 0.8611, + "step": 321 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.9413725733757019, + "learning_rate": 4.956378873252293e-06, + "loss": 0.8056, + "step": 322 + }, + { + "epoch": 0.505741127348643, + "grad_norm": 0.9058850407600403, + "learning_rate": 4.955986164813008e-06, + "loss": 0.8162, + "step": 323 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 0.893100917339325, + "learning_rate": 4.955591712263901e-06, + "loss": 0.7554, + "step": 324 + }, + { + "epoch": 0.5088726513569938, + "grad_norm": 0.8967075943946838, + "learning_rate": 4.955195515885089e-06, + "loss": 0.8143, + "step": 325 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.9542785286903381, + "learning_rate": 4.954797575957929e-06, + "loss": 0.8102, + "step": 326 + }, + { + "epoch": 0.5120041753653445, + "grad_norm": 1.006451964378357, + "learning_rate": 4.954397892765018e-06, + "loss": 0.86, + "step": 327 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 0.9964715838432312, + "learning_rate": 4.953996466590189e-06, + "loss": 0.7166, + "step": 328 + }, + { + "epoch": 0.5151356993736952, + "grad_norm": 0.9568808674812317, + "learning_rate": 4.953593297718512e-06, + "loss": 0.8753, + "step": 329 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.9435825347900391, + "learning_rate": 4.953188386436298e-06, + "loss": 0.8961, + "step": 330 + }, + { + "epoch": 0.5182672233820459, + "grad_norm": 0.9386457800865173, + "learning_rate": 4.952781733031092e-06, + "loss": 0.8468, + "step": 331 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 0.9916554689407349, + "learning_rate": 4.952373337791678e-06, + "loss": 0.8274, + "step": 332 + }, + { + "epoch": 0.5213987473903967, + "grad_norm": 0.9351572394371033, + "learning_rate": 4.9519632010080765e-06, + "loss": 0.7964, + "step": 333 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9936269521713257, + "learning_rate": 4.951551322971545e-06, + "loss": 0.8354, + "step": 334 + }, + { + "epoch": 0.5245302713987474, + "grad_norm": 0.9678221940994263, + "learning_rate": 4.951137703974576e-06, + "loss": 0.9115, + "step": 335 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 0.9453133940696716, + "learning_rate": 4.9507223443109005e-06, + "loss": 0.7725, + "step": 336 + }, + { + "epoch": 0.5276617954070981, + "grad_norm": 0.9577226638793945, + "learning_rate": 4.950305244275484e-06, + "loss": 0.8679, + "step": 337 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 0.998489260673523, + "learning_rate": 4.94988640416453e-06, + "loss": 0.7505, + "step": 338 + }, + { + "epoch": 0.5307933194154488, + "grad_norm": 0.976026713848114, + "learning_rate": 4.949465824275475e-06, + "loss": 0.9083, + "step": 339 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.0041502714157104, + "learning_rate": 4.9490435049069925e-06, + "loss": 0.8815, + "step": 340 + }, + { + "epoch": 0.5339248434237995, + "grad_norm": 0.9839801788330078, + "learning_rate": 4.948619446358991e-06, + "loss": 0.8517, + "step": 341 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 0.9995140433311462, + "learning_rate": 4.948193648932616e-06, + "loss": 0.8157, + "step": 342 + }, + { + "epoch": 0.5370563674321504, + "grad_norm": 1.0782225131988525, + "learning_rate": 4.947766112930243e-06, + "loss": 0.9206, + "step": 343 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9981802105903625, + "learning_rate": 4.947336838655487e-06, + "loss": 0.8389, + "step": 344 + }, + { + "epoch": 0.5401878914405011, + "grad_norm": 0.9261236190795898, + "learning_rate": 4.946905826413195e-06, + "loss": 0.8035, + "step": 345 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.9195284247398376, + "learning_rate": 4.946473076509449e-06, + "loss": 0.8265, + "step": 346 + }, + { + "epoch": 0.5433194154488518, + "grad_norm": 0.9535123109817505, + "learning_rate": 4.946038589251566e-06, + "loss": 0.7528, + "step": 347 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 0.9969532489776611, + "learning_rate": 4.9456023649480935e-06, + "loss": 0.7787, + "step": 348 + }, + { + "epoch": 0.5464509394572025, + "grad_norm": 0.9612399935722351, + "learning_rate": 4.945164403908816e-06, + "loss": 0.8298, + "step": 349 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 0.9541627764701843, + "learning_rate": 4.944724706444749e-06, + "loss": 0.8039, + "step": 350 + }, + { + "epoch": 0.5495824634655533, + "grad_norm": 0.9916557669639587, + "learning_rate": 4.944283272868143e-06, + "loss": 0.9012, + "step": 351 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 0.8807661533355713, + "learning_rate": 4.94384010349248e-06, + "loss": 0.9005, + "step": 352 + }, + { + "epoch": 0.552713987473904, + "grad_norm": 1.0179708003997803, + "learning_rate": 4.943395198632476e-06, + "loss": 0.8845, + "step": 353 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.923014223575592, + "learning_rate": 4.942948558604077e-06, + "loss": 0.8173, + "step": 354 + }, + { + "epoch": 0.5558455114822547, + "grad_norm": 0.9318749308586121, + "learning_rate": 4.942500183724463e-06, + "loss": 0.8828, + "step": 355 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.0907000303268433, + "learning_rate": 4.942050074312048e-06, + "loss": 0.7951, + "step": 356 + }, + { + "epoch": 0.5589770354906054, + "grad_norm": 0.9811620712280273, + "learning_rate": 4.9415982306864744e-06, + "loss": 0.8826, + "step": 357 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9888150691986084, + "learning_rate": 4.9411446531686156e-06, + "loss": 0.8338, + "step": 358 + }, + { + "epoch": 0.5621085594989561, + "grad_norm": 0.9389257431030273, + "learning_rate": 4.9406893420805804e-06, + "loss": 0.8316, + "step": 359 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.9500603675842285, + "learning_rate": 4.940232297745705e-06, + "loss": 0.8764, + "step": 360 + }, + { + "epoch": 0.5652400835073069, + "grad_norm": 0.9748769998550415, + "learning_rate": 4.939773520488559e-06, + "loss": 0.8308, + "step": 361 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 0.9179681539535522, + "learning_rate": 4.93931301063494e-06, + "loss": 0.789, + "step": 362 + }, + { + "epoch": 0.5683716075156576, + "grad_norm": 0.8609325289726257, + "learning_rate": 4.938850768511878e-06, + "loss": 0.7368, + "step": 363 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 0.9375531673431396, + "learning_rate": 4.9383867944476325e-06, + "loss": 0.8164, + "step": 364 + }, + { + "epoch": 0.5715031315240083, + "grad_norm": 0.9386453032493591, + "learning_rate": 4.937921088771694e-06, + "loss": 0.7898, + "step": 365 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.025040864944458, + "learning_rate": 4.937453651814781e-06, + "loss": 0.8465, + "step": 366 + }, + { + "epoch": 0.574634655532359, + "grad_norm": 1.1037245988845825, + "learning_rate": 4.936984483908842e-06, + "loss": 0.7935, + "step": 367 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.230821132659912, + "learning_rate": 4.936513585387055e-06, + "loss": 0.8043, + "step": 368 + }, + { + "epoch": 0.5777661795407099, + "grad_norm": 1.090235948562622, + "learning_rate": 4.9360409565838265e-06, + "loss": 0.8726, + "step": 369 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8977810144424438, + "learning_rate": 4.935566597834793e-06, + "loss": 0.8667, + "step": 370 + }, + { + "epoch": 0.5808977035490606, + "grad_norm": 0.96401447057724, + "learning_rate": 4.935090509476818e-06, + "loss": 0.8494, + "step": 371 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 0.9290149211883545, + "learning_rate": 4.934612691847995e-06, + "loss": 0.7678, + "step": 372 + }, + { + "epoch": 0.5840292275574113, + "grad_norm": 0.9765584468841553, + "learning_rate": 4.934133145287644e-06, + "loss": 0.7947, + "step": 373 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 0.8939430117607117, + "learning_rate": 4.933651870136313e-06, + "loss": 0.885, + "step": 374 + }, + { + "epoch": 0.587160751565762, + "grad_norm": 0.9512760043144226, + "learning_rate": 4.933168866735777e-06, + "loss": 0.8863, + "step": 375 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 1.0334441661834717, + "learning_rate": 4.932684135429041e-06, + "loss": 0.8362, + "step": 376 + }, + { + "epoch": 0.5902922755741128, + "grad_norm": 0.9402525424957275, + "learning_rate": 4.932197676560334e-06, + "loss": 0.8758, + "step": 377 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 0.9615777730941772, + "learning_rate": 4.931709490475113e-06, + "loss": 0.8324, + "step": 378 + }, + { + "epoch": 0.5934237995824635, + "grad_norm": 0.9407138228416443, + "learning_rate": 4.931219577520061e-06, + "loss": 0.8454, + "step": 379 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.0415815114974976, + "learning_rate": 4.930727938043091e-06, + "loss": 0.8649, + "step": 380 + }, + { + "epoch": 0.5965553235908142, + "grad_norm": 1.0946165323257446, + "learning_rate": 4.9302345723933344e-06, + "loss": 0.7794, + "step": 381 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.0532523393630981, + "learning_rate": 4.929739480921157e-06, + "loss": 0.8555, + "step": 382 + }, + { + "epoch": 0.5996868475991649, + "grad_norm": 0.9441399574279785, + "learning_rate": 4.929242663978144e-06, + "loss": 0.7613, + "step": 383 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 0.9970793128013611, + "learning_rate": 4.92874412191711e-06, + "loss": 0.9037, + "step": 384 + }, + { + "epoch": 0.6028183716075156, + "grad_norm": 0.9777354001998901, + "learning_rate": 4.928243855092091e-06, + "loss": 0.9044, + "step": 385 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 0.9507855772972107, + "learning_rate": 4.92774186385835e-06, + "loss": 0.8359, + "step": 386 + }, + { + "epoch": 0.6059498956158664, + "grad_norm": 0.9827553033828735, + "learning_rate": 4.9272381485723765e-06, + "loss": 0.7938, + "step": 387 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 0.8896475434303284, + "learning_rate": 4.926732709591879e-06, + "loss": 0.8545, + "step": 388 + }, + { + "epoch": 0.6090814196242171, + "grad_norm": 0.9817705750465393, + "learning_rate": 4.926225547275795e-06, + "loss": 0.8456, + "step": 389 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 0.9646008610725403, + "learning_rate": 4.925716661984283e-06, + "loss": 0.7906, + "step": 390 + }, + { + "epoch": 0.6122129436325678, + "grad_norm": 0.8940444588661194, + "learning_rate": 4.9252060540787275e-06, + "loss": 0.8371, + "step": 391 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.0812394618988037, + "learning_rate": 4.924693723921734e-06, + "loss": 0.8056, + "step": 392 + }, + { + "epoch": 0.6153444676409185, + "grad_norm": 0.9953856468200684, + "learning_rate": 4.924179671877132e-06, + "loss": 0.8651, + "step": 393 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 0.9760703444480896, + "learning_rate": 4.9236638983099735e-06, + "loss": 0.845, + "step": 394 + }, + { + "epoch": 0.6184759916492694, + "grad_norm": 1.0127615928649902, + "learning_rate": 4.923146403586533e-06, + "loss": 0.7987, + "step": 395 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 0.8771396279335022, + "learning_rate": 4.9226271880743086e-06, + "loss": 0.8092, + "step": 396 + }, + { + "epoch": 0.6216075156576201, + "grad_norm": 0.8906831741333008, + "learning_rate": 4.9221062521420196e-06, + "loss": 0.8039, + "step": 397 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.0218392610549927, + "learning_rate": 4.9215835961596045e-06, + "loss": 0.8354, + "step": 398 + }, + { + "epoch": 0.6247390396659708, + "grad_norm": 1.0382038354873657, + "learning_rate": 4.921059220498227e-06, + "loss": 0.8021, + "step": 399 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 0.8997588157653809, + "learning_rate": 4.920533125530272e-06, + "loss": 0.8716, + "step": 400 + }, + { + "epoch": 0.6278705636743215, + "grad_norm": 0.8725724220275879, + "learning_rate": 4.920005311629342e-06, + "loss": 0.7934, + "step": 401 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.9289800524711609, + "learning_rate": 4.9194757791702625e-06, + "loss": 0.812, + "step": 402 + }, + { + "epoch": 0.6310020876826722, + "grad_norm": 0.9969078898429871, + "learning_rate": 4.918944528529079e-06, + "loss": 0.866, + "step": 403 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 0.9544941186904907, + "learning_rate": 4.918411560083058e-06, + "loss": 0.8035, + "step": 404 + }, + { + "epoch": 0.634133611691023, + "grad_norm": 0.9678053855895996, + "learning_rate": 4.917876874210686e-06, + "loss": 0.8254, + "step": 405 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 0.9782965779304504, + "learning_rate": 4.917340471291666e-06, + "loss": 0.7954, + "step": 406 + }, + { + "epoch": 0.6372651356993737, + "grad_norm": 0.9380600452423096, + "learning_rate": 4.916802351706923e-06, + "loss": 0.8287, + "step": 407 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 0.9455891847610474, + "learning_rate": 4.916262515838603e-06, + "loss": 0.8714, + "step": 408 + }, + { + "epoch": 0.6403966597077244, + "grad_norm": 0.8937395811080933, + "learning_rate": 4.915720964070065e-06, + "loss": 0.7618, + "step": 409 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 1.0304689407348633, + "learning_rate": 4.915177696785894e-06, + "loss": 0.8436, + "step": 410 + }, + { + "epoch": 0.6435281837160751, + "grad_norm": 0.9623110294342041, + "learning_rate": 4.914632714371885e-06, + "loss": 0.865, + "step": 411 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 0.9212867021560669, + "learning_rate": 4.914086017215059e-06, + "loss": 0.8232, + "step": 412 + }, + { + "epoch": 0.6466597077244259, + "grad_norm": 0.9817205667495728, + "learning_rate": 4.9135376057036485e-06, + "loss": 0.7926, + "step": 413 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 0.9576719403266907, + "learning_rate": 4.912987480227108e-06, + "loss": 0.8812, + "step": 414 + }, + { + "epoch": 0.6497912317327766, + "grad_norm": 0.9415407180786133, + "learning_rate": 4.912435641176106e-06, + "loss": 0.8985, + "step": 415 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 0.9084275364875793, + "learning_rate": 4.91188208894253e-06, + "loss": 0.8085, + "step": 416 + }, + { + "epoch": 0.6529227557411273, + "grad_norm": 0.8890969753265381, + "learning_rate": 4.911326823919482e-06, + "loss": 0.8127, + "step": 417 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 0.8727678656578064, + "learning_rate": 4.9107698465012815e-06, + "loss": 0.8656, + "step": 418 + }, + { + "epoch": 0.656054279749478, + "grad_norm": 1.0174667835235596, + "learning_rate": 4.910211157083465e-06, + "loss": 0.8358, + "step": 419 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.045121431350708, + "learning_rate": 4.909650756062782e-06, + "loss": 0.8239, + "step": 420 + }, + { + "epoch": 0.6591858037578288, + "grad_norm": 1.0688676834106445, + "learning_rate": 4.909088643837202e-06, + "loss": 0.9246, + "step": 421 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.9371877908706665, + "learning_rate": 4.908524820805905e-06, + "loss": 0.7862, + "step": 422 + }, + { + "epoch": 0.6623173277661796, + "grad_norm": 0.9567264318466187, + "learning_rate": 4.907959287369288e-06, + "loss": 0.8164, + "step": 423 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 0.9109166860580444, + "learning_rate": 4.907392043928964e-06, + "loss": 0.7016, + "step": 424 + }, + { + "epoch": 0.6654488517745303, + "grad_norm": 1.0378203392028809, + "learning_rate": 4.906823090887757e-06, + "loss": 0.8986, + "step": 425 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 0.9871951341629028, + "learning_rate": 4.906252428649708e-06, + "loss": 0.8366, + "step": 426 + }, + { + "epoch": 0.668580375782881, + "grad_norm": 0.8682780265808105, + "learning_rate": 4.905680057620072e-06, + "loss": 0.738, + "step": 427 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.8834647536277771, + "learning_rate": 4.9051059782053125e-06, + "loss": 0.8304, + "step": 428 + }, + { + "epoch": 0.6717118997912317, + "grad_norm": 1.0133378505706787, + "learning_rate": 4.904530190813113e-06, + "loss": 0.9369, + "step": 429 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 0.9967373013496399, + "learning_rate": 4.9039526958523676e-06, + "loss": 0.782, + "step": 430 + }, + { + "epoch": 0.6748434237995825, + "grad_norm": 1.0258032083511353, + "learning_rate": 4.90337349373318e-06, + "loss": 0.8214, + "step": 431 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 0.9814366102218628, + "learning_rate": 4.90279258486687e-06, + "loss": 0.8173, + "step": 432 + }, + { + "epoch": 0.6779749478079332, + "grad_norm": 0.9351440668106079, + "learning_rate": 4.902209969665966e-06, + "loss": 0.8815, + "step": 433 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 0.9490398168563843, + "learning_rate": 4.901625648544212e-06, + "loss": 0.8295, + "step": 434 + }, + { + "epoch": 0.6811064718162839, + "grad_norm": 1.0183534622192383, + "learning_rate": 4.9010396219165614e-06, + "loss": 0.8538, + "step": 435 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 0.9546058773994446, + "learning_rate": 4.900451890199179e-06, + "loss": 0.8385, + "step": 436 + }, + { + "epoch": 0.6842379958246346, + "grad_norm": 1.076120138168335, + "learning_rate": 4.8998624538094394e-06, + "loss": 0.8958, + "step": 437 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.8935732841491699, + "learning_rate": 4.899271313165929e-06, + "loss": 0.8674, + "step": 438 + }, + { + "epoch": 0.6873695198329853, + "grad_norm": 0.9574604630470276, + "learning_rate": 4.898678468688445e-06, + "loss": 0.8245, + "step": 439 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.931751549243927, + "learning_rate": 4.898083920797993e-06, + "loss": 0.8409, + "step": 440 + }, + { + "epoch": 0.6905010438413361, + "grad_norm": 0.937874436378479, + "learning_rate": 4.89748766991679e-06, + "loss": 0.8222, + "step": 441 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9606375694274902, + "learning_rate": 4.896889716468259e-06, + "loss": 0.8123, + "step": 442 + }, + { + "epoch": 0.6936325678496869, + "grad_norm": 0.8692249059677124, + "learning_rate": 4.896290060877038e-06, + "loss": 0.8532, + "step": 443 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 0.9597892165184021, + "learning_rate": 4.895688703568968e-06, + "loss": 0.8365, + "step": 444 + }, + { + "epoch": 0.6967640918580376, + "grad_norm": 0.958342432975769, + "learning_rate": 4.895085644971102e-06, + "loss": 0.8535, + "step": 445 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.9603601098060608, + "learning_rate": 4.894480885511699e-06, + "loss": 0.8227, + "step": 446 + }, + { + "epoch": 0.6998956158663883, + "grad_norm": 1.0010526180267334, + "learning_rate": 4.893874425620226e-06, + "loss": 0.8785, + "step": 447 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.0061219930648804, + "learning_rate": 4.893266265727361e-06, + "loss": 0.8336, + "step": 448 + }, + { + "epoch": 0.7030271398747391, + "grad_norm": 0.8967173099517822, + "learning_rate": 4.892656406264984e-06, + "loss": 0.8305, + "step": 449 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.962729275226593, + "learning_rate": 4.892044847666186e-06, + "loss": 0.7843, + "step": 450 + }, + { + "epoch": 0.7061586638830898, + "grad_norm": 1.0067342519760132, + "learning_rate": 4.891431590365262e-06, + "loss": 0.8111, + "step": 451 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 1.0252268314361572, + "learning_rate": 4.890816634797716e-06, + "loss": 0.8253, + "step": 452 + }, + { + "epoch": 0.7092901878914405, + "grad_norm": 0.9098564386367798, + "learning_rate": 4.890199981400255e-06, + "loss": 0.8526, + "step": 453 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 0.893135130405426, + "learning_rate": 4.889581630610793e-06, + "loss": 0.8217, + "step": 454 + }, + { + "epoch": 0.7124217118997912, + "grad_norm": 0.9225817322731018, + "learning_rate": 4.888961582868453e-06, + "loss": 0.8296, + "step": 455 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 0.973386824131012, + "learning_rate": 4.888339838613555e-06, + "loss": 0.7429, + "step": 456 + }, + { + "epoch": 0.715553235908142, + "grad_norm": 0.9829993844032288, + "learning_rate": 4.8877163982876306e-06, + "loss": 0.7405, + "step": 457 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 0.927728533744812, + "learning_rate": 4.887091262333414e-06, + "loss": 0.8257, + "step": 458 + }, + { + "epoch": 0.7186847599164927, + "grad_norm": 0.9419897794723511, + "learning_rate": 4.886464431194844e-06, + "loss": 0.7762, + "step": 459 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 0.9358823299407959, + "learning_rate": 4.885835905317061e-06, + "loss": 0.7794, + "step": 460 + }, + { + "epoch": 0.7218162839248434, + "grad_norm": 1.0290783643722534, + "learning_rate": 4.88520568514641e-06, + "loss": 0.8723, + "step": 461 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 0.9909068942070007, + "learning_rate": 4.884573771130442e-06, + "loss": 0.8302, + "step": 462 + }, + { + "epoch": 0.7249478079331941, + "grad_norm": 0.9972653985023499, + "learning_rate": 4.883940163717906e-06, + "loss": 0.7961, + "step": 463 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.9245638251304626, + "learning_rate": 4.883304863358757e-06, + "loss": 0.8057, + "step": 464 + }, + { + "epoch": 0.7280793319415448, + "grad_norm": 1.0097148418426514, + "learning_rate": 4.882667870504153e-06, + "loss": 0.7296, + "step": 465 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 0.9792670607566833, + "learning_rate": 4.882029185606448e-06, + "loss": 0.8006, + "step": 466 + }, + { + "epoch": 0.7312108559498957, + "grad_norm": 0.9308120608329773, + "learning_rate": 4.881388809119206e-06, + "loss": 0.7957, + "step": 467 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 0.9472301602363586, + "learning_rate": 4.880746741497187e-06, + "loss": 0.8984, + "step": 468 + }, + { + "epoch": 0.7343423799582464, + "grad_norm": 0.9898937344551086, + "learning_rate": 4.880102983196351e-06, + "loss": 0.8365, + "step": 469 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 0.9376612901687622, + "learning_rate": 4.879457534673863e-06, + "loss": 0.7838, + "step": 470 + }, + { + "epoch": 0.7374739039665971, + "grad_norm": 1.030616283416748, + "learning_rate": 4.878810396388085e-06, + "loss": 0.7698, + "step": 471 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.9583387970924377, + "learning_rate": 4.87816156879858e-06, + "loss": 0.7381, + "step": 472 + }, + { + "epoch": 0.7406054279749478, + "grad_norm": 1.00180184841156, + "learning_rate": 4.87751105236611e-06, + "loss": 0.8497, + "step": 473 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.9382562637329102, + "learning_rate": 4.876858847552638e-06, + "loss": 0.8698, + "step": 474 + }, + { + "epoch": 0.7437369519832986, + "grad_norm": 1.0104063749313354, + "learning_rate": 4.8762049548213255e-06, + "loss": 0.8444, + "step": 475 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 0.9321765303611755, + "learning_rate": 4.87554937463653e-06, + "loss": 0.8027, + "step": 476 + }, + { + "epoch": 0.7468684759916493, + "grad_norm": 0.9527666568756104, + "learning_rate": 4.874892107463811e-06, + "loss": 0.8069, + "step": 477 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 0.9576907157897949, + "learning_rate": 4.874233153769924e-06, + "loss": 0.774, + "step": 478 + }, + { + "epoch": 0.75, + "grad_norm": 1.0516161918640137, + "learning_rate": 4.873572514022822e-06, + "loss": 0.8176, + "step": 479 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.026928186416626, + "learning_rate": 4.872910188691657e-06, + "loss": 0.8797, + "step": 480 + }, + { + "epoch": 0.7531315240083507, + "grad_norm": 0.9312179088592529, + "learning_rate": 4.872246178246776e-06, + "loss": 0.7747, + "step": 481 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0676255226135254, + "learning_rate": 4.871580483159725e-06, + "loss": 0.7801, + "step": 482 + }, + { + "epoch": 0.7562630480167014, + "grad_norm": 1.066056489944458, + "learning_rate": 4.870913103903245e-06, + "loss": 0.7837, + "step": 483 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9388124346733093, + "learning_rate": 4.87024404095127e-06, + "loss": 0.7801, + "step": 484 + }, + { + "epoch": 0.7593945720250522, + "grad_norm": 0.9879791140556335, + "learning_rate": 4.8695732947789356e-06, + "loss": 0.7849, + "step": 485 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0002318620681763, + "learning_rate": 4.868900865862569e-06, + "loss": 0.8219, + "step": 486 + }, + { + "epoch": 0.7625260960334029, + "grad_norm": 0.9599125385284424, + "learning_rate": 4.8682267546796945e-06, + "loss": 0.8888, + "step": 487 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 0.9099973440170288, + "learning_rate": 4.867550961709027e-06, + "loss": 0.7585, + "step": 488 + }, + { + "epoch": 0.7656576200417536, + "grad_norm": 0.9715955257415771, + "learning_rate": 4.866873487430481e-06, + "loss": 0.8237, + "step": 489 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 1.0194706916809082, + "learning_rate": 4.86619433232516e-06, + "loss": 0.7708, + "step": 490 + }, + { + "epoch": 0.7687891440501043, + "grad_norm": 1.044005036354065, + "learning_rate": 4.865513496875367e-06, + "loss": 0.8, + "step": 491 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.9776361584663391, + "learning_rate": 4.8648309815645915e-06, + "loss": 0.7696, + "step": 492 + }, + { + "epoch": 0.7719206680584552, + "grad_norm": 0.9213030934333801, + "learning_rate": 4.864146786877521e-06, + "loss": 0.8558, + "step": 493 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.9361440539360046, + "learning_rate": 4.8634609133000336e-06, + "loss": 0.7851, + "step": 494 + }, + { + "epoch": 0.7750521920668059, + "grad_norm": 1.0199540853500366, + "learning_rate": 4.8627733613192e-06, + "loss": 0.8229, + "step": 495 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 1.0244604349136353, + "learning_rate": 4.862084131423283e-06, + "loss": 0.772, + "step": 496 + }, + { + "epoch": 0.7781837160751566, + "grad_norm": 0.9414368271827698, + "learning_rate": 4.861393224101736e-06, + "loss": 0.7579, + "step": 497 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9198965430259705, + "learning_rate": 4.860700639845205e-06, + "loss": 0.855, + "step": 498 + }, + { + "epoch": 0.7813152400835073, + "grad_norm": 0.9398712515830994, + "learning_rate": 4.8600063791455275e-06, + "loss": 0.7824, + "step": 499 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.9544969201087952, + "learning_rate": 4.8593104424957275e-06, + "loss": 0.8104, + "step": 500 + }, + { + "epoch": 0.784446764091858, + "grad_norm": 0.9522370100021362, + "learning_rate": 4.858612830390024e-06, + "loss": 0.813, + "step": 501 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.9830234050750732, + "learning_rate": 4.857913543323823e-06, + "loss": 0.8656, + "step": 502 + }, + { + "epoch": 0.7875782881002088, + "grad_norm": 1.0527008771896362, + "learning_rate": 4.8572125817937195e-06, + "loss": 0.8417, + "step": 503 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 0.9772782921791077, + "learning_rate": 4.856509946297501e-06, + "loss": 0.7983, + "step": 504 + }, + { + "epoch": 0.7907098121085595, + "grad_norm": 0.9250578880310059, + "learning_rate": 4.855805637334141e-06, + "loss": 0.7391, + "step": 505 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9792859554290771, + "learning_rate": 4.855099655403802e-06, + "loss": 0.7753, + "step": 506 + }, + { + "epoch": 0.7938413361169102, + "grad_norm": 0.8876479864120483, + "learning_rate": 4.854392001007834e-06, + "loss": 0.8249, + "step": 507 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.9336771368980408, + "learning_rate": 4.853682674648775e-06, + "loss": 0.8462, + "step": 508 + }, + { + "epoch": 0.7969728601252609, + "grad_norm": 0.9732388854026794, + "learning_rate": 4.852971676830352e-06, + "loss": 0.8791, + "step": 509 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.9135870337486267, + "learning_rate": 4.852259008057476e-06, + "loss": 0.7279, + "step": 510 + }, + { + "epoch": 0.8001043841336117, + "grad_norm": 0.9782464504241943, + "learning_rate": 4.8515446688362476e-06, + "loss": 0.7797, + "step": 511 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 0.9571824073791504, + "learning_rate": 4.85082865967395e-06, + "loss": 0.8349, + "step": 512 + }, + { + "epoch": 0.8032359081419624, + "grad_norm": 1.011786937713623, + "learning_rate": 4.850110981079057e-06, + "loss": 0.8057, + "step": 513 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 0.9418787956237793, + "learning_rate": 4.8493916335612255e-06, + "loss": 0.8592, + "step": 514 + }, + { + "epoch": 0.8063674321503131, + "grad_norm": 0.9292604923248291, + "learning_rate": 4.848670617631296e-06, + "loss": 0.7471, + "step": 515 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.9303188323974609, + "learning_rate": 4.847947933801296e-06, + "loss": 0.8768, + "step": 516 + }, + { + "epoch": 0.8094989561586639, + "grad_norm": 0.9897359609603882, + "learning_rate": 4.847223582584437e-06, + "loss": 0.8343, + "step": 517 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.9695891737937927, + "learning_rate": 4.846497564495114e-06, + "loss": 0.8318, + "step": 518 + }, + { + "epoch": 0.8126304801670147, + "grad_norm": 1.0175273418426514, + "learning_rate": 4.845769880048906e-06, + "loss": 0.9137, + "step": 519 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9497922658920288, + "learning_rate": 4.845040529762577e-06, + "loss": 0.6549, + "step": 520 + }, + { + "epoch": 0.8157620041753654, + "grad_norm": 1.0647574663162231, + "learning_rate": 4.844309514154071e-06, + "loss": 0.8243, + "step": 521 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.9710965156555176, + "learning_rate": 4.843576833742517e-06, + "loss": 0.7864, + "step": 522 + }, + { + "epoch": 0.8188935281837161, + "grad_norm": 0.9881441593170166, + "learning_rate": 4.842842489048225e-06, + "loss": 0.7799, + "step": 523 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 0.9455597400665283, + "learning_rate": 4.842106480592687e-06, + "loss": 0.7943, + "step": 524 + }, + { + "epoch": 0.8220250521920668, + "grad_norm": 0.9948442578315735, + "learning_rate": 4.841368808898577e-06, + "loss": 0.8587, + "step": 525 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 0.9959186315536499, + "learning_rate": 4.84062947448975e-06, + "loss": 0.8477, + "step": 526 + }, + { + "epoch": 0.8251565762004175, + "grad_norm": 0.9791848659515381, + "learning_rate": 4.839888477891243e-06, + "loss": 0.7568, + "step": 527 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 1.1499426364898682, + "learning_rate": 4.839145819629269e-06, + "loss": 0.8296, + "step": 528 + }, + { + "epoch": 0.8282881002087683, + "grad_norm": 0.9269506335258484, + "learning_rate": 4.8384015002312265e-06, + "loss": 0.7935, + "step": 529 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.970565140247345, + "learning_rate": 4.837655520225692e-06, + "loss": 0.8138, + "step": 530 + }, + { + "epoch": 0.831419624217119, + "grad_norm": 0.9852119088172913, + "learning_rate": 4.836907880142418e-06, + "loss": 0.7909, + "step": 531 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.927129328250885, + "learning_rate": 4.836158580512339e-06, + "loss": 0.8105, + "step": 532 + }, + { + "epoch": 0.8345511482254697, + "grad_norm": 1.0711781978607178, + "learning_rate": 4.835407621867569e-06, + "loss": 0.8241, + "step": 533 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 1.0472421646118164, + "learning_rate": 4.834655004741397e-06, + "loss": 0.7923, + "step": 534 + }, + { + "epoch": 0.8376826722338204, + "grad_norm": 0.9761579632759094, + "learning_rate": 4.833900729668293e-06, + "loss": 0.8203, + "step": 535 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 1.0559637546539307, + "learning_rate": 4.833144797183902e-06, + "loss": 0.8915, + "step": 536 + }, + { + "epoch": 0.8408141962421712, + "grad_norm": 0.9951052069664001, + "learning_rate": 4.832387207825045e-06, + "loss": 0.8253, + "step": 537 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.894079864025116, + "learning_rate": 4.831627962129725e-06, + "loss": 0.7479, + "step": 538 + }, + { + "epoch": 0.843945720250522, + "grad_norm": 0.9832531213760376, + "learning_rate": 4.830867060637115e-06, + "loss": 0.8484, + "step": 539 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0477591753005981, + "learning_rate": 4.8301045038875665e-06, + "loss": 0.8052, + "step": 540 + }, + { + "epoch": 0.8470772442588727, + "grad_norm": 0.9388875365257263, + "learning_rate": 4.829340292422607e-06, + "loss": 0.7998, + "step": 541 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 0.8975532054901123, + "learning_rate": 4.8285744267849395e-06, + "loss": 0.848, + "step": 542 + }, + { + "epoch": 0.8502087682672234, + "grad_norm": 0.9211076498031616, + "learning_rate": 4.8278069075184396e-06, + "loss": 0.8337, + "step": 543 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.970134437084198, + "learning_rate": 4.827037735168159e-06, + "loss": 0.7843, + "step": 544 + }, + { + "epoch": 0.8533402922755741, + "grad_norm": 0.9714353084564209, + "learning_rate": 4.826266910280322e-06, + "loss": 0.6991, + "step": 545 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 0.8607689142227173, + "learning_rate": 4.8254944334023265e-06, + "loss": 0.7314, + "step": 546 + }, + { + "epoch": 0.8564718162839249, + "grad_norm": 0.9843030571937561, + "learning_rate": 4.824720305082747e-06, + "loss": 0.7497, + "step": 547 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 0.9638474583625793, + "learning_rate": 4.823944525871324e-06, + "loss": 0.776, + "step": 548 + }, + { + "epoch": 0.8596033402922756, + "grad_norm": 0.9342365860939026, + "learning_rate": 4.823167096318979e-06, + "loss": 0.77, + "step": 549 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9608579874038696, + "learning_rate": 4.822388016977797e-06, + "loss": 0.8281, + "step": 550 + }, + { + "epoch": 0.8627348643006263, + "grad_norm": 0.8996383547782898, + "learning_rate": 4.8216072884010395e-06, + "loss": 0.7519, + "step": 551 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 0.9922810196876526, + "learning_rate": 4.820824911143139e-06, + "loss": 0.8714, + "step": 552 + }, + { + "epoch": 0.865866388308977, + "grad_norm": 0.9861842393875122, + "learning_rate": 4.820040885759697e-06, + "loss": 0.7998, + "step": 553 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 1.013168215751648, + "learning_rate": 4.819255212807486e-06, + "loss": 0.8291, + "step": 554 + }, + { + "epoch": 0.8689979123173278, + "grad_norm": 0.9695484638214111, + "learning_rate": 4.818467892844451e-06, + "loss": 0.8257, + "step": 555 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 1.0018948316574097, + "learning_rate": 4.817678926429702e-06, + "loss": 0.7551, + "step": 556 + }, + { + "epoch": 0.8721294363256785, + "grad_norm": 1.0259422063827515, + "learning_rate": 4.816888314123521e-06, + "loss": 0.8582, + "step": 557 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9045076966285706, + "learning_rate": 4.81609605648736e-06, + "loss": 0.811, + "step": 558 + }, + { + "epoch": 0.8752609603340292, + "grad_norm": 0.9112316370010376, + "learning_rate": 4.815302154083837e-06, + "loss": 0.818, + "step": 559 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0060032606124878, + "learning_rate": 4.814506607476739e-06, + "loss": 0.766, + "step": 560 + }, + { + "epoch": 0.8783924843423799, + "grad_norm": 0.9126970171928406, + "learning_rate": 4.813709417231021e-06, + "loss": 0.8903, + "step": 561 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.0825906991958618, + "learning_rate": 4.812910583912804e-06, + "loss": 0.8466, + "step": 562 + }, + { + "epoch": 0.8815240083507306, + "grad_norm": 0.9530975818634033, + "learning_rate": 4.812110108089377e-06, + "loss": 0.8163, + "step": 563 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.9566405415534973, + "learning_rate": 4.8113079903291955e-06, + "loss": 0.8244, + "step": 564 + }, + { + "epoch": 0.8846555323590815, + "grad_norm": 0.919980525970459, + "learning_rate": 4.81050423120188e-06, + "loss": 0.8335, + "step": 565 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.9123868346214294, + "learning_rate": 4.809698831278217e-06, + "loss": 0.7712, + "step": 566 + }, + { + "epoch": 0.8877870563674322, + "grad_norm": 0.9589663743972778, + "learning_rate": 4.8088917911301595e-06, + "loss": 0.8467, + "step": 567 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 0.9052798748016357, + "learning_rate": 4.808083111330823e-06, + "loss": 0.8078, + "step": 568 + }, + { + "epoch": 0.8909185803757829, + "grad_norm": 0.9250494241714478, + "learning_rate": 4.807272792454489e-06, + "loss": 0.8536, + "step": 569 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.8451298475265503, + "learning_rate": 4.806460835076603e-06, + "loss": 0.7752, + "step": 570 + }, + { + "epoch": 0.8940501043841336, + "grad_norm": 1.0324240922927856, + "learning_rate": 4.805647239773773e-06, + "loss": 0.7871, + "step": 571 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 0.9338528513908386, + "learning_rate": 4.804832007123771e-06, + "loss": 0.8082, + "step": 572 + }, + { + "epoch": 0.8971816283924844, + "grad_norm": 1.0080739259719849, + "learning_rate": 4.804015137705531e-06, + "loss": 0.7636, + "step": 573 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9817036986351013, + "learning_rate": 4.803196632099152e-06, + "loss": 0.8599, + "step": 574 + }, + { + "epoch": 0.9003131524008351, + "grad_norm": 0.9284219145774841, + "learning_rate": 4.80237649088589e-06, + "loss": 0.8152, + "step": 575 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.8922317028045654, + "learning_rate": 4.801554714648166e-06, + "loss": 0.8146, + "step": 576 + }, + { + "epoch": 0.9034446764091858, + "grad_norm": 0.9465014338493347, + "learning_rate": 4.800731303969565e-06, + "loss": 0.7663, + "step": 577 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.0325525999069214, + "learning_rate": 4.799906259434824e-06, + "loss": 0.8458, + "step": 578 + }, + { + "epoch": 0.9065762004175365, + "grad_norm": 0.9366307258605957, + "learning_rate": 4.7990795816298485e-06, + "loss": 0.7992, + "step": 579 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 0.9923847317695618, + "learning_rate": 4.7982512711416995e-06, + "loss": 0.7256, + "step": 580 + }, + { + "epoch": 0.9097077244258872, + "grad_norm": 1.0387133359909058, + "learning_rate": 4.7974213285586e-06, + "loss": 0.7766, + "step": 581 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.9617043733596802, + "learning_rate": 4.79658975446993e-06, + "loss": 0.7602, + "step": 582 + }, + { + "epoch": 0.912839248434238, + "grad_norm": 0.9530353546142578, + "learning_rate": 4.795756549466229e-06, + "loss": 0.7922, + "step": 583 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.9305918216705322, + "learning_rate": 4.794921714139194e-06, + "loss": 0.8377, + "step": 584 + }, + { + "epoch": 0.9159707724425887, + "grad_norm": 0.899860143661499, + "learning_rate": 4.794085249081682e-06, + "loss": 0.8343, + "step": 585 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.9024940729141235, + "learning_rate": 4.793247154887703e-06, + "loss": 0.8271, + "step": 586 + }, + { + "epoch": 0.9191022964509394, + "grad_norm": 0.980869472026825, + "learning_rate": 4.7924074321524286e-06, + "loss": 0.7468, + "step": 587 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.0506541728973389, + "learning_rate": 4.791566081472185e-06, + "loss": 0.9207, + "step": 588 + }, + { + "epoch": 0.9222338204592901, + "grad_norm": 1.0723917484283447, + "learning_rate": 4.790723103444454e-06, + "loss": 0.7615, + "step": 589 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 1.0455081462860107, + "learning_rate": 4.789878498667873e-06, + "loss": 0.8115, + "step": 590 + }, + { + "epoch": 0.925365344467641, + "grad_norm": 0.9729639291763306, + "learning_rate": 4.789032267742234e-06, + "loss": 0.8622, + "step": 591 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.9695239067077637, + "learning_rate": 4.788184411268488e-06, + "loss": 0.8165, + "step": 592 + }, + { + "epoch": 0.9284968684759917, + "grad_norm": 1.0099278688430786, + "learning_rate": 4.7873349298487345e-06, + "loss": 0.759, + "step": 593 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 0.9377149939537048, + "learning_rate": 4.786483824086231e-06, + "loss": 0.8323, + "step": 594 + }, + { + "epoch": 0.9316283924843424, + "grad_norm": 0.8885690569877625, + "learning_rate": 4.785631094585387e-06, + "loss": 0.8448, + "step": 595 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 0.9133430123329163, + "learning_rate": 4.784776741951766e-06, + "loss": 0.8135, + "step": 596 + }, + { + "epoch": 0.9347599164926931, + "grad_norm": 1.015851378440857, + "learning_rate": 4.783920766792082e-06, + "loss": 0.8464, + "step": 597 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.8883825540542603, + "learning_rate": 4.783063169714203e-06, + "loss": 0.7445, + "step": 598 + }, + { + "epoch": 0.9378914405010439, + "grad_norm": 0.9872002005577087, + "learning_rate": 4.782203951327149e-06, + "loss": 0.8293, + "step": 599 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 1.0305813550949097, + "learning_rate": 4.781343112241091e-06, + "loss": 0.8606, + "step": 600 + }, + { + "epoch": 0.9410229645093946, + "grad_norm": 0.9667104482650757, + "learning_rate": 4.780480653067351e-06, + "loss": 0.8057, + "step": 601 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 1.067185640335083, + "learning_rate": 4.7796165744184e-06, + "loss": 0.849, + "step": 602 + }, + { + "epoch": 0.9441544885177453, + "grad_norm": 0.9017406105995178, + "learning_rate": 4.778750876907862e-06, + "loss": 0.7927, + "step": 603 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.9827666878700256, + "learning_rate": 4.777883561150508e-06, + "loss": 0.8087, + "step": 604 + }, + { + "epoch": 0.947286012526096, + "grad_norm": 0.9627811908721924, + "learning_rate": 4.77701462776226e-06, + "loss": 0.8488, + "step": 605 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0321338176727295, + "learning_rate": 4.776144077360186e-06, + "loss": 0.8289, + "step": 606 + }, + { + "epoch": 0.9504175365344467, + "grad_norm": 1.0012633800506592, + "learning_rate": 4.775271910562508e-06, + "loss": 0.7184, + "step": 607 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.9186056852340698, + "learning_rate": 4.7743981279885875e-06, + "loss": 0.8727, + "step": 608 + }, + { + "epoch": 0.9535490605427975, + "grad_norm": 0.910044252872467, + "learning_rate": 4.773522730258942e-06, + "loss": 0.8455, + "step": 609 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.9342944025993347, + "learning_rate": 4.77264571799523e-06, + "loss": 0.7548, + "step": 610 + }, + { + "epoch": 0.9566805845511482, + "grad_norm": 0.9609628319740295, + "learning_rate": 4.771767091820259e-06, + "loss": 0.8294, + "step": 611 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9708508253097534, + "learning_rate": 4.770886852357983e-06, + "loss": 0.7857, + "step": 612 + }, + { + "epoch": 0.9598121085594989, + "grad_norm": 1.0910807847976685, + "learning_rate": 4.7700050002334995e-06, + "loss": 0.8162, + "step": 613 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 0.9703771471977234, + "learning_rate": 4.769121536073054e-06, + "loss": 0.7205, + "step": 614 + }, + { + "epoch": 0.9629436325678496, + "grad_norm": 1.027938961982727, + "learning_rate": 4.768236460504035e-06, + "loss": 0.7641, + "step": 615 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 1.0003989934921265, + "learning_rate": 4.767349774154974e-06, + "loss": 0.7588, + "step": 616 + }, + { + "epoch": 0.9660751565762005, + "grad_norm": 0.9572570323944092, + "learning_rate": 4.766461477655552e-06, + "loss": 0.8274, + "step": 617 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.977287769317627, + "learning_rate": 4.765571571636586e-06, + "loss": 0.7856, + "step": 618 + }, + { + "epoch": 0.9692066805845512, + "grad_norm": 0.9582562446594238, + "learning_rate": 4.764680056730042e-06, + "loss": 0.8102, + "step": 619 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.9842986464500427, + "learning_rate": 4.763786933569025e-06, + "loss": 0.7515, + "step": 620 + }, + { + "epoch": 0.9723382045929019, + "grad_norm": 0.9955743551254272, + "learning_rate": 4.762892202787782e-06, + "loss": 0.7398, + "step": 621 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 1.0011107921600342, + "learning_rate": 4.761995865021706e-06, + "loss": 0.8182, + "step": 622 + }, + { + "epoch": 0.9754697286012526, + "grad_norm": 0.9586448669433594, + "learning_rate": 4.761097920907325e-06, + "loss": 0.8434, + "step": 623 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.979123055934906, + "learning_rate": 4.760198371082312e-06, + "loss": 0.7954, + "step": 624 + }, + { + "epoch": 0.9786012526096033, + "grad_norm": 1.0081250667572021, + "learning_rate": 4.75929721618548e-06, + "loss": 0.7931, + "step": 625 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 1.027185320854187, + "learning_rate": 4.75839445685678e-06, + "loss": 0.7339, + "step": 626 + }, + { + "epoch": 0.9817327766179541, + "grad_norm": 1.0553215742111206, + "learning_rate": 4.757490093737303e-06, + "loss": 0.7639, + "step": 627 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 0.8578981757164001, + "learning_rate": 4.75658412746928e-06, + "loss": 0.8128, + "step": 628 + }, + { + "epoch": 0.9848643006263048, + "grad_norm": 1.0143071413040161, + "learning_rate": 4.75567655869608e-06, + "loss": 0.8169, + "step": 629 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 0.9743554592132568, + "learning_rate": 4.754767388062208e-06, + "loss": 0.8179, + "step": 630 + }, + { + "epoch": 0.9879958246346555, + "grad_norm": 1.0276602506637573, + "learning_rate": 4.7538566162133105e-06, + "loss": 0.8009, + "step": 631 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.9296634793281555, + "learning_rate": 4.752944243796167e-06, + "loss": 0.8262, + "step": 632 + }, + { + "epoch": 0.9911273486430062, + "grad_norm": 0.9216791987419128, + "learning_rate": 4.752030271458696e-06, + "loss": 0.8929, + "step": 633 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 0.9272832274436951, + "learning_rate": 4.751114699849954e-06, + "loss": 0.7396, + "step": 634 + }, + { + "epoch": 0.994258872651357, + "grad_norm": 0.9121203422546387, + "learning_rate": 4.750197529620127e-06, + "loss": 0.8311, + "step": 635 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.999600887298584, + "learning_rate": 4.7492787614205425e-06, + "loss": 0.8125, + "step": 636 + }, + { + "epoch": 0.9973903966597077, + "grad_norm": 1.012978196144104, + "learning_rate": 4.748358395903661e-06, + "loss": 0.7966, + "step": 637 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.9767918586730957, + "learning_rate": 4.747436433723075e-06, + "loss": 0.8024, + "step": 638 + }, + { + "epoch": 1.0015657620041754, + "grad_norm": 1.8302738666534424, + "learning_rate": 4.7465128755335135e-06, + "loss": 1.6427, + "step": 639 + }, + { + "epoch": 1.0031315240083507, + "grad_norm": 0.8704890012741089, + "learning_rate": 4.7455877219908386e-06, + "loss": 0.7766, + "step": 640 + }, + { + "epoch": 1.004697286012526, + "grad_norm": 0.9848434925079346, + "learning_rate": 4.744660973752044e-06, + "loss": 0.8314, + "step": 641 + }, + { + "epoch": 1.0062630480167014, + "grad_norm": 1.054824709892273, + "learning_rate": 4.743732631475258e-06, + "loss": 0.797, + "step": 642 + }, + { + "epoch": 1.0078288100208768, + "grad_norm": 0.9286114573478699, + "learning_rate": 4.742802695819737e-06, + "loss": 0.764, + "step": 643 + }, + { + "epoch": 1.0093945720250521, + "grad_norm": 0.8625781536102295, + "learning_rate": 4.7418711674458735e-06, + "loss": 0.6691, + "step": 644 + }, + { + "epoch": 1.0109603340292275, + "grad_norm": 0.8710452914237976, + "learning_rate": 4.740938047015188e-06, + "loss": 0.7277, + "step": 645 + }, + { + "epoch": 1.0125260960334028, + "grad_norm": 0.8873130679130554, + "learning_rate": 4.740003335190334e-06, + "loss": 0.6531, + "step": 646 + }, + { + "epoch": 1.0140918580375784, + "grad_norm": 0.9777164459228516, + "learning_rate": 4.7390670326350925e-06, + "loss": 0.8047, + "step": 647 + }, + { + "epoch": 1.0156576200417538, + "grad_norm": 0.886025071144104, + "learning_rate": 4.738129140014377e-06, + "loss": 0.715, + "step": 648 + }, + { + "epoch": 1.017223382045929, + "grad_norm": 0.9825332164764404, + "learning_rate": 4.737189657994226e-06, + "loss": 0.7783, + "step": 649 + }, + { + "epoch": 1.0187891440501045, + "grad_norm": 0.9331428408622742, + "learning_rate": 4.736248587241812e-06, + "loss": 0.7598, + "step": 650 + }, + { + "epoch": 1.0203549060542798, + "grad_norm": 0.9406535625457764, + "learning_rate": 4.735305928425431e-06, + "loss": 0.7989, + "step": 651 + }, + { + "epoch": 1.0219206680584552, + "grad_norm": 1.0644745826721191, + "learning_rate": 4.734361682214511e-06, + "loss": 0.8135, + "step": 652 + }, + { + "epoch": 1.0234864300626305, + "grad_norm": 0.9037052392959595, + "learning_rate": 4.733415849279603e-06, + "loss": 0.7051, + "step": 653 + }, + { + "epoch": 1.0250521920668059, + "grad_norm": 1.000563144683838, + "learning_rate": 4.7324684302923865e-06, + "loss": 0.7427, + "step": 654 + }, + { + "epoch": 1.0266179540709812, + "grad_norm": 0.9181354641914368, + "learning_rate": 4.731519425925671e-06, + "loss": 0.7236, + "step": 655 + }, + { + "epoch": 1.0281837160751566, + "grad_norm": 1.0283046960830688, + "learning_rate": 4.730568836853384e-06, + "loss": 0.7998, + "step": 656 + }, + { + "epoch": 1.029749478079332, + "grad_norm": 0.8692315220832825, + "learning_rate": 4.729616663750587e-06, + "loss": 0.7974, + "step": 657 + }, + { + "epoch": 1.0313152400835073, + "grad_norm": 0.928920567035675, + "learning_rate": 4.72866290729346e-06, + "loss": 0.7199, + "step": 658 + }, + { + "epoch": 1.0328810020876826, + "grad_norm": 0.9066957235336304, + "learning_rate": 4.727707568159311e-06, + "loss": 0.791, + "step": 659 + }, + { + "epoch": 1.034446764091858, + "grad_norm": 1.0086729526519775, + "learning_rate": 4.726750647026569e-06, + "loss": 0.7113, + "step": 660 + }, + { + "epoch": 1.0360125260960333, + "grad_norm": 0.931856632232666, + "learning_rate": 4.72579214457479e-06, + "loss": 0.7977, + "step": 661 + }, + { + "epoch": 1.0375782881002087, + "grad_norm": 0.9191203713417053, + "learning_rate": 4.724832061484651e-06, + "loss": 0.7979, + "step": 662 + }, + { + "epoch": 1.039144050104384, + "grad_norm": 0.9614707827568054, + "learning_rate": 4.72387039843795e-06, + "loss": 0.6994, + "step": 663 + }, + { + "epoch": 1.0407098121085594, + "grad_norm": 1.014823317527771, + "learning_rate": 4.72290715611761e-06, + "loss": 0.7539, + "step": 664 + }, + { + "epoch": 1.042275574112735, + "grad_norm": 0.9177424311637878, + "learning_rate": 4.721942335207673e-06, + "loss": 0.6625, + "step": 665 + }, + { + "epoch": 1.0438413361169103, + "grad_norm": 1.0253031253814697, + "learning_rate": 4.720975936393305e-06, + "loss": 0.7747, + "step": 666 + }, + { + "epoch": 1.0454070981210857, + "grad_norm": 1.0004568099975586, + "learning_rate": 4.720007960360788e-06, + "loss": 0.7008, + "step": 667 + }, + { + "epoch": 1.046972860125261, + "grad_norm": 1.1344349384307861, + "learning_rate": 4.719038407797529e-06, + "loss": 0.7493, + "step": 668 + }, + { + "epoch": 1.0485386221294364, + "grad_norm": 0.9944799542427063, + "learning_rate": 4.718067279392052e-06, + "loss": 0.7823, + "step": 669 + }, + { + "epoch": 1.0501043841336117, + "grad_norm": 0.9469346404075623, + "learning_rate": 4.717094575833999e-06, + "loss": 0.7419, + "step": 670 + }, + { + "epoch": 1.051670146137787, + "grad_norm": 1.015339732170105, + "learning_rate": 4.716120297814133e-06, + "loss": 0.7168, + "step": 671 + }, + { + "epoch": 1.0532359081419624, + "grad_norm": 0.9544686675071716, + "learning_rate": 4.715144446024333e-06, + "loss": 0.791, + "step": 672 + }, + { + "epoch": 1.0548016701461378, + "grad_norm": 1.0285652875900269, + "learning_rate": 4.714167021157599e-06, + "loss": 0.7649, + "step": 673 + }, + { + "epoch": 1.0563674321503131, + "grad_norm": 0.9366872906684875, + "learning_rate": 4.713188023908044e-06, + "loss": 0.7524, + "step": 674 + }, + { + "epoch": 1.0579331941544885, + "grad_norm": 1.0348883867263794, + "learning_rate": 4.712207454970899e-06, + "loss": 0.6965, + "step": 675 + }, + { + "epoch": 1.0594989561586639, + "grad_norm": 0.8716043829917908, + "learning_rate": 4.711225315042513e-06, + "loss": 0.6264, + "step": 676 + }, + { + "epoch": 1.0610647181628392, + "grad_norm": 1.0421313047409058, + "learning_rate": 4.710241604820348e-06, + "loss": 0.7548, + "step": 677 + }, + { + "epoch": 1.0626304801670146, + "grad_norm": 1.0886270999908447, + "learning_rate": 4.709256325002983e-06, + "loss": 0.6352, + "step": 678 + }, + { + "epoch": 1.06419624217119, + "grad_norm": 0.9752539396286011, + "learning_rate": 4.708269476290111e-06, + "loss": 0.7764, + "step": 679 + }, + { + "epoch": 1.0657620041753653, + "grad_norm": 1.0658339262008667, + "learning_rate": 4.7072810593825385e-06, + "loss": 0.766, + "step": 680 + }, + { + "epoch": 1.0673277661795406, + "grad_norm": 1.0299617052078247, + "learning_rate": 4.706291074982187e-06, + "loss": 0.7558, + "step": 681 + }, + { + "epoch": 1.068893528183716, + "grad_norm": 0.9171326160430908, + "learning_rate": 4.705299523792089e-06, + "loss": 0.7084, + "step": 682 + }, + { + "epoch": 1.0704592901878915, + "grad_norm": 0.9924243688583374, + "learning_rate": 4.7043064065163925e-06, + "loss": 0.8393, + "step": 683 + }, + { + "epoch": 1.072025052192067, + "grad_norm": 1.0243877172470093, + "learning_rate": 4.703311723860356e-06, + "loss": 0.7755, + "step": 684 + }, + { + "epoch": 1.0735908141962422, + "grad_norm": 1.0019137859344482, + "learning_rate": 4.702315476530349e-06, + "loss": 0.7254, + "step": 685 + }, + { + "epoch": 1.0751565762004176, + "grad_norm": 0.8541996479034424, + "learning_rate": 4.701317665233853e-06, + "loss": 0.7389, + "step": 686 + }, + { + "epoch": 1.076722338204593, + "grad_norm": 1.1119922399520874, + "learning_rate": 4.700318290679462e-06, + "loss": 0.7803, + "step": 687 + }, + { + "epoch": 1.0782881002087683, + "grad_norm": 1.078452706336975, + "learning_rate": 4.699317353576876e-06, + "loss": 0.756, + "step": 688 + }, + { + "epoch": 1.0798538622129437, + "grad_norm": 1.0079457759857178, + "learning_rate": 4.698314854636908e-06, + "loss": 0.7365, + "step": 689 + }, + { + "epoch": 1.081419624217119, + "grad_norm": 0.931423544883728, + "learning_rate": 4.6973107945714795e-06, + "loss": 0.7881, + "step": 690 + }, + { + "epoch": 1.0829853862212944, + "grad_norm": 1.0135142803192139, + "learning_rate": 4.696305174093619e-06, + "loss": 0.7884, + "step": 691 + }, + { + "epoch": 1.0845511482254697, + "grad_norm": 1.1194610595703125, + "learning_rate": 4.695297993917465e-06, + "loss": 0.7969, + "step": 692 + }, + { + "epoch": 1.086116910229645, + "grad_norm": 0.9972780346870422, + "learning_rate": 4.6942892547582625e-06, + "loss": 0.5873, + "step": 693 + }, + { + "epoch": 1.0876826722338204, + "grad_norm": 1.0711641311645508, + "learning_rate": 4.693278957332364e-06, + "loss": 0.7273, + "step": 694 + }, + { + "epoch": 1.0892484342379958, + "grad_norm": 0.9708715081214905, + "learning_rate": 4.6922671023572305e-06, + "loss": 0.6786, + "step": 695 + }, + { + "epoch": 1.0908141962421711, + "grad_norm": 0.9773644804954529, + "learning_rate": 4.691253690551424e-06, + "loss": 0.7839, + "step": 696 + }, + { + "epoch": 1.0923799582463465, + "grad_norm": 1.039009690284729, + "learning_rate": 4.690238722634618e-06, + "loss": 0.6807, + "step": 697 + }, + { + "epoch": 1.0939457202505218, + "grad_norm": 1.140777349472046, + "learning_rate": 4.689222199327586e-06, + "loss": 0.7506, + "step": 698 + }, + { + "epoch": 1.0955114822546972, + "grad_norm": 0.9779882431030273, + "learning_rate": 4.688204121352211e-06, + "loss": 0.7732, + "step": 699 + }, + { + "epoch": 1.0970772442588728, + "grad_norm": 1.0412248373031616, + "learning_rate": 4.687184489431476e-06, + "loss": 0.7034, + "step": 700 + }, + { + "epoch": 1.0986430062630481, + "grad_norm": 0.9802400469779968, + "learning_rate": 4.686163304289469e-06, + "loss": 0.729, + "step": 701 + }, + { + "epoch": 1.1002087682672235, + "grad_norm": 1.0531584024429321, + "learning_rate": 4.6851405666513804e-06, + "loss": 0.7635, + "step": 702 + }, + { + "epoch": 1.1017745302713988, + "grad_norm": 0.983923077583313, + "learning_rate": 4.684116277243505e-06, + "loss": 0.7062, + "step": 703 + }, + { + "epoch": 1.1033402922755742, + "grad_norm": 0.9886301159858704, + "learning_rate": 4.683090436793238e-06, + "loss": 0.7299, + "step": 704 + }, + { + "epoch": 1.1049060542797495, + "grad_norm": 1.0247817039489746, + "learning_rate": 4.682063046029075e-06, + "loss": 0.729, + "step": 705 + }, + { + "epoch": 1.1064718162839249, + "grad_norm": 0.9860950708389282, + "learning_rate": 4.681034105680616e-06, + "loss": 0.7392, + "step": 706 + }, + { + "epoch": 1.1080375782881002, + "grad_norm": 0.9543380737304688, + "learning_rate": 4.680003616478556e-06, + "loss": 0.7061, + "step": 707 + }, + { + "epoch": 1.1096033402922756, + "grad_norm": 1.0253225564956665, + "learning_rate": 4.678971579154698e-06, + "loss": 0.6577, + "step": 708 + }, + { + "epoch": 1.111169102296451, + "grad_norm": 0.9013616442680359, + "learning_rate": 4.677937994441935e-06, + "loss": 0.7422, + "step": 709 + }, + { + "epoch": 1.1127348643006263, + "grad_norm": 0.9716471433639526, + "learning_rate": 4.676902863074268e-06, + "loss": 0.7353, + "step": 710 + }, + { + "epoch": 1.1143006263048016, + "grad_norm": 1.040898084640503, + "learning_rate": 4.675866185786787e-06, + "loss": 0.7396, + "step": 711 + }, + { + "epoch": 1.115866388308977, + "grad_norm": 0.9931808114051819, + "learning_rate": 4.674827963315688e-06, + "loss": 0.7354, + "step": 712 + }, + { + "epoch": 1.1174321503131524, + "grad_norm": 1.020512580871582, + "learning_rate": 4.6737881963982605e-06, + "loss": 0.7366, + "step": 713 + }, + { + "epoch": 1.1189979123173277, + "grad_norm": 1.0201045274734497, + "learning_rate": 4.67274688577289e-06, + "loss": 0.7907, + "step": 714 + }, + { + "epoch": 1.120563674321503, + "grad_norm": 1.0163429975509644, + "learning_rate": 4.671704032179061e-06, + "loss": 0.7452, + "step": 715 + }, + { + "epoch": 1.1221294363256784, + "grad_norm": 0.9667377471923828, + "learning_rate": 4.670659636357352e-06, + "loss": 0.7983, + "step": 716 + }, + { + "epoch": 1.1236951983298538, + "grad_norm": 1.0788301229476929, + "learning_rate": 4.669613699049436e-06, + "loss": 0.7861, + "step": 717 + }, + { + "epoch": 1.1252609603340291, + "grad_norm": 1.0512142181396484, + "learning_rate": 4.668566220998082e-06, + "loss": 0.7582, + "step": 718 + }, + { + "epoch": 1.1268267223382047, + "grad_norm": 0.9225794076919556, + "learning_rate": 4.667517202947154e-06, + "loss": 0.7801, + "step": 719 + }, + { + "epoch": 1.12839248434238, + "grad_norm": 1.1193747520446777, + "learning_rate": 4.666466645641607e-06, + "loss": 0.8073, + "step": 720 + }, + { + "epoch": 1.1299582463465554, + "grad_norm": 0.934438943862915, + "learning_rate": 4.6654145498274915e-06, + "loss": 0.774, + "step": 721 + }, + { + "epoch": 1.1315240083507307, + "grad_norm": 1.0614968538284302, + "learning_rate": 4.664360916251949e-06, + "loss": 0.7216, + "step": 722 + }, + { + "epoch": 1.133089770354906, + "grad_norm": 1.0807424783706665, + "learning_rate": 4.663305745663212e-06, + "loss": 0.6434, + "step": 723 + }, + { + "epoch": 1.1346555323590815, + "grad_norm": 0.9515354633331299, + "learning_rate": 4.66224903881061e-06, + "loss": 0.7336, + "step": 724 + }, + { + "epoch": 1.1362212943632568, + "grad_norm": 0.9852489233016968, + "learning_rate": 4.661190796444557e-06, + "loss": 0.8156, + "step": 725 + }, + { + "epoch": 1.1377870563674322, + "grad_norm": 1.0071501731872559, + "learning_rate": 4.66013101931656e-06, + "loss": 0.8836, + "step": 726 + }, + { + "epoch": 1.1393528183716075, + "grad_norm": 0.9622960686683655, + "learning_rate": 4.6590697081792176e-06, + "loss": 0.7631, + "step": 727 + }, + { + "epoch": 1.1409185803757829, + "grad_norm": 1.0588105916976929, + "learning_rate": 4.6580068637862144e-06, + "loss": 0.7874, + "step": 728 + }, + { + "epoch": 1.1424843423799582, + "grad_norm": 0.9920447468757629, + "learning_rate": 4.656942486892326e-06, + "loss": 0.7161, + "step": 729 + }, + { + "epoch": 1.1440501043841336, + "grad_norm": 1.0036561489105225, + "learning_rate": 4.655876578253416e-06, + "loss": 0.7116, + "step": 730 + }, + { + "epoch": 1.145615866388309, + "grad_norm": 1.0208396911621094, + "learning_rate": 4.654809138626436e-06, + "loss": 0.818, + "step": 731 + }, + { + "epoch": 1.1471816283924843, + "grad_norm": 1.1095491647720337, + "learning_rate": 4.653740168769424e-06, + "loss": 0.7319, + "step": 732 + }, + { + "epoch": 1.1487473903966596, + "grad_norm": 0.8151328563690186, + "learning_rate": 4.652669669441505e-06, + "loss": 0.774, + "step": 733 + }, + { + "epoch": 1.150313152400835, + "grad_norm": 1.0747915506362915, + "learning_rate": 4.651597641402891e-06, + "loss": 0.8275, + "step": 734 + }, + { + "epoch": 1.1518789144050103, + "grad_norm": 0.9265272617340088, + "learning_rate": 4.650524085414878e-06, + "loss": 0.6946, + "step": 735 + }, + { + "epoch": 1.153444676409186, + "grad_norm": 0.9531672596931458, + "learning_rate": 4.649449002239849e-06, + "loss": 0.7899, + "step": 736 + }, + { + "epoch": 1.1550104384133613, + "grad_norm": 1.0496262311935425, + "learning_rate": 4.648372392641269e-06, + "loss": 0.8237, + "step": 737 + }, + { + "epoch": 1.1565762004175366, + "grad_norm": 0.8967193961143494, + "learning_rate": 4.64729425738369e-06, + "loss": 0.736, + "step": 738 + }, + { + "epoch": 1.158141962421712, + "grad_norm": 1.061038851737976, + "learning_rate": 4.646214597232745e-06, + "loss": 0.7308, + "step": 739 + }, + { + "epoch": 1.1597077244258873, + "grad_norm": 0.9556939601898193, + "learning_rate": 4.64513341295515e-06, + "loss": 0.7526, + "step": 740 + }, + { + "epoch": 1.1612734864300627, + "grad_norm": 1.000093936920166, + "learning_rate": 4.6440507053187075e-06, + "loss": 0.7488, + "step": 741 + }, + { + "epoch": 1.162839248434238, + "grad_norm": 0.9552261233329773, + "learning_rate": 4.642966475092294e-06, + "loss": 0.8291, + "step": 742 + }, + { + "epoch": 1.1644050104384134, + "grad_norm": 1.0215847492218018, + "learning_rate": 4.641880723045875e-06, + "loss": 0.756, + "step": 743 + }, + { + "epoch": 1.1659707724425887, + "grad_norm": 1.0543124675750732, + "learning_rate": 4.640793449950493e-06, + "loss": 0.7668, + "step": 744 + }, + { + "epoch": 1.167536534446764, + "grad_norm": 0.9648609161376953, + "learning_rate": 4.639704656578269e-06, + "loss": 0.6531, + "step": 745 + }, + { + "epoch": 1.1691022964509394, + "grad_norm": 1.0117427110671997, + "learning_rate": 4.638614343702408e-06, + "loss": 0.7084, + "step": 746 + }, + { + "epoch": 1.1706680584551148, + "grad_norm": 0.9475041627883911, + "learning_rate": 4.6375225120971915e-06, + "loss": 0.8578, + "step": 747 + }, + { + "epoch": 1.1722338204592901, + "grad_norm": 1.0122299194335938, + "learning_rate": 4.6364291625379785e-06, + "loss": 0.7541, + "step": 748 + }, + { + "epoch": 1.1737995824634655, + "grad_norm": 0.9487535357475281, + "learning_rate": 4.635334295801208e-06, + "loss": 0.7203, + "step": 749 + }, + { + "epoch": 1.1753653444676408, + "grad_norm": 1.0628844499588013, + "learning_rate": 4.6342379126643965e-06, + "loss": 0.7532, + "step": 750 + }, + { + "epoch": 1.1769311064718162, + "grad_norm": 0.9647781848907471, + "learning_rate": 4.633140013906136e-06, + "loss": 0.5918, + "step": 751 + }, + { + "epoch": 1.1784968684759916, + "grad_norm": 0.9011932611465454, + "learning_rate": 4.632040600306095e-06, + "loss": 0.7395, + "step": 752 + }, + { + "epoch": 1.1800626304801671, + "grad_norm": 1.0524142980575562, + "learning_rate": 4.630939672645018e-06, + "loss": 0.7487, + "step": 753 + }, + { + "epoch": 1.1816283924843423, + "grad_norm": 1.165077567100525, + "learning_rate": 4.6298372317047265e-06, + "loss": 0.6817, + "step": 754 + }, + { + "epoch": 1.1831941544885178, + "grad_norm": 0.9301269054412842, + "learning_rate": 4.628733278268113e-06, + "loss": 0.8446, + "step": 755 + }, + { + "epoch": 1.1847599164926932, + "grad_norm": 0.9762471318244934, + "learning_rate": 4.627627813119147e-06, + "loss": 0.7527, + "step": 756 + }, + { + "epoch": 1.1863256784968685, + "grad_norm": 0.8634734153747559, + "learning_rate": 4.626520837042871e-06, + "loss": 0.7177, + "step": 757 + }, + { + "epoch": 1.187891440501044, + "grad_norm": 1.0592188835144043, + "learning_rate": 4.625412350825399e-06, + "loss": 0.7293, + "step": 758 + }, + { + "epoch": 1.1894572025052192, + "grad_norm": 1.1318559646606445, + "learning_rate": 4.62430235525392e-06, + "loss": 0.6592, + "step": 759 + }, + { + "epoch": 1.1910229645093946, + "grad_norm": 0.9689339399337769, + "learning_rate": 4.623190851116692e-06, + "loss": 0.7667, + "step": 760 + }, + { + "epoch": 1.19258872651357, + "grad_norm": 0.9965885281562805, + "learning_rate": 4.6220778392030464e-06, + "loss": 0.6982, + "step": 761 + }, + { + "epoch": 1.1941544885177453, + "grad_norm": 1.1428629159927368, + "learning_rate": 4.6209633203033846e-06, + "loss": 0.7342, + "step": 762 + }, + { + "epoch": 1.1957202505219207, + "grad_norm": 0.901360809803009, + "learning_rate": 4.6198472952091785e-06, + "loss": 0.8371, + "step": 763 + }, + { + "epoch": 1.197286012526096, + "grad_norm": 0.9882062077522278, + "learning_rate": 4.618729764712969e-06, + "loss": 0.6882, + "step": 764 + }, + { + "epoch": 1.1988517745302714, + "grad_norm": 1.0692517757415771, + "learning_rate": 4.617610729608366e-06, + "loss": 0.7327, + "step": 765 + }, + { + "epoch": 1.2004175365344467, + "grad_norm": 1.1056455373764038, + "learning_rate": 4.616490190690049e-06, + "loss": 0.781, + "step": 766 + }, + { + "epoch": 1.201983298538622, + "grad_norm": 0.9928885698318481, + "learning_rate": 4.615368148753765e-06, + "loss": 0.7463, + "step": 767 + }, + { + "epoch": 1.2035490605427974, + "grad_norm": 1.0006698369979858, + "learning_rate": 4.614244604596327e-06, + "loss": 0.6994, + "step": 768 + }, + { + "epoch": 1.2051148225469728, + "grad_norm": 0.907508909702301, + "learning_rate": 4.613119559015616e-06, + "loss": 0.7117, + "step": 769 + }, + { + "epoch": 1.2066805845511483, + "grad_norm": 1.1037627458572388, + "learning_rate": 4.61199301281058e-06, + "loss": 0.772, + "step": 770 + }, + { + "epoch": 1.2082463465553235, + "grad_norm": 0.9474150538444519, + "learning_rate": 4.610864966781231e-06, + "loss": 0.7601, + "step": 771 + }, + { + "epoch": 1.209812108559499, + "grad_norm": 1.003667950630188, + "learning_rate": 4.609735421728647e-06, + "loss": 0.7141, + "step": 772 + }, + { + "epoch": 1.2113778705636744, + "grad_norm": 0.9724787473678589, + "learning_rate": 4.608604378454971e-06, + "loss": 0.7841, + "step": 773 + }, + { + "epoch": 1.2129436325678498, + "grad_norm": 1.0460760593414307, + "learning_rate": 4.607471837763409e-06, + "loss": 0.7563, + "step": 774 + }, + { + "epoch": 1.2145093945720251, + "grad_norm": 0.9798458218574524, + "learning_rate": 4.606337800458231e-06, + "loss": 0.6812, + "step": 775 + }, + { + "epoch": 1.2160751565762005, + "grad_norm": 1.1637914180755615, + "learning_rate": 4.605202267344768e-06, + "loss": 0.7571, + "step": 776 + }, + { + "epoch": 1.2176409185803758, + "grad_norm": 0.9384608864784241, + "learning_rate": 4.6040652392294175e-06, + "loss": 0.7842, + "step": 777 + }, + { + "epoch": 1.2192066805845512, + "grad_norm": 1.0490702390670776, + "learning_rate": 4.602926716919633e-06, + "loss": 0.7157, + "step": 778 + }, + { + "epoch": 1.2207724425887265, + "grad_norm": 1.131402611732483, + "learning_rate": 4.601786701223934e-06, + "loss": 0.7724, + "step": 779 + }, + { + "epoch": 1.2223382045929019, + "grad_norm": 1.0159722566604614, + "learning_rate": 4.600645192951898e-06, + "loss": 0.7666, + "step": 780 + }, + { + "epoch": 1.2239039665970772, + "grad_norm": 1.039806604385376, + "learning_rate": 4.599502192914164e-06, + "loss": 0.7845, + "step": 781 + }, + { + "epoch": 1.2254697286012526, + "grad_norm": 1.0722614526748657, + "learning_rate": 4.598357701922428e-06, + "loss": 0.782, + "step": 782 + }, + { + "epoch": 1.227035490605428, + "grad_norm": 0.9443033933639526, + "learning_rate": 4.5972117207894465e-06, + "loss": 0.6573, + "step": 783 + }, + { + "epoch": 1.2286012526096033, + "grad_norm": 0.9738026857376099, + "learning_rate": 4.596064250329035e-06, + "loss": 0.7475, + "step": 784 + }, + { + "epoch": 1.2301670146137786, + "grad_norm": 0.9932008385658264, + "learning_rate": 4.594915291356063e-06, + "loss": 0.757, + "step": 785 + }, + { + "epoch": 1.231732776617954, + "grad_norm": 1.0612564086914062, + "learning_rate": 4.5937648446864625e-06, + "loss": 0.7824, + "step": 786 + }, + { + "epoch": 1.2332985386221293, + "grad_norm": 1.0500893592834473, + "learning_rate": 4.592612911137218e-06, + "loss": 0.7646, + "step": 787 + }, + { + "epoch": 1.2348643006263047, + "grad_norm": 0.9607855081558228, + "learning_rate": 4.591459491526371e-06, + "loss": 0.754, + "step": 788 + }, + { + "epoch": 1.2364300626304803, + "grad_norm": 1.0871951580047607, + "learning_rate": 4.590304586673018e-06, + "loss": 0.7226, + "step": 789 + }, + { + "epoch": 1.2379958246346556, + "grad_norm": 1.0030092000961304, + "learning_rate": 4.589148197397311e-06, + "loss": 0.7357, + "step": 790 + }, + { + "epoch": 1.239561586638831, + "grad_norm": 1.0297991037368774, + "learning_rate": 4.587990324520455e-06, + "loss": 0.7724, + "step": 791 + }, + { + "epoch": 1.2411273486430063, + "grad_norm": 1.0192303657531738, + "learning_rate": 4.586830968864711e-06, + "loss": 0.6921, + "step": 792 + }, + { + "epoch": 1.2426931106471817, + "grad_norm": 0.9617307186126709, + "learning_rate": 4.585670131253389e-06, + "loss": 0.7443, + "step": 793 + }, + { + "epoch": 1.244258872651357, + "grad_norm": 1.0355238914489746, + "learning_rate": 4.584507812510856e-06, + "loss": 0.7741, + "step": 794 + }, + { + "epoch": 1.2458246346555324, + "grad_norm": 0.9938539862632751, + "learning_rate": 4.583344013462526e-06, + "loss": 0.7372, + "step": 795 + }, + { + "epoch": 1.2473903966597077, + "grad_norm": 1.0573335886001587, + "learning_rate": 4.582178734934869e-06, + "loss": 0.7191, + "step": 796 + }, + { + "epoch": 1.248956158663883, + "grad_norm": 0.963642954826355, + "learning_rate": 4.581011977755402e-06, + "loss": 0.6283, + "step": 797 + }, + { + "epoch": 1.2505219206680585, + "grad_norm": 1.0119287967681885, + "learning_rate": 4.579843742752694e-06, + "loss": 0.7599, + "step": 798 + }, + { + "epoch": 1.2520876826722338, + "grad_norm": 1.0628082752227783, + "learning_rate": 4.578674030756364e-06, + "loss": 0.7741, + "step": 799 + }, + { + "epoch": 1.2536534446764092, + "grad_norm": 1.0510241985321045, + "learning_rate": 4.577502842597077e-06, + "loss": 0.7615, + "step": 800 + }, + { + "epoch": 1.2552192066805845, + "grad_norm": 1.1180559396743774, + "learning_rate": 4.5763301791065474e-06, + "loss": 0.7119, + "step": 801 + }, + { + "epoch": 1.2567849686847599, + "grad_norm": 1.0198049545288086, + "learning_rate": 4.575156041117541e-06, + "loss": 0.7487, + "step": 802 + }, + { + "epoch": 1.2583507306889352, + "grad_norm": 1.018902063369751, + "learning_rate": 4.5739804294638655e-06, + "loss": 0.7461, + "step": 803 + }, + { + "epoch": 1.2599164926931106, + "grad_norm": 0.9230141639709473, + "learning_rate": 4.572803344980378e-06, + "loss": 0.7295, + "step": 804 + }, + { + "epoch": 1.261482254697286, + "grad_norm": 1.0011619329452515, + "learning_rate": 4.571624788502981e-06, + "loss": 0.7325, + "step": 805 + }, + { + "epoch": 1.2630480167014615, + "grad_norm": 0.9569756388664246, + "learning_rate": 4.570444760868622e-06, + "loss": 0.7494, + "step": 806 + }, + { + "epoch": 1.2646137787056366, + "grad_norm": 0.9860491156578064, + "learning_rate": 4.569263262915293e-06, + "loss": 0.6536, + "step": 807 + }, + { + "epoch": 1.2661795407098122, + "grad_norm": 1.0986074209213257, + "learning_rate": 4.568080295482032e-06, + "loss": 0.7337, + "step": 808 + }, + { + "epoch": 1.2677453027139876, + "grad_norm": 1.0782687664031982, + "learning_rate": 4.566895859408916e-06, + "loss": 0.7193, + "step": 809 + }, + { + "epoch": 1.269311064718163, + "grad_norm": 1.1424410343170166, + "learning_rate": 4.565709955537071e-06, + "loss": 0.6153, + "step": 810 + }, + { + "epoch": 1.2708768267223383, + "grad_norm": 0.9178900718688965, + "learning_rate": 4.564522584708662e-06, + "loss": 0.7802, + "step": 811 + }, + { + "epoch": 1.2724425887265136, + "grad_norm": 1.0006638765335083, + "learning_rate": 4.563333747766896e-06, + "loss": 0.7882, + "step": 812 + }, + { + "epoch": 1.274008350730689, + "grad_norm": 1.0139516592025757, + "learning_rate": 4.56214344555602e-06, + "loss": 0.6562, + "step": 813 + }, + { + "epoch": 1.2755741127348643, + "grad_norm": 1.0360203981399536, + "learning_rate": 4.560951678921324e-06, + "loss": 0.7951, + "step": 814 + }, + { + "epoch": 1.2771398747390397, + "grad_norm": 1.042457103729248, + "learning_rate": 4.559758448709138e-06, + "loss": 0.7572, + "step": 815 + }, + { + "epoch": 1.278705636743215, + "grad_norm": 1.0056653022766113, + "learning_rate": 4.558563755766827e-06, + "loss": 0.7679, + "step": 816 + }, + { + "epoch": 1.2802713987473904, + "grad_norm": 0.9838577508926392, + "learning_rate": 4.5573676009428e-06, + "loss": 0.7245, + "step": 817 + }, + { + "epoch": 1.2818371607515657, + "grad_norm": 1.0474631786346436, + "learning_rate": 4.556169985086503e-06, + "loss": 0.7823, + "step": 818 + }, + { + "epoch": 1.283402922755741, + "grad_norm": 1.0177569389343262, + "learning_rate": 4.554970909048417e-06, + "loss": 0.7599, + "step": 819 + }, + { + "epoch": 1.2849686847599164, + "grad_norm": 1.0120493173599243, + "learning_rate": 4.553770373680062e-06, + "loss": 0.7842, + "step": 820 + }, + { + "epoch": 1.2865344467640918, + "grad_norm": 1.1045918464660645, + "learning_rate": 4.5525683798339946e-06, + "loss": 0.7075, + "step": 821 + }, + { + "epoch": 1.2881002087682671, + "grad_norm": 1.097714900970459, + "learning_rate": 4.551364928363805e-06, + "loss": 0.7571, + "step": 822 + }, + { + "epoch": 1.2896659707724427, + "grad_norm": 0.9125490188598633, + "learning_rate": 4.550160020124121e-06, + "loss": 0.7626, + "step": 823 + }, + { + "epoch": 1.2912317327766178, + "grad_norm": 1.1127262115478516, + "learning_rate": 4.548953655970604e-06, + "loss": 0.6618, + "step": 824 + }, + { + "epoch": 1.2927974947807934, + "grad_norm": 1.0372540950775146, + "learning_rate": 4.547745836759949e-06, + "loss": 0.7467, + "step": 825 + }, + { + "epoch": 1.2943632567849686, + "grad_norm": 1.0158213376998901, + "learning_rate": 4.546536563349885e-06, + "loss": 0.6421, + "step": 826 + }, + { + "epoch": 1.2959290187891441, + "grad_norm": 1.0057018995285034, + "learning_rate": 4.545325836599171e-06, + "loss": 0.7252, + "step": 827 + }, + { + "epoch": 1.2974947807933195, + "grad_norm": 1.0552566051483154, + "learning_rate": 4.544113657367604e-06, + "loss": 0.6645, + "step": 828 + }, + { + "epoch": 1.2990605427974948, + "grad_norm": 1.0798696279525757, + "learning_rate": 4.542900026516006e-06, + "loss": 0.7627, + "step": 829 + }, + { + "epoch": 1.3006263048016702, + "grad_norm": 1.0450047254562378, + "learning_rate": 4.541684944906233e-06, + "loss": 0.705, + "step": 830 + }, + { + "epoch": 1.3021920668058455, + "grad_norm": 0.9975817203521729, + "learning_rate": 4.540468413401171e-06, + "loss": 0.7639, + "step": 831 + }, + { + "epoch": 1.303757828810021, + "grad_norm": 0.9081487059593201, + "learning_rate": 4.539250432864736e-06, + "loss": 0.8129, + "step": 832 + }, + { + "epoch": 1.3053235908141962, + "grad_norm": 1.228684902191162, + "learning_rate": 4.538031004161872e-06, + "loss": 0.6984, + "step": 833 + }, + { + "epoch": 1.3068893528183716, + "grad_norm": 1.0577794313430786, + "learning_rate": 4.536810128158552e-06, + "loss": 0.7436, + "step": 834 + }, + { + "epoch": 1.308455114822547, + "grad_norm": 1.1270350217819214, + "learning_rate": 4.535587805721777e-06, + "loss": 0.7589, + "step": 835 + }, + { + "epoch": 1.3100208768267223, + "grad_norm": 1.1473027467727661, + "learning_rate": 4.5343640377195766e-06, + "loss": 0.7988, + "step": 836 + }, + { + "epoch": 1.3115866388308977, + "grad_norm": 0.9700750112533569, + "learning_rate": 4.533138825021001e-06, + "loss": 0.7388, + "step": 837 + }, + { + "epoch": 1.313152400835073, + "grad_norm": 0.9596260786056519, + "learning_rate": 4.531912168496135e-06, + "loss": 0.8177, + "step": 838 + }, + { + "epoch": 1.3147181628392484, + "grad_norm": 0.9888957738876343, + "learning_rate": 4.530684069016081e-06, + "loss": 0.809, + "step": 839 + }, + { + "epoch": 1.316283924843424, + "grad_norm": 1.219224214553833, + "learning_rate": 4.529454527452972e-06, + "loss": 0.7761, + "step": 840 + }, + { + "epoch": 1.317849686847599, + "grad_norm": 0.9730296730995178, + "learning_rate": 4.5282235446799624e-06, + "loss": 0.8044, + "step": 841 + }, + { + "epoch": 1.3194154488517746, + "grad_norm": 1.1046907901763916, + "learning_rate": 4.526991121571228e-06, + "loss": 0.7452, + "step": 842 + }, + { + "epoch": 1.3209812108559498, + "grad_norm": 0.9707220196723938, + "learning_rate": 4.525757259001972e-06, + "loss": 0.7112, + "step": 843 + }, + { + "epoch": 1.3225469728601253, + "grad_norm": 1.0085593461990356, + "learning_rate": 4.524521957848416e-06, + "loss": 0.6266, + "step": 844 + }, + { + "epoch": 1.3241127348643007, + "grad_norm": 1.131439447402954, + "learning_rate": 4.5232852189878055e-06, + "loss": 0.7393, + "step": 845 + }, + { + "epoch": 1.325678496868476, + "grad_norm": 1.1238782405853271, + "learning_rate": 4.522047043298406e-06, + "loss": 0.7541, + "step": 846 + }, + { + "epoch": 1.3272442588726514, + "grad_norm": 0.9149152636528015, + "learning_rate": 4.520807431659504e-06, + "loss": 0.8162, + "step": 847 + }, + { + "epoch": 1.3288100208768268, + "grad_norm": 1.0428732633590698, + "learning_rate": 4.519566384951403e-06, + "loss": 0.7511, + "step": 848 + }, + { + "epoch": 1.3303757828810021, + "grad_norm": 1.0263090133666992, + "learning_rate": 4.518323904055432e-06, + "loss": 0.7482, + "step": 849 + }, + { + "epoch": 1.3319415448851775, + "grad_norm": 1.113810420036316, + "learning_rate": 4.517079989853931e-06, + "loss": 0.797, + "step": 850 + }, + { + "epoch": 1.3335073068893528, + "grad_norm": 1.0090736150741577, + "learning_rate": 4.515834643230263e-06, + "loss": 0.688, + "step": 851 + }, + { + "epoch": 1.3350730688935282, + "grad_norm": 1.012824296951294, + "learning_rate": 4.514587865068806e-06, + "loss": 0.6683, + "step": 852 + }, + { + "epoch": 1.3366388308977035, + "grad_norm": 0.9442676305770874, + "learning_rate": 4.513339656254954e-06, + "loss": 0.7812, + "step": 853 + }, + { + "epoch": 1.3382045929018789, + "grad_norm": 1.0870879888534546, + "learning_rate": 4.512090017675119e-06, + "loss": 0.6943, + "step": 854 + }, + { + "epoch": 1.3397703549060542, + "grad_norm": 0.9352210164070129, + "learning_rate": 4.510838950216727e-06, + "loss": 0.7571, + "step": 855 + }, + { + "epoch": 1.3413361169102296, + "grad_norm": 1.173621654510498, + "learning_rate": 4.509586454768221e-06, + "loss": 0.6985, + "step": 856 + }, + { + "epoch": 1.342901878914405, + "grad_norm": 0.9790180921554565, + "learning_rate": 4.508332532219053e-06, + "loss": 0.7132, + "step": 857 + }, + { + "epoch": 1.3444676409185803, + "grad_norm": 1.0999163389205933, + "learning_rate": 4.507077183459694e-06, + "loss": 0.7858, + "step": 858 + }, + { + "epoch": 1.3460334029227559, + "grad_norm": 1.018182635307312, + "learning_rate": 4.505820409381625e-06, + "loss": 0.712, + "step": 859 + }, + { + "epoch": 1.347599164926931, + "grad_norm": 1.1015701293945312, + "learning_rate": 4.504562210877338e-06, + "loss": 0.8004, + "step": 860 + }, + { + "epoch": 1.3491649269311066, + "grad_norm": 1.0872273445129395, + "learning_rate": 4.50330258884034e-06, + "loss": 0.7478, + "step": 861 + }, + { + "epoch": 1.350730688935282, + "grad_norm": 1.1098262071609497, + "learning_rate": 4.502041544165148e-06, + "loss": 0.7894, + "step": 862 + }, + { + "epoch": 1.3522964509394573, + "grad_norm": 1.0535573959350586, + "learning_rate": 4.500779077747285e-06, + "loss": 0.7185, + "step": 863 + }, + { + "epoch": 1.3538622129436326, + "grad_norm": 1.0424474477767944, + "learning_rate": 4.499515190483291e-06, + "loss": 0.748, + "step": 864 + }, + { + "epoch": 1.355427974947808, + "grad_norm": 1.005065679550171, + "learning_rate": 4.498249883270708e-06, + "loss": 0.7948, + "step": 865 + }, + { + "epoch": 1.3569937369519833, + "grad_norm": 1.0052721500396729, + "learning_rate": 4.49698315700809e-06, + "loss": 0.7619, + "step": 866 + }, + { + "epoch": 1.3585594989561587, + "grad_norm": 1.0953692197799683, + "learning_rate": 4.495715012594999e-06, + "loss": 0.7363, + "step": 867 + }, + { + "epoch": 1.360125260960334, + "grad_norm": 1.1238093376159668, + "learning_rate": 4.494445450932003e-06, + "loss": 0.6576, + "step": 868 + }, + { + "epoch": 1.3616910229645094, + "grad_norm": 0.8546008467674255, + "learning_rate": 4.493174472920676e-06, + "loss": 0.7513, + "step": 869 + }, + { + "epoch": 1.3632567849686847, + "grad_norm": 1.0863126516342163, + "learning_rate": 4.4919020794635995e-06, + "loss": 0.6886, + "step": 870 + }, + { + "epoch": 1.36482254697286, + "grad_norm": 0.9943820834159851, + "learning_rate": 4.490628271464359e-06, + "loss": 0.7253, + "step": 871 + }, + { + "epoch": 1.3663883089770354, + "grad_norm": 0.9377979636192322, + "learning_rate": 4.4893530498275436e-06, + "loss": 0.7026, + "step": 872 + }, + { + "epoch": 1.3679540709812108, + "grad_norm": 1.100799798965454, + "learning_rate": 4.488076415458749e-06, + "loss": 0.719, + "step": 873 + }, + { + "epoch": 1.3695198329853862, + "grad_norm": 1.015074610710144, + "learning_rate": 4.486798369264572e-06, + "loss": 0.7513, + "step": 874 + }, + { + "epoch": 1.3710855949895615, + "grad_norm": 1.047530174255371, + "learning_rate": 4.485518912152611e-06, + "loss": 0.7757, + "step": 875 + }, + { + "epoch": 1.372651356993737, + "grad_norm": 1.0264610052108765, + "learning_rate": 4.484238045031471e-06, + "loss": 0.7712, + "step": 876 + }, + { + "epoch": 1.3742171189979122, + "grad_norm": 0.9848430156707764, + "learning_rate": 4.482955768810752e-06, + "loss": 0.7569, + "step": 877 + }, + { + "epoch": 1.3757828810020878, + "grad_norm": 1.070816159248352, + "learning_rate": 4.48167208440106e-06, + "loss": 0.7857, + "step": 878 + }, + { + "epoch": 1.377348643006263, + "grad_norm": 1.0134745836257935, + "learning_rate": 4.480386992713998e-06, + "loss": 0.7372, + "step": 879 + }, + { + "epoch": 1.3789144050104385, + "grad_norm": 1.0736140012741089, + "learning_rate": 4.47910049466217e-06, + "loss": 0.7794, + "step": 880 + }, + { + "epoch": 1.3804801670146138, + "grad_norm": 1.0163054466247559, + "learning_rate": 4.477812591159176e-06, + "loss": 0.7341, + "step": 881 + }, + { + "epoch": 1.3820459290187892, + "grad_norm": 1.074578881263733, + "learning_rate": 4.476523283119619e-06, + "loss": 0.6924, + "step": 882 + }, + { + "epoch": 1.3836116910229646, + "grad_norm": 1.183667540550232, + "learning_rate": 4.475232571459095e-06, + "loss": 0.7866, + "step": 883 + }, + { + "epoch": 1.38517745302714, + "grad_norm": 1.030827522277832, + "learning_rate": 4.473940457094199e-06, + "loss": 0.7845, + "step": 884 + }, + { + "epoch": 1.3867432150313153, + "grad_norm": 1.0043658018112183, + "learning_rate": 4.47264694094252e-06, + "loss": 0.7399, + "step": 885 + }, + { + "epoch": 1.3883089770354906, + "grad_norm": 1.1135472059249878, + "learning_rate": 4.471352023922645e-06, + "loss": 0.771, + "step": 886 + }, + { + "epoch": 1.389874739039666, + "grad_norm": 1.0338670015335083, + "learning_rate": 4.470055706954154e-06, + "loss": 0.7282, + "step": 887 + }, + { + "epoch": 1.3914405010438413, + "grad_norm": 0.918855607509613, + "learning_rate": 4.468757990957623e-06, + "loss": 0.728, + "step": 888 + }, + { + "epoch": 1.3930062630480167, + "grad_norm": 1.06833016872406, + "learning_rate": 4.467458876854619e-06, + "loss": 0.7012, + "step": 889 + }, + { + "epoch": 1.394572025052192, + "grad_norm": 1.0103005170822144, + "learning_rate": 4.4661583655677045e-06, + "loss": 0.7074, + "step": 890 + }, + { + "epoch": 1.3961377870563674, + "grad_norm": 1.0418543815612793, + "learning_rate": 4.464856458020433e-06, + "loss": 0.6858, + "step": 891 + }, + { + "epoch": 1.3977035490605427, + "grad_norm": 0.9720668196678162, + "learning_rate": 4.463553155137348e-06, + "loss": 0.7024, + "step": 892 + }, + { + "epoch": 1.399269311064718, + "grad_norm": 1.1870561838150024, + "learning_rate": 4.462248457843986e-06, + "loss": 0.8069, + "step": 893 + }, + { + "epoch": 1.4008350730688934, + "grad_norm": 0.9888094663619995, + "learning_rate": 4.460942367066874e-06, + "loss": 0.7281, + "step": 894 + }, + { + "epoch": 1.402400835073069, + "grad_norm": 0.9120670557022095, + "learning_rate": 4.459634883733527e-06, + "loss": 0.7992, + "step": 895 + }, + { + "epoch": 1.4039665970772441, + "grad_norm": 1.0239481925964355, + "learning_rate": 4.45832600877245e-06, + "loss": 0.7312, + "step": 896 + }, + { + "epoch": 1.4055323590814197, + "grad_norm": 0.9379130005836487, + "learning_rate": 4.4570157431131355e-06, + "loss": 0.7019, + "step": 897 + }, + { + "epoch": 1.407098121085595, + "grad_norm": 1.0789109468460083, + "learning_rate": 4.455704087686063e-06, + "loss": 0.8149, + "step": 898 + }, + { + "epoch": 1.4086638830897704, + "grad_norm": 1.0521917343139648, + "learning_rate": 4.454391043422703e-06, + "loss": 0.7873, + "step": 899 + }, + { + "epoch": 1.4102296450939458, + "grad_norm": 0.9582643508911133, + "learning_rate": 4.453076611255507e-06, + "loss": 0.7815, + "step": 900 + }, + { + "epoch": 1.4117954070981211, + "grad_norm": 1.1096739768981934, + "learning_rate": 4.451760792117914e-06, + "loss": 0.7636, + "step": 901 + }, + { + "epoch": 1.4133611691022965, + "grad_norm": 1.043574333190918, + "learning_rate": 4.450443586944349e-06, + "loss": 0.7411, + "step": 902 + }, + { + "epoch": 1.4149269311064718, + "grad_norm": 1.0280033349990845, + "learning_rate": 4.449124996670221e-06, + "loss": 0.6525, + "step": 903 + }, + { + "epoch": 1.4164926931106472, + "grad_norm": 1.0329992771148682, + "learning_rate": 4.44780502223192e-06, + "loss": 0.7133, + "step": 904 + }, + { + "epoch": 1.4180584551148225, + "grad_norm": 1.016168475151062, + "learning_rate": 4.446483664566825e-06, + "loss": 0.7098, + "step": 905 + }, + { + "epoch": 1.4196242171189979, + "grad_norm": 1.2022786140441895, + "learning_rate": 4.4451609246132895e-06, + "loss": 0.7303, + "step": 906 + }, + { + "epoch": 1.4211899791231732, + "grad_norm": 1.0311683416366577, + "learning_rate": 4.443836803310655e-06, + "loss": 0.7475, + "step": 907 + }, + { + "epoch": 1.4227557411273486, + "grad_norm": 1.0073823928833008, + "learning_rate": 4.442511301599241e-06, + "loss": 0.7513, + "step": 908 + }, + { + "epoch": 1.424321503131524, + "grad_norm": 1.0702322721481323, + "learning_rate": 4.441184420420347e-06, + "loss": 0.6918, + "step": 909 + }, + { + "epoch": 1.4258872651356993, + "grad_norm": 1.1416171789169312, + "learning_rate": 4.439856160716254e-06, + "loss": 0.6324, + "step": 910 + }, + { + "epoch": 1.4274530271398747, + "grad_norm": 1.0560253858566284, + "learning_rate": 4.438526523430221e-06, + "loss": 0.749, + "step": 911 + }, + { + "epoch": 1.4290187891440502, + "grad_norm": 1.021008849143982, + "learning_rate": 4.4371955095064835e-06, + "loss": 0.7666, + "step": 912 + }, + { + "epoch": 1.4305845511482254, + "grad_norm": 1.151904821395874, + "learning_rate": 4.435863119890258e-06, + "loss": 0.6904, + "step": 913 + }, + { + "epoch": 1.432150313152401, + "grad_norm": 1.1216702461242676, + "learning_rate": 4.434529355527736e-06, + "loss": 0.7068, + "step": 914 + }, + { + "epoch": 1.433716075156576, + "grad_norm": 1.1130588054656982, + "learning_rate": 4.433194217366085e-06, + "loss": 0.6524, + "step": 915 + }, + { + "epoch": 1.4352818371607516, + "grad_norm": 1.07612144947052, + "learning_rate": 4.431857706353449e-06, + "loss": 0.8078, + "step": 916 + }, + { + "epoch": 1.436847599164927, + "grad_norm": 1.0220447778701782, + "learning_rate": 4.430519823438946e-06, + "loss": 0.6736, + "step": 917 + }, + { + "epoch": 1.4384133611691023, + "grad_norm": 1.0657141208648682, + "learning_rate": 4.429180569572669e-06, + "loss": 0.7457, + "step": 918 + }, + { + "epoch": 1.4399791231732777, + "grad_norm": 1.0646854639053345, + "learning_rate": 4.427839945705684e-06, + "loss": 0.7925, + "step": 919 + }, + { + "epoch": 1.441544885177453, + "grad_norm": 1.0619502067565918, + "learning_rate": 4.426497952790031e-06, + "loss": 0.6968, + "step": 920 + }, + { + "epoch": 1.4431106471816284, + "grad_norm": 0.9684338569641113, + "learning_rate": 4.425154591778722e-06, + "loss": 0.7411, + "step": 921 + }, + { + "epoch": 1.4446764091858038, + "grad_norm": 1.1225167512893677, + "learning_rate": 4.423809863625739e-06, + "loss": 0.7623, + "step": 922 + }, + { + "epoch": 1.446242171189979, + "grad_norm": 0.9310746788978577, + "learning_rate": 4.422463769286036e-06, + "loss": 0.8219, + "step": 923 + }, + { + "epoch": 1.4478079331941545, + "grad_norm": 1.1107957363128662, + "learning_rate": 4.4211163097155375e-06, + "loss": 0.7755, + "step": 924 + }, + { + "epoch": 1.4493736951983298, + "grad_norm": 0.9827596545219421, + "learning_rate": 4.419767485871136e-06, + "loss": 0.7102, + "step": 925 + }, + { + "epoch": 1.4509394572025052, + "grad_norm": 1.1075087785720825, + "learning_rate": 4.418417298710695e-06, + "loss": 0.6399, + "step": 926 + }, + { + "epoch": 1.4525052192066805, + "grad_norm": 0.9575802087783813, + "learning_rate": 4.417065749193045e-06, + "loss": 0.721, + "step": 927 + }, + { + "epoch": 1.4540709812108559, + "grad_norm": 1.0016006231307983, + "learning_rate": 4.415712838277985e-06, + "loss": 0.762, + "step": 928 + }, + { + "epoch": 1.4556367432150314, + "grad_norm": 1.1907106637954712, + "learning_rate": 4.414358566926279e-06, + "loss": 0.7904, + "step": 929 + }, + { + "epoch": 1.4572025052192066, + "grad_norm": 1.0011543035507202, + "learning_rate": 4.413002936099656e-06, + "loss": 0.7179, + "step": 930 + }, + { + "epoch": 1.4587682672233822, + "grad_norm": 0.980684757232666, + "learning_rate": 4.411645946760817e-06, + "loss": 0.6845, + "step": 931 + }, + { + "epoch": 1.4603340292275573, + "grad_norm": 0.9458135366439819, + "learning_rate": 4.4102875998734176e-06, + "loss": 0.6911, + "step": 932 + }, + { + "epoch": 1.4618997912317329, + "grad_norm": 1.020966649055481, + "learning_rate": 4.408927896402087e-06, + "loss": 0.7284, + "step": 933 + }, + { + "epoch": 1.4634655532359082, + "grad_norm": 1.041833519935608, + "learning_rate": 4.407566837312413e-06, + "loss": 0.7451, + "step": 934 + }, + { + "epoch": 1.4650313152400836, + "grad_norm": 0.9076465964317322, + "learning_rate": 4.406204423570946e-06, + "loss": 0.7553, + "step": 935 + }, + { + "epoch": 1.466597077244259, + "grad_norm": 1.1963952779769897, + "learning_rate": 4.404840656145199e-06, + "loss": 0.7437, + "step": 936 + }, + { + "epoch": 1.4681628392484343, + "grad_norm": 1.0181536674499512, + "learning_rate": 4.403475536003648e-06, + "loss": 0.7283, + "step": 937 + }, + { + "epoch": 1.4697286012526096, + "grad_norm": 0.9563003778457642, + "learning_rate": 4.402109064115727e-06, + "loss": 0.7375, + "step": 938 + }, + { + "epoch": 1.471294363256785, + "grad_norm": 1.05129873752594, + "learning_rate": 4.400741241451832e-06, + "loss": 0.6805, + "step": 939 + }, + { + "epoch": 1.4728601252609603, + "grad_norm": 1.0912166833877563, + "learning_rate": 4.399372068983317e-06, + "loss": 0.7076, + "step": 940 + }, + { + "epoch": 1.4744258872651357, + "grad_norm": 1.108500599861145, + "learning_rate": 4.398001547682494e-06, + "loss": 0.6836, + "step": 941 + }, + { + "epoch": 1.475991649269311, + "grad_norm": 1.0741651058197021, + "learning_rate": 4.396629678522636e-06, + "loss": 0.7405, + "step": 942 + }, + { + "epoch": 1.4775574112734864, + "grad_norm": 1.0603930950164795, + "learning_rate": 4.39525646247797e-06, + "loss": 0.7602, + "step": 943 + }, + { + "epoch": 1.4791231732776617, + "grad_norm": 0.9814990162849426, + "learning_rate": 4.39388190052368e-06, + "loss": 0.7324, + "step": 944 + }, + { + "epoch": 1.480688935281837, + "grad_norm": 0.9643695950508118, + "learning_rate": 4.392505993635906e-06, + "loss": 0.7034, + "step": 945 + }, + { + "epoch": 1.4822546972860124, + "grad_norm": 1.0490994453430176, + "learning_rate": 4.391128742791743e-06, + "loss": 0.7358, + "step": 946 + }, + { + "epoch": 1.4838204592901878, + "grad_norm": 0.9223467111587524, + "learning_rate": 4.389750148969244e-06, + "loss": 0.698, + "step": 947 + }, + { + "epoch": 1.4853862212943634, + "grad_norm": 1.1107758283615112, + "learning_rate": 4.388370213147409e-06, + "loss": 0.7264, + "step": 948 + }, + { + "epoch": 1.4869519832985385, + "grad_norm": 1.05939781665802, + "learning_rate": 4.386988936306196e-06, + "loss": 0.7209, + "step": 949 + }, + { + "epoch": 1.488517745302714, + "grad_norm": 1.024686574935913, + "learning_rate": 4.385606319426512e-06, + "loss": 0.7798, + "step": 950 + }, + { + "epoch": 1.4900835073068894, + "grad_norm": 0.9920977354049683, + "learning_rate": 4.3842223634902205e-06, + "loss": 0.8045, + "step": 951 + }, + { + "epoch": 1.4916492693110648, + "grad_norm": 0.950927197933197, + "learning_rate": 4.382837069480131e-06, + "loss": 0.6777, + "step": 952 + }, + { + "epoch": 1.4932150313152401, + "grad_norm": 1.0072081089019775, + "learning_rate": 4.3814504383800056e-06, + "loss": 0.7458, + "step": 953 + }, + { + "epoch": 1.4947807933194155, + "grad_norm": 0.9982030987739563, + "learning_rate": 4.3800624711745546e-06, + "loss": 0.8215, + "step": 954 + }, + { + "epoch": 1.4963465553235908, + "grad_norm": 0.9682126641273499, + "learning_rate": 4.3786731688494396e-06, + "loss": 0.7157, + "step": 955 + }, + { + "epoch": 1.4979123173277662, + "grad_norm": 1.1599830389022827, + "learning_rate": 4.377282532391267e-06, + "loss": 0.8091, + "step": 956 + }, + { + "epoch": 1.4994780793319415, + "grad_norm": 1.0624948740005493, + "learning_rate": 4.3758905627875934e-06, + "loss": 0.7428, + "step": 957 + }, + { + "epoch": 1.501043841336117, + "grad_norm": 1.14694082736969, + "learning_rate": 4.374497261026921e-06, + "loss": 0.7752, + "step": 958 + }, + { + "epoch": 1.5026096033402923, + "grad_norm": 1.0355825424194336, + "learning_rate": 4.373102628098697e-06, + "loss": 0.7348, + "step": 959 + }, + { + "epoch": 1.5041753653444676, + "grad_norm": 1.036026954650879, + "learning_rate": 4.371706664993316e-06, + "loss": 0.7536, + "step": 960 + }, + { + "epoch": 1.505741127348643, + "grad_norm": 1.2319790124893188, + "learning_rate": 4.370309372702116e-06, + "loss": 0.6985, + "step": 961 + }, + { + "epoch": 1.5073068893528183, + "grad_norm": 0.908687949180603, + "learning_rate": 4.36891075221738e-06, + "loss": 0.7173, + "step": 962 + }, + { + "epoch": 1.5088726513569939, + "grad_norm": 1.0482099056243896, + "learning_rate": 4.367510804532331e-06, + "loss": 0.7514, + "step": 963 + }, + { + "epoch": 1.510438413361169, + "grad_norm": 1.0434144735336304, + "learning_rate": 4.36610953064114e-06, + "loss": 0.754, + "step": 964 + }, + { + "epoch": 1.5120041753653446, + "grad_norm": 1.0122450590133667, + "learning_rate": 4.364706931538914e-06, + "loss": 0.7767, + "step": 965 + }, + { + "epoch": 1.5135699373695197, + "grad_norm": 1.0471900701522827, + "learning_rate": 4.363303008221705e-06, + "loss": 0.7337, + "step": 966 + }, + { + "epoch": 1.5151356993736953, + "grad_norm": 1.0691189765930176, + "learning_rate": 4.3618977616865045e-06, + "loss": 0.8047, + "step": 967 + }, + { + "epoch": 1.5167014613778704, + "grad_norm": 1.015114188194275, + "learning_rate": 4.360491192931242e-06, + "loss": 0.726, + "step": 968 + }, + { + "epoch": 1.518267223382046, + "grad_norm": 0.9770560264587402, + "learning_rate": 4.359083302954788e-06, + "loss": 0.7692, + "step": 969 + }, + { + "epoch": 1.5198329853862211, + "grad_norm": 1.079991340637207, + "learning_rate": 4.357674092756951e-06, + "loss": 0.8424, + "step": 970 + }, + { + "epoch": 1.5213987473903967, + "grad_norm": 1.068708896636963, + "learning_rate": 4.356263563338476e-06, + "loss": 0.758, + "step": 971 + }, + { + "epoch": 1.522964509394572, + "grad_norm": 1.1005477905273438, + "learning_rate": 4.354851715701046e-06, + "loss": 0.7209, + "step": 972 + }, + { + "epoch": 1.5245302713987474, + "grad_norm": 0.9870902299880981, + "learning_rate": 4.3534385508472786e-06, + "loss": 0.7272, + "step": 973 + }, + { + "epoch": 1.5260960334029228, + "grad_norm": 0.9616238474845886, + "learning_rate": 4.352024069780729e-06, + "loss": 0.8124, + "step": 974 + }, + { + "epoch": 1.5276617954070981, + "grad_norm": 1.0341622829437256, + "learning_rate": 4.3506082735058855e-06, + "loss": 0.7434, + "step": 975 + }, + { + "epoch": 1.5292275574112735, + "grad_norm": 1.0157302618026733, + "learning_rate": 4.349191163028171e-06, + "loss": 0.7534, + "step": 976 + }, + { + "epoch": 1.5307933194154488, + "grad_norm": 1.041379690170288, + "learning_rate": 4.347772739353939e-06, + "loss": 0.6942, + "step": 977 + }, + { + "epoch": 1.5323590814196242, + "grad_norm": 1.016205072402954, + "learning_rate": 4.346353003490481e-06, + "loss": 0.6707, + "step": 978 + }, + { + "epoch": 1.5339248434237995, + "grad_norm": 1.0034714937210083, + "learning_rate": 4.344931956446017e-06, + "loss": 0.6358, + "step": 979 + }, + { + "epoch": 1.535490605427975, + "grad_norm": 0.9671767354011536, + "learning_rate": 4.343509599229697e-06, + "loss": 0.6333, + "step": 980 + }, + { + "epoch": 1.5370563674321502, + "grad_norm": 1.1384506225585938, + "learning_rate": 4.342085932851605e-06, + "loss": 0.803, + "step": 981 + }, + { + "epoch": 1.5386221294363258, + "grad_norm": 1.023031234741211, + "learning_rate": 4.34066095832275e-06, + "loss": 0.7801, + "step": 982 + }, + { + "epoch": 1.540187891440501, + "grad_norm": 1.0675241947174072, + "learning_rate": 4.339234676655075e-06, + "loss": 0.7335, + "step": 983 + }, + { + "epoch": 1.5417536534446765, + "grad_norm": 1.0397543907165527, + "learning_rate": 4.337807088861447e-06, + "loss": 0.7722, + "step": 984 + }, + { + "epoch": 1.5433194154488517, + "grad_norm": 1.0892809629440308, + "learning_rate": 4.3363781959556635e-06, + "loss": 0.7603, + "step": 985 + }, + { + "epoch": 1.5448851774530272, + "grad_norm": 0.9335037469863892, + "learning_rate": 4.334947998952448e-06, + "loss": 0.7678, + "step": 986 + }, + { + "epoch": 1.5464509394572024, + "grad_norm": 1.0991342067718506, + "learning_rate": 4.333516498867449e-06, + "loss": 0.7461, + "step": 987 + }, + { + "epoch": 1.548016701461378, + "grad_norm": 1.112265944480896, + "learning_rate": 4.332083696717242e-06, + "loss": 0.7593, + "step": 988 + }, + { + "epoch": 1.5495824634655533, + "grad_norm": 0.9525822997093201, + "learning_rate": 4.330649593519325e-06, + "loss": 0.7629, + "step": 989 + }, + { + "epoch": 1.5511482254697286, + "grad_norm": 1.0624464750289917, + "learning_rate": 4.329214190292122e-06, + "loss": 0.7261, + "step": 990 + }, + { + "epoch": 1.552713987473904, + "grad_norm": 1.1969516277313232, + "learning_rate": 4.327777488054979e-06, + "loss": 0.7483, + "step": 991 + }, + { + "epoch": 1.5542797494780793, + "grad_norm": 1.0239863395690918, + "learning_rate": 4.326339487828167e-06, + "loss": 0.7265, + "step": 992 + }, + { + "epoch": 1.5558455114822547, + "grad_norm": 1.0881083011627197, + "learning_rate": 4.324900190632874e-06, + "loss": 0.724, + "step": 993 + }, + { + "epoch": 1.55741127348643, + "grad_norm": 1.098827838897705, + "learning_rate": 4.323459597491214e-06, + "loss": 0.7222, + "step": 994 + }, + { + "epoch": 1.5589770354906054, + "grad_norm": 1.105057716369629, + "learning_rate": 4.322017709426217e-06, + "loss": 0.7827, + "step": 995 + }, + { + "epoch": 1.5605427974947808, + "grad_norm": 0.9830679297447205, + "learning_rate": 4.3205745274618365e-06, + "loss": 0.7316, + "step": 996 + }, + { + "epoch": 1.562108559498956, + "grad_norm": 0.9951156973838806, + "learning_rate": 4.319130052622942e-06, + "loss": 0.7181, + "step": 997 + }, + { + "epoch": 1.5636743215031315, + "grad_norm": 1.0374284982681274, + "learning_rate": 4.317684285935323e-06, + "loss": 0.7355, + "step": 998 + }, + { + "epoch": 1.565240083507307, + "grad_norm": 1.0139906406402588, + "learning_rate": 4.316237228425683e-06, + "loss": 0.7249, + "step": 999 + }, + { + "epoch": 1.5668058455114822, + "grad_norm": 0.9555441737174988, + "learning_rate": 4.3147888811216496e-06, + "loss": 0.7387, + "step": 1000 + }, + { + "epoch": 1.5683716075156577, + "grad_norm": 1.0600265264511108, + "learning_rate": 4.313339245051758e-06, + "loss": 0.7445, + "step": 1001 + }, + { + "epoch": 1.5699373695198329, + "grad_norm": 1.0070478916168213, + "learning_rate": 4.311888321245461e-06, + "loss": 0.7605, + "step": 1002 + }, + { + "epoch": 1.5715031315240084, + "grad_norm": 1.019710898399353, + "learning_rate": 4.31043611073313e-06, + "loss": 0.7262, + "step": 1003 + }, + { + "epoch": 1.5730688935281836, + "grad_norm": 1.0476088523864746, + "learning_rate": 4.308982614546045e-06, + "loss": 0.6766, + "step": 1004 + }, + { + "epoch": 1.5746346555323592, + "grad_norm": 1.0670830011367798, + "learning_rate": 4.307527833716403e-06, + "loss": 0.7465, + "step": 1005 + }, + { + "epoch": 1.5762004175365343, + "grad_norm": 1.1807290315628052, + "learning_rate": 4.30607176927731e-06, + "loss": 0.7032, + "step": 1006 + }, + { + "epoch": 1.5777661795407099, + "grad_norm": 1.0482438802719116, + "learning_rate": 4.3046144222627855e-06, + "loss": 0.7156, + "step": 1007 + }, + { + "epoch": 1.5793319415448852, + "grad_norm": 0.9434391856193542, + "learning_rate": 4.303155793707759e-06, + "loss": 0.7816, + "step": 1008 + }, + { + "epoch": 1.5808977035490606, + "grad_norm": 1.0497010946273804, + "learning_rate": 4.3016958846480706e-06, + "loss": 0.735, + "step": 1009 + }, + { + "epoch": 1.582463465553236, + "grad_norm": 0.955949068069458, + "learning_rate": 4.300234696120469e-06, + "loss": 0.6414, + "step": 1010 + }, + { + "epoch": 1.5840292275574113, + "grad_norm": 1.0462027788162231, + "learning_rate": 4.298772229162613e-06, + "loss": 0.6949, + "step": 1011 + }, + { + "epoch": 1.5855949895615866, + "grad_norm": 1.0634411573410034, + "learning_rate": 4.297308484813067e-06, + "loss": 0.7736, + "step": 1012 + }, + { + "epoch": 1.587160751565762, + "grad_norm": 1.0463027954101562, + "learning_rate": 4.295843464111306e-06, + "loss": 0.6812, + "step": 1013 + }, + { + "epoch": 1.5887265135699373, + "grad_norm": 1.0409252643585205, + "learning_rate": 4.294377168097706e-06, + "loss": 0.7319, + "step": 1014 + }, + { + "epoch": 1.5902922755741127, + "grad_norm": 0.9714667201042175, + "learning_rate": 4.292909597813555e-06, + "loss": 0.7473, + "step": 1015 + }, + { + "epoch": 1.5918580375782883, + "grad_norm": 1.03997802734375, + "learning_rate": 4.291440754301041e-06, + "loss": 0.7261, + "step": 1016 + }, + { + "epoch": 1.5934237995824634, + "grad_norm": 1.063293218612671, + "learning_rate": 4.289970638603258e-06, + "loss": 0.7909, + "step": 1017 + }, + { + "epoch": 1.594989561586639, + "grad_norm": 1.110007643699646, + "learning_rate": 4.288499251764205e-06, + "loss": 0.7, + "step": 1018 + }, + { + "epoch": 1.596555323590814, + "grad_norm": 0.981261134147644, + "learning_rate": 4.2870265948287806e-06, + "loss": 0.7516, + "step": 1019 + }, + { + "epoch": 1.5981210855949897, + "grad_norm": 1.0653973817825317, + "learning_rate": 4.2855526688427875e-06, + "loss": 0.7278, + "step": 1020 + }, + { + "epoch": 1.5996868475991648, + "grad_norm": 0.9212517738342285, + "learning_rate": 4.284077474852928e-06, + "loss": 0.7251, + "step": 1021 + }, + { + "epoch": 1.6012526096033404, + "grad_norm": 1.0887659788131714, + "learning_rate": 4.282601013906808e-06, + "loss": 0.7271, + "step": 1022 + }, + { + "epoch": 1.6028183716075155, + "grad_norm": 1.0753779411315918, + "learning_rate": 4.281123287052928e-06, + "loss": 0.7679, + "step": 1023 + }, + { + "epoch": 1.604384133611691, + "grad_norm": 1.0282115936279297, + "learning_rate": 4.2796442953406935e-06, + "loss": 0.695, + "step": 1024 + }, + { + "epoch": 1.6059498956158664, + "grad_norm": 0.9782281517982483, + "learning_rate": 4.2781640398204036e-06, + "loss": 0.7717, + "step": 1025 + }, + { + "epoch": 1.6075156576200418, + "grad_norm": 1.06228768825531, + "learning_rate": 4.276682521543256e-06, + "loss": 0.8069, + "step": 1026 + }, + { + "epoch": 1.6090814196242171, + "grad_norm": 0.8853397965431213, + "learning_rate": 4.2751997415613465e-06, + "loss": 0.8174, + "step": 1027 + }, + { + "epoch": 1.6106471816283925, + "grad_norm": 1.1525455713272095, + "learning_rate": 4.273715700927666e-06, + "loss": 0.7149, + "step": 1028 + }, + { + "epoch": 1.6122129436325678, + "grad_norm": 0.8743720650672913, + "learning_rate": 4.272230400696101e-06, + "loss": 0.7296, + "step": 1029 + }, + { + "epoch": 1.6137787056367432, + "grad_norm": 1.1136301755905151, + "learning_rate": 4.270743841921431e-06, + "loss": 0.7539, + "step": 1030 + }, + { + "epoch": 1.6153444676409185, + "grad_norm": 0.9830031991004944, + "learning_rate": 4.269256025659332e-06, + "loss": 0.6545, + "step": 1031 + }, + { + "epoch": 1.616910229645094, + "grad_norm": 1.1221113204956055, + "learning_rate": 4.267766952966369e-06, + "loss": 0.6144, + "step": 1032 + }, + { + "epoch": 1.6184759916492695, + "grad_norm": 0.975713849067688, + "learning_rate": 4.266276624900004e-06, + "loss": 0.746, + "step": 1033 + }, + { + "epoch": 1.6200417536534446, + "grad_norm": 1.1131149530410767, + "learning_rate": 4.264785042518587e-06, + "loss": 0.705, + "step": 1034 + }, + { + "epoch": 1.6216075156576202, + "grad_norm": 1.0574990510940552, + "learning_rate": 4.263292206881361e-06, + "loss": 0.7681, + "step": 1035 + }, + { + "epoch": 1.6231732776617953, + "grad_norm": 0.9931074976921082, + "learning_rate": 4.261798119048456e-06, + "loss": 0.7854, + "step": 1036 + }, + { + "epoch": 1.6247390396659709, + "grad_norm": 1.1598386764526367, + "learning_rate": 4.260302780080896e-06, + "loss": 0.7406, + "step": 1037 + }, + { + "epoch": 1.626304801670146, + "grad_norm": 0.9910677671432495, + "learning_rate": 4.258806191040587e-06, + "loss": 0.7107, + "step": 1038 + }, + { + "epoch": 1.6278705636743216, + "grad_norm": 1.1023072004318237, + "learning_rate": 4.2573083529903294e-06, + "loss": 0.7964, + "step": 1039 + }, + { + "epoch": 1.6294363256784967, + "grad_norm": 0.991082489490509, + "learning_rate": 4.255809266993806e-06, + "loss": 0.762, + "step": 1040 + }, + { + "epoch": 1.6310020876826723, + "grad_norm": 0.9281193017959595, + "learning_rate": 4.254308934115589e-06, + "loss": 0.7272, + "step": 1041 + }, + { + "epoch": 1.6325678496868476, + "grad_norm": 1.20292067527771, + "learning_rate": 4.252807355421133e-06, + "loss": 0.7535, + "step": 1042 + }, + { + "epoch": 1.634133611691023, + "grad_norm": 1.0362178087234497, + "learning_rate": 4.251304531976778e-06, + "loss": 0.7524, + "step": 1043 + }, + { + "epoch": 1.6356993736951984, + "grad_norm": 1.0267268419265747, + "learning_rate": 4.249800464849751e-06, + "loss": 0.6588, + "step": 1044 + }, + { + "epoch": 1.6372651356993737, + "grad_norm": 1.0531187057495117, + "learning_rate": 4.248295155108158e-06, + "loss": 0.7328, + "step": 1045 + }, + { + "epoch": 1.638830897703549, + "grad_norm": 1.0008069276809692, + "learning_rate": 4.246788603820989e-06, + "loss": 0.719, + "step": 1046 + }, + { + "epoch": 1.6403966597077244, + "grad_norm": 1.064780831336975, + "learning_rate": 4.245280812058119e-06, + "loss": 0.7388, + "step": 1047 + }, + { + "epoch": 1.6419624217118998, + "grad_norm": 1.06492018699646, + "learning_rate": 4.243771780890298e-06, + "loss": 0.7495, + "step": 1048 + }, + { + "epoch": 1.6435281837160751, + "grad_norm": 0.9472903609275818, + "learning_rate": 4.24226151138916e-06, + "loss": 0.7409, + "step": 1049 + }, + { + "epoch": 1.6450939457202505, + "grad_norm": 1.0612895488739014, + "learning_rate": 4.240750004627217e-06, + "loss": 0.7605, + "step": 1050 + }, + { + "epoch": 1.6466597077244258, + "grad_norm": 0.9634467959403992, + "learning_rate": 4.239237261677863e-06, + "loss": 0.7109, + "step": 1051 + }, + { + "epoch": 1.6482254697286014, + "grad_norm": 1.1347044706344604, + "learning_rate": 4.2377232836153635e-06, + "loss": 0.693, + "step": 1052 + }, + { + "epoch": 1.6497912317327765, + "grad_norm": 1.0814287662506104, + "learning_rate": 4.236208071514865e-06, + "loss": 0.7159, + "step": 1053 + }, + { + "epoch": 1.651356993736952, + "grad_norm": 0.9427600502967834, + "learning_rate": 4.234691626452392e-06, + "loss": 0.7014, + "step": 1054 + }, + { + "epoch": 1.6529227557411272, + "grad_norm": 1.0177332162857056, + "learning_rate": 4.233173949504842e-06, + "loss": 0.7717, + "step": 1055 + }, + { + "epoch": 1.6544885177453028, + "grad_norm": 1.0621635913848877, + "learning_rate": 4.231655041749985e-06, + "loss": 0.7078, + "step": 1056 + }, + { + "epoch": 1.656054279749478, + "grad_norm": 1.0474066734313965, + "learning_rate": 4.230134904266472e-06, + "loss": 0.7869, + "step": 1057 + }, + { + "epoch": 1.6576200417536535, + "grad_norm": 0.9170332551002502, + "learning_rate": 4.228613538133821e-06, + "loss": 0.7806, + "step": 1058 + }, + { + "epoch": 1.6591858037578286, + "grad_norm": 1.010465383529663, + "learning_rate": 4.227090944432426e-06, + "loss": 0.7352, + "step": 1059 + }, + { + "epoch": 1.6607515657620042, + "grad_norm": 0.988068163394928, + "learning_rate": 4.22556712424355e-06, + "loss": 0.7173, + "step": 1060 + }, + { + "epoch": 1.6623173277661796, + "grad_norm": 1.0753575563430786, + "learning_rate": 4.22404207864933e-06, + "loss": 0.6981, + "step": 1061 + }, + { + "epoch": 1.663883089770355, + "grad_norm": 1.0075691938400269, + "learning_rate": 4.222515808732769e-06, + "loss": 0.7279, + "step": 1062 + }, + { + "epoch": 1.6654488517745303, + "grad_norm": 1.0954983234405518, + "learning_rate": 4.220988315577745e-06, + "loss": 0.7675, + "step": 1063 + }, + { + "epoch": 1.6670146137787056, + "grad_norm": 1.0368746519088745, + "learning_rate": 4.2194596002690016e-06, + "loss": 0.7466, + "step": 1064 + }, + { + "epoch": 1.668580375782881, + "grad_norm": 1.083906650543213, + "learning_rate": 4.217929663892149e-06, + "loss": 0.7517, + "step": 1065 + }, + { + "epoch": 1.6701461377870563, + "grad_norm": 1.1356595754623413, + "learning_rate": 4.216398507533667e-06, + "loss": 0.7601, + "step": 1066 + }, + { + "epoch": 1.6717118997912317, + "grad_norm": 1.1610008478164673, + "learning_rate": 4.2148661322809e-06, + "loss": 0.7356, + "step": 1067 + }, + { + "epoch": 1.673277661795407, + "grad_norm": 1.019673466682434, + "learning_rate": 4.21333253922206e-06, + "loss": 0.7618, + "step": 1068 + }, + { + "epoch": 1.6748434237995826, + "grad_norm": 1.1470829248428345, + "learning_rate": 4.2117977294462235e-06, + "loss": 0.8001, + "step": 1069 + }, + { + "epoch": 1.6764091858037578, + "grad_norm": 0.9750358462333679, + "learning_rate": 4.210261704043327e-06, + "loss": 0.6736, + "step": 1070 + }, + { + "epoch": 1.6779749478079333, + "grad_norm": 1.061862826347351, + "learning_rate": 4.208724464104176e-06, + "loss": 0.7751, + "step": 1071 + }, + { + "epoch": 1.6795407098121085, + "grad_norm": 1.0311503410339355, + "learning_rate": 4.2071860107204346e-06, + "loss": 0.7629, + "step": 1072 + }, + { + "epoch": 1.681106471816284, + "grad_norm": 1.0023622512817383, + "learning_rate": 4.2056463449846305e-06, + "loss": 0.6935, + "step": 1073 + }, + { + "epoch": 1.6826722338204592, + "grad_norm": 1.1624165773391724, + "learning_rate": 4.204105467990153e-06, + "loss": 0.7451, + "step": 1074 + }, + { + "epoch": 1.6842379958246347, + "grad_norm": 0.9852432608604431, + "learning_rate": 4.202563380831249e-06, + "loss": 0.7446, + "step": 1075 + }, + { + "epoch": 1.6858037578288099, + "grad_norm": 1.0009260177612305, + "learning_rate": 4.201020084603027e-06, + "loss": 0.726, + "step": 1076 + }, + { + "epoch": 1.6873695198329854, + "grad_norm": 1.0374701023101807, + "learning_rate": 4.199475580401452e-06, + "loss": 0.6687, + "step": 1077 + }, + { + "epoch": 1.6889352818371608, + "grad_norm": 1.1437046527862549, + "learning_rate": 4.19792986932335e-06, + "loss": 0.7802, + "step": 1078 + }, + { + "epoch": 1.6905010438413361, + "grad_norm": 0.9866487383842468, + "learning_rate": 4.1963829524664e-06, + "loss": 0.7117, + "step": 1079 + }, + { + "epoch": 1.6920668058455115, + "grad_norm": 0.9720652103424072, + "learning_rate": 4.194834830929142e-06, + "loss": 0.778, + "step": 1080 + }, + { + "epoch": 1.6936325678496869, + "grad_norm": 1.009422779083252, + "learning_rate": 4.1932855058109675e-06, + "loss": 0.6783, + "step": 1081 + }, + { + "epoch": 1.6951983298538622, + "grad_norm": 1.024781346321106, + "learning_rate": 4.191734978212124e-06, + "loss": 0.6697, + "step": 1082 + }, + { + "epoch": 1.6967640918580376, + "grad_norm": 1.060354471206665, + "learning_rate": 4.190183249233715e-06, + "loss": 0.7151, + "step": 1083 + }, + { + "epoch": 1.698329853862213, + "grad_norm": 1.118486762046814, + "learning_rate": 4.1886303199776924e-06, + "loss": 0.7408, + "step": 1084 + }, + { + "epoch": 1.6998956158663883, + "grad_norm": 0.9334209561347961, + "learning_rate": 4.187076191546865e-06, + "loss": 0.7646, + "step": 1085 + }, + { + "epoch": 1.7014613778705638, + "grad_norm": 1.0506161451339722, + "learning_rate": 4.185520865044892e-06, + "loss": 0.7299, + "step": 1086 + }, + { + "epoch": 1.703027139874739, + "grad_norm": 1.1261264085769653, + "learning_rate": 4.1839643415762825e-06, + "loss": 0.6813, + "step": 1087 + }, + { + "epoch": 1.7045929018789145, + "grad_norm": 0.9690954685211182, + "learning_rate": 4.182406622246396e-06, + "loss": 0.7417, + "step": 1088 + }, + { + "epoch": 1.7061586638830897, + "grad_norm": 1.0872795581817627, + "learning_rate": 4.180847708161442e-06, + "loss": 0.676, + "step": 1089 + }, + { + "epoch": 1.7077244258872653, + "grad_norm": 1.054091453552246, + "learning_rate": 4.179287600428476e-06, + "loss": 0.785, + "step": 1090 + }, + { + "epoch": 1.7092901878914404, + "grad_norm": 1.109984278678894, + "learning_rate": 4.177726300155404e-06, + "loss": 0.8033, + "step": 1091 + }, + { + "epoch": 1.710855949895616, + "grad_norm": 1.0984348058700562, + "learning_rate": 4.176163808450978e-06, + "loss": 0.6964, + "step": 1092 + }, + { + "epoch": 1.712421711899791, + "grad_norm": 1.0788499116897583, + "learning_rate": 4.174600126424797e-06, + "loss": 0.7596, + "step": 1093 + }, + { + "epoch": 1.7139874739039667, + "grad_norm": 1.0846409797668457, + "learning_rate": 4.173035255187302e-06, + "loss": 0.7322, + "step": 1094 + }, + { + "epoch": 1.715553235908142, + "grad_norm": 1.1376627683639526, + "learning_rate": 4.171469195849782e-06, + "loss": 0.6934, + "step": 1095 + }, + { + "epoch": 1.7171189979123174, + "grad_norm": 1.003432035446167, + "learning_rate": 4.169901949524368e-06, + "loss": 0.7663, + "step": 1096 + }, + { + "epoch": 1.7186847599164927, + "grad_norm": 0.9824592471122742, + "learning_rate": 4.168333517324036e-06, + "loss": 0.8253, + "step": 1097 + }, + { + "epoch": 1.720250521920668, + "grad_norm": 1.0256636142730713, + "learning_rate": 4.1667639003626e-06, + "loss": 0.6614, + "step": 1098 + }, + { + "epoch": 1.7218162839248434, + "grad_norm": 1.0290439128875732, + "learning_rate": 4.16519309975472e-06, + "loss": 0.7403, + "step": 1099 + }, + { + "epoch": 1.7233820459290188, + "grad_norm": 1.0358202457427979, + "learning_rate": 4.163621116615892e-06, + "loss": 0.7332, + "step": 1100 + }, + { + "epoch": 1.7249478079331941, + "grad_norm": 1.1046626567840576, + "learning_rate": 4.162047952062456e-06, + "loss": 0.6672, + "step": 1101 + }, + { + "epoch": 1.7265135699373695, + "grad_norm": 1.116408109664917, + "learning_rate": 4.16047360721159e-06, + "loss": 0.6997, + "step": 1102 + }, + { + "epoch": 1.7280793319415448, + "grad_norm": 0.98244309425354, + "learning_rate": 4.158898083181308e-06, + "loss": 0.754, + "step": 1103 + }, + { + "epoch": 1.7296450939457202, + "grad_norm": 1.033761978149414, + "learning_rate": 4.1573213810904635e-06, + "loss": 0.7317, + "step": 1104 + }, + { + "epoch": 1.7312108559498958, + "grad_norm": 1.1415889263153076, + "learning_rate": 4.155743502058745e-06, + "loss": 0.8361, + "step": 1105 + }, + { + "epoch": 1.732776617954071, + "grad_norm": 0.9897739291191101, + "learning_rate": 4.154164447206679e-06, + "loss": 0.7361, + "step": 1106 + }, + { + "epoch": 1.7343423799582465, + "grad_norm": 1.0837332010269165, + "learning_rate": 4.1525842176556235e-06, + "loss": 0.7007, + "step": 1107 + }, + { + "epoch": 1.7359081419624216, + "grad_norm": 1.021332025527954, + "learning_rate": 4.151002814527774e-06, + "loss": 0.723, + "step": 1108 + }, + { + "epoch": 1.7374739039665972, + "grad_norm": 0.9808626770973206, + "learning_rate": 4.14942023894616e-06, + "loss": 0.6121, + "step": 1109 + }, + { + "epoch": 1.7390396659707723, + "grad_norm": 1.0266727209091187, + "learning_rate": 4.14783649203464e-06, + "loss": 0.8221, + "step": 1110 + }, + { + "epoch": 1.7406054279749479, + "grad_norm": 1.0880744457244873, + "learning_rate": 4.146251574917907e-06, + "loss": 0.7159, + "step": 1111 + }, + { + "epoch": 1.742171189979123, + "grad_norm": 1.0125631093978882, + "learning_rate": 4.144665488721483e-06, + "loss": 0.7228, + "step": 1112 + }, + { + "epoch": 1.7437369519832986, + "grad_norm": 0.9339308142662048, + "learning_rate": 4.143078234571723e-06, + "loss": 0.7472, + "step": 1113 + }, + { + "epoch": 1.745302713987474, + "grad_norm": 1.1234889030456543, + "learning_rate": 4.141489813595809e-06, + "loss": 0.7875, + "step": 1114 + }, + { + "epoch": 1.7468684759916493, + "grad_norm": 1.0862795114517212, + "learning_rate": 4.139900226921753e-06, + "loss": 0.7219, + "step": 1115 + }, + { + "epoch": 1.7484342379958246, + "grad_norm": 1.027912974357605, + "learning_rate": 4.138309475678393e-06, + "loss": 0.6616, + "step": 1116 + }, + { + "epoch": 1.75, + "grad_norm": 1.0271416902542114, + "learning_rate": 4.136717560995398e-06, + "loss": 0.7245, + "step": 1117 + }, + { + "epoch": 1.7515657620041754, + "grad_norm": 0.9430286288261414, + "learning_rate": 4.135124484003258e-06, + "loss": 0.7347, + "step": 1118 + }, + { + "epoch": 1.7531315240083507, + "grad_norm": 1.1191871166229248, + "learning_rate": 4.133530245833292e-06, + "loss": 0.6713, + "step": 1119 + }, + { + "epoch": 1.754697286012526, + "grad_norm": 1.0734976530075073, + "learning_rate": 4.131934847617642e-06, + "loss": 0.7268, + "step": 1120 + }, + { + "epoch": 1.7562630480167014, + "grad_norm": 1.0727618932724, + "learning_rate": 4.130338290489276e-06, + "loss": 0.7855, + "step": 1121 + }, + { + "epoch": 1.757828810020877, + "grad_norm": 1.0351287126541138, + "learning_rate": 4.1287405755819805e-06, + "loss": 0.7297, + "step": 1122 + }, + { + "epoch": 1.7593945720250521, + "grad_norm": 1.046382188796997, + "learning_rate": 4.127141704030371e-06, + "loss": 0.6508, + "step": 1123 + }, + { + "epoch": 1.7609603340292277, + "grad_norm": 1.0040491819381714, + "learning_rate": 4.125541676969876e-06, + "loss": 0.7707, + "step": 1124 + }, + { + "epoch": 1.7625260960334028, + "grad_norm": 1.1626756191253662, + "learning_rate": 4.123940495536752e-06, + "loss": 0.7366, + "step": 1125 + }, + { + "epoch": 1.7640918580375784, + "grad_norm": 0.9720292091369629, + "learning_rate": 4.122338160868071e-06, + "loss": 0.7641, + "step": 1126 + }, + { + "epoch": 1.7656576200417535, + "grad_norm": 0.9957835078239441, + "learning_rate": 4.1207346741017265e-06, + "loss": 0.7398, + "step": 1127 + }, + { + "epoch": 1.767223382045929, + "grad_norm": 0.972172737121582, + "learning_rate": 4.1191300363764276e-06, + "loss": 0.6565, + "step": 1128 + }, + { + "epoch": 1.7687891440501042, + "grad_norm": 1.0128676891326904, + "learning_rate": 4.117524248831702e-06, + "loss": 0.7433, + "step": 1129 + }, + { + "epoch": 1.7703549060542798, + "grad_norm": 1.1161550283432007, + "learning_rate": 4.115917312607895e-06, + "loss": 0.8241, + "step": 1130 + }, + { + "epoch": 1.7719206680584552, + "grad_norm": 1.009810209274292, + "learning_rate": 4.114309228846167e-06, + "loss": 0.6939, + "step": 1131 + }, + { + "epoch": 1.7734864300626305, + "grad_norm": 1.0082894563674927, + "learning_rate": 4.112699998688492e-06, + "loss": 0.6772, + "step": 1132 + }, + { + "epoch": 1.7750521920668059, + "grad_norm": 0.9453067183494568, + "learning_rate": 4.111089623277658e-06, + "loss": 0.6982, + "step": 1133 + }, + { + "epoch": 1.7766179540709812, + "grad_norm": 1.074747920036316, + "learning_rate": 4.1094781037572696e-06, + "loss": 0.7296, + "step": 1134 + }, + { + "epoch": 1.7781837160751566, + "grad_norm": 1.0477592945098877, + "learning_rate": 4.1078654412717404e-06, + "loss": 0.8162, + "step": 1135 + }, + { + "epoch": 1.779749478079332, + "grad_norm": 1.1362730264663696, + "learning_rate": 4.106251636966297e-06, + "loss": 0.6942, + "step": 1136 + }, + { + "epoch": 1.7813152400835073, + "grad_norm": 0.9564191102981567, + "learning_rate": 4.104636691986977e-06, + "loss": 0.7613, + "step": 1137 + }, + { + "epoch": 1.7828810020876826, + "grad_norm": 1.0384364128112793, + "learning_rate": 4.103020607480628e-06, + "loss": 0.6482, + "step": 1138 + }, + { + "epoch": 1.784446764091858, + "grad_norm": 1.1258740425109863, + "learning_rate": 4.101403384594905e-06, + "loss": 0.7736, + "step": 1139 + }, + { + "epoch": 1.7860125260960333, + "grad_norm": 1.0740283727645874, + "learning_rate": 4.099785024478276e-06, + "loss": 0.7219, + "step": 1140 + }, + { + "epoch": 1.787578288100209, + "grad_norm": 1.0377696752548218, + "learning_rate": 4.098165528280012e-06, + "loss": 0.7132, + "step": 1141 + }, + { + "epoch": 1.789144050104384, + "grad_norm": 1.0551276206970215, + "learning_rate": 4.096544897150192e-06, + "loss": 0.7879, + "step": 1142 + }, + { + "epoch": 1.7907098121085596, + "grad_norm": 1.096389651298523, + "learning_rate": 4.0949231322397036e-06, + "loss": 0.7516, + "step": 1143 + }, + { + "epoch": 1.7922755741127347, + "grad_norm": 0.9602596163749695, + "learning_rate": 4.093300234700234e-06, + "loss": 0.7197, + "step": 1144 + }, + { + "epoch": 1.7938413361169103, + "grad_norm": 1.1918690204620361, + "learning_rate": 4.09167620568428e-06, + "loss": 0.7634, + "step": 1145 + }, + { + "epoch": 1.7954070981210855, + "grad_norm": 1.1249059438705444, + "learning_rate": 4.09005104634514e-06, + "loss": 0.7718, + "step": 1146 + }, + { + "epoch": 1.796972860125261, + "grad_norm": 1.0596246719360352, + "learning_rate": 4.0884247578369164e-06, + "loss": 0.7258, + "step": 1147 + }, + { + "epoch": 1.7985386221294362, + "grad_norm": 1.026201844215393, + "learning_rate": 4.086797341314509e-06, + "loss": 0.6426, + "step": 1148 + }, + { + "epoch": 1.8001043841336117, + "grad_norm": 1.0791343450546265, + "learning_rate": 4.085168797933625e-06, + "loss": 0.7147, + "step": 1149 + }, + { + "epoch": 1.801670146137787, + "grad_norm": 0.9660344123840332, + "learning_rate": 4.083539128850766e-06, + "loss": 0.7335, + "step": 1150 + }, + { + "epoch": 1.8032359081419624, + "grad_norm": 1.0022811889648438, + "learning_rate": 4.081908335223238e-06, + "loss": 0.6824, + "step": 1151 + }, + { + "epoch": 1.8048016701461378, + "grad_norm": 0.9261506199836731, + "learning_rate": 4.080276418209143e-06, + "loss": 0.7538, + "step": 1152 + }, + { + "epoch": 1.8063674321503131, + "grad_norm": 1.088552713394165, + "learning_rate": 4.078643378967379e-06, + "loss": 0.6988, + "step": 1153 + }, + { + "epoch": 1.8079331941544885, + "grad_norm": 1.0581461191177368, + "learning_rate": 4.077009218657645e-06, + "loss": 0.7075, + "step": 1154 + }, + { + "epoch": 1.8094989561586639, + "grad_norm": 1.0722301006317139, + "learning_rate": 4.0753739384404335e-06, + "loss": 0.6974, + "step": 1155 + }, + { + "epoch": 1.8110647181628392, + "grad_norm": 0.9519115090370178, + "learning_rate": 4.073737539477033e-06, + "loss": 0.6898, + "step": 1156 + }, + { + "epoch": 1.8126304801670146, + "grad_norm": 1.1201752424240112, + "learning_rate": 4.072100022929526e-06, + "loss": 0.8572, + "step": 1157 + }, + { + "epoch": 1.8141962421711901, + "grad_norm": 1.24104642868042, + "learning_rate": 4.070461389960789e-06, + "loss": 0.7407, + "step": 1158 + }, + { + "epoch": 1.8157620041753653, + "grad_norm": 1.0625057220458984, + "learning_rate": 4.068821641734492e-06, + "loss": 0.655, + "step": 1159 + }, + { + "epoch": 1.8173277661795408, + "grad_norm": 1.1174049377441406, + "learning_rate": 4.067180779415097e-06, + "loss": 0.7214, + "step": 1160 + }, + { + "epoch": 1.818893528183716, + "grad_norm": 1.0683097839355469, + "learning_rate": 4.065538804167855e-06, + "loss": 0.7863, + "step": 1161 + }, + { + "epoch": 1.8204592901878915, + "grad_norm": 1.031198263168335, + "learning_rate": 4.063895717158809e-06, + "loss": 0.7663, + "step": 1162 + }, + { + "epoch": 1.8220250521920667, + "grad_norm": 1.1091554164886475, + "learning_rate": 4.062251519554791e-06, + "loss": 0.7234, + "step": 1163 + }, + { + "epoch": 1.8235908141962422, + "grad_norm": 1.0818198919296265, + "learning_rate": 4.060606212523425e-06, + "loss": 0.7709, + "step": 1164 + }, + { + "epoch": 1.8251565762004174, + "grad_norm": 1.0104292631149292, + "learning_rate": 4.058959797233116e-06, + "loss": 0.7225, + "step": 1165 + }, + { + "epoch": 1.826722338204593, + "grad_norm": 1.1726491451263428, + "learning_rate": 4.057312274853063e-06, + "loss": 0.6831, + "step": 1166 + }, + { + "epoch": 1.8282881002087683, + "grad_norm": 0.9735565185546875, + "learning_rate": 4.055663646553248e-06, + "loss": 0.7123, + "step": 1167 + }, + { + "epoch": 1.8298538622129437, + "grad_norm": 1.011709213256836, + "learning_rate": 4.054013913504436e-06, + "loss": 0.6885, + "step": 1168 + }, + { + "epoch": 1.831419624217119, + "grad_norm": 1.0374364852905273, + "learning_rate": 4.05236307687818e-06, + "loss": 0.782, + "step": 1169 + }, + { + "epoch": 1.8329853862212944, + "grad_norm": 1.011202096939087, + "learning_rate": 4.050711137846817e-06, + "loss": 0.7381, + "step": 1170 + }, + { + "epoch": 1.8345511482254697, + "grad_norm": 1.0984755754470825, + "learning_rate": 4.049058097583464e-06, + "loss": 0.7472, + "step": 1171 + }, + { + "epoch": 1.836116910229645, + "grad_norm": 1.090078353881836, + "learning_rate": 4.047403957262024e-06, + "loss": 0.6773, + "step": 1172 + }, + { + "epoch": 1.8376826722338204, + "grad_norm": 0.9866096377372742, + "learning_rate": 4.045748718057176e-06, + "loss": 0.6812, + "step": 1173 + }, + { + "epoch": 1.8392484342379958, + "grad_norm": 1.1006139516830444, + "learning_rate": 4.0440923811443846e-06, + "loss": 0.7953, + "step": 1174 + }, + { + "epoch": 1.8408141962421714, + "grad_norm": 0.9621697664260864, + "learning_rate": 4.04243494769989e-06, + "loss": 0.7557, + "step": 1175 + }, + { + "epoch": 1.8423799582463465, + "grad_norm": 1.1520715951919556, + "learning_rate": 4.040776418900714e-06, + "loss": 0.7422, + "step": 1176 + }, + { + "epoch": 1.843945720250522, + "grad_norm": 1.0679237842559814, + "learning_rate": 4.039116795924653e-06, + "loss": 0.6738, + "step": 1177 + }, + { + "epoch": 1.8455114822546972, + "grad_norm": 1.026510238647461, + "learning_rate": 4.037456079950284e-06, + "loss": 0.7741, + "step": 1178 + }, + { + "epoch": 1.8470772442588728, + "grad_norm": 1.0111522674560547, + "learning_rate": 4.0357942721569585e-06, + "loss": 0.6968, + "step": 1179 + }, + { + "epoch": 1.848643006263048, + "grad_norm": 1.0290493965148926, + "learning_rate": 4.034131373724802e-06, + "loss": 0.7131, + "step": 1180 + }, + { + "epoch": 1.8502087682672235, + "grad_norm": 0.9878649711608887, + "learning_rate": 4.032467385834718e-06, + "loss": 0.7477, + "step": 1181 + }, + { + "epoch": 1.8517745302713986, + "grad_norm": 1.0317823886871338, + "learning_rate": 4.030802309668379e-06, + "loss": 0.6318, + "step": 1182 + }, + { + "epoch": 1.8533402922755742, + "grad_norm": 1.0163817405700684, + "learning_rate": 4.029136146408235e-06, + "loss": 0.7027, + "step": 1183 + }, + { + "epoch": 1.8549060542797495, + "grad_norm": 0.9590600728988647, + "learning_rate": 4.027468897237505e-06, + "loss": 0.6988, + "step": 1184 + }, + { + "epoch": 1.8564718162839249, + "grad_norm": 1.0645931959152222, + "learning_rate": 4.0258005633401795e-06, + "loss": 0.7588, + "step": 1185 + }, + { + "epoch": 1.8580375782881002, + "grad_norm": 1.0380210876464844, + "learning_rate": 4.024131145901022e-06, + "loss": 0.6984, + "step": 1186 + }, + { + "epoch": 1.8596033402922756, + "grad_norm": 1.0558087825775146, + "learning_rate": 4.022460646105561e-06, + "loss": 0.6729, + "step": 1187 + }, + { + "epoch": 1.861169102296451, + "grad_norm": 1.0865014791488647, + "learning_rate": 4.020789065140097e-06, + "loss": 0.7208, + "step": 1188 + }, + { + "epoch": 1.8627348643006263, + "grad_norm": 1.0456374883651733, + "learning_rate": 4.019116404191697e-06, + "loss": 0.7759, + "step": 1189 + }, + { + "epoch": 1.8643006263048016, + "grad_norm": 0.9724026918411255, + "learning_rate": 4.017442664448197e-06, + "loss": 0.7475, + "step": 1190 + }, + { + "epoch": 1.865866388308977, + "grad_norm": 1.1170289516448975, + "learning_rate": 4.015767847098194e-06, + "loss": 0.6993, + "step": 1191 + }, + { + "epoch": 1.8674321503131524, + "grad_norm": 1.1052978038787842, + "learning_rate": 4.014091953331056e-06, + "loss": 0.748, + "step": 1192 + }, + { + "epoch": 1.8689979123173277, + "grad_norm": 1.1223454475402832, + "learning_rate": 4.012414984336912e-06, + "loss": 0.6921, + "step": 1193 + }, + { + "epoch": 1.8705636743215033, + "grad_norm": 1.0611138343811035, + "learning_rate": 4.010736941306655e-06, + "loss": 0.7859, + "step": 1194 + }, + { + "epoch": 1.8721294363256784, + "grad_norm": 1.169189691543579, + "learning_rate": 4.009057825431944e-06, + "loss": 0.5874, + "step": 1195 + }, + { + "epoch": 1.873695198329854, + "grad_norm": 1.036626935005188, + "learning_rate": 4.0073776379051945e-06, + "loss": 0.6801, + "step": 1196 + }, + { + "epoch": 1.8752609603340291, + "grad_norm": 0.9431620240211487, + "learning_rate": 4.005696379919586e-06, + "loss": 0.7476, + "step": 1197 + }, + { + "epoch": 1.8768267223382047, + "grad_norm": 1.0381059646606445, + "learning_rate": 4.0040140526690585e-06, + "loss": 0.6691, + "step": 1198 + }, + { + "epoch": 1.8783924843423798, + "grad_norm": 1.120033860206604, + "learning_rate": 4.0023306573483105e-06, + "loss": 0.6669, + "step": 1199 + }, + { + "epoch": 1.8799582463465554, + "grad_norm": 1.0298967361450195, + "learning_rate": 4.000646195152798e-06, + "loss": 0.7355, + "step": 1200 + }, + { + "epoch": 1.8815240083507305, + "grad_norm": 1.0036594867706299, + "learning_rate": 3.99896066727874e-06, + "loss": 0.7545, + "step": 1201 + }, + { + "epoch": 1.883089770354906, + "grad_norm": 1.1166342496871948, + "learning_rate": 3.997274074923103e-06, + "loss": 0.7468, + "step": 1202 + }, + { + "epoch": 1.8846555323590815, + "grad_norm": 1.0512094497680664, + "learning_rate": 3.995586419283617e-06, + "loss": 0.7647, + "step": 1203 + }, + { + "epoch": 1.8862212943632568, + "grad_norm": 1.0115973949432373, + "learning_rate": 3.993897701558764e-06, + "loss": 0.7893, + "step": 1204 + }, + { + "epoch": 1.8877870563674322, + "grad_norm": 1.0839310884475708, + "learning_rate": 3.992207922947783e-06, + "loss": 0.6824, + "step": 1205 + }, + { + "epoch": 1.8893528183716075, + "grad_norm": 0.9987578988075256, + "learning_rate": 3.99051708465066e-06, + "loss": 0.8045, + "step": 1206 + }, + { + "epoch": 1.8909185803757829, + "grad_norm": 1.0759059190750122, + "learning_rate": 3.9888251878681425e-06, + "loss": 0.7756, + "step": 1207 + }, + { + "epoch": 1.8924843423799582, + "grad_norm": 1.1055301427841187, + "learning_rate": 3.987132233801722e-06, + "loss": 0.6947, + "step": 1208 + }, + { + "epoch": 1.8940501043841336, + "grad_norm": 1.042575478553772, + "learning_rate": 3.985438223653644e-06, + "loss": 0.6669, + "step": 1209 + }, + { + "epoch": 1.895615866388309, + "grad_norm": 1.0595242977142334, + "learning_rate": 3.983743158626904e-06, + "loss": 0.6937, + "step": 1210 + }, + { + "epoch": 1.8971816283924845, + "grad_norm": 1.0498789548873901, + "learning_rate": 3.982047039925247e-06, + "loss": 0.7244, + "step": 1211 + }, + { + "epoch": 1.8987473903966596, + "grad_norm": 1.0960590839385986, + "learning_rate": 3.980349868753166e-06, + "loss": 0.7661, + "step": 1212 + }, + { + "epoch": 1.9003131524008352, + "grad_norm": 1.108823299407959, + "learning_rate": 3.9786516463159e-06, + "loss": 0.6937, + "step": 1213 + }, + { + "epoch": 1.9018789144050103, + "grad_norm": 1.051416039466858, + "learning_rate": 3.976952373819436e-06, + "loss": 0.7263, + "step": 1214 + }, + { + "epoch": 1.903444676409186, + "grad_norm": 0.9884061217308044, + "learning_rate": 3.9752520524705065e-06, + "loss": 0.6577, + "step": 1215 + }, + { + "epoch": 1.905010438413361, + "grad_norm": 1.0950369834899902, + "learning_rate": 3.9735506834765885e-06, + "loss": 0.7376, + "step": 1216 + }, + { + "epoch": 1.9065762004175366, + "grad_norm": 1.1140044927597046, + "learning_rate": 3.971848268045902e-06, + "loss": 0.7125, + "step": 1217 + }, + { + "epoch": 1.9081419624217117, + "grad_norm": 1.0908716917037964, + "learning_rate": 3.970144807387414e-06, + "loss": 0.7796, + "step": 1218 + }, + { + "epoch": 1.9097077244258873, + "grad_norm": 1.1068549156188965, + "learning_rate": 3.968440302710829e-06, + "loss": 0.7454, + "step": 1219 + }, + { + "epoch": 1.9112734864300627, + "grad_norm": 1.102594017982483, + "learning_rate": 3.9667347552265945e-06, + "loss": 0.757, + "step": 1220 + }, + { + "epoch": 1.912839248434238, + "grad_norm": 1.1588326692581177, + "learning_rate": 3.9650281661459015e-06, + "loss": 0.6491, + "step": 1221 + }, + { + "epoch": 1.9144050104384134, + "grad_norm": 0.9896461367607117, + "learning_rate": 3.963320536680675e-06, + "loss": 0.773, + "step": 1222 + }, + { + "epoch": 1.9159707724425887, + "grad_norm": 1.0855952501296997, + "learning_rate": 3.961611868043583e-06, + "loss": 0.6906, + "step": 1223 + }, + { + "epoch": 1.917536534446764, + "grad_norm": 1.0556972026824951, + "learning_rate": 3.959902161448032e-06, + "loss": 0.7577, + "step": 1224 + }, + { + "epoch": 1.9191022964509394, + "grad_norm": 1.0284528732299805, + "learning_rate": 3.958191418108162e-06, + "loss": 0.6844, + "step": 1225 + }, + { + "epoch": 1.9206680584551148, + "grad_norm": 0.9578842520713806, + "learning_rate": 3.956479639238852e-06, + "loss": 0.7216, + "step": 1226 + }, + { + "epoch": 1.9222338204592901, + "grad_norm": 1.0366833209991455, + "learning_rate": 3.954766826055717e-06, + "loss": 0.7365, + "step": 1227 + }, + { + "epoch": 1.9237995824634657, + "grad_norm": 1.0498946905136108, + "learning_rate": 3.953052979775103e-06, + "loss": 0.6788, + "step": 1228 + }, + { + "epoch": 1.9253653444676408, + "grad_norm": 1.010860562324524, + "learning_rate": 3.951338101614094e-06, + "loss": 0.7181, + "step": 1229 + }, + { + "epoch": 1.9269311064718164, + "grad_norm": 1.0203677415847778, + "learning_rate": 3.949622192790502e-06, + "loss": 0.7277, + "step": 1230 + }, + { + "epoch": 1.9284968684759916, + "grad_norm": 0.9907971024513245, + "learning_rate": 3.947905254522876e-06, + "loss": 0.6858, + "step": 1231 + }, + { + "epoch": 1.9300626304801671, + "grad_norm": 0.9824337363243103, + "learning_rate": 3.946187288030493e-06, + "loss": 0.7517, + "step": 1232 + }, + { + "epoch": 1.9316283924843423, + "grad_norm": 1.0206327438354492, + "learning_rate": 3.944468294533359e-06, + "loss": 0.7225, + "step": 1233 + }, + { + "epoch": 1.9331941544885178, + "grad_norm": 1.0529900789260864, + "learning_rate": 3.942748275252214e-06, + "loss": 0.7965, + "step": 1234 + }, + { + "epoch": 1.934759916492693, + "grad_norm": 1.0501983165740967, + "learning_rate": 3.941027231408521e-06, + "loss": 0.7646, + "step": 1235 + }, + { + "epoch": 1.9363256784968685, + "grad_norm": 0.9482662081718445, + "learning_rate": 3.939305164224474e-06, + "loss": 0.7449, + "step": 1236 + }, + { + "epoch": 1.937891440501044, + "grad_norm": 1.096723198890686, + "learning_rate": 3.937582074922993e-06, + "loss": 0.7464, + "step": 1237 + }, + { + "epoch": 1.9394572025052192, + "grad_norm": 1.0417059659957886, + "learning_rate": 3.935857964727725e-06, + "loss": 0.6586, + "step": 1238 + }, + { + "epoch": 1.9410229645093946, + "grad_norm": 1.1008849143981934, + "learning_rate": 3.934132834863039e-06, + "loss": 0.6236, + "step": 1239 + }, + { + "epoch": 1.94258872651357, + "grad_norm": 0.9107227921485901, + "learning_rate": 3.93240668655403e-06, + "loss": 0.7069, + "step": 1240 + }, + { + "epoch": 1.9441544885177453, + "grad_norm": 1.0516945123672485, + "learning_rate": 3.930679521026517e-06, + "loss": 0.773, + "step": 1241 + }, + { + "epoch": 1.9457202505219207, + "grad_norm": 0.9941951632499695, + "learning_rate": 3.92895133950704e-06, + "loss": 0.6002, + "step": 1242 + }, + { + "epoch": 1.947286012526096, + "grad_norm": 1.031784176826477, + "learning_rate": 3.927222143222859e-06, + "loss": 0.7737, + "step": 1243 + }, + { + "epoch": 1.9488517745302714, + "grad_norm": 0.986473798751831, + "learning_rate": 3.925491933401961e-06, + "loss": 0.7455, + "step": 1244 + }, + { + "epoch": 1.9504175365344467, + "grad_norm": 1.109964370727539, + "learning_rate": 3.923760711273044e-06, + "loss": 0.7048, + "step": 1245 + }, + { + "epoch": 1.951983298538622, + "grad_norm": 1.1303372383117676, + "learning_rate": 3.922028478065532e-06, + "loss": 0.749, + "step": 1246 + }, + { + "epoch": 1.9535490605427976, + "grad_norm": 0.999824583530426, + "learning_rate": 3.920295235009562e-06, + "loss": 0.7596, + "step": 1247 + }, + { + "epoch": 1.9551148225469728, + "grad_norm": 1.0440820455551147, + "learning_rate": 3.918560983335993e-06, + "loss": 0.7841, + "step": 1248 + }, + { + "epoch": 1.9566805845511483, + "grad_norm": 1.0420323610305786, + "learning_rate": 3.916825724276395e-06, + "loss": 0.7837, + "step": 1249 + }, + { + "epoch": 1.9582463465553235, + "grad_norm": 1.001997947692871, + "learning_rate": 3.915089459063058e-06, + "loss": 0.6956, + "step": 1250 + }, + { + "epoch": 1.959812108559499, + "grad_norm": 1.030626893043518, + "learning_rate": 3.913352188928984e-06, + "loss": 0.74, + "step": 1251 + }, + { + "epoch": 1.9613778705636742, + "grad_norm": 1.0981760025024414, + "learning_rate": 3.911613915107888e-06, + "loss": 0.713, + "step": 1252 + }, + { + "epoch": 1.9629436325678498, + "grad_norm": 1.020772933959961, + "learning_rate": 3.9098746388342006e-06, + "loss": 0.7605, + "step": 1253 + }, + { + "epoch": 1.964509394572025, + "grad_norm": 1.0206503868103027, + "learning_rate": 3.908134361343062e-06, + "loss": 0.7176, + "step": 1254 + }, + { + "epoch": 1.9660751565762005, + "grad_norm": 1.0471689701080322, + "learning_rate": 3.906393083870325e-06, + "loss": 0.7698, + "step": 1255 + }, + { + "epoch": 1.9676409185803758, + "grad_norm": 1.0658202171325684, + "learning_rate": 3.904650807652549e-06, + "loss": 0.7179, + "step": 1256 + }, + { + "epoch": 1.9692066805845512, + "grad_norm": 1.196192741394043, + "learning_rate": 3.902907533927007e-06, + "loss": 0.7801, + "step": 1257 + }, + { + "epoch": 1.9707724425887265, + "grad_norm": 1.0874967575073242, + "learning_rate": 3.90116326393168e-06, + "loss": 0.7672, + "step": 1258 + }, + { + "epoch": 1.9723382045929019, + "grad_norm": 1.153617024421692, + "learning_rate": 3.899417998905253e-06, + "loss": 0.752, + "step": 1259 + }, + { + "epoch": 1.9739039665970772, + "grad_norm": 1.0750453472137451, + "learning_rate": 3.89767174008712e-06, + "loss": 0.7471, + "step": 1260 + }, + { + "epoch": 1.9754697286012526, + "grad_norm": 1.0693799257278442, + "learning_rate": 3.895924488717381e-06, + "loss": 0.6704, + "step": 1261 + }, + { + "epoch": 1.977035490605428, + "grad_norm": 0.9552210569381714, + "learning_rate": 3.89417624603684e-06, + "loss": 0.7793, + "step": 1262 + }, + { + "epoch": 1.9786012526096033, + "grad_norm": 0.96712726354599, + "learning_rate": 3.892427013287007e-06, + "loss": 0.6843, + "step": 1263 + }, + { + "epoch": 1.9801670146137789, + "grad_norm": 1.1453518867492676, + "learning_rate": 3.890676791710091e-06, + "loss": 0.7532, + "step": 1264 + }, + { + "epoch": 1.981732776617954, + "grad_norm": 1.0294159650802612, + "learning_rate": 3.888925582549006e-06, + "loss": 0.703, + "step": 1265 + }, + { + "epoch": 1.9832985386221296, + "grad_norm": 1.0790715217590332, + "learning_rate": 3.8871733870473686e-06, + "loss": 0.6731, + "step": 1266 + }, + { + "epoch": 1.9848643006263047, + "grad_norm": 1.052460789680481, + "learning_rate": 3.8854202064494915e-06, + "loss": 0.748, + "step": 1267 + }, + { + "epoch": 1.9864300626304803, + "grad_norm": 1.112462043762207, + "learning_rate": 3.883666042000392e-06, + "loss": 0.7408, + "step": 1268 + }, + { + "epoch": 1.9879958246346554, + "grad_norm": 1.0738476514816284, + "learning_rate": 3.881910894945783e-06, + "loss": 0.759, + "step": 1269 + }, + { + "epoch": 1.989561586638831, + "grad_norm": 1.136472463607788, + "learning_rate": 3.8801547665320775e-06, + "loss": 0.754, + "step": 1270 + }, + { + "epoch": 1.9911273486430061, + "grad_norm": 1.0640734434127808, + "learning_rate": 3.87839765800638e-06, + "loss": 0.719, + "step": 1271 + }, + { + "epoch": 1.9926931106471817, + "grad_norm": 0.9155470132827759, + "learning_rate": 3.876639570616498e-06, + "loss": 0.762, + "step": 1272 + }, + { + "epoch": 1.994258872651357, + "grad_norm": 1.015743374824524, + "learning_rate": 3.87488050561093e-06, + "loss": 0.6926, + "step": 1273 + }, + { + "epoch": 1.9958246346555324, + "grad_norm": 1.0829862356185913, + "learning_rate": 3.87312046423887e-06, + "loss": 0.7644, + "step": 1274 + }, + { + "epoch": 1.9973903966597077, + "grad_norm": 1.0962239503860474, + "learning_rate": 3.871359447750205e-06, + "loss": 0.7038, + "step": 1275 + }, + { + "epoch": 1.998956158663883, + "grad_norm": 1.0470675230026245, + "learning_rate": 3.869597457395514e-06, + "loss": 0.7629, + "step": 1276 + }, + { + "epoch": 2.0015657620041756, + "grad_norm": 1.962398648262024, + "learning_rate": 3.867834494426069e-06, + "loss": 1.4251, + "step": 1277 + }, + { + "epoch": 2.0031315240083507, + "grad_norm": 1.0987200736999512, + "learning_rate": 3.866070560093832e-06, + "loss": 0.6873, + "step": 1278 + }, + { + "epoch": 2.0046972860125263, + "grad_norm": 1.0096511840820312, + "learning_rate": 3.864305655651453e-06, + "loss": 0.7113, + "step": 1279 + }, + { + "epoch": 2.0062630480167014, + "grad_norm": 0.9988889098167419, + "learning_rate": 3.8625397823522755e-06, + "loss": 0.6555, + "step": 1280 + }, + { + "epoch": 2.007828810020877, + "grad_norm": 0.9960927963256836, + "learning_rate": 3.860772941450326e-06, + "loss": 0.6458, + "step": 1281 + }, + { + "epoch": 2.009394572025052, + "grad_norm": 0.971942126750946, + "learning_rate": 3.85900513420032e-06, + "loss": 0.6821, + "step": 1282 + }, + { + "epoch": 2.0109603340292277, + "grad_norm": 1.0935462713241577, + "learning_rate": 3.857236361857662e-06, + "loss": 0.5862, + "step": 1283 + }, + { + "epoch": 2.012526096033403, + "grad_norm": 1.0295177698135376, + "learning_rate": 3.855466625678435e-06, + "loss": 0.723, + "step": 1284 + }, + { + "epoch": 2.0140918580375784, + "grad_norm": 0.9750491976737976, + "learning_rate": 3.853695926919415e-06, + "loss": 0.7105, + "step": 1285 + }, + { + "epoch": 2.0156576200417535, + "grad_norm": 1.1649892330169678, + "learning_rate": 3.851924266838056e-06, + "loss": 0.7106, + "step": 1286 + }, + { + "epoch": 2.017223382045929, + "grad_norm": 1.068339467048645, + "learning_rate": 3.850151646692496e-06, + "loss": 0.668, + "step": 1287 + }, + { + "epoch": 2.0187891440501042, + "grad_norm": 0.9659110903739929, + "learning_rate": 3.848378067741555e-06, + "loss": 0.6201, + "step": 1288 + }, + { + "epoch": 2.02035490605428, + "grad_norm": 1.0711618661880493, + "learning_rate": 3.846603531244734e-06, + "loss": 0.6379, + "step": 1289 + }, + { + "epoch": 2.021920668058455, + "grad_norm": 1.092109203338623, + "learning_rate": 3.844828038462213e-06, + "loss": 0.7105, + "step": 1290 + }, + { + "epoch": 2.0234864300626305, + "grad_norm": 1.1485183238983154, + "learning_rate": 3.843051590654852e-06, + "loss": 0.724, + "step": 1291 + }, + { + "epoch": 2.0250521920668056, + "grad_norm": 1.1114039421081543, + "learning_rate": 3.841274189084189e-06, + "loss": 0.6928, + "step": 1292 + }, + { + "epoch": 2.026617954070981, + "grad_norm": 1.1090906858444214, + "learning_rate": 3.839495835012441e-06, + "loss": 0.6988, + "step": 1293 + }, + { + "epoch": 2.028183716075157, + "grad_norm": 1.0808806419372559, + "learning_rate": 3.837716529702498e-06, + "loss": 0.6374, + "step": 1294 + }, + { + "epoch": 2.029749478079332, + "grad_norm": 1.0822157859802246, + "learning_rate": 3.835936274417927e-06, + "loss": 0.6249, + "step": 1295 + }, + { + "epoch": 2.0313152400835075, + "grad_norm": 1.0366182327270508, + "learning_rate": 3.834155070422972e-06, + "loss": 0.702, + "step": 1296 + }, + { + "epoch": 2.0328810020876826, + "grad_norm": 1.0268456935882568, + "learning_rate": 3.832372918982547e-06, + "loss": 0.6809, + "step": 1297 + }, + { + "epoch": 2.034446764091858, + "grad_norm": 1.1484824419021606, + "learning_rate": 3.830589821362242e-06, + "loss": 0.6442, + "step": 1298 + }, + { + "epoch": 2.0360125260960333, + "grad_norm": 1.0607216358184814, + "learning_rate": 3.828805778828315e-06, + "loss": 0.6469, + "step": 1299 + }, + { + "epoch": 2.037578288100209, + "grad_norm": 1.0950021743774414, + "learning_rate": 3.8270207926477e-06, + "loss": 0.6756, + "step": 1300 + }, + { + "epoch": 2.039144050104384, + "grad_norm": 0.9932938814163208, + "learning_rate": 3.825234864087998e-06, + "loss": 0.706, + "step": 1301 + }, + { + "epoch": 2.0407098121085596, + "grad_norm": 1.0706391334533691, + "learning_rate": 3.823447994417477e-06, + "loss": 0.6799, + "step": 1302 + }, + { + "epoch": 2.0422755741127347, + "grad_norm": 1.0093334913253784, + "learning_rate": 3.821660184905081e-06, + "loss": 0.676, + "step": 1303 + }, + { + "epoch": 2.0438413361169103, + "grad_norm": 1.0283772945404053, + "learning_rate": 3.819871436820413e-06, + "loss": 0.6738, + "step": 1304 + }, + { + "epoch": 2.0454070981210855, + "grad_norm": 1.0533452033996582, + "learning_rate": 3.818081751433746e-06, + "loss": 0.7053, + "step": 1305 + }, + { + "epoch": 2.046972860125261, + "grad_norm": 1.1219292879104614, + "learning_rate": 3.816291130016021e-06, + "loss": 0.6563, + "step": 1306 + }, + { + "epoch": 2.048538622129436, + "grad_norm": 1.025702953338623, + "learning_rate": 3.8144995738388397e-06, + "loss": 0.7264, + "step": 1307 + }, + { + "epoch": 2.0501043841336117, + "grad_norm": 1.0799325704574585, + "learning_rate": 3.8127070841744695e-06, + "loss": 0.6541, + "step": 1308 + }, + { + "epoch": 2.051670146137787, + "grad_norm": 1.043544888496399, + "learning_rate": 3.810913662295841e-06, + "loss": 0.627, + "step": 1309 + }, + { + "epoch": 2.0532359081419624, + "grad_norm": 0.9250067472457886, + "learning_rate": 3.809119309476547e-06, + "loss": 0.7447, + "step": 1310 + }, + { + "epoch": 2.0548016701461376, + "grad_norm": 1.0885530710220337, + "learning_rate": 3.8073240269908385e-06, + "loss": 0.6509, + "step": 1311 + }, + { + "epoch": 2.056367432150313, + "grad_norm": 1.2108066082000732, + "learning_rate": 3.805527816113631e-06, + "loss": 0.6844, + "step": 1312 + }, + { + "epoch": 2.0579331941544887, + "grad_norm": 1.1016451120376587, + "learning_rate": 3.8037306781204974e-06, + "loss": 0.6942, + "step": 1313 + }, + { + "epoch": 2.059498956158664, + "grad_norm": 1.080930233001709, + "learning_rate": 3.801932614287668e-06, + "loss": 0.7379, + "step": 1314 + }, + { + "epoch": 2.0610647181628394, + "grad_norm": 1.1148338317871094, + "learning_rate": 3.800133625892031e-06, + "loss": 0.6817, + "step": 1315 + }, + { + "epoch": 2.0626304801670146, + "grad_norm": 1.0792765617370605, + "learning_rate": 3.798333714211132e-06, + "loss": 0.6787, + "step": 1316 + }, + { + "epoch": 2.06419624217119, + "grad_norm": 1.219314455986023, + "learning_rate": 3.7965328805231727e-06, + "loss": 0.6451, + "step": 1317 + }, + { + "epoch": 2.0657620041753653, + "grad_norm": 1.0637946128845215, + "learning_rate": 3.7947311261070073e-06, + "loss": 0.6869, + "step": 1318 + }, + { + "epoch": 2.067327766179541, + "grad_norm": 1.112642765045166, + "learning_rate": 3.792928452242146e-06, + "loss": 0.6785, + "step": 1319 + }, + { + "epoch": 2.068893528183716, + "grad_norm": 1.006885051727295, + "learning_rate": 3.791124860208752e-06, + "loss": 0.7289, + "step": 1320 + }, + { + "epoch": 2.0704592901878915, + "grad_norm": 1.1578105688095093, + "learning_rate": 3.7893203512876385e-06, + "loss": 0.6472, + "step": 1321 + }, + { + "epoch": 2.0720250521920667, + "grad_norm": 1.0113657712936401, + "learning_rate": 3.7875149267602725e-06, + "loss": 0.6314, + "step": 1322 + }, + { + "epoch": 2.0735908141962422, + "grad_norm": 1.0363519191741943, + "learning_rate": 3.785708587908767e-06, + "loss": 0.6949, + "step": 1323 + }, + { + "epoch": 2.0751565762004174, + "grad_norm": 1.0536452531814575, + "learning_rate": 3.7839013360158904e-06, + "loss": 0.6817, + "step": 1324 + }, + { + "epoch": 2.076722338204593, + "grad_norm": 1.2700945138931274, + "learning_rate": 3.7820931723650547e-06, + "loss": 0.7364, + "step": 1325 + }, + { + "epoch": 2.078288100208768, + "grad_norm": 0.979896068572998, + "learning_rate": 3.7802840982403204e-06, + "loss": 0.6583, + "step": 1326 + }, + { + "epoch": 2.0798538622129437, + "grad_norm": 1.1849523782730103, + "learning_rate": 3.778474114926396e-06, + "loss": 0.6623, + "step": 1327 + }, + { + "epoch": 2.081419624217119, + "grad_norm": 1.0124396085739136, + "learning_rate": 3.7766632237086336e-06, + "loss": 0.7018, + "step": 1328 + }, + { + "epoch": 2.0829853862212944, + "grad_norm": 1.0735864639282227, + "learning_rate": 3.774851425873032e-06, + "loss": 0.6178, + "step": 1329 + }, + { + "epoch": 2.08455114822547, + "grad_norm": 1.1188064813613892, + "learning_rate": 3.773038722706234e-06, + "loss": 0.6603, + "step": 1330 + }, + { + "epoch": 2.086116910229645, + "grad_norm": 1.1021878719329834, + "learning_rate": 3.771225115495522e-06, + "loss": 0.7027, + "step": 1331 + }, + { + "epoch": 2.0876826722338206, + "grad_norm": 1.067917823791504, + "learning_rate": 3.769410605528824e-06, + "loss": 0.5733, + "step": 1332 + }, + { + "epoch": 2.0892484342379958, + "grad_norm": 1.199495553970337, + "learning_rate": 3.7675951940947075e-06, + "loss": 0.7376, + "step": 1333 + }, + { + "epoch": 2.0908141962421714, + "grad_norm": 1.104844093322754, + "learning_rate": 3.7657788824823805e-06, + "loss": 0.7099, + "step": 1334 + }, + { + "epoch": 2.0923799582463465, + "grad_norm": 1.1797598600387573, + "learning_rate": 3.7639616719816907e-06, + "loss": 0.6264, + "step": 1335 + }, + { + "epoch": 2.093945720250522, + "grad_norm": 1.0947678089141846, + "learning_rate": 3.7621435638831226e-06, + "loss": 0.7066, + "step": 1336 + }, + { + "epoch": 2.095511482254697, + "grad_norm": 1.1054636240005493, + "learning_rate": 3.7603245594778e-06, + "loss": 0.7439, + "step": 1337 + }, + { + "epoch": 2.0970772442588728, + "grad_norm": 1.106891393661499, + "learning_rate": 3.758504660057482e-06, + "loss": 0.6446, + "step": 1338 + }, + { + "epoch": 2.098643006263048, + "grad_norm": 1.043440818786621, + "learning_rate": 3.7566838669145644e-06, + "loss": 0.6885, + "step": 1339 + }, + { + "epoch": 2.1002087682672235, + "grad_norm": 1.0602154731750488, + "learning_rate": 3.7548621813420765e-06, + "loss": 0.6808, + "step": 1340 + }, + { + "epoch": 2.1017745302713986, + "grad_norm": 0.9517614841461182, + "learning_rate": 3.7530396046336814e-06, + "loss": 0.6369, + "step": 1341 + }, + { + "epoch": 2.103340292275574, + "grad_norm": 1.1436609029769897, + "learning_rate": 3.7512161380836763e-06, + "loss": 0.7427, + "step": 1342 + }, + { + "epoch": 2.1049060542797493, + "grad_norm": 1.00765860080719, + "learning_rate": 3.7493917829869887e-06, + "loss": 0.7012, + "step": 1343 + }, + { + "epoch": 2.106471816283925, + "grad_norm": 1.1718883514404297, + "learning_rate": 3.747566540639178e-06, + "loss": 0.6689, + "step": 1344 + }, + { + "epoch": 2.1080375782881, + "grad_norm": 1.0489915609359741, + "learning_rate": 3.745740412336434e-06, + "loss": 0.6735, + "step": 1345 + }, + { + "epoch": 2.1096033402922756, + "grad_norm": 1.1287360191345215, + "learning_rate": 3.743913399375575e-06, + "loss": 0.6211, + "step": 1346 + }, + { + "epoch": 2.1111691022964507, + "grad_norm": 1.0830544233322144, + "learning_rate": 3.7420855030540476e-06, + "loss": 0.7219, + "step": 1347 + }, + { + "epoch": 2.1127348643006263, + "grad_norm": 1.1535146236419678, + "learning_rate": 3.7402567246699257e-06, + "loss": 0.6995, + "step": 1348 + }, + { + "epoch": 2.114300626304802, + "grad_norm": 1.1392090320587158, + "learning_rate": 3.7384270655219102e-06, + "loss": 0.7118, + "step": 1349 + }, + { + "epoch": 2.115866388308977, + "grad_norm": 1.009692668914795, + "learning_rate": 3.736596526909327e-06, + "loss": 0.7211, + "step": 1350 + }, + { + "epoch": 2.1174321503131526, + "grad_norm": 1.2041988372802734, + "learning_rate": 3.7347651101321258e-06, + "loss": 0.5969, + "step": 1351 + }, + { + "epoch": 2.1189979123173277, + "grad_norm": 1.0518646240234375, + "learning_rate": 3.7329328164908818e-06, + "loss": 0.6624, + "step": 1352 + }, + { + "epoch": 2.1205636743215033, + "grad_norm": 1.1757079362869263, + "learning_rate": 3.7310996472867916e-06, + "loss": 0.7201, + "step": 1353 + }, + { + "epoch": 2.1221294363256784, + "grad_norm": 1.0831660032272339, + "learning_rate": 3.7292656038216722e-06, + "loss": 0.6302, + "step": 1354 + }, + { + "epoch": 2.123695198329854, + "grad_norm": 1.1849291324615479, + "learning_rate": 3.7274306873979654e-06, + "loss": 0.7365, + "step": 1355 + }, + { + "epoch": 2.125260960334029, + "grad_norm": 1.1130744218826294, + "learning_rate": 3.72559489931873e-06, + "loss": 0.6759, + "step": 1356 + }, + { + "epoch": 2.1268267223382047, + "grad_norm": 1.0014575719833374, + "learning_rate": 3.7237582408876426e-06, + "loss": 0.7179, + "step": 1357 + }, + { + "epoch": 2.12839248434238, + "grad_norm": 1.1398662328720093, + "learning_rate": 3.7219207134090013e-06, + "loss": 0.6678, + "step": 1358 + }, + { + "epoch": 2.1299582463465554, + "grad_norm": 1.1094143390655518, + "learning_rate": 3.720082318187719e-06, + "loss": 0.6962, + "step": 1359 + }, + { + "epoch": 2.1315240083507305, + "grad_norm": 1.0060160160064697, + "learning_rate": 3.718243056529326e-06, + "loss": 0.688, + "step": 1360 + }, + { + "epoch": 2.133089770354906, + "grad_norm": 1.0628772974014282, + "learning_rate": 3.716402929739967e-06, + "loss": 0.7124, + "step": 1361 + }, + { + "epoch": 2.1346555323590812, + "grad_norm": 1.1098920106887817, + "learning_rate": 3.714561939126402e-06, + "loss": 0.7016, + "step": 1362 + }, + { + "epoch": 2.136221294363257, + "grad_norm": 0.9880667328834534, + "learning_rate": 3.712720085996003e-06, + "loss": 0.7333, + "step": 1363 + }, + { + "epoch": 2.137787056367432, + "grad_norm": 1.1795120239257812, + "learning_rate": 3.710877371656757e-06, + "loss": 0.6729, + "step": 1364 + }, + { + "epoch": 2.1393528183716075, + "grad_norm": 1.1155129671096802, + "learning_rate": 3.7090337974172595e-06, + "loss": 0.6676, + "step": 1365 + }, + { + "epoch": 2.140918580375783, + "grad_norm": 1.1053357124328613, + "learning_rate": 3.707189364586719e-06, + "loss": 0.7043, + "step": 1366 + }, + { + "epoch": 2.142484342379958, + "grad_norm": 1.1484512090682983, + "learning_rate": 3.705344074474952e-06, + "loss": 0.7166, + "step": 1367 + }, + { + "epoch": 2.144050104384134, + "grad_norm": 1.056976318359375, + "learning_rate": 3.703497928392386e-06, + "loss": 0.7034, + "step": 1368 + }, + { + "epoch": 2.145615866388309, + "grad_norm": 1.093001365661621, + "learning_rate": 3.7016509276500546e-06, + "loss": 0.5952, + "step": 1369 + }, + { + "epoch": 2.1471816283924845, + "grad_norm": 1.094588041305542, + "learning_rate": 3.699803073559598e-06, + "loss": 0.7126, + "step": 1370 + }, + { + "epoch": 2.1487473903966596, + "grad_norm": 1.126368522644043, + "learning_rate": 3.6979543674332634e-06, + "loss": 0.714, + "step": 1371 + }, + { + "epoch": 2.150313152400835, + "grad_norm": 1.0853040218353271, + "learning_rate": 3.696104810583904e-06, + "loss": 0.721, + "step": 1372 + }, + { + "epoch": 2.1518789144050103, + "grad_norm": 1.091913104057312, + "learning_rate": 3.694254404324975e-06, + "loss": 0.6752, + "step": 1373 + }, + { + "epoch": 2.153444676409186, + "grad_norm": 1.1095316410064697, + "learning_rate": 3.6924031499705366e-06, + "loss": 0.6988, + "step": 1374 + }, + { + "epoch": 2.155010438413361, + "grad_norm": 1.1327425241470337, + "learning_rate": 3.690551048835249e-06, + "loss": 0.6563, + "step": 1375 + }, + { + "epoch": 2.1565762004175366, + "grad_norm": 1.1065635681152344, + "learning_rate": 3.688698102234377e-06, + "loss": 0.683, + "step": 1376 + }, + { + "epoch": 2.1581419624217117, + "grad_norm": 1.050032138824463, + "learning_rate": 3.6868443114837828e-06, + "loss": 0.6471, + "step": 1377 + }, + { + "epoch": 2.1597077244258873, + "grad_norm": 1.2008466720581055, + "learning_rate": 3.6849896778999293e-06, + "loss": 0.7257, + "step": 1378 + }, + { + "epoch": 2.1612734864300625, + "grad_norm": 1.101426601409912, + "learning_rate": 3.6831342027998796e-06, + "loss": 0.6542, + "step": 1379 + }, + { + "epoch": 2.162839248434238, + "grad_norm": 1.0931123495101929, + "learning_rate": 3.68127788750129e-06, + "loss": 0.665, + "step": 1380 + }, + { + "epoch": 2.164405010438413, + "grad_norm": 1.2418708801269531, + "learning_rate": 3.6794207333224184e-06, + "loss": 0.6399, + "step": 1381 + }, + { + "epoch": 2.1659707724425887, + "grad_norm": 1.1205544471740723, + "learning_rate": 3.6775627415821158e-06, + "loss": 0.7029, + "step": 1382 + }, + { + "epoch": 2.1675365344467643, + "grad_norm": 1.0413703918457031, + "learning_rate": 3.6757039135998276e-06, + "loss": 0.7521, + "step": 1383 + }, + { + "epoch": 2.1691022964509394, + "grad_norm": 1.0559121370315552, + "learning_rate": 3.673844250695594e-06, + "loss": 0.6664, + "step": 1384 + }, + { + "epoch": 2.170668058455115, + "grad_norm": 1.1533021926879883, + "learning_rate": 3.6719837541900495e-06, + "loss": 0.6425, + "step": 1385 + }, + { + "epoch": 2.17223382045929, + "grad_norm": 1.246163010597229, + "learning_rate": 3.6701224254044173e-06, + "loss": 0.6683, + "step": 1386 + }, + { + "epoch": 2.1737995824634657, + "grad_norm": 1.1539034843444824, + "learning_rate": 3.6682602656605136e-06, + "loss": 0.6699, + "step": 1387 + }, + { + "epoch": 2.175365344467641, + "grad_norm": 1.021262526512146, + "learning_rate": 3.6663972762807453e-06, + "loss": 0.6409, + "step": 1388 + }, + { + "epoch": 2.1769311064718164, + "grad_norm": 1.09273099899292, + "learning_rate": 3.664533458588107e-06, + "loss": 0.7315, + "step": 1389 + }, + { + "epoch": 2.1784968684759916, + "grad_norm": 1.169747233390808, + "learning_rate": 3.662668813906183e-06, + "loss": 0.6675, + "step": 1390 + }, + { + "epoch": 2.180062630480167, + "grad_norm": 1.189447045326233, + "learning_rate": 3.6608033435591435e-06, + "loss": 0.7016, + "step": 1391 + }, + { + "epoch": 2.1816283924843423, + "grad_norm": 1.0325034856796265, + "learning_rate": 3.658937048871746e-06, + "loss": 0.74, + "step": 1392 + }, + { + "epoch": 2.183194154488518, + "grad_norm": 1.1599889993667603, + "learning_rate": 3.6570699311693316e-06, + "loss": 0.6525, + "step": 1393 + }, + { + "epoch": 2.184759916492693, + "grad_norm": 1.018955945968628, + "learning_rate": 3.6552019917778294e-06, + "loss": 0.6779, + "step": 1394 + }, + { + "epoch": 2.1863256784968685, + "grad_norm": 1.1707937717437744, + "learning_rate": 3.653333232023748e-06, + "loss": 0.6321, + "step": 1395 + }, + { + "epoch": 2.1878914405010437, + "grad_norm": 1.1663832664489746, + "learning_rate": 3.6514636532341825e-06, + "loss": 0.645, + "step": 1396 + }, + { + "epoch": 2.1894572025052192, + "grad_norm": 1.0475002527236938, + "learning_rate": 3.6495932567368054e-06, + "loss": 0.6209, + "step": 1397 + }, + { + "epoch": 2.1910229645093944, + "grad_norm": 1.052640438079834, + "learning_rate": 3.647722043859874e-06, + "loss": 0.7383, + "step": 1398 + }, + { + "epoch": 2.19258872651357, + "grad_norm": 1.1010091304779053, + "learning_rate": 3.6458500159322226e-06, + "loss": 0.669, + "step": 1399 + }, + { + "epoch": 2.1941544885177455, + "grad_norm": 1.0842148065567017, + "learning_rate": 3.6439771742832645e-06, + "loss": 0.7248, + "step": 1400 + }, + { + "epoch": 2.1957202505219207, + "grad_norm": 1.1516163349151611, + "learning_rate": 3.642103520242993e-06, + "loss": 0.6501, + "step": 1401 + }, + { + "epoch": 2.1972860125260962, + "grad_norm": 1.1424976587295532, + "learning_rate": 3.6402290551419754e-06, + "loss": 0.7191, + "step": 1402 + }, + { + "epoch": 2.1988517745302714, + "grad_norm": 1.132794737815857, + "learning_rate": 3.6383537803113577e-06, + "loss": 0.676, + "step": 1403 + }, + { + "epoch": 2.200417536534447, + "grad_norm": 1.0671226978302002, + "learning_rate": 3.6364776970828586e-06, + "loss": 0.6878, + "step": 1404 + }, + { + "epoch": 2.201983298538622, + "grad_norm": 1.0326734781265259, + "learning_rate": 3.6346008067887724e-06, + "loss": 0.6559, + "step": 1405 + }, + { + "epoch": 2.2035490605427976, + "grad_norm": 1.0672800540924072, + "learning_rate": 3.6327231107619664e-06, + "loss": 0.6826, + "step": 1406 + }, + { + "epoch": 2.2051148225469728, + "grad_norm": 1.0678563117980957, + "learning_rate": 3.6308446103358785e-06, + "loss": 0.6772, + "step": 1407 + }, + { + "epoch": 2.2066805845511483, + "grad_norm": 1.1288729906082153, + "learning_rate": 3.628965306844521e-06, + "loss": 0.7248, + "step": 1408 + }, + { + "epoch": 2.2082463465553235, + "grad_norm": 1.0807738304138184, + "learning_rate": 3.6270852016224735e-06, + "loss": 0.6994, + "step": 1409 + }, + { + "epoch": 2.209812108559499, + "grad_norm": 1.2594057321548462, + "learning_rate": 3.625204296004885e-06, + "loss": 0.6429, + "step": 1410 + }, + { + "epoch": 2.211377870563674, + "grad_norm": 1.159574031829834, + "learning_rate": 3.623322591327476e-06, + "loss": 0.6292, + "step": 1411 + }, + { + "epoch": 2.2129436325678498, + "grad_norm": 1.2090522050857544, + "learning_rate": 3.621440088926531e-06, + "loss": 0.663, + "step": 1412 + }, + { + "epoch": 2.214509394572025, + "grad_norm": 1.176761507987976, + "learning_rate": 3.619556790138903e-06, + "loss": 0.5766, + "step": 1413 + }, + { + "epoch": 2.2160751565762005, + "grad_norm": 1.0924803018569946, + "learning_rate": 3.6176726963020085e-06, + "loss": 0.6838, + "step": 1414 + }, + { + "epoch": 2.2176409185803756, + "grad_norm": 1.0821013450622559, + "learning_rate": 3.6157878087538322e-06, + "loss": 0.6464, + "step": 1415 + }, + { + "epoch": 2.219206680584551, + "grad_norm": 1.0731130838394165, + "learning_rate": 3.613902128832919e-06, + "loss": 0.5345, + "step": 1416 + }, + { + "epoch": 2.2207724425887267, + "grad_norm": 1.0433294773101807, + "learning_rate": 3.612015657878378e-06, + "loss": 0.6799, + "step": 1417 + }, + { + "epoch": 2.222338204592902, + "grad_norm": 1.2107945680618286, + "learning_rate": 3.61012839722988e-06, + "loss": 0.746, + "step": 1418 + }, + { + "epoch": 2.223903966597077, + "grad_norm": 1.2256337404251099, + "learning_rate": 3.608240348227656e-06, + "loss": 0.6438, + "step": 1419 + }, + { + "epoch": 2.2254697286012526, + "grad_norm": 1.0490953922271729, + "learning_rate": 3.6063515122124975e-06, + "loss": 0.6618, + "step": 1420 + }, + { + "epoch": 2.227035490605428, + "grad_norm": 1.134993553161621, + "learning_rate": 3.6044618905257546e-06, + "loss": 0.6669, + "step": 1421 + }, + { + "epoch": 2.2286012526096033, + "grad_norm": 1.120219349861145, + "learning_rate": 3.602571484509336e-06, + "loss": 0.7164, + "step": 1422 + }, + { + "epoch": 2.230167014613779, + "grad_norm": 1.2099931240081787, + "learning_rate": 3.6006802955057057e-06, + "loss": 0.6499, + "step": 1423 + }, + { + "epoch": 2.231732776617954, + "grad_norm": 1.276110291481018, + "learning_rate": 3.5987883248578857e-06, + "loss": 0.7182, + "step": 1424 + }, + { + "epoch": 2.2332985386221296, + "grad_norm": 1.12351655960083, + "learning_rate": 3.5968955739094522e-06, + "loss": 0.6672, + "step": 1425 + }, + { + "epoch": 2.2348643006263047, + "grad_norm": 1.1756179332733154, + "learning_rate": 3.595002044004535e-06, + "loss": 0.6417, + "step": 1426 + }, + { + "epoch": 2.2364300626304803, + "grad_norm": 1.1440640687942505, + "learning_rate": 3.593107736487818e-06, + "loss": 0.6694, + "step": 1427 + }, + { + "epoch": 2.2379958246346554, + "grad_norm": 1.160673975944519, + "learning_rate": 3.5912126527045368e-06, + "loss": 0.7067, + "step": 1428 + }, + { + "epoch": 2.239561586638831, + "grad_norm": 1.1037743091583252, + "learning_rate": 3.589316794000478e-06, + "loss": 0.7258, + "step": 1429 + }, + { + "epoch": 2.241127348643006, + "grad_norm": 1.1246981620788574, + "learning_rate": 3.58742016172198e-06, + "loss": 0.6715, + "step": 1430 + }, + { + "epoch": 2.2426931106471817, + "grad_norm": 1.185558795928955, + "learning_rate": 3.5855227572159283e-06, + "loss": 0.6932, + "step": 1431 + }, + { + "epoch": 2.244258872651357, + "grad_norm": 1.0668050050735474, + "learning_rate": 3.583624581829759e-06, + "loss": 0.6836, + "step": 1432 + }, + { + "epoch": 2.2458246346555324, + "grad_norm": 1.1782646179199219, + "learning_rate": 3.581725636911454e-06, + "loss": 0.6787, + "step": 1433 + }, + { + "epoch": 2.2473903966597075, + "grad_norm": 1.2447540760040283, + "learning_rate": 3.579825923809542e-06, + "loss": 0.6702, + "step": 1434 + }, + { + "epoch": 2.248956158663883, + "grad_norm": 1.029355525970459, + "learning_rate": 3.577925443873098e-06, + "loss": 0.637, + "step": 1435 + }, + { + "epoch": 2.2505219206680582, + "grad_norm": 1.1154545545578003, + "learning_rate": 3.5760241984517397e-06, + "loss": 0.7274, + "step": 1436 + }, + { + "epoch": 2.252087682672234, + "grad_norm": 1.116469144821167, + "learning_rate": 3.5741221888956325e-06, + "loss": 0.6321, + "step": 1437 + }, + { + "epoch": 2.2536534446764094, + "grad_norm": 1.2985416650772095, + "learning_rate": 3.5722194165554795e-06, + "loss": 0.648, + "step": 1438 + }, + { + "epoch": 2.2552192066805845, + "grad_norm": 1.2045787572860718, + "learning_rate": 3.5703158827825277e-06, + "loss": 0.6138, + "step": 1439 + }, + { + "epoch": 2.25678496868476, + "grad_norm": 1.0173790454864502, + "learning_rate": 3.5684115889285663e-06, + "loss": 0.6144, + "step": 1440 + }, + { + "epoch": 2.258350730688935, + "grad_norm": 1.1876413822174072, + "learning_rate": 3.5665065363459215e-06, + "loss": 0.5951, + "step": 1441 + }, + { + "epoch": 2.259916492693111, + "grad_norm": 1.0441265106201172, + "learning_rate": 3.564600726387461e-06, + "loss": 0.6681, + "step": 1442 + }, + { + "epoch": 2.261482254697286, + "grad_norm": 1.2096240520477295, + "learning_rate": 3.5626941604065866e-06, + "loss": 0.7213, + "step": 1443 + }, + { + "epoch": 2.2630480167014615, + "grad_norm": 1.0928987264633179, + "learning_rate": 3.560786839757242e-06, + "loss": 0.6129, + "step": 1444 + }, + { + "epoch": 2.2646137787056366, + "grad_norm": 1.169466495513916, + "learning_rate": 3.558878765793902e-06, + "loss": 0.7212, + "step": 1445 + }, + { + "epoch": 2.266179540709812, + "grad_norm": 1.0599427223205566, + "learning_rate": 3.5569699398715794e-06, + "loss": 0.683, + "step": 1446 + }, + { + "epoch": 2.2677453027139873, + "grad_norm": 1.126380443572998, + "learning_rate": 3.5550603633458202e-06, + "loss": 0.6886, + "step": 1447 + }, + { + "epoch": 2.269311064718163, + "grad_norm": 1.2540115118026733, + "learning_rate": 3.553150037572704e-06, + "loss": 0.6515, + "step": 1448 + }, + { + "epoch": 2.270876826722338, + "grad_norm": 1.1289552450180054, + "learning_rate": 3.55123896390884e-06, + "loss": 0.7067, + "step": 1449 + }, + { + "epoch": 2.2724425887265136, + "grad_norm": 1.0645952224731445, + "learning_rate": 3.549327143711372e-06, + "loss": 0.7059, + "step": 1450 + }, + { + "epoch": 2.2740083507306887, + "grad_norm": 0.9427555799484253, + "learning_rate": 3.5474145783379728e-06, + "loss": 0.6472, + "step": 1451 + }, + { + "epoch": 2.2755741127348643, + "grad_norm": 1.1604477167129517, + "learning_rate": 3.5455012691468417e-06, + "loss": 0.6602, + "step": 1452 + }, + { + "epoch": 2.2771398747390394, + "grad_norm": 1.1867820024490356, + "learning_rate": 3.5435872174967102e-06, + "loss": 0.7162, + "step": 1453 + }, + { + "epoch": 2.278705636743215, + "grad_norm": 1.0727450847625732, + "learning_rate": 3.541672424746834e-06, + "loss": 0.7171, + "step": 1454 + }, + { + "epoch": 2.2802713987473906, + "grad_norm": 1.1541563272476196, + "learning_rate": 3.539756892256997e-06, + "loss": 0.7214, + "step": 1455 + }, + { + "epoch": 2.2818371607515657, + "grad_norm": 1.0292634963989258, + "learning_rate": 3.5378406213875084e-06, + "loss": 0.702, + "step": 1456 + }, + { + "epoch": 2.2834029227557413, + "grad_norm": 1.1958640813827515, + "learning_rate": 3.5359236134991996e-06, + "loss": 0.6456, + "step": 1457 + }, + { + "epoch": 2.2849686847599164, + "grad_norm": 1.191214680671692, + "learning_rate": 3.5340058699534287e-06, + "loss": 0.7149, + "step": 1458 + }, + { + "epoch": 2.286534446764092, + "grad_norm": 1.0933841466903687, + "learning_rate": 3.5320873921120735e-06, + "loss": 0.657, + "step": 1459 + }, + { + "epoch": 2.288100208768267, + "grad_norm": 1.0500751733779907, + "learning_rate": 3.5301681813375343e-06, + "loss": 0.6708, + "step": 1460 + }, + { + "epoch": 2.2896659707724427, + "grad_norm": 1.0765804052352905, + "learning_rate": 3.5282482389927313e-06, + "loss": 0.6528, + "step": 1461 + }, + { + "epoch": 2.291231732776618, + "grad_norm": 1.0991063117980957, + "learning_rate": 3.526327566441105e-06, + "loss": 0.7167, + "step": 1462 + }, + { + "epoch": 2.2927974947807934, + "grad_norm": 1.096111536026001, + "learning_rate": 3.5244061650466155e-06, + "loss": 0.6983, + "step": 1463 + }, + { + "epoch": 2.2943632567849686, + "grad_norm": 1.0495761632919312, + "learning_rate": 3.522484036173737e-06, + "loss": 0.6019, + "step": 1464 + }, + { + "epoch": 2.295929018789144, + "grad_norm": 1.204898476600647, + "learning_rate": 3.520561181187464e-06, + "loss": 0.7198, + "step": 1465 + }, + { + "epoch": 2.2974947807933193, + "grad_norm": 1.0273269414901733, + "learning_rate": 3.518637601453304e-06, + "loss": 0.7077, + "step": 1466 + }, + { + "epoch": 2.299060542797495, + "grad_norm": 1.0621248483657837, + "learning_rate": 3.5167132983372808e-06, + "loss": 0.703, + "step": 1467 + }, + { + "epoch": 2.30062630480167, + "grad_norm": 1.216728687286377, + "learning_rate": 3.5147882732059323e-06, + "loss": 0.5898, + "step": 1468 + }, + { + "epoch": 2.3021920668058455, + "grad_norm": 1.065699815750122, + "learning_rate": 3.5128625274263074e-06, + "loss": 0.611, + "step": 1469 + }, + { + "epoch": 2.3037578288100207, + "grad_norm": 1.1529189348220825, + "learning_rate": 3.5109360623659672e-06, + "loss": 0.6778, + "step": 1470 + }, + { + "epoch": 2.3053235908141962, + "grad_norm": 1.1462815999984741, + "learning_rate": 3.5090088793929842e-06, + "loss": 0.6366, + "step": 1471 + }, + { + "epoch": 2.306889352818372, + "grad_norm": 1.0135819911956787, + "learning_rate": 3.507080979875941e-06, + "loss": 0.6393, + "step": 1472 + }, + { + "epoch": 2.308455114822547, + "grad_norm": 1.112345814704895, + "learning_rate": 3.5051523651839288e-06, + "loss": 0.595, + "step": 1473 + }, + { + "epoch": 2.3100208768267225, + "grad_norm": 1.0799155235290527, + "learning_rate": 3.503223036686545e-06, + "loss": 0.7024, + "step": 1474 + }, + { + "epoch": 2.3115866388308977, + "grad_norm": 1.2584905624389648, + "learning_rate": 3.501292995753897e-06, + "loss": 0.6308, + "step": 1475 + }, + { + "epoch": 2.3131524008350732, + "grad_norm": 1.0637624263763428, + "learning_rate": 3.4993622437565955e-06, + "loss": 0.6502, + "step": 1476 + }, + { + "epoch": 2.3147181628392484, + "grad_norm": 1.2071688175201416, + "learning_rate": 3.4974307820657584e-06, + "loss": 0.6389, + "step": 1477 + }, + { + "epoch": 2.316283924843424, + "grad_norm": 1.2328526973724365, + "learning_rate": 3.4954986120530042e-06, + "loss": 0.6748, + "step": 1478 + }, + { + "epoch": 2.317849686847599, + "grad_norm": 1.1051819324493408, + "learning_rate": 3.493565735090458e-06, + "loss": 0.6675, + "step": 1479 + }, + { + "epoch": 2.3194154488517746, + "grad_norm": 1.1707861423492432, + "learning_rate": 3.491632152550746e-06, + "loss": 0.6305, + "step": 1480 + }, + { + "epoch": 2.3209812108559498, + "grad_norm": 1.0759583711624146, + "learning_rate": 3.4896978658069937e-06, + "loss": 0.683, + "step": 1481 + }, + { + "epoch": 2.3225469728601253, + "grad_norm": 1.1699646711349487, + "learning_rate": 3.487762876232828e-06, + "loss": 0.7231, + "step": 1482 + }, + { + "epoch": 2.3241127348643005, + "grad_norm": 1.0097750425338745, + "learning_rate": 3.4858271852023756e-06, + "loss": 0.5897, + "step": 1483 + }, + { + "epoch": 2.325678496868476, + "grad_norm": 1.0517746210098267, + "learning_rate": 3.4838907940902607e-06, + "loss": 0.6386, + "step": 1484 + }, + { + "epoch": 2.327244258872651, + "grad_norm": 1.2001336812973022, + "learning_rate": 3.4819537042716033e-06, + "loss": 0.6854, + "step": 1485 + }, + { + "epoch": 2.3288100208768268, + "grad_norm": 1.1421329975128174, + "learning_rate": 3.480015917122023e-06, + "loss": 0.7186, + "step": 1486 + }, + { + "epoch": 2.330375782881002, + "grad_norm": 1.0638641119003296, + "learning_rate": 3.47807743401763e-06, + "loss": 0.6848, + "step": 1487 + }, + { + "epoch": 2.3319415448851775, + "grad_norm": 1.019140601158142, + "learning_rate": 3.476138256335033e-06, + "loss": 0.6904, + "step": 1488 + }, + { + "epoch": 2.333507306889353, + "grad_norm": 1.2467342615127563, + "learning_rate": 3.4741983854513306e-06, + "loss": 0.753, + "step": 1489 + }, + { + "epoch": 2.335073068893528, + "grad_norm": 1.1455196142196655, + "learning_rate": 3.472257822744116e-06, + "loss": 0.6544, + "step": 1490 + }, + { + "epoch": 2.3366388308977033, + "grad_norm": 1.1508904695510864, + "learning_rate": 3.4703165695914737e-06, + "loss": 0.6216, + "step": 1491 + }, + { + "epoch": 2.338204592901879, + "grad_norm": 1.1695144176483154, + "learning_rate": 3.4683746273719754e-06, + "loss": 0.6478, + "step": 1492 + }, + { + "epoch": 2.3397703549060545, + "grad_norm": 1.041002631187439, + "learning_rate": 3.4664319974646872e-06, + "loss": 0.7142, + "step": 1493 + }, + { + "epoch": 2.3413361169102296, + "grad_norm": 1.0497469902038574, + "learning_rate": 3.4644886812491598e-06, + "loss": 0.7028, + "step": 1494 + }, + { + "epoch": 2.342901878914405, + "grad_norm": 1.0564930438995361, + "learning_rate": 3.4625446801054312e-06, + "loss": 0.6796, + "step": 1495 + }, + { + "epoch": 2.3444676409185803, + "grad_norm": 1.2151904106140137, + "learning_rate": 3.460599995414028e-06, + "loss": 0.6933, + "step": 1496 + }, + { + "epoch": 2.346033402922756, + "grad_norm": 1.197263240814209, + "learning_rate": 3.458654628555961e-06, + "loss": 0.7615, + "step": 1497 + }, + { + "epoch": 2.347599164926931, + "grad_norm": 1.037292718887329, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.7014, + "step": 1498 + }, + { + "epoch": 2.3491649269311066, + "grad_norm": 1.0831308364868164, + "learning_rate": 3.454761853866299e-06, + "loss": 0.6682, + "step": 1499 + }, + { + "epoch": 2.3507306889352817, + "grad_norm": 1.0729385614395142, + "learning_rate": 3.4528144487991448e-06, + "loss": 0.725, + "step": 1500 + }, + { + "epoch": 2.3522964509394573, + "grad_norm": 1.010629415512085, + "learning_rate": 3.4508663670942044e-06, + "loss": 0.6579, + "step": 1501 + }, + { + "epoch": 2.3538622129436324, + "grad_norm": 1.0924932956695557, + "learning_rate": 3.448917610134902e-06, + "loss": 0.6892, + "step": 1502 + }, + { + "epoch": 2.355427974947808, + "grad_norm": 1.1776976585388184, + "learning_rate": 3.4469681793051403e-06, + "loss": 0.6366, + "step": 1503 + }, + { + "epoch": 2.356993736951983, + "grad_norm": 1.062983751296997, + "learning_rate": 3.4450180759892994e-06, + "loss": 0.6518, + "step": 1504 + }, + { + "epoch": 2.3585594989561587, + "grad_norm": 1.1387819051742554, + "learning_rate": 3.4430673015722393e-06, + "loss": 0.623, + "step": 1505 + }, + { + "epoch": 2.3601252609603343, + "grad_norm": 1.035445213317871, + "learning_rate": 3.4411158574392956e-06, + "loss": 0.6876, + "step": 1506 + }, + { + "epoch": 2.3616910229645094, + "grad_norm": 1.0623654127120972, + "learning_rate": 3.439163744976279e-06, + "loss": 0.6795, + "step": 1507 + }, + { + "epoch": 2.3632567849686845, + "grad_norm": 1.0046356916427612, + "learning_rate": 3.437210965569475e-06, + "loss": 0.7043, + "step": 1508 + }, + { + "epoch": 2.36482254697286, + "grad_norm": 1.183174729347229, + "learning_rate": 3.4352575206056426e-06, + "loss": 0.7059, + "step": 1509 + }, + { + "epoch": 2.3663883089770357, + "grad_norm": 1.0844060182571411, + "learning_rate": 3.4333034114720155e-06, + "loss": 0.6811, + "step": 1510 + }, + { + "epoch": 2.367954070981211, + "grad_norm": 1.058577299118042, + "learning_rate": 3.431348639556296e-06, + "loss": 0.7258, + "step": 1511 + }, + { + "epoch": 2.3695198329853864, + "grad_norm": 1.0441113710403442, + "learning_rate": 3.429393206246659e-06, + "loss": 0.6659, + "step": 1512 + }, + { + "epoch": 2.3710855949895615, + "grad_norm": 1.111039400100708, + "learning_rate": 3.4274371129317476e-06, + "loss": 0.6113, + "step": 1513 + }, + { + "epoch": 2.372651356993737, + "grad_norm": 1.1454105377197266, + "learning_rate": 3.425480361000675e-06, + "loss": 0.727, + "step": 1514 + }, + { + "epoch": 2.374217118997912, + "grad_norm": 1.1241672039031982, + "learning_rate": 3.4235229518430212e-06, + "loss": 0.5266, + "step": 1515 + }, + { + "epoch": 2.375782881002088, + "grad_norm": 1.0866069793701172, + "learning_rate": 3.421564886848835e-06, + "loss": 0.6871, + "step": 1516 + }, + { + "epoch": 2.377348643006263, + "grad_norm": 1.211893081665039, + "learning_rate": 3.4196061674086276e-06, + "loss": 0.7064, + "step": 1517 + }, + { + "epoch": 2.3789144050104385, + "grad_norm": 1.1039608716964722, + "learning_rate": 3.417646794913376e-06, + "loss": 0.6821, + "step": 1518 + }, + { + "epoch": 2.3804801670146136, + "grad_norm": 1.0984389781951904, + "learning_rate": 3.4156867707545237e-06, + "loss": 0.6443, + "step": 1519 + }, + { + "epoch": 2.382045929018789, + "grad_norm": 1.11539888381958, + "learning_rate": 3.413726096323975e-06, + "loss": 0.6896, + "step": 1520 + }, + { + "epoch": 2.3836116910229643, + "grad_norm": 1.1566188335418701, + "learning_rate": 3.4117647730140935e-06, + "loss": 0.699, + "step": 1521 + }, + { + "epoch": 2.38517745302714, + "grad_norm": 1.0620112419128418, + "learning_rate": 3.409802802217707e-06, + "loss": 0.6741, + "step": 1522 + }, + { + "epoch": 2.3867432150313155, + "grad_norm": 1.117667317390442, + "learning_rate": 3.4078401853281022e-06, + "loss": 0.7027, + "step": 1523 + }, + { + "epoch": 2.3883089770354906, + "grad_norm": 1.0997360944747925, + "learning_rate": 3.4058769237390254e-06, + "loss": 0.7249, + "step": 1524 + }, + { + "epoch": 2.3898747390396657, + "grad_norm": 1.1101152896881104, + "learning_rate": 3.403913018844678e-06, + "loss": 0.6794, + "step": 1525 + }, + { + "epoch": 2.3914405010438413, + "grad_norm": 1.1798306703567505, + "learning_rate": 3.4019484720397217e-06, + "loss": 0.6219, + "step": 1526 + }, + { + "epoch": 2.393006263048017, + "grad_norm": 1.0730222463607788, + "learning_rate": 3.3999832847192716e-06, + "loss": 0.6622, + "step": 1527 + }, + { + "epoch": 2.394572025052192, + "grad_norm": 1.0943461656570435, + "learning_rate": 3.3980174582788982e-06, + "loss": 0.6999, + "step": 1528 + }, + { + "epoch": 2.3961377870563676, + "grad_norm": 1.1592621803283691, + "learning_rate": 3.396050994114628e-06, + "loss": 0.6419, + "step": 1529 + }, + { + "epoch": 2.3977035490605427, + "grad_norm": 1.1652766466140747, + "learning_rate": 3.3940838936229357e-06, + "loss": 0.6952, + "step": 1530 + }, + { + "epoch": 2.3992693110647183, + "grad_norm": 1.1424206495285034, + "learning_rate": 3.3921161582007524e-06, + "loss": 0.6521, + "step": 1531 + }, + { + "epoch": 2.4008350730688934, + "grad_norm": 1.0935698747634888, + "learning_rate": 3.3901477892454583e-06, + "loss": 0.6707, + "step": 1532 + }, + { + "epoch": 2.402400835073069, + "grad_norm": 1.2164827585220337, + "learning_rate": 3.3881787881548835e-06, + "loss": 0.6485, + "step": 1533 + }, + { + "epoch": 2.403966597077244, + "grad_norm": 1.0536198616027832, + "learning_rate": 3.386209156327307e-06, + "loss": 0.6997, + "step": 1534 + }, + { + "epoch": 2.4055323590814197, + "grad_norm": 1.1083351373672485, + "learning_rate": 3.3842388951614545e-06, + "loss": 0.6264, + "step": 1535 + }, + { + "epoch": 2.407098121085595, + "grad_norm": 1.133180856704712, + "learning_rate": 3.3822680060565026e-06, + "loss": 0.6905, + "step": 1536 + }, + { + "epoch": 2.4086638830897704, + "grad_norm": 1.0531203746795654, + "learning_rate": 3.3802964904120703e-06, + "loss": 0.6492, + "step": 1537 + }, + { + "epoch": 2.4102296450939455, + "grad_norm": 1.0389595031738281, + "learning_rate": 3.3783243496282215e-06, + "loss": 0.6506, + "step": 1538 + }, + { + "epoch": 2.411795407098121, + "grad_norm": 1.1839019060134888, + "learning_rate": 3.376351585105465e-06, + "loss": 0.6933, + "step": 1539 + }, + { + "epoch": 2.4133611691022967, + "grad_norm": 1.0715898275375366, + "learning_rate": 3.3743781982447533e-06, + "loss": 0.7043, + "step": 1540 + }, + { + "epoch": 2.414926931106472, + "grad_norm": 1.1898013353347778, + "learning_rate": 3.3724041904474793e-06, + "loss": 0.7054, + "step": 1541 + }, + { + "epoch": 2.416492693110647, + "grad_norm": 1.026489019393921, + "learning_rate": 3.370429563115478e-06, + "loss": 0.6458, + "step": 1542 + }, + { + "epoch": 2.4180584551148225, + "grad_norm": 1.2772443294525146, + "learning_rate": 3.368454317651024e-06, + "loss": 0.6647, + "step": 1543 + }, + { + "epoch": 2.419624217118998, + "grad_norm": 1.0573022365570068, + "learning_rate": 3.366478455456831e-06, + "loss": 0.6723, + "step": 1544 + }, + { + "epoch": 2.4211899791231732, + "grad_norm": 1.2983055114746094, + "learning_rate": 3.3645019779360493e-06, + "loss": 0.6922, + "step": 1545 + }, + { + "epoch": 2.422755741127349, + "grad_norm": 1.0773214101791382, + "learning_rate": 3.36252488649227e-06, + "loss": 0.6538, + "step": 1546 + }, + { + "epoch": 2.424321503131524, + "grad_norm": 1.1467341184616089, + "learning_rate": 3.3605471825295143e-06, + "loss": 0.7274, + "step": 1547 + }, + { + "epoch": 2.4258872651356995, + "grad_norm": 1.1962897777557373, + "learning_rate": 3.3585688674522438e-06, + "loss": 0.7113, + "step": 1548 + }, + { + "epoch": 2.4274530271398747, + "grad_norm": 0.997977614402771, + "learning_rate": 3.3565899426653515e-06, + "loss": 0.6731, + "step": 1549 + }, + { + "epoch": 2.4290187891440502, + "grad_norm": 1.095198392868042, + "learning_rate": 3.3546104095741634e-06, + "loss": 0.6353, + "step": 1550 + }, + { + "epoch": 2.4305845511482254, + "grad_norm": 1.0401874780654907, + "learning_rate": 3.3526302695844382e-06, + "loss": 0.7459, + "step": 1551 + }, + { + "epoch": 2.432150313152401, + "grad_norm": 1.174526333808899, + "learning_rate": 3.3506495241023646e-06, + "loss": 0.6435, + "step": 1552 + }, + { + "epoch": 2.433716075156576, + "grad_norm": 1.0898199081420898, + "learning_rate": 3.3486681745345627e-06, + "loss": 0.6991, + "step": 1553 + }, + { + "epoch": 2.4352818371607516, + "grad_norm": 1.0757240056991577, + "learning_rate": 3.346686222288081e-06, + "loss": 0.6447, + "step": 1554 + }, + { + "epoch": 2.4368475991649268, + "grad_norm": 1.0817614793777466, + "learning_rate": 3.3447036687703956e-06, + "loss": 0.6546, + "step": 1555 + }, + { + "epoch": 2.4384133611691023, + "grad_norm": 1.181427240371704, + "learning_rate": 3.3427205153894088e-06, + "loss": 0.6592, + "step": 1556 + }, + { + "epoch": 2.4399791231732775, + "grad_norm": 1.0263524055480957, + "learning_rate": 3.340736763553451e-06, + "loss": 0.6766, + "step": 1557 + }, + { + "epoch": 2.441544885177453, + "grad_norm": 1.0632904767990112, + "learning_rate": 3.3387524146712762e-06, + "loss": 0.6368, + "step": 1558 + }, + { + "epoch": 2.443110647181628, + "grad_norm": 1.165933609008789, + "learning_rate": 3.3367674701520627e-06, + "loss": 0.692, + "step": 1559 + }, + { + "epoch": 2.4446764091858038, + "grad_norm": 1.109712839126587, + "learning_rate": 3.3347819314054124e-06, + "loss": 0.6721, + "step": 1560 + }, + { + "epoch": 2.4462421711899793, + "grad_norm": 1.0221387147903442, + "learning_rate": 3.332795799841347e-06, + "loss": 0.6547, + "step": 1561 + }, + { + "epoch": 2.4478079331941545, + "grad_norm": 1.1391754150390625, + "learning_rate": 3.3308090768703118e-06, + "loss": 0.68, + "step": 1562 + }, + { + "epoch": 2.44937369519833, + "grad_norm": 1.0794706344604492, + "learning_rate": 3.328821763903172e-06, + "loss": 0.6531, + "step": 1563 + }, + { + "epoch": 2.450939457202505, + "grad_norm": 1.0358234643936157, + "learning_rate": 3.3268338623512094e-06, + "loss": 0.6638, + "step": 1564 + }, + { + "epoch": 2.4525052192066807, + "grad_norm": 1.1351442337036133, + "learning_rate": 3.3248453736261247e-06, + "loss": 0.6356, + "step": 1565 + }, + { + "epoch": 2.454070981210856, + "grad_norm": 1.010600209236145, + "learning_rate": 3.322856299140038e-06, + "loss": 0.6722, + "step": 1566 + }, + { + "epoch": 2.4556367432150314, + "grad_norm": 1.216776728630066, + "learning_rate": 3.320866640305482e-06, + "loss": 0.658, + "step": 1567 + }, + { + "epoch": 2.4572025052192066, + "grad_norm": 1.0781611204147339, + "learning_rate": 3.3188763985354055e-06, + "loss": 0.711, + "step": 1568 + }, + { + "epoch": 2.458768267223382, + "grad_norm": 1.1265203952789307, + "learning_rate": 3.3168855752431732e-06, + "loss": 0.6746, + "step": 1569 + }, + { + "epoch": 2.4603340292275573, + "grad_norm": 1.1726053953170776, + "learning_rate": 3.3148941718425606e-06, + "loss": 0.6836, + "step": 1570 + }, + { + "epoch": 2.461899791231733, + "grad_norm": 1.1113516092300415, + "learning_rate": 3.3129021897477546e-06, + "loss": 0.6322, + "step": 1571 + }, + { + "epoch": 2.463465553235908, + "grad_norm": 1.0234968662261963, + "learning_rate": 3.3109096303733564e-06, + "loss": 0.7179, + "step": 1572 + }, + { + "epoch": 2.4650313152400836, + "grad_norm": 1.1797139644622803, + "learning_rate": 3.308916495134372e-06, + "loss": 0.6996, + "step": 1573 + }, + { + "epoch": 2.4665970772442587, + "grad_norm": 0.9623079895973206, + "learning_rate": 3.3069227854462216e-06, + "loss": 0.6583, + "step": 1574 + }, + { + "epoch": 2.4681628392484343, + "grad_norm": 1.1546630859375, + "learning_rate": 3.30492850272473e-06, + "loss": 0.7457, + "step": 1575 + }, + { + "epoch": 2.4697286012526094, + "grad_norm": 1.0559042692184448, + "learning_rate": 3.302933648386131e-06, + "loss": 0.692, + "step": 1576 + }, + { + "epoch": 2.471294363256785, + "grad_norm": 1.0812047719955444, + "learning_rate": 3.300938223847062e-06, + "loss": 0.6795, + "step": 1577 + }, + { + "epoch": 2.4728601252609606, + "grad_norm": 1.0247067213058472, + "learning_rate": 3.298942230524567e-06, + "loss": 0.695, + "step": 1578 + }, + { + "epoch": 2.4744258872651357, + "grad_norm": 1.354367971420288, + "learning_rate": 3.296945669836094e-06, + "loss": 0.7014, + "step": 1579 + }, + { + "epoch": 2.4759916492693113, + "grad_norm": 1.0897941589355469, + "learning_rate": 3.2949485431994945e-06, + "loss": 0.6338, + "step": 1580 + }, + { + "epoch": 2.4775574112734864, + "grad_norm": 0.9883112907409668, + "learning_rate": 3.292950852033018e-06, + "loss": 0.7102, + "step": 1581 + }, + { + "epoch": 2.479123173277662, + "grad_norm": 1.1789237260818481, + "learning_rate": 3.290952597755319e-06, + "loss": 0.6802, + "step": 1582 + }, + { + "epoch": 2.480688935281837, + "grad_norm": 1.1877659559249878, + "learning_rate": 3.288953781785452e-06, + "loss": 0.6846, + "step": 1583 + }, + { + "epoch": 2.4822546972860127, + "grad_norm": 1.0334861278533936, + "learning_rate": 3.2869544055428678e-06, + "loss": 0.7138, + "step": 1584 + }, + { + "epoch": 2.483820459290188, + "grad_norm": 1.2335622310638428, + "learning_rate": 3.284954470447416e-06, + "loss": 0.6125, + "step": 1585 + }, + { + "epoch": 2.4853862212943634, + "grad_norm": 1.1405894756317139, + "learning_rate": 3.282953977919345e-06, + "loss": 0.6801, + "step": 1586 + }, + { + "epoch": 2.4869519832985385, + "grad_norm": 1.1099573373794556, + "learning_rate": 3.2809529293792953e-06, + "loss": 0.7071, + "step": 1587 + }, + { + "epoch": 2.488517745302714, + "grad_norm": 1.1282474994659424, + "learning_rate": 3.2789513262483053e-06, + "loss": 0.7082, + "step": 1588 + }, + { + "epoch": 2.490083507306889, + "grad_norm": 1.1693743467330933, + "learning_rate": 3.2769491699478077e-06, + "loss": 0.7137, + "step": 1589 + }, + { + "epoch": 2.491649269311065, + "grad_norm": 1.1324049234390259, + "learning_rate": 3.2749464618996244e-06, + "loss": 0.6402, + "step": 1590 + }, + { + "epoch": 2.49321503131524, + "grad_norm": 1.1645187139511108, + "learning_rate": 3.2729432035259733e-06, + "loss": 0.6941, + "step": 1591 + }, + { + "epoch": 2.4947807933194155, + "grad_norm": 1.1647381782531738, + "learning_rate": 3.2709393962494597e-06, + "loss": 0.6583, + "step": 1592 + }, + { + "epoch": 2.4963465553235906, + "grad_norm": 1.1749935150146484, + "learning_rate": 3.268935041493081e-06, + "loss": 0.7099, + "step": 1593 + }, + { + "epoch": 2.497912317327766, + "grad_norm": 1.1859298944473267, + "learning_rate": 3.266930140680222e-06, + "loss": 0.6551, + "step": 1594 + }, + { + "epoch": 2.4994780793319418, + "grad_norm": 1.1642869710922241, + "learning_rate": 3.264924695234657e-06, + "loss": 0.6712, + "step": 1595 + }, + { + "epoch": 2.501043841336117, + "grad_norm": 1.1522237062454224, + "learning_rate": 3.2629187065805445e-06, + "loss": 0.6773, + "step": 1596 + }, + { + "epoch": 2.502609603340292, + "grad_norm": 1.142119288444519, + "learning_rate": 3.2609121761424312e-06, + "loss": 0.6077, + "step": 1597 + }, + { + "epoch": 2.5041753653444676, + "grad_norm": 1.137640357017517, + "learning_rate": 3.258905105345248e-06, + "loss": 0.7086, + "step": 1598 + }, + { + "epoch": 2.505741127348643, + "grad_norm": 1.1106284856796265, + "learning_rate": 3.2568974956143073e-06, + "loss": 0.6795, + "step": 1599 + }, + { + "epoch": 2.5073068893528183, + "grad_norm": 1.0668182373046875, + "learning_rate": 3.254889348375307e-06, + "loss": 0.7163, + "step": 1600 + }, + { + "epoch": 2.508872651356994, + "grad_norm": 1.2222528457641602, + "learning_rate": 3.2528806650543264e-06, + "loss": 0.6589, + "step": 1601 + }, + { + "epoch": 2.510438413361169, + "grad_norm": 1.0996103286743164, + "learning_rate": 3.250871447077824e-06, + "loss": 0.6097, + "step": 1602 + }, + { + "epoch": 2.5120041753653446, + "grad_norm": 1.2791520357131958, + "learning_rate": 3.248861695872639e-06, + "loss": 0.6367, + "step": 1603 + }, + { + "epoch": 2.5135699373695197, + "grad_norm": 1.1636749505996704, + "learning_rate": 3.2468514128659884e-06, + "loss": 0.6271, + "step": 1604 + }, + { + "epoch": 2.5151356993736953, + "grad_norm": 1.1000317335128784, + "learning_rate": 3.2448405994854683e-06, + "loss": 0.6746, + "step": 1605 + }, + { + "epoch": 2.5167014613778704, + "grad_norm": 1.163297414779663, + "learning_rate": 3.2428292571590514e-06, + "loss": 0.6248, + "step": 1606 + }, + { + "epoch": 2.518267223382046, + "grad_norm": 1.0616698265075684, + "learning_rate": 3.240817387315083e-06, + "loss": 0.6249, + "step": 1607 + }, + { + "epoch": 2.519832985386221, + "grad_norm": 1.2272953987121582, + "learning_rate": 3.2388049913822874e-06, + "loss": 0.6856, + "step": 1608 + }, + { + "epoch": 2.5213987473903967, + "grad_norm": 1.0715241432189941, + "learning_rate": 3.2367920707897594e-06, + "loss": 0.6348, + "step": 1609 + }, + { + "epoch": 2.522964509394572, + "grad_norm": 1.069440245628357, + "learning_rate": 3.234778626966968e-06, + "loss": 0.698, + "step": 1610 + }, + { + "epoch": 2.5245302713987474, + "grad_norm": 1.0824602842330933, + "learning_rate": 3.2327646613437525e-06, + "loss": 0.6809, + "step": 1611 + }, + { + "epoch": 2.526096033402923, + "grad_norm": 1.1153802871704102, + "learning_rate": 3.230750175350324e-06, + "loss": 0.6562, + "step": 1612 + }, + { + "epoch": 2.527661795407098, + "grad_norm": 1.1219229698181152, + "learning_rate": 3.228735170417261e-06, + "loss": 0.6819, + "step": 1613 + }, + { + "epoch": 2.5292275574112733, + "grad_norm": 0.9705930352210999, + "learning_rate": 3.226719647975514e-06, + "loss": 0.7457, + "step": 1614 + }, + { + "epoch": 2.530793319415449, + "grad_norm": 1.0846235752105713, + "learning_rate": 3.2247036094563992e-06, + "loss": 0.6122, + "step": 1615 + }, + { + "epoch": 2.5323590814196244, + "grad_norm": 1.0858566761016846, + "learning_rate": 3.222687056291598e-06, + "loss": 0.6141, + "step": 1616 + }, + { + "epoch": 2.5339248434237995, + "grad_norm": 1.0649619102478027, + "learning_rate": 3.2206699899131576e-06, + "loss": 0.6813, + "step": 1617 + }, + { + "epoch": 2.535490605427975, + "grad_norm": 1.2037075757980347, + "learning_rate": 3.2186524117534935e-06, + "loss": 0.7131, + "step": 1618 + }, + { + "epoch": 2.5370563674321502, + "grad_norm": 1.217882513999939, + "learning_rate": 3.216634323245379e-06, + "loss": 0.6977, + "step": 1619 + }, + { + "epoch": 2.538622129436326, + "grad_norm": 1.1430108547210693, + "learning_rate": 3.2146157258219534e-06, + "loss": 0.6799, + "step": 1620 + }, + { + "epoch": 2.540187891440501, + "grad_norm": 1.1914441585540771, + "learning_rate": 3.2125966209167166e-06, + "loss": 0.7284, + "step": 1621 + }, + { + "epoch": 2.5417536534446765, + "grad_norm": 1.1642671823501587, + "learning_rate": 3.210577009963528e-06, + "loss": 0.7203, + "step": 1622 + }, + { + "epoch": 2.5433194154488517, + "grad_norm": 1.0995980501174927, + "learning_rate": 3.208556894396608e-06, + "loss": 0.6635, + "step": 1623 + }, + { + "epoch": 2.5448851774530272, + "grad_norm": 1.1318360567092896, + "learning_rate": 3.2065362756505345e-06, + "loss": 0.6537, + "step": 1624 + }, + { + "epoch": 2.5464509394572024, + "grad_norm": 1.2233246564865112, + "learning_rate": 3.204515155160242e-06, + "loss": 0.7006, + "step": 1625 + }, + { + "epoch": 2.548016701461378, + "grad_norm": 0.9861915707588196, + "learning_rate": 3.202493534361022e-06, + "loss": 0.6647, + "step": 1626 + }, + { + "epoch": 2.549582463465553, + "grad_norm": 1.2189120054244995, + "learning_rate": 3.2004714146885225e-06, + "loss": 0.6374, + "step": 1627 + }, + { + "epoch": 2.5511482254697286, + "grad_norm": 1.165047526359558, + "learning_rate": 3.1984487975787433e-06, + "loss": 0.6287, + "step": 1628 + }, + { + "epoch": 2.552713987473904, + "grad_norm": 1.1922754049301147, + "learning_rate": 3.196425684468039e-06, + "loss": 0.7103, + "step": 1629 + }, + { + "epoch": 2.5542797494780793, + "grad_norm": 1.1970739364624023, + "learning_rate": 3.194402076793116e-06, + "loss": 0.6951, + "step": 1630 + }, + { + "epoch": 2.5558455114822545, + "grad_norm": 1.1461929082870483, + "learning_rate": 3.1923779759910327e-06, + "loss": 0.6989, + "step": 1631 + }, + { + "epoch": 2.55741127348643, + "grad_norm": 1.1536473035812378, + "learning_rate": 3.190353383499198e-06, + "loss": 0.6433, + "step": 1632 + }, + { + "epoch": 2.5589770354906056, + "grad_norm": 1.3109079599380493, + "learning_rate": 3.188328300755367e-06, + "loss": 0.6873, + "step": 1633 + }, + { + "epoch": 2.5605427974947808, + "grad_norm": 1.1532987356185913, + "learning_rate": 3.186302729197645e-06, + "loss": 0.6818, + "step": 1634 + }, + { + "epoch": 2.562108559498956, + "grad_norm": 1.1802146434783936, + "learning_rate": 3.184276670264486e-06, + "loss": 0.7233, + "step": 1635 + }, + { + "epoch": 2.5636743215031315, + "grad_norm": 1.1282404661178589, + "learning_rate": 3.1822501253946875e-06, + "loss": 0.7203, + "step": 1636 + }, + { + "epoch": 2.565240083507307, + "grad_norm": 1.0870387554168701, + "learning_rate": 3.180223096027394e-06, + "loss": 0.6472, + "step": 1637 + }, + { + "epoch": 2.566805845511482, + "grad_norm": 1.038499355316162, + "learning_rate": 3.178195583602092e-06, + "loss": 0.7197, + "step": 1638 + }, + { + "epoch": 2.5683716075156577, + "grad_norm": 1.0605285167694092, + "learning_rate": 3.176167589558613e-06, + "loss": 0.7222, + "step": 1639 + }, + { + "epoch": 2.569937369519833, + "grad_norm": 1.1737507581710815, + "learning_rate": 3.174139115337129e-06, + "loss": 0.6215, + "step": 1640 + }, + { + "epoch": 2.5715031315240084, + "grad_norm": 1.0304044485092163, + "learning_rate": 3.1721101623781556e-06, + "loss": 0.7113, + "step": 1641 + }, + { + "epoch": 2.5730688935281836, + "grad_norm": 1.1509263515472412, + "learning_rate": 3.170080732122545e-06, + "loss": 0.6521, + "step": 1642 + }, + { + "epoch": 2.574634655532359, + "grad_norm": 1.2465925216674805, + "learning_rate": 3.1680508260114882e-06, + "loss": 0.6921, + "step": 1643 + }, + { + "epoch": 2.5762004175365343, + "grad_norm": 1.0584062337875366, + "learning_rate": 3.1660204454865203e-06, + "loss": 0.7041, + "step": 1644 + }, + { + "epoch": 2.57776617954071, + "grad_norm": 1.1217284202575684, + "learning_rate": 3.163989591989505e-06, + "loss": 0.7177, + "step": 1645 + }, + { + "epoch": 2.5793319415448854, + "grad_norm": 1.1175196170806885, + "learning_rate": 3.1619582669626457e-06, + "loss": 0.7192, + "step": 1646 + }, + { + "epoch": 2.5808977035490606, + "grad_norm": 1.1421573162078857, + "learning_rate": 3.1599264718484835e-06, + "loss": 0.7199, + "step": 1647 + }, + { + "epoch": 2.5824634655532357, + "grad_norm": 1.0428645610809326, + "learning_rate": 3.1578942080898876e-06, + "loss": 0.6496, + "step": 1648 + }, + { + "epoch": 2.5840292275574113, + "grad_norm": 0.9507089257240295, + "learning_rate": 3.1558614771300645e-06, + "loss": 0.6271, + "step": 1649 + }, + { + "epoch": 2.585594989561587, + "grad_norm": 1.2682905197143555, + "learning_rate": 3.1538282804125507e-06, + "loss": 0.6635, + "step": 1650 + }, + { + "epoch": 2.587160751565762, + "grad_norm": 1.098144769668579, + "learning_rate": 3.1517946193812132e-06, + "loss": 0.6631, + "step": 1651 + }, + { + "epoch": 2.588726513569937, + "grad_norm": 1.1256821155548096, + "learning_rate": 3.1497604954802485e-06, + "loss": 0.5353, + "step": 1652 + }, + { + "epoch": 2.5902922755741127, + "grad_norm": 1.1058018207550049, + "learning_rate": 3.1477259101541834e-06, + "loss": 0.6324, + "step": 1653 + }, + { + "epoch": 2.5918580375782883, + "grad_norm": 1.200184941291809, + "learning_rate": 3.145690864847871e-06, + "loss": 0.6484, + "step": 1654 + }, + { + "epoch": 2.5934237995824634, + "grad_norm": 1.0515025854110718, + "learning_rate": 3.1436553610064923e-06, + "loss": 0.6774, + "step": 1655 + }, + { + "epoch": 2.594989561586639, + "grad_norm": 1.0765128135681152, + "learning_rate": 3.1416194000755517e-06, + "loss": 0.6838, + "step": 1656 + }, + { + "epoch": 2.596555323590814, + "grad_norm": 1.157872200012207, + "learning_rate": 3.139582983500881e-06, + "loss": 0.7306, + "step": 1657 + }, + { + "epoch": 2.5981210855949897, + "grad_norm": 1.056947946548462, + "learning_rate": 3.1375461127286334e-06, + "loss": 0.628, + "step": 1658 + }, + { + "epoch": 2.599686847599165, + "grad_norm": 1.1467186212539673, + "learning_rate": 3.135508789205285e-06, + "loss": 0.6829, + "step": 1659 + }, + { + "epoch": 2.6012526096033404, + "grad_norm": 1.0062966346740723, + "learning_rate": 3.1334710143776346e-06, + "loss": 0.654, + "step": 1660 + }, + { + "epoch": 2.6028183716075155, + "grad_norm": 1.1658058166503906, + "learning_rate": 3.1314327896928006e-06, + "loss": 0.7067, + "step": 1661 + }, + { + "epoch": 2.604384133611691, + "grad_norm": 1.2473435401916504, + "learning_rate": 3.12939411659822e-06, + "loss": 0.668, + "step": 1662 + }, + { + "epoch": 2.6059498956158667, + "grad_norm": 1.171295404434204, + "learning_rate": 3.127354996541651e-06, + "loss": 0.6847, + "step": 1663 + }, + { + "epoch": 2.607515657620042, + "grad_norm": 1.118043303489685, + "learning_rate": 3.1253154309711663e-06, + "loss": 0.6948, + "step": 1664 + }, + { + "epoch": 2.609081419624217, + "grad_norm": 1.0427628755569458, + "learning_rate": 3.123275421335157e-06, + "loss": 0.7186, + "step": 1665 + }, + { + "epoch": 2.6106471816283925, + "grad_norm": 1.231054425239563, + "learning_rate": 3.1212349690823273e-06, + "loss": 0.6574, + "step": 1666 + }, + { + "epoch": 2.612212943632568, + "grad_norm": 1.0482161045074463, + "learning_rate": 3.1191940756616983e-06, + "loss": 0.7376, + "step": 1667 + }, + { + "epoch": 2.613778705636743, + "grad_norm": 1.2325738668441772, + "learning_rate": 3.1171527425226027e-06, + "loss": 0.6741, + "step": 1668 + }, + { + "epoch": 2.6153444676409183, + "grad_norm": 1.183379888534546, + "learning_rate": 3.1151109711146854e-06, + "loss": 0.7424, + "step": 1669 + }, + { + "epoch": 2.616910229645094, + "grad_norm": 1.1731956005096436, + "learning_rate": 3.1130687628879047e-06, + "loss": 0.7913, + "step": 1670 + }, + { + "epoch": 2.6184759916492695, + "grad_norm": 1.155126690864563, + "learning_rate": 3.111026119292526e-06, + "loss": 0.6724, + "step": 1671 + }, + { + "epoch": 2.6200417536534446, + "grad_norm": 1.0947262048721313, + "learning_rate": 3.1089830417791255e-06, + "loss": 0.6959, + "step": 1672 + }, + { + "epoch": 2.62160751565762, + "grad_norm": 1.073900580406189, + "learning_rate": 3.1069395317985884e-06, + "loss": 0.7274, + "step": 1673 + }, + { + "epoch": 2.6231732776617953, + "grad_norm": 1.1424334049224854, + "learning_rate": 3.104895590802105e-06, + "loss": 0.657, + "step": 1674 + }, + { + "epoch": 2.624739039665971, + "grad_norm": 1.100588321685791, + "learning_rate": 3.1028512202411736e-06, + "loss": 0.6451, + "step": 1675 + }, + { + "epoch": 2.626304801670146, + "grad_norm": 1.0387758016586304, + "learning_rate": 3.100806421567596e-06, + "loss": 0.7331, + "step": 1676 + }, + { + "epoch": 2.6278705636743216, + "grad_norm": 1.2019600868225098, + "learning_rate": 3.0987611962334784e-06, + "loss": 0.6224, + "step": 1677 + }, + { + "epoch": 2.6294363256784967, + "grad_norm": 1.0992989540100098, + "learning_rate": 3.0967155456912313e-06, + "loss": 0.6557, + "step": 1678 + }, + { + "epoch": 2.6310020876826723, + "grad_norm": 1.259507417678833, + "learning_rate": 3.0946694713935655e-06, + "loss": 0.6668, + "step": 1679 + }, + { + "epoch": 2.632567849686848, + "grad_norm": 1.1520506143569946, + "learning_rate": 3.0926229747934932e-06, + "loss": 0.7199, + "step": 1680 + }, + { + "epoch": 2.634133611691023, + "grad_norm": 1.0905909538269043, + "learning_rate": 3.0905760573443278e-06, + "loss": 0.6927, + "step": 1681 + }, + { + "epoch": 2.635699373695198, + "grad_norm": 1.1551228761672974, + "learning_rate": 3.0885287204996784e-06, + "loss": 0.6872, + "step": 1682 + }, + { + "epoch": 2.6372651356993737, + "grad_norm": 1.177915096282959, + "learning_rate": 3.0864809657134553e-06, + "loss": 0.6102, + "step": 1683 + }, + { + "epoch": 2.6388308977035493, + "grad_norm": 1.2892553806304932, + "learning_rate": 3.084432794439865e-06, + "loss": 0.6576, + "step": 1684 + }, + { + "epoch": 2.6403966597077244, + "grad_norm": 1.1799864768981934, + "learning_rate": 3.0823842081334075e-06, + "loss": 0.7214, + "step": 1685 + }, + { + "epoch": 2.6419624217118995, + "grad_norm": 1.11050283908844, + "learning_rate": 3.08033520824888e-06, + "loss": 0.6156, + "step": 1686 + }, + { + "epoch": 2.643528183716075, + "grad_norm": 1.15585196018219, + "learning_rate": 3.078285796241372e-06, + "loss": 0.6311, + "step": 1687 + }, + { + "epoch": 2.6450939457202507, + "grad_norm": 1.1588464975357056, + "learning_rate": 3.0762359735662663e-06, + "loss": 0.6804, + "step": 1688 + }, + { + "epoch": 2.646659707724426, + "grad_norm": 1.1857856512069702, + "learning_rate": 3.0741857416792374e-06, + "loss": 0.6737, + "step": 1689 + }, + { + "epoch": 2.6482254697286014, + "grad_norm": 1.1420296430587769, + "learning_rate": 3.0721351020362495e-06, + "loss": 0.6241, + "step": 1690 + }, + { + "epoch": 2.6497912317327765, + "grad_norm": 1.2236177921295166, + "learning_rate": 3.070084056093558e-06, + "loss": 0.7142, + "step": 1691 + }, + { + "epoch": 2.651356993736952, + "grad_norm": 1.1257219314575195, + "learning_rate": 3.0680326053077047e-06, + "loss": 0.5786, + "step": 1692 + }, + { + "epoch": 2.6529227557411272, + "grad_norm": 1.1258952617645264, + "learning_rate": 3.065980751135522e-06, + "loss": 0.7306, + "step": 1693 + }, + { + "epoch": 2.654488517745303, + "grad_norm": 1.1650607585906982, + "learning_rate": 3.0639284950341246e-06, + "loss": 0.6776, + "step": 1694 + }, + { + "epoch": 2.656054279749478, + "grad_norm": 1.1687644720077515, + "learning_rate": 3.0618758384609156e-06, + "loss": 0.6602, + "step": 1695 + }, + { + "epoch": 2.6576200417536535, + "grad_norm": 1.160516381263733, + "learning_rate": 3.0598227828735825e-06, + "loss": 0.6676, + "step": 1696 + }, + { + "epoch": 2.6591858037578286, + "grad_norm": 1.0864505767822266, + "learning_rate": 3.0577693297300947e-06, + "loss": 0.6843, + "step": 1697 + }, + { + "epoch": 2.6607515657620042, + "grad_norm": 1.0814275741577148, + "learning_rate": 3.055715480488706e-06, + "loss": 0.7214, + "step": 1698 + }, + { + "epoch": 2.6623173277661794, + "grad_norm": 1.1549509763717651, + "learning_rate": 3.0536612366079465e-06, + "loss": 0.7756, + "step": 1699 + }, + { + "epoch": 2.663883089770355, + "grad_norm": 1.125248908996582, + "learning_rate": 3.0516065995466336e-06, + "loss": 0.7133, + "step": 1700 + }, + { + "epoch": 2.6654488517745305, + "grad_norm": 1.0562750101089478, + "learning_rate": 3.0495515707638605e-06, + "loss": 0.6655, + "step": 1701 + }, + { + "epoch": 2.6670146137787056, + "grad_norm": 1.1991479396820068, + "learning_rate": 3.0474961517189965e-06, + "loss": 0.7382, + "step": 1702 + }, + { + "epoch": 2.6685803757828808, + "grad_norm": 1.0789530277252197, + "learning_rate": 3.0454403438716918e-06, + "loss": 0.6608, + "step": 1703 + }, + { + "epoch": 2.6701461377870563, + "grad_norm": 1.1776909828186035, + "learning_rate": 3.04338414868187e-06, + "loss": 0.7493, + "step": 1704 + }, + { + "epoch": 2.671711899791232, + "grad_norm": 1.1587637662887573, + "learning_rate": 3.041327567609731e-06, + "loss": 0.7042, + "step": 1705 + }, + { + "epoch": 2.673277661795407, + "grad_norm": 1.151100516319275, + "learning_rate": 3.0392706021157485e-06, + "loss": 0.5642, + "step": 1706 + }, + { + "epoch": 2.6748434237995826, + "grad_norm": 1.098358392715454, + "learning_rate": 3.03721325366067e-06, + "loss": 0.6177, + "step": 1707 + }, + { + "epoch": 2.6764091858037578, + "grad_norm": 1.1176345348358154, + "learning_rate": 3.0351555237055135e-06, + "loss": 0.7474, + "step": 1708 + }, + { + "epoch": 2.6779749478079333, + "grad_norm": 1.1730945110321045, + "learning_rate": 3.0330974137115683e-06, + "loss": 0.5899, + "step": 1709 + }, + { + "epoch": 2.6795407098121085, + "grad_norm": 1.1680086851119995, + "learning_rate": 3.031038925140395e-06, + "loss": 0.6576, + "step": 1710 + }, + { + "epoch": 2.681106471816284, + "grad_norm": 1.1488760709762573, + "learning_rate": 3.028980059453821e-06, + "loss": 0.6541, + "step": 1711 + }, + { + "epoch": 2.682672233820459, + "grad_norm": 1.1394031047821045, + "learning_rate": 3.0269208181139413e-06, + "loss": 0.6725, + "step": 1712 + }, + { + "epoch": 2.6842379958246347, + "grad_norm": 1.105613112449646, + "learning_rate": 3.024861202583122e-06, + "loss": 0.7088, + "step": 1713 + }, + { + "epoch": 2.68580375782881, + "grad_norm": 1.1166354417800903, + "learning_rate": 3.022801214323989e-06, + "loss": 0.7273, + "step": 1714 + }, + { + "epoch": 2.6873695198329854, + "grad_norm": 1.05818772315979, + "learning_rate": 3.020740854799436e-06, + "loss": 0.591, + "step": 1715 + }, + { + "epoch": 2.6889352818371606, + "grad_norm": 1.019797682762146, + "learning_rate": 3.0186801254726213e-06, + "loss": 0.6676, + "step": 1716 + }, + { + "epoch": 2.690501043841336, + "grad_norm": 1.1530799865722656, + "learning_rate": 3.016619027806963e-06, + "loss": 0.6181, + "step": 1717 + }, + { + "epoch": 2.6920668058455117, + "grad_norm": 1.132246971130371, + "learning_rate": 3.0145575632661426e-06, + "loss": 0.6932, + "step": 1718 + }, + { + "epoch": 2.693632567849687, + "grad_norm": 1.2640838623046875, + "learning_rate": 3.0124957333141036e-06, + "loss": 0.7029, + "step": 1719 + }, + { + "epoch": 2.695198329853862, + "grad_norm": 1.1203837394714355, + "learning_rate": 3.0104335394150447e-06, + "loss": 0.6522, + "step": 1720 + }, + { + "epoch": 2.6967640918580376, + "grad_norm": 1.1490654945373535, + "learning_rate": 3.008370983033426e-06, + "loss": 0.6837, + "step": 1721 + }, + { + "epoch": 2.698329853862213, + "grad_norm": 1.0226919651031494, + "learning_rate": 3.006308065633966e-06, + "loss": 0.6344, + "step": 1722 + }, + { + "epoch": 2.6998956158663883, + "grad_norm": 1.2632393836975098, + "learning_rate": 3.004244788681637e-06, + "loss": 0.7157, + "step": 1723 + }, + { + "epoch": 2.701461377870564, + "grad_norm": 1.0831620693206787, + "learning_rate": 3.0021811536416676e-06, + "loss": 0.6671, + "step": 1724 + }, + { + "epoch": 2.703027139874739, + "grad_norm": 1.1284756660461426, + "learning_rate": 3.0001171619795417e-06, + "loss": 0.6669, + "step": 1725 + }, + { + "epoch": 2.7045929018789145, + "grad_norm": 1.0045720338821411, + "learning_rate": 2.9980528151609945e-06, + "loss": 0.7276, + "step": 1726 + }, + { + "epoch": 2.7061586638830897, + "grad_norm": 1.1337859630584717, + "learning_rate": 2.9959881146520164e-06, + "loss": 0.6734, + "step": 1727 + }, + { + "epoch": 2.7077244258872653, + "grad_norm": 1.0121667385101318, + "learning_rate": 2.993923061918845e-06, + "loss": 0.6201, + "step": 1728 + }, + { + "epoch": 2.7092901878914404, + "grad_norm": 1.2311614751815796, + "learning_rate": 2.991857658427972e-06, + "loss": 0.6392, + "step": 1729 + }, + { + "epoch": 2.710855949895616, + "grad_norm": 1.0462316274642944, + "learning_rate": 2.989791905646135e-06, + "loss": 0.6736, + "step": 1730 + }, + { + "epoch": 2.712421711899791, + "grad_norm": 1.1915360689163208, + "learning_rate": 2.9877258050403214e-06, + "loss": 0.5977, + "step": 1731 + }, + { + "epoch": 2.7139874739039667, + "grad_norm": 1.0216567516326904, + "learning_rate": 2.985659358077765e-06, + "loss": 0.6393, + "step": 1732 + }, + { + "epoch": 2.715553235908142, + "grad_norm": 1.114071011543274, + "learning_rate": 2.9835925662259457e-06, + "loss": 0.6854, + "step": 1733 + }, + { + "epoch": 2.7171189979123174, + "grad_norm": 1.2007335424423218, + "learning_rate": 2.981525430952589e-06, + "loss": 0.584, + "step": 1734 + }, + { + "epoch": 2.718684759916493, + "grad_norm": 1.221422791481018, + "learning_rate": 2.979457953725663e-06, + "loss": 0.742, + "step": 1735 + }, + { + "epoch": 2.720250521920668, + "grad_norm": 1.0720511674880981, + "learning_rate": 2.9773901360133804e-06, + "loss": 0.6722, + "step": 1736 + }, + { + "epoch": 2.721816283924843, + "grad_norm": 1.1078705787658691, + "learning_rate": 2.975321979284193e-06, + "loss": 0.6837, + "step": 1737 + }, + { + "epoch": 2.723382045929019, + "grad_norm": 1.1647889614105225, + "learning_rate": 2.973253485006795e-06, + "loss": 0.7406, + "step": 1738 + }, + { + "epoch": 2.7249478079331944, + "grad_norm": 1.1482031345367432, + "learning_rate": 2.971184654650122e-06, + "loss": 0.7109, + "step": 1739 + }, + { + "epoch": 2.7265135699373695, + "grad_norm": 1.095131278038025, + "learning_rate": 2.9691154896833454e-06, + "loss": 0.6795, + "step": 1740 + }, + { + "epoch": 2.7280793319415446, + "grad_norm": 1.0917677879333496, + "learning_rate": 2.9670459915758746e-06, + "loss": 0.7142, + "step": 1741 + }, + { + "epoch": 2.72964509394572, + "grad_norm": 1.0911260843276978, + "learning_rate": 2.9649761617973572e-06, + "loss": 0.6917, + "step": 1742 + }, + { + "epoch": 2.7312108559498958, + "grad_norm": 1.2418321371078491, + "learning_rate": 2.9629060018176757e-06, + "loss": 0.6562, + "step": 1743 + }, + { + "epoch": 2.732776617954071, + "grad_norm": 1.114390254020691, + "learning_rate": 2.9608355131069464e-06, + "loss": 0.6812, + "step": 1744 + }, + { + "epoch": 2.7343423799582465, + "grad_norm": 1.0684012174606323, + "learning_rate": 2.9587646971355194e-06, + "loss": 0.688, + "step": 1745 + }, + { + "epoch": 2.7359081419624216, + "grad_norm": 1.1974529027938843, + "learning_rate": 2.9566935553739773e-06, + "loss": 0.6697, + "step": 1746 + }, + { + "epoch": 2.737473903966597, + "grad_norm": 1.0794286727905273, + "learning_rate": 2.9546220892931336e-06, + "loss": 0.6587, + "step": 1747 + }, + { + "epoch": 2.7390396659707723, + "grad_norm": 1.1143841743469238, + "learning_rate": 2.9525503003640336e-06, + "loss": 0.6814, + "step": 1748 + }, + { + "epoch": 2.740605427974948, + "grad_norm": 1.1970027685165405, + "learning_rate": 2.9504781900579504e-06, + "loss": 0.6714, + "step": 1749 + }, + { + "epoch": 2.742171189979123, + "grad_norm": 1.0299553871154785, + "learning_rate": 2.948405759846386e-06, + "loss": 0.6707, + "step": 1750 + }, + { + "epoch": 2.7437369519832986, + "grad_norm": 1.103028416633606, + "learning_rate": 2.946333011201068e-06, + "loss": 0.6163, + "step": 1751 + }, + { + "epoch": 2.745302713987474, + "grad_norm": 1.1230628490447998, + "learning_rate": 2.944259945593953e-06, + "loss": 0.5917, + "step": 1752 + }, + { + "epoch": 2.7468684759916493, + "grad_norm": 1.1009881496429443, + "learning_rate": 2.942186564497221e-06, + "loss": 0.7048, + "step": 1753 + }, + { + "epoch": 2.7484342379958244, + "grad_norm": 1.1123842000961304, + "learning_rate": 2.9401128693832752e-06, + "loss": 0.6063, + "step": 1754 + }, + { + "epoch": 2.75, + "grad_norm": 1.0923182964324951, + "learning_rate": 2.9380388617247433e-06, + "loss": 0.663, + "step": 1755 + }, + { + "epoch": 2.7515657620041756, + "grad_norm": 1.1761754751205444, + "learning_rate": 2.935964542994475e-06, + "loss": 0.6494, + "step": 1756 + }, + { + "epoch": 2.7531315240083507, + "grad_norm": 1.0580326318740845, + "learning_rate": 2.9338899146655396e-06, + "loss": 0.7391, + "step": 1757 + }, + { + "epoch": 2.754697286012526, + "grad_norm": 1.1655254364013672, + "learning_rate": 2.931814978211227e-06, + "loss": 0.6595, + "step": 1758 + }, + { + "epoch": 2.7562630480167014, + "grad_norm": 1.1162359714508057, + "learning_rate": 2.929739735105047e-06, + "loss": 0.6438, + "step": 1759 + }, + { + "epoch": 2.757828810020877, + "grad_norm": 1.173933744430542, + "learning_rate": 2.9276641868207244e-06, + "loss": 0.5591, + "step": 1760 + }, + { + "epoch": 2.759394572025052, + "grad_norm": 1.2183250188827515, + "learning_rate": 2.9255883348322038e-06, + "loss": 0.7317, + "step": 1761 + }, + { + "epoch": 2.7609603340292277, + "grad_norm": 1.0836644172668457, + "learning_rate": 2.9235121806136446e-06, + "loss": 0.7539, + "step": 1762 + }, + { + "epoch": 2.762526096033403, + "grad_norm": 1.1108825206756592, + "learning_rate": 2.921435725639419e-06, + "loss": 0.6598, + "step": 1763 + }, + { + "epoch": 2.7640918580375784, + "grad_norm": 1.090196967124939, + "learning_rate": 2.9193589713841132e-06, + "loss": 0.7412, + "step": 1764 + }, + { + "epoch": 2.7656576200417535, + "grad_norm": 1.1506152153015137, + "learning_rate": 2.9172819193225296e-06, + "loss": 0.6857, + "step": 1765 + }, + { + "epoch": 2.767223382045929, + "grad_norm": 1.127022385597229, + "learning_rate": 2.9152045709296766e-06, + "loss": 0.6383, + "step": 1766 + }, + { + "epoch": 2.7687891440501042, + "grad_norm": 1.1096911430358887, + "learning_rate": 2.9131269276807782e-06, + "loss": 0.6363, + "step": 1767 + }, + { + "epoch": 2.77035490605428, + "grad_norm": 1.1879534721374512, + "learning_rate": 2.911048991051263e-06, + "loss": 0.7313, + "step": 1768 + }, + { + "epoch": 2.7719206680584554, + "grad_norm": 1.0864057540893555, + "learning_rate": 2.908970762516772e-06, + "loss": 0.6602, + "step": 1769 + }, + { + "epoch": 2.7734864300626305, + "grad_norm": 1.174880027770996, + "learning_rate": 2.9068922435531515e-06, + "loss": 0.6731, + "step": 1770 + }, + { + "epoch": 2.7750521920668056, + "grad_norm": 1.1670523881912231, + "learning_rate": 2.904813435636453e-06, + "loss": 0.5982, + "step": 1771 + }, + { + "epoch": 2.776617954070981, + "grad_norm": 1.016558289527893, + "learning_rate": 2.902734340242937e-06, + "loss": 0.5549, + "step": 1772 + }, + { + "epoch": 2.778183716075157, + "grad_norm": 1.2644622325897217, + "learning_rate": 2.9006549588490636e-06, + "loss": 0.6209, + "step": 1773 + }, + { + "epoch": 2.779749478079332, + "grad_norm": 1.1689661741256714, + "learning_rate": 2.8985752929314985e-06, + "loss": 0.6629, + "step": 1774 + }, + { + "epoch": 2.781315240083507, + "grad_norm": 1.123618721961975, + "learning_rate": 2.8964953439671095e-06, + "loss": 0.6986, + "step": 1775 + }, + { + "epoch": 2.7828810020876826, + "grad_norm": 1.1276949644088745, + "learning_rate": 2.8944151134329646e-06, + "loss": 0.6125, + "step": 1776 + }, + { + "epoch": 2.784446764091858, + "grad_norm": 1.1804574728012085, + "learning_rate": 2.8923346028063328e-06, + "loss": 0.5912, + "step": 1777 + }, + { + "epoch": 2.7860125260960333, + "grad_norm": 1.1222927570343018, + "learning_rate": 2.89025381356468e-06, + "loss": 0.5837, + "step": 1778 + }, + { + "epoch": 2.787578288100209, + "grad_norm": 1.063511848449707, + "learning_rate": 2.888172747185673e-06, + "loss": 0.7243, + "step": 1779 + }, + { + "epoch": 2.789144050104384, + "grad_norm": 1.1183874607086182, + "learning_rate": 2.8860914051471722e-06, + "loss": 0.7204, + "step": 1780 + }, + { + "epoch": 2.7907098121085596, + "grad_norm": 1.135906457901001, + "learning_rate": 2.884009788927235e-06, + "loss": 0.7308, + "step": 1781 + }, + { + "epoch": 2.7922755741127347, + "grad_norm": 1.2013477087020874, + "learning_rate": 2.8819279000041156e-06, + "loss": 0.7004, + "step": 1782 + }, + { + "epoch": 2.7938413361169103, + "grad_norm": 1.074418306350708, + "learning_rate": 2.8798457398562584e-06, + "loss": 0.6404, + "step": 1783 + }, + { + "epoch": 2.7954070981210855, + "grad_norm": 1.0827600955963135, + "learning_rate": 2.8777633099623027e-06, + "loss": 0.7488, + "step": 1784 + }, + { + "epoch": 2.796972860125261, + "grad_norm": 1.1087040901184082, + "learning_rate": 2.8756806118010784e-06, + "loss": 0.6554, + "step": 1785 + }, + { + "epoch": 2.798538622129436, + "grad_norm": 1.0850439071655273, + "learning_rate": 2.8735976468516057e-06, + "loss": 0.6501, + "step": 1786 + }, + { + "epoch": 2.8001043841336117, + "grad_norm": 1.129045844078064, + "learning_rate": 2.871514416593096e-06, + "loss": 0.6697, + "step": 1787 + }, + { + "epoch": 2.801670146137787, + "grad_norm": 1.0854123830795288, + "learning_rate": 2.869430922504947e-06, + "loss": 0.6787, + "step": 1788 + }, + { + "epoch": 2.8032359081419624, + "grad_norm": 1.2950549125671387, + "learning_rate": 2.8673471660667448e-06, + "loss": 0.6682, + "step": 1789 + }, + { + "epoch": 2.804801670146138, + "grad_norm": 1.0760269165039062, + "learning_rate": 2.865263148758261e-06, + "loss": 0.6846, + "step": 1790 + }, + { + "epoch": 2.806367432150313, + "grad_norm": 1.1861540079116821, + "learning_rate": 2.863178872059454e-06, + "loss": 0.6029, + "step": 1791 + }, + { + "epoch": 2.8079331941544883, + "grad_norm": 1.081518292427063, + "learning_rate": 2.8610943374504645e-06, + "loss": 0.6801, + "step": 1792 + }, + { + "epoch": 2.809498956158664, + "grad_norm": 1.1814477443695068, + "learning_rate": 2.8590095464116192e-06, + "loss": 0.68, + "step": 1793 + }, + { + "epoch": 2.8110647181628394, + "grad_norm": 1.154828667640686, + "learning_rate": 2.856924500423422e-06, + "loss": 0.6164, + "step": 1794 + }, + { + "epoch": 2.8126304801670146, + "grad_norm": 1.0605709552764893, + "learning_rate": 2.8548392009665638e-06, + "loss": 0.6146, + "step": 1795 + }, + { + "epoch": 2.81419624217119, + "grad_norm": 1.1769119501113892, + "learning_rate": 2.852753649521911e-06, + "loss": 0.6789, + "step": 1796 + }, + { + "epoch": 2.8157620041753653, + "grad_norm": 1.006847858428955, + "learning_rate": 2.8506678475705114e-06, + "loss": 0.6719, + "step": 1797 + }, + { + "epoch": 2.817327766179541, + "grad_norm": 1.1305062770843506, + "learning_rate": 2.8485817965935886e-06, + "loss": 0.6548, + "step": 1798 + }, + { + "epoch": 2.818893528183716, + "grad_norm": 1.0815331935882568, + "learning_rate": 2.846495498072545e-06, + "loss": 0.7306, + "step": 1799 + }, + { + "epoch": 2.8204592901878915, + "grad_norm": 1.1019850969314575, + "learning_rate": 2.8444089534889587e-06, + "loss": 0.7188, + "step": 1800 + }, + { + "epoch": 2.8220250521920667, + "grad_norm": 1.1405457258224487, + "learning_rate": 2.8423221643245807e-06, + "loss": 0.6482, + "step": 1801 + }, + { + "epoch": 2.8235908141962422, + "grad_norm": 1.1295759677886963, + "learning_rate": 2.8402351320613374e-06, + "loss": 0.5882, + "step": 1802 + }, + { + "epoch": 2.8251565762004174, + "grad_norm": 1.1165974140167236, + "learning_rate": 2.8381478581813275e-06, + "loss": 0.6714, + "step": 1803 + }, + { + "epoch": 2.826722338204593, + "grad_norm": 1.0910004377365112, + "learning_rate": 2.836060344166821e-06, + "loss": 0.6971, + "step": 1804 + }, + { + "epoch": 2.828288100208768, + "grad_norm": 1.093596339225769, + "learning_rate": 2.833972591500259e-06, + "loss": 0.7041, + "step": 1805 + }, + { + "epoch": 2.8298538622129437, + "grad_norm": 1.1507813930511475, + "learning_rate": 2.8318846016642507e-06, + "loss": 0.6455, + "step": 1806 + }, + { + "epoch": 2.8314196242171192, + "grad_norm": 1.055422306060791, + "learning_rate": 2.8297963761415748e-06, + "loss": 0.7499, + "step": 1807 + }, + { + "epoch": 2.8329853862212944, + "grad_norm": 1.0835453271865845, + "learning_rate": 2.8277079164151782e-06, + "loss": 0.6739, + "step": 1808 + }, + { + "epoch": 2.8345511482254695, + "grad_norm": 1.247808575630188, + "learning_rate": 2.8256192239681723e-06, + "loss": 0.6774, + "step": 1809 + }, + { + "epoch": 2.836116910229645, + "grad_norm": 1.0836130380630493, + "learning_rate": 2.8235303002838343e-06, + "loss": 0.6592, + "step": 1810 + }, + { + "epoch": 2.8376826722338206, + "grad_norm": 1.199852466583252, + "learning_rate": 2.8214411468456067e-06, + "loss": 0.6728, + "step": 1811 + }, + { + "epoch": 2.8392484342379958, + "grad_norm": 1.0706546306610107, + "learning_rate": 2.8193517651370934e-06, + "loss": 0.7438, + "step": 1812 + }, + { + "epoch": 2.8408141962421714, + "grad_norm": 1.301465630531311, + "learning_rate": 2.8172621566420633e-06, + "loss": 0.6736, + "step": 1813 + }, + { + "epoch": 2.8423799582463465, + "grad_norm": 1.0916422605514526, + "learning_rate": 2.815172322844442e-06, + "loss": 0.7258, + "step": 1814 + }, + { + "epoch": 2.843945720250522, + "grad_norm": 1.1184017658233643, + "learning_rate": 2.8130822652283183e-06, + "loss": 0.6831, + "step": 1815 + }, + { + "epoch": 2.845511482254697, + "grad_norm": 1.1714346408843994, + "learning_rate": 2.8109919852779406e-06, + "loss": 0.6866, + "step": 1816 + }, + { + "epoch": 2.8470772442588728, + "grad_norm": 1.0131067037582397, + "learning_rate": 2.808901484477712e-06, + "loss": 0.6574, + "step": 1817 + }, + { + "epoch": 2.848643006263048, + "grad_norm": 1.2358931303024292, + "learning_rate": 2.8068107643121952e-06, + "loss": 0.5596, + "step": 1818 + }, + { + "epoch": 2.8502087682672235, + "grad_norm": 0.9713279604911804, + "learning_rate": 2.8047198262661074e-06, + "loss": 0.6502, + "step": 1819 + }, + { + "epoch": 2.8517745302713986, + "grad_norm": 1.0890235900878906, + "learning_rate": 2.80262867182432e-06, + "loss": 0.6566, + "step": 1820 + }, + { + "epoch": 2.853340292275574, + "grad_norm": 1.084876298904419, + "learning_rate": 2.8005373024718597e-06, + "loss": 0.6356, + "step": 1821 + }, + { + "epoch": 2.8549060542797493, + "grad_norm": 1.1338250637054443, + "learning_rate": 2.7984457196939056e-06, + "loss": 0.7226, + "step": 1822 + }, + { + "epoch": 2.856471816283925, + "grad_norm": 1.1295956373214722, + "learning_rate": 2.7963539249757872e-06, + "loss": 0.724, + "step": 1823 + }, + { + "epoch": 2.8580375782881005, + "grad_norm": 1.0954629182815552, + "learning_rate": 2.794261919802984e-06, + "loss": 0.685, + "step": 1824 + }, + { + "epoch": 2.8596033402922756, + "grad_norm": 1.0435383319854736, + "learning_rate": 2.792169705661128e-06, + "loss": 0.7532, + "step": 1825 + }, + { + "epoch": 2.8611691022964507, + "grad_norm": 1.1580495834350586, + "learning_rate": 2.790077284035996e-06, + "loss": 0.6745, + "step": 1826 + }, + { + "epoch": 2.8627348643006263, + "grad_norm": 1.0154657363891602, + "learning_rate": 2.787984656413514e-06, + "loss": 0.7312, + "step": 1827 + }, + { + "epoch": 2.864300626304802, + "grad_norm": 1.1673710346221924, + "learning_rate": 2.785891824279755e-06, + "loss": 0.6769, + "step": 1828 + }, + { + "epoch": 2.865866388308977, + "grad_norm": 1.1436103582382202, + "learning_rate": 2.7837987891209352e-06, + "loss": 0.6987, + "step": 1829 + }, + { + "epoch": 2.867432150313152, + "grad_norm": 1.2176343202590942, + "learning_rate": 2.7817055524234167e-06, + "loss": 0.6768, + "step": 1830 + }, + { + "epoch": 2.8689979123173277, + "grad_norm": 1.1192113161087036, + "learning_rate": 2.7796121156737042e-06, + "loss": 0.6468, + "step": 1831 + }, + { + "epoch": 2.8705636743215033, + "grad_norm": 1.085511326789856, + "learning_rate": 2.777518480358444e-06, + "loss": 0.6923, + "step": 1832 + }, + { + "epoch": 2.8721294363256784, + "grad_norm": 1.1534816026687622, + "learning_rate": 2.7754246479644227e-06, + "loss": 0.7897, + "step": 1833 + }, + { + "epoch": 2.873695198329854, + "grad_norm": 1.2280727624893188, + "learning_rate": 2.7733306199785704e-06, + "loss": 0.5913, + "step": 1834 + }, + { + "epoch": 2.875260960334029, + "grad_norm": 1.1843805313110352, + "learning_rate": 2.771236397887952e-06, + "loss": 0.6554, + "step": 1835 + }, + { + "epoch": 2.8768267223382047, + "grad_norm": 1.1090878248214722, + "learning_rate": 2.7691419831797724e-06, + "loss": 0.7155, + "step": 1836 + }, + { + "epoch": 2.87839248434238, + "grad_norm": 1.3048434257507324, + "learning_rate": 2.767047377341371e-06, + "loss": 0.689, + "step": 1837 + }, + { + "epoch": 2.8799582463465554, + "grad_norm": 1.0179991722106934, + "learning_rate": 2.764952581860227e-06, + "loss": 0.6919, + "step": 1838 + }, + { + "epoch": 2.8815240083507305, + "grad_norm": 1.0855897665023804, + "learning_rate": 2.7628575982239513e-06, + "loss": 0.7377, + "step": 1839 + }, + { + "epoch": 2.883089770354906, + "grad_norm": 1.0885690450668335, + "learning_rate": 2.7607624279202887e-06, + "loss": 0.7436, + "step": 1840 + }, + { + "epoch": 2.8846555323590817, + "grad_norm": 0.9842485189437866, + "learning_rate": 2.758667072437117e-06, + "loss": 0.6818, + "step": 1841 + }, + { + "epoch": 2.886221294363257, + "grad_norm": 1.2647449970245361, + "learning_rate": 2.7565715332624455e-06, + "loss": 0.6892, + "step": 1842 + }, + { + "epoch": 2.887787056367432, + "grad_norm": 1.0584148168563843, + "learning_rate": 2.7544758118844134e-06, + "loss": 0.6945, + "step": 1843 + }, + { + "epoch": 2.8893528183716075, + "grad_norm": 1.2626157999038696, + "learning_rate": 2.7523799097912905e-06, + "loss": 0.6617, + "step": 1844 + }, + { + "epoch": 2.890918580375783, + "grad_norm": 1.1559371948242188, + "learning_rate": 2.7502838284714746e-06, + "loss": 0.679, + "step": 1845 + }, + { + "epoch": 2.892484342379958, + "grad_norm": 1.1553645133972168, + "learning_rate": 2.7481875694134884e-06, + "loss": 0.7216, + "step": 1846 + }, + { + "epoch": 2.8940501043841333, + "grad_norm": 1.1300830841064453, + "learning_rate": 2.7460911341059847e-06, + "loss": 0.6761, + "step": 1847 + }, + { + "epoch": 2.895615866388309, + "grad_norm": 1.1680023670196533, + "learning_rate": 2.7439945240377395e-06, + "loss": 0.675, + "step": 1848 + }, + { + "epoch": 2.8971816283924845, + "grad_norm": 1.0944335460662842, + "learning_rate": 2.7418977406976526e-06, + "loss": 0.6959, + "step": 1849 + }, + { + "epoch": 2.8987473903966596, + "grad_norm": 1.1949477195739746, + "learning_rate": 2.7398007855747455e-06, + "loss": 0.6822, + "step": 1850 + }, + { + "epoch": 2.900313152400835, + "grad_norm": 1.2108348608016968, + "learning_rate": 2.737703660158167e-06, + "loss": 0.6568, + "step": 1851 + }, + { + "epoch": 2.9018789144050103, + "grad_norm": 1.119547963142395, + "learning_rate": 2.73560636593718e-06, + "loss": 0.6508, + "step": 1852 + }, + { + "epoch": 2.903444676409186, + "grad_norm": 1.1591404676437378, + "learning_rate": 2.733508904401172e-06, + "loss": 0.6462, + "step": 1853 + }, + { + "epoch": 2.905010438413361, + "grad_norm": 1.1378757953643799, + "learning_rate": 2.7314112770396474e-06, + "loss": 0.6202, + "step": 1854 + }, + { + "epoch": 2.9065762004175366, + "grad_norm": 1.1146254539489746, + "learning_rate": 2.729313485342229e-06, + "loss": 0.6618, + "step": 1855 + }, + { + "epoch": 2.9081419624217117, + "grad_norm": 1.1829131841659546, + "learning_rate": 2.727215530798656e-06, + "loss": 0.6708, + "step": 1856 + }, + { + "epoch": 2.9097077244258873, + "grad_norm": 1.0372153520584106, + "learning_rate": 2.725117414898784e-06, + "loss": 0.6666, + "step": 1857 + }, + { + "epoch": 2.911273486430063, + "grad_norm": 1.1896759271621704, + "learning_rate": 2.723019139132581e-06, + "loss": 0.7169, + "step": 1858 + }, + { + "epoch": 2.912839248434238, + "grad_norm": 1.137988805770874, + "learning_rate": 2.720920704990131e-06, + "loss": 0.6872, + "step": 1859 + }, + { + "epoch": 2.914405010438413, + "grad_norm": 1.1891788244247437, + "learning_rate": 2.7188221139616303e-06, + "loss": 0.7105, + "step": 1860 + }, + { + "epoch": 2.9159707724425887, + "grad_norm": 1.0640588998794556, + "learning_rate": 2.716723367537385e-06, + "loss": 0.6905, + "step": 1861 + }, + { + "epoch": 2.9175365344467643, + "grad_norm": 1.2714836597442627, + "learning_rate": 2.7146244672078125e-06, + "loss": 0.6713, + "step": 1862 + }, + { + "epoch": 2.9191022964509394, + "grad_norm": 1.222401738166809, + "learning_rate": 2.7125254144634388e-06, + "loss": 0.6638, + "step": 1863 + }, + { + "epoch": 2.9206680584551146, + "grad_norm": 1.0254578590393066, + "learning_rate": 2.7104262107948993e-06, + "loss": 0.7374, + "step": 1864 + }, + { + "epoch": 2.92223382045929, + "grad_norm": 1.089826226234436, + "learning_rate": 2.7083268576929368e-06, + "loss": 0.6742, + "step": 1865 + }, + { + "epoch": 2.9237995824634657, + "grad_norm": 1.1508773565292358, + "learning_rate": 2.7062273566483984e-06, + "loss": 0.5628, + "step": 1866 + }, + { + "epoch": 2.925365344467641, + "grad_norm": 1.1214534044265747, + "learning_rate": 2.7041277091522365e-06, + "loss": 0.5819, + "step": 1867 + }, + { + "epoch": 2.9269311064718164, + "grad_norm": 1.2162528038024902, + "learning_rate": 2.70202791669551e-06, + "loss": 0.6774, + "step": 1868 + }, + { + "epoch": 2.9284968684759916, + "grad_norm": 1.130321741104126, + "learning_rate": 2.6999279807693783e-06, + "loss": 0.6617, + "step": 1869 + }, + { + "epoch": 2.930062630480167, + "grad_norm": 1.1052041053771973, + "learning_rate": 2.697827902865103e-06, + "loss": 0.6348, + "step": 1870 + }, + { + "epoch": 2.9316283924843423, + "grad_norm": 1.243033766746521, + "learning_rate": 2.695727684474048e-06, + "loss": 0.671, + "step": 1871 + }, + { + "epoch": 2.933194154488518, + "grad_norm": 1.163845419883728, + "learning_rate": 2.693627327087675e-06, + "loss": 0.6605, + "step": 1872 + }, + { + "epoch": 2.934759916492693, + "grad_norm": 1.0230295658111572, + "learning_rate": 2.6915268321975454e-06, + "loss": 0.6511, + "step": 1873 + }, + { + "epoch": 2.9363256784968685, + "grad_norm": 1.0699726343154907, + "learning_rate": 2.6894262012953194e-06, + "loss": 0.5641, + "step": 1874 + }, + { + "epoch": 2.937891440501044, + "grad_norm": 1.139090657234192, + "learning_rate": 2.6873254358727503e-06, + "loss": 0.7327, + "step": 1875 + }, + { + "epoch": 2.9394572025052192, + "grad_norm": 1.0480340719223022, + "learning_rate": 2.68522453742169e-06, + "loss": 0.6559, + "step": 1876 + }, + { + "epoch": 2.9410229645093944, + "grad_norm": 1.1894673109054565, + "learning_rate": 2.6831235074340854e-06, + "loss": 0.6235, + "step": 1877 + }, + { + "epoch": 2.94258872651357, + "grad_norm": 1.1843831539154053, + "learning_rate": 2.681022347401975e-06, + "loss": 0.7161, + "step": 1878 + }, + { + "epoch": 2.9441544885177455, + "grad_norm": 1.278631329536438, + "learning_rate": 2.6789210588174887e-06, + "loss": 0.6651, + "step": 1879 + }, + { + "epoch": 2.9457202505219207, + "grad_norm": 1.1330087184906006, + "learning_rate": 2.67681964317285e-06, + "loss": 0.682, + "step": 1880 + }, + { + "epoch": 2.947286012526096, + "grad_norm": 1.1718090772628784, + "learning_rate": 2.6747181019603718e-06, + "loss": 0.6656, + "step": 1881 + }, + { + "epoch": 2.9488517745302714, + "grad_norm": 1.0476222038269043, + "learning_rate": 2.672616436672456e-06, + "loss": 0.543, + "step": 1882 + }, + { + "epoch": 2.950417536534447, + "grad_norm": 1.1602777242660522, + "learning_rate": 2.6705146488015933e-06, + "loss": 0.6887, + "step": 1883 + }, + { + "epoch": 2.951983298538622, + "grad_norm": 1.1091994047164917, + "learning_rate": 2.66841273984036e-06, + "loss": 0.7068, + "step": 1884 + }, + { + "epoch": 2.9535490605427976, + "grad_norm": 1.1577891111373901, + "learning_rate": 2.666310711281419e-06, + "loss": 0.7285, + "step": 1885 + }, + { + "epoch": 2.9551148225469728, + "grad_norm": 1.1392951011657715, + "learning_rate": 2.6642085646175196e-06, + "loss": 0.6308, + "step": 1886 + }, + { + "epoch": 2.9566805845511483, + "grad_norm": 1.3026043176651, + "learning_rate": 2.6621063013414934e-06, + "loss": 0.7092, + "step": 1887 + }, + { + "epoch": 2.9582463465553235, + "grad_norm": 1.0123710632324219, + "learning_rate": 2.660003922946255e-06, + "loss": 0.7045, + "step": 1888 + }, + { + "epoch": 2.959812108559499, + "grad_norm": 1.0754269361495972, + "learning_rate": 2.6579014309247998e-06, + "loss": 0.7301, + "step": 1889 + }, + { + "epoch": 2.961377870563674, + "grad_norm": 1.19624662399292, + "learning_rate": 2.655798826770206e-06, + "loss": 0.699, + "step": 1890 + }, + { + "epoch": 2.9629436325678498, + "grad_norm": 1.0728397369384766, + "learning_rate": 2.6536961119756317e-06, + "loss": 0.7188, + "step": 1891 + }, + { + "epoch": 2.964509394572025, + "grad_norm": 1.231490135192871, + "learning_rate": 2.6515932880343103e-06, + "loss": 0.7309, + "step": 1892 + }, + { + "epoch": 2.9660751565762005, + "grad_norm": 1.2304288148880005, + "learning_rate": 2.6494903564395563e-06, + "loss": 0.7267, + "step": 1893 + }, + { + "epoch": 2.9676409185803756, + "grad_norm": 1.0533643960952759, + "learning_rate": 2.6473873186847577e-06, + "loss": 0.7011, + "step": 1894 + }, + { + "epoch": 2.969206680584551, + "grad_norm": 1.1855214834213257, + "learning_rate": 2.6452841762633795e-06, + "loss": 0.707, + "step": 1895 + }, + { + "epoch": 2.9707724425887267, + "grad_norm": 1.1279842853546143, + "learning_rate": 2.643180930668962e-06, + "loss": 0.6033, + "step": 1896 + }, + { + "epoch": 2.972338204592902, + "grad_norm": 1.102318525314331, + "learning_rate": 2.6410775833951164e-06, + "loss": 0.6696, + "step": 1897 + }, + { + "epoch": 2.973903966597077, + "grad_norm": 1.1372147798538208, + "learning_rate": 2.6389741359355275e-06, + "loss": 0.7164, + "step": 1898 + }, + { + "epoch": 2.9754697286012526, + "grad_norm": 1.2800220251083374, + "learning_rate": 2.6368705897839515e-06, + "loss": 0.733, + "step": 1899 + }, + { + "epoch": 2.977035490605428, + "grad_norm": 1.0855633020401, + "learning_rate": 2.634766946434214e-06, + "loss": 0.5894, + "step": 1900 + }, + { + "epoch": 2.9786012526096033, + "grad_norm": 1.1212704181671143, + "learning_rate": 2.6326632073802092e-06, + "loss": 0.6753, + "step": 1901 + }, + { + "epoch": 2.980167014613779, + "grad_norm": 1.0885487794876099, + "learning_rate": 2.6305593741159004e-06, + "loss": 0.6802, + "step": 1902 + }, + { + "epoch": 2.981732776617954, + "grad_norm": 1.1605111360549927, + "learning_rate": 2.6284554481353182e-06, + "loss": 0.5837, + "step": 1903 + }, + { + "epoch": 2.9832985386221296, + "grad_norm": 1.1119557619094849, + "learning_rate": 2.626351430932556e-06, + "loss": 0.6767, + "step": 1904 + }, + { + "epoch": 2.9848643006263047, + "grad_norm": 1.1678719520568848, + "learning_rate": 2.6242473240017757e-06, + "loss": 0.7052, + "step": 1905 + }, + { + "epoch": 2.9864300626304803, + "grad_norm": 1.1222507953643799, + "learning_rate": 2.622143128837201e-06, + "loss": 0.6751, + "step": 1906 + }, + { + "epoch": 2.9879958246346554, + "grad_norm": 1.1091363430023193, + "learning_rate": 2.620038846933119e-06, + "loss": 0.6427, + "step": 1907 + }, + { + "epoch": 2.989561586638831, + "grad_norm": 1.0949740409851074, + "learning_rate": 2.6179344797838775e-06, + "loss": 0.6353, + "step": 1908 + }, + { + "epoch": 2.991127348643006, + "grad_norm": 1.2391382455825806, + "learning_rate": 2.615830028883885e-06, + "loss": 0.7133, + "step": 1909 + }, + { + "epoch": 2.9926931106471817, + "grad_norm": 1.1227985620498657, + "learning_rate": 2.6137254957276105e-06, + "loss": 0.6942, + "step": 1910 + }, + { + "epoch": 2.994258872651357, + "grad_norm": 1.2233500480651855, + "learning_rate": 2.61162088180958e-06, + "loss": 0.7006, + "step": 1911 + }, + { + "epoch": 2.9958246346555324, + "grad_norm": 1.1352185010910034, + "learning_rate": 2.609516188624378e-06, + "loss": 0.6304, + "step": 1912 + }, + { + "epoch": 2.997390396659708, + "grad_norm": 1.2024694681167603, + "learning_rate": 2.6074114176666455e-06, + "loss": 0.6911, + "step": 1913 + }, + { + "epoch": 2.998956158663883, + "grad_norm": 1.1703287363052368, + "learning_rate": 2.605306570431077e-06, + "loss": 0.6528, + "step": 1914 + }, + { + "epoch": 3.0015657620041756, + "grad_norm": 1.1240348815917969, + "learning_rate": 2.6032016484124224e-06, + "loss": 1.3204, + "step": 1915 + }, + { + "epoch": 3.0031315240083507, + "grad_norm": 1.0691215991973877, + "learning_rate": 2.6010966531054852e-06, + "loss": 0.643, + "step": 1916 + }, + { + "epoch": 3.0046972860125263, + "grad_norm": 1.1520895957946777, + "learning_rate": 2.5989915860051205e-06, + "loss": 0.6326, + "step": 1917 + }, + { + "epoch": 3.0062630480167014, + "grad_norm": 1.1557644605636597, + "learning_rate": 2.5968864486062334e-06, + "loss": 0.6654, + "step": 1918 + }, + { + "epoch": 3.007828810020877, + "grad_norm": 1.216338872909546, + "learning_rate": 2.59478124240378e-06, + "loss": 0.6109, + "step": 1919 + }, + { + "epoch": 3.009394572025052, + "grad_norm": 1.168593406677246, + "learning_rate": 2.592675968892765e-06, + "loss": 0.6167, + "step": 1920 + }, + { + "epoch": 3.0109603340292277, + "grad_norm": 1.103684902191162, + "learning_rate": 2.590570629568241e-06, + "loss": 0.5883, + "step": 1921 + }, + { + "epoch": 3.012526096033403, + "grad_norm": 1.1893324851989746, + "learning_rate": 2.5884652259253067e-06, + "loss": 0.6295, + "step": 1922 + }, + { + "epoch": 3.0140918580375784, + "grad_norm": 1.046500325202942, + "learning_rate": 2.5863597594591077e-06, + "loss": 0.6338, + "step": 1923 + }, + { + "epoch": 3.0156576200417535, + "grad_norm": 1.1588988304138184, + "learning_rate": 2.5842542316648333e-06, + "loss": 0.6208, + "step": 1924 + }, + { + "epoch": 3.017223382045929, + "grad_norm": 1.0734959840774536, + "learning_rate": 2.5821486440377168e-06, + "loss": 0.5612, + "step": 1925 + }, + { + "epoch": 3.0187891440501042, + "grad_norm": 1.128631353378296, + "learning_rate": 2.5800429980730335e-06, + "loss": 0.6208, + "step": 1926 + }, + { + "epoch": 3.02035490605428, + "grad_norm": 1.1834310293197632, + "learning_rate": 2.5779372952661003e-06, + "loss": 0.6418, + "step": 1927 + }, + { + "epoch": 3.021920668058455, + "grad_norm": 1.217034101486206, + "learning_rate": 2.5758315371122733e-06, + "loss": 0.7117, + "step": 1928 + }, + { + "epoch": 3.0234864300626305, + "grad_norm": 1.060900330543518, + "learning_rate": 2.573725725106952e-06, + "loss": 0.6104, + "step": 1929 + }, + { + "epoch": 3.0250521920668056, + "grad_norm": 1.1937845945358276, + "learning_rate": 2.571619860745569e-06, + "loss": 0.672, + "step": 1930 + }, + { + "epoch": 3.026617954070981, + "grad_norm": 1.0636308193206787, + "learning_rate": 2.569513945523598e-06, + "loss": 0.6525, + "step": 1931 + }, + { + "epoch": 3.028183716075157, + "grad_norm": 1.2644332647323608, + "learning_rate": 2.5674079809365443e-06, + "loss": 0.6109, + "step": 1932 + }, + { + "epoch": 3.029749478079332, + "grad_norm": 1.1538089513778687, + "learning_rate": 2.5653019684799545e-06, + "loss": 0.6813, + "step": 1933 + }, + { + "epoch": 3.0313152400835075, + "grad_norm": 1.1390219926834106, + "learning_rate": 2.5631959096494047e-06, + "loss": 0.6725, + "step": 1934 + }, + { + "epoch": 3.0328810020876826, + "grad_norm": 1.17123281955719, + "learning_rate": 2.5610898059405044e-06, + "loss": 0.5332, + "step": 1935 + }, + { + "epoch": 3.034446764091858, + "grad_norm": 1.1410160064697266, + "learning_rate": 2.5589836588488958e-06, + "loss": 0.6445, + "step": 1936 + }, + { + "epoch": 3.0360125260960333, + "grad_norm": 1.134597897529602, + "learning_rate": 2.5568774698702526e-06, + "loss": 0.6513, + "step": 1937 + }, + { + "epoch": 3.037578288100209, + "grad_norm": 1.1042146682739258, + "learning_rate": 2.5547712405002762e-06, + "loss": 0.6622, + "step": 1938 + }, + { + "epoch": 3.039144050104384, + "grad_norm": 1.129697561264038, + "learning_rate": 2.5526649722346986e-06, + "loss": 0.674, + "step": 1939 + }, + { + "epoch": 3.0407098121085596, + "grad_norm": 1.1788016557693481, + "learning_rate": 2.550558666569279e-06, + "loss": 0.6768, + "step": 1940 + }, + { + "epoch": 3.0422755741127347, + "grad_norm": 1.0913262367248535, + "learning_rate": 2.5484523249998022e-06, + "loss": 0.6817, + "step": 1941 + }, + { + "epoch": 3.0438413361169103, + "grad_norm": 1.1731491088867188, + "learning_rate": 2.5463459490220794e-06, + "loss": 0.652, + "step": 1942 + }, + { + "epoch": 3.0454070981210855, + "grad_norm": 1.139214038848877, + "learning_rate": 2.544239540131947e-06, + "loss": 0.6462, + "step": 1943 + }, + { + "epoch": 3.046972860125261, + "grad_norm": 1.163565754890442, + "learning_rate": 2.5421330998252617e-06, + "loss": 0.611, + "step": 1944 + }, + { + "epoch": 3.048538622129436, + "grad_norm": 1.136677861213684, + "learning_rate": 2.5400266295979055e-06, + "loss": 0.6502, + "step": 1945 + }, + { + "epoch": 3.0501043841336117, + "grad_norm": 1.1309559345245361, + "learning_rate": 2.537920130945782e-06, + "loss": 0.5777, + "step": 1946 + }, + { + "epoch": 3.051670146137787, + "grad_norm": 1.0693000555038452, + "learning_rate": 2.535813605364812e-06, + "loss": 0.6596, + "step": 1947 + }, + { + "epoch": 3.0532359081419624, + "grad_norm": 1.2112295627593994, + "learning_rate": 2.533707054350938e-06, + "loss": 0.6033, + "step": 1948 + }, + { + "epoch": 3.0548016701461376, + "grad_norm": 1.186867594718933, + "learning_rate": 2.531600479400119e-06, + "loss": 0.6399, + "step": 1949 + }, + { + "epoch": 3.056367432150313, + "grad_norm": 1.146904468536377, + "learning_rate": 2.5294938820083324e-06, + "loss": 0.6431, + "step": 1950 + }, + { + "epoch": 3.0579331941544887, + "grad_norm": 1.0728389024734497, + "learning_rate": 2.5273872636715707e-06, + "loss": 0.6754, + "step": 1951 + }, + { + "epoch": 3.059498956158664, + "grad_norm": 1.219117522239685, + "learning_rate": 2.5252806258858416e-06, + "loss": 0.5961, + "step": 1952 + }, + { + "epoch": 3.0610647181628394, + "grad_norm": 1.1819669008255005, + "learning_rate": 2.5231739701471657e-06, + "loss": 0.6809, + "step": 1953 + }, + { + "epoch": 3.0626304801670146, + "grad_norm": 1.181559443473816, + "learning_rate": 2.5210672979515773e-06, + "loss": 0.5754, + "step": 1954 + }, + { + "epoch": 3.06419624217119, + "grad_norm": 1.1187459230422974, + "learning_rate": 2.5189606107951225e-06, + "loss": 0.6324, + "step": 1955 + }, + { + "epoch": 3.0657620041753653, + "grad_norm": 1.20028555393219, + "learning_rate": 2.5168539101738576e-06, + "loss": 0.6178, + "step": 1956 + }, + { + "epoch": 3.067327766179541, + "grad_norm": 1.1661895513534546, + "learning_rate": 2.514747197583849e-06, + "loss": 0.5646, + "step": 1957 + }, + { + "epoch": 3.068893528183716, + "grad_norm": 1.1893212795257568, + "learning_rate": 2.512640474521169e-06, + "loss": 0.6387, + "step": 1958 + }, + { + "epoch": 3.0704592901878915, + "grad_norm": 1.1453262567520142, + "learning_rate": 2.510533742481902e-06, + "loss": 0.6416, + "step": 1959 + }, + { + "epoch": 3.0720250521920667, + "grad_norm": 1.2452423572540283, + "learning_rate": 2.5084270029621353e-06, + "loss": 0.6435, + "step": 1960 + }, + { + "epoch": 3.0735908141962422, + "grad_norm": 1.1707086563110352, + "learning_rate": 2.5063202574579625e-06, + "loss": 0.6328, + "step": 1961 + }, + { + "epoch": 3.0751565762004174, + "grad_norm": 1.2181516885757446, + "learning_rate": 2.5042135074654815e-06, + "loss": 0.6084, + "step": 1962 + }, + { + "epoch": 3.076722338204593, + "grad_norm": 1.1872752904891968, + "learning_rate": 2.502106754480793e-06, + "loss": 0.5128, + "step": 1963 + }, + { + "epoch": 3.078288100208768, + "grad_norm": 1.1264151334762573, + "learning_rate": 2.5e-06, + "loss": 0.6391, + "step": 1964 + }, + { + "epoch": 3.0798538622129437, + "grad_norm": 1.1406108140945435, + "learning_rate": 2.497893245519208e-06, + "loss": 0.6686, + "step": 1965 + }, + { + "epoch": 3.081419624217119, + "grad_norm": 1.1387429237365723, + "learning_rate": 2.495786492534519e-06, + "loss": 0.6449, + "step": 1966 + }, + { + "epoch": 3.0829853862212944, + "grad_norm": 1.1874876022338867, + "learning_rate": 2.493679742542038e-06, + "loss": 0.6471, + "step": 1967 + }, + { + "epoch": 3.08455114822547, + "grad_norm": 1.2808914184570312, + "learning_rate": 2.4915729970378655e-06, + "loss": 0.5295, + "step": 1968 + }, + { + "epoch": 3.086116910229645, + "grad_norm": 1.203076720237732, + "learning_rate": 2.4894662575180982e-06, + "loss": 0.6161, + "step": 1969 + }, + { + "epoch": 3.0876826722338206, + "grad_norm": 1.0870691537857056, + "learning_rate": 2.4873595254788317e-06, + "loss": 0.6344, + "step": 1970 + }, + { + "epoch": 3.0892484342379958, + "grad_norm": 1.174511194229126, + "learning_rate": 2.4852528024161526e-06, + "loss": 0.5958, + "step": 1971 + }, + { + "epoch": 3.0908141962421714, + "grad_norm": 1.2232667207717896, + "learning_rate": 2.4831460898261428e-06, + "loss": 0.5745, + "step": 1972 + }, + { + "epoch": 3.0923799582463465, + "grad_norm": 1.1356587409973145, + "learning_rate": 2.4810393892048783e-06, + "loss": 0.645, + "step": 1973 + }, + { + "epoch": 3.093945720250522, + "grad_norm": 1.2031471729278564, + "learning_rate": 2.478932702048424e-06, + "loss": 0.6246, + "step": 1974 + }, + { + "epoch": 3.095511482254697, + "grad_norm": 1.1597861051559448, + "learning_rate": 2.476826029852835e-06, + "loss": 0.6527, + "step": 1975 + }, + { + "epoch": 3.0970772442588728, + "grad_norm": 1.106753945350647, + "learning_rate": 2.4747193741141593e-06, + "loss": 0.7164, + "step": 1976 + }, + { + "epoch": 3.098643006263048, + "grad_norm": 1.147586703300476, + "learning_rate": 2.47261273632843e-06, + "loss": 0.6295, + "step": 1977 + }, + { + "epoch": 3.1002087682672235, + "grad_norm": 1.2318609952926636, + "learning_rate": 2.470506117991668e-06, + "loss": 0.5468, + "step": 1978 + }, + { + "epoch": 3.1017745302713986, + "grad_norm": 1.1728826761245728, + "learning_rate": 2.4683995205998818e-06, + "loss": 0.6144, + "step": 1979 + }, + { + "epoch": 3.103340292275574, + "grad_norm": 1.1983925104141235, + "learning_rate": 2.4662929456490633e-06, + "loss": 0.6214, + "step": 1980 + }, + { + "epoch": 3.1049060542797493, + "grad_norm": 1.134922981262207, + "learning_rate": 2.4641863946351884e-06, + "loss": 0.631, + "step": 1981 + }, + { + "epoch": 3.106471816283925, + "grad_norm": 1.2184958457946777, + "learning_rate": 2.4620798690542184e-06, + "loss": 0.6642, + "step": 1982 + }, + { + "epoch": 3.1080375782881, + "grad_norm": 1.1365389823913574, + "learning_rate": 2.459973370402095e-06, + "loss": 0.631, + "step": 1983 + }, + { + "epoch": 3.1096033402922756, + "grad_norm": 1.107224464416504, + "learning_rate": 2.4578669001747387e-06, + "loss": 0.6455, + "step": 1984 + }, + { + "epoch": 3.1111691022964507, + "grad_norm": 1.09954833984375, + "learning_rate": 2.455760459868054e-06, + "loss": 0.6674, + "step": 1985 + }, + { + "epoch": 3.1127348643006263, + "grad_norm": 1.1493922472000122, + "learning_rate": 2.453654050977921e-06, + "loss": 0.6144, + "step": 1986 + }, + { + "epoch": 3.114300626304802, + "grad_norm": 1.1649893522262573, + "learning_rate": 2.4515476750001978e-06, + "loss": 0.6621, + "step": 1987 + }, + { + "epoch": 3.115866388308977, + "grad_norm": 1.16617751121521, + "learning_rate": 2.449441333430722e-06, + "loss": 0.7146, + "step": 1988 + }, + { + "epoch": 3.1174321503131526, + "grad_norm": 1.137216567993164, + "learning_rate": 2.447335027765302e-06, + "loss": 0.6294, + "step": 1989 + }, + { + "epoch": 3.1189979123173277, + "grad_norm": 1.1343148946762085, + "learning_rate": 2.445228759499725e-06, + "loss": 0.6289, + "step": 1990 + }, + { + "epoch": 3.1205636743215033, + "grad_norm": 1.113688588142395, + "learning_rate": 2.4431225301297482e-06, + "loss": 0.6331, + "step": 1991 + }, + { + "epoch": 3.1221294363256784, + "grad_norm": 1.1988941431045532, + "learning_rate": 2.4410163411511046e-06, + "loss": 0.5989, + "step": 1992 + }, + { + "epoch": 3.123695198329854, + "grad_norm": 1.1929535865783691, + "learning_rate": 2.438910194059497e-06, + "loss": 0.6789, + "step": 1993 + }, + { + "epoch": 3.125260960334029, + "grad_norm": 1.2655651569366455, + "learning_rate": 2.436804090350596e-06, + "loss": 0.5709, + "step": 1994 + }, + { + "epoch": 3.1268267223382047, + "grad_norm": 1.187930941581726, + "learning_rate": 2.4346980315200464e-06, + "loss": 0.6324, + "step": 1995 + }, + { + "epoch": 3.12839248434238, + "grad_norm": 1.1696043014526367, + "learning_rate": 2.432592019063456e-06, + "loss": 0.5987, + "step": 1996 + }, + { + "epoch": 3.1299582463465554, + "grad_norm": 1.228115439414978, + "learning_rate": 2.430486054476403e-06, + "loss": 0.6581, + "step": 1997 + }, + { + "epoch": 3.1315240083507305, + "grad_norm": 1.1672264337539673, + "learning_rate": 2.4283801392544317e-06, + "loss": 0.6412, + "step": 1998 + }, + { + "epoch": 3.133089770354906, + "grad_norm": 1.2138696908950806, + "learning_rate": 2.4262742748930493e-06, + "loss": 0.6761, + "step": 1999 + }, + { + "epoch": 3.1346555323590812, + "grad_norm": 1.215649962425232, + "learning_rate": 2.4241684628877267e-06, + "loss": 0.6074, + "step": 2000 + }, + { + "epoch": 3.136221294363257, + "grad_norm": 1.2324877977371216, + "learning_rate": 2.4220627047339006e-06, + "loss": 0.6402, + "step": 2001 + }, + { + "epoch": 3.137787056367432, + "grad_norm": 1.230507254600525, + "learning_rate": 2.4199570019269678e-06, + "loss": 0.6784, + "step": 2002 + }, + { + "epoch": 3.1393528183716075, + "grad_norm": 1.056394100189209, + "learning_rate": 2.4178513559622836e-06, + "loss": 0.6209, + "step": 2003 + }, + { + "epoch": 3.140918580375783, + "grad_norm": 1.178450107574463, + "learning_rate": 2.415745768335167e-06, + "loss": 0.6463, + "step": 2004 + }, + { + "epoch": 3.142484342379958, + "grad_norm": 1.2509713172912598, + "learning_rate": 2.4136402405408927e-06, + "loss": 0.5886, + "step": 2005 + }, + { + "epoch": 3.144050104384134, + "grad_norm": 1.1313499212265015, + "learning_rate": 2.411534774074693e-06, + "loss": 0.6083, + "step": 2006 + }, + { + "epoch": 3.145615866388309, + "grad_norm": 1.208187460899353, + "learning_rate": 2.4094293704317596e-06, + "loss": 0.6488, + "step": 2007 + }, + { + "epoch": 3.1471816283924845, + "grad_norm": 1.1712708473205566, + "learning_rate": 2.4073240311072363e-06, + "loss": 0.6386, + "step": 2008 + }, + { + "epoch": 3.1487473903966596, + "grad_norm": 1.2195419073104858, + "learning_rate": 2.4052187575962206e-06, + "loss": 0.6016, + "step": 2009 + }, + { + "epoch": 3.150313152400835, + "grad_norm": 1.2239069938659668, + "learning_rate": 2.4031135513937675e-06, + "loss": 0.6095, + "step": 2010 + }, + { + "epoch": 3.1518789144050103, + "grad_norm": 1.274491786956787, + "learning_rate": 2.4010084139948807e-06, + "loss": 0.5551, + "step": 2011 + }, + { + "epoch": 3.153444676409186, + "grad_norm": 1.104736089706421, + "learning_rate": 2.398903346894515e-06, + "loss": 0.6555, + "step": 2012 + }, + { + "epoch": 3.155010438413361, + "grad_norm": 1.159359097480774, + "learning_rate": 2.396798351587578e-06, + "loss": 0.7065, + "step": 2013 + }, + { + "epoch": 3.1565762004175366, + "grad_norm": 1.2481755018234253, + "learning_rate": 2.394693429568924e-06, + "loss": 0.6085, + "step": 2014 + }, + { + "epoch": 3.1581419624217117, + "grad_norm": 1.2071157693862915, + "learning_rate": 2.392588582333355e-06, + "loss": 0.6871, + "step": 2015 + }, + { + "epoch": 3.1597077244258873, + "grad_norm": 1.1674847602844238, + "learning_rate": 2.3904838113756228e-06, + "loss": 0.5756, + "step": 2016 + }, + { + "epoch": 3.1612734864300625, + "grad_norm": 1.1972459554672241, + "learning_rate": 2.3883791181904216e-06, + "loss": 0.6307, + "step": 2017 + }, + { + "epoch": 3.162839248434238, + "grad_norm": 1.2457903623580933, + "learning_rate": 2.3862745042723907e-06, + "loss": 0.5552, + "step": 2018 + }, + { + "epoch": 3.164405010438413, + "grad_norm": 1.2475699186325073, + "learning_rate": 2.3841699711161155e-06, + "loss": 0.6054, + "step": 2019 + }, + { + "epoch": 3.1659707724425887, + "grad_norm": 1.2934919595718384, + "learning_rate": 2.3820655202161237e-06, + "loss": 0.6986, + "step": 2020 + }, + { + "epoch": 3.1675365344467643, + "grad_norm": 1.1634780168533325, + "learning_rate": 2.3799611530668814e-06, + "loss": 0.6697, + "step": 2021 + }, + { + "epoch": 3.1691022964509394, + "grad_norm": 1.1987229585647583, + "learning_rate": 2.3778568711628e-06, + "loss": 0.6376, + "step": 2022 + }, + { + "epoch": 3.170668058455115, + "grad_norm": 1.0842245817184448, + "learning_rate": 2.375752675998225e-06, + "loss": 0.629, + "step": 2023 + }, + { + "epoch": 3.17223382045929, + "grad_norm": 1.1573731899261475, + "learning_rate": 2.373648569067444e-06, + "loss": 0.6231, + "step": 2024 + }, + { + "epoch": 3.1737995824634657, + "grad_norm": 1.1786472797393799, + "learning_rate": 2.371544551864683e-06, + "loss": 0.7082, + "step": 2025 + }, + { + "epoch": 3.175365344467641, + "grad_norm": 1.2212613821029663, + "learning_rate": 2.3694406258841004e-06, + "loss": 0.6382, + "step": 2026 + }, + { + "epoch": 3.1769311064718164, + "grad_norm": 1.0942133665084839, + "learning_rate": 2.367336792619791e-06, + "loss": 0.6157, + "step": 2027 + }, + { + "epoch": 3.1784968684759916, + "grad_norm": 1.198861837387085, + "learning_rate": 2.365233053565787e-06, + "loss": 0.6172, + "step": 2028 + }, + { + "epoch": 3.180062630480167, + "grad_norm": 1.2345949411392212, + "learning_rate": 2.3631294102160493e-06, + "loss": 0.6561, + "step": 2029 + }, + { + "epoch": 3.1816283924843423, + "grad_norm": 1.1807947158813477, + "learning_rate": 2.3610258640644725e-06, + "loss": 0.6323, + "step": 2030 + }, + { + "epoch": 3.183194154488518, + "grad_norm": 1.272806167602539, + "learning_rate": 2.3589224166048844e-06, + "loss": 0.6286, + "step": 2031 + }, + { + "epoch": 3.184759916492693, + "grad_norm": 1.1519666910171509, + "learning_rate": 2.3568190693310388e-06, + "loss": 0.6351, + "step": 2032 + }, + { + "epoch": 3.1863256784968685, + "grad_norm": 1.3199100494384766, + "learning_rate": 2.35471582373662e-06, + "loss": 0.6282, + "step": 2033 + }, + { + "epoch": 3.1878914405010437, + "grad_norm": 1.1525390148162842, + "learning_rate": 2.352612681315243e-06, + "loss": 0.5453, + "step": 2034 + }, + { + "epoch": 3.1894572025052192, + "grad_norm": 1.1851332187652588, + "learning_rate": 2.3505096435604446e-06, + "loss": 0.5935, + "step": 2035 + }, + { + "epoch": 3.1910229645093944, + "grad_norm": 1.1144623756408691, + "learning_rate": 2.3484067119656905e-06, + "loss": 0.6756, + "step": 2036 + }, + { + "epoch": 3.19258872651357, + "grad_norm": 1.2141358852386475, + "learning_rate": 2.3463038880243687e-06, + "loss": 0.649, + "step": 2037 + }, + { + "epoch": 3.1941544885177455, + "grad_norm": 1.206795573234558, + "learning_rate": 2.3442011732297944e-06, + "loss": 0.7051, + "step": 2038 + }, + { + "epoch": 3.1957202505219207, + "grad_norm": 1.1212406158447266, + "learning_rate": 2.3420985690752015e-06, + "loss": 0.6279, + "step": 2039 + }, + { + "epoch": 3.1972860125260962, + "grad_norm": 1.1145480871200562, + "learning_rate": 2.339996077053746e-06, + "loss": 0.6885, + "step": 2040 + }, + { + "epoch": 3.1988517745302714, + "grad_norm": 1.235465168952942, + "learning_rate": 2.337893698658508e-06, + "loss": 0.6146, + "step": 2041 + }, + { + "epoch": 3.200417536534447, + "grad_norm": 1.2589614391326904, + "learning_rate": 2.3357914353824816e-06, + "loss": 0.6892, + "step": 2042 + }, + { + "epoch": 3.201983298538622, + "grad_norm": 1.3488636016845703, + "learning_rate": 2.3336892887185816e-06, + "loss": 0.5532, + "step": 2043 + }, + { + "epoch": 3.2035490605427976, + "grad_norm": 1.218481183052063, + "learning_rate": 2.331587260159641e-06, + "loss": 0.6554, + "step": 2044 + }, + { + "epoch": 3.2051148225469728, + "grad_norm": 1.207727074623108, + "learning_rate": 2.329485351198408e-06, + "loss": 0.5789, + "step": 2045 + }, + { + "epoch": 3.2066805845511483, + "grad_norm": 1.143080711364746, + "learning_rate": 2.3273835633275444e-06, + "loss": 0.6171, + "step": 2046 + }, + { + "epoch": 3.2082463465553235, + "grad_norm": 1.1894160509109497, + "learning_rate": 2.325281898039629e-06, + "loss": 0.6514, + "step": 2047 + }, + { + "epoch": 3.209812108559499, + "grad_norm": 1.1722674369812012, + "learning_rate": 2.323180356827151e-06, + "loss": 0.6305, + "step": 2048 + }, + { + "epoch": 3.211377870563674, + "grad_norm": 1.131166934967041, + "learning_rate": 2.3210789411825117e-06, + "loss": 0.651, + "step": 2049 + }, + { + "epoch": 3.2129436325678498, + "grad_norm": 1.0506892204284668, + "learning_rate": 2.318977652598026e-06, + "loss": 0.6041, + "step": 2050 + }, + { + "epoch": 3.214509394572025, + "grad_norm": 1.2478179931640625, + "learning_rate": 2.3168764925659155e-06, + "loss": 0.6418, + "step": 2051 + }, + { + "epoch": 3.2160751565762005, + "grad_norm": 1.1366816759109497, + "learning_rate": 2.31477546257831e-06, + "loss": 0.6269, + "step": 2052 + }, + { + "epoch": 3.2176409185803756, + "grad_norm": 1.1247450113296509, + "learning_rate": 2.3126745641272506e-06, + "loss": 0.6706, + "step": 2053 + }, + { + "epoch": 3.219206680584551, + "grad_norm": 1.166385531425476, + "learning_rate": 2.3105737987046823e-06, + "loss": 0.6267, + "step": 2054 + }, + { + "epoch": 3.2207724425887267, + "grad_norm": 1.134945273399353, + "learning_rate": 2.3084731678024546e-06, + "loss": 0.6367, + "step": 2055 + }, + { + "epoch": 3.222338204592902, + "grad_norm": 1.129788875579834, + "learning_rate": 2.306372672912326e-06, + "loss": 0.6327, + "step": 2056 + }, + { + "epoch": 3.223903966597077, + "grad_norm": 1.0779662132263184, + "learning_rate": 2.3042723155259526e-06, + "loss": 0.6261, + "step": 2057 + }, + { + "epoch": 3.2254697286012526, + "grad_norm": 1.2631971836090088, + "learning_rate": 2.302172097134897e-06, + "loss": 0.5617, + "step": 2058 + }, + { + "epoch": 3.227035490605428, + "grad_norm": 1.180594801902771, + "learning_rate": 2.3000720192306226e-06, + "loss": 0.585, + "step": 2059 + }, + { + "epoch": 3.2286012526096033, + "grad_norm": 1.1890621185302734, + "learning_rate": 2.297972083304491e-06, + "loss": 0.6525, + "step": 2060 + }, + { + "epoch": 3.230167014613779, + "grad_norm": 1.199937343597412, + "learning_rate": 2.295872290847764e-06, + "loss": 0.6024, + "step": 2061 + }, + { + "epoch": 3.231732776617954, + "grad_norm": 1.2697222232818604, + "learning_rate": 2.293772643351603e-06, + "loss": 0.6007, + "step": 2062 + }, + { + "epoch": 3.2332985386221296, + "grad_norm": 1.173552393913269, + "learning_rate": 2.2916731423070645e-06, + "loss": 0.6576, + "step": 2063 + }, + { + "epoch": 3.2348643006263047, + "grad_norm": 1.2051188945770264, + "learning_rate": 2.2895737892051007e-06, + "loss": 0.6862, + "step": 2064 + }, + { + "epoch": 3.2364300626304803, + "grad_norm": 1.1974767446517944, + "learning_rate": 2.287474585536562e-06, + "loss": 0.6201, + "step": 2065 + }, + { + "epoch": 3.2379958246346554, + "grad_norm": 1.2346913814544678, + "learning_rate": 2.2853755327921883e-06, + "loss": 0.5762, + "step": 2066 + }, + { + "epoch": 3.239561586638831, + "grad_norm": 1.1864664554595947, + "learning_rate": 2.2832766324626153e-06, + "loss": 0.6389, + "step": 2067 + }, + { + "epoch": 3.241127348643006, + "grad_norm": 1.300667405128479, + "learning_rate": 2.28117788603837e-06, + "loss": 0.6645, + "step": 2068 + }, + { + "epoch": 3.2426931106471817, + "grad_norm": 1.2889467477798462, + "learning_rate": 2.2790792950098695e-06, + "loss": 0.6399, + "step": 2069 + }, + { + "epoch": 3.244258872651357, + "grad_norm": 1.1807681322097778, + "learning_rate": 2.2769808608674193e-06, + "loss": 0.5974, + "step": 2070 + }, + { + "epoch": 3.2458246346555324, + "grad_norm": 1.2693206071853638, + "learning_rate": 2.274882585101217e-06, + "loss": 0.7067, + "step": 2071 + }, + { + "epoch": 3.2473903966597075, + "grad_norm": 1.1668689250946045, + "learning_rate": 2.2727844692013448e-06, + "loss": 0.5987, + "step": 2072 + }, + { + "epoch": 3.248956158663883, + "grad_norm": 1.2399280071258545, + "learning_rate": 2.2706865146577716e-06, + "loss": 0.6821, + "step": 2073 + }, + { + "epoch": 3.2505219206680582, + "grad_norm": 1.2630313634872437, + "learning_rate": 2.2685887229603534e-06, + "loss": 0.5846, + "step": 2074 + }, + { + "epoch": 3.252087682672234, + "grad_norm": 1.1724474430084229, + "learning_rate": 2.266491095598829e-06, + "loss": 0.6272, + "step": 2075 + }, + { + "epoch": 3.2536534446764094, + "grad_norm": 1.3041969537734985, + "learning_rate": 2.2643936340628205e-06, + "loss": 0.6246, + "step": 2076 + }, + { + "epoch": 3.2552192066805845, + "grad_norm": 1.152488350868225, + "learning_rate": 2.262296339841834e-06, + "loss": 0.6827, + "step": 2077 + }, + { + "epoch": 3.25678496868476, + "grad_norm": 1.1879392862319946, + "learning_rate": 2.260199214425255e-06, + "loss": 0.681, + "step": 2078 + }, + { + "epoch": 3.258350730688935, + "grad_norm": 1.1999655961990356, + "learning_rate": 2.2581022593023483e-06, + "loss": 0.7044, + "step": 2079 + }, + { + "epoch": 3.259916492693111, + "grad_norm": 1.2020212411880493, + "learning_rate": 2.256005475962261e-06, + "loss": 0.6206, + "step": 2080 + }, + { + "epoch": 3.261482254697286, + "grad_norm": 1.312864065170288, + "learning_rate": 2.2539088658940158e-06, + "loss": 0.6385, + "step": 2081 + }, + { + "epoch": 3.2630480167014615, + "grad_norm": 1.1875925064086914, + "learning_rate": 2.251812430586512e-06, + "loss": 0.5999, + "step": 2082 + }, + { + "epoch": 3.2646137787056366, + "grad_norm": 1.0950852632522583, + "learning_rate": 2.249716171528526e-06, + "loss": 0.7346, + "step": 2083 + }, + { + "epoch": 3.266179540709812, + "grad_norm": 1.2490246295928955, + "learning_rate": 2.24762009020871e-06, + "loss": 0.5889, + "step": 2084 + }, + { + "epoch": 3.2677453027139873, + "grad_norm": 1.1770070791244507, + "learning_rate": 2.2455241881155874e-06, + "loss": 0.681, + "step": 2085 + }, + { + "epoch": 3.269311064718163, + "grad_norm": 1.1888784170150757, + "learning_rate": 2.2434284667375553e-06, + "loss": 0.566, + "step": 2086 + }, + { + "epoch": 3.270876826722338, + "grad_norm": 1.290913701057434, + "learning_rate": 2.2413329275628836e-06, + "loss": 0.6737, + "step": 2087 + }, + { + "epoch": 3.2724425887265136, + "grad_norm": 1.1826350688934326, + "learning_rate": 2.2392375720797126e-06, + "loss": 0.6149, + "step": 2088 + }, + { + "epoch": 3.2740083507306887, + "grad_norm": 1.1560386419296265, + "learning_rate": 2.237142401776049e-06, + "loss": 0.6214, + "step": 2089 + }, + { + "epoch": 3.2755741127348643, + "grad_norm": 1.1863820552825928, + "learning_rate": 2.2350474181397736e-06, + "loss": 0.592, + "step": 2090 + }, + { + "epoch": 3.2771398747390394, + "grad_norm": 1.2404558658599854, + "learning_rate": 2.2329526226586294e-06, + "loss": 0.6648, + "step": 2091 + }, + { + "epoch": 3.278705636743215, + "grad_norm": 1.1192048788070679, + "learning_rate": 2.2308580168202284e-06, + "loss": 0.6657, + "step": 2092 + }, + { + "epoch": 3.2802713987473906, + "grad_norm": 1.2412128448486328, + "learning_rate": 2.228763602112049e-06, + "loss": 0.5999, + "step": 2093 + }, + { + "epoch": 3.2818371607515657, + "grad_norm": 1.2265843152999878, + "learning_rate": 2.2266693800214305e-06, + "loss": 0.6335, + "step": 2094 + }, + { + "epoch": 3.2834029227557413, + "grad_norm": 1.1712675094604492, + "learning_rate": 2.2245753520355773e-06, + "loss": 0.6533, + "step": 2095 + }, + { + "epoch": 3.2849686847599164, + "grad_norm": 1.2727960348129272, + "learning_rate": 2.222481519641557e-06, + "loss": 0.6233, + "step": 2096 + }, + { + "epoch": 3.286534446764092, + "grad_norm": 1.0983766317367554, + "learning_rate": 2.220387884326297e-06, + "loss": 0.6268, + "step": 2097 + }, + { + "epoch": 3.288100208768267, + "grad_norm": 1.2702240943908691, + "learning_rate": 2.2182944475765837e-06, + "loss": 0.6453, + "step": 2098 + }, + { + "epoch": 3.2896659707724427, + "grad_norm": 1.2952048778533936, + "learning_rate": 2.2162012108790656e-06, + "loss": 0.6177, + "step": 2099 + }, + { + "epoch": 3.291231732776618, + "grad_norm": 1.1496988534927368, + "learning_rate": 2.214108175720246e-06, + "loss": 0.6123, + "step": 2100 + }, + { + "epoch": 3.2927974947807934, + "grad_norm": 1.1527506113052368, + "learning_rate": 2.212015343586486e-06, + "loss": 0.5732, + "step": 2101 + }, + { + "epoch": 3.2943632567849686, + "grad_norm": 1.1944189071655273, + "learning_rate": 2.209922715964005e-06, + "loss": 0.6502, + "step": 2102 + }, + { + "epoch": 3.295929018789144, + "grad_norm": 1.2035129070281982, + "learning_rate": 2.2078302943388736e-06, + "loss": 0.6233, + "step": 2103 + }, + { + "epoch": 3.2974947807933193, + "grad_norm": 1.111652135848999, + "learning_rate": 2.2057380801970164e-06, + "loss": 0.6082, + "step": 2104 + }, + { + "epoch": 3.299060542797495, + "grad_norm": 1.223018765449524, + "learning_rate": 2.2036460750242136e-06, + "loss": 0.6537, + "step": 2105 + }, + { + "epoch": 3.30062630480167, + "grad_norm": 1.2612967491149902, + "learning_rate": 2.2015542803060952e-06, + "loss": 0.6857, + "step": 2106 + }, + { + "epoch": 3.3021920668058455, + "grad_norm": 1.0949403047561646, + "learning_rate": 2.1994626975281403e-06, + "loss": 0.591, + "step": 2107 + }, + { + "epoch": 3.3037578288100207, + "grad_norm": 1.1419187784194946, + "learning_rate": 2.197371328175681e-06, + "loss": 0.6729, + "step": 2108 + }, + { + "epoch": 3.3053235908141962, + "grad_norm": 1.1853833198547363, + "learning_rate": 2.195280173733894e-06, + "loss": 0.6015, + "step": 2109 + }, + { + "epoch": 3.306889352818372, + "grad_norm": 1.2181181907653809, + "learning_rate": 2.193189235687805e-06, + "loss": 0.6378, + "step": 2110 + }, + { + "epoch": 3.308455114822547, + "grad_norm": 1.291874885559082, + "learning_rate": 2.191098515522289e-06, + "loss": 0.6883, + "step": 2111 + }, + { + "epoch": 3.3100208768267225, + "grad_norm": 1.1752232313156128, + "learning_rate": 2.1890080147220606e-06, + "loss": 0.5764, + "step": 2112 + }, + { + "epoch": 3.3115866388308977, + "grad_norm": 1.2201036214828491, + "learning_rate": 2.186917734771682e-06, + "loss": 0.6593, + "step": 2113 + }, + { + "epoch": 3.3131524008350732, + "grad_norm": 1.2628161907196045, + "learning_rate": 2.1848276771555586e-06, + "loss": 0.5571, + "step": 2114 + }, + { + "epoch": 3.3147181628392484, + "grad_norm": 1.2265757322311401, + "learning_rate": 2.1827378433579384e-06, + "loss": 0.5375, + "step": 2115 + }, + { + "epoch": 3.316283924843424, + "grad_norm": 1.2193812131881714, + "learning_rate": 2.1806482348629065e-06, + "loss": 0.6333, + "step": 2116 + }, + { + "epoch": 3.317849686847599, + "grad_norm": 1.1655728816986084, + "learning_rate": 2.1785588531543945e-06, + "loss": 0.6113, + "step": 2117 + }, + { + "epoch": 3.3194154488517746, + "grad_norm": 1.2318220138549805, + "learning_rate": 2.1764696997161665e-06, + "loss": 0.659, + "step": 2118 + }, + { + "epoch": 3.3209812108559498, + "grad_norm": 1.1320031881332397, + "learning_rate": 2.174380776031828e-06, + "loss": 0.656, + "step": 2119 + }, + { + "epoch": 3.3225469728601253, + "grad_norm": 1.1628968715667725, + "learning_rate": 2.1722920835848226e-06, + "loss": 0.6614, + "step": 2120 + }, + { + "epoch": 3.3241127348643005, + "grad_norm": 1.2760236263275146, + "learning_rate": 2.170203623858426e-06, + "loss": 0.5455, + "step": 2121 + }, + { + "epoch": 3.325678496868476, + "grad_norm": 1.1655547618865967, + "learning_rate": 2.1681153983357497e-06, + "loss": 0.6167, + "step": 2122 + }, + { + "epoch": 3.327244258872651, + "grad_norm": 1.1703981161117554, + "learning_rate": 2.1660274084997414e-06, + "loss": 0.6874, + "step": 2123 + }, + { + "epoch": 3.3288100208768268, + "grad_norm": 1.2664302587509155, + "learning_rate": 2.1639396558331794e-06, + "loss": 0.6355, + "step": 2124 + }, + { + "epoch": 3.330375782881002, + "grad_norm": 1.1683710813522339, + "learning_rate": 2.1618521418186725e-06, + "loss": 0.6236, + "step": 2125 + }, + { + "epoch": 3.3319415448851775, + "grad_norm": 1.219686508178711, + "learning_rate": 2.159764867938663e-06, + "loss": 0.622, + "step": 2126 + }, + { + "epoch": 3.333507306889353, + "grad_norm": 1.1496042013168335, + "learning_rate": 2.15767783567542e-06, + "loss": 0.5684, + "step": 2127 + }, + { + "epoch": 3.335073068893528, + "grad_norm": 1.1844868659973145, + "learning_rate": 2.155591046511043e-06, + "loss": 0.5401, + "step": 2128 + }, + { + "epoch": 3.3366388308977033, + "grad_norm": 1.0850353240966797, + "learning_rate": 2.1535045019274557e-06, + "loss": 0.7124, + "step": 2129 + }, + { + "epoch": 3.338204592901879, + "grad_norm": 1.1659971475601196, + "learning_rate": 2.1514182034064123e-06, + "loss": 0.6502, + "step": 2130 + }, + { + "epoch": 3.3397703549060545, + "grad_norm": 1.1342421770095825, + "learning_rate": 2.1493321524294902e-06, + "loss": 0.6352, + "step": 2131 + }, + { + "epoch": 3.3413361169102296, + "grad_norm": 1.1351344585418701, + "learning_rate": 2.1472463504780893e-06, + "loss": 0.7151, + "step": 2132 + }, + { + "epoch": 3.342901878914405, + "grad_norm": 1.2182797193527222, + "learning_rate": 2.145160799033437e-06, + "loss": 0.6671, + "step": 2133 + }, + { + "epoch": 3.3444676409185803, + "grad_norm": 1.0785852670669556, + "learning_rate": 2.1430754995765783e-06, + "loss": 0.6595, + "step": 2134 + }, + { + "epoch": 3.346033402922756, + "grad_norm": 1.2832294702529907, + "learning_rate": 2.1409904535883816e-06, + "loss": 0.6094, + "step": 2135 + }, + { + "epoch": 3.347599164926931, + "grad_norm": 1.1902724504470825, + "learning_rate": 2.138905662549536e-06, + "loss": 0.6205, + "step": 2136 + }, + { + "epoch": 3.3491649269311066, + "grad_norm": 1.1487525701522827, + "learning_rate": 2.136821127940547e-06, + "loss": 0.6812, + "step": 2137 + }, + { + "epoch": 3.3507306889352817, + "grad_norm": 1.214062213897705, + "learning_rate": 2.1347368512417394e-06, + "loss": 0.6414, + "step": 2138 + }, + { + "epoch": 3.3522964509394573, + "grad_norm": 1.1467952728271484, + "learning_rate": 2.132652833933256e-06, + "loss": 0.6825, + "step": 2139 + }, + { + "epoch": 3.3538622129436324, + "grad_norm": 1.1381312608718872, + "learning_rate": 2.1305690774950543e-06, + "loss": 0.5342, + "step": 2140 + }, + { + "epoch": 3.355427974947808, + "grad_norm": 1.2287116050720215, + "learning_rate": 2.1284855834069044e-06, + "loss": 0.6098, + "step": 2141 + }, + { + "epoch": 3.356993736951983, + "grad_norm": 1.1297441720962524, + "learning_rate": 2.1264023531483947e-06, + "loss": 0.6527, + "step": 2142 + }, + { + "epoch": 3.3585594989561587, + "grad_norm": 1.1599788665771484, + "learning_rate": 2.124319388198923e-06, + "loss": 0.6767, + "step": 2143 + }, + { + "epoch": 3.3601252609603343, + "grad_norm": 1.2442164421081543, + "learning_rate": 2.1222366900376977e-06, + "loss": 0.618, + "step": 2144 + }, + { + "epoch": 3.3616910229645094, + "grad_norm": 1.2075343132019043, + "learning_rate": 2.120154260143742e-06, + "loss": 0.6413, + "step": 2145 + }, + { + "epoch": 3.3632567849686845, + "grad_norm": 1.173141360282898, + "learning_rate": 2.1180720999958857e-06, + "loss": 0.6272, + "step": 2146 + }, + { + "epoch": 3.36482254697286, + "grad_norm": 1.2015706300735474, + "learning_rate": 2.115990211072765e-06, + "loss": 0.6148, + "step": 2147 + }, + { + "epoch": 3.3663883089770357, + "grad_norm": 1.2163746356964111, + "learning_rate": 2.1139085948528286e-06, + "loss": 0.697, + "step": 2148 + }, + { + "epoch": 3.367954070981211, + "grad_norm": 1.2009632587432861, + "learning_rate": 2.111827252814328e-06, + "loss": 0.6202, + "step": 2149 + }, + { + "epoch": 3.3695198329853864, + "grad_norm": 1.1929175853729248, + "learning_rate": 2.1097461864353202e-06, + "loss": 0.5118, + "step": 2150 + }, + { + "epoch": 3.3710855949895615, + "grad_norm": 1.1137278079986572, + "learning_rate": 2.1076653971936685e-06, + "loss": 0.6489, + "step": 2151 + }, + { + "epoch": 3.372651356993737, + "grad_norm": 1.248371958732605, + "learning_rate": 2.1055848865670363e-06, + "loss": 0.6153, + "step": 2152 + }, + { + "epoch": 3.374217118997912, + "grad_norm": 1.1446887254714966, + "learning_rate": 2.103504656032891e-06, + "loss": 0.586, + "step": 2153 + }, + { + "epoch": 3.375782881002088, + "grad_norm": 1.3197392225265503, + "learning_rate": 2.1014247070685024e-06, + "loss": 0.6469, + "step": 2154 + }, + { + "epoch": 3.377348643006263, + "grad_norm": 1.1482443809509277, + "learning_rate": 2.099345041150938e-06, + "loss": 0.5673, + "step": 2155 + }, + { + "epoch": 3.3789144050104385, + "grad_norm": 1.3151997327804565, + "learning_rate": 2.097265659757064e-06, + "loss": 0.6684, + "step": 2156 + }, + { + "epoch": 3.3804801670146136, + "grad_norm": 1.1619030237197876, + "learning_rate": 2.0951865643635472e-06, + "loss": 0.6521, + "step": 2157 + }, + { + "epoch": 3.382045929018789, + "grad_norm": 1.2650394439697266, + "learning_rate": 2.09310775644685e-06, + "loss": 0.5612, + "step": 2158 + }, + { + "epoch": 3.3836116910229643, + "grad_norm": 1.1590896844863892, + "learning_rate": 2.0910292374832285e-06, + "loss": 0.6129, + "step": 2159 + }, + { + "epoch": 3.38517745302714, + "grad_norm": 1.1127238273620605, + "learning_rate": 2.088951008948738e-06, + "loss": 0.5928, + "step": 2160 + }, + { + "epoch": 3.3867432150313155, + "grad_norm": 1.203165888786316, + "learning_rate": 2.086873072319223e-06, + "loss": 0.6821, + "step": 2161 + }, + { + "epoch": 3.3883089770354906, + "grad_norm": 1.186804175376892, + "learning_rate": 2.0847954290703234e-06, + "loss": 0.6913, + "step": 2162 + }, + { + "epoch": 3.3898747390396657, + "grad_norm": 1.2237906455993652, + "learning_rate": 2.0827180806774717e-06, + "loss": 0.635, + "step": 2163 + }, + { + "epoch": 3.3914405010438413, + "grad_norm": 1.1457163095474243, + "learning_rate": 2.080641028615888e-06, + "loss": 0.6628, + "step": 2164 + }, + { + "epoch": 3.393006263048017, + "grad_norm": 1.2211480140686035, + "learning_rate": 2.078564274360582e-06, + "loss": 0.6259, + "step": 2165 + }, + { + "epoch": 3.394572025052192, + "grad_norm": 1.1769284009933472, + "learning_rate": 2.076487819386356e-06, + "loss": 0.6458, + "step": 2166 + }, + { + "epoch": 3.3961377870563676, + "grad_norm": 1.1229456663131714, + "learning_rate": 2.0744116651677967e-06, + "loss": 0.6518, + "step": 2167 + }, + { + "epoch": 3.3977035490605427, + "grad_norm": 1.2193604707717896, + "learning_rate": 2.072335813179275e-06, + "loss": 0.636, + "step": 2168 + }, + { + "epoch": 3.3992693110647183, + "grad_norm": 1.2844295501708984, + "learning_rate": 2.0702602648949536e-06, + "loss": 0.6275, + "step": 2169 + }, + { + "epoch": 3.4008350730688934, + "grad_norm": 1.2357110977172852, + "learning_rate": 2.0681850217887737e-06, + "loss": 0.6535, + "step": 2170 + }, + { + "epoch": 3.402400835073069, + "grad_norm": 1.191166877746582, + "learning_rate": 2.0661100853344608e-06, + "loss": 0.63, + "step": 2171 + }, + { + "epoch": 3.403966597077244, + "grad_norm": 1.1162155866622925, + "learning_rate": 2.064035457005526e-06, + "loss": 0.6438, + "step": 2172 + }, + { + "epoch": 3.4055323590814197, + "grad_norm": 1.1535841226577759, + "learning_rate": 2.0619611382752575e-06, + "loss": 0.6106, + "step": 2173 + }, + { + "epoch": 3.407098121085595, + "grad_norm": 1.2141941785812378, + "learning_rate": 2.059887130616726e-06, + "loss": 0.6009, + "step": 2174 + }, + { + "epoch": 3.4086638830897704, + "grad_norm": 1.1389793157577515, + "learning_rate": 2.05781343550278e-06, + "loss": 0.6691, + "step": 2175 + }, + { + "epoch": 3.4102296450939455, + "grad_norm": 1.2746304273605347, + "learning_rate": 2.055740054406048e-06, + "loss": 0.5752, + "step": 2176 + }, + { + "epoch": 3.411795407098121, + "grad_norm": 1.250404953956604, + "learning_rate": 2.053666988798933e-06, + "loss": 0.6839, + "step": 2177 + }, + { + "epoch": 3.4133611691022967, + "grad_norm": 1.2025327682495117, + "learning_rate": 2.0515942401536148e-06, + "loss": 0.6806, + "step": 2178 + }, + { + "epoch": 3.414926931106472, + "grad_norm": 1.1835360527038574, + "learning_rate": 2.0495218099420504e-06, + "loss": 0.6353, + "step": 2179 + }, + { + "epoch": 3.416492693110647, + "grad_norm": 1.3114181756973267, + "learning_rate": 2.0474496996359676e-06, + "loss": 0.6561, + "step": 2180 + }, + { + "epoch": 3.4180584551148225, + "grad_norm": 1.0880002975463867, + "learning_rate": 2.045377910706867e-06, + "loss": 0.6646, + "step": 2181 + }, + { + "epoch": 3.419624217118998, + "grad_norm": 1.1725163459777832, + "learning_rate": 2.0433064446260235e-06, + "loss": 0.5227, + "step": 2182 + }, + { + "epoch": 3.4211899791231732, + "grad_norm": 1.1769042015075684, + "learning_rate": 2.041235302864482e-06, + "loss": 0.6465, + "step": 2183 + }, + { + "epoch": 3.422755741127349, + "grad_norm": 1.2654465436935425, + "learning_rate": 2.039164486893054e-06, + "loss": 0.6407, + "step": 2184 + }, + { + "epoch": 3.424321503131524, + "grad_norm": 1.2088981866836548, + "learning_rate": 2.037093998182325e-06, + "loss": 0.6607, + "step": 2185 + }, + { + "epoch": 3.4258872651356995, + "grad_norm": 1.2232211828231812, + "learning_rate": 2.035023838202643e-06, + "loss": 0.6702, + "step": 2186 + }, + { + "epoch": 3.4274530271398747, + "grad_norm": 1.214909553527832, + "learning_rate": 2.032954008424126e-06, + "loss": 0.6322, + "step": 2187 + }, + { + "epoch": 3.4290187891440502, + "grad_norm": 1.1763339042663574, + "learning_rate": 2.0308845103166555e-06, + "loss": 0.6287, + "step": 2188 + }, + { + "epoch": 3.4305845511482254, + "grad_norm": 1.3962702751159668, + "learning_rate": 2.028815345349879e-06, + "loss": 0.547, + "step": 2189 + }, + { + "epoch": 3.432150313152401, + "grad_norm": 1.2137348651885986, + "learning_rate": 2.026746514993205e-06, + "loss": 0.6401, + "step": 2190 + }, + { + "epoch": 3.433716075156576, + "grad_norm": 1.143447756767273, + "learning_rate": 2.024678020715808e-06, + "loss": 0.6563, + "step": 2191 + }, + { + "epoch": 3.4352818371607516, + "grad_norm": 1.1660840511322021, + "learning_rate": 2.022609863986621e-06, + "loss": 0.6908, + "step": 2192 + }, + { + "epoch": 3.4368475991649268, + "grad_norm": 1.2545430660247803, + "learning_rate": 2.020542046274337e-06, + "loss": 0.6209, + "step": 2193 + }, + { + "epoch": 3.4384133611691023, + "grad_norm": 1.3274216651916504, + "learning_rate": 2.0184745690474116e-06, + "loss": 0.6243, + "step": 2194 + }, + { + "epoch": 3.4399791231732775, + "grad_norm": 1.2278743982315063, + "learning_rate": 2.0164074337740548e-06, + "loss": 0.6487, + "step": 2195 + }, + { + "epoch": 3.441544885177453, + "grad_norm": 1.1612262725830078, + "learning_rate": 2.0143406419222354e-06, + "loss": 0.5982, + "step": 2196 + }, + { + "epoch": 3.443110647181628, + "grad_norm": 1.2417254447937012, + "learning_rate": 2.01227419495968e-06, + "loss": 0.6292, + "step": 2197 + }, + { + "epoch": 3.4446764091858038, + "grad_norm": 1.1197260618209839, + "learning_rate": 2.0102080943538664e-06, + "loss": 0.6478, + "step": 2198 + }, + { + "epoch": 3.4462421711899793, + "grad_norm": 1.113791584968567, + "learning_rate": 2.008142341572029e-06, + "loss": 0.5747, + "step": 2199 + }, + { + "epoch": 3.4478079331941545, + "grad_norm": 1.1108101606369019, + "learning_rate": 2.0060769380811552e-06, + "loss": 0.6325, + "step": 2200 + }, + { + "epoch": 3.44937369519833, + "grad_norm": 1.2102993726730347, + "learning_rate": 2.004011885347985e-06, + "loss": 0.6066, + "step": 2201 + }, + { + "epoch": 3.450939457202505, + "grad_norm": 1.304426908493042, + "learning_rate": 2.0019471848390055e-06, + "loss": 0.6503, + "step": 2202 + }, + { + "epoch": 3.4525052192066807, + "grad_norm": 1.244708776473999, + "learning_rate": 1.9998828380204596e-06, + "loss": 0.6723, + "step": 2203 + }, + { + "epoch": 3.454070981210856, + "grad_norm": 1.1601401567459106, + "learning_rate": 1.997818846358333e-06, + "loss": 0.684, + "step": 2204 + }, + { + "epoch": 3.4556367432150314, + "grad_norm": 1.1886024475097656, + "learning_rate": 1.9957552113183635e-06, + "loss": 0.6277, + "step": 2205 + }, + { + "epoch": 3.4572025052192066, + "grad_norm": 1.2624013423919678, + "learning_rate": 1.993691934366035e-06, + "loss": 0.6569, + "step": 2206 + }, + { + "epoch": 3.458768267223382, + "grad_norm": 1.272737979888916, + "learning_rate": 1.991629016966575e-06, + "loss": 0.6481, + "step": 2207 + }, + { + "epoch": 3.4603340292275573, + "grad_norm": 1.2370678186416626, + "learning_rate": 1.989566460584956e-06, + "loss": 0.6598, + "step": 2208 + }, + { + "epoch": 3.461899791231733, + "grad_norm": 1.1626964807510376, + "learning_rate": 1.9875042666858973e-06, + "loss": 0.6582, + "step": 2209 + }, + { + "epoch": 3.463465553235908, + "grad_norm": 1.1608844995498657, + "learning_rate": 1.9854424367338578e-06, + "loss": 0.5931, + "step": 2210 + }, + { + "epoch": 3.4650313152400836, + "grad_norm": 1.2470782995224, + "learning_rate": 1.983380972193037e-06, + "loss": 0.6186, + "step": 2211 + }, + { + "epoch": 3.4665970772442587, + "grad_norm": 1.1697843074798584, + "learning_rate": 1.98131987452738e-06, + "loss": 0.6373, + "step": 2212 + }, + { + "epoch": 3.4681628392484343, + "grad_norm": 1.1374456882476807, + "learning_rate": 1.9792591452005645e-06, + "loss": 0.6518, + "step": 2213 + }, + { + "epoch": 3.4697286012526094, + "grad_norm": 1.1979857683181763, + "learning_rate": 1.9771987856760115e-06, + "loss": 0.6174, + "step": 2214 + }, + { + "epoch": 3.471294363256785, + "grad_norm": 1.2085691690444946, + "learning_rate": 1.975138797416879e-06, + "loss": 0.6296, + "step": 2215 + }, + { + "epoch": 3.4728601252609606, + "grad_norm": 1.1843719482421875, + "learning_rate": 1.973079181886059e-06, + "loss": 0.6796, + "step": 2216 + }, + { + "epoch": 3.4744258872651357, + "grad_norm": 1.230175256729126, + "learning_rate": 1.97101994054618e-06, + "loss": 0.6626, + "step": 2217 + }, + { + "epoch": 3.4759916492693113, + "grad_norm": 1.14729642868042, + "learning_rate": 1.9689610748596056e-06, + "loss": 0.5719, + "step": 2218 + }, + { + "epoch": 3.4775574112734864, + "grad_norm": 1.3141698837280273, + "learning_rate": 1.966902586288432e-06, + "loss": 0.6455, + "step": 2219 + }, + { + "epoch": 3.479123173277662, + "grad_norm": 1.15304434299469, + "learning_rate": 1.964844476294487e-06, + "loss": 0.6873, + "step": 2220 + }, + { + "epoch": 3.480688935281837, + "grad_norm": 1.193190336227417, + "learning_rate": 1.9627867463393308e-06, + "loss": 0.6656, + "step": 2221 + }, + { + "epoch": 3.4822546972860127, + "grad_norm": 1.2585062980651855, + "learning_rate": 1.960729397884252e-06, + "loss": 0.673, + "step": 2222 + }, + { + "epoch": 3.483820459290188, + "grad_norm": 1.2261855602264404, + "learning_rate": 1.9586724323902704e-06, + "loss": 0.6563, + "step": 2223 + }, + { + "epoch": 3.4853862212943634, + "grad_norm": 1.1301530599594116, + "learning_rate": 1.9566158513181312e-06, + "loss": 0.6392, + "step": 2224 + }, + { + "epoch": 3.4869519832985385, + "grad_norm": 1.1394745111465454, + "learning_rate": 1.954559656128309e-06, + "loss": 0.5913, + "step": 2225 + }, + { + "epoch": 3.488517745302714, + "grad_norm": 1.2290774583816528, + "learning_rate": 1.9525038482810044e-06, + "loss": 0.6117, + "step": 2226 + }, + { + "epoch": 3.490083507306889, + "grad_norm": 1.1881879568099976, + "learning_rate": 1.9504484292361403e-06, + "loss": 0.6634, + "step": 2227 + }, + { + "epoch": 3.491649269311065, + "grad_norm": 1.157094955444336, + "learning_rate": 1.948393400453367e-06, + "loss": 0.608, + "step": 2228 + }, + { + "epoch": 3.49321503131524, + "grad_norm": 1.1708686351776123, + "learning_rate": 1.9463387633920543e-06, + "loss": 0.6324, + "step": 2229 + }, + { + "epoch": 3.4947807933194155, + "grad_norm": 1.1796098947525024, + "learning_rate": 1.9442845195112954e-06, + "loss": 0.6601, + "step": 2230 + }, + { + "epoch": 3.4963465553235906, + "grad_norm": 1.2861493825912476, + "learning_rate": 1.942230670269906e-06, + "loss": 0.615, + "step": 2231 + }, + { + "epoch": 3.497912317327766, + "grad_norm": 1.2033058404922485, + "learning_rate": 1.9401772171264187e-06, + "loss": 0.6148, + "step": 2232 + }, + { + "epoch": 3.4994780793319418, + "grad_norm": 1.2392219305038452, + "learning_rate": 1.938124161539085e-06, + "loss": 0.5715, + "step": 2233 + }, + { + "epoch": 3.501043841336117, + "grad_norm": 1.1681913137435913, + "learning_rate": 1.936071504965876e-06, + "loss": 0.5809, + "step": 2234 + }, + { + "epoch": 3.502609603340292, + "grad_norm": 1.2426320314407349, + "learning_rate": 1.9340192488644793e-06, + "loss": 0.6475, + "step": 2235 + }, + { + "epoch": 3.5041753653444676, + "grad_norm": 1.2066962718963623, + "learning_rate": 1.9319673946922953e-06, + "loss": 0.5845, + "step": 2236 + }, + { + "epoch": 3.505741127348643, + "grad_norm": 1.3178675174713135, + "learning_rate": 1.929915943906443e-06, + "loss": 0.6422, + "step": 2237 + }, + { + "epoch": 3.5073068893528183, + "grad_norm": 1.1881810426712036, + "learning_rate": 1.927864897963751e-06, + "loss": 0.6726, + "step": 2238 + }, + { + "epoch": 3.508872651356994, + "grad_norm": 1.130273699760437, + "learning_rate": 1.925814258320763e-06, + "loss": 0.6027, + "step": 2239 + }, + { + "epoch": 3.510438413361169, + "grad_norm": 1.1589324474334717, + "learning_rate": 1.923764026433734e-06, + "loss": 0.5668, + "step": 2240 + }, + { + "epoch": 3.5120041753653446, + "grad_norm": 1.2391080856323242, + "learning_rate": 1.921714203758629e-06, + "loss": 0.6494, + "step": 2241 + }, + { + "epoch": 3.5135699373695197, + "grad_norm": 1.2799794673919678, + "learning_rate": 1.9196647917511207e-06, + "loss": 0.6579, + "step": 2242 + }, + { + "epoch": 3.5151356993736953, + "grad_norm": 1.1656668186187744, + "learning_rate": 1.917615791866593e-06, + "loss": 0.6021, + "step": 2243 + }, + { + "epoch": 3.5167014613778704, + "grad_norm": 1.142409086227417, + "learning_rate": 1.9155672055601364e-06, + "loss": 0.6258, + "step": 2244 + }, + { + "epoch": 3.518267223382046, + "grad_norm": 1.1905341148376465, + "learning_rate": 1.9135190342865447e-06, + "loss": 0.6112, + "step": 2245 + }, + { + "epoch": 3.519832985386221, + "grad_norm": 1.143678903579712, + "learning_rate": 1.911471279500323e-06, + "loss": 0.6136, + "step": 2246 + }, + { + "epoch": 3.5213987473903967, + "grad_norm": 1.240808367729187, + "learning_rate": 1.909423942655674e-06, + "loss": 0.6403, + "step": 2247 + }, + { + "epoch": 3.522964509394572, + "grad_norm": 1.2007147073745728, + "learning_rate": 1.907377025206507e-06, + "loss": 0.6564, + "step": 2248 + }, + { + "epoch": 3.5245302713987474, + "grad_norm": 1.1846612691879272, + "learning_rate": 1.9053305286064356e-06, + "loss": 0.6839, + "step": 2249 + }, + { + "epoch": 3.526096033402923, + "grad_norm": 1.1430362462997437, + "learning_rate": 1.90328445430877e-06, + "loss": 0.6447, + "step": 2250 + }, + { + "epoch": 3.527661795407098, + "grad_norm": 1.1141353845596313, + "learning_rate": 1.9012388037665222e-06, + "loss": 0.637, + "step": 2251 + }, + { + "epoch": 3.5292275574112733, + "grad_norm": 1.329208254814148, + "learning_rate": 1.8991935784324048e-06, + "loss": 0.7307, + "step": 2252 + }, + { + "epoch": 3.530793319415449, + "grad_norm": 1.3457560539245605, + "learning_rate": 1.8971487797588276e-06, + "loss": 0.6345, + "step": 2253 + }, + { + "epoch": 3.5323590814196244, + "grad_norm": 1.1213788986206055, + "learning_rate": 1.8951044091978951e-06, + "loss": 0.5832, + "step": 2254 + }, + { + "epoch": 3.5339248434237995, + "grad_norm": 1.2542991638183594, + "learning_rate": 1.8930604682014125e-06, + "loss": 0.7679, + "step": 2255 + }, + { + "epoch": 3.535490605427975, + "grad_norm": 1.20833158493042, + "learning_rate": 1.891016958220875e-06, + "loss": 0.6436, + "step": 2256 + }, + { + "epoch": 3.5370563674321502, + "grad_norm": 1.2129275798797607, + "learning_rate": 1.8889738807074743e-06, + "loss": 0.6603, + "step": 2257 + }, + { + "epoch": 3.538622129436326, + "grad_norm": 1.2521908283233643, + "learning_rate": 1.886931237112096e-06, + "loss": 0.635, + "step": 2258 + }, + { + "epoch": 3.540187891440501, + "grad_norm": 1.248688817024231, + "learning_rate": 1.8848890288853152e-06, + "loss": 0.6358, + "step": 2259 + }, + { + "epoch": 3.5417536534446765, + "grad_norm": 1.1389580965042114, + "learning_rate": 1.882847257477398e-06, + "loss": 0.621, + "step": 2260 + }, + { + "epoch": 3.5433194154488517, + "grad_norm": 1.2111111879348755, + "learning_rate": 1.880805924338302e-06, + "loss": 0.6486, + "step": 2261 + }, + { + "epoch": 3.5448851774530272, + "grad_norm": 1.1926789283752441, + "learning_rate": 1.8787650309176735e-06, + "loss": 0.6613, + "step": 2262 + }, + { + "epoch": 3.5464509394572024, + "grad_norm": 1.144853949546814, + "learning_rate": 1.8767245786648435e-06, + "loss": 0.6353, + "step": 2263 + }, + { + "epoch": 3.548016701461378, + "grad_norm": 1.2768452167510986, + "learning_rate": 1.874684569028834e-06, + "loss": 0.6479, + "step": 2264 + }, + { + "epoch": 3.549582463465553, + "grad_norm": 1.2346423864364624, + "learning_rate": 1.8726450034583496e-06, + "loss": 0.6315, + "step": 2265 + }, + { + "epoch": 3.5511482254697286, + "grad_norm": 1.1488447189331055, + "learning_rate": 1.87060588340178e-06, + "loss": 0.6493, + "step": 2266 + }, + { + "epoch": 3.552713987473904, + "grad_norm": 1.1464084386825562, + "learning_rate": 1.8685672103072005e-06, + "loss": 0.6181, + "step": 2267 + }, + { + "epoch": 3.5542797494780793, + "grad_norm": 1.289974570274353, + "learning_rate": 1.8665289856223662e-06, + "loss": 0.6522, + "step": 2268 + }, + { + "epoch": 3.5558455114822545, + "grad_norm": 1.2859828472137451, + "learning_rate": 1.8644912107947164e-06, + "loss": 0.57, + "step": 2269 + }, + { + "epoch": 3.55741127348643, + "grad_norm": 1.176608920097351, + "learning_rate": 1.8624538872713676e-06, + "loss": 0.6546, + "step": 2270 + }, + { + "epoch": 3.5589770354906056, + "grad_norm": 1.335280418395996, + "learning_rate": 1.86041701649912e-06, + "loss": 0.6659, + "step": 2271 + }, + { + "epoch": 3.5605427974947808, + "grad_norm": 1.2693639993667603, + "learning_rate": 1.858380599924449e-06, + "loss": 0.6348, + "step": 2272 + }, + { + "epoch": 3.562108559498956, + "grad_norm": 1.1791412830352783, + "learning_rate": 1.856344638993508e-06, + "loss": 0.6755, + "step": 2273 + }, + { + "epoch": 3.5636743215031315, + "grad_norm": 1.1750035285949707, + "learning_rate": 1.8543091351521292e-06, + "loss": 0.6508, + "step": 2274 + }, + { + "epoch": 3.565240083507307, + "grad_norm": 1.1812372207641602, + "learning_rate": 1.8522740898458178e-06, + "loss": 0.5811, + "step": 2275 + }, + { + "epoch": 3.566805845511482, + "grad_norm": 1.1619631052017212, + "learning_rate": 1.8502395045197522e-06, + "loss": 0.6227, + "step": 2276 + }, + { + "epoch": 3.5683716075156577, + "grad_norm": 1.2588560581207275, + "learning_rate": 1.8482053806187878e-06, + "loss": 0.6806, + "step": 2277 + }, + { + "epoch": 3.569937369519833, + "grad_norm": 1.0842543840408325, + "learning_rate": 1.8461717195874504e-06, + "loss": 0.6315, + "step": 2278 + }, + { + "epoch": 3.5715031315240084, + "grad_norm": 1.1269354820251465, + "learning_rate": 1.8441385228699355e-06, + "loss": 0.6387, + "step": 2279 + }, + { + "epoch": 3.5730688935281836, + "grad_norm": 1.196655511856079, + "learning_rate": 1.8421057919101128e-06, + "loss": 0.6872, + "step": 2280 + }, + { + "epoch": 3.574634655532359, + "grad_norm": 1.3561420440673828, + "learning_rate": 1.8400735281515175e-06, + "loss": 0.6538, + "step": 2281 + }, + { + "epoch": 3.5762004175365343, + "grad_norm": 1.1209763288497925, + "learning_rate": 1.838041733037354e-06, + "loss": 0.6934, + "step": 2282 + }, + { + "epoch": 3.57776617954071, + "grad_norm": 1.0936529636383057, + "learning_rate": 1.8360104080104962e-06, + "loss": 0.7136, + "step": 2283 + }, + { + "epoch": 3.5793319415448854, + "grad_norm": 1.2066773176193237, + "learning_rate": 1.8339795545134814e-06, + "loss": 0.6259, + "step": 2284 + }, + { + "epoch": 3.5808977035490606, + "grad_norm": 1.0662962198257446, + "learning_rate": 1.8319491739885115e-06, + "loss": 0.6195, + "step": 2285 + }, + { + "epoch": 3.5824634655532357, + "grad_norm": 1.284588098526001, + "learning_rate": 1.8299192678774564e-06, + "loss": 0.6154, + "step": 2286 + }, + { + "epoch": 3.5840292275574113, + "grad_norm": 1.1617132425308228, + "learning_rate": 1.8278898376218456e-06, + "loss": 0.6574, + "step": 2287 + }, + { + "epoch": 3.585594989561587, + "grad_norm": 1.1630313396453857, + "learning_rate": 1.8258608846628708e-06, + "loss": 0.577, + "step": 2288 + }, + { + "epoch": 3.587160751565762, + "grad_norm": 1.1374728679656982, + "learning_rate": 1.8238324104413879e-06, + "loss": 0.6908, + "step": 2289 + }, + { + "epoch": 3.588726513569937, + "grad_norm": 1.1832219362258911, + "learning_rate": 1.8218044163979088e-06, + "loss": 0.6409, + "step": 2290 + }, + { + "epoch": 3.5902922755741127, + "grad_norm": 1.4350699186325073, + "learning_rate": 1.8197769039726065e-06, + "loss": 0.6043, + "step": 2291 + }, + { + "epoch": 3.5918580375782883, + "grad_norm": 1.0792481899261475, + "learning_rate": 1.8177498746053129e-06, + "loss": 0.5953, + "step": 2292 + }, + { + "epoch": 3.5934237995824634, + "grad_norm": 1.2297285795211792, + "learning_rate": 1.8157233297355154e-06, + "loss": 0.646, + "step": 2293 + }, + { + "epoch": 3.594989561586639, + "grad_norm": 1.1795411109924316, + "learning_rate": 1.8136972708023557e-06, + "loss": 0.6187, + "step": 2294 + }, + { + "epoch": 3.596555323590814, + "grad_norm": 1.2370169162750244, + "learning_rate": 1.8116716992446342e-06, + "loss": 0.6899, + "step": 2295 + }, + { + "epoch": 3.5981210855949897, + "grad_norm": 1.1697649955749512, + "learning_rate": 1.8096466165008037e-06, + "loss": 0.6495, + "step": 2296 + }, + { + "epoch": 3.599686847599165, + "grad_norm": 1.1952897310256958, + "learning_rate": 1.807622024008967e-06, + "loss": 0.6007, + "step": 2297 + }, + { + "epoch": 3.6012526096033404, + "grad_norm": 1.2342556715011597, + "learning_rate": 1.8055979232068845e-06, + "loss": 0.5842, + "step": 2298 + }, + { + "epoch": 3.6028183716075155, + "grad_norm": 1.2582154273986816, + "learning_rate": 1.8035743155319618e-06, + "loss": 0.6939, + "step": 2299 + }, + { + "epoch": 3.604384133611691, + "grad_norm": 1.1596347093582153, + "learning_rate": 1.8015512024212573e-06, + "loss": 0.6609, + "step": 2300 + }, + { + "epoch": 3.6059498956158667, + "grad_norm": 1.2669062614440918, + "learning_rate": 1.7995285853114785e-06, + "loss": 0.6747, + "step": 2301 + }, + { + "epoch": 3.607515657620042, + "grad_norm": 1.2295914888381958, + "learning_rate": 1.797506465638979e-06, + "loss": 0.6374, + "step": 2302 + }, + { + "epoch": 3.609081419624217, + "grad_norm": 1.2566529512405396, + "learning_rate": 1.795484844839759e-06, + "loss": 0.6034, + "step": 2303 + }, + { + "epoch": 3.6106471816283925, + "grad_norm": 1.2655516862869263, + "learning_rate": 1.7934637243494663e-06, + "loss": 0.6394, + "step": 2304 + }, + { + "epoch": 3.612212943632568, + "grad_norm": 1.2168526649475098, + "learning_rate": 1.7914431056033927e-06, + "loss": 0.639, + "step": 2305 + }, + { + "epoch": 3.613778705636743, + "grad_norm": 1.309919834136963, + "learning_rate": 1.7894229900364721e-06, + "loss": 0.648, + "step": 2306 + }, + { + "epoch": 3.6153444676409183, + "grad_norm": 1.3890948295593262, + "learning_rate": 1.7874033790832846e-06, + "loss": 0.5951, + "step": 2307 + }, + { + "epoch": 3.616910229645094, + "grad_norm": 1.1613552570343018, + "learning_rate": 1.7853842741780474e-06, + "loss": 0.6083, + "step": 2308 + }, + { + "epoch": 3.6184759916492695, + "grad_norm": 1.3169668912887573, + "learning_rate": 1.783365676754621e-06, + "loss": 0.6439, + "step": 2309 + }, + { + "epoch": 3.6200417536534446, + "grad_norm": 1.21388578414917, + "learning_rate": 1.781347588246507e-06, + "loss": 0.6702, + "step": 2310 + }, + { + "epoch": 3.62160751565762, + "grad_norm": 1.202735185623169, + "learning_rate": 1.7793300100868428e-06, + "loss": 0.6417, + "step": 2311 + }, + { + "epoch": 3.6231732776617953, + "grad_norm": 1.2919602394104004, + "learning_rate": 1.7773129437084028e-06, + "loss": 0.6139, + "step": 2312 + }, + { + "epoch": 3.624739039665971, + "grad_norm": 1.1894437074661255, + "learning_rate": 1.7752963905436014e-06, + "loss": 0.6689, + "step": 2313 + }, + { + "epoch": 3.626304801670146, + "grad_norm": 1.1630206108093262, + "learning_rate": 1.7732803520244866e-06, + "loss": 0.6381, + "step": 2314 + }, + { + "epoch": 3.6278705636743216, + "grad_norm": 1.2039670944213867, + "learning_rate": 1.7712648295827395e-06, + "loss": 0.6972, + "step": 2315 + }, + { + "epoch": 3.6294363256784967, + "grad_norm": 1.2051432132720947, + "learning_rate": 1.769249824649677e-06, + "loss": 0.6404, + "step": 2316 + }, + { + "epoch": 3.6310020876826723, + "grad_norm": 1.2452560663223267, + "learning_rate": 1.7672353386562486e-06, + "loss": 0.6057, + "step": 2317 + }, + { + "epoch": 3.632567849686848, + "grad_norm": 1.1576342582702637, + "learning_rate": 1.7652213730330336e-06, + "loss": 0.6235, + "step": 2318 + }, + { + "epoch": 3.634133611691023, + "grad_norm": 1.3302104473114014, + "learning_rate": 1.7632079292102412e-06, + "loss": 0.6464, + "step": 2319 + }, + { + "epoch": 3.635699373695198, + "grad_norm": 1.1526870727539062, + "learning_rate": 1.7611950086177132e-06, + "loss": 0.6619, + "step": 2320 + }, + { + "epoch": 3.6372651356993737, + "grad_norm": 1.1325047016143799, + "learning_rate": 1.7591826126849179e-06, + "loss": 0.6815, + "step": 2321 + }, + { + "epoch": 3.6388308977035493, + "grad_norm": 1.2510312795639038, + "learning_rate": 1.7571707428409496e-06, + "loss": 0.6344, + "step": 2322 + }, + { + "epoch": 3.6403966597077244, + "grad_norm": 1.3015072345733643, + "learning_rate": 1.7551594005145323e-06, + "loss": 0.6082, + "step": 2323 + }, + { + "epoch": 3.6419624217118995, + "grad_norm": 1.1332790851593018, + "learning_rate": 1.7531485871340122e-06, + "loss": 0.6854, + "step": 2324 + }, + { + "epoch": 3.643528183716075, + "grad_norm": 1.1714104413986206, + "learning_rate": 1.7511383041273617e-06, + "loss": 0.6497, + "step": 2325 + }, + { + "epoch": 3.6450939457202507, + "grad_norm": 1.1853827238082886, + "learning_rate": 1.7491285529221768e-06, + "loss": 0.5655, + "step": 2326 + }, + { + "epoch": 3.646659707724426, + "grad_norm": 1.2981195449829102, + "learning_rate": 1.7471193349456746e-06, + "loss": 0.6327, + "step": 2327 + }, + { + "epoch": 3.6482254697286014, + "grad_norm": 1.2444530725479126, + "learning_rate": 1.7451106516246932e-06, + "loss": 0.643, + "step": 2328 + }, + { + "epoch": 3.6497912317327765, + "grad_norm": 1.1355667114257812, + "learning_rate": 1.7431025043856931e-06, + "loss": 0.6439, + "step": 2329 + }, + { + "epoch": 3.651356993736952, + "grad_norm": 1.1224238872528076, + "learning_rate": 1.7410948946547536e-06, + "loss": 0.6606, + "step": 2330 + }, + { + "epoch": 3.6529227557411272, + "grad_norm": 1.163265347480774, + "learning_rate": 1.739087823857569e-06, + "loss": 0.5989, + "step": 2331 + }, + { + "epoch": 3.654488517745303, + "grad_norm": 1.3157579898834229, + "learning_rate": 1.7370812934194565e-06, + "loss": 0.6818, + "step": 2332 + }, + { + "epoch": 3.656054279749478, + "grad_norm": 1.1727818250656128, + "learning_rate": 1.735075304765344e-06, + "loss": 0.6271, + "step": 2333 + }, + { + "epoch": 3.6576200417536535, + "grad_norm": 1.2371114492416382, + "learning_rate": 1.733069859319778e-06, + "loss": 0.667, + "step": 2334 + }, + { + "epoch": 3.6591858037578286, + "grad_norm": 1.210724115371704, + "learning_rate": 1.73106495850692e-06, + "loss": 0.725, + "step": 2335 + }, + { + "epoch": 3.6607515657620042, + "grad_norm": 1.1199427843093872, + "learning_rate": 1.7290606037505417e-06, + "loss": 0.6634, + "step": 2336 + }, + { + "epoch": 3.6623173277661794, + "grad_norm": 1.2661418914794922, + "learning_rate": 1.7270567964740275e-06, + "loss": 0.6259, + "step": 2337 + }, + { + "epoch": 3.663883089770355, + "grad_norm": 1.2222251892089844, + "learning_rate": 1.7250535381003758e-06, + "loss": 0.6082, + "step": 2338 + }, + { + "epoch": 3.6654488517745305, + "grad_norm": 1.1748756170272827, + "learning_rate": 1.7230508300521935e-06, + "loss": 0.6017, + "step": 2339 + }, + { + "epoch": 3.6670146137787056, + "grad_norm": 1.2061327695846558, + "learning_rate": 1.7210486737516947e-06, + "loss": 0.6196, + "step": 2340 + }, + { + "epoch": 3.6685803757828808, + "grad_norm": 1.168503999710083, + "learning_rate": 1.7190470706207058e-06, + "loss": 0.6709, + "step": 2341 + }, + { + "epoch": 3.6701461377870563, + "grad_norm": 1.1397614479064941, + "learning_rate": 1.7170460220806563e-06, + "loss": 0.6043, + "step": 2342 + }, + { + "epoch": 3.671711899791232, + "grad_norm": 1.1928285360336304, + "learning_rate": 1.7150455295525842e-06, + "loss": 0.5985, + "step": 2343 + }, + { + "epoch": 3.673277661795407, + "grad_norm": 1.205994725227356, + "learning_rate": 1.713045594457133e-06, + "loss": 0.6461, + "step": 2344 + }, + { + "epoch": 3.6748434237995826, + "grad_norm": 1.192114233970642, + "learning_rate": 1.7110462182145494e-06, + "loss": 0.6461, + "step": 2345 + }, + { + "epoch": 3.6764091858037578, + "grad_norm": 1.2672173976898193, + "learning_rate": 1.7090474022446812e-06, + "loss": 0.6623, + "step": 2346 + }, + { + "epoch": 3.6779749478079333, + "grad_norm": 1.1894220113754272, + "learning_rate": 1.707049147966983e-06, + "loss": 0.6687, + "step": 2347 + }, + { + "epoch": 3.6795407098121085, + "grad_norm": 1.170541763305664, + "learning_rate": 1.7050514568005072e-06, + "loss": 0.6258, + "step": 2348 + }, + { + "epoch": 3.681106471816284, + "grad_norm": 1.2120347023010254, + "learning_rate": 1.7030543301639057e-06, + "loss": 0.5854, + "step": 2349 + }, + { + "epoch": 3.682672233820459, + "grad_norm": 1.1434928178787231, + "learning_rate": 1.7010577694754332e-06, + "loss": 0.6177, + "step": 2350 + }, + { + "epoch": 3.6842379958246347, + "grad_norm": 1.2849304676055908, + "learning_rate": 1.6990617761529388e-06, + "loss": 0.5642, + "step": 2351 + }, + { + "epoch": 3.68580375782881, + "grad_norm": 1.1992167234420776, + "learning_rate": 1.6970663516138692e-06, + "loss": 0.6594, + "step": 2352 + }, + { + "epoch": 3.6873695198329854, + "grad_norm": 1.2290526628494263, + "learning_rate": 1.6950714972752702e-06, + "loss": 0.6506, + "step": 2353 + }, + { + "epoch": 3.6889352818371606, + "grad_norm": 1.1701496839523315, + "learning_rate": 1.6930772145537794e-06, + "loss": 0.5947, + "step": 2354 + }, + { + "epoch": 3.690501043841336, + "grad_norm": 1.2298767566680908, + "learning_rate": 1.6910835048656282e-06, + "loss": 0.6824, + "step": 2355 + }, + { + "epoch": 3.6920668058455117, + "grad_norm": 1.1752413511276245, + "learning_rate": 1.6890903696266447e-06, + "loss": 0.5763, + "step": 2356 + }, + { + "epoch": 3.693632567849687, + "grad_norm": 1.174170732498169, + "learning_rate": 1.6870978102522462e-06, + "loss": 0.5934, + "step": 2357 + }, + { + "epoch": 3.695198329853862, + "grad_norm": 1.1316381692886353, + "learning_rate": 1.68510582815744e-06, + "loss": 0.5857, + "step": 2358 + }, + { + "epoch": 3.6967640918580376, + "grad_norm": 1.2187336683273315, + "learning_rate": 1.6831144247568274e-06, + "loss": 0.6832, + "step": 2359 + }, + { + "epoch": 3.698329853862213, + "grad_norm": 1.2257745265960693, + "learning_rate": 1.6811236014645949e-06, + "loss": 0.6001, + "step": 2360 + }, + { + "epoch": 3.6998956158663883, + "grad_norm": 1.2153297662734985, + "learning_rate": 1.6791333596945197e-06, + "loss": 0.6452, + "step": 2361 + }, + { + "epoch": 3.701461377870564, + "grad_norm": 1.203791618347168, + "learning_rate": 1.677143700859963e-06, + "loss": 0.6032, + "step": 2362 + }, + { + "epoch": 3.703027139874739, + "grad_norm": 1.215394139289856, + "learning_rate": 1.6751546263738761e-06, + "loss": 0.694, + "step": 2363 + }, + { + "epoch": 3.7045929018789145, + "grad_norm": 1.1475002765655518, + "learning_rate": 1.6731661376487923e-06, + "loss": 0.6627, + "step": 2364 + }, + { + "epoch": 3.7061586638830897, + "grad_norm": 1.2240904569625854, + "learning_rate": 1.6711782360968288e-06, + "loss": 0.6605, + "step": 2365 + }, + { + "epoch": 3.7077244258872653, + "grad_norm": 1.1701643466949463, + "learning_rate": 1.6691909231296888e-06, + "loss": 0.5705, + "step": 2366 + }, + { + "epoch": 3.7092901878914404, + "grad_norm": 1.1641205549240112, + "learning_rate": 1.6672042001586538e-06, + "loss": 0.6604, + "step": 2367 + }, + { + "epoch": 3.710855949895616, + "grad_norm": 1.120191216468811, + "learning_rate": 1.6652180685945884e-06, + "loss": 0.567, + "step": 2368 + }, + { + "epoch": 3.712421711899791, + "grad_norm": 1.1677566766738892, + "learning_rate": 1.663232529847938e-06, + "loss": 0.6484, + "step": 2369 + }, + { + "epoch": 3.7139874739039667, + "grad_norm": 1.1286911964416504, + "learning_rate": 1.6612475853287252e-06, + "loss": 0.5604, + "step": 2370 + }, + { + "epoch": 3.715553235908142, + "grad_norm": 1.2938463687896729, + "learning_rate": 1.65926323644655e-06, + "loss": 0.683, + "step": 2371 + }, + { + "epoch": 3.7171189979123174, + "grad_norm": 1.160536527633667, + "learning_rate": 1.6572794846105919e-06, + "loss": 0.6467, + "step": 2372 + }, + { + "epoch": 3.718684759916493, + "grad_norm": 1.1587703227996826, + "learning_rate": 1.655296331229606e-06, + "loss": 0.6386, + "step": 2373 + }, + { + "epoch": 3.720250521920668, + "grad_norm": 1.2075384855270386, + "learning_rate": 1.6533137777119196e-06, + "loss": 0.6331, + "step": 2374 + }, + { + "epoch": 3.721816283924843, + "grad_norm": 1.2803996801376343, + "learning_rate": 1.651331825465438e-06, + "loss": 0.6421, + "step": 2375 + }, + { + "epoch": 3.723382045929019, + "grad_norm": 1.1343461275100708, + "learning_rate": 1.6493504758976358e-06, + "loss": 0.6807, + "step": 2376 + }, + { + "epoch": 3.7249478079331944, + "grad_norm": 1.211895227432251, + "learning_rate": 1.6473697304155622e-06, + "loss": 0.6533, + "step": 2377 + }, + { + "epoch": 3.7265135699373695, + "grad_norm": 1.191689372062683, + "learning_rate": 1.6453895904258372e-06, + "loss": 0.6559, + "step": 2378 + }, + { + "epoch": 3.7280793319415446, + "grad_norm": 1.2990779876708984, + "learning_rate": 1.6434100573346496e-06, + "loss": 0.6354, + "step": 2379 + }, + { + "epoch": 3.72964509394572, + "grad_norm": 1.2150678634643555, + "learning_rate": 1.6414311325477567e-06, + "loss": 0.6438, + "step": 2380 + }, + { + "epoch": 3.7312108559498958, + "grad_norm": 1.1220753192901611, + "learning_rate": 1.639452817470486e-06, + "loss": 0.6315, + "step": 2381 + }, + { + "epoch": 3.732776617954071, + "grad_norm": 1.096580147743225, + "learning_rate": 1.6374751135077315e-06, + "loss": 0.6525, + "step": 2382 + }, + { + "epoch": 3.7343423799582465, + "grad_norm": 1.2590006589889526, + "learning_rate": 1.6354980220639504e-06, + "loss": 0.6609, + "step": 2383 + }, + { + "epoch": 3.7359081419624216, + "grad_norm": 1.2841553688049316, + "learning_rate": 1.6335215445431702e-06, + "loss": 0.6861, + "step": 2384 + }, + { + "epoch": 3.737473903966597, + "grad_norm": 1.1383644342422485, + "learning_rate": 1.6315456823489766e-06, + "loss": 0.6707, + "step": 2385 + }, + { + "epoch": 3.7390396659707723, + "grad_norm": 1.3325951099395752, + "learning_rate": 1.6295704368845221e-06, + "loss": 0.5672, + "step": 2386 + }, + { + "epoch": 3.740605427974948, + "grad_norm": 1.2384133338928223, + "learning_rate": 1.6275958095525217e-06, + "loss": 0.6274, + "step": 2387 + }, + { + "epoch": 3.742171189979123, + "grad_norm": 1.173836588859558, + "learning_rate": 1.6256218017552484e-06, + "loss": 0.5975, + "step": 2388 + }, + { + "epoch": 3.7437369519832986, + "grad_norm": 1.1901642084121704, + "learning_rate": 1.623648414894536e-06, + "loss": 0.6735, + "step": 2389 + }, + { + "epoch": 3.745302713987474, + "grad_norm": 1.2956198453903198, + "learning_rate": 1.6216756503717795e-06, + "loss": 0.6171, + "step": 2390 + }, + { + "epoch": 3.7468684759916493, + "grad_norm": 1.173884391784668, + "learning_rate": 1.619703509587931e-06, + "loss": 0.6375, + "step": 2391 + }, + { + "epoch": 3.7484342379958244, + "grad_norm": 1.1478065252304077, + "learning_rate": 1.6177319939434974e-06, + "loss": 0.6806, + "step": 2392 + }, + { + "epoch": 3.75, + "grad_norm": 1.1560224294662476, + "learning_rate": 1.6157611048385457e-06, + "loss": 0.5867, + "step": 2393 + }, + { + "epoch": 3.7515657620041756, + "grad_norm": 1.1529253721237183, + "learning_rate": 1.6137908436726943e-06, + "loss": 0.686, + "step": 2394 + }, + { + "epoch": 3.7531315240083507, + "grad_norm": 1.2027194499969482, + "learning_rate": 1.6118212118451171e-06, + "loss": 0.6579, + "step": 2395 + }, + { + "epoch": 3.754697286012526, + "grad_norm": 1.1857298612594604, + "learning_rate": 1.6098522107545426e-06, + "loss": 0.6274, + "step": 2396 + }, + { + "epoch": 3.7562630480167014, + "grad_norm": 1.2853041887283325, + "learning_rate": 1.6078838417992487e-06, + "loss": 0.6866, + "step": 2397 + }, + { + "epoch": 3.757828810020877, + "grad_norm": 1.1234532594680786, + "learning_rate": 1.6059161063770651e-06, + "loss": 0.5872, + "step": 2398 + }, + { + "epoch": 3.759394572025052, + "grad_norm": 1.2463502883911133, + "learning_rate": 1.603949005885373e-06, + "loss": 0.6566, + "step": 2399 + }, + { + "epoch": 3.7609603340292277, + "grad_norm": 1.1617919206619263, + "learning_rate": 1.6019825417211022e-06, + "loss": 0.6428, + "step": 2400 + }, + { + "epoch": 3.762526096033403, + "grad_norm": 1.1768699884414673, + "learning_rate": 1.6000167152807288e-06, + "loss": 0.6147, + "step": 2401 + }, + { + "epoch": 3.7640918580375784, + "grad_norm": 1.2775728702545166, + "learning_rate": 1.598051527960279e-06, + "loss": 0.6594, + "step": 2402 + }, + { + "epoch": 3.7656576200417535, + "grad_norm": 1.2250229120254517, + "learning_rate": 1.5960869811553225e-06, + "loss": 0.6844, + "step": 2403 + }, + { + "epoch": 3.767223382045929, + "grad_norm": 1.3038227558135986, + "learning_rate": 1.594123076260975e-06, + "loss": 0.6515, + "step": 2404 + }, + { + "epoch": 3.7687891440501042, + "grad_norm": 1.2010571956634521, + "learning_rate": 1.592159814671898e-06, + "loss": 0.6293, + "step": 2405 + }, + { + "epoch": 3.77035490605428, + "grad_norm": 1.2710856199264526, + "learning_rate": 1.590197197782294e-06, + "loss": 0.6096, + "step": 2406 + }, + { + "epoch": 3.7719206680584554, + "grad_norm": 1.2024561166763306, + "learning_rate": 1.588235226985908e-06, + "loss": 0.6227, + "step": 2407 + }, + { + "epoch": 3.7734864300626305, + "grad_norm": 1.1803232431411743, + "learning_rate": 1.5862739036760262e-06, + "loss": 0.6251, + "step": 2408 + }, + { + "epoch": 3.7750521920668056, + "grad_norm": 1.2758967876434326, + "learning_rate": 1.5843132292454765e-06, + "loss": 0.6948, + "step": 2409 + }, + { + "epoch": 3.776617954070981, + "grad_norm": 1.2238482236862183, + "learning_rate": 1.5823532050866242e-06, + "loss": 0.6186, + "step": 2410 + }, + { + "epoch": 3.778183716075157, + "grad_norm": 1.24042546749115, + "learning_rate": 1.5803938325913732e-06, + "loss": 0.6261, + "step": 2411 + }, + { + "epoch": 3.779749478079332, + "grad_norm": 1.137797474861145, + "learning_rate": 1.578435113151166e-06, + "loss": 0.5794, + "step": 2412 + }, + { + "epoch": 3.781315240083507, + "grad_norm": 1.2196292877197266, + "learning_rate": 1.5764770481569796e-06, + "loss": 0.64, + "step": 2413 + }, + { + "epoch": 3.7828810020876826, + "grad_norm": 1.1684080362319946, + "learning_rate": 1.5745196389993259e-06, + "loss": 0.5861, + "step": 2414 + }, + { + "epoch": 3.784446764091858, + "grad_norm": 1.2441654205322266, + "learning_rate": 1.5725628870682532e-06, + "loss": 0.6084, + "step": 2415 + }, + { + "epoch": 3.7860125260960333, + "grad_norm": 1.187728762626648, + "learning_rate": 1.5706067937533427e-06, + "loss": 0.6436, + "step": 2416 + }, + { + "epoch": 3.787578288100209, + "grad_norm": 1.2837245464324951, + "learning_rate": 1.5686513604437043e-06, + "loss": 0.5967, + "step": 2417 + }, + { + "epoch": 3.789144050104384, + "grad_norm": 1.1892480850219727, + "learning_rate": 1.566696588527985e-06, + "loss": 0.6547, + "step": 2418 + }, + { + "epoch": 3.7907098121085596, + "grad_norm": 1.2308993339538574, + "learning_rate": 1.5647424793943579e-06, + "loss": 0.6512, + "step": 2419 + }, + { + "epoch": 3.7922755741127347, + "grad_norm": 1.3206483125686646, + "learning_rate": 1.5627890344305256e-06, + "loss": 0.6845, + "step": 2420 + }, + { + "epoch": 3.7938413361169103, + "grad_norm": 1.151696801185608, + "learning_rate": 1.560836255023722e-06, + "loss": 0.597, + "step": 2421 + }, + { + "epoch": 3.7954070981210855, + "grad_norm": 1.2277007102966309, + "learning_rate": 1.5588841425607056e-06, + "loss": 0.6035, + "step": 2422 + }, + { + "epoch": 3.796972860125261, + "grad_norm": 1.2150050401687622, + "learning_rate": 1.556932698427761e-06, + "loss": 0.6679, + "step": 2423 + }, + { + "epoch": 3.798538622129436, + "grad_norm": 1.075766921043396, + "learning_rate": 1.5549819240107012e-06, + "loss": 0.6112, + "step": 2424 + }, + { + "epoch": 3.8001043841336117, + "grad_norm": 1.2026962041854858, + "learning_rate": 1.5530318206948607e-06, + "loss": 0.6104, + "step": 2425 + }, + { + "epoch": 3.801670146137787, + "grad_norm": 1.2209258079528809, + "learning_rate": 1.551082389865098e-06, + "loss": 0.6391, + "step": 2426 + }, + { + "epoch": 3.8032359081419624, + "grad_norm": 1.1685758829116821, + "learning_rate": 1.549133632905796e-06, + "loss": 0.6352, + "step": 2427 + }, + { + "epoch": 3.804801670146138, + "grad_norm": 1.1990925073623657, + "learning_rate": 1.547185551200856e-06, + "loss": 0.631, + "step": 2428 + }, + { + "epoch": 3.806367432150313, + "grad_norm": 1.1937427520751953, + "learning_rate": 1.5452381461337013e-06, + "loss": 0.6393, + "step": 2429 + }, + { + "epoch": 3.8079331941544883, + "grad_norm": 1.1711888313293457, + "learning_rate": 1.5432914190872757e-06, + "loss": 0.6415, + "step": 2430 + }, + { + "epoch": 3.809498956158664, + "grad_norm": 1.3688654899597168, + "learning_rate": 1.5413453714440404e-06, + "loss": 0.6711, + "step": 2431 + }, + { + "epoch": 3.8110647181628394, + "grad_norm": 1.1071898937225342, + "learning_rate": 1.5394000045859728e-06, + "loss": 0.5966, + "step": 2432 + }, + { + "epoch": 3.8126304801670146, + "grad_norm": 1.1659204959869385, + "learning_rate": 1.5374553198945696e-06, + "loss": 0.6672, + "step": 2433 + }, + { + "epoch": 3.81419624217119, + "grad_norm": 1.1961894035339355, + "learning_rate": 1.5355113187508413e-06, + "loss": 0.6229, + "step": 2434 + }, + { + "epoch": 3.8157620041753653, + "grad_norm": 1.2803820371627808, + "learning_rate": 1.5335680025353132e-06, + "loss": 0.657, + "step": 2435 + }, + { + "epoch": 3.817327766179541, + "grad_norm": 1.1563304662704468, + "learning_rate": 1.531625372628025e-06, + "loss": 0.6386, + "step": 2436 + }, + { + "epoch": 3.818893528183716, + "grad_norm": 1.2539798021316528, + "learning_rate": 1.5296834304085276e-06, + "loss": 0.5733, + "step": 2437 + }, + { + "epoch": 3.8204592901878915, + "grad_norm": 1.2615854740142822, + "learning_rate": 1.5277421772558842e-06, + "loss": 0.6073, + "step": 2438 + }, + { + "epoch": 3.8220250521920667, + "grad_norm": 1.2182807922363281, + "learning_rate": 1.5258016145486705e-06, + "loss": 0.6719, + "step": 2439 + }, + { + "epoch": 3.8235908141962422, + "grad_norm": 1.2267918586730957, + "learning_rate": 1.5238617436649689e-06, + "loss": 0.616, + "step": 2440 + }, + { + "epoch": 3.8251565762004174, + "grad_norm": 1.2033759355545044, + "learning_rate": 1.5219225659823706e-06, + "loss": 0.6759, + "step": 2441 + }, + { + "epoch": 3.826722338204593, + "grad_norm": 1.3071197271347046, + "learning_rate": 1.5199840828779777e-06, + "loss": 0.6672, + "step": 2442 + }, + { + "epoch": 3.828288100208768, + "grad_norm": 1.2482445240020752, + "learning_rate": 1.5180462957283971e-06, + "loss": 0.6363, + "step": 2443 + }, + { + "epoch": 3.8298538622129437, + "grad_norm": 1.2446298599243164, + "learning_rate": 1.5161092059097399e-06, + "loss": 0.6506, + "step": 2444 + }, + { + "epoch": 3.8314196242171192, + "grad_norm": 1.217212438583374, + "learning_rate": 1.5141728147976248e-06, + "loss": 0.7131, + "step": 2445 + }, + { + "epoch": 3.8329853862212944, + "grad_norm": 1.2086701393127441, + "learning_rate": 1.5122371237671726e-06, + "loss": 0.6107, + "step": 2446 + }, + { + "epoch": 3.8345511482254695, + "grad_norm": 1.244437336921692, + "learning_rate": 1.510302134193007e-06, + "loss": 0.5585, + "step": 2447 + }, + { + "epoch": 3.836116910229645, + "grad_norm": 1.1509884595870972, + "learning_rate": 1.5083678474492546e-06, + "loss": 0.5905, + "step": 2448 + }, + { + "epoch": 3.8376826722338206, + "grad_norm": 1.188522458076477, + "learning_rate": 1.5064342649095428e-06, + "loss": 0.6584, + "step": 2449 + }, + { + "epoch": 3.8392484342379958, + "grad_norm": 1.2903960943222046, + "learning_rate": 1.504501387946996e-06, + "loss": 0.6286, + "step": 2450 + }, + { + "epoch": 3.8408141962421714, + "grad_norm": 1.135545253753662, + "learning_rate": 1.5025692179342425e-06, + "loss": 0.6769, + "step": 2451 + }, + { + "epoch": 3.8423799582463465, + "grad_norm": 1.2375710010528564, + "learning_rate": 1.500637756243405e-06, + "loss": 0.6506, + "step": 2452 + }, + { + "epoch": 3.843945720250522, + "grad_norm": 1.7769895792007446, + "learning_rate": 1.4987070042461032e-06, + "loss": 0.6302, + "step": 2453 + }, + { + "epoch": 3.845511482254697, + "grad_norm": 1.0754798650741577, + "learning_rate": 1.4967769633134555e-06, + "loss": 0.6019, + "step": 2454 + }, + { + "epoch": 3.8470772442588728, + "grad_norm": 1.2362276315689087, + "learning_rate": 1.4948476348160723e-06, + "loss": 0.578, + "step": 2455 + }, + { + "epoch": 3.848643006263048, + "grad_norm": 1.1476119756698608, + "learning_rate": 1.4929190201240601e-06, + "loss": 0.5866, + "step": 2456 + }, + { + "epoch": 3.8502087682672235, + "grad_norm": 1.14879310131073, + "learning_rate": 1.4909911206070162e-06, + "loss": 0.6431, + "step": 2457 + }, + { + "epoch": 3.8517745302713986, + "grad_norm": 1.1493771076202393, + "learning_rate": 1.4890639376340336e-06, + "loss": 0.6297, + "step": 2458 + }, + { + "epoch": 3.853340292275574, + "grad_norm": 1.2373193502426147, + "learning_rate": 1.4871374725736936e-06, + "loss": 0.6387, + "step": 2459 + }, + { + "epoch": 3.8549060542797493, + "grad_norm": 1.2413831949234009, + "learning_rate": 1.485211726794068e-06, + "loss": 0.6569, + "step": 2460 + }, + { + "epoch": 3.856471816283925, + "grad_norm": 1.2246074676513672, + "learning_rate": 1.4832867016627194e-06, + "loss": 0.5908, + "step": 2461 + }, + { + "epoch": 3.8580375782881005, + "grad_norm": 1.1326402425765991, + "learning_rate": 1.4813623985466968e-06, + "loss": 0.6488, + "step": 2462 + }, + { + "epoch": 3.8596033402922756, + "grad_norm": 1.2424553632736206, + "learning_rate": 1.4794388188125364e-06, + "loss": 0.6299, + "step": 2463 + }, + { + "epoch": 3.8611691022964507, + "grad_norm": 1.2312356233596802, + "learning_rate": 1.4775159638262637e-06, + "loss": 0.6359, + "step": 2464 + }, + { + "epoch": 3.8627348643006263, + "grad_norm": 1.049404263496399, + "learning_rate": 1.475593834953386e-06, + "loss": 0.5736, + "step": 2465 + }, + { + "epoch": 3.864300626304802, + "grad_norm": 1.2294747829437256, + "learning_rate": 1.4736724335588952e-06, + "loss": 0.6342, + "step": 2466 + }, + { + "epoch": 3.865866388308977, + "grad_norm": 1.2269598245620728, + "learning_rate": 1.471751761007269e-06, + "loss": 0.6316, + "step": 2467 + }, + { + "epoch": 3.867432150313152, + "grad_norm": 1.138282299041748, + "learning_rate": 1.469831818662467e-06, + "loss": 0.6402, + "step": 2468 + }, + { + "epoch": 3.8689979123173277, + "grad_norm": 1.2380733489990234, + "learning_rate": 1.467912607887927e-06, + "loss": 0.639, + "step": 2469 + }, + { + "epoch": 3.8705636743215033, + "grad_norm": 1.109584093093872, + "learning_rate": 1.465994130046572e-06, + "loss": 0.6178, + "step": 2470 + }, + { + "epoch": 3.8721294363256784, + "grad_norm": 1.3927913904190063, + "learning_rate": 1.4640763865008006e-06, + "loss": 0.5844, + "step": 2471 + }, + { + "epoch": 3.873695198329854, + "grad_norm": 1.1995818614959717, + "learning_rate": 1.462159378612492e-06, + "loss": 0.6532, + "step": 2472 + }, + { + "epoch": 3.875260960334029, + "grad_norm": 1.244868278503418, + "learning_rate": 1.4602431077430035e-06, + "loss": 0.5823, + "step": 2473 + }, + { + "epoch": 3.8768267223382047, + "grad_norm": 1.2030248641967773, + "learning_rate": 1.4583275752531672e-06, + "loss": 0.6096, + "step": 2474 + }, + { + "epoch": 3.87839248434238, + "grad_norm": 1.1157633066177368, + "learning_rate": 1.4564127825032904e-06, + "loss": 0.6433, + "step": 2475 + }, + { + "epoch": 3.8799582463465554, + "grad_norm": 1.1652717590332031, + "learning_rate": 1.4544987308531594e-06, + "loss": 0.5917, + "step": 2476 + }, + { + "epoch": 3.8815240083507305, + "grad_norm": 1.2521084547042847, + "learning_rate": 1.452585421662029e-06, + "loss": 0.6254, + "step": 2477 + }, + { + "epoch": 3.883089770354906, + "grad_norm": 1.2177410125732422, + "learning_rate": 1.4506728562886285e-06, + "loss": 0.6565, + "step": 2478 + }, + { + "epoch": 3.8846555323590817, + "grad_norm": 1.2197600603103638, + "learning_rate": 1.4487610360911606e-06, + "loss": 0.6001, + "step": 2479 + }, + { + "epoch": 3.886221294363257, + "grad_norm": 1.186809778213501, + "learning_rate": 1.4468499624272972e-06, + "loss": 0.611, + "step": 2480 + }, + { + "epoch": 3.887787056367432, + "grad_norm": 1.2346223592758179, + "learning_rate": 1.4449396366541802e-06, + "loss": 0.7188, + "step": 2481 + }, + { + "epoch": 3.8893528183716075, + "grad_norm": 1.148040533065796, + "learning_rate": 1.4430300601284214e-06, + "loss": 0.6711, + "step": 2482 + }, + { + "epoch": 3.890918580375783, + "grad_norm": 1.2020211219787598, + "learning_rate": 1.4411212342060988e-06, + "loss": 0.6765, + "step": 2483 + }, + { + "epoch": 3.892484342379958, + "grad_norm": 1.2378863096237183, + "learning_rate": 1.439213160242759e-06, + "loss": 0.6814, + "step": 2484 + }, + { + "epoch": 3.8940501043841333, + "grad_norm": 1.2072421312332153, + "learning_rate": 1.4373058395934136e-06, + "loss": 0.6178, + "step": 2485 + }, + { + "epoch": 3.895615866388309, + "grad_norm": 1.3510164022445679, + "learning_rate": 1.4353992736125396e-06, + "loss": 0.6775, + "step": 2486 + }, + { + "epoch": 3.8971816283924845, + "grad_norm": 1.2395039796829224, + "learning_rate": 1.4334934636540782e-06, + "loss": 0.5773, + "step": 2487 + }, + { + "epoch": 3.8987473903966596, + "grad_norm": 1.208178162574768, + "learning_rate": 1.4315884110714337e-06, + "loss": 0.5234, + "step": 2488 + }, + { + "epoch": 3.900313152400835, + "grad_norm": 1.2231117486953735, + "learning_rate": 1.429684117217473e-06, + "loss": 0.6405, + "step": 2489 + }, + { + "epoch": 3.9018789144050103, + "grad_norm": 1.1081833839416504, + "learning_rate": 1.4277805834445207e-06, + "loss": 0.6349, + "step": 2490 + }, + { + "epoch": 3.903444676409186, + "grad_norm": 1.3395150899887085, + "learning_rate": 1.4258778111043674e-06, + "loss": 0.6236, + "step": 2491 + }, + { + "epoch": 3.905010438413361, + "grad_norm": 1.2174855470657349, + "learning_rate": 1.4239758015482607e-06, + "loss": 0.667, + "step": 2492 + }, + { + "epoch": 3.9065762004175366, + "grad_norm": 1.3547643423080444, + "learning_rate": 1.4220745561269022e-06, + "loss": 0.6654, + "step": 2493 + }, + { + "epoch": 3.9081419624217117, + "grad_norm": 1.1408005952835083, + "learning_rate": 1.4201740761904591e-06, + "loss": 0.638, + "step": 2494 + }, + { + "epoch": 3.9097077244258873, + "grad_norm": 1.299989938735962, + "learning_rate": 1.418274363088547e-06, + "loss": 0.5875, + "step": 2495 + }, + { + "epoch": 3.911273486430063, + "grad_norm": 1.1767139434814453, + "learning_rate": 1.4163754181702408e-06, + "loss": 0.6723, + "step": 2496 + }, + { + "epoch": 3.912839248434238, + "grad_norm": 1.2104045152664185, + "learning_rate": 1.414477242784072e-06, + "loss": 0.5944, + "step": 2497 + }, + { + "epoch": 3.914405010438413, + "grad_norm": 1.2391446828842163, + "learning_rate": 1.4125798382780207e-06, + "loss": 0.6747, + "step": 2498 + }, + { + "epoch": 3.9159707724425887, + "grad_norm": 1.1312291622161865, + "learning_rate": 1.4106832059995224e-06, + "loss": 0.6705, + "step": 2499 + }, + { + "epoch": 3.9175365344467643, + "grad_norm": 1.2023922204971313, + "learning_rate": 1.4087873472954638e-06, + "loss": 0.6417, + "step": 2500 + }, + { + "epoch": 3.9191022964509394, + "grad_norm": 1.2477304935455322, + "learning_rate": 1.4068922635121828e-06, + "loss": 0.6471, + "step": 2501 + }, + { + "epoch": 3.9206680584551146, + "grad_norm": 1.140338659286499, + "learning_rate": 1.4049979559954666e-06, + "loss": 0.6042, + "step": 2502 + }, + { + "epoch": 3.92223382045929, + "grad_norm": 1.1643264293670654, + "learning_rate": 1.4031044260905486e-06, + "loss": 0.7038, + "step": 2503 + }, + { + "epoch": 3.9237995824634657, + "grad_norm": 1.2577917575836182, + "learning_rate": 1.4012116751421145e-06, + "loss": 0.6826, + "step": 2504 + }, + { + "epoch": 3.925365344467641, + "grad_norm": 1.1324503421783447, + "learning_rate": 1.3993197044942958e-06, + "loss": 0.6337, + "step": 2505 + }, + { + "epoch": 3.9269311064718164, + "grad_norm": 1.3015128374099731, + "learning_rate": 1.3974285154906647e-06, + "loss": 0.6443, + "step": 2506 + }, + { + "epoch": 3.9284968684759916, + "grad_norm": 1.139436960220337, + "learning_rate": 1.3955381094742464e-06, + "loss": 0.6569, + "step": 2507 + }, + { + "epoch": 3.930062630480167, + "grad_norm": 1.2854009866714478, + "learning_rate": 1.393648487787504e-06, + "loss": 0.6191, + "step": 2508 + }, + { + "epoch": 3.9316283924843423, + "grad_norm": 1.2040328979492188, + "learning_rate": 1.3917596517723441e-06, + "loss": 0.6355, + "step": 2509 + }, + { + "epoch": 3.933194154488518, + "grad_norm": 1.2754182815551758, + "learning_rate": 1.3898716027701214e-06, + "loss": 0.658, + "step": 2510 + }, + { + "epoch": 3.934759916492693, + "grad_norm": 1.2027299404144287, + "learning_rate": 1.387984342121623e-06, + "loss": 0.5478, + "step": 2511 + }, + { + "epoch": 3.9363256784968685, + "grad_norm": 1.1748281717300415, + "learning_rate": 1.3860978711670819e-06, + "loss": 0.6427, + "step": 2512 + }, + { + "epoch": 3.937891440501044, + "grad_norm": 1.209522008895874, + "learning_rate": 1.3842121912461686e-06, + "loss": 0.6234, + "step": 2513 + }, + { + "epoch": 3.9394572025052192, + "grad_norm": 1.1207284927368164, + "learning_rate": 1.382327303697992e-06, + "loss": 0.6092, + "step": 2514 + }, + { + "epoch": 3.9410229645093944, + "grad_norm": 1.2785614728927612, + "learning_rate": 1.3804432098610982e-06, + "loss": 0.6507, + "step": 2515 + }, + { + "epoch": 3.94258872651357, + "grad_norm": 1.2241697311401367, + "learning_rate": 1.37855991107347e-06, + "loss": 0.5863, + "step": 2516 + }, + { + "epoch": 3.9441544885177455, + "grad_norm": 1.1810356378555298, + "learning_rate": 1.3766774086725246e-06, + "loss": 0.6718, + "step": 2517 + }, + { + "epoch": 3.9457202505219207, + "grad_norm": 1.1314932107925415, + "learning_rate": 1.3747957039951152e-06, + "loss": 0.6726, + "step": 2518 + }, + { + "epoch": 3.947286012526096, + "grad_norm": 1.2174315452575684, + "learning_rate": 1.3729147983775272e-06, + "loss": 0.6201, + "step": 2519 + }, + { + "epoch": 3.9488517745302714, + "grad_norm": 1.2153561115264893, + "learning_rate": 1.371034693155479e-06, + "loss": 0.6499, + "step": 2520 + }, + { + "epoch": 3.950417536534447, + "grad_norm": 1.219773292541504, + "learning_rate": 1.3691553896641213e-06, + "loss": 0.6046, + "step": 2521 + }, + { + "epoch": 3.951983298538622, + "grad_norm": 1.284940481185913, + "learning_rate": 1.367276889238034e-06, + "loss": 0.6899, + "step": 2522 + }, + { + "epoch": 3.9535490605427976, + "grad_norm": 1.356799602508545, + "learning_rate": 1.3653991932112287e-06, + "loss": 0.6861, + "step": 2523 + }, + { + "epoch": 3.9551148225469728, + "grad_norm": 1.166818618774414, + "learning_rate": 1.3635223029171418e-06, + "loss": 0.5631, + "step": 2524 + }, + { + "epoch": 3.9566805845511483, + "grad_norm": 1.2792567014694214, + "learning_rate": 1.3616462196886427e-06, + "loss": 0.6471, + "step": 2525 + }, + { + "epoch": 3.9582463465553235, + "grad_norm": 1.2961795330047607, + "learning_rate": 1.3597709448580257e-06, + "loss": 0.6074, + "step": 2526 + }, + { + "epoch": 3.959812108559499, + "grad_norm": 1.2090468406677246, + "learning_rate": 1.3578964797570077e-06, + "loss": 0.6225, + "step": 2527 + }, + { + "epoch": 3.961377870563674, + "grad_norm": 1.1482014656066895, + "learning_rate": 1.3560228257167365e-06, + "loss": 0.5968, + "step": 2528 + }, + { + "epoch": 3.9629436325678498, + "grad_norm": 1.2505629062652588, + "learning_rate": 1.354149984067779e-06, + "loss": 0.6198, + "step": 2529 + }, + { + "epoch": 3.964509394572025, + "grad_norm": 1.1473637819290161, + "learning_rate": 1.3522779561401262e-06, + "loss": 0.5933, + "step": 2530 + }, + { + "epoch": 3.9660751565762005, + "grad_norm": 1.2111499309539795, + "learning_rate": 1.3504067432631954e-06, + "loss": 0.6173, + "step": 2531 + }, + { + "epoch": 3.9676409185803756, + "grad_norm": 1.2065558433532715, + "learning_rate": 1.3485363467658186e-06, + "loss": 0.6819, + "step": 2532 + }, + { + "epoch": 3.969206680584551, + "grad_norm": 1.3822040557861328, + "learning_rate": 1.3466667679762525e-06, + "loss": 0.6258, + "step": 2533 + }, + { + "epoch": 3.9707724425887267, + "grad_norm": 1.2183210849761963, + "learning_rate": 1.3447980082221719e-06, + "loss": 0.6501, + "step": 2534 + }, + { + "epoch": 3.972338204592902, + "grad_norm": 1.1406798362731934, + "learning_rate": 1.342930068830669e-06, + "loss": 0.5535, + "step": 2535 + }, + { + "epoch": 3.973903966597077, + "grad_norm": 1.2553014755249023, + "learning_rate": 1.3410629511282553e-06, + "loss": 0.6809, + "step": 2536 + }, + { + "epoch": 3.9754697286012526, + "grad_norm": 1.2471221685409546, + "learning_rate": 1.3391966564408571e-06, + "loss": 0.635, + "step": 2537 + }, + { + "epoch": 3.977035490605428, + "grad_norm": 1.318433165550232, + "learning_rate": 1.3373311860938177e-06, + "loss": 0.6571, + "step": 2538 + }, + { + "epoch": 3.9786012526096033, + "grad_norm": 1.2130142450332642, + "learning_rate": 1.335466541411893e-06, + "loss": 0.6643, + "step": 2539 + }, + { + "epoch": 3.980167014613779, + "grad_norm": 1.1460458040237427, + "learning_rate": 1.3336027237192551e-06, + "loss": 0.5837, + "step": 2540 + }, + { + "epoch": 3.981732776617954, + "grad_norm": 1.1360803842544556, + "learning_rate": 1.3317397343394879e-06, + "loss": 0.6432, + "step": 2541 + }, + { + "epoch": 3.9832985386221296, + "grad_norm": 1.1411229372024536, + "learning_rate": 1.3298775745955833e-06, + "loss": 0.5568, + "step": 2542 + }, + { + "epoch": 3.9848643006263047, + "grad_norm": 1.3322495222091675, + "learning_rate": 1.328016245809951e-06, + "loss": 0.7113, + "step": 2543 + }, + { + "epoch": 3.9864300626304803, + "grad_norm": 1.1256479024887085, + "learning_rate": 1.3261557493044067e-06, + "loss": 0.5875, + "step": 2544 + }, + { + "epoch": 3.9879958246346554, + "grad_norm": 1.1804993152618408, + "learning_rate": 1.3242960864001724e-06, + "loss": 0.6372, + "step": 2545 + }, + { + "epoch": 3.989561586638831, + "grad_norm": 1.1208686828613281, + "learning_rate": 1.3224372584178855e-06, + "loss": 0.4952, + "step": 2546 + }, + { + "epoch": 3.991127348643006, + "grad_norm": 1.2875077724456787, + "learning_rate": 1.3205792666775824e-06, + "loss": 0.6465, + "step": 2547 + }, + { + "epoch": 3.9926931106471817, + "grad_norm": 1.2553578615188599, + "learning_rate": 1.3187221124987107e-06, + "loss": 0.6053, + "step": 2548 + }, + { + "epoch": 3.994258872651357, + "grad_norm": 1.1742136478424072, + "learning_rate": 1.3168657972001214e-06, + "loss": 0.6711, + "step": 2549 + }, + { + "epoch": 3.9958246346555324, + "grad_norm": 1.184242844581604, + "learning_rate": 1.3150103221000709e-06, + "loss": 0.6405, + "step": 2550 + }, + { + "epoch": 3.997390396659708, + "grad_norm": 1.1263542175292969, + "learning_rate": 1.313155688516218e-06, + "loss": 0.5631, + "step": 2551 + }, + { + "epoch": 3.998956158663883, + "grad_norm": 1.2463663816452026, + "learning_rate": 1.3113018977656238e-06, + "loss": 0.5974, + "step": 2552 + }, + { + "epoch": 4.001565762004176, + "grad_norm": 2.222411870956421, + "learning_rate": 1.3094489511647512e-06, + "loss": 1.1917, + "step": 2553 + }, + { + "epoch": 4.003131524008351, + "grad_norm": 1.1565176248550415, + "learning_rate": 1.307596850029465e-06, + "loss": 0.6157, + "step": 2554 + }, + { + "epoch": 4.004697286012526, + "grad_norm": 1.3374539613723755, + "learning_rate": 1.3057455956750253e-06, + "loss": 0.548, + "step": 2555 + }, + { + "epoch": 4.006263048016701, + "grad_norm": 1.0805665254592896, + "learning_rate": 1.3038951894160962e-06, + "loss": 0.5639, + "step": 2556 + }, + { + "epoch": 4.007828810020877, + "grad_norm": 1.2551461458206177, + "learning_rate": 1.3020456325667374e-06, + "loss": 0.6393, + "step": 2557 + }, + { + "epoch": 4.009394572025053, + "grad_norm": 1.3844956159591675, + "learning_rate": 1.3001969264404024e-06, + "loss": 0.5836, + "step": 2558 + }, + { + "epoch": 4.010960334029227, + "grad_norm": 1.155824065208435, + "learning_rate": 1.298349072349947e-06, + "loss": 0.6122, + "step": 2559 + }, + { + "epoch": 4.012526096033403, + "grad_norm": 1.2663135528564453, + "learning_rate": 1.2965020716076148e-06, + "loss": 0.5728, + "step": 2560 + }, + { + "epoch": 4.014091858037578, + "grad_norm": 1.1352088451385498, + "learning_rate": 1.294655925525048e-06, + "loss": 0.6713, + "step": 2561 + }, + { + "epoch": 4.015657620041754, + "grad_norm": 1.179442286491394, + "learning_rate": 1.2928106354132823e-06, + "loss": 0.5363, + "step": 2562 + }, + { + "epoch": 4.017223382045929, + "grad_norm": 1.2838183641433716, + "learning_rate": 1.2909662025827413e-06, + "loss": 0.5278, + "step": 2563 + }, + { + "epoch": 4.018789144050104, + "grad_norm": 1.186926007270813, + "learning_rate": 1.289122628343244e-06, + "loss": 0.581, + "step": 2564 + }, + { + "epoch": 4.02035490605428, + "grad_norm": 1.1647127866744995, + "learning_rate": 1.2872799140039973e-06, + "loss": 0.5944, + "step": 2565 + }, + { + "epoch": 4.021920668058455, + "grad_norm": 1.1807323694229126, + "learning_rate": 1.2854380608735989e-06, + "loss": 0.6162, + "step": 2566 + }, + { + "epoch": 4.02348643006263, + "grad_norm": 1.1833971738815308, + "learning_rate": 1.2835970702600337e-06, + "loss": 0.6241, + "step": 2567 + }, + { + "epoch": 4.025052192066806, + "grad_norm": 1.1918176412582397, + "learning_rate": 1.2817569434706748e-06, + "loss": 0.6184, + "step": 2568 + }, + { + "epoch": 4.026617954070981, + "grad_norm": 1.275282859802246, + "learning_rate": 1.2799176818122816e-06, + "loss": 0.6285, + "step": 2569 + }, + { + "epoch": 4.028183716075157, + "grad_norm": 1.2189844846725464, + "learning_rate": 1.2780792865909996e-06, + "loss": 0.5431, + "step": 2570 + }, + { + "epoch": 4.029749478079332, + "grad_norm": 1.235913634300232, + "learning_rate": 1.2762417591123582e-06, + "loss": 0.6149, + "step": 2571 + }, + { + "epoch": 4.031315240083507, + "grad_norm": 1.160127878189087, + "learning_rate": 1.2744051006812712e-06, + "loss": 0.5061, + "step": 2572 + }, + { + "epoch": 4.032881002087683, + "grad_norm": 1.2041453123092651, + "learning_rate": 1.272569312602035e-06, + "loss": 0.636, + "step": 2573 + }, + { + "epoch": 4.034446764091858, + "grad_norm": 1.318177342414856, + "learning_rate": 1.2707343961783278e-06, + "loss": 0.6081, + "step": 2574 + }, + { + "epoch": 4.036012526096034, + "grad_norm": 1.0629124641418457, + "learning_rate": 1.2689003527132103e-06, + "loss": 0.6312, + "step": 2575 + }, + { + "epoch": 4.0375782881002085, + "grad_norm": 1.2917940616607666, + "learning_rate": 1.2670671835091187e-06, + "loss": 0.6204, + "step": 2576 + }, + { + "epoch": 4.039144050104384, + "grad_norm": 1.1925923824310303, + "learning_rate": 1.2652348898678742e-06, + "loss": 0.6608, + "step": 2577 + }, + { + "epoch": 4.04070981210856, + "grad_norm": 1.296000361442566, + "learning_rate": 1.2634034730906745e-06, + "loss": 0.613, + "step": 2578 + }, + { + "epoch": 4.042275574112735, + "grad_norm": 1.194916009902954, + "learning_rate": 1.26157293447809e-06, + "loss": 0.5946, + "step": 2579 + }, + { + "epoch": 4.04384133611691, + "grad_norm": 1.239607334136963, + "learning_rate": 1.2597432753300753e-06, + "loss": 0.6065, + "step": 2580 + }, + { + "epoch": 4.0454070981210855, + "grad_norm": 1.2236425876617432, + "learning_rate": 1.2579144969459536e-06, + "loss": 0.5722, + "step": 2581 + }, + { + "epoch": 4.046972860125261, + "grad_norm": 1.3422942161560059, + "learning_rate": 1.2560866006244249e-06, + "loss": 0.6717, + "step": 2582 + }, + { + "epoch": 4.048538622129437, + "grad_norm": 1.1085593700408936, + "learning_rate": 1.254259587663567e-06, + "loss": 0.6199, + "step": 2583 + }, + { + "epoch": 4.050104384133611, + "grad_norm": 1.3027156591415405, + "learning_rate": 1.252433459360823e-06, + "loss": 0.6361, + "step": 2584 + }, + { + "epoch": 4.051670146137787, + "grad_norm": 1.2550909519195557, + "learning_rate": 1.2506082170130121e-06, + "loss": 0.5996, + "step": 2585 + }, + { + "epoch": 4.053235908141962, + "grad_norm": 1.2809008359909058, + "learning_rate": 1.2487838619163245e-06, + "loss": 0.555, + "step": 2586 + }, + { + "epoch": 4.054801670146138, + "grad_norm": 1.0087321996688843, + "learning_rate": 1.246960395366319e-06, + "loss": 0.4473, + "step": 2587 + }, + { + "epoch": 4.056367432150314, + "grad_norm": 1.2337830066680908, + "learning_rate": 1.245137818657924e-06, + "loss": 0.5699, + "step": 2588 + }, + { + "epoch": 4.057933194154488, + "grad_norm": 1.1802204847335815, + "learning_rate": 1.243316133085436e-06, + "loss": 0.6677, + "step": 2589 + }, + { + "epoch": 4.059498956158664, + "grad_norm": 1.2031978368759155, + "learning_rate": 1.2414953399425182e-06, + "loss": 0.6389, + "step": 2590 + }, + { + "epoch": 4.061064718162839, + "grad_norm": 1.3534972667694092, + "learning_rate": 1.2396754405222003e-06, + "loss": 0.6177, + "step": 2591 + }, + { + "epoch": 4.062630480167015, + "grad_norm": 1.1125755310058594, + "learning_rate": 1.2378564361168776e-06, + "loss": 0.6383, + "step": 2592 + }, + { + "epoch": 4.06419624217119, + "grad_norm": 1.1830637454986572, + "learning_rate": 1.2360383280183108e-06, + "loss": 0.6341, + "step": 2593 + }, + { + "epoch": 4.065762004175365, + "grad_norm": 1.1648894548416138, + "learning_rate": 1.2342211175176206e-06, + "loss": 0.6327, + "step": 2594 + }, + { + "epoch": 4.067327766179541, + "grad_norm": 1.1454265117645264, + "learning_rate": 1.232404805905293e-06, + "loss": 0.6376, + "step": 2595 + }, + { + "epoch": 4.068893528183716, + "grad_norm": 1.153611660003662, + "learning_rate": 1.2305893944711773e-06, + "loss": 0.6032, + "step": 2596 + }, + { + "epoch": 4.070459290187891, + "grad_norm": 1.3515207767486572, + "learning_rate": 1.2287748845044792e-06, + "loss": 0.6064, + "step": 2597 + }, + { + "epoch": 4.072025052192067, + "grad_norm": 1.1163814067840576, + "learning_rate": 1.2269612772937672e-06, + "loss": 0.6582, + "step": 2598 + }, + { + "epoch": 4.073590814196242, + "grad_norm": 1.2641352415084839, + "learning_rate": 1.2251485741269683e-06, + "loss": 0.6221, + "step": 2599 + }, + { + "epoch": 4.075156576200418, + "grad_norm": 1.1293519735336304, + "learning_rate": 1.223336776291367e-06, + "loss": 0.5346, + "step": 2600 + }, + { + "epoch": 4.0767223382045925, + "grad_norm": 1.3557345867156982, + "learning_rate": 1.2215258850736051e-06, + "loss": 0.6602, + "step": 2601 + }, + { + "epoch": 4.078288100208768, + "grad_norm": 1.2822085618972778, + "learning_rate": 1.2197159017596803e-06, + "loss": 0.5706, + "step": 2602 + }, + { + "epoch": 4.079853862212944, + "grad_norm": 1.3728432655334473, + "learning_rate": 1.2179068276349462e-06, + "loss": 0.631, + "step": 2603 + }, + { + "epoch": 4.081419624217119, + "grad_norm": 1.1968104839324951, + "learning_rate": 1.21609866398411e-06, + "loss": 0.5861, + "step": 2604 + }, + { + "epoch": 4.082985386221294, + "grad_norm": 1.1587492227554321, + "learning_rate": 1.214291412091233e-06, + "loss": 0.6104, + "step": 2605 + }, + { + "epoch": 4.0845511482254695, + "grad_norm": 1.2656822204589844, + "learning_rate": 1.2124850732397294e-06, + "loss": 0.6327, + "step": 2606 + }, + { + "epoch": 4.086116910229645, + "grad_norm": 1.1670796871185303, + "learning_rate": 1.2106796487123615e-06, + "loss": 0.5624, + "step": 2607 + }, + { + "epoch": 4.087682672233821, + "grad_norm": 1.1668428182601929, + "learning_rate": 1.208875139791248e-06, + "loss": 0.6524, + "step": 2608 + }, + { + "epoch": 4.089248434237996, + "grad_norm": 1.2208465337753296, + "learning_rate": 1.2070715477578546e-06, + "loss": 0.5131, + "step": 2609 + }, + { + "epoch": 4.090814196242171, + "grad_norm": 1.1894104480743408, + "learning_rate": 1.2052688738929927e-06, + "loss": 0.6169, + "step": 2610 + }, + { + "epoch": 4.0923799582463465, + "grad_norm": 1.325358510017395, + "learning_rate": 1.2034671194768286e-06, + "loss": 0.5342, + "step": 2611 + }, + { + "epoch": 4.093945720250522, + "grad_norm": 1.1475489139556885, + "learning_rate": 1.201666285788869e-06, + "loss": 0.6409, + "step": 2612 + }, + { + "epoch": 4.095511482254698, + "grad_norm": 1.3095086812973022, + "learning_rate": 1.199866374107969e-06, + "loss": 0.6314, + "step": 2613 + }, + { + "epoch": 4.097077244258872, + "grad_norm": 1.2492880821228027, + "learning_rate": 1.198067385712333e-06, + "loss": 0.639, + "step": 2614 + }, + { + "epoch": 4.098643006263048, + "grad_norm": 1.179559588432312, + "learning_rate": 1.1962693218795036e-06, + "loss": 0.6343, + "step": 2615 + }, + { + "epoch": 4.1002087682672235, + "grad_norm": 1.2899236679077148, + "learning_rate": 1.1944721838863693e-06, + "loss": 0.6627, + "step": 2616 + }, + { + "epoch": 4.101774530271399, + "grad_norm": 1.3395811319351196, + "learning_rate": 1.192675973009162e-06, + "loss": 0.6222, + "step": 2617 + }, + { + "epoch": 4.103340292275574, + "grad_norm": 1.2343623638153076, + "learning_rate": 1.1908806905234544e-06, + "loss": 0.606, + "step": 2618 + }, + { + "epoch": 4.104906054279749, + "grad_norm": 1.300563931465149, + "learning_rate": 1.1890863377041595e-06, + "loss": 0.5701, + "step": 2619 + }, + { + "epoch": 4.106471816283925, + "grad_norm": 1.462306261062622, + "learning_rate": 1.187292915825531e-06, + "loss": 0.5738, + "step": 2620 + }, + { + "epoch": 4.1080375782881005, + "grad_norm": 1.21458101272583, + "learning_rate": 1.1855004261611611e-06, + "loss": 0.6335, + "step": 2621 + }, + { + "epoch": 4.109603340292275, + "grad_norm": 1.1758227348327637, + "learning_rate": 1.1837088699839798e-06, + "loss": 0.6325, + "step": 2622 + }, + { + "epoch": 4.111169102296451, + "grad_norm": 1.2418543100357056, + "learning_rate": 1.1819182485662542e-06, + "loss": 0.6509, + "step": 2623 + }, + { + "epoch": 4.112734864300626, + "grad_norm": 1.235479474067688, + "learning_rate": 1.180128563179588e-06, + "loss": 0.594, + "step": 2624 + }, + { + "epoch": 4.114300626304802, + "grad_norm": 1.2463157176971436, + "learning_rate": 1.17833981509492e-06, + "loss": 0.5918, + "step": 2625 + }, + { + "epoch": 4.115866388308977, + "grad_norm": 1.2002460956573486, + "learning_rate": 1.1765520055825225e-06, + "loss": 0.5981, + "step": 2626 + }, + { + "epoch": 4.117432150313152, + "grad_norm": 1.349730372428894, + "learning_rate": 1.1747651359120038e-06, + "loss": 0.6084, + "step": 2627 + }, + { + "epoch": 4.118997912317328, + "grad_norm": 1.277337908744812, + "learning_rate": 1.1729792073523e-06, + "loss": 0.595, + "step": 2628 + }, + { + "epoch": 4.120563674321503, + "grad_norm": 1.2968724966049194, + "learning_rate": 1.1711942211716845e-06, + "loss": 0.6708, + "step": 2629 + }, + { + "epoch": 4.122129436325679, + "grad_norm": 1.1646991968154907, + "learning_rate": 1.1694101786377593e-06, + "loss": 0.6009, + "step": 2630 + }, + { + "epoch": 4.1236951983298535, + "grad_norm": 1.392899513244629, + "learning_rate": 1.1676270810174526e-06, + "loss": 0.6038, + "step": 2631 + }, + { + "epoch": 4.125260960334029, + "grad_norm": 1.085792899131775, + "learning_rate": 1.1658449295770287e-06, + "loss": 0.5864, + "step": 2632 + }, + { + "epoch": 4.126826722338205, + "grad_norm": 1.2588834762573242, + "learning_rate": 1.1640637255820732e-06, + "loss": 0.633, + "step": 2633 + }, + { + "epoch": 4.12839248434238, + "grad_norm": 1.2557780742645264, + "learning_rate": 1.1622834702975022e-06, + "loss": 0.6003, + "step": 2634 + }, + { + "epoch": 4.129958246346555, + "grad_norm": 1.4421063661575317, + "learning_rate": 1.1605041649875602e-06, + "loss": 0.5322, + "step": 2635 + }, + { + "epoch": 4.1315240083507305, + "grad_norm": 1.1980674266815186, + "learning_rate": 1.1587258109158114e-06, + "loss": 0.5461, + "step": 2636 + }, + { + "epoch": 4.133089770354906, + "grad_norm": 1.2281138896942139, + "learning_rate": 1.156948409345149e-06, + "loss": 0.6324, + "step": 2637 + }, + { + "epoch": 4.134655532359082, + "grad_norm": 1.25807523727417, + "learning_rate": 1.1551719615377882e-06, + "loss": 0.6451, + "step": 2638 + }, + { + "epoch": 4.136221294363256, + "grad_norm": 1.1664526462554932, + "learning_rate": 1.153396468755267e-06, + "loss": 0.5765, + "step": 2639 + }, + { + "epoch": 4.137787056367432, + "grad_norm": 1.226771593093872, + "learning_rate": 1.1516219322584463e-06, + "loss": 0.6863, + "step": 2640 + }, + { + "epoch": 4.1393528183716075, + "grad_norm": 1.1369787454605103, + "learning_rate": 1.1498483533075042e-06, + "loss": 0.6013, + "step": 2641 + }, + { + "epoch": 4.140918580375783, + "grad_norm": 1.2024613618850708, + "learning_rate": 1.148075733161944e-06, + "loss": 0.6674, + "step": 2642 + }, + { + "epoch": 4.142484342379959, + "grad_norm": 1.1067324876785278, + "learning_rate": 1.1463040730805859e-06, + "loss": 0.5326, + "step": 2643 + }, + { + "epoch": 4.144050104384133, + "grad_norm": 1.2651442289352417, + "learning_rate": 1.1445333743215648e-06, + "loss": 0.6634, + "step": 2644 + }, + { + "epoch": 4.145615866388309, + "grad_norm": 1.22459876537323, + "learning_rate": 1.1427636381423402e-06, + "loss": 0.6078, + "step": 2645 + }, + { + "epoch": 4.1471816283924845, + "grad_norm": 1.200835943222046, + "learning_rate": 1.140994865799681e-06, + "loss": 0.6767, + "step": 2646 + }, + { + "epoch": 4.14874739039666, + "grad_norm": 1.2004497051239014, + "learning_rate": 1.1392270585496746e-06, + "loss": 0.6043, + "step": 2647 + }, + { + "epoch": 4.150313152400835, + "grad_norm": 1.2646071910858154, + "learning_rate": 1.1374602176477256e-06, + "loss": 0.5793, + "step": 2648 + }, + { + "epoch": 4.15187891440501, + "grad_norm": 1.2798433303833008, + "learning_rate": 1.1356943443485475e-06, + "loss": 0.6171, + "step": 2649 + }, + { + "epoch": 4.153444676409186, + "grad_norm": 1.2004300355911255, + "learning_rate": 1.1339294399061692e-06, + "loss": 0.6218, + "step": 2650 + }, + { + "epoch": 4.1550104384133615, + "grad_norm": 1.3948124647140503, + "learning_rate": 1.1321655055739316e-06, + "loss": 0.6047, + "step": 2651 + }, + { + "epoch": 4.156576200417536, + "grad_norm": 1.2331879138946533, + "learning_rate": 1.1304025426044869e-06, + "loss": 0.6276, + "step": 2652 + }, + { + "epoch": 4.158141962421712, + "grad_norm": 1.3041292428970337, + "learning_rate": 1.1286405522497962e-06, + "loss": 0.5769, + "step": 2653 + }, + { + "epoch": 4.159707724425887, + "grad_norm": 1.2389923334121704, + "learning_rate": 1.1268795357611308e-06, + "loss": 0.5811, + "step": 2654 + }, + { + "epoch": 4.161273486430063, + "grad_norm": 1.3386740684509277, + "learning_rate": 1.1251194943890706e-06, + "loss": 0.5769, + "step": 2655 + }, + { + "epoch": 4.162839248434238, + "grad_norm": 1.129999041557312, + "learning_rate": 1.1233604293835027e-06, + "loss": 0.6376, + "step": 2656 + }, + { + "epoch": 4.164405010438413, + "grad_norm": 1.2060986757278442, + "learning_rate": 1.1216023419936206e-06, + "loss": 0.6338, + "step": 2657 + }, + { + "epoch": 4.165970772442589, + "grad_norm": 1.2099809646606445, + "learning_rate": 1.1198452334679246e-06, + "loss": 0.6374, + "step": 2658 + }, + { + "epoch": 4.167536534446764, + "grad_norm": 1.1674392223358154, + "learning_rate": 1.1180891050542168e-06, + "loss": 0.6486, + "step": 2659 + }, + { + "epoch": 4.16910229645094, + "grad_norm": 1.2688627243041992, + "learning_rate": 1.116333957999608e-06, + "loss": 0.5927, + "step": 2660 + }, + { + "epoch": 4.170668058455115, + "grad_norm": 1.36686372756958, + "learning_rate": 1.1145797935505095e-06, + "loss": 0.6127, + "step": 2661 + }, + { + "epoch": 4.17223382045929, + "grad_norm": 1.32498300075531, + "learning_rate": 1.1128266129526325e-06, + "loss": 0.67, + "step": 2662 + }, + { + "epoch": 4.173799582463466, + "grad_norm": 1.247199535369873, + "learning_rate": 1.1110744174509952e-06, + "loss": 0.6419, + "step": 2663 + }, + { + "epoch": 4.175365344467641, + "grad_norm": 1.2895094156265259, + "learning_rate": 1.1093232082899106e-06, + "loss": 0.6274, + "step": 2664 + }, + { + "epoch": 4.176931106471816, + "grad_norm": 1.2122670412063599, + "learning_rate": 1.1075729867129934e-06, + "loss": 0.6086, + "step": 2665 + }, + { + "epoch": 4.1784968684759916, + "grad_norm": 1.2727183103561401, + "learning_rate": 1.1058237539631603e-06, + "loss": 0.6258, + "step": 2666 + }, + { + "epoch": 4.180062630480167, + "grad_norm": 1.295921802520752, + "learning_rate": 1.1040755112826197e-06, + "loss": 0.6066, + "step": 2667 + }, + { + "epoch": 4.181628392484343, + "grad_norm": 1.3321471214294434, + "learning_rate": 1.1023282599128797e-06, + "loss": 0.5826, + "step": 2668 + }, + { + "epoch": 4.183194154488517, + "grad_norm": 1.190930962562561, + "learning_rate": 1.100582001094748e-06, + "loss": 0.5981, + "step": 2669 + }, + { + "epoch": 4.184759916492693, + "grad_norm": 1.2681376934051514, + "learning_rate": 1.0988367360683209e-06, + "loss": 0.6093, + "step": 2670 + }, + { + "epoch": 4.1863256784968685, + "grad_norm": 1.2412774562835693, + "learning_rate": 1.097092466072993e-06, + "loss": 0.5771, + "step": 2671 + }, + { + "epoch": 4.187891440501044, + "grad_norm": 1.077805757522583, + "learning_rate": 1.0953491923474515e-06, + "loss": 0.6112, + "step": 2672 + }, + { + "epoch": 4.189457202505219, + "grad_norm": 1.1926559209823608, + "learning_rate": 1.0936069161296764e-06, + "loss": 0.5909, + "step": 2673 + }, + { + "epoch": 4.191022964509394, + "grad_norm": 1.104425072669983, + "learning_rate": 1.0918656386569384e-06, + "loss": 0.4628, + "step": 2674 + }, + { + "epoch": 4.19258872651357, + "grad_norm": 1.3284837007522583, + "learning_rate": 1.0901253611657996e-06, + "loss": 0.654, + "step": 2675 + }, + { + "epoch": 4.1941544885177455, + "grad_norm": 1.249345302581787, + "learning_rate": 1.0883860848921122e-06, + "loss": 0.5535, + "step": 2676 + }, + { + "epoch": 4.195720250521921, + "grad_norm": 1.233498454093933, + "learning_rate": 1.086647811071017e-06, + "loss": 0.6309, + "step": 2677 + }, + { + "epoch": 4.197286012526096, + "grad_norm": 1.1552709341049194, + "learning_rate": 1.0849105409369424e-06, + "loss": 0.5742, + "step": 2678 + }, + { + "epoch": 4.198851774530271, + "grad_norm": 1.1115463972091675, + "learning_rate": 1.0831742757236059e-06, + "loss": 0.6373, + "step": 2679 + }, + { + "epoch": 4.200417536534447, + "grad_norm": 1.2297323942184448, + "learning_rate": 1.0814390166640073e-06, + "loss": 0.6526, + "step": 2680 + }, + { + "epoch": 4.2019832985386225, + "grad_norm": 1.4014023542404175, + "learning_rate": 1.0797047649904377e-06, + "loss": 0.5836, + "step": 2681 + }, + { + "epoch": 4.203549060542797, + "grad_norm": 1.242067813873291, + "learning_rate": 1.0779715219344692e-06, + "loss": 0.6644, + "step": 2682 + }, + { + "epoch": 4.205114822546973, + "grad_norm": 1.3167695999145508, + "learning_rate": 1.0762392887269556e-06, + "loss": 0.637, + "step": 2683 + }, + { + "epoch": 4.206680584551148, + "grad_norm": 1.229784369468689, + "learning_rate": 1.07450806659804e-06, + "loss": 0.6116, + "step": 2684 + }, + { + "epoch": 4.208246346555324, + "grad_norm": 1.2310740947723389, + "learning_rate": 1.0727778567771407e-06, + "loss": 0.5437, + "step": 2685 + }, + { + "epoch": 4.209812108559499, + "grad_norm": 1.277980089187622, + "learning_rate": 1.0710486604929602e-06, + "loss": 0.6151, + "step": 2686 + }, + { + "epoch": 4.211377870563674, + "grad_norm": 1.2275346517562866, + "learning_rate": 1.0693204789734834e-06, + "loss": 0.5514, + "step": 2687 + }, + { + "epoch": 4.21294363256785, + "grad_norm": 1.2537457942962646, + "learning_rate": 1.0675933134459702e-06, + "loss": 0.5858, + "step": 2688 + }, + { + "epoch": 4.214509394572025, + "grad_norm": 1.2391557693481445, + "learning_rate": 1.0658671651369615e-06, + "loss": 0.5432, + "step": 2689 + }, + { + "epoch": 4.2160751565762, + "grad_norm": 1.2173511981964111, + "learning_rate": 1.0641420352722754e-06, + "loss": 0.651, + "step": 2690 + }, + { + "epoch": 4.217640918580376, + "grad_norm": 1.227282166481018, + "learning_rate": 1.0624179250770066e-06, + "loss": 0.5668, + "step": 2691 + }, + { + "epoch": 4.219206680584551, + "grad_norm": 1.195631980895996, + "learning_rate": 1.060694835775527e-06, + "loss": 0.6318, + "step": 2692 + }, + { + "epoch": 4.220772442588727, + "grad_norm": 1.2749062776565552, + "learning_rate": 1.0589727685914798e-06, + "loss": 0.6389, + "step": 2693 + }, + { + "epoch": 4.222338204592901, + "grad_norm": 1.3092809915542603, + "learning_rate": 1.057251724747787e-06, + "loss": 0.5782, + "step": 2694 + }, + { + "epoch": 4.223903966597077, + "grad_norm": 1.222758173942566, + "learning_rate": 1.055531705466642e-06, + "loss": 0.6047, + "step": 2695 + }, + { + "epoch": 4.225469728601253, + "grad_norm": 1.2806799411773682, + "learning_rate": 1.0538127119695076e-06, + "loss": 0.5916, + "step": 2696 + }, + { + "epoch": 4.227035490605428, + "grad_norm": 1.2441686391830444, + "learning_rate": 1.052094745477125e-06, + "loss": 0.6405, + "step": 2697 + }, + { + "epoch": 4.228601252609604, + "grad_norm": 1.2265493869781494, + "learning_rate": 1.0503778072094986e-06, + "loss": 0.624, + "step": 2698 + }, + { + "epoch": 4.230167014613778, + "grad_norm": 1.2378262281417847, + "learning_rate": 1.0486618983859064e-06, + "loss": 0.6267, + "step": 2699 + }, + { + "epoch": 4.231732776617954, + "grad_norm": 1.1325656175613403, + "learning_rate": 1.0469470202248976e-06, + "loss": 0.5143, + "step": 2700 + }, + { + "epoch": 4.23329853862213, + "grad_norm": 1.1236976385116577, + "learning_rate": 1.0452331739442839e-06, + "loss": 0.5564, + "step": 2701 + }, + { + "epoch": 4.234864300626305, + "grad_norm": 1.3158353567123413, + "learning_rate": 1.0435203607611483e-06, + "loss": 0.5853, + "step": 2702 + }, + { + "epoch": 4.23643006263048, + "grad_norm": 1.1873832941055298, + "learning_rate": 1.041808581891839e-06, + "loss": 0.6496, + "step": 2703 + }, + { + "epoch": 4.237995824634655, + "grad_norm": 1.2505794763565063, + "learning_rate": 1.0400978385519692e-06, + "loss": 0.5656, + "step": 2704 + }, + { + "epoch": 4.239561586638831, + "grad_norm": 1.230660319328308, + "learning_rate": 1.0383881319564176e-06, + "loss": 0.5938, + "step": 2705 + }, + { + "epoch": 4.241127348643007, + "grad_norm": 1.349231481552124, + "learning_rate": 1.036679463319326e-06, + "loss": 0.6108, + "step": 2706 + }, + { + "epoch": 4.242693110647181, + "grad_norm": 1.275140404701233, + "learning_rate": 1.0349718338540998e-06, + "loss": 0.6672, + "step": 2707 + }, + { + "epoch": 4.244258872651357, + "grad_norm": 1.3713583946228027, + "learning_rate": 1.0332652447734057e-06, + "loss": 0.5547, + "step": 2708 + }, + { + "epoch": 4.245824634655532, + "grad_norm": 1.2152209281921387, + "learning_rate": 1.0315596972891714e-06, + "loss": 0.5614, + "step": 2709 + }, + { + "epoch": 4.247390396659708, + "grad_norm": 1.302657127380371, + "learning_rate": 1.0298551926125863e-06, + "loss": 0.6417, + "step": 2710 + }, + { + "epoch": 4.2489561586638835, + "grad_norm": 1.1909019947052002, + "learning_rate": 1.0281517319540976e-06, + "loss": 0.6784, + "step": 2711 + }, + { + "epoch": 4.250521920668058, + "grad_norm": 1.202182650566101, + "learning_rate": 1.026449316523412e-06, + "loss": 0.6202, + "step": 2712 + }, + { + "epoch": 4.252087682672234, + "grad_norm": 1.1392393112182617, + "learning_rate": 1.0247479475294948e-06, + "loss": 0.5573, + "step": 2713 + }, + { + "epoch": 4.253653444676409, + "grad_norm": 1.2918214797973633, + "learning_rate": 1.023047626180564e-06, + "loss": 0.6204, + "step": 2714 + }, + { + "epoch": 4.255219206680585, + "grad_norm": 1.249887228012085, + "learning_rate": 1.0213483536841e-06, + "loss": 0.6308, + "step": 2715 + }, + { + "epoch": 4.25678496868476, + "grad_norm": 1.156160593032837, + "learning_rate": 1.019650131246835e-06, + "loss": 0.5617, + "step": 2716 + }, + { + "epoch": 4.258350730688935, + "grad_norm": 1.2980233430862427, + "learning_rate": 1.0179529600747526e-06, + "loss": 0.619, + "step": 2717 + }, + { + "epoch": 4.259916492693111, + "grad_norm": 1.1643873453140259, + "learning_rate": 1.0162568413730966e-06, + "loss": 0.6488, + "step": 2718 + }, + { + "epoch": 4.261482254697286, + "grad_norm": 1.3735922574996948, + "learning_rate": 1.0145617763463573e-06, + "loss": 0.6022, + "step": 2719 + }, + { + "epoch": 4.263048016701461, + "grad_norm": 1.2653254270553589, + "learning_rate": 1.0128677661982786e-06, + "loss": 0.6285, + "step": 2720 + }, + { + "epoch": 4.264613778705637, + "grad_norm": 1.3981285095214844, + "learning_rate": 1.0111748121318587e-06, + "loss": 0.6381, + "step": 2721 + }, + { + "epoch": 4.266179540709812, + "grad_norm": 1.2364410161972046, + "learning_rate": 1.00948291534934e-06, + "loss": 0.5779, + "step": 2722 + }, + { + "epoch": 4.267745302713988, + "grad_norm": 1.0957547426223755, + "learning_rate": 1.0077920770522184e-06, + "loss": 0.6646, + "step": 2723 + }, + { + "epoch": 4.2693110647181625, + "grad_norm": 1.1948002576828003, + "learning_rate": 1.006102298441236e-06, + "loss": 0.6691, + "step": 2724 + }, + { + "epoch": 4.270876826722338, + "grad_norm": 1.3255823850631714, + "learning_rate": 1.0044135807163834e-06, + "loss": 0.5499, + "step": 2725 + }, + { + "epoch": 4.272442588726514, + "grad_norm": 1.2720073461532593, + "learning_rate": 1.0027259250768975e-06, + "loss": 0.628, + "step": 2726 + }, + { + "epoch": 4.274008350730689, + "grad_norm": 1.1583763360977173, + "learning_rate": 1.0010393327212612e-06, + "loss": 0.6253, + "step": 2727 + }, + { + "epoch": 4.275574112734864, + "grad_norm": 1.1668976545333862, + "learning_rate": 9.993538048472013e-07, + "loss": 0.6063, + "step": 2728 + }, + { + "epoch": 4.2771398747390394, + "grad_norm": 1.2782100439071655, + "learning_rate": 9.976693426516902e-07, + "loss": 0.6156, + "step": 2729 + }, + { + "epoch": 4.278705636743215, + "grad_norm": 1.2681186199188232, + "learning_rate": 9.959859473309422e-07, + "loss": 0.6426, + "step": 2730 + }, + { + "epoch": 4.280271398747391, + "grad_norm": 1.1437429189682007, + "learning_rate": 9.943036200804154e-07, + "loss": 0.5888, + "step": 2731 + }, + { + "epoch": 4.281837160751566, + "grad_norm": 1.3140877485275269, + "learning_rate": 9.926223620948061e-07, + "loss": 0.5913, + "step": 2732 + }, + { + "epoch": 4.283402922755741, + "grad_norm": 1.1629775762557983, + "learning_rate": 9.909421745680565e-07, + "loss": 0.5928, + "step": 2733 + }, + { + "epoch": 4.284968684759916, + "grad_norm": 1.3503170013427734, + "learning_rate": 9.892630586933452e-07, + "loss": 0.6137, + "step": 2734 + }, + { + "epoch": 4.286534446764092, + "grad_norm": 1.1408926248550415, + "learning_rate": 9.875850156630892e-07, + "loss": 0.5934, + "step": 2735 + }, + { + "epoch": 4.288100208768268, + "grad_norm": 1.3211277723312378, + "learning_rate": 9.859080466689452e-07, + "loss": 0.5723, + "step": 2736 + }, + { + "epoch": 4.289665970772442, + "grad_norm": 1.2893445491790771, + "learning_rate": 9.84232152901807e-07, + "loss": 0.5508, + "step": 2737 + }, + { + "epoch": 4.291231732776618, + "grad_norm": 1.268737554550171, + "learning_rate": 9.825573355518041e-07, + "loss": 0.5846, + "step": 2738 + }, + { + "epoch": 4.292797494780793, + "grad_norm": 1.145947813987732, + "learning_rate": 9.808835958083029e-07, + "loss": 0.5976, + "step": 2739 + }, + { + "epoch": 4.294363256784969, + "grad_norm": 1.2387874126434326, + "learning_rate": 9.792109348599036e-07, + "loss": 0.6492, + "step": 2740 + }, + { + "epoch": 4.295929018789144, + "grad_norm": 1.212338924407959, + "learning_rate": 9.775393538944394e-07, + "loss": 0.6379, + "step": 2741 + }, + { + "epoch": 4.297494780793319, + "grad_norm": 1.263137936592102, + "learning_rate": 9.75868854098979e-07, + "loss": 0.5711, + "step": 2742 + }, + { + "epoch": 4.299060542797495, + "grad_norm": 1.2036539316177368, + "learning_rate": 9.741994366598207e-07, + "loss": 0.6217, + "step": 2743 + }, + { + "epoch": 4.30062630480167, + "grad_norm": 1.3662471771240234, + "learning_rate": 9.725311027624967e-07, + "loss": 0.6184, + "step": 2744 + }, + { + "epoch": 4.302192066805846, + "grad_norm": 1.1384464502334595, + "learning_rate": 9.708638535917658e-07, + "loss": 0.6273, + "step": 2745 + }, + { + "epoch": 4.303757828810021, + "grad_norm": 1.3032082319259644, + "learning_rate": 9.691976903316213e-07, + "loss": 0.6393, + "step": 2746 + }, + { + "epoch": 4.305323590814196, + "grad_norm": 1.3347432613372803, + "learning_rate": 9.675326141652839e-07, + "loss": 0.6131, + "step": 2747 + }, + { + "epoch": 4.306889352818372, + "grad_norm": 1.2203959226608276, + "learning_rate": 9.65868626275198e-07, + "loss": 0.5778, + "step": 2748 + }, + { + "epoch": 4.3084551148225465, + "grad_norm": 1.2628487348556519, + "learning_rate": 9.642057278430431e-07, + "loss": 0.6296, + "step": 2749 + }, + { + "epoch": 4.310020876826722, + "grad_norm": 1.208202838897705, + "learning_rate": 9.62543920049717e-07, + "loss": 0.6245, + "step": 2750 + }, + { + "epoch": 4.311586638830898, + "grad_norm": 1.1364301443099976, + "learning_rate": 9.608832040753472e-07, + "loss": 0.6183, + "step": 2751 + }, + { + "epoch": 4.313152400835073, + "grad_norm": 1.3837535381317139, + "learning_rate": 9.592235810992875e-07, + "loss": 0.6547, + "step": 2752 + }, + { + "epoch": 4.314718162839249, + "grad_norm": 1.222366452217102, + "learning_rate": 9.575650523001106e-07, + "loss": 0.6415, + "step": 2753 + }, + { + "epoch": 4.3162839248434235, + "grad_norm": 1.3130534887313843, + "learning_rate": 9.559076188556163e-07, + "loss": 0.6968, + "step": 2754 + }, + { + "epoch": 4.317849686847599, + "grad_norm": 1.3480855226516724, + "learning_rate": 9.542512819428246e-07, + "loss": 0.6532, + "step": 2755 + }, + { + "epoch": 4.319415448851775, + "grad_norm": 1.4194175004959106, + "learning_rate": 9.525960427379772e-07, + "loss": 0.6413, + "step": 2756 + }, + { + "epoch": 4.32098121085595, + "grad_norm": 1.1544206142425537, + "learning_rate": 9.509419024165364e-07, + "loss": 0.6062, + "step": 2757 + }, + { + "epoch": 4.322546972860125, + "grad_norm": 1.343361735343933, + "learning_rate": 9.49288862153184e-07, + "loss": 0.682, + "step": 2758 + }, + { + "epoch": 4.3241127348643005, + "grad_norm": 1.2994462251663208, + "learning_rate": 9.476369231218207e-07, + "loss": 0.6081, + "step": 2759 + }, + { + "epoch": 4.325678496868476, + "grad_norm": 1.2797702550888062, + "learning_rate": 9.459860864955653e-07, + "loss": 0.5675, + "step": 2760 + }, + { + "epoch": 4.327244258872652, + "grad_norm": 1.322637915611267, + "learning_rate": 9.443363534467534e-07, + "loss": 0.5972, + "step": 2761 + }, + { + "epoch": 4.328810020876826, + "grad_norm": 1.3239537477493286, + "learning_rate": 9.426877251469372e-07, + "loss": 0.5902, + "step": 2762 + }, + { + "epoch": 4.330375782881002, + "grad_norm": 1.213059663772583, + "learning_rate": 9.410402027668835e-07, + "loss": 0.6537, + "step": 2763 + }, + { + "epoch": 4.3319415448851775, + "grad_norm": 1.2869120836257935, + "learning_rate": 9.393937874765754e-07, + "loss": 0.5573, + "step": 2764 + }, + { + "epoch": 4.333507306889353, + "grad_norm": 1.2183059453964233, + "learning_rate": 9.377484804452094e-07, + "loss": 0.6328, + "step": 2765 + }, + { + "epoch": 4.335073068893529, + "grad_norm": 1.2206132411956787, + "learning_rate": 9.361042828411915e-07, + "loss": 0.5814, + "step": 2766 + }, + { + "epoch": 4.336638830897703, + "grad_norm": 1.1965863704681396, + "learning_rate": 9.344611958321459e-07, + "loss": 0.5821, + "step": 2767 + }, + { + "epoch": 4.338204592901879, + "grad_norm": 1.29924738407135, + "learning_rate": 9.328192205849043e-07, + "loss": 0.6607, + "step": 2768 + }, + { + "epoch": 4.3397703549060545, + "grad_norm": 1.0587471723556519, + "learning_rate": 9.311783582655076e-07, + "loss": 0.6591, + "step": 2769 + }, + { + "epoch": 4.34133611691023, + "grad_norm": 1.2471832036972046, + "learning_rate": 9.295386100392117e-07, + "loss": 0.6371, + "step": 2770 + }, + { + "epoch": 4.342901878914405, + "grad_norm": 1.2618634700775146, + "learning_rate": 9.27899977070475e-07, + "loss": 0.5275, + "step": 2771 + }, + { + "epoch": 4.34446764091858, + "grad_norm": 1.2841023206710815, + "learning_rate": 9.262624605229673e-07, + "loss": 0.5328, + "step": 2772 + }, + { + "epoch": 4.346033402922756, + "grad_norm": 1.3328698873519897, + "learning_rate": 9.246260615595673e-07, + "loss": 0.5379, + "step": 2773 + }, + { + "epoch": 4.347599164926931, + "grad_norm": 1.1054933071136475, + "learning_rate": 9.229907813423556e-07, + "loss": 0.5078, + "step": 2774 + }, + { + "epoch": 4.349164926931106, + "grad_norm": 1.2079209089279175, + "learning_rate": 9.213566210326213e-07, + "loss": 0.6467, + "step": 2775 + }, + { + "epoch": 4.350730688935282, + "grad_norm": 1.2839919328689575, + "learning_rate": 9.197235817908579e-07, + "loss": 0.5954, + "step": 2776 + }, + { + "epoch": 4.352296450939457, + "grad_norm": 1.2439117431640625, + "learning_rate": 9.180916647767621e-07, + "loss": 0.6098, + "step": 2777 + }, + { + "epoch": 4.353862212943633, + "grad_norm": 1.2913306951522827, + "learning_rate": 9.164608711492337e-07, + "loss": 0.5375, + "step": 2778 + }, + { + "epoch": 4.3554279749478075, + "grad_norm": 1.1591486930847168, + "learning_rate": 9.148312020663755e-07, + "loss": 0.5844, + "step": 2779 + }, + { + "epoch": 4.356993736951983, + "grad_norm": 1.2469542026519775, + "learning_rate": 9.132026586854909e-07, + "loss": 0.6684, + "step": 2780 + }, + { + "epoch": 4.358559498956159, + "grad_norm": 1.310805320739746, + "learning_rate": 9.115752421630852e-07, + "loss": 0.6514, + "step": 2781 + }, + { + "epoch": 4.360125260960334, + "grad_norm": 1.2318816184997559, + "learning_rate": 9.099489536548598e-07, + "loss": 0.5453, + "step": 2782 + }, + { + "epoch": 4.361691022964509, + "grad_norm": 1.23331618309021, + "learning_rate": 9.083237943157208e-07, + "loss": 0.5484, + "step": 2783 + }, + { + "epoch": 4.3632567849686845, + "grad_norm": 1.390173077583313, + "learning_rate": 9.066997652997672e-07, + "loss": 0.6245, + "step": 2784 + }, + { + "epoch": 4.36482254697286, + "grad_norm": 1.3007646799087524, + "learning_rate": 9.050768677602973e-07, + "loss": 0.6415, + "step": 2785 + }, + { + "epoch": 4.366388308977036, + "grad_norm": 1.2786531448364258, + "learning_rate": 9.034551028498084e-07, + "loss": 0.6174, + "step": 2786 + }, + { + "epoch": 4.367954070981211, + "grad_norm": 1.4713473320007324, + "learning_rate": 9.018344717199887e-07, + "loss": 0.5183, + "step": 2787 + }, + { + "epoch": 4.369519832985386, + "grad_norm": 1.112297773361206, + "learning_rate": 9.002149755217246e-07, + "loss": 0.6369, + "step": 2788 + }, + { + "epoch": 4.3710855949895615, + "grad_norm": 1.2173497676849365, + "learning_rate": 8.985966154050952e-07, + "loss": 0.6195, + "step": 2789 + }, + { + "epoch": 4.372651356993737, + "grad_norm": 1.2108879089355469, + "learning_rate": 8.969793925193731e-07, + "loss": 0.6155, + "step": 2790 + }, + { + "epoch": 4.374217118997913, + "grad_norm": 1.409883975982666, + "learning_rate": 8.953633080130239e-07, + "loss": 0.5926, + "step": 2791 + }, + { + "epoch": 4.375782881002087, + "grad_norm": 1.2854543924331665, + "learning_rate": 8.937483630337038e-07, + "loss": 0.6337, + "step": 2792 + }, + { + "epoch": 4.377348643006263, + "grad_norm": 1.2990738153457642, + "learning_rate": 8.921345587282604e-07, + "loss": 0.6159, + "step": 2793 + }, + { + "epoch": 4.3789144050104385, + "grad_norm": 1.230802059173584, + "learning_rate": 8.905218962427312e-07, + "loss": 0.5945, + "step": 2794 + }, + { + "epoch": 4.380480167014614, + "grad_norm": 1.2086981534957886, + "learning_rate": 8.889103767223425e-07, + "loss": 0.5439, + "step": 2795 + }, + { + "epoch": 4.382045929018789, + "grad_norm": 1.215370774269104, + "learning_rate": 8.873000013115099e-07, + "loss": 0.6095, + "step": 2796 + }, + { + "epoch": 4.383611691022964, + "grad_norm": 1.2571933269500732, + "learning_rate": 8.856907711538335e-07, + "loss": 0.6036, + "step": 2797 + }, + { + "epoch": 4.38517745302714, + "grad_norm": 1.1642861366271973, + "learning_rate": 8.840826873921049e-07, + "loss": 0.582, + "step": 2798 + }, + { + "epoch": 4.3867432150313155, + "grad_norm": 1.23332941532135, + "learning_rate": 8.824757511682988e-07, + "loss": 0.6014, + "step": 2799 + }, + { + "epoch": 4.388308977035491, + "grad_norm": 1.2016581296920776, + "learning_rate": 8.80869963623573e-07, + "loss": 0.642, + "step": 2800 + }, + { + "epoch": 4.389874739039666, + "grad_norm": 1.2199851274490356, + "learning_rate": 8.792653258982747e-07, + "loss": 0.6413, + "step": 2801 + }, + { + "epoch": 4.391440501043841, + "grad_norm": 1.3119784593582153, + "learning_rate": 8.776618391319297e-07, + "loss": 0.5756, + "step": 2802 + }, + { + "epoch": 4.393006263048017, + "grad_norm": 1.267626404762268, + "learning_rate": 8.760595044632484e-07, + "loss": 0.5972, + "step": 2803 + }, + { + "epoch": 4.3945720250521925, + "grad_norm": 1.0666111707687378, + "learning_rate": 8.744583230301248e-07, + "loss": 0.6459, + "step": 2804 + }, + { + "epoch": 4.396137787056367, + "grad_norm": 1.276969313621521, + "learning_rate": 8.728582959696305e-07, + "loss": 0.567, + "step": 2805 + }, + { + "epoch": 4.397703549060543, + "grad_norm": 1.313031554222107, + "learning_rate": 8.712594244180198e-07, + "loss": 0.562, + "step": 2806 + }, + { + "epoch": 4.399269311064718, + "grad_norm": 1.164860486984253, + "learning_rate": 8.696617095107249e-07, + "loss": 0.5849, + "step": 2807 + }, + { + "epoch": 4.400835073068894, + "grad_norm": 1.2664166688919067, + "learning_rate": 8.680651523823583e-07, + "loss": 0.533, + "step": 2808 + }, + { + "epoch": 4.402400835073069, + "grad_norm": 1.2811753749847412, + "learning_rate": 8.664697541667085e-07, + "loss": 0.6285, + "step": 2809 + }, + { + "epoch": 4.403966597077244, + "grad_norm": 1.2239644527435303, + "learning_rate": 8.648755159967426e-07, + "loss": 0.5728, + "step": 2810 + }, + { + "epoch": 4.40553235908142, + "grad_norm": 1.2360845804214478, + "learning_rate": 8.632824390046027e-07, + "loss": 0.6006, + "step": 2811 + }, + { + "epoch": 4.407098121085595, + "grad_norm": 1.1378620862960815, + "learning_rate": 8.61690524321607e-07, + "loss": 0.5738, + "step": 2812 + }, + { + "epoch": 4.40866388308977, + "grad_norm": 1.2592755556106567, + "learning_rate": 8.600997730782476e-07, + "loss": 0.538, + "step": 2813 + }, + { + "epoch": 4.4102296450939455, + "grad_norm": 1.359745979309082, + "learning_rate": 8.585101864041914e-07, + "loss": 0.6385, + "step": 2814 + }, + { + "epoch": 4.411795407098121, + "grad_norm": 1.1939847469329834, + "learning_rate": 8.569217654282774e-07, + "loss": 0.6262, + "step": 2815 + }, + { + "epoch": 4.413361169102297, + "grad_norm": 1.2629926204681396, + "learning_rate": 8.553345112785172e-07, + "loss": 0.5972, + "step": 2816 + }, + { + "epoch": 4.414926931106471, + "grad_norm": 1.177193284034729, + "learning_rate": 8.537484250820943e-07, + "loss": 0.6511, + "step": 2817 + }, + { + "epoch": 4.416492693110647, + "grad_norm": 1.2521287202835083, + "learning_rate": 8.521635079653601e-07, + "loss": 0.5487, + "step": 2818 + }, + { + "epoch": 4.4180584551148225, + "grad_norm": 1.2183914184570312, + "learning_rate": 8.505797610538397e-07, + "loss": 0.6102, + "step": 2819 + }, + { + "epoch": 4.419624217118998, + "grad_norm": 1.4728264808654785, + "learning_rate": 8.48997185472226e-07, + "loss": 0.6378, + "step": 2820 + }, + { + "epoch": 4.421189979123174, + "grad_norm": 1.1505528688430786, + "learning_rate": 8.474157823443766e-07, + "loss": 0.6167, + "step": 2821 + }, + { + "epoch": 4.422755741127348, + "grad_norm": 1.1801762580871582, + "learning_rate": 8.458355527933224e-07, + "loss": 0.622, + "step": 2822 + }, + { + "epoch": 4.424321503131524, + "grad_norm": 1.3376151323318481, + "learning_rate": 8.442564979412557e-07, + "loss": 0.6606, + "step": 2823 + }, + { + "epoch": 4.4258872651356995, + "grad_norm": 1.3469598293304443, + "learning_rate": 8.426786189095362e-07, + "loss": 0.5997, + "step": 2824 + }, + { + "epoch": 4.427453027139875, + "grad_norm": 1.230717420578003, + "learning_rate": 8.411019168186921e-07, + "loss": 0.5861, + "step": 2825 + }, + { + "epoch": 4.42901878914405, + "grad_norm": 1.2221441268920898, + "learning_rate": 8.395263927884101e-07, + "loss": 0.6121, + "step": 2826 + }, + { + "epoch": 4.430584551148225, + "grad_norm": 1.409119963645935, + "learning_rate": 8.379520479375436e-07, + "loss": 0.5807, + "step": 2827 + }, + { + "epoch": 4.432150313152401, + "grad_norm": 1.076403260231018, + "learning_rate": 8.363788833841083e-07, + "loss": 0.6234, + "step": 2828 + }, + { + "epoch": 4.4337160751565765, + "grad_norm": 1.3182142972946167, + "learning_rate": 8.34806900245281e-07, + "loss": 0.6223, + "step": 2829 + }, + { + "epoch": 4.435281837160751, + "grad_norm": 1.265117883682251, + "learning_rate": 8.332360996374014e-07, + "loss": 0.5916, + "step": 2830 + }, + { + "epoch": 4.436847599164927, + "grad_norm": 1.243553876876831, + "learning_rate": 8.316664826759646e-07, + "loss": 0.6201, + "step": 2831 + }, + { + "epoch": 4.438413361169102, + "grad_norm": 1.3207454681396484, + "learning_rate": 8.300980504756318e-07, + "loss": 0.6287, + "step": 2832 + }, + { + "epoch": 4.439979123173278, + "grad_norm": 1.1349838972091675, + "learning_rate": 8.285308041502188e-07, + "loss": 0.5632, + "step": 2833 + }, + { + "epoch": 4.4415448851774535, + "grad_norm": 1.1458604335784912, + "learning_rate": 8.269647448126982e-07, + "loss": 0.6769, + "step": 2834 + }, + { + "epoch": 4.443110647181628, + "grad_norm": 1.2419075965881348, + "learning_rate": 8.253998735752044e-07, + "loss": 0.5475, + "step": 2835 + }, + { + "epoch": 4.444676409185804, + "grad_norm": 1.3041012287139893, + "learning_rate": 8.238361915490226e-07, + "loss": 0.5608, + "step": 2836 + }, + { + "epoch": 4.446242171189979, + "grad_norm": 1.1919783353805542, + "learning_rate": 8.222736998445963e-07, + "loss": 0.5819, + "step": 2837 + }, + { + "epoch": 4.447807933194154, + "grad_norm": 1.126298427581787, + "learning_rate": 8.207123995715254e-07, + "loss": 0.6019, + "step": 2838 + }, + { + "epoch": 4.44937369519833, + "grad_norm": 1.2007626295089722, + "learning_rate": 8.191522918385597e-07, + "loss": 0.606, + "step": 2839 + }, + { + "epoch": 4.450939457202505, + "grad_norm": 1.2129793167114258, + "learning_rate": 8.175933777536049e-07, + "loss": 0.5726, + "step": 2840 + }, + { + "epoch": 4.452505219206681, + "grad_norm": 1.2193784713745117, + "learning_rate": 8.160356584237184e-07, + "loss": 0.6005, + "step": 2841 + }, + { + "epoch": 4.454070981210856, + "grad_norm": 1.2187601327896118, + "learning_rate": 8.144791349551084e-07, + "loss": 0.6322, + "step": 2842 + }, + { + "epoch": 4.455636743215031, + "grad_norm": 1.3521336317062378, + "learning_rate": 8.129238084531355e-07, + "loss": 0.5851, + "step": 2843 + }, + { + "epoch": 4.457202505219207, + "grad_norm": 1.32852303981781, + "learning_rate": 8.113696800223084e-07, + "loss": 0.6288, + "step": 2844 + }, + { + "epoch": 4.458768267223382, + "grad_norm": 1.2852197885513306, + "learning_rate": 8.098167507662863e-07, + "loss": 0.6271, + "step": 2845 + }, + { + "epoch": 4.460334029227558, + "grad_norm": 1.3286292552947998, + "learning_rate": 8.08265021787876e-07, + "loss": 0.6584, + "step": 2846 + }, + { + "epoch": 4.461899791231732, + "grad_norm": 1.1803559064865112, + "learning_rate": 8.067144941890329e-07, + "loss": 0.6344, + "step": 2847 + }, + { + "epoch": 4.463465553235908, + "grad_norm": 1.2986726760864258, + "learning_rate": 8.051651690708581e-07, + "loss": 0.5878, + "step": 2848 + }, + { + "epoch": 4.465031315240084, + "grad_norm": 1.2154364585876465, + "learning_rate": 8.036170475335997e-07, + "loss": 0.6315, + "step": 2849 + }, + { + "epoch": 4.466597077244259, + "grad_norm": 1.2473022937774658, + "learning_rate": 8.020701306766504e-07, + "loss": 0.6432, + "step": 2850 + }, + { + "epoch": 4.468162839248434, + "grad_norm": 1.336057186126709, + "learning_rate": 8.005244195985487e-07, + "loss": 0.5991, + "step": 2851 + }, + { + "epoch": 4.469728601252609, + "grad_norm": 1.2605669498443604, + "learning_rate": 7.989799153969735e-07, + "loss": 0.595, + "step": 2852 + }, + { + "epoch": 4.471294363256785, + "grad_norm": 1.1553322076797485, + "learning_rate": 7.97436619168751e-07, + "loss": 0.6515, + "step": 2853 + }, + { + "epoch": 4.4728601252609606, + "grad_norm": 1.1928796768188477, + "learning_rate": 7.958945320098476e-07, + "loss": 0.5477, + "step": 2854 + }, + { + "epoch": 4.474425887265136, + "grad_norm": 1.1510995626449585, + "learning_rate": 7.943536550153689e-07, + "loss": 0.6296, + "step": 2855 + }, + { + "epoch": 4.475991649269311, + "grad_norm": 1.210997223854065, + "learning_rate": 7.928139892795661e-07, + "loss": 0.5907, + "step": 2856 + }, + { + "epoch": 4.477557411273486, + "grad_norm": 1.2773675918579102, + "learning_rate": 7.912755358958251e-07, + "loss": 0.5547, + "step": 2857 + }, + { + "epoch": 4.479123173277662, + "grad_norm": 1.248292326927185, + "learning_rate": 7.897382959566732e-07, + "loss": 0.6521, + "step": 2858 + }, + { + "epoch": 4.4806889352818375, + "grad_norm": 1.3389019966125488, + "learning_rate": 7.882022705537779e-07, + "loss": 0.6355, + "step": 2859 + }, + { + "epoch": 4.482254697286012, + "grad_norm": 1.1564635038375854, + "learning_rate": 7.866674607779401e-07, + "loss": 0.5526, + "step": 2860 + }, + { + "epoch": 4.483820459290188, + "grad_norm": 1.2170965671539307, + "learning_rate": 7.851338677191001e-07, + "loss": 0.5251, + "step": 2861 + }, + { + "epoch": 4.485386221294363, + "grad_norm": 1.2128232717514038, + "learning_rate": 7.836014924663338e-07, + "loss": 0.6117, + "step": 2862 + }, + { + "epoch": 4.486951983298539, + "grad_norm": 1.3145650625228882, + "learning_rate": 7.820703361078516e-07, + "loss": 0.5499, + "step": 2863 + }, + { + "epoch": 4.488517745302714, + "grad_norm": 1.2057058811187744, + "learning_rate": 7.805403997309991e-07, + "loss": 0.5714, + "step": 2864 + }, + { + "epoch": 4.490083507306889, + "grad_norm": 1.2229523658752441, + "learning_rate": 7.790116844222548e-07, + "loss": 0.6145, + "step": 2865 + }, + { + "epoch": 4.491649269311065, + "grad_norm": 1.3067467212677002, + "learning_rate": 7.77484191267231e-07, + "loss": 0.6141, + "step": 2866 + }, + { + "epoch": 4.49321503131524, + "grad_norm": 1.2613667249679565, + "learning_rate": 7.75957921350671e-07, + "loss": 0.6222, + "step": 2867 + }, + { + "epoch": 4.494780793319415, + "grad_norm": 1.1608632802963257, + "learning_rate": 7.744328757564501e-07, + "loss": 0.6279, + "step": 2868 + }, + { + "epoch": 4.496346555323591, + "grad_norm": 1.3013458251953125, + "learning_rate": 7.72909055567575e-07, + "loss": 0.5413, + "step": 2869 + }, + { + "epoch": 4.497912317327766, + "grad_norm": 1.3727405071258545, + "learning_rate": 7.713864618661787e-07, + "loss": 0.6053, + "step": 2870 + }, + { + "epoch": 4.499478079331942, + "grad_norm": 1.2287840843200684, + "learning_rate": 7.698650957335277e-07, + "loss": 0.6531, + "step": 2871 + }, + { + "epoch": 4.5010438413361165, + "grad_norm": 1.2294139862060547, + "learning_rate": 7.683449582500152e-07, + "loss": 0.6021, + "step": 2872 + }, + { + "epoch": 4.502609603340292, + "grad_norm": 1.23564875125885, + "learning_rate": 7.668260504951596e-07, + "loss": 0.6128, + "step": 2873 + }, + { + "epoch": 4.504175365344468, + "grad_norm": 1.1630370616912842, + "learning_rate": 7.653083735476088e-07, + "loss": 0.6183, + "step": 2874 + }, + { + "epoch": 4.505741127348643, + "grad_norm": 1.2798091173171997, + "learning_rate": 7.637919284851353e-07, + "loss": 0.6534, + "step": 2875 + }, + { + "epoch": 4.507306889352819, + "grad_norm": 1.2523179054260254, + "learning_rate": 7.622767163846376e-07, + "loss": 0.6357, + "step": 2876 + }, + { + "epoch": 4.508872651356993, + "grad_norm": 1.2298544645309448, + "learning_rate": 7.607627383221381e-07, + "loss": 0.5961, + "step": 2877 + }, + { + "epoch": 4.510438413361169, + "grad_norm": 1.2161098718643188, + "learning_rate": 7.592499953727828e-07, + "loss": 0.6332, + "step": 2878 + }, + { + "epoch": 4.512004175365345, + "grad_norm": 1.2523022890090942, + "learning_rate": 7.577384886108405e-07, + "loss": 0.6439, + "step": 2879 + }, + { + "epoch": 4.51356993736952, + "grad_norm": 1.30768883228302, + "learning_rate": 7.562282191097026e-07, + "loss": 0.6736, + "step": 2880 + }, + { + "epoch": 4.515135699373695, + "grad_norm": 1.2421822547912598, + "learning_rate": 7.547191879418817e-07, + "loss": 0.6004, + "step": 2881 + }, + { + "epoch": 4.51670146137787, + "grad_norm": 1.1340116262435913, + "learning_rate": 7.532113961790116e-07, + "loss": 0.6065, + "step": 2882 + }, + { + "epoch": 4.518267223382046, + "grad_norm": 1.319522738456726, + "learning_rate": 7.517048448918429e-07, + "loss": 0.5796, + "step": 2883 + }, + { + "epoch": 4.519832985386222, + "grad_norm": 1.1828511953353882, + "learning_rate": 7.501995351502497e-07, + "loss": 0.6417, + "step": 2884 + }, + { + "epoch": 4.521398747390396, + "grad_norm": 1.302408218383789, + "learning_rate": 7.486954680232231e-07, + "loss": 0.6234, + "step": 2885 + }, + { + "epoch": 4.522964509394572, + "grad_norm": 1.1842252016067505, + "learning_rate": 7.471926445788677e-07, + "loss": 0.6115, + "step": 2886 + }, + { + "epoch": 4.524530271398747, + "grad_norm": 1.2214895486831665, + "learning_rate": 7.456910658844119e-07, + "loss": 0.6104, + "step": 2887 + }, + { + "epoch": 4.526096033402923, + "grad_norm": 1.2608014345169067, + "learning_rate": 7.44190733006194e-07, + "loss": 0.5499, + "step": 2888 + }, + { + "epoch": 4.527661795407099, + "grad_norm": 1.259846806526184, + "learning_rate": 7.426916470096701e-07, + "loss": 0.6065, + "step": 2889 + }, + { + "epoch": 4.529227557411273, + "grad_norm": 1.4206489324569702, + "learning_rate": 7.411938089594131e-07, + "loss": 0.6266, + "step": 2890 + }, + { + "epoch": 4.530793319415449, + "grad_norm": 1.0867127180099487, + "learning_rate": 7.396972199191052e-07, + "loss": 0.6088, + "step": 2891 + }, + { + "epoch": 4.532359081419624, + "grad_norm": 1.2204397916793823, + "learning_rate": 7.38201880951544e-07, + "loss": 0.6255, + "step": 2892 + }, + { + "epoch": 4.533924843423799, + "grad_norm": 1.2188392877578735, + "learning_rate": 7.367077931186397e-07, + "loss": 0.6001, + "step": 2893 + }, + { + "epoch": 4.535490605427975, + "grad_norm": 1.22636079788208, + "learning_rate": 7.352149574814132e-07, + "loss": 0.6662, + "step": 2894 + }, + { + "epoch": 4.53705636743215, + "grad_norm": 1.3532042503356934, + "learning_rate": 7.337233750999964e-07, + "loss": 0.6121, + "step": 2895 + }, + { + "epoch": 4.538622129436326, + "grad_norm": 1.2929681539535522, + "learning_rate": 7.322330470336314e-07, + "loss": 0.5343, + "step": 2896 + }, + { + "epoch": 4.540187891440501, + "grad_norm": 1.243160605430603, + "learning_rate": 7.307439743406691e-07, + "loss": 0.5873, + "step": 2897 + }, + { + "epoch": 4.541753653444676, + "grad_norm": 1.190952181816101, + "learning_rate": 7.292561580785696e-07, + "loss": 0.648, + "step": 2898 + }, + { + "epoch": 4.543319415448852, + "grad_norm": 1.27082359790802, + "learning_rate": 7.277695993038996e-07, + "loss": 0.5988, + "step": 2899 + }, + { + "epoch": 4.544885177453027, + "grad_norm": 1.2360665798187256, + "learning_rate": 7.26284299072334e-07, + "loss": 0.607, + "step": 2900 + }, + { + "epoch": 4.546450939457203, + "grad_norm": 1.1328120231628418, + "learning_rate": 7.248002584386535e-07, + "loss": 0.658, + "step": 2901 + }, + { + "epoch": 4.5480167014613775, + "grad_norm": 1.173593282699585, + "learning_rate": 7.233174784567443e-07, + "loss": 0.6941, + "step": 2902 + }, + { + "epoch": 4.549582463465553, + "grad_norm": 1.2456272840499878, + "learning_rate": 7.218359601795979e-07, + "loss": 0.6372, + "step": 2903 + }, + { + "epoch": 4.551148225469729, + "grad_norm": 1.2233866453170776, + "learning_rate": 7.203557046593068e-07, + "loss": 0.5841, + "step": 2904 + }, + { + "epoch": 4.552713987473904, + "grad_norm": 1.4194185733795166, + "learning_rate": 7.188767129470717e-07, + "loss": 0.6086, + "step": 2905 + }, + { + "epoch": 4.554279749478079, + "grad_norm": 1.1789617538452148, + "learning_rate": 7.173989860931935e-07, + "loss": 0.5785, + "step": 2906 + }, + { + "epoch": 4.5558455114822545, + "grad_norm": 1.2355077266693115, + "learning_rate": 7.15922525147072e-07, + "loss": 0.6122, + "step": 2907 + }, + { + "epoch": 4.55741127348643, + "grad_norm": 1.1994844675064087, + "learning_rate": 7.144473311572136e-07, + "loss": 0.6097, + "step": 2908 + }, + { + "epoch": 4.558977035490606, + "grad_norm": 1.4132137298583984, + "learning_rate": 7.129734051712203e-07, + "loss": 0.6377, + "step": 2909 + }, + { + "epoch": 4.560542797494781, + "grad_norm": 1.1752851009368896, + "learning_rate": 7.115007482357953e-07, + "loss": 0.649, + "step": 2910 + }, + { + "epoch": 4.562108559498956, + "grad_norm": 1.2555209398269653, + "learning_rate": 7.100293613967424e-07, + "loss": 0.6313, + "step": 2911 + }, + { + "epoch": 4.5636743215031315, + "grad_norm": 1.2966251373291016, + "learning_rate": 7.085592456989599e-07, + "loss": 0.5969, + "step": 2912 + }, + { + "epoch": 4.565240083507307, + "grad_norm": 1.2769136428833008, + "learning_rate": 7.070904021864461e-07, + "loss": 0.6198, + "step": 2913 + }, + { + "epoch": 4.566805845511483, + "grad_norm": 1.304149866104126, + "learning_rate": 7.056228319022945e-07, + "loss": 0.4576, + "step": 2914 + }, + { + "epoch": 4.568371607515657, + "grad_norm": 1.3524707555770874, + "learning_rate": 7.041565358886954e-07, + "loss": 0.6173, + "step": 2915 + }, + { + "epoch": 4.569937369519833, + "grad_norm": 1.2801849842071533, + "learning_rate": 7.026915151869335e-07, + "loss": 0.6357, + "step": 2916 + }, + { + "epoch": 4.5715031315240084, + "grad_norm": 1.234526515007019, + "learning_rate": 7.012277708373877e-07, + "loss": 0.5631, + "step": 2917 + }, + { + "epoch": 4.573068893528184, + "grad_norm": 1.1721588373184204, + "learning_rate": 6.997653038795316e-07, + "loss": 0.5776, + "step": 2918 + }, + { + "epoch": 4.574634655532359, + "grad_norm": 1.2359575033187866, + "learning_rate": 6.983041153519302e-07, + "loss": 0.5767, + "step": 2919 + }, + { + "epoch": 4.576200417536534, + "grad_norm": 1.2173316478729248, + "learning_rate": 6.968442062922415e-07, + "loss": 0.6165, + "step": 2920 + }, + { + "epoch": 4.57776617954071, + "grad_norm": 1.3483846187591553, + "learning_rate": 6.953855777372157e-07, + "loss": 0.5667, + "step": 2921 + }, + { + "epoch": 4.579331941544885, + "grad_norm": 1.376077651977539, + "learning_rate": 6.939282307226908e-07, + "loss": 0.6666, + "step": 2922 + }, + { + "epoch": 4.580897703549061, + "grad_norm": 1.2124888896942139, + "learning_rate": 6.924721662835971e-07, + "loss": 0.4712, + "step": 2923 + }, + { + "epoch": 4.582463465553236, + "grad_norm": 1.1178332567214966, + "learning_rate": 6.910173854539551e-07, + "loss": 0.5903, + "step": 2924 + }, + { + "epoch": 4.584029227557411, + "grad_norm": 1.1721351146697998, + "learning_rate": 6.895638892668708e-07, + "loss": 0.6249, + "step": 2925 + }, + { + "epoch": 4.585594989561587, + "grad_norm": 1.2229433059692383, + "learning_rate": 6.881116787545394e-07, + "loss": 0.6408, + "step": 2926 + }, + { + "epoch": 4.5871607515657615, + "grad_norm": 1.378939151763916, + "learning_rate": 6.866607549482435e-07, + "loss": 0.6422, + "step": 2927 + }, + { + "epoch": 4.588726513569937, + "grad_norm": 1.3970625400543213, + "learning_rate": 6.852111188783512e-07, + "loss": 0.5783, + "step": 2928 + }, + { + "epoch": 4.590292275574113, + "grad_norm": 1.2715126276016235, + "learning_rate": 6.837627715743165e-07, + "loss": 0.607, + "step": 2929 + }, + { + "epoch": 4.591858037578288, + "grad_norm": 1.2084861993789673, + "learning_rate": 6.82315714064678e-07, + "loss": 0.5667, + "step": 2930 + }, + { + "epoch": 4.593423799582464, + "grad_norm": 1.2971402406692505, + "learning_rate": 6.808699473770586e-07, + "loss": 0.6281, + "step": 2931 + }, + { + "epoch": 4.5949895615866385, + "grad_norm": 1.2204023599624634, + "learning_rate": 6.794254725381641e-07, + "loss": 0.6225, + "step": 2932 + }, + { + "epoch": 4.596555323590814, + "grad_norm": 1.2303305864334106, + "learning_rate": 6.779822905737834e-07, + "loss": 0.6175, + "step": 2933 + }, + { + "epoch": 4.59812108559499, + "grad_norm": 1.252663254737854, + "learning_rate": 6.765404025087876e-07, + "loss": 0.6163, + "step": 2934 + }, + { + "epoch": 4.599686847599165, + "grad_norm": 1.3693299293518066, + "learning_rate": 6.750998093671263e-07, + "loss": 0.6057, + "step": 2935 + }, + { + "epoch": 4.60125260960334, + "grad_norm": 1.2067402601242065, + "learning_rate": 6.736605121718337e-07, + "loss": 0.6249, + "step": 2936 + }, + { + "epoch": 4.6028183716075155, + "grad_norm": 1.249069333076477, + "learning_rate": 6.722225119450216e-07, + "loss": 0.6229, + "step": 2937 + }, + { + "epoch": 4.604384133611691, + "grad_norm": 1.3225828409194946, + "learning_rate": 6.707858097078784e-07, + "loss": 0.6726, + "step": 2938 + }, + { + "epoch": 4.605949895615867, + "grad_norm": 1.2723355293273926, + "learning_rate": 6.69350406480676e-07, + "loss": 0.589, + "step": 2939 + }, + { + "epoch": 4.607515657620041, + "grad_norm": 1.2256345748901367, + "learning_rate": 6.679163032827593e-07, + "loss": 0.5873, + "step": 2940 + }, + { + "epoch": 4.609081419624217, + "grad_norm": 1.1579097509384155, + "learning_rate": 6.664835011325508e-07, + "loss": 0.616, + "step": 2941 + }, + { + "epoch": 4.6106471816283925, + "grad_norm": 1.2893751859664917, + "learning_rate": 6.650520010475525e-07, + "loss": 0.6022, + "step": 2942 + }, + { + "epoch": 4.612212943632568, + "grad_norm": 1.3577022552490234, + "learning_rate": 6.636218040443366e-07, + "loss": 0.6644, + "step": 2943 + }, + { + "epoch": 4.613778705636744, + "grad_norm": 1.2338060140609741, + "learning_rate": 6.621929111385533e-07, + "loss": 0.6048, + "step": 2944 + }, + { + "epoch": 4.615344467640918, + "grad_norm": 1.2044262886047363, + "learning_rate": 6.607653233449257e-07, + "loss": 0.6546, + "step": 2945 + }, + { + "epoch": 4.616910229645094, + "grad_norm": 1.4120864868164062, + "learning_rate": 6.593390416772505e-07, + "loss": 0.6433, + "step": 2946 + }, + { + "epoch": 4.6184759916492695, + "grad_norm": 1.2405333518981934, + "learning_rate": 6.57914067148396e-07, + "loss": 0.5755, + "step": 2947 + }, + { + "epoch": 4.620041753653445, + "grad_norm": 1.2720845937728882, + "learning_rate": 6.564904007703032e-07, + "loss": 0.5998, + "step": 2948 + }, + { + "epoch": 4.62160751565762, + "grad_norm": 1.316292405128479, + "learning_rate": 6.550680435539836e-07, + "loss": 0.5557, + "step": 2949 + }, + { + "epoch": 4.623173277661795, + "grad_norm": 1.274463176727295, + "learning_rate": 6.536469965095193e-07, + "loss": 0.5593, + "step": 2950 + }, + { + "epoch": 4.624739039665971, + "grad_norm": 1.177530288696289, + "learning_rate": 6.522272606460614e-07, + "loss": 0.6317, + "step": 2951 + }, + { + "epoch": 4.6263048016701465, + "grad_norm": 1.313736915588379, + "learning_rate": 6.508088369718305e-07, + "loss": 0.652, + "step": 2952 + }, + { + "epoch": 4.627870563674321, + "grad_norm": 1.2262637615203857, + "learning_rate": 6.493917264941149e-07, + "loss": 0.5847, + "step": 2953 + }, + { + "epoch": 4.629436325678497, + "grad_norm": 1.1686118841171265, + "learning_rate": 6.479759302192712e-07, + "loss": 0.6166, + "step": 2954 + }, + { + "epoch": 4.631002087682672, + "grad_norm": 1.2170262336730957, + "learning_rate": 6.46561449152722e-07, + "loss": 0.6207, + "step": 2955 + }, + { + "epoch": 4.632567849686848, + "grad_norm": 1.2834949493408203, + "learning_rate": 6.45148284298954e-07, + "loss": 0.5227, + "step": 2956 + }, + { + "epoch": 4.6341336116910234, + "grad_norm": 1.237107515335083, + "learning_rate": 6.43736436661524e-07, + "loss": 0.6574, + "step": 2957 + }, + { + "epoch": 4.635699373695198, + "grad_norm": 1.2491086721420288, + "learning_rate": 6.423259072430496e-07, + "loss": 0.6345, + "step": 2958 + }, + { + "epoch": 4.637265135699374, + "grad_norm": 1.2701539993286133, + "learning_rate": 6.409166970452118e-07, + "loss": 0.5826, + "step": 2959 + }, + { + "epoch": 4.638830897703549, + "grad_norm": 1.3169174194335938, + "learning_rate": 6.395088070687588e-07, + "loss": 0.6026, + "step": 2960 + }, + { + "epoch": 4.640396659707724, + "grad_norm": 1.1995527744293213, + "learning_rate": 6.381022383134966e-07, + "loss": 0.6577, + "step": 2961 + }, + { + "epoch": 4.6419624217118995, + "grad_norm": 1.2342932224273682, + "learning_rate": 6.36696991778295e-07, + "loss": 0.6074, + "step": 2962 + }, + { + "epoch": 4.643528183716075, + "grad_norm": 1.2346426248550415, + "learning_rate": 6.352930684610867e-07, + "loss": 0.594, + "step": 2963 + }, + { + "epoch": 4.645093945720251, + "grad_norm": 1.1832835674285889, + "learning_rate": 6.33890469358861e-07, + "loss": 0.6149, + "step": 2964 + }, + { + "epoch": 4.646659707724426, + "grad_norm": 1.2718520164489746, + "learning_rate": 6.324891954676693e-07, + "loss": 0.5602, + "step": 2965 + }, + { + "epoch": 4.648225469728601, + "grad_norm": 1.2307515144348145, + "learning_rate": 6.310892477826211e-07, + "loss": 0.6094, + "step": 2966 + }, + { + "epoch": 4.6497912317327765, + "grad_norm": 1.3244339227676392, + "learning_rate": 6.296906272978845e-07, + "loss": 0.6468, + "step": 2967 + }, + { + "epoch": 4.651356993736952, + "grad_norm": 1.3322306871414185, + "learning_rate": 6.282933350066853e-07, + "loss": 0.5508, + "step": 2968 + }, + { + "epoch": 4.652922755741128, + "grad_norm": 1.316904067993164, + "learning_rate": 6.268973719013036e-07, + "loss": 0.641, + "step": 2969 + }, + { + "epoch": 4.654488517745302, + "grad_norm": 1.1402695178985596, + "learning_rate": 6.2550273897308e-07, + "loss": 0.5811, + "step": 2970 + }, + { + "epoch": 4.656054279749478, + "grad_norm": 1.331150770187378, + "learning_rate": 6.241094372124079e-07, + "loss": 0.6364, + "step": 2971 + }, + { + "epoch": 4.6576200417536535, + "grad_norm": 1.3078680038452148, + "learning_rate": 6.227174676087333e-07, + "loss": 0.595, + "step": 2972 + }, + { + "epoch": 4.659185803757829, + "grad_norm": 1.1026049852371216, + "learning_rate": 6.213268311505616e-07, + "loss": 0.6616, + "step": 2973 + }, + { + "epoch": 4.660751565762004, + "grad_norm": 1.2403117418289185, + "learning_rate": 6.199375288254461e-07, + "loss": 0.5761, + "step": 2974 + }, + { + "epoch": 4.662317327766179, + "grad_norm": 1.258516788482666, + "learning_rate": 6.185495616199949e-07, + "loss": 0.5506, + "step": 2975 + }, + { + "epoch": 4.663883089770355, + "grad_norm": 1.2604827880859375, + "learning_rate": 6.171629305198698e-07, + "loss": 0.5802, + "step": 2976 + }, + { + "epoch": 4.6654488517745305, + "grad_norm": 1.1909523010253906, + "learning_rate": 6.157776365097803e-07, + "loss": 0.6558, + "step": 2977 + }, + { + "epoch": 4.667014613778706, + "grad_norm": 1.2381128072738647, + "learning_rate": 6.143936805734881e-07, + "loss": 0.6345, + "step": 2978 + }, + { + "epoch": 4.668580375782881, + "grad_norm": 1.3379584550857544, + "learning_rate": 6.130110636938055e-07, + "loss": 0.5533, + "step": 2979 + }, + { + "epoch": 4.670146137787056, + "grad_norm": 1.335717797279358, + "learning_rate": 6.11629786852592e-07, + "loss": 0.6319, + "step": 2980 + }, + { + "epoch": 4.671711899791232, + "grad_norm": 1.246821641921997, + "learning_rate": 6.102498510307569e-07, + "loss": 0.6484, + "step": 2981 + }, + { + "epoch": 4.673277661795407, + "grad_norm": 1.2727938890457153, + "learning_rate": 6.088712572082566e-07, + "loss": 0.6236, + "step": 2982 + }, + { + "epoch": 4.674843423799582, + "grad_norm": 1.0983227491378784, + "learning_rate": 6.074940063640946e-07, + "loss": 0.595, + "step": 2983 + }, + { + "epoch": 4.676409185803758, + "grad_norm": 1.2299860715866089, + "learning_rate": 6.061180994763211e-07, + "loss": 0.592, + "step": 2984 + }, + { + "epoch": 4.677974947807933, + "grad_norm": 1.316436529159546, + "learning_rate": 6.047435375220306e-07, + "loss": 0.5987, + "step": 2985 + }, + { + "epoch": 4.679540709812109, + "grad_norm": 1.2510743141174316, + "learning_rate": 6.03370321477364e-07, + "loss": 0.6229, + "step": 2986 + }, + { + "epoch": 4.681106471816284, + "grad_norm": 1.2618223428726196, + "learning_rate": 6.019984523175057e-07, + "loss": 0.6101, + "step": 2987 + }, + { + "epoch": 4.682672233820459, + "grad_norm": 1.23671293258667, + "learning_rate": 6.006279310166835e-07, + "loss": 0.6364, + "step": 2988 + }, + { + "epoch": 4.684237995824635, + "grad_norm": 1.2197179794311523, + "learning_rate": 5.99258758548169e-07, + "loss": 0.5805, + "step": 2989 + }, + { + "epoch": 4.68580375782881, + "grad_norm": 1.2514771223068237, + "learning_rate": 5.978909358842732e-07, + "loss": 0.6141, + "step": 2990 + }, + { + "epoch": 4.687369519832985, + "grad_norm": 1.2339932918548584, + "learning_rate": 5.965244639963524e-07, + "loss": 0.5687, + "step": 2991 + }, + { + "epoch": 4.688935281837161, + "grad_norm": 1.3678441047668457, + "learning_rate": 5.95159343854802e-07, + "loss": 0.6436, + "step": 2992 + }, + { + "epoch": 4.690501043841336, + "grad_norm": 1.1420538425445557, + "learning_rate": 5.937955764290549e-07, + "loss": 0.6132, + "step": 2993 + }, + { + "epoch": 4.692066805845512, + "grad_norm": 1.3040863275527954, + "learning_rate": 5.924331626875885e-07, + "loss": 0.6149, + "step": 2994 + }, + { + "epoch": 4.693632567849686, + "grad_norm": 1.190412998199463, + "learning_rate": 5.910721035979139e-07, + "loss": 0.5449, + "step": 2995 + }, + { + "epoch": 4.695198329853862, + "grad_norm": 1.354242205619812, + "learning_rate": 5.897124001265822e-07, + "loss": 0.5768, + "step": 2996 + }, + { + "epoch": 4.696764091858038, + "grad_norm": 1.3582462072372437, + "learning_rate": 5.883540532391846e-07, + "loss": 0.5956, + "step": 2997 + }, + { + "epoch": 4.698329853862213, + "grad_norm": 1.1748749017715454, + "learning_rate": 5.869970639003439e-07, + "loss": 0.5381, + "step": 2998 + }, + { + "epoch": 4.699895615866389, + "grad_norm": 1.226905345916748, + "learning_rate": 5.85641433073722e-07, + "loss": 0.6287, + "step": 2999 + }, + { + "epoch": 4.701461377870563, + "grad_norm": 1.3289525508880615, + "learning_rate": 5.842871617220153e-07, + "loss": 0.6064, + "step": 3000 + }, + { + "epoch": 4.703027139874739, + "grad_norm": 1.2593392133712769, + "learning_rate": 5.829342508069546e-07, + "loss": 0.6537, + "step": 3001 + }, + { + "epoch": 4.7045929018789145, + "grad_norm": 1.2690147161483765, + "learning_rate": 5.81582701289305e-07, + "loss": 0.6201, + "step": 3002 + }, + { + "epoch": 4.70615866388309, + "grad_norm": 1.2982971668243408, + "learning_rate": 5.802325141288643e-07, + "loss": 0.5869, + "step": 3003 + }, + { + "epoch": 4.707724425887265, + "grad_norm": 1.2125791311264038, + "learning_rate": 5.788836902844633e-07, + "loss": 0.6744, + "step": 3004 + }, + { + "epoch": 4.70929018789144, + "grad_norm": 1.2710962295532227, + "learning_rate": 5.775362307139645e-07, + "loss": 0.5961, + "step": 3005 + }, + { + "epoch": 4.710855949895616, + "grad_norm": 1.3088948726654053, + "learning_rate": 5.761901363742614e-07, + "loss": 0.5572, + "step": 3006 + }, + { + "epoch": 4.7124217118997915, + "grad_norm": 1.2394931316375732, + "learning_rate": 5.74845408221279e-07, + "loss": 0.6722, + "step": 3007 + }, + { + "epoch": 4.713987473903966, + "grad_norm": 1.390460729598999, + "learning_rate": 5.735020472099687e-07, + "loss": 0.6166, + "step": 3008 + }, + { + "epoch": 4.715553235908142, + "grad_norm": 1.2904356718063354, + "learning_rate": 5.721600542943159e-07, + "loss": 0.5555, + "step": 3009 + }, + { + "epoch": 4.717118997912317, + "grad_norm": 1.1953577995300293, + "learning_rate": 5.708194304273323e-07, + "loss": 0.5141, + "step": 3010 + }, + { + "epoch": 4.718684759916493, + "grad_norm": 1.3318582773208618, + "learning_rate": 5.694801765610547e-07, + "loss": 0.557, + "step": 3011 + }, + { + "epoch": 4.7202505219206685, + "grad_norm": 1.344522476196289, + "learning_rate": 5.681422936465522e-07, + "loss": 0.5855, + "step": 3012 + }, + { + "epoch": 4.721816283924843, + "grad_norm": 1.270960807800293, + "learning_rate": 5.66805782633916e-07, + "loss": 0.6152, + "step": 3013 + }, + { + "epoch": 4.723382045929019, + "grad_norm": 1.4696820974349976, + "learning_rate": 5.654706444722646e-07, + "loss": 0.5986, + "step": 3014 + }, + { + "epoch": 4.724947807933194, + "grad_norm": 1.2178717851638794, + "learning_rate": 5.641368801097424e-07, + "loss": 0.653, + "step": 3015 + }, + { + "epoch": 4.726513569937369, + "grad_norm": 1.2786749601364136, + "learning_rate": 5.628044904935168e-07, + "loss": 0.6124, + "step": 3016 + }, + { + "epoch": 4.728079331941545, + "grad_norm": 1.2376960515975952, + "learning_rate": 5.614734765697799e-07, + "loss": 0.6467, + "step": 3017 + }, + { + "epoch": 4.72964509394572, + "grad_norm": 1.2517948150634766, + "learning_rate": 5.60143839283746e-07, + "loss": 0.6772, + "step": 3018 + }, + { + "epoch": 4.731210855949896, + "grad_norm": 1.1662477254867554, + "learning_rate": 5.588155795796533e-07, + "loss": 0.6071, + "step": 3019 + }, + { + "epoch": 4.732776617954071, + "grad_norm": 1.2507755756378174, + "learning_rate": 5.574886984007602e-07, + "loss": 0.5908, + "step": 3020 + }, + { + "epoch": 4.734342379958246, + "grad_norm": 1.1487728357315063, + "learning_rate": 5.561631966893452e-07, + "loss": 0.5582, + "step": 3021 + }, + { + "epoch": 4.735908141962422, + "grad_norm": 1.185463786125183, + "learning_rate": 5.548390753867106e-07, + "loss": 0.6161, + "step": 3022 + }, + { + "epoch": 4.737473903966597, + "grad_norm": 1.255568265914917, + "learning_rate": 5.535163354331763e-07, + "loss": 0.6313, + "step": 3023 + }, + { + "epoch": 4.739039665970773, + "grad_norm": 1.231364130973816, + "learning_rate": 5.521949777680796e-07, + "loss": 0.6123, + "step": 3024 + }, + { + "epoch": 4.740605427974947, + "grad_norm": 1.2934824228286743, + "learning_rate": 5.508750033297805e-07, + "loss": 0.6017, + "step": 3025 + }, + { + "epoch": 4.742171189979123, + "grad_norm": 1.20609450340271, + "learning_rate": 5.495564130556519e-07, + "loss": 0.586, + "step": 3026 + }, + { + "epoch": 4.743736951983299, + "grad_norm": 1.3231621980667114, + "learning_rate": 5.482392078820864e-07, + "loss": 0.5804, + "step": 3027 + }, + { + "epoch": 4.745302713987474, + "grad_norm": 1.168235182762146, + "learning_rate": 5.469233887444941e-07, + "loss": 0.602, + "step": 3028 + }, + { + "epoch": 4.746868475991649, + "grad_norm": 1.235970377922058, + "learning_rate": 5.456089565772976e-07, + "loss": 0.5859, + "step": 3029 + }, + { + "epoch": 4.748434237995824, + "grad_norm": 1.3735214471817017, + "learning_rate": 5.442959123139369e-07, + "loss": 0.4943, + "step": 3030 + }, + { + "epoch": 4.75, + "grad_norm": 1.2391469478607178, + "learning_rate": 5.429842568868654e-07, + "loss": 0.5896, + "step": 3031 + }, + { + "epoch": 4.751565762004176, + "grad_norm": 1.3225210905075073, + "learning_rate": 5.41673991227551e-07, + "loss": 0.5941, + "step": 3032 + }, + { + "epoch": 4.753131524008351, + "grad_norm": 1.2474992275238037, + "learning_rate": 5.403651162664738e-07, + "loss": 0.6647, + "step": 3033 + }, + { + "epoch": 4.754697286012526, + "grad_norm": 1.2459207773208618, + "learning_rate": 5.390576329331267e-07, + "loss": 0.5771, + "step": 3034 + }, + { + "epoch": 4.756263048016701, + "grad_norm": 1.2632790803909302, + "learning_rate": 5.377515421560147e-07, + "loss": 0.6667, + "step": 3035 + }, + { + "epoch": 4.757828810020877, + "grad_norm": 1.316978096961975, + "learning_rate": 5.36446844862653e-07, + "loss": 0.5392, + "step": 3036 + }, + { + "epoch": 4.759394572025053, + "grad_norm": 1.3076261281967163, + "learning_rate": 5.351435419795681e-07, + "loss": 0.6489, + "step": 3037 + }, + { + "epoch": 4.760960334029227, + "grad_norm": 1.197774052619934, + "learning_rate": 5.338416344322958e-07, + "loss": 0.5918, + "step": 3038 + }, + { + "epoch": 4.762526096033403, + "grad_norm": 1.1374410390853882, + "learning_rate": 5.325411231453811e-07, + "loss": 0.6153, + "step": 3039 + }, + { + "epoch": 4.764091858037578, + "grad_norm": 1.2959661483764648, + "learning_rate": 5.312420090423775e-07, + "loss": 0.487, + "step": 3040 + }, + { + "epoch": 4.765657620041754, + "grad_norm": 1.263738989830017, + "learning_rate": 5.299442930458468e-07, + "loss": 0.5549, + "step": 3041 + }, + { + "epoch": 4.767223382045929, + "grad_norm": 1.333878755569458, + "learning_rate": 5.286479760773553e-07, + "loss": 0.6261, + "step": 3042 + }, + { + "epoch": 4.768789144050104, + "grad_norm": 1.3264974355697632, + "learning_rate": 5.273530590574802e-07, + "loss": 0.5951, + "step": 3043 + }, + { + "epoch": 4.77035490605428, + "grad_norm": 1.252334475517273, + "learning_rate": 5.260595429058021e-07, + "loss": 0.5427, + "step": 3044 + }, + { + "epoch": 4.771920668058455, + "grad_norm": 1.233492374420166, + "learning_rate": 5.247674285409046e-07, + "loss": 0.6738, + "step": 3045 + }, + { + "epoch": 4.773486430062631, + "grad_norm": 1.2096290588378906, + "learning_rate": 5.234767168803815e-07, + "loss": 0.5748, + "step": 3046 + }, + { + "epoch": 4.775052192066806, + "grad_norm": 1.2133762836456299, + "learning_rate": 5.221874088408241e-07, + "loss": 0.6234, + "step": 3047 + }, + { + "epoch": 4.776617954070981, + "grad_norm": 1.2251299619674683, + "learning_rate": 5.208995053378305e-07, + "loss": 0.6235, + "step": 3048 + }, + { + "epoch": 4.778183716075157, + "grad_norm": 1.2990975379943848, + "learning_rate": 5.196130072860028e-07, + "loss": 0.6395, + "step": 3049 + }, + { + "epoch": 4.7797494780793315, + "grad_norm": 1.2998696565628052, + "learning_rate": 5.183279155989409e-07, + "loss": 0.6271, + "step": 3050 + }, + { + "epoch": 4.781315240083507, + "grad_norm": 1.2194361686706543, + "learning_rate": 5.170442311892488e-07, + "loss": 0.6563, + "step": 3051 + }, + { + "epoch": 4.782881002087683, + "grad_norm": 1.3821429014205933, + "learning_rate": 5.1576195496853e-07, + "loss": 0.6717, + "step": 3052 + }, + { + "epoch": 4.784446764091858, + "grad_norm": 1.3171210289001465, + "learning_rate": 5.144810878473891e-07, + "loss": 0.5997, + "step": 3053 + }, + { + "epoch": 4.786012526096034, + "grad_norm": 1.251981258392334, + "learning_rate": 5.132016307354287e-07, + "loss": 0.6034, + "step": 3054 + }, + { + "epoch": 4.7875782881002085, + "grad_norm": 1.3084864616394043, + "learning_rate": 5.119235845412512e-07, + "loss": 0.5528, + "step": 3055 + }, + { + "epoch": 4.789144050104384, + "grad_norm": 1.2730029821395874, + "learning_rate": 5.106469501724564e-07, + "loss": 0.5746, + "step": 3056 + }, + { + "epoch": 4.79070981210856, + "grad_norm": 1.2262660264968872, + "learning_rate": 5.093717285356414e-07, + "loss": 0.6208, + "step": 3057 + }, + { + "epoch": 4.792275574112735, + "grad_norm": 1.2117376327514648, + "learning_rate": 5.080979205364006e-07, + "loss": 0.5987, + "step": 3058 + }, + { + "epoch": 4.79384133611691, + "grad_norm": 1.2486417293548584, + "learning_rate": 5.068255270793245e-07, + "loss": 0.5414, + "step": 3059 + }, + { + "epoch": 4.7954070981210855, + "grad_norm": 1.231568455696106, + "learning_rate": 5.055545490679981e-07, + "loss": 0.6141, + "step": 3060 + }, + { + "epoch": 4.796972860125261, + "grad_norm": 1.3257794380187988, + "learning_rate": 5.042849874050012e-07, + "loss": 0.6121, + "step": 3061 + }, + { + "epoch": 4.798538622129437, + "grad_norm": 1.2174944877624512, + "learning_rate": 5.030168429919108e-07, + "loss": 0.6314, + "step": 3062 + }, + { + "epoch": 4.800104384133611, + "grad_norm": 1.3967257738113403, + "learning_rate": 5.017501167292931e-07, + "loss": 0.614, + "step": 3063 + }, + { + "epoch": 4.801670146137787, + "grad_norm": 1.0998293161392212, + "learning_rate": 5.004848095167103e-07, + "loss": 0.614, + "step": 3064 + }, + { + "epoch": 4.803235908141962, + "grad_norm": 1.228244662284851, + "learning_rate": 4.99220922252715e-07, + "loss": 0.5525, + "step": 3065 + }, + { + "epoch": 4.804801670146138, + "grad_norm": 1.257345199584961, + "learning_rate": 4.979584558348529e-07, + "loss": 0.6794, + "step": 3066 + }, + { + "epoch": 4.806367432150314, + "grad_norm": 1.240714192390442, + "learning_rate": 4.966974111596598e-07, + "loss": 0.626, + "step": 3067 + }, + { + "epoch": 4.807933194154488, + "grad_norm": 1.2045975923538208, + "learning_rate": 4.954377891226623e-07, + "loss": 0.6175, + "step": 3068 + }, + { + "epoch": 4.809498956158664, + "grad_norm": 1.317455530166626, + "learning_rate": 4.94179590618376e-07, + "loss": 0.5919, + "step": 3069 + }, + { + "epoch": 4.811064718162839, + "grad_norm": 1.2258387804031372, + "learning_rate": 4.929228165403066e-07, + "loss": 0.6348, + "step": 3070 + }, + { + "epoch": 4.812630480167015, + "grad_norm": 1.1854313611984253, + "learning_rate": 4.916674677809475e-07, + "loss": 0.6007, + "step": 3071 + }, + { + "epoch": 4.81419624217119, + "grad_norm": 1.3538085222244263, + "learning_rate": 4.904135452317807e-07, + "loss": 0.5999, + "step": 3072 + }, + { + "epoch": 4.815762004175365, + "grad_norm": 1.1934870481491089, + "learning_rate": 4.89161049783273e-07, + "loss": 0.5818, + "step": 3073 + }, + { + "epoch": 4.817327766179541, + "grad_norm": 1.3245843648910522, + "learning_rate": 4.879099823248814e-07, + "loss": 0.6442, + "step": 3074 + }, + { + "epoch": 4.818893528183716, + "grad_norm": 1.1364349126815796, + "learning_rate": 4.866603437450473e-07, + "loss": 0.5472, + "step": 3075 + }, + { + "epoch": 4.820459290187891, + "grad_norm": 1.250402808189392, + "learning_rate": 4.854121349311949e-07, + "loss": 0.5438, + "step": 3076 + }, + { + "epoch": 4.822025052192067, + "grad_norm": 1.2647368907928467, + "learning_rate": 4.84165356769738e-07, + "loss": 0.6582, + "step": 3077 + }, + { + "epoch": 4.823590814196242, + "grad_norm": 1.2903023958206177, + "learning_rate": 4.829200101460696e-07, + "loss": 0.6657, + "step": 3078 + }, + { + "epoch": 4.825156576200418, + "grad_norm": 1.1874885559082031, + "learning_rate": 4.816760959445682e-07, + "loss": 0.5539, + "step": 3079 + }, + { + "epoch": 4.826722338204593, + "grad_norm": 1.1794021129608154, + "learning_rate": 4.804336150485969e-07, + "loss": 0.6093, + "step": 3080 + }, + { + "epoch": 4.828288100208768, + "grad_norm": 1.2572773694992065, + "learning_rate": 4.791925683404971e-07, + "loss": 0.6647, + "step": 3081 + }, + { + "epoch": 4.829853862212944, + "grad_norm": 1.2959238290786743, + "learning_rate": 4.779529567015948e-07, + "loss": 0.6719, + "step": 3082 + }, + { + "epoch": 4.831419624217119, + "grad_norm": 1.2031400203704834, + "learning_rate": 4.767147810121953e-07, + "loss": 0.6111, + "step": 3083 + }, + { + "epoch": 4.832985386221294, + "grad_norm": 1.2516471147537231, + "learning_rate": 4.7547804215158476e-07, + "loss": 0.5653, + "step": 3084 + }, + { + "epoch": 4.8345511482254695, + "grad_norm": 1.305241584777832, + "learning_rate": 4.7424274099802883e-07, + "loss": 0.607, + "step": 3085 + }, + { + "epoch": 4.836116910229645, + "grad_norm": 1.3069261312484741, + "learning_rate": 4.730088784287723e-07, + "loss": 0.6446, + "step": 3086 + }, + { + "epoch": 4.837682672233821, + "grad_norm": 1.1565463542938232, + "learning_rate": 4.717764553200385e-07, + "loss": 0.6627, + "step": 3087 + }, + { + "epoch": 4.839248434237996, + "grad_norm": 1.3230305910110474, + "learning_rate": 4.7054547254702793e-07, + "loss": 0.5788, + "step": 3088 + }, + { + "epoch": 4.840814196242171, + "grad_norm": 1.3158049583435059, + "learning_rate": 4.6931593098391874e-07, + "loss": 0.6677, + "step": 3089 + }, + { + "epoch": 4.8423799582463465, + "grad_norm": 1.222174048423767, + "learning_rate": 4.680878315038656e-07, + "loss": 0.554, + "step": 3090 + }, + { + "epoch": 4.843945720250522, + "grad_norm": 1.339778184890747, + "learning_rate": 4.668611749789989e-07, + "loss": 0.6612, + "step": 3091 + }, + { + "epoch": 4.845511482254698, + "grad_norm": 1.3295081853866577, + "learning_rate": 4.6563596228042433e-07, + "loss": 0.6592, + "step": 3092 + }, + { + "epoch": 4.847077244258872, + "grad_norm": 1.2802320718765259, + "learning_rate": 4.6441219427822314e-07, + "loss": 0.5961, + "step": 3093 + }, + { + "epoch": 4.848643006263048, + "grad_norm": 1.2144367694854736, + "learning_rate": 4.6318987184144795e-07, + "loss": 0.5894, + "step": 3094 + }, + { + "epoch": 4.8502087682672235, + "grad_norm": 1.2537792921066284, + "learning_rate": 4.619689958381282e-07, + "loss": 0.6544, + "step": 3095 + }, + { + "epoch": 4.851774530271399, + "grad_norm": 1.2162483930587769, + "learning_rate": 4.607495671352652e-07, + "loss": 0.6051, + "step": 3096 + }, + { + "epoch": 4.853340292275574, + "grad_norm": 1.2548906803131104, + "learning_rate": 4.5953158659882923e-07, + "loss": 0.6362, + "step": 3097 + }, + { + "epoch": 4.854906054279749, + "grad_norm": 1.2650794982910156, + "learning_rate": 4.583150550937682e-07, + "loss": 0.5981, + "step": 3098 + }, + { + "epoch": 4.856471816283925, + "grad_norm": 1.1979948282241821, + "learning_rate": 4.570999734839951e-07, + "loss": 0.6477, + "step": 3099 + }, + { + "epoch": 4.8580375782881005, + "grad_norm": 1.3879836797714233, + "learning_rate": 4.558863426323962e-07, + "loss": 0.6126, + "step": 3100 + }, + { + "epoch": 4.859603340292276, + "grad_norm": 1.2028707265853882, + "learning_rate": 4.5467416340082864e-07, + "loss": 0.6146, + "step": 3101 + }, + { + "epoch": 4.861169102296451, + "grad_norm": 1.2319929599761963, + "learning_rate": 4.534634366501159e-07, + "loss": 0.6523, + "step": 3102 + }, + { + "epoch": 4.862734864300626, + "grad_norm": 1.2286549806594849, + "learning_rate": 4.522541632400512e-07, + "loss": 0.6516, + "step": 3103 + }, + { + "epoch": 4.864300626304802, + "grad_norm": 1.1734979152679443, + "learning_rate": 4.5104634402939605e-07, + "loss": 0.5758, + "step": 3104 + }, + { + "epoch": 4.8658663883089766, + "grad_norm": 1.3007832765579224, + "learning_rate": 4.49839979875879e-07, + "loss": 0.5611, + "step": 3105 + }, + { + "epoch": 4.867432150313152, + "grad_norm": 1.2126667499542236, + "learning_rate": 4.486350716361959e-07, + "loss": 0.5952, + "step": 3106 + }, + { + "epoch": 4.868997912317328, + "grad_norm": 1.2437777519226074, + "learning_rate": 4.4743162016600615e-07, + "loss": 0.6277, + "step": 3107 + }, + { + "epoch": 4.870563674321503, + "grad_norm": 1.1449244022369385, + "learning_rate": 4.462296263199381e-07, + "loss": 0.5828, + "step": 3108 + }, + { + "epoch": 4.872129436325679, + "grad_norm": 1.3628195524215698, + "learning_rate": 4.450290909515839e-07, + "loss": 0.622, + "step": 3109 + }, + { + "epoch": 4.8736951983298535, + "grad_norm": 1.1359556913375854, + "learning_rate": 4.438300149134975e-07, + "loss": 0.6068, + "step": 3110 + }, + { + "epoch": 4.875260960334029, + "grad_norm": 1.2504936456680298, + "learning_rate": 4.426323990572004e-07, + "loss": 0.6443, + "step": 3111 + }, + { + "epoch": 4.876826722338205, + "grad_norm": 1.1604193449020386, + "learning_rate": 4.4143624423317403e-07, + "loss": 0.6291, + "step": 3112 + }, + { + "epoch": 4.87839248434238, + "grad_norm": 1.3639174699783325, + "learning_rate": 4.402415512908631e-07, + "loss": 0.6624, + "step": 3113 + }, + { + "epoch": 4.879958246346555, + "grad_norm": 1.3654690980911255, + "learning_rate": 4.3904832107867666e-07, + "loss": 0.5647, + "step": 3114 + }, + { + "epoch": 4.8815240083507305, + "grad_norm": 1.0890389680862427, + "learning_rate": 4.3785655444398087e-07, + "loss": 0.6207, + "step": 3115 + }, + { + "epoch": 4.883089770354906, + "grad_norm": 1.3411756753921509, + "learning_rate": 4.366662522331053e-07, + "loss": 0.5585, + "step": 3116 + }, + { + "epoch": 4.884655532359082, + "grad_norm": 1.1507508754730225, + "learning_rate": 4.3547741529133854e-07, + "loss": 0.682, + "step": 3117 + }, + { + "epoch": 4.886221294363256, + "grad_norm": 1.3855395317077637, + "learning_rate": 4.3429004446292945e-07, + "loss": 0.5744, + "step": 3118 + }, + { + "epoch": 4.887787056367432, + "grad_norm": 1.211582064628601, + "learning_rate": 4.3310414059108465e-07, + "loss": 0.6033, + "step": 3119 + }, + { + "epoch": 4.8893528183716075, + "grad_norm": 1.209214687347412, + "learning_rate": 4.319197045179696e-07, + "loss": 0.5845, + "step": 3120 + }, + { + "epoch": 4.890918580375783, + "grad_norm": 1.2045565843582153, + "learning_rate": 4.3073673708470766e-07, + "loss": 0.5807, + "step": 3121 + }, + { + "epoch": 4.892484342379959, + "grad_norm": 1.264220118522644, + "learning_rate": 4.295552391313787e-07, + "loss": 0.6604, + "step": 3122 + }, + { + "epoch": 4.894050104384133, + "grad_norm": 1.2991132736206055, + "learning_rate": 4.2837521149701937e-07, + "loss": 0.7245, + "step": 3123 + }, + { + "epoch": 4.895615866388309, + "grad_norm": 1.3955750465393066, + "learning_rate": 4.27196655019623e-07, + "loss": 0.6339, + "step": 3124 + }, + { + "epoch": 4.8971816283924845, + "grad_norm": 1.3281763792037964, + "learning_rate": 4.260195705361347e-07, + "loss": 0.6138, + "step": 3125 + }, + { + "epoch": 4.89874739039666, + "grad_norm": 1.2890803813934326, + "learning_rate": 4.2484395888245935e-07, + "loss": 0.6588, + "step": 3126 + }, + { + "epoch": 4.900313152400835, + "grad_norm": 1.2077101469039917, + "learning_rate": 4.236698208934531e-07, + "loss": 0.626, + "step": 3127 + }, + { + "epoch": 4.90187891440501, + "grad_norm": 1.1384289264678955, + "learning_rate": 4.224971574029241e-07, + "loss": 0.6468, + "step": 3128 + }, + { + "epoch": 4.903444676409186, + "grad_norm": 1.319321632385254, + "learning_rate": 4.2132596924363666e-07, + "loss": 0.6104, + "step": 3129 + }, + { + "epoch": 4.9050104384133615, + "grad_norm": 1.3868818283081055, + "learning_rate": 4.201562572473064e-07, + "loss": 0.6325, + "step": 3130 + }, + { + "epoch": 4.906576200417536, + "grad_norm": 1.281776785850525, + "learning_rate": 4.18988022244598e-07, + "loss": 0.5742, + "step": 3131 + }, + { + "epoch": 4.908141962421712, + "grad_norm": 1.2451962232589722, + "learning_rate": 4.1782126506513196e-07, + "loss": 0.5826, + "step": 3132 + }, + { + "epoch": 4.909707724425887, + "grad_norm": 1.189673662185669, + "learning_rate": 4.166559865374747e-07, + "loss": 0.6543, + "step": 3133 + }, + { + "epoch": 4.911273486430063, + "grad_norm": 1.3293418884277344, + "learning_rate": 4.154921874891446e-07, + "loss": 0.6165, + "step": 3134 + }, + { + "epoch": 4.9128392484342385, + "grad_norm": 1.3895233869552612, + "learning_rate": 4.1432986874661156e-07, + "loss": 0.6627, + "step": 3135 + }, + { + "epoch": 4.914405010438413, + "grad_norm": 1.3557475805282593, + "learning_rate": 4.131690311352898e-07, + "loss": 0.598, + "step": 3136 + }, + { + "epoch": 4.915970772442589, + "grad_norm": 1.2514588832855225, + "learning_rate": 4.1200967547954525e-07, + "loss": 0.5649, + "step": 3137 + }, + { + "epoch": 4.917536534446764, + "grad_norm": 1.248897910118103, + "learning_rate": 4.1085180260268975e-07, + "loss": 0.6182, + "step": 3138 + }, + { + "epoch": 4.919102296450939, + "grad_norm": 1.254090666770935, + "learning_rate": 4.0969541332698257e-07, + "loss": 0.523, + "step": 3139 + }, + { + "epoch": 4.920668058455115, + "grad_norm": 1.2779461145401, + "learning_rate": 4.0854050847362966e-07, + "loss": 0.6123, + "step": 3140 + }, + { + "epoch": 4.92223382045929, + "grad_norm": 1.2242066860198975, + "learning_rate": 4.0738708886278233e-07, + "loss": 0.6661, + "step": 3141 + }, + { + "epoch": 4.923799582463466, + "grad_norm": 1.0954583883285522, + "learning_rate": 4.062351553135377e-07, + "loss": 0.5705, + "step": 3142 + }, + { + "epoch": 4.925365344467641, + "grad_norm": 1.4686397314071655, + "learning_rate": 4.050847086439369e-07, + "loss": 0.6784, + "step": 3143 + }, + { + "epoch": 4.926931106471816, + "grad_norm": 1.094277262687683, + "learning_rate": 4.0393574967096586e-07, + "loss": 0.598, + "step": 3144 + }, + { + "epoch": 4.9284968684759916, + "grad_norm": 1.2568544149398804, + "learning_rate": 4.0278827921055447e-07, + "loss": 0.6422, + "step": 3145 + }, + { + "epoch": 4.930062630480167, + "grad_norm": 1.3163423538208008, + "learning_rate": 4.016422980775725e-07, + "loss": 0.6132, + "step": 3146 + }, + { + "epoch": 4.931628392484343, + "grad_norm": 1.2578885555267334, + "learning_rate": 4.0049780708583647e-07, + "loss": 0.6313, + "step": 3147 + }, + { + "epoch": 4.933194154488517, + "grad_norm": 1.1483683586120605, + "learning_rate": 3.9935480704810237e-07, + "loss": 0.5864, + "step": 3148 + }, + { + "epoch": 4.934759916492693, + "grad_norm": 1.239707589149475, + "learning_rate": 3.982132987760659e-07, + "loss": 0.5784, + "step": 3149 + }, + { + "epoch": 4.9363256784968685, + "grad_norm": 1.2554842233657837, + "learning_rate": 3.970732830803678e-07, + "loss": 0.5633, + "step": 3150 + }, + { + "epoch": 4.937891440501044, + "grad_norm": 1.1999422311782837, + "learning_rate": 3.959347607705838e-07, + "loss": 0.6405, + "step": 3151 + }, + { + "epoch": 4.939457202505219, + "grad_norm": 1.3461389541625977, + "learning_rate": 3.9479773265523186e-07, + "loss": 0.6856, + "step": 3152 + }, + { + "epoch": 4.941022964509394, + "grad_norm": 1.3426072597503662, + "learning_rate": 3.9366219954177e-07, + "loss": 0.6218, + "step": 3153 + }, + { + "epoch": 4.94258872651357, + "grad_norm": 1.2127430438995361, + "learning_rate": 3.925281622365915e-07, + "loss": 0.5842, + "step": 3154 + }, + { + "epoch": 4.9441544885177455, + "grad_norm": 1.2543373107910156, + "learning_rate": 3.913956215450293e-07, + "loss": 0.6345, + "step": 3155 + }, + { + "epoch": 4.945720250521921, + "grad_norm": 1.3203067779541016, + "learning_rate": 3.9026457827135324e-07, + "loss": 0.6336, + "step": 3156 + }, + { + "epoch": 4.947286012526096, + "grad_norm": 1.2462279796600342, + "learning_rate": 3.891350332187696e-07, + "loss": 0.5859, + "step": 3157 + }, + { + "epoch": 4.948851774530271, + "grad_norm": 1.3469984531402588, + "learning_rate": 3.8800698718942126e-07, + "loss": 0.5761, + "step": 3158 + }, + { + "epoch": 4.950417536534447, + "grad_norm": 1.1929227113723755, + "learning_rate": 3.868804409843846e-07, + "loss": 0.5895, + "step": 3159 + }, + { + "epoch": 4.9519832985386225, + "grad_norm": 1.2928577661514282, + "learning_rate": 3.8575539540367383e-07, + "loss": 0.6293, + "step": 3160 + }, + { + "epoch": 4.953549060542797, + "grad_norm": 1.2409569025039673, + "learning_rate": 3.8463185124623657e-07, + "loss": 0.5194, + "step": 3161 + }, + { + "epoch": 4.955114822546973, + "grad_norm": 1.1920491456985474, + "learning_rate": 3.8350980930995124e-07, + "loss": 0.6085, + "step": 3162 + }, + { + "epoch": 4.956680584551148, + "grad_norm": 1.1986509561538696, + "learning_rate": 3.8238927039163487e-07, + "loss": 0.5688, + "step": 3163 + }, + { + "epoch": 4.958246346555324, + "grad_norm": 1.313879370689392, + "learning_rate": 3.812702352870321e-07, + "loss": 0.622, + "step": 3164 + }, + { + "epoch": 4.959812108559499, + "grad_norm": 1.205352544784546, + "learning_rate": 3.801527047908218e-07, + "loss": 0.604, + "step": 3165 + }, + { + "epoch": 4.961377870563674, + "grad_norm": 1.2026801109313965, + "learning_rate": 3.7903667969661605e-07, + "loss": 0.5921, + "step": 3166 + }, + { + "epoch": 4.96294363256785, + "grad_norm": 1.324131965637207, + "learning_rate": 3.779221607969541e-07, + "loss": 0.6334, + "step": 3167 + }, + { + "epoch": 4.964509394572025, + "grad_norm": 1.314497947692871, + "learning_rate": 3.7680914888330865e-07, + "loss": 0.6152, + "step": 3168 + }, + { + "epoch": 4.966075156576201, + "grad_norm": 1.3103183507919312, + "learning_rate": 3.7569764474608085e-07, + "loss": 0.649, + "step": 3169 + }, + { + "epoch": 4.967640918580376, + "grad_norm": 1.3193435668945312, + "learning_rate": 3.7458764917460146e-07, + "loss": 0.688, + "step": 3170 + }, + { + "epoch": 4.969206680584551, + "grad_norm": 1.3641090393066406, + "learning_rate": 3.7347916295712967e-07, + "loss": 0.6085, + "step": 3171 + }, + { + "epoch": 4.970772442588727, + "grad_norm": 1.1966512203216553, + "learning_rate": 3.723721868808533e-07, + "loss": 0.573, + "step": 3172 + }, + { + "epoch": 4.972338204592901, + "grad_norm": 1.12050199508667, + "learning_rate": 3.712667217318874e-07, + "loss": 0.6238, + "step": 3173 + }, + { + "epoch": 4.973903966597077, + "grad_norm": 1.4070502519607544, + "learning_rate": 3.701627682952741e-07, + "loss": 0.63, + "step": 3174 + }, + { + "epoch": 4.975469728601253, + "grad_norm": 1.147275686264038, + "learning_rate": 3.690603273549817e-07, + "loss": 0.6603, + "step": 3175 + }, + { + "epoch": 4.977035490605428, + "grad_norm": 1.2202069759368896, + "learning_rate": 3.6795939969390544e-07, + "loss": 0.6294, + "step": 3176 + }, + { + "epoch": 4.978601252609604, + "grad_norm": 1.2288408279418945, + "learning_rate": 3.668599860938643e-07, + "loss": 0.6612, + "step": 3177 + }, + { + "epoch": 4.980167014613778, + "grad_norm": 1.2829551696777344, + "learning_rate": 3.657620873356035e-07, + "loss": 0.652, + "step": 3178 + }, + { + "epoch": 4.981732776617954, + "grad_norm": 1.230684757232666, + "learning_rate": 3.646657041987922e-07, + "loss": 0.5631, + "step": 3179 + }, + { + "epoch": 4.98329853862213, + "grad_norm": 1.1531054973602295, + "learning_rate": 3.6357083746202173e-07, + "loss": 0.5369, + "step": 3180 + }, + { + "epoch": 4.984864300626305, + "grad_norm": 1.212597131729126, + "learning_rate": 3.6247748790280895e-07, + "loss": 0.562, + "step": 3181 + }, + { + "epoch": 4.98643006263048, + "grad_norm": 1.318061351776123, + "learning_rate": 3.613856562975923e-07, + "loss": 0.5543, + "step": 3182 + }, + { + "epoch": 4.987995824634655, + "grad_norm": 1.1155426502227783, + "learning_rate": 3.6029534342173073e-07, + "loss": 0.6585, + "step": 3183 + }, + { + "epoch": 4.989561586638831, + "grad_norm": 1.4041640758514404, + "learning_rate": 3.592065500495079e-07, + "loss": 0.5962, + "step": 3184 + }, + { + "epoch": 4.991127348643007, + "grad_norm": 1.3740663528442383, + "learning_rate": 3.581192769541253e-07, + "loss": 0.6219, + "step": 3185 + }, + { + "epoch": 4.992693110647181, + "grad_norm": 1.2645107507705688, + "learning_rate": 3.570335249077056e-07, + "loss": 0.6126, + "step": 3186 + }, + { + "epoch": 4.994258872651357, + "grad_norm": 1.3721750974655151, + "learning_rate": 3.559492946812934e-07, + "loss": 0.6416, + "step": 3187 + }, + { + "epoch": 4.995824634655532, + "grad_norm": 1.257168173789978, + "learning_rate": 3.5486658704484977e-07, + "loss": 0.6237, + "step": 3188 + }, + { + "epoch": 4.997390396659708, + "grad_norm": 1.2237387895584106, + "learning_rate": 3.537854027672558e-07, + "loss": 0.5944, + "step": 3189 + }, + { + "epoch": 4.9989561586638835, + "grad_norm": 1.3416454792022705, + "learning_rate": 3.5270574261631054e-07, + "loss": 0.614, + "step": 3190 + } + ], + "logging_steps": 1, + "max_steps": 3828, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 638, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 8.061214608536371e+18, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-3190/training_args.bin b/checkpoint-3190/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..07c8c9aae2a66d9fb6a8b65759b1cd86d2f7d08b --- /dev/null +++ b/checkpoint-3190/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81aeca374ba1e1a5384aca380c5dd9da735cc6d6164e0b5b0681e227cf61f41c +size 7992 diff --git a/checkpoint-3190/zero_to_fp32.py b/checkpoint-3190/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-3190/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-3828/README.md b/checkpoint-3828/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f07164552af92bf3072e4e4326b5e572dea01ae8 --- /dev/null +++ b/checkpoint-3828/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-Nano-8B-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 0.14.0 \ No newline at end of file diff --git a/checkpoint-3828/adapter_config.json b/checkpoint-3828/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..3db75f3ecbeb5f4990760e56c7957796d83680c5 --- /dev/null +++ b/checkpoint-3828/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1", + "bias": "none", + "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": [ + "q_proj", + "down_proj", + "o_proj", + "v_proj", + "k_proj", + "gate_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-3828/adapter_model.safetensors b/checkpoint-3828/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b7828c64e3e693a901fefa43b86429b0bf262f8d --- /dev/null +++ b/checkpoint-3828/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c045b55e37980f1987824870ab9c9912e9bbf700733b076ba36f6b536e1393f +size 3443586272 diff --git a/checkpoint-3828/global_step3829/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-3828/global_step3829/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..2cb1f6a017d99500d3ddf2c0a50b1cf38ed5e74f --- /dev/null +++ b/checkpoint-3828/global_step3829/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31491dcc439a7df26efd05508aa0a5e5c0db4319f58616d93b1b6c7416ed14e9 +size 20661195036 diff --git a/checkpoint-3828/global_step3829/mp_rank_00_model_states.pt b/checkpoint-3828/global_step3829/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a04203fb2539cf2c81e7634ad84855644f00caf1 --- /dev/null +++ b/checkpoint-3828/global_step3829/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c61ec93734051248fb47a1407a92d3e506b508dacbee9236b2aad2680af7e78b +size 3555326777 diff --git a/checkpoint-3828/latest b/checkpoint-3828/latest new file mode 100644 index 0000000000000000000000000000000000000000..8c9837d08e85fdb41cd4289e8c50cc1b2f03d2e8 --- /dev/null +++ b/checkpoint-3828/latest @@ -0,0 +1 @@ +global_step3829 \ No newline at end of file diff --git a/checkpoint-3828/rng_state.pth b/checkpoint-3828/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..fe190006e3e2da227889f9e9e064f6d806ef054b --- /dev/null +++ b/checkpoint-3828/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05f982f1cedf261cdfc42a4f984ce732383e4936b3eee2324d2d58956e4062ef +size 14244 diff --git a/checkpoint-3828/scheduler.pt b/checkpoint-3828/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..831912246e69c4f700299f407586820f855c3ce1 --- /dev/null +++ b/checkpoint-3828/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e923faad506389b5360a6f8a635a927c728dea9d32095698e603c6450b0447d0 +size 1064 diff --git a/checkpoint-3828/special_tokens_map.json b/checkpoint-3828/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-3828/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-3828/tokenizer.json b/checkpoint-3828/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-3828/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-3828/tokenizer_config.json b/checkpoint-3828/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b480d149dc274c0a23db079bfd38032d90082e49 --- /dev/null +++ b/checkpoint-3828/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": "{%- if messages[0]['role'] == 'system' -%}{%- set system_message = messages[0]['content'] | trim -%}{%- set messages = messages[1:] -%}{%- else -%}{%- set system_message = '' -%}{%- endif -%}{%- if tools is not none -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message -}} {{- '\n\n' if system_message else '' -}} {{- '[' -}} {% for t in tools %}{{- (t.function if t.function is defined else t) | tojson() -}}{{- ', ' if not loop.last else '' -}}{%- endfor -%} {{- ']' -}} {{- '<|eot_id|>' -}}{%- else -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message + '<|eot_id|>' -}}{%- endif -%}{%- for message in messages -%}{%- if (message['role'] in ['user', 'tool']) != (loop.index0 % 2 == 0) -%}{{- raise_exception('Conversation roles must alternate between user/tool and assistant') -}}{%- elif message['role'] == 'user' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- elif message['role'] == 'tool' -%}{%- set tool_response = '[' + message['content'] | trim + ']' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + tool_response + '<|eot_id|>' -}}{%- elif message['role'] == 'assistant' and message.get('tool_calls') is not none -%}{%- set tool_calls = message['tool_calls'] -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + '[' -}}{%- for tool_call in tool_calls -%}{{ '{' + '\"name\": \"' + tool_call.function.name + '\", \"arguments\": ' + tool_call.function.arguments | tojson + '}' }}{%- if not loop.last -%}{{ ', ' }}{%- else -%}{{ ']' + '<|eot_id|>' }}{%- endif -%}{%- endfor -%}{%- elif message['role'] == 'assistant' -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- endif -%}{%- 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-3828/trainer_state.json b/checkpoint-3828/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..7f8db6f1442791ee56d612500436388d66437d85 --- /dev/null +++ b/checkpoint-3828/trainer_state.json @@ -0,0 +1,26830 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 5.9989561586638835, + "eval_steps": 500, + "global_step": 3828, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0015657620041753654, + "grad_norm": 658.1201782226562, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.3039, + "step": 1 + }, + { + "epoch": 0.003131524008350731, + "grad_norm": 749.5796508789062, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.4164, + "step": 2 + }, + { + "epoch": 0.004697286012526096, + "grad_norm": 394.2350769042969, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.3367, + "step": 3 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 384.08319091796875, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.3192, + "step": 4 + }, + { + "epoch": 0.007828810020876827, + "grad_norm": 376.9790954589844, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.2756, + "step": 5 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 175.25262451171875, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.3992, + "step": 6 + }, + { + "epoch": 0.010960334029227558, + "grad_norm": 100.1114730834961, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.2348, + "step": 7 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 131.44432067871094, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.2685, + "step": 8 + }, + { + "epoch": 0.014091858037578288, + "grad_norm": 67.03341674804688, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.3272, + "step": 9 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 32.0869140625, + "learning_rate": 5.000000000000001e-07, + "loss": 1.3458, + "step": 10 + }, + { + "epoch": 0.01722338204592902, + "grad_norm": 17.701913833618164, + "learning_rate": 5.5e-07, + "loss": 1.3608, + "step": 11 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 9.164765357971191, + "learning_rate": 6.000000000000001e-07, + "loss": 1.2961, + "step": 12 + }, + { + "epoch": 0.02035490605427975, + "grad_norm": 4.315349578857422, + "learning_rate": 6.5e-07, + "loss": 1.2069, + "step": 13 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 6.652385234832764, + "learning_rate": 7.000000000000001e-07, + "loss": 1.2588, + "step": 14 + }, + { + "epoch": 0.02348643006263048, + "grad_norm": 3.3322548866271973, + "learning_rate": 7.5e-07, + "loss": 1.2505, + "step": 15 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 9.599331855773926, + "learning_rate": 8.000000000000001e-07, + "loss": 1.1343, + "step": 16 + }, + { + "epoch": 0.02661795407098121, + "grad_norm": 10.33918285369873, + "learning_rate": 8.500000000000001e-07, + "loss": 1.3334, + "step": 17 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 9.453723907470703, + "learning_rate": 9.000000000000001e-07, + "loss": 1.3099, + "step": 18 + }, + { + "epoch": 0.029749478079331943, + "grad_norm": 7.075769901275635, + "learning_rate": 9.500000000000001e-07, + "loss": 1.1121, + "step": 19 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 7.770864009857178, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.1836, + "step": 20 + }, + { + "epoch": 0.03288100208768267, + "grad_norm": 9.582944869995117, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.1939, + "step": 21 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 9.496583938598633, + "learning_rate": 1.1e-06, + "loss": 1.2395, + "step": 22 + }, + { + "epoch": 0.0360125260960334, + "grad_norm": 5.912876129150391, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.1769, + "step": 23 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 2.6761083602905273, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.204, + "step": 24 + }, + { + "epoch": 0.03914405010438413, + "grad_norm": 11.136463165283203, + "learning_rate": 1.25e-06, + "loss": 1.2271, + "step": 25 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 23.912893295288086, + "learning_rate": 1.3e-06, + "loss": 1.1771, + "step": 26 + }, + { + "epoch": 0.042275574112734866, + "grad_norm": 57.08906555175781, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.2253, + "step": 27 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 36.11408615112305, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.1014, + "step": 28 + }, + { + "epoch": 0.04540709812108559, + "grad_norm": 51.39445495605469, + "learning_rate": 1.45e-06, + "loss": 1.1978, + "step": 29 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 27.394500732421875, + "learning_rate": 1.5e-06, + "loss": 1.1155, + "step": 30 + }, + { + "epoch": 0.048538622129436326, + "grad_norm": 6.844716548919678, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.1798, + "step": 31 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 2.397536277770996, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.1846, + "step": 32 + }, + { + "epoch": 0.05167014613778706, + "grad_norm": 7.821156024932861, + "learning_rate": 1.6500000000000003e-06, + "loss": 1.1312, + "step": 33 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 5.455922603607178, + "learning_rate": 1.7000000000000002e-06, + "loss": 1.2153, + "step": 34 + }, + { + "epoch": 0.054801670146137786, + "grad_norm": 5.377617835998535, + "learning_rate": 1.75e-06, + "loss": 1.1292, + "step": 35 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 7.172031879425049, + "learning_rate": 1.8000000000000001e-06, + "loss": 1.1459, + "step": 36 + }, + { + "epoch": 0.05793319415448852, + "grad_norm": 2.1464905738830566, + "learning_rate": 1.85e-06, + "loss": 1.1649, + "step": 37 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 2.7782726287841797, + "learning_rate": 1.9000000000000002e-06, + "loss": 1.159, + "step": 38 + }, + { + "epoch": 0.061064718162839246, + "grad_norm": 20.818374633789062, + "learning_rate": 1.9500000000000004e-06, + "loss": 1.0174, + "step": 39 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 22.776456832885742, + "learning_rate": 2.0000000000000003e-06, + "loss": 1.1425, + "step": 40 + }, + { + "epoch": 0.06419624217118998, + "grad_norm": 33.601287841796875, + "learning_rate": 2.05e-06, + "loss": 1.0498, + "step": 41 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.884684562683105, + "learning_rate": 2.1000000000000002e-06, + "loss": 1.1231, + "step": 42 + }, + { + "epoch": 0.06732776617954071, + "grad_norm": 1.951470971107483, + "learning_rate": 2.15e-06, + "loss": 1.1256, + "step": 43 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 2.584127426147461, + "learning_rate": 2.2e-06, + "loss": 1.1242, + "step": 44 + }, + { + "epoch": 0.07045929018789145, + "grad_norm": 6.982508659362793, + "learning_rate": 2.25e-06, + "loss": 1.0769, + "step": 45 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 7.940657138824463, + "learning_rate": 2.3000000000000004e-06, + "loss": 1.0516, + "step": 46 + }, + { + "epoch": 0.07359081419624217, + "grad_norm": 5.456567287445068, + "learning_rate": 2.35e-06, + "loss": 0.9894, + "step": 47 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 4.623636722564697, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9912, + "step": 48 + }, + { + "epoch": 0.0767223382045929, + "grad_norm": 2.182708978652954, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.1303, + "step": 49 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 2.388091802597046, + "learning_rate": 2.5e-06, + "loss": 1.0725, + "step": 50 + }, + { + "epoch": 0.07985386221294363, + "grad_norm": 2.2816522121429443, + "learning_rate": 2.55e-06, + "loss": 1.0895, + "step": 51 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 3.8391623497009277, + "learning_rate": 2.6e-06, + "loss": 1.0587, + "step": 52 + }, + { + "epoch": 0.08298538622129437, + "grad_norm": 3.700389862060547, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9795, + "step": 53 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 3.180649995803833, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.0434, + "step": 54 + }, + { + "epoch": 0.0861169102296451, + "grad_norm": 3.790774345397949, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9652, + "step": 55 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 1.8411438465118408, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.947, + "step": 56 + }, + { + "epoch": 0.08924843423799582, + "grad_norm": 2.100008010864258, + "learning_rate": 2.85e-06, + "loss": 1.0129, + "step": 57 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 1.103857159614563, + "learning_rate": 2.9e-06, + "loss": 0.9993, + "step": 58 + }, + { + "epoch": 0.09237995824634655, + "grad_norm": 1.232283353805542, + "learning_rate": 2.95e-06, + "loss": 0.9988, + "step": 59 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 2.4731297492980957, + "learning_rate": 3e-06, + "loss": 0.9873, + "step": 60 + }, + { + "epoch": 0.09551148225469729, + "grad_norm": 1.1080410480499268, + "learning_rate": 3.05e-06, + "loss": 1.0657, + "step": 61 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 1.1155121326446533, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.974, + "step": 62 + }, + { + "epoch": 0.09864300626304802, + "grad_norm": 1.763231873512268, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.0215, + "step": 63 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 1.1961321830749512, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.9955, + "step": 64 + }, + { + "epoch": 0.10177453027139875, + "grad_norm": 1.377040982246399, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.9834, + "step": 65 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 1.373260498046875, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.984, + "step": 66 + }, + { + "epoch": 0.10490605427974947, + "grad_norm": 2.034064292907715, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.973, + "step": 67 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 1.41798734664917, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.925, + "step": 68 + }, + { + "epoch": 0.1080375782881002, + "grad_norm": 1.7425439357757568, + "learning_rate": 3.45e-06, + "loss": 0.9884, + "step": 69 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 2.369657278060913, + "learning_rate": 3.5e-06, + "loss": 1.0048, + "step": 70 + }, + { + "epoch": 0.11116910229645094, + "grad_norm": 1.4699137210845947, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.9155, + "step": 71 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 1.1473263502120972, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8445, + "step": 72 + }, + { + "epoch": 0.11430062630480167, + "grad_norm": 2.315568685531616, + "learning_rate": 3.65e-06, + "loss": 0.977, + "step": 73 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 1.8554916381835938, + "learning_rate": 3.7e-06, + "loss": 0.841, + "step": 74 + }, + { + "epoch": 0.1174321503131524, + "grad_norm": 1.1517397165298462, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.9478, + "step": 75 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 2.799316883087158, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.9873, + "step": 76 + }, + { + "epoch": 0.12056367432150313, + "grad_norm": 1.11930251121521, + "learning_rate": 3.85e-06, + "loss": 0.9214, + "step": 77 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 1.0082950592041016, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9115, + "step": 78 + }, + { + "epoch": 0.12369519832985386, + "grad_norm": 1.5534541606903076, + "learning_rate": 3.95e-06, + "loss": 1.04, + "step": 79 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 1.3469066619873047, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9091, + "step": 80 + }, + { + "epoch": 0.1268267223382046, + "grad_norm": 1.1149636507034302, + "learning_rate": 4.05e-06, + "loss": 0.8829, + "step": 81 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 2.5933806896209717, + "learning_rate": 4.1e-06, + "loss": 0.9343, + "step": 82 + }, + { + "epoch": 0.1299582463465553, + "grad_norm": 1.980901837348938, + "learning_rate": 4.15e-06, + "loss": 0.9028, + "step": 83 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 0.8476768732070923, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.9038, + "step": 84 + }, + { + "epoch": 0.13308977035490605, + "grad_norm": 1.1239815950393677, + "learning_rate": 4.25e-06, + "loss": 1.0359, + "step": 85 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 1.0159156322479248, + "learning_rate": 4.3e-06, + "loss": 0.8705, + "step": 86 + }, + { + "epoch": 0.13622129436325678, + "grad_norm": 2.180856466293335, + "learning_rate": 4.350000000000001e-06, + "loss": 0.9834, + "step": 87 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 1.5725276470184326, + "learning_rate": 4.4e-06, + "loss": 0.9449, + "step": 88 + }, + { + "epoch": 0.1393528183716075, + "grad_norm": 1.5560917854309082, + "learning_rate": 4.450000000000001e-06, + "loss": 1.0109, + "step": 89 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 1.1096960306167603, + "learning_rate": 4.5e-06, + "loss": 0.9081, + "step": 90 + }, + { + "epoch": 0.14248434237995825, + "grad_norm": 1.2525160312652588, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.9907, + "step": 91 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 0.9198770523071289, + "learning_rate": 4.600000000000001e-06, + "loss": 0.9266, + "step": 92 + }, + { + "epoch": 0.14561586638830898, + "grad_norm": 0.992057740688324, + "learning_rate": 4.65e-06, + "loss": 0.9556, + "step": 93 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 1.00990891456604, + "learning_rate": 4.7e-06, + "loss": 0.9712, + "step": 94 + }, + { + "epoch": 0.1487473903966597, + "grad_norm": 1.1176848411560059, + "learning_rate": 4.75e-06, + "loss": 0.9604, + "step": 95 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 0.9323613047599792, + "learning_rate": 4.800000000000001e-06, + "loss": 0.9708, + "step": 96 + }, + { + "epoch": 0.15187891440501045, + "grad_norm": 0.9800642728805542, + "learning_rate": 4.85e-06, + "loss": 0.9868, + "step": 97 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 1.487841010093689, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8679, + "step": 98 + }, + { + "epoch": 0.15501043841336118, + "grad_norm": 0.95517897605896, + "learning_rate": 4.95e-06, + "loss": 0.7949, + "step": 99 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 0.9686336517333984, + "learning_rate": 5e-06, + "loss": 0.8655, + "step": 100 + }, + { + "epoch": 0.1581419624217119, + "grad_norm": 1.9481035470962524, + "learning_rate": 4.999999112316954e-06, + "loss": 0.9328, + "step": 101 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 1.272631287574768, + "learning_rate": 4.9999964492684465e-06, + "loss": 0.7869, + "step": 102 + }, + { + "epoch": 0.16127348643006262, + "grad_norm": 1.1117815971374512, + "learning_rate": 4.999992010856368e-06, + "loss": 0.949, + "step": 103 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 0.9681039452552795, + "learning_rate": 4.999985797083871e-06, + "loss": 1.0396, + "step": 104 + }, + { + "epoch": 0.16440501043841335, + "grad_norm": 0.9797878265380859, + "learning_rate": 4.999977807955368e-06, + "loss": 0.9393, + "step": 105 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 1.0239158868789673, + "learning_rate": 4.999968043476532e-06, + "loss": 0.8766, + "step": 106 + }, + { + "epoch": 0.16753653444676408, + "grad_norm": 0.9314751029014587, + "learning_rate": 4.999956503654299e-06, + "loss": 0.8545, + "step": 107 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 1.1627179384231567, + "learning_rate": 4.999943188496862e-06, + "loss": 0.9241, + "step": 108 + }, + { + "epoch": 0.17066805845511482, + "grad_norm": 1.022752285003662, + "learning_rate": 4.9999280980136765e-06, + "loss": 0.9662, + "step": 109 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 1.0513653755187988, + "learning_rate": 4.99991123221546e-06, + "loss": 0.9054, + "step": 110 + }, + { + "epoch": 0.17379958246346555, + "grad_norm": 0.9488444328308105, + "learning_rate": 4.99989259111419e-06, + "loss": 0.8921, + "step": 111 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 0.9934154748916626, + "learning_rate": 4.999872174723104e-06, + "loss": 0.9848, + "step": 112 + }, + { + "epoch": 0.17693110647181629, + "grad_norm": 1.2035515308380127, + "learning_rate": 4.9998499830567e-06, + "loss": 0.9833, + "step": 113 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 1.027538537979126, + "learning_rate": 4.999826016130739e-06, + "loss": 1.0089, + "step": 114 + }, + { + "epoch": 0.18006263048016702, + "grad_norm": 1.0635300874710083, + "learning_rate": 4.999800273962238e-06, + "loss": 0.8863, + "step": 115 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 0.9978796243667603, + "learning_rate": 4.999772756569482e-06, + "loss": 0.899, + "step": 116 + }, + { + "epoch": 0.18319415448851775, + "grad_norm": 0.9971838593482971, + "learning_rate": 4.999743463972008e-06, + "loss": 0.9515, + "step": 117 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 1.2436699867248535, + "learning_rate": 4.999712396190622e-06, + "loss": 0.9101, + "step": 118 + }, + { + "epoch": 0.18632567849686849, + "grad_norm": 0.9573557376861572, + "learning_rate": 4.999679553247383e-06, + "loss": 0.8817, + "step": 119 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 0.9105668067932129, + "learning_rate": 4.9996449351656165e-06, + "loss": 1.0077, + "step": 120 + }, + { + "epoch": 0.18945720250521922, + "grad_norm": 0.8741400241851807, + "learning_rate": 4.999608541969905e-06, + "loss": 0.912, + "step": 121 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 1.1172688007354736, + "learning_rate": 4.999570373686095e-06, + "loss": 0.8062, + "step": 122 + }, + { + "epoch": 0.19258872651356992, + "grad_norm": 0.940539538860321, + "learning_rate": 4.9995304303412905e-06, + "loss": 0.8944, + "step": 123 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 2.1082208156585693, + "learning_rate": 4.999488711963857e-06, + "loss": 0.9668, + "step": 124 + }, + { + "epoch": 0.19572025052192066, + "grad_norm": 1.0358761548995972, + "learning_rate": 4.99944521858342e-06, + "loss": 0.9032, + "step": 125 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 2.0804238319396973, + "learning_rate": 4.999399950230867e-06, + "loss": 0.9819, + "step": 126 + }, + { + "epoch": 0.1988517745302714, + "grad_norm": 1.0986278057098389, + "learning_rate": 4.9993529069383465e-06, + "loss": 0.9034, + "step": 127 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 1.2374722957611084, + "learning_rate": 4.999304088739263e-06, + "loss": 0.9255, + "step": 128 + }, + { + "epoch": 0.20198329853862212, + "grad_norm": 1.0443146228790283, + "learning_rate": 4.999253495668287e-06, + "loss": 0.9352, + "step": 129 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 1.009800910949707, + "learning_rate": 4.999201127761346e-06, + "loss": 0.8324, + "step": 130 + }, + { + "epoch": 0.20511482254697286, + "grad_norm": 0.892861008644104, + "learning_rate": 4.9991469850556295e-06, + "loss": 0.886, + "step": 131 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 1.1675922870635986, + "learning_rate": 4.999091067589587e-06, + "loss": 0.967, + "step": 132 + }, + { + "epoch": 0.2082463465553236, + "grad_norm": 1.0169954299926758, + "learning_rate": 4.999033375402926e-06, + "loss": 0.8498, + "step": 133 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 1.0006227493286133, + "learning_rate": 4.99897390853662e-06, + "loss": 0.8481, + "step": 134 + }, + { + "epoch": 0.21137787056367432, + "grad_norm": 0.9049347639083862, + "learning_rate": 4.9989126670328945e-06, + "loss": 0.9186, + "step": 135 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 1.2151603698730469, + "learning_rate": 4.9988496509352444e-06, + "loss": 0.9067, + "step": 136 + }, + { + "epoch": 0.21450939457202506, + "grad_norm": 1.0112006664276123, + "learning_rate": 4.998784860288417e-06, + "loss": 0.8843, + "step": 137 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 0.9316662549972534, + "learning_rate": 4.9987182951384264e-06, + "loss": 0.8635, + "step": 138 + }, + { + "epoch": 0.2176409185803758, + "grad_norm": 0.942743718624115, + "learning_rate": 4.9986499555325406e-06, + "loss": 0.8481, + "step": 139 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 1.0298817157745361, + "learning_rate": 4.998579841519292e-06, + "loss": 0.865, + "step": 140 + }, + { + "epoch": 0.22077244258872653, + "grad_norm": 1.1556729078292847, + "learning_rate": 4.998507953148472e-06, + "loss": 0.8731, + "step": 141 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 0.997958779335022, + "learning_rate": 4.9984342904711315e-06, + "loss": 0.7949, + "step": 142 + }, + { + "epoch": 0.22390396659707723, + "grad_norm": 0.9565998315811157, + "learning_rate": 4.998358853539582e-06, + "loss": 0.8875, + "step": 143 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 0.9667196273803711, + "learning_rate": 4.998281642407394e-06, + "loss": 0.8537, + "step": 144 + }, + { + "epoch": 0.22703549060542796, + "grad_norm": 0.9701303839683533, + "learning_rate": 4.998202657129401e-06, + "loss": 0.873, + "step": 145 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 0.9936159253120422, + "learning_rate": 4.998121897761692e-06, + "loss": 0.8915, + "step": 146 + }, + { + "epoch": 0.2301670146137787, + "grad_norm": 1.5054311752319336, + "learning_rate": 4.998039364361617e-06, + "loss": 0.9944, + "step": 147 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 1.5895891189575195, + "learning_rate": 4.99795505698779e-06, + "loss": 0.9282, + "step": 148 + }, + { + "epoch": 0.23329853862212943, + "grad_norm": 1.0141469240188599, + "learning_rate": 4.9978689757000785e-06, + "loss": 0.9166, + "step": 149 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 1.1720311641693115, + "learning_rate": 4.997781120559615e-06, + "loss": 0.9404, + "step": 150 + }, + { + "epoch": 0.23643006263048016, + "grad_norm": 0.9536433815956116, + "learning_rate": 4.997691491628789e-06, + "loss": 0.9339, + "step": 151 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 0.9407872557640076, + "learning_rate": 4.997600088971249e-06, + "loss": 0.9019, + "step": 152 + }, + { + "epoch": 0.2395615866388309, + "grad_norm": 0.9202889204025269, + "learning_rate": 4.997506912651906e-06, + "loss": 0.9598, + "step": 153 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 1.0630513429641724, + "learning_rate": 4.997411962736927e-06, + "loss": 0.8554, + "step": 154 + }, + { + "epoch": 0.24269311064718163, + "grad_norm": 0.8378801941871643, + "learning_rate": 4.997315239293742e-06, + "loss": 0.8583, + "step": 155 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 0.9442815184593201, + "learning_rate": 4.997216742391038e-06, + "loss": 0.8518, + "step": 156 + }, + { + "epoch": 0.24582463465553236, + "grad_norm": 0.9273308515548706, + "learning_rate": 4.997116472098763e-06, + "loss": 0.8942, + "step": 157 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 0.9028997421264648, + "learning_rate": 4.997014428488121e-06, + "loss": 0.8592, + "step": 158 + }, + { + "epoch": 0.2489561586638831, + "grad_norm": 0.9877567887306213, + "learning_rate": 4.996910611631582e-06, + "loss": 0.9316, + "step": 159 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 0.961592435836792, + "learning_rate": 4.996805021602868e-06, + "loss": 0.8875, + "step": 160 + }, + { + "epoch": 0.2520876826722338, + "grad_norm": 0.9923007488250732, + "learning_rate": 4.996697658476965e-06, + "loss": 0.8554, + "step": 161 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 0.956763744354248, + "learning_rate": 4.996588522330114e-06, + "loss": 0.9381, + "step": 162 + }, + { + "epoch": 0.25521920668058456, + "grad_norm": 0.919992208480835, + "learning_rate": 4.996477613239822e-06, + "loss": 0.9016, + "step": 163 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 0.9655581712722778, + "learning_rate": 4.996364931284847e-06, + "loss": 0.896, + "step": 164 + }, + { + "epoch": 0.25835073068893527, + "grad_norm": 0.9159606099128723, + "learning_rate": 4.996250476545212e-06, + "loss": 0.9475, + "step": 165 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 1.1147626638412476, + "learning_rate": 4.996134249102195e-06, + "loss": 0.8793, + "step": 166 + }, + { + "epoch": 0.26148225469728603, + "grad_norm": 1.112069010734558, + "learning_rate": 4.996016249038334e-06, + "loss": 0.8765, + "step": 167 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 1.0366944074630737, + "learning_rate": 4.995896476437428e-06, + "loss": 0.9356, + "step": 168 + }, + { + "epoch": 0.26461377870563674, + "grad_norm": 0.9494239091873169, + "learning_rate": 4.995774931384534e-06, + "loss": 0.8246, + "step": 169 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 0.9819200038909912, + "learning_rate": 4.995651613965964e-06, + "loss": 0.9403, + "step": 170 + }, + { + "epoch": 0.2677453027139875, + "grad_norm": 1.7329354286193848, + "learning_rate": 4.995526524269293e-06, + "loss": 0.9091, + "step": 171 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.1415139436721802, + "learning_rate": 4.995399662383352e-06, + "loss": 0.9106, + "step": 172 + }, + { + "epoch": 0.2708768267223382, + "grad_norm": 0.9471906423568726, + "learning_rate": 4.995271028398233e-06, + "loss": 0.8858, + "step": 173 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 1.096053123474121, + "learning_rate": 4.995140622405284e-06, + "loss": 0.7502, + "step": 174 + }, + { + "epoch": 0.2740083507306889, + "grad_norm": 1.285410761833191, + "learning_rate": 4.995008444497113e-06, + "loss": 0.9288, + "step": 175 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 0.9714595079421997, + "learning_rate": 4.994874494767585e-06, + "loss": 0.8824, + "step": 176 + }, + { + "epoch": 0.27713987473903967, + "grad_norm": 0.8399011492729187, + "learning_rate": 4.994738773311824e-06, + "loss": 0.8703, + "step": 177 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 0.9454224109649658, + "learning_rate": 4.994601280226212e-06, + "loss": 0.9627, + "step": 178 + }, + { + "epoch": 0.2802713987473904, + "grad_norm": 0.8983661532402039, + "learning_rate": 4.99446201560839e-06, + "loss": 0.9086, + "step": 179 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 0.9814153909683228, + "learning_rate": 4.994320979557256e-06, + "loss": 0.908, + "step": 180 + }, + { + "epoch": 0.28340292275574114, + "grad_norm": 0.8405355215072632, + "learning_rate": 4.9941781721729645e-06, + "loss": 0.8414, + "step": 181 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 0.950950026512146, + "learning_rate": 4.994033593556933e-06, + "loss": 0.9091, + "step": 182 + }, + { + "epoch": 0.28653444676409184, + "grad_norm": 1.1019747257232666, + "learning_rate": 4.99388724381183e-06, + "loss": 0.8305, + "step": 183 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 0.9339331984519958, + "learning_rate": 4.9937391230415875e-06, + "loss": 0.8542, + "step": 184 + }, + { + "epoch": 0.2896659707724426, + "grad_norm": 0.8813616633415222, + "learning_rate": 4.993589231351394e-06, + "loss": 0.848, + "step": 185 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 0.8916418552398682, + "learning_rate": 4.9934375688476905e-06, + "loss": 0.8032, + "step": 186 + }, + { + "epoch": 0.2927974947807933, + "grad_norm": 0.9891005754470825, + "learning_rate": 4.993284135638182e-06, + "loss": 0.7937, + "step": 187 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 0.9842082262039185, + "learning_rate": 4.99312893183183e-06, + "loss": 0.9532, + "step": 188 + }, + { + "epoch": 0.29592901878914407, + "grad_norm": 0.993478536605835, + "learning_rate": 4.99297195753885e-06, + "loss": 0.9406, + "step": 189 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 1.0654981136322021, + "learning_rate": 4.992813212870718e-06, + "loss": 0.907, + "step": 190 + }, + { + "epoch": 0.2990605427974948, + "grad_norm": 0.8937190771102905, + "learning_rate": 4.992652697940164e-06, + "loss": 0.9307, + "step": 191 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 0.9006052613258362, + "learning_rate": 4.992490412861178e-06, + "loss": 0.8808, + "step": 192 + }, + { + "epoch": 0.30219206680584554, + "grad_norm": 0.9601522088050842, + "learning_rate": 4.992326357749007e-06, + "loss": 0.8495, + "step": 193 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 0.9814243316650391, + "learning_rate": 4.992160532720153e-06, + "loss": 0.8204, + "step": 194 + }, + { + "epoch": 0.30532359081419624, + "grad_norm": 0.9660456776618958, + "learning_rate": 4.991992937892377e-06, + "loss": 0.8209, + "step": 195 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 0.8504270315170288, + "learning_rate": 4.991823573384695e-06, + "loss": 0.8415, + "step": 196 + }, + { + "epoch": 0.30845511482254695, + "grad_norm": 0.8733190894126892, + "learning_rate": 4.991652439317382e-06, + "loss": 0.8753, + "step": 197 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 0.9761794805526733, + "learning_rate": 4.991479535811967e-06, + "loss": 0.8923, + "step": 198 + }, + { + "epoch": 0.3115866388308977, + "grad_norm": 1.0400285720825195, + "learning_rate": 4.991304862991238e-06, + "loss": 0.899, + "step": 199 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 0.8561192750930786, + "learning_rate": 4.991128420979237e-06, + "loss": 0.9117, + "step": 200 + }, + { + "epoch": 0.3147181628392484, + "grad_norm": 0.8839634656906128, + "learning_rate": 4.990950209901263e-06, + "loss": 0.9022, + "step": 201 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 0.9569464921951294, + "learning_rate": 4.990770229883873e-06, + "loss": 0.9013, + "step": 202 + }, + { + "epoch": 0.3178496868475992, + "grad_norm": 1.0600861310958862, + "learning_rate": 4.990588481054881e-06, + "loss": 0.8387, + "step": 203 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 0.9492761492729187, + "learning_rate": 4.990404963543352e-06, + "loss": 0.8521, + "step": 204 + }, + { + "epoch": 0.3209812108559499, + "grad_norm": 0.9534817934036255, + "learning_rate": 4.990219677479612e-06, + "loss": 0.8009, + "step": 205 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 0.9918960928916931, + "learning_rate": 4.990032622995242e-06, + "loss": 0.8215, + "step": 206 + }, + { + "epoch": 0.32411273486430064, + "grad_norm": 0.8751630783081055, + "learning_rate": 4.989843800223077e-06, + "loss": 0.8744, + "step": 207 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 1.0034785270690918, + "learning_rate": 4.9896532092972085e-06, + "loss": 0.884, + "step": 208 + }, + { + "epoch": 0.32724425887265135, + "grad_norm": 1.2693432569503784, + "learning_rate": 4.989460850352984e-06, + "loss": 0.9689, + "step": 209 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 0.9923811554908752, + "learning_rate": 4.989266723527007e-06, + "loss": 0.9622, + "step": 210 + }, + { + "epoch": 0.3303757828810021, + "grad_norm": 0.9439074397087097, + "learning_rate": 4.989070828957137e-06, + "loss": 0.9155, + "step": 211 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.2564488649368286, + "learning_rate": 4.988873166782485e-06, + "loss": 0.8712, + "step": 212 + }, + { + "epoch": 0.3335073068893528, + "grad_norm": 1.1441881656646729, + "learning_rate": 4.988673737143423e-06, + "loss": 0.8895, + "step": 213 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 0.972155749797821, + "learning_rate": 4.988472540181574e-06, + "loss": 0.9158, + "step": 214 + }, + { + "epoch": 0.3366388308977035, + "grad_norm": 0.9536120295524597, + "learning_rate": 4.988269576039817e-06, + "loss": 0.7963, + "step": 215 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 0.8956594467163086, + "learning_rate": 4.988064844862288e-06, + "loss": 0.8947, + "step": 216 + }, + { + "epoch": 0.3397703549060543, + "grad_norm": 0.9503691792488098, + "learning_rate": 4.987858346794373e-06, + "loss": 0.7857, + "step": 217 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 0.9471456408500671, + "learning_rate": 4.987650081982719e-06, + "loss": 0.9118, + "step": 218 + }, + { + "epoch": 0.342901878914405, + "grad_norm": 0.9485713839530945, + "learning_rate": 4.987440050575222e-06, + "loss": 0.8188, + "step": 219 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 1.0775927305221558, + "learning_rate": 4.987228252721037e-06, + "loss": 0.9469, + "step": 220 + }, + { + "epoch": 0.34603340292275575, + "grad_norm": 0.9173988103866577, + "learning_rate": 4.9870146885705705e-06, + "loss": 0.8566, + "step": 221 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.008735179901123, + "learning_rate": 4.986799358275485e-06, + "loss": 0.8637, + "step": 222 + }, + { + "epoch": 0.34916492693110646, + "grad_norm": 0.8183198571205139, + "learning_rate": 4.986582261988695e-06, + "loss": 0.9321, + "step": 223 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 0.8951883316040039, + "learning_rate": 4.986363399864372e-06, + "loss": 0.8015, + "step": 224 + }, + { + "epoch": 0.3522964509394572, + "grad_norm": 0.9275704622268677, + "learning_rate": 4.986142772057941e-06, + "loss": 0.8242, + "step": 225 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 0.9548516273498535, + "learning_rate": 4.985920378726077e-06, + "loss": 0.7889, + "step": 226 + }, + { + "epoch": 0.3554279749478079, + "grad_norm": 1.1203585863113403, + "learning_rate": 4.985696220026714e-06, + "loss": 0.835, + "step": 227 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 0.9525771141052246, + "learning_rate": 4.985470296119038e-06, + "loss": 0.9162, + "step": 228 + }, + { + "epoch": 0.3585594989561587, + "grad_norm": 1.150404930114746, + "learning_rate": 4.985242607163488e-06, + "loss": 0.9456, + "step": 229 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.01652991771698, + "learning_rate": 4.985013153321755e-06, + "loss": 0.87, + "step": 230 + }, + { + "epoch": 0.3616910229645094, + "grad_norm": 0.9455218315124512, + "learning_rate": 4.984781934756786e-06, + "loss": 0.8466, + "step": 231 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 0.9125218391418457, + "learning_rate": 4.984548951632779e-06, + "loss": 0.8764, + "step": 232 + }, + { + "epoch": 0.36482254697286015, + "grad_norm": 2.0192980766296387, + "learning_rate": 4.984314204115187e-06, + "loss": 0.7866, + "step": 233 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.8118515014648438, + "learning_rate": 4.984077692370716e-06, + "loss": 0.8703, + "step": 234 + }, + { + "epoch": 0.36795407098121086, + "grad_norm": 1.2995024919509888, + "learning_rate": 4.983839416567322e-06, + "loss": 0.9459, + "step": 235 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 1.0611528158187866, + "learning_rate": 4.983599376874216e-06, + "loss": 0.8456, + "step": 236 + }, + { + "epoch": 0.37108559498956156, + "grad_norm": 1.2018043994903564, + "learning_rate": 4.983357573461863e-06, + "loss": 0.914, + "step": 237 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9091911315917969, + "learning_rate": 4.983114006501978e-06, + "loss": 0.791, + "step": 238 + }, + { + "epoch": 0.3742171189979123, + "grad_norm": 0.980873167514801, + "learning_rate": 4.982868676167528e-06, + "loss": 0.77, + "step": 239 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 0.8945695161819458, + "learning_rate": 4.982621582632735e-06, + "loss": 0.8038, + "step": 240 + }, + { + "epoch": 0.37734864300626303, + "grad_norm": 0.9325360655784607, + "learning_rate": 4.9823727260730705e-06, + "loss": 0.8908, + "step": 241 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 0.987630307674408, + "learning_rate": 4.98212210666526e-06, + "loss": 0.8642, + "step": 242 + }, + { + "epoch": 0.3804801670146138, + "grad_norm": 0.9151357412338257, + "learning_rate": 4.98186972458728e-06, + "loss": 0.8437, + "step": 243 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 0.9084850549697876, + "learning_rate": 4.981615580018358e-06, + "loss": 0.8775, + "step": 244 + }, + { + "epoch": 0.3836116910229645, + "grad_norm": 0.9786385297775269, + "learning_rate": 4.981359673138974e-06, + "loss": 0.8251, + "step": 245 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 0.8947877883911133, + "learning_rate": 4.98110200413086e-06, + "loss": 0.7709, + "step": 246 + }, + { + "epoch": 0.38674321503131526, + "grad_norm": 0.967019259929657, + "learning_rate": 4.9808425731769985e-06, + "loss": 0.8954, + "step": 247 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 1.1066704988479614, + "learning_rate": 4.980581380461622e-06, + "loss": 0.8544, + "step": 248 + }, + { + "epoch": 0.38987473903966596, + "grad_norm": 0.9750981330871582, + "learning_rate": 4.980318426170218e-06, + "loss": 0.8354, + "step": 249 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 0.8727232217788696, + "learning_rate": 4.980053710489521e-06, + "loss": 0.8911, + "step": 250 + }, + { + "epoch": 0.3930062630480167, + "grad_norm": 0.9337188601493835, + "learning_rate": 4.979787233607518e-06, + "loss": 0.8567, + "step": 251 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 0.890701174736023, + "learning_rate": 4.979518995713448e-06, + "loss": 0.8327, + "step": 252 + }, + { + "epoch": 0.39613778705636743, + "grad_norm": 0.9973195195198059, + "learning_rate": 4.979248996997797e-06, + "loss": 0.9065, + "step": 253 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 0.9894055128097534, + "learning_rate": 4.978977237652306e-06, + "loss": 0.8729, + "step": 254 + }, + { + "epoch": 0.39926931106471814, + "grad_norm": 1.0398536920547485, + "learning_rate": 4.978703717869961e-06, + "loss": 0.861, + "step": 255 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.1713942289352417, + "learning_rate": 4.978428437845003e-06, + "loss": 0.8183, + "step": 256 + }, + { + "epoch": 0.4024008350730689, + "grad_norm": 0.9868776202201843, + "learning_rate": 4.9781513977729225e-06, + "loss": 0.941, + "step": 257 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 0.9260863065719604, + "learning_rate": 4.977872597850455e-06, + "loss": 0.8963, + "step": 258 + }, + { + "epoch": 0.4055323590814196, + "grad_norm": 0.9553239941596985, + "learning_rate": 4.977592038275592e-06, + "loss": 0.8102, + "step": 259 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 0.9426907300949097, + "learning_rate": 4.977309719247571e-06, + "loss": 0.89, + "step": 260 + }, + { + "epoch": 0.40866388308977036, + "grad_norm": 0.9691527485847473, + "learning_rate": 4.97702564096688e-06, + "loss": 0.7967, + "step": 261 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 0.9917487502098083, + "learning_rate": 4.976739803635256e-06, + "loss": 0.8327, + "step": 262 + }, + { + "epoch": 0.41179540709812107, + "grad_norm": 0.992340087890625, + "learning_rate": 4.976452207455686e-06, + "loss": 0.9462, + "step": 263 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 0.9554483890533447, + "learning_rate": 4.9761628526324035e-06, + "loss": 0.8294, + "step": 264 + }, + { + "epoch": 0.41492693110647183, + "grad_norm": 0.9318789839744568, + "learning_rate": 4.975871739370895e-06, + "loss": 0.8079, + "step": 265 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 0.9158215522766113, + "learning_rate": 4.975578867877893e-06, + "loss": 0.8632, + "step": 266 + }, + { + "epoch": 0.41805845511482254, + "grad_norm": 0.9444370865821838, + "learning_rate": 4.975284238361378e-06, + "loss": 0.8897, + "step": 267 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 0.9949231743812561, + "learning_rate": 4.974987851030581e-06, + "loss": 0.7905, + "step": 268 + }, + { + "epoch": 0.4211899791231733, + "grad_norm": 0.9819440245628357, + "learning_rate": 4.974689706095981e-06, + "loss": 0.9183, + "step": 269 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0787206888198853, + "learning_rate": 4.974389803769304e-06, + "loss": 0.7739, + "step": 270 + }, + { + "epoch": 0.424321503131524, + "grad_norm": 0.8871361017227173, + "learning_rate": 4.9740881442635235e-06, + "loss": 0.9454, + "step": 271 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 0.9125893115997314, + "learning_rate": 4.973784727792863e-06, + "loss": 0.8017, + "step": 272 + }, + { + "epoch": 0.42745302713987476, + "grad_norm": 0.9102838635444641, + "learning_rate": 4.973479554572793e-06, + "loss": 0.8307, + "step": 273 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 0.8780474662780762, + "learning_rate": 4.973172624820031e-06, + "loss": 0.9402, + "step": 274 + }, + { + "epoch": 0.43058455114822547, + "grad_norm": 1.0059378147125244, + "learning_rate": 4.972863938752541e-06, + "loss": 0.7438, + "step": 275 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.9684615135192871, + "learning_rate": 4.972553496589537e-06, + "loss": 0.8897, + "step": 276 + }, + { + "epoch": 0.4337160751565762, + "grad_norm": 0.9615057110786438, + "learning_rate": 4.972241298551477e-06, + "loss": 0.8157, + "step": 277 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8589702844619751, + "learning_rate": 4.971927344860068e-06, + "loss": 0.7813, + "step": 278 + }, + { + "epoch": 0.43684759916492694, + "grad_norm": 0.900324285030365, + "learning_rate": 4.971611635738263e-06, + "loss": 0.7534, + "step": 279 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 0.8881833553314209, + "learning_rate": 4.971294171410261e-06, + "loss": 0.8255, + "step": 280 + }, + { + "epoch": 0.43997912317327764, + "grad_norm": 0.9723751544952393, + "learning_rate": 4.9709749521015085e-06, + "loss": 0.8796, + "step": 281 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 0.9341713190078735, + "learning_rate": 4.9706539780386995e-06, + "loss": 0.7429, + "step": 282 + }, + { + "epoch": 0.4431106471816284, + "grad_norm": 0.9354205131530762, + "learning_rate": 4.97033124944977e-06, + "loss": 0.8545, + "step": 283 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 0.9294624328613281, + "learning_rate": 4.970006766563906e-06, + "loss": 0.7783, + "step": 284 + }, + { + "epoch": 0.4462421711899791, + "grad_norm": 0.9584887027740479, + "learning_rate": 4.969680529611538e-06, + "loss": 0.8068, + "step": 285 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 0.9525735378265381, + "learning_rate": 4.9693525388243424e-06, + "loss": 0.8055, + "step": 286 + }, + { + "epoch": 0.44937369519832987, + "grad_norm": 0.8845317959785461, + "learning_rate": 4.96902279443524e-06, + "loss": 0.8318, + "step": 287 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 0.8932276964187622, + "learning_rate": 4.9686912966783965e-06, + "loss": 0.8185, + "step": 288 + }, + { + "epoch": 0.4525052192066806, + "grad_norm": 0.9881951808929443, + "learning_rate": 4.968358045789225e-06, + "loss": 0.9001, + "step": 289 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 0.995212197303772, + "learning_rate": 4.968023042004384e-06, + "loss": 0.75, + "step": 290 + }, + { + "epoch": 0.45563674321503134, + "grad_norm": 0.9997846484184265, + "learning_rate": 4.967686285561773e-06, + "loss": 0.8993, + "step": 291 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 0.9438545107841492, + "learning_rate": 4.967347776700538e-06, + "loss": 0.8964, + "step": 292 + }, + { + "epoch": 0.45876826722338204, + "grad_norm": 0.9164395332336426, + "learning_rate": 4.967007515661072e-06, + "loss": 0.8671, + "step": 293 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.895302951335907, + "learning_rate": 4.966665502685009e-06, + "loss": 0.8785, + "step": 294 + }, + { + "epoch": 0.4618997912317328, + "grad_norm": 0.9122863411903381, + "learning_rate": 4.966321738015227e-06, + "loss": 0.8499, + "step": 295 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.0097336769104004, + "learning_rate": 4.965976221895852e-06, + "loss": 0.83, + "step": 296 + }, + { + "epoch": 0.4650313152400835, + "grad_norm": 1.0664499998092651, + "learning_rate": 4.965628954572249e-06, + "loss": 0.7799, + "step": 297 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 0.9602341055870056, + "learning_rate": 4.96527993629103e-06, + "loss": 0.8061, + "step": 298 + }, + { + "epoch": 0.4681628392484342, + "grad_norm": 1.0133620500564575, + "learning_rate": 4.964929167300048e-06, + "loss": 0.9148, + "step": 299 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9351868033409119, + "learning_rate": 4.964576647848401e-06, + "loss": 0.83, + "step": 300 + }, + { + "epoch": 0.471294363256785, + "grad_norm": 1.0078860521316528, + "learning_rate": 4.964222378186429e-06, + "loss": 0.8391, + "step": 301 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 0.8966493606567383, + "learning_rate": 4.963866358565716e-06, + "loss": 0.8677, + "step": 302 + }, + { + "epoch": 0.4744258872651357, + "grad_norm": 0.9283657670021057, + "learning_rate": 4.963508589239087e-06, + "loss": 0.8787, + "step": 303 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 0.9540023803710938, + "learning_rate": 4.963149070460611e-06, + "loss": 0.8667, + "step": 304 + }, + { + "epoch": 0.47755741127348644, + "grad_norm": 0.9580283164978027, + "learning_rate": 4.9627878024855995e-06, + "loss": 0.8442, + "step": 305 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.9008278846740723, + "learning_rate": 4.962424785570605e-06, + "loss": 0.8605, + "step": 306 + }, + { + "epoch": 0.48068893528183715, + "grad_norm": 0.9681883454322815, + "learning_rate": 4.962060019973423e-06, + "loss": 0.7865, + "step": 307 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 0.8492597341537476, + "learning_rate": 4.9616935059530915e-06, + "loss": 0.8578, + "step": 308 + }, + { + "epoch": 0.4838204592901879, + "grad_norm": 0.9323164820671082, + "learning_rate": 4.9613252437698865e-06, + "loss": 0.8491, + "step": 309 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 0.9417833685874939, + "learning_rate": 4.960955233685331e-06, + "loss": 0.8108, + "step": 310 + }, + { + "epoch": 0.4869519832985386, + "grad_norm": 0.901364803314209, + "learning_rate": 4.960583475962185e-06, + "loss": 0.8958, + "step": 311 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8929979205131531, + "learning_rate": 4.96020997086445e-06, + "loss": 0.8237, + "step": 312 + }, + { + "epoch": 0.4900835073068894, + "grad_norm": 0.9422337412834167, + "learning_rate": 4.9598347186573716e-06, + "loss": 0.8024, + "step": 313 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 0.9165504574775696, + "learning_rate": 4.959457719607432e-06, + "loss": 0.8124, + "step": 314 + }, + { + "epoch": 0.4932150313152401, + "grad_norm": 1.0233370065689087, + "learning_rate": 4.959078973982356e-06, + "loss": 0.788, + "step": 315 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9375972151756287, + "learning_rate": 4.958698482051109e-06, + "loss": 0.839, + "step": 316 + }, + { + "epoch": 0.4963465553235908, + "grad_norm": 0.8805437684059143, + "learning_rate": 4.958316244083895e-06, + "loss": 0.7445, + "step": 317 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.0231893062591553, + "learning_rate": 4.95793226035216e-06, + "loss": 0.8257, + "step": 318 + }, + { + "epoch": 0.49947807933194155, + "grad_norm": 0.9125928282737732, + "learning_rate": 4.957546531128589e-06, + "loss": 0.7699, + "step": 319 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.8844016790390015, + "learning_rate": 4.957159056687104e-06, + "loss": 0.8784, + "step": 320 + }, + { + "epoch": 0.5026096033402923, + "grad_norm": 0.9770572781562805, + "learning_rate": 4.956769837302871e-06, + "loss": 0.8611, + "step": 321 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.9413725733757019, + "learning_rate": 4.956378873252293e-06, + "loss": 0.8056, + "step": 322 + }, + { + "epoch": 0.505741127348643, + "grad_norm": 0.9058850407600403, + "learning_rate": 4.955986164813008e-06, + "loss": 0.8162, + "step": 323 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 0.893100917339325, + "learning_rate": 4.955591712263901e-06, + "loss": 0.7554, + "step": 324 + }, + { + "epoch": 0.5088726513569938, + "grad_norm": 0.8967075943946838, + "learning_rate": 4.955195515885089e-06, + "loss": 0.8143, + "step": 325 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.9542785286903381, + "learning_rate": 4.954797575957929e-06, + "loss": 0.8102, + "step": 326 + }, + { + "epoch": 0.5120041753653445, + "grad_norm": 1.006451964378357, + "learning_rate": 4.954397892765018e-06, + "loss": 0.86, + "step": 327 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 0.9964715838432312, + "learning_rate": 4.953996466590189e-06, + "loss": 0.7166, + "step": 328 + }, + { + "epoch": 0.5151356993736952, + "grad_norm": 0.9568808674812317, + "learning_rate": 4.953593297718512e-06, + "loss": 0.8753, + "step": 329 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.9435825347900391, + "learning_rate": 4.953188386436298e-06, + "loss": 0.8961, + "step": 330 + }, + { + "epoch": 0.5182672233820459, + "grad_norm": 0.9386457800865173, + "learning_rate": 4.952781733031092e-06, + "loss": 0.8468, + "step": 331 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 0.9916554689407349, + "learning_rate": 4.952373337791678e-06, + "loss": 0.8274, + "step": 332 + }, + { + "epoch": 0.5213987473903967, + "grad_norm": 0.9351572394371033, + "learning_rate": 4.9519632010080765e-06, + "loss": 0.7964, + "step": 333 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9936269521713257, + "learning_rate": 4.951551322971545e-06, + "loss": 0.8354, + "step": 334 + }, + { + "epoch": 0.5245302713987474, + "grad_norm": 0.9678221940994263, + "learning_rate": 4.951137703974576e-06, + "loss": 0.9115, + "step": 335 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 0.9453133940696716, + "learning_rate": 4.9507223443109005e-06, + "loss": 0.7725, + "step": 336 + }, + { + "epoch": 0.5276617954070981, + "grad_norm": 0.9577226638793945, + "learning_rate": 4.950305244275484e-06, + "loss": 0.8679, + "step": 337 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 0.998489260673523, + "learning_rate": 4.94988640416453e-06, + "loss": 0.7505, + "step": 338 + }, + { + "epoch": 0.5307933194154488, + "grad_norm": 0.976026713848114, + "learning_rate": 4.949465824275475e-06, + "loss": 0.9083, + "step": 339 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.0041502714157104, + "learning_rate": 4.9490435049069925e-06, + "loss": 0.8815, + "step": 340 + }, + { + "epoch": 0.5339248434237995, + "grad_norm": 0.9839801788330078, + "learning_rate": 4.948619446358991e-06, + "loss": 0.8517, + "step": 341 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 0.9995140433311462, + "learning_rate": 4.948193648932616e-06, + "loss": 0.8157, + "step": 342 + }, + { + "epoch": 0.5370563674321504, + "grad_norm": 1.0782225131988525, + "learning_rate": 4.947766112930243e-06, + "loss": 0.9206, + "step": 343 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9981802105903625, + "learning_rate": 4.947336838655487e-06, + "loss": 0.8389, + "step": 344 + }, + { + "epoch": 0.5401878914405011, + "grad_norm": 0.9261236190795898, + "learning_rate": 4.946905826413195e-06, + "loss": 0.8035, + "step": 345 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.9195284247398376, + "learning_rate": 4.946473076509449e-06, + "loss": 0.8265, + "step": 346 + }, + { + "epoch": 0.5433194154488518, + "grad_norm": 0.9535123109817505, + "learning_rate": 4.946038589251566e-06, + "loss": 0.7528, + "step": 347 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 0.9969532489776611, + "learning_rate": 4.9456023649480935e-06, + "loss": 0.7787, + "step": 348 + }, + { + "epoch": 0.5464509394572025, + "grad_norm": 0.9612399935722351, + "learning_rate": 4.945164403908816e-06, + "loss": 0.8298, + "step": 349 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 0.9541627764701843, + "learning_rate": 4.944724706444749e-06, + "loss": 0.8039, + "step": 350 + }, + { + "epoch": 0.5495824634655533, + "grad_norm": 0.9916557669639587, + "learning_rate": 4.944283272868143e-06, + "loss": 0.9012, + "step": 351 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 0.8807661533355713, + "learning_rate": 4.94384010349248e-06, + "loss": 0.9005, + "step": 352 + }, + { + "epoch": 0.552713987473904, + "grad_norm": 1.0179708003997803, + "learning_rate": 4.943395198632476e-06, + "loss": 0.8845, + "step": 353 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.923014223575592, + "learning_rate": 4.942948558604077e-06, + "loss": 0.8173, + "step": 354 + }, + { + "epoch": 0.5558455114822547, + "grad_norm": 0.9318749308586121, + "learning_rate": 4.942500183724463e-06, + "loss": 0.8828, + "step": 355 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.0907000303268433, + "learning_rate": 4.942050074312048e-06, + "loss": 0.7951, + "step": 356 + }, + { + "epoch": 0.5589770354906054, + "grad_norm": 0.9811620712280273, + "learning_rate": 4.9415982306864744e-06, + "loss": 0.8826, + "step": 357 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9888150691986084, + "learning_rate": 4.9411446531686156e-06, + "loss": 0.8338, + "step": 358 + }, + { + "epoch": 0.5621085594989561, + "grad_norm": 0.9389257431030273, + "learning_rate": 4.9406893420805804e-06, + "loss": 0.8316, + "step": 359 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.9500603675842285, + "learning_rate": 4.940232297745705e-06, + "loss": 0.8764, + "step": 360 + }, + { + "epoch": 0.5652400835073069, + "grad_norm": 0.9748769998550415, + "learning_rate": 4.939773520488559e-06, + "loss": 0.8308, + "step": 361 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 0.9179681539535522, + "learning_rate": 4.93931301063494e-06, + "loss": 0.789, + "step": 362 + }, + { + "epoch": 0.5683716075156576, + "grad_norm": 0.8609325289726257, + "learning_rate": 4.938850768511878e-06, + "loss": 0.7368, + "step": 363 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 0.9375531673431396, + "learning_rate": 4.9383867944476325e-06, + "loss": 0.8164, + "step": 364 + }, + { + "epoch": 0.5715031315240083, + "grad_norm": 0.9386453032493591, + "learning_rate": 4.937921088771694e-06, + "loss": 0.7898, + "step": 365 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.025040864944458, + "learning_rate": 4.937453651814781e-06, + "loss": 0.8465, + "step": 366 + }, + { + "epoch": 0.574634655532359, + "grad_norm": 1.1037245988845825, + "learning_rate": 4.936984483908842e-06, + "loss": 0.7935, + "step": 367 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.230821132659912, + "learning_rate": 4.936513585387055e-06, + "loss": 0.8043, + "step": 368 + }, + { + "epoch": 0.5777661795407099, + "grad_norm": 1.090235948562622, + "learning_rate": 4.9360409565838265e-06, + "loss": 0.8726, + "step": 369 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8977810144424438, + "learning_rate": 4.935566597834793e-06, + "loss": 0.8667, + "step": 370 + }, + { + "epoch": 0.5808977035490606, + "grad_norm": 0.96401447057724, + "learning_rate": 4.935090509476818e-06, + "loss": 0.8494, + "step": 371 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 0.9290149211883545, + "learning_rate": 4.934612691847995e-06, + "loss": 0.7678, + "step": 372 + }, + { + "epoch": 0.5840292275574113, + "grad_norm": 0.9765584468841553, + "learning_rate": 4.934133145287644e-06, + "loss": 0.7947, + "step": 373 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 0.8939430117607117, + "learning_rate": 4.933651870136313e-06, + "loss": 0.885, + "step": 374 + }, + { + "epoch": 0.587160751565762, + "grad_norm": 0.9512760043144226, + "learning_rate": 4.933168866735777e-06, + "loss": 0.8863, + "step": 375 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 1.0334441661834717, + "learning_rate": 4.932684135429041e-06, + "loss": 0.8362, + "step": 376 + }, + { + "epoch": 0.5902922755741128, + "grad_norm": 0.9402525424957275, + "learning_rate": 4.932197676560334e-06, + "loss": 0.8758, + "step": 377 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 0.9615777730941772, + "learning_rate": 4.931709490475113e-06, + "loss": 0.8324, + "step": 378 + }, + { + "epoch": 0.5934237995824635, + "grad_norm": 0.9407138228416443, + "learning_rate": 4.931219577520061e-06, + "loss": 0.8454, + "step": 379 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.0415815114974976, + "learning_rate": 4.930727938043091e-06, + "loss": 0.8649, + "step": 380 + }, + { + "epoch": 0.5965553235908142, + "grad_norm": 1.0946165323257446, + "learning_rate": 4.9302345723933344e-06, + "loss": 0.7794, + "step": 381 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.0532523393630981, + "learning_rate": 4.929739480921157e-06, + "loss": 0.8555, + "step": 382 + }, + { + "epoch": 0.5996868475991649, + "grad_norm": 0.9441399574279785, + "learning_rate": 4.929242663978144e-06, + "loss": 0.7613, + "step": 383 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 0.9970793128013611, + "learning_rate": 4.92874412191711e-06, + "loss": 0.9037, + "step": 384 + }, + { + "epoch": 0.6028183716075156, + "grad_norm": 0.9777354001998901, + "learning_rate": 4.928243855092091e-06, + "loss": 0.9044, + "step": 385 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 0.9507855772972107, + "learning_rate": 4.92774186385835e-06, + "loss": 0.8359, + "step": 386 + }, + { + "epoch": 0.6059498956158664, + "grad_norm": 0.9827553033828735, + "learning_rate": 4.9272381485723765e-06, + "loss": 0.7938, + "step": 387 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 0.8896475434303284, + "learning_rate": 4.926732709591879e-06, + "loss": 0.8545, + "step": 388 + }, + { + "epoch": 0.6090814196242171, + "grad_norm": 0.9817705750465393, + "learning_rate": 4.926225547275795e-06, + "loss": 0.8456, + "step": 389 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 0.9646008610725403, + "learning_rate": 4.925716661984283e-06, + "loss": 0.7906, + "step": 390 + }, + { + "epoch": 0.6122129436325678, + "grad_norm": 0.8940444588661194, + "learning_rate": 4.9252060540787275e-06, + "loss": 0.8371, + "step": 391 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.0812394618988037, + "learning_rate": 4.924693723921734e-06, + "loss": 0.8056, + "step": 392 + }, + { + "epoch": 0.6153444676409185, + "grad_norm": 0.9953856468200684, + "learning_rate": 4.924179671877132e-06, + "loss": 0.8651, + "step": 393 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 0.9760703444480896, + "learning_rate": 4.9236638983099735e-06, + "loss": 0.845, + "step": 394 + }, + { + "epoch": 0.6184759916492694, + "grad_norm": 1.0127615928649902, + "learning_rate": 4.923146403586533e-06, + "loss": 0.7987, + "step": 395 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 0.8771396279335022, + "learning_rate": 4.9226271880743086e-06, + "loss": 0.8092, + "step": 396 + }, + { + "epoch": 0.6216075156576201, + "grad_norm": 0.8906831741333008, + "learning_rate": 4.9221062521420196e-06, + "loss": 0.8039, + "step": 397 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.0218392610549927, + "learning_rate": 4.9215835961596045e-06, + "loss": 0.8354, + "step": 398 + }, + { + "epoch": 0.6247390396659708, + "grad_norm": 1.0382038354873657, + "learning_rate": 4.921059220498227e-06, + "loss": 0.8021, + "step": 399 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 0.8997588157653809, + "learning_rate": 4.920533125530272e-06, + "loss": 0.8716, + "step": 400 + }, + { + "epoch": 0.6278705636743215, + "grad_norm": 0.8725724220275879, + "learning_rate": 4.920005311629342e-06, + "loss": 0.7934, + "step": 401 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.9289800524711609, + "learning_rate": 4.9194757791702625e-06, + "loss": 0.812, + "step": 402 + }, + { + "epoch": 0.6310020876826722, + "grad_norm": 0.9969078898429871, + "learning_rate": 4.918944528529079e-06, + "loss": 0.866, + "step": 403 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 0.9544941186904907, + "learning_rate": 4.918411560083058e-06, + "loss": 0.8035, + "step": 404 + }, + { + "epoch": 0.634133611691023, + "grad_norm": 0.9678053855895996, + "learning_rate": 4.917876874210686e-06, + "loss": 0.8254, + "step": 405 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 0.9782965779304504, + "learning_rate": 4.917340471291666e-06, + "loss": 0.7954, + "step": 406 + }, + { + "epoch": 0.6372651356993737, + "grad_norm": 0.9380600452423096, + "learning_rate": 4.916802351706923e-06, + "loss": 0.8287, + "step": 407 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 0.9455891847610474, + "learning_rate": 4.916262515838603e-06, + "loss": 0.8714, + "step": 408 + }, + { + "epoch": 0.6403966597077244, + "grad_norm": 0.8937395811080933, + "learning_rate": 4.915720964070065e-06, + "loss": 0.7618, + "step": 409 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 1.0304689407348633, + "learning_rate": 4.915177696785894e-06, + "loss": 0.8436, + "step": 410 + }, + { + "epoch": 0.6435281837160751, + "grad_norm": 0.9623110294342041, + "learning_rate": 4.914632714371885e-06, + "loss": 0.865, + "step": 411 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 0.9212867021560669, + "learning_rate": 4.914086017215059e-06, + "loss": 0.8232, + "step": 412 + }, + { + "epoch": 0.6466597077244259, + "grad_norm": 0.9817205667495728, + "learning_rate": 4.9135376057036485e-06, + "loss": 0.7926, + "step": 413 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 0.9576719403266907, + "learning_rate": 4.912987480227108e-06, + "loss": 0.8812, + "step": 414 + }, + { + "epoch": 0.6497912317327766, + "grad_norm": 0.9415407180786133, + "learning_rate": 4.912435641176106e-06, + "loss": 0.8985, + "step": 415 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 0.9084275364875793, + "learning_rate": 4.91188208894253e-06, + "loss": 0.8085, + "step": 416 + }, + { + "epoch": 0.6529227557411273, + "grad_norm": 0.8890969753265381, + "learning_rate": 4.911326823919482e-06, + "loss": 0.8127, + "step": 417 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 0.8727678656578064, + "learning_rate": 4.9107698465012815e-06, + "loss": 0.8656, + "step": 418 + }, + { + "epoch": 0.656054279749478, + "grad_norm": 1.0174667835235596, + "learning_rate": 4.910211157083465e-06, + "loss": 0.8358, + "step": 419 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.045121431350708, + "learning_rate": 4.909650756062782e-06, + "loss": 0.8239, + "step": 420 + }, + { + "epoch": 0.6591858037578288, + "grad_norm": 1.0688676834106445, + "learning_rate": 4.909088643837202e-06, + "loss": 0.9246, + "step": 421 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.9371877908706665, + "learning_rate": 4.908524820805905e-06, + "loss": 0.7862, + "step": 422 + }, + { + "epoch": 0.6623173277661796, + "grad_norm": 0.9567264318466187, + "learning_rate": 4.907959287369288e-06, + "loss": 0.8164, + "step": 423 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 0.9109166860580444, + "learning_rate": 4.907392043928964e-06, + "loss": 0.7016, + "step": 424 + }, + { + "epoch": 0.6654488517745303, + "grad_norm": 1.0378203392028809, + "learning_rate": 4.906823090887757e-06, + "loss": 0.8986, + "step": 425 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 0.9871951341629028, + "learning_rate": 4.906252428649708e-06, + "loss": 0.8366, + "step": 426 + }, + { + "epoch": 0.668580375782881, + "grad_norm": 0.8682780265808105, + "learning_rate": 4.905680057620072e-06, + "loss": 0.738, + "step": 427 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.8834647536277771, + "learning_rate": 4.9051059782053125e-06, + "loss": 0.8304, + "step": 428 + }, + { + "epoch": 0.6717118997912317, + "grad_norm": 1.0133378505706787, + "learning_rate": 4.904530190813113e-06, + "loss": 0.9369, + "step": 429 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 0.9967373013496399, + "learning_rate": 4.9039526958523676e-06, + "loss": 0.782, + "step": 430 + }, + { + "epoch": 0.6748434237995825, + "grad_norm": 1.0258032083511353, + "learning_rate": 4.90337349373318e-06, + "loss": 0.8214, + "step": 431 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 0.9814366102218628, + "learning_rate": 4.90279258486687e-06, + "loss": 0.8173, + "step": 432 + }, + { + "epoch": 0.6779749478079332, + "grad_norm": 0.9351440668106079, + "learning_rate": 4.902209969665966e-06, + "loss": 0.8815, + "step": 433 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 0.9490398168563843, + "learning_rate": 4.901625648544212e-06, + "loss": 0.8295, + "step": 434 + }, + { + "epoch": 0.6811064718162839, + "grad_norm": 1.0183534622192383, + "learning_rate": 4.9010396219165614e-06, + "loss": 0.8538, + "step": 435 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 0.9546058773994446, + "learning_rate": 4.900451890199179e-06, + "loss": 0.8385, + "step": 436 + }, + { + "epoch": 0.6842379958246346, + "grad_norm": 1.076120138168335, + "learning_rate": 4.8998624538094394e-06, + "loss": 0.8958, + "step": 437 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.8935732841491699, + "learning_rate": 4.899271313165929e-06, + "loss": 0.8674, + "step": 438 + }, + { + "epoch": 0.6873695198329853, + "grad_norm": 0.9574604630470276, + "learning_rate": 4.898678468688445e-06, + "loss": 0.8245, + "step": 439 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.931751549243927, + "learning_rate": 4.898083920797993e-06, + "loss": 0.8409, + "step": 440 + }, + { + "epoch": 0.6905010438413361, + "grad_norm": 0.937874436378479, + "learning_rate": 4.89748766991679e-06, + "loss": 0.8222, + "step": 441 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9606375694274902, + "learning_rate": 4.896889716468259e-06, + "loss": 0.8123, + "step": 442 + }, + { + "epoch": 0.6936325678496869, + "grad_norm": 0.8692249059677124, + "learning_rate": 4.896290060877038e-06, + "loss": 0.8532, + "step": 443 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 0.9597892165184021, + "learning_rate": 4.895688703568968e-06, + "loss": 0.8365, + "step": 444 + }, + { + "epoch": 0.6967640918580376, + "grad_norm": 0.958342432975769, + "learning_rate": 4.895085644971102e-06, + "loss": 0.8535, + "step": 445 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.9603601098060608, + "learning_rate": 4.894480885511699e-06, + "loss": 0.8227, + "step": 446 + }, + { + "epoch": 0.6998956158663883, + "grad_norm": 1.0010526180267334, + "learning_rate": 4.893874425620226e-06, + "loss": 0.8785, + "step": 447 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.0061219930648804, + "learning_rate": 4.893266265727361e-06, + "loss": 0.8336, + "step": 448 + }, + { + "epoch": 0.7030271398747391, + "grad_norm": 0.8967173099517822, + "learning_rate": 4.892656406264984e-06, + "loss": 0.8305, + "step": 449 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.962729275226593, + "learning_rate": 4.892044847666186e-06, + "loss": 0.7843, + "step": 450 + }, + { + "epoch": 0.7061586638830898, + "grad_norm": 1.0067342519760132, + "learning_rate": 4.891431590365262e-06, + "loss": 0.8111, + "step": 451 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 1.0252268314361572, + "learning_rate": 4.890816634797716e-06, + "loss": 0.8253, + "step": 452 + }, + { + "epoch": 0.7092901878914405, + "grad_norm": 0.9098564386367798, + "learning_rate": 4.890199981400255e-06, + "loss": 0.8526, + "step": 453 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 0.893135130405426, + "learning_rate": 4.889581630610793e-06, + "loss": 0.8217, + "step": 454 + }, + { + "epoch": 0.7124217118997912, + "grad_norm": 0.9225817322731018, + "learning_rate": 4.888961582868453e-06, + "loss": 0.8296, + "step": 455 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 0.973386824131012, + "learning_rate": 4.888339838613555e-06, + "loss": 0.7429, + "step": 456 + }, + { + "epoch": 0.715553235908142, + "grad_norm": 0.9829993844032288, + "learning_rate": 4.8877163982876306e-06, + "loss": 0.7405, + "step": 457 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 0.927728533744812, + "learning_rate": 4.887091262333414e-06, + "loss": 0.8257, + "step": 458 + }, + { + "epoch": 0.7186847599164927, + "grad_norm": 0.9419897794723511, + "learning_rate": 4.886464431194844e-06, + "loss": 0.7762, + "step": 459 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 0.9358823299407959, + "learning_rate": 4.885835905317061e-06, + "loss": 0.7794, + "step": 460 + }, + { + "epoch": 0.7218162839248434, + "grad_norm": 1.0290783643722534, + "learning_rate": 4.88520568514641e-06, + "loss": 0.8723, + "step": 461 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 0.9909068942070007, + "learning_rate": 4.884573771130442e-06, + "loss": 0.8302, + "step": 462 + }, + { + "epoch": 0.7249478079331941, + "grad_norm": 0.9972653985023499, + "learning_rate": 4.883940163717906e-06, + "loss": 0.7961, + "step": 463 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.9245638251304626, + "learning_rate": 4.883304863358757e-06, + "loss": 0.8057, + "step": 464 + }, + { + "epoch": 0.7280793319415448, + "grad_norm": 1.0097148418426514, + "learning_rate": 4.882667870504153e-06, + "loss": 0.7296, + "step": 465 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 0.9792670607566833, + "learning_rate": 4.882029185606448e-06, + "loss": 0.8006, + "step": 466 + }, + { + "epoch": 0.7312108559498957, + "grad_norm": 0.9308120608329773, + "learning_rate": 4.881388809119206e-06, + "loss": 0.7957, + "step": 467 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 0.9472301602363586, + "learning_rate": 4.880746741497187e-06, + "loss": 0.8984, + "step": 468 + }, + { + "epoch": 0.7343423799582464, + "grad_norm": 0.9898937344551086, + "learning_rate": 4.880102983196351e-06, + "loss": 0.8365, + "step": 469 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 0.9376612901687622, + "learning_rate": 4.879457534673863e-06, + "loss": 0.7838, + "step": 470 + }, + { + "epoch": 0.7374739039665971, + "grad_norm": 1.030616283416748, + "learning_rate": 4.878810396388085e-06, + "loss": 0.7698, + "step": 471 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.9583387970924377, + "learning_rate": 4.87816156879858e-06, + "loss": 0.7381, + "step": 472 + }, + { + "epoch": 0.7406054279749478, + "grad_norm": 1.00180184841156, + "learning_rate": 4.87751105236611e-06, + "loss": 0.8497, + "step": 473 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.9382562637329102, + "learning_rate": 4.876858847552638e-06, + "loss": 0.8698, + "step": 474 + }, + { + "epoch": 0.7437369519832986, + "grad_norm": 1.0104063749313354, + "learning_rate": 4.8762049548213255e-06, + "loss": 0.8444, + "step": 475 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 0.9321765303611755, + "learning_rate": 4.87554937463653e-06, + "loss": 0.8027, + "step": 476 + }, + { + "epoch": 0.7468684759916493, + "grad_norm": 0.9527666568756104, + "learning_rate": 4.874892107463811e-06, + "loss": 0.8069, + "step": 477 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 0.9576907157897949, + "learning_rate": 4.874233153769924e-06, + "loss": 0.774, + "step": 478 + }, + { + "epoch": 0.75, + "grad_norm": 1.0516161918640137, + "learning_rate": 4.873572514022822e-06, + "loss": 0.8176, + "step": 479 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.026928186416626, + "learning_rate": 4.872910188691657e-06, + "loss": 0.8797, + "step": 480 + }, + { + "epoch": 0.7531315240083507, + "grad_norm": 0.9312179088592529, + "learning_rate": 4.872246178246776e-06, + "loss": 0.7747, + "step": 481 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0676255226135254, + "learning_rate": 4.871580483159725e-06, + "loss": 0.7801, + "step": 482 + }, + { + "epoch": 0.7562630480167014, + "grad_norm": 1.066056489944458, + "learning_rate": 4.870913103903245e-06, + "loss": 0.7837, + "step": 483 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9388124346733093, + "learning_rate": 4.87024404095127e-06, + "loss": 0.7801, + "step": 484 + }, + { + "epoch": 0.7593945720250522, + "grad_norm": 0.9879791140556335, + "learning_rate": 4.8695732947789356e-06, + "loss": 0.7849, + "step": 485 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0002318620681763, + "learning_rate": 4.868900865862569e-06, + "loss": 0.8219, + "step": 486 + }, + { + "epoch": 0.7625260960334029, + "grad_norm": 0.9599125385284424, + "learning_rate": 4.8682267546796945e-06, + "loss": 0.8888, + "step": 487 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 0.9099973440170288, + "learning_rate": 4.867550961709027e-06, + "loss": 0.7585, + "step": 488 + }, + { + "epoch": 0.7656576200417536, + "grad_norm": 0.9715955257415771, + "learning_rate": 4.866873487430481e-06, + "loss": 0.8237, + "step": 489 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 1.0194706916809082, + "learning_rate": 4.86619433232516e-06, + "loss": 0.7708, + "step": 490 + }, + { + "epoch": 0.7687891440501043, + "grad_norm": 1.044005036354065, + "learning_rate": 4.865513496875367e-06, + "loss": 0.8, + "step": 491 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.9776361584663391, + "learning_rate": 4.8648309815645915e-06, + "loss": 0.7696, + "step": 492 + }, + { + "epoch": 0.7719206680584552, + "grad_norm": 0.9213030934333801, + "learning_rate": 4.864146786877521e-06, + "loss": 0.8558, + "step": 493 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.9361440539360046, + "learning_rate": 4.8634609133000336e-06, + "loss": 0.7851, + "step": 494 + }, + { + "epoch": 0.7750521920668059, + "grad_norm": 1.0199540853500366, + "learning_rate": 4.8627733613192e-06, + "loss": 0.8229, + "step": 495 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 1.0244604349136353, + "learning_rate": 4.862084131423283e-06, + "loss": 0.772, + "step": 496 + }, + { + "epoch": 0.7781837160751566, + "grad_norm": 0.9414368271827698, + "learning_rate": 4.861393224101736e-06, + "loss": 0.7579, + "step": 497 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9198965430259705, + "learning_rate": 4.860700639845205e-06, + "loss": 0.855, + "step": 498 + }, + { + "epoch": 0.7813152400835073, + "grad_norm": 0.9398712515830994, + "learning_rate": 4.8600063791455275e-06, + "loss": 0.7824, + "step": 499 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.9544969201087952, + "learning_rate": 4.8593104424957275e-06, + "loss": 0.8104, + "step": 500 + }, + { + "epoch": 0.784446764091858, + "grad_norm": 0.9522370100021362, + "learning_rate": 4.858612830390024e-06, + "loss": 0.813, + "step": 501 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.9830234050750732, + "learning_rate": 4.857913543323823e-06, + "loss": 0.8656, + "step": 502 + }, + { + "epoch": 0.7875782881002088, + "grad_norm": 1.0527008771896362, + "learning_rate": 4.8572125817937195e-06, + "loss": 0.8417, + "step": 503 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 0.9772782921791077, + "learning_rate": 4.856509946297501e-06, + "loss": 0.7983, + "step": 504 + }, + { + "epoch": 0.7907098121085595, + "grad_norm": 0.9250578880310059, + "learning_rate": 4.855805637334141e-06, + "loss": 0.7391, + "step": 505 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9792859554290771, + "learning_rate": 4.855099655403802e-06, + "loss": 0.7753, + "step": 506 + }, + { + "epoch": 0.7938413361169102, + "grad_norm": 0.8876479864120483, + "learning_rate": 4.854392001007834e-06, + "loss": 0.8249, + "step": 507 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.9336771368980408, + "learning_rate": 4.853682674648775e-06, + "loss": 0.8462, + "step": 508 + }, + { + "epoch": 0.7969728601252609, + "grad_norm": 0.9732388854026794, + "learning_rate": 4.852971676830352e-06, + "loss": 0.8791, + "step": 509 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.9135870337486267, + "learning_rate": 4.852259008057476e-06, + "loss": 0.7279, + "step": 510 + }, + { + "epoch": 0.8001043841336117, + "grad_norm": 0.9782464504241943, + "learning_rate": 4.8515446688362476e-06, + "loss": 0.7797, + "step": 511 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 0.9571824073791504, + "learning_rate": 4.85082865967395e-06, + "loss": 0.8349, + "step": 512 + }, + { + "epoch": 0.8032359081419624, + "grad_norm": 1.011786937713623, + "learning_rate": 4.850110981079057e-06, + "loss": 0.8057, + "step": 513 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 0.9418787956237793, + "learning_rate": 4.8493916335612255e-06, + "loss": 0.8592, + "step": 514 + }, + { + "epoch": 0.8063674321503131, + "grad_norm": 0.9292604923248291, + "learning_rate": 4.848670617631296e-06, + "loss": 0.7471, + "step": 515 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.9303188323974609, + "learning_rate": 4.847947933801296e-06, + "loss": 0.8768, + "step": 516 + }, + { + "epoch": 0.8094989561586639, + "grad_norm": 0.9897359609603882, + "learning_rate": 4.847223582584437e-06, + "loss": 0.8343, + "step": 517 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.9695891737937927, + "learning_rate": 4.846497564495114e-06, + "loss": 0.8318, + "step": 518 + }, + { + "epoch": 0.8126304801670147, + "grad_norm": 1.0175273418426514, + "learning_rate": 4.845769880048906e-06, + "loss": 0.9137, + "step": 519 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9497922658920288, + "learning_rate": 4.845040529762577e-06, + "loss": 0.6549, + "step": 520 + }, + { + "epoch": 0.8157620041753654, + "grad_norm": 1.0647574663162231, + "learning_rate": 4.844309514154071e-06, + "loss": 0.8243, + "step": 521 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.9710965156555176, + "learning_rate": 4.843576833742517e-06, + "loss": 0.7864, + "step": 522 + }, + { + "epoch": 0.8188935281837161, + "grad_norm": 0.9881441593170166, + "learning_rate": 4.842842489048225e-06, + "loss": 0.7799, + "step": 523 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 0.9455597400665283, + "learning_rate": 4.842106480592687e-06, + "loss": 0.7943, + "step": 524 + }, + { + "epoch": 0.8220250521920668, + "grad_norm": 0.9948442578315735, + "learning_rate": 4.841368808898577e-06, + "loss": 0.8587, + "step": 525 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 0.9959186315536499, + "learning_rate": 4.84062947448975e-06, + "loss": 0.8477, + "step": 526 + }, + { + "epoch": 0.8251565762004175, + "grad_norm": 0.9791848659515381, + "learning_rate": 4.839888477891243e-06, + "loss": 0.7568, + "step": 527 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 1.1499426364898682, + "learning_rate": 4.839145819629269e-06, + "loss": 0.8296, + "step": 528 + }, + { + "epoch": 0.8282881002087683, + "grad_norm": 0.9269506335258484, + "learning_rate": 4.8384015002312265e-06, + "loss": 0.7935, + "step": 529 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.970565140247345, + "learning_rate": 4.837655520225692e-06, + "loss": 0.8138, + "step": 530 + }, + { + "epoch": 0.831419624217119, + "grad_norm": 0.9852119088172913, + "learning_rate": 4.836907880142418e-06, + "loss": 0.7909, + "step": 531 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.927129328250885, + "learning_rate": 4.836158580512339e-06, + "loss": 0.8105, + "step": 532 + }, + { + "epoch": 0.8345511482254697, + "grad_norm": 1.0711781978607178, + "learning_rate": 4.835407621867569e-06, + "loss": 0.8241, + "step": 533 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 1.0472421646118164, + "learning_rate": 4.834655004741397e-06, + "loss": 0.7923, + "step": 534 + }, + { + "epoch": 0.8376826722338204, + "grad_norm": 0.9761579632759094, + "learning_rate": 4.833900729668293e-06, + "loss": 0.8203, + "step": 535 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 1.0559637546539307, + "learning_rate": 4.833144797183902e-06, + "loss": 0.8915, + "step": 536 + }, + { + "epoch": 0.8408141962421712, + "grad_norm": 0.9951052069664001, + "learning_rate": 4.832387207825045e-06, + "loss": 0.8253, + "step": 537 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.894079864025116, + "learning_rate": 4.831627962129725e-06, + "loss": 0.7479, + "step": 538 + }, + { + "epoch": 0.843945720250522, + "grad_norm": 0.9832531213760376, + "learning_rate": 4.830867060637115e-06, + "loss": 0.8484, + "step": 539 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0477591753005981, + "learning_rate": 4.8301045038875665e-06, + "loss": 0.8052, + "step": 540 + }, + { + "epoch": 0.8470772442588727, + "grad_norm": 0.9388875365257263, + "learning_rate": 4.829340292422607e-06, + "loss": 0.7998, + "step": 541 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 0.8975532054901123, + "learning_rate": 4.8285744267849395e-06, + "loss": 0.848, + "step": 542 + }, + { + "epoch": 0.8502087682672234, + "grad_norm": 0.9211076498031616, + "learning_rate": 4.8278069075184396e-06, + "loss": 0.8337, + "step": 543 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.970134437084198, + "learning_rate": 4.827037735168159e-06, + "loss": 0.7843, + "step": 544 + }, + { + "epoch": 0.8533402922755741, + "grad_norm": 0.9714353084564209, + "learning_rate": 4.826266910280322e-06, + "loss": 0.6991, + "step": 545 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 0.8607689142227173, + "learning_rate": 4.8254944334023265e-06, + "loss": 0.7314, + "step": 546 + }, + { + "epoch": 0.8564718162839249, + "grad_norm": 0.9843030571937561, + "learning_rate": 4.824720305082747e-06, + "loss": 0.7497, + "step": 547 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 0.9638474583625793, + "learning_rate": 4.823944525871324e-06, + "loss": 0.776, + "step": 548 + }, + { + "epoch": 0.8596033402922756, + "grad_norm": 0.9342365860939026, + "learning_rate": 4.823167096318979e-06, + "loss": 0.77, + "step": 549 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9608579874038696, + "learning_rate": 4.822388016977797e-06, + "loss": 0.8281, + "step": 550 + }, + { + "epoch": 0.8627348643006263, + "grad_norm": 0.8996383547782898, + "learning_rate": 4.8216072884010395e-06, + "loss": 0.7519, + "step": 551 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 0.9922810196876526, + "learning_rate": 4.820824911143139e-06, + "loss": 0.8714, + "step": 552 + }, + { + "epoch": 0.865866388308977, + "grad_norm": 0.9861842393875122, + "learning_rate": 4.820040885759697e-06, + "loss": 0.7998, + "step": 553 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 1.013168215751648, + "learning_rate": 4.819255212807486e-06, + "loss": 0.8291, + "step": 554 + }, + { + "epoch": 0.8689979123173278, + "grad_norm": 0.9695484638214111, + "learning_rate": 4.818467892844451e-06, + "loss": 0.8257, + "step": 555 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 1.0018948316574097, + "learning_rate": 4.817678926429702e-06, + "loss": 0.7551, + "step": 556 + }, + { + "epoch": 0.8721294363256785, + "grad_norm": 1.0259422063827515, + "learning_rate": 4.816888314123521e-06, + "loss": 0.8582, + "step": 557 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9045076966285706, + "learning_rate": 4.81609605648736e-06, + "loss": 0.811, + "step": 558 + }, + { + "epoch": 0.8752609603340292, + "grad_norm": 0.9112316370010376, + "learning_rate": 4.815302154083837e-06, + "loss": 0.818, + "step": 559 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0060032606124878, + "learning_rate": 4.814506607476739e-06, + "loss": 0.766, + "step": 560 + }, + { + "epoch": 0.8783924843423799, + "grad_norm": 0.9126970171928406, + "learning_rate": 4.813709417231021e-06, + "loss": 0.8903, + "step": 561 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.0825906991958618, + "learning_rate": 4.812910583912804e-06, + "loss": 0.8466, + "step": 562 + }, + { + "epoch": 0.8815240083507306, + "grad_norm": 0.9530975818634033, + "learning_rate": 4.812110108089377e-06, + "loss": 0.8163, + "step": 563 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.9566405415534973, + "learning_rate": 4.8113079903291955e-06, + "loss": 0.8244, + "step": 564 + }, + { + "epoch": 0.8846555323590815, + "grad_norm": 0.919980525970459, + "learning_rate": 4.81050423120188e-06, + "loss": 0.8335, + "step": 565 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.9123868346214294, + "learning_rate": 4.809698831278217e-06, + "loss": 0.7712, + "step": 566 + }, + { + "epoch": 0.8877870563674322, + "grad_norm": 0.9589663743972778, + "learning_rate": 4.8088917911301595e-06, + "loss": 0.8467, + "step": 567 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 0.9052798748016357, + "learning_rate": 4.808083111330823e-06, + "loss": 0.8078, + "step": 568 + }, + { + "epoch": 0.8909185803757829, + "grad_norm": 0.9250494241714478, + "learning_rate": 4.807272792454489e-06, + "loss": 0.8536, + "step": 569 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.8451298475265503, + "learning_rate": 4.806460835076603e-06, + "loss": 0.7752, + "step": 570 + }, + { + "epoch": 0.8940501043841336, + "grad_norm": 1.0324240922927856, + "learning_rate": 4.805647239773773e-06, + "loss": 0.7871, + "step": 571 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 0.9338528513908386, + "learning_rate": 4.804832007123771e-06, + "loss": 0.8082, + "step": 572 + }, + { + "epoch": 0.8971816283924844, + "grad_norm": 1.0080739259719849, + "learning_rate": 4.804015137705531e-06, + "loss": 0.7636, + "step": 573 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9817036986351013, + "learning_rate": 4.803196632099152e-06, + "loss": 0.8599, + "step": 574 + }, + { + "epoch": 0.9003131524008351, + "grad_norm": 0.9284219145774841, + "learning_rate": 4.80237649088589e-06, + "loss": 0.8152, + "step": 575 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.8922317028045654, + "learning_rate": 4.801554714648166e-06, + "loss": 0.8146, + "step": 576 + }, + { + "epoch": 0.9034446764091858, + "grad_norm": 0.9465014338493347, + "learning_rate": 4.800731303969565e-06, + "loss": 0.7663, + "step": 577 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.0325525999069214, + "learning_rate": 4.799906259434824e-06, + "loss": 0.8458, + "step": 578 + }, + { + "epoch": 0.9065762004175365, + "grad_norm": 0.9366307258605957, + "learning_rate": 4.7990795816298485e-06, + "loss": 0.7992, + "step": 579 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 0.9923847317695618, + "learning_rate": 4.7982512711416995e-06, + "loss": 0.7256, + "step": 580 + }, + { + "epoch": 0.9097077244258872, + "grad_norm": 1.0387133359909058, + "learning_rate": 4.7974213285586e-06, + "loss": 0.7766, + "step": 581 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.9617043733596802, + "learning_rate": 4.79658975446993e-06, + "loss": 0.7602, + "step": 582 + }, + { + "epoch": 0.912839248434238, + "grad_norm": 0.9530353546142578, + "learning_rate": 4.795756549466229e-06, + "loss": 0.7922, + "step": 583 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.9305918216705322, + "learning_rate": 4.794921714139194e-06, + "loss": 0.8377, + "step": 584 + }, + { + "epoch": 0.9159707724425887, + "grad_norm": 0.899860143661499, + "learning_rate": 4.794085249081682e-06, + "loss": 0.8343, + "step": 585 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.9024940729141235, + "learning_rate": 4.793247154887703e-06, + "loss": 0.8271, + "step": 586 + }, + { + "epoch": 0.9191022964509394, + "grad_norm": 0.980869472026825, + "learning_rate": 4.7924074321524286e-06, + "loss": 0.7468, + "step": 587 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.0506541728973389, + "learning_rate": 4.791566081472185e-06, + "loss": 0.9207, + "step": 588 + }, + { + "epoch": 0.9222338204592901, + "grad_norm": 1.0723917484283447, + "learning_rate": 4.790723103444454e-06, + "loss": 0.7615, + "step": 589 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 1.0455081462860107, + "learning_rate": 4.789878498667873e-06, + "loss": 0.8115, + "step": 590 + }, + { + "epoch": 0.925365344467641, + "grad_norm": 0.9729639291763306, + "learning_rate": 4.789032267742234e-06, + "loss": 0.8622, + "step": 591 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.9695239067077637, + "learning_rate": 4.788184411268488e-06, + "loss": 0.8165, + "step": 592 + }, + { + "epoch": 0.9284968684759917, + "grad_norm": 1.0099278688430786, + "learning_rate": 4.7873349298487345e-06, + "loss": 0.759, + "step": 593 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 0.9377149939537048, + "learning_rate": 4.786483824086231e-06, + "loss": 0.8323, + "step": 594 + }, + { + "epoch": 0.9316283924843424, + "grad_norm": 0.8885690569877625, + "learning_rate": 4.785631094585387e-06, + "loss": 0.8448, + "step": 595 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 0.9133430123329163, + "learning_rate": 4.784776741951766e-06, + "loss": 0.8135, + "step": 596 + }, + { + "epoch": 0.9347599164926931, + "grad_norm": 1.015851378440857, + "learning_rate": 4.783920766792082e-06, + "loss": 0.8464, + "step": 597 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.8883825540542603, + "learning_rate": 4.783063169714203e-06, + "loss": 0.7445, + "step": 598 + }, + { + "epoch": 0.9378914405010439, + "grad_norm": 0.9872002005577087, + "learning_rate": 4.782203951327149e-06, + "loss": 0.8293, + "step": 599 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 1.0305813550949097, + "learning_rate": 4.781343112241091e-06, + "loss": 0.8606, + "step": 600 + }, + { + "epoch": 0.9410229645093946, + "grad_norm": 0.9667104482650757, + "learning_rate": 4.780480653067351e-06, + "loss": 0.8057, + "step": 601 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 1.067185640335083, + "learning_rate": 4.7796165744184e-06, + "loss": 0.849, + "step": 602 + }, + { + "epoch": 0.9441544885177453, + "grad_norm": 0.9017406105995178, + "learning_rate": 4.778750876907862e-06, + "loss": 0.7927, + "step": 603 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.9827666878700256, + "learning_rate": 4.777883561150508e-06, + "loss": 0.8087, + "step": 604 + }, + { + "epoch": 0.947286012526096, + "grad_norm": 0.9627811908721924, + "learning_rate": 4.77701462776226e-06, + "loss": 0.8488, + "step": 605 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0321338176727295, + "learning_rate": 4.776144077360186e-06, + "loss": 0.8289, + "step": 606 + }, + { + "epoch": 0.9504175365344467, + "grad_norm": 1.0012633800506592, + "learning_rate": 4.775271910562508e-06, + "loss": 0.7184, + "step": 607 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.9186056852340698, + "learning_rate": 4.7743981279885875e-06, + "loss": 0.8727, + "step": 608 + }, + { + "epoch": 0.9535490605427975, + "grad_norm": 0.910044252872467, + "learning_rate": 4.773522730258942e-06, + "loss": 0.8455, + "step": 609 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.9342944025993347, + "learning_rate": 4.77264571799523e-06, + "loss": 0.7548, + "step": 610 + }, + { + "epoch": 0.9566805845511482, + "grad_norm": 0.9609628319740295, + "learning_rate": 4.771767091820259e-06, + "loss": 0.8294, + "step": 611 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9708508253097534, + "learning_rate": 4.770886852357983e-06, + "loss": 0.7857, + "step": 612 + }, + { + "epoch": 0.9598121085594989, + "grad_norm": 1.0910807847976685, + "learning_rate": 4.7700050002334995e-06, + "loss": 0.8162, + "step": 613 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 0.9703771471977234, + "learning_rate": 4.769121536073054e-06, + "loss": 0.7205, + "step": 614 + }, + { + "epoch": 0.9629436325678496, + "grad_norm": 1.027938961982727, + "learning_rate": 4.768236460504035e-06, + "loss": 0.7641, + "step": 615 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 1.0003989934921265, + "learning_rate": 4.767349774154974e-06, + "loss": 0.7588, + "step": 616 + }, + { + "epoch": 0.9660751565762005, + "grad_norm": 0.9572570323944092, + "learning_rate": 4.766461477655552e-06, + "loss": 0.8274, + "step": 617 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.977287769317627, + "learning_rate": 4.765571571636586e-06, + "loss": 0.7856, + "step": 618 + }, + { + "epoch": 0.9692066805845512, + "grad_norm": 0.9582562446594238, + "learning_rate": 4.764680056730042e-06, + "loss": 0.8102, + "step": 619 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.9842986464500427, + "learning_rate": 4.763786933569025e-06, + "loss": 0.7515, + "step": 620 + }, + { + "epoch": 0.9723382045929019, + "grad_norm": 0.9955743551254272, + "learning_rate": 4.762892202787782e-06, + "loss": 0.7398, + "step": 621 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 1.0011107921600342, + "learning_rate": 4.761995865021706e-06, + "loss": 0.8182, + "step": 622 + }, + { + "epoch": 0.9754697286012526, + "grad_norm": 0.9586448669433594, + "learning_rate": 4.761097920907325e-06, + "loss": 0.8434, + "step": 623 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.979123055934906, + "learning_rate": 4.760198371082312e-06, + "loss": 0.7954, + "step": 624 + }, + { + "epoch": 0.9786012526096033, + "grad_norm": 1.0081250667572021, + "learning_rate": 4.75929721618548e-06, + "loss": 0.7931, + "step": 625 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 1.027185320854187, + "learning_rate": 4.75839445685678e-06, + "loss": 0.7339, + "step": 626 + }, + { + "epoch": 0.9817327766179541, + "grad_norm": 1.0553215742111206, + "learning_rate": 4.757490093737303e-06, + "loss": 0.7639, + "step": 627 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 0.8578981757164001, + "learning_rate": 4.75658412746928e-06, + "loss": 0.8128, + "step": 628 + }, + { + "epoch": 0.9848643006263048, + "grad_norm": 1.0143071413040161, + "learning_rate": 4.75567655869608e-06, + "loss": 0.8169, + "step": 629 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 0.9743554592132568, + "learning_rate": 4.754767388062208e-06, + "loss": 0.8179, + "step": 630 + }, + { + "epoch": 0.9879958246346555, + "grad_norm": 1.0276602506637573, + "learning_rate": 4.7538566162133105e-06, + "loss": 0.8009, + "step": 631 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.9296634793281555, + "learning_rate": 4.752944243796167e-06, + "loss": 0.8262, + "step": 632 + }, + { + "epoch": 0.9911273486430062, + "grad_norm": 0.9216791987419128, + "learning_rate": 4.752030271458696e-06, + "loss": 0.8929, + "step": 633 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 0.9272832274436951, + "learning_rate": 4.751114699849954e-06, + "loss": 0.7396, + "step": 634 + }, + { + "epoch": 0.994258872651357, + "grad_norm": 0.9121203422546387, + "learning_rate": 4.750197529620127e-06, + "loss": 0.8311, + "step": 635 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.999600887298584, + "learning_rate": 4.7492787614205425e-06, + "loss": 0.8125, + "step": 636 + }, + { + "epoch": 0.9973903966597077, + "grad_norm": 1.012978196144104, + "learning_rate": 4.748358395903661e-06, + "loss": 0.7966, + "step": 637 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.9767918586730957, + "learning_rate": 4.747436433723075e-06, + "loss": 0.8024, + "step": 638 + }, + { + "epoch": 1.0015657620041754, + "grad_norm": 1.8302738666534424, + "learning_rate": 4.7465128755335135e-06, + "loss": 1.6427, + "step": 639 + }, + { + "epoch": 1.0031315240083507, + "grad_norm": 0.8704890012741089, + "learning_rate": 4.7455877219908386e-06, + "loss": 0.7766, + "step": 640 + }, + { + "epoch": 1.004697286012526, + "grad_norm": 0.9848434925079346, + "learning_rate": 4.744660973752044e-06, + "loss": 0.8314, + "step": 641 + }, + { + "epoch": 1.0062630480167014, + "grad_norm": 1.054824709892273, + "learning_rate": 4.743732631475258e-06, + "loss": 0.797, + "step": 642 + }, + { + "epoch": 1.0078288100208768, + "grad_norm": 0.9286114573478699, + "learning_rate": 4.742802695819737e-06, + "loss": 0.764, + "step": 643 + }, + { + "epoch": 1.0093945720250521, + "grad_norm": 0.8625781536102295, + "learning_rate": 4.7418711674458735e-06, + "loss": 0.6691, + "step": 644 + }, + { + "epoch": 1.0109603340292275, + "grad_norm": 0.8710452914237976, + "learning_rate": 4.740938047015188e-06, + "loss": 0.7277, + "step": 645 + }, + { + "epoch": 1.0125260960334028, + "grad_norm": 0.8873130679130554, + "learning_rate": 4.740003335190334e-06, + "loss": 0.6531, + "step": 646 + }, + { + "epoch": 1.0140918580375784, + "grad_norm": 0.9777164459228516, + "learning_rate": 4.7390670326350925e-06, + "loss": 0.8047, + "step": 647 + }, + { + "epoch": 1.0156576200417538, + "grad_norm": 0.886025071144104, + "learning_rate": 4.738129140014377e-06, + "loss": 0.715, + "step": 648 + }, + { + "epoch": 1.017223382045929, + "grad_norm": 0.9825332164764404, + "learning_rate": 4.737189657994226e-06, + "loss": 0.7783, + "step": 649 + }, + { + "epoch": 1.0187891440501045, + "grad_norm": 0.9331428408622742, + "learning_rate": 4.736248587241812e-06, + "loss": 0.7598, + "step": 650 + }, + { + "epoch": 1.0203549060542798, + "grad_norm": 0.9406535625457764, + "learning_rate": 4.735305928425431e-06, + "loss": 0.7989, + "step": 651 + }, + { + "epoch": 1.0219206680584552, + "grad_norm": 1.0644745826721191, + "learning_rate": 4.734361682214511e-06, + "loss": 0.8135, + "step": 652 + }, + { + "epoch": 1.0234864300626305, + "grad_norm": 0.9037052392959595, + "learning_rate": 4.733415849279603e-06, + "loss": 0.7051, + "step": 653 + }, + { + "epoch": 1.0250521920668059, + "grad_norm": 1.000563144683838, + "learning_rate": 4.7324684302923865e-06, + "loss": 0.7427, + "step": 654 + }, + { + "epoch": 1.0266179540709812, + "grad_norm": 0.9181354641914368, + "learning_rate": 4.731519425925671e-06, + "loss": 0.7236, + "step": 655 + }, + { + "epoch": 1.0281837160751566, + "grad_norm": 1.0283046960830688, + "learning_rate": 4.730568836853384e-06, + "loss": 0.7998, + "step": 656 + }, + { + "epoch": 1.029749478079332, + "grad_norm": 0.8692315220832825, + "learning_rate": 4.729616663750587e-06, + "loss": 0.7974, + "step": 657 + }, + { + "epoch": 1.0313152400835073, + "grad_norm": 0.928920567035675, + "learning_rate": 4.72866290729346e-06, + "loss": 0.7199, + "step": 658 + }, + { + "epoch": 1.0328810020876826, + "grad_norm": 0.9066957235336304, + "learning_rate": 4.727707568159311e-06, + "loss": 0.791, + "step": 659 + }, + { + "epoch": 1.034446764091858, + "grad_norm": 1.0086729526519775, + "learning_rate": 4.726750647026569e-06, + "loss": 0.7113, + "step": 660 + }, + { + "epoch": 1.0360125260960333, + "grad_norm": 0.931856632232666, + "learning_rate": 4.72579214457479e-06, + "loss": 0.7977, + "step": 661 + }, + { + "epoch": 1.0375782881002087, + "grad_norm": 0.9191203713417053, + "learning_rate": 4.724832061484651e-06, + "loss": 0.7979, + "step": 662 + }, + { + "epoch": 1.039144050104384, + "grad_norm": 0.9614707827568054, + "learning_rate": 4.72387039843795e-06, + "loss": 0.6994, + "step": 663 + }, + { + "epoch": 1.0407098121085594, + "grad_norm": 1.014823317527771, + "learning_rate": 4.72290715611761e-06, + "loss": 0.7539, + "step": 664 + }, + { + "epoch": 1.042275574112735, + "grad_norm": 0.9177424311637878, + "learning_rate": 4.721942335207673e-06, + "loss": 0.6625, + "step": 665 + }, + { + "epoch": 1.0438413361169103, + "grad_norm": 1.0253031253814697, + "learning_rate": 4.720975936393305e-06, + "loss": 0.7747, + "step": 666 + }, + { + "epoch": 1.0454070981210857, + "grad_norm": 1.0004568099975586, + "learning_rate": 4.720007960360788e-06, + "loss": 0.7008, + "step": 667 + }, + { + "epoch": 1.046972860125261, + "grad_norm": 1.1344349384307861, + "learning_rate": 4.719038407797529e-06, + "loss": 0.7493, + "step": 668 + }, + { + "epoch": 1.0485386221294364, + "grad_norm": 0.9944799542427063, + "learning_rate": 4.718067279392052e-06, + "loss": 0.7823, + "step": 669 + }, + { + "epoch": 1.0501043841336117, + "grad_norm": 0.9469346404075623, + "learning_rate": 4.717094575833999e-06, + "loss": 0.7419, + "step": 670 + }, + { + "epoch": 1.051670146137787, + "grad_norm": 1.015339732170105, + "learning_rate": 4.716120297814133e-06, + "loss": 0.7168, + "step": 671 + }, + { + "epoch": 1.0532359081419624, + "grad_norm": 0.9544686675071716, + "learning_rate": 4.715144446024333e-06, + "loss": 0.791, + "step": 672 + }, + { + "epoch": 1.0548016701461378, + "grad_norm": 1.0285652875900269, + "learning_rate": 4.714167021157599e-06, + "loss": 0.7649, + "step": 673 + }, + { + "epoch": 1.0563674321503131, + "grad_norm": 0.9366872906684875, + "learning_rate": 4.713188023908044e-06, + "loss": 0.7524, + "step": 674 + }, + { + "epoch": 1.0579331941544885, + "grad_norm": 1.0348883867263794, + "learning_rate": 4.712207454970899e-06, + "loss": 0.6965, + "step": 675 + }, + { + "epoch": 1.0594989561586639, + "grad_norm": 0.8716043829917908, + "learning_rate": 4.711225315042513e-06, + "loss": 0.6264, + "step": 676 + }, + { + "epoch": 1.0610647181628392, + "grad_norm": 1.0421313047409058, + "learning_rate": 4.710241604820348e-06, + "loss": 0.7548, + "step": 677 + }, + { + "epoch": 1.0626304801670146, + "grad_norm": 1.0886270999908447, + "learning_rate": 4.709256325002983e-06, + "loss": 0.6352, + "step": 678 + }, + { + "epoch": 1.06419624217119, + "grad_norm": 0.9752539396286011, + "learning_rate": 4.708269476290111e-06, + "loss": 0.7764, + "step": 679 + }, + { + "epoch": 1.0657620041753653, + "grad_norm": 1.0658339262008667, + "learning_rate": 4.7072810593825385e-06, + "loss": 0.766, + "step": 680 + }, + { + "epoch": 1.0673277661795406, + "grad_norm": 1.0299617052078247, + "learning_rate": 4.706291074982187e-06, + "loss": 0.7558, + "step": 681 + }, + { + "epoch": 1.068893528183716, + "grad_norm": 0.9171326160430908, + "learning_rate": 4.705299523792089e-06, + "loss": 0.7084, + "step": 682 + }, + { + "epoch": 1.0704592901878915, + "grad_norm": 0.9924243688583374, + "learning_rate": 4.7043064065163925e-06, + "loss": 0.8393, + "step": 683 + }, + { + "epoch": 1.072025052192067, + "grad_norm": 1.0243877172470093, + "learning_rate": 4.703311723860356e-06, + "loss": 0.7755, + "step": 684 + }, + { + "epoch": 1.0735908141962422, + "grad_norm": 1.0019137859344482, + "learning_rate": 4.702315476530349e-06, + "loss": 0.7254, + "step": 685 + }, + { + "epoch": 1.0751565762004176, + "grad_norm": 0.8541996479034424, + "learning_rate": 4.701317665233853e-06, + "loss": 0.7389, + "step": 686 + }, + { + "epoch": 1.076722338204593, + "grad_norm": 1.1119922399520874, + "learning_rate": 4.700318290679462e-06, + "loss": 0.7803, + "step": 687 + }, + { + "epoch": 1.0782881002087683, + "grad_norm": 1.078452706336975, + "learning_rate": 4.699317353576876e-06, + "loss": 0.756, + "step": 688 + }, + { + "epoch": 1.0798538622129437, + "grad_norm": 1.0079457759857178, + "learning_rate": 4.698314854636908e-06, + "loss": 0.7365, + "step": 689 + }, + { + "epoch": 1.081419624217119, + "grad_norm": 0.931423544883728, + "learning_rate": 4.6973107945714795e-06, + "loss": 0.7881, + "step": 690 + }, + { + "epoch": 1.0829853862212944, + "grad_norm": 1.0135142803192139, + "learning_rate": 4.696305174093619e-06, + "loss": 0.7884, + "step": 691 + }, + { + "epoch": 1.0845511482254697, + "grad_norm": 1.1194610595703125, + "learning_rate": 4.695297993917465e-06, + "loss": 0.7969, + "step": 692 + }, + { + "epoch": 1.086116910229645, + "grad_norm": 0.9972780346870422, + "learning_rate": 4.6942892547582625e-06, + "loss": 0.5873, + "step": 693 + }, + { + "epoch": 1.0876826722338204, + "grad_norm": 1.0711641311645508, + "learning_rate": 4.693278957332364e-06, + "loss": 0.7273, + "step": 694 + }, + { + "epoch": 1.0892484342379958, + "grad_norm": 0.9708715081214905, + "learning_rate": 4.6922671023572305e-06, + "loss": 0.6786, + "step": 695 + }, + { + "epoch": 1.0908141962421711, + "grad_norm": 0.9773644804954529, + "learning_rate": 4.691253690551424e-06, + "loss": 0.7839, + "step": 696 + }, + { + "epoch": 1.0923799582463465, + "grad_norm": 1.039009690284729, + "learning_rate": 4.690238722634618e-06, + "loss": 0.6807, + "step": 697 + }, + { + "epoch": 1.0939457202505218, + "grad_norm": 1.140777349472046, + "learning_rate": 4.689222199327586e-06, + "loss": 0.7506, + "step": 698 + }, + { + "epoch": 1.0955114822546972, + "grad_norm": 0.9779882431030273, + "learning_rate": 4.688204121352211e-06, + "loss": 0.7732, + "step": 699 + }, + { + "epoch": 1.0970772442588728, + "grad_norm": 1.0412248373031616, + "learning_rate": 4.687184489431476e-06, + "loss": 0.7034, + "step": 700 + }, + { + "epoch": 1.0986430062630481, + "grad_norm": 0.9802400469779968, + "learning_rate": 4.686163304289469e-06, + "loss": 0.729, + "step": 701 + }, + { + "epoch": 1.1002087682672235, + "grad_norm": 1.0531584024429321, + "learning_rate": 4.6851405666513804e-06, + "loss": 0.7635, + "step": 702 + }, + { + "epoch": 1.1017745302713988, + "grad_norm": 0.983923077583313, + "learning_rate": 4.684116277243505e-06, + "loss": 0.7062, + "step": 703 + }, + { + "epoch": 1.1033402922755742, + "grad_norm": 0.9886301159858704, + "learning_rate": 4.683090436793238e-06, + "loss": 0.7299, + "step": 704 + }, + { + "epoch": 1.1049060542797495, + "grad_norm": 1.0247817039489746, + "learning_rate": 4.682063046029075e-06, + "loss": 0.729, + "step": 705 + }, + { + "epoch": 1.1064718162839249, + "grad_norm": 0.9860950708389282, + "learning_rate": 4.681034105680616e-06, + "loss": 0.7392, + "step": 706 + }, + { + "epoch": 1.1080375782881002, + "grad_norm": 0.9543380737304688, + "learning_rate": 4.680003616478556e-06, + "loss": 0.7061, + "step": 707 + }, + { + "epoch": 1.1096033402922756, + "grad_norm": 1.0253225564956665, + "learning_rate": 4.678971579154698e-06, + "loss": 0.6577, + "step": 708 + }, + { + "epoch": 1.111169102296451, + "grad_norm": 0.9013616442680359, + "learning_rate": 4.677937994441935e-06, + "loss": 0.7422, + "step": 709 + }, + { + "epoch": 1.1127348643006263, + "grad_norm": 0.9716471433639526, + "learning_rate": 4.676902863074268e-06, + "loss": 0.7353, + "step": 710 + }, + { + "epoch": 1.1143006263048016, + "grad_norm": 1.040898084640503, + "learning_rate": 4.675866185786787e-06, + "loss": 0.7396, + "step": 711 + }, + { + "epoch": 1.115866388308977, + "grad_norm": 0.9931808114051819, + "learning_rate": 4.674827963315688e-06, + "loss": 0.7354, + "step": 712 + }, + { + "epoch": 1.1174321503131524, + "grad_norm": 1.020512580871582, + "learning_rate": 4.6737881963982605e-06, + "loss": 0.7366, + "step": 713 + }, + { + "epoch": 1.1189979123173277, + "grad_norm": 1.0201045274734497, + "learning_rate": 4.67274688577289e-06, + "loss": 0.7907, + "step": 714 + }, + { + "epoch": 1.120563674321503, + "grad_norm": 1.0163429975509644, + "learning_rate": 4.671704032179061e-06, + "loss": 0.7452, + "step": 715 + }, + { + "epoch": 1.1221294363256784, + "grad_norm": 0.9667377471923828, + "learning_rate": 4.670659636357352e-06, + "loss": 0.7983, + "step": 716 + }, + { + "epoch": 1.1236951983298538, + "grad_norm": 1.0788301229476929, + "learning_rate": 4.669613699049436e-06, + "loss": 0.7861, + "step": 717 + }, + { + "epoch": 1.1252609603340291, + "grad_norm": 1.0512142181396484, + "learning_rate": 4.668566220998082e-06, + "loss": 0.7582, + "step": 718 + }, + { + "epoch": 1.1268267223382047, + "grad_norm": 0.9225794076919556, + "learning_rate": 4.667517202947154e-06, + "loss": 0.7801, + "step": 719 + }, + { + "epoch": 1.12839248434238, + "grad_norm": 1.1193747520446777, + "learning_rate": 4.666466645641607e-06, + "loss": 0.8073, + "step": 720 + }, + { + "epoch": 1.1299582463465554, + "grad_norm": 0.934438943862915, + "learning_rate": 4.6654145498274915e-06, + "loss": 0.774, + "step": 721 + }, + { + "epoch": 1.1315240083507307, + "grad_norm": 1.0614968538284302, + "learning_rate": 4.664360916251949e-06, + "loss": 0.7216, + "step": 722 + }, + { + "epoch": 1.133089770354906, + "grad_norm": 1.0807424783706665, + "learning_rate": 4.663305745663212e-06, + "loss": 0.6434, + "step": 723 + }, + { + "epoch": 1.1346555323590815, + "grad_norm": 0.9515354633331299, + "learning_rate": 4.66224903881061e-06, + "loss": 0.7336, + "step": 724 + }, + { + "epoch": 1.1362212943632568, + "grad_norm": 0.9852489233016968, + "learning_rate": 4.661190796444557e-06, + "loss": 0.8156, + "step": 725 + }, + { + "epoch": 1.1377870563674322, + "grad_norm": 1.0071501731872559, + "learning_rate": 4.66013101931656e-06, + "loss": 0.8836, + "step": 726 + }, + { + "epoch": 1.1393528183716075, + "grad_norm": 0.9622960686683655, + "learning_rate": 4.6590697081792176e-06, + "loss": 0.7631, + "step": 727 + }, + { + "epoch": 1.1409185803757829, + "grad_norm": 1.0588105916976929, + "learning_rate": 4.6580068637862144e-06, + "loss": 0.7874, + "step": 728 + }, + { + "epoch": 1.1424843423799582, + "grad_norm": 0.9920447468757629, + "learning_rate": 4.656942486892326e-06, + "loss": 0.7161, + "step": 729 + }, + { + "epoch": 1.1440501043841336, + "grad_norm": 1.0036561489105225, + "learning_rate": 4.655876578253416e-06, + "loss": 0.7116, + "step": 730 + }, + { + "epoch": 1.145615866388309, + "grad_norm": 1.0208396911621094, + "learning_rate": 4.654809138626436e-06, + "loss": 0.818, + "step": 731 + }, + { + "epoch": 1.1471816283924843, + "grad_norm": 1.1095491647720337, + "learning_rate": 4.653740168769424e-06, + "loss": 0.7319, + "step": 732 + }, + { + "epoch": 1.1487473903966596, + "grad_norm": 0.8151328563690186, + "learning_rate": 4.652669669441505e-06, + "loss": 0.774, + "step": 733 + }, + { + "epoch": 1.150313152400835, + "grad_norm": 1.0747915506362915, + "learning_rate": 4.651597641402891e-06, + "loss": 0.8275, + "step": 734 + }, + { + "epoch": 1.1518789144050103, + "grad_norm": 0.9265272617340088, + "learning_rate": 4.650524085414878e-06, + "loss": 0.6946, + "step": 735 + }, + { + "epoch": 1.153444676409186, + "grad_norm": 0.9531672596931458, + "learning_rate": 4.649449002239849e-06, + "loss": 0.7899, + "step": 736 + }, + { + "epoch": 1.1550104384133613, + "grad_norm": 1.0496262311935425, + "learning_rate": 4.648372392641269e-06, + "loss": 0.8237, + "step": 737 + }, + { + "epoch": 1.1565762004175366, + "grad_norm": 0.8967193961143494, + "learning_rate": 4.64729425738369e-06, + "loss": 0.736, + "step": 738 + }, + { + "epoch": 1.158141962421712, + "grad_norm": 1.061038851737976, + "learning_rate": 4.646214597232745e-06, + "loss": 0.7308, + "step": 739 + }, + { + "epoch": 1.1597077244258873, + "grad_norm": 0.9556939601898193, + "learning_rate": 4.64513341295515e-06, + "loss": 0.7526, + "step": 740 + }, + { + "epoch": 1.1612734864300627, + "grad_norm": 1.000093936920166, + "learning_rate": 4.6440507053187075e-06, + "loss": 0.7488, + "step": 741 + }, + { + "epoch": 1.162839248434238, + "grad_norm": 0.9552261233329773, + "learning_rate": 4.642966475092294e-06, + "loss": 0.8291, + "step": 742 + }, + { + "epoch": 1.1644050104384134, + "grad_norm": 1.0215847492218018, + "learning_rate": 4.641880723045875e-06, + "loss": 0.756, + "step": 743 + }, + { + "epoch": 1.1659707724425887, + "grad_norm": 1.0543124675750732, + "learning_rate": 4.640793449950493e-06, + "loss": 0.7668, + "step": 744 + }, + { + "epoch": 1.167536534446764, + "grad_norm": 0.9648609161376953, + "learning_rate": 4.639704656578269e-06, + "loss": 0.6531, + "step": 745 + }, + { + "epoch": 1.1691022964509394, + "grad_norm": 1.0117427110671997, + "learning_rate": 4.638614343702408e-06, + "loss": 0.7084, + "step": 746 + }, + { + "epoch": 1.1706680584551148, + "grad_norm": 0.9475041627883911, + "learning_rate": 4.6375225120971915e-06, + "loss": 0.8578, + "step": 747 + }, + { + "epoch": 1.1722338204592901, + "grad_norm": 1.0122299194335938, + "learning_rate": 4.6364291625379785e-06, + "loss": 0.7541, + "step": 748 + }, + { + "epoch": 1.1737995824634655, + "grad_norm": 0.9487535357475281, + "learning_rate": 4.635334295801208e-06, + "loss": 0.7203, + "step": 749 + }, + { + "epoch": 1.1753653444676408, + "grad_norm": 1.0628844499588013, + "learning_rate": 4.6342379126643965e-06, + "loss": 0.7532, + "step": 750 + }, + { + "epoch": 1.1769311064718162, + "grad_norm": 0.9647781848907471, + "learning_rate": 4.633140013906136e-06, + "loss": 0.5918, + "step": 751 + }, + { + "epoch": 1.1784968684759916, + "grad_norm": 0.9011932611465454, + "learning_rate": 4.632040600306095e-06, + "loss": 0.7395, + "step": 752 + }, + { + "epoch": 1.1800626304801671, + "grad_norm": 1.0524142980575562, + "learning_rate": 4.630939672645018e-06, + "loss": 0.7487, + "step": 753 + }, + { + "epoch": 1.1816283924843423, + "grad_norm": 1.165077567100525, + "learning_rate": 4.6298372317047265e-06, + "loss": 0.6817, + "step": 754 + }, + { + "epoch": 1.1831941544885178, + "grad_norm": 0.9301269054412842, + "learning_rate": 4.628733278268113e-06, + "loss": 0.8446, + "step": 755 + }, + { + "epoch": 1.1847599164926932, + "grad_norm": 0.9762471318244934, + "learning_rate": 4.627627813119147e-06, + "loss": 0.7527, + "step": 756 + }, + { + "epoch": 1.1863256784968685, + "grad_norm": 0.8634734153747559, + "learning_rate": 4.626520837042871e-06, + "loss": 0.7177, + "step": 757 + }, + { + "epoch": 1.187891440501044, + "grad_norm": 1.0592188835144043, + "learning_rate": 4.625412350825399e-06, + "loss": 0.7293, + "step": 758 + }, + { + "epoch": 1.1894572025052192, + "grad_norm": 1.1318559646606445, + "learning_rate": 4.62430235525392e-06, + "loss": 0.6592, + "step": 759 + }, + { + "epoch": 1.1910229645093946, + "grad_norm": 0.9689339399337769, + "learning_rate": 4.623190851116692e-06, + "loss": 0.7667, + "step": 760 + }, + { + "epoch": 1.19258872651357, + "grad_norm": 0.9965885281562805, + "learning_rate": 4.6220778392030464e-06, + "loss": 0.6982, + "step": 761 + }, + { + "epoch": 1.1941544885177453, + "grad_norm": 1.1428629159927368, + "learning_rate": 4.6209633203033846e-06, + "loss": 0.7342, + "step": 762 + }, + { + "epoch": 1.1957202505219207, + "grad_norm": 0.901360809803009, + "learning_rate": 4.6198472952091785e-06, + "loss": 0.8371, + "step": 763 + }, + { + "epoch": 1.197286012526096, + "grad_norm": 0.9882062077522278, + "learning_rate": 4.618729764712969e-06, + "loss": 0.6882, + "step": 764 + }, + { + "epoch": 1.1988517745302714, + "grad_norm": 1.0692517757415771, + "learning_rate": 4.617610729608366e-06, + "loss": 0.7327, + "step": 765 + }, + { + "epoch": 1.2004175365344467, + "grad_norm": 1.1056455373764038, + "learning_rate": 4.616490190690049e-06, + "loss": 0.781, + "step": 766 + }, + { + "epoch": 1.201983298538622, + "grad_norm": 0.9928885698318481, + "learning_rate": 4.615368148753765e-06, + "loss": 0.7463, + "step": 767 + }, + { + "epoch": 1.2035490605427974, + "grad_norm": 1.0006698369979858, + "learning_rate": 4.614244604596327e-06, + "loss": 0.6994, + "step": 768 + }, + { + "epoch": 1.2051148225469728, + "grad_norm": 0.907508909702301, + "learning_rate": 4.613119559015616e-06, + "loss": 0.7117, + "step": 769 + }, + { + "epoch": 1.2066805845511483, + "grad_norm": 1.1037627458572388, + "learning_rate": 4.61199301281058e-06, + "loss": 0.772, + "step": 770 + }, + { + "epoch": 1.2082463465553235, + "grad_norm": 0.9474150538444519, + "learning_rate": 4.610864966781231e-06, + "loss": 0.7601, + "step": 771 + }, + { + "epoch": 1.209812108559499, + "grad_norm": 1.003667950630188, + "learning_rate": 4.609735421728647e-06, + "loss": 0.7141, + "step": 772 + }, + { + "epoch": 1.2113778705636744, + "grad_norm": 0.9724787473678589, + "learning_rate": 4.608604378454971e-06, + "loss": 0.7841, + "step": 773 + }, + { + "epoch": 1.2129436325678498, + "grad_norm": 1.0460760593414307, + "learning_rate": 4.607471837763409e-06, + "loss": 0.7563, + "step": 774 + }, + { + "epoch": 1.2145093945720251, + "grad_norm": 0.9798458218574524, + "learning_rate": 4.606337800458231e-06, + "loss": 0.6812, + "step": 775 + }, + { + "epoch": 1.2160751565762005, + "grad_norm": 1.1637914180755615, + "learning_rate": 4.605202267344768e-06, + "loss": 0.7571, + "step": 776 + }, + { + "epoch": 1.2176409185803758, + "grad_norm": 0.9384608864784241, + "learning_rate": 4.6040652392294175e-06, + "loss": 0.7842, + "step": 777 + }, + { + "epoch": 1.2192066805845512, + "grad_norm": 1.0490702390670776, + "learning_rate": 4.602926716919633e-06, + "loss": 0.7157, + "step": 778 + }, + { + "epoch": 1.2207724425887265, + "grad_norm": 1.131402611732483, + "learning_rate": 4.601786701223934e-06, + "loss": 0.7724, + "step": 779 + }, + { + "epoch": 1.2223382045929019, + "grad_norm": 1.0159722566604614, + "learning_rate": 4.600645192951898e-06, + "loss": 0.7666, + "step": 780 + }, + { + "epoch": 1.2239039665970772, + "grad_norm": 1.039806604385376, + "learning_rate": 4.599502192914164e-06, + "loss": 0.7845, + "step": 781 + }, + { + "epoch": 1.2254697286012526, + "grad_norm": 1.0722614526748657, + "learning_rate": 4.598357701922428e-06, + "loss": 0.782, + "step": 782 + }, + { + "epoch": 1.227035490605428, + "grad_norm": 0.9443033933639526, + "learning_rate": 4.5972117207894465e-06, + "loss": 0.6573, + "step": 783 + }, + { + "epoch": 1.2286012526096033, + "grad_norm": 0.9738026857376099, + "learning_rate": 4.596064250329035e-06, + "loss": 0.7475, + "step": 784 + }, + { + "epoch": 1.2301670146137786, + "grad_norm": 0.9932008385658264, + "learning_rate": 4.594915291356063e-06, + "loss": 0.757, + "step": 785 + }, + { + "epoch": 1.231732776617954, + "grad_norm": 1.0612564086914062, + "learning_rate": 4.5937648446864625e-06, + "loss": 0.7824, + "step": 786 + }, + { + "epoch": 1.2332985386221293, + "grad_norm": 1.0500893592834473, + "learning_rate": 4.592612911137218e-06, + "loss": 0.7646, + "step": 787 + }, + { + "epoch": 1.2348643006263047, + "grad_norm": 0.9607855081558228, + "learning_rate": 4.591459491526371e-06, + "loss": 0.754, + "step": 788 + }, + { + "epoch": 1.2364300626304803, + "grad_norm": 1.0871951580047607, + "learning_rate": 4.590304586673018e-06, + "loss": 0.7226, + "step": 789 + }, + { + "epoch": 1.2379958246346556, + "grad_norm": 1.0030092000961304, + "learning_rate": 4.589148197397311e-06, + "loss": 0.7357, + "step": 790 + }, + { + "epoch": 1.239561586638831, + "grad_norm": 1.0297991037368774, + "learning_rate": 4.587990324520455e-06, + "loss": 0.7724, + "step": 791 + }, + { + "epoch": 1.2411273486430063, + "grad_norm": 1.0192303657531738, + "learning_rate": 4.586830968864711e-06, + "loss": 0.6921, + "step": 792 + }, + { + "epoch": 1.2426931106471817, + "grad_norm": 0.9617307186126709, + "learning_rate": 4.585670131253389e-06, + "loss": 0.7443, + "step": 793 + }, + { + "epoch": 1.244258872651357, + "grad_norm": 1.0355238914489746, + "learning_rate": 4.584507812510856e-06, + "loss": 0.7741, + "step": 794 + }, + { + "epoch": 1.2458246346555324, + "grad_norm": 0.9938539862632751, + "learning_rate": 4.583344013462526e-06, + "loss": 0.7372, + "step": 795 + }, + { + "epoch": 1.2473903966597077, + "grad_norm": 1.0573335886001587, + "learning_rate": 4.582178734934869e-06, + "loss": 0.7191, + "step": 796 + }, + { + "epoch": 1.248956158663883, + "grad_norm": 0.963642954826355, + "learning_rate": 4.581011977755402e-06, + "loss": 0.6283, + "step": 797 + }, + { + "epoch": 1.2505219206680585, + "grad_norm": 1.0119287967681885, + "learning_rate": 4.579843742752694e-06, + "loss": 0.7599, + "step": 798 + }, + { + "epoch": 1.2520876826722338, + "grad_norm": 1.0628082752227783, + "learning_rate": 4.578674030756364e-06, + "loss": 0.7741, + "step": 799 + }, + { + "epoch": 1.2536534446764092, + "grad_norm": 1.0510241985321045, + "learning_rate": 4.577502842597077e-06, + "loss": 0.7615, + "step": 800 + }, + { + "epoch": 1.2552192066805845, + "grad_norm": 1.1180559396743774, + "learning_rate": 4.5763301791065474e-06, + "loss": 0.7119, + "step": 801 + }, + { + "epoch": 1.2567849686847599, + "grad_norm": 1.0198049545288086, + "learning_rate": 4.575156041117541e-06, + "loss": 0.7487, + "step": 802 + }, + { + "epoch": 1.2583507306889352, + "grad_norm": 1.018902063369751, + "learning_rate": 4.5739804294638655e-06, + "loss": 0.7461, + "step": 803 + }, + { + "epoch": 1.2599164926931106, + "grad_norm": 0.9230141639709473, + "learning_rate": 4.572803344980378e-06, + "loss": 0.7295, + "step": 804 + }, + { + "epoch": 1.261482254697286, + "grad_norm": 1.0011619329452515, + "learning_rate": 4.571624788502981e-06, + "loss": 0.7325, + "step": 805 + }, + { + "epoch": 1.2630480167014615, + "grad_norm": 0.9569756388664246, + "learning_rate": 4.570444760868622e-06, + "loss": 0.7494, + "step": 806 + }, + { + "epoch": 1.2646137787056366, + "grad_norm": 0.9860491156578064, + "learning_rate": 4.569263262915293e-06, + "loss": 0.6536, + "step": 807 + }, + { + "epoch": 1.2661795407098122, + "grad_norm": 1.0986074209213257, + "learning_rate": 4.568080295482032e-06, + "loss": 0.7337, + "step": 808 + }, + { + "epoch": 1.2677453027139876, + "grad_norm": 1.0782687664031982, + "learning_rate": 4.566895859408916e-06, + "loss": 0.7193, + "step": 809 + }, + { + "epoch": 1.269311064718163, + "grad_norm": 1.1424410343170166, + "learning_rate": 4.565709955537071e-06, + "loss": 0.6153, + "step": 810 + }, + { + "epoch": 1.2708768267223383, + "grad_norm": 0.9178900718688965, + "learning_rate": 4.564522584708662e-06, + "loss": 0.7802, + "step": 811 + }, + { + "epoch": 1.2724425887265136, + "grad_norm": 1.0006638765335083, + "learning_rate": 4.563333747766896e-06, + "loss": 0.7882, + "step": 812 + }, + { + "epoch": 1.274008350730689, + "grad_norm": 1.0139516592025757, + "learning_rate": 4.56214344555602e-06, + "loss": 0.6562, + "step": 813 + }, + { + "epoch": 1.2755741127348643, + "grad_norm": 1.0360203981399536, + "learning_rate": 4.560951678921324e-06, + "loss": 0.7951, + "step": 814 + }, + { + "epoch": 1.2771398747390397, + "grad_norm": 1.042457103729248, + "learning_rate": 4.559758448709138e-06, + "loss": 0.7572, + "step": 815 + }, + { + "epoch": 1.278705636743215, + "grad_norm": 1.0056653022766113, + "learning_rate": 4.558563755766827e-06, + "loss": 0.7679, + "step": 816 + }, + { + "epoch": 1.2802713987473904, + "grad_norm": 0.9838577508926392, + "learning_rate": 4.5573676009428e-06, + "loss": 0.7245, + "step": 817 + }, + { + "epoch": 1.2818371607515657, + "grad_norm": 1.0474631786346436, + "learning_rate": 4.556169985086503e-06, + "loss": 0.7823, + "step": 818 + }, + { + "epoch": 1.283402922755741, + "grad_norm": 1.0177569389343262, + "learning_rate": 4.554970909048417e-06, + "loss": 0.7599, + "step": 819 + }, + { + "epoch": 1.2849686847599164, + "grad_norm": 1.0120493173599243, + "learning_rate": 4.553770373680062e-06, + "loss": 0.7842, + "step": 820 + }, + { + "epoch": 1.2865344467640918, + "grad_norm": 1.1045918464660645, + "learning_rate": 4.5525683798339946e-06, + "loss": 0.7075, + "step": 821 + }, + { + "epoch": 1.2881002087682671, + "grad_norm": 1.097714900970459, + "learning_rate": 4.551364928363805e-06, + "loss": 0.7571, + "step": 822 + }, + { + "epoch": 1.2896659707724427, + "grad_norm": 0.9125490188598633, + "learning_rate": 4.550160020124121e-06, + "loss": 0.7626, + "step": 823 + }, + { + "epoch": 1.2912317327766178, + "grad_norm": 1.1127262115478516, + "learning_rate": 4.548953655970604e-06, + "loss": 0.6618, + "step": 824 + }, + { + "epoch": 1.2927974947807934, + "grad_norm": 1.0372540950775146, + "learning_rate": 4.547745836759949e-06, + "loss": 0.7467, + "step": 825 + }, + { + "epoch": 1.2943632567849686, + "grad_norm": 1.0158213376998901, + "learning_rate": 4.546536563349885e-06, + "loss": 0.6421, + "step": 826 + }, + { + "epoch": 1.2959290187891441, + "grad_norm": 1.0057018995285034, + "learning_rate": 4.545325836599171e-06, + "loss": 0.7252, + "step": 827 + }, + { + "epoch": 1.2974947807933195, + "grad_norm": 1.0552566051483154, + "learning_rate": 4.544113657367604e-06, + "loss": 0.6645, + "step": 828 + }, + { + "epoch": 1.2990605427974948, + "grad_norm": 1.0798696279525757, + "learning_rate": 4.542900026516006e-06, + "loss": 0.7627, + "step": 829 + }, + { + "epoch": 1.3006263048016702, + "grad_norm": 1.0450047254562378, + "learning_rate": 4.541684944906233e-06, + "loss": 0.705, + "step": 830 + }, + { + "epoch": 1.3021920668058455, + "grad_norm": 0.9975817203521729, + "learning_rate": 4.540468413401171e-06, + "loss": 0.7639, + "step": 831 + }, + { + "epoch": 1.303757828810021, + "grad_norm": 0.9081487059593201, + "learning_rate": 4.539250432864736e-06, + "loss": 0.8129, + "step": 832 + }, + { + "epoch": 1.3053235908141962, + "grad_norm": 1.228684902191162, + "learning_rate": 4.538031004161872e-06, + "loss": 0.6984, + "step": 833 + }, + { + "epoch": 1.3068893528183716, + "grad_norm": 1.0577794313430786, + "learning_rate": 4.536810128158552e-06, + "loss": 0.7436, + "step": 834 + }, + { + "epoch": 1.308455114822547, + "grad_norm": 1.1270350217819214, + "learning_rate": 4.535587805721777e-06, + "loss": 0.7589, + "step": 835 + }, + { + "epoch": 1.3100208768267223, + "grad_norm": 1.1473027467727661, + "learning_rate": 4.5343640377195766e-06, + "loss": 0.7988, + "step": 836 + }, + { + "epoch": 1.3115866388308977, + "grad_norm": 0.9700750112533569, + "learning_rate": 4.533138825021001e-06, + "loss": 0.7388, + "step": 837 + }, + { + "epoch": 1.313152400835073, + "grad_norm": 0.9596260786056519, + "learning_rate": 4.531912168496135e-06, + "loss": 0.8177, + "step": 838 + }, + { + "epoch": 1.3147181628392484, + "grad_norm": 0.9888957738876343, + "learning_rate": 4.530684069016081e-06, + "loss": 0.809, + "step": 839 + }, + { + "epoch": 1.316283924843424, + "grad_norm": 1.219224214553833, + "learning_rate": 4.529454527452972e-06, + "loss": 0.7761, + "step": 840 + }, + { + "epoch": 1.317849686847599, + "grad_norm": 0.9730296730995178, + "learning_rate": 4.5282235446799624e-06, + "loss": 0.8044, + "step": 841 + }, + { + "epoch": 1.3194154488517746, + "grad_norm": 1.1046907901763916, + "learning_rate": 4.526991121571228e-06, + "loss": 0.7452, + "step": 842 + }, + { + "epoch": 1.3209812108559498, + "grad_norm": 0.9707220196723938, + "learning_rate": 4.525757259001972e-06, + "loss": 0.7112, + "step": 843 + }, + { + "epoch": 1.3225469728601253, + "grad_norm": 1.0085593461990356, + "learning_rate": 4.524521957848416e-06, + "loss": 0.6266, + "step": 844 + }, + { + "epoch": 1.3241127348643007, + "grad_norm": 1.131439447402954, + "learning_rate": 4.5232852189878055e-06, + "loss": 0.7393, + "step": 845 + }, + { + "epoch": 1.325678496868476, + "grad_norm": 1.1238782405853271, + "learning_rate": 4.522047043298406e-06, + "loss": 0.7541, + "step": 846 + }, + { + "epoch": 1.3272442588726514, + "grad_norm": 0.9149152636528015, + "learning_rate": 4.520807431659504e-06, + "loss": 0.8162, + "step": 847 + }, + { + "epoch": 1.3288100208768268, + "grad_norm": 1.0428732633590698, + "learning_rate": 4.519566384951403e-06, + "loss": 0.7511, + "step": 848 + }, + { + "epoch": 1.3303757828810021, + "grad_norm": 1.0263090133666992, + "learning_rate": 4.518323904055432e-06, + "loss": 0.7482, + "step": 849 + }, + { + "epoch": 1.3319415448851775, + "grad_norm": 1.113810420036316, + "learning_rate": 4.517079989853931e-06, + "loss": 0.797, + "step": 850 + }, + { + "epoch": 1.3335073068893528, + "grad_norm": 1.0090736150741577, + "learning_rate": 4.515834643230263e-06, + "loss": 0.688, + "step": 851 + }, + { + "epoch": 1.3350730688935282, + "grad_norm": 1.012824296951294, + "learning_rate": 4.514587865068806e-06, + "loss": 0.6683, + "step": 852 + }, + { + "epoch": 1.3366388308977035, + "grad_norm": 0.9442676305770874, + "learning_rate": 4.513339656254954e-06, + "loss": 0.7812, + "step": 853 + }, + { + "epoch": 1.3382045929018789, + "grad_norm": 1.0870879888534546, + "learning_rate": 4.512090017675119e-06, + "loss": 0.6943, + "step": 854 + }, + { + "epoch": 1.3397703549060542, + "grad_norm": 0.9352210164070129, + "learning_rate": 4.510838950216727e-06, + "loss": 0.7571, + "step": 855 + }, + { + "epoch": 1.3413361169102296, + "grad_norm": 1.173621654510498, + "learning_rate": 4.509586454768221e-06, + "loss": 0.6985, + "step": 856 + }, + { + "epoch": 1.342901878914405, + "grad_norm": 0.9790180921554565, + "learning_rate": 4.508332532219053e-06, + "loss": 0.7132, + "step": 857 + }, + { + "epoch": 1.3444676409185803, + "grad_norm": 1.0999163389205933, + "learning_rate": 4.507077183459694e-06, + "loss": 0.7858, + "step": 858 + }, + { + "epoch": 1.3460334029227559, + "grad_norm": 1.018182635307312, + "learning_rate": 4.505820409381625e-06, + "loss": 0.712, + "step": 859 + }, + { + "epoch": 1.347599164926931, + "grad_norm": 1.1015701293945312, + "learning_rate": 4.504562210877338e-06, + "loss": 0.8004, + "step": 860 + }, + { + "epoch": 1.3491649269311066, + "grad_norm": 1.0872273445129395, + "learning_rate": 4.50330258884034e-06, + "loss": 0.7478, + "step": 861 + }, + { + "epoch": 1.350730688935282, + "grad_norm": 1.1098262071609497, + "learning_rate": 4.502041544165148e-06, + "loss": 0.7894, + "step": 862 + }, + { + "epoch": 1.3522964509394573, + "grad_norm": 1.0535573959350586, + "learning_rate": 4.500779077747285e-06, + "loss": 0.7185, + "step": 863 + }, + { + "epoch": 1.3538622129436326, + "grad_norm": 1.0424474477767944, + "learning_rate": 4.499515190483291e-06, + "loss": 0.748, + "step": 864 + }, + { + "epoch": 1.355427974947808, + "grad_norm": 1.005065679550171, + "learning_rate": 4.498249883270708e-06, + "loss": 0.7948, + "step": 865 + }, + { + "epoch": 1.3569937369519833, + "grad_norm": 1.0052721500396729, + "learning_rate": 4.49698315700809e-06, + "loss": 0.7619, + "step": 866 + }, + { + "epoch": 1.3585594989561587, + "grad_norm": 1.0953692197799683, + "learning_rate": 4.495715012594999e-06, + "loss": 0.7363, + "step": 867 + }, + { + "epoch": 1.360125260960334, + "grad_norm": 1.1238093376159668, + "learning_rate": 4.494445450932003e-06, + "loss": 0.6576, + "step": 868 + }, + { + "epoch": 1.3616910229645094, + "grad_norm": 0.8546008467674255, + "learning_rate": 4.493174472920676e-06, + "loss": 0.7513, + "step": 869 + }, + { + "epoch": 1.3632567849686847, + "grad_norm": 1.0863126516342163, + "learning_rate": 4.4919020794635995e-06, + "loss": 0.6886, + "step": 870 + }, + { + "epoch": 1.36482254697286, + "grad_norm": 0.9943820834159851, + "learning_rate": 4.490628271464359e-06, + "loss": 0.7253, + "step": 871 + }, + { + "epoch": 1.3663883089770354, + "grad_norm": 0.9377979636192322, + "learning_rate": 4.4893530498275436e-06, + "loss": 0.7026, + "step": 872 + }, + { + "epoch": 1.3679540709812108, + "grad_norm": 1.100799798965454, + "learning_rate": 4.488076415458749e-06, + "loss": 0.719, + "step": 873 + }, + { + "epoch": 1.3695198329853862, + "grad_norm": 1.015074610710144, + "learning_rate": 4.486798369264572e-06, + "loss": 0.7513, + "step": 874 + }, + { + "epoch": 1.3710855949895615, + "grad_norm": 1.047530174255371, + "learning_rate": 4.485518912152611e-06, + "loss": 0.7757, + "step": 875 + }, + { + "epoch": 1.372651356993737, + "grad_norm": 1.0264610052108765, + "learning_rate": 4.484238045031471e-06, + "loss": 0.7712, + "step": 876 + }, + { + "epoch": 1.3742171189979122, + "grad_norm": 0.9848430156707764, + "learning_rate": 4.482955768810752e-06, + "loss": 0.7569, + "step": 877 + }, + { + "epoch": 1.3757828810020878, + "grad_norm": 1.070816159248352, + "learning_rate": 4.48167208440106e-06, + "loss": 0.7857, + "step": 878 + }, + { + "epoch": 1.377348643006263, + "grad_norm": 1.0134745836257935, + "learning_rate": 4.480386992713998e-06, + "loss": 0.7372, + "step": 879 + }, + { + "epoch": 1.3789144050104385, + "grad_norm": 1.0736140012741089, + "learning_rate": 4.47910049466217e-06, + "loss": 0.7794, + "step": 880 + }, + { + "epoch": 1.3804801670146138, + "grad_norm": 1.0163054466247559, + "learning_rate": 4.477812591159176e-06, + "loss": 0.7341, + "step": 881 + }, + { + "epoch": 1.3820459290187892, + "grad_norm": 1.074578881263733, + "learning_rate": 4.476523283119619e-06, + "loss": 0.6924, + "step": 882 + }, + { + "epoch": 1.3836116910229646, + "grad_norm": 1.183667540550232, + "learning_rate": 4.475232571459095e-06, + "loss": 0.7866, + "step": 883 + }, + { + "epoch": 1.38517745302714, + "grad_norm": 1.030827522277832, + "learning_rate": 4.473940457094199e-06, + "loss": 0.7845, + "step": 884 + }, + { + "epoch": 1.3867432150313153, + "grad_norm": 1.0043658018112183, + "learning_rate": 4.47264694094252e-06, + "loss": 0.7399, + "step": 885 + }, + { + "epoch": 1.3883089770354906, + "grad_norm": 1.1135472059249878, + "learning_rate": 4.471352023922645e-06, + "loss": 0.771, + "step": 886 + }, + { + "epoch": 1.389874739039666, + "grad_norm": 1.0338670015335083, + "learning_rate": 4.470055706954154e-06, + "loss": 0.7282, + "step": 887 + }, + { + "epoch": 1.3914405010438413, + "grad_norm": 0.918855607509613, + "learning_rate": 4.468757990957623e-06, + "loss": 0.728, + "step": 888 + }, + { + "epoch": 1.3930062630480167, + "grad_norm": 1.06833016872406, + "learning_rate": 4.467458876854619e-06, + "loss": 0.7012, + "step": 889 + }, + { + "epoch": 1.394572025052192, + "grad_norm": 1.0103005170822144, + "learning_rate": 4.4661583655677045e-06, + "loss": 0.7074, + "step": 890 + }, + { + "epoch": 1.3961377870563674, + "grad_norm": 1.0418543815612793, + "learning_rate": 4.464856458020433e-06, + "loss": 0.6858, + "step": 891 + }, + { + "epoch": 1.3977035490605427, + "grad_norm": 0.9720668196678162, + "learning_rate": 4.463553155137348e-06, + "loss": 0.7024, + "step": 892 + }, + { + "epoch": 1.399269311064718, + "grad_norm": 1.1870561838150024, + "learning_rate": 4.462248457843986e-06, + "loss": 0.8069, + "step": 893 + }, + { + "epoch": 1.4008350730688934, + "grad_norm": 0.9888094663619995, + "learning_rate": 4.460942367066874e-06, + "loss": 0.7281, + "step": 894 + }, + { + "epoch": 1.402400835073069, + "grad_norm": 0.9120670557022095, + "learning_rate": 4.459634883733527e-06, + "loss": 0.7992, + "step": 895 + }, + { + "epoch": 1.4039665970772441, + "grad_norm": 1.0239481925964355, + "learning_rate": 4.45832600877245e-06, + "loss": 0.7312, + "step": 896 + }, + { + "epoch": 1.4055323590814197, + "grad_norm": 0.9379130005836487, + "learning_rate": 4.4570157431131355e-06, + "loss": 0.7019, + "step": 897 + }, + { + "epoch": 1.407098121085595, + "grad_norm": 1.0789109468460083, + "learning_rate": 4.455704087686063e-06, + "loss": 0.8149, + "step": 898 + }, + { + "epoch": 1.4086638830897704, + "grad_norm": 1.0521917343139648, + "learning_rate": 4.454391043422703e-06, + "loss": 0.7873, + "step": 899 + }, + { + "epoch": 1.4102296450939458, + "grad_norm": 0.9582643508911133, + "learning_rate": 4.453076611255507e-06, + "loss": 0.7815, + "step": 900 + }, + { + "epoch": 1.4117954070981211, + "grad_norm": 1.1096739768981934, + "learning_rate": 4.451760792117914e-06, + "loss": 0.7636, + "step": 901 + }, + { + "epoch": 1.4133611691022965, + "grad_norm": 1.043574333190918, + "learning_rate": 4.450443586944349e-06, + "loss": 0.7411, + "step": 902 + }, + { + "epoch": 1.4149269311064718, + "grad_norm": 1.0280033349990845, + "learning_rate": 4.449124996670221e-06, + "loss": 0.6525, + "step": 903 + }, + { + "epoch": 1.4164926931106472, + "grad_norm": 1.0329992771148682, + "learning_rate": 4.44780502223192e-06, + "loss": 0.7133, + "step": 904 + }, + { + "epoch": 1.4180584551148225, + "grad_norm": 1.016168475151062, + "learning_rate": 4.446483664566825e-06, + "loss": 0.7098, + "step": 905 + }, + { + "epoch": 1.4196242171189979, + "grad_norm": 1.2022786140441895, + "learning_rate": 4.4451609246132895e-06, + "loss": 0.7303, + "step": 906 + }, + { + "epoch": 1.4211899791231732, + "grad_norm": 1.0311683416366577, + "learning_rate": 4.443836803310655e-06, + "loss": 0.7475, + "step": 907 + }, + { + "epoch": 1.4227557411273486, + "grad_norm": 1.0073823928833008, + "learning_rate": 4.442511301599241e-06, + "loss": 0.7513, + "step": 908 + }, + { + "epoch": 1.424321503131524, + "grad_norm": 1.0702322721481323, + "learning_rate": 4.441184420420347e-06, + "loss": 0.6918, + "step": 909 + }, + { + "epoch": 1.4258872651356993, + "grad_norm": 1.1416171789169312, + "learning_rate": 4.439856160716254e-06, + "loss": 0.6324, + "step": 910 + }, + { + "epoch": 1.4274530271398747, + "grad_norm": 1.0560253858566284, + "learning_rate": 4.438526523430221e-06, + "loss": 0.749, + "step": 911 + }, + { + "epoch": 1.4290187891440502, + "grad_norm": 1.021008849143982, + "learning_rate": 4.4371955095064835e-06, + "loss": 0.7666, + "step": 912 + }, + { + "epoch": 1.4305845511482254, + "grad_norm": 1.151904821395874, + "learning_rate": 4.435863119890258e-06, + "loss": 0.6904, + "step": 913 + }, + { + "epoch": 1.432150313152401, + "grad_norm": 1.1216702461242676, + "learning_rate": 4.434529355527736e-06, + "loss": 0.7068, + "step": 914 + }, + { + "epoch": 1.433716075156576, + "grad_norm": 1.1130588054656982, + "learning_rate": 4.433194217366085e-06, + "loss": 0.6524, + "step": 915 + }, + { + "epoch": 1.4352818371607516, + "grad_norm": 1.07612144947052, + "learning_rate": 4.431857706353449e-06, + "loss": 0.8078, + "step": 916 + }, + { + "epoch": 1.436847599164927, + "grad_norm": 1.0220447778701782, + "learning_rate": 4.430519823438946e-06, + "loss": 0.6736, + "step": 917 + }, + { + "epoch": 1.4384133611691023, + "grad_norm": 1.0657141208648682, + "learning_rate": 4.429180569572669e-06, + "loss": 0.7457, + "step": 918 + }, + { + "epoch": 1.4399791231732777, + "grad_norm": 1.0646854639053345, + "learning_rate": 4.427839945705684e-06, + "loss": 0.7925, + "step": 919 + }, + { + "epoch": 1.441544885177453, + "grad_norm": 1.0619502067565918, + "learning_rate": 4.426497952790031e-06, + "loss": 0.6968, + "step": 920 + }, + { + "epoch": 1.4431106471816284, + "grad_norm": 0.9684338569641113, + "learning_rate": 4.425154591778722e-06, + "loss": 0.7411, + "step": 921 + }, + { + "epoch": 1.4446764091858038, + "grad_norm": 1.1225167512893677, + "learning_rate": 4.423809863625739e-06, + "loss": 0.7623, + "step": 922 + }, + { + "epoch": 1.446242171189979, + "grad_norm": 0.9310746788978577, + "learning_rate": 4.422463769286036e-06, + "loss": 0.8219, + "step": 923 + }, + { + "epoch": 1.4478079331941545, + "grad_norm": 1.1107957363128662, + "learning_rate": 4.4211163097155375e-06, + "loss": 0.7755, + "step": 924 + }, + { + "epoch": 1.4493736951983298, + "grad_norm": 0.9827596545219421, + "learning_rate": 4.419767485871136e-06, + "loss": 0.7102, + "step": 925 + }, + { + "epoch": 1.4509394572025052, + "grad_norm": 1.1075087785720825, + "learning_rate": 4.418417298710695e-06, + "loss": 0.6399, + "step": 926 + }, + { + "epoch": 1.4525052192066805, + "grad_norm": 0.9575802087783813, + "learning_rate": 4.417065749193045e-06, + "loss": 0.721, + "step": 927 + }, + { + "epoch": 1.4540709812108559, + "grad_norm": 1.0016006231307983, + "learning_rate": 4.415712838277985e-06, + "loss": 0.762, + "step": 928 + }, + { + "epoch": 1.4556367432150314, + "grad_norm": 1.1907106637954712, + "learning_rate": 4.414358566926279e-06, + "loss": 0.7904, + "step": 929 + }, + { + "epoch": 1.4572025052192066, + "grad_norm": 1.0011543035507202, + "learning_rate": 4.413002936099656e-06, + "loss": 0.7179, + "step": 930 + }, + { + "epoch": 1.4587682672233822, + "grad_norm": 0.980684757232666, + "learning_rate": 4.411645946760817e-06, + "loss": 0.6845, + "step": 931 + }, + { + "epoch": 1.4603340292275573, + "grad_norm": 0.9458135366439819, + "learning_rate": 4.4102875998734176e-06, + "loss": 0.6911, + "step": 932 + }, + { + "epoch": 1.4618997912317329, + "grad_norm": 1.020966649055481, + "learning_rate": 4.408927896402087e-06, + "loss": 0.7284, + "step": 933 + }, + { + "epoch": 1.4634655532359082, + "grad_norm": 1.041833519935608, + "learning_rate": 4.407566837312413e-06, + "loss": 0.7451, + "step": 934 + }, + { + "epoch": 1.4650313152400836, + "grad_norm": 0.9076465964317322, + "learning_rate": 4.406204423570946e-06, + "loss": 0.7553, + "step": 935 + }, + { + "epoch": 1.466597077244259, + "grad_norm": 1.1963952779769897, + "learning_rate": 4.404840656145199e-06, + "loss": 0.7437, + "step": 936 + }, + { + "epoch": 1.4681628392484343, + "grad_norm": 1.0181536674499512, + "learning_rate": 4.403475536003648e-06, + "loss": 0.7283, + "step": 937 + }, + { + "epoch": 1.4697286012526096, + "grad_norm": 0.9563003778457642, + "learning_rate": 4.402109064115727e-06, + "loss": 0.7375, + "step": 938 + }, + { + "epoch": 1.471294363256785, + "grad_norm": 1.05129873752594, + "learning_rate": 4.400741241451832e-06, + "loss": 0.6805, + "step": 939 + }, + { + "epoch": 1.4728601252609603, + "grad_norm": 1.0912166833877563, + "learning_rate": 4.399372068983317e-06, + "loss": 0.7076, + "step": 940 + }, + { + "epoch": 1.4744258872651357, + "grad_norm": 1.108500599861145, + "learning_rate": 4.398001547682494e-06, + "loss": 0.6836, + "step": 941 + }, + { + "epoch": 1.475991649269311, + "grad_norm": 1.0741651058197021, + "learning_rate": 4.396629678522636e-06, + "loss": 0.7405, + "step": 942 + }, + { + "epoch": 1.4775574112734864, + "grad_norm": 1.0603930950164795, + "learning_rate": 4.39525646247797e-06, + "loss": 0.7602, + "step": 943 + }, + { + "epoch": 1.4791231732776617, + "grad_norm": 0.9814990162849426, + "learning_rate": 4.39388190052368e-06, + "loss": 0.7324, + "step": 944 + }, + { + "epoch": 1.480688935281837, + "grad_norm": 0.9643695950508118, + "learning_rate": 4.392505993635906e-06, + "loss": 0.7034, + "step": 945 + }, + { + "epoch": 1.4822546972860124, + "grad_norm": 1.0490994453430176, + "learning_rate": 4.391128742791743e-06, + "loss": 0.7358, + "step": 946 + }, + { + "epoch": 1.4838204592901878, + "grad_norm": 0.9223467111587524, + "learning_rate": 4.389750148969244e-06, + "loss": 0.698, + "step": 947 + }, + { + "epoch": 1.4853862212943634, + "grad_norm": 1.1107758283615112, + "learning_rate": 4.388370213147409e-06, + "loss": 0.7264, + "step": 948 + }, + { + "epoch": 1.4869519832985385, + "grad_norm": 1.05939781665802, + "learning_rate": 4.386988936306196e-06, + "loss": 0.7209, + "step": 949 + }, + { + "epoch": 1.488517745302714, + "grad_norm": 1.024686574935913, + "learning_rate": 4.385606319426512e-06, + "loss": 0.7798, + "step": 950 + }, + { + "epoch": 1.4900835073068894, + "grad_norm": 0.9920977354049683, + "learning_rate": 4.3842223634902205e-06, + "loss": 0.8045, + "step": 951 + }, + { + "epoch": 1.4916492693110648, + "grad_norm": 0.950927197933197, + "learning_rate": 4.382837069480131e-06, + "loss": 0.6777, + "step": 952 + }, + { + "epoch": 1.4932150313152401, + "grad_norm": 1.0072081089019775, + "learning_rate": 4.3814504383800056e-06, + "loss": 0.7458, + "step": 953 + }, + { + "epoch": 1.4947807933194155, + "grad_norm": 0.9982030987739563, + "learning_rate": 4.3800624711745546e-06, + "loss": 0.8215, + "step": 954 + }, + { + "epoch": 1.4963465553235908, + "grad_norm": 0.9682126641273499, + "learning_rate": 4.3786731688494396e-06, + "loss": 0.7157, + "step": 955 + }, + { + "epoch": 1.4979123173277662, + "grad_norm": 1.1599830389022827, + "learning_rate": 4.377282532391267e-06, + "loss": 0.8091, + "step": 956 + }, + { + "epoch": 1.4994780793319415, + "grad_norm": 1.0624948740005493, + "learning_rate": 4.3758905627875934e-06, + "loss": 0.7428, + "step": 957 + }, + { + "epoch": 1.501043841336117, + "grad_norm": 1.14694082736969, + "learning_rate": 4.374497261026921e-06, + "loss": 0.7752, + "step": 958 + }, + { + "epoch": 1.5026096033402923, + "grad_norm": 1.0355825424194336, + "learning_rate": 4.373102628098697e-06, + "loss": 0.7348, + "step": 959 + }, + { + "epoch": 1.5041753653444676, + "grad_norm": 1.036026954650879, + "learning_rate": 4.371706664993316e-06, + "loss": 0.7536, + "step": 960 + }, + { + "epoch": 1.505741127348643, + "grad_norm": 1.2319790124893188, + "learning_rate": 4.370309372702116e-06, + "loss": 0.6985, + "step": 961 + }, + { + "epoch": 1.5073068893528183, + "grad_norm": 0.908687949180603, + "learning_rate": 4.36891075221738e-06, + "loss": 0.7173, + "step": 962 + }, + { + "epoch": 1.5088726513569939, + "grad_norm": 1.0482099056243896, + "learning_rate": 4.367510804532331e-06, + "loss": 0.7514, + "step": 963 + }, + { + "epoch": 1.510438413361169, + "grad_norm": 1.0434144735336304, + "learning_rate": 4.36610953064114e-06, + "loss": 0.754, + "step": 964 + }, + { + "epoch": 1.5120041753653446, + "grad_norm": 1.0122450590133667, + "learning_rate": 4.364706931538914e-06, + "loss": 0.7767, + "step": 965 + }, + { + "epoch": 1.5135699373695197, + "grad_norm": 1.0471900701522827, + "learning_rate": 4.363303008221705e-06, + "loss": 0.7337, + "step": 966 + }, + { + "epoch": 1.5151356993736953, + "grad_norm": 1.0691189765930176, + "learning_rate": 4.3618977616865045e-06, + "loss": 0.8047, + "step": 967 + }, + { + "epoch": 1.5167014613778704, + "grad_norm": 1.015114188194275, + "learning_rate": 4.360491192931242e-06, + "loss": 0.726, + "step": 968 + }, + { + "epoch": 1.518267223382046, + "grad_norm": 0.9770560264587402, + "learning_rate": 4.359083302954788e-06, + "loss": 0.7692, + "step": 969 + }, + { + "epoch": 1.5198329853862211, + "grad_norm": 1.079991340637207, + "learning_rate": 4.357674092756951e-06, + "loss": 0.8424, + "step": 970 + }, + { + "epoch": 1.5213987473903967, + "grad_norm": 1.068708896636963, + "learning_rate": 4.356263563338476e-06, + "loss": 0.758, + "step": 971 + }, + { + "epoch": 1.522964509394572, + "grad_norm": 1.1005477905273438, + "learning_rate": 4.354851715701046e-06, + "loss": 0.7209, + "step": 972 + }, + { + "epoch": 1.5245302713987474, + "grad_norm": 0.9870902299880981, + "learning_rate": 4.3534385508472786e-06, + "loss": 0.7272, + "step": 973 + }, + { + "epoch": 1.5260960334029228, + "grad_norm": 0.9616238474845886, + "learning_rate": 4.352024069780729e-06, + "loss": 0.8124, + "step": 974 + }, + { + "epoch": 1.5276617954070981, + "grad_norm": 1.0341622829437256, + "learning_rate": 4.3506082735058855e-06, + "loss": 0.7434, + "step": 975 + }, + { + "epoch": 1.5292275574112735, + "grad_norm": 1.0157302618026733, + "learning_rate": 4.349191163028171e-06, + "loss": 0.7534, + "step": 976 + }, + { + "epoch": 1.5307933194154488, + "grad_norm": 1.041379690170288, + "learning_rate": 4.347772739353939e-06, + "loss": 0.6942, + "step": 977 + }, + { + "epoch": 1.5323590814196242, + "grad_norm": 1.016205072402954, + "learning_rate": 4.346353003490481e-06, + "loss": 0.6707, + "step": 978 + }, + { + "epoch": 1.5339248434237995, + "grad_norm": 1.0034714937210083, + "learning_rate": 4.344931956446017e-06, + "loss": 0.6358, + "step": 979 + }, + { + "epoch": 1.535490605427975, + "grad_norm": 0.9671767354011536, + "learning_rate": 4.343509599229697e-06, + "loss": 0.6333, + "step": 980 + }, + { + "epoch": 1.5370563674321502, + "grad_norm": 1.1384506225585938, + "learning_rate": 4.342085932851605e-06, + "loss": 0.803, + "step": 981 + }, + { + "epoch": 1.5386221294363258, + "grad_norm": 1.023031234741211, + "learning_rate": 4.34066095832275e-06, + "loss": 0.7801, + "step": 982 + }, + { + "epoch": 1.540187891440501, + "grad_norm": 1.0675241947174072, + "learning_rate": 4.339234676655075e-06, + "loss": 0.7335, + "step": 983 + }, + { + "epoch": 1.5417536534446765, + "grad_norm": 1.0397543907165527, + "learning_rate": 4.337807088861447e-06, + "loss": 0.7722, + "step": 984 + }, + { + "epoch": 1.5433194154488517, + "grad_norm": 1.0892809629440308, + "learning_rate": 4.3363781959556635e-06, + "loss": 0.7603, + "step": 985 + }, + { + "epoch": 1.5448851774530272, + "grad_norm": 0.9335037469863892, + "learning_rate": 4.334947998952448e-06, + "loss": 0.7678, + "step": 986 + }, + { + "epoch": 1.5464509394572024, + "grad_norm": 1.0991342067718506, + "learning_rate": 4.333516498867449e-06, + "loss": 0.7461, + "step": 987 + }, + { + "epoch": 1.548016701461378, + "grad_norm": 1.112265944480896, + "learning_rate": 4.332083696717242e-06, + "loss": 0.7593, + "step": 988 + }, + { + "epoch": 1.5495824634655533, + "grad_norm": 0.9525822997093201, + "learning_rate": 4.330649593519325e-06, + "loss": 0.7629, + "step": 989 + }, + { + "epoch": 1.5511482254697286, + "grad_norm": 1.0624464750289917, + "learning_rate": 4.329214190292122e-06, + "loss": 0.7261, + "step": 990 + }, + { + "epoch": 1.552713987473904, + "grad_norm": 1.1969516277313232, + "learning_rate": 4.327777488054979e-06, + "loss": 0.7483, + "step": 991 + }, + { + "epoch": 1.5542797494780793, + "grad_norm": 1.0239863395690918, + "learning_rate": 4.326339487828167e-06, + "loss": 0.7265, + "step": 992 + }, + { + "epoch": 1.5558455114822547, + "grad_norm": 1.0881083011627197, + "learning_rate": 4.324900190632874e-06, + "loss": 0.724, + "step": 993 + }, + { + "epoch": 1.55741127348643, + "grad_norm": 1.098827838897705, + "learning_rate": 4.323459597491214e-06, + "loss": 0.7222, + "step": 994 + }, + { + "epoch": 1.5589770354906054, + "grad_norm": 1.105057716369629, + "learning_rate": 4.322017709426217e-06, + "loss": 0.7827, + "step": 995 + }, + { + "epoch": 1.5605427974947808, + "grad_norm": 0.9830679297447205, + "learning_rate": 4.3205745274618365e-06, + "loss": 0.7316, + "step": 996 + }, + { + "epoch": 1.562108559498956, + "grad_norm": 0.9951156973838806, + "learning_rate": 4.319130052622942e-06, + "loss": 0.7181, + "step": 997 + }, + { + "epoch": 1.5636743215031315, + "grad_norm": 1.0374284982681274, + "learning_rate": 4.317684285935323e-06, + "loss": 0.7355, + "step": 998 + }, + { + "epoch": 1.565240083507307, + "grad_norm": 1.0139906406402588, + "learning_rate": 4.316237228425683e-06, + "loss": 0.7249, + "step": 999 + }, + { + "epoch": 1.5668058455114822, + "grad_norm": 0.9555441737174988, + "learning_rate": 4.3147888811216496e-06, + "loss": 0.7387, + "step": 1000 + }, + { + "epoch": 1.5683716075156577, + "grad_norm": 1.0600265264511108, + "learning_rate": 4.313339245051758e-06, + "loss": 0.7445, + "step": 1001 + }, + { + "epoch": 1.5699373695198329, + "grad_norm": 1.0070478916168213, + "learning_rate": 4.311888321245461e-06, + "loss": 0.7605, + "step": 1002 + }, + { + "epoch": 1.5715031315240084, + "grad_norm": 1.019710898399353, + "learning_rate": 4.31043611073313e-06, + "loss": 0.7262, + "step": 1003 + }, + { + "epoch": 1.5730688935281836, + "grad_norm": 1.0476088523864746, + "learning_rate": 4.308982614546045e-06, + "loss": 0.6766, + "step": 1004 + }, + { + "epoch": 1.5746346555323592, + "grad_norm": 1.0670830011367798, + "learning_rate": 4.307527833716403e-06, + "loss": 0.7465, + "step": 1005 + }, + { + "epoch": 1.5762004175365343, + "grad_norm": 1.1807290315628052, + "learning_rate": 4.30607176927731e-06, + "loss": 0.7032, + "step": 1006 + }, + { + "epoch": 1.5777661795407099, + "grad_norm": 1.0482438802719116, + "learning_rate": 4.3046144222627855e-06, + "loss": 0.7156, + "step": 1007 + }, + { + "epoch": 1.5793319415448852, + "grad_norm": 0.9434391856193542, + "learning_rate": 4.303155793707759e-06, + "loss": 0.7816, + "step": 1008 + }, + { + "epoch": 1.5808977035490606, + "grad_norm": 1.0497010946273804, + "learning_rate": 4.3016958846480706e-06, + "loss": 0.735, + "step": 1009 + }, + { + "epoch": 1.582463465553236, + "grad_norm": 0.955949068069458, + "learning_rate": 4.300234696120469e-06, + "loss": 0.6414, + "step": 1010 + }, + { + "epoch": 1.5840292275574113, + "grad_norm": 1.0462027788162231, + "learning_rate": 4.298772229162613e-06, + "loss": 0.6949, + "step": 1011 + }, + { + "epoch": 1.5855949895615866, + "grad_norm": 1.0634411573410034, + "learning_rate": 4.297308484813067e-06, + "loss": 0.7736, + "step": 1012 + }, + { + "epoch": 1.587160751565762, + "grad_norm": 1.0463027954101562, + "learning_rate": 4.295843464111306e-06, + "loss": 0.6812, + "step": 1013 + }, + { + "epoch": 1.5887265135699373, + "grad_norm": 1.0409252643585205, + "learning_rate": 4.294377168097706e-06, + "loss": 0.7319, + "step": 1014 + }, + { + "epoch": 1.5902922755741127, + "grad_norm": 0.9714667201042175, + "learning_rate": 4.292909597813555e-06, + "loss": 0.7473, + "step": 1015 + }, + { + "epoch": 1.5918580375782883, + "grad_norm": 1.03997802734375, + "learning_rate": 4.291440754301041e-06, + "loss": 0.7261, + "step": 1016 + }, + { + "epoch": 1.5934237995824634, + "grad_norm": 1.063293218612671, + "learning_rate": 4.289970638603258e-06, + "loss": 0.7909, + "step": 1017 + }, + { + "epoch": 1.594989561586639, + "grad_norm": 1.110007643699646, + "learning_rate": 4.288499251764205e-06, + "loss": 0.7, + "step": 1018 + }, + { + "epoch": 1.596555323590814, + "grad_norm": 0.981261134147644, + "learning_rate": 4.2870265948287806e-06, + "loss": 0.7516, + "step": 1019 + }, + { + "epoch": 1.5981210855949897, + "grad_norm": 1.0653973817825317, + "learning_rate": 4.2855526688427875e-06, + "loss": 0.7278, + "step": 1020 + }, + { + "epoch": 1.5996868475991648, + "grad_norm": 0.9212517738342285, + "learning_rate": 4.284077474852928e-06, + "loss": 0.7251, + "step": 1021 + }, + { + "epoch": 1.6012526096033404, + "grad_norm": 1.0887659788131714, + "learning_rate": 4.282601013906808e-06, + "loss": 0.7271, + "step": 1022 + }, + { + "epoch": 1.6028183716075155, + "grad_norm": 1.0753779411315918, + "learning_rate": 4.281123287052928e-06, + "loss": 0.7679, + "step": 1023 + }, + { + "epoch": 1.604384133611691, + "grad_norm": 1.0282115936279297, + "learning_rate": 4.2796442953406935e-06, + "loss": 0.695, + "step": 1024 + }, + { + "epoch": 1.6059498956158664, + "grad_norm": 0.9782281517982483, + "learning_rate": 4.2781640398204036e-06, + "loss": 0.7717, + "step": 1025 + }, + { + "epoch": 1.6075156576200418, + "grad_norm": 1.06228768825531, + "learning_rate": 4.276682521543256e-06, + "loss": 0.8069, + "step": 1026 + }, + { + "epoch": 1.6090814196242171, + "grad_norm": 0.8853397965431213, + "learning_rate": 4.2751997415613465e-06, + "loss": 0.8174, + "step": 1027 + }, + { + "epoch": 1.6106471816283925, + "grad_norm": 1.1525455713272095, + "learning_rate": 4.273715700927666e-06, + "loss": 0.7149, + "step": 1028 + }, + { + "epoch": 1.6122129436325678, + "grad_norm": 0.8743720650672913, + "learning_rate": 4.272230400696101e-06, + "loss": 0.7296, + "step": 1029 + }, + { + "epoch": 1.6137787056367432, + "grad_norm": 1.1136301755905151, + "learning_rate": 4.270743841921431e-06, + "loss": 0.7539, + "step": 1030 + }, + { + "epoch": 1.6153444676409185, + "grad_norm": 0.9830031991004944, + "learning_rate": 4.269256025659332e-06, + "loss": 0.6545, + "step": 1031 + }, + { + "epoch": 1.616910229645094, + "grad_norm": 1.1221113204956055, + "learning_rate": 4.267766952966369e-06, + "loss": 0.6144, + "step": 1032 + }, + { + "epoch": 1.6184759916492695, + "grad_norm": 0.975713849067688, + "learning_rate": 4.266276624900004e-06, + "loss": 0.746, + "step": 1033 + }, + { + "epoch": 1.6200417536534446, + "grad_norm": 1.1131149530410767, + "learning_rate": 4.264785042518587e-06, + "loss": 0.705, + "step": 1034 + }, + { + "epoch": 1.6216075156576202, + "grad_norm": 1.0574990510940552, + "learning_rate": 4.263292206881361e-06, + "loss": 0.7681, + "step": 1035 + }, + { + "epoch": 1.6231732776617953, + "grad_norm": 0.9931074976921082, + "learning_rate": 4.261798119048456e-06, + "loss": 0.7854, + "step": 1036 + }, + { + "epoch": 1.6247390396659709, + "grad_norm": 1.1598386764526367, + "learning_rate": 4.260302780080896e-06, + "loss": 0.7406, + "step": 1037 + }, + { + "epoch": 1.626304801670146, + "grad_norm": 0.9910677671432495, + "learning_rate": 4.258806191040587e-06, + "loss": 0.7107, + "step": 1038 + }, + { + "epoch": 1.6278705636743216, + "grad_norm": 1.1023072004318237, + "learning_rate": 4.2573083529903294e-06, + "loss": 0.7964, + "step": 1039 + }, + { + "epoch": 1.6294363256784967, + "grad_norm": 0.991082489490509, + "learning_rate": 4.255809266993806e-06, + "loss": 0.762, + "step": 1040 + }, + { + "epoch": 1.6310020876826723, + "grad_norm": 0.9281193017959595, + "learning_rate": 4.254308934115589e-06, + "loss": 0.7272, + "step": 1041 + }, + { + "epoch": 1.6325678496868476, + "grad_norm": 1.20292067527771, + "learning_rate": 4.252807355421133e-06, + "loss": 0.7535, + "step": 1042 + }, + { + "epoch": 1.634133611691023, + "grad_norm": 1.0362178087234497, + "learning_rate": 4.251304531976778e-06, + "loss": 0.7524, + "step": 1043 + }, + { + "epoch": 1.6356993736951984, + "grad_norm": 1.0267268419265747, + "learning_rate": 4.249800464849751e-06, + "loss": 0.6588, + "step": 1044 + }, + { + "epoch": 1.6372651356993737, + "grad_norm": 1.0531187057495117, + "learning_rate": 4.248295155108158e-06, + "loss": 0.7328, + "step": 1045 + }, + { + "epoch": 1.638830897703549, + "grad_norm": 1.0008069276809692, + "learning_rate": 4.246788603820989e-06, + "loss": 0.719, + "step": 1046 + }, + { + "epoch": 1.6403966597077244, + "grad_norm": 1.064780831336975, + "learning_rate": 4.245280812058119e-06, + "loss": 0.7388, + "step": 1047 + }, + { + "epoch": 1.6419624217118998, + "grad_norm": 1.06492018699646, + "learning_rate": 4.243771780890298e-06, + "loss": 0.7495, + "step": 1048 + }, + { + "epoch": 1.6435281837160751, + "grad_norm": 0.9472903609275818, + "learning_rate": 4.24226151138916e-06, + "loss": 0.7409, + "step": 1049 + }, + { + "epoch": 1.6450939457202505, + "grad_norm": 1.0612895488739014, + "learning_rate": 4.240750004627217e-06, + "loss": 0.7605, + "step": 1050 + }, + { + "epoch": 1.6466597077244258, + "grad_norm": 0.9634467959403992, + "learning_rate": 4.239237261677863e-06, + "loss": 0.7109, + "step": 1051 + }, + { + "epoch": 1.6482254697286014, + "grad_norm": 1.1347044706344604, + "learning_rate": 4.2377232836153635e-06, + "loss": 0.693, + "step": 1052 + }, + { + "epoch": 1.6497912317327765, + "grad_norm": 1.0814287662506104, + "learning_rate": 4.236208071514865e-06, + "loss": 0.7159, + "step": 1053 + }, + { + "epoch": 1.651356993736952, + "grad_norm": 0.9427600502967834, + "learning_rate": 4.234691626452392e-06, + "loss": 0.7014, + "step": 1054 + }, + { + "epoch": 1.6529227557411272, + "grad_norm": 1.0177332162857056, + "learning_rate": 4.233173949504842e-06, + "loss": 0.7717, + "step": 1055 + }, + { + "epoch": 1.6544885177453028, + "grad_norm": 1.0621635913848877, + "learning_rate": 4.231655041749985e-06, + "loss": 0.7078, + "step": 1056 + }, + { + "epoch": 1.656054279749478, + "grad_norm": 1.0474066734313965, + "learning_rate": 4.230134904266472e-06, + "loss": 0.7869, + "step": 1057 + }, + { + "epoch": 1.6576200417536535, + "grad_norm": 0.9170332551002502, + "learning_rate": 4.228613538133821e-06, + "loss": 0.7806, + "step": 1058 + }, + { + "epoch": 1.6591858037578286, + "grad_norm": 1.010465383529663, + "learning_rate": 4.227090944432426e-06, + "loss": 0.7352, + "step": 1059 + }, + { + "epoch": 1.6607515657620042, + "grad_norm": 0.988068163394928, + "learning_rate": 4.22556712424355e-06, + "loss": 0.7173, + "step": 1060 + }, + { + "epoch": 1.6623173277661796, + "grad_norm": 1.0753575563430786, + "learning_rate": 4.22404207864933e-06, + "loss": 0.6981, + "step": 1061 + }, + { + "epoch": 1.663883089770355, + "grad_norm": 1.0075691938400269, + "learning_rate": 4.222515808732769e-06, + "loss": 0.7279, + "step": 1062 + }, + { + "epoch": 1.6654488517745303, + "grad_norm": 1.0954983234405518, + "learning_rate": 4.220988315577745e-06, + "loss": 0.7675, + "step": 1063 + }, + { + "epoch": 1.6670146137787056, + "grad_norm": 1.0368746519088745, + "learning_rate": 4.2194596002690016e-06, + "loss": 0.7466, + "step": 1064 + }, + { + "epoch": 1.668580375782881, + "grad_norm": 1.083906650543213, + "learning_rate": 4.217929663892149e-06, + "loss": 0.7517, + "step": 1065 + }, + { + "epoch": 1.6701461377870563, + "grad_norm": 1.1356595754623413, + "learning_rate": 4.216398507533667e-06, + "loss": 0.7601, + "step": 1066 + }, + { + "epoch": 1.6717118997912317, + "grad_norm": 1.1610008478164673, + "learning_rate": 4.2148661322809e-06, + "loss": 0.7356, + "step": 1067 + }, + { + "epoch": 1.673277661795407, + "grad_norm": 1.019673466682434, + "learning_rate": 4.21333253922206e-06, + "loss": 0.7618, + "step": 1068 + }, + { + "epoch": 1.6748434237995826, + "grad_norm": 1.1470829248428345, + "learning_rate": 4.2117977294462235e-06, + "loss": 0.8001, + "step": 1069 + }, + { + "epoch": 1.6764091858037578, + "grad_norm": 0.9750358462333679, + "learning_rate": 4.210261704043327e-06, + "loss": 0.6736, + "step": 1070 + }, + { + "epoch": 1.6779749478079333, + "grad_norm": 1.061862826347351, + "learning_rate": 4.208724464104176e-06, + "loss": 0.7751, + "step": 1071 + }, + { + "epoch": 1.6795407098121085, + "grad_norm": 1.0311503410339355, + "learning_rate": 4.2071860107204346e-06, + "loss": 0.7629, + "step": 1072 + }, + { + "epoch": 1.681106471816284, + "grad_norm": 1.0023622512817383, + "learning_rate": 4.2056463449846305e-06, + "loss": 0.6935, + "step": 1073 + }, + { + "epoch": 1.6826722338204592, + "grad_norm": 1.1624165773391724, + "learning_rate": 4.204105467990153e-06, + "loss": 0.7451, + "step": 1074 + }, + { + "epoch": 1.6842379958246347, + "grad_norm": 0.9852432608604431, + "learning_rate": 4.202563380831249e-06, + "loss": 0.7446, + "step": 1075 + }, + { + "epoch": 1.6858037578288099, + "grad_norm": 1.0009260177612305, + "learning_rate": 4.201020084603027e-06, + "loss": 0.726, + "step": 1076 + }, + { + "epoch": 1.6873695198329854, + "grad_norm": 1.0374701023101807, + "learning_rate": 4.199475580401452e-06, + "loss": 0.6687, + "step": 1077 + }, + { + "epoch": 1.6889352818371608, + "grad_norm": 1.1437046527862549, + "learning_rate": 4.19792986932335e-06, + "loss": 0.7802, + "step": 1078 + }, + { + "epoch": 1.6905010438413361, + "grad_norm": 0.9866487383842468, + "learning_rate": 4.1963829524664e-06, + "loss": 0.7117, + "step": 1079 + }, + { + "epoch": 1.6920668058455115, + "grad_norm": 0.9720652103424072, + "learning_rate": 4.194834830929142e-06, + "loss": 0.778, + "step": 1080 + }, + { + "epoch": 1.6936325678496869, + "grad_norm": 1.009422779083252, + "learning_rate": 4.1932855058109675e-06, + "loss": 0.6783, + "step": 1081 + }, + { + "epoch": 1.6951983298538622, + "grad_norm": 1.024781346321106, + "learning_rate": 4.191734978212124e-06, + "loss": 0.6697, + "step": 1082 + }, + { + "epoch": 1.6967640918580376, + "grad_norm": 1.060354471206665, + "learning_rate": 4.190183249233715e-06, + "loss": 0.7151, + "step": 1083 + }, + { + "epoch": 1.698329853862213, + "grad_norm": 1.118486762046814, + "learning_rate": 4.1886303199776924e-06, + "loss": 0.7408, + "step": 1084 + }, + { + "epoch": 1.6998956158663883, + "grad_norm": 0.9334209561347961, + "learning_rate": 4.187076191546865e-06, + "loss": 0.7646, + "step": 1085 + }, + { + "epoch": 1.7014613778705638, + "grad_norm": 1.0506161451339722, + "learning_rate": 4.185520865044892e-06, + "loss": 0.7299, + "step": 1086 + }, + { + "epoch": 1.703027139874739, + "grad_norm": 1.1261264085769653, + "learning_rate": 4.1839643415762825e-06, + "loss": 0.6813, + "step": 1087 + }, + { + "epoch": 1.7045929018789145, + "grad_norm": 0.9690954685211182, + "learning_rate": 4.182406622246396e-06, + "loss": 0.7417, + "step": 1088 + }, + { + "epoch": 1.7061586638830897, + "grad_norm": 1.0872795581817627, + "learning_rate": 4.180847708161442e-06, + "loss": 0.676, + "step": 1089 + }, + { + "epoch": 1.7077244258872653, + "grad_norm": 1.054091453552246, + "learning_rate": 4.179287600428476e-06, + "loss": 0.785, + "step": 1090 + }, + { + "epoch": 1.7092901878914404, + "grad_norm": 1.109984278678894, + "learning_rate": 4.177726300155404e-06, + "loss": 0.8033, + "step": 1091 + }, + { + "epoch": 1.710855949895616, + "grad_norm": 1.0984348058700562, + "learning_rate": 4.176163808450978e-06, + "loss": 0.6964, + "step": 1092 + }, + { + "epoch": 1.712421711899791, + "grad_norm": 1.0788499116897583, + "learning_rate": 4.174600126424797e-06, + "loss": 0.7596, + "step": 1093 + }, + { + "epoch": 1.7139874739039667, + "grad_norm": 1.0846409797668457, + "learning_rate": 4.173035255187302e-06, + "loss": 0.7322, + "step": 1094 + }, + { + "epoch": 1.715553235908142, + "grad_norm": 1.1376627683639526, + "learning_rate": 4.171469195849782e-06, + "loss": 0.6934, + "step": 1095 + }, + { + "epoch": 1.7171189979123174, + "grad_norm": 1.003432035446167, + "learning_rate": 4.169901949524368e-06, + "loss": 0.7663, + "step": 1096 + }, + { + "epoch": 1.7186847599164927, + "grad_norm": 0.9824592471122742, + "learning_rate": 4.168333517324036e-06, + "loss": 0.8253, + "step": 1097 + }, + { + "epoch": 1.720250521920668, + "grad_norm": 1.0256636142730713, + "learning_rate": 4.1667639003626e-06, + "loss": 0.6614, + "step": 1098 + }, + { + "epoch": 1.7218162839248434, + "grad_norm": 1.0290439128875732, + "learning_rate": 4.16519309975472e-06, + "loss": 0.7403, + "step": 1099 + }, + { + "epoch": 1.7233820459290188, + "grad_norm": 1.0358202457427979, + "learning_rate": 4.163621116615892e-06, + "loss": 0.7332, + "step": 1100 + }, + { + "epoch": 1.7249478079331941, + "grad_norm": 1.1046626567840576, + "learning_rate": 4.162047952062456e-06, + "loss": 0.6672, + "step": 1101 + }, + { + "epoch": 1.7265135699373695, + "grad_norm": 1.116408109664917, + "learning_rate": 4.16047360721159e-06, + "loss": 0.6997, + "step": 1102 + }, + { + "epoch": 1.7280793319415448, + "grad_norm": 0.98244309425354, + "learning_rate": 4.158898083181308e-06, + "loss": 0.754, + "step": 1103 + }, + { + "epoch": 1.7296450939457202, + "grad_norm": 1.033761978149414, + "learning_rate": 4.1573213810904635e-06, + "loss": 0.7317, + "step": 1104 + }, + { + "epoch": 1.7312108559498958, + "grad_norm": 1.1415889263153076, + "learning_rate": 4.155743502058745e-06, + "loss": 0.8361, + "step": 1105 + }, + { + "epoch": 1.732776617954071, + "grad_norm": 0.9897739291191101, + "learning_rate": 4.154164447206679e-06, + "loss": 0.7361, + "step": 1106 + }, + { + "epoch": 1.7343423799582465, + "grad_norm": 1.0837332010269165, + "learning_rate": 4.1525842176556235e-06, + "loss": 0.7007, + "step": 1107 + }, + { + "epoch": 1.7359081419624216, + "grad_norm": 1.021332025527954, + "learning_rate": 4.151002814527774e-06, + "loss": 0.723, + "step": 1108 + }, + { + "epoch": 1.7374739039665972, + "grad_norm": 0.9808626770973206, + "learning_rate": 4.14942023894616e-06, + "loss": 0.6121, + "step": 1109 + }, + { + "epoch": 1.7390396659707723, + "grad_norm": 1.0266727209091187, + "learning_rate": 4.14783649203464e-06, + "loss": 0.8221, + "step": 1110 + }, + { + "epoch": 1.7406054279749479, + "grad_norm": 1.0880744457244873, + "learning_rate": 4.146251574917907e-06, + "loss": 0.7159, + "step": 1111 + }, + { + "epoch": 1.742171189979123, + "grad_norm": 1.0125631093978882, + "learning_rate": 4.144665488721483e-06, + "loss": 0.7228, + "step": 1112 + }, + { + "epoch": 1.7437369519832986, + "grad_norm": 0.9339308142662048, + "learning_rate": 4.143078234571723e-06, + "loss": 0.7472, + "step": 1113 + }, + { + "epoch": 1.745302713987474, + "grad_norm": 1.1234889030456543, + "learning_rate": 4.141489813595809e-06, + "loss": 0.7875, + "step": 1114 + }, + { + "epoch": 1.7468684759916493, + "grad_norm": 1.0862795114517212, + "learning_rate": 4.139900226921753e-06, + "loss": 0.7219, + "step": 1115 + }, + { + "epoch": 1.7484342379958246, + "grad_norm": 1.027912974357605, + "learning_rate": 4.138309475678393e-06, + "loss": 0.6616, + "step": 1116 + }, + { + "epoch": 1.75, + "grad_norm": 1.0271416902542114, + "learning_rate": 4.136717560995398e-06, + "loss": 0.7245, + "step": 1117 + }, + { + "epoch": 1.7515657620041754, + "grad_norm": 0.9430286288261414, + "learning_rate": 4.135124484003258e-06, + "loss": 0.7347, + "step": 1118 + }, + { + "epoch": 1.7531315240083507, + "grad_norm": 1.1191871166229248, + "learning_rate": 4.133530245833292e-06, + "loss": 0.6713, + "step": 1119 + }, + { + "epoch": 1.754697286012526, + "grad_norm": 1.0734976530075073, + "learning_rate": 4.131934847617642e-06, + "loss": 0.7268, + "step": 1120 + }, + { + "epoch": 1.7562630480167014, + "grad_norm": 1.0727618932724, + "learning_rate": 4.130338290489276e-06, + "loss": 0.7855, + "step": 1121 + }, + { + "epoch": 1.757828810020877, + "grad_norm": 1.0351287126541138, + "learning_rate": 4.1287405755819805e-06, + "loss": 0.7297, + "step": 1122 + }, + { + "epoch": 1.7593945720250521, + "grad_norm": 1.046382188796997, + "learning_rate": 4.127141704030371e-06, + "loss": 0.6508, + "step": 1123 + }, + { + "epoch": 1.7609603340292277, + "grad_norm": 1.0040491819381714, + "learning_rate": 4.125541676969876e-06, + "loss": 0.7707, + "step": 1124 + }, + { + "epoch": 1.7625260960334028, + "grad_norm": 1.1626756191253662, + "learning_rate": 4.123940495536752e-06, + "loss": 0.7366, + "step": 1125 + }, + { + "epoch": 1.7640918580375784, + "grad_norm": 0.9720292091369629, + "learning_rate": 4.122338160868071e-06, + "loss": 0.7641, + "step": 1126 + }, + { + "epoch": 1.7656576200417535, + "grad_norm": 0.9957835078239441, + "learning_rate": 4.1207346741017265e-06, + "loss": 0.7398, + "step": 1127 + }, + { + "epoch": 1.767223382045929, + "grad_norm": 0.972172737121582, + "learning_rate": 4.1191300363764276e-06, + "loss": 0.6565, + "step": 1128 + }, + { + "epoch": 1.7687891440501042, + "grad_norm": 1.0128676891326904, + "learning_rate": 4.117524248831702e-06, + "loss": 0.7433, + "step": 1129 + }, + { + "epoch": 1.7703549060542798, + "grad_norm": 1.1161550283432007, + "learning_rate": 4.115917312607895e-06, + "loss": 0.8241, + "step": 1130 + }, + { + "epoch": 1.7719206680584552, + "grad_norm": 1.009810209274292, + "learning_rate": 4.114309228846167e-06, + "loss": 0.6939, + "step": 1131 + }, + { + "epoch": 1.7734864300626305, + "grad_norm": 1.0082894563674927, + "learning_rate": 4.112699998688492e-06, + "loss": 0.6772, + "step": 1132 + }, + { + "epoch": 1.7750521920668059, + "grad_norm": 0.9453067183494568, + "learning_rate": 4.111089623277658e-06, + "loss": 0.6982, + "step": 1133 + }, + { + "epoch": 1.7766179540709812, + "grad_norm": 1.074747920036316, + "learning_rate": 4.1094781037572696e-06, + "loss": 0.7296, + "step": 1134 + }, + { + "epoch": 1.7781837160751566, + "grad_norm": 1.0477592945098877, + "learning_rate": 4.1078654412717404e-06, + "loss": 0.8162, + "step": 1135 + }, + { + "epoch": 1.779749478079332, + "grad_norm": 1.1362730264663696, + "learning_rate": 4.106251636966297e-06, + "loss": 0.6942, + "step": 1136 + }, + { + "epoch": 1.7813152400835073, + "grad_norm": 0.9564191102981567, + "learning_rate": 4.104636691986977e-06, + "loss": 0.7613, + "step": 1137 + }, + { + "epoch": 1.7828810020876826, + "grad_norm": 1.0384364128112793, + "learning_rate": 4.103020607480628e-06, + "loss": 0.6482, + "step": 1138 + }, + { + "epoch": 1.784446764091858, + "grad_norm": 1.1258740425109863, + "learning_rate": 4.101403384594905e-06, + "loss": 0.7736, + "step": 1139 + }, + { + "epoch": 1.7860125260960333, + "grad_norm": 1.0740283727645874, + "learning_rate": 4.099785024478276e-06, + "loss": 0.7219, + "step": 1140 + }, + { + "epoch": 1.787578288100209, + "grad_norm": 1.0377696752548218, + "learning_rate": 4.098165528280012e-06, + "loss": 0.7132, + "step": 1141 + }, + { + "epoch": 1.789144050104384, + "grad_norm": 1.0551276206970215, + "learning_rate": 4.096544897150192e-06, + "loss": 0.7879, + "step": 1142 + }, + { + "epoch": 1.7907098121085596, + "grad_norm": 1.096389651298523, + "learning_rate": 4.0949231322397036e-06, + "loss": 0.7516, + "step": 1143 + }, + { + "epoch": 1.7922755741127347, + "grad_norm": 0.9602596163749695, + "learning_rate": 4.093300234700234e-06, + "loss": 0.7197, + "step": 1144 + }, + { + "epoch": 1.7938413361169103, + "grad_norm": 1.1918690204620361, + "learning_rate": 4.09167620568428e-06, + "loss": 0.7634, + "step": 1145 + }, + { + "epoch": 1.7954070981210855, + "grad_norm": 1.1249059438705444, + "learning_rate": 4.09005104634514e-06, + "loss": 0.7718, + "step": 1146 + }, + { + "epoch": 1.796972860125261, + "grad_norm": 1.0596246719360352, + "learning_rate": 4.0884247578369164e-06, + "loss": 0.7258, + "step": 1147 + }, + { + "epoch": 1.7985386221294362, + "grad_norm": 1.026201844215393, + "learning_rate": 4.086797341314509e-06, + "loss": 0.6426, + "step": 1148 + }, + { + "epoch": 1.8001043841336117, + "grad_norm": 1.0791343450546265, + "learning_rate": 4.085168797933625e-06, + "loss": 0.7147, + "step": 1149 + }, + { + "epoch": 1.801670146137787, + "grad_norm": 0.9660344123840332, + "learning_rate": 4.083539128850766e-06, + "loss": 0.7335, + "step": 1150 + }, + { + "epoch": 1.8032359081419624, + "grad_norm": 1.0022811889648438, + "learning_rate": 4.081908335223238e-06, + "loss": 0.6824, + "step": 1151 + }, + { + "epoch": 1.8048016701461378, + "grad_norm": 0.9261506199836731, + "learning_rate": 4.080276418209143e-06, + "loss": 0.7538, + "step": 1152 + }, + { + "epoch": 1.8063674321503131, + "grad_norm": 1.088552713394165, + "learning_rate": 4.078643378967379e-06, + "loss": 0.6988, + "step": 1153 + }, + { + "epoch": 1.8079331941544885, + "grad_norm": 1.0581461191177368, + "learning_rate": 4.077009218657645e-06, + "loss": 0.7075, + "step": 1154 + }, + { + "epoch": 1.8094989561586639, + "grad_norm": 1.0722301006317139, + "learning_rate": 4.0753739384404335e-06, + "loss": 0.6974, + "step": 1155 + }, + { + "epoch": 1.8110647181628392, + "grad_norm": 0.9519115090370178, + "learning_rate": 4.073737539477033e-06, + "loss": 0.6898, + "step": 1156 + }, + { + "epoch": 1.8126304801670146, + "grad_norm": 1.1201752424240112, + "learning_rate": 4.072100022929526e-06, + "loss": 0.8572, + "step": 1157 + }, + { + "epoch": 1.8141962421711901, + "grad_norm": 1.24104642868042, + "learning_rate": 4.070461389960789e-06, + "loss": 0.7407, + "step": 1158 + }, + { + "epoch": 1.8157620041753653, + "grad_norm": 1.0625057220458984, + "learning_rate": 4.068821641734492e-06, + "loss": 0.655, + "step": 1159 + }, + { + "epoch": 1.8173277661795408, + "grad_norm": 1.1174049377441406, + "learning_rate": 4.067180779415097e-06, + "loss": 0.7214, + "step": 1160 + }, + { + "epoch": 1.818893528183716, + "grad_norm": 1.0683097839355469, + "learning_rate": 4.065538804167855e-06, + "loss": 0.7863, + "step": 1161 + }, + { + "epoch": 1.8204592901878915, + "grad_norm": 1.031198263168335, + "learning_rate": 4.063895717158809e-06, + "loss": 0.7663, + "step": 1162 + }, + { + "epoch": 1.8220250521920667, + "grad_norm": 1.1091554164886475, + "learning_rate": 4.062251519554791e-06, + "loss": 0.7234, + "step": 1163 + }, + { + "epoch": 1.8235908141962422, + "grad_norm": 1.0818198919296265, + "learning_rate": 4.060606212523425e-06, + "loss": 0.7709, + "step": 1164 + }, + { + "epoch": 1.8251565762004174, + "grad_norm": 1.0104292631149292, + "learning_rate": 4.058959797233116e-06, + "loss": 0.7225, + "step": 1165 + }, + { + "epoch": 1.826722338204593, + "grad_norm": 1.1726491451263428, + "learning_rate": 4.057312274853063e-06, + "loss": 0.6831, + "step": 1166 + }, + { + "epoch": 1.8282881002087683, + "grad_norm": 0.9735565185546875, + "learning_rate": 4.055663646553248e-06, + "loss": 0.7123, + "step": 1167 + }, + { + "epoch": 1.8298538622129437, + "grad_norm": 1.011709213256836, + "learning_rate": 4.054013913504436e-06, + "loss": 0.6885, + "step": 1168 + }, + { + "epoch": 1.831419624217119, + "grad_norm": 1.0374364852905273, + "learning_rate": 4.05236307687818e-06, + "loss": 0.782, + "step": 1169 + }, + { + "epoch": 1.8329853862212944, + "grad_norm": 1.011202096939087, + "learning_rate": 4.050711137846817e-06, + "loss": 0.7381, + "step": 1170 + }, + { + "epoch": 1.8345511482254697, + "grad_norm": 1.0984755754470825, + "learning_rate": 4.049058097583464e-06, + "loss": 0.7472, + "step": 1171 + }, + { + "epoch": 1.836116910229645, + "grad_norm": 1.090078353881836, + "learning_rate": 4.047403957262024e-06, + "loss": 0.6773, + "step": 1172 + }, + { + "epoch": 1.8376826722338204, + "grad_norm": 0.9866096377372742, + "learning_rate": 4.045748718057176e-06, + "loss": 0.6812, + "step": 1173 + }, + { + "epoch": 1.8392484342379958, + "grad_norm": 1.1006139516830444, + "learning_rate": 4.0440923811443846e-06, + "loss": 0.7953, + "step": 1174 + }, + { + "epoch": 1.8408141962421714, + "grad_norm": 0.9621697664260864, + "learning_rate": 4.04243494769989e-06, + "loss": 0.7557, + "step": 1175 + }, + { + "epoch": 1.8423799582463465, + "grad_norm": 1.1520715951919556, + "learning_rate": 4.040776418900714e-06, + "loss": 0.7422, + "step": 1176 + }, + { + "epoch": 1.843945720250522, + "grad_norm": 1.0679237842559814, + "learning_rate": 4.039116795924653e-06, + "loss": 0.6738, + "step": 1177 + }, + { + "epoch": 1.8455114822546972, + "grad_norm": 1.026510238647461, + "learning_rate": 4.037456079950284e-06, + "loss": 0.7741, + "step": 1178 + }, + { + "epoch": 1.8470772442588728, + "grad_norm": 1.0111522674560547, + "learning_rate": 4.0357942721569585e-06, + "loss": 0.6968, + "step": 1179 + }, + { + "epoch": 1.848643006263048, + "grad_norm": 1.0290493965148926, + "learning_rate": 4.034131373724802e-06, + "loss": 0.7131, + "step": 1180 + }, + { + "epoch": 1.8502087682672235, + "grad_norm": 0.9878649711608887, + "learning_rate": 4.032467385834718e-06, + "loss": 0.7477, + "step": 1181 + }, + { + "epoch": 1.8517745302713986, + "grad_norm": 1.0317823886871338, + "learning_rate": 4.030802309668379e-06, + "loss": 0.6318, + "step": 1182 + }, + { + "epoch": 1.8533402922755742, + "grad_norm": 1.0163817405700684, + "learning_rate": 4.029136146408235e-06, + "loss": 0.7027, + "step": 1183 + }, + { + "epoch": 1.8549060542797495, + "grad_norm": 0.9590600728988647, + "learning_rate": 4.027468897237505e-06, + "loss": 0.6988, + "step": 1184 + }, + { + "epoch": 1.8564718162839249, + "grad_norm": 1.0645931959152222, + "learning_rate": 4.0258005633401795e-06, + "loss": 0.7588, + "step": 1185 + }, + { + "epoch": 1.8580375782881002, + "grad_norm": 1.0380210876464844, + "learning_rate": 4.024131145901022e-06, + "loss": 0.6984, + "step": 1186 + }, + { + "epoch": 1.8596033402922756, + "grad_norm": 1.0558087825775146, + "learning_rate": 4.022460646105561e-06, + "loss": 0.6729, + "step": 1187 + }, + { + "epoch": 1.861169102296451, + "grad_norm": 1.0865014791488647, + "learning_rate": 4.020789065140097e-06, + "loss": 0.7208, + "step": 1188 + }, + { + "epoch": 1.8627348643006263, + "grad_norm": 1.0456374883651733, + "learning_rate": 4.019116404191697e-06, + "loss": 0.7759, + "step": 1189 + }, + { + "epoch": 1.8643006263048016, + "grad_norm": 0.9724026918411255, + "learning_rate": 4.017442664448197e-06, + "loss": 0.7475, + "step": 1190 + }, + { + "epoch": 1.865866388308977, + "grad_norm": 1.1170289516448975, + "learning_rate": 4.015767847098194e-06, + "loss": 0.6993, + "step": 1191 + }, + { + "epoch": 1.8674321503131524, + "grad_norm": 1.1052978038787842, + "learning_rate": 4.014091953331056e-06, + "loss": 0.748, + "step": 1192 + }, + { + "epoch": 1.8689979123173277, + "grad_norm": 1.1223454475402832, + "learning_rate": 4.012414984336912e-06, + "loss": 0.6921, + "step": 1193 + }, + { + "epoch": 1.8705636743215033, + "grad_norm": 1.0611138343811035, + "learning_rate": 4.010736941306655e-06, + "loss": 0.7859, + "step": 1194 + }, + { + "epoch": 1.8721294363256784, + "grad_norm": 1.169189691543579, + "learning_rate": 4.009057825431944e-06, + "loss": 0.5874, + "step": 1195 + }, + { + "epoch": 1.873695198329854, + "grad_norm": 1.036626935005188, + "learning_rate": 4.0073776379051945e-06, + "loss": 0.6801, + "step": 1196 + }, + { + "epoch": 1.8752609603340291, + "grad_norm": 0.9431620240211487, + "learning_rate": 4.005696379919586e-06, + "loss": 0.7476, + "step": 1197 + }, + { + "epoch": 1.8768267223382047, + "grad_norm": 1.0381059646606445, + "learning_rate": 4.0040140526690585e-06, + "loss": 0.6691, + "step": 1198 + }, + { + "epoch": 1.8783924843423798, + "grad_norm": 1.120033860206604, + "learning_rate": 4.0023306573483105e-06, + "loss": 0.6669, + "step": 1199 + }, + { + "epoch": 1.8799582463465554, + "grad_norm": 1.0298967361450195, + "learning_rate": 4.000646195152798e-06, + "loss": 0.7355, + "step": 1200 + }, + { + "epoch": 1.8815240083507305, + "grad_norm": 1.0036594867706299, + "learning_rate": 3.99896066727874e-06, + "loss": 0.7545, + "step": 1201 + }, + { + "epoch": 1.883089770354906, + "grad_norm": 1.1166342496871948, + "learning_rate": 3.997274074923103e-06, + "loss": 0.7468, + "step": 1202 + }, + { + "epoch": 1.8846555323590815, + "grad_norm": 1.0512094497680664, + "learning_rate": 3.995586419283617e-06, + "loss": 0.7647, + "step": 1203 + }, + { + "epoch": 1.8862212943632568, + "grad_norm": 1.0115973949432373, + "learning_rate": 3.993897701558764e-06, + "loss": 0.7893, + "step": 1204 + }, + { + "epoch": 1.8877870563674322, + "grad_norm": 1.0839310884475708, + "learning_rate": 3.992207922947783e-06, + "loss": 0.6824, + "step": 1205 + }, + { + "epoch": 1.8893528183716075, + "grad_norm": 0.9987578988075256, + "learning_rate": 3.99051708465066e-06, + "loss": 0.8045, + "step": 1206 + }, + { + "epoch": 1.8909185803757829, + "grad_norm": 1.0759059190750122, + "learning_rate": 3.9888251878681425e-06, + "loss": 0.7756, + "step": 1207 + }, + { + "epoch": 1.8924843423799582, + "grad_norm": 1.1055301427841187, + "learning_rate": 3.987132233801722e-06, + "loss": 0.6947, + "step": 1208 + }, + { + "epoch": 1.8940501043841336, + "grad_norm": 1.042575478553772, + "learning_rate": 3.985438223653644e-06, + "loss": 0.6669, + "step": 1209 + }, + { + "epoch": 1.895615866388309, + "grad_norm": 1.0595242977142334, + "learning_rate": 3.983743158626904e-06, + "loss": 0.6937, + "step": 1210 + }, + { + "epoch": 1.8971816283924845, + "grad_norm": 1.0498789548873901, + "learning_rate": 3.982047039925247e-06, + "loss": 0.7244, + "step": 1211 + }, + { + "epoch": 1.8987473903966596, + "grad_norm": 1.0960590839385986, + "learning_rate": 3.980349868753166e-06, + "loss": 0.7661, + "step": 1212 + }, + { + "epoch": 1.9003131524008352, + "grad_norm": 1.108823299407959, + "learning_rate": 3.9786516463159e-06, + "loss": 0.6937, + "step": 1213 + }, + { + "epoch": 1.9018789144050103, + "grad_norm": 1.051416039466858, + "learning_rate": 3.976952373819436e-06, + "loss": 0.7263, + "step": 1214 + }, + { + "epoch": 1.903444676409186, + "grad_norm": 0.9884061217308044, + "learning_rate": 3.9752520524705065e-06, + "loss": 0.6577, + "step": 1215 + }, + { + "epoch": 1.905010438413361, + "grad_norm": 1.0950369834899902, + "learning_rate": 3.9735506834765885e-06, + "loss": 0.7376, + "step": 1216 + }, + { + "epoch": 1.9065762004175366, + "grad_norm": 1.1140044927597046, + "learning_rate": 3.971848268045902e-06, + "loss": 0.7125, + "step": 1217 + }, + { + "epoch": 1.9081419624217117, + "grad_norm": 1.0908716917037964, + "learning_rate": 3.970144807387414e-06, + "loss": 0.7796, + "step": 1218 + }, + { + "epoch": 1.9097077244258873, + "grad_norm": 1.1068549156188965, + "learning_rate": 3.968440302710829e-06, + "loss": 0.7454, + "step": 1219 + }, + { + "epoch": 1.9112734864300627, + "grad_norm": 1.102594017982483, + "learning_rate": 3.9667347552265945e-06, + "loss": 0.757, + "step": 1220 + }, + { + "epoch": 1.912839248434238, + "grad_norm": 1.1588326692581177, + "learning_rate": 3.9650281661459015e-06, + "loss": 0.6491, + "step": 1221 + }, + { + "epoch": 1.9144050104384134, + "grad_norm": 0.9896461367607117, + "learning_rate": 3.963320536680675e-06, + "loss": 0.773, + "step": 1222 + }, + { + "epoch": 1.9159707724425887, + "grad_norm": 1.0855952501296997, + "learning_rate": 3.961611868043583e-06, + "loss": 0.6906, + "step": 1223 + }, + { + "epoch": 1.917536534446764, + "grad_norm": 1.0556972026824951, + "learning_rate": 3.959902161448032e-06, + "loss": 0.7577, + "step": 1224 + }, + { + "epoch": 1.9191022964509394, + "grad_norm": 1.0284528732299805, + "learning_rate": 3.958191418108162e-06, + "loss": 0.6844, + "step": 1225 + }, + { + "epoch": 1.9206680584551148, + "grad_norm": 0.9578842520713806, + "learning_rate": 3.956479639238852e-06, + "loss": 0.7216, + "step": 1226 + }, + { + "epoch": 1.9222338204592901, + "grad_norm": 1.0366833209991455, + "learning_rate": 3.954766826055717e-06, + "loss": 0.7365, + "step": 1227 + }, + { + "epoch": 1.9237995824634657, + "grad_norm": 1.0498946905136108, + "learning_rate": 3.953052979775103e-06, + "loss": 0.6788, + "step": 1228 + }, + { + "epoch": 1.9253653444676408, + "grad_norm": 1.010860562324524, + "learning_rate": 3.951338101614094e-06, + "loss": 0.7181, + "step": 1229 + }, + { + "epoch": 1.9269311064718164, + "grad_norm": 1.0203677415847778, + "learning_rate": 3.949622192790502e-06, + "loss": 0.7277, + "step": 1230 + }, + { + "epoch": 1.9284968684759916, + "grad_norm": 0.9907971024513245, + "learning_rate": 3.947905254522876e-06, + "loss": 0.6858, + "step": 1231 + }, + { + "epoch": 1.9300626304801671, + "grad_norm": 0.9824337363243103, + "learning_rate": 3.946187288030493e-06, + "loss": 0.7517, + "step": 1232 + }, + { + "epoch": 1.9316283924843423, + "grad_norm": 1.0206327438354492, + "learning_rate": 3.944468294533359e-06, + "loss": 0.7225, + "step": 1233 + }, + { + "epoch": 1.9331941544885178, + "grad_norm": 1.0529900789260864, + "learning_rate": 3.942748275252214e-06, + "loss": 0.7965, + "step": 1234 + }, + { + "epoch": 1.934759916492693, + "grad_norm": 1.0501983165740967, + "learning_rate": 3.941027231408521e-06, + "loss": 0.7646, + "step": 1235 + }, + { + "epoch": 1.9363256784968685, + "grad_norm": 0.9482662081718445, + "learning_rate": 3.939305164224474e-06, + "loss": 0.7449, + "step": 1236 + }, + { + "epoch": 1.937891440501044, + "grad_norm": 1.096723198890686, + "learning_rate": 3.937582074922993e-06, + "loss": 0.7464, + "step": 1237 + }, + { + "epoch": 1.9394572025052192, + "grad_norm": 1.0417059659957886, + "learning_rate": 3.935857964727725e-06, + "loss": 0.6586, + "step": 1238 + }, + { + "epoch": 1.9410229645093946, + "grad_norm": 1.1008849143981934, + "learning_rate": 3.934132834863039e-06, + "loss": 0.6236, + "step": 1239 + }, + { + "epoch": 1.94258872651357, + "grad_norm": 0.9107227921485901, + "learning_rate": 3.93240668655403e-06, + "loss": 0.7069, + "step": 1240 + }, + { + "epoch": 1.9441544885177453, + "grad_norm": 1.0516945123672485, + "learning_rate": 3.930679521026517e-06, + "loss": 0.773, + "step": 1241 + }, + { + "epoch": 1.9457202505219207, + "grad_norm": 0.9941951632499695, + "learning_rate": 3.92895133950704e-06, + "loss": 0.6002, + "step": 1242 + }, + { + "epoch": 1.947286012526096, + "grad_norm": 1.031784176826477, + "learning_rate": 3.927222143222859e-06, + "loss": 0.7737, + "step": 1243 + }, + { + "epoch": 1.9488517745302714, + "grad_norm": 0.986473798751831, + "learning_rate": 3.925491933401961e-06, + "loss": 0.7455, + "step": 1244 + }, + { + "epoch": 1.9504175365344467, + "grad_norm": 1.109964370727539, + "learning_rate": 3.923760711273044e-06, + "loss": 0.7048, + "step": 1245 + }, + { + "epoch": 1.951983298538622, + "grad_norm": 1.1303372383117676, + "learning_rate": 3.922028478065532e-06, + "loss": 0.749, + "step": 1246 + }, + { + "epoch": 1.9535490605427976, + "grad_norm": 0.999824583530426, + "learning_rate": 3.920295235009562e-06, + "loss": 0.7596, + "step": 1247 + }, + { + "epoch": 1.9551148225469728, + "grad_norm": 1.0440820455551147, + "learning_rate": 3.918560983335993e-06, + "loss": 0.7841, + "step": 1248 + }, + { + "epoch": 1.9566805845511483, + "grad_norm": 1.0420323610305786, + "learning_rate": 3.916825724276395e-06, + "loss": 0.7837, + "step": 1249 + }, + { + "epoch": 1.9582463465553235, + "grad_norm": 1.001997947692871, + "learning_rate": 3.915089459063058e-06, + "loss": 0.6956, + "step": 1250 + }, + { + "epoch": 1.959812108559499, + "grad_norm": 1.030626893043518, + "learning_rate": 3.913352188928984e-06, + "loss": 0.74, + "step": 1251 + }, + { + "epoch": 1.9613778705636742, + "grad_norm": 1.0981760025024414, + "learning_rate": 3.911613915107888e-06, + "loss": 0.713, + "step": 1252 + }, + { + "epoch": 1.9629436325678498, + "grad_norm": 1.020772933959961, + "learning_rate": 3.9098746388342006e-06, + "loss": 0.7605, + "step": 1253 + }, + { + "epoch": 1.964509394572025, + "grad_norm": 1.0206503868103027, + "learning_rate": 3.908134361343062e-06, + "loss": 0.7176, + "step": 1254 + }, + { + "epoch": 1.9660751565762005, + "grad_norm": 1.0471689701080322, + "learning_rate": 3.906393083870325e-06, + "loss": 0.7698, + "step": 1255 + }, + { + "epoch": 1.9676409185803758, + "grad_norm": 1.0658202171325684, + "learning_rate": 3.904650807652549e-06, + "loss": 0.7179, + "step": 1256 + }, + { + "epoch": 1.9692066805845512, + "grad_norm": 1.196192741394043, + "learning_rate": 3.902907533927007e-06, + "loss": 0.7801, + "step": 1257 + }, + { + "epoch": 1.9707724425887265, + "grad_norm": 1.0874967575073242, + "learning_rate": 3.90116326393168e-06, + "loss": 0.7672, + "step": 1258 + }, + { + "epoch": 1.9723382045929019, + "grad_norm": 1.153617024421692, + "learning_rate": 3.899417998905253e-06, + "loss": 0.752, + "step": 1259 + }, + { + "epoch": 1.9739039665970772, + "grad_norm": 1.0750453472137451, + "learning_rate": 3.89767174008712e-06, + "loss": 0.7471, + "step": 1260 + }, + { + "epoch": 1.9754697286012526, + "grad_norm": 1.0693799257278442, + "learning_rate": 3.895924488717381e-06, + "loss": 0.6704, + "step": 1261 + }, + { + "epoch": 1.977035490605428, + "grad_norm": 0.9552210569381714, + "learning_rate": 3.89417624603684e-06, + "loss": 0.7793, + "step": 1262 + }, + { + "epoch": 1.9786012526096033, + "grad_norm": 0.96712726354599, + "learning_rate": 3.892427013287007e-06, + "loss": 0.6843, + "step": 1263 + }, + { + "epoch": 1.9801670146137789, + "grad_norm": 1.1453518867492676, + "learning_rate": 3.890676791710091e-06, + "loss": 0.7532, + "step": 1264 + }, + { + "epoch": 1.981732776617954, + "grad_norm": 1.0294159650802612, + "learning_rate": 3.888925582549006e-06, + "loss": 0.703, + "step": 1265 + }, + { + "epoch": 1.9832985386221296, + "grad_norm": 1.0790715217590332, + "learning_rate": 3.8871733870473686e-06, + "loss": 0.6731, + "step": 1266 + }, + { + "epoch": 1.9848643006263047, + "grad_norm": 1.052460789680481, + "learning_rate": 3.8854202064494915e-06, + "loss": 0.748, + "step": 1267 + }, + { + "epoch": 1.9864300626304803, + "grad_norm": 1.112462043762207, + "learning_rate": 3.883666042000392e-06, + "loss": 0.7408, + "step": 1268 + }, + { + "epoch": 1.9879958246346554, + "grad_norm": 1.0738476514816284, + "learning_rate": 3.881910894945783e-06, + "loss": 0.759, + "step": 1269 + }, + { + "epoch": 1.989561586638831, + "grad_norm": 1.136472463607788, + "learning_rate": 3.8801547665320775e-06, + "loss": 0.754, + "step": 1270 + }, + { + "epoch": 1.9911273486430061, + "grad_norm": 1.0640734434127808, + "learning_rate": 3.87839765800638e-06, + "loss": 0.719, + "step": 1271 + }, + { + "epoch": 1.9926931106471817, + "grad_norm": 0.9155470132827759, + "learning_rate": 3.876639570616498e-06, + "loss": 0.762, + "step": 1272 + }, + { + "epoch": 1.994258872651357, + "grad_norm": 1.015743374824524, + "learning_rate": 3.87488050561093e-06, + "loss": 0.6926, + "step": 1273 + }, + { + "epoch": 1.9958246346555324, + "grad_norm": 1.0829862356185913, + "learning_rate": 3.87312046423887e-06, + "loss": 0.7644, + "step": 1274 + }, + { + "epoch": 1.9973903966597077, + "grad_norm": 1.0962239503860474, + "learning_rate": 3.871359447750205e-06, + "loss": 0.7038, + "step": 1275 + }, + { + "epoch": 1.998956158663883, + "grad_norm": 1.0470675230026245, + "learning_rate": 3.869597457395514e-06, + "loss": 0.7629, + "step": 1276 + }, + { + "epoch": 2.0015657620041756, + "grad_norm": 1.962398648262024, + "learning_rate": 3.867834494426069e-06, + "loss": 1.4251, + "step": 1277 + }, + { + "epoch": 2.0031315240083507, + "grad_norm": 1.0987200736999512, + "learning_rate": 3.866070560093832e-06, + "loss": 0.6873, + "step": 1278 + }, + { + "epoch": 2.0046972860125263, + "grad_norm": 1.0096511840820312, + "learning_rate": 3.864305655651453e-06, + "loss": 0.7113, + "step": 1279 + }, + { + "epoch": 2.0062630480167014, + "grad_norm": 0.9988889098167419, + "learning_rate": 3.8625397823522755e-06, + "loss": 0.6555, + "step": 1280 + }, + { + "epoch": 2.007828810020877, + "grad_norm": 0.9960927963256836, + "learning_rate": 3.860772941450326e-06, + "loss": 0.6458, + "step": 1281 + }, + { + "epoch": 2.009394572025052, + "grad_norm": 0.971942126750946, + "learning_rate": 3.85900513420032e-06, + "loss": 0.6821, + "step": 1282 + }, + { + "epoch": 2.0109603340292277, + "grad_norm": 1.0935462713241577, + "learning_rate": 3.857236361857662e-06, + "loss": 0.5862, + "step": 1283 + }, + { + "epoch": 2.012526096033403, + "grad_norm": 1.0295177698135376, + "learning_rate": 3.855466625678435e-06, + "loss": 0.723, + "step": 1284 + }, + { + "epoch": 2.0140918580375784, + "grad_norm": 0.9750491976737976, + "learning_rate": 3.853695926919415e-06, + "loss": 0.7105, + "step": 1285 + }, + { + "epoch": 2.0156576200417535, + "grad_norm": 1.1649892330169678, + "learning_rate": 3.851924266838056e-06, + "loss": 0.7106, + "step": 1286 + }, + { + "epoch": 2.017223382045929, + "grad_norm": 1.068339467048645, + "learning_rate": 3.850151646692496e-06, + "loss": 0.668, + "step": 1287 + }, + { + "epoch": 2.0187891440501042, + "grad_norm": 0.9659110903739929, + "learning_rate": 3.848378067741555e-06, + "loss": 0.6201, + "step": 1288 + }, + { + "epoch": 2.02035490605428, + "grad_norm": 1.0711618661880493, + "learning_rate": 3.846603531244734e-06, + "loss": 0.6379, + "step": 1289 + }, + { + "epoch": 2.021920668058455, + "grad_norm": 1.092109203338623, + "learning_rate": 3.844828038462213e-06, + "loss": 0.7105, + "step": 1290 + }, + { + "epoch": 2.0234864300626305, + "grad_norm": 1.1485183238983154, + "learning_rate": 3.843051590654852e-06, + "loss": 0.724, + "step": 1291 + }, + { + "epoch": 2.0250521920668056, + "grad_norm": 1.1114039421081543, + "learning_rate": 3.841274189084189e-06, + "loss": 0.6928, + "step": 1292 + }, + { + "epoch": 2.026617954070981, + "grad_norm": 1.1090906858444214, + "learning_rate": 3.839495835012441e-06, + "loss": 0.6988, + "step": 1293 + }, + { + "epoch": 2.028183716075157, + "grad_norm": 1.0808806419372559, + "learning_rate": 3.837716529702498e-06, + "loss": 0.6374, + "step": 1294 + }, + { + "epoch": 2.029749478079332, + "grad_norm": 1.0822157859802246, + "learning_rate": 3.835936274417927e-06, + "loss": 0.6249, + "step": 1295 + }, + { + "epoch": 2.0313152400835075, + "grad_norm": 1.0366182327270508, + "learning_rate": 3.834155070422972e-06, + "loss": 0.702, + "step": 1296 + }, + { + "epoch": 2.0328810020876826, + "grad_norm": 1.0268456935882568, + "learning_rate": 3.832372918982547e-06, + "loss": 0.6809, + "step": 1297 + }, + { + "epoch": 2.034446764091858, + "grad_norm": 1.1484824419021606, + "learning_rate": 3.830589821362242e-06, + "loss": 0.6442, + "step": 1298 + }, + { + "epoch": 2.0360125260960333, + "grad_norm": 1.0607216358184814, + "learning_rate": 3.828805778828315e-06, + "loss": 0.6469, + "step": 1299 + }, + { + "epoch": 2.037578288100209, + "grad_norm": 1.0950021743774414, + "learning_rate": 3.8270207926477e-06, + "loss": 0.6756, + "step": 1300 + }, + { + "epoch": 2.039144050104384, + "grad_norm": 0.9932938814163208, + "learning_rate": 3.825234864087998e-06, + "loss": 0.706, + "step": 1301 + }, + { + "epoch": 2.0407098121085596, + "grad_norm": 1.0706391334533691, + "learning_rate": 3.823447994417477e-06, + "loss": 0.6799, + "step": 1302 + }, + { + "epoch": 2.0422755741127347, + "grad_norm": 1.0093334913253784, + "learning_rate": 3.821660184905081e-06, + "loss": 0.676, + "step": 1303 + }, + { + "epoch": 2.0438413361169103, + "grad_norm": 1.0283772945404053, + "learning_rate": 3.819871436820413e-06, + "loss": 0.6738, + "step": 1304 + }, + { + "epoch": 2.0454070981210855, + "grad_norm": 1.0533452033996582, + "learning_rate": 3.818081751433746e-06, + "loss": 0.7053, + "step": 1305 + }, + { + "epoch": 2.046972860125261, + "grad_norm": 1.1219292879104614, + "learning_rate": 3.816291130016021e-06, + "loss": 0.6563, + "step": 1306 + }, + { + "epoch": 2.048538622129436, + "grad_norm": 1.025702953338623, + "learning_rate": 3.8144995738388397e-06, + "loss": 0.7264, + "step": 1307 + }, + { + "epoch": 2.0501043841336117, + "grad_norm": 1.0799325704574585, + "learning_rate": 3.8127070841744695e-06, + "loss": 0.6541, + "step": 1308 + }, + { + "epoch": 2.051670146137787, + "grad_norm": 1.043544888496399, + "learning_rate": 3.810913662295841e-06, + "loss": 0.627, + "step": 1309 + }, + { + "epoch": 2.0532359081419624, + "grad_norm": 0.9250067472457886, + "learning_rate": 3.809119309476547e-06, + "loss": 0.7447, + "step": 1310 + }, + { + "epoch": 2.0548016701461376, + "grad_norm": 1.0885530710220337, + "learning_rate": 3.8073240269908385e-06, + "loss": 0.6509, + "step": 1311 + }, + { + "epoch": 2.056367432150313, + "grad_norm": 1.2108066082000732, + "learning_rate": 3.805527816113631e-06, + "loss": 0.6844, + "step": 1312 + }, + { + "epoch": 2.0579331941544887, + "grad_norm": 1.1016451120376587, + "learning_rate": 3.8037306781204974e-06, + "loss": 0.6942, + "step": 1313 + }, + { + "epoch": 2.059498956158664, + "grad_norm": 1.080930233001709, + "learning_rate": 3.801932614287668e-06, + "loss": 0.7379, + "step": 1314 + }, + { + "epoch": 2.0610647181628394, + "grad_norm": 1.1148338317871094, + "learning_rate": 3.800133625892031e-06, + "loss": 0.6817, + "step": 1315 + }, + { + "epoch": 2.0626304801670146, + "grad_norm": 1.0792765617370605, + "learning_rate": 3.798333714211132e-06, + "loss": 0.6787, + "step": 1316 + }, + { + "epoch": 2.06419624217119, + "grad_norm": 1.219314455986023, + "learning_rate": 3.7965328805231727e-06, + "loss": 0.6451, + "step": 1317 + }, + { + "epoch": 2.0657620041753653, + "grad_norm": 1.0637946128845215, + "learning_rate": 3.7947311261070073e-06, + "loss": 0.6869, + "step": 1318 + }, + { + "epoch": 2.067327766179541, + "grad_norm": 1.112642765045166, + "learning_rate": 3.792928452242146e-06, + "loss": 0.6785, + "step": 1319 + }, + { + "epoch": 2.068893528183716, + "grad_norm": 1.006885051727295, + "learning_rate": 3.791124860208752e-06, + "loss": 0.7289, + "step": 1320 + }, + { + "epoch": 2.0704592901878915, + "grad_norm": 1.1578105688095093, + "learning_rate": 3.7893203512876385e-06, + "loss": 0.6472, + "step": 1321 + }, + { + "epoch": 2.0720250521920667, + "grad_norm": 1.0113657712936401, + "learning_rate": 3.7875149267602725e-06, + "loss": 0.6314, + "step": 1322 + }, + { + "epoch": 2.0735908141962422, + "grad_norm": 1.0363519191741943, + "learning_rate": 3.785708587908767e-06, + "loss": 0.6949, + "step": 1323 + }, + { + "epoch": 2.0751565762004174, + "grad_norm": 1.0536452531814575, + "learning_rate": 3.7839013360158904e-06, + "loss": 0.6817, + "step": 1324 + }, + { + "epoch": 2.076722338204593, + "grad_norm": 1.2700945138931274, + "learning_rate": 3.7820931723650547e-06, + "loss": 0.7364, + "step": 1325 + }, + { + "epoch": 2.078288100208768, + "grad_norm": 0.979896068572998, + "learning_rate": 3.7802840982403204e-06, + "loss": 0.6583, + "step": 1326 + }, + { + "epoch": 2.0798538622129437, + "grad_norm": 1.1849523782730103, + "learning_rate": 3.778474114926396e-06, + "loss": 0.6623, + "step": 1327 + }, + { + "epoch": 2.081419624217119, + "grad_norm": 1.0124396085739136, + "learning_rate": 3.7766632237086336e-06, + "loss": 0.7018, + "step": 1328 + }, + { + "epoch": 2.0829853862212944, + "grad_norm": 1.0735864639282227, + "learning_rate": 3.774851425873032e-06, + "loss": 0.6178, + "step": 1329 + }, + { + "epoch": 2.08455114822547, + "grad_norm": 1.1188064813613892, + "learning_rate": 3.773038722706234e-06, + "loss": 0.6603, + "step": 1330 + }, + { + "epoch": 2.086116910229645, + "grad_norm": 1.1021878719329834, + "learning_rate": 3.771225115495522e-06, + "loss": 0.7027, + "step": 1331 + }, + { + "epoch": 2.0876826722338206, + "grad_norm": 1.067917823791504, + "learning_rate": 3.769410605528824e-06, + "loss": 0.5733, + "step": 1332 + }, + { + "epoch": 2.0892484342379958, + "grad_norm": 1.199495553970337, + "learning_rate": 3.7675951940947075e-06, + "loss": 0.7376, + "step": 1333 + }, + { + "epoch": 2.0908141962421714, + "grad_norm": 1.104844093322754, + "learning_rate": 3.7657788824823805e-06, + "loss": 0.7099, + "step": 1334 + }, + { + "epoch": 2.0923799582463465, + "grad_norm": 1.1797598600387573, + "learning_rate": 3.7639616719816907e-06, + "loss": 0.6264, + "step": 1335 + }, + { + "epoch": 2.093945720250522, + "grad_norm": 1.0947678089141846, + "learning_rate": 3.7621435638831226e-06, + "loss": 0.7066, + "step": 1336 + }, + { + "epoch": 2.095511482254697, + "grad_norm": 1.1054636240005493, + "learning_rate": 3.7603245594778e-06, + "loss": 0.7439, + "step": 1337 + }, + { + "epoch": 2.0970772442588728, + "grad_norm": 1.106891393661499, + "learning_rate": 3.758504660057482e-06, + "loss": 0.6446, + "step": 1338 + }, + { + "epoch": 2.098643006263048, + "grad_norm": 1.043440818786621, + "learning_rate": 3.7566838669145644e-06, + "loss": 0.6885, + "step": 1339 + }, + { + "epoch": 2.1002087682672235, + "grad_norm": 1.0602154731750488, + "learning_rate": 3.7548621813420765e-06, + "loss": 0.6808, + "step": 1340 + }, + { + "epoch": 2.1017745302713986, + "grad_norm": 0.9517614841461182, + "learning_rate": 3.7530396046336814e-06, + "loss": 0.6369, + "step": 1341 + }, + { + "epoch": 2.103340292275574, + "grad_norm": 1.1436609029769897, + "learning_rate": 3.7512161380836763e-06, + "loss": 0.7427, + "step": 1342 + }, + { + "epoch": 2.1049060542797493, + "grad_norm": 1.00765860080719, + "learning_rate": 3.7493917829869887e-06, + "loss": 0.7012, + "step": 1343 + }, + { + "epoch": 2.106471816283925, + "grad_norm": 1.1718883514404297, + "learning_rate": 3.747566540639178e-06, + "loss": 0.6689, + "step": 1344 + }, + { + "epoch": 2.1080375782881, + "grad_norm": 1.0489915609359741, + "learning_rate": 3.745740412336434e-06, + "loss": 0.6735, + "step": 1345 + }, + { + "epoch": 2.1096033402922756, + "grad_norm": 1.1287360191345215, + "learning_rate": 3.743913399375575e-06, + "loss": 0.6211, + "step": 1346 + }, + { + "epoch": 2.1111691022964507, + "grad_norm": 1.0830544233322144, + "learning_rate": 3.7420855030540476e-06, + "loss": 0.7219, + "step": 1347 + }, + { + "epoch": 2.1127348643006263, + "grad_norm": 1.1535146236419678, + "learning_rate": 3.7402567246699257e-06, + "loss": 0.6995, + "step": 1348 + }, + { + "epoch": 2.114300626304802, + "grad_norm": 1.1392090320587158, + "learning_rate": 3.7384270655219102e-06, + "loss": 0.7118, + "step": 1349 + }, + { + "epoch": 2.115866388308977, + "grad_norm": 1.009692668914795, + "learning_rate": 3.736596526909327e-06, + "loss": 0.7211, + "step": 1350 + }, + { + "epoch": 2.1174321503131526, + "grad_norm": 1.2041988372802734, + "learning_rate": 3.7347651101321258e-06, + "loss": 0.5969, + "step": 1351 + }, + { + "epoch": 2.1189979123173277, + "grad_norm": 1.0518646240234375, + "learning_rate": 3.7329328164908818e-06, + "loss": 0.6624, + "step": 1352 + }, + { + "epoch": 2.1205636743215033, + "grad_norm": 1.1757079362869263, + "learning_rate": 3.7310996472867916e-06, + "loss": 0.7201, + "step": 1353 + }, + { + "epoch": 2.1221294363256784, + "grad_norm": 1.0831660032272339, + "learning_rate": 3.7292656038216722e-06, + "loss": 0.6302, + "step": 1354 + }, + { + "epoch": 2.123695198329854, + "grad_norm": 1.1849291324615479, + "learning_rate": 3.7274306873979654e-06, + "loss": 0.7365, + "step": 1355 + }, + { + "epoch": 2.125260960334029, + "grad_norm": 1.1130744218826294, + "learning_rate": 3.72559489931873e-06, + "loss": 0.6759, + "step": 1356 + }, + { + "epoch": 2.1268267223382047, + "grad_norm": 1.0014575719833374, + "learning_rate": 3.7237582408876426e-06, + "loss": 0.7179, + "step": 1357 + }, + { + "epoch": 2.12839248434238, + "grad_norm": 1.1398662328720093, + "learning_rate": 3.7219207134090013e-06, + "loss": 0.6678, + "step": 1358 + }, + { + "epoch": 2.1299582463465554, + "grad_norm": 1.1094143390655518, + "learning_rate": 3.720082318187719e-06, + "loss": 0.6962, + "step": 1359 + }, + { + "epoch": 2.1315240083507305, + "grad_norm": 1.0060160160064697, + "learning_rate": 3.718243056529326e-06, + "loss": 0.688, + "step": 1360 + }, + { + "epoch": 2.133089770354906, + "grad_norm": 1.0628772974014282, + "learning_rate": 3.716402929739967e-06, + "loss": 0.7124, + "step": 1361 + }, + { + "epoch": 2.1346555323590812, + "grad_norm": 1.1098920106887817, + "learning_rate": 3.714561939126402e-06, + "loss": 0.7016, + "step": 1362 + }, + { + "epoch": 2.136221294363257, + "grad_norm": 0.9880667328834534, + "learning_rate": 3.712720085996003e-06, + "loss": 0.7333, + "step": 1363 + }, + { + "epoch": 2.137787056367432, + "grad_norm": 1.1795120239257812, + "learning_rate": 3.710877371656757e-06, + "loss": 0.6729, + "step": 1364 + }, + { + "epoch": 2.1393528183716075, + "grad_norm": 1.1155129671096802, + "learning_rate": 3.7090337974172595e-06, + "loss": 0.6676, + "step": 1365 + }, + { + "epoch": 2.140918580375783, + "grad_norm": 1.1053357124328613, + "learning_rate": 3.707189364586719e-06, + "loss": 0.7043, + "step": 1366 + }, + { + "epoch": 2.142484342379958, + "grad_norm": 1.1484512090682983, + "learning_rate": 3.705344074474952e-06, + "loss": 0.7166, + "step": 1367 + }, + { + "epoch": 2.144050104384134, + "grad_norm": 1.056976318359375, + "learning_rate": 3.703497928392386e-06, + "loss": 0.7034, + "step": 1368 + }, + { + "epoch": 2.145615866388309, + "grad_norm": 1.093001365661621, + "learning_rate": 3.7016509276500546e-06, + "loss": 0.5952, + "step": 1369 + }, + { + "epoch": 2.1471816283924845, + "grad_norm": 1.094588041305542, + "learning_rate": 3.699803073559598e-06, + "loss": 0.7126, + "step": 1370 + }, + { + "epoch": 2.1487473903966596, + "grad_norm": 1.126368522644043, + "learning_rate": 3.6979543674332634e-06, + "loss": 0.714, + "step": 1371 + }, + { + "epoch": 2.150313152400835, + "grad_norm": 1.0853040218353271, + "learning_rate": 3.696104810583904e-06, + "loss": 0.721, + "step": 1372 + }, + { + "epoch": 2.1518789144050103, + "grad_norm": 1.091913104057312, + "learning_rate": 3.694254404324975e-06, + "loss": 0.6752, + "step": 1373 + }, + { + "epoch": 2.153444676409186, + "grad_norm": 1.1095316410064697, + "learning_rate": 3.6924031499705366e-06, + "loss": 0.6988, + "step": 1374 + }, + { + "epoch": 2.155010438413361, + "grad_norm": 1.1327425241470337, + "learning_rate": 3.690551048835249e-06, + "loss": 0.6563, + "step": 1375 + }, + { + "epoch": 2.1565762004175366, + "grad_norm": 1.1065635681152344, + "learning_rate": 3.688698102234377e-06, + "loss": 0.683, + "step": 1376 + }, + { + "epoch": 2.1581419624217117, + "grad_norm": 1.050032138824463, + "learning_rate": 3.6868443114837828e-06, + "loss": 0.6471, + "step": 1377 + }, + { + "epoch": 2.1597077244258873, + "grad_norm": 1.2008466720581055, + "learning_rate": 3.6849896778999293e-06, + "loss": 0.7257, + "step": 1378 + }, + { + "epoch": 2.1612734864300625, + "grad_norm": 1.101426601409912, + "learning_rate": 3.6831342027998796e-06, + "loss": 0.6542, + "step": 1379 + }, + { + "epoch": 2.162839248434238, + "grad_norm": 1.0931123495101929, + "learning_rate": 3.68127788750129e-06, + "loss": 0.665, + "step": 1380 + }, + { + "epoch": 2.164405010438413, + "grad_norm": 1.2418708801269531, + "learning_rate": 3.6794207333224184e-06, + "loss": 0.6399, + "step": 1381 + }, + { + "epoch": 2.1659707724425887, + "grad_norm": 1.1205544471740723, + "learning_rate": 3.6775627415821158e-06, + "loss": 0.7029, + "step": 1382 + }, + { + "epoch": 2.1675365344467643, + "grad_norm": 1.0413703918457031, + "learning_rate": 3.6757039135998276e-06, + "loss": 0.7521, + "step": 1383 + }, + { + "epoch": 2.1691022964509394, + "grad_norm": 1.0559121370315552, + "learning_rate": 3.673844250695594e-06, + "loss": 0.6664, + "step": 1384 + }, + { + "epoch": 2.170668058455115, + "grad_norm": 1.1533021926879883, + "learning_rate": 3.6719837541900495e-06, + "loss": 0.6425, + "step": 1385 + }, + { + "epoch": 2.17223382045929, + "grad_norm": 1.246163010597229, + "learning_rate": 3.6701224254044173e-06, + "loss": 0.6683, + "step": 1386 + }, + { + "epoch": 2.1737995824634657, + "grad_norm": 1.1539034843444824, + "learning_rate": 3.6682602656605136e-06, + "loss": 0.6699, + "step": 1387 + }, + { + "epoch": 2.175365344467641, + "grad_norm": 1.021262526512146, + "learning_rate": 3.6663972762807453e-06, + "loss": 0.6409, + "step": 1388 + }, + { + "epoch": 2.1769311064718164, + "grad_norm": 1.09273099899292, + "learning_rate": 3.664533458588107e-06, + "loss": 0.7315, + "step": 1389 + }, + { + "epoch": 2.1784968684759916, + "grad_norm": 1.169747233390808, + "learning_rate": 3.662668813906183e-06, + "loss": 0.6675, + "step": 1390 + }, + { + "epoch": 2.180062630480167, + "grad_norm": 1.189447045326233, + "learning_rate": 3.6608033435591435e-06, + "loss": 0.7016, + "step": 1391 + }, + { + "epoch": 2.1816283924843423, + "grad_norm": 1.0325034856796265, + "learning_rate": 3.658937048871746e-06, + "loss": 0.74, + "step": 1392 + }, + { + "epoch": 2.183194154488518, + "grad_norm": 1.1599889993667603, + "learning_rate": 3.6570699311693316e-06, + "loss": 0.6525, + "step": 1393 + }, + { + "epoch": 2.184759916492693, + "grad_norm": 1.018955945968628, + "learning_rate": 3.6552019917778294e-06, + "loss": 0.6779, + "step": 1394 + }, + { + "epoch": 2.1863256784968685, + "grad_norm": 1.1707937717437744, + "learning_rate": 3.653333232023748e-06, + "loss": 0.6321, + "step": 1395 + }, + { + "epoch": 2.1878914405010437, + "grad_norm": 1.1663832664489746, + "learning_rate": 3.6514636532341825e-06, + "loss": 0.645, + "step": 1396 + }, + { + "epoch": 2.1894572025052192, + "grad_norm": 1.0475002527236938, + "learning_rate": 3.6495932567368054e-06, + "loss": 0.6209, + "step": 1397 + }, + { + "epoch": 2.1910229645093944, + "grad_norm": 1.052640438079834, + "learning_rate": 3.647722043859874e-06, + "loss": 0.7383, + "step": 1398 + }, + { + "epoch": 2.19258872651357, + "grad_norm": 1.1010091304779053, + "learning_rate": 3.6458500159322226e-06, + "loss": 0.669, + "step": 1399 + }, + { + "epoch": 2.1941544885177455, + "grad_norm": 1.0842148065567017, + "learning_rate": 3.6439771742832645e-06, + "loss": 0.7248, + "step": 1400 + }, + { + "epoch": 2.1957202505219207, + "grad_norm": 1.1516163349151611, + "learning_rate": 3.642103520242993e-06, + "loss": 0.6501, + "step": 1401 + }, + { + "epoch": 2.1972860125260962, + "grad_norm": 1.1424976587295532, + "learning_rate": 3.6402290551419754e-06, + "loss": 0.7191, + "step": 1402 + }, + { + "epoch": 2.1988517745302714, + "grad_norm": 1.132794737815857, + "learning_rate": 3.6383537803113577e-06, + "loss": 0.676, + "step": 1403 + }, + { + "epoch": 2.200417536534447, + "grad_norm": 1.0671226978302002, + "learning_rate": 3.6364776970828586e-06, + "loss": 0.6878, + "step": 1404 + }, + { + "epoch": 2.201983298538622, + "grad_norm": 1.0326734781265259, + "learning_rate": 3.6346008067887724e-06, + "loss": 0.6559, + "step": 1405 + }, + { + "epoch": 2.2035490605427976, + "grad_norm": 1.0672800540924072, + "learning_rate": 3.6327231107619664e-06, + "loss": 0.6826, + "step": 1406 + }, + { + "epoch": 2.2051148225469728, + "grad_norm": 1.0678563117980957, + "learning_rate": 3.6308446103358785e-06, + "loss": 0.6772, + "step": 1407 + }, + { + "epoch": 2.2066805845511483, + "grad_norm": 1.1288729906082153, + "learning_rate": 3.628965306844521e-06, + "loss": 0.7248, + "step": 1408 + }, + { + "epoch": 2.2082463465553235, + "grad_norm": 1.0807738304138184, + "learning_rate": 3.6270852016224735e-06, + "loss": 0.6994, + "step": 1409 + }, + { + "epoch": 2.209812108559499, + "grad_norm": 1.2594057321548462, + "learning_rate": 3.625204296004885e-06, + "loss": 0.6429, + "step": 1410 + }, + { + "epoch": 2.211377870563674, + "grad_norm": 1.159574031829834, + "learning_rate": 3.623322591327476e-06, + "loss": 0.6292, + "step": 1411 + }, + { + "epoch": 2.2129436325678498, + "grad_norm": 1.2090522050857544, + "learning_rate": 3.621440088926531e-06, + "loss": 0.663, + "step": 1412 + }, + { + "epoch": 2.214509394572025, + "grad_norm": 1.176761507987976, + "learning_rate": 3.619556790138903e-06, + "loss": 0.5766, + "step": 1413 + }, + { + "epoch": 2.2160751565762005, + "grad_norm": 1.0924803018569946, + "learning_rate": 3.6176726963020085e-06, + "loss": 0.6838, + "step": 1414 + }, + { + "epoch": 2.2176409185803756, + "grad_norm": 1.0821013450622559, + "learning_rate": 3.6157878087538322e-06, + "loss": 0.6464, + "step": 1415 + }, + { + "epoch": 2.219206680584551, + "grad_norm": 1.0731130838394165, + "learning_rate": 3.613902128832919e-06, + "loss": 0.5345, + "step": 1416 + }, + { + "epoch": 2.2207724425887267, + "grad_norm": 1.0433294773101807, + "learning_rate": 3.612015657878378e-06, + "loss": 0.6799, + "step": 1417 + }, + { + "epoch": 2.222338204592902, + "grad_norm": 1.2107945680618286, + "learning_rate": 3.61012839722988e-06, + "loss": 0.746, + "step": 1418 + }, + { + "epoch": 2.223903966597077, + "grad_norm": 1.2256337404251099, + "learning_rate": 3.608240348227656e-06, + "loss": 0.6438, + "step": 1419 + }, + { + "epoch": 2.2254697286012526, + "grad_norm": 1.0490953922271729, + "learning_rate": 3.6063515122124975e-06, + "loss": 0.6618, + "step": 1420 + }, + { + "epoch": 2.227035490605428, + "grad_norm": 1.134993553161621, + "learning_rate": 3.6044618905257546e-06, + "loss": 0.6669, + "step": 1421 + }, + { + "epoch": 2.2286012526096033, + "grad_norm": 1.120219349861145, + "learning_rate": 3.602571484509336e-06, + "loss": 0.7164, + "step": 1422 + }, + { + "epoch": 2.230167014613779, + "grad_norm": 1.2099931240081787, + "learning_rate": 3.6006802955057057e-06, + "loss": 0.6499, + "step": 1423 + }, + { + "epoch": 2.231732776617954, + "grad_norm": 1.276110291481018, + "learning_rate": 3.5987883248578857e-06, + "loss": 0.7182, + "step": 1424 + }, + { + "epoch": 2.2332985386221296, + "grad_norm": 1.12351655960083, + "learning_rate": 3.5968955739094522e-06, + "loss": 0.6672, + "step": 1425 + }, + { + "epoch": 2.2348643006263047, + "grad_norm": 1.1756179332733154, + "learning_rate": 3.595002044004535e-06, + "loss": 0.6417, + "step": 1426 + }, + { + "epoch": 2.2364300626304803, + "grad_norm": 1.1440640687942505, + "learning_rate": 3.593107736487818e-06, + "loss": 0.6694, + "step": 1427 + }, + { + "epoch": 2.2379958246346554, + "grad_norm": 1.160673975944519, + "learning_rate": 3.5912126527045368e-06, + "loss": 0.7067, + "step": 1428 + }, + { + "epoch": 2.239561586638831, + "grad_norm": 1.1037743091583252, + "learning_rate": 3.589316794000478e-06, + "loss": 0.7258, + "step": 1429 + }, + { + "epoch": 2.241127348643006, + "grad_norm": 1.1246981620788574, + "learning_rate": 3.58742016172198e-06, + "loss": 0.6715, + "step": 1430 + }, + { + "epoch": 2.2426931106471817, + "grad_norm": 1.185558795928955, + "learning_rate": 3.5855227572159283e-06, + "loss": 0.6932, + "step": 1431 + }, + { + "epoch": 2.244258872651357, + "grad_norm": 1.0668050050735474, + "learning_rate": 3.583624581829759e-06, + "loss": 0.6836, + "step": 1432 + }, + { + "epoch": 2.2458246346555324, + "grad_norm": 1.1782646179199219, + "learning_rate": 3.581725636911454e-06, + "loss": 0.6787, + "step": 1433 + }, + { + "epoch": 2.2473903966597075, + "grad_norm": 1.2447540760040283, + "learning_rate": 3.579825923809542e-06, + "loss": 0.6702, + "step": 1434 + }, + { + "epoch": 2.248956158663883, + "grad_norm": 1.029355525970459, + "learning_rate": 3.577925443873098e-06, + "loss": 0.637, + "step": 1435 + }, + { + "epoch": 2.2505219206680582, + "grad_norm": 1.1154545545578003, + "learning_rate": 3.5760241984517397e-06, + "loss": 0.7274, + "step": 1436 + }, + { + "epoch": 2.252087682672234, + "grad_norm": 1.116469144821167, + "learning_rate": 3.5741221888956325e-06, + "loss": 0.6321, + "step": 1437 + }, + { + "epoch": 2.2536534446764094, + "grad_norm": 1.2985416650772095, + "learning_rate": 3.5722194165554795e-06, + "loss": 0.648, + "step": 1438 + }, + { + "epoch": 2.2552192066805845, + "grad_norm": 1.2045787572860718, + "learning_rate": 3.5703158827825277e-06, + "loss": 0.6138, + "step": 1439 + }, + { + "epoch": 2.25678496868476, + "grad_norm": 1.0173790454864502, + "learning_rate": 3.5684115889285663e-06, + "loss": 0.6144, + "step": 1440 + }, + { + "epoch": 2.258350730688935, + "grad_norm": 1.1876413822174072, + "learning_rate": 3.5665065363459215e-06, + "loss": 0.5951, + "step": 1441 + }, + { + "epoch": 2.259916492693111, + "grad_norm": 1.0441265106201172, + "learning_rate": 3.564600726387461e-06, + "loss": 0.6681, + "step": 1442 + }, + { + "epoch": 2.261482254697286, + "grad_norm": 1.2096240520477295, + "learning_rate": 3.5626941604065866e-06, + "loss": 0.7213, + "step": 1443 + }, + { + "epoch": 2.2630480167014615, + "grad_norm": 1.0928987264633179, + "learning_rate": 3.560786839757242e-06, + "loss": 0.6129, + "step": 1444 + }, + { + "epoch": 2.2646137787056366, + "grad_norm": 1.169466495513916, + "learning_rate": 3.558878765793902e-06, + "loss": 0.7212, + "step": 1445 + }, + { + "epoch": 2.266179540709812, + "grad_norm": 1.0599427223205566, + "learning_rate": 3.5569699398715794e-06, + "loss": 0.683, + "step": 1446 + }, + { + "epoch": 2.2677453027139873, + "grad_norm": 1.126380443572998, + "learning_rate": 3.5550603633458202e-06, + "loss": 0.6886, + "step": 1447 + }, + { + "epoch": 2.269311064718163, + "grad_norm": 1.2540115118026733, + "learning_rate": 3.553150037572704e-06, + "loss": 0.6515, + "step": 1448 + }, + { + "epoch": 2.270876826722338, + "grad_norm": 1.1289552450180054, + "learning_rate": 3.55123896390884e-06, + "loss": 0.7067, + "step": 1449 + }, + { + "epoch": 2.2724425887265136, + "grad_norm": 1.0645952224731445, + "learning_rate": 3.549327143711372e-06, + "loss": 0.7059, + "step": 1450 + }, + { + "epoch": 2.2740083507306887, + "grad_norm": 0.9427555799484253, + "learning_rate": 3.5474145783379728e-06, + "loss": 0.6472, + "step": 1451 + }, + { + "epoch": 2.2755741127348643, + "grad_norm": 1.1604477167129517, + "learning_rate": 3.5455012691468417e-06, + "loss": 0.6602, + "step": 1452 + }, + { + "epoch": 2.2771398747390394, + "grad_norm": 1.1867820024490356, + "learning_rate": 3.5435872174967102e-06, + "loss": 0.7162, + "step": 1453 + }, + { + "epoch": 2.278705636743215, + "grad_norm": 1.0727450847625732, + "learning_rate": 3.541672424746834e-06, + "loss": 0.7171, + "step": 1454 + }, + { + "epoch": 2.2802713987473906, + "grad_norm": 1.1541563272476196, + "learning_rate": 3.539756892256997e-06, + "loss": 0.7214, + "step": 1455 + }, + { + "epoch": 2.2818371607515657, + "grad_norm": 1.0292634963989258, + "learning_rate": 3.5378406213875084e-06, + "loss": 0.702, + "step": 1456 + }, + { + "epoch": 2.2834029227557413, + "grad_norm": 1.1958640813827515, + "learning_rate": 3.5359236134991996e-06, + "loss": 0.6456, + "step": 1457 + }, + { + "epoch": 2.2849686847599164, + "grad_norm": 1.191214680671692, + "learning_rate": 3.5340058699534287e-06, + "loss": 0.7149, + "step": 1458 + }, + { + "epoch": 2.286534446764092, + "grad_norm": 1.0933841466903687, + "learning_rate": 3.5320873921120735e-06, + "loss": 0.657, + "step": 1459 + }, + { + "epoch": 2.288100208768267, + "grad_norm": 1.0500751733779907, + "learning_rate": 3.5301681813375343e-06, + "loss": 0.6708, + "step": 1460 + }, + { + "epoch": 2.2896659707724427, + "grad_norm": 1.0765804052352905, + "learning_rate": 3.5282482389927313e-06, + "loss": 0.6528, + "step": 1461 + }, + { + "epoch": 2.291231732776618, + "grad_norm": 1.0991063117980957, + "learning_rate": 3.526327566441105e-06, + "loss": 0.7167, + "step": 1462 + }, + { + "epoch": 2.2927974947807934, + "grad_norm": 1.096111536026001, + "learning_rate": 3.5244061650466155e-06, + "loss": 0.6983, + "step": 1463 + }, + { + "epoch": 2.2943632567849686, + "grad_norm": 1.0495761632919312, + "learning_rate": 3.522484036173737e-06, + "loss": 0.6019, + "step": 1464 + }, + { + "epoch": 2.295929018789144, + "grad_norm": 1.204898476600647, + "learning_rate": 3.520561181187464e-06, + "loss": 0.7198, + "step": 1465 + }, + { + "epoch": 2.2974947807933193, + "grad_norm": 1.0273269414901733, + "learning_rate": 3.518637601453304e-06, + "loss": 0.7077, + "step": 1466 + }, + { + "epoch": 2.299060542797495, + "grad_norm": 1.0621248483657837, + "learning_rate": 3.5167132983372808e-06, + "loss": 0.703, + "step": 1467 + }, + { + "epoch": 2.30062630480167, + "grad_norm": 1.216728687286377, + "learning_rate": 3.5147882732059323e-06, + "loss": 0.5898, + "step": 1468 + }, + { + "epoch": 2.3021920668058455, + "grad_norm": 1.065699815750122, + "learning_rate": 3.5128625274263074e-06, + "loss": 0.611, + "step": 1469 + }, + { + "epoch": 2.3037578288100207, + "grad_norm": 1.1529189348220825, + "learning_rate": 3.5109360623659672e-06, + "loss": 0.6778, + "step": 1470 + }, + { + "epoch": 2.3053235908141962, + "grad_norm": 1.1462815999984741, + "learning_rate": 3.5090088793929842e-06, + "loss": 0.6366, + "step": 1471 + }, + { + "epoch": 2.306889352818372, + "grad_norm": 1.0135819911956787, + "learning_rate": 3.507080979875941e-06, + "loss": 0.6393, + "step": 1472 + }, + { + "epoch": 2.308455114822547, + "grad_norm": 1.112345814704895, + "learning_rate": 3.5051523651839288e-06, + "loss": 0.595, + "step": 1473 + }, + { + "epoch": 2.3100208768267225, + "grad_norm": 1.0799155235290527, + "learning_rate": 3.503223036686545e-06, + "loss": 0.7024, + "step": 1474 + }, + { + "epoch": 2.3115866388308977, + "grad_norm": 1.2584905624389648, + "learning_rate": 3.501292995753897e-06, + "loss": 0.6308, + "step": 1475 + }, + { + "epoch": 2.3131524008350732, + "grad_norm": 1.0637624263763428, + "learning_rate": 3.4993622437565955e-06, + "loss": 0.6502, + "step": 1476 + }, + { + "epoch": 2.3147181628392484, + "grad_norm": 1.2071688175201416, + "learning_rate": 3.4974307820657584e-06, + "loss": 0.6389, + "step": 1477 + }, + { + "epoch": 2.316283924843424, + "grad_norm": 1.2328526973724365, + "learning_rate": 3.4954986120530042e-06, + "loss": 0.6748, + "step": 1478 + }, + { + "epoch": 2.317849686847599, + "grad_norm": 1.1051819324493408, + "learning_rate": 3.493565735090458e-06, + "loss": 0.6675, + "step": 1479 + }, + { + "epoch": 2.3194154488517746, + "grad_norm": 1.1707861423492432, + "learning_rate": 3.491632152550746e-06, + "loss": 0.6305, + "step": 1480 + }, + { + "epoch": 2.3209812108559498, + "grad_norm": 1.0759583711624146, + "learning_rate": 3.4896978658069937e-06, + "loss": 0.683, + "step": 1481 + }, + { + "epoch": 2.3225469728601253, + "grad_norm": 1.1699646711349487, + "learning_rate": 3.487762876232828e-06, + "loss": 0.7231, + "step": 1482 + }, + { + "epoch": 2.3241127348643005, + "grad_norm": 1.0097750425338745, + "learning_rate": 3.4858271852023756e-06, + "loss": 0.5897, + "step": 1483 + }, + { + "epoch": 2.325678496868476, + "grad_norm": 1.0517746210098267, + "learning_rate": 3.4838907940902607e-06, + "loss": 0.6386, + "step": 1484 + }, + { + "epoch": 2.327244258872651, + "grad_norm": 1.2001336812973022, + "learning_rate": 3.4819537042716033e-06, + "loss": 0.6854, + "step": 1485 + }, + { + "epoch": 2.3288100208768268, + "grad_norm": 1.1421329975128174, + "learning_rate": 3.480015917122023e-06, + "loss": 0.7186, + "step": 1486 + }, + { + "epoch": 2.330375782881002, + "grad_norm": 1.0638641119003296, + "learning_rate": 3.47807743401763e-06, + "loss": 0.6848, + "step": 1487 + }, + { + "epoch": 2.3319415448851775, + "grad_norm": 1.019140601158142, + "learning_rate": 3.476138256335033e-06, + "loss": 0.6904, + "step": 1488 + }, + { + "epoch": 2.333507306889353, + "grad_norm": 1.2467342615127563, + "learning_rate": 3.4741983854513306e-06, + "loss": 0.753, + "step": 1489 + }, + { + "epoch": 2.335073068893528, + "grad_norm": 1.1455196142196655, + "learning_rate": 3.472257822744116e-06, + "loss": 0.6544, + "step": 1490 + }, + { + "epoch": 2.3366388308977033, + "grad_norm": 1.1508904695510864, + "learning_rate": 3.4703165695914737e-06, + "loss": 0.6216, + "step": 1491 + }, + { + "epoch": 2.338204592901879, + "grad_norm": 1.1695144176483154, + "learning_rate": 3.4683746273719754e-06, + "loss": 0.6478, + "step": 1492 + }, + { + "epoch": 2.3397703549060545, + "grad_norm": 1.041002631187439, + "learning_rate": 3.4664319974646872e-06, + "loss": 0.7142, + "step": 1493 + }, + { + "epoch": 2.3413361169102296, + "grad_norm": 1.0497469902038574, + "learning_rate": 3.4644886812491598e-06, + "loss": 0.7028, + "step": 1494 + }, + { + "epoch": 2.342901878914405, + "grad_norm": 1.0564930438995361, + "learning_rate": 3.4625446801054312e-06, + "loss": 0.6796, + "step": 1495 + }, + { + "epoch": 2.3444676409185803, + "grad_norm": 1.2151904106140137, + "learning_rate": 3.460599995414028e-06, + "loss": 0.6933, + "step": 1496 + }, + { + "epoch": 2.346033402922756, + "grad_norm": 1.197263240814209, + "learning_rate": 3.458654628555961e-06, + "loss": 0.7615, + "step": 1497 + }, + { + "epoch": 2.347599164926931, + "grad_norm": 1.037292718887329, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.7014, + "step": 1498 + }, + { + "epoch": 2.3491649269311066, + "grad_norm": 1.0831308364868164, + "learning_rate": 3.454761853866299e-06, + "loss": 0.6682, + "step": 1499 + }, + { + "epoch": 2.3507306889352817, + "grad_norm": 1.0729385614395142, + "learning_rate": 3.4528144487991448e-06, + "loss": 0.725, + "step": 1500 + }, + { + "epoch": 2.3522964509394573, + "grad_norm": 1.010629415512085, + "learning_rate": 3.4508663670942044e-06, + "loss": 0.6579, + "step": 1501 + }, + { + "epoch": 2.3538622129436324, + "grad_norm": 1.0924932956695557, + "learning_rate": 3.448917610134902e-06, + "loss": 0.6892, + "step": 1502 + }, + { + "epoch": 2.355427974947808, + "grad_norm": 1.1776976585388184, + "learning_rate": 3.4469681793051403e-06, + "loss": 0.6366, + "step": 1503 + }, + { + "epoch": 2.356993736951983, + "grad_norm": 1.062983751296997, + "learning_rate": 3.4450180759892994e-06, + "loss": 0.6518, + "step": 1504 + }, + { + "epoch": 2.3585594989561587, + "grad_norm": 1.1387819051742554, + "learning_rate": 3.4430673015722393e-06, + "loss": 0.623, + "step": 1505 + }, + { + "epoch": 2.3601252609603343, + "grad_norm": 1.035445213317871, + "learning_rate": 3.4411158574392956e-06, + "loss": 0.6876, + "step": 1506 + }, + { + "epoch": 2.3616910229645094, + "grad_norm": 1.0623654127120972, + "learning_rate": 3.439163744976279e-06, + "loss": 0.6795, + "step": 1507 + }, + { + "epoch": 2.3632567849686845, + "grad_norm": 1.0046356916427612, + "learning_rate": 3.437210965569475e-06, + "loss": 0.7043, + "step": 1508 + }, + { + "epoch": 2.36482254697286, + "grad_norm": 1.183174729347229, + "learning_rate": 3.4352575206056426e-06, + "loss": 0.7059, + "step": 1509 + }, + { + "epoch": 2.3663883089770357, + "grad_norm": 1.0844060182571411, + "learning_rate": 3.4333034114720155e-06, + "loss": 0.6811, + "step": 1510 + }, + { + "epoch": 2.367954070981211, + "grad_norm": 1.058577299118042, + "learning_rate": 3.431348639556296e-06, + "loss": 0.7258, + "step": 1511 + }, + { + "epoch": 2.3695198329853864, + "grad_norm": 1.0441113710403442, + "learning_rate": 3.429393206246659e-06, + "loss": 0.6659, + "step": 1512 + }, + { + "epoch": 2.3710855949895615, + "grad_norm": 1.111039400100708, + "learning_rate": 3.4274371129317476e-06, + "loss": 0.6113, + "step": 1513 + }, + { + "epoch": 2.372651356993737, + "grad_norm": 1.1454105377197266, + "learning_rate": 3.425480361000675e-06, + "loss": 0.727, + "step": 1514 + }, + { + "epoch": 2.374217118997912, + "grad_norm": 1.1241672039031982, + "learning_rate": 3.4235229518430212e-06, + "loss": 0.5266, + "step": 1515 + }, + { + "epoch": 2.375782881002088, + "grad_norm": 1.0866069793701172, + "learning_rate": 3.421564886848835e-06, + "loss": 0.6871, + "step": 1516 + }, + { + "epoch": 2.377348643006263, + "grad_norm": 1.211893081665039, + "learning_rate": 3.4196061674086276e-06, + "loss": 0.7064, + "step": 1517 + }, + { + "epoch": 2.3789144050104385, + "grad_norm": 1.1039608716964722, + "learning_rate": 3.417646794913376e-06, + "loss": 0.6821, + "step": 1518 + }, + { + "epoch": 2.3804801670146136, + "grad_norm": 1.0984389781951904, + "learning_rate": 3.4156867707545237e-06, + "loss": 0.6443, + "step": 1519 + }, + { + "epoch": 2.382045929018789, + "grad_norm": 1.11539888381958, + "learning_rate": 3.413726096323975e-06, + "loss": 0.6896, + "step": 1520 + }, + { + "epoch": 2.3836116910229643, + "grad_norm": 1.1566188335418701, + "learning_rate": 3.4117647730140935e-06, + "loss": 0.699, + "step": 1521 + }, + { + "epoch": 2.38517745302714, + "grad_norm": 1.0620112419128418, + "learning_rate": 3.409802802217707e-06, + "loss": 0.6741, + "step": 1522 + }, + { + "epoch": 2.3867432150313155, + "grad_norm": 1.117667317390442, + "learning_rate": 3.4078401853281022e-06, + "loss": 0.7027, + "step": 1523 + }, + { + "epoch": 2.3883089770354906, + "grad_norm": 1.0997360944747925, + "learning_rate": 3.4058769237390254e-06, + "loss": 0.7249, + "step": 1524 + }, + { + "epoch": 2.3898747390396657, + "grad_norm": 1.1101152896881104, + "learning_rate": 3.403913018844678e-06, + "loss": 0.6794, + "step": 1525 + }, + { + "epoch": 2.3914405010438413, + "grad_norm": 1.1798306703567505, + "learning_rate": 3.4019484720397217e-06, + "loss": 0.6219, + "step": 1526 + }, + { + "epoch": 2.393006263048017, + "grad_norm": 1.0730222463607788, + "learning_rate": 3.3999832847192716e-06, + "loss": 0.6622, + "step": 1527 + }, + { + "epoch": 2.394572025052192, + "grad_norm": 1.0943461656570435, + "learning_rate": 3.3980174582788982e-06, + "loss": 0.6999, + "step": 1528 + }, + { + "epoch": 2.3961377870563676, + "grad_norm": 1.1592621803283691, + "learning_rate": 3.396050994114628e-06, + "loss": 0.6419, + "step": 1529 + }, + { + "epoch": 2.3977035490605427, + "grad_norm": 1.1652766466140747, + "learning_rate": 3.3940838936229357e-06, + "loss": 0.6952, + "step": 1530 + }, + { + "epoch": 2.3992693110647183, + "grad_norm": 1.1424206495285034, + "learning_rate": 3.3921161582007524e-06, + "loss": 0.6521, + "step": 1531 + }, + { + "epoch": 2.4008350730688934, + "grad_norm": 1.0935698747634888, + "learning_rate": 3.3901477892454583e-06, + "loss": 0.6707, + "step": 1532 + }, + { + "epoch": 2.402400835073069, + "grad_norm": 1.2164827585220337, + "learning_rate": 3.3881787881548835e-06, + "loss": 0.6485, + "step": 1533 + }, + { + "epoch": 2.403966597077244, + "grad_norm": 1.0536198616027832, + "learning_rate": 3.386209156327307e-06, + "loss": 0.6997, + "step": 1534 + }, + { + "epoch": 2.4055323590814197, + "grad_norm": 1.1083351373672485, + "learning_rate": 3.3842388951614545e-06, + "loss": 0.6264, + "step": 1535 + }, + { + "epoch": 2.407098121085595, + "grad_norm": 1.133180856704712, + "learning_rate": 3.3822680060565026e-06, + "loss": 0.6905, + "step": 1536 + }, + { + "epoch": 2.4086638830897704, + "grad_norm": 1.0531203746795654, + "learning_rate": 3.3802964904120703e-06, + "loss": 0.6492, + "step": 1537 + }, + { + "epoch": 2.4102296450939455, + "grad_norm": 1.0389595031738281, + "learning_rate": 3.3783243496282215e-06, + "loss": 0.6506, + "step": 1538 + }, + { + "epoch": 2.411795407098121, + "grad_norm": 1.1839019060134888, + "learning_rate": 3.376351585105465e-06, + "loss": 0.6933, + "step": 1539 + }, + { + "epoch": 2.4133611691022967, + "grad_norm": 1.0715898275375366, + "learning_rate": 3.3743781982447533e-06, + "loss": 0.7043, + "step": 1540 + }, + { + "epoch": 2.414926931106472, + "grad_norm": 1.1898013353347778, + "learning_rate": 3.3724041904474793e-06, + "loss": 0.7054, + "step": 1541 + }, + { + "epoch": 2.416492693110647, + "grad_norm": 1.026489019393921, + "learning_rate": 3.370429563115478e-06, + "loss": 0.6458, + "step": 1542 + }, + { + "epoch": 2.4180584551148225, + "grad_norm": 1.2772443294525146, + "learning_rate": 3.368454317651024e-06, + "loss": 0.6647, + "step": 1543 + }, + { + "epoch": 2.419624217118998, + "grad_norm": 1.0573022365570068, + "learning_rate": 3.366478455456831e-06, + "loss": 0.6723, + "step": 1544 + }, + { + "epoch": 2.4211899791231732, + "grad_norm": 1.2983055114746094, + "learning_rate": 3.3645019779360493e-06, + "loss": 0.6922, + "step": 1545 + }, + { + "epoch": 2.422755741127349, + "grad_norm": 1.0773214101791382, + "learning_rate": 3.36252488649227e-06, + "loss": 0.6538, + "step": 1546 + }, + { + "epoch": 2.424321503131524, + "grad_norm": 1.1467341184616089, + "learning_rate": 3.3605471825295143e-06, + "loss": 0.7274, + "step": 1547 + }, + { + "epoch": 2.4258872651356995, + "grad_norm": 1.1962897777557373, + "learning_rate": 3.3585688674522438e-06, + "loss": 0.7113, + "step": 1548 + }, + { + "epoch": 2.4274530271398747, + "grad_norm": 0.997977614402771, + "learning_rate": 3.3565899426653515e-06, + "loss": 0.6731, + "step": 1549 + }, + { + "epoch": 2.4290187891440502, + "grad_norm": 1.095198392868042, + "learning_rate": 3.3546104095741634e-06, + "loss": 0.6353, + "step": 1550 + }, + { + "epoch": 2.4305845511482254, + "grad_norm": 1.0401874780654907, + "learning_rate": 3.3526302695844382e-06, + "loss": 0.7459, + "step": 1551 + }, + { + "epoch": 2.432150313152401, + "grad_norm": 1.174526333808899, + "learning_rate": 3.3506495241023646e-06, + "loss": 0.6435, + "step": 1552 + }, + { + "epoch": 2.433716075156576, + "grad_norm": 1.0898199081420898, + "learning_rate": 3.3486681745345627e-06, + "loss": 0.6991, + "step": 1553 + }, + { + "epoch": 2.4352818371607516, + "grad_norm": 1.0757240056991577, + "learning_rate": 3.346686222288081e-06, + "loss": 0.6447, + "step": 1554 + }, + { + "epoch": 2.4368475991649268, + "grad_norm": 1.0817614793777466, + "learning_rate": 3.3447036687703956e-06, + "loss": 0.6546, + "step": 1555 + }, + { + "epoch": 2.4384133611691023, + "grad_norm": 1.181427240371704, + "learning_rate": 3.3427205153894088e-06, + "loss": 0.6592, + "step": 1556 + }, + { + "epoch": 2.4399791231732775, + "grad_norm": 1.0263524055480957, + "learning_rate": 3.340736763553451e-06, + "loss": 0.6766, + "step": 1557 + }, + { + "epoch": 2.441544885177453, + "grad_norm": 1.0632904767990112, + "learning_rate": 3.3387524146712762e-06, + "loss": 0.6368, + "step": 1558 + }, + { + "epoch": 2.443110647181628, + "grad_norm": 1.165933609008789, + "learning_rate": 3.3367674701520627e-06, + "loss": 0.692, + "step": 1559 + }, + { + "epoch": 2.4446764091858038, + "grad_norm": 1.109712839126587, + "learning_rate": 3.3347819314054124e-06, + "loss": 0.6721, + "step": 1560 + }, + { + "epoch": 2.4462421711899793, + "grad_norm": 1.0221387147903442, + "learning_rate": 3.332795799841347e-06, + "loss": 0.6547, + "step": 1561 + }, + { + "epoch": 2.4478079331941545, + "grad_norm": 1.1391754150390625, + "learning_rate": 3.3308090768703118e-06, + "loss": 0.68, + "step": 1562 + }, + { + "epoch": 2.44937369519833, + "grad_norm": 1.0794706344604492, + "learning_rate": 3.328821763903172e-06, + "loss": 0.6531, + "step": 1563 + }, + { + "epoch": 2.450939457202505, + "grad_norm": 1.0358234643936157, + "learning_rate": 3.3268338623512094e-06, + "loss": 0.6638, + "step": 1564 + }, + { + "epoch": 2.4525052192066807, + "grad_norm": 1.1351442337036133, + "learning_rate": 3.3248453736261247e-06, + "loss": 0.6356, + "step": 1565 + }, + { + "epoch": 2.454070981210856, + "grad_norm": 1.010600209236145, + "learning_rate": 3.322856299140038e-06, + "loss": 0.6722, + "step": 1566 + }, + { + "epoch": 2.4556367432150314, + "grad_norm": 1.216776728630066, + "learning_rate": 3.320866640305482e-06, + "loss": 0.658, + "step": 1567 + }, + { + "epoch": 2.4572025052192066, + "grad_norm": 1.0781611204147339, + "learning_rate": 3.3188763985354055e-06, + "loss": 0.711, + "step": 1568 + }, + { + "epoch": 2.458768267223382, + "grad_norm": 1.1265203952789307, + "learning_rate": 3.3168855752431732e-06, + "loss": 0.6746, + "step": 1569 + }, + { + "epoch": 2.4603340292275573, + "grad_norm": 1.1726053953170776, + "learning_rate": 3.3148941718425606e-06, + "loss": 0.6836, + "step": 1570 + }, + { + "epoch": 2.461899791231733, + "grad_norm": 1.1113516092300415, + "learning_rate": 3.3129021897477546e-06, + "loss": 0.6322, + "step": 1571 + }, + { + "epoch": 2.463465553235908, + "grad_norm": 1.0234968662261963, + "learning_rate": 3.3109096303733564e-06, + "loss": 0.7179, + "step": 1572 + }, + { + "epoch": 2.4650313152400836, + "grad_norm": 1.1797139644622803, + "learning_rate": 3.308916495134372e-06, + "loss": 0.6996, + "step": 1573 + }, + { + "epoch": 2.4665970772442587, + "grad_norm": 0.9623079895973206, + "learning_rate": 3.3069227854462216e-06, + "loss": 0.6583, + "step": 1574 + }, + { + "epoch": 2.4681628392484343, + "grad_norm": 1.1546630859375, + "learning_rate": 3.30492850272473e-06, + "loss": 0.7457, + "step": 1575 + }, + { + "epoch": 2.4697286012526094, + "grad_norm": 1.0559042692184448, + "learning_rate": 3.302933648386131e-06, + "loss": 0.692, + "step": 1576 + }, + { + "epoch": 2.471294363256785, + "grad_norm": 1.0812047719955444, + "learning_rate": 3.300938223847062e-06, + "loss": 0.6795, + "step": 1577 + }, + { + "epoch": 2.4728601252609606, + "grad_norm": 1.0247067213058472, + "learning_rate": 3.298942230524567e-06, + "loss": 0.695, + "step": 1578 + }, + { + "epoch": 2.4744258872651357, + "grad_norm": 1.354367971420288, + "learning_rate": 3.296945669836094e-06, + "loss": 0.7014, + "step": 1579 + }, + { + "epoch": 2.4759916492693113, + "grad_norm": 1.0897941589355469, + "learning_rate": 3.2949485431994945e-06, + "loss": 0.6338, + "step": 1580 + }, + { + "epoch": 2.4775574112734864, + "grad_norm": 0.9883112907409668, + "learning_rate": 3.292950852033018e-06, + "loss": 0.7102, + "step": 1581 + }, + { + "epoch": 2.479123173277662, + "grad_norm": 1.1789237260818481, + "learning_rate": 3.290952597755319e-06, + "loss": 0.6802, + "step": 1582 + }, + { + "epoch": 2.480688935281837, + "grad_norm": 1.1877659559249878, + "learning_rate": 3.288953781785452e-06, + "loss": 0.6846, + "step": 1583 + }, + { + "epoch": 2.4822546972860127, + "grad_norm": 1.0334861278533936, + "learning_rate": 3.2869544055428678e-06, + "loss": 0.7138, + "step": 1584 + }, + { + "epoch": 2.483820459290188, + "grad_norm": 1.2335622310638428, + "learning_rate": 3.284954470447416e-06, + "loss": 0.6125, + "step": 1585 + }, + { + "epoch": 2.4853862212943634, + "grad_norm": 1.1405894756317139, + "learning_rate": 3.282953977919345e-06, + "loss": 0.6801, + "step": 1586 + }, + { + "epoch": 2.4869519832985385, + "grad_norm": 1.1099573373794556, + "learning_rate": 3.2809529293792953e-06, + "loss": 0.7071, + "step": 1587 + }, + { + "epoch": 2.488517745302714, + "grad_norm": 1.1282474994659424, + "learning_rate": 3.2789513262483053e-06, + "loss": 0.7082, + "step": 1588 + }, + { + "epoch": 2.490083507306889, + "grad_norm": 1.1693743467330933, + "learning_rate": 3.2769491699478077e-06, + "loss": 0.7137, + "step": 1589 + }, + { + "epoch": 2.491649269311065, + "grad_norm": 1.1324049234390259, + "learning_rate": 3.2749464618996244e-06, + "loss": 0.6402, + "step": 1590 + }, + { + "epoch": 2.49321503131524, + "grad_norm": 1.1645187139511108, + "learning_rate": 3.2729432035259733e-06, + "loss": 0.6941, + "step": 1591 + }, + { + "epoch": 2.4947807933194155, + "grad_norm": 1.1647381782531738, + "learning_rate": 3.2709393962494597e-06, + "loss": 0.6583, + "step": 1592 + }, + { + "epoch": 2.4963465553235906, + "grad_norm": 1.1749935150146484, + "learning_rate": 3.268935041493081e-06, + "loss": 0.7099, + "step": 1593 + }, + { + "epoch": 2.497912317327766, + "grad_norm": 1.1859298944473267, + "learning_rate": 3.266930140680222e-06, + "loss": 0.6551, + "step": 1594 + }, + { + "epoch": 2.4994780793319418, + "grad_norm": 1.1642869710922241, + "learning_rate": 3.264924695234657e-06, + "loss": 0.6712, + "step": 1595 + }, + { + "epoch": 2.501043841336117, + "grad_norm": 1.1522237062454224, + "learning_rate": 3.2629187065805445e-06, + "loss": 0.6773, + "step": 1596 + }, + { + "epoch": 2.502609603340292, + "grad_norm": 1.142119288444519, + "learning_rate": 3.2609121761424312e-06, + "loss": 0.6077, + "step": 1597 + }, + { + "epoch": 2.5041753653444676, + "grad_norm": 1.137640357017517, + "learning_rate": 3.258905105345248e-06, + "loss": 0.7086, + "step": 1598 + }, + { + "epoch": 2.505741127348643, + "grad_norm": 1.1106284856796265, + "learning_rate": 3.2568974956143073e-06, + "loss": 0.6795, + "step": 1599 + }, + { + "epoch": 2.5073068893528183, + "grad_norm": 1.0668182373046875, + "learning_rate": 3.254889348375307e-06, + "loss": 0.7163, + "step": 1600 + }, + { + "epoch": 2.508872651356994, + "grad_norm": 1.2222528457641602, + "learning_rate": 3.2528806650543264e-06, + "loss": 0.6589, + "step": 1601 + }, + { + "epoch": 2.510438413361169, + "grad_norm": 1.0996103286743164, + "learning_rate": 3.250871447077824e-06, + "loss": 0.6097, + "step": 1602 + }, + { + "epoch": 2.5120041753653446, + "grad_norm": 1.2791520357131958, + "learning_rate": 3.248861695872639e-06, + "loss": 0.6367, + "step": 1603 + }, + { + "epoch": 2.5135699373695197, + "grad_norm": 1.1636749505996704, + "learning_rate": 3.2468514128659884e-06, + "loss": 0.6271, + "step": 1604 + }, + { + "epoch": 2.5151356993736953, + "grad_norm": 1.1000317335128784, + "learning_rate": 3.2448405994854683e-06, + "loss": 0.6746, + "step": 1605 + }, + { + "epoch": 2.5167014613778704, + "grad_norm": 1.163297414779663, + "learning_rate": 3.2428292571590514e-06, + "loss": 0.6248, + "step": 1606 + }, + { + "epoch": 2.518267223382046, + "grad_norm": 1.0616698265075684, + "learning_rate": 3.240817387315083e-06, + "loss": 0.6249, + "step": 1607 + }, + { + "epoch": 2.519832985386221, + "grad_norm": 1.2272953987121582, + "learning_rate": 3.2388049913822874e-06, + "loss": 0.6856, + "step": 1608 + }, + { + "epoch": 2.5213987473903967, + "grad_norm": 1.0715241432189941, + "learning_rate": 3.2367920707897594e-06, + "loss": 0.6348, + "step": 1609 + }, + { + "epoch": 2.522964509394572, + "grad_norm": 1.069440245628357, + "learning_rate": 3.234778626966968e-06, + "loss": 0.698, + "step": 1610 + }, + { + "epoch": 2.5245302713987474, + "grad_norm": 1.0824602842330933, + "learning_rate": 3.2327646613437525e-06, + "loss": 0.6809, + "step": 1611 + }, + { + "epoch": 2.526096033402923, + "grad_norm": 1.1153802871704102, + "learning_rate": 3.230750175350324e-06, + "loss": 0.6562, + "step": 1612 + }, + { + "epoch": 2.527661795407098, + "grad_norm": 1.1219229698181152, + "learning_rate": 3.228735170417261e-06, + "loss": 0.6819, + "step": 1613 + }, + { + "epoch": 2.5292275574112733, + "grad_norm": 0.9705930352210999, + "learning_rate": 3.226719647975514e-06, + "loss": 0.7457, + "step": 1614 + }, + { + "epoch": 2.530793319415449, + "grad_norm": 1.0846235752105713, + "learning_rate": 3.2247036094563992e-06, + "loss": 0.6122, + "step": 1615 + }, + { + "epoch": 2.5323590814196244, + "grad_norm": 1.0858566761016846, + "learning_rate": 3.222687056291598e-06, + "loss": 0.6141, + "step": 1616 + }, + { + "epoch": 2.5339248434237995, + "grad_norm": 1.0649619102478027, + "learning_rate": 3.2206699899131576e-06, + "loss": 0.6813, + "step": 1617 + }, + { + "epoch": 2.535490605427975, + "grad_norm": 1.2037075757980347, + "learning_rate": 3.2186524117534935e-06, + "loss": 0.7131, + "step": 1618 + }, + { + "epoch": 2.5370563674321502, + "grad_norm": 1.217882513999939, + "learning_rate": 3.216634323245379e-06, + "loss": 0.6977, + "step": 1619 + }, + { + "epoch": 2.538622129436326, + "grad_norm": 1.1430108547210693, + "learning_rate": 3.2146157258219534e-06, + "loss": 0.6799, + "step": 1620 + }, + { + "epoch": 2.540187891440501, + "grad_norm": 1.1914441585540771, + "learning_rate": 3.2125966209167166e-06, + "loss": 0.7284, + "step": 1621 + }, + { + "epoch": 2.5417536534446765, + "grad_norm": 1.1642671823501587, + "learning_rate": 3.210577009963528e-06, + "loss": 0.7203, + "step": 1622 + }, + { + "epoch": 2.5433194154488517, + "grad_norm": 1.0995980501174927, + "learning_rate": 3.208556894396608e-06, + "loss": 0.6635, + "step": 1623 + }, + { + "epoch": 2.5448851774530272, + "grad_norm": 1.1318360567092896, + "learning_rate": 3.2065362756505345e-06, + "loss": 0.6537, + "step": 1624 + }, + { + "epoch": 2.5464509394572024, + "grad_norm": 1.2233246564865112, + "learning_rate": 3.204515155160242e-06, + "loss": 0.7006, + "step": 1625 + }, + { + "epoch": 2.548016701461378, + "grad_norm": 0.9861915707588196, + "learning_rate": 3.202493534361022e-06, + "loss": 0.6647, + "step": 1626 + }, + { + "epoch": 2.549582463465553, + "grad_norm": 1.2189120054244995, + "learning_rate": 3.2004714146885225e-06, + "loss": 0.6374, + "step": 1627 + }, + { + "epoch": 2.5511482254697286, + "grad_norm": 1.165047526359558, + "learning_rate": 3.1984487975787433e-06, + "loss": 0.6287, + "step": 1628 + }, + { + "epoch": 2.552713987473904, + "grad_norm": 1.1922754049301147, + "learning_rate": 3.196425684468039e-06, + "loss": 0.7103, + "step": 1629 + }, + { + "epoch": 2.5542797494780793, + "grad_norm": 1.1970739364624023, + "learning_rate": 3.194402076793116e-06, + "loss": 0.6951, + "step": 1630 + }, + { + "epoch": 2.5558455114822545, + "grad_norm": 1.1461929082870483, + "learning_rate": 3.1923779759910327e-06, + "loss": 0.6989, + "step": 1631 + }, + { + "epoch": 2.55741127348643, + "grad_norm": 1.1536473035812378, + "learning_rate": 3.190353383499198e-06, + "loss": 0.6433, + "step": 1632 + }, + { + "epoch": 2.5589770354906056, + "grad_norm": 1.3109079599380493, + "learning_rate": 3.188328300755367e-06, + "loss": 0.6873, + "step": 1633 + }, + { + "epoch": 2.5605427974947808, + "grad_norm": 1.1532987356185913, + "learning_rate": 3.186302729197645e-06, + "loss": 0.6818, + "step": 1634 + }, + { + "epoch": 2.562108559498956, + "grad_norm": 1.1802146434783936, + "learning_rate": 3.184276670264486e-06, + "loss": 0.7233, + "step": 1635 + }, + { + "epoch": 2.5636743215031315, + "grad_norm": 1.1282404661178589, + "learning_rate": 3.1822501253946875e-06, + "loss": 0.7203, + "step": 1636 + }, + { + "epoch": 2.565240083507307, + "grad_norm": 1.0870387554168701, + "learning_rate": 3.180223096027394e-06, + "loss": 0.6472, + "step": 1637 + }, + { + "epoch": 2.566805845511482, + "grad_norm": 1.038499355316162, + "learning_rate": 3.178195583602092e-06, + "loss": 0.7197, + "step": 1638 + }, + { + "epoch": 2.5683716075156577, + "grad_norm": 1.0605285167694092, + "learning_rate": 3.176167589558613e-06, + "loss": 0.7222, + "step": 1639 + }, + { + "epoch": 2.569937369519833, + "grad_norm": 1.1737507581710815, + "learning_rate": 3.174139115337129e-06, + "loss": 0.6215, + "step": 1640 + }, + { + "epoch": 2.5715031315240084, + "grad_norm": 1.0304044485092163, + "learning_rate": 3.1721101623781556e-06, + "loss": 0.7113, + "step": 1641 + }, + { + "epoch": 2.5730688935281836, + "grad_norm": 1.1509263515472412, + "learning_rate": 3.170080732122545e-06, + "loss": 0.6521, + "step": 1642 + }, + { + "epoch": 2.574634655532359, + "grad_norm": 1.2465925216674805, + "learning_rate": 3.1680508260114882e-06, + "loss": 0.6921, + "step": 1643 + }, + { + "epoch": 2.5762004175365343, + "grad_norm": 1.0584062337875366, + "learning_rate": 3.1660204454865203e-06, + "loss": 0.7041, + "step": 1644 + }, + { + "epoch": 2.57776617954071, + "grad_norm": 1.1217284202575684, + "learning_rate": 3.163989591989505e-06, + "loss": 0.7177, + "step": 1645 + }, + { + "epoch": 2.5793319415448854, + "grad_norm": 1.1175196170806885, + "learning_rate": 3.1619582669626457e-06, + "loss": 0.7192, + "step": 1646 + }, + { + "epoch": 2.5808977035490606, + "grad_norm": 1.1421573162078857, + "learning_rate": 3.1599264718484835e-06, + "loss": 0.7199, + "step": 1647 + }, + { + "epoch": 2.5824634655532357, + "grad_norm": 1.0428645610809326, + "learning_rate": 3.1578942080898876e-06, + "loss": 0.6496, + "step": 1648 + }, + { + "epoch": 2.5840292275574113, + "grad_norm": 0.9507089257240295, + "learning_rate": 3.1558614771300645e-06, + "loss": 0.6271, + "step": 1649 + }, + { + "epoch": 2.585594989561587, + "grad_norm": 1.2682905197143555, + "learning_rate": 3.1538282804125507e-06, + "loss": 0.6635, + "step": 1650 + }, + { + "epoch": 2.587160751565762, + "grad_norm": 1.098144769668579, + "learning_rate": 3.1517946193812132e-06, + "loss": 0.6631, + "step": 1651 + }, + { + "epoch": 2.588726513569937, + "grad_norm": 1.1256821155548096, + "learning_rate": 3.1497604954802485e-06, + "loss": 0.5353, + "step": 1652 + }, + { + "epoch": 2.5902922755741127, + "grad_norm": 1.1058018207550049, + "learning_rate": 3.1477259101541834e-06, + "loss": 0.6324, + "step": 1653 + }, + { + "epoch": 2.5918580375782883, + "grad_norm": 1.200184941291809, + "learning_rate": 3.145690864847871e-06, + "loss": 0.6484, + "step": 1654 + }, + { + "epoch": 2.5934237995824634, + "grad_norm": 1.0515025854110718, + "learning_rate": 3.1436553610064923e-06, + "loss": 0.6774, + "step": 1655 + }, + { + "epoch": 2.594989561586639, + "grad_norm": 1.0765128135681152, + "learning_rate": 3.1416194000755517e-06, + "loss": 0.6838, + "step": 1656 + }, + { + "epoch": 2.596555323590814, + "grad_norm": 1.157872200012207, + "learning_rate": 3.139582983500881e-06, + "loss": 0.7306, + "step": 1657 + }, + { + "epoch": 2.5981210855949897, + "grad_norm": 1.056947946548462, + "learning_rate": 3.1375461127286334e-06, + "loss": 0.628, + "step": 1658 + }, + { + "epoch": 2.599686847599165, + "grad_norm": 1.1467186212539673, + "learning_rate": 3.135508789205285e-06, + "loss": 0.6829, + "step": 1659 + }, + { + "epoch": 2.6012526096033404, + "grad_norm": 1.0062966346740723, + "learning_rate": 3.1334710143776346e-06, + "loss": 0.654, + "step": 1660 + }, + { + "epoch": 2.6028183716075155, + "grad_norm": 1.1658058166503906, + "learning_rate": 3.1314327896928006e-06, + "loss": 0.7067, + "step": 1661 + }, + { + "epoch": 2.604384133611691, + "grad_norm": 1.2473435401916504, + "learning_rate": 3.12939411659822e-06, + "loss": 0.668, + "step": 1662 + }, + { + "epoch": 2.6059498956158667, + "grad_norm": 1.171295404434204, + "learning_rate": 3.127354996541651e-06, + "loss": 0.6847, + "step": 1663 + }, + { + "epoch": 2.607515657620042, + "grad_norm": 1.118043303489685, + "learning_rate": 3.1253154309711663e-06, + "loss": 0.6948, + "step": 1664 + }, + { + "epoch": 2.609081419624217, + "grad_norm": 1.0427628755569458, + "learning_rate": 3.123275421335157e-06, + "loss": 0.7186, + "step": 1665 + }, + { + "epoch": 2.6106471816283925, + "grad_norm": 1.231054425239563, + "learning_rate": 3.1212349690823273e-06, + "loss": 0.6574, + "step": 1666 + }, + { + "epoch": 2.612212943632568, + "grad_norm": 1.0482161045074463, + "learning_rate": 3.1191940756616983e-06, + "loss": 0.7376, + "step": 1667 + }, + { + "epoch": 2.613778705636743, + "grad_norm": 1.2325738668441772, + "learning_rate": 3.1171527425226027e-06, + "loss": 0.6741, + "step": 1668 + }, + { + "epoch": 2.6153444676409183, + "grad_norm": 1.183379888534546, + "learning_rate": 3.1151109711146854e-06, + "loss": 0.7424, + "step": 1669 + }, + { + "epoch": 2.616910229645094, + "grad_norm": 1.1731956005096436, + "learning_rate": 3.1130687628879047e-06, + "loss": 0.7913, + "step": 1670 + }, + { + "epoch": 2.6184759916492695, + "grad_norm": 1.155126690864563, + "learning_rate": 3.111026119292526e-06, + "loss": 0.6724, + "step": 1671 + }, + { + "epoch": 2.6200417536534446, + "grad_norm": 1.0947262048721313, + "learning_rate": 3.1089830417791255e-06, + "loss": 0.6959, + "step": 1672 + }, + { + "epoch": 2.62160751565762, + "grad_norm": 1.073900580406189, + "learning_rate": 3.1069395317985884e-06, + "loss": 0.7274, + "step": 1673 + }, + { + "epoch": 2.6231732776617953, + "grad_norm": 1.1424334049224854, + "learning_rate": 3.104895590802105e-06, + "loss": 0.657, + "step": 1674 + }, + { + "epoch": 2.624739039665971, + "grad_norm": 1.100588321685791, + "learning_rate": 3.1028512202411736e-06, + "loss": 0.6451, + "step": 1675 + }, + { + "epoch": 2.626304801670146, + "grad_norm": 1.0387758016586304, + "learning_rate": 3.100806421567596e-06, + "loss": 0.7331, + "step": 1676 + }, + { + "epoch": 2.6278705636743216, + "grad_norm": 1.2019600868225098, + "learning_rate": 3.0987611962334784e-06, + "loss": 0.6224, + "step": 1677 + }, + { + "epoch": 2.6294363256784967, + "grad_norm": 1.0992989540100098, + "learning_rate": 3.0967155456912313e-06, + "loss": 0.6557, + "step": 1678 + }, + { + "epoch": 2.6310020876826723, + "grad_norm": 1.259507417678833, + "learning_rate": 3.0946694713935655e-06, + "loss": 0.6668, + "step": 1679 + }, + { + "epoch": 2.632567849686848, + "grad_norm": 1.1520506143569946, + "learning_rate": 3.0926229747934932e-06, + "loss": 0.7199, + "step": 1680 + }, + { + "epoch": 2.634133611691023, + "grad_norm": 1.0905909538269043, + "learning_rate": 3.0905760573443278e-06, + "loss": 0.6927, + "step": 1681 + }, + { + "epoch": 2.635699373695198, + "grad_norm": 1.1551228761672974, + "learning_rate": 3.0885287204996784e-06, + "loss": 0.6872, + "step": 1682 + }, + { + "epoch": 2.6372651356993737, + "grad_norm": 1.177915096282959, + "learning_rate": 3.0864809657134553e-06, + "loss": 0.6102, + "step": 1683 + }, + { + "epoch": 2.6388308977035493, + "grad_norm": 1.2892553806304932, + "learning_rate": 3.084432794439865e-06, + "loss": 0.6576, + "step": 1684 + }, + { + "epoch": 2.6403966597077244, + "grad_norm": 1.1799864768981934, + "learning_rate": 3.0823842081334075e-06, + "loss": 0.7214, + "step": 1685 + }, + { + "epoch": 2.6419624217118995, + "grad_norm": 1.11050283908844, + "learning_rate": 3.08033520824888e-06, + "loss": 0.6156, + "step": 1686 + }, + { + "epoch": 2.643528183716075, + "grad_norm": 1.15585196018219, + "learning_rate": 3.078285796241372e-06, + "loss": 0.6311, + "step": 1687 + }, + { + "epoch": 2.6450939457202507, + "grad_norm": 1.1588464975357056, + "learning_rate": 3.0762359735662663e-06, + "loss": 0.6804, + "step": 1688 + }, + { + "epoch": 2.646659707724426, + "grad_norm": 1.1857856512069702, + "learning_rate": 3.0741857416792374e-06, + "loss": 0.6737, + "step": 1689 + }, + { + "epoch": 2.6482254697286014, + "grad_norm": 1.1420296430587769, + "learning_rate": 3.0721351020362495e-06, + "loss": 0.6241, + "step": 1690 + }, + { + "epoch": 2.6497912317327765, + "grad_norm": 1.2236177921295166, + "learning_rate": 3.070084056093558e-06, + "loss": 0.7142, + "step": 1691 + }, + { + "epoch": 2.651356993736952, + "grad_norm": 1.1257219314575195, + "learning_rate": 3.0680326053077047e-06, + "loss": 0.5786, + "step": 1692 + }, + { + "epoch": 2.6529227557411272, + "grad_norm": 1.1258952617645264, + "learning_rate": 3.065980751135522e-06, + "loss": 0.7306, + "step": 1693 + }, + { + "epoch": 2.654488517745303, + "grad_norm": 1.1650607585906982, + "learning_rate": 3.0639284950341246e-06, + "loss": 0.6776, + "step": 1694 + }, + { + "epoch": 2.656054279749478, + "grad_norm": 1.1687644720077515, + "learning_rate": 3.0618758384609156e-06, + "loss": 0.6602, + "step": 1695 + }, + { + "epoch": 2.6576200417536535, + "grad_norm": 1.160516381263733, + "learning_rate": 3.0598227828735825e-06, + "loss": 0.6676, + "step": 1696 + }, + { + "epoch": 2.6591858037578286, + "grad_norm": 1.0864505767822266, + "learning_rate": 3.0577693297300947e-06, + "loss": 0.6843, + "step": 1697 + }, + { + "epoch": 2.6607515657620042, + "grad_norm": 1.0814275741577148, + "learning_rate": 3.055715480488706e-06, + "loss": 0.7214, + "step": 1698 + }, + { + "epoch": 2.6623173277661794, + "grad_norm": 1.1549509763717651, + "learning_rate": 3.0536612366079465e-06, + "loss": 0.7756, + "step": 1699 + }, + { + "epoch": 2.663883089770355, + "grad_norm": 1.125248908996582, + "learning_rate": 3.0516065995466336e-06, + "loss": 0.7133, + "step": 1700 + }, + { + "epoch": 2.6654488517745305, + "grad_norm": 1.0562750101089478, + "learning_rate": 3.0495515707638605e-06, + "loss": 0.6655, + "step": 1701 + }, + { + "epoch": 2.6670146137787056, + "grad_norm": 1.1991479396820068, + "learning_rate": 3.0474961517189965e-06, + "loss": 0.7382, + "step": 1702 + }, + { + "epoch": 2.6685803757828808, + "grad_norm": 1.0789530277252197, + "learning_rate": 3.0454403438716918e-06, + "loss": 0.6608, + "step": 1703 + }, + { + "epoch": 2.6701461377870563, + "grad_norm": 1.1776909828186035, + "learning_rate": 3.04338414868187e-06, + "loss": 0.7493, + "step": 1704 + }, + { + "epoch": 2.671711899791232, + "grad_norm": 1.1587637662887573, + "learning_rate": 3.041327567609731e-06, + "loss": 0.7042, + "step": 1705 + }, + { + "epoch": 2.673277661795407, + "grad_norm": 1.151100516319275, + "learning_rate": 3.0392706021157485e-06, + "loss": 0.5642, + "step": 1706 + }, + { + "epoch": 2.6748434237995826, + "grad_norm": 1.098358392715454, + "learning_rate": 3.03721325366067e-06, + "loss": 0.6177, + "step": 1707 + }, + { + "epoch": 2.6764091858037578, + "grad_norm": 1.1176345348358154, + "learning_rate": 3.0351555237055135e-06, + "loss": 0.7474, + "step": 1708 + }, + { + "epoch": 2.6779749478079333, + "grad_norm": 1.1730945110321045, + "learning_rate": 3.0330974137115683e-06, + "loss": 0.5899, + "step": 1709 + }, + { + "epoch": 2.6795407098121085, + "grad_norm": 1.1680086851119995, + "learning_rate": 3.031038925140395e-06, + "loss": 0.6576, + "step": 1710 + }, + { + "epoch": 2.681106471816284, + "grad_norm": 1.1488760709762573, + "learning_rate": 3.028980059453821e-06, + "loss": 0.6541, + "step": 1711 + }, + { + "epoch": 2.682672233820459, + "grad_norm": 1.1394031047821045, + "learning_rate": 3.0269208181139413e-06, + "loss": 0.6725, + "step": 1712 + }, + { + "epoch": 2.6842379958246347, + "grad_norm": 1.105613112449646, + "learning_rate": 3.024861202583122e-06, + "loss": 0.7088, + "step": 1713 + }, + { + "epoch": 2.68580375782881, + "grad_norm": 1.1166354417800903, + "learning_rate": 3.022801214323989e-06, + "loss": 0.7273, + "step": 1714 + }, + { + "epoch": 2.6873695198329854, + "grad_norm": 1.05818772315979, + "learning_rate": 3.020740854799436e-06, + "loss": 0.591, + "step": 1715 + }, + { + "epoch": 2.6889352818371606, + "grad_norm": 1.019797682762146, + "learning_rate": 3.0186801254726213e-06, + "loss": 0.6676, + "step": 1716 + }, + { + "epoch": 2.690501043841336, + "grad_norm": 1.1530799865722656, + "learning_rate": 3.016619027806963e-06, + "loss": 0.6181, + "step": 1717 + }, + { + "epoch": 2.6920668058455117, + "grad_norm": 1.132246971130371, + "learning_rate": 3.0145575632661426e-06, + "loss": 0.6932, + "step": 1718 + }, + { + "epoch": 2.693632567849687, + "grad_norm": 1.2640838623046875, + "learning_rate": 3.0124957333141036e-06, + "loss": 0.7029, + "step": 1719 + }, + { + "epoch": 2.695198329853862, + "grad_norm": 1.1203837394714355, + "learning_rate": 3.0104335394150447e-06, + "loss": 0.6522, + "step": 1720 + }, + { + "epoch": 2.6967640918580376, + "grad_norm": 1.1490654945373535, + "learning_rate": 3.008370983033426e-06, + "loss": 0.6837, + "step": 1721 + }, + { + "epoch": 2.698329853862213, + "grad_norm": 1.0226919651031494, + "learning_rate": 3.006308065633966e-06, + "loss": 0.6344, + "step": 1722 + }, + { + "epoch": 2.6998956158663883, + "grad_norm": 1.2632393836975098, + "learning_rate": 3.004244788681637e-06, + "loss": 0.7157, + "step": 1723 + }, + { + "epoch": 2.701461377870564, + "grad_norm": 1.0831620693206787, + "learning_rate": 3.0021811536416676e-06, + "loss": 0.6671, + "step": 1724 + }, + { + "epoch": 2.703027139874739, + "grad_norm": 1.1284756660461426, + "learning_rate": 3.0001171619795417e-06, + "loss": 0.6669, + "step": 1725 + }, + { + "epoch": 2.7045929018789145, + "grad_norm": 1.0045720338821411, + "learning_rate": 2.9980528151609945e-06, + "loss": 0.7276, + "step": 1726 + }, + { + "epoch": 2.7061586638830897, + "grad_norm": 1.1337859630584717, + "learning_rate": 2.9959881146520164e-06, + "loss": 0.6734, + "step": 1727 + }, + { + "epoch": 2.7077244258872653, + "grad_norm": 1.0121667385101318, + "learning_rate": 2.993923061918845e-06, + "loss": 0.6201, + "step": 1728 + }, + { + "epoch": 2.7092901878914404, + "grad_norm": 1.2311614751815796, + "learning_rate": 2.991857658427972e-06, + "loss": 0.6392, + "step": 1729 + }, + { + "epoch": 2.710855949895616, + "grad_norm": 1.0462316274642944, + "learning_rate": 2.989791905646135e-06, + "loss": 0.6736, + "step": 1730 + }, + { + "epoch": 2.712421711899791, + "grad_norm": 1.1915360689163208, + "learning_rate": 2.9877258050403214e-06, + "loss": 0.5977, + "step": 1731 + }, + { + "epoch": 2.7139874739039667, + "grad_norm": 1.0216567516326904, + "learning_rate": 2.985659358077765e-06, + "loss": 0.6393, + "step": 1732 + }, + { + "epoch": 2.715553235908142, + "grad_norm": 1.114071011543274, + "learning_rate": 2.9835925662259457e-06, + "loss": 0.6854, + "step": 1733 + }, + { + "epoch": 2.7171189979123174, + "grad_norm": 1.2007335424423218, + "learning_rate": 2.981525430952589e-06, + "loss": 0.584, + "step": 1734 + }, + { + "epoch": 2.718684759916493, + "grad_norm": 1.221422791481018, + "learning_rate": 2.979457953725663e-06, + "loss": 0.742, + "step": 1735 + }, + { + "epoch": 2.720250521920668, + "grad_norm": 1.0720511674880981, + "learning_rate": 2.9773901360133804e-06, + "loss": 0.6722, + "step": 1736 + }, + { + "epoch": 2.721816283924843, + "grad_norm": 1.1078705787658691, + "learning_rate": 2.975321979284193e-06, + "loss": 0.6837, + "step": 1737 + }, + { + "epoch": 2.723382045929019, + "grad_norm": 1.1647889614105225, + "learning_rate": 2.973253485006795e-06, + "loss": 0.7406, + "step": 1738 + }, + { + "epoch": 2.7249478079331944, + "grad_norm": 1.1482031345367432, + "learning_rate": 2.971184654650122e-06, + "loss": 0.7109, + "step": 1739 + }, + { + "epoch": 2.7265135699373695, + "grad_norm": 1.095131278038025, + "learning_rate": 2.9691154896833454e-06, + "loss": 0.6795, + "step": 1740 + }, + { + "epoch": 2.7280793319415446, + "grad_norm": 1.0917677879333496, + "learning_rate": 2.9670459915758746e-06, + "loss": 0.7142, + "step": 1741 + }, + { + "epoch": 2.72964509394572, + "grad_norm": 1.0911260843276978, + "learning_rate": 2.9649761617973572e-06, + "loss": 0.6917, + "step": 1742 + }, + { + "epoch": 2.7312108559498958, + "grad_norm": 1.2418321371078491, + "learning_rate": 2.9629060018176757e-06, + "loss": 0.6562, + "step": 1743 + }, + { + "epoch": 2.732776617954071, + "grad_norm": 1.114390254020691, + "learning_rate": 2.9608355131069464e-06, + "loss": 0.6812, + "step": 1744 + }, + { + "epoch": 2.7343423799582465, + "grad_norm": 1.0684012174606323, + "learning_rate": 2.9587646971355194e-06, + "loss": 0.688, + "step": 1745 + }, + { + "epoch": 2.7359081419624216, + "grad_norm": 1.1974529027938843, + "learning_rate": 2.9566935553739773e-06, + "loss": 0.6697, + "step": 1746 + }, + { + "epoch": 2.737473903966597, + "grad_norm": 1.0794286727905273, + "learning_rate": 2.9546220892931336e-06, + "loss": 0.6587, + "step": 1747 + }, + { + "epoch": 2.7390396659707723, + "grad_norm": 1.1143841743469238, + "learning_rate": 2.9525503003640336e-06, + "loss": 0.6814, + "step": 1748 + }, + { + "epoch": 2.740605427974948, + "grad_norm": 1.1970027685165405, + "learning_rate": 2.9504781900579504e-06, + "loss": 0.6714, + "step": 1749 + }, + { + "epoch": 2.742171189979123, + "grad_norm": 1.0299553871154785, + "learning_rate": 2.948405759846386e-06, + "loss": 0.6707, + "step": 1750 + }, + { + "epoch": 2.7437369519832986, + "grad_norm": 1.103028416633606, + "learning_rate": 2.946333011201068e-06, + "loss": 0.6163, + "step": 1751 + }, + { + "epoch": 2.745302713987474, + "grad_norm": 1.1230628490447998, + "learning_rate": 2.944259945593953e-06, + "loss": 0.5917, + "step": 1752 + }, + { + "epoch": 2.7468684759916493, + "grad_norm": 1.1009881496429443, + "learning_rate": 2.942186564497221e-06, + "loss": 0.7048, + "step": 1753 + }, + { + "epoch": 2.7484342379958244, + "grad_norm": 1.1123842000961304, + "learning_rate": 2.9401128693832752e-06, + "loss": 0.6063, + "step": 1754 + }, + { + "epoch": 2.75, + "grad_norm": 1.0923182964324951, + "learning_rate": 2.9380388617247433e-06, + "loss": 0.663, + "step": 1755 + }, + { + "epoch": 2.7515657620041756, + "grad_norm": 1.1761754751205444, + "learning_rate": 2.935964542994475e-06, + "loss": 0.6494, + "step": 1756 + }, + { + "epoch": 2.7531315240083507, + "grad_norm": 1.0580326318740845, + "learning_rate": 2.9338899146655396e-06, + "loss": 0.7391, + "step": 1757 + }, + { + "epoch": 2.754697286012526, + "grad_norm": 1.1655254364013672, + "learning_rate": 2.931814978211227e-06, + "loss": 0.6595, + "step": 1758 + }, + { + "epoch": 2.7562630480167014, + "grad_norm": 1.1162359714508057, + "learning_rate": 2.929739735105047e-06, + "loss": 0.6438, + "step": 1759 + }, + { + "epoch": 2.757828810020877, + "grad_norm": 1.173933744430542, + "learning_rate": 2.9276641868207244e-06, + "loss": 0.5591, + "step": 1760 + }, + { + "epoch": 2.759394572025052, + "grad_norm": 1.2183250188827515, + "learning_rate": 2.9255883348322038e-06, + "loss": 0.7317, + "step": 1761 + }, + { + "epoch": 2.7609603340292277, + "grad_norm": 1.0836644172668457, + "learning_rate": 2.9235121806136446e-06, + "loss": 0.7539, + "step": 1762 + }, + { + "epoch": 2.762526096033403, + "grad_norm": 1.1108825206756592, + "learning_rate": 2.921435725639419e-06, + "loss": 0.6598, + "step": 1763 + }, + { + "epoch": 2.7640918580375784, + "grad_norm": 1.090196967124939, + "learning_rate": 2.9193589713841132e-06, + "loss": 0.7412, + "step": 1764 + }, + { + "epoch": 2.7656576200417535, + "grad_norm": 1.1506152153015137, + "learning_rate": 2.9172819193225296e-06, + "loss": 0.6857, + "step": 1765 + }, + { + "epoch": 2.767223382045929, + "grad_norm": 1.127022385597229, + "learning_rate": 2.9152045709296766e-06, + "loss": 0.6383, + "step": 1766 + }, + { + "epoch": 2.7687891440501042, + "grad_norm": 1.1096911430358887, + "learning_rate": 2.9131269276807782e-06, + "loss": 0.6363, + "step": 1767 + }, + { + "epoch": 2.77035490605428, + "grad_norm": 1.1879534721374512, + "learning_rate": 2.911048991051263e-06, + "loss": 0.7313, + "step": 1768 + }, + { + "epoch": 2.7719206680584554, + "grad_norm": 1.0864057540893555, + "learning_rate": 2.908970762516772e-06, + "loss": 0.6602, + "step": 1769 + }, + { + "epoch": 2.7734864300626305, + "grad_norm": 1.174880027770996, + "learning_rate": 2.9068922435531515e-06, + "loss": 0.6731, + "step": 1770 + }, + { + "epoch": 2.7750521920668056, + "grad_norm": 1.1670523881912231, + "learning_rate": 2.904813435636453e-06, + "loss": 0.5982, + "step": 1771 + }, + { + "epoch": 2.776617954070981, + "grad_norm": 1.016558289527893, + "learning_rate": 2.902734340242937e-06, + "loss": 0.5549, + "step": 1772 + }, + { + "epoch": 2.778183716075157, + "grad_norm": 1.2644622325897217, + "learning_rate": 2.9006549588490636e-06, + "loss": 0.6209, + "step": 1773 + }, + { + "epoch": 2.779749478079332, + "grad_norm": 1.1689661741256714, + "learning_rate": 2.8985752929314985e-06, + "loss": 0.6629, + "step": 1774 + }, + { + "epoch": 2.781315240083507, + "grad_norm": 1.123618721961975, + "learning_rate": 2.8964953439671095e-06, + "loss": 0.6986, + "step": 1775 + }, + { + "epoch": 2.7828810020876826, + "grad_norm": 1.1276949644088745, + "learning_rate": 2.8944151134329646e-06, + "loss": 0.6125, + "step": 1776 + }, + { + "epoch": 2.784446764091858, + "grad_norm": 1.1804574728012085, + "learning_rate": 2.8923346028063328e-06, + "loss": 0.5912, + "step": 1777 + }, + { + "epoch": 2.7860125260960333, + "grad_norm": 1.1222927570343018, + "learning_rate": 2.89025381356468e-06, + "loss": 0.5837, + "step": 1778 + }, + { + "epoch": 2.787578288100209, + "grad_norm": 1.063511848449707, + "learning_rate": 2.888172747185673e-06, + "loss": 0.7243, + "step": 1779 + }, + { + "epoch": 2.789144050104384, + "grad_norm": 1.1183874607086182, + "learning_rate": 2.8860914051471722e-06, + "loss": 0.7204, + "step": 1780 + }, + { + "epoch": 2.7907098121085596, + "grad_norm": 1.135906457901001, + "learning_rate": 2.884009788927235e-06, + "loss": 0.7308, + "step": 1781 + }, + { + "epoch": 2.7922755741127347, + "grad_norm": 1.2013477087020874, + "learning_rate": 2.8819279000041156e-06, + "loss": 0.7004, + "step": 1782 + }, + { + "epoch": 2.7938413361169103, + "grad_norm": 1.074418306350708, + "learning_rate": 2.8798457398562584e-06, + "loss": 0.6404, + "step": 1783 + }, + { + "epoch": 2.7954070981210855, + "grad_norm": 1.0827600955963135, + "learning_rate": 2.8777633099623027e-06, + "loss": 0.7488, + "step": 1784 + }, + { + "epoch": 2.796972860125261, + "grad_norm": 1.1087040901184082, + "learning_rate": 2.8756806118010784e-06, + "loss": 0.6554, + "step": 1785 + }, + { + "epoch": 2.798538622129436, + "grad_norm": 1.0850439071655273, + "learning_rate": 2.8735976468516057e-06, + "loss": 0.6501, + "step": 1786 + }, + { + "epoch": 2.8001043841336117, + "grad_norm": 1.129045844078064, + "learning_rate": 2.871514416593096e-06, + "loss": 0.6697, + "step": 1787 + }, + { + "epoch": 2.801670146137787, + "grad_norm": 1.0854123830795288, + "learning_rate": 2.869430922504947e-06, + "loss": 0.6787, + "step": 1788 + }, + { + "epoch": 2.8032359081419624, + "grad_norm": 1.2950549125671387, + "learning_rate": 2.8673471660667448e-06, + "loss": 0.6682, + "step": 1789 + }, + { + "epoch": 2.804801670146138, + "grad_norm": 1.0760269165039062, + "learning_rate": 2.865263148758261e-06, + "loss": 0.6846, + "step": 1790 + }, + { + "epoch": 2.806367432150313, + "grad_norm": 1.1861540079116821, + "learning_rate": 2.863178872059454e-06, + "loss": 0.6029, + "step": 1791 + }, + { + "epoch": 2.8079331941544883, + "grad_norm": 1.081518292427063, + "learning_rate": 2.8610943374504645e-06, + "loss": 0.6801, + "step": 1792 + }, + { + "epoch": 2.809498956158664, + "grad_norm": 1.1814477443695068, + "learning_rate": 2.8590095464116192e-06, + "loss": 0.68, + "step": 1793 + }, + { + "epoch": 2.8110647181628394, + "grad_norm": 1.154828667640686, + "learning_rate": 2.856924500423422e-06, + "loss": 0.6164, + "step": 1794 + }, + { + "epoch": 2.8126304801670146, + "grad_norm": 1.0605709552764893, + "learning_rate": 2.8548392009665638e-06, + "loss": 0.6146, + "step": 1795 + }, + { + "epoch": 2.81419624217119, + "grad_norm": 1.1769119501113892, + "learning_rate": 2.852753649521911e-06, + "loss": 0.6789, + "step": 1796 + }, + { + "epoch": 2.8157620041753653, + "grad_norm": 1.006847858428955, + "learning_rate": 2.8506678475705114e-06, + "loss": 0.6719, + "step": 1797 + }, + { + "epoch": 2.817327766179541, + "grad_norm": 1.1305062770843506, + "learning_rate": 2.8485817965935886e-06, + "loss": 0.6548, + "step": 1798 + }, + { + "epoch": 2.818893528183716, + "grad_norm": 1.0815331935882568, + "learning_rate": 2.846495498072545e-06, + "loss": 0.7306, + "step": 1799 + }, + { + "epoch": 2.8204592901878915, + "grad_norm": 1.1019850969314575, + "learning_rate": 2.8444089534889587e-06, + "loss": 0.7188, + "step": 1800 + }, + { + "epoch": 2.8220250521920667, + "grad_norm": 1.1405457258224487, + "learning_rate": 2.8423221643245807e-06, + "loss": 0.6482, + "step": 1801 + }, + { + "epoch": 2.8235908141962422, + "grad_norm": 1.1295759677886963, + "learning_rate": 2.8402351320613374e-06, + "loss": 0.5882, + "step": 1802 + }, + { + "epoch": 2.8251565762004174, + "grad_norm": 1.1165974140167236, + "learning_rate": 2.8381478581813275e-06, + "loss": 0.6714, + "step": 1803 + }, + { + "epoch": 2.826722338204593, + "grad_norm": 1.0910004377365112, + "learning_rate": 2.836060344166821e-06, + "loss": 0.6971, + "step": 1804 + }, + { + "epoch": 2.828288100208768, + "grad_norm": 1.093596339225769, + "learning_rate": 2.833972591500259e-06, + "loss": 0.7041, + "step": 1805 + }, + { + "epoch": 2.8298538622129437, + "grad_norm": 1.1507813930511475, + "learning_rate": 2.8318846016642507e-06, + "loss": 0.6455, + "step": 1806 + }, + { + "epoch": 2.8314196242171192, + "grad_norm": 1.055422306060791, + "learning_rate": 2.8297963761415748e-06, + "loss": 0.7499, + "step": 1807 + }, + { + "epoch": 2.8329853862212944, + "grad_norm": 1.0835453271865845, + "learning_rate": 2.8277079164151782e-06, + "loss": 0.6739, + "step": 1808 + }, + { + "epoch": 2.8345511482254695, + "grad_norm": 1.247808575630188, + "learning_rate": 2.8256192239681723e-06, + "loss": 0.6774, + "step": 1809 + }, + { + "epoch": 2.836116910229645, + "grad_norm": 1.0836130380630493, + "learning_rate": 2.8235303002838343e-06, + "loss": 0.6592, + "step": 1810 + }, + { + "epoch": 2.8376826722338206, + "grad_norm": 1.199852466583252, + "learning_rate": 2.8214411468456067e-06, + "loss": 0.6728, + "step": 1811 + }, + { + "epoch": 2.8392484342379958, + "grad_norm": 1.0706546306610107, + "learning_rate": 2.8193517651370934e-06, + "loss": 0.7438, + "step": 1812 + }, + { + "epoch": 2.8408141962421714, + "grad_norm": 1.301465630531311, + "learning_rate": 2.8172621566420633e-06, + "loss": 0.6736, + "step": 1813 + }, + { + "epoch": 2.8423799582463465, + "grad_norm": 1.0916422605514526, + "learning_rate": 2.815172322844442e-06, + "loss": 0.7258, + "step": 1814 + }, + { + "epoch": 2.843945720250522, + "grad_norm": 1.1184017658233643, + "learning_rate": 2.8130822652283183e-06, + "loss": 0.6831, + "step": 1815 + }, + { + "epoch": 2.845511482254697, + "grad_norm": 1.1714346408843994, + "learning_rate": 2.8109919852779406e-06, + "loss": 0.6866, + "step": 1816 + }, + { + "epoch": 2.8470772442588728, + "grad_norm": 1.0131067037582397, + "learning_rate": 2.808901484477712e-06, + "loss": 0.6574, + "step": 1817 + }, + { + "epoch": 2.848643006263048, + "grad_norm": 1.2358931303024292, + "learning_rate": 2.8068107643121952e-06, + "loss": 0.5596, + "step": 1818 + }, + { + "epoch": 2.8502087682672235, + "grad_norm": 0.9713279604911804, + "learning_rate": 2.8047198262661074e-06, + "loss": 0.6502, + "step": 1819 + }, + { + "epoch": 2.8517745302713986, + "grad_norm": 1.0890235900878906, + "learning_rate": 2.80262867182432e-06, + "loss": 0.6566, + "step": 1820 + }, + { + "epoch": 2.853340292275574, + "grad_norm": 1.084876298904419, + "learning_rate": 2.8005373024718597e-06, + "loss": 0.6356, + "step": 1821 + }, + { + "epoch": 2.8549060542797493, + "grad_norm": 1.1338250637054443, + "learning_rate": 2.7984457196939056e-06, + "loss": 0.7226, + "step": 1822 + }, + { + "epoch": 2.856471816283925, + "grad_norm": 1.1295956373214722, + "learning_rate": 2.7963539249757872e-06, + "loss": 0.724, + "step": 1823 + }, + { + "epoch": 2.8580375782881005, + "grad_norm": 1.0954629182815552, + "learning_rate": 2.794261919802984e-06, + "loss": 0.685, + "step": 1824 + }, + { + "epoch": 2.8596033402922756, + "grad_norm": 1.0435383319854736, + "learning_rate": 2.792169705661128e-06, + "loss": 0.7532, + "step": 1825 + }, + { + "epoch": 2.8611691022964507, + "grad_norm": 1.1580495834350586, + "learning_rate": 2.790077284035996e-06, + "loss": 0.6745, + "step": 1826 + }, + { + "epoch": 2.8627348643006263, + "grad_norm": 1.0154657363891602, + "learning_rate": 2.787984656413514e-06, + "loss": 0.7312, + "step": 1827 + }, + { + "epoch": 2.864300626304802, + "grad_norm": 1.1673710346221924, + "learning_rate": 2.785891824279755e-06, + "loss": 0.6769, + "step": 1828 + }, + { + "epoch": 2.865866388308977, + "grad_norm": 1.1436103582382202, + "learning_rate": 2.7837987891209352e-06, + "loss": 0.6987, + "step": 1829 + }, + { + "epoch": 2.867432150313152, + "grad_norm": 1.2176343202590942, + "learning_rate": 2.7817055524234167e-06, + "loss": 0.6768, + "step": 1830 + }, + { + "epoch": 2.8689979123173277, + "grad_norm": 1.1192113161087036, + "learning_rate": 2.7796121156737042e-06, + "loss": 0.6468, + "step": 1831 + }, + { + "epoch": 2.8705636743215033, + "grad_norm": 1.085511326789856, + "learning_rate": 2.777518480358444e-06, + "loss": 0.6923, + "step": 1832 + }, + { + "epoch": 2.8721294363256784, + "grad_norm": 1.1534816026687622, + "learning_rate": 2.7754246479644227e-06, + "loss": 0.7897, + "step": 1833 + }, + { + "epoch": 2.873695198329854, + "grad_norm": 1.2280727624893188, + "learning_rate": 2.7733306199785704e-06, + "loss": 0.5913, + "step": 1834 + }, + { + "epoch": 2.875260960334029, + "grad_norm": 1.1843805313110352, + "learning_rate": 2.771236397887952e-06, + "loss": 0.6554, + "step": 1835 + }, + { + "epoch": 2.8768267223382047, + "grad_norm": 1.1090878248214722, + "learning_rate": 2.7691419831797724e-06, + "loss": 0.7155, + "step": 1836 + }, + { + "epoch": 2.87839248434238, + "grad_norm": 1.3048434257507324, + "learning_rate": 2.767047377341371e-06, + "loss": 0.689, + "step": 1837 + }, + { + "epoch": 2.8799582463465554, + "grad_norm": 1.0179991722106934, + "learning_rate": 2.764952581860227e-06, + "loss": 0.6919, + "step": 1838 + }, + { + "epoch": 2.8815240083507305, + "grad_norm": 1.0855897665023804, + "learning_rate": 2.7628575982239513e-06, + "loss": 0.7377, + "step": 1839 + }, + { + "epoch": 2.883089770354906, + "grad_norm": 1.0885690450668335, + "learning_rate": 2.7607624279202887e-06, + "loss": 0.7436, + "step": 1840 + }, + { + "epoch": 2.8846555323590817, + "grad_norm": 0.9842485189437866, + "learning_rate": 2.758667072437117e-06, + "loss": 0.6818, + "step": 1841 + }, + { + "epoch": 2.886221294363257, + "grad_norm": 1.2647449970245361, + "learning_rate": 2.7565715332624455e-06, + "loss": 0.6892, + "step": 1842 + }, + { + "epoch": 2.887787056367432, + "grad_norm": 1.0584148168563843, + "learning_rate": 2.7544758118844134e-06, + "loss": 0.6945, + "step": 1843 + }, + { + "epoch": 2.8893528183716075, + "grad_norm": 1.2626157999038696, + "learning_rate": 2.7523799097912905e-06, + "loss": 0.6617, + "step": 1844 + }, + { + "epoch": 2.890918580375783, + "grad_norm": 1.1559371948242188, + "learning_rate": 2.7502838284714746e-06, + "loss": 0.679, + "step": 1845 + }, + { + "epoch": 2.892484342379958, + "grad_norm": 1.1553645133972168, + "learning_rate": 2.7481875694134884e-06, + "loss": 0.7216, + "step": 1846 + }, + { + "epoch": 2.8940501043841333, + "grad_norm": 1.1300830841064453, + "learning_rate": 2.7460911341059847e-06, + "loss": 0.6761, + "step": 1847 + }, + { + "epoch": 2.895615866388309, + "grad_norm": 1.1680023670196533, + "learning_rate": 2.7439945240377395e-06, + "loss": 0.675, + "step": 1848 + }, + { + "epoch": 2.8971816283924845, + "grad_norm": 1.0944335460662842, + "learning_rate": 2.7418977406976526e-06, + "loss": 0.6959, + "step": 1849 + }, + { + "epoch": 2.8987473903966596, + "grad_norm": 1.1949477195739746, + "learning_rate": 2.7398007855747455e-06, + "loss": 0.6822, + "step": 1850 + }, + { + "epoch": 2.900313152400835, + "grad_norm": 1.2108348608016968, + "learning_rate": 2.737703660158167e-06, + "loss": 0.6568, + "step": 1851 + }, + { + "epoch": 2.9018789144050103, + "grad_norm": 1.119547963142395, + "learning_rate": 2.73560636593718e-06, + "loss": 0.6508, + "step": 1852 + }, + { + "epoch": 2.903444676409186, + "grad_norm": 1.1591404676437378, + "learning_rate": 2.733508904401172e-06, + "loss": 0.6462, + "step": 1853 + }, + { + "epoch": 2.905010438413361, + "grad_norm": 1.1378757953643799, + "learning_rate": 2.7314112770396474e-06, + "loss": 0.6202, + "step": 1854 + }, + { + "epoch": 2.9065762004175366, + "grad_norm": 1.1146254539489746, + "learning_rate": 2.729313485342229e-06, + "loss": 0.6618, + "step": 1855 + }, + { + "epoch": 2.9081419624217117, + "grad_norm": 1.1829131841659546, + "learning_rate": 2.727215530798656e-06, + "loss": 0.6708, + "step": 1856 + }, + { + "epoch": 2.9097077244258873, + "grad_norm": 1.0372153520584106, + "learning_rate": 2.725117414898784e-06, + "loss": 0.6666, + "step": 1857 + }, + { + "epoch": 2.911273486430063, + "grad_norm": 1.1896759271621704, + "learning_rate": 2.723019139132581e-06, + "loss": 0.7169, + "step": 1858 + }, + { + "epoch": 2.912839248434238, + "grad_norm": 1.137988805770874, + "learning_rate": 2.720920704990131e-06, + "loss": 0.6872, + "step": 1859 + }, + { + "epoch": 2.914405010438413, + "grad_norm": 1.1891788244247437, + "learning_rate": 2.7188221139616303e-06, + "loss": 0.7105, + "step": 1860 + }, + { + "epoch": 2.9159707724425887, + "grad_norm": 1.0640588998794556, + "learning_rate": 2.716723367537385e-06, + "loss": 0.6905, + "step": 1861 + }, + { + "epoch": 2.9175365344467643, + "grad_norm": 1.2714836597442627, + "learning_rate": 2.7146244672078125e-06, + "loss": 0.6713, + "step": 1862 + }, + { + "epoch": 2.9191022964509394, + "grad_norm": 1.222401738166809, + "learning_rate": 2.7125254144634388e-06, + "loss": 0.6638, + "step": 1863 + }, + { + "epoch": 2.9206680584551146, + "grad_norm": 1.0254578590393066, + "learning_rate": 2.7104262107948993e-06, + "loss": 0.7374, + "step": 1864 + }, + { + "epoch": 2.92223382045929, + "grad_norm": 1.089826226234436, + "learning_rate": 2.7083268576929368e-06, + "loss": 0.6742, + "step": 1865 + }, + { + "epoch": 2.9237995824634657, + "grad_norm": 1.1508773565292358, + "learning_rate": 2.7062273566483984e-06, + "loss": 0.5628, + "step": 1866 + }, + { + "epoch": 2.925365344467641, + "grad_norm": 1.1214534044265747, + "learning_rate": 2.7041277091522365e-06, + "loss": 0.5819, + "step": 1867 + }, + { + "epoch": 2.9269311064718164, + "grad_norm": 1.2162528038024902, + "learning_rate": 2.70202791669551e-06, + "loss": 0.6774, + "step": 1868 + }, + { + "epoch": 2.9284968684759916, + "grad_norm": 1.130321741104126, + "learning_rate": 2.6999279807693783e-06, + "loss": 0.6617, + "step": 1869 + }, + { + "epoch": 2.930062630480167, + "grad_norm": 1.1052041053771973, + "learning_rate": 2.697827902865103e-06, + "loss": 0.6348, + "step": 1870 + }, + { + "epoch": 2.9316283924843423, + "grad_norm": 1.243033766746521, + "learning_rate": 2.695727684474048e-06, + "loss": 0.671, + "step": 1871 + }, + { + "epoch": 2.933194154488518, + "grad_norm": 1.163845419883728, + "learning_rate": 2.693627327087675e-06, + "loss": 0.6605, + "step": 1872 + }, + { + "epoch": 2.934759916492693, + "grad_norm": 1.0230295658111572, + "learning_rate": 2.6915268321975454e-06, + "loss": 0.6511, + "step": 1873 + }, + { + "epoch": 2.9363256784968685, + "grad_norm": 1.0699726343154907, + "learning_rate": 2.6894262012953194e-06, + "loss": 0.5641, + "step": 1874 + }, + { + "epoch": 2.937891440501044, + "grad_norm": 1.139090657234192, + "learning_rate": 2.6873254358727503e-06, + "loss": 0.7327, + "step": 1875 + }, + { + "epoch": 2.9394572025052192, + "grad_norm": 1.0480340719223022, + "learning_rate": 2.68522453742169e-06, + "loss": 0.6559, + "step": 1876 + }, + { + "epoch": 2.9410229645093944, + "grad_norm": 1.1894673109054565, + "learning_rate": 2.6831235074340854e-06, + "loss": 0.6235, + "step": 1877 + }, + { + "epoch": 2.94258872651357, + "grad_norm": 1.1843831539154053, + "learning_rate": 2.681022347401975e-06, + "loss": 0.7161, + "step": 1878 + }, + { + "epoch": 2.9441544885177455, + "grad_norm": 1.278631329536438, + "learning_rate": 2.6789210588174887e-06, + "loss": 0.6651, + "step": 1879 + }, + { + "epoch": 2.9457202505219207, + "grad_norm": 1.1330087184906006, + "learning_rate": 2.67681964317285e-06, + "loss": 0.682, + "step": 1880 + }, + { + "epoch": 2.947286012526096, + "grad_norm": 1.1718090772628784, + "learning_rate": 2.6747181019603718e-06, + "loss": 0.6656, + "step": 1881 + }, + { + "epoch": 2.9488517745302714, + "grad_norm": 1.0476222038269043, + "learning_rate": 2.672616436672456e-06, + "loss": 0.543, + "step": 1882 + }, + { + "epoch": 2.950417536534447, + "grad_norm": 1.1602777242660522, + "learning_rate": 2.6705146488015933e-06, + "loss": 0.6887, + "step": 1883 + }, + { + "epoch": 2.951983298538622, + "grad_norm": 1.1091994047164917, + "learning_rate": 2.66841273984036e-06, + "loss": 0.7068, + "step": 1884 + }, + { + "epoch": 2.9535490605427976, + "grad_norm": 1.1577891111373901, + "learning_rate": 2.666310711281419e-06, + "loss": 0.7285, + "step": 1885 + }, + { + "epoch": 2.9551148225469728, + "grad_norm": 1.1392951011657715, + "learning_rate": 2.6642085646175196e-06, + "loss": 0.6308, + "step": 1886 + }, + { + "epoch": 2.9566805845511483, + "grad_norm": 1.3026043176651, + "learning_rate": 2.6621063013414934e-06, + "loss": 0.7092, + "step": 1887 + }, + { + "epoch": 2.9582463465553235, + "grad_norm": 1.0123710632324219, + "learning_rate": 2.660003922946255e-06, + "loss": 0.7045, + "step": 1888 + }, + { + "epoch": 2.959812108559499, + "grad_norm": 1.0754269361495972, + "learning_rate": 2.6579014309247998e-06, + "loss": 0.7301, + "step": 1889 + }, + { + "epoch": 2.961377870563674, + "grad_norm": 1.19624662399292, + "learning_rate": 2.655798826770206e-06, + "loss": 0.699, + "step": 1890 + }, + { + "epoch": 2.9629436325678498, + "grad_norm": 1.0728397369384766, + "learning_rate": 2.6536961119756317e-06, + "loss": 0.7188, + "step": 1891 + }, + { + "epoch": 2.964509394572025, + "grad_norm": 1.231490135192871, + "learning_rate": 2.6515932880343103e-06, + "loss": 0.7309, + "step": 1892 + }, + { + "epoch": 2.9660751565762005, + "grad_norm": 1.2304288148880005, + "learning_rate": 2.6494903564395563e-06, + "loss": 0.7267, + "step": 1893 + }, + { + "epoch": 2.9676409185803756, + "grad_norm": 1.0533643960952759, + "learning_rate": 2.6473873186847577e-06, + "loss": 0.7011, + "step": 1894 + }, + { + "epoch": 2.969206680584551, + "grad_norm": 1.1855214834213257, + "learning_rate": 2.6452841762633795e-06, + "loss": 0.707, + "step": 1895 + }, + { + "epoch": 2.9707724425887267, + "grad_norm": 1.1279842853546143, + "learning_rate": 2.643180930668962e-06, + "loss": 0.6033, + "step": 1896 + }, + { + "epoch": 2.972338204592902, + "grad_norm": 1.102318525314331, + "learning_rate": 2.6410775833951164e-06, + "loss": 0.6696, + "step": 1897 + }, + { + "epoch": 2.973903966597077, + "grad_norm": 1.1372147798538208, + "learning_rate": 2.6389741359355275e-06, + "loss": 0.7164, + "step": 1898 + }, + { + "epoch": 2.9754697286012526, + "grad_norm": 1.2800220251083374, + "learning_rate": 2.6368705897839515e-06, + "loss": 0.733, + "step": 1899 + }, + { + "epoch": 2.977035490605428, + "grad_norm": 1.0855633020401, + "learning_rate": 2.634766946434214e-06, + "loss": 0.5894, + "step": 1900 + }, + { + "epoch": 2.9786012526096033, + "grad_norm": 1.1212704181671143, + "learning_rate": 2.6326632073802092e-06, + "loss": 0.6753, + "step": 1901 + }, + { + "epoch": 2.980167014613779, + "grad_norm": 1.0885487794876099, + "learning_rate": 2.6305593741159004e-06, + "loss": 0.6802, + "step": 1902 + }, + { + "epoch": 2.981732776617954, + "grad_norm": 1.1605111360549927, + "learning_rate": 2.6284554481353182e-06, + "loss": 0.5837, + "step": 1903 + }, + { + "epoch": 2.9832985386221296, + "grad_norm": 1.1119557619094849, + "learning_rate": 2.626351430932556e-06, + "loss": 0.6767, + "step": 1904 + }, + { + "epoch": 2.9848643006263047, + "grad_norm": 1.1678719520568848, + "learning_rate": 2.6242473240017757e-06, + "loss": 0.7052, + "step": 1905 + }, + { + "epoch": 2.9864300626304803, + "grad_norm": 1.1222507953643799, + "learning_rate": 2.622143128837201e-06, + "loss": 0.6751, + "step": 1906 + }, + { + "epoch": 2.9879958246346554, + "grad_norm": 1.1091363430023193, + "learning_rate": 2.620038846933119e-06, + "loss": 0.6427, + "step": 1907 + }, + { + "epoch": 2.989561586638831, + "grad_norm": 1.0949740409851074, + "learning_rate": 2.6179344797838775e-06, + "loss": 0.6353, + "step": 1908 + }, + { + "epoch": 2.991127348643006, + "grad_norm": 1.2391382455825806, + "learning_rate": 2.615830028883885e-06, + "loss": 0.7133, + "step": 1909 + }, + { + "epoch": 2.9926931106471817, + "grad_norm": 1.1227985620498657, + "learning_rate": 2.6137254957276105e-06, + "loss": 0.6942, + "step": 1910 + }, + { + "epoch": 2.994258872651357, + "grad_norm": 1.2233500480651855, + "learning_rate": 2.61162088180958e-06, + "loss": 0.7006, + "step": 1911 + }, + { + "epoch": 2.9958246346555324, + "grad_norm": 1.1352185010910034, + "learning_rate": 2.609516188624378e-06, + "loss": 0.6304, + "step": 1912 + }, + { + "epoch": 2.997390396659708, + "grad_norm": 1.2024694681167603, + "learning_rate": 2.6074114176666455e-06, + "loss": 0.6911, + "step": 1913 + }, + { + "epoch": 2.998956158663883, + "grad_norm": 1.1703287363052368, + "learning_rate": 2.605306570431077e-06, + "loss": 0.6528, + "step": 1914 + }, + { + "epoch": 3.0015657620041756, + "grad_norm": 1.1240348815917969, + "learning_rate": 2.6032016484124224e-06, + "loss": 1.3204, + "step": 1915 + }, + { + "epoch": 3.0031315240083507, + "grad_norm": 1.0691215991973877, + "learning_rate": 2.6010966531054852e-06, + "loss": 0.643, + "step": 1916 + }, + { + "epoch": 3.0046972860125263, + "grad_norm": 1.1520895957946777, + "learning_rate": 2.5989915860051205e-06, + "loss": 0.6326, + "step": 1917 + }, + { + "epoch": 3.0062630480167014, + "grad_norm": 1.1557644605636597, + "learning_rate": 2.5968864486062334e-06, + "loss": 0.6654, + "step": 1918 + }, + { + "epoch": 3.007828810020877, + "grad_norm": 1.216338872909546, + "learning_rate": 2.59478124240378e-06, + "loss": 0.6109, + "step": 1919 + }, + { + "epoch": 3.009394572025052, + "grad_norm": 1.168593406677246, + "learning_rate": 2.592675968892765e-06, + "loss": 0.6167, + "step": 1920 + }, + { + "epoch": 3.0109603340292277, + "grad_norm": 1.103684902191162, + "learning_rate": 2.590570629568241e-06, + "loss": 0.5883, + "step": 1921 + }, + { + "epoch": 3.012526096033403, + "grad_norm": 1.1893324851989746, + "learning_rate": 2.5884652259253067e-06, + "loss": 0.6295, + "step": 1922 + }, + { + "epoch": 3.0140918580375784, + "grad_norm": 1.046500325202942, + "learning_rate": 2.5863597594591077e-06, + "loss": 0.6338, + "step": 1923 + }, + { + "epoch": 3.0156576200417535, + "grad_norm": 1.1588988304138184, + "learning_rate": 2.5842542316648333e-06, + "loss": 0.6208, + "step": 1924 + }, + { + "epoch": 3.017223382045929, + "grad_norm": 1.0734959840774536, + "learning_rate": 2.5821486440377168e-06, + "loss": 0.5612, + "step": 1925 + }, + { + "epoch": 3.0187891440501042, + "grad_norm": 1.128631353378296, + "learning_rate": 2.5800429980730335e-06, + "loss": 0.6208, + "step": 1926 + }, + { + "epoch": 3.02035490605428, + "grad_norm": 1.1834310293197632, + "learning_rate": 2.5779372952661003e-06, + "loss": 0.6418, + "step": 1927 + }, + { + "epoch": 3.021920668058455, + "grad_norm": 1.217034101486206, + "learning_rate": 2.5758315371122733e-06, + "loss": 0.7117, + "step": 1928 + }, + { + "epoch": 3.0234864300626305, + "grad_norm": 1.060900330543518, + "learning_rate": 2.573725725106952e-06, + "loss": 0.6104, + "step": 1929 + }, + { + "epoch": 3.0250521920668056, + "grad_norm": 1.1937845945358276, + "learning_rate": 2.571619860745569e-06, + "loss": 0.672, + "step": 1930 + }, + { + "epoch": 3.026617954070981, + "grad_norm": 1.0636308193206787, + "learning_rate": 2.569513945523598e-06, + "loss": 0.6525, + "step": 1931 + }, + { + "epoch": 3.028183716075157, + "grad_norm": 1.2644332647323608, + "learning_rate": 2.5674079809365443e-06, + "loss": 0.6109, + "step": 1932 + }, + { + "epoch": 3.029749478079332, + "grad_norm": 1.1538089513778687, + "learning_rate": 2.5653019684799545e-06, + "loss": 0.6813, + "step": 1933 + }, + { + "epoch": 3.0313152400835075, + "grad_norm": 1.1390219926834106, + "learning_rate": 2.5631959096494047e-06, + "loss": 0.6725, + "step": 1934 + }, + { + "epoch": 3.0328810020876826, + "grad_norm": 1.17123281955719, + "learning_rate": 2.5610898059405044e-06, + "loss": 0.5332, + "step": 1935 + }, + { + "epoch": 3.034446764091858, + "grad_norm": 1.1410160064697266, + "learning_rate": 2.5589836588488958e-06, + "loss": 0.6445, + "step": 1936 + }, + { + "epoch": 3.0360125260960333, + "grad_norm": 1.134597897529602, + "learning_rate": 2.5568774698702526e-06, + "loss": 0.6513, + "step": 1937 + }, + { + "epoch": 3.037578288100209, + "grad_norm": 1.1042146682739258, + "learning_rate": 2.5547712405002762e-06, + "loss": 0.6622, + "step": 1938 + }, + { + "epoch": 3.039144050104384, + "grad_norm": 1.129697561264038, + "learning_rate": 2.5526649722346986e-06, + "loss": 0.674, + "step": 1939 + }, + { + "epoch": 3.0407098121085596, + "grad_norm": 1.1788016557693481, + "learning_rate": 2.550558666569279e-06, + "loss": 0.6768, + "step": 1940 + }, + { + "epoch": 3.0422755741127347, + "grad_norm": 1.0913262367248535, + "learning_rate": 2.5484523249998022e-06, + "loss": 0.6817, + "step": 1941 + }, + { + "epoch": 3.0438413361169103, + "grad_norm": 1.1731491088867188, + "learning_rate": 2.5463459490220794e-06, + "loss": 0.652, + "step": 1942 + }, + { + "epoch": 3.0454070981210855, + "grad_norm": 1.139214038848877, + "learning_rate": 2.544239540131947e-06, + "loss": 0.6462, + "step": 1943 + }, + { + "epoch": 3.046972860125261, + "grad_norm": 1.163565754890442, + "learning_rate": 2.5421330998252617e-06, + "loss": 0.611, + "step": 1944 + }, + { + "epoch": 3.048538622129436, + "grad_norm": 1.136677861213684, + "learning_rate": 2.5400266295979055e-06, + "loss": 0.6502, + "step": 1945 + }, + { + "epoch": 3.0501043841336117, + "grad_norm": 1.1309559345245361, + "learning_rate": 2.537920130945782e-06, + "loss": 0.5777, + "step": 1946 + }, + { + "epoch": 3.051670146137787, + "grad_norm": 1.0693000555038452, + "learning_rate": 2.535813605364812e-06, + "loss": 0.6596, + "step": 1947 + }, + { + "epoch": 3.0532359081419624, + "grad_norm": 1.2112295627593994, + "learning_rate": 2.533707054350938e-06, + "loss": 0.6033, + "step": 1948 + }, + { + "epoch": 3.0548016701461376, + "grad_norm": 1.186867594718933, + "learning_rate": 2.531600479400119e-06, + "loss": 0.6399, + "step": 1949 + }, + { + "epoch": 3.056367432150313, + "grad_norm": 1.146904468536377, + "learning_rate": 2.5294938820083324e-06, + "loss": 0.6431, + "step": 1950 + }, + { + "epoch": 3.0579331941544887, + "grad_norm": 1.0728389024734497, + "learning_rate": 2.5273872636715707e-06, + "loss": 0.6754, + "step": 1951 + }, + { + "epoch": 3.059498956158664, + "grad_norm": 1.219117522239685, + "learning_rate": 2.5252806258858416e-06, + "loss": 0.5961, + "step": 1952 + }, + { + "epoch": 3.0610647181628394, + "grad_norm": 1.1819669008255005, + "learning_rate": 2.5231739701471657e-06, + "loss": 0.6809, + "step": 1953 + }, + { + "epoch": 3.0626304801670146, + "grad_norm": 1.181559443473816, + "learning_rate": 2.5210672979515773e-06, + "loss": 0.5754, + "step": 1954 + }, + { + "epoch": 3.06419624217119, + "grad_norm": 1.1187459230422974, + "learning_rate": 2.5189606107951225e-06, + "loss": 0.6324, + "step": 1955 + }, + { + "epoch": 3.0657620041753653, + "grad_norm": 1.20028555393219, + "learning_rate": 2.5168539101738576e-06, + "loss": 0.6178, + "step": 1956 + }, + { + "epoch": 3.067327766179541, + "grad_norm": 1.1661895513534546, + "learning_rate": 2.514747197583849e-06, + "loss": 0.5646, + "step": 1957 + }, + { + "epoch": 3.068893528183716, + "grad_norm": 1.1893212795257568, + "learning_rate": 2.512640474521169e-06, + "loss": 0.6387, + "step": 1958 + }, + { + "epoch": 3.0704592901878915, + "grad_norm": 1.1453262567520142, + "learning_rate": 2.510533742481902e-06, + "loss": 0.6416, + "step": 1959 + }, + { + "epoch": 3.0720250521920667, + "grad_norm": 1.2452423572540283, + "learning_rate": 2.5084270029621353e-06, + "loss": 0.6435, + "step": 1960 + }, + { + "epoch": 3.0735908141962422, + "grad_norm": 1.1707086563110352, + "learning_rate": 2.5063202574579625e-06, + "loss": 0.6328, + "step": 1961 + }, + { + "epoch": 3.0751565762004174, + "grad_norm": 1.2181516885757446, + "learning_rate": 2.5042135074654815e-06, + "loss": 0.6084, + "step": 1962 + }, + { + "epoch": 3.076722338204593, + "grad_norm": 1.1872752904891968, + "learning_rate": 2.502106754480793e-06, + "loss": 0.5128, + "step": 1963 + }, + { + "epoch": 3.078288100208768, + "grad_norm": 1.1264151334762573, + "learning_rate": 2.5e-06, + "loss": 0.6391, + "step": 1964 + }, + { + "epoch": 3.0798538622129437, + "grad_norm": 1.1406108140945435, + "learning_rate": 2.497893245519208e-06, + "loss": 0.6686, + "step": 1965 + }, + { + "epoch": 3.081419624217119, + "grad_norm": 1.1387429237365723, + "learning_rate": 2.495786492534519e-06, + "loss": 0.6449, + "step": 1966 + }, + { + "epoch": 3.0829853862212944, + "grad_norm": 1.1874876022338867, + "learning_rate": 2.493679742542038e-06, + "loss": 0.6471, + "step": 1967 + }, + { + "epoch": 3.08455114822547, + "grad_norm": 1.2808914184570312, + "learning_rate": 2.4915729970378655e-06, + "loss": 0.5295, + "step": 1968 + }, + { + "epoch": 3.086116910229645, + "grad_norm": 1.203076720237732, + "learning_rate": 2.4894662575180982e-06, + "loss": 0.6161, + "step": 1969 + }, + { + "epoch": 3.0876826722338206, + "grad_norm": 1.0870691537857056, + "learning_rate": 2.4873595254788317e-06, + "loss": 0.6344, + "step": 1970 + }, + { + "epoch": 3.0892484342379958, + "grad_norm": 1.174511194229126, + "learning_rate": 2.4852528024161526e-06, + "loss": 0.5958, + "step": 1971 + }, + { + "epoch": 3.0908141962421714, + "grad_norm": 1.2232667207717896, + "learning_rate": 2.4831460898261428e-06, + "loss": 0.5745, + "step": 1972 + }, + { + "epoch": 3.0923799582463465, + "grad_norm": 1.1356587409973145, + "learning_rate": 2.4810393892048783e-06, + "loss": 0.645, + "step": 1973 + }, + { + "epoch": 3.093945720250522, + "grad_norm": 1.2031471729278564, + "learning_rate": 2.478932702048424e-06, + "loss": 0.6246, + "step": 1974 + }, + { + "epoch": 3.095511482254697, + "grad_norm": 1.1597861051559448, + "learning_rate": 2.476826029852835e-06, + "loss": 0.6527, + "step": 1975 + }, + { + "epoch": 3.0970772442588728, + "grad_norm": 1.106753945350647, + "learning_rate": 2.4747193741141593e-06, + "loss": 0.7164, + "step": 1976 + }, + { + "epoch": 3.098643006263048, + "grad_norm": 1.147586703300476, + "learning_rate": 2.47261273632843e-06, + "loss": 0.6295, + "step": 1977 + }, + { + "epoch": 3.1002087682672235, + "grad_norm": 1.2318609952926636, + "learning_rate": 2.470506117991668e-06, + "loss": 0.5468, + "step": 1978 + }, + { + "epoch": 3.1017745302713986, + "grad_norm": 1.1728826761245728, + "learning_rate": 2.4683995205998818e-06, + "loss": 0.6144, + "step": 1979 + }, + { + "epoch": 3.103340292275574, + "grad_norm": 1.1983925104141235, + "learning_rate": 2.4662929456490633e-06, + "loss": 0.6214, + "step": 1980 + }, + { + "epoch": 3.1049060542797493, + "grad_norm": 1.134922981262207, + "learning_rate": 2.4641863946351884e-06, + "loss": 0.631, + "step": 1981 + }, + { + "epoch": 3.106471816283925, + "grad_norm": 1.2184958457946777, + "learning_rate": 2.4620798690542184e-06, + "loss": 0.6642, + "step": 1982 + }, + { + "epoch": 3.1080375782881, + "grad_norm": 1.1365389823913574, + "learning_rate": 2.459973370402095e-06, + "loss": 0.631, + "step": 1983 + }, + { + "epoch": 3.1096033402922756, + "grad_norm": 1.107224464416504, + "learning_rate": 2.4578669001747387e-06, + "loss": 0.6455, + "step": 1984 + }, + { + "epoch": 3.1111691022964507, + "grad_norm": 1.09954833984375, + "learning_rate": 2.455760459868054e-06, + "loss": 0.6674, + "step": 1985 + }, + { + "epoch": 3.1127348643006263, + "grad_norm": 1.1493922472000122, + "learning_rate": 2.453654050977921e-06, + "loss": 0.6144, + "step": 1986 + }, + { + "epoch": 3.114300626304802, + "grad_norm": 1.1649893522262573, + "learning_rate": 2.4515476750001978e-06, + "loss": 0.6621, + "step": 1987 + }, + { + "epoch": 3.115866388308977, + "grad_norm": 1.16617751121521, + "learning_rate": 2.449441333430722e-06, + "loss": 0.7146, + "step": 1988 + }, + { + "epoch": 3.1174321503131526, + "grad_norm": 1.137216567993164, + "learning_rate": 2.447335027765302e-06, + "loss": 0.6294, + "step": 1989 + }, + { + "epoch": 3.1189979123173277, + "grad_norm": 1.1343148946762085, + "learning_rate": 2.445228759499725e-06, + "loss": 0.6289, + "step": 1990 + }, + { + "epoch": 3.1205636743215033, + "grad_norm": 1.113688588142395, + "learning_rate": 2.4431225301297482e-06, + "loss": 0.6331, + "step": 1991 + }, + { + "epoch": 3.1221294363256784, + "grad_norm": 1.1988941431045532, + "learning_rate": 2.4410163411511046e-06, + "loss": 0.5989, + "step": 1992 + }, + { + "epoch": 3.123695198329854, + "grad_norm": 1.1929535865783691, + "learning_rate": 2.438910194059497e-06, + "loss": 0.6789, + "step": 1993 + }, + { + "epoch": 3.125260960334029, + "grad_norm": 1.2655651569366455, + "learning_rate": 2.436804090350596e-06, + "loss": 0.5709, + "step": 1994 + }, + { + "epoch": 3.1268267223382047, + "grad_norm": 1.187930941581726, + "learning_rate": 2.4346980315200464e-06, + "loss": 0.6324, + "step": 1995 + }, + { + "epoch": 3.12839248434238, + "grad_norm": 1.1696043014526367, + "learning_rate": 2.432592019063456e-06, + "loss": 0.5987, + "step": 1996 + }, + { + "epoch": 3.1299582463465554, + "grad_norm": 1.228115439414978, + "learning_rate": 2.430486054476403e-06, + "loss": 0.6581, + "step": 1997 + }, + { + "epoch": 3.1315240083507305, + "grad_norm": 1.1672264337539673, + "learning_rate": 2.4283801392544317e-06, + "loss": 0.6412, + "step": 1998 + }, + { + "epoch": 3.133089770354906, + "grad_norm": 1.2138696908950806, + "learning_rate": 2.4262742748930493e-06, + "loss": 0.6761, + "step": 1999 + }, + { + "epoch": 3.1346555323590812, + "grad_norm": 1.215649962425232, + "learning_rate": 2.4241684628877267e-06, + "loss": 0.6074, + "step": 2000 + }, + { + "epoch": 3.136221294363257, + "grad_norm": 1.2324877977371216, + "learning_rate": 2.4220627047339006e-06, + "loss": 0.6402, + "step": 2001 + }, + { + "epoch": 3.137787056367432, + "grad_norm": 1.230507254600525, + "learning_rate": 2.4199570019269678e-06, + "loss": 0.6784, + "step": 2002 + }, + { + "epoch": 3.1393528183716075, + "grad_norm": 1.056394100189209, + "learning_rate": 2.4178513559622836e-06, + "loss": 0.6209, + "step": 2003 + }, + { + "epoch": 3.140918580375783, + "grad_norm": 1.178450107574463, + "learning_rate": 2.415745768335167e-06, + "loss": 0.6463, + "step": 2004 + }, + { + "epoch": 3.142484342379958, + "grad_norm": 1.2509713172912598, + "learning_rate": 2.4136402405408927e-06, + "loss": 0.5886, + "step": 2005 + }, + { + "epoch": 3.144050104384134, + "grad_norm": 1.1313499212265015, + "learning_rate": 2.411534774074693e-06, + "loss": 0.6083, + "step": 2006 + }, + { + "epoch": 3.145615866388309, + "grad_norm": 1.208187460899353, + "learning_rate": 2.4094293704317596e-06, + "loss": 0.6488, + "step": 2007 + }, + { + "epoch": 3.1471816283924845, + "grad_norm": 1.1712708473205566, + "learning_rate": 2.4073240311072363e-06, + "loss": 0.6386, + "step": 2008 + }, + { + "epoch": 3.1487473903966596, + "grad_norm": 1.2195419073104858, + "learning_rate": 2.4052187575962206e-06, + "loss": 0.6016, + "step": 2009 + }, + { + "epoch": 3.150313152400835, + "grad_norm": 1.2239069938659668, + "learning_rate": 2.4031135513937675e-06, + "loss": 0.6095, + "step": 2010 + }, + { + "epoch": 3.1518789144050103, + "grad_norm": 1.274491786956787, + "learning_rate": 2.4010084139948807e-06, + "loss": 0.5551, + "step": 2011 + }, + { + "epoch": 3.153444676409186, + "grad_norm": 1.104736089706421, + "learning_rate": 2.398903346894515e-06, + "loss": 0.6555, + "step": 2012 + }, + { + "epoch": 3.155010438413361, + "grad_norm": 1.159359097480774, + "learning_rate": 2.396798351587578e-06, + "loss": 0.7065, + "step": 2013 + }, + { + "epoch": 3.1565762004175366, + "grad_norm": 1.2481755018234253, + "learning_rate": 2.394693429568924e-06, + "loss": 0.6085, + "step": 2014 + }, + { + "epoch": 3.1581419624217117, + "grad_norm": 1.2071157693862915, + "learning_rate": 2.392588582333355e-06, + "loss": 0.6871, + "step": 2015 + }, + { + "epoch": 3.1597077244258873, + "grad_norm": 1.1674847602844238, + "learning_rate": 2.3904838113756228e-06, + "loss": 0.5756, + "step": 2016 + }, + { + "epoch": 3.1612734864300625, + "grad_norm": 1.1972459554672241, + "learning_rate": 2.3883791181904216e-06, + "loss": 0.6307, + "step": 2017 + }, + { + "epoch": 3.162839248434238, + "grad_norm": 1.2457903623580933, + "learning_rate": 2.3862745042723907e-06, + "loss": 0.5552, + "step": 2018 + }, + { + "epoch": 3.164405010438413, + "grad_norm": 1.2475699186325073, + "learning_rate": 2.3841699711161155e-06, + "loss": 0.6054, + "step": 2019 + }, + { + "epoch": 3.1659707724425887, + "grad_norm": 1.2934919595718384, + "learning_rate": 2.3820655202161237e-06, + "loss": 0.6986, + "step": 2020 + }, + { + "epoch": 3.1675365344467643, + "grad_norm": 1.1634780168533325, + "learning_rate": 2.3799611530668814e-06, + "loss": 0.6697, + "step": 2021 + }, + { + "epoch": 3.1691022964509394, + "grad_norm": 1.1987229585647583, + "learning_rate": 2.3778568711628e-06, + "loss": 0.6376, + "step": 2022 + }, + { + "epoch": 3.170668058455115, + "grad_norm": 1.0842245817184448, + "learning_rate": 2.375752675998225e-06, + "loss": 0.629, + "step": 2023 + }, + { + "epoch": 3.17223382045929, + "grad_norm": 1.1573731899261475, + "learning_rate": 2.373648569067444e-06, + "loss": 0.6231, + "step": 2024 + }, + { + "epoch": 3.1737995824634657, + "grad_norm": 1.1786472797393799, + "learning_rate": 2.371544551864683e-06, + "loss": 0.7082, + "step": 2025 + }, + { + "epoch": 3.175365344467641, + "grad_norm": 1.2212613821029663, + "learning_rate": 2.3694406258841004e-06, + "loss": 0.6382, + "step": 2026 + }, + { + "epoch": 3.1769311064718164, + "grad_norm": 1.0942133665084839, + "learning_rate": 2.367336792619791e-06, + "loss": 0.6157, + "step": 2027 + }, + { + "epoch": 3.1784968684759916, + "grad_norm": 1.198861837387085, + "learning_rate": 2.365233053565787e-06, + "loss": 0.6172, + "step": 2028 + }, + { + "epoch": 3.180062630480167, + "grad_norm": 1.2345949411392212, + "learning_rate": 2.3631294102160493e-06, + "loss": 0.6561, + "step": 2029 + }, + { + "epoch": 3.1816283924843423, + "grad_norm": 1.1807947158813477, + "learning_rate": 2.3610258640644725e-06, + "loss": 0.6323, + "step": 2030 + }, + { + "epoch": 3.183194154488518, + "grad_norm": 1.272806167602539, + "learning_rate": 2.3589224166048844e-06, + "loss": 0.6286, + "step": 2031 + }, + { + "epoch": 3.184759916492693, + "grad_norm": 1.1519666910171509, + "learning_rate": 2.3568190693310388e-06, + "loss": 0.6351, + "step": 2032 + }, + { + "epoch": 3.1863256784968685, + "grad_norm": 1.3199100494384766, + "learning_rate": 2.35471582373662e-06, + "loss": 0.6282, + "step": 2033 + }, + { + "epoch": 3.1878914405010437, + "grad_norm": 1.1525390148162842, + "learning_rate": 2.352612681315243e-06, + "loss": 0.5453, + "step": 2034 + }, + { + "epoch": 3.1894572025052192, + "grad_norm": 1.1851332187652588, + "learning_rate": 2.3505096435604446e-06, + "loss": 0.5935, + "step": 2035 + }, + { + "epoch": 3.1910229645093944, + "grad_norm": 1.1144623756408691, + "learning_rate": 2.3484067119656905e-06, + "loss": 0.6756, + "step": 2036 + }, + { + "epoch": 3.19258872651357, + "grad_norm": 1.2141358852386475, + "learning_rate": 2.3463038880243687e-06, + "loss": 0.649, + "step": 2037 + }, + { + "epoch": 3.1941544885177455, + "grad_norm": 1.206795573234558, + "learning_rate": 2.3442011732297944e-06, + "loss": 0.7051, + "step": 2038 + }, + { + "epoch": 3.1957202505219207, + "grad_norm": 1.1212406158447266, + "learning_rate": 2.3420985690752015e-06, + "loss": 0.6279, + "step": 2039 + }, + { + "epoch": 3.1972860125260962, + "grad_norm": 1.1145480871200562, + "learning_rate": 2.339996077053746e-06, + "loss": 0.6885, + "step": 2040 + }, + { + "epoch": 3.1988517745302714, + "grad_norm": 1.235465168952942, + "learning_rate": 2.337893698658508e-06, + "loss": 0.6146, + "step": 2041 + }, + { + "epoch": 3.200417536534447, + "grad_norm": 1.2589614391326904, + "learning_rate": 2.3357914353824816e-06, + "loss": 0.6892, + "step": 2042 + }, + { + "epoch": 3.201983298538622, + "grad_norm": 1.3488636016845703, + "learning_rate": 2.3336892887185816e-06, + "loss": 0.5532, + "step": 2043 + }, + { + "epoch": 3.2035490605427976, + "grad_norm": 1.218481183052063, + "learning_rate": 2.331587260159641e-06, + "loss": 0.6554, + "step": 2044 + }, + { + "epoch": 3.2051148225469728, + "grad_norm": 1.207727074623108, + "learning_rate": 2.329485351198408e-06, + "loss": 0.5789, + "step": 2045 + }, + { + "epoch": 3.2066805845511483, + "grad_norm": 1.143080711364746, + "learning_rate": 2.3273835633275444e-06, + "loss": 0.6171, + "step": 2046 + }, + { + "epoch": 3.2082463465553235, + "grad_norm": 1.1894160509109497, + "learning_rate": 2.325281898039629e-06, + "loss": 0.6514, + "step": 2047 + }, + { + "epoch": 3.209812108559499, + "grad_norm": 1.1722674369812012, + "learning_rate": 2.323180356827151e-06, + "loss": 0.6305, + "step": 2048 + }, + { + "epoch": 3.211377870563674, + "grad_norm": 1.131166934967041, + "learning_rate": 2.3210789411825117e-06, + "loss": 0.651, + "step": 2049 + }, + { + "epoch": 3.2129436325678498, + "grad_norm": 1.0506892204284668, + "learning_rate": 2.318977652598026e-06, + "loss": 0.6041, + "step": 2050 + }, + { + "epoch": 3.214509394572025, + "grad_norm": 1.2478179931640625, + "learning_rate": 2.3168764925659155e-06, + "loss": 0.6418, + "step": 2051 + }, + { + "epoch": 3.2160751565762005, + "grad_norm": 1.1366816759109497, + "learning_rate": 2.31477546257831e-06, + "loss": 0.6269, + "step": 2052 + }, + { + "epoch": 3.2176409185803756, + "grad_norm": 1.1247450113296509, + "learning_rate": 2.3126745641272506e-06, + "loss": 0.6706, + "step": 2053 + }, + { + "epoch": 3.219206680584551, + "grad_norm": 1.166385531425476, + "learning_rate": 2.3105737987046823e-06, + "loss": 0.6267, + "step": 2054 + }, + { + "epoch": 3.2207724425887267, + "grad_norm": 1.134945273399353, + "learning_rate": 2.3084731678024546e-06, + "loss": 0.6367, + "step": 2055 + }, + { + "epoch": 3.222338204592902, + "grad_norm": 1.129788875579834, + "learning_rate": 2.306372672912326e-06, + "loss": 0.6327, + "step": 2056 + }, + { + "epoch": 3.223903966597077, + "grad_norm": 1.0779662132263184, + "learning_rate": 2.3042723155259526e-06, + "loss": 0.6261, + "step": 2057 + }, + { + "epoch": 3.2254697286012526, + "grad_norm": 1.2631971836090088, + "learning_rate": 2.302172097134897e-06, + "loss": 0.5617, + "step": 2058 + }, + { + "epoch": 3.227035490605428, + "grad_norm": 1.180594801902771, + "learning_rate": 2.3000720192306226e-06, + "loss": 0.585, + "step": 2059 + }, + { + "epoch": 3.2286012526096033, + "grad_norm": 1.1890621185302734, + "learning_rate": 2.297972083304491e-06, + "loss": 0.6525, + "step": 2060 + }, + { + "epoch": 3.230167014613779, + "grad_norm": 1.199937343597412, + "learning_rate": 2.295872290847764e-06, + "loss": 0.6024, + "step": 2061 + }, + { + "epoch": 3.231732776617954, + "grad_norm": 1.2697222232818604, + "learning_rate": 2.293772643351603e-06, + "loss": 0.6007, + "step": 2062 + }, + { + "epoch": 3.2332985386221296, + "grad_norm": 1.173552393913269, + "learning_rate": 2.2916731423070645e-06, + "loss": 0.6576, + "step": 2063 + }, + { + "epoch": 3.2348643006263047, + "grad_norm": 1.2051188945770264, + "learning_rate": 2.2895737892051007e-06, + "loss": 0.6862, + "step": 2064 + }, + { + "epoch": 3.2364300626304803, + "grad_norm": 1.1974767446517944, + "learning_rate": 2.287474585536562e-06, + "loss": 0.6201, + "step": 2065 + }, + { + "epoch": 3.2379958246346554, + "grad_norm": 1.2346913814544678, + "learning_rate": 2.2853755327921883e-06, + "loss": 0.5762, + "step": 2066 + }, + { + "epoch": 3.239561586638831, + "grad_norm": 1.1864664554595947, + "learning_rate": 2.2832766324626153e-06, + "loss": 0.6389, + "step": 2067 + }, + { + "epoch": 3.241127348643006, + "grad_norm": 1.300667405128479, + "learning_rate": 2.28117788603837e-06, + "loss": 0.6645, + "step": 2068 + }, + { + "epoch": 3.2426931106471817, + "grad_norm": 1.2889467477798462, + "learning_rate": 2.2790792950098695e-06, + "loss": 0.6399, + "step": 2069 + }, + { + "epoch": 3.244258872651357, + "grad_norm": 1.1807681322097778, + "learning_rate": 2.2769808608674193e-06, + "loss": 0.5974, + "step": 2070 + }, + { + "epoch": 3.2458246346555324, + "grad_norm": 1.2693206071853638, + "learning_rate": 2.274882585101217e-06, + "loss": 0.7067, + "step": 2071 + }, + { + "epoch": 3.2473903966597075, + "grad_norm": 1.1668689250946045, + "learning_rate": 2.2727844692013448e-06, + "loss": 0.5987, + "step": 2072 + }, + { + "epoch": 3.248956158663883, + "grad_norm": 1.2399280071258545, + "learning_rate": 2.2706865146577716e-06, + "loss": 0.6821, + "step": 2073 + }, + { + "epoch": 3.2505219206680582, + "grad_norm": 1.2630313634872437, + "learning_rate": 2.2685887229603534e-06, + "loss": 0.5846, + "step": 2074 + }, + { + "epoch": 3.252087682672234, + "grad_norm": 1.1724474430084229, + "learning_rate": 2.266491095598829e-06, + "loss": 0.6272, + "step": 2075 + }, + { + "epoch": 3.2536534446764094, + "grad_norm": 1.3041969537734985, + "learning_rate": 2.2643936340628205e-06, + "loss": 0.6246, + "step": 2076 + }, + { + "epoch": 3.2552192066805845, + "grad_norm": 1.152488350868225, + "learning_rate": 2.262296339841834e-06, + "loss": 0.6827, + "step": 2077 + }, + { + "epoch": 3.25678496868476, + "grad_norm": 1.1879392862319946, + "learning_rate": 2.260199214425255e-06, + "loss": 0.681, + "step": 2078 + }, + { + "epoch": 3.258350730688935, + "grad_norm": 1.1999655961990356, + "learning_rate": 2.2581022593023483e-06, + "loss": 0.7044, + "step": 2079 + }, + { + "epoch": 3.259916492693111, + "grad_norm": 1.2020212411880493, + "learning_rate": 2.256005475962261e-06, + "loss": 0.6206, + "step": 2080 + }, + { + "epoch": 3.261482254697286, + "grad_norm": 1.312864065170288, + "learning_rate": 2.2539088658940158e-06, + "loss": 0.6385, + "step": 2081 + }, + { + "epoch": 3.2630480167014615, + "grad_norm": 1.1875925064086914, + "learning_rate": 2.251812430586512e-06, + "loss": 0.5999, + "step": 2082 + }, + { + "epoch": 3.2646137787056366, + "grad_norm": 1.0950852632522583, + "learning_rate": 2.249716171528526e-06, + "loss": 0.7346, + "step": 2083 + }, + { + "epoch": 3.266179540709812, + "grad_norm": 1.2490246295928955, + "learning_rate": 2.24762009020871e-06, + "loss": 0.5889, + "step": 2084 + }, + { + "epoch": 3.2677453027139873, + "grad_norm": 1.1770070791244507, + "learning_rate": 2.2455241881155874e-06, + "loss": 0.681, + "step": 2085 + }, + { + "epoch": 3.269311064718163, + "grad_norm": 1.1888784170150757, + "learning_rate": 2.2434284667375553e-06, + "loss": 0.566, + "step": 2086 + }, + { + "epoch": 3.270876826722338, + "grad_norm": 1.290913701057434, + "learning_rate": 2.2413329275628836e-06, + "loss": 0.6737, + "step": 2087 + }, + { + "epoch": 3.2724425887265136, + "grad_norm": 1.1826350688934326, + "learning_rate": 2.2392375720797126e-06, + "loss": 0.6149, + "step": 2088 + }, + { + "epoch": 3.2740083507306887, + "grad_norm": 1.1560386419296265, + "learning_rate": 2.237142401776049e-06, + "loss": 0.6214, + "step": 2089 + }, + { + "epoch": 3.2755741127348643, + "grad_norm": 1.1863820552825928, + "learning_rate": 2.2350474181397736e-06, + "loss": 0.592, + "step": 2090 + }, + { + "epoch": 3.2771398747390394, + "grad_norm": 1.2404558658599854, + "learning_rate": 2.2329526226586294e-06, + "loss": 0.6648, + "step": 2091 + }, + { + "epoch": 3.278705636743215, + "grad_norm": 1.1192048788070679, + "learning_rate": 2.2308580168202284e-06, + "loss": 0.6657, + "step": 2092 + }, + { + "epoch": 3.2802713987473906, + "grad_norm": 1.2412128448486328, + "learning_rate": 2.228763602112049e-06, + "loss": 0.5999, + "step": 2093 + }, + { + "epoch": 3.2818371607515657, + "grad_norm": 1.2265843152999878, + "learning_rate": 2.2266693800214305e-06, + "loss": 0.6335, + "step": 2094 + }, + { + "epoch": 3.2834029227557413, + "grad_norm": 1.1712675094604492, + "learning_rate": 2.2245753520355773e-06, + "loss": 0.6533, + "step": 2095 + }, + { + "epoch": 3.2849686847599164, + "grad_norm": 1.2727960348129272, + "learning_rate": 2.222481519641557e-06, + "loss": 0.6233, + "step": 2096 + }, + { + "epoch": 3.286534446764092, + "grad_norm": 1.0983766317367554, + "learning_rate": 2.220387884326297e-06, + "loss": 0.6268, + "step": 2097 + }, + { + "epoch": 3.288100208768267, + "grad_norm": 1.2702240943908691, + "learning_rate": 2.2182944475765837e-06, + "loss": 0.6453, + "step": 2098 + }, + { + "epoch": 3.2896659707724427, + "grad_norm": 1.2952048778533936, + "learning_rate": 2.2162012108790656e-06, + "loss": 0.6177, + "step": 2099 + }, + { + "epoch": 3.291231732776618, + "grad_norm": 1.1496988534927368, + "learning_rate": 2.214108175720246e-06, + "loss": 0.6123, + "step": 2100 + }, + { + "epoch": 3.2927974947807934, + "grad_norm": 1.1527506113052368, + "learning_rate": 2.212015343586486e-06, + "loss": 0.5732, + "step": 2101 + }, + { + "epoch": 3.2943632567849686, + "grad_norm": 1.1944189071655273, + "learning_rate": 2.209922715964005e-06, + "loss": 0.6502, + "step": 2102 + }, + { + "epoch": 3.295929018789144, + "grad_norm": 1.2035129070281982, + "learning_rate": 2.2078302943388736e-06, + "loss": 0.6233, + "step": 2103 + }, + { + "epoch": 3.2974947807933193, + "grad_norm": 1.111652135848999, + "learning_rate": 2.2057380801970164e-06, + "loss": 0.6082, + "step": 2104 + }, + { + "epoch": 3.299060542797495, + "grad_norm": 1.223018765449524, + "learning_rate": 2.2036460750242136e-06, + "loss": 0.6537, + "step": 2105 + }, + { + "epoch": 3.30062630480167, + "grad_norm": 1.2612967491149902, + "learning_rate": 2.2015542803060952e-06, + "loss": 0.6857, + "step": 2106 + }, + { + "epoch": 3.3021920668058455, + "grad_norm": 1.0949403047561646, + "learning_rate": 2.1994626975281403e-06, + "loss": 0.591, + "step": 2107 + }, + { + "epoch": 3.3037578288100207, + "grad_norm": 1.1419187784194946, + "learning_rate": 2.197371328175681e-06, + "loss": 0.6729, + "step": 2108 + }, + { + "epoch": 3.3053235908141962, + "grad_norm": 1.1853833198547363, + "learning_rate": 2.195280173733894e-06, + "loss": 0.6015, + "step": 2109 + }, + { + "epoch": 3.306889352818372, + "grad_norm": 1.2181181907653809, + "learning_rate": 2.193189235687805e-06, + "loss": 0.6378, + "step": 2110 + }, + { + "epoch": 3.308455114822547, + "grad_norm": 1.291874885559082, + "learning_rate": 2.191098515522289e-06, + "loss": 0.6883, + "step": 2111 + }, + { + "epoch": 3.3100208768267225, + "grad_norm": 1.1752232313156128, + "learning_rate": 2.1890080147220606e-06, + "loss": 0.5764, + "step": 2112 + }, + { + "epoch": 3.3115866388308977, + "grad_norm": 1.2201036214828491, + "learning_rate": 2.186917734771682e-06, + "loss": 0.6593, + "step": 2113 + }, + { + "epoch": 3.3131524008350732, + "grad_norm": 1.2628161907196045, + "learning_rate": 2.1848276771555586e-06, + "loss": 0.5571, + "step": 2114 + }, + { + "epoch": 3.3147181628392484, + "grad_norm": 1.2265757322311401, + "learning_rate": 2.1827378433579384e-06, + "loss": 0.5375, + "step": 2115 + }, + { + "epoch": 3.316283924843424, + "grad_norm": 1.2193812131881714, + "learning_rate": 2.1806482348629065e-06, + "loss": 0.6333, + "step": 2116 + }, + { + "epoch": 3.317849686847599, + "grad_norm": 1.1655728816986084, + "learning_rate": 2.1785588531543945e-06, + "loss": 0.6113, + "step": 2117 + }, + { + "epoch": 3.3194154488517746, + "grad_norm": 1.2318220138549805, + "learning_rate": 2.1764696997161665e-06, + "loss": 0.659, + "step": 2118 + }, + { + "epoch": 3.3209812108559498, + "grad_norm": 1.1320031881332397, + "learning_rate": 2.174380776031828e-06, + "loss": 0.656, + "step": 2119 + }, + { + "epoch": 3.3225469728601253, + "grad_norm": 1.1628968715667725, + "learning_rate": 2.1722920835848226e-06, + "loss": 0.6614, + "step": 2120 + }, + { + "epoch": 3.3241127348643005, + "grad_norm": 1.2760236263275146, + "learning_rate": 2.170203623858426e-06, + "loss": 0.5455, + "step": 2121 + }, + { + "epoch": 3.325678496868476, + "grad_norm": 1.1655547618865967, + "learning_rate": 2.1681153983357497e-06, + "loss": 0.6167, + "step": 2122 + }, + { + "epoch": 3.327244258872651, + "grad_norm": 1.1703981161117554, + "learning_rate": 2.1660274084997414e-06, + "loss": 0.6874, + "step": 2123 + }, + { + "epoch": 3.3288100208768268, + "grad_norm": 1.2664302587509155, + "learning_rate": 2.1639396558331794e-06, + "loss": 0.6355, + "step": 2124 + }, + { + "epoch": 3.330375782881002, + "grad_norm": 1.1683710813522339, + "learning_rate": 2.1618521418186725e-06, + "loss": 0.6236, + "step": 2125 + }, + { + "epoch": 3.3319415448851775, + "grad_norm": 1.219686508178711, + "learning_rate": 2.159764867938663e-06, + "loss": 0.622, + "step": 2126 + }, + { + "epoch": 3.333507306889353, + "grad_norm": 1.1496042013168335, + "learning_rate": 2.15767783567542e-06, + "loss": 0.5684, + "step": 2127 + }, + { + "epoch": 3.335073068893528, + "grad_norm": 1.1844868659973145, + "learning_rate": 2.155591046511043e-06, + "loss": 0.5401, + "step": 2128 + }, + { + "epoch": 3.3366388308977033, + "grad_norm": 1.0850353240966797, + "learning_rate": 2.1535045019274557e-06, + "loss": 0.7124, + "step": 2129 + }, + { + "epoch": 3.338204592901879, + "grad_norm": 1.1659971475601196, + "learning_rate": 2.1514182034064123e-06, + "loss": 0.6502, + "step": 2130 + }, + { + "epoch": 3.3397703549060545, + "grad_norm": 1.1342421770095825, + "learning_rate": 2.1493321524294902e-06, + "loss": 0.6352, + "step": 2131 + }, + { + "epoch": 3.3413361169102296, + "grad_norm": 1.1351344585418701, + "learning_rate": 2.1472463504780893e-06, + "loss": 0.7151, + "step": 2132 + }, + { + "epoch": 3.342901878914405, + "grad_norm": 1.2182797193527222, + "learning_rate": 2.145160799033437e-06, + "loss": 0.6671, + "step": 2133 + }, + { + "epoch": 3.3444676409185803, + "grad_norm": 1.0785852670669556, + "learning_rate": 2.1430754995765783e-06, + "loss": 0.6595, + "step": 2134 + }, + { + "epoch": 3.346033402922756, + "grad_norm": 1.2832294702529907, + "learning_rate": 2.1409904535883816e-06, + "loss": 0.6094, + "step": 2135 + }, + { + "epoch": 3.347599164926931, + "grad_norm": 1.1902724504470825, + "learning_rate": 2.138905662549536e-06, + "loss": 0.6205, + "step": 2136 + }, + { + "epoch": 3.3491649269311066, + "grad_norm": 1.1487525701522827, + "learning_rate": 2.136821127940547e-06, + "loss": 0.6812, + "step": 2137 + }, + { + "epoch": 3.3507306889352817, + "grad_norm": 1.214062213897705, + "learning_rate": 2.1347368512417394e-06, + "loss": 0.6414, + "step": 2138 + }, + { + "epoch": 3.3522964509394573, + "grad_norm": 1.1467952728271484, + "learning_rate": 2.132652833933256e-06, + "loss": 0.6825, + "step": 2139 + }, + { + "epoch": 3.3538622129436324, + "grad_norm": 1.1381312608718872, + "learning_rate": 2.1305690774950543e-06, + "loss": 0.5342, + "step": 2140 + }, + { + "epoch": 3.355427974947808, + "grad_norm": 1.2287116050720215, + "learning_rate": 2.1284855834069044e-06, + "loss": 0.6098, + "step": 2141 + }, + { + "epoch": 3.356993736951983, + "grad_norm": 1.1297441720962524, + "learning_rate": 2.1264023531483947e-06, + "loss": 0.6527, + "step": 2142 + }, + { + "epoch": 3.3585594989561587, + "grad_norm": 1.1599788665771484, + "learning_rate": 2.124319388198923e-06, + "loss": 0.6767, + "step": 2143 + }, + { + "epoch": 3.3601252609603343, + "grad_norm": 1.2442164421081543, + "learning_rate": 2.1222366900376977e-06, + "loss": 0.618, + "step": 2144 + }, + { + "epoch": 3.3616910229645094, + "grad_norm": 1.2075343132019043, + "learning_rate": 2.120154260143742e-06, + "loss": 0.6413, + "step": 2145 + }, + { + "epoch": 3.3632567849686845, + "grad_norm": 1.173141360282898, + "learning_rate": 2.1180720999958857e-06, + "loss": 0.6272, + "step": 2146 + }, + { + "epoch": 3.36482254697286, + "grad_norm": 1.2015706300735474, + "learning_rate": 2.115990211072765e-06, + "loss": 0.6148, + "step": 2147 + }, + { + "epoch": 3.3663883089770357, + "grad_norm": 1.2163746356964111, + "learning_rate": 2.1139085948528286e-06, + "loss": 0.697, + "step": 2148 + }, + { + "epoch": 3.367954070981211, + "grad_norm": 1.2009632587432861, + "learning_rate": 2.111827252814328e-06, + "loss": 0.6202, + "step": 2149 + }, + { + "epoch": 3.3695198329853864, + "grad_norm": 1.1929175853729248, + "learning_rate": 2.1097461864353202e-06, + "loss": 0.5118, + "step": 2150 + }, + { + "epoch": 3.3710855949895615, + "grad_norm": 1.1137278079986572, + "learning_rate": 2.1076653971936685e-06, + "loss": 0.6489, + "step": 2151 + }, + { + "epoch": 3.372651356993737, + "grad_norm": 1.248371958732605, + "learning_rate": 2.1055848865670363e-06, + "loss": 0.6153, + "step": 2152 + }, + { + "epoch": 3.374217118997912, + "grad_norm": 1.1446887254714966, + "learning_rate": 2.103504656032891e-06, + "loss": 0.586, + "step": 2153 + }, + { + "epoch": 3.375782881002088, + "grad_norm": 1.3197392225265503, + "learning_rate": 2.1014247070685024e-06, + "loss": 0.6469, + "step": 2154 + }, + { + "epoch": 3.377348643006263, + "grad_norm": 1.1482443809509277, + "learning_rate": 2.099345041150938e-06, + "loss": 0.5673, + "step": 2155 + }, + { + "epoch": 3.3789144050104385, + "grad_norm": 1.3151997327804565, + "learning_rate": 2.097265659757064e-06, + "loss": 0.6684, + "step": 2156 + }, + { + "epoch": 3.3804801670146136, + "grad_norm": 1.1619030237197876, + "learning_rate": 2.0951865643635472e-06, + "loss": 0.6521, + "step": 2157 + }, + { + "epoch": 3.382045929018789, + "grad_norm": 1.2650394439697266, + "learning_rate": 2.09310775644685e-06, + "loss": 0.5612, + "step": 2158 + }, + { + "epoch": 3.3836116910229643, + "grad_norm": 1.1590896844863892, + "learning_rate": 2.0910292374832285e-06, + "loss": 0.6129, + "step": 2159 + }, + { + "epoch": 3.38517745302714, + "grad_norm": 1.1127238273620605, + "learning_rate": 2.088951008948738e-06, + "loss": 0.5928, + "step": 2160 + }, + { + "epoch": 3.3867432150313155, + "grad_norm": 1.203165888786316, + "learning_rate": 2.086873072319223e-06, + "loss": 0.6821, + "step": 2161 + }, + { + "epoch": 3.3883089770354906, + "grad_norm": 1.186804175376892, + "learning_rate": 2.0847954290703234e-06, + "loss": 0.6913, + "step": 2162 + }, + { + "epoch": 3.3898747390396657, + "grad_norm": 1.2237906455993652, + "learning_rate": 2.0827180806774717e-06, + "loss": 0.635, + "step": 2163 + }, + { + "epoch": 3.3914405010438413, + "grad_norm": 1.1457163095474243, + "learning_rate": 2.080641028615888e-06, + "loss": 0.6628, + "step": 2164 + }, + { + "epoch": 3.393006263048017, + "grad_norm": 1.2211480140686035, + "learning_rate": 2.078564274360582e-06, + "loss": 0.6259, + "step": 2165 + }, + { + "epoch": 3.394572025052192, + "grad_norm": 1.1769284009933472, + "learning_rate": 2.076487819386356e-06, + "loss": 0.6458, + "step": 2166 + }, + { + "epoch": 3.3961377870563676, + "grad_norm": 1.1229456663131714, + "learning_rate": 2.0744116651677967e-06, + "loss": 0.6518, + "step": 2167 + }, + { + "epoch": 3.3977035490605427, + "grad_norm": 1.2193604707717896, + "learning_rate": 2.072335813179275e-06, + "loss": 0.636, + "step": 2168 + }, + { + "epoch": 3.3992693110647183, + "grad_norm": 1.2844295501708984, + "learning_rate": 2.0702602648949536e-06, + "loss": 0.6275, + "step": 2169 + }, + { + "epoch": 3.4008350730688934, + "grad_norm": 1.2357110977172852, + "learning_rate": 2.0681850217887737e-06, + "loss": 0.6535, + "step": 2170 + }, + { + "epoch": 3.402400835073069, + "grad_norm": 1.191166877746582, + "learning_rate": 2.0661100853344608e-06, + "loss": 0.63, + "step": 2171 + }, + { + "epoch": 3.403966597077244, + "grad_norm": 1.1162155866622925, + "learning_rate": 2.064035457005526e-06, + "loss": 0.6438, + "step": 2172 + }, + { + "epoch": 3.4055323590814197, + "grad_norm": 1.1535841226577759, + "learning_rate": 2.0619611382752575e-06, + "loss": 0.6106, + "step": 2173 + }, + { + "epoch": 3.407098121085595, + "grad_norm": 1.2141941785812378, + "learning_rate": 2.059887130616726e-06, + "loss": 0.6009, + "step": 2174 + }, + { + "epoch": 3.4086638830897704, + "grad_norm": 1.1389793157577515, + "learning_rate": 2.05781343550278e-06, + "loss": 0.6691, + "step": 2175 + }, + { + "epoch": 3.4102296450939455, + "grad_norm": 1.2746304273605347, + "learning_rate": 2.055740054406048e-06, + "loss": 0.5752, + "step": 2176 + }, + { + "epoch": 3.411795407098121, + "grad_norm": 1.250404953956604, + "learning_rate": 2.053666988798933e-06, + "loss": 0.6839, + "step": 2177 + }, + { + "epoch": 3.4133611691022967, + "grad_norm": 1.2025327682495117, + "learning_rate": 2.0515942401536148e-06, + "loss": 0.6806, + "step": 2178 + }, + { + "epoch": 3.414926931106472, + "grad_norm": 1.1835360527038574, + "learning_rate": 2.0495218099420504e-06, + "loss": 0.6353, + "step": 2179 + }, + { + "epoch": 3.416492693110647, + "grad_norm": 1.3114181756973267, + "learning_rate": 2.0474496996359676e-06, + "loss": 0.6561, + "step": 2180 + }, + { + "epoch": 3.4180584551148225, + "grad_norm": 1.0880002975463867, + "learning_rate": 2.045377910706867e-06, + "loss": 0.6646, + "step": 2181 + }, + { + "epoch": 3.419624217118998, + "grad_norm": 1.1725163459777832, + "learning_rate": 2.0433064446260235e-06, + "loss": 0.5227, + "step": 2182 + }, + { + "epoch": 3.4211899791231732, + "grad_norm": 1.1769042015075684, + "learning_rate": 2.041235302864482e-06, + "loss": 0.6465, + "step": 2183 + }, + { + "epoch": 3.422755741127349, + "grad_norm": 1.2654465436935425, + "learning_rate": 2.039164486893054e-06, + "loss": 0.6407, + "step": 2184 + }, + { + "epoch": 3.424321503131524, + "grad_norm": 1.2088981866836548, + "learning_rate": 2.037093998182325e-06, + "loss": 0.6607, + "step": 2185 + }, + { + "epoch": 3.4258872651356995, + "grad_norm": 1.2232211828231812, + "learning_rate": 2.035023838202643e-06, + "loss": 0.6702, + "step": 2186 + }, + { + "epoch": 3.4274530271398747, + "grad_norm": 1.214909553527832, + "learning_rate": 2.032954008424126e-06, + "loss": 0.6322, + "step": 2187 + }, + { + "epoch": 3.4290187891440502, + "grad_norm": 1.1763339042663574, + "learning_rate": 2.0308845103166555e-06, + "loss": 0.6287, + "step": 2188 + }, + { + "epoch": 3.4305845511482254, + "grad_norm": 1.3962702751159668, + "learning_rate": 2.028815345349879e-06, + "loss": 0.547, + "step": 2189 + }, + { + "epoch": 3.432150313152401, + "grad_norm": 1.2137348651885986, + "learning_rate": 2.026746514993205e-06, + "loss": 0.6401, + "step": 2190 + }, + { + "epoch": 3.433716075156576, + "grad_norm": 1.143447756767273, + "learning_rate": 2.024678020715808e-06, + "loss": 0.6563, + "step": 2191 + }, + { + "epoch": 3.4352818371607516, + "grad_norm": 1.1660840511322021, + "learning_rate": 2.022609863986621e-06, + "loss": 0.6908, + "step": 2192 + }, + { + "epoch": 3.4368475991649268, + "grad_norm": 1.2545430660247803, + "learning_rate": 2.020542046274337e-06, + "loss": 0.6209, + "step": 2193 + }, + { + "epoch": 3.4384133611691023, + "grad_norm": 1.3274216651916504, + "learning_rate": 2.0184745690474116e-06, + "loss": 0.6243, + "step": 2194 + }, + { + "epoch": 3.4399791231732775, + "grad_norm": 1.2278743982315063, + "learning_rate": 2.0164074337740548e-06, + "loss": 0.6487, + "step": 2195 + }, + { + "epoch": 3.441544885177453, + "grad_norm": 1.1612262725830078, + "learning_rate": 2.0143406419222354e-06, + "loss": 0.5982, + "step": 2196 + }, + { + "epoch": 3.443110647181628, + "grad_norm": 1.2417254447937012, + "learning_rate": 2.01227419495968e-06, + "loss": 0.6292, + "step": 2197 + }, + { + "epoch": 3.4446764091858038, + "grad_norm": 1.1197260618209839, + "learning_rate": 2.0102080943538664e-06, + "loss": 0.6478, + "step": 2198 + }, + { + "epoch": 3.4462421711899793, + "grad_norm": 1.113791584968567, + "learning_rate": 2.008142341572029e-06, + "loss": 0.5747, + "step": 2199 + }, + { + "epoch": 3.4478079331941545, + "grad_norm": 1.1108101606369019, + "learning_rate": 2.0060769380811552e-06, + "loss": 0.6325, + "step": 2200 + }, + { + "epoch": 3.44937369519833, + "grad_norm": 1.2102993726730347, + "learning_rate": 2.004011885347985e-06, + "loss": 0.6066, + "step": 2201 + }, + { + "epoch": 3.450939457202505, + "grad_norm": 1.304426908493042, + "learning_rate": 2.0019471848390055e-06, + "loss": 0.6503, + "step": 2202 + }, + { + "epoch": 3.4525052192066807, + "grad_norm": 1.244708776473999, + "learning_rate": 1.9998828380204596e-06, + "loss": 0.6723, + "step": 2203 + }, + { + "epoch": 3.454070981210856, + "grad_norm": 1.1601401567459106, + "learning_rate": 1.997818846358333e-06, + "loss": 0.684, + "step": 2204 + }, + { + "epoch": 3.4556367432150314, + "grad_norm": 1.1886024475097656, + "learning_rate": 1.9957552113183635e-06, + "loss": 0.6277, + "step": 2205 + }, + { + "epoch": 3.4572025052192066, + "grad_norm": 1.2624013423919678, + "learning_rate": 1.993691934366035e-06, + "loss": 0.6569, + "step": 2206 + }, + { + "epoch": 3.458768267223382, + "grad_norm": 1.272737979888916, + "learning_rate": 1.991629016966575e-06, + "loss": 0.6481, + "step": 2207 + }, + { + "epoch": 3.4603340292275573, + "grad_norm": 1.2370678186416626, + "learning_rate": 1.989566460584956e-06, + "loss": 0.6598, + "step": 2208 + }, + { + "epoch": 3.461899791231733, + "grad_norm": 1.1626964807510376, + "learning_rate": 1.9875042666858973e-06, + "loss": 0.6582, + "step": 2209 + }, + { + "epoch": 3.463465553235908, + "grad_norm": 1.1608844995498657, + "learning_rate": 1.9854424367338578e-06, + "loss": 0.5931, + "step": 2210 + }, + { + "epoch": 3.4650313152400836, + "grad_norm": 1.2470782995224, + "learning_rate": 1.983380972193037e-06, + "loss": 0.6186, + "step": 2211 + }, + { + "epoch": 3.4665970772442587, + "grad_norm": 1.1697843074798584, + "learning_rate": 1.98131987452738e-06, + "loss": 0.6373, + "step": 2212 + }, + { + "epoch": 3.4681628392484343, + "grad_norm": 1.1374456882476807, + "learning_rate": 1.9792591452005645e-06, + "loss": 0.6518, + "step": 2213 + }, + { + "epoch": 3.4697286012526094, + "grad_norm": 1.1979857683181763, + "learning_rate": 1.9771987856760115e-06, + "loss": 0.6174, + "step": 2214 + }, + { + "epoch": 3.471294363256785, + "grad_norm": 1.2085691690444946, + "learning_rate": 1.975138797416879e-06, + "loss": 0.6296, + "step": 2215 + }, + { + "epoch": 3.4728601252609606, + "grad_norm": 1.1843719482421875, + "learning_rate": 1.973079181886059e-06, + "loss": 0.6796, + "step": 2216 + }, + { + "epoch": 3.4744258872651357, + "grad_norm": 1.230175256729126, + "learning_rate": 1.97101994054618e-06, + "loss": 0.6626, + "step": 2217 + }, + { + "epoch": 3.4759916492693113, + "grad_norm": 1.14729642868042, + "learning_rate": 1.9689610748596056e-06, + "loss": 0.5719, + "step": 2218 + }, + { + "epoch": 3.4775574112734864, + "grad_norm": 1.3141698837280273, + "learning_rate": 1.966902586288432e-06, + "loss": 0.6455, + "step": 2219 + }, + { + "epoch": 3.479123173277662, + "grad_norm": 1.15304434299469, + "learning_rate": 1.964844476294487e-06, + "loss": 0.6873, + "step": 2220 + }, + { + "epoch": 3.480688935281837, + "grad_norm": 1.193190336227417, + "learning_rate": 1.9627867463393308e-06, + "loss": 0.6656, + "step": 2221 + }, + { + "epoch": 3.4822546972860127, + "grad_norm": 1.2585062980651855, + "learning_rate": 1.960729397884252e-06, + "loss": 0.673, + "step": 2222 + }, + { + "epoch": 3.483820459290188, + "grad_norm": 1.2261855602264404, + "learning_rate": 1.9586724323902704e-06, + "loss": 0.6563, + "step": 2223 + }, + { + "epoch": 3.4853862212943634, + "grad_norm": 1.1301530599594116, + "learning_rate": 1.9566158513181312e-06, + "loss": 0.6392, + "step": 2224 + }, + { + "epoch": 3.4869519832985385, + "grad_norm": 1.1394745111465454, + "learning_rate": 1.954559656128309e-06, + "loss": 0.5913, + "step": 2225 + }, + { + "epoch": 3.488517745302714, + "grad_norm": 1.2290774583816528, + "learning_rate": 1.9525038482810044e-06, + "loss": 0.6117, + "step": 2226 + }, + { + "epoch": 3.490083507306889, + "grad_norm": 1.1881879568099976, + "learning_rate": 1.9504484292361403e-06, + "loss": 0.6634, + "step": 2227 + }, + { + "epoch": 3.491649269311065, + "grad_norm": 1.157094955444336, + "learning_rate": 1.948393400453367e-06, + "loss": 0.608, + "step": 2228 + }, + { + "epoch": 3.49321503131524, + "grad_norm": 1.1708686351776123, + "learning_rate": 1.9463387633920543e-06, + "loss": 0.6324, + "step": 2229 + }, + { + "epoch": 3.4947807933194155, + "grad_norm": 1.1796098947525024, + "learning_rate": 1.9442845195112954e-06, + "loss": 0.6601, + "step": 2230 + }, + { + "epoch": 3.4963465553235906, + "grad_norm": 1.2861493825912476, + "learning_rate": 1.942230670269906e-06, + "loss": 0.615, + "step": 2231 + }, + { + "epoch": 3.497912317327766, + "grad_norm": 1.2033058404922485, + "learning_rate": 1.9401772171264187e-06, + "loss": 0.6148, + "step": 2232 + }, + { + "epoch": 3.4994780793319418, + "grad_norm": 1.2392219305038452, + "learning_rate": 1.938124161539085e-06, + "loss": 0.5715, + "step": 2233 + }, + { + "epoch": 3.501043841336117, + "grad_norm": 1.1681913137435913, + "learning_rate": 1.936071504965876e-06, + "loss": 0.5809, + "step": 2234 + }, + { + "epoch": 3.502609603340292, + "grad_norm": 1.2426320314407349, + "learning_rate": 1.9340192488644793e-06, + "loss": 0.6475, + "step": 2235 + }, + { + "epoch": 3.5041753653444676, + "grad_norm": 1.2066962718963623, + "learning_rate": 1.9319673946922953e-06, + "loss": 0.5845, + "step": 2236 + }, + { + "epoch": 3.505741127348643, + "grad_norm": 1.3178675174713135, + "learning_rate": 1.929915943906443e-06, + "loss": 0.6422, + "step": 2237 + }, + { + "epoch": 3.5073068893528183, + "grad_norm": 1.1881810426712036, + "learning_rate": 1.927864897963751e-06, + "loss": 0.6726, + "step": 2238 + }, + { + "epoch": 3.508872651356994, + "grad_norm": 1.130273699760437, + "learning_rate": 1.925814258320763e-06, + "loss": 0.6027, + "step": 2239 + }, + { + "epoch": 3.510438413361169, + "grad_norm": 1.1589324474334717, + "learning_rate": 1.923764026433734e-06, + "loss": 0.5668, + "step": 2240 + }, + { + "epoch": 3.5120041753653446, + "grad_norm": 1.2391080856323242, + "learning_rate": 1.921714203758629e-06, + "loss": 0.6494, + "step": 2241 + }, + { + "epoch": 3.5135699373695197, + "grad_norm": 1.2799794673919678, + "learning_rate": 1.9196647917511207e-06, + "loss": 0.6579, + "step": 2242 + }, + { + "epoch": 3.5151356993736953, + "grad_norm": 1.1656668186187744, + "learning_rate": 1.917615791866593e-06, + "loss": 0.6021, + "step": 2243 + }, + { + "epoch": 3.5167014613778704, + "grad_norm": 1.142409086227417, + "learning_rate": 1.9155672055601364e-06, + "loss": 0.6258, + "step": 2244 + }, + { + "epoch": 3.518267223382046, + "grad_norm": 1.1905341148376465, + "learning_rate": 1.9135190342865447e-06, + "loss": 0.6112, + "step": 2245 + }, + { + "epoch": 3.519832985386221, + "grad_norm": 1.143678903579712, + "learning_rate": 1.911471279500323e-06, + "loss": 0.6136, + "step": 2246 + }, + { + "epoch": 3.5213987473903967, + "grad_norm": 1.240808367729187, + "learning_rate": 1.909423942655674e-06, + "loss": 0.6403, + "step": 2247 + }, + { + "epoch": 3.522964509394572, + "grad_norm": 1.2007147073745728, + "learning_rate": 1.907377025206507e-06, + "loss": 0.6564, + "step": 2248 + }, + { + "epoch": 3.5245302713987474, + "grad_norm": 1.1846612691879272, + "learning_rate": 1.9053305286064356e-06, + "loss": 0.6839, + "step": 2249 + }, + { + "epoch": 3.526096033402923, + "grad_norm": 1.1430362462997437, + "learning_rate": 1.90328445430877e-06, + "loss": 0.6447, + "step": 2250 + }, + { + "epoch": 3.527661795407098, + "grad_norm": 1.1141353845596313, + "learning_rate": 1.9012388037665222e-06, + "loss": 0.637, + "step": 2251 + }, + { + "epoch": 3.5292275574112733, + "grad_norm": 1.329208254814148, + "learning_rate": 1.8991935784324048e-06, + "loss": 0.7307, + "step": 2252 + }, + { + "epoch": 3.530793319415449, + "grad_norm": 1.3457560539245605, + "learning_rate": 1.8971487797588276e-06, + "loss": 0.6345, + "step": 2253 + }, + { + "epoch": 3.5323590814196244, + "grad_norm": 1.1213788986206055, + "learning_rate": 1.8951044091978951e-06, + "loss": 0.5832, + "step": 2254 + }, + { + "epoch": 3.5339248434237995, + "grad_norm": 1.2542991638183594, + "learning_rate": 1.8930604682014125e-06, + "loss": 0.7679, + "step": 2255 + }, + { + "epoch": 3.535490605427975, + "grad_norm": 1.20833158493042, + "learning_rate": 1.891016958220875e-06, + "loss": 0.6436, + "step": 2256 + }, + { + "epoch": 3.5370563674321502, + "grad_norm": 1.2129275798797607, + "learning_rate": 1.8889738807074743e-06, + "loss": 0.6603, + "step": 2257 + }, + { + "epoch": 3.538622129436326, + "grad_norm": 1.2521908283233643, + "learning_rate": 1.886931237112096e-06, + "loss": 0.635, + "step": 2258 + }, + { + "epoch": 3.540187891440501, + "grad_norm": 1.248688817024231, + "learning_rate": 1.8848890288853152e-06, + "loss": 0.6358, + "step": 2259 + }, + { + "epoch": 3.5417536534446765, + "grad_norm": 1.1389580965042114, + "learning_rate": 1.882847257477398e-06, + "loss": 0.621, + "step": 2260 + }, + { + "epoch": 3.5433194154488517, + "grad_norm": 1.2111111879348755, + "learning_rate": 1.880805924338302e-06, + "loss": 0.6486, + "step": 2261 + }, + { + "epoch": 3.5448851774530272, + "grad_norm": 1.1926789283752441, + "learning_rate": 1.8787650309176735e-06, + "loss": 0.6613, + "step": 2262 + }, + { + "epoch": 3.5464509394572024, + "grad_norm": 1.144853949546814, + "learning_rate": 1.8767245786648435e-06, + "loss": 0.6353, + "step": 2263 + }, + { + "epoch": 3.548016701461378, + "grad_norm": 1.2768452167510986, + "learning_rate": 1.874684569028834e-06, + "loss": 0.6479, + "step": 2264 + }, + { + "epoch": 3.549582463465553, + "grad_norm": 1.2346423864364624, + "learning_rate": 1.8726450034583496e-06, + "loss": 0.6315, + "step": 2265 + }, + { + "epoch": 3.5511482254697286, + "grad_norm": 1.1488447189331055, + "learning_rate": 1.87060588340178e-06, + "loss": 0.6493, + "step": 2266 + }, + { + "epoch": 3.552713987473904, + "grad_norm": 1.1464084386825562, + "learning_rate": 1.8685672103072005e-06, + "loss": 0.6181, + "step": 2267 + }, + { + "epoch": 3.5542797494780793, + "grad_norm": 1.289974570274353, + "learning_rate": 1.8665289856223662e-06, + "loss": 0.6522, + "step": 2268 + }, + { + "epoch": 3.5558455114822545, + "grad_norm": 1.2859828472137451, + "learning_rate": 1.8644912107947164e-06, + "loss": 0.57, + "step": 2269 + }, + { + "epoch": 3.55741127348643, + "grad_norm": 1.176608920097351, + "learning_rate": 1.8624538872713676e-06, + "loss": 0.6546, + "step": 2270 + }, + { + "epoch": 3.5589770354906056, + "grad_norm": 1.335280418395996, + "learning_rate": 1.86041701649912e-06, + "loss": 0.6659, + "step": 2271 + }, + { + "epoch": 3.5605427974947808, + "grad_norm": 1.2693639993667603, + "learning_rate": 1.858380599924449e-06, + "loss": 0.6348, + "step": 2272 + }, + { + "epoch": 3.562108559498956, + "grad_norm": 1.1791412830352783, + "learning_rate": 1.856344638993508e-06, + "loss": 0.6755, + "step": 2273 + }, + { + "epoch": 3.5636743215031315, + "grad_norm": 1.1750035285949707, + "learning_rate": 1.8543091351521292e-06, + "loss": 0.6508, + "step": 2274 + }, + { + "epoch": 3.565240083507307, + "grad_norm": 1.1812372207641602, + "learning_rate": 1.8522740898458178e-06, + "loss": 0.5811, + "step": 2275 + }, + { + "epoch": 3.566805845511482, + "grad_norm": 1.1619631052017212, + "learning_rate": 1.8502395045197522e-06, + "loss": 0.6227, + "step": 2276 + }, + { + "epoch": 3.5683716075156577, + "grad_norm": 1.2588560581207275, + "learning_rate": 1.8482053806187878e-06, + "loss": 0.6806, + "step": 2277 + }, + { + "epoch": 3.569937369519833, + "grad_norm": 1.0842543840408325, + "learning_rate": 1.8461717195874504e-06, + "loss": 0.6315, + "step": 2278 + }, + { + "epoch": 3.5715031315240084, + "grad_norm": 1.1269354820251465, + "learning_rate": 1.8441385228699355e-06, + "loss": 0.6387, + "step": 2279 + }, + { + "epoch": 3.5730688935281836, + "grad_norm": 1.196655511856079, + "learning_rate": 1.8421057919101128e-06, + "loss": 0.6872, + "step": 2280 + }, + { + "epoch": 3.574634655532359, + "grad_norm": 1.3561420440673828, + "learning_rate": 1.8400735281515175e-06, + "loss": 0.6538, + "step": 2281 + }, + { + "epoch": 3.5762004175365343, + "grad_norm": 1.1209763288497925, + "learning_rate": 1.838041733037354e-06, + "loss": 0.6934, + "step": 2282 + }, + { + "epoch": 3.57776617954071, + "grad_norm": 1.0936529636383057, + "learning_rate": 1.8360104080104962e-06, + "loss": 0.7136, + "step": 2283 + }, + { + "epoch": 3.5793319415448854, + "grad_norm": 1.2066773176193237, + "learning_rate": 1.8339795545134814e-06, + "loss": 0.6259, + "step": 2284 + }, + { + "epoch": 3.5808977035490606, + "grad_norm": 1.0662962198257446, + "learning_rate": 1.8319491739885115e-06, + "loss": 0.6195, + "step": 2285 + }, + { + "epoch": 3.5824634655532357, + "grad_norm": 1.284588098526001, + "learning_rate": 1.8299192678774564e-06, + "loss": 0.6154, + "step": 2286 + }, + { + "epoch": 3.5840292275574113, + "grad_norm": 1.1617132425308228, + "learning_rate": 1.8278898376218456e-06, + "loss": 0.6574, + "step": 2287 + }, + { + "epoch": 3.585594989561587, + "grad_norm": 1.1630313396453857, + "learning_rate": 1.8258608846628708e-06, + "loss": 0.577, + "step": 2288 + }, + { + "epoch": 3.587160751565762, + "grad_norm": 1.1374728679656982, + "learning_rate": 1.8238324104413879e-06, + "loss": 0.6908, + "step": 2289 + }, + { + "epoch": 3.588726513569937, + "grad_norm": 1.1832219362258911, + "learning_rate": 1.8218044163979088e-06, + "loss": 0.6409, + "step": 2290 + }, + { + "epoch": 3.5902922755741127, + "grad_norm": 1.4350699186325073, + "learning_rate": 1.8197769039726065e-06, + "loss": 0.6043, + "step": 2291 + }, + { + "epoch": 3.5918580375782883, + "grad_norm": 1.0792481899261475, + "learning_rate": 1.8177498746053129e-06, + "loss": 0.5953, + "step": 2292 + }, + { + "epoch": 3.5934237995824634, + "grad_norm": 1.2297285795211792, + "learning_rate": 1.8157233297355154e-06, + "loss": 0.646, + "step": 2293 + }, + { + "epoch": 3.594989561586639, + "grad_norm": 1.1795411109924316, + "learning_rate": 1.8136972708023557e-06, + "loss": 0.6187, + "step": 2294 + }, + { + "epoch": 3.596555323590814, + "grad_norm": 1.2370169162750244, + "learning_rate": 1.8116716992446342e-06, + "loss": 0.6899, + "step": 2295 + }, + { + "epoch": 3.5981210855949897, + "grad_norm": 1.1697649955749512, + "learning_rate": 1.8096466165008037e-06, + "loss": 0.6495, + "step": 2296 + }, + { + "epoch": 3.599686847599165, + "grad_norm": 1.1952897310256958, + "learning_rate": 1.807622024008967e-06, + "loss": 0.6007, + "step": 2297 + }, + { + "epoch": 3.6012526096033404, + "grad_norm": 1.2342556715011597, + "learning_rate": 1.8055979232068845e-06, + "loss": 0.5842, + "step": 2298 + }, + { + "epoch": 3.6028183716075155, + "grad_norm": 1.2582154273986816, + "learning_rate": 1.8035743155319618e-06, + "loss": 0.6939, + "step": 2299 + }, + { + "epoch": 3.604384133611691, + "grad_norm": 1.1596347093582153, + "learning_rate": 1.8015512024212573e-06, + "loss": 0.6609, + "step": 2300 + }, + { + "epoch": 3.6059498956158667, + "grad_norm": 1.2669062614440918, + "learning_rate": 1.7995285853114785e-06, + "loss": 0.6747, + "step": 2301 + }, + { + "epoch": 3.607515657620042, + "grad_norm": 1.2295914888381958, + "learning_rate": 1.797506465638979e-06, + "loss": 0.6374, + "step": 2302 + }, + { + "epoch": 3.609081419624217, + "grad_norm": 1.2566529512405396, + "learning_rate": 1.795484844839759e-06, + "loss": 0.6034, + "step": 2303 + }, + { + "epoch": 3.6106471816283925, + "grad_norm": 1.2655516862869263, + "learning_rate": 1.7934637243494663e-06, + "loss": 0.6394, + "step": 2304 + }, + { + "epoch": 3.612212943632568, + "grad_norm": 1.2168526649475098, + "learning_rate": 1.7914431056033927e-06, + "loss": 0.639, + "step": 2305 + }, + { + "epoch": 3.613778705636743, + "grad_norm": 1.309919834136963, + "learning_rate": 1.7894229900364721e-06, + "loss": 0.648, + "step": 2306 + }, + { + "epoch": 3.6153444676409183, + "grad_norm": 1.3890948295593262, + "learning_rate": 1.7874033790832846e-06, + "loss": 0.5951, + "step": 2307 + }, + { + "epoch": 3.616910229645094, + "grad_norm": 1.1613552570343018, + "learning_rate": 1.7853842741780474e-06, + "loss": 0.6083, + "step": 2308 + }, + { + "epoch": 3.6184759916492695, + "grad_norm": 1.3169668912887573, + "learning_rate": 1.783365676754621e-06, + "loss": 0.6439, + "step": 2309 + }, + { + "epoch": 3.6200417536534446, + "grad_norm": 1.21388578414917, + "learning_rate": 1.781347588246507e-06, + "loss": 0.6702, + "step": 2310 + }, + { + "epoch": 3.62160751565762, + "grad_norm": 1.202735185623169, + "learning_rate": 1.7793300100868428e-06, + "loss": 0.6417, + "step": 2311 + }, + { + "epoch": 3.6231732776617953, + "grad_norm": 1.2919602394104004, + "learning_rate": 1.7773129437084028e-06, + "loss": 0.6139, + "step": 2312 + }, + { + "epoch": 3.624739039665971, + "grad_norm": 1.1894437074661255, + "learning_rate": 1.7752963905436014e-06, + "loss": 0.6689, + "step": 2313 + }, + { + "epoch": 3.626304801670146, + "grad_norm": 1.1630206108093262, + "learning_rate": 1.7732803520244866e-06, + "loss": 0.6381, + "step": 2314 + }, + { + "epoch": 3.6278705636743216, + "grad_norm": 1.2039670944213867, + "learning_rate": 1.7712648295827395e-06, + "loss": 0.6972, + "step": 2315 + }, + { + "epoch": 3.6294363256784967, + "grad_norm": 1.2051432132720947, + "learning_rate": 1.769249824649677e-06, + "loss": 0.6404, + "step": 2316 + }, + { + "epoch": 3.6310020876826723, + "grad_norm": 1.2452560663223267, + "learning_rate": 1.7672353386562486e-06, + "loss": 0.6057, + "step": 2317 + }, + { + "epoch": 3.632567849686848, + "grad_norm": 1.1576342582702637, + "learning_rate": 1.7652213730330336e-06, + "loss": 0.6235, + "step": 2318 + }, + { + "epoch": 3.634133611691023, + "grad_norm": 1.3302104473114014, + "learning_rate": 1.7632079292102412e-06, + "loss": 0.6464, + "step": 2319 + }, + { + "epoch": 3.635699373695198, + "grad_norm": 1.1526870727539062, + "learning_rate": 1.7611950086177132e-06, + "loss": 0.6619, + "step": 2320 + }, + { + "epoch": 3.6372651356993737, + "grad_norm": 1.1325047016143799, + "learning_rate": 1.7591826126849179e-06, + "loss": 0.6815, + "step": 2321 + }, + { + "epoch": 3.6388308977035493, + "grad_norm": 1.2510312795639038, + "learning_rate": 1.7571707428409496e-06, + "loss": 0.6344, + "step": 2322 + }, + { + "epoch": 3.6403966597077244, + "grad_norm": 1.3015072345733643, + "learning_rate": 1.7551594005145323e-06, + "loss": 0.6082, + "step": 2323 + }, + { + "epoch": 3.6419624217118995, + "grad_norm": 1.1332790851593018, + "learning_rate": 1.7531485871340122e-06, + "loss": 0.6854, + "step": 2324 + }, + { + "epoch": 3.643528183716075, + "grad_norm": 1.1714104413986206, + "learning_rate": 1.7511383041273617e-06, + "loss": 0.6497, + "step": 2325 + }, + { + "epoch": 3.6450939457202507, + "grad_norm": 1.1853827238082886, + "learning_rate": 1.7491285529221768e-06, + "loss": 0.5655, + "step": 2326 + }, + { + "epoch": 3.646659707724426, + "grad_norm": 1.2981195449829102, + "learning_rate": 1.7471193349456746e-06, + "loss": 0.6327, + "step": 2327 + }, + { + "epoch": 3.6482254697286014, + "grad_norm": 1.2444530725479126, + "learning_rate": 1.7451106516246932e-06, + "loss": 0.643, + "step": 2328 + }, + { + "epoch": 3.6497912317327765, + "grad_norm": 1.1355667114257812, + "learning_rate": 1.7431025043856931e-06, + "loss": 0.6439, + "step": 2329 + }, + { + "epoch": 3.651356993736952, + "grad_norm": 1.1224238872528076, + "learning_rate": 1.7410948946547536e-06, + "loss": 0.6606, + "step": 2330 + }, + { + "epoch": 3.6529227557411272, + "grad_norm": 1.163265347480774, + "learning_rate": 1.739087823857569e-06, + "loss": 0.5989, + "step": 2331 + }, + { + "epoch": 3.654488517745303, + "grad_norm": 1.3157579898834229, + "learning_rate": 1.7370812934194565e-06, + "loss": 0.6818, + "step": 2332 + }, + { + "epoch": 3.656054279749478, + "grad_norm": 1.1727818250656128, + "learning_rate": 1.735075304765344e-06, + "loss": 0.6271, + "step": 2333 + }, + { + "epoch": 3.6576200417536535, + "grad_norm": 1.2371114492416382, + "learning_rate": 1.733069859319778e-06, + "loss": 0.667, + "step": 2334 + }, + { + "epoch": 3.6591858037578286, + "grad_norm": 1.210724115371704, + "learning_rate": 1.73106495850692e-06, + "loss": 0.725, + "step": 2335 + }, + { + "epoch": 3.6607515657620042, + "grad_norm": 1.1199427843093872, + "learning_rate": 1.7290606037505417e-06, + "loss": 0.6634, + "step": 2336 + }, + { + "epoch": 3.6623173277661794, + "grad_norm": 1.2661418914794922, + "learning_rate": 1.7270567964740275e-06, + "loss": 0.6259, + "step": 2337 + }, + { + "epoch": 3.663883089770355, + "grad_norm": 1.2222251892089844, + "learning_rate": 1.7250535381003758e-06, + "loss": 0.6082, + "step": 2338 + }, + { + "epoch": 3.6654488517745305, + "grad_norm": 1.1748756170272827, + "learning_rate": 1.7230508300521935e-06, + "loss": 0.6017, + "step": 2339 + }, + { + "epoch": 3.6670146137787056, + "grad_norm": 1.2061327695846558, + "learning_rate": 1.7210486737516947e-06, + "loss": 0.6196, + "step": 2340 + }, + { + "epoch": 3.6685803757828808, + "grad_norm": 1.168503999710083, + "learning_rate": 1.7190470706207058e-06, + "loss": 0.6709, + "step": 2341 + }, + { + "epoch": 3.6701461377870563, + "grad_norm": 1.1397614479064941, + "learning_rate": 1.7170460220806563e-06, + "loss": 0.6043, + "step": 2342 + }, + { + "epoch": 3.671711899791232, + "grad_norm": 1.1928285360336304, + "learning_rate": 1.7150455295525842e-06, + "loss": 0.5985, + "step": 2343 + }, + { + "epoch": 3.673277661795407, + "grad_norm": 1.205994725227356, + "learning_rate": 1.713045594457133e-06, + "loss": 0.6461, + "step": 2344 + }, + { + "epoch": 3.6748434237995826, + "grad_norm": 1.192114233970642, + "learning_rate": 1.7110462182145494e-06, + "loss": 0.6461, + "step": 2345 + }, + { + "epoch": 3.6764091858037578, + "grad_norm": 1.2672173976898193, + "learning_rate": 1.7090474022446812e-06, + "loss": 0.6623, + "step": 2346 + }, + { + "epoch": 3.6779749478079333, + "grad_norm": 1.1894220113754272, + "learning_rate": 1.707049147966983e-06, + "loss": 0.6687, + "step": 2347 + }, + { + "epoch": 3.6795407098121085, + "grad_norm": 1.170541763305664, + "learning_rate": 1.7050514568005072e-06, + "loss": 0.6258, + "step": 2348 + }, + { + "epoch": 3.681106471816284, + "grad_norm": 1.2120347023010254, + "learning_rate": 1.7030543301639057e-06, + "loss": 0.5854, + "step": 2349 + }, + { + "epoch": 3.682672233820459, + "grad_norm": 1.1434928178787231, + "learning_rate": 1.7010577694754332e-06, + "loss": 0.6177, + "step": 2350 + }, + { + "epoch": 3.6842379958246347, + "grad_norm": 1.2849304676055908, + "learning_rate": 1.6990617761529388e-06, + "loss": 0.5642, + "step": 2351 + }, + { + "epoch": 3.68580375782881, + "grad_norm": 1.1992167234420776, + "learning_rate": 1.6970663516138692e-06, + "loss": 0.6594, + "step": 2352 + }, + { + "epoch": 3.6873695198329854, + "grad_norm": 1.2290526628494263, + "learning_rate": 1.6950714972752702e-06, + "loss": 0.6506, + "step": 2353 + }, + { + "epoch": 3.6889352818371606, + "grad_norm": 1.1701496839523315, + "learning_rate": 1.6930772145537794e-06, + "loss": 0.5947, + "step": 2354 + }, + { + "epoch": 3.690501043841336, + "grad_norm": 1.2298767566680908, + "learning_rate": 1.6910835048656282e-06, + "loss": 0.6824, + "step": 2355 + }, + { + "epoch": 3.6920668058455117, + "grad_norm": 1.1752413511276245, + "learning_rate": 1.6890903696266447e-06, + "loss": 0.5763, + "step": 2356 + }, + { + "epoch": 3.693632567849687, + "grad_norm": 1.174170732498169, + "learning_rate": 1.6870978102522462e-06, + "loss": 0.5934, + "step": 2357 + }, + { + "epoch": 3.695198329853862, + "grad_norm": 1.1316381692886353, + "learning_rate": 1.68510582815744e-06, + "loss": 0.5857, + "step": 2358 + }, + { + "epoch": 3.6967640918580376, + "grad_norm": 1.2187336683273315, + "learning_rate": 1.6831144247568274e-06, + "loss": 0.6832, + "step": 2359 + }, + { + "epoch": 3.698329853862213, + "grad_norm": 1.2257745265960693, + "learning_rate": 1.6811236014645949e-06, + "loss": 0.6001, + "step": 2360 + }, + { + "epoch": 3.6998956158663883, + "grad_norm": 1.2153297662734985, + "learning_rate": 1.6791333596945197e-06, + "loss": 0.6452, + "step": 2361 + }, + { + "epoch": 3.701461377870564, + "grad_norm": 1.203791618347168, + "learning_rate": 1.677143700859963e-06, + "loss": 0.6032, + "step": 2362 + }, + { + "epoch": 3.703027139874739, + "grad_norm": 1.215394139289856, + "learning_rate": 1.6751546263738761e-06, + "loss": 0.694, + "step": 2363 + }, + { + "epoch": 3.7045929018789145, + "grad_norm": 1.1475002765655518, + "learning_rate": 1.6731661376487923e-06, + "loss": 0.6627, + "step": 2364 + }, + { + "epoch": 3.7061586638830897, + "grad_norm": 1.2240904569625854, + "learning_rate": 1.6711782360968288e-06, + "loss": 0.6605, + "step": 2365 + }, + { + "epoch": 3.7077244258872653, + "grad_norm": 1.1701643466949463, + "learning_rate": 1.6691909231296888e-06, + "loss": 0.5705, + "step": 2366 + }, + { + "epoch": 3.7092901878914404, + "grad_norm": 1.1641205549240112, + "learning_rate": 1.6672042001586538e-06, + "loss": 0.6604, + "step": 2367 + }, + { + "epoch": 3.710855949895616, + "grad_norm": 1.120191216468811, + "learning_rate": 1.6652180685945884e-06, + "loss": 0.567, + "step": 2368 + }, + { + "epoch": 3.712421711899791, + "grad_norm": 1.1677566766738892, + "learning_rate": 1.663232529847938e-06, + "loss": 0.6484, + "step": 2369 + }, + { + "epoch": 3.7139874739039667, + "grad_norm": 1.1286911964416504, + "learning_rate": 1.6612475853287252e-06, + "loss": 0.5604, + "step": 2370 + }, + { + "epoch": 3.715553235908142, + "grad_norm": 1.2938463687896729, + "learning_rate": 1.65926323644655e-06, + "loss": 0.683, + "step": 2371 + }, + { + "epoch": 3.7171189979123174, + "grad_norm": 1.160536527633667, + "learning_rate": 1.6572794846105919e-06, + "loss": 0.6467, + "step": 2372 + }, + { + "epoch": 3.718684759916493, + "grad_norm": 1.1587703227996826, + "learning_rate": 1.655296331229606e-06, + "loss": 0.6386, + "step": 2373 + }, + { + "epoch": 3.720250521920668, + "grad_norm": 1.2075384855270386, + "learning_rate": 1.6533137777119196e-06, + "loss": 0.6331, + "step": 2374 + }, + { + "epoch": 3.721816283924843, + "grad_norm": 1.2803996801376343, + "learning_rate": 1.651331825465438e-06, + "loss": 0.6421, + "step": 2375 + }, + { + "epoch": 3.723382045929019, + "grad_norm": 1.1343461275100708, + "learning_rate": 1.6493504758976358e-06, + "loss": 0.6807, + "step": 2376 + }, + { + "epoch": 3.7249478079331944, + "grad_norm": 1.211895227432251, + "learning_rate": 1.6473697304155622e-06, + "loss": 0.6533, + "step": 2377 + }, + { + "epoch": 3.7265135699373695, + "grad_norm": 1.191689372062683, + "learning_rate": 1.6453895904258372e-06, + "loss": 0.6559, + "step": 2378 + }, + { + "epoch": 3.7280793319415446, + "grad_norm": 1.2990779876708984, + "learning_rate": 1.6434100573346496e-06, + "loss": 0.6354, + "step": 2379 + }, + { + "epoch": 3.72964509394572, + "grad_norm": 1.2150678634643555, + "learning_rate": 1.6414311325477567e-06, + "loss": 0.6438, + "step": 2380 + }, + { + "epoch": 3.7312108559498958, + "grad_norm": 1.1220753192901611, + "learning_rate": 1.639452817470486e-06, + "loss": 0.6315, + "step": 2381 + }, + { + "epoch": 3.732776617954071, + "grad_norm": 1.096580147743225, + "learning_rate": 1.6374751135077315e-06, + "loss": 0.6525, + "step": 2382 + }, + { + "epoch": 3.7343423799582465, + "grad_norm": 1.2590006589889526, + "learning_rate": 1.6354980220639504e-06, + "loss": 0.6609, + "step": 2383 + }, + { + "epoch": 3.7359081419624216, + "grad_norm": 1.2841553688049316, + "learning_rate": 1.6335215445431702e-06, + "loss": 0.6861, + "step": 2384 + }, + { + "epoch": 3.737473903966597, + "grad_norm": 1.1383644342422485, + "learning_rate": 1.6315456823489766e-06, + "loss": 0.6707, + "step": 2385 + }, + { + "epoch": 3.7390396659707723, + "grad_norm": 1.3325951099395752, + "learning_rate": 1.6295704368845221e-06, + "loss": 0.5672, + "step": 2386 + }, + { + "epoch": 3.740605427974948, + "grad_norm": 1.2384133338928223, + "learning_rate": 1.6275958095525217e-06, + "loss": 0.6274, + "step": 2387 + }, + { + "epoch": 3.742171189979123, + "grad_norm": 1.173836588859558, + "learning_rate": 1.6256218017552484e-06, + "loss": 0.5975, + "step": 2388 + }, + { + "epoch": 3.7437369519832986, + "grad_norm": 1.1901642084121704, + "learning_rate": 1.623648414894536e-06, + "loss": 0.6735, + "step": 2389 + }, + { + "epoch": 3.745302713987474, + "grad_norm": 1.2956198453903198, + "learning_rate": 1.6216756503717795e-06, + "loss": 0.6171, + "step": 2390 + }, + { + "epoch": 3.7468684759916493, + "grad_norm": 1.173884391784668, + "learning_rate": 1.619703509587931e-06, + "loss": 0.6375, + "step": 2391 + }, + { + "epoch": 3.7484342379958244, + "grad_norm": 1.1478065252304077, + "learning_rate": 1.6177319939434974e-06, + "loss": 0.6806, + "step": 2392 + }, + { + "epoch": 3.75, + "grad_norm": 1.1560224294662476, + "learning_rate": 1.6157611048385457e-06, + "loss": 0.5867, + "step": 2393 + }, + { + "epoch": 3.7515657620041756, + "grad_norm": 1.1529253721237183, + "learning_rate": 1.6137908436726943e-06, + "loss": 0.686, + "step": 2394 + }, + { + "epoch": 3.7531315240083507, + "grad_norm": 1.2027194499969482, + "learning_rate": 1.6118212118451171e-06, + "loss": 0.6579, + "step": 2395 + }, + { + "epoch": 3.754697286012526, + "grad_norm": 1.1857298612594604, + "learning_rate": 1.6098522107545426e-06, + "loss": 0.6274, + "step": 2396 + }, + { + "epoch": 3.7562630480167014, + "grad_norm": 1.2853041887283325, + "learning_rate": 1.6078838417992487e-06, + "loss": 0.6866, + "step": 2397 + }, + { + "epoch": 3.757828810020877, + "grad_norm": 1.1234532594680786, + "learning_rate": 1.6059161063770651e-06, + "loss": 0.5872, + "step": 2398 + }, + { + "epoch": 3.759394572025052, + "grad_norm": 1.2463502883911133, + "learning_rate": 1.603949005885373e-06, + "loss": 0.6566, + "step": 2399 + }, + { + "epoch": 3.7609603340292277, + "grad_norm": 1.1617919206619263, + "learning_rate": 1.6019825417211022e-06, + "loss": 0.6428, + "step": 2400 + }, + { + "epoch": 3.762526096033403, + "grad_norm": 1.1768699884414673, + "learning_rate": 1.6000167152807288e-06, + "loss": 0.6147, + "step": 2401 + }, + { + "epoch": 3.7640918580375784, + "grad_norm": 1.2775728702545166, + "learning_rate": 1.598051527960279e-06, + "loss": 0.6594, + "step": 2402 + }, + { + "epoch": 3.7656576200417535, + "grad_norm": 1.2250229120254517, + "learning_rate": 1.5960869811553225e-06, + "loss": 0.6844, + "step": 2403 + }, + { + "epoch": 3.767223382045929, + "grad_norm": 1.3038227558135986, + "learning_rate": 1.594123076260975e-06, + "loss": 0.6515, + "step": 2404 + }, + { + "epoch": 3.7687891440501042, + "grad_norm": 1.2010571956634521, + "learning_rate": 1.592159814671898e-06, + "loss": 0.6293, + "step": 2405 + }, + { + "epoch": 3.77035490605428, + "grad_norm": 1.2710856199264526, + "learning_rate": 1.590197197782294e-06, + "loss": 0.6096, + "step": 2406 + }, + { + "epoch": 3.7719206680584554, + "grad_norm": 1.2024561166763306, + "learning_rate": 1.588235226985908e-06, + "loss": 0.6227, + "step": 2407 + }, + { + "epoch": 3.7734864300626305, + "grad_norm": 1.1803232431411743, + "learning_rate": 1.5862739036760262e-06, + "loss": 0.6251, + "step": 2408 + }, + { + "epoch": 3.7750521920668056, + "grad_norm": 1.2758967876434326, + "learning_rate": 1.5843132292454765e-06, + "loss": 0.6948, + "step": 2409 + }, + { + "epoch": 3.776617954070981, + "grad_norm": 1.2238482236862183, + "learning_rate": 1.5823532050866242e-06, + "loss": 0.6186, + "step": 2410 + }, + { + "epoch": 3.778183716075157, + "grad_norm": 1.24042546749115, + "learning_rate": 1.5803938325913732e-06, + "loss": 0.6261, + "step": 2411 + }, + { + "epoch": 3.779749478079332, + "grad_norm": 1.137797474861145, + "learning_rate": 1.578435113151166e-06, + "loss": 0.5794, + "step": 2412 + }, + { + "epoch": 3.781315240083507, + "grad_norm": 1.2196292877197266, + "learning_rate": 1.5764770481569796e-06, + "loss": 0.64, + "step": 2413 + }, + { + "epoch": 3.7828810020876826, + "grad_norm": 1.1684080362319946, + "learning_rate": 1.5745196389993259e-06, + "loss": 0.5861, + "step": 2414 + }, + { + "epoch": 3.784446764091858, + "grad_norm": 1.2441654205322266, + "learning_rate": 1.5725628870682532e-06, + "loss": 0.6084, + "step": 2415 + }, + { + "epoch": 3.7860125260960333, + "grad_norm": 1.187728762626648, + "learning_rate": 1.5706067937533427e-06, + "loss": 0.6436, + "step": 2416 + }, + { + "epoch": 3.787578288100209, + "grad_norm": 1.2837245464324951, + "learning_rate": 1.5686513604437043e-06, + "loss": 0.5967, + "step": 2417 + }, + { + "epoch": 3.789144050104384, + "grad_norm": 1.1892480850219727, + "learning_rate": 1.566696588527985e-06, + "loss": 0.6547, + "step": 2418 + }, + { + "epoch": 3.7907098121085596, + "grad_norm": 1.2308993339538574, + "learning_rate": 1.5647424793943579e-06, + "loss": 0.6512, + "step": 2419 + }, + { + "epoch": 3.7922755741127347, + "grad_norm": 1.3206483125686646, + "learning_rate": 1.5627890344305256e-06, + "loss": 0.6845, + "step": 2420 + }, + { + "epoch": 3.7938413361169103, + "grad_norm": 1.151696801185608, + "learning_rate": 1.560836255023722e-06, + "loss": 0.597, + "step": 2421 + }, + { + "epoch": 3.7954070981210855, + "grad_norm": 1.2277007102966309, + "learning_rate": 1.5588841425607056e-06, + "loss": 0.6035, + "step": 2422 + }, + { + "epoch": 3.796972860125261, + "grad_norm": 1.2150050401687622, + "learning_rate": 1.556932698427761e-06, + "loss": 0.6679, + "step": 2423 + }, + { + "epoch": 3.798538622129436, + "grad_norm": 1.075766921043396, + "learning_rate": 1.5549819240107012e-06, + "loss": 0.6112, + "step": 2424 + }, + { + "epoch": 3.8001043841336117, + "grad_norm": 1.2026962041854858, + "learning_rate": 1.5530318206948607e-06, + "loss": 0.6104, + "step": 2425 + }, + { + "epoch": 3.801670146137787, + "grad_norm": 1.2209258079528809, + "learning_rate": 1.551082389865098e-06, + "loss": 0.6391, + "step": 2426 + }, + { + "epoch": 3.8032359081419624, + "grad_norm": 1.1685758829116821, + "learning_rate": 1.549133632905796e-06, + "loss": 0.6352, + "step": 2427 + }, + { + "epoch": 3.804801670146138, + "grad_norm": 1.1990925073623657, + "learning_rate": 1.547185551200856e-06, + "loss": 0.631, + "step": 2428 + }, + { + "epoch": 3.806367432150313, + "grad_norm": 1.1937427520751953, + "learning_rate": 1.5452381461337013e-06, + "loss": 0.6393, + "step": 2429 + }, + { + "epoch": 3.8079331941544883, + "grad_norm": 1.1711888313293457, + "learning_rate": 1.5432914190872757e-06, + "loss": 0.6415, + "step": 2430 + }, + { + "epoch": 3.809498956158664, + "grad_norm": 1.3688654899597168, + "learning_rate": 1.5413453714440404e-06, + "loss": 0.6711, + "step": 2431 + }, + { + "epoch": 3.8110647181628394, + "grad_norm": 1.1071898937225342, + "learning_rate": 1.5394000045859728e-06, + "loss": 0.5966, + "step": 2432 + }, + { + "epoch": 3.8126304801670146, + "grad_norm": 1.1659204959869385, + "learning_rate": 1.5374553198945696e-06, + "loss": 0.6672, + "step": 2433 + }, + { + "epoch": 3.81419624217119, + "grad_norm": 1.1961894035339355, + "learning_rate": 1.5355113187508413e-06, + "loss": 0.6229, + "step": 2434 + }, + { + "epoch": 3.8157620041753653, + "grad_norm": 1.2803820371627808, + "learning_rate": 1.5335680025353132e-06, + "loss": 0.657, + "step": 2435 + }, + { + "epoch": 3.817327766179541, + "grad_norm": 1.1563304662704468, + "learning_rate": 1.531625372628025e-06, + "loss": 0.6386, + "step": 2436 + }, + { + "epoch": 3.818893528183716, + "grad_norm": 1.2539798021316528, + "learning_rate": 1.5296834304085276e-06, + "loss": 0.5733, + "step": 2437 + }, + { + "epoch": 3.8204592901878915, + "grad_norm": 1.2615854740142822, + "learning_rate": 1.5277421772558842e-06, + "loss": 0.6073, + "step": 2438 + }, + { + "epoch": 3.8220250521920667, + "grad_norm": 1.2182807922363281, + "learning_rate": 1.5258016145486705e-06, + "loss": 0.6719, + "step": 2439 + }, + { + "epoch": 3.8235908141962422, + "grad_norm": 1.2267918586730957, + "learning_rate": 1.5238617436649689e-06, + "loss": 0.616, + "step": 2440 + }, + { + "epoch": 3.8251565762004174, + "grad_norm": 1.2033759355545044, + "learning_rate": 1.5219225659823706e-06, + "loss": 0.6759, + "step": 2441 + }, + { + "epoch": 3.826722338204593, + "grad_norm": 1.3071197271347046, + "learning_rate": 1.5199840828779777e-06, + "loss": 0.6672, + "step": 2442 + }, + { + "epoch": 3.828288100208768, + "grad_norm": 1.2482445240020752, + "learning_rate": 1.5180462957283971e-06, + "loss": 0.6363, + "step": 2443 + }, + { + "epoch": 3.8298538622129437, + "grad_norm": 1.2446298599243164, + "learning_rate": 1.5161092059097399e-06, + "loss": 0.6506, + "step": 2444 + }, + { + "epoch": 3.8314196242171192, + "grad_norm": 1.217212438583374, + "learning_rate": 1.5141728147976248e-06, + "loss": 0.7131, + "step": 2445 + }, + { + "epoch": 3.8329853862212944, + "grad_norm": 1.2086701393127441, + "learning_rate": 1.5122371237671726e-06, + "loss": 0.6107, + "step": 2446 + }, + { + "epoch": 3.8345511482254695, + "grad_norm": 1.244437336921692, + "learning_rate": 1.510302134193007e-06, + "loss": 0.5585, + "step": 2447 + }, + { + "epoch": 3.836116910229645, + "grad_norm": 1.1509884595870972, + "learning_rate": 1.5083678474492546e-06, + "loss": 0.5905, + "step": 2448 + }, + { + "epoch": 3.8376826722338206, + "grad_norm": 1.188522458076477, + "learning_rate": 1.5064342649095428e-06, + "loss": 0.6584, + "step": 2449 + }, + { + "epoch": 3.8392484342379958, + "grad_norm": 1.2903960943222046, + "learning_rate": 1.504501387946996e-06, + "loss": 0.6286, + "step": 2450 + }, + { + "epoch": 3.8408141962421714, + "grad_norm": 1.135545253753662, + "learning_rate": 1.5025692179342425e-06, + "loss": 0.6769, + "step": 2451 + }, + { + "epoch": 3.8423799582463465, + "grad_norm": 1.2375710010528564, + "learning_rate": 1.500637756243405e-06, + "loss": 0.6506, + "step": 2452 + }, + { + "epoch": 3.843945720250522, + "grad_norm": 1.7769895792007446, + "learning_rate": 1.4987070042461032e-06, + "loss": 0.6302, + "step": 2453 + }, + { + "epoch": 3.845511482254697, + "grad_norm": 1.0754798650741577, + "learning_rate": 1.4967769633134555e-06, + "loss": 0.6019, + "step": 2454 + }, + { + "epoch": 3.8470772442588728, + "grad_norm": 1.2362276315689087, + "learning_rate": 1.4948476348160723e-06, + "loss": 0.578, + "step": 2455 + }, + { + "epoch": 3.848643006263048, + "grad_norm": 1.1476119756698608, + "learning_rate": 1.4929190201240601e-06, + "loss": 0.5866, + "step": 2456 + }, + { + "epoch": 3.8502087682672235, + "grad_norm": 1.14879310131073, + "learning_rate": 1.4909911206070162e-06, + "loss": 0.6431, + "step": 2457 + }, + { + "epoch": 3.8517745302713986, + "grad_norm": 1.1493771076202393, + "learning_rate": 1.4890639376340336e-06, + "loss": 0.6297, + "step": 2458 + }, + { + "epoch": 3.853340292275574, + "grad_norm": 1.2373193502426147, + "learning_rate": 1.4871374725736936e-06, + "loss": 0.6387, + "step": 2459 + }, + { + "epoch": 3.8549060542797493, + "grad_norm": 1.2413831949234009, + "learning_rate": 1.485211726794068e-06, + "loss": 0.6569, + "step": 2460 + }, + { + "epoch": 3.856471816283925, + "grad_norm": 1.2246074676513672, + "learning_rate": 1.4832867016627194e-06, + "loss": 0.5908, + "step": 2461 + }, + { + "epoch": 3.8580375782881005, + "grad_norm": 1.1326402425765991, + "learning_rate": 1.4813623985466968e-06, + "loss": 0.6488, + "step": 2462 + }, + { + "epoch": 3.8596033402922756, + "grad_norm": 1.2424553632736206, + "learning_rate": 1.4794388188125364e-06, + "loss": 0.6299, + "step": 2463 + }, + { + "epoch": 3.8611691022964507, + "grad_norm": 1.2312356233596802, + "learning_rate": 1.4775159638262637e-06, + "loss": 0.6359, + "step": 2464 + }, + { + "epoch": 3.8627348643006263, + "grad_norm": 1.049404263496399, + "learning_rate": 1.475593834953386e-06, + "loss": 0.5736, + "step": 2465 + }, + { + "epoch": 3.864300626304802, + "grad_norm": 1.2294747829437256, + "learning_rate": 1.4736724335588952e-06, + "loss": 0.6342, + "step": 2466 + }, + { + "epoch": 3.865866388308977, + "grad_norm": 1.2269598245620728, + "learning_rate": 1.471751761007269e-06, + "loss": 0.6316, + "step": 2467 + }, + { + "epoch": 3.867432150313152, + "grad_norm": 1.138282299041748, + "learning_rate": 1.469831818662467e-06, + "loss": 0.6402, + "step": 2468 + }, + { + "epoch": 3.8689979123173277, + "grad_norm": 1.2380733489990234, + "learning_rate": 1.467912607887927e-06, + "loss": 0.639, + "step": 2469 + }, + { + "epoch": 3.8705636743215033, + "grad_norm": 1.109584093093872, + "learning_rate": 1.465994130046572e-06, + "loss": 0.6178, + "step": 2470 + }, + { + "epoch": 3.8721294363256784, + "grad_norm": 1.3927913904190063, + "learning_rate": 1.4640763865008006e-06, + "loss": 0.5844, + "step": 2471 + }, + { + "epoch": 3.873695198329854, + "grad_norm": 1.1995818614959717, + "learning_rate": 1.462159378612492e-06, + "loss": 0.6532, + "step": 2472 + }, + { + "epoch": 3.875260960334029, + "grad_norm": 1.244868278503418, + "learning_rate": 1.4602431077430035e-06, + "loss": 0.5823, + "step": 2473 + }, + { + "epoch": 3.8768267223382047, + "grad_norm": 1.2030248641967773, + "learning_rate": 1.4583275752531672e-06, + "loss": 0.6096, + "step": 2474 + }, + { + "epoch": 3.87839248434238, + "grad_norm": 1.1157633066177368, + "learning_rate": 1.4564127825032904e-06, + "loss": 0.6433, + "step": 2475 + }, + { + "epoch": 3.8799582463465554, + "grad_norm": 1.1652717590332031, + "learning_rate": 1.4544987308531594e-06, + "loss": 0.5917, + "step": 2476 + }, + { + "epoch": 3.8815240083507305, + "grad_norm": 1.2521084547042847, + "learning_rate": 1.452585421662029e-06, + "loss": 0.6254, + "step": 2477 + }, + { + "epoch": 3.883089770354906, + "grad_norm": 1.2177410125732422, + "learning_rate": 1.4506728562886285e-06, + "loss": 0.6565, + "step": 2478 + }, + { + "epoch": 3.8846555323590817, + "grad_norm": 1.2197600603103638, + "learning_rate": 1.4487610360911606e-06, + "loss": 0.6001, + "step": 2479 + }, + { + "epoch": 3.886221294363257, + "grad_norm": 1.186809778213501, + "learning_rate": 1.4468499624272972e-06, + "loss": 0.611, + "step": 2480 + }, + { + "epoch": 3.887787056367432, + "grad_norm": 1.2346223592758179, + "learning_rate": 1.4449396366541802e-06, + "loss": 0.7188, + "step": 2481 + }, + { + "epoch": 3.8893528183716075, + "grad_norm": 1.148040533065796, + "learning_rate": 1.4430300601284214e-06, + "loss": 0.6711, + "step": 2482 + }, + { + "epoch": 3.890918580375783, + "grad_norm": 1.2020211219787598, + "learning_rate": 1.4411212342060988e-06, + "loss": 0.6765, + "step": 2483 + }, + { + "epoch": 3.892484342379958, + "grad_norm": 1.2378863096237183, + "learning_rate": 1.439213160242759e-06, + "loss": 0.6814, + "step": 2484 + }, + { + "epoch": 3.8940501043841333, + "grad_norm": 1.2072421312332153, + "learning_rate": 1.4373058395934136e-06, + "loss": 0.6178, + "step": 2485 + }, + { + "epoch": 3.895615866388309, + "grad_norm": 1.3510164022445679, + "learning_rate": 1.4353992736125396e-06, + "loss": 0.6775, + "step": 2486 + }, + { + "epoch": 3.8971816283924845, + "grad_norm": 1.2395039796829224, + "learning_rate": 1.4334934636540782e-06, + "loss": 0.5773, + "step": 2487 + }, + { + "epoch": 3.8987473903966596, + "grad_norm": 1.208178162574768, + "learning_rate": 1.4315884110714337e-06, + "loss": 0.5234, + "step": 2488 + }, + { + "epoch": 3.900313152400835, + "grad_norm": 1.2231117486953735, + "learning_rate": 1.429684117217473e-06, + "loss": 0.6405, + "step": 2489 + }, + { + "epoch": 3.9018789144050103, + "grad_norm": 1.1081833839416504, + "learning_rate": 1.4277805834445207e-06, + "loss": 0.6349, + "step": 2490 + }, + { + "epoch": 3.903444676409186, + "grad_norm": 1.3395150899887085, + "learning_rate": 1.4258778111043674e-06, + "loss": 0.6236, + "step": 2491 + }, + { + "epoch": 3.905010438413361, + "grad_norm": 1.2174855470657349, + "learning_rate": 1.4239758015482607e-06, + "loss": 0.667, + "step": 2492 + }, + { + "epoch": 3.9065762004175366, + "grad_norm": 1.3547643423080444, + "learning_rate": 1.4220745561269022e-06, + "loss": 0.6654, + "step": 2493 + }, + { + "epoch": 3.9081419624217117, + "grad_norm": 1.1408005952835083, + "learning_rate": 1.4201740761904591e-06, + "loss": 0.638, + "step": 2494 + }, + { + "epoch": 3.9097077244258873, + "grad_norm": 1.299989938735962, + "learning_rate": 1.418274363088547e-06, + "loss": 0.5875, + "step": 2495 + }, + { + "epoch": 3.911273486430063, + "grad_norm": 1.1767139434814453, + "learning_rate": 1.4163754181702408e-06, + "loss": 0.6723, + "step": 2496 + }, + { + "epoch": 3.912839248434238, + "grad_norm": 1.2104045152664185, + "learning_rate": 1.414477242784072e-06, + "loss": 0.5944, + "step": 2497 + }, + { + "epoch": 3.914405010438413, + "grad_norm": 1.2391446828842163, + "learning_rate": 1.4125798382780207e-06, + "loss": 0.6747, + "step": 2498 + }, + { + "epoch": 3.9159707724425887, + "grad_norm": 1.1312291622161865, + "learning_rate": 1.4106832059995224e-06, + "loss": 0.6705, + "step": 2499 + }, + { + "epoch": 3.9175365344467643, + "grad_norm": 1.2023922204971313, + "learning_rate": 1.4087873472954638e-06, + "loss": 0.6417, + "step": 2500 + }, + { + "epoch": 3.9191022964509394, + "grad_norm": 1.2477304935455322, + "learning_rate": 1.4068922635121828e-06, + "loss": 0.6471, + "step": 2501 + }, + { + "epoch": 3.9206680584551146, + "grad_norm": 1.140338659286499, + "learning_rate": 1.4049979559954666e-06, + "loss": 0.6042, + "step": 2502 + }, + { + "epoch": 3.92223382045929, + "grad_norm": 1.1643264293670654, + "learning_rate": 1.4031044260905486e-06, + "loss": 0.7038, + "step": 2503 + }, + { + "epoch": 3.9237995824634657, + "grad_norm": 1.2577917575836182, + "learning_rate": 1.4012116751421145e-06, + "loss": 0.6826, + "step": 2504 + }, + { + "epoch": 3.925365344467641, + "grad_norm": 1.1324503421783447, + "learning_rate": 1.3993197044942958e-06, + "loss": 0.6337, + "step": 2505 + }, + { + "epoch": 3.9269311064718164, + "grad_norm": 1.3015128374099731, + "learning_rate": 1.3974285154906647e-06, + "loss": 0.6443, + "step": 2506 + }, + { + "epoch": 3.9284968684759916, + "grad_norm": 1.139436960220337, + "learning_rate": 1.3955381094742464e-06, + "loss": 0.6569, + "step": 2507 + }, + { + "epoch": 3.930062630480167, + "grad_norm": 1.2854009866714478, + "learning_rate": 1.393648487787504e-06, + "loss": 0.6191, + "step": 2508 + }, + { + "epoch": 3.9316283924843423, + "grad_norm": 1.2040328979492188, + "learning_rate": 1.3917596517723441e-06, + "loss": 0.6355, + "step": 2509 + }, + { + "epoch": 3.933194154488518, + "grad_norm": 1.2754182815551758, + "learning_rate": 1.3898716027701214e-06, + "loss": 0.658, + "step": 2510 + }, + { + "epoch": 3.934759916492693, + "grad_norm": 1.2027299404144287, + "learning_rate": 1.387984342121623e-06, + "loss": 0.5478, + "step": 2511 + }, + { + "epoch": 3.9363256784968685, + "grad_norm": 1.1748281717300415, + "learning_rate": 1.3860978711670819e-06, + "loss": 0.6427, + "step": 2512 + }, + { + "epoch": 3.937891440501044, + "grad_norm": 1.209522008895874, + "learning_rate": 1.3842121912461686e-06, + "loss": 0.6234, + "step": 2513 + }, + { + "epoch": 3.9394572025052192, + "grad_norm": 1.1207284927368164, + "learning_rate": 1.382327303697992e-06, + "loss": 0.6092, + "step": 2514 + }, + { + "epoch": 3.9410229645093944, + "grad_norm": 1.2785614728927612, + "learning_rate": 1.3804432098610982e-06, + "loss": 0.6507, + "step": 2515 + }, + { + "epoch": 3.94258872651357, + "grad_norm": 1.2241697311401367, + "learning_rate": 1.37855991107347e-06, + "loss": 0.5863, + "step": 2516 + }, + { + "epoch": 3.9441544885177455, + "grad_norm": 1.1810356378555298, + "learning_rate": 1.3766774086725246e-06, + "loss": 0.6718, + "step": 2517 + }, + { + "epoch": 3.9457202505219207, + "grad_norm": 1.1314932107925415, + "learning_rate": 1.3747957039951152e-06, + "loss": 0.6726, + "step": 2518 + }, + { + "epoch": 3.947286012526096, + "grad_norm": 1.2174315452575684, + "learning_rate": 1.3729147983775272e-06, + "loss": 0.6201, + "step": 2519 + }, + { + "epoch": 3.9488517745302714, + "grad_norm": 1.2153561115264893, + "learning_rate": 1.371034693155479e-06, + "loss": 0.6499, + "step": 2520 + }, + { + "epoch": 3.950417536534447, + "grad_norm": 1.219773292541504, + "learning_rate": 1.3691553896641213e-06, + "loss": 0.6046, + "step": 2521 + }, + { + "epoch": 3.951983298538622, + "grad_norm": 1.284940481185913, + "learning_rate": 1.367276889238034e-06, + "loss": 0.6899, + "step": 2522 + }, + { + "epoch": 3.9535490605427976, + "grad_norm": 1.356799602508545, + "learning_rate": 1.3653991932112287e-06, + "loss": 0.6861, + "step": 2523 + }, + { + "epoch": 3.9551148225469728, + "grad_norm": 1.166818618774414, + "learning_rate": 1.3635223029171418e-06, + "loss": 0.5631, + "step": 2524 + }, + { + "epoch": 3.9566805845511483, + "grad_norm": 1.2792567014694214, + "learning_rate": 1.3616462196886427e-06, + "loss": 0.6471, + "step": 2525 + }, + { + "epoch": 3.9582463465553235, + "grad_norm": 1.2961795330047607, + "learning_rate": 1.3597709448580257e-06, + "loss": 0.6074, + "step": 2526 + }, + { + "epoch": 3.959812108559499, + "grad_norm": 1.2090468406677246, + "learning_rate": 1.3578964797570077e-06, + "loss": 0.6225, + "step": 2527 + }, + { + "epoch": 3.961377870563674, + "grad_norm": 1.1482014656066895, + "learning_rate": 1.3560228257167365e-06, + "loss": 0.5968, + "step": 2528 + }, + { + "epoch": 3.9629436325678498, + "grad_norm": 1.2505629062652588, + "learning_rate": 1.354149984067779e-06, + "loss": 0.6198, + "step": 2529 + }, + { + "epoch": 3.964509394572025, + "grad_norm": 1.1473637819290161, + "learning_rate": 1.3522779561401262e-06, + "loss": 0.5933, + "step": 2530 + }, + { + "epoch": 3.9660751565762005, + "grad_norm": 1.2111499309539795, + "learning_rate": 1.3504067432631954e-06, + "loss": 0.6173, + "step": 2531 + }, + { + "epoch": 3.9676409185803756, + "grad_norm": 1.2065558433532715, + "learning_rate": 1.3485363467658186e-06, + "loss": 0.6819, + "step": 2532 + }, + { + "epoch": 3.969206680584551, + "grad_norm": 1.3822040557861328, + "learning_rate": 1.3466667679762525e-06, + "loss": 0.6258, + "step": 2533 + }, + { + "epoch": 3.9707724425887267, + "grad_norm": 1.2183210849761963, + "learning_rate": 1.3447980082221719e-06, + "loss": 0.6501, + "step": 2534 + }, + { + "epoch": 3.972338204592902, + "grad_norm": 1.1406798362731934, + "learning_rate": 1.342930068830669e-06, + "loss": 0.5535, + "step": 2535 + }, + { + "epoch": 3.973903966597077, + "grad_norm": 1.2553014755249023, + "learning_rate": 1.3410629511282553e-06, + "loss": 0.6809, + "step": 2536 + }, + { + "epoch": 3.9754697286012526, + "grad_norm": 1.2471221685409546, + "learning_rate": 1.3391966564408571e-06, + "loss": 0.635, + "step": 2537 + }, + { + "epoch": 3.977035490605428, + "grad_norm": 1.318433165550232, + "learning_rate": 1.3373311860938177e-06, + "loss": 0.6571, + "step": 2538 + }, + { + "epoch": 3.9786012526096033, + "grad_norm": 1.2130142450332642, + "learning_rate": 1.335466541411893e-06, + "loss": 0.6643, + "step": 2539 + }, + { + "epoch": 3.980167014613779, + "grad_norm": 1.1460458040237427, + "learning_rate": 1.3336027237192551e-06, + "loss": 0.5837, + "step": 2540 + }, + { + "epoch": 3.981732776617954, + "grad_norm": 1.1360803842544556, + "learning_rate": 1.3317397343394879e-06, + "loss": 0.6432, + "step": 2541 + }, + { + "epoch": 3.9832985386221296, + "grad_norm": 1.1411229372024536, + "learning_rate": 1.3298775745955833e-06, + "loss": 0.5568, + "step": 2542 + }, + { + "epoch": 3.9848643006263047, + "grad_norm": 1.3322495222091675, + "learning_rate": 1.328016245809951e-06, + "loss": 0.7113, + "step": 2543 + }, + { + "epoch": 3.9864300626304803, + "grad_norm": 1.1256479024887085, + "learning_rate": 1.3261557493044067e-06, + "loss": 0.5875, + "step": 2544 + }, + { + "epoch": 3.9879958246346554, + "grad_norm": 1.1804993152618408, + "learning_rate": 1.3242960864001724e-06, + "loss": 0.6372, + "step": 2545 + }, + { + "epoch": 3.989561586638831, + "grad_norm": 1.1208686828613281, + "learning_rate": 1.3224372584178855e-06, + "loss": 0.4952, + "step": 2546 + }, + { + "epoch": 3.991127348643006, + "grad_norm": 1.2875077724456787, + "learning_rate": 1.3205792666775824e-06, + "loss": 0.6465, + "step": 2547 + }, + { + "epoch": 3.9926931106471817, + "grad_norm": 1.2553578615188599, + "learning_rate": 1.3187221124987107e-06, + "loss": 0.6053, + "step": 2548 + }, + { + "epoch": 3.994258872651357, + "grad_norm": 1.1742136478424072, + "learning_rate": 1.3168657972001214e-06, + "loss": 0.6711, + "step": 2549 + }, + { + "epoch": 3.9958246346555324, + "grad_norm": 1.184242844581604, + "learning_rate": 1.3150103221000709e-06, + "loss": 0.6405, + "step": 2550 + }, + { + "epoch": 3.997390396659708, + "grad_norm": 1.1263542175292969, + "learning_rate": 1.313155688516218e-06, + "loss": 0.5631, + "step": 2551 + }, + { + "epoch": 3.998956158663883, + "grad_norm": 1.2463663816452026, + "learning_rate": 1.3113018977656238e-06, + "loss": 0.5974, + "step": 2552 + }, + { + "epoch": 4.001565762004176, + "grad_norm": 2.222411870956421, + "learning_rate": 1.3094489511647512e-06, + "loss": 1.1917, + "step": 2553 + }, + { + "epoch": 4.003131524008351, + "grad_norm": 1.1565176248550415, + "learning_rate": 1.307596850029465e-06, + "loss": 0.6157, + "step": 2554 + }, + { + "epoch": 4.004697286012526, + "grad_norm": 1.3374539613723755, + "learning_rate": 1.3057455956750253e-06, + "loss": 0.548, + "step": 2555 + }, + { + "epoch": 4.006263048016701, + "grad_norm": 1.0805665254592896, + "learning_rate": 1.3038951894160962e-06, + "loss": 0.5639, + "step": 2556 + }, + { + "epoch": 4.007828810020877, + "grad_norm": 1.2551461458206177, + "learning_rate": 1.3020456325667374e-06, + "loss": 0.6393, + "step": 2557 + }, + { + "epoch": 4.009394572025053, + "grad_norm": 1.3844956159591675, + "learning_rate": 1.3001969264404024e-06, + "loss": 0.5836, + "step": 2558 + }, + { + "epoch": 4.010960334029227, + "grad_norm": 1.155824065208435, + "learning_rate": 1.298349072349947e-06, + "loss": 0.6122, + "step": 2559 + }, + { + "epoch": 4.012526096033403, + "grad_norm": 1.2663135528564453, + "learning_rate": 1.2965020716076148e-06, + "loss": 0.5728, + "step": 2560 + }, + { + "epoch": 4.014091858037578, + "grad_norm": 1.1352088451385498, + "learning_rate": 1.294655925525048e-06, + "loss": 0.6713, + "step": 2561 + }, + { + "epoch": 4.015657620041754, + "grad_norm": 1.179442286491394, + "learning_rate": 1.2928106354132823e-06, + "loss": 0.5363, + "step": 2562 + }, + { + "epoch": 4.017223382045929, + "grad_norm": 1.2838183641433716, + "learning_rate": 1.2909662025827413e-06, + "loss": 0.5278, + "step": 2563 + }, + { + "epoch": 4.018789144050104, + "grad_norm": 1.186926007270813, + "learning_rate": 1.289122628343244e-06, + "loss": 0.581, + "step": 2564 + }, + { + "epoch": 4.02035490605428, + "grad_norm": 1.1647127866744995, + "learning_rate": 1.2872799140039973e-06, + "loss": 0.5944, + "step": 2565 + }, + { + "epoch": 4.021920668058455, + "grad_norm": 1.1807323694229126, + "learning_rate": 1.2854380608735989e-06, + "loss": 0.6162, + "step": 2566 + }, + { + "epoch": 4.02348643006263, + "grad_norm": 1.1833971738815308, + "learning_rate": 1.2835970702600337e-06, + "loss": 0.6241, + "step": 2567 + }, + { + "epoch": 4.025052192066806, + "grad_norm": 1.1918176412582397, + "learning_rate": 1.2817569434706748e-06, + "loss": 0.6184, + "step": 2568 + }, + { + "epoch": 4.026617954070981, + "grad_norm": 1.275282859802246, + "learning_rate": 1.2799176818122816e-06, + "loss": 0.6285, + "step": 2569 + }, + { + "epoch": 4.028183716075157, + "grad_norm": 1.2189844846725464, + "learning_rate": 1.2780792865909996e-06, + "loss": 0.5431, + "step": 2570 + }, + { + "epoch": 4.029749478079332, + "grad_norm": 1.235913634300232, + "learning_rate": 1.2762417591123582e-06, + "loss": 0.6149, + "step": 2571 + }, + { + "epoch": 4.031315240083507, + "grad_norm": 1.160127878189087, + "learning_rate": 1.2744051006812712e-06, + "loss": 0.5061, + "step": 2572 + }, + { + "epoch": 4.032881002087683, + "grad_norm": 1.2041453123092651, + "learning_rate": 1.272569312602035e-06, + "loss": 0.636, + "step": 2573 + }, + { + "epoch": 4.034446764091858, + "grad_norm": 1.318177342414856, + "learning_rate": 1.2707343961783278e-06, + "loss": 0.6081, + "step": 2574 + }, + { + "epoch": 4.036012526096034, + "grad_norm": 1.0629124641418457, + "learning_rate": 1.2689003527132103e-06, + "loss": 0.6312, + "step": 2575 + }, + { + "epoch": 4.0375782881002085, + "grad_norm": 1.2917940616607666, + "learning_rate": 1.2670671835091187e-06, + "loss": 0.6204, + "step": 2576 + }, + { + "epoch": 4.039144050104384, + "grad_norm": 1.1925923824310303, + "learning_rate": 1.2652348898678742e-06, + "loss": 0.6608, + "step": 2577 + }, + { + "epoch": 4.04070981210856, + "grad_norm": 1.296000361442566, + "learning_rate": 1.2634034730906745e-06, + "loss": 0.613, + "step": 2578 + }, + { + "epoch": 4.042275574112735, + "grad_norm": 1.194916009902954, + "learning_rate": 1.26157293447809e-06, + "loss": 0.5946, + "step": 2579 + }, + { + "epoch": 4.04384133611691, + "grad_norm": 1.239607334136963, + "learning_rate": 1.2597432753300753e-06, + "loss": 0.6065, + "step": 2580 + }, + { + "epoch": 4.0454070981210855, + "grad_norm": 1.2236425876617432, + "learning_rate": 1.2579144969459536e-06, + "loss": 0.5722, + "step": 2581 + }, + { + "epoch": 4.046972860125261, + "grad_norm": 1.3422942161560059, + "learning_rate": 1.2560866006244249e-06, + "loss": 0.6717, + "step": 2582 + }, + { + "epoch": 4.048538622129437, + "grad_norm": 1.1085593700408936, + "learning_rate": 1.254259587663567e-06, + "loss": 0.6199, + "step": 2583 + }, + { + "epoch": 4.050104384133611, + "grad_norm": 1.3027156591415405, + "learning_rate": 1.252433459360823e-06, + "loss": 0.6361, + "step": 2584 + }, + { + "epoch": 4.051670146137787, + "grad_norm": 1.2550909519195557, + "learning_rate": 1.2506082170130121e-06, + "loss": 0.5996, + "step": 2585 + }, + { + "epoch": 4.053235908141962, + "grad_norm": 1.2809008359909058, + "learning_rate": 1.2487838619163245e-06, + "loss": 0.555, + "step": 2586 + }, + { + "epoch": 4.054801670146138, + "grad_norm": 1.0087321996688843, + "learning_rate": 1.246960395366319e-06, + "loss": 0.4473, + "step": 2587 + }, + { + "epoch": 4.056367432150314, + "grad_norm": 1.2337830066680908, + "learning_rate": 1.245137818657924e-06, + "loss": 0.5699, + "step": 2588 + }, + { + "epoch": 4.057933194154488, + "grad_norm": 1.1802204847335815, + "learning_rate": 1.243316133085436e-06, + "loss": 0.6677, + "step": 2589 + }, + { + "epoch": 4.059498956158664, + "grad_norm": 1.2031978368759155, + "learning_rate": 1.2414953399425182e-06, + "loss": 0.6389, + "step": 2590 + }, + { + "epoch": 4.061064718162839, + "grad_norm": 1.3534972667694092, + "learning_rate": 1.2396754405222003e-06, + "loss": 0.6177, + "step": 2591 + }, + { + "epoch": 4.062630480167015, + "grad_norm": 1.1125755310058594, + "learning_rate": 1.2378564361168776e-06, + "loss": 0.6383, + "step": 2592 + }, + { + "epoch": 4.06419624217119, + "grad_norm": 1.1830637454986572, + "learning_rate": 1.2360383280183108e-06, + "loss": 0.6341, + "step": 2593 + }, + { + "epoch": 4.065762004175365, + "grad_norm": 1.1648894548416138, + "learning_rate": 1.2342211175176206e-06, + "loss": 0.6327, + "step": 2594 + }, + { + "epoch": 4.067327766179541, + "grad_norm": 1.1454265117645264, + "learning_rate": 1.232404805905293e-06, + "loss": 0.6376, + "step": 2595 + }, + { + "epoch": 4.068893528183716, + "grad_norm": 1.153611660003662, + "learning_rate": 1.2305893944711773e-06, + "loss": 0.6032, + "step": 2596 + }, + { + "epoch": 4.070459290187891, + "grad_norm": 1.3515207767486572, + "learning_rate": 1.2287748845044792e-06, + "loss": 0.6064, + "step": 2597 + }, + { + "epoch": 4.072025052192067, + "grad_norm": 1.1163814067840576, + "learning_rate": 1.2269612772937672e-06, + "loss": 0.6582, + "step": 2598 + }, + { + "epoch": 4.073590814196242, + "grad_norm": 1.2641352415084839, + "learning_rate": 1.2251485741269683e-06, + "loss": 0.6221, + "step": 2599 + }, + { + "epoch": 4.075156576200418, + "grad_norm": 1.1293519735336304, + "learning_rate": 1.223336776291367e-06, + "loss": 0.5346, + "step": 2600 + }, + { + "epoch": 4.0767223382045925, + "grad_norm": 1.3557345867156982, + "learning_rate": 1.2215258850736051e-06, + "loss": 0.6602, + "step": 2601 + }, + { + "epoch": 4.078288100208768, + "grad_norm": 1.2822085618972778, + "learning_rate": 1.2197159017596803e-06, + "loss": 0.5706, + "step": 2602 + }, + { + "epoch": 4.079853862212944, + "grad_norm": 1.3728432655334473, + "learning_rate": 1.2179068276349462e-06, + "loss": 0.631, + "step": 2603 + }, + { + "epoch": 4.081419624217119, + "grad_norm": 1.1968104839324951, + "learning_rate": 1.21609866398411e-06, + "loss": 0.5861, + "step": 2604 + }, + { + "epoch": 4.082985386221294, + "grad_norm": 1.1587492227554321, + "learning_rate": 1.214291412091233e-06, + "loss": 0.6104, + "step": 2605 + }, + { + "epoch": 4.0845511482254695, + "grad_norm": 1.2656822204589844, + "learning_rate": 1.2124850732397294e-06, + "loss": 0.6327, + "step": 2606 + }, + { + "epoch": 4.086116910229645, + "grad_norm": 1.1670796871185303, + "learning_rate": 1.2106796487123615e-06, + "loss": 0.5624, + "step": 2607 + }, + { + "epoch": 4.087682672233821, + "grad_norm": 1.1668428182601929, + "learning_rate": 1.208875139791248e-06, + "loss": 0.6524, + "step": 2608 + }, + { + "epoch": 4.089248434237996, + "grad_norm": 1.2208465337753296, + "learning_rate": 1.2070715477578546e-06, + "loss": 0.5131, + "step": 2609 + }, + { + "epoch": 4.090814196242171, + "grad_norm": 1.1894104480743408, + "learning_rate": 1.2052688738929927e-06, + "loss": 0.6169, + "step": 2610 + }, + { + "epoch": 4.0923799582463465, + "grad_norm": 1.325358510017395, + "learning_rate": 1.2034671194768286e-06, + "loss": 0.5342, + "step": 2611 + }, + { + "epoch": 4.093945720250522, + "grad_norm": 1.1475489139556885, + "learning_rate": 1.201666285788869e-06, + "loss": 0.6409, + "step": 2612 + }, + { + "epoch": 4.095511482254698, + "grad_norm": 1.3095086812973022, + "learning_rate": 1.199866374107969e-06, + "loss": 0.6314, + "step": 2613 + }, + { + "epoch": 4.097077244258872, + "grad_norm": 1.2492880821228027, + "learning_rate": 1.198067385712333e-06, + "loss": 0.639, + "step": 2614 + }, + { + "epoch": 4.098643006263048, + "grad_norm": 1.179559588432312, + "learning_rate": 1.1962693218795036e-06, + "loss": 0.6343, + "step": 2615 + }, + { + "epoch": 4.1002087682672235, + "grad_norm": 1.2899236679077148, + "learning_rate": 1.1944721838863693e-06, + "loss": 0.6627, + "step": 2616 + }, + { + "epoch": 4.101774530271399, + "grad_norm": 1.3395811319351196, + "learning_rate": 1.192675973009162e-06, + "loss": 0.6222, + "step": 2617 + }, + { + "epoch": 4.103340292275574, + "grad_norm": 1.2343623638153076, + "learning_rate": 1.1908806905234544e-06, + "loss": 0.606, + "step": 2618 + }, + { + "epoch": 4.104906054279749, + "grad_norm": 1.300563931465149, + "learning_rate": 1.1890863377041595e-06, + "loss": 0.5701, + "step": 2619 + }, + { + "epoch": 4.106471816283925, + "grad_norm": 1.462306261062622, + "learning_rate": 1.187292915825531e-06, + "loss": 0.5738, + "step": 2620 + }, + { + "epoch": 4.1080375782881005, + "grad_norm": 1.21458101272583, + "learning_rate": 1.1855004261611611e-06, + "loss": 0.6335, + "step": 2621 + }, + { + "epoch": 4.109603340292275, + "grad_norm": 1.1758227348327637, + "learning_rate": 1.1837088699839798e-06, + "loss": 0.6325, + "step": 2622 + }, + { + "epoch": 4.111169102296451, + "grad_norm": 1.2418543100357056, + "learning_rate": 1.1819182485662542e-06, + "loss": 0.6509, + "step": 2623 + }, + { + "epoch": 4.112734864300626, + "grad_norm": 1.235479474067688, + "learning_rate": 1.180128563179588e-06, + "loss": 0.594, + "step": 2624 + }, + { + "epoch": 4.114300626304802, + "grad_norm": 1.2463157176971436, + "learning_rate": 1.17833981509492e-06, + "loss": 0.5918, + "step": 2625 + }, + { + "epoch": 4.115866388308977, + "grad_norm": 1.2002460956573486, + "learning_rate": 1.1765520055825225e-06, + "loss": 0.5981, + "step": 2626 + }, + { + "epoch": 4.117432150313152, + "grad_norm": 1.349730372428894, + "learning_rate": 1.1747651359120038e-06, + "loss": 0.6084, + "step": 2627 + }, + { + "epoch": 4.118997912317328, + "grad_norm": 1.277337908744812, + "learning_rate": 1.1729792073523e-06, + "loss": 0.595, + "step": 2628 + }, + { + "epoch": 4.120563674321503, + "grad_norm": 1.2968724966049194, + "learning_rate": 1.1711942211716845e-06, + "loss": 0.6708, + "step": 2629 + }, + { + "epoch": 4.122129436325679, + "grad_norm": 1.1646991968154907, + "learning_rate": 1.1694101786377593e-06, + "loss": 0.6009, + "step": 2630 + }, + { + "epoch": 4.1236951983298535, + "grad_norm": 1.392899513244629, + "learning_rate": 1.1676270810174526e-06, + "loss": 0.6038, + "step": 2631 + }, + { + "epoch": 4.125260960334029, + "grad_norm": 1.085792899131775, + "learning_rate": 1.1658449295770287e-06, + "loss": 0.5864, + "step": 2632 + }, + { + "epoch": 4.126826722338205, + "grad_norm": 1.2588834762573242, + "learning_rate": 1.1640637255820732e-06, + "loss": 0.633, + "step": 2633 + }, + { + "epoch": 4.12839248434238, + "grad_norm": 1.2557780742645264, + "learning_rate": 1.1622834702975022e-06, + "loss": 0.6003, + "step": 2634 + }, + { + "epoch": 4.129958246346555, + "grad_norm": 1.4421063661575317, + "learning_rate": 1.1605041649875602e-06, + "loss": 0.5322, + "step": 2635 + }, + { + "epoch": 4.1315240083507305, + "grad_norm": 1.1980674266815186, + "learning_rate": 1.1587258109158114e-06, + "loss": 0.5461, + "step": 2636 + }, + { + "epoch": 4.133089770354906, + "grad_norm": 1.2281138896942139, + "learning_rate": 1.156948409345149e-06, + "loss": 0.6324, + "step": 2637 + }, + { + "epoch": 4.134655532359082, + "grad_norm": 1.25807523727417, + "learning_rate": 1.1551719615377882e-06, + "loss": 0.6451, + "step": 2638 + }, + { + "epoch": 4.136221294363256, + "grad_norm": 1.1664526462554932, + "learning_rate": 1.153396468755267e-06, + "loss": 0.5765, + "step": 2639 + }, + { + "epoch": 4.137787056367432, + "grad_norm": 1.226771593093872, + "learning_rate": 1.1516219322584463e-06, + "loss": 0.6863, + "step": 2640 + }, + { + "epoch": 4.1393528183716075, + "grad_norm": 1.1369787454605103, + "learning_rate": 1.1498483533075042e-06, + "loss": 0.6013, + "step": 2641 + }, + { + "epoch": 4.140918580375783, + "grad_norm": 1.2024613618850708, + "learning_rate": 1.148075733161944e-06, + "loss": 0.6674, + "step": 2642 + }, + { + "epoch": 4.142484342379959, + "grad_norm": 1.1067324876785278, + "learning_rate": 1.1463040730805859e-06, + "loss": 0.5326, + "step": 2643 + }, + { + "epoch": 4.144050104384133, + "grad_norm": 1.2651442289352417, + "learning_rate": 1.1445333743215648e-06, + "loss": 0.6634, + "step": 2644 + }, + { + "epoch": 4.145615866388309, + "grad_norm": 1.22459876537323, + "learning_rate": 1.1427636381423402e-06, + "loss": 0.6078, + "step": 2645 + }, + { + "epoch": 4.1471816283924845, + "grad_norm": 1.200835943222046, + "learning_rate": 1.140994865799681e-06, + "loss": 0.6767, + "step": 2646 + }, + { + "epoch": 4.14874739039666, + "grad_norm": 1.2004497051239014, + "learning_rate": 1.1392270585496746e-06, + "loss": 0.6043, + "step": 2647 + }, + { + "epoch": 4.150313152400835, + "grad_norm": 1.2646071910858154, + "learning_rate": 1.1374602176477256e-06, + "loss": 0.5793, + "step": 2648 + }, + { + "epoch": 4.15187891440501, + "grad_norm": 1.2798433303833008, + "learning_rate": 1.1356943443485475e-06, + "loss": 0.6171, + "step": 2649 + }, + { + "epoch": 4.153444676409186, + "grad_norm": 1.2004300355911255, + "learning_rate": 1.1339294399061692e-06, + "loss": 0.6218, + "step": 2650 + }, + { + "epoch": 4.1550104384133615, + "grad_norm": 1.3948124647140503, + "learning_rate": 1.1321655055739316e-06, + "loss": 0.6047, + "step": 2651 + }, + { + "epoch": 4.156576200417536, + "grad_norm": 1.2331879138946533, + "learning_rate": 1.1304025426044869e-06, + "loss": 0.6276, + "step": 2652 + }, + { + "epoch": 4.158141962421712, + "grad_norm": 1.3041292428970337, + "learning_rate": 1.1286405522497962e-06, + "loss": 0.5769, + "step": 2653 + }, + { + "epoch": 4.159707724425887, + "grad_norm": 1.2389923334121704, + "learning_rate": 1.1268795357611308e-06, + "loss": 0.5811, + "step": 2654 + }, + { + "epoch": 4.161273486430063, + "grad_norm": 1.3386740684509277, + "learning_rate": 1.1251194943890706e-06, + "loss": 0.5769, + "step": 2655 + }, + { + "epoch": 4.162839248434238, + "grad_norm": 1.129999041557312, + "learning_rate": 1.1233604293835027e-06, + "loss": 0.6376, + "step": 2656 + }, + { + "epoch": 4.164405010438413, + "grad_norm": 1.2060986757278442, + "learning_rate": 1.1216023419936206e-06, + "loss": 0.6338, + "step": 2657 + }, + { + "epoch": 4.165970772442589, + "grad_norm": 1.2099809646606445, + "learning_rate": 1.1198452334679246e-06, + "loss": 0.6374, + "step": 2658 + }, + { + "epoch": 4.167536534446764, + "grad_norm": 1.1674392223358154, + "learning_rate": 1.1180891050542168e-06, + "loss": 0.6486, + "step": 2659 + }, + { + "epoch": 4.16910229645094, + "grad_norm": 1.2688627243041992, + "learning_rate": 1.116333957999608e-06, + "loss": 0.5927, + "step": 2660 + }, + { + "epoch": 4.170668058455115, + "grad_norm": 1.36686372756958, + "learning_rate": 1.1145797935505095e-06, + "loss": 0.6127, + "step": 2661 + }, + { + "epoch": 4.17223382045929, + "grad_norm": 1.32498300075531, + "learning_rate": 1.1128266129526325e-06, + "loss": 0.67, + "step": 2662 + }, + { + "epoch": 4.173799582463466, + "grad_norm": 1.247199535369873, + "learning_rate": 1.1110744174509952e-06, + "loss": 0.6419, + "step": 2663 + }, + { + "epoch": 4.175365344467641, + "grad_norm": 1.2895094156265259, + "learning_rate": 1.1093232082899106e-06, + "loss": 0.6274, + "step": 2664 + }, + { + "epoch": 4.176931106471816, + "grad_norm": 1.2122670412063599, + "learning_rate": 1.1075729867129934e-06, + "loss": 0.6086, + "step": 2665 + }, + { + "epoch": 4.1784968684759916, + "grad_norm": 1.2727183103561401, + "learning_rate": 1.1058237539631603e-06, + "loss": 0.6258, + "step": 2666 + }, + { + "epoch": 4.180062630480167, + "grad_norm": 1.295921802520752, + "learning_rate": 1.1040755112826197e-06, + "loss": 0.6066, + "step": 2667 + }, + { + "epoch": 4.181628392484343, + "grad_norm": 1.3321471214294434, + "learning_rate": 1.1023282599128797e-06, + "loss": 0.5826, + "step": 2668 + }, + { + "epoch": 4.183194154488517, + "grad_norm": 1.190930962562561, + "learning_rate": 1.100582001094748e-06, + "loss": 0.5981, + "step": 2669 + }, + { + "epoch": 4.184759916492693, + "grad_norm": 1.2681376934051514, + "learning_rate": 1.0988367360683209e-06, + "loss": 0.6093, + "step": 2670 + }, + { + "epoch": 4.1863256784968685, + "grad_norm": 1.2412774562835693, + "learning_rate": 1.097092466072993e-06, + "loss": 0.5771, + "step": 2671 + }, + { + "epoch": 4.187891440501044, + "grad_norm": 1.077805757522583, + "learning_rate": 1.0953491923474515e-06, + "loss": 0.6112, + "step": 2672 + }, + { + "epoch": 4.189457202505219, + "grad_norm": 1.1926559209823608, + "learning_rate": 1.0936069161296764e-06, + "loss": 0.5909, + "step": 2673 + }, + { + "epoch": 4.191022964509394, + "grad_norm": 1.104425072669983, + "learning_rate": 1.0918656386569384e-06, + "loss": 0.4628, + "step": 2674 + }, + { + "epoch": 4.19258872651357, + "grad_norm": 1.3284837007522583, + "learning_rate": 1.0901253611657996e-06, + "loss": 0.654, + "step": 2675 + }, + { + "epoch": 4.1941544885177455, + "grad_norm": 1.249345302581787, + "learning_rate": 1.0883860848921122e-06, + "loss": 0.5535, + "step": 2676 + }, + { + "epoch": 4.195720250521921, + "grad_norm": 1.233498454093933, + "learning_rate": 1.086647811071017e-06, + "loss": 0.6309, + "step": 2677 + }, + { + "epoch": 4.197286012526096, + "grad_norm": 1.1552709341049194, + "learning_rate": 1.0849105409369424e-06, + "loss": 0.5742, + "step": 2678 + }, + { + "epoch": 4.198851774530271, + "grad_norm": 1.1115463972091675, + "learning_rate": 1.0831742757236059e-06, + "loss": 0.6373, + "step": 2679 + }, + { + "epoch": 4.200417536534447, + "grad_norm": 1.2297323942184448, + "learning_rate": 1.0814390166640073e-06, + "loss": 0.6526, + "step": 2680 + }, + { + "epoch": 4.2019832985386225, + "grad_norm": 1.4014023542404175, + "learning_rate": 1.0797047649904377e-06, + "loss": 0.5836, + "step": 2681 + }, + { + "epoch": 4.203549060542797, + "grad_norm": 1.242067813873291, + "learning_rate": 1.0779715219344692e-06, + "loss": 0.6644, + "step": 2682 + }, + { + "epoch": 4.205114822546973, + "grad_norm": 1.3167695999145508, + "learning_rate": 1.0762392887269556e-06, + "loss": 0.637, + "step": 2683 + }, + { + "epoch": 4.206680584551148, + "grad_norm": 1.229784369468689, + "learning_rate": 1.07450806659804e-06, + "loss": 0.6116, + "step": 2684 + }, + { + "epoch": 4.208246346555324, + "grad_norm": 1.2310740947723389, + "learning_rate": 1.0727778567771407e-06, + "loss": 0.5437, + "step": 2685 + }, + { + "epoch": 4.209812108559499, + "grad_norm": 1.277980089187622, + "learning_rate": 1.0710486604929602e-06, + "loss": 0.6151, + "step": 2686 + }, + { + "epoch": 4.211377870563674, + "grad_norm": 1.2275346517562866, + "learning_rate": 1.0693204789734834e-06, + "loss": 0.5514, + "step": 2687 + }, + { + "epoch": 4.21294363256785, + "grad_norm": 1.2537457942962646, + "learning_rate": 1.0675933134459702e-06, + "loss": 0.5858, + "step": 2688 + }, + { + "epoch": 4.214509394572025, + "grad_norm": 1.2391557693481445, + "learning_rate": 1.0658671651369615e-06, + "loss": 0.5432, + "step": 2689 + }, + { + "epoch": 4.2160751565762, + "grad_norm": 1.2173511981964111, + "learning_rate": 1.0641420352722754e-06, + "loss": 0.651, + "step": 2690 + }, + { + "epoch": 4.217640918580376, + "grad_norm": 1.227282166481018, + "learning_rate": 1.0624179250770066e-06, + "loss": 0.5668, + "step": 2691 + }, + { + "epoch": 4.219206680584551, + "grad_norm": 1.195631980895996, + "learning_rate": 1.060694835775527e-06, + "loss": 0.6318, + "step": 2692 + }, + { + "epoch": 4.220772442588727, + "grad_norm": 1.2749062776565552, + "learning_rate": 1.0589727685914798e-06, + "loss": 0.6389, + "step": 2693 + }, + { + "epoch": 4.222338204592901, + "grad_norm": 1.3092809915542603, + "learning_rate": 1.057251724747787e-06, + "loss": 0.5782, + "step": 2694 + }, + { + "epoch": 4.223903966597077, + "grad_norm": 1.222758173942566, + "learning_rate": 1.055531705466642e-06, + "loss": 0.6047, + "step": 2695 + }, + { + "epoch": 4.225469728601253, + "grad_norm": 1.2806799411773682, + "learning_rate": 1.0538127119695076e-06, + "loss": 0.5916, + "step": 2696 + }, + { + "epoch": 4.227035490605428, + "grad_norm": 1.2441686391830444, + "learning_rate": 1.052094745477125e-06, + "loss": 0.6405, + "step": 2697 + }, + { + "epoch": 4.228601252609604, + "grad_norm": 1.2265493869781494, + "learning_rate": 1.0503778072094986e-06, + "loss": 0.624, + "step": 2698 + }, + { + "epoch": 4.230167014613778, + "grad_norm": 1.2378262281417847, + "learning_rate": 1.0486618983859064e-06, + "loss": 0.6267, + "step": 2699 + }, + { + "epoch": 4.231732776617954, + "grad_norm": 1.1325656175613403, + "learning_rate": 1.0469470202248976e-06, + "loss": 0.5143, + "step": 2700 + }, + { + "epoch": 4.23329853862213, + "grad_norm": 1.1236976385116577, + "learning_rate": 1.0452331739442839e-06, + "loss": 0.5564, + "step": 2701 + }, + { + "epoch": 4.234864300626305, + "grad_norm": 1.3158353567123413, + "learning_rate": 1.0435203607611483e-06, + "loss": 0.5853, + "step": 2702 + }, + { + "epoch": 4.23643006263048, + "grad_norm": 1.1873832941055298, + "learning_rate": 1.041808581891839e-06, + "loss": 0.6496, + "step": 2703 + }, + { + "epoch": 4.237995824634655, + "grad_norm": 1.2505794763565063, + "learning_rate": 1.0400978385519692e-06, + "loss": 0.5656, + "step": 2704 + }, + { + "epoch": 4.239561586638831, + "grad_norm": 1.230660319328308, + "learning_rate": 1.0383881319564176e-06, + "loss": 0.5938, + "step": 2705 + }, + { + "epoch": 4.241127348643007, + "grad_norm": 1.349231481552124, + "learning_rate": 1.036679463319326e-06, + "loss": 0.6108, + "step": 2706 + }, + { + "epoch": 4.242693110647181, + "grad_norm": 1.275140404701233, + "learning_rate": 1.0349718338540998e-06, + "loss": 0.6672, + "step": 2707 + }, + { + "epoch": 4.244258872651357, + "grad_norm": 1.3713583946228027, + "learning_rate": 1.0332652447734057e-06, + "loss": 0.5547, + "step": 2708 + }, + { + "epoch": 4.245824634655532, + "grad_norm": 1.2152209281921387, + "learning_rate": 1.0315596972891714e-06, + "loss": 0.5614, + "step": 2709 + }, + { + "epoch": 4.247390396659708, + "grad_norm": 1.302657127380371, + "learning_rate": 1.0298551926125863e-06, + "loss": 0.6417, + "step": 2710 + }, + { + "epoch": 4.2489561586638835, + "grad_norm": 1.1909019947052002, + "learning_rate": 1.0281517319540976e-06, + "loss": 0.6784, + "step": 2711 + }, + { + "epoch": 4.250521920668058, + "grad_norm": 1.202182650566101, + "learning_rate": 1.026449316523412e-06, + "loss": 0.6202, + "step": 2712 + }, + { + "epoch": 4.252087682672234, + "grad_norm": 1.1392393112182617, + "learning_rate": 1.0247479475294948e-06, + "loss": 0.5573, + "step": 2713 + }, + { + "epoch": 4.253653444676409, + "grad_norm": 1.2918214797973633, + "learning_rate": 1.023047626180564e-06, + "loss": 0.6204, + "step": 2714 + }, + { + "epoch": 4.255219206680585, + "grad_norm": 1.249887228012085, + "learning_rate": 1.0213483536841e-06, + "loss": 0.6308, + "step": 2715 + }, + { + "epoch": 4.25678496868476, + "grad_norm": 1.156160593032837, + "learning_rate": 1.019650131246835e-06, + "loss": 0.5617, + "step": 2716 + }, + { + "epoch": 4.258350730688935, + "grad_norm": 1.2980233430862427, + "learning_rate": 1.0179529600747526e-06, + "loss": 0.619, + "step": 2717 + }, + { + "epoch": 4.259916492693111, + "grad_norm": 1.1643873453140259, + "learning_rate": 1.0162568413730966e-06, + "loss": 0.6488, + "step": 2718 + }, + { + "epoch": 4.261482254697286, + "grad_norm": 1.3735922574996948, + "learning_rate": 1.0145617763463573e-06, + "loss": 0.6022, + "step": 2719 + }, + { + "epoch": 4.263048016701461, + "grad_norm": 1.2653254270553589, + "learning_rate": 1.0128677661982786e-06, + "loss": 0.6285, + "step": 2720 + }, + { + "epoch": 4.264613778705637, + "grad_norm": 1.3981285095214844, + "learning_rate": 1.0111748121318587e-06, + "loss": 0.6381, + "step": 2721 + }, + { + "epoch": 4.266179540709812, + "grad_norm": 1.2364410161972046, + "learning_rate": 1.00948291534934e-06, + "loss": 0.5779, + "step": 2722 + }, + { + "epoch": 4.267745302713988, + "grad_norm": 1.0957547426223755, + "learning_rate": 1.0077920770522184e-06, + "loss": 0.6646, + "step": 2723 + }, + { + "epoch": 4.2693110647181625, + "grad_norm": 1.1948002576828003, + "learning_rate": 1.006102298441236e-06, + "loss": 0.6691, + "step": 2724 + }, + { + "epoch": 4.270876826722338, + "grad_norm": 1.3255823850631714, + "learning_rate": 1.0044135807163834e-06, + "loss": 0.5499, + "step": 2725 + }, + { + "epoch": 4.272442588726514, + "grad_norm": 1.2720073461532593, + "learning_rate": 1.0027259250768975e-06, + "loss": 0.628, + "step": 2726 + }, + { + "epoch": 4.274008350730689, + "grad_norm": 1.1583763360977173, + "learning_rate": 1.0010393327212612e-06, + "loss": 0.6253, + "step": 2727 + }, + { + "epoch": 4.275574112734864, + "grad_norm": 1.1668976545333862, + "learning_rate": 9.993538048472013e-07, + "loss": 0.6063, + "step": 2728 + }, + { + "epoch": 4.2771398747390394, + "grad_norm": 1.2782100439071655, + "learning_rate": 9.976693426516902e-07, + "loss": 0.6156, + "step": 2729 + }, + { + "epoch": 4.278705636743215, + "grad_norm": 1.2681186199188232, + "learning_rate": 9.959859473309422e-07, + "loss": 0.6426, + "step": 2730 + }, + { + "epoch": 4.280271398747391, + "grad_norm": 1.1437429189682007, + "learning_rate": 9.943036200804154e-07, + "loss": 0.5888, + "step": 2731 + }, + { + "epoch": 4.281837160751566, + "grad_norm": 1.3140877485275269, + "learning_rate": 9.926223620948061e-07, + "loss": 0.5913, + "step": 2732 + }, + { + "epoch": 4.283402922755741, + "grad_norm": 1.1629775762557983, + "learning_rate": 9.909421745680565e-07, + "loss": 0.5928, + "step": 2733 + }, + { + "epoch": 4.284968684759916, + "grad_norm": 1.3503170013427734, + "learning_rate": 9.892630586933452e-07, + "loss": 0.6137, + "step": 2734 + }, + { + "epoch": 4.286534446764092, + "grad_norm": 1.1408926248550415, + "learning_rate": 9.875850156630892e-07, + "loss": 0.5934, + "step": 2735 + }, + { + "epoch": 4.288100208768268, + "grad_norm": 1.3211277723312378, + "learning_rate": 9.859080466689452e-07, + "loss": 0.5723, + "step": 2736 + }, + { + "epoch": 4.289665970772442, + "grad_norm": 1.2893445491790771, + "learning_rate": 9.84232152901807e-07, + "loss": 0.5508, + "step": 2737 + }, + { + "epoch": 4.291231732776618, + "grad_norm": 1.268737554550171, + "learning_rate": 9.825573355518041e-07, + "loss": 0.5846, + "step": 2738 + }, + { + "epoch": 4.292797494780793, + "grad_norm": 1.145947813987732, + "learning_rate": 9.808835958083029e-07, + "loss": 0.5976, + "step": 2739 + }, + { + "epoch": 4.294363256784969, + "grad_norm": 1.2387874126434326, + "learning_rate": 9.792109348599036e-07, + "loss": 0.6492, + "step": 2740 + }, + { + "epoch": 4.295929018789144, + "grad_norm": 1.212338924407959, + "learning_rate": 9.775393538944394e-07, + "loss": 0.6379, + "step": 2741 + }, + { + "epoch": 4.297494780793319, + "grad_norm": 1.263137936592102, + "learning_rate": 9.75868854098979e-07, + "loss": 0.5711, + "step": 2742 + }, + { + "epoch": 4.299060542797495, + "grad_norm": 1.2036539316177368, + "learning_rate": 9.741994366598207e-07, + "loss": 0.6217, + "step": 2743 + }, + { + "epoch": 4.30062630480167, + "grad_norm": 1.3662471771240234, + "learning_rate": 9.725311027624967e-07, + "loss": 0.6184, + "step": 2744 + }, + { + "epoch": 4.302192066805846, + "grad_norm": 1.1384464502334595, + "learning_rate": 9.708638535917658e-07, + "loss": 0.6273, + "step": 2745 + }, + { + "epoch": 4.303757828810021, + "grad_norm": 1.3032082319259644, + "learning_rate": 9.691976903316213e-07, + "loss": 0.6393, + "step": 2746 + }, + { + "epoch": 4.305323590814196, + "grad_norm": 1.3347432613372803, + "learning_rate": 9.675326141652839e-07, + "loss": 0.6131, + "step": 2747 + }, + { + "epoch": 4.306889352818372, + "grad_norm": 1.2203959226608276, + "learning_rate": 9.65868626275198e-07, + "loss": 0.5778, + "step": 2748 + }, + { + "epoch": 4.3084551148225465, + "grad_norm": 1.2628487348556519, + "learning_rate": 9.642057278430431e-07, + "loss": 0.6296, + "step": 2749 + }, + { + "epoch": 4.310020876826722, + "grad_norm": 1.208202838897705, + "learning_rate": 9.62543920049717e-07, + "loss": 0.6245, + "step": 2750 + }, + { + "epoch": 4.311586638830898, + "grad_norm": 1.1364301443099976, + "learning_rate": 9.608832040753472e-07, + "loss": 0.6183, + "step": 2751 + }, + { + "epoch": 4.313152400835073, + "grad_norm": 1.3837535381317139, + "learning_rate": 9.592235810992875e-07, + "loss": 0.6547, + "step": 2752 + }, + { + "epoch": 4.314718162839249, + "grad_norm": 1.222366452217102, + "learning_rate": 9.575650523001106e-07, + "loss": 0.6415, + "step": 2753 + }, + { + "epoch": 4.3162839248434235, + "grad_norm": 1.3130534887313843, + "learning_rate": 9.559076188556163e-07, + "loss": 0.6968, + "step": 2754 + }, + { + "epoch": 4.317849686847599, + "grad_norm": 1.3480855226516724, + "learning_rate": 9.542512819428246e-07, + "loss": 0.6532, + "step": 2755 + }, + { + "epoch": 4.319415448851775, + "grad_norm": 1.4194175004959106, + "learning_rate": 9.525960427379772e-07, + "loss": 0.6413, + "step": 2756 + }, + { + "epoch": 4.32098121085595, + "grad_norm": 1.1544206142425537, + "learning_rate": 9.509419024165364e-07, + "loss": 0.6062, + "step": 2757 + }, + { + "epoch": 4.322546972860125, + "grad_norm": 1.343361735343933, + "learning_rate": 9.49288862153184e-07, + "loss": 0.682, + "step": 2758 + }, + { + "epoch": 4.3241127348643005, + "grad_norm": 1.2994462251663208, + "learning_rate": 9.476369231218207e-07, + "loss": 0.6081, + "step": 2759 + }, + { + "epoch": 4.325678496868476, + "grad_norm": 1.2797702550888062, + "learning_rate": 9.459860864955653e-07, + "loss": 0.5675, + "step": 2760 + }, + { + "epoch": 4.327244258872652, + "grad_norm": 1.322637915611267, + "learning_rate": 9.443363534467534e-07, + "loss": 0.5972, + "step": 2761 + }, + { + "epoch": 4.328810020876826, + "grad_norm": 1.3239537477493286, + "learning_rate": 9.426877251469372e-07, + "loss": 0.5902, + "step": 2762 + }, + { + "epoch": 4.330375782881002, + "grad_norm": 1.213059663772583, + "learning_rate": 9.410402027668835e-07, + "loss": 0.6537, + "step": 2763 + }, + { + "epoch": 4.3319415448851775, + "grad_norm": 1.2869120836257935, + "learning_rate": 9.393937874765754e-07, + "loss": 0.5573, + "step": 2764 + }, + { + "epoch": 4.333507306889353, + "grad_norm": 1.2183059453964233, + "learning_rate": 9.377484804452094e-07, + "loss": 0.6328, + "step": 2765 + }, + { + "epoch": 4.335073068893529, + "grad_norm": 1.2206132411956787, + "learning_rate": 9.361042828411915e-07, + "loss": 0.5814, + "step": 2766 + }, + { + "epoch": 4.336638830897703, + "grad_norm": 1.1965863704681396, + "learning_rate": 9.344611958321459e-07, + "loss": 0.5821, + "step": 2767 + }, + { + "epoch": 4.338204592901879, + "grad_norm": 1.29924738407135, + "learning_rate": 9.328192205849043e-07, + "loss": 0.6607, + "step": 2768 + }, + { + "epoch": 4.3397703549060545, + "grad_norm": 1.0587471723556519, + "learning_rate": 9.311783582655076e-07, + "loss": 0.6591, + "step": 2769 + }, + { + "epoch": 4.34133611691023, + "grad_norm": 1.2471832036972046, + "learning_rate": 9.295386100392117e-07, + "loss": 0.6371, + "step": 2770 + }, + { + "epoch": 4.342901878914405, + "grad_norm": 1.2618634700775146, + "learning_rate": 9.27899977070475e-07, + "loss": 0.5275, + "step": 2771 + }, + { + "epoch": 4.34446764091858, + "grad_norm": 1.2841023206710815, + "learning_rate": 9.262624605229673e-07, + "loss": 0.5328, + "step": 2772 + }, + { + "epoch": 4.346033402922756, + "grad_norm": 1.3328698873519897, + "learning_rate": 9.246260615595673e-07, + "loss": 0.5379, + "step": 2773 + }, + { + "epoch": 4.347599164926931, + "grad_norm": 1.1054933071136475, + "learning_rate": 9.229907813423556e-07, + "loss": 0.5078, + "step": 2774 + }, + { + "epoch": 4.349164926931106, + "grad_norm": 1.2079209089279175, + "learning_rate": 9.213566210326213e-07, + "loss": 0.6467, + "step": 2775 + }, + { + "epoch": 4.350730688935282, + "grad_norm": 1.2839919328689575, + "learning_rate": 9.197235817908579e-07, + "loss": 0.5954, + "step": 2776 + }, + { + "epoch": 4.352296450939457, + "grad_norm": 1.2439117431640625, + "learning_rate": 9.180916647767621e-07, + "loss": 0.6098, + "step": 2777 + }, + { + "epoch": 4.353862212943633, + "grad_norm": 1.2913306951522827, + "learning_rate": 9.164608711492337e-07, + "loss": 0.5375, + "step": 2778 + }, + { + "epoch": 4.3554279749478075, + "grad_norm": 1.1591486930847168, + "learning_rate": 9.148312020663755e-07, + "loss": 0.5844, + "step": 2779 + }, + { + "epoch": 4.356993736951983, + "grad_norm": 1.2469542026519775, + "learning_rate": 9.132026586854909e-07, + "loss": 0.6684, + "step": 2780 + }, + { + "epoch": 4.358559498956159, + "grad_norm": 1.310805320739746, + "learning_rate": 9.115752421630852e-07, + "loss": 0.6514, + "step": 2781 + }, + { + "epoch": 4.360125260960334, + "grad_norm": 1.2318816184997559, + "learning_rate": 9.099489536548598e-07, + "loss": 0.5453, + "step": 2782 + }, + { + "epoch": 4.361691022964509, + "grad_norm": 1.23331618309021, + "learning_rate": 9.083237943157208e-07, + "loss": 0.5484, + "step": 2783 + }, + { + "epoch": 4.3632567849686845, + "grad_norm": 1.390173077583313, + "learning_rate": 9.066997652997672e-07, + "loss": 0.6245, + "step": 2784 + }, + { + "epoch": 4.36482254697286, + "grad_norm": 1.3007646799087524, + "learning_rate": 9.050768677602973e-07, + "loss": 0.6415, + "step": 2785 + }, + { + "epoch": 4.366388308977036, + "grad_norm": 1.2786531448364258, + "learning_rate": 9.034551028498084e-07, + "loss": 0.6174, + "step": 2786 + }, + { + "epoch": 4.367954070981211, + "grad_norm": 1.4713473320007324, + "learning_rate": 9.018344717199887e-07, + "loss": 0.5183, + "step": 2787 + }, + { + "epoch": 4.369519832985386, + "grad_norm": 1.112297773361206, + "learning_rate": 9.002149755217246e-07, + "loss": 0.6369, + "step": 2788 + }, + { + "epoch": 4.3710855949895615, + "grad_norm": 1.2173497676849365, + "learning_rate": 8.985966154050952e-07, + "loss": 0.6195, + "step": 2789 + }, + { + "epoch": 4.372651356993737, + "grad_norm": 1.2108879089355469, + "learning_rate": 8.969793925193731e-07, + "loss": 0.6155, + "step": 2790 + }, + { + "epoch": 4.374217118997913, + "grad_norm": 1.409883975982666, + "learning_rate": 8.953633080130239e-07, + "loss": 0.5926, + "step": 2791 + }, + { + "epoch": 4.375782881002087, + "grad_norm": 1.2854543924331665, + "learning_rate": 8.937483630337038e-07, + "loss": 0.6337, + "step": 2792 + }, + { + "epoch": 4.377348643006263, + "grad_norm": 1.2990738153457642, + "learning_rate": 8.921345587282604e-07, + "loss": 0.6159, + "step": 2793 + }, + { + "epoch": 4.3789144050104385, + "grad_norm": 1.230802059173584, + "learning_rate": 8.905218962427312e-07, + "loss": 0.5945, + "step": 2794 + }, + { + "epoch": 4.380480167014614, + "grad_norm": 1.2086981534957886, + "learning_rate": 8.889103767223425e-07, + "loss": 0.5439, + "step": 2795 + }, + { + "epoch": 4.382045929018789, + "grad_norm": 1.215370774269104, + "learning_rate": 8.873000013115099e-07, + "loss": 0.6095, + "step": 2796 + }, + { + "epoch": 4.383611691022964, + "grad_norm": 1.2571933269500732, + "learning_rate": 8.856907711538335e-07, + "loss": 0.6036, + "step": 2797 + }, + { + "epoch": 4.38517745302714, + "grad_norm": 1.1642861366271973, + "learning_rate": 8.840826873921049e-07, + "loss": 0.582, + "step": 2798 + }, + { + "epoch": 4.3867432150313155, + "grad_norm": 1.23332941532135, + "learning_rate": 8.824757511682988e-07, + "loss": 0.6014, + "step": 2799 + }, + { + "epoch": 4.388308977035491, + "grad_norm": 1.2016581296920776, + "learning_rate": 8.80869963623573e-07, + "loss": 0.642, + "step": 2800 + }, + { + "epoch": 4.389874739039666, + "grad_norm": 1.2199851274490356, + "learning_rate": 8.792653258982747e-07, + "loss": 0.6413, + "step": 2801 + }, + { + "epoch": 4.391440501043841, + "grad_norm": 1.3119784593582153, + "learning_rate": 8.776618391319297e-07, + "loss": 0.5756, + "step": 2802 + }, + { + "epoch": 4.393006263048017, + "grad_norm": 1.267626404762268, + "learning_rate": 8.760595044632484e-07, + "loss": 0.5972, + "step": 2803 + }, + { + "epoch": 4.3945720250521925, + "grad_norm": 1.0666111707687378, + "learning_rate": 8.744583230301248e-07, + "loss": 0.6459, + "step": 2804 + }, + { + "epoch": 4.396137787056367, + "grad_norm": 1.276969313621521, + "learning_rate": 8.728582959696305e-07, + "loss": 0.567, + "step": 2805 + }, + { + "epoch": 4.397703549060543, + "grad_norm": 1.313031554222107, + "learning_rate": 8.712594244180198e-07, + "loss": 0.562, + "step": 2806 + }, + { + "epoch": 4.399269311064718, + "grad_norm": 1.164860486984253, + "learning_rate": 8.696617095107249e-07, + "loss": 0.5849, + "step": 2807 + }, + { + "epoch": 4.400835073068894, + "grad_norm": 1.2664166688919067, + "learning_rate": 8.680651523823583e-07, + "loss": 0.533, + "step": 2808 + }, + { + "epoch": 4.402400835073069, + "grad_norm": 1.2811753749847412, + "learning_rate": 8.664697541667085e-07, + "loss": 0.6285, + "step": 2809 + }, + { + "epoch": 4.403966597077244, + "grad_norm": 1.2239644527435303, + "learning_rate": 8.648755159967426e-07, + "loss": 0.5728, + "step": 2810 + }, + { + "epoch": 4.40553235908142, + "grad_norm": 1.2360845804214478, + "learning_rate": 8.632824390046027e-07, + "loss": 0.6006, + "step": 2811 + }, + { + "epoch": 4.407098121085595, + "grad_norm": 1.1378620862960815, + "learning_rate": 8.61690524321607e-07, + "loss": 0.5738, + "step": 2812 + }, + { + "epoch": 4.40866388308977, + "grad_norm": 1.2592755556106567, + "learning_rate": 8.600997730782476e-07, + "loss": 0.538, + "step": 2813 + }, + { + "epoch": 4.4102296450939455, + "grad_norm": 1.359745979309082, + "learning_rate": 8.585101864041914e-07, + "loss": 0.6385, + "step": 2814 + }, + { + "epoch": 4.411795407098121, + "grad_norm": 1.1939847469329834, + "learning_rate": 8.569217654282774e-07, + "loss": 0.6262, + "step": 2815 + }, + { + "epoch": 4.413361169102297, + "grad_norm": 1.2629926204681396, + "learning_rate": 8.553345112785172e-07, + "loss": 0.5972, + "step": 2816 + }, + { + "epoch": 4.414926931106471, + "grad_norm": 1.177193284034729, + "learning_rate": 8.537484250820943e-07, + "loss": 0.6511, + "step": 2817 + }, + { + "epoch": 4.416492693110647, + "grad_norm": 1.2521287202835083, + "learning_rate": 8.521635079653601e-07, + "loss": 0.5487, + "step": 2818 + }, + { + "epoch": 4.4180584551148225, + "grad_norm": 1.2183914184570312, + "learning_rate": 8.505797610538397e-07, + "loss": 0.6102, + "step": 2819 + }, + { + "epoch": 4.419624217118998, + "grad_norm": 1.4728264808654785, + "learning_rate": 8.48997185472226e-07, + "loss": 0.6378, + "step": 2820 + }, + { + "epoch": 4.421189979123174, + "grad_norm": 1.1505528688430786, + "learning_rate": 8.474157823443766e-07, + "loss": 0.6167, + "step": 2821 + }, + { + "epoch": 4.422755741127348, + "grad_norm": 1.1801762580871582, + "learning_rate": 8.458355527933224e-07, + "loss": 0.622, + "step": 2822 + }, + { + "epoch": 4.424321503131524, + "grad_norm": 1.3376151323318481, + "learning_rate": 8.442564979412557e-07, + "loss": 0.6606, + "step": 2823 + }, + { + "epoch": 4.4258872651356995, + "grad_norm": 1.3469598293304443, + "learning_rate": 8.426786189095362e-07, + "loss": 0.5997, + "step": 2824 + }, + { + "epoch": 4.427453027139875, + "grad_norm": 1.230717420578003, + "learning_rate": 8.411019168186921e-07, + "loss": 0.5861, + "step": 2825 + }, + { + "epoch": 4.42901878914405, + "grad_norm": 1.2221441268920898, + "learning_rate": 8.395263927884101e-07, + "loss": 0.6121, + "step": 2826 + }, + { + "epoch": 4.430584551148225, + "grad_norm": 1.409119963645935, + "learning_rate": 8.379520479375436e-07, + "loss": 0.5807, + "step": 2827 + }, + { + "epoch": 4.432150313152401, + "grad_norm": 1.076403260231018, + "learning_rate": 8.363788833841083e-07, + "loss": 0.6234, + "step": 2828 + }, + { + "epoch": 4.4337160751565765, + "grad_norm": 1.3182142972946167, + "learning_rate": 8.34806900245281e-07, + "loss": 0.6223, + "step": 2829 + }, + { + "epoch": 4.435281837160751, + "grad_norm": 1.265117883682251, + "learning_rate": 8.332360996374014e-07, + "loss": 0.5916, + "step": 2830 + }, + { + "epoch": 4.436847599164927, + "grad_norm": 1.243553876876831, + "learning_rate": 8.316664826759646e-07, + "loss": 0.6201, + "step": 2831 + }, + { + "epoch": 4.438413361169102, + "grad_norm": 1.3207454681396484, + "learning_rate": 8.300980504756318e-07, + "loss": 0.6287, + "step": 2832 + }, + { + "epoch": 4.439979123173278, + "grad_norm": 1.1349838972091675, + "learning_rate": 8.285308041502188e-07, + "loss": 0.5632, + "step": 2833 + }, + { + "epoch": 4.4415448851774535, + "grad_norm": 1.1458604335784912, + "learning_rate": 8.269647448126982e-07, + "loss": 0.6769, + "step": 2834 + }, + { + "epoch": 4.443110647181628, + "grad_norm": 1.2419075965881348, + "learning_rate": 8.253998735752044e-07, + "loss": 0.5475, + "step": 2835 + }, + { + "epoch": 4.444676409185804, + "grad_norm": 1.3041012287139893, + "learning_rate": 8.238361915490226e-07, + "loss": 0.5608, + "step": 2836 + }, + { + "epoch": 4.446242171189979, + "grad_norm": 1.1919783353805542, + "learning_rate": 8.222736998445963e-07, + "loss": 0.5819, + "step": 2837 + }, + { + "epoch": 4.447807933194154, + "grad_norm": 1.126298427581787, + "learning_rate": 8.207123995715254e-07, + "loss": 0.6019, + "step": 2838 + }, + { + "epoch": 4.44937369519833, + "grad_norm": 1.2007626295089722, + "learning_rate": 8.191522918385597e-07, + "loss": 0.606, + "step": 2839 + }, + { + "epoch": 4.450939457202505, + "grad_norm": 1.2129793167114258, + "learning_rate": 8.175933777536049e-07, + "loss": 0.5726, + "step": 2840 + }, + { + "epoch": 4.452505219206681, + "grad_norm": 1.2193784713745117, + "learning_rate": 8.160356584237184e-07, + "loss": 0.6005, + "step": 2841 + }, + { + "epoch": 4.454070981210856, + "grad_norm": 1.2187601327896118, + "learning_rate": 8.144791349551084e-07, + "loss": 0.6322, + "step": 2842 + }, + { + "epoch": 4.455636743215031, + "grad_norm": 1.3521336317062378, + "learning_rate": 8.129238084531355e-07, + "loss": 0.5851, + "step": 2843 + }, + { + "epoch": 4.457202505219207, + "grad_norm": 1.32852303981781, + "learning_rate": 8.113696800223084e-07, + "loss": 0.6288, + "step": 2844 + }, + { + "epoch": 4.458768267223382, + "grad_norm": 1.2852197885513306, + "learning_rate": 8.098167507662863e-07, + "loss": 0.6271, + "step": 2845 + }, + { + "epoch": 4.460334029227558, + "grad_norm": 1.3286292552947998, + "learning_rate": 8.08265021787876e-07, + "loss": 0.6584, + "step": 2846 + }, + { + "epoch": 4.461899791231732, + "grad_norm": 1.1803559064865112, + "learning_rate": 8.067144941890329e-07, + "loss": 0.6344, + "step": 2847 + }, + { + "epoch": 4.463465553235908, + "grad_norm": 1.2986726760864258, + "learning_rate": 8.051651690708581e-07, + "loss": 0.5878, + "step": 2848 + }, + { + "epoch": 4.465031315240084, + "grad_norm": 1.2154364585876465, + "learning_rate": 8.036170475335997e-07, + "loss": 0.6315, + "step": 2849 + }, + { + "epoch": 4.466597077244259, + "grad_norm": 1.2473022937774658, + "learning_rate": 8.020701306766504e-07, + "loss": 0.6432, + "step": 2850 + }, + { + "epoch": 4.468162839248434, + "grad_norm": 1.336057186126709, + "learning_rate": 8.005244195985487e-07, + "loss": 0.5991, + "step": 2851 + }, + { + "epoch": 4.469728601252609, + "grad_norm": 1.2605669498443604, + "learning_rate": 7.989799153969735e-07, + "loss": 0.595, + "step": 2852 + }, + { + "epoch": 4.471294363256785, + "grad_norm": 1.1553322076797485, + "learning_rate": 7.97436619168751e-07, + "loss": 0.6515, + "step": 2853 + }, + { + "epoch": 4.4728601252609606, + "grad_norm": 1.1928796768188477, + "learning_rate": 7.958945320098476e-07, + "loss": 0.5477, + "step": 2854 + }, + { + "epoch": 4.474425887265136, + "grad_norm": 1.1510995626449585, + "learning_rate": 7.943536550153689e-07, + "loss": 0.6296, + "step": 2855 + }, + { + "epoch": 4.475991649269311, + "grad_norm": 1.210997223854065, + "learning_rate": 7.928139892795661e-07, + "loss": 0.5907, + "step": 2856 + }, + { + "epoch": 4.477557411273486, + "grad_norm": 1.2773675918579102, + "learning_rate": 7.912755358958251e-07, + "loss": 0.5547, + "step": 2857 + }, + { + "epoch": 4.479123173277662, + "grad_norm": 1.248292326927185, + "learning_rate": 7.897382959566732e-07, + "loss": 0.6521, + "step": 2858 + }, + { + "epoch": 4.4806889352818375, + "grad_norm": 1.3389019966125488, + "learning_rate": 7.882022705537779e-07, + "loss": 0.6355, + "step": 2859 + }, + { + "epoch": 4.482254697286012, + "grad_norm": 1.1564635038375854, + "learning_rate": 7.866674607779401e-07, + "loss": 0.5526, + "step": 2860 + }, + { + "epoch": 4.483820459290188, + "grad_norm": 1.2170965671539307, + "learning_rate": 7.851338677191001e-07, + "loss": 0.5251, + "step": 2861 + }, + { + "epoch": 4.485386221294363, + "grad_norm": 1.2128232717514038, + "learning_rate": 7.836014924663338e-07, + "loss": 0.6117, + "step": 2862 + }, + { + "epoch": 4.486951983298539, + "grad_norm": 1.3145650625228882, + "learning_rate": 7.820703361078516e-07, + "loss": 0.5499, + "step": 2863 + }, + { + "epoch": 4.488517745302714, + "grad_norm": 1.2057058811187744, + "learning_rate": 7.805403997309991e-07, + "loss": 0.5714, + "step": 2864 + }, + { + "epoch": 4.490083507306889, + "grad_norm": 1.2229523658752441, + "learning_rate": 7.790116844222548e-07, + "loss": 0.6145, + "step": 2865 + }, + { + "epoch": 4.491649269311065, + "grad_norm": 1.3067467212677002, + "learning_rate": 7.77484191267231e-07, + "loss": 0.6141, + "step": 2866 + }, + { + "epoch": 4.49321503131524, + "grad_norm": 1.2613667249679565, + "learning_rate": 7.75957921350671e-07, + "loss": 0.6222, + "step": 2867 + }, + { + "epoch": 4.494780793319415, + "grad_norm": 1.1608632802963257, + "learning_rate": 7.744328757564501e-07, + "loss": 0.6279, + "step": 2868 + }, + { + "epoch": 4.496346555323591, + "grad_norm": 1.3013458251953125, + "learning_rate": 7.72909055567575e-07, + "loss": 0.5413, + "step": 2869 + }, + { + "epoch": 4.497912317327766, + "grad_norm": 1.3727405071258545, + "learning_rate": 7.713864618661787e-07, + "loss": 0.6053, + "step": 2870 + }, + { + "epoch": 4.499478079331942, + "grad_norm": 1.2287840843200684, + "learning_rate": 7.698650957335277e-07, + "loss": 0.6531, + "step": 2871 + }, + { + "epoch": 4.5010438413361165, + "grad_norm": 1.2294139862060547, + "learning_rate": 7.683449582500152e-07, + "loss": 0.6021, + "step": 2872 + }, + { + "epoch": 4.502609603340292, + "grad_norm": 1.23564875125885, + "learning_rate": 7.668260504951596e-07, + "loss": 0.6128, + "step": 2873 + }, + { + "epoch": 4.504175365344468, + "grad_norm": 1.1630370616912842, + "learning_rate": 7.653083735476088e-07, + "loss": 0.6183, + "step": 2874 + }, + { + "epoch": 4.505741127348643, + "grad_norm": 1.2798091173171997, + "learning_rate": 7.637919284851353e-07, + "loss": 0.6534, + "step": 2875 + }, + { + "epoch": 4.507306889352819, + "grad_norm": 1.2523179054260254, + "learning_rate": 7.622767163846376e-07, + "loss": 0.6357, + "step": 2876 + }, + { + "epoch": 4.508872651356993, + "grad_norm": 1.2298544645309448, + "learning_rate": 7.607627383221381e-07, + "loss": 0.5961, + "step": 2877 + }, + { + "epoch": 4.510438413361169, + "grad_norm": 1.2161098718643188, + "learning_rate": 7.592499953727828e-07, + "loss": 0.6332, + "step": 2878 + }, + { + "epoch": 4.512004175365345, + "grad_norm": 1.2523022890090942, + "learning_rate": 7.577384886108405e-07, + "loss": 0.6439, + "step": 2879 + }, + { + "epoch": 4.51356993736952, + "grad_norm": 1.30768883228302, + "learning_rate": 7.562282191097026e-07, + "loss": 0.6736, + "step": 2880 + }, + { + "epoch": 4.515135699373695, + "grad_norm": 1.2421822547912598, + "learning_rate": 7.547191879418817e-07, + "loss": 0.6004, + "step": 2881 + }, + { + "epoch": 4.51670146137787, + "grad_norm": 1.1340116262435913, + "learning_rate": 7.532113961790116e-07, + "loss": 0.6065, + "step": 2882 + }, + { + "epoch": 4.518267223382046, + "grad_norm": 1.319522738456726, + "learning_rate": 7.517048448918429e-07, + "loss": 0.5796, + "step": 2883 + }, + { + "epoch": 4.519832985386222, + "grad_norm": 1.1828511953353882, + "learning_rate": 7.501995351502497e-07, + "loss": 0.6417, + "step": 2884 + }, + { + "epoch": 4.521398747390396, + "grad_norm": 1.302408218383789, + "learning_rate": 7.486954680232231e-07, + "loss": 0.6234, + "step": 2885 + }, + { + "epoch": 4.522964509394572, + "grad_norm": 1.1842252016067505, + "learning_rate": 7.471926445788677e-07, + "loss": 0.6115, + "step": 2886 + }, + { + "epoch": 4.524530271398747, + "grad_norm": 1.2214895486831665, + "learning_rate": 7.456910658844119e-07, + "loss": 0.6104, + "step": 2887 + }, + { + "epoch": 4.526096033402923, + "grad_norm": 1.2608014345169067, + "learning_rate": 7.44190733006194e-07, + "loss": 0.5499, + "step": 2888 + }, + { + "epoch": 4.527661795407099, + "grad_norm": 1.259846806526184, + "learning_rate": 7.426916470096701e-07, + "loss": 0.6065, + "step": 2889 + }, + { + "epoch": 4.529227557411273, + "grad_norm": 1.4206489324569702, + "learning_rate": 7.411938089594131e-07, + "loss": 0.6266, + "step": 2890 + }, + { + "epoch": 4.530793319415449, + "grad_norm": 1.0867127180099487, + "learning_rate": 7.396972199191052e-07, + "loss": 0.6088, + "step": 2891 + }, + { + "epoch": 4.532359081419624, + "grad_norm": 1.2204397916793823, + "learning_rate": 7.38201880951544e-07, + "loss": 0.6255, + "step": 2892 + }, + { + "epoch": 4.533924843423799, + "grad_norm": 1.2188392877578735, + "learning_rate": 7.367077931186397e-07, + "loss": 0.6001, + "step": 2893 + }, + { + "epoch": 4.535490605427975, + "grad_norm": 1.22636079788208, + "learning_rate": 7.352149574814132e-07, + "loss": 0.6662, + "step": 2894 + }, + { + "epoch": 4.53705636743215, + "grad_norm": 1.3532042503356934, + "learning_rate": 7.337233750999964e-07, + "loss": 0.6121, + "step": 2895 + }, + { + "epoch": 4.538622129436326, + "grad_norm": 1.2929681539535522, + "learning_rate": 7.322330470336314e-07, + "loss": 0.5343, + "step": 2896 + }, + { + "epoch": 4.540187891440501, + "grad_norm": 1.243160605430603, + "learning_rate": 7.307439743406691e-07, + "loss": 0.5873, + "step": 2897 + }, + { + "epoch": 4.541753653444676, + "grad_norm": 1.190952181816101, + "learning_rate": 7.292561580785696e-07, + "loss": 0.648, + "step": 2898 + }, + { + "epoch": 4.543319415448852, + "grad_norm": 1.27082359790802, + "learning_rate": 7.277695993038996e-07, + "loss": 0.5988, + "step": 2899 + }, + { + "epoch": 4.544885177453027, + "grad_norm": 1.2360665798187256, + "learning_rate": 7.26284299072334e-07, + "loss": 0.607, + "step": 2900 + }, + { + "epoch": 4.546450939457203, + "grad_norm": 1.1328120231628418, + "learning_rate": 7.248002584386535e-07, + "loss": 0.658, + "step": 2901 + }, + { + "epoch": 4.5480167014613775, + "grad_norm": 1.173593282699585, + "learning_rate": 7.233174784567443e-07, + "loss": 0.6941, + "step": 2902 + }, + { + "epoch": 4.549582463465553, + "grad_norm": 1.2456272840499878, + "learning_rate": 7.218359601795979e-07, + "loss": 0.6372, + "step": 2903 + }, + { + "epoch": 4.551148225469729, + "grad_norm": 1.2233866453170776, + "learning_rate": 7.203557046593068e-07, + "loss": 0.5841, + "step": 2904 + }, + { + "epoch": 4.552713987473904, + "grad_norm": 1.4194185733795166, + "learning_rate": 7.188767129470717e-07, + "loss": 0.6086, + "step": 2905 + }, + { + "epoch": 4.554279749478079, + "grad_norm": 1.1789617538452148, + "learning_rate": 7.173989860931935e-07, + "loss": 0.5785, + "step": 2906 + }, + { + "epoch": 4.5558455114822545, + "grad_norm": 1.2355077266693115, + "learning_rate": 7.15922525147072e-07, + "loss": 0.6122, + "step": 2907 + }, + { + "epoch": 4.55741127348643, + "grad_norm": 1.1994844675064087, + "learning_rate": 7.144473311572136e-07, + "loss": 0.6097, + "step": 2908 + }, + { + "epoch": 4.558977035490606, + "grad_norm": 1.4132137298583984, + "learning_rate": 7.129734051712203e-07, + "loss": 0.6377, + "step": 2909 + }, + { + "epoch": 4.560542797494781, + "grad_norm": 1.1752851009368896, + "learning_rate": 7.115007482357953e-07, + "loss": 0.649, + "step": 2910 + }, + { + "epoch": 4.562108559498956, + "grad_norm": 1.2555209398269653, + "learning_rate": 7.100293613967424e-07, + "loss": 0.6313, + "step": 2911 + }, + { + "epoch": 4.5636743215031315, + "grad_norm": 1.2966251373291016, + "learning_rate": 7.085592456989599e-07, + "loss": 0.5969, + "step": 2912 + }, + { + "epoch": 4.565240083507307, + "grad_norm": 1.2769136428833008, + "learning_rate": 7.070904021864461e-07, + "loss": 0.6198, + "step": 2913 + }, + { + "epoch": 4.566805845511483, + "grad_norm": 1.304149866104126, + "learning_rate": 7.056228319022945e-07, + "loss": 0.4576, + "step": 2914 + }, + { + "epoch": 4.568371607515657, + "grad_norm": 1.3524707555770874, + "learning_rate": 7.041565358886954e-07, + "loss": 0.6173, + "step": 2915 + }, + { + "epoch": 4.569937369519833, + "grad_norm": 1.2801849842071533, + "learning_rate": 7.026915151869335e-07, + "loss": 0.6357, + "step": 2916 + }, + { + "epoch": 4.5715031315240084, + "grad_norm": 1.234526515007019, + "learning_rate": 7.012277708373877e-07, + "loss": 0.5631, + "step": 2917 + }, + { + "epoch": 4.573068893528184, + "grad_norm": 1.1721588373184204, + "learning_rate": 6.997653038795316e-07, + "loss": 0.5776, + "step": 2918 + }, + { + "epoch": 4.574634655532359, + "grad_norm": 1.2359575033187866, + "learning_rate": 6.983041153519302e-07, + "loss": 0.5767, + "step": 2919 + }, + { + "epoch": 4.576200417536534, + "grad_norm": 1.2173316478729248, + "learning_rate": 6.968442062922415e-07, + "loss": 0.6165, + "step": 2920 + }, + { + "epoch": 4.57776617954071, + "grad_norm": 1.3483846187591553, + "learning_rate": 6.953855777372157e-07, + "loss": 0.5667, + "step": 2921 + }, + { + "epoch": 4.579331941544885, + "grad_norm": 1.376077651977539, + "learning_rate": 6.939282307226908e-07, + "loss": 0.6666, + "step": 2922 + }, + { + "epoch": 4.580897703549061, + "grad_norm": 1.2124888896942139, + "learning_rate": 6.924721662835971e-07, + "loss": 0.4712, + "step": 2923 + }, + { + "epoch": 4.582463465553236, + "grad_norm": 1.1178332567214966, + "learning_rate": 6.910173854539551e-07, + "loss": 0.5903, + "step": 2924 + }, + { + "epoch": 4.584029227557411, + "grad_norm": 1.1721351146697998, + "learning_rate": 6.895638892668708e-07, + "loss": 0.6249, + "step": 2925 + }, + { + "epoch": 4.585594989561587, + "grad_norm": 1.2229433059692383, + "learning_rate": 6.881116787545394e-07, + "loss": 0.6408, + "step": 2926 + }, + { + "epoch": 4.5871607515657615, + "grad_norm": 1.378939151763916, + "learning_rate": 6.866607549482435e-07, + "loss": 0.6422, + "step": 2927 + }, + { + "epoch": 4.588726513569937, + "grad_norm": 1.3970625400543213, + "learning_rate": 6.852111188783512e-07, + "loss": 0.5783, + "step": 2928 + }, + { + "epoch": 4.590292275574113, + "grad_norm": 1.2715126276016235, + "learning_rate": 6.837627715743165e-07, + "loss": 0.607, + "step": 2929 + }, + { + "epoch": 4.591858037578288, + "grad_norm": 1.2084861993789673, + "learning_rate": 6.82315714064678e-07, + "loss": 0.5667, + "step": 2930 + }, + { + "epoch": 4.593423799582464, + "grad_norm": 1.2971402406692505, + "learning_rate": 6.808699473770586e-07, + "loss": 0.6281, + "step": 2931 + }, + { + "epoch": 4.5949895615866385, + "grad_norm": 1.2204023599624634, + "learning_rate": 6.794254725381641e-07, + "loss": 0.6225, + "step": 2932 + }, + { + "epoch": 4.596555323590814, + "grad_norm": 1.2303305864334106, + "learning_rate": 6.779822905737834e-07, + "loss": 0.6175, + "step": 2933 + }, + { + "epoch": 4.59812108559499, + "grad_norm": 1.252663254737854, + "learning_rate": 6.765404025087876e-07, + "loss": 0.6163, + "step": 2934 + }, + { + "epoch": 4.599686847599165, + "grad_norm": 1.3693299293518066, + "learning_rate": 6.750998093671263e-07, + "loss": 0.6057, + "step": 2935 + }, + { + "epoch": 4.60125260960334, + "grad_norm": 1.2067402601242065, + "learning_rate": 6.736605121718337e-07, + "loss": 0.6249, + "step": 2936 + }, + { + "epoch": 4.6028183716075155, + "grad_norm": 1.249069333076477, + "learning_rate": 6.722225119450216e-07, + "loss": 0.6229, + "step": 2937 + }, + { + "epoch": 4.604384133611691, + "grad_norm": 1.3225828409194946, + "learning_rate": 6.707858097078784e-07, + "loss": 0.6726, + "step": 2938 + }, + { + "epoch": 4.605949895615867, + "grad_norm": 1.2723355293273926, + "learning_rate": 6.69350406480676e-07, + "loss": 0.589, + "step": 2939 + }, + { + "epoch": 4.607515657620041, + "grad_norm": 1.2256345748901367, + "learning_rate": 6.679163032827593e-07, + "loss": 0.5873, + "step": 2940 + }, + { + "epoch": 4.609081419624217, + "grad_norm": 1.1579097509384155, + "learning_rate": 6.664835011325508e-07, + "loss": 0.616, + "step": 2941 + }, + { + "epoch": 4.6106471816283925, + "grad_norm": 1.2893751859664917, + "learning_rate": 6.650520010475525e-07, + "loss": 0.6022, + "step": 2942 + }, + { + "epoch": 4.612212943632568, + "grad_norm": 1.3577022552490234, + "learning_rate": 6.636218040443366e-07, + "loss": 0.6644, + "step": 2943 + }, + { + "epoch": 4.613778705636744, + "grad_norm": 1.2338060140609741, + "learning_rate": 6.621929111385533e-07, + "loss": 0.6048, + "step": 2944 + }, + { + "epoch": 4.615344467640918, + "grad_norm": 1.2044262886047363, + "learning_rate": 6.607653233449257e-07, + "loss": 0.6546, + "step": 2945 + }, + { + "epoch": 4.616910229645094, + "grad_norm": 1.4120864868164062, + "learning_rate": 6.593390416772505e-07, + "loss": 0.6433, + "step": 2946 + }, + { + "epoch": 4.6184759916492695, + "grad_norm": 1.2405333518981934, + "learning_rate": 6.57914067148396e-07, + "loss": 0.5755, + "step": 2947 + }, + { + "epoch": 4.620041753653445, + "grad_norm": 1.2720845937728882, + "learning_rate": 6.564904007703032e-07, + "loss": 0.5998, + "step": 2948 + }, + { + "epoch": 4.62160751565762, + "grad_norm": 1.316292405128479, + "learning_rate": 6.550680435539836e-07, + "loss": 0.5557, + "step": 2949 + }, + { + "epoch": 4.623173277661795, + "grad_norm": 1.274463176727295, + "learning_rate": 6.536469965095193e-07, + "loss": 0.5593, + "step": 2950 + }, + { + "epoch": 4.624739039665971, + "grad_norm": 1.177530288696289, + "learning_rate": 6.522272606460614e-07, + "loss": 0.6317, + "step": 2951 + }, + { + "epoch": 4.6263048016701465, + "grad_norm": 1.313736915588379, + "learning_rate": 6.508088369718305e-07, + "loss": 0.652, + "step": 2952 + }, + { + "epoch": 4.627870563674321, + "grad_norm": 1.2262637615203857, + "learning_rate": 6.493917264941149e-07, + "loss": 0.5847, + "step": 2953 + }, + { + "epoch": 4.629436325678497, + "grad_norm": 1.1686118841171265, + "learning_rate": 6.479759302192712e-07, + "loss": 0.6166, + "step": 2954 + }, + { + "epoch": 4.631002087682672, + "grad_norm": 1.2170262336730957, + "learning_rate": 6.46561449152722e-07, + "loss": 0.6207, + "step": 2955 + }, + { + "epoch": 4.632567849686848, + "grad_norm": 1.2834949493408203, + "learning_rate": 6.45148284298954e-07, + "loss": 0.5227, + "step": 2956 + }, + { + "epoch": 4.6341336116910234, + "grad_norm": 1.237107515335083, + "learning_rate": 6.43736436661524e-07, + "loss": 0.6574, + "step": 2957 + }, + { + "epoch": 4.635699373695198, + "grad_norm": 1.2491086721420288, + "learning_rate": 6.423259072430496e-07, + "loss": 0.6345, + "step": 2958 + }, + { + "epoch": 4.637265135699374, + "grad_norm": 1.2701539993286133, + "learning_rate": 6.409166970452118e-07, + "loss": 0.5826, + "step": 2959 + }, + { + "epoch": 4.638830897703549, + "grad_norm": 1.3169174194335938, + "learning_rate": 6.395088070687588e-07, + "loss": 0.6026, + "step": 2960 + }, + { + "epoch": 4.640396659707724, + "grad_norm": 1.1995527744293213, + "learning_rate": 6.381022383134966e-07, + "loss": 0.6577, + "step": 2961 + }, + { + "epoch": 4.6419624217118995, + "grad_norm": 1.2342932224273682, + "learning_rate": 6.36696991778295e-07, + "loss": 0.6074, + "step": 2962 + }, + { + "epoch": 4.643528183716075, + "grad_norm": 1.2346426248550415, + "learning_rate": 6.352930684610867e-07, + "loss": 0.594, + "step": 2963 + }, + { + "epoch": 4.645093945720251, + "grad_norm": 1.1832835674285889, + "learning_rate": 6.33890469358861e-07, + "loss": 0.6149, + "step": 2964 + }, + { + "epoch": 4.646659707724426, + "grad_norm": 1.2718520164489746, + "learning_rate": 6.324891954676693e-07, + "loss": 0.5602, + "step": 2965 + }, + { + "epoch": 4.648225469728601, + "grad_norm": 1.2307515144348145, + "learning_rate": 6.310892477826211e-07, + "loss": 0.6094, + "step": 2966 + }, + { + "epoch": 4.6497912317327765, + "grad_norm": 1.3244339227676392, + "learning_rate": 6.296906272978845e-07, + "loss": 0.6468, + "step": 2967 + }, + { + "epoch": 4.651356993736952, + "grad_norm": 1.3322306871414185, + "learning_rate": 6.282933350066853e-07, + "loss": 0.5508, + "step": 2968 + }, + { + "epoch": 4.652922755741128, + "grad_norm": 1.316904067993164, + "learning_rate": 6.268973719013036e-07, + "loss": 0.641, + "step": 2969 + }, + { + "epoch": 4.654488517745302, + "grad_norm": 1.1402695178985596, + "learning_rate": 6.2550273897308e-07, + "loss": 0.5811, + "step": 2970 + }, + { + "epoch": 4.656054279749478, + "grad_norm": 1.331150770187378, + "learning_rate": 6.241094372124079e-07, + "loss": 0.6364, + "step": 2971 + }, + { + "epoch": 4.6576200417536535, + "grad_norm": 1.3078680038452148, + "learning_rate": 6.227174676087333e-07, + "loss": 0.595, + "step": 2972 + }, + { + "epoch": 4.659185803757829, + "grad_norm": 1.1026049852371216, + "learning_rate": 6.213268311505616e-07, + "loss": 0.6616, + "step": 2973 + }, + { + "epoch": 4.660751565762004, + "grad_norm": 1.2403117418289185, + "learning_rate": 6.199375288254461e-07, + "loss": 0.5761, + "step": 2974 + }, + { + "epoch": 4.662317327766179, + "grad_norm": 1.258516788482666, + "learning_rate": 6.185495616199949e-07, + "loss": 0.5506, + "step": 2975 + }, + { + "epoch": 4.663883089770355, + "grad_norm": 1.2604827880859375, + "learning_rate": 6.171629305198698e-07, + "loss": 0.5802, + "step": 2976 + }, + { + "epoch": 4.6654488517745305, + "grad_norm": 1.1909523010253906, + "learning_rate": 6.157776365097803e-07, + "loss": 0.6558, + "step": 2977 + }, + { + "epoch": 4.667014613778706, + "grad_norm": 1.2381128072738647, + "learning_rate": 6.143936805734881e-07, + "loss": 0.6345, + "step": 2978 + }, + { + "epoch": 4.668580375782881, + "grad_norm": 1.3379584550857544, + "learning_rate": 6.130110636938055e-07, + "loss": 0.5533, + "step": 2979 + }, + { + "epoch": 4.670146137787056, + "grad_norm": 1.335717797279358, + "learning_rate": 6.11629786852592e-07, + "loss": 0.6319, + "step": 2980 + }, + { + "epoch": 4.671711899791232, + "grad_norm": 1.246821641921997, + "learning_rate": 6.102498510307569e-07, + "loss": 0.6484, + "step": 2981 + }, + { + "epoch": 4.673277661795407, + "grad_norm": 1.2727938890457153, + "learning_rate": 6.088712572082566e-07, + "loss": 0.6236, + "step": 2982 + }, + { + "epoch": 4.674843423799582, + "grad_norm": 1.0983227491378784, + "learning_rate": 6.074940063640946e-07, + "loss": 0.595, + "step": 2983 + }, + { + "epoch": 4.676409185803758, + "grad_norm": 1.2299860715866089, + "learning_rate": 6.061180994763211e-07, + "loss": 0.592, + "step": 2984 + }, + { + "epoch": 4.677974947807933, + "grad_norm": 1.316436529159546, + "learning_rate": 6.047435375220306e-07, + "loss": 0.5987, + "step": 2985 + }, + { + "epoch": 4.679540709812109, + "grad_norm": 1.2510743141174316, + "learning_rate": 6.03370321477364e-07, + "loss": 0.6229, + "step": 2986 + }, + { + "epoch": 4.681106471816284, + "grad_norm": 1.2618223428726196, + "learning_rate": 6.019984523175057e-07, + "loss": 0.6101, + "step": 2987 + }, + { + "epoch": 4.682672233820459, + "grad_norm": 1.23671293258667, + "learning_rate": 6.006279310166835e-07, + "loss": 0.6364, + "step": 2988 + }, + { + "epoch": 4.684237995824635, + "grad_norm": 1.2197179794311523, + "learning_rate": 5.99258758548169e-07, + "loss": 0.5805, + "step": 2989 + }, + { + "epoch": 4.68580375782881, + "grad_norm": 1.2514771223068237, + "learning_rate": 5.978909358842732e-07, + "loss": 0.6141, + "step": 2990 + }, + { + "epoch": 4.687369519832985, + "grad_norm": 1.2339932918548584, + "learning_rate": 5.965244639963524e-07, + "loss": 0.5687, + "step": 2991 + }, + { + "epoch": 4.688935281837161, + "grad_norm": 1.3678441047668457, + "learning_rate": 5.95159343854802e-07, + "loss": 0.6436, + "step": 2992 + }, + { + "epoch": 4.690501043841336, + "grad_norm": 1.1420538425445557, + "learning_rate": 5.937955764290549e-07, + "loss": 0.6132, + "step": 2993 + }, + { + "epoch": 4.692066805845512, + "grad_norm": 1.3040863275527954, + "learning_rate": 5.924331626875885e-07, + "loss": 0.6149, + "step": 2994 + }, + { + "epoch": 4.693632567849686, + "grad_norm": 1.190412998199463, + "learning_rate": 5.910721035979139e-07, + "loss": 0.5449, + "step": 2995 + }, + { + "epoch": 4.695198329853862, + "grad_norm": 1.354242205619812, + "learning_rate": 5.897124001265822e-07, + "loss": 0.5768, + "step": 2996 + }, + { + "epoch": 4.696764091858038, + "grad_norm": 1.3582462072372437, + "learning_rate": 5.883540532391846e-07, + "loss": 0.5956, + "step": 2997 + }, + { + "epoch": 4.698329853862213, + "grad_norm": 1.1748749017715454, + "learning_rate": 5.869970639003439e-07, + "loss": 0.5381, + "step": 2998 + }, + { + "epoch": 4.699895615866389, + "grad_norm": 1.226905345916748, + "learning_rate": 5.85641433073722e-07, + "loss": 0.6287, + "step": 2999 + }, + { + "epoch": 4.701461377870563, + "grad_norm": 1.3289525508880615, + "learning_rate": 5.842871617220153e-07, + "loss": 0.6064, + "step": 3000 + }, + { + "epoch": 4.703027139874739, + "grad_norm": 1.2593392133712769, + "learning_rate": 5.829342508069546e-07, + "loss": 0.6537, + "step": 3001 + }, + { + "epoch": 4.7045929018789145, + "grad_norm": 1.2690147161483765, + "learning_rate": 5.81582701289305e-07, + "loss": 0.6201, + "step": 3002 + }, + { + "epoch": 4.70615866388309, + "grad_norm": 1.2982971668243408, + "learning_rate": 5.802325141288643e-07, + "loss": 0.5869, + "step": 3003 + }, + { + "epoch": 4.707724425887265, + "grad_norm": 1.2125791311264038, + "learning_rate": 5.788836902844633e-07, + "loss": 0.6744, + "step": 3004 + }, + { + "epoch": 4.70929018789144, + "grad_norm": 1.2710962295532227, + "learning_rate": 5.775362307139645e-07, + "loss": 0.5961, + "step": 3005 + }, + { + "epoch": 4.710855949895616, + "grad_norm": 1.3088948726654053, + "learning_rate": 5.761901363742614e-07, + "loss": 0.5572, + "step": 3006 + }, + { + "epoch": 4.7124217118997915, + "grad_norm": 1.2394931316375732, + "learning_rate": 5.74845408221279e-07, + "loss": 0.6722, + "step": 3007 + }, + { + "epoch": 4.713987473903966, + "grad_norm": 1.390460729598999, + "learning_rate": 5.735020472099687e-07, + "loss": 0.6166, + "step": 3008 + }, + { + "epoch": 4.715553235908142, + "grad_norm": 1.2904356718063354, + "learning_rate": 5.721600542943159e-07, + "loss": 0.5555, + "step": 3009 + }, + { + "epoch": 4.717118997912317, + "grad_norm": 1.1953577995300293, + "learning_rate": 5.708194304273323e-07, + "loss": 0.5141, + "step": 3010 + }, + { + "epoch": 4.718684759916493, + "grad_norm": 1.3318582773208618, + "learning_rate": 5.694801765610547e-07, + "loss": 0.557, + "step": 3011 + }, + { + "epoch": 4.7202505219206685, + "grad_norm": 1.344522476196289, + "learning_rate": 5.681422936465522e-07, + "loss": 0.5855, + "step": 3012 + }, + { + "epoch": 4.721816283924843, + "grad_norm": 1.270960807800293, + "learning_rate": 5.66805782633916e-07, + "loss": 0.6152, + "step": 3013 + }, + { + "epoch": 4.723382045929019, + "grad_norm": 1.4696820974349976, + "learning_rate": 5.654706444722646e-07, + "loss": 0.5986, + "step": 3014 + }, + { + "epoch": 4.724947807933194, + "grad_norm": 1.2178717851638794, + "learning_rate": 5.641368801097424e-07, + "loss": 0.653, + "step": 3015 + }, + { + "epoch": 4.726513569937369, + "grad_norm": 1.2786749601364136, + "learning_rate": 5.628044904935168e-07, + "loss": 0.6124, + "step": 3016 + }, + { + "epoch": 4.728079331941545, + "grad_norm": 1.2376960515975952, + "learning_rate": 5.614734765697799e-07, + "loss": 0.6467, + "step": 3017 + }, + { + "epoch": 4.72964509394572, + "grad_norm": 1.2517948150634766, + "learning_rate": 5.60143839283746e-07, + "loss": 0.6772, + "step": 3018 + }, + { + "epoch": 4.731210855949896, + "grad_norm": 1.1662477254867554, + "learning_rate": 5.588155795796533e-07, + "loss": 0.6071, + "step": 3019 + }, + { + "epoch": 4.732776617954071, + "grad_norm": 1.2507755756378174, + "learning_rate": 5.574886984007602e-07, + "loss": 0.5908, + "step": 3020 + }, + { + "epoch": 4.734342379958246, + "grad_norm": 1.1487728357315063, + "learning_rate": 5.561631966893452e-07, + "loss": 0.5582, + "step": 3021 + }, + { + "epoch": 4.735908141962422, + "grad_norm": 1.185463786125183, + "learning_rate": 5.548390753867106e-07, + "loss": 0.6161, + "step": 3022 + }, + { + "epoch": 4.737473903966597, + "grad_norm": 1.255568265914917, + "learning_rate": 5.535163354331763e-07, + "loss": 0.6313, + "step": 3023 + }, + { + "epoch": 4.739039665970773, + "grad_norm": 1.231364130973816, + "learning_rate": 5.521949777680796e-07, + "loss": 0.6123, + "step": 3024 + }, + { + "epoch": 4.740605427974947, + "grad_norm": 1.2934824228286743, + "learning_rate": 5.508750033297805e-07, + "loss": 0.6017, + "step": 3025 + }, + { + "epoch": 4.742171189979123, + "grad_norm": 1.20609450340271, + "learning_rate": 5.495564130556519e-07, + "loss": 0.586, + "step": 3026 + }, + { + "epoch": 4.743736951983299, + "grad_norm": 1.3231621980667114, + "learning_rate": 5.482392078820864e-07, + "loss": 0.5804, + "step": 3027 + }, + { + "epoch": 4.745302713987474, + "grad_norm": 1.168235182762146, + "learning_rate": 5.469233887444941e-07, + "loss": 0.602, + "step": 3028 + }, + { + "epoch": 4.746868475991649, + "grad_norm": 1.235970377922058, + "learning_rate": 5.456089565772976e-07, + "loss": 0.5859, + "step": 3029 + }, + { + "epoch": 4.748434237995824, + "grad_norm": 1.3735214471817017, + "learning_rate": 5.442959123139369e-07, + "loss": 0.4943, + "step": 3030 + }, + { + "epoch": 4.75, + "grad_norm": 1.2391469478607178, + "learning_rate": 5.429842568868654e-07, + "loss": 0.5896, + "step": 3031 + }, + { + "epoch": 4.751565762004176, + "grad_norm": 1.3225210905075073, + "learning_rate": 5.41673991227551e-07, + "loss": 0.5941, + "step": 3032 + }, + { + "epoch": 4.753131524008351, + "grad_norm": 1.2474992275238037, + "learning_rate": 5.403651162664738e-07, + "loss": 0.6647, + "step": 3033 + }, + { + "epoch": 4.754697286012526, + "grad_norm": 1.2459207773208618, + "learning_rate": 5.390576329331267e-07, + "loss": 0.5771, + "step": 3034 + }, + { + "epoch": 4.756263048016701, + "grad_norm": 1.2632790803909302, + "learning_rate": 5.377515421560147e-07, + "loss": 0.6667, + "step": 3035 + }, + { + "epoch": 4.757828810020877, + "grad_norm": 1.316978096961975, + "learning_rate": 5.36446844862653e-07, + "loss": 0.5392, + "step": 3036 + }, + { + "epoch": 4.759394572025053, + "grad_norm": 1.3076261281967163, + "learning_rate": 5.351435419795681e-07, + "loss": 0.6489, + "step": 3037 + }, + { + "epoch": 4.760960334029227, + "grad_norm": 1.197774052619934, + "learning_rate": 5.338416344322958e-07, + "loss": 0.5918, + "step": 3038 + }, + { + "epoch": 4.762526096033403, + "grad_norm": 1.1374410390853882, + "learning_rate": 5.325411231453811e-07, + "loss": 0.6153, + "step": 3039 + }, + { + "epoch": 4.764091858037578, + "grad_norm": 1.2959661483764648, + "learning_rate": 5.312420090423775e-07, + "loss": 0.487, + "step": 3040 + }, + { + "epoch": 4.765657620041754, + "grad_norm": 1.263738989830017, + "learning_rate": 5.299442930458468e-07, + "loss": 0.5549, + "step": 3041 + }, + { + "epoch": 4.767223382045929, + "grad_norm": 1.333878755569458, + "learning_rate": 5.286479760773553e-07, + "loss": 0.6261, + "step": 3042 + }, + { + "epoch": 4.768789144050104, + "grad_norm": 1.3264974355697632, + "learning_rate": 5.273530590574802e-07, + "loss": 0.5951, + "step": 3043 + }, + { + "epoch": 4.77035490605428, + "grad_norm": 1.252334475517273, + "learning_rate": 5.260595429058021e-07, + "loss": 0.5427, + "step": 3044 + }, + { + "epoch": 4.771920668058455, + "grad_norm": 1.233492374420166, + "learning_rate": 5.247674285409046e-07, + "loss": 0.6738, + "step": 3045 + }, + { + "epoch": 4.773486430062631, + "grad_norm": 1.2096290588378906, + "learning_rate": 5.234767168803815e-07, + "loss": 0.5748, + "step": 3046 + }, + { + "epoch": 4.775052192066806, + "grad_norm": 1.2133762836456299, + "learning_rate": 5.221874088408241e-07, + "loss": 0.6234, + "step": 3047 + }, + { + "epoch": 4.776617954070981, + "grad_norm": 1.2251299619674683, + "learning_rate": 5.208995053378305e-07, + "loss": 0.6235, + "step": 3048 + }, + { + "epoch": 4.778183716075157, + "grad_norm": 1.2990975379943848, + "learning_rate": 5.196130072860028e-07, + "loss": 0.6395, + "step": 3049 + }, + { + "epoch": 4.7797494780793315, + "grad_norm": 1.2998696565628052, + "learning_rate": 5.183279155989409e-07, + "loss": 0.6271, + "step": 3050 + }, + { + "epoch": 4.781315240083507, + "grad_norm": 1.2194361686706543, + "learning_rate": 5.170442311892488e-07, + "loss": 0.6563, + "step": 3051 + }, + { + "epoch": 4.782881002087683, + "grad_norm": 1.3821429014205933, + "learning_rate": 5.1576195496853e-07, + "loss": 0.6717, + "step": 3052 + }, + { + "epoch": 4.784446764091858, + "grad_norm": 1.3171210289001465, + "learning_rate": 5.144810878473891e-07, + "loss": 0.5997, + "step": 3053 + }, + { + "epoch": 4.786012526096034, + "grad_norm": 1.251981258392334, + "learning_rate": 5.132016307354287e-07, + "loss": 0.6034, + "step": 3054 + }, + { + "epoch": 4.7875782881002085, + "grad_norm": 1.3084864616394043, + "learning_rate": 5.119235845412512e-07, + "loss": 0.5528, + "step": 3055 + }, + { + "epoch": 4.789144050104384, + "grad_norm": 1.2730029821395874, + "learning_rate": 5.106469501724564e-07, + "loss": 0.5746, + "step": 3056 + }, + { + "epoch": 4.79070981210856, + "grad_norm": 1.2262660264968872, + "learning_rate": 5.093717285356414e-07, + "loss": 0.6208, + "step": 3057 + }, + { + "epoch": 4.792275574112735, + "grad_norm": 1.2117376327514648, + "learning_rate": 5.080979205364006e-07, + "loss": 0.5987, + "step": 3058 + }, + { + "epoch": 4.79384133611691, + "grad_norm": 1.2486417293548584, + "learning_rate": 5.068255270793245e-07, + "loss": 0.5414, + "step": 3059 + }, + { + "epoch": 4.7954070981210855, + "grad_norm": 1.231568455696106, + "learning_rate": 5.055545490679981e-07, + "loss": 0.6141, + "step": 3060 + }, + { + "epoch": 4.796972860125261, + "grad_norm": 1.3257794380187988, + "learning_rate": 5.042849874050012e-07, + "loss": 0.6121, + "step": 3061 + }, + { + "epoch": 4.798538622129437, + "grad_norm": 1.2174944877624512, + "learning_rate": 5.030168429919108e-07, + "loss": 0.6314, + "step": 3062 + }, + { + "epoch": 4.800104384133611, + "grad_norm": 1.3967257738113403, + "learning_rate": 5.017501167292931e-07, + "loss": 0.614, + "step": 3063 + }, + { + "epoch": 4.801670146137787, + "grad_norm": 1.0998293161392212, + "learning_rate": 5.004848095167103e-07, + "loss": 0.614, + "step": 3064 + }, + { + "epoch": 4.803235908141962, + "grad_norm": 1.228244662284851, + "learning_rate": 4.99220922252715e-07, + "loss": 0.5525, + "step": 3065 + }, + { + "epoch": 4.804801670146138, + "grad_norm": 1.257345199584961, + "learning_rate": 4.979584558348529e-07, + "loss": 0.6794, + "step": 3066 + }, + { + "epoch": 4.806367432150314, + "grad_norm": 1.240714192390442, + "learning_rate": 4.966974111596598e-07, + "loss": 0.626, + "step": 3067 + }, + { + "epoch": 4.807933194154488, + "grad_norm": 1.2045975923538208, + "learning_rate": 4.954377891226623e-07, + "loss": 0.6175, + "step": 3068 + }, + { + "epoch": 4.809498956158664, + "grad_norm": 1.317455530166626, + "learning_rate": 4.94179590618376e-07, + "loss": 0.5919, + "step": 3069 + }, + { + "epoch": 4.811064718162839, + "grad_norm": 1.2258387804031372, + "learning_rate": 4.929228165403066e-07, + "loss": 0.6348, + "step": 3070 + }, + { + "epoch": 4.812630480167015, + "grad_norm": 1.1854313611984253, + "learning_rate": 4.916674677809475e-07, + "loss": 0.6007, + "step": 3071 + }, + { + "epoch": 4.81419624217119, + "grad_norm": 1.3538085222244263, + "learning_rate": 4.904135452317807e-07, + "loss": 0.5999, + "step": 3072 + }, + { + "epoch": 4.815762004175365, + "grad_norm": 1.1934870481491089, + "learning_rate": 4.89161049783273e-07, + "loss": 0.5818, + "step": 3073 + }, + { + "epoch": 4.817327766179541, + "grad_norm": 1.3245843648910522, + "learning_rate": 4.879099823248814e-07, + "loss": 0.6442, + "step": 3074 + }, + { + "epoch": 4.818893528183716, + "grad_norm": 1.1364349126815796, + "learning_rate": 4.866603437450473e-07, + "loss": 0.5472, + "step": 3075 + }, + { + "epoch": 4.820459290187891, + "grad_norm": 1.250402808189392, + "learning_rate": 4.854121349311949e-07, + "loss": 0.5438, + "step": 3076 + }, + { + "epoch": 4.822025052192067, + "grad_norm": 1.2647368907928467, + "learning_rate": 4.84165356769738e-07, + "loss": 0.6582, + "step": 3077 + }, + { + "epoch": 4.823590814196242, + "grad_norm": 1.2903023958206177, + "learning_rate": 4.829200101460696e-07, + "loss": 0.6657, + "step": 3078 + }, + { + "epoch": 4.825156576200418, + "grad_norm": 1.1874885559082031, + "learning_rate": 4.816760959445682e-07, + "loss": 0.5539, + "step": 3079 + }, + { + "epoch": 4.826722338204593, + "grad_norm": 1.1794021129608154, + "learning_rate": 4.804336150485969e-07, + "loss": 0.6093, + "step": 3080 + }, + { + "epoch": 4.828288100208768, + "grad_norm": 1.2572773694992065, + "learning_rate": 4.791925683404971e-07, + "loss": 0.6647, + "step": 3081 + }, + { + "epoch": 4.829853862212944, + "grad_norm": 1.2959238290786743, + "learning_rate": 4.779529567015948e-07, + "loss": 0.6719, + "step": 3082 + }, + { + "epoch": 4.831419624217119, + "grad_norm": 1.2031400203704834, + "learning_rate": 4.767147810121953e-07, + "loss": 0.6111, + "step": 3083 + }, + { + "epoch": 4.832985386221294, + "grad_norm": 1.2516471147537231, + "learning_rate": 4.7547804215158476e-07, + "loss": 0.5653, + "step": 3084 + }, + { + "epoch": 4.8345511482254695, + "grad_norm": 1.305241584777832, + "learning_rate": 4.7424274099802883e-07, + "loss": 0.607, + "step": 3085 + }, + { + "epoch": 4.836116910229645, + "grad_norm": 1.3069261312484741, + "learning_rate": 4.730088784287723e-07, + "loss": 0.6446, + "step": 3086 + }, + { + "epoch": 4.837682672233821, + "grad_norm": 1.1565463542938232, + "learning_rate": 4.717764553200385e-07, + "loss": 0.6627, + "step": 3087 + }, + { + "epoch": 4.839248434237996, + "grad_norm": 1.3230305910110474, + "learning_rate": 4.7054547254702793e-07, + "loss": 0.5788, + "step": 3088 + }, + { + "epoch": 4.840814196242171, + "grad_norm": 1.3158049583435059, + "learning_rate": 4.6931593098391874e-07, + "loss": 0.6677, + "step": 3089 + }, + { + "epoch": 4.8423799582463465, + "grad_norm": 1.222174048423767, + "learning_rate": 4.680878315038656e-07, + "loss": 0.554, + "step": 3090 + }, + { + "epoch": 4.843945720250522, + "grad_norm": 1.339778184890747, + "learning_rate": 4.668611749789989e-07, + "loss": 0.6612, + "step": 3091 + }, + { + "epoch": 4.845511482254698, + "grad_norm": 1.3295081853866577, + "learning_rate": 4.6563596228042433e-07, + "loss": 0.6592, + "step": 3092 + }, + { + "epoch": 4.847077244258872, + "grad_norm": 1.2802320718765259, + "learning_rate": 4.6441219427822314e-07, + "loss": 0.5961, + "step": 3093 + }, + { + "epoch": 4.848643006263048, + "grad_norm": 1.2144367694854736, + "learning_rate": 4.6318987184144795e-07, + "loss": 0.5894, + "step": 3094 + }, + { + "epoch": 4.8502087682672235, + "grad_norm": 1.2537792921066284, + "learning_rate": 4.619689958381282e-07, + "loss": 0.6544, + "step": 3095 + }, + { + "epoch": 4.851774530271399, + "grad_norm": 1.2162483930587769, + "learning_rate": 4.607495671352652e-07, + "loss": 0.6051, + "step": 3096 + }, + { + "epoch": 4.853340292275574, + "grad_norm": 1.2548906803131104, + "learning_rate": 4.5953158659882923e-07, + "loss": 0.6362, + "step": 3097 + }, + { + "epoch": 4.854906054279749, + "grad_norm": 1.2650794982910156, + "learning_rate": 4.583150550937682e-07, + "loss": 0.5981, + "step": 3098 + }, + { + "epoch": 4.856471816283925, + "grad_norm": 1.1979948282241821, + "learning_rate": 4.570999734839951e-07, + "loss": 0.6477, + "step": 3099 + }, + { + "epoch": 4.8580375782881005, + "grad_norm": 1.3879836797714233, + "learning_rate": 4.558863426323962e-07, + "loss": 0.6126, + "step": 3100 + }, + { + "epoch": 4.859603340292276, + "grad_norm": 1.2028707265853882, + "learning_rate": 4.5467416340082864e-07, + "loss": 0.6146, + "step": 3101 + }, + { + "epoch": 4.861169102296451, + "grad_norm": 1.2319929599761963, + "learning_rate": 4.534634366501159e-07, + "loss": 0.6523, + "step": 3102 + }, + { + "epoch": 4.862734864300626, + "grad_norm": 1.2286549806594849, + "learning_rate": 4.522541632400512e-07, + "loss": 0.6516, + "step": 3103 + }, + { + "epoch": 4.864300626304802, + "grad_norm": 1.1734979152679443, + "learning_rate": 4.5104634402939605e-07, + "loss": 0.5758, + "step": 3104 + }, + { + "epoch": 4.8658663883089766, + "grad_norm": 1.3007832765579224, + "learning_rate": 4.49839979875879e-07, + "loss": 0.5611, + "step": 3105 + }, + { + "epoch": 4.867432150313152, + "grad_norm": 1.2126667499542236, + "learning_rate": 4.486350716361959e-07, + "loss": 0.5952, + "step": 3106 + }, + { + "epoch": 4.868997912317328, + "grad_norm": 1.2437777519226074, + "learning_rate": 4.4743162016600615e-07, + "loss": 0.6277, + "step": 3107 + }, + { + "epoch": 4.870563674321503, + "grad_norm": 1.1449244022369385, + "learning_rate": 4.462296263199381e-07, + "loss": 0.5828, + "step": 3108 + }, + { + "epoch": 4.872129436325679, + "grad_norm": 1.3628195524215698, + "learning_rate": 4.450290909515839e-07, + "loss": 0.622, + "step": 3109 + }, + { + "epoch": 4.8736951983298535, + "grad_norm": 1.1359556913375854, + "learning_rate": 4.438300149134975e-07, + "loss": 0.6068, + "step": 3110 + }, + { + "epoch": 4.875260960334029, + "grad_norm": 1.2504936456680298, + "learning_rate": 4.426323990572004e-07, + "loss": 0.6443, + "step": 3111 + }, + { + "epoch": 4.876826722338205, + "grad_norm": 1.1604193449020386, + "learning_rate": 4.4143624423317403e-07, + "loss": 0.6291, + "step": 3112 + }, + { + "epoch": 4.87839248434238, + "grad_norm": 1.3639174699783325, + "learning_rate": 4.402415512908631e-07, + "loss": 0.6624, + "step": 3113 + }, + { + "epoch": 4.879958246346555, + "grad_norm": 1.3654690980911255, + "learning_rate": 4.3904832107867666e-07, + "loss": 0.5647, + "step": 3114 + }, + { + "epoch": 4.8815240083507305, + "grad_norm": 1.0890389680862427, + "learning_rate": 4.3785655444398087e-07, + "loss": 0.6207, + "step": 3115 + }, + { + "epoch": 4.883089770354906, + "grad_norm": 1.3411756753921509, + "learning_rate": 4.366662522331053e-07, + "loss": 0.5585, + "step": 3116 + }, + { + "epoch": 4.884655532359082, + "grad_norm": 1.1507508754730225, + "learning_rate": 4.3547741529133854e-07, + "loss": 0.682, + "step": 3117 + }, + { + "epoch": 4.886221294363256, + "grad_norm": 1.3855395317077637, + "learning_rate": 4.3429004446292945e-07, + "loss": 0.5744, + "step": 3118 + }, + { + "epoch": 4.887787056367432, + "grad_norm": 1.211582064628601, + "learning_rate": 4.3310414059108465e-07, + "loss": 0.6033, + "step": 3119 + }, + { + "epoch": 4.8893528183716075, + "grad_norm": 1.209214687347412, + "learning_rate": 4.319197045179696e-07, + "loss": 0.5845, + "step": 3120 + }, + { + "epoch": 4.890918580375783, + "grad_norm": 1.2045565843582153, + "learning_rate": 4.3073673708470766e-07, + "loss": 0.5807, + "step": 3121 + }, + { + "epoch": 4.892484342379959, + "grad_norm": 1.264220118522644, + "learning_rate": 4.295552391313787e-07, + "loss": 0.6604, + "step": 3122 + }, + { + "epoch": 4.894050104384133, + "grad_norm": 1.2991132736206055, + "learning_rate": 4.2837521149701937e-07, + "loss": 0.7245, + "step": 3123 + }, + { + "epoch": 4.895615866388309, + "grad_norm": 1.3955750465393066, + "learning_rate": 4.27196655019623e-07, + "loss": 0.6339, + "step": 3124 + }, + { + "epoch": 4.8971816283924845, + "grad_norm": 1.3281763792037964, + "learning_rate": 4.260195705361347e-07, + "loss": 0.6138, + "step": 3125 + }, + { + "epoch": 4.89874739039666, + "grad_norm": 1.2890803813934326, + "learning_rate": 4.2484395888245935e-07, + "loss": 0.6588, + "step": 3126 + }, + { + "epoch": 4.900313152400835, + "grad_norm": 1.2077101469039917, + "learning_rate": 4.236698208934531e-07, + "loss": 0.626, + "step": 3127 + }, + { + "epoch": 4.90187891440501, + "grad_norm": 1.1384289264678955, + "learning_rate": 4.224971574029241e-07, + "loss": 0.6468, + "step": 3128 + }, + { + "epoch": 4.903444676409186, + "grad_norm": 1.319321632385254, + "learning_rate": 4.2132596924363666e-07, + "loss": 0.6104, + "step": 3129 + }, + { + "epoch": 4.9050104384133615, + "grad_norm": 1.3868818283081055, + "learning_rate": 4.201562572473064e-07, + "loss": 0.6325, + "step": 3130 + }, + { + "epoch": 4.906576200417536, + "grad_norm": 1.281776785850525, + "learning_rate": 4.18988022244598e-07, + "loss": 0.5742, + "step": 3131 + }, + { + "epoch": 4.908141962421712, + "grad_norm": 1.2451962232589722, + "learning_rate": 4.1782126506513196e-07, + "loss": 0.5826, + "step": 3132 + }, + { + "epoch": 4.909707724425887, + "grad_norm": 1.189673662185669, + "learning_rate": 4.166559865374747e-07, + "loss": 0.6543, + "step": 3133 + }, + { + "epoch": 4.911273486430063, + "grad_norm": 1.3293418884277344, + "learning_rate": 4.154921874891446e-07, + "loss": 0.6165, + "step": 3134 + }, + { + "epoch": 4.9128392484342385, + "grad_norm": 1.3895233869552612, + "learning_rate": 4.1432986874661156e-07, + "loss": 0.6627, + "step": 3135 + }, + { + "epoch": 4.914405010438413, + "grad_norm": 1.3557475805282593, + "learning_rate": 4.131690311352898e-07, + "loss": 0.598, + "step": 3136 + }, + { + "epoch": 4.915970772442589, + "grad_norm": 1.2514588832855225, + "learning_rate": 4.1200967547954525e-07, + "loss": 0.5649, + "step": 3137 + }, + { + "epoch": 4.917536534446764, + "grad_norm": 1.248897910118103, + "learning_rate": 4.1085180260268975e-07, + "loss": 0.6182, + "step": 3138 + }, + { + "epoch": 4.919102296450939, + "grad_norm": 1.254090666770935, + "learning_rate": 4.0969541332698257e-07, + "loss": 0.523, + "step": 3139 + }, + { + "epoch": 4.920668058455115, + "grad_norm": 1.2779461145401, + "learning_rate": 4.0854050847362966e-07, + "loss": 0.6123, + "step": 3140 + }, + { + "epoch": 4.92223382045929, + "grad_norm": 1.2242066860198975, + "learning_rate": 4.0738708886278233e-07, + "loss": 0.6661, + "step": 3141 + }, + { + "epoch": 4.923799582463466, + "grad_norm": 1.0954583883285522, + "learning_rate": 4.062351553135377e-07, + "loss": 0.5705, + "step": 3142 + }, + { + "epoch": 4.925365344467641, + "grad_norm": 1.4686397314071655, + "learning_rate": 4.050847086439369e-07, + "loss": 0.6784, + "step": 3143 + }, + { + "epoch": 4.926931106471816, + "grad_norm": 1.094277262687683, + "learning_rate": 4.0393574967096586e-07, + "loss": 0.598, + "step": 3144 + }, + { + "epoch": 4.9284968684759916, + "grad_norm": 1.2568544149398804, + "learning_rate": 4.0278827921055447e-07, + "loss": 0.6422, + "step": 3145 + }, + { + "epoch": 4.930062630480167, + "grad_norm": 1.3163423538208008, + "learning_rate": 4.016422980775725e-07, + "loss": 0.6132, + "step": 3146 + }, + { + "epoch": 4.931628392484343, + "grad_norm": 1.2578885555267334, + "learning_rate": 4.0049780708583647e-07, + "loss": 0.6313, + "step": 3147 + }, + { + "epoch": 4.933194154488517, + "grad_norm": 1.1483683586120605, + "learning_rate": 3.9935480704810237e-07, + "loss": 0.5864, + "step": 3148 + }, + { + "epoch": 4.934759916492693, + "grad_norm": 1.239707589149475, + "learning_rate": 3.982132987760659e-07, + "loss": 0.5784, + "step": 3149 + }, + { + "epoch": 4.9363256784968685, + "grad_norm": 1.2554842233657837, + "learning_rate": 3.970732830803678e-07, + "loss": 0.5633, + "step": 3150 + }, + { + "epoch": 4.937891440501044, + "grad_norm": 1.1999422311782837, + "learning_rate": 3.959347607705838e-07, + "loss": 0.6405, + "step": 3151 + }, + { + "epoch": 4.939457202505219, + "grad_norm": 1.3461389541625977, + "learning_rate": 3.9479773265523186e-07, + "loss": 0.6856, + "step": 3152 + }, + { + "epoch": 4.941022964509394, + "grad_norm": 1.3426072597503662, + "learning_rate": 3.9366219954177e-07, + "loss": 0.6218, + "step": 3153 + }, + { + "epoch": 4.94258872651357, + "grad_norm": 1.2127430438995361, + "learning_rate": 3.925281622365915e-07, + "loss": 0.5842, + "step": 3154 + }, + { + "epoch": 4.9441544885177455, + "grad_norm": 1.2543373107910156, + "learning_rate": 3.913956215450293e-07, + "loss": 0.6345, + "step": 3155 + }, + { + "epoch": 4.945720250521921, + "grad_norm": 1.3203067779541016, + "learning_rate": 3.9026457827135324e-07, + "loss": 0.6336, + "step": 3156 + }, + { + "epoch": 4.947286012526096, + "grad_norm": 1.2462279796600342, + "learning_rate": 3.891350332187696e-07, + "loss": 0.5859, + "step": 3157 + }, + { + "epoch": 4.948851774530271, + "grad_norm": 1.3469984531402588, + "learning_rate": 3.8800698718942126e-07, + "loss": 0.5761, + "step": 3158 + }, + { + "epoch": 4.950417536534447, + "grad_norm": 1.1929227113723755, + "learning_rate": 3.868804409843846e-07, + "loss": 0.5895, + "step": 3159 + }, + { + "epoch": 4.9519832985386225, + "grad_norm": 1.2928577661514282, + "learning_rate": 3.8575539540367383e-07, + "loss": 0.6293, + "step": 3160 + }, + { + "epoch": 4.953549060542797, + "grad_norm": 1.2409569025039673, + "learning_rate": 3.8463185124623657e-07, + "loss": 0.5194, + "step": 3161 + }, + { + "epoch": 4.955114822546973, + "grad_norm": 1.1920491456985474, + "learning_rate": 3.8350980930995124e-07, + "loss": 0.6085, + "step": 3162 + }, + { + "epoch": 4.956680584551148, + "grad_norm": 1.1986509561538696, + "learning_rate": 3.8238927039163487e-07, + "loss": 0.5688, + "step": 3163 + }, + { + "epoch": 4.958246346555324, + "grad_norm": 1.313879370689392, + "learning_rate": 3.812702352870321e-07, + "loss": 0.622, + "step": 3164 + }, + { + "epoch": 4.959812108559499, + "grad_norm": 1.205352544784546, + "learning_rate": 3.801527047908218e-07, + "loss": 0.604, + "step": 3165 + }, + { + "epoch": 4.961377870563674, + "grad_norm": 1.2026801109313965, + "learning_rate": 3.7903667969661605e-07, + "loss": 0.5921, + "step": 3166 + }, + { + "epoch": 4.96294363256785, + "grad_norm": 1.324131965637207, + "learning_rate": 3.779221607969541e-07, + "loss": 0.6334, + "step": 3167 + }, + { + "epoch": 4.964509394572025, + "grad_norm": 1.314497947692871, + "learning_rate": 3.7680914888330865e-07, + "loss": 0.6152, + "step": 3168 + }, + { + "epoch": 4.966075156576201, + "grad_norm": 1.3103183507919312, + "learning_rate": 3.7569764474608085e-07, + "loss": 0.649, + "step": 3169 + }, + { + "epoch": 4.967640918580376, + "grad_norm": 1.3193435668945312, + "learning_rate": 3.7458764917460146e-07, + "loss": 0.688, + "step": 3170 + }, + { + "epoch": 4.969206680584551, + "grad_norm": 1.3641090393066406, + "learning_rate": 3.7347916295712967e-07, + "loss": 0.6085, + "step": 3171 + }, + { + "epoch": 4.970772442588727, + "grad_norm": 1.1966512203216553, + "learning_rate": 3.723721868808533e-07, + "loss": 0.573, + "step": 3172 + }, + { + "epoch": 4.972338204592901, + "grad_norm": 1.12050199508667, + "learning_rate": 3.712667217318874e-07, + "loss": 0.6238, + "step": 3173 + }, + { + "epoch": 4.973903966597077, + "grad_norm": 1.4070502519607544, + "learning_rate": 3.701627682952741e-07, + "loss": 0.63, + "step": 3174 + }, + { + "epoch": 4.975469728601253, + "grad_norm": 1.147275686264038, + "learning_rate": 3.690603273549817e-07, + "loss": 0.6603, + "step": 3175 + }, + { + "epoch": 4.977035490605428, + "grad_norm": 1.2202069759368896, + "learning_rate": 3.6795939969390544e-07, + "loss": 0.6294, + "step": 3176 + }, + { + "epoch": 4.978601252609604, + "grad_norm": 1.2288408279418945, + "learning_rate": 3.668599860938643e-07, + "loss": 0.6612, + "step": 3177 + }, + { + "epoch": 4.980167014613778, + "grad_norm": 1.2829551696777344, + "learning_rate": 3.657620873356035e-07, + "loss": 0.652, + "step": 3178 + }, + { + "epoch": 4.981732776617954, + "grad_norm": 1.230684757232666, + "learning_rate": 3.646657041987922e-07, + "loss": 0.5631, + "step": 3179 + }, + { + "epoch": 4.98329853862213, + "grad_norm": 1.1531054973602295, + "learning_rate": 3.6357083746202173e-07, + "loss": 0.5369, + "step": 3180 + }, + { + "epoch": 4.984864300626305, + "grad_norm": 1.212597131729126, + "learning_rate": 3.6247748790280895e-07, + "loss": 0.562, + "step": 3181 + }, + { + "epoch": 4.98643006263048, + "grad_norm": 1.318061351776123, + "learning_rate": 3.613856562975923e-07, + "loss": 0.5543, + "step": 3182 + }, + { + "epoch": 4.987995824634655, + "grad_norm": 1.1155426502227783, + "learning_rate": 3.6029534342173073e-07, + "loss": 0.6585, + "step": 3183 + }, + { + "epoch": 4.989561586638831, + "grad_norm": 1.4041640758514404, + "learning_rate": 3.592065500495079e-07, + "loss": 0.5962, + "step": 3184 + }, + { + "epoch": 4.991127348643007, + "grad_norm": 1.3740663528442383, + "learning_rate": 3.581192769541253e-07, + "loss": 0.6219, + "step": 3185 + }, + { + "epoch": 4.992693110647181, + "grad_norm": 1.2645107507705688, + "learning_rate": 3.570335249077056e-07, + "loss": 0.6126, + "step": 3186 + }, + { + "epoch": 4.994258872651357, + "grad_norm": 1.3721750974655151, + "learning_rate": 3.559492946812934e-07, + "loss": 0.6416, + "step": 3187 + }, + { + "epoch": 4.995824634655532, + "grad_norm": 1.257168173789978, + "learning_rate": 3.5486658704484977e-07, + "loss": 0.6237, + "step": 3188 + }, + { + "epoch": 4.997390396659708, + "grad_norm": 1.2237387895584106, + "learning_rate": 3.537854027672558e-07, + "loss": 0.5944, + "step": 3189 + }, + { + "epoch": 4.9989561586638835, + "grad_norm": 1.3416454792022705, + "learning_rate": 3.5270574261631054e-07, + "loss": 0.614, + "step": 3190 + }, + { + "epoch": 5.001565762004176, + "grad_norm": 2.7915163040161133, + "learning_rate": 3.516276073587313e-07, + "loss": 1.1565, + "step": 3191 + }, + { + "epoch": 5.003131524008351, + "grad_norm": 1.23666512966156, + "learning_rate": 3.5055099776015176e-07, + "loss": 0.6385, + "step": 3192 + }, + { + "epoch": 5.004697286012526, + "grad_norm": 1.2803066968917847, + "learning_rate": 3.494759145851223e-07, + "loss": 0.6114, + "step": 3193 + }, + { + "epoch": 5.006263048016701, + "grad_norm": 1.2515772581100464, + "learning_rate": 3.484023585971094e-07, + "loss": 0.5448, + "step": 3194 + }, + { + "epoch": 5.007828810020877, + "grad_norm": 1.2075544595718384, + "learning_rate": 3.473303305584949e-07, + "loss": 0.6052, + "step": 3195 + }, + { + "epoch": 5.009394572025053, + "grad_norm": 1.189892053604126, + "learning_rate": 3.4625983123057624e-07, + "loss": 0.5926, + "step": 3196 + }, + { + "epoch": 5.010960334029227, + "grad_norm": 1.4415851831436157, + "learning_rate": 3.4519086137356465e-07, + "loss": 0.5865, + "step": 3197 + }, + { + "epoch": 5.012526096033403, + "grad_norm": 1.1303807497024536, + "learning_rate": 3.4412342174658385e-07, + "loss": 0.574, + "step": 3198 + }, + { + "epoch": 5.014091858037578, + "grad_norm": 1.1930426359176636, + "learning_rate": 3.430575131076741e-07, + "loss": 0.6401, + "step": 3199 + }, + { + "epoch": 5.015657620041754, + "grad_norm": 1.166683316230774, + "learning_rate": 3.4199313621378625e-07, + "loss": 0.6505, + "step": 3200 + }, + { + "epoch": 5.017223382045929, + "grad_norm": 1.0946024656295776, + "learning_rate": 3.4093029182078326e-07, + "loss": 0.6337, + "step": 3201 + }, + { + "epoch": 5.018789144050104, + "grad_norm": 1.3278331756591797, + "learning_rate": 3.3986898068344017e-07, + "loss": 0.6693, + "step": 3202 + }, + { + "epoch": 5.02035490605428, + "grad_norm": 1.1627013683319092, + "learning_rate": 3.3880920355544357e-07, + "loss": 0.5224, + "step": 3203 + }, + { + "epoch": 5.021920668058455, + "grad_norm": 1.2656400203704834, + "learning_rate": 3.3775096118939033e-07, + "loss": 0.5572, + "step": 3204 + }, + { + "epoch": 5.02348643006263, + "grad_norm": 1.334549903869629, + "learning_rate": 3.3669425433678766e-07, + "loss": 0.4875, + "step": 3205 + }, + { + "epoch": 5.025052192066806, + "grad_norm": 1.0639917850494385, + "learning_rate": 3.356390837480522e-07, + "loss": 0.6216, + "step": 3206 + }, + { + "epoch": 5.026617954070981, + "grad_norm": 1.3102185726165771, + "learning_rate": 3.345854501725093e-07, + "loss": 0.5622, + "step": 3207 + }, + { + "epoch": 5.028183716075157, + "grad_norm": 1.3327429294586182, + "learning_rate": 3.335333543583935e-07, + "loss": 0.5919, + "step": 3208 + }, + { + "epoch": 5.029749478079332, + "grad_norm": 1.1817301511764526, + "learning_rate": 3.324827970528463e-07, + "loss": 0.5751, + "step": 3209 + }, + { + "epoch": 5.031315240083507, + "grad_norm": 1.443637728691101, + "learning_rate": 3.314337790019184e-07, + "loss": 0.5826, + "step": 3210 + }, + { + "epoch": 5.032881002087683, + "grad_norm": 1.223143219947815, + "learning_rate": 3.3038630095056426e-07, + "loss": 0.5676, + "step": 3211 + }, + { + "epoch": 5.034446764091858, + "grad_norm": 1.1428143978118896, + "learning_rate": 3.2934036364264845e-07, + "loss": 0.5272, + "step": 3212 + }, + { + "epoch": 5.036012526096034, + "grad_norm": 1.2499232292175293, + "learning_rate": 3.2829596782093965e-07, + "loss": 0.5976, + "step": 3213 + }, + { + "epoch": 5.0375782881002085, + "grad_norm": 1.3135937452316284, + "learning_rate": 3.2725311422710985e-07, + "loss": 0.5609, + "step": 3214 + }, + { + "epoch": 5.039144050104384, + "grad_norm": 1.1771061420440674, + "learning_rate": 3.262118036017406e-07, + "loss": 0.5371, + "step": 3215 + }, + { + "epoch": 5.04070981210856, + "grad_norm": 1.2707849740982056, + "learning_rate": 3.2517203668431254e-07, + "loss": 0.5451, + "step": 3216 + }, + { + "epoch": 5.042275574112735, + "grad_norm": 1.2625596523284912, + "learning_rate": 3.2413381421321307e-07, + "loss": 0.5235, + "step": 3217 + }, + { + "epoch": 5.04384133611691, + "grad_norm": 1.1677110195159912, + "learning_rate": 3.2309713692573355e-07, + "loss": 0.5649, + "step": 3218 + }, + { + "epoch": 5.0454070981210855, + "grad_norm": 1.265136957168579, + "learning_rate": 3.2206200555806504e-07, + "loss": 0.6077, + "step": 3219 + }, + { + "epoch": 5.046972860125261, + "grad_norm": 1.318882942199707, + "learning_rate": 3.2102842084530293e-07, + "loss": 0.6005, + "step": 3220 + }, + { + "epoch": 5.048538622129437, + "grad_norm": 1.1624666452407837, + "learning_rate": 3.19996383521444e-07, + "loss": 0.5583, + "step": 3221 + }, + { + "epoch": 5.050104384133611, + "grad_norm": 1.2922674417495728, + "learning_rate": 3.1896589431938546e-07, + "loss": 0.6166, + "step": 3222 + }, + { + "epoch": 5.051670146137787, + "grad_norm": 1.3477156162261963, + "learning_rate": 3.1793695397092566e-07, + "loss": 0.6443, + "step": 3223 + }, + { + "epoch": 5.053235908141962, + "grad_norm": 1.2025333642959595, + "learning_rate": 3.169095632067631e-07, + "loss": 0.5686, + "step": 3224 + }, + { + "epoch": 5.054801670146138, + "grad_norm": 1.347083568572998, + "learning_rate": 3.1588372275649566e-07, + "loss": 0.5893, + "step": 3225 + }, + { + "epoch": 5.056367432150314, + "grad_norm": 1.1417858600616455, + "learning_rate": 3.148594333486199e-07, + "loss": 0.6308, + "step": 3226 + }, + { + "epoch": 5.057933194154488, + "grad_norm": 1.140698790550232, + "learning_rate": 3.1383669571053156e-07, + "loss": 0.5876, + "step": 3227 + }, + { + "epoch": 5.059498956158664, + "grad_norm": 1.2835367918014526, + "learning_rate": 3.128155105685243e-07, + "loss": 0.6531, + "step": 3228 + }, + { + "epoch": 5.061064718162839, + "grad_norm": 1.2449157238006592, + "learning_rate": 3.11795878647789e-07, + "loss": 0.5998, + "step": 3229 + }, + { + "epoch": 5.062630480167015, + "grad_norm": 1.3161486387252808, + "learning_rate": 3.1077780067241355e-07, + "loss": 0.5826, + "step": 3230 + }, + { + "epoch": 5.06419624217119, + "grad_norm": 1.2425956726074219, + "learning_rate": 3.097612773653827e-07, + "loss": 0.5579, + "step": 3231 + }, + { + "epoch": 5.065762004175365, + "grad_norm": 1.1990351676940918, + "learning_rate": 3.0874630944857596e-07, + "loss": 0.6454, + "step": 3232 + }, + { + "epoch": 5.067327766179541, + "grad_norm": 1.3438416719436646, + "learning_rate": 3.077328976427699e-07, + "loss": 0.539, + "step": 3233 + }, + { + "epoch": 5.068893528183716, + "grad_norm": 1.247770071029663, + "learning_rate": 3.067210426676359e-07, + "loss": 0.5977, + "step": 3234 + }, + { + "epoch": 5.070459290187891, + "grad_norm": 1.256011724472046, + "learning_rate": 3.0571074524173765e-07, + "loss": 0.512, + "step": 3235 + }, + { + "epoch": 5.072025052192067, + "grad_norm": 1.1846981048583984, + "learning_rate": 3.0470200608253594e-07, + "loss": 0.5655, + "step": 3236 + }, + { + "epoch": 5.073590814196242, + "grad_norm": 1.3106032609939575, + "learning_rate": 3.036948259063818e-07, + "loss": 0.6444, + "step": 3237 + }, + { + "epoch": 5.075156576200418, + "grad_norm": 1.1632797718048096, + "learning_rate": 3.0268920542852074e-07, + "loss": 0.6264, + "step": 3238 + }, + { + "epoch": 5.0767223382045925, + "grad_norm": 1.1612005233764648, + "learning_rate": 3.016851453630923e-07, + "loss": 0.5708, + "step": 3239 + }, + { + "epoch": 5.078288100208768, + "grad_norm": 1.3502079248428345, + "learning_rate": 3.006826464231244e-07, + "loss": 0.5929, + "step": 3240 + }, + { + "epoch": 5.079853862212944, + "grad_norm": 1.255334496498108, + "learning_rate": 2.996817093205387e-07, + "loss": 0.5924, + "step": 3241 + }, + { + "epoch": 5.081419624217119, + "grad_norm": 1.3202862739562988, + "learning_rate": 2.9868233476614715e-07, + "loss": 0.6686, + "step": 3242 + }, + { + "epoch": 5.082985386221294, + "grad_norm": 1.3197559118270874, + "learning_rate": 2.976845234696518e-07, + "loss": 0.5892, + "step": 3243 + }, + { + "epoch": 5.0845511482254695, + "grad_norm": 1.1784745454788208, + "learning_rate": 2.96688276139645e-07, + "loss": 0.6715, + "step": 3244 + }, + { + "epoch": 5.086116910229645, + "grad_norm": 1.3473583459854126, + "learning_rate": 2.9569359348360804e-07, + "loss": 0.5744, + "step": 3245 + }, + { + "epoch": 5.087682672233821, + "grad_norm": 1.1897343397140503, + "learning_rate": 2.9470047620791143e-07, + "loss": 0.5942, + "step": 3246 + }, + { + "epoch": 5.089248434237996, + "grad_norm": 1.2343817949295044, + "learning_rate": 2.9370892501781434e-07, + "loss": 0.6083, + "step": 3247 + }, + { + "epoch": 5.090814196242171, + "grad_norm": 1.1905215978622437, + "learning_rate": 2.927189406174619e-07, + "loss": 0.5538, + "step": 3248 + }, + { + "epoch": 5.0923799582463465, + "grad_norm": 1.4438196420669556, + "learning_rate": 2.9173052370988997e-07, + "loss": 0.494, + "step": 3249 + }, + { + "epoch": 5.093945720250522, + "grad_norm": 1.2399394512176514, + "learning_rate": 2.9074367499701745e-07, + "loss": 0.5808, + "step": 3250 + }, + { + "epoch": 5.095511482254698, + "grad_norm": 1.1947413682937622, + "learning_rate": 2.8975839517965174e-07, + "loss": 0.5425, + "step": 3251 + }, + { + "epoch": 5.097077244258872, + "grad_norm": 1.347930669784546, + "learning_rate": 2.887746849574877e-07, + "loss": 0.6383, + "step": 3252 + }, + { + "epoch": 5.098643006263048, + "grad_norm": 1.1950582265853882, + "learning_rate": 2.877925450291014e-07, + "loss": 0.6735, + "step": 3253 + }, + { + "epoch": 5.1002087682672235, + "grad_norm": 1.2808043956756592, + "learning_rate": 2.868119760919569e-07, + "loss": 0.546, + "step": 3254 + }, + { + "epoch": 5.101774530271399, + "grad_norm": 1.3777042627334595, + "learning_rate": 2.8583297884240133e-07, + "loss": 0.5633, + "step": 3255 + }, + { + "epoch": 5.103340292275574, + "grad_norm": 1.141104817390442, + "learning_rate": 2.84855553975667e-07, + "loss": 0.6137, + "step": 3256 + }, + { + "epoch": 5.104906054279749, + "grad_norm": 1.376975178718567, + "learning_rate": 2.838797021858677e-07, + "loss": 0.5511, + "step": 3257 + }, + { + "epoch": 5.106471816283925, + "grad_norm": 1.1074012517929077, + "learning_rate": 2.8290542416600174e-07, + "loss": 0.5533, + "step": 3258 + }, + { + "epoch": 5.1080375782881005, + "grad_norm": 1.1775327920913696, + "learning_rate": 2.819327206079489e-07, + "loss": 0.6083, + "step": 3259 + }, + { + "epoch": 5.109603340292275, + "grad_norm": 1.1862449645996094, + "learning_rate": 2.809615922024711e-07, + "loss": 0.5849, + "step": 3260 + }, + { + "epoch": 5.111169102296451, + "grad_norm": 1.3680003881454468, + "learning_rate": 2.79992039639212e-07, + "loss": 0.5678, + "step": 3261 + }, + { + "epoch": 5.112734864300626, + "grad_norm": 1.2132647037506104, + "learning_rate": 2.790240636066963e-07, + "loss": 0.5872, + "step": 3262 + }, + { + "epoch": 5.114300626304802, + "grad_norm": 1.2291581630706787, + "learning_rate": 2.7805766479232694e-07, + "loss": 0.6209, + "step": 3263 + }, + { + "epoch": 5.115866388308977, + "grad_norm": 1.2505005598068237, + "learning_rate": 2.7709284388239026e-07, + "loss": 0.6064, + "step": 3264 + }, + { + "epoch": 5.117432150313152, + "grad_norm": 1.3502693176269531, + "learning_rate": 2.7612960156205073e-07, + "loss": 0.5915, + "step": 3265 + }, + { + "epoch": 5.118997912317328, + "grad_norm": 1.2608944177627563, + "learning_rate": 2.751679385153497e-07, + "loss": 0.5983, + "step": 3266 + }, + { + "epoch": 5.120563674321503, + "grad_norm": 1.1505229473114014, + "learning_rate": 2.7420785542521035e-07, + "loss": 0.5956, + "step": 3267 + }, + { + "epoch": 5.122129436325679, + "grad_norm": 1.265816569328308, + "learning_rate": 2.7324935297343146e-07, + "loss": 0.5514, + "step": 3268 + }, + { + "epoch": 5.1236951983298535, + "grad_norm": 1.3940019607543945, + "learning_rate": 2.722924318406897e-07, + "loss": 0.5997, + "step": 3269 + }, + { + "epoch": 5.125260960334029, + "grad_norm": 1.2447338104248047, + "learning_rate": 2.713370927065409e-07, + "loss": 0.5494, + "step": 3270 + }, + { + "epoch": 5.126826722338205, + "grad_norm": 1.2866191864013672, + "learning_rate": 2.7038333624941403e-07, + "loss": 0.5323, + "step": 3271 + }, + { + "epoch": 5.12839248434238, + "grad_norm": 1.135482668876648, + "learning_rate": 2.694311631466165e-07, + "loss": 0.6302, + "step": 3272 + }, + { + "epoch": 5.129958246346555, + "grad_norm": 1.2680634260177612, + "learning_rate": 2.684805740743307e-07, + "loss": 0.5755, + "step": 3273 + }, + { + "epoch": 5.1315240083507305, + "grad_norm": 1.254333257675171, + "learning_rate": 2.675315697076139e-07, + "loss": 0.5805, + "step": 3274 + }, + { + "epoch": 5.133089770354906, + "grad_norm": 1.1351419687271118, + "learning_rate": 2.665841507203984e-07, + "loss": 0.6145, + "step": 3275 + }, + { + "epoch": 5.134655532359082, + "grad_norm": 1.3801335096359253, + "learning_rate": 2.6563831778549015e-07, + "loss": 0.6033, + "step": 3276 + }, + { + "epoch": 5.136221294363256, + "grad_norm": 1.3474493026733398, + "learning_rate": 2.646940715745691e-07, + "loss": 0.5696, + "step": 3277 + }, + { + "epoch": 5.137787056367432, + "grad_norm": 1.229743242263794, + "learning_rate": 2.6375141275818835e-07, + "loss": 0.621, + "step": 3278 + }, + { + "epoch": 5.1393528183716075, + "grad_norm": 1.401671051979065, + "learning_rate": 2.62810342005774e-07, + "loss": 0.6163, + "step": 3279 + }, + { + "epoch": 5.140918580375783, + "grad_norm": 1.2571163177490234, + "learning_rate": 2.6187085998562373e-07, + "loss": 0.6026, + "step": 3280 + }, + { + "epoch": 5.142484342379959, + "grad_norm": 1.2606779336929321, + "learning_rate": 2.609329673649072e-07, + "loss": 0.473, + "step": 3281 + }, + { + "epoch": 5.144050104384133, + "grad_norm": 1.1929835081100464, + "learning_rate": 2.5999666480966606e-07, + "loss": 0.6195, + "step": 3282 + }, + { + "epoch": 5.145615866388309, + "grad_norm": 1.167556643486023, + "learning_rate": 2.5906195298481236e-07, + "loss": 0.5363, + "step": 3283 + }, + { + "epoch": 5.1471816283924845, + "grad_norm": 1.2752892971038818, + "learning_rate": 2.5812883255412704e-07, + "loss": 0.5821, + "step": 3284 + }, + { + "epoch": 5.14874739039666, + "grad_norm": 1.1993733644485474, + "learning_rate": 2.571973041802636e-07, + "loss": 0.6036, + "step": 3285 + }, + { + "epoch": 5.150313152400835, + "grad_norm": 1.3206144571304321, + "learning_rate": 2.5626736852474357e-07, + "loss": 0.5621, + "step": 3286 + }, + { + "epoch": 5.15187891440501, + "grad_norm": 1.292662501335144, + "learning_rate": 2.5533902624795605e-07, + "loss": 0.5751, + "step": 3287 + }, + { + "epoch": 5.153444676409186, + "grad_norm": 1.2682784795761108, + "learning_rate": 2.5441227800916206e-07, + "loss": 0.5896, + "step": 3288 + }, + { + "epoch": 5.1550104384133615, + "grad_norm": 1.3132630586624146, + "learning_rate": 2.534871244664869e-07, + "loss": 0.6055, + "step": 3289 + }, + { + "epoch": 5.156576200417536, + "grad_norm": 1.2162318229675293, + "learning_rate": 2.525635662769252e-07, + "loss": 0.6599, + "step": 3290 + }, + { + "epoch": 5.158141962421712, + "grad_norm": 1.2406466007232666, + "learning_rate": 2.516416040963399e-07, + "loss": 0.5714, + "step": 3291 + }, + { + "epoch": 5.159707724425887, + "grad_norm": 1.2157220840454102, + "learning_rate": 2.5072123857945773e-07, + "loss": 0.5679, + "step": 3292 + }, + { + "epoch": 5.161273486430063, + "grad_norm": 1.1547855138778687, + "learning_rate": 2.498024703798735e-07, + "loss": 0.536, + "step": 3293 + }, + { + "epoch": 5.162839248434238, + "grad_norm": 1.3439668416976929, + "learning_rate": 2.488853001500471e-07, + "loss": 0.5524, + "step": 3294 + }, + { + "epoch": 5.164405010438413, + "grad_norm": 1.1343441009521484, + "learning_rate": 2.479697285413038e-07, + "loss": 0.5694, + "step": 3295 + }, + { + "epoch": 5.165970772442589, + "grad_norm": 1.2414132356643677, + "learning_rate": 2.470557562038339e-07, + "loss": 0.618, + "step": 3296 + }, + { + "epoch": 5.167536534446764, + "grad_norm": 1.235949993133545, + "learning_rate": 2.4614338378669026e-07, + "loss": 0.6293, + "step": 3297 + }, + { + "epoch": 5.16910229645094, + "grad_norm": 1.3037298917770386, + "learning_rate": 2.4523261193779234e-07, + "loss": 0.6268, + "step": 3298 + }, + { + "epoch": 5.170668058455115, + "grad_norm": 1.248589277267456, + "learning_rate": 2.443234413039214e-07, + "loss": 0.59, + "step": 3299 + }, + { + "epoch": 5.17223382045929, + "grad_norm": 1.354189395904541, + "learning_rate": 2.4341587253072035e-07, + "loss": 0.5633, + "step": 3300 + }, + { + "epoch": 5.173799582463466, + "grad_norm": 1.332455039024353, + "learning_rate": 2.4250990626269763e-07, + "loss": 0.5885, + "step": 3301 + }, + { + "epoch": 5.175365344467641, + "grad_norm": 1.3065413236618042, + "learning_rate": 2.4160554314322097e-07, + "loss": 0.6089, + "step": 3302 + }, + { + "epoch": 5.176931106471816, + "grad_norm": 1.2889814376831055, + "learning_rate": 2.4070278381452013e-07, + "loss": 0.6184, + "step": 3303 + }, + { + "epoch": 5.1784968684759916, + "grad_norm": 1.1198309659957886, + "learning_rate": 2.3980162891768825e-07, + "loss": 0.5978, + "step": 3304 + }, + { + "epoch": 5.180062630480167, + "grad_norm": 1.2639782428741455, + "learning_rate": 2.3890207909267575e-07, + "loss": 0.6467, + "step": 3305 + }, + { + "epoch": 5.181628392484343, + "grad_norm": 1.3044084310531616, + "learning_rate": 2.380041349782952e-07, + "loss": 0.6309, + "step": 3306 + }, + { + "epoch": 5.183194154488517, + "grad_norm": 1.2028284072875977, + "learning_rate": 2.3710779721221821e-07, + "loss": 0.6076, + "step": 3307 + }, + { + "epoch": 5.184759916492693, + "grad_norm": 1.2832525968551636, + "learning_rate": 2.3621306643097613e-07, + "loss": 0.5625, + "step": 3308 + }, + { + "epoch": 5.1863256784968685, + "grad_norm": 1.1777451038360596, + "learning_rate": 2.3531994326995877e-07, + "loss": 0.584, + "step": 3309 + }, + { + "epoch": 5.187891440501044, + "grad_norm": 1.1548548936843872, + "learning_rate": 2.3442842836341406e-07, + "loss": 0.5425, + "step": 3310 + }, + { + "epoch": 5.189457202505219, + "grad_norm": 1.268331527709961, + "learning_rate": 2.3353852234444862e-07, + "loss": 0.6001, + "step": 3311 + }, + { + "epoch": 5.191022964509394, + "grad_norm": 1.369941234588623, + "learning_rate": 2.3265022584502556e-07, + "loss": 0.5804, + "step": 3312 + }, + { + "epoch": 5.19258872651357, + "grad_norm": 1.1923965215682983, + "learning_rate": 2.317635394959658e-07, + "loss": 0.5907, + "step": 3313 + }, + { + "epoch": 5.1941544885177455, + "grad_norm": 1.3358242511749268, + "learning_rate": 2.308784639269465e-07, + "loss": 0.6358, + "step": 3314 + }, + { + "epoch": 5.195720250521921, + "grad_norm": 1.2181466817855835, + "learning_rate": 2.2999499976650075e-07, + "loss": 0.5648, + "step": 3315 + }, + { + "epoch": 5.197286012526096, + "grad_norm": 1.3371413946151733, + "learning_rate": 2.2911314764201775e-07, + "loss": 0.5807, + "step": 3316 + }, + { + "epoch": 5.198851774530271, + "grad_norm": 1.3621243238449097, + "learning_rate": 2.282329081797416e-07, + "loss": 0.5705, + "step": 3317 + }, + { + "epoch": 5.200417536534447, + "grad_norm": 1.3168977499008179, + "learning_rate": 2.2735428200477028e-07, + "loss": 0.6016, + "step": 3318 + }, + { + "epoch": 5.2019832985386225, + "grad_norm": 1.3411412239074707, + "learning_rate": 2.2647726974105832e-07, + "loss": 0.5838, + "step": 3319 + }, + { + "epoch": 5.203549060542797, + "grad_norm": 1.3580838441848755, + "learning_rate": 2.256018720114128e-07, + "loss": 0.6054, + "step": 3320 + }, + { + "epoch": 5.205114822546973, + "grad_norm": 1.3712846040725708, + "learning_rate": 2.2472808943749287e-07, + "loss": 0.6094, + "step": 3321 + }, + { + "epoch": 5.206680584551148, + "grad_norm": 1.189433217048645, + "learning_rate": 2.2385592263981404e-07, + "loss": 0.6573, + "step": 3322 + }, + { + "epoch": 5.208246346555324, + "grad_norm": 1.3507299423217773, + "learning_rate": 2.2298537223774093e-07, + "loss": 0.5025, + "step": 3323 + }, + { + "epoch": 5.209812108559499, + "grad_norm": 1.0847887992858887, + "learning_rate": 2.221164388494923e-07, + "loss": 0.5786, + "step": 3324 + }, + { + "epoch": 5.211377870563674, + "grad_norm": 1.236777901649475, + "learning_rate": 2.2124912309213876e-07, + "loss": 0.5772, + "step": 3325 + }, + { + "epoch": 5.21294363256785, + "grad_norm": 1.246819257736206, + "learning_rate": 2.2038342558160064e-07, + "loss": 0.5479, + "step": 3326 + }, + { + "epoch": 5.214509394572025, + "grad_norm": 1.2399665117263794, + "learning_rate": 2.195193469326501e-07, + "loss": 0.5093, + "step": 3327 + }, + { + "epoch": 5.2160751565762, + "grad_norm": 1.1499581336975098, + "learning_rate": 2.1865688775890958e-07, + "loss": 0.5939, + "step": 3328 + }, + { + "epoch": 5.217640918580376, + "grad_norm": 1.20567786693573, + "learning_rate": 2.177960486728517e-07, + "loss": 0.5956, + "step": 3329 + }, + { + "epoch": 5.219206680584551, + "grad_norm": 1.3368107080459595, + "learning_rate": 2.169368302857977e-07, + "loss": 0.6054, + "step": 3330 + }, + { + "epoch": 5.220772442588727, + "grad_norm": 1.325939655303955, + "learning_rate": 2.1607923320791896e-07, + "loss": 0.6585, + "step": 3331 + }, + { + "epoch": 5.222338204592901, + "grad_norm": 1.3356062173843384, + "learning_rate": 2.1522325804823496e-07, + "loss": 0.5791, + "step": 3332 + }, + { + "epoch": 5.223903966597077, + "grad_norm": 1.2414422035217285, + "learning_rate": 2.1436890541461308e-07, + "loss": 0.5934, + "step": 3333 + }, + { + "epoch": 5.225469728601253, + "grad_norm": 1.160778522491455, + "learning_rate": 2.135161759137691e-07, + "loss": 0.5827, + "step": 3334 + }, + { + "epoch": 5.227035490605428, + "grad_norm": 1.2166001796722412, + "learning_rate": 2.1266507015126586e-07, + "loss": 0.6131, + "step": 3335 + }, + { + "epoch": 5.228601252609604, + "grad_norm": 1.4281837940216064, + "learning_rate": 2.1181558873151238e-07, + "loss": 0.6125, + "step": 3336 + }, + { + "epoch": 5.230167014613778, + "grad_norm": 1.3111658096313477, + "learning_rate": 2.1096773225776578e-07, + "loss": 0.6023, + "step": 3337 + }, + { + "epoch": 5.231732776617954, + "grad_norm": 1.4153447151184082, + "learning_rate": 2.101215013321284e-07, + "loss": 0.5552, + "step": 3338 + }, + { + "epoch": 5.23329853862213, + "grad_norm": 1.1794157028198242, + "learning_rate": 2.0927689655554723e-07, + "loss": 0.6257, + "step": 3339 + }, + { + "epoch": 5.234864300626305, + "grad_norm": 1.4494861364364624, + "learning_rate": 2.0843391852781558e-07, + "loss": 0.6027, + "step": 3340 + }, + { + "epoch": 5.23643006263048, + "grad_norm": 1.1171547174453735, + "learning_rate": 2.075925678475718e-07, + "loss": 0.6064, + "step": 3341 + }, + { + "epoch": 5.237995824634655, + "grad_norm": 1.2824617624282837, + "learning_rate": 2.0675284511229742e-07, + "loss": 0.6428, + "step": 3342 + }, + { + "epoch": 5.239561586638831, + "grad_norm": 1.2023221254348755, + "learning_rate": 2.0591475091831931e-07, + "loss": 0.6168, + "step": 3343 + }, + { + "epoch": 5.241127348643007, + "grad_norm": 1.3137130737304688, + "learning_rate": 2.0507828586080637e-07, + "loss": 0.6077, + "step": 3344 + }, + { + "epoch": 5.242693110647181, + "grad_norm": 1.3151475191116333, + "learning_rate": 2.0424345053377176e-07, + "loss": 0.677, + "step": 3345 + }, + { + "epoch": 5.244258872651357, + "grad_norm": 1.338149070739746, + "learning_rate": 2.0341024553007045e-07, + "loss": 0.6186, + "step": 3346 + }, + { + "epoch": 5.245824634655532, + "grad_norm": 1.2673715353012085, + "learning_rate": 2.0257867144140046e-07, + "loss": 0.6379, + "step": 3347 + }, + { + "epoch": 5.247390396659708, + "grad_norm": 1.19679856300354, + "learning_rate": 2.0174872885830117e-07, + "loss": 0.6038, + "step": 3348 + }, + { + "epoch": 5.2489561586638835, + "grad_norm": 1.3132461309432983, + "learning_rate": 2.0092041837015214e-07, + "loss": 0.5586, + "step": 3349 + }, + { + "epoch": 5.250521920668058, + "grad_norm": 1.1302937269210815, + "learning_rate": 2.000937405651765e-07, + "loss": 0.61, + "step": 3350 + }, + { + "epoch": 5.252087682672234, + "grad_norm": 1.208752989768982, + "learning_rate": 1.9926869603043659e-07, + "loss": 0.6054, + "step": 3351 + }, + { + "epoch": 5.253653444676409, + "grad_norm": 1.3355869054794312, + "learning_rate": 1.9844528535183343e-07, + "loss": 0.5996, + "step": 3352 + }, + { + "epoch": 5.255219206680585, + "grad_norm": 1.2324848175048828, + "learning_rate": 1.97623509114111e-07, + "loss": 0.6499, + "step": 3353 + }, + { + "epoch": 5.25678496868476, + "grad_norm": 1.2028673887252808, + "learning_rate": 1.9680336790084937e-07, + "loss": 0.5923, + "step": 3354 + }, + { + "epoch": 5.258350730688935, + "grad_norm": 1.1964545249938965, + "learning_rate": 1.9598486229446907e-07, + "loss": 0.6524, + "step": 3355 + }, + { + "epoch": 5.259916492693111, + "grad_norm": 1.4048811197280884, + "learning_rate": 1.9516799287622984e-07, + "loss": 0.5758, + "step": 3356 + }, + { + "epoch": 5.261482254697286, + "grad_norm": 1.1697945594787598, + "learning_rate": 1.9435276022622796e-07, + "loss": 0.5579, + "step": 3357 + }, + { + "epoch": 5.263048016701461, + "grad_norm": 1.3509764671325684, + "learning_rate": 1.935391649233978e-07, + "loss": 0.5641, + "step": 3358 + }, + { + "epoch": 5.264613778705637, + "grad_norm": 1.359906792640686, + "learning_rate": 1.9272720754551139e-07, + "loss": 0.5689, + "step": 3359 + }, + { + "epoch": 5.266179540709812, + "grad_norm": 1.36385178565979, + "learning_rate": 1.9191688866917747e-07, + "loss": 0.6205, + "step": 3360 + }, + { + "epoch": 5.267745302713988, + "grad_norm": 1.300125241279602, + "learning_rate": 1.9110820886984098e-07, + "loss": 0.6685, + "step": 3361 + }, + { + "epoch": 5.2693110647181625, + "grad_norm": 1.3445782661437988, + "learning_rate": 1.9030116872178317e-07, + "loss": 0.6333, + "step": 3362 + }, + { + "epoch": 5.270876826722338, + "grad_norm": 1.3689970970153809, + "learning_rate": 1.8949576879812055e-07, + "loss": 0.6039, + "step": 3363 + }, + { + "epoch": 5.272442588726514, + "grad_norm": 1.2647545337677002, + "learning_rate": 1.8869200967080503e-07, + "loss": 0.5857, + "step": 3364 + }, + { + "epoch": 5.274008350730689, + "grad_norm": 1.164014220237732, + "learning_rate": 1.8788989191062312e-07, + "loss": 0.6071, + "step": 3365 + }, + { + "epoch": 5.275574112734864, + "grad_norm": 1.2692615985870361, + "learning_rate": 1.870894160871964e-07, + "loss": 0.6164, + "step": 3366 + }, + { + "epoch": 5.2771398747390394, + "grad_norm": 1.144971489906311, + "learning_rate": 1.8629058276897955e-07, + "loss": 0.6325, + "step": 3367 + }, + { + "epoch": 5.278705636743215, + "grad_norm": 1.3754347562789917, + "learning_rate": 1.8549339252326103e-07, + "loss": 0.5394, + "step": 3368 + }, + { + "epoch": 5.280271398747391, + "grad_norm": 1.2376211881637573, + "learning_rate": 1.846978459161633e-07, + "loss": 0.5702, + "step": 3369 + }, + { + "epoch": 5.281837160751566, + "grad_norm": 1.4409040212631226, + "learning_rate": 1.8390394351263997e-07, + "loss": 0.619, + "step": 3370 + }, + { + "epoch": 5.283402922755741, + "grad_norm": 1.2496124505996704, + "learning_rate": 1.8311168587647865e-07, + "loss": 0.6094, + "step": 3371 + }, + { + "epoch": 5.284968684759916, + "grad_norm": 1.2824435234069824, + "learning_rate": 1.8232107357029877e-07, + "loss": 0.5281, + "step": 3372 + }, + { + "epoch": 5.286534446764092, + "grad_norm": 1.2305798530578613, + "learning_rate": 1.8153210715554953e-07, + "loss": 0.5538, + "step": 3373 + }, + { + "epoch": 5.288100208768268, + "grad_norm": 1.2084088325500488, + "learning_rate": 1.8074478719251414e-07, + "loss": 0.6187, + "step": 3374 + }, + { + "epoch": 5.289665970772442, + "grad_norm": 1.3996961116790771, + "learning_rate": 1.7995911424030372e-07, + "loss": 0.5601, + "step": 3375 + }, + { + "epoch": 5.291231732776618, + "grad_norm": 1.3905977010726929, + "learning_rate": 1.7917508885686135e-07, + "loss": 0.6463, + "step": 3376 + }, + { + "epoch": 5.292797494780793, + "grad_norm": 1.3874620199203491, + "learning_rate": 1.7839271159896088e-07, + "loss": 0.6219, + "step": 3377 + }, + { + "epoch": 5.294363256784969, + "grad_norm": 1.2685861587524414, + "learning_rate": 1.7761198302220368e-07, + "loss": 0.5364, + "step": 3378 + }, + { + "epoch": 5.295929018789144, + "grad_norm": 1.2921316623687744, + "learning_rate": 1.768329036810218e-07, + "loss": 0.5776, + "step": 3379 + }, + { + "epoch": 5.297494780793319, + "grad_norm": 1.2747477293014526, + "learning_rate": 1.7605547412867574e-07, + "loss": 0.5923, + "step": 3380 + }, + { + "epoch": 5.299060542797495, + "grad_norm": 1.261170506477356, + "learning_rate": 1.7527969491725388e-07, + "loss": 0.6459, + "step": 3381 + }, + { + "epoch": 5.30062630480167, + "grad_norm": 1.3607902526855469, + "learning_rate": 1.745055665976733e-07, + "loss": 0.6348, + "step": 3382 + }, + { + "epoch": 5.302192066805846, + "grad_norm": 1.3094136714935303, + "learning_rate": 1.7373308971967873e-07, + "loss": 0.7125, + "step": 3383 + }, + { + "epoch": 5.303757828810021, + "grad_norm": 1.2425636053085327, + "learning_rate": 1.7296226483184159e-07, + "loss": 0.6405, + "step": 3384 + }, + { + "epoch": 5.305323590814196, + "grad_norm": 1.3874858617782593, + "learning_rate": 1.7219309248156068e-07, + "loss": 0.6487, + "step": 3385 + }, + { + "epoch": 5.306889352818372, + "grad_norm": 1.1943751573562622, + "learning_rate": 1.7142557321506048e-07, + "loss": 0.5778, + "step": 3386 + }, + { + "epoch": 5.3084551148225465, + "grad_norm": 1.4116579294204712, + "learning_rate": 1.7065970757739302e-07, + "loss": 0.6459, + "step": 3387 + }, + { + "epoch": 5.310020876826722, + "grad_norm": 1.2798161506652832, + "learning_rate": 1.6989549611243412e-07, + "loss": 0.6387, + "step": 3388 + }, + { + "epoch": 5.311586638830898, + "grad_norm": 1.3025716543197632, + "learning_rate": 1.6913293936288543e-07, + "loss": 0.5692, + "step": 3389 + }, + { + "epoch": 5.313152400835073, + "grad_norm": 1.2523722648620605, + "learning_rate": 1.683720378702758e-07, + "loss": 0.6187, + "step": 3390 + }, + { + "epoch": 5.314718162839249, + "grad_norm": 1.3035849332809448, + "learning_rate": 1.6761279217495486e-07, + "loss": 0.6172, + "step": 3391 + }, + { + "epoch": 5.3162839248434235, + "grad_norm": 1.3494850397109985, + "learning_rate": 1.6685520281609913e-07, + "loss": 0.6287, + "step": 3392 + }, + { + "epoch": 5.317849686847599, + "grad_norm": 1.3379586935043335, + "learning_rate": 1.660992703317077e-07, + "loss": 0.5732, + "step": 3393 + }, + { + "epoch": 5.319415448851775, + "grad_norm": 1.316839575767517, + "learning_rate": 1.6534499525860338e-07, + "loss": 0.6073, + "step": 3394 + }, + { + "epoch": 5.32098121085595, + "grad_norm": 1.3319520950317383, + "learning_rate": 1.6459237813243184e-07, + "loss": 0.5583, + "step": 3395 + }, + { + "epoch": 5.322546972860125, + "grad_norm": 1.242706060409546, + "learning_rate": 1.638414194876617e-07, + "loss": 0.5879, + "step": 3396 + }, + { + "epoch": 5.3241127348643005, + "grad_norm": 1.3216476440429688, + "learning_rate": 1.6309211985758327e-07, + "loss": 0.6424, + "step": 3397 + }, + { + "epoch": 5.325678496868476, + "grad_norm": 1.2512398958206177, + "learning_rate": 1.6234447977430927e-07, + "loss": 0.5595, + "step": 3398 + }, + { + "epoch": 5.327244258872652, + "grad_norm": 1.3934729099273682, + "learning_rate": 1.6159849976877361e-07, + "loss": 0.5821, + "step": 3399 + }, + { + "epoch": 5.328810020876826, + "grad_norm": 1.228318452835083, + "learning_rate": 1.608541803707317e-07, + "loss": 0.5471, + "step": 3400 + }, + { + "epoch": 5.330375782881002, + "grad_norm": 1.3927935361862183, + "learning_rate": 1.601115221087582e-07, + "loss": 0.6739, + "step": 3401 + }, + { + "epoch": 5.3319415448851775, + "grad_norm": 1.1782944202423096, + "learning_rate": 1.5937052551024984e-07, + "loss": 0.6394, + "step": 3402 + }, + { + "epoch": 5.333507306889353, + "grad_norm": 1.1474275588989258, + "learning_rate": 1.586311911014235e-07, + "loss": 0.5717, + "step": 3403 + }, + { + "epoch": 5.335073068893529, + "grad_norm": 1.2519960403442383, + "learning_rate": 1.5789351940731334e-07, + "loss": 0.5492, + "step": 3404 + }, + { + "epoch": 5.336638830897703, + "grad_norm": 1.1698763370513916, + "learning_rate": 1.5715751095177588e-07, + "loss": 0.6226, + "step": 3405 + }, + { + "epoch": 5.338204592901879, + "grad_norm": 1.2749401330947876, + "learning_rate": 1.564231662574836e-07, + "loss": 0.6209, + "step": 3406 + }, + { + "epoch": 5.3397703549060545, + "grad_norm": 1.2376207113265991, + "learning_rate": 1.5569048584592905e-07, + "loss": 0.6255, + "step": 3407 + }, + { + "epoch": 5.34133611691023, + "grad_norm": 1.2711988687515259, + "learning_rate": 1.549594702374238e-07, + "loss": 0.5721, + "step": 3408 + }, + { + "epoch": 5.342901878914405, + "grad_norm": 1.3186453580856323, + "learning_rate": 1.5423011995109426e-07, + "loss": 0.5542, + "step": 3409 + }, + { + "epoch": 5.34446764091858, + "grad_norm": 1.1754095554351807, + "learning_rate": 1.5350243550488697e-07, + "loss": 0.5816, + "step": 3410 + }, + { + "epoch": 5.346033402922756, + "grad_norm": 1.2622367143630981, + "learning_rate": 1.527764174155641e-07, + "loss": 0.5997, + "step": 3411 + }, + { + "epoch": 5.347599164926931, + "grad_norm": 1.361920952796936, + "learning_rate": 1.520520661987049e-07, + "loss": 0.6131, + "step": 3412 + }, + { + "epoch": 5.349164926931106, + "grad_norm": 1.2565398216247559, + "learning_rate": 1.5132938236870481e-07, + "loss": 0.595, + "step": 3413 + }, + { + "epoch": 5.350730688935282, + "grad_norm": 1.203333854675293, + "learning_rate": 1.5060836643877503e-07, + "loss": 0.6203, + "step": 3414 + }, + { + "epoch": 5.352296450939457, + "grad_norm": 1.2418543100357056, + "learning_rate": 1.4988901892094288e-07, + "loss": 0.6364, + "step": 3415 + }, + { + "epoch": 5.353862212943633, + "grad_norm": 1.3369126319885254, + "learning_rate": 1.4917134032605003e-07, + "loss": 0.5941, + "step": 3416 + }, + { + "epoch": 5.3554279749478075, + "grad_norm": 1.1760553121566772, + "learning_rate": 1.4845533116375322e-07, + "loss": 0.6713, + "step": 3417 + }, + { + "epoch": 5.356993736951983, + "grad_norm": 1.2983871698379517, + "learning_rate": 1.4774099194252434e-07, + "loss": 0.5012, + "step": 3418 + }, + { + "epoch": 5.358559498956159, + "grad_norm": 1.2111915349960327, + "learning_rate": 1.470283231696487e-07, + "loss": 0.6084, + "step": 3419 + }, + { + "epoch": 5.360125260960334, + "grad_norm": 1.2792680263519287, + "learning_rate": 1.463173253512251e-07, + "loss": 0.4793, + "step": 3420 + }, + { + "epoch": 5.361691022964509, + "grad_norm": 1.1497719287872314, + "learning_rate": 1.456079989921669e-07, + "loss": 0.5297, + "step": 3421 + }, + { + "epoch": 5.3632567849686845, + "grad_norm": 1.3330738544464111, + "learning_rate": 1.4490034459619855e-07, + "loss": 0.5177, + "step": 3422 + }, + { + "epoch": 5.36482254697286, + "grad_norm": 1.2797367572784424, + "learning_rate": 1.4419436266585902e-07, + "loss": 0.5736, + "step": 3423 + }, + { + "epoch": 5.366388308977036, + "grad_norm": 1.2632088661193848, + "learning_rate": 1.434900537024994e-07, + "loss": 0.6572, + "step": 3424 + }, + { + "epoch": 5.367954070981211, + "grad_norm": 1.1909950971603394, + "learning_rate": 1.427874182062805e-07, + "loss": 0.608, + "step": 3425 + }, + { + "epoch": 5.369519832985386, + "grad_norm": 1.3106014728546143, + "learning_rate": 1.4208645667617844e-07, + "loss": 0.6373, + "step": 3426 + }, + { + "epoch": 5.3710855949895615, + "grad_norm": 1.2101953029632568, + "learning_rate": 1.4138716960997722e-07, + "loss": 0.6414, + "step": 3427 + }, + { + "epoch": 5.372651356993737, + "grad_norm": 1.316143274307251, + "learning_rate": 1.406895575042727e-07, + "loss": 0.5866, + "step": 3428 + }, + { + "epoch": 5.374217118997913, + "grad_norm": 1.4197582006454468, + "learning_rate": 1.399936208544736e-07, + "loss": 0.5069, + "step": 3429 + }, + { + "epoch": 5.375782881002087, + "grad_norm": 1.180269718170166, + "learning_rate": 1.3929936015479484e-07, + "loss": 0.6513, + "step": 3430 + }, + { + "epoch": 5.377348643006263, + "grad_norm": 1.2665772438049316, + "learning_rate": 1.3860677589826398e-07, + "loss": 0.6205, + "step": 3431 + }, + { + "epoch": 5.3789144050104385, + "grad_norm": 1.350320816040039, + "learning_rate": 1.3791586857671734e-07, + "loss": 0.6491, + "step": 3432 + }, + { + "epoch": 5.380480167014614, + "grad_norm": 1.2478621006011963, + "learning_rate": 1.3722663868080022e-07, + "loss": 0.6076, + "step": 3433 + }, + { + "epoch": 5.382045929018789, + "grad_norm": 1.2639954090118408, + "learning_rate": 1.365390866999669e-07, + "loss": 0.5593, + "step": 3434 + }, + { + "epoch": 5.383611691022964, + "grad_norm": 1.2016838788986206, + "learning_rate": 1.3585321312247935e-07, + "loss": 0.5601, + "step": 3435 + }, + { + "epoch": 5.38517745302714, + "grad_norm": 1.163132667541504, + "learning_rate": 1.3516901843540876e-07, + "loss": 0.6067, + "step": 3436 + }, + { + "epoch": 5.3867432150313155, + "grad_norm": 1.3532346487045288, + "learning_rate": 1.3448650312463397e-07, + "loss": 0.5975, + "step": 3437 + }, + { + "epoch": 5.388308977035491, + "grad_norm": 1.213822364807129, + "learning_rate": 1.338056676748395e-07, + "loss": 0.6136, + "step": 3438 + }, + { + "epoch": 5.389874739039666, + "grad_norm": 1.2295317649841309, + "learning_rate": 1.3312651256951997e-07, + "loss": 0.5449, + "step": 3439 + }, + { + "epoch": 5.391440501043841, + "grad_norm": 1.2767428159713745, + "learning_rate": 1.324490382909735e-07, + "loss": 0.5929, + "step": 3440 + }, + { + "epoch": 5.393006263048017, + "grad_norm": 1.360330581665039, + "learning_rate": 1.3177324532030634e-07, + "loss": 0.651, + "step": 3441 + }, + { + "epoch": 5.3945720250521925, + "grad_norm": 1.3756864070892334, + "learning_rate": 1.3109913413743124e-07, + "loss": 0.6081, + "step": 3442 + }, + { + "epoch": 5.396137787056367, + "grad_norm": 1.1370102167129517, + "learning_rate": 1.3042670522106477e-07, + "loss": 0.6117, + "step": 3443 + }, + { + "epoch": 5.397703549060543, + "grad_norm": 1.3401273488998413, + "learning_rate": 1.2975595904873073e-07, + "loss": 0.5781, + "step": 3444 + }, + { + "epoch": 5.399269311064718, + "grad_norm": 1.2364094257354736, + "learning_rate": 1.2908689609675672e-07, + "loss": 0.6045, + "step": 3445 + }, + { + "epoch": 5.400835073068894, + "grad_norm": 1.3034133911132812, + "learning_rate": 1.2841951684027548e-07, + "loss": 0.6042, + "step": 3446 + }, + { + "epoch": 5.402400835073069, + "grad_norm": 1.3812674283981323, + "learning_rate": 1.2775382175322427e-07, + "loss": 0.6254, + "step": 3447 + }, + { + "epoch": 5.403966597077244, + "grad_norm": 1.291243314743042, + "learning_rate": 1.270898113083438e-07, + "loss": 0.5226, + "step": 3448 + }, + { + "epoch": 5.40553235908142, + "grad_norm": 1.1997270584106445, + "learning_rate": 1.2642748597717856e-07, + "loss": 0.6492, + "step": 3449 + }, + { + "epoch": 5.407098121085595, + "grad_norm": 1.3700882196426392, + "learning_rate": 1.2576684623007706e-07, + "loss": 0.596, + "step": 3450 + }, + { + "epoch": 5.40866388308977, + "grad_norm": 1.2742419242858887, + "learning_rate": 1.2510789253618982e-07, + "loss": 0.6278, + "step": 3451 + }, + { + "epoch": 5.4102296450939455, + "grad_norm": 1.1523911952972412, + "learning_rate": 1.2445062536347057e-07, + "loss": 0.5814, + "step": 3452 + }, + { + "epoch": 5.411795407098121, + "grad_norm": 1.2164733409881592, + "learning_rate": 1.2379504517867542e-07, + "loss": 0.6192, + "step": 3453 + }, + { + "epoch": 5.413361169102297, + "grad_norm": 1.2368628978729248, + "learning_rate": 1.231411524473619e-07, + "loss": 0.5599, + "step": 3454 + }, + { + "epoch": 5.414926931106471, + "grad_norm": 1.2275792360305786, + "learning_rate": 1.2248894763389058e-07, + "loss": 0.6439, + "step": 3455 + }, + { + "epoch": 5.416492693110647, + "grad_norm": 1.2999566793441772, + "learning_rate": 1.2183843120142063e-07, + "loss": 0.6067, + "step": 3456 + }, + { + "epoch": 5.4180584551148225, + "grad_norm": 1.224501371383667, + "learning_rate": 1.211896036119156e-07, + "loss": 0.6795, + "step": 3457 + }, + { + "epoch": 5.419624217118998, + "grad_norm": 1.3449523448944092, + "learning_rate": 1.2054246532613777e-07, + "loss": 0.6082, + "step": 3458 + }, + { + "epoch": 5.421189979123174, + "grad_norm": 1.3984527587890625, + "learning_rate": 1.1989701680364923e-07, + "loss": 0.5778, + "step": 3459 + }, + { + "epoch": 5.422755741127348, + "grad_norm": 1.1819032430648804, + "learning_rate": 1.1925325850281416e-07, + "loss": 0.5941, + "step": 3460 + }, + { + "epoch": 5.424321503131524, + "grad_norm": 1.2696601152420044, + "learning_rate": 1.186111908807941e-07, + "loss": 0.5279, + "step": 3461 + }, + { + "epoch": 5.4258872651356995, + "grad_norm": 1.263803482055664, + "learning_rate": 1.1797081439355151e-07, + "loss": 0.6207, + "step": 3462 + }, + { + "epoch": 5.427453027139875, + "grad_norm": 1.2001559734344482, + "learning_rate": 1.1733212949584822e-07, + "loss": 0.543, + "step": 3463 + }, + { + "epoch": 5.42901878914405, + "grad_norm": 1.369549036026001, + "learning_rate": 1.1669513664124304e-07, + "loss": 0.5688, + "step": 3464 + }, + { + "epoch": 5.430584551148225, + "grad_norm": 1.46800696849823, + "learning_rate": 1.1605983628209417e-07, + "loss": 0.6093, + "step": 3465 + }, + { + "epoch": 5.432150313152401, + "grad_norm": 1.161247730255127, + "learning_rate": 1.1542622886955873e-07, + "loss": 0.5142, + "step": 3466 + }, + { + "epoch": 5.4337160751565765, + "grad_norm": 1.234609842300415, + "learning_rate": 1.1479431485358989e-07, + "loss": 0.5878, + "step": 3467 + }, + { + "epoch": 5.435281837160751, + "grad_norm": 1.268078327178955, + "learning_rate": 1.1416409468293977e-07, + "loss": 0.6159, + "step": 3468 + }, + { + "epoch": 5.436847599164927, + "grad_norm": 1.2104543447494507, + "learning_rate": 1.1353556880515648e-07, + "loss": 0.5088, + "step": 3469 + }, + { + "epoch": 5.438413361169102, + "grad_norm": 1.2559877634048462, + "learning_rate": 1.1290873766658578e-07, + "loss": 0.6159, + "step": 3470 + }, + { + "epoch": 5.439979123173278, + "grad_norm": 1.2496120929718018, + "learning_rate": 1.1228360171236936e-07, + "loss": 0.5015, + "step": 3471 + }, + { + "epoch": 5.4415448851774535, + "grad_norm": 1.1984806060791016, + "learning_rate": 1.116601613864457e-07, + "loss": 0.608, + "step": 3472 + }, + { + "epoch": 5.443110647181628, + "grad_norm": 1.2826205492019653, + "learning_rate": 1.1103841713154845e-07, + "loss": 0.621, + "step": 3473 + }, + { + "epoch": 5.444676409185804, + "grad_norm": 1.1606104373931885, + "learning_rate": 1.1041836938920664e-07, + "loss": 0.6085, + "step": 3474 + }, + { + "epoch": 5.446242171189979, + "grad_norm": 1.1810858249664307, + "learning_rate": 1.0980001859974555e-07, + "loss": 0.616, + "step": 3475 + }, + { + "epoch": 5.447807933194154, + "grad_norm": 1.3992196321487427, + "learning_rate": 1.0918336520228474e-07, + "loss": 0.5193, + "step": 3476 + }, + { + "epoch": 5.44937369519833, + "grad_norm": 1.2406203746795654, + "learning_rate": 1.0856840963473809e-07, + "loss": 0.6593, + "step": 3477 + }, + { + "epoch": 5.450939457202505, + "grad_norm": 1.2792857885360718, + "learning_rate": 1.079551523338146e-07, + "loss": 0.5844, + "step": 3478 + }, + { + "epoch": 5.452505219206681, + "grad_norm": 1.3366221189498901, + "learning_rate": 1.0734359373501618e-07, + "loss": 0.6411, + "step": 3479 + }, + { + "epoch": 5.454070981210856, + "grad_norm": 1.3238123655319214, + "learning_rate": 1.0673373427263961e-07, + "loss": 0.6046, + "step": 3480 + }, + { + "epoch": 5.455636743215031, + "grad_norm": 1.3767836093902588, + "learning_rate": 1.0612557437977372e-07, + "loss": 0.564, + "step": 3481 + }, + { + "epoch": 5.457202505219207, + "grad_norm": 1.3002339601516724, + "learning_rate": 1.0551911448830165e-07, + "loss": 0.6272, + "step": 3482 + }, + { + "epoch": 5.458768267223382, + "grad_norm": 1.3575612306594849, + "learning_rate": 1.0491435502889836e-07, + "loss": 0.5842, + "step": 3483 + }, + { + "epoch": 5.460334029227558, + "grad_norm": 1.1978793144226074, + "learning_rate": 1.0431129643103193e-07, + "loss": 0.5535, + "step": 3484 + }, + { + "epoch": 5.461899791231732, + "grad_norm": 1.1925379037857056, + "learning_rate": 1.0370993912296206e-07, + "loss": 0.6265, + "step": 3485 + }, + { + "epoch": 5.463465553235908, + "grad_norm": 1.262814998626709, + "learning_rate": 1.0311028353174102e-07, + "loss": 0.5894, + "step": 3486 + }, + { + "epoch": 5.465031315240084, + "grad_norm": 1.3055074214935303, + "learning_rate": 1.0251233008321093e-07, + "loss": 0.6333, + "step": 3487 + }, + { + "epoch": 5.466597077244259, + "grad_norm": 1.3066006898880005, + "learning_rate": 1.0191607920200741e-07, + "loss": 0.5586, + "step": 3488 + }, + { + "epoch": 5.468162839248434, + "grad_norm": 1.2938679456710815, + "learning_rate": 1.0132153131155592e-07, + "loss": 0.6103, + "step": 3489 + }, + { + "epoch": 5.469728601252609, + "grad_norm": 1.246503233909607, + "learning_rate": 1.0072868683407123e-07, + "loss": 0.6242, + "step": 3490 + }, + { + "epoch": 5.471294363256785, + "grad_norm": 1.2741035223007202, + "learning_rate": 1.0013754619056154e-07, + "loss": 0.6777, + "step": 3491 + }, + { + "epoch": 5.4728601252609606, + "grad_norm": 1.244509220123291, + "learning_rate": 9.954810980082191e-08, + "loss": 0.623, + "step": 3492 + }, + { + "epoch": 5.474425887265136, + "grad_norm": 1.3073979616165161, + "learning_rate": 9.896037808343861e-08, + "loss": 0.622, + "step": 3493 + }, + { + "epoch": 5.475991649269311, + "grad_norm": 1.2299357652664185, + "learning_rate": 9.837435145578805e-08, + "loss": 0.5826, + "step": 3494 + }, + { + "epoch": 5.477557411273486, + "grad_norm": 1.2250887155532837, + "learning_rate": 9.779003033403428e-08, + "loss": 0.6827, + "step": 3495 + }, + { + "epoch": 5.479123173277662, + "grad_norm": 1.3190733194351196, + "learning_rate": 9.720741513313092e-08, + "loss": 0.6338, + "step": 3496 + }, + { + "epoch": 5.4806889352818375, + "grad_norm": 1.2337404489517212, + "learning_rate": 9.662650626682036e-08, + "loss": 0.5672, + "step": 3497 + }, + { + "epoch": 5.482254697286012, + "grad_norm": 1.1437941789627075, + "learning_rate": 9.604730414763286e-08, + "loss": 0.6203, + "step": 3498 + }, + { + "epoch": 5.483820459290188, + "grad_norm": 1.1348474025726318, + "learning_rate": 9.546980918688636e-08, + "loss": 0.5708, + "step": 3499 + }, + { + "epoch": 5.485386221294363, + "grad_norm": 1.2211592197418213, + "learning_rate": 9.489402179468754e-08, + "loss": 0.5924, + "step": 3500 + }, + { + "epoch": 5.486951983298539, + "grad_norm": 1.250848650932312, + "learning_rate": 9.431994237992931e-08, + "loss": 0.6249, + "step": 3501 + }, + { + "epoch": 5.488517745302714, + "grad_norm": 1.2227661609649658, + "learning_rate": 9.374757135029195e-08, + "loss": 0.5943, + "step": 3502 + }, + { + "epoch": 5.490083507306889, + "grad_norm": 1.3056727647781372, + "learning_rate": 9.317690911224309e-08, + "loss": 0.5965, + "step": 3503 + }, + { + "epoch": 5.491649269311065, + "grad_norm": 1.2243983745574951, + "learning_rate": 9.260795607103634e-08, + "loss": 0.6399, + "step": 3504 + }, + { + "epoch": 5.49321503131524, + "grad_norm": 1.2149602174758911, + "learning_rate": 9.204071263071185e-08, + "loss": 0.5343, + "step": 3505 + }, + { + "epoch": 5.494780793319415, + "grad_norm": 1.360884189605713, + "learning_rate": 9.147517919409516e-08, + "loss": 0.6298, + "step": 3506 + }, + { + "epoch": 5.496346555323591, + "grad_norm": 1.267560362815857, + "learning_rate": 9.091135616279861e-08, + "loss": 0.6095, + "step": 3507 + }, + { + "epoch": 5.497912317327766, + "grad_norm": 1.2687102556228638, + "learning_rate": 9.034924393721778e-08, + "loss": 0.6232, + "step": 3508 + }, + { + "epoch": 5.499478079331942, + "grad_norm": 1.3354545831680298, + "learning_rate": 8.978884291653555e-08, + "loss": 0.6037, + "step": 3509 + }, + { + "epoch": 5.5010438413361165, + "grad_norm": 1.28574800491333, + "learning_rate": 8.923015349871917e-08, + "loss": 0.6113, + "step": 3510 + }, + { + "epoch": 5.502609603340292, + "grad_norm": 1.2026561498641968, + "learning_rate": 8.867317608051845e-08, + "loss": 0.6035, + "step": 3511 + }, + { + "epoch": 5.504175365344468, + "grad_norm": 1.300851583480835, + "learning_rate": 8.811791105747065e-08, + "loss": 0.6382, + "step": 3512 + }, + { + "epoch": 5.505741127348643, + "grad_norm": 1.2547096014022827, + "learning_rate": 8.75643588238942e-08, + "loss": 0.6162, + "step": 3513 + }, + { + "epoch": 5.507306889352819, + "grad_norm": 1.314209222793579, + "learning_rate": 8.701251977289188e-08, + "loss": 0.5838, + "step": 3514 + }, + { + "epoch": 5.508872651356993, + "grad_norm": 1.3353283405303955, + "learning_rate": 8.646239429635161e-08, + "loss": 0.5619, + "step": 3515 + }, + { + "epoch": 5.510438413361169, + "grad_norm": 1.2824180126190186, + "learning_rate": 8.5913982784942e-08, + "loss": 0.6169, + "step": 3516 + }, + { + "epoch": 5.512004175365345, + "grad_norm": 1.187626600265503, + "learning_rate": 8.536728562811542e-08, + "loss": 0.5788, + "step": 3517 + }, + { + "epoch": 5.51356993736952, + "grad_norm": 1.2610911130905151, + "learning_rate": 8.482230321410767e-08, + "loss": 0.633, + "step": 3518 + }, + { + "epoch": 5.515135699373695, + "grad_norm": 1.3033767938613892, + "learning_rate": 8.427903592993531e-08, + "loss": 0.5973, + "step": 3519 + }, + { + "epoch": 5.51670146137787, + "grad_norm": 1.2476587295532227, + "learning_rate": 8.373748416139832e-08, + "loss": 0.6117, + "step": 3520 + }, + { + "epoch": 5.518267223382046, + "grad_norm": 1.1832818984985352, + "learning_rate": 8.319764829307714e-08, + "loss": 0.602, + "step": 3521 + }, + { + "epoch": 5.519832985386222, + "grad_norm": 1.322898030281067, + "learning_rate": 8.265952870833482e-08, + "loss": 0.61, + "step": 3522 + }, + { + "epoch": 5.521398747390396, + "grad_norm": 1.3117808103561401, + "learning_rate": 8.212312578931486e-08, + "loss": 0.6212, + "step": 3523 + }, + { + "epoch": 5.522964509394572, + "grad_norm": 1.2084193229675293, + "learning_rate": 8.15884399169417e-08, + "loss": 0.5849, + "step": 3524 + }, + { + "epoch": 5.524530271398747, + "grad_norm": 1.1629563570022583, + "learning_rate": 8.105547147092108e-08, + "loss": 0.644, + "step": 3525 + }, + { + "epoch": 5.526096033402923, + "grad_norm": 1.316291093826294, + "learning_rate": 8.052422082973826e-08, + "loss": 0.6115, + "step": 3526 + }, + { + "epoch": 5.527661795407099, + "grad_norm": 1.249566912651062, + "learning_rate": 7.999468837065871e-08, + "loss": 0.5564, + "step": 3527 + }, + { + "epoch": 5.529227557411273, + "grad_norm": 1.2442042827606201, + "learning_rate": 7.946687446972884e-08, + "loss": 0.5902, + "step": 3528 + }, + { + "epoch": 5.530793319415449, + "grad_norm": 1.199759840965271, + "learning_rate": 7.894077950177299e-08, + "loss": 0.563, + "step": 3529 + }, + { + "epoch": 5.532359081419624, + "grad_norm": 1.2485020160675049, + "learning_rate": 7.841640384039618e-08, + "loss": 0.616, + "step": 3530 + }, + { + "epoch": 5.533924843423799, + "grad_norm": 1.3423585891723633, + "learning_rate": 7.789374785798165e-08, + "loss": 0.6182, + "step": 3531 + }, + { + "epoch": 5.535490605427975, + "grad_norm": 1.2344738245010376, + "learning_rate": 7.737281192569169e-08, + "loss": 0.5452, + "step": 3532 + }, + { + "epoch": 5.53705636743215, + "grad_norm": 1.311337947845459, + "learning_rate": 7.685359641346729e-08, + "loss": 0.6066, + "step": 3533 + }, + { + "epoch": 5.538622129436326, + "grad_norm": 1.194998860359192, + "learning_rate": 7.633610169002714e-08, + "loss": 0.6015, + "step": 3534 + }, + { + "epoch": 5.540187891440501, + "grad_norm": 1.2133246660232544, + "learning_rate": 7.582032812286865e-08, + "loss": 0.5721, + "step": 3535 + }, + { + "epoch": 5.541753653444676, + "grad_norm": 1.2820360660552979, + "learning_rate": 7.530627607826636e-08, + "loss": 0.5868, + "step": 3536 + }, + { + "epoch": 5.543319415448852, + "grad_norm": 1.3318873643875122, + "learning_rate": 7.479394592127265e-08, + "loss": 0.6209, + "step": 3537 + }, + { + "epoch": 5.544885177453027, + "grad_norm": 1.315002679824829, + "learning_rate": 7.428333801571736e-08, + "loss": 0.5449, + "step": 3538 + }, + { + "epoch": 5.546450939457203, + "grad_norm": 1.1738154888153076, + "learning_rate": 7.377445272420541e-08, + "loss": 0.5907, + "step": 3539 + }, + { + "epoch": 5.5480167014613775, + "grad_norm": 1.3171212673187256, + "learning_rate": 7.326729040812136e-08, + "loss": 0.6319, + "step": 3540 + }, + { + "epoch": 5.549582463465553, + "grad_norm": 1.5304313898086548, + "learning_rate": 7.27618514276246e-08, + "loss": 0.6066, + "step": 3541 + }, + { + "epoch": 5.551148225469729, + "grad_norm": 1.158853530883789, + "learning_rate": 7.225813614164967e-08, + "loss": 0.6008, + "step": 3542 + }, + { + "epoch": 5.552713987473904, + "grad_norm": 1.2796374559402466, + "learning_rate": 7.17561449079096e-08, + "loss": 0.6436, + "step": 3543 + }, + { + "epoch": 5.554279749478079, + "grad_norm": 1.2964136600494385, + "learning_rate": 7.125587808289064e-08, + "loss": 0.5942, + "step": 3544 + }, + { + "epoch": 5.5558455114822545, + "grad_norm": 1.2614834308624268, + "learning_rate": 7.075733602185581e-08, + "loss": 0.6811, + "step": 3545 + }, + { + "epoch": 5.55741127348643, + "grad_norm": 1.303294062614441, + "learning_rate": 7.026051907884357e-08, + "loss": 0.6176, + "step": 3546 + }, + { + "epoch": 5.558977035490606, + "grad_norm": 1.2901746034622192, + "learning_rate": 6.976542760666561e-08, + "loss": 0.5787, + "step": 3547 + }, + { + "epoch": 5.560542797494781, + "grad_norm": 1.3341819047927856, + "learning_rate": 6.927206195691039e-08, + "loss": 0.5674, + "step": 3548 + }, + { + "epoch": 5.562108559498956, + "grad_norm": 1.1802226305007935, + "learning_rate": 6.878042247993905e-08, + "loss": 0.6144, + "step": 3549 + }, + { + "epoch": 5.5636743215031315, + "grad_norm": 1.3665695190429688, + "learning_rate": 6.829050952488786e-08, + "loss": 0.5759, + "step": 3550 + }, + { + "epoch": 5.565240083507307, + "grad_norm": 1.1364253759384155, + "learning_rate": 6.780232343966686e-08, + "loss": 0.593, + "step": 3551 + }, + { + "epoch": 5.566805845511483, + "grad_norm": 1.303017258644104, + "learning_rate": 6.731586457095984e-08, + "loss": 0.5766, + "step": 3552 + }, + { + "epoch": 5.568371607515657, + "grad_norm": 1.075515627861023, + "learning_rate": 6.683113326422325e-08, + "loss": 0.5717, + "step": 3553 + }, + { + "epoch": 5.569937369519833, + "grad_norm": 1.3725066184997559, + "learning_rate": 6.634812986368782e-08, + "loss": 0.6188, + "step": 3554 + }, + { + "epoch": 5.5715031315240084, + "grad_norm": 1.332984209060669, + "learning_rate": 6.586685471235615e-08, + "loss": 0.6241, + "step": 3555 + }, + { + "epoch": 5.573068893528184, + "grad_norm": 1.2360045909881592, + "learning_rate": 6.538730815200483e-08, + "loss": 0.6322, + "step": 3556 + }, + { + "epoch": 5.574634655532359, + "grad_norm": 1.2404091358184814, + "learning_rate": 6.490949052318174e-08, + "loss": 0.6253, + "step": 3557 + }, + { + "epoch": 5.576200417536534, + "grad_norm": 1.2391175031661987, + "learning_rate": 6.44334021652071e-08, + "loss": 0.588, + "step": 3558 + }, + { + "epoch": 5.57776617954071, + "grad_norm": 1.334181785583496, + "learning_rate": 6.39590434161741e-08, + "loss": 0.6009, + "step": 3559 + }, + { + "epoch": 5.579331941544885, + "grad_norm": 1.1880091428756714, + "learning_rate": 6.348641461294574e-08, + "loss": 0.5814, + "step": 3560 + }, + { + "epoch": 5.580897703549061, + "grad_norm": 1.2914915084838867, + "learning_rate": 6.301551609115885e-08, + "loss": 0.5824, + "step": 3561 + }, + { + "epoch": 5.582463465553236, + "grad_norm": 1.3836625814437866, + "learning_rate": 6.254634818521982e-08, + "loss": 0.595, + "step": 3562 + }, + { + "epoch": 5.584029227557411, + "grad_norm": 1.2656164169311523, + "learning_rate": 6.207891122830629e-08, + "loss": 0.5796, + "step": 3563 + }, + { + "epoch": 5.585594989561587, + "grad_norm": 1.3809819221496582, + "learning_rate": 6.1613205552368e-08, + "loss": 0.5903, + "step": 3564 + }, + { + "epoch": 5.5871607515657615, + "grad_norm": 1.1625736951828003, + "learning_rate": 6.114923148812291e-08, + "loss": 0.5964, + "step": 3565 + }, + { + "epoch": 5.588726513569937, + "grad_norm": 1.3540935516357422, + "learning_rate": 6.068698936506079e-08, + "loss": 0.5654, + "step": 3566 + }, + { + "epoch": 5.590292275574113, + "grad_norm": 1.1908572912216187, + "learning_rate": 6.022647951144239e-08, + "loss": 0.5679, + "step": 3567 + }, + { + "epoch": 5.591858037578288, + "grad_norm": 1.363163948059082, + "learning_rate": 5.976770225429557e-08, + "loss": 0.5762, + "step": 3568 + }, + { + "epoch": 5.593423799582464, + "grad_norm": 1.3758068084716797, + "learning_rate": 5.9310657919420536e-08, + "loss": 0.5644, + "step": 3569 + }, + { + "epoch": 5.5949895615866385, + "grad_norm": 1.2351075410842896, + "learning_rate": 5.8855346831384885e-08, + "loss": 0.6617, + "step": 3570 + }, + { + "epoch": 5.596555323590814, + "grad_norm": 1.2445124387741089, + "learning_rate": 5.840176931352664e-08, + "loss": 0.5523, + "step": 3571 + }, + { + "epoch": 5.59812108559499, + "grad_norm": 1.26252019405365, + "learning_rate": 5.79499256879526e-08, + "loss": 0.5441, + "step": 3572 + }, + { + "epoch": 5.599686847599165, + "grad_norm": 1.2237505912780762, + "learning_rate": 5.7499816275536634e-08, + "loss": 0.5844, + "step": 3573 + }, + { + "epoch": 5.60125260960334, + "grad_norm": 1.1804537773132324, + "learning_rate": 5.705144139592361e-08, + "loss": 0.6157, + "step": 3574 + }, + { + "epoch": 5.6028183716075155, + "grad_norm": 1.3245272636413574, + "learning_rate": 5.6604801367524945e-08, + "loss": 0.6149, + "step": 3575 + }, + { + "epoch": 5.604384133611691, + "grad_norm": 1.3412095308303833, + "learning_rate": 5.615989650752024e-08, + "loss": 0.649, + "step": 3576 + }, + { + "epoch": 5.605949895615867, + "grad_norm": 1.3259727954864502, + "learning_rate": 5.571672713185733e-08, + "loss": 0.6055, + "step": 3577 + }, + { + "epoch": 5.607515657620041, + "grad_norm": 1.196613073348999, + "learning_rate": 5.5275293555251406e-08, + "loss": 0.5687, + "step": 3578 + }, + { + "epoch": 5.609081419624217, + "grad_norm": 1.3341057300567627, + "learning_rate": 5.483559609118449e-08, + "loss": 0.6077, + "step": 3579 + }, + { + "epoch": 5.6106471816283925, + "grad_norm": 1.3463157415390015, + "learning_rate": 5.4397635051907093e-08, + "loss": 0.6132, + "step": 3580 + }, + { + "epoch": 5.612212943632568, + "grad_norm": 1.3728768825531006, + "learning_rate": 5.396141074843458e-08, + "loss": 0.6001, + "step": 3581 + }, + { + "epoch": 5.613778705636744, + "grad_norm": 1.3943250179290771, + "learning_rate": 5.352692349055083e-08, + "loss": 0.5692, + "step": 3582 + }, + { + "epoch": 5.615344467640918, + "grad_norm": 1.3521478176116943, + "learning_rate": 5.309417358680541e-08, + "loss": 0.5243, + "step": 3583 + }, + { + "epoch": 5.616910229645094, + "grad_norm": 1.1756808757781982, + "learning_rate": 5.26631613445136e-08, + "loss": 0.5871, + "step": 3584 + }, + { + "epoch": 5.6184759916492695, + "grad_norm": 1.424647331237793, + "learning_rate": 5.223388706975779e-08, + "loss": 0.6115, + "step": 3585 + }, + { + "epoch": 5.620041753653445, + "grad_norm": 1.2350553274154663, + "learning_rate": 5.180635106738496e-08, + "loss": 0.6281, + "step": 3586 + }, + { + "epoch": 5.62160751565762, + "grad_norm": 1.2100744247436523, + "learning_rate": 5.1380553641008913e-08, + "loss": 0.6549, + "step": 3587 + }, + { + "epoch": 5.623173277661795, + "grad_norm": 1.2485849857330322, + "learning_rate": 5.095649509300804e-08, + "loss": 0.5817, + "step": 3588 + }, + { + "epoch": 5.624739039665971, + "grad_norm": 1.223224401473999, + "learning_rate": 5.0534175724525626e-08, + "loss": 0.5605, + "step": 3589 + }, + { + "epoch": 5.6263048016701465, + "grad_norm": 1.1859829425811768, + "learning_rate": 5.011359583547065e-08, + "loss": 0.6304, + "step": 3590 + }, + { + "epoch": 5.627870563674321, + "grad_norm": 1.2232156991958618, + "learning_rate": 4.969475572451615e-08, + "loss": 0.592, + "step": 3591 + }, + { + "epoch": 5.629436325678497, + "grad_norm": 1.276767373085022, + "learning_rate": 4.9277655689100033e-08, + "loss": 0.5121, + "step": 3592 + }, + { + "epoch": 5.631002087682672, + "grad_norm": 1.2463449239730835, + "learning_rate": 4.8862296025424805e-08, + "loss": 0.5715, + "step": 3593 + }, + { + "epoch": 5.632567849686848, + "grad_norm": 1.1216026544570923, + "learning_rate": 4.8448677028455905e-08, + "loss": 0.6393, + "step": 3594 + }, + { + "epoch": 5.6341336116910234, + "grad_norm": 1.346016764640808, + "learning_rate": 4.8036798991923925e-08, + "loss": 0.5423, + "step": 3595 + }, + { + "epoch": 5.635699373695198, + "grad_norm": 1.2266265153884888, + "learning_rate": 4.7626662208322405e-08, + "loss": 0.5975, + "step": 3596 + }, + { + "epoch": 5.637265135699374, + "grad_norm": 1.241432785987854, + "learning_rate": 4.7218266968908354e-08, + "loss": 0.5893, + "step": 3597 + }, + { + "epoch": 5.638830897703549, + "grad_norm": 1.354709506034851, + "learning_rate": 4.6811613563702843e-08, + "loss": 0.5263, + "step": 3598 + }, + { + "epoch": 5.640396659707724, + "grad_norm": 1.177618145942688, + "learning_rate": 4.6406702281488746e-08, + "loss": 0.579, + "step": 3599 + }, + { + "epoch": 5.6419624217118995, + "grad_norm": 1.3579180240631104, + "learning_rate": 4.6003533409812175e-08, + "loss": 0.6922, + "step": 3600 + }, + { + "epoch": 5.643528183716075, + "grad_norm": 1.1778520345687866, + "learning_rate": 4.560210723498271e-08, + "loss": 0.6131, + "step": 3601 + }, + { + "epoch": 5.645093945720251, + "grad_norm": 1.1965405941009521, + "learning_rate": 4.52024240420712e-08, + "loss": 0.6065, + "step": 3602 + }, + { + "epoch": 5.646659707724426, + "grad_norm": 1.3655333518981934, + "learning_rate": 4.480448411491173e-08, + "loss": 0.6123, + "step": 3603 + }, + { + "epoch": 5.648225469728601, + "grad_norm": 1.301324725151062, + "learning_rate": 4.4408287736099344e-08, + "loss": 0.5914, + "step": 3604 + }, + { + "epoch": 5.6497912317327765, + "grad_norm": 1.2587555646896362, + "learning_rate": 4.4013835186992036e-08, + "loss": 0.6054, + "step": 3605 + }, + { + "epoch": 5.651356993736952, + "grad_norm": 1.2624011039733887, + "learning_rate": 4.362112674770824e-08, + "loss": 0.5834, + "step": 3606 + }, + { + "epoch": 5.652922755741128, + "grad_norm": 1.2550424337387085, + "learning_rate": 4.323016269712932e-08, + "loss": 0.5747, + "step": 3607 + }, + { + "epoch": 5.654488517745302, + "grad_norm": 1.2000874280929565, + "learning_rate": 4.284094331289623e-08, + "loss": 0.6272, + "step": 3608 + }, + { + "epoch": 5.656054279749478, + "grad_norm": 1.308048963546753, + "learning_rate": 4.245346887141205e-08, + "loss": 0.5124, + "step": 3609 + }, + { + "epoch": 5.6576200417536535, + "grad_norm": 1.385587453842163, + "learning_rate": 4.2067739647840275e-08, + "loss": 0.6139, + "step": 3610 + }, + { + "epoch": 5.659185803757829, + "grad_norm": 1.1399497985839844, + "learning_rate": 4.168375591610568e-08, + "loss": 0.5989, + "step": 3611 + }, + { + "epoch": 5.660751565762004, + "grad_norm": 1.2604403495788574, + "learning_rate": 4.130151794889181e-08, + "loss": 0.6327, + "step": 3612 + }, + { + "epoch": 5.662317327766179, + "grad_norm": 1.2142161130905151, + "learning_rate": 4.0921026017644584e-08, + "loss": 0.6093, + "step": 3613 + }, + { + "epoch": 5.663883089770355, + "grad_norm": 1.2195683717727661, + "learning_rate": 4.0542280392568963e-08, + "loss": 0.5434, + "step": 3614 + }, + { + "epoch": 5.6654488517745305, + "grad_norm": 1.4029487371444702, + "learning_rate": 4.016528134262898e-08, + "loss": 0.4949, + "step": 3615 + }, + { + "epoch": 5.667014613778706, + "grad_norm": 1.0893887281417847, + "learning_rate": 3.979002913555019e-08, + "loss": 0.5672, + "step": 3616 + }, + { + "epoch": 5.668580375782881, + "grad_norm": 1.259775996208191, + "learning_rate": 3.9416524037815814e-08, + "loss": 0.6246, + "step": 3617 + }, + { + "epoch": 5.670146137787056, + "grad_norm": 1.2941632270812988, + "learning_rate": 3.904476631466925e-08, + "loss": 0.5921, + "step": 3618 + }, + { + "epoch": 5.671711899791232, + "grad_norm": 1.3075370788574219, + "learning_rate": 3.867475623011346e-08, + "loss": 0.5677, + "step": 3619 + }, + { + "epoch": 5.673277661795407, + "grad_norm": 1.214609980583191, + "learning_rate": 3.830649404690939e-08, + "loss": 0.5747, + "step": 3620 + }, + { + "epoch": 5.674843423799582, + "grad_norm": 1.4172025918960571, + "learning_rate": 3.793998002657728e-08, + "loss": 0.5631, + "step": 3621 + }, + { + "epoch": 5.676409185803758, + "grad_norm": 1.2903814315795898, + "learning_rate": 3.75752144293956e-08, + "loss": 0.6276, + "step": 3622 + }, + { + "epoch": 5.677974947807933, + "grad_norm": 1.1860072612762451, + "learning_rate": 3.7212197514401305e-08, + "loss": 0.5379, + "step": 3623 + }, + { + "epoch": 5.679540709812109, + "grad_norm": 1.314446210861206, + "learning_rate": 3.6850929539389844e-08, + "loss": 0.6343, + "step": 3624 + }, + { + "epoch": 5.681106471816284, + "grad_norm": 1.229719877243042, + "learning_rate": 3.649141076091406e-08, + "loss": 0.6024, + "step": 3625 + }, + { + "epoch": 5.682672233820459, + "grad_norm": 1.1693522930145264, + "learning_rate": 3.6133641434284995e-08, + "loss": 0.6114, + "step": 3626 + }, + { + "epoch": 5.684237995824635, + "grad_norm": 1.2493668794631958, + "learning_rate": 3.577762181357164e-08, + "loss": 0.5479, + "step": 3627 + }, + { + "epoch": 5.68580375782881, + "grad_norm": 1.3555315732955933, + "learning_rate": 3.5423352151599534e-08, + "loss": 0.5869, + "step": 3628 + }, + { + "epoch": 5.687369519832985, + "grad_norm": 1.2836302518844604, + "learning_rate": 3.507083269995243e-08, + "loss": 0.6059, + "step": 3629 + }, + { + "epoch": 5.688935281837161, + "grad_norm": 1.2259082794189453, + "learning_rate": 3.4720063708970355e-08, + "loss": 0.6411, + "step": 3630 + }, + { + "epoch": 5.690501043841336, + "grad_norm": 1.3272029161453247, + "learning_rate": 3.437104542775099e-08, + "loss": 0.678, + "step": 3631 + }, + { + "epoch": 5.692066805845512, + "grad_norm": 1.1469253301620483, + "learning_rate": 3.402377810414831e-08, + "loss": 0.4935, + "step": 3632 + }, + { + "epoch": 5.693632567849686, + "grad_norm": 1.2290380001068115, + "learning_rate": 3.3678261984773084e-08, + "loss": 0.5988, + "step": 3633 + }, + { + "epoch": 5.695198329853862, + "grad_norm": 1.3161014318466187, + "learning_rate": 3.3334497314992107e-08, + "loss": 0.5523, + "step": 3634 + }, + { + "epoch": 5.696764091858038, + "grad_norm": 1.2987585067749023, + "learning_rate": 3.299248433892871e-08, + "loss": 0.6287, + "step": 3635 + }, + { + "epoch": 5.698329853862213, + "grad_norm": 1.1819089651107788, + "learning_rate": 3.2652223299462214e-08, + "loss": 0.5626, + "step": 3636 + }, + { + "epoch": 5.699895615866389, + "grad_norm": 1.1800951957702637, + "learning_rate": 3.231371443822795e-08, + "loss": 0.6143, + "step": 3637 + }, + { + "epoch": 5.701461377870563, + "grad_norm": 1.30130934715271, + "learning_rate": 3.197695799561667e-08, + "loss": 0.6191, + "step": 3638 + }, + { + "epoch": 5.703027139874739, + "grad_norm": 1.2036466598510742, + "learning_rate": 3.164195421077459e-08, + "loss": 0.5666, + "step": 3639 + }, + { + "epoch": 5.7045929018789145, + "grad_norm": 1.2085832357406616, + "learning_rate": 3.130870332160363e-08, + "loss": 0.6359, + "step": 3640 + }, + { + "epoch": 5.70615866388309, + "grad_norm": 1.220350980758667, + "learning_rate": 3.097720556476086e-08, + "loss": 0.597, + "step": 3641 + }, + { + "epoch": 5.707724425887265, + "grad_norm": 1.2355475425720215, + "learning_rate": 3.064746117565798e-08, + "loss": 0.6487, + "step": 3642 + }, + { + "epoch": 5.70929018789144, + "grad_norm": 1.224936842918396, + "learning_rate": 3.0319470388461836e-08, + "loss": 0.5994, + "step": 3643 + }, + { + "epoch": 5.710855949895616, + "grad_norm": 1.2243340015411377, + "learning_rate": 2.9993233436093895e-08, + "loss": 0.5779, + "step": 3644 + }, + { + "epoch": 5.7124217118997915, + "grad_norm": 1.4743517637252808, + "learning_rate": 2.9668750550230497e-08, + "loss": 0.5459, + "step": 3645 + }, + { + "epoch": 5.713987473903966, + "grad_norm": 1.2513774633407593, + "learning_rate": 2.9346021961301206e-08, + "loss": 0.6223, + "step": 3646 + }, + { + "epoch": 5.715553235908142, + "grad_norm": 1.2895479202270508, + "learning_rate": 2.9025047898491587e-08, + "loss": 0.6008, + "step": 3647 + }, + { + "epoch": 5.717118997912317, + "grad_norm": 1.281267762184143, + "learning_rate": 2.8705828589739583e-08, + "loss": 0.5816, + "step": 3648 + }, + { + "epoch": 5.718684759916493, + "grad_norm": 1.1376088857650757, + "learning_rate": 2.8388364261737743e-08, + "loss": 0.6479, + "step": 3649 + }, + { + "epoch": 5.7202505219206685, + "grad_norm": 1.2615599632263184, + "learning_rate": 2.807265513993268e-08, + "loss": 0.581, + "step": 3650 + }, + { + "epoch": 5.721816283924843, + "grad_norm": 1.4371247291564941, + "learning_rate": 2.7758701448523386e-08, + "loss": 0.5355, + "step": 3651 + }, + { + "epoch": 5.723382045929019, + "grad_norm": 1.2366218566894531, + "learning_rate": 2.7446503410463178e-08, + "loss": 0.6099, + "step": 3652 + }, + { + "epoch": 5.724947807933194, + "grad_norm": 1.4122114181518555, + "learning_rate": 2.7136061247458877e-08, + "loss": 0.5788, + "step": 3653 + }, + { + "epoch": 5.726513569937369, + "grad_norm": 1.2621092796325684, + "learning_rate": 2.682737517996914e-08, + "loss": 0.5571, + "step": 3654 + }, + { + "epoch": 5.728079331941545, + "grad_norm": 1.2963430881500244, + "learning_rate": 2.652044542720694e-08, + "loss": 0.5646, + "step": 3655 + }, + { + "epoch": 5.72964509394572, + "grad_norm": 1.24502694606781, + "learning_rate": 2.6215272207136823e-08, + "loss": 0.6011, + "step": 3656 + }, + { + "epoch": 5.731210855949896, + "grad_norm": 1.2528014183044434, + "learning_rate": 2.5911855736476808e-08, + "loss": 0.5667, + "step": 3657 + }, + { + "epoch": 5.732776617954071, + "grad_norm": 1.1972178220748901, + "learning_rate": 2.561019623069677e-08, + "loss": 0.488, + "step": 3658 + }, + { + "epoch": 5.734342379958246, + "grad_norm": 1.2752708196640015, + "learning_rate": 2.531029390401951e-08, + "loss": 0.5932, + "step": 3659 + }, + { + "epoch": 5.735908141962422, + "grad_norm": 1.3707789182662964, + "learning_rate": 2.5012148969419113e-08, + "loss": 0.6097, + "step": 3660 + }, + { + "epoch": 5.737473903966597, + "grad_norm": 1.1388009786605835, + "learning_rate": 2.4715761638622328e-08, + "loss": 0.6094, + "step": 3661 + }, + { + "epoch": 5.739039665970773, + "grad_norm": 1.3091226816177368, + "learning_rate": 2.4421132122107727e-08, + "loss": 0.5649, + "step": 3662 + }, + { + "epoch": 5.740605427974947, + "grad_norm": 1.2672806978225708, + "learning_rate": 2.4128260629105448e-08, + "loss": 0.5438, + "step": 3663 + }, + { + "epoch": 5.742171189979123, + "grad_norm": 1.2817412614822388, + "learning_rate": 2.3837147367596623e-08, + "loss": 0.6496, + "step": 3664 + }, + { + "epoch": 5.743736951983299, + "grad_norm": 1.1511565446853638, + "learning_rate": 2.3547792544315052e-08, + "loss": 0.5159, + "step": 3665 + }, + { + "epoch": 5.745302713987474, + "grad_norm": 1.2575610876083374, + "learning_rate": 2.3260196364744415e-08, + "loss": 0.6478, + "step": 3666 + }, + { + "epoch": 5.746868475991649, + "grad_norm": 1.2787147760391235, + "learning_rate": 2.2974359033120507e-08, + "loss": 0.6039, + "step": 3667 + }, + { + "epoch": 5.748434237995824, + "grad_norm": 1.2893658876419067, + "learning_rate": 2.2690280752429293e-08, + "loss": 0.5978, + "step": 3668 + }, + { + "epoch": 5.75, + "grad_norm": 1.3511700630187988, + "learning_rate": 2.240796172440829e-08, + "loss": 0.6243, + "step": 3669 + }, + { + "epoch": 5.751565762004176, + "grad_norm": 1.2458500862121582, + "learning_rate": 2.2127402149545175e-08, + "loss": 0.5435, + "step": 3670 + }, + { + "epoch": 5.753131524008351, + "grad_norm": 1.1775976419448853, + "learning_rate": 2.184860222707863e-08, + "loss": 0.6445, + "step": 3671 + }, + { + "epoch": 5.754697286012526, + "grad_norm": 1.1638679504394531, + "learning_rate": 2.157156215499695e-08, + "loss": 0.6052, + "step": 3672 + }, + { + "epoch": 5.756263048016701, + "grad_norm": 1.1701326370239258, + "learning_rate": 2.1296282130039424e-08, + "loss": 0.5373, + "step": 3673 + }, + { + "epoch": 5.757828810020877, + "grad_norm": 1.3792227506637573, + "learning_rate": 2.1022762347695235e-08, + "loss": 0.572, + "step": 3674 + }, + { + "epoch": 5.759394572025053, + "grad_norm": 1.2260279655456543, + "learning_rate": 2.0751003002203175e-08, + "loss": 0.5232, + "step": 3675 + }, + { + "epoch": 5.760960334029227, + "grad_norm": 1.400583267211914, + "learning_rate": 2.0481004286552753e-08, + "loss": 0.6289, + "step": 3676 + }, + { + "epoch": 5.762526096033403, + "grad_norm": 1.4171677827835083, + "learning_rate": 2.0212766392481986e-08, + "loss": 0.6057, + "step": 3677 + }, + { + "epoch": 5.764091858037578, + "grad_norm": 1.2289799451828003, + "learning_rate": 1.9946289510479333e-08, + "loss": 0.6556, + "step": 3678 + }, + { + "epoch": 5.765657620041754, + "grad_norm": 1.3580001592636108, + "learning_rate": 1.9681573829782585e-08, + "loss": 0.624, + "step": 3679 + }, + { + "epoch": 5.767223382045929, + "grad_norm": 1.279936671257019, + "learning_rate": 1.9418619538378037e-08, + "loss": 0.5537, + "step": 3680 + }, + { + "epoch": 5.768789144050104, + "grad_norm": 1.3206218481063843, + "learning_rate": 1.91574268230027e-08, + "loss": 0.5577, + "step": 3681 + }, + { + "epoch": 5.77035490605428, + "grad_norm": 1.2767149209976196, + "learning_rate": 1.8897995869140706e-08, + "loss": 0.5936, + "step": 3682 + }, + { + "epoch": 5.771920668058455, + "grad_norm": 1.2772310972213745, + "learning_rate": 1.8640326861026347e-08, + "loss": 0.594, + "step": 3683 + }, + { + "epoch": 5.773486430062631, + "grad_norm": 1.1934822797775269, + "learning_rate": 1.8384419981642698e-08, + "loss": 0.6199, + "step": 3684 + }, + { + "epoch": 5.775052192066806, + "grad_norm": 1.19818913936615, + "learning_rate": 1.8130275412720777e-08, + "loss": 0.619, + "step": 3685 + }, + { + "epoch": 5.776617954070981, + "grad_norm": 1.3573529720306396, + "learning_rate": 1.7877893334740382e-08, + "loss": 0.6109, + "step": 3686 + }, + { + "epoch": 5.778183716075157, + "grad_norm": 1.2722681760787964, + "learning_rate": 1.762727392692981e-08, + "loss": 0.6363, + "step": 3687 + }, + { + "epoch": 5.7797494780793315, + "grad_norm": 1.3638943433761597, + "learning_rate": 1.7378417367265865e-08, + "loss": 0.5585, + "step": 3688 + }, + { + "epoch": 5.781315240083507, + "grad_norm": 1.2096654176712036, + "learning_rate": 1.7131323832472736e-08, + "loss": 0.565, + "step": 3689 + }, + { + "epoch": 5.782881002087683, + "grad_norm": 1.3079156875610352, + "learning_rate": 1.6885993498023114e-08, + "loss": 0.6581, + "step": 3690 + }, + { + "epoch": 5.784446764091858, + "grad_norm": 1.255713939666748, + "learning_rate": 1.6642426538137357e-08, + "loss": 0.6287, + "step": 3691 + }, + { + "epoch": 5.786012526096034, + "grad_norm": 1.2188702821731567, + "learning_rate": 1.6400623125784053e-08, + "loss": 0.5933, + "step": 3692 + }, + { + "epoch": 5.7875782881002085, + "grad_norm": 1.1591229438781738, + "learning_rate": 1.616058343267862e-08, + "loss": 0.6189, + "step": 3693 + }, + { + "epoch": 5.789144050104384, + "grad_norm": 1.2696372270584106, + "learning_rate": 1.5922307629284706e-08, + "loss": 0.6217, + "step": 3694 + }, + { + "epoch": 5.79070981210856, + "grad_norm": 1.1796928644180298, + "learning_rate": 1.5685795884813058e-08, + "loss": 0.6551, + "step": 3695 + }, + { + "epoch": 5.792275574112735, + "grad_norm": 1.263322353363037, + "learning_rate": 1.5451048367221277e-08, + "loss": 0.5967, + "step": 3696 + }, + { + "epoch": 5.79384133611691, + "grad_norm": 1.3144618272781372, + "learning_rate": 1.521806524321462e-08, + "loss": 0.6416, + "step": 3697 + }, + { + "epoch": 5.7954070981210855, + "grad_norm": 1.3365296125411987, + "learning_rate": 1.498684667824546e-08, + "loss": 0.6141, + "step": 3698 + }, + { + "epoch": 5.796972860125261, + "grad_norm": 1.2244493961334229, + "learning_rate": 1.4757392836512453e-08, + "loss": 0.6023, + "step": 3699 + }, + { + "epoch": 5.798538622129437, + "grad_norm": 1.2851767539978027, + "learning_rate": 1.452970388096192e-08, + "loss": 0.5985, + "step": 3700 + }, + { + "epoch": 5.800104384133611, + "grad_norm": 1.1774119138717651, + "learning_rate": 1.4303779973285636e-08, + "loss": 0.6022, + "step": 3701 + }, + { + "epoch": 5.801670146137787, + "grad_norm": 1.2201687097549438, + "learning_rate": 1.4079621273923594e-08, + "loss": 0.5928, + "step": 3702 + }, + { + "epoch": 5.803235908141962, + "grad_norm": 1.325063705444336, + "learning_rate": 1.3857227942060403e-08, + "loss": 0.6069, + "step": 3703 + }, + { + "epoch": 5.804801670146138, + "grad_norm": 1.2128392457962036, + "learning_rate": 1.3636600135628341e-08, + "loss": 0.6065, + "step": 3704 + }, + { + "epoch": 5.806367432150314, + "grad_norm": 1.2512086629867554, + "learning_rate": 1.3417738011305692e-08, + "loss": 0.6101, + "step": 3705 + }, + { + "epoch": 5.807933194154488, + "grad_norm": 1.3149466514587402, + "learning_rate": 1.3200641724516183e-08, + "loss": 0.5675, + "step": 3706 + }, + { + "epoch": 5.809498956158664, + "grad_norm": 1.2250738143920898, + "learning_rate": 1.2985311429430104e-08, + "loss": 0.5691, + "step": 3707 + }, + { + "epoch": 5.811064718162839, + "grad_norm": 1.2220633029937744, + "learning_rate": 1.2771747278963464e-08, + "loss": 0.5247, + "step": 3708 + }, + { + "epoch": 5.812630480167015, + "grad_norm": 1.3097003698349, + "learning_rate": 1.2559949424778285e-08, + "loss": 0.6224, + "step": 3709 + }, + { + "epoch": 5.81419624217119, + "grad_norm": 1.234337329864502, + "learning_rate": 1.2349918017281749e-08, + "loss": 0.6116, + "step": 3710 + }, + { + "epoch": 5.815762004175365, + "grad_norm": 1.3012564182281494, + "learning_rate": 1.2141653205627046e-08, + "loss": 0.6343, + "step": 3711 + }, + { + "epoch": 5.817327766179541, + "grad_norm": 1.3063634634017944, + "learning_rate": 1.1935155137712817e-08, + "loss": 0.6238, + "step": 3712 + }, + { + "epoch": 5.818893528183716, + "grad_norm": 1.2745128870010376, + "learning_rate": 1.1730423960182868e-08, + "loss": 0.6165, + "step": 3713 + }, + { + "epoch": 5.820459290187891, + "grad_norm": 1.2855013608932495, + "learning_rate": 1.1527459818426178e-08, + "loss": 0.6302, + "step": 3714 + }, + { + "epoch": 5.822025052192067, + "grad_norm": 1.2457293272018433, + "learning_rate": 1.1326262856577175e-08, + "loss": 0.5439, + "step": 3715 + }, + { + "epoch": 5.823590814196242, + "grad_norm": 1.1844635009765625, + "learning_rate": 1.1126833217514898e-08, + "loss": 0.6226, + "step": 3716 + }, + { + "epoch": 5.825156576200418, + "grad_norm": 1.1715151071548462, + "learning_rate": 1.0929171042863839e-08, + "loss": 0.6079, + "step": 3717 + }, + { + "epoch": 5.826722338204593, + "grad_norm": 1.205751657485962, + "learning_rate": 1.0733276472993382e-08, + "loss": 0.5398, + "step": 3718 + }, + { + "epoch": 5.828288100208768, + "grad_norm": 1.3817614316940308, + "learning_rate": 1.0539149647016689e-08, + "loss": 0.6456, + "step": 3719 + }, + { + "epoch": 5.829853862212944, + "grad_norm": 1.3757718801498413, + "learning_rate": 1.0346790702792376e-08, + "loss": 0.5977, + "step": 3720 + }, + { + "epoch": 5.831419624217119, + "grad_norm": 1.242540717124939, + "learning_rate": 1.0156199776923946e-08, + "loss": 0.6123, + "step": 3721 + }, + { + "epoch": 5.832985386221294, + "grad_norm": 1.378861904144287, + "learning_rate": 9.967377004758416e-09, + "loss": 0.6191, + "step": 3722 + }, + { + "epoch": 5.8345511482254695, + "grad_norm": 1.2724937200546265, + "learning_rate": 9.780322520387963e-09, + "loss": 0.5997, + "step": 3723 + }, + { + "epoch": 5.836116910229645, + "grad_norm": 1.2619316577911377, + "learning_rate": 9.595036456648277e-09, + "loss": 0.6567, + "step": 3724 + }, + { + "epoch": 5.837682672233821, + "grad_norm": 1.163804054260254, + "learning_rate": 9.411518945119657e-09, + "loss": 0.5332, + "step": 3725 + }, + { + "epoch": 5.839248434237996, + "grad_norm": 1.2588956356048584, + "learning_rate": 9.229770116126746e-09, + "loss": 0.6085, + "step": 3726 + }, + { + "epoch": 5.840814196242171, + "grad_norm": 1.338892936706543, + "learning_rate": 9.04979009873741e-09, + "loss": 0.6442, + "step": 3727 + }, + { + "epoch": 5.8423799582463465, + "grad_norm": 1.223652958869934, + "learning_rate": 8.871579020764409e-09, + "loss": 0.5496, + "step": 3728 + }, + { + "epoch": 5.843945720250522, + "grad_norm": 1.3303650617599487, + "learning_rate": 8.6951370087629e-09, + "loss": 0.6114, + "step": 3729 + }, + { + "epoch": 5.845511482254698, + "grad_norm": 1.1986969709396362, + "learning_rate": 8.520464188032929e-09, + "loss": 0.5758, + "step": 3730 + }, + { + "epoch": 5.847077244258872, + "grad_norm": 1.1804407835006714, + "learning_rate": 8.34756068261805e-09, + "loss": 0.622, + "step": 3731 + }, + { + "epoch": 5.848643006263048, + "grad_norm": 1.2349408864974976, + "learning_rate": 8.176426615304767e-09, + "loss": 0.6497, + "step": 3732 + }, + { + "epoch": 5.8502087682672235, + "grad_norm": 1.3370027542114258, + "learning_rate": 8.007062107623364e-09, + "loss": 0.6261, + "step": 3733 + }, + { + "epoch": 5.851774530271399, + "grad_norm": 1.2989472150802612, + "learning_rate": 7.839467279847357e-09, + "loss": 0.6008, + "step": 3734 + }, + { + "epoch": 5.853340292275574, + "grad_norm": 1.2511489391326904, + "learning_rate": 7.673642250993762e-09, + "loss": 0.6268, + "step": 3735 + }, + { + "epoch": 5.854906054279749, + "grad_norm": 1.288687825202942, + "learning_rate": 7.509587138822549e-09, + "loss": 0.6125, + "step": 3736 + }, + { + "epoch": 5.856471816283925, + "grad_norm": 1.3703006505966187, + "learning_rate": 7.3473020598371915e-09, + "loss": 0.621, + "step": 3737 + }, + { + "epoch": 5.8580375782881005, + "grad_norm": 1.2187769412994385, + "learning_rate": 7.186787129283279e-09, + "loss": 0.5594, + "step": 3738 + }, + { + "epoch": 5.859603340292276, + "grad_norm": 1.2823399305343628, + "learning_rate": 7.028042461150187e-09, + "loss": 0.5837, + "step": 3739 + }, + { + "epoch": 5.861169102296451, + "grad_norm": 1.2807185649871826, + "learning_rate": 6.871068168170237e-09, + "loss": 0.6046, + "step": 3740 + }, + { + "epoch": 5.862734864300626, + "grad_norm": 1.1859006881713867, + "learning_rate": 6.715864361817592e-09, + "loss": 0.6362, + "step": 3741 + }, + { + "epoch": 5.864300626304802, + "grad_norm": 1.4097034931182861, + "learning_rate": 6.562431152309923e-09, + "loss": 0.5823, + "step": 3742 + }, + { + "epoch": 5.8658663883089766, + "grad_norm": 1.1742969751358032, + "learning_rate": 6.41076864860729e-09, + "loss": 0.5698, + "step": 3743 + }, + { + "epoch": 5.867432150313152, + "grad_norm": 1.1863396167755127, + "learning_rate": 6.260876958412432e-09, + "loss": 0.5944, + "step": 3744 + }, + { + "epoch": 5.868997912317328, + "grad_norm": 1.206772804260254, + "learning_rate": 6.112756188170199e-09, + "loss": 0.5208, + "step": 3745 + }, + { + "epoch": 5.870563674321503, + "grad_norm": 1.3259044885635376, + "learning_rate": 5.96640644306784e-09, + "loss": 0.5765, + "step": 3746 + }, + { + "epoch": 5.872129436325679, + "grad_norm": 1.4725146293640137, + "learning_rate": 5.8218278270355534e-09, + "loss": 0.5978, + "step": 3747 + }, + { + "epoch": 5.8736951983298535, + "grad_norm": 1.302071213722229, + "learning_rate": 5.679020442745098e-09, + "loss": 0.5235, + "step": 3748 + }, + { + "epoch": 5.875260960334029, + "grad_norm": 1.236410140991211, + "learning_rate": 5.537984391610629e-09, + "loss": 0.6485, + "step": 3749 + }, + { + "epoch": 5.876826722338205, + "grad_norm": 1.2594784498214722, + "learning_rate": 5.398719773788419e-09, + "loss": 0.6174, + "step": 3750 + }, + { + "epoch": 5.87839248434238, + "grad_norm": 1.1691596508026123, + "learning_rate": 5.261226688176579e-09, + "loss": 0.6461, + "step": 3751 + }, + { + "epoch": 5.879958246346555, + "grad_norm": 1.3323622941970825, + "learning_rate": 5.125505232415618e-09, + "loss": 0.5811, + "step": 3752 + }, + { + "epoch": 5.8815240083507305, + "grad_norm": 1.1331349611282349, + "learning_rate": 4.991555502887601e-09, + "loss": 0.6315, + "step": 3753 + }, + { + "epoch": 5.883089770354906, + "grad_norm": 1.2791389226913452, + "learning_rate": 4.859377594716163e-09, + "loss": 0.5812, + "step": 3754 + }, + { + "epoch": 5.884655532359082, + "grad_norm": 1.2167613506317139, + "learning_rate": 4.728971601767329e-09, + "loss": 0.6705, + "step": 3755 + }, + { + "epoch": 5.886221294363256, + "grad_norm": 1.230459213256836, + "learning_rate": 4.600337616648131e-09, + "loss": 0.4992, + "step": 3756 + }, + { + "epoch": 5.887787056367432, + "grad_norm": 1.3930604457855225, + "learning_rate": 4.473475730707721e-09, + "loss": 0.6456, + "step": 3757 + }, + { + "epoch": 5.8893528183716075, + "grad_norm": 1.3491474390029907, + "learning_rate": 4.3483860340365335e-09, + "loss": 0.6355, + "step": 3758 + }, + { + "epoch": 5.890918580375783, + "grad_norm": 1.2870649099349976, + "learning_rate": 4.225068615466843e-09, + "loss": 0.6118, + "step": 3759 + }, + { + "epoch": 5.892484342379959, + "grad_norm": 1.2994024753570557, + "learning_rate": 4.103523562571654e-09, + "loss": 0.557, + "step": 3760 + }, + { + "epoch": 5.894050104384133, + "grad_norm": 1.2077248096466064, + "learning_rate": 3.983750961666088e-09, + "loss": 0.5848, + "step": 3761 + }, + { + "epoch": 5.895615866388309, + "grad_norm": 1.396703839302063, + "learning_rate": 3.865750897806275e-09, + "loss": 0.6662, + "step": 3762 + }, + { + "epoch": 5.8971816283924845, + "grad_norm": 1.2555094957351685, + "learning_rate": 3.749523454789072e-09, + "loss": 0.6243, + "step": 3763 + }, + { + "epoch": 5.89874739039666, + "grad_norm": 1.2748899459838867, + "learning_rate": 3.6350687151531782e-09, + "loss": 0.5068, + "step": 3764 + }, + { + "epoch": 5.900313152400835, + "grad_norm": 1.2701084613800049, + "learning_rate": 3.5223867601785754e-09, + "loss": 0.5863, + "step": 3765 + }, + { + "epoch": 5.90187891440501, + "grad_norm": 1.338657259941101, + "learning_rate": 3.4114776698854215e-09, + "loss": 0.5791, + "step": 3766 + }, + { + "epoch": 5.903444676409186, + "grad_norm": 1.2731455564498901, + "learning_rate": 3.30234152303599e-09, + "loss": 0.5464, + "step": 3767 + }, + { + "epoch": 5.9050104384133615, + "grad_norm": 1.3091557025909424, + "learning_rate": 3.1949783971324535e-09, + "loss": 0.6203, + "step": 3768 + }, + { + "epoch": 5.906576200417536, + "grad_norm": 1.232047438621521, + "learning_rate": 3.089388368418544e-09, + "loss": 0.6483, + "step": 3769 + }, + { + "epoch": 5.908141962421712, + "grad_norm": 1.3137093782424927, + "learning_rate": 2.9855715118787265e-09, + "loss": 0.6578, + "step": 3770 + }, + { + "epoch": 5.909707724425887, + "grad_norm": 1.3319385051727295, + "learning_rate": 2.8835279012379147e-09, + "loss": 0.5578, + "step": 3771 + }, + { + "epoch": 5.911273486430063, + "grad_norm": 1.2313734292984009, + "learning_rate": 2.7832576089623086e-09, + "loss": 0.5776, + "step": 3772 + }, + { + "epoch": 5.9128392484342385, + "grad_norm": 1.230716347694397, + "learning_rate": 2.6847607062582826e-09, + "loss": 0.5903, + "step": 3773 + }, + { + "epoch": 5.914405010438413, + "grad_norm": 1.2998744249343872, + "learning_rate": 2.5880372630732174e-09, + "loss": 0.6593, + "step": 3774 + }, + { + "epoch": 5.915970772442589, + "grad_norm": 1.3287510871887207, + "learning_rate": 2.4930873480946694e-09, + "loss": 0.5698, + "step": 3775 + }, + { + "epoch": 5.917536534446764, + "grad_norm": 1.1340152025222778, + "learning_rate": 2.3999110287512005e-09, + "loss": 0.5893, + "step": 3776 + }, + { + "epoch": 5.919102296450939, + "grad_norm": 1.2747018337249756, + "learning_rate": 2.3085083712115486e-09, + "loss": 0.6281, + "step": 3777 + }, + { + "epoch": 5.920668058455115, + "grad_norm": 1.3687758445739746, + "learning_rate": 2.21887944038518e-09, + "loss": 0.5984, + "step": 3778 + }, + { + "epoch": 5.92223382045929, + "grad_norm": 1.3160481452941895, + "learning_rate": 2.1310242999217356e-09, + "loss": 0.5382, + "step": 3779 + }, + { + "epoch": 5.923799582463466, + "grad_norm": 1.3093929290771484, + "learning_rate": 2.044943012210754e-09, + "loss": 0.6056, + "step": 3780 + }, + { + "epoch": 5.925365344467641, + "grad_norm": 1.3370702266693115, + "learning_rate": 1.960635638383057e-09, + "loss": 0.5711, + "step": 3781 + }, + { + "epoch": 5.926931106471816, + "grad_norm": 1.2634838819503784, + "learning_rate": 1.8781022383090873e-09, + "loss": 0.573, + "step": 3782 + }, + { + "epoch": 5.9284968684759916, + "grad_norm": 1.4597774744033813, + "learning_rate": 1.7973428705997387e-09, + "loss": 0.6024, + "step": 3783 + }, + { + "epoch": 5.930062630480167, + "grad_norm": 1.1795402765274048, + "learning_rate": 1.7183575926058016e-09, + "loss": 0.6044, + "step": 3784 + }, + { + "epoch": 5.931628392484343, + "grad_norm": 1.1035513877868652, + "learning_rate": 1.641146460418519e-09, + "loss": 0.6468, + "step": 3785 + }, + { + "epoch": 5.933194154488517, + "grad_norm": 1.5152153968811035, + "learning_rate": 1.5657095288690304e-09, + "loss": 0.5662, + "step": 3786 + }, + { + "epoch": 5.934759916492693, + "grad_norm": 1.2346514463424683, + "learning_rate": 1.4920468515283726e-09, + "loss": 0.5593, + "step": 3787 + }, + { + "epoch": 5.9363256784968685, + "grad_norm": 1.4137080907821655, + "learning_rate": 1.4201584807083113e-09, + "loss": 0.6073, + "step": 3788 + }, + { + "epoch": 5.937891440501044, + "grad_norm": 1.1767393350601196, + "learning_rate": 1.3500444674599544e-09, + "loss": 0.6304, + "step": 3789 + }, + { + "epoch": 5.939457202505219, + "grad_norm": 1.3576456308364868, + "learning_rate": 1.2817048615743067e-09, + "loss": 0.5507, + "step": 3790 + }, + { + "epoch": 5.941022964509394, + "grad_norm": 1.2209057807922363, + "learning_rate": 1.2151397115828245e-09, + "loss": 0.6118, + "step": 3791 + }, + { + "epoch": 5.94258872651357, + "grad_norm": 1.226086139678955, + "learning_rate": 1.1503490647563065e-09, + "loss": 0.6193, + "step": 3792 + }, + { + "epoch": 5.9441544885177455, + "grad_norm": 1.1844003200531006, + "learning_rate": 1.0873329671054477e-09, + "loss": 0.5964, + "step": 3793 + }, + { + "epoch": 5.945720250521921, + "grad_norm": 1.342863917350769, + "learning_rate": 1.0260914633811181e-09, + "loss": 0.5849, + "step": 3794 + }, + { + "epoch": 5.947286012526096, + "grad_norm": 1.1411577463150024, + "learning_rate": 9.666245970738063e-10, + "loss": 0.5742, + "step": 3795 + }, + { + "epoch": 5.948851774530271, + "grad_norm": 1.2785258293151855, + "learning_rate": 9.08932410413621e-10, + "loss": 0.558, + "step": 3796 + }, + { + "epoch": 5.950417536534447, + "grad_norm": 1.4084155559539795, + "learning_rate": 8.530149443705671e-10, + "loss": 0.6329, + "step": 3797 + }, + { + "epoch": 5.9519832985386225, + "grad_norm": 1.2690620422363281, + "learning_rate": 7.988722386539915e-10, + "loss": 0.6673, + "step": 3798 + }, + { + "epoch": 5.953549060542797, + "grad_norm": 1.3061922788619995, + "learning_rate": 7.465043317131382e-10, + "loss": 0.6458, + "step": 3799 + }, + { + "epoch": 5.955114822546973, + "grad_norm": 1.323809027671814, + "learning_rate": 6.959112607371476e-10, + "loss": 0.6327, + "step": 3800 + }, + { + "epoch": 5.956680584551148, + "grad_norm": 1.2874526977539062, + "learning_rate": 6.470930616545023e-10, + "loss": 0.5957, + "step": 3801 + }, + { + "epoch": 5.958246346555324, + "grad_norm": 1.1838451623916626, + "learning_rate": 6.000497691330264e-10, + "loss": 0.6247, + "step": 3802 + }, + { + "epoch": 5.959812108559499, + "grad_norm": 1.2807035446166992, + "learning_rate": 5.547814165804411e-10, + "loss": 0.6116, + "step": 3803 + }, + { + "epoch": 5.961377870563674, + "grad_norm": 1.2280174493789673, + "learning_rate": 5.112880361438088e-10, + "loss": 0.6431, + "step": 3804 + }, + { + "epoch": 5.96294363256785, + "grad_norm": 1.1873998641967773, + "learning_rate": 4.695696587100895e-10, + "loss": 0.5909, + "step": 3805 + }, + { + "epoch": 5.964509394572025, + "grad_norm": 1.33540940284729, + "learning_rate": 4.2962631390530697e-10, + "loss": 0.6109, + "step": 3806 + }, + { + "epoch": 5.966075156576201, + "grad_norm": 1.2234054803848267, + "learning_rate": 3.914580300951043e-10, + "loss": 0.6348, + "step": 3807 + }, + { + "epoch": 5.967640918580376, + "grad_norm": 1.3723591566085815, + "learning_rate": 3.5506483438446646e-10, + "loss": 0.5201, + "step": 3808 + }, + { + "epoch": 5.969206680584551, + "grad_norm": 1.1671289205551147, + "learning_rate": 3.204467526177202e-10, + "loss": 0.5831, + "step": 3809 + }, + { + "epoch": 5.970772442588727, + "grad_norm": 1.2280356884002686, + "learning_rate": 2.8760380937936647e-10, + "loss": 0.6227, + "step": 3810 + }, + { + "epoch": 5.972338204592901, + "grad_norm": 1.1912339925765991, + "learning_rate": 2.5653602799213804e-10, + "loss": 0.5989, + "step": 3811 + }, + { + "epoch": 5.973903966597077, + "grad_norm": 1.2407976388931274, + "learning_rate": 2.2724343051866438e-10, + "loss": 0.6175, + "step": 3812 + }, + { + "epoch": 5.975469728601253, + "grad_norm": 1.4313815832138062, + "learning_rate": 1.997260377614718e-10, + "loss": 0.6205, + "step": 3813 + }, + { + "epoch": 5.977035490605428, + "grad_norm": 1.2734475135803223, + "learning_rate": 1.7398386926159582e-10, + "loss": 0.5919, + "step": 3814 + }, + { + "epoch": 5.978601252609604, + "grad_norm": 1.3108834028244019, + "learning_rate": 1.5001694329996874e-10, + "loss": 0.5807, + "step": 3815 + }, + { + "epoch": 5.980167014613778, + "grad_norm": 1.3993126153945923, + "learning_rate": 1.2782527689630952e-10, + "loss": 0.6291, + "step": 3816 + }, + { + "epoch": 5.981732776617954, + "grad_norm": 1.280991554260254, + "learning_rate": 1.0740888581023401e-10, + "loss": 0.6294, + "step": 3817 + }, + { + "epoch": 5.98329853862213, + "grad_norm": 1.3276211023330688, + "learning_rate": 8.876778454014467e-11, + "loss": 0.6025, + "step": 3818 + }, + { + "epoch": 5.984864300626305, + "grad_norm": 1.183289647102356, + "learning_rate": 7.190198632406331e-11, + "loss": 0.6169, + "step": 3819 + }, + { + "epoch": 5.98643006263048, + "grad_norm": 1.1286604404449463, + "learning_rate": 5.681150313907591e-11, + "loss": 0.5932, + "step": 3820 + }, + { + "epoch": 5.987995824634655, + "grad_norm": 1.380447268486023, + "learning_rate": 4.349634570188777e-11, + "loss": 0.5574, + "step": 3821 + }, + { + "epoch": 5.989561586638831, + "grad_norm": 1.1990282535552979, + "learning_rate": 3.195652346799083e-11, + "loss": 0.6394, + "step": 3822 + }, + { + "epoch": 5.991127348643007, + "grad_norm": 1.2492790222167969, + "learning_rate": 2.2192044632496356e-11, + "loss": 0.5568, + "step": 3823 + }, + { + "epoch": 5.992693110647181, + "grad_norm": 1.3902348279953003, + "learning_rate": 1.4202916129302246e-11, + "loss": 0.6358, + "step": 3824 + }, + { + "epoch": 5.994258872651357, + "grad_norm": 1.300254225730896, + "learning_rate": 7.989143632203267e-12, + "loss": 0.5995, + "step": 3825 + }, + { + "epoch": 5.995824634655532, + "grad_norm": 1.166805386543274, + "learning_rate": 3.5507315537808284e-12, + "loss": 0.5363, + "step": 3826 + }, + { + "epoch": 5.997390396659708, + "grad_norm": 1.3551692962646484, + "learning_rate": 8.876830459580987e-13, + "loss": 0.5666, + "step": 3827 + }, + { + "epoch": 5.9989561586638835, + "grad_norm": 1.3476738929748535, + "learning_rate": 0.0, + "loss": 0.6208, + "step": 3828 + } + ], + "logging_steps": 1, + "max_steps": 3828, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 638, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 9.673668033301709e+18, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-3828/training_args.bin b/checkpoint-3828/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..07c8c9aae2a66d9fb6a8b65759b1cd86d2f7d08b --- /dev/null +++ b/checkpoint-3828/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81aeca374ba1e1a5384aca380c5dd9da735cc6d6164e0b5b0681e227cf61f41c +size 7992 diff --git a/checkpoint-3828/zero_to_fp32.py b/checkpoint-3828/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-3828/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..f07164552af92bf3072e4e4326b5e572dea01ae8 --- /dev/null +++ b/checkpoint-638/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-Nano-8B-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 0.14.0 \ 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..3db75f3ecbeb5f4990760e56c7957796d83680c5 --- /dev/null +++ b/checkpoint-638/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1", + "bias": "none", + "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": [ + "q_proj", + "down_proj", + "o_proj", + "v_proj", + "k_proj", + "gate_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "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..5d7875864030fbad1b3f28983899c3d7d72a2c47 --- /dev/null +++ b/checkpoint-638/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b6be6f6827329f7b6b5d346b15b149a8c62b8a19a707141ebee0f6d2395ccbb2 +size 3443586272 diff --git a/checkpoint-638/global_step638/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-638/global_step638/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..63f77c7058f973e2a4070c00743ff0b12058fb9c --- /dev/null +++ b/checkpoint-638/global_step638/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:ebcde64930aba772c7b20fdff0aadea52e8fb6d0c90d079e88b76c776ad5bc27 +size 20661195036 diff --git a/checkpoint-638/global_step638/mp_rank_00_model_states.pt b/checkpoint-638/global_step638/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..5f6a65c9e44e89aa886d5c9c5ee8658e2aa63ef2 --- /dev/null +++ b/checkpoint-638/global_step638/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f74fffdb3805e61ba61d4d0f023a34d19c85022674541922d1601eabba38d8bb +size 3555326777 diff --git a/checkpoint-638/latest b/checkpoint-638/latest new file mode 100644 index 0000000000000000000000000000000000000000..9ad9c31e72daba355fe30d3751281b5a377bb997 --- /dev/null +++ b/checkpoint-638/latest @@ -0,0 +1 @@ +global_step638 \ No newline at end of file diff --git a/checkpoint-638/rng_state.pth b/checkpoint-638/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..db83635162ae8cba9fc255c1434ea1dcaf1cd086 --- /dev/null +++ b/checkpoint-638/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:54a1527b2da471128d8149521961d653304c6a48ce1d93ee52f61c2673e8a605 +size 14244 diff --git a/checkpoint-638/scheduler.pt b/checkpoint-638/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..f4557923d19edbdb24cad1b3af8c741ba2a93be0 --- /dev/null +++ b/checkpoint-638/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b7e4eea191dde345c636b5183607ae39a5e00b97ed5fec9e783b9ae2f08e7ac +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..b480d149dc274c0a23db079bfd38032d90082e49 --- /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": "{%- if messages[0]['role'] == 'system' -%}{%- set system_message = messages[0]['content'] | trim -%}{%- set messages = messages[1:] -%}{%- else -%}{%- set system_message = '' -%}{%- endif -%}{%- if tools is not none -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message -}} {{- '\n\n' if system_message else '' -}} {{- '[' -}} {% for t in tools %}{{- (t.function if t.function is defined else t) | tojson() -}}{{- ', ' if not loop.last else '' -}}{%- endfor -%} {{- ']' -}} {{- '<|eot_id|>' -}}{%- else -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message + '<|eot_id|>' -}}{%- endif -%}{%- for message in messages -%}{%- if (message['role'] in ['user', 'tool']) != (loop.index0 % 2 == 0) -%}{{- raise_exception('Conversation roles must alternate between user/tool and assistant') -}}{%- elif message['role'] == 'user' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- elif message['role'] == 'tool' -%}{%- set tool_response = '[' + message['content'] | trim + ']' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + tool_response + '<|eot_id|>' -}}{%- elif message['role'] == 'assistant' and message.get('tool_calls') is not none -%}{%- set tool_calls = message['tool_calls'] -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + '[' -}}{%- for tool_call in tool_calls -%}{{ '{' + '\"name\": \"' + tool_call.function.name + '\", \"arguments\": ' + tool_call.function.arguments | tojson + '}' }}{%- if not loop.last -%}{{ ', ' }}{%- else -%}{{ ']' + '<|eot_id|>' }}{%- endif -%}{%- endfor -%}{%- elif message['role'] == 'assistant' -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- endif -%}{%- 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..50ca456da67f3a032fdc9ae13b70822da0d449bf --- /dev/null +++ b/checkpoint-638/trainer_state.json @@ -0,0 +1,4500 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.9989561586638831, + "eval_steps": 500, + "global_step": 638, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0015657620041753654, + "grad_norm": 658.1201782226562, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.3039, + "step": 1 + }, + { + "epoch": 0.003131524008350731, + "grad_norm": 749.5796508789062, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.4164, + "step": 2 + }, + { + "epoch": 0.004697286012526096, + "grad_norm": 394.2350769042969, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.3367, + "step": 3 + }, + { + "epoch": 0.006263048016701462, + "grad_norm": 384.08319091796875, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.3192, + "step": 4 + }, + { + "epoch": 0.007828810020876827, + "grad_norm": 376.9790954589844, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.2756, + "step": 5 + }, + { + "epoch": 0.009394572025052192, + "grad_norm": 175.25262451171875, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.3992, + "step": 6 + }, + { + "epoch": 0.010960334029227558, + "grad_norm": 100.1114730834961, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.2348, + "step": 7 + }, + { + "epoch": 0.012526096033402923, + "grad_norm": 131.44432067871094, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.2685, + "step": 8 + }, + { + "epoch": 0.014091858037578288, + "grad_norm": 67.03341674804688, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.3272, + "step": 9 + }, + { + "epoch": 0.015657620041753653, + "grad_norm": 32.0869140625, + "learning_rate": 5.000000000000001e-07, + "loss": 1.3458, + "step": 10 + }, + { + "epoch": 0.01722338204592902, + "grad_norm": 17.701913833618164, + "learning_rate": 5.5e-07, + "loss": 1.3608, + "step": 11 + }, + { + "epoch": 0.018789144050104383, + "grad_norm": 9.164765357971191, + "learning_rate": 6.000000000000001e-07, + "loss": 1.2961, + "step": 12 + }, + { + "epoch": 0.02035490605427975, + "grad_norm": 4.315349578857422, + "learning_rate": 6.5e-07, + "loss": 1.2069, + "step": 13 + }, + { + "epoch": 0.021920668058455117, + "grad_norm": 6.652385234832764, + "learning_rate": 7.000000000000001e-07, + "loss": 1.2588, + "step": 14 + }, + { + "epoch": 0.02348643006263048, + "grad_norm": 3.3322548866271973, + "learning_rate": 7.5e-07, + "loss": 1.2505, + "step": 15 + }, + { + "epoch": 0.025052192066805846, + "grad_norm": 9.599331855773926, + "learning_rate": 8.000000000000001e-07, + "loss": 1.1343, + "step": 16 + }, + { + "epoch": 0.02661795407098121, + "grad_norm": 10.33918285369873, + "learning_rate": 8.500000000000001e-07, + "loss": 1.3334, + "step": 17 + }, + { + "epoch": 0.028183716075156576, + "grad_norm": 9.453723907470703, + "learning_rate": 9.000000000000001e-07, + "loss": 1.3099, + "step": 18 + }, + { + "epoch": 0.029749478079331943, + "grad_norm": 7.075769901275635, + "learning_rate": 9.500000000000001e-07, + "loss": 1.1121, + "step": 19 + }, + { + "epoch": 0.031315240083507306, + "grad_norm": 7.770864009857178, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.1836, + "step": 20 + }, + { + "epoch": 0.03288100208768267, + "grad_norm": 9.582944869995117, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.1939, + "step": 21 + }, + { + "epoch": 0.03444676409185804, + "grad_norm": 9.496583938598633, + "learning_rate": 1.1e-06, + "loss": 1.2395, + "step": 22 + }, + { + "epoch": 0.0360125260960334, + "grad_norm": 5.912876129150391, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.1769, + "step": 23 + }, + { + "epoch": 0.037578288100208766, + "grad_norm": 2.6761083602905273, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.204, + "step": 24 + }, + { + "epoch": 0.03914405010438413, + "grad_norm": 11.136463165283203, + "learning_rate": 1.25e-06, + "loss": 1.2271, + "step": 25 + }, + { + "epoch": 0.0407098121085595, + "grad_norm": 23.912893295288086, + "learning_rate": 1.3e-06, + "loss": 1.1771, + "step": 26 + }, + { + "epoch": 0.042275574112734866, + "grad_norm": 57.08906555175781, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.2253, + "step": 27 + }, + { + "epoch": 0.04384133611691023, + "grad_norm": 36.11408615112305, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.1014, + "step": 28 + }, + { + "epoch": 0.04540709812108559, + "grad_norm": 51.39445495605469, + "learning_rate": 1.45e-06, + "loss": 1.1978, + "step": 29 + }, + { + "epoch": 0.04697286012526096, + "grad_norm": 27.394500732421875, + "learning_rate": 1.5e-06, + "loss": 1.1155, + "step": 30 + }, + { + "epoch": 0.048538622129436326, + "grad_norm": 6.844716548919678, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.1798, + "step": 31 + }, + { + "epoch": 0.05010438413361169, + "grad_norm": 2.397536277770996, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.1846, + "step": 32 + }, + { + "epoch": 0.05167014613778706, + "grad_norm": 7.821156024932861, + "learning_rate": 1.6500000000000003e-06, + "loss": 1.1312, + "step": 33 + }, + { + "epoch": 0.05323590814196242, + "grad_norm": 5.455922603607178, + "learning_rate": 1.7000000000000002e-06, + "loss": 1.2153, + "step": 34 + }, + { + "epoch": 0.054801670146137786, + "grad_norm": 5.377617835998535, + "learning_rate": 1.75e-06, + "loss": 1.1292, + "step": 35 + }, + { + "epoch": 0.05636743215031315, + "grad_norm": 7.172031879425049, + "learning_rate": 1.8000000000000001e-06, + "loss": 1.1459, + "step": 36 + }, + { + "epoch": 0.05793319415448852, + "grad_norm": 2.1464905738830566, + "learning_rate": 1.85e-06, + "loss": 1.1649, + "step": 37 + }, + { + "epoch": 0.059498956158663886, + "grad_norm": 2.7782726287841797, + "learning_rate": 1.9000000000000002e-06, + "loss": 1.159, + "step": 38 + }, + { + "epoch": 0.061064718162839246, + "grad_norm": 20.818374633789062, + "learning_rate": 1.9500000000000004e-06, + "loss": 1.0174, + "step": 39 + }, + { + "epoch": 0.06263048016701461, + "grad_norm": 22.776456832885742, + "learning_rate": 2.0000000000000003e-06, + "loss": 1.1425, + "step": 40 + }, + { + "epoch": 0.06419624217118998, + "grad_norm": 33.601287841796875, + "learning_rate": 2.05e-06, + "loss": 1.0498, + "step": 41 + }, + { + "epoch": 0.06576200417536535, + "grad_norm": 12.884684562683105, + "learning_rate": 2.1000000000000002e-06, + "loss": 1.1231, + "step": 42 + }, + { + "epoch": 0.06732776617954071, + "grad_norm": 1.951470971107483, + "learning_rate": 2.15e-06, + "loss": 1.1256, + "step": 43 + }, + { + "epoch": 0.06889352818371608, + "grad_norm": 2.584127426147461, + "learning_rate": 2.2e-06, + "loss": 1.1242, + "step": 44 + }, + { + "epoch": 0.07045929018789145, + "grad_norm": 6.982508659362793, + "learning_rate": 2.25e-06, + "loss": 1.0769, + "step": 45 + }, + { + "epoch": 0.0720250521920668, + "grad_norm": 7.940657138824463, + "learning_rate": 2.3000000000000004e-06, + "loss": 1.0516, + "step": 46 + }, + { + "epoch": 0.07359081419624217, + "grad_norm": 5.456567287445068, + "learning_rate": 2.35e-06, + "loss": 0.9894, + "step": 47 + }, + { + "epoch": 0.07515657620041753, + "grad_norm": 4.623636722564697, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9912, + "step": 48 + }, + { + "epoch": 0.0767223382045929, + "grad_norm": 2.182708978652954, + "learning_rate": 2.4500000000000003e-06, + "loss": 1.1303, + "step": 49 + }, + { + "epoch": 0.07828810020876827, + "grad_norm": 2.388091802597046, + "learning_rate": 2.5e-06, + "loss": 1.0725, + "step": 50 + }, + { + "epoch": 0.07985386221294363, + "grad_norm": 2.2816522121429443, + "learning_rate": 2.55e-06, + "loss": 1.0895, + "step": 51 + }, + { + "epoch": 0.081419624217119, + "grad_norm": 3.8391623497009277, + "learning_rate": 2.6e-06, + "loss": 1.0587, + "step": 52 + }, + { + "epoch": 0.08298538622129437, + "grad_norm": 3.700389862060547, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9795, + "step": 53 + }, + { + "epoch": 0.08455114822546973, + "grad_norm": 3.180649995803833, + "learning_rate": 2.7000000000000004e-06, + "loss": 1.0434, + "step": 54 + }, + { + "epoch": 0.0861169102296451, + "grad_norm": 3.790774345397949, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9652, + "step": 55 + }, + { + "epoch": 0.08768267223382047, + "grad_norm": 1.8411438465118408, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.947, + "step": 56 + }, + { + "epoch": 0.08924843423799582, + "grad_norm": 2.100008010864258, + "learning_rate": 2.85e-06, + "loss": 1.0129, + "step": 57 + }, + { + "epoch": 0.09081419624217119, + "grad_norm": 1.103857159614563, + "learning_rate": 2.9e-06, + "loss": 0.9993, + "step": 58 + }, + { + "epoch": 0.09237995824634655, + "grad_norm": 1.232283353805542, + "learning_rate": 2.95e-06, + "loss": 0.9988, + "step": 59 + }, + { + "epoch": 0.09394572025052192, + "grad_norm": 2.4731297492980957, + "learning_rate": 3e-06, + "loss": 0.9873, + "step": 60 + }, + { + "epoch": 0.09551148225469729, + "grad_norm": 1.1080410480499268, + "learning_rate": 3.05e-06, + "loss": 1.0657, + "step": 61 + }, + { + "epoch": 0.09707724425887265, + "grad_norm": 1.1155121326446533, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.974, + "step": 62 + }, + { + "epoch": 0.09864300626304802, + "grad_norm": 1.763231873512268, + "learning_rate": 3.1500000000000003e-06, + "loss": 1.0215, + "step": 63 + }, + { + "epoch": 0.10020876826722339, + "grad_norm": 1.1961321830749512, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.9955, + "step": 64 + }, + { + "epoch": 0.10177453027139875, + "grad_norm": 1.377040982246399, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.9834, + "step": 65 + }, + { + "epoch": 0.10334029227557412, + "grad_norm": 1.373260498046875, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.984, + "step": 66 + }, + { + "epoch": 0.10490605427974947, + "grad_norm": 2.034064292907715, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.973, + "step": 67 + }, + { + "epoch": 0.10647181628392484, + "grad_norm": 1.41798734664917, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.925, + "step": 68 + }, + { + "epoch": 0.1080375782881002, + "grad_norm": 1.7425439357757568, + "learning_rate": 3.45e-06, + "loss": 0.9884, + "step": 69 + }, + { + "epoch": 0.10960334029227557, + "grad_norm": 2.369657278060913, + "learning_rate": 3.5e-06, + "loss": 1.0048, + "step": 70 + }, + { + "epoch": 0.11116910229645094, + "grad_norm": 1.4699137210845947, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.9155, + "step": 71 + }, + { + "epoch": 0.1127348643006263, + "grad_norm": 1.1473263502120972, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8445, + "step": 72 + }, + { + "epoch": 0.11430062630480167, + "grad_norm": 2.315568685531616, + "learning_rate": 3.65e-06, + "loss": 0.977, + "step": 73 + }, + { + "epoch": 0.11586638830897704, + "grad_norm": 1.8554916381835938, + "learning_rate": 3.7e-06, + "loss": 0.841, + "step": 74 + }, + { + "epoch": 0.1174321503131524, + "grad_norm": 1.1517397165298462, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.9478, + "step": 75 + }, + { + "epoch": 0.11899791231732777, + "grad_norm": 2.799316883087158, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.9873, + "step": 76 + }, + { + "epoch": 0.12056367432150313, + "grad_norm": 1.11930251121521, + "learning_rate": 3.85e-06, + "loss": 0.9214, + "step": 77 + }, + { + "epoch": 0.12212943632567849, + "grad_norm": 1.0082950592041016, + "learning_rate": 3.900000000000001e-06, + "loss": 0.9115, + "step": 78 + }, + { + "epoch": 0.12369519832985386, + "grad_norm": 1.5534541606903076, + "learning_rate": 3.95e-06, + "loss": 1.04, + "step": 79 + }, + { + "epoch": 0.12526096033402923, + "grad_norm": 1.3469066619873047, + "learning_rate": 4.000000000000001e-06, + "loss": 0.9091, + "step": 80 + }, + { + "epoch": 0.1268267223382046, + "grad_norm": 1.1149636507034302, + "learning_rate": 4.05e-06, + "loss": 0.8829, + "step": 81 + }, + { + "epoch": 0.12839248434237996, + "grad_norm": 2.5933806896209717, + "learning_rate": 4.1e-06, + "loss": 0.9343, + "step": 82 + }, + { + "epoch": 0.1299582463465553, + "grad_norm": 1.980901837348938, + "learning_rate": 4.15e-06, + "loss": 0.9028, + "step": 83 + }, + { + "epoch": 0.1315240083507307, + "grad_norm": 0.8476768732070923, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.9038, + "step": 84 + }, + { + "epoch": 0.13308977035490605, + "grad_norm": 1.1239815950393677, + "learning_rate": 4.25e-06, + "loss": 1.0359, + "step": 85 + }, + { + "epoch": 0.13465553235908143, + "grad_norm": 1.0159156322479248, + "learning_rate": 4.3e-06, + "loss": 0.8705, + "step": 86 + }, + { + "epoch": 0.13622129436325678, + "grad_norm": 2.180856466293335, + "learning_rate": 4.350000000000001e-06, + "loss": 0.9834, + "step": 87 + }, + { + "epoch": 0.13778705636743216, + "grad_norm": 1.5725276470184326, + "learning_rate": 4.4e-06, + "loss": 0.9449, + "step": 88 + }, + { + "epoch": 0.1393528183716075, + "grad_norm": 1.5560917854309082, + "learning_rate": 4.450000000000001e-06, + "loss": 1.0109, + "step": 89 + }, + { + "epoch": 0.1409185803757829, + "grad_norm": 1.1096960306167603, + "learning_rate": 4.5e-06, + "loss": 0.9081, + "step": 90 + }, + { + "epoch": 0.14248434237995825, + "grad_norm": 1.2525160312652588, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.9907, + "step": 91 + }, + { + "epoch": 0.1440501043841336, + "grad_norm": 0.9198770523071289, + "learning_rate": 4.600000000000001e-06, + "loss": 0.9266, + "step": 92 + }, + { + "epoch": 0.14561586638830898, + "grad_norm": 0.992057740688324, + "learning_rate": 4.65e-06, + "loss": 0.9556, + "step": 93 + }, + { + "epoch": 0.14718162839248433, + "grad_norm": 1.00990891456604, + "learning_rate": 4.7e-06, + "loss": 0.9712, + "step": 94 + }, + { + "epoch": 0.1487473903966597, + "grad_norm": 1.1176848411560059, + "learning_rate": 4.75e-06, + "loss": 0.9604, + "step": 95 + }, + { + "epoch": 0.15031315240083507, + "grad_norm": 0.9323613047599792, + "learning_rate": 4.800000000000001e-06, + "loss": 0.9708, + "step": 96 + }, + { + "epoch": 0.15187891440501045, + "grad_norm": 0.9800642728805542, + "learning_rate": 4.85e-06, + "loss": 0.9868, + "step": 97 + }, + { + "epoch": 0.1534446764091858, + "grad_norm": 1.487841010093689, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8679, + "step": 98 + }, + { + "epoch": 0.15501043841336118, + "grad_norm": 0.95517897605896, + "learning_rate": 4.95e-06, + "loss": 0.7949, + "step": 99 + }, + { + "epoch": 0.15657620041753653, + "grad_norm": 0.9686336517333984, + "learning_rate": 5e-06, + "loss": 0.8655, + "step": 100 + }, + { + "epoch": 0.1581419624217119, + "grad_norm": 1.9481035470962524, + "learning_rate": 4.999999112316954e-06, + "loss": 0.9328, + "step": 101 + }, + { + "epoch": 0.15970772442588727, + "grad_norm": 1.272631287574768, + "learning_rate": 4.9999964492684465e-06, + "loss": 0.7869, + "step": 102 + }, + { + "epoch": 0.16127348643006262, + "grad_norm": 1.1117815971374512, + "learning_rate": 4.999992010856368e-06, + "loss": 0.949, + "step": 103 + }, + { + "epoch": 0.162839248434238, + "grad_norm": 0.9681039452552795, + "learning_rate": 4.999985797083871e-06, + "loss": 1.0396, + "step": 104 + }, + { + "epoch": 0.16440501043841335, + "grad_norm": 0.9797878265380859, + "learning_rate": 4.999977807955368e-06, + "loss": 0.9393, + "step": 105 + }, + { + "epoch": 0.16597077244258873, + "grad_norm": 1.0239158868789673, + "learning_rate": 4.999968043476532e-06, + "loss": 0.8766, + "step": 106 + }, + { + "epoch": 0.16753653444676408, + "grad_norm": 0.9314751029014587, + "learning_rate": 4.999956503654299e-06, + "loss": 0.8545, + "step": 107 + }, + { + "epoch": 0.16910229645093947, + "grad_norm": 1.1627179384231567, + "learning_rate": 4.999943188496862e-06, + "loss": 0.9241, + "step": 108 + }, + { + "epoch": 0.17066805845511482, + "grad_norm": 1.022752285003662, + "learning_rate": 4.9999280980136765e-06, + "loss": 0.9662, + "step": 109 + }, + { + "epoch": 0.1722338204592902, + "grad_norm": 1.0513653755187988, + "learning_rate": 4.99991123221546e-06, + "loss": 0.9054, + "step": 110 + }, + { + "epoch": 0.17379958246346555, + "grad_norm": 0.9488444328308105, + "learning_rate": 4.99989259111419e-06, + "loss": 0.8921, + "step": 111 + }, + { + "epoch": 0.17536534446764093, + "grad_norm": 0.9934154748916626, + "learning_rate": 4.999872174723104e-06, + "loss": 0.9848, + "step": 112 + }, + { + "epoch": 0.17693110647181629, + "grad_norm": 1.2035515308380127, + "learning_rate": 4.9998499830567e-06, + "loss": 0.9833, + "step": 113 + }, + { + "epoch": 0.17849686847599164, + "grad_norm": 1.027538537979126, + "learning_rate": 4.999826016130739e-06, + "loss": 1.0089, + "step": 114 + }, + { + "epoch": 0.18006263048016702, + "grad_norm": 1.0635300874710083, + "learning_rate": 4.999800273962238e-06, + "loss": 0.8863, + "step": 115 + }, + { + "epoch": 0.18162839248434237, + "grad_norm": 0.9978796243667603, + "learning_rate": 4.999772756569482e-06, + "loss": 0.899, + "step": 116 + }, + { + "epoch": 0.18319415448851775, + "grad_norm": 0.9971838593482971, + "learning_rate": 4.999743463972008e-06, + "loss": 0.9515, + "step": 117 + }, + { + "epoch": 0.1847599164926931, + "grad_norm": 1.2436699867248535, + "learning_rate": 4.999712396190622e-06, + "loss": 0.9101, + "step": 118 + }, + { + "epoch": 0.18632567849686849, + "grad_norm": 0.9573557376861572, + "learning_rate": 4.999679553247383e-06, + "loss": 0.8817, + "step": 119 + }, + { + "epoch": 0.18789144050104384, + "grad_norm": 0.9105668067932129, + "learning_rate": 4.9996449351656165e-06, + "loss": 1.0077, + "step": 120 + }, + { + "epoch": 0.18945720250521922, + "grad_norm": 0.8741400241851807, + "learning_rate": 4.999608541969905e-06, + "loss": 0.912, + "step": 121 + }, + { + "epoch": 0.19102296450939457, + "grad_norm": 1.1172688007354736, + "learning_rate": 4.999570373686095e-06, + "loss": 0.8062, + "step": 122 + }, + { + "epoch": 0.19258872651356992, + "grad_norm": 0.940539538860321, + "learning_rate": 4.9995304303412905e-06, + "loss": 0.8944, + "step": 123 + }, + { + "epoch": 0.1941544885177453, + "grad_norm": 2.1082208156585693, + "learning_rate": 4.999488711963857e-06, + "loss": 0.9668, + "step": 124 + }, + { + "epoch": 0.19572025052192066, + "grad_norm": 1.0358761548995972, + "learning_rate": 4.99944521858342e-06, + "loss": 0.9032, + "step": 125 + }, + { + "epoch": 0.19728601252609604, + "grad_norm": 2.0804238319396973, + "learning_rate": 4.999399950230867e-06, + "loss": 0.9819, + "step": 126 + }, + { + "epoch": 0.1988517745302714, + "grad_norm": 1.0986278057098389, + "learning_rate": 4.9993529069383465e-06, + "loss": 0.9034, + "step": 127 + }, + { + "epoch": 0.20041753653444677, + "grad_norm": 1.2374722957611084, + "learning_rate": 4.999304088739263e-06, + "loss": 0.9255, + "step": 128 + }, + { + "epoch": 0.20198329853862212, + "grad_norm": 1.0443146228790283, + "learning_rate": 4.999253495668287e-06, + "loss": 0.9352, + "step": 129 + }, + { + "epoch": 0.2035490605427975, + "grad_norm": 1.009800910949707, + "learning_rate": 4.999201127761346e-06, + "loss": 0.8324, + "step": 130 + }, + { + "epoch": 0.20511482254697286, + "grad_norm": 0.892861008644104, + "learning_rate": 4.9991469850556295e-06, + "loss": 0.886, + "step": 131 + }, + { + "epoch": 0.20668058455114824, + "grad_norm": 1.1675922870635986, + "learning_rate": 4.999091067589587e-06, + "loss": 0.967, + "step": 132 + }, + { + "epoch": 0.2082463465553236, + "grad_norm": 1.0169954299926758, + "learning_rate": 4.999033375402926e-06, + "loss": 0.8498, + "step": 133 + }, + { + "epoch": 0.20981210855949894, + "grad_norm": 1.0006227493286133, + "learning_rate": 4.99897390853662e-06, + "loss": 0.8481, + "step": 134 + }, + { + "epoch": 0.21137787056367432, + "grad_norm": 0.9049347639083862, + "learning_rate": 4.9989126670328945e-06, + "loss": 0.9186, + "step": 135 + }, + { + "epoch": 0.21294363256784968, + "grad_norm": 1.2151603698730469, + "learning_rate": 4.9988496509352444e-06, + "loss": 0.9067, + "step": 136 + }, + { + "epoch": 0.21450939457202506, + "grad_norm": 1.0112006664276123, + "learning_rate": 4.998784860288417e-06, + "loss": 0.8843, + "step": 137 + }, + { + "epoch": 0.2160751565762004, + "grad_norm": 0.9316662549972534, + "learning_rate": 4.9987182951384264e-06, + "loss": 0.8635, + "step": 138 + }, + { + "epoch": 0.2176409185803758, + "grad_norm": 0.942743718624115, + "learning_rate": 4.9986499555325406e-06, + "loss": 0.8481, + "step": 139 + }, + { + "epoch": 0.21920668058455114, + "grad_norm": 1.0298817157745361, + "learning_rate": 4.998579841519292e-06, + "loss": 0.865, + "step": 140 + }, + { + "epoch": 0.22077244258872653, + "grad_norm": 1.1556729078292847, + "learning_rate": 4.998507953148472e-06, + "loss": 0.8731, + "step": 141 + }, + { + "epoch": 0.22233820459290188, + "grad_norm": 0.997958779335022, + "learning_rate": 4.9984342904711315e-06, + "loss": 0.7949, + "step": 142 + }, + { + "epoch": 0.22390396659707723, + "grad_norm": 0.9565998315811157, + "learning_rate": 4.998358853539582e-06, + "loss": 0.8875, + "step": 143 + }, + { + "epoch": 0.2254697286012526, + "grad_norm": 0.9667196273803711, + "learning_rate": 4.998281642407394e-06, + "loss": 0.8537, + "step": 144 + }, + { + "epoch": 0.22703549060542796, + "grad_norm": 0.9701303839683533, + "learning_rate": 4.998202657129401e-06, + "loss": 0.873, + "step": 145 + }, + { + "epoch": 0.22860125260960334, + "grad_norm": 0.9936159253120422, + "learning_rate": 4.998121897761692e-06, + "loss": 0.8915, + "step": 146 + }, + { + "epoch": 0.2301670146137787, + "grad_norm": 1.5054311752319336, + "learning_rate": 4.998039364361617e-06, + "loss": 0.9944, + "step": 147 + }, + { + "epoch": 0.23173277661795408, + "grad_norm": 1.5895891189575195, + "learning_rate": 4.99795505698779e-06, + "loss": 0.9282, + "step": 148 + }, + { + "epoch": 0.23329853862212943, + "grad_norm": 1.0141469240188599, + "learning_rate": 4.9978689757000785e-06, + "loss": 0.9166, + "step": 149 + }, + { + "epoch": 0.2348643006263048, + "grad_norm": 1.1720311641693115, + "learning_rate": 4.997781120559615e-06, + "loss": 0.9404, + "step": 150 + }, + { + "epoch": 0.23643006263048016, + "grad_norm": 0.9536433815956116, + "learning_rate": 4.997691491628789e-06, + "loss": 0.9339, + "step": 151 + }, + { + "epoch": 0.23799582463465555, + "grad_norm": 0.9407872557640076, + "learning_rate": 4.997600088971249e-06, + "loss": 0.9019, + "step": 152 + }, + { + "epoch": 0.2395615866388309, + "grad_norm": 0.9202889204025269, + "learning_rate": 4.997506912651906e-06, + "loss": 0.9598, + "step": 153 + }, + { + "epoch": 0.24112734864300625, + "grad_norm": 1.0630513429641724, + "learning_rate": 4.997411962736927e-06, + "loss": 0.8554, + "step": 154 + }, + { + "epoch": 0.24269311064718163, + "grad_norm": 0.8378801941871643, + "learning_rate": 4.997315239293742e-06, + "loss": 0.8583, + "step": 155 + }, + { + "epoch": 0.24425887265135698, + "grad_norm": 0.9442815184593201, + "learning_rate": 4.997216742391038e-06, + "loss": 0.8518, + "step": 156 + }, + { + "epoch": 0.24582463465553236, + "grad_norm": 0.9273308515548706, + "learning_rate": 4.997116472098763e-06, + "loss": 0.8942, + "step": 157 + }, + { + "epoch": 0.24739039665970772, + "grad_norm": 0.9028997421264648, + "learning_rate": 4.997014428488121e-06, + "loss": 0.8592, + "step": 158 + }, + { + "epoch": 0.2489561586638831, + "grad_norm": 0.9877567887306213, + "learning_rate": 4.996910611631582e-06, + "loss": 0.9316, + "step": 159 + }, + { + "epoch": 0.25052192066805845, + "grad_norm": 0.961592435836792, + "learning_rate": 4.996805021602868e-06, + "loss": 0.8875, + "step": 160 + }, + { + "epoch": 0.2520876826722338, + "grad_norm": 0.9923007488250732, + "learning_rate": 4.996697658476965e-06, + "loss": 0.8554, + "step": 161 + }, + { + "epoch": 0.2536534446764092, + "grad_norm": 0.956763744354248, + "learning_rate": 4.996588522330114e-06, + "loss": 0.9381, + "step": 162 + }, + { + "epoch": 0.25521920668058456, + "grad_norm": 0.919992208480835, + "learning_rate": 4.996477613239822e-06, + "loss": 0.9016, + "step": 163 + }, + { + "epoch": 0.2567849686847599, + "grad_norm": 0.9655581712722778, + "learning_rate": 4.996364931284847e-06, + "loss": 0.896, + "step": 164 + }, + { + "epoch": 0.25835073068893527, + "grad_norm": 0.9159606099128723, + "learning_rate": 4.996250476545212e-06, + "loss": 0.9475, + "step": 165 + }, + { + "epoch": 0.2599164926931106, + "grad_norm": 1.1147626638412476, + "learning_rate": 4.996134249102195e-06, + "loss": 0.8793, + "step": 166 + }, + { + "epoch": 0.26148225469728603, + "grad_norm": 1.112069010734558, + "learning_rate": 4.996016249038334e-06, + "loss": 0.8765, + "step": 167 + }, + { + "epoch": 0.2630480167014614, + "grad_norm": 1.0366944074630737, + "learning_rate": 4.995896476437428e-06, + "loss": 0.9356, + "step": 168 + }, + { + "epoch": 0.26461377870563674, + "grad_norm": 0.9494239091873169, + "learning_rate": 4.995774931384534e-06, + "loss": 0.8246, + "step": 169 + }, + { + "epoch": 0.2661795407098121, + "grad_norm": 0.9819200038909912, + "learning_rate": 4.995651613965964e-06, + "loss": 0.9403, + "step": 170 + }, + { + "epoch": 0.2677453027139875, + "grad_norm": 1.7329354286193848, + "learning_rate": 4.995526524269293e-06, + "loss": 0.9091, + "step": 171 + }, + { + "epoch": 0.26931106471816285, + "grad_norm": 1.1415139436721802, + "learning_rate": 4.995399662383352e-06, + "loss": 0.9106, + "step": 172 + }, + { + "epoch": 0.2708768267223382, + "grad_norm": 0.9471906423568726, + "learning_rate": 4.995271028398233e-06, + "loss": 0.8858, + "step": 173 + }, + { + "epoch": 0.27244258872651356, + "grad_norm": 1.096053123474121, + "learning_rate": 4.995140622405284e-06, + "loss": 0.7502, + "step": 174 + }, + { + "epoch": 0.2740083507306889, + "grad_norm": 1.285410761833191, + "learning_rate": 4.995008444497113e-06, + "loss": 0.9288, + "step": 175 + }, + { + "epoch": 0.2755741127348643, + "grad_norm": 0.9714595079421997, + "learning_rate": 4.994874494767585e-06, + "loss": 0.8824, + "step": 176 + }, + { + "epoch": 0.27713987473903967, + "grad_norm": 0.8399011492729187, + "learning_rate": 4.994738773311824e-06, + "loss": 0.8703, + "step": 177 + }, + { + "epoch": 0.278705636743215, + "grad_norm": 0.9454224109649658, + "learning_rate": 4.994601280226212e-06, + "loss": 0.9627, + "step": 178 + }, + { + "epoch": 0.2802713987473904, + "grad_norm": 0.8983661532402039, + "learning_rate": 4.99446201560839e-06, + "loss": 0.9086, + "step": 179 + }, + { + "epoch": 0.2818371607515658, + "grad_norm": 0.9814153909683228, + "learning_rate": 4.994320979557256e-06, + "loss": 0.908, + "step": 180 + }, + { + "epoch": 0.28340292275574114, + "grad_norm": 0.8405355215072632, + "learning_rate": 4.9941781721729645e-06, + "loss": 0.8414, + "step": 181 + }, + { + "epoch": 0.2849686847599165, + "grad_norm": 0.950950026512146, + "learning_rate": 4.994033593556933e-06, + "loss": 0.9091, + "step": 182 + }, + { + "epoch": 0.28653444676409184, + "grad_norm": 1.1019747257232666, + "learning_rate": 4.99388724381183e-06, + "loss": 0.8305, + "step": 183 + }, + { + "epoch": 0.2881002087682672, + "grad_norm": 0.9339331984519958, + "learning_rate": 4.9937391230415875e-06, + "loss": 0.8542, + "step": 184 + }, + { + "epoch": 0.2896659707724426, + "grad_norm": 0.8813616633415222, + "learning_rate": 4.993589231351394e-06, + "loss": 0.848, + "step": 185 + }, + { + "epoch": 0.29123173277661796, + "grad_norm": 0.8916418552398682, + "learning_rate": 4.9934375688476905e-06, + "loss": 0.8032, + "step": 186 + }, + { + "epoch": 0.2927974947807933, + "grad_norm": 0.9891005754470825, + "learning_rate": 4.993284135638182e-06, + "loss": 0.7937, + "step": 187 + }, + { + "epoch": 0.29436325678496866, + "grad_norm": 0.9842082262039185, + "learning_rate": 4.99312893183183e-06, + "loss": 0.9532, + "step": 188 + }, + { + "epoch": 0.29592901878914407, + "grad_norm": 0.993478536605835, + "learning_rate": 4.99297195753885e-06, + "loss": 0.9406, + "step": 189 + }, + { + "epoch": 0.2974947807933194, + "grad_norm": 1.0654981136322021, + "learning_rate": 4.992813212870718e-06, + "loss": 0.907, + "step": 190 + }, + { + "epoch": 0.2990605427974948, + "grad_norm": 0.8937190771102905, + "learning_rate": 4.992652697940164e-06, + "loss": 0.9307, + "step": 191 + }, + { + "epoch": 0.30062630480167013, + "grad_norm": 0.9006052613258362, + "learning_rate": 4.992490412861178e-06, + "loss": 0.8808, + "step": 192 + }, + { + "epoch": 0.30219206680584554, + "grad_norm": 0.9601522088050842, + "learning_rate": 4.992326357749007e-06, + "loss": 0.8495, + "step": 193 + }, + { + "epoch": 0.3037578288100209, + "grad_norm": 0.9814243316650391, + "learning_rate": 4.992160532720153e-06, + "loss": 0.8204, + "step": 194 + }, + { + "epoch": 0.30532359081419624, + "grad_norm": 0.9660456776618958, + "learning_rate": 4.991992937892377e-06, + "loss": 0.8209, + "step": 195 + }, + { + "epoch": 0.3068893528183716, + "grad_norm": 0.8504270315170288, + "learning_rate": 4.991823573384695e-06, + "loss": 0.8415, + "step": 196 + }, + { + "epoch": 0.30845511482254695, + "grad_norm": 0.8733190894126892, + "learning_rate": 4.991652439317382e-06, + "loss": 0.8753, + "step": 197 + }, + { + "epoch": 0.31002087682672236, + "grad_norm": 0.9761794805526733, + "learning_rate": 4.991479535811967e-06, + "loss": 0.8923, + "step": 198 + }, + { + "epoch": 0.3115866388308977, + "grad_norm": 1.0400285720825195, + "learning_rate": 4.991304862991238e-06, + "loss": 0.899, + "step": 199 + }, + { + "epoch": 0.31315240083507306, + "grad_norm": 0.8561192750930786, + "learning_rate": 4.991128420979237e-06, + "loss": 0.9117, + "step": 200 + }, + { + "epoch": 0.3147181628392484, + "grad_norm": 0.8839634656906128, + "learning_rate": 4.990950209901263e-06, + "loss": 0.9022, + "step": 201 + }, + { + "epoch": 0.3162839248434238, + "grad_norm": 0.9569464921951294, + "learning_rate": 4.990770229883873e-06, + "loss": 0.9013, + "step": 202 + }, + { + "epoch": 0.3178496868475992, + "grad_norm": 1.0600861310958862, + "learning_rate": 4.990588481054881e-06, + "loss": 0.8387, + "step": 203 + }, + { + "epoch": 0.31941544885177453, + "grad_norm": 0.9492761492729187, + "learning_rate": 4.990404963543352e-06, + "loss": 0.8521, + "step": 204 + }, + { + "epoch": 0.3209812108559499, + "grad_norm": 0.9534817934036255, + "learning_rate": 4.990219677479612e-06, + "loss": 0.8009, + "step": 205 + }, + { + "epoch": 0.32254697286012524, + "grad_norm": 0.9918960928916931, + "learning_rate": 4.990032622995242e-06, + "loss": 0.8215, + "step": 206 + }, + { + "epoch": 0.32411273486430064, + "grad_norm": 0.8751630783081055, + "learning_rate": 4.989843800223077e-06, + "loss": 0.8744, + "step": 207 + }, + { + "epoch": 0.325678496868476, + "grad_norm": 1.0034785270690918, + "learning_rate": 4.9896532092972085e-06, + "loss": 0.884, + "step": 208 + }, + { + "epoch": 0.32724425887265135, + "grad_norm": 1.2693432569503784, + "learning_rate": 4.989460850352984e-06, + "loss": 0.9689, + "step": 209 + }, + { + "epoch": 0.3288100208768267, + "grad_norm": 0.9923811554908752, + "learning_rate": 4.989266723527007e-06, + "loss": 0.9622, + "step": 210 + }, + { + "epoch": 0.3303757828810021, + "grad_norm": 0.9439074397087097, + "learning_rate": 4.989070828957137e-06, + "loss": 0.9155, + "step": 211 + }, + { + "epoch": 0.33194154488517746, + "grad_norm": 1.2564488649368286, + "learning_rate": 4.988873166782485e-06, + "loss": 0.8712, + "step": 212 + }, + { + "epoch": 0.3335073068893528, + "grad_norm": 1.1441881656646729, + "learning_rate": 4.988673737143423e-06, + "loss": 0.8895, + "step": 213 + }, + { + "epoch": 0.33507306889352817, + "grad_norm": 0.972155749797821, + "learning_rate": 4.988472540181574e-06, + "loss": 0.9158, + "step": 214 + }, + { + "epoch": 0.3366388308977035, + "grad_norm": 0.9536120295524597, + "learning_rate": 4.988269576039817e-06, + "loss": 0.7963, + "step": 215 + }, + { + "epoch": 0.33820459290187893, + "grad_norm": 0.8956594467163086, + "learning_rate": 4.988064844862288e-06, + "loss": 0.8947, + "step": 216 + }, + { + "epoch": 0.3397703549060543, + "grad_norm": 0.9503691792488098, + "learning_rate": 4.987858346794373e-06, + "loss": 0.7857, + "step": 217 + }, + { + "epoch": 0.34133611691022964, + "grad_norm": 0.9471456408500671, + "learning_rate": 4.987650081982719e-06, + "loss": 0.9118, + "step": 218 + }, + { + "epoch": 0.342901878914405, + "grad_norm": 0.9485713839530945, + "learning_rate": 4.987440050575222e-06, + "loss": 0.8188, + "step": 219 + }, + { + "epoch": 0.3444676409185804, + "grad_norm": 1.0775927305221558, + "learning_rate": 4.987228252721037e-06, + "loss": 0.9469, + "step": 220 + }, + { + "epoch": 0.34603340292275575, + "grad_norm": 0.9173988103866577, + "learning_rate": 4.9870146885705705e-06, + "loss": 0.8566, + "step": 221 + }, + { + "epoch": 0.3475991649269311, + "grad_norm": 1.008735179901123, + "learning_rate": 4.986799358275485e-06, + "loss": 0.8637, + "step": 222 + }, + { + "epoch": 0.34916492693110646, + "grad_norm": 0.8183198571205139, + "learning_rate": 4.986582261988695e-06, + "loss": 0.9321, + "step": 223 + }, + { + "epoch": 0.35073068893528186, + "grad_norm": 0.8951883316040039, + "learning_rate": 4.986363399864372e-06, + "loss": 0.8015, + "step": 224 + }, + { + "epoch": 0.3522964509394572, + "grad_norm": 0.9275704622268677, + "learning_rate": 4.986142772057941e-06, + "loss": 0.8242, + "step": 225 + }, + { + "epoch": 0.35386221294363257, + "grad_norm": 0.9548516273498535, + "learning_rate": 4.985920378726077e-06, + "loss": 0.7889, + "step": 226 + }, + { + "epoch": 0.3554279749478079, + "grad_norm": 1.1203585863113403, + "learning_rate": 4.985696220026714e-06, + "loss": 0.835, + "step": 227 + }, + { + "epoch": 0.3569937369519833, + "grad_norm": 0.9525771141052246, + "learning_rate": 4.985470296119038e-06, + "loss": 0.9162, + "step": 228 + }, + { + "epoch": 0.3585594989561587, + "grad_norm": 1.150404930114746, + "learning_rate": 4.985242607163488e-06, + "loss": 0.9456, + "step": 229 + }, + { + "epoch": 0.36012526096033404, + "grad_norm": 1.01652991771698, + "learning_rate": 4.985013153321755e-06, + "loss": 0.87, + "step": 230 + }, + { + "epoch": 0.3616910229645094, + "grad_norm": 0.9455218315124512, + "learning_rate": 4.984781934756786e-06, + "loss": 0.8466, + "step": 231 + }, + { + "epoch": 0.36325678496868474, + "grad_norm": 0.9125218391418457, + "learning_rate": 4.984548951632779e-06, + "loss": 0.8764, + "step": 232 + }, + { + "epoch": 0.36482254697286015, + "grad_norm": 2.0192980766296387, + "learning_rate": 4.984314204115187e-06, + "loss": 0.7866, + "step": 233 + }, + { + "epoch": 0.3663883089770355, + "grad_norm": 1.8118515014648438, + "learning_rate": 4.984077692370716e-06, + "loss": 0.8703, + "step": 234 + }, + { + "epoch": 0.36795407098121086, + "grad_norm": 1.2995024919509888, + "learning_rate": 4.983839416567322e-06, + "loss": 0.9459, + "step": 235 + }, + { + "epoch": 0.3695198329853862, + "grad_norm": 1.0611528158187866, + "learning_rate": 4.983599376874216e-06, + "loss": 0.8456, + "step": 236 + }, + { + "epoch": 0.37108559498956156, + "grad_norm": 1.2018043994903564, + "learning_rate": 4.983357573461863e-06, + "loss": 0.914, + "step": 237 + }, + { + "epoch": 0.37265135699373697, + "grad_norm": 0.9091911315917969, + "learning_rate": 4.983114006501978e-06, + "loss": 0.791, + "step": 238 + }, + { + "epoch": 0.3742171189979123, + "grad_norm": 0.980873167514801, + "learning_rate": 4.982868676167528e-06, + "loss": 0.77, + "step": 239 + }, + { + "epoch": 0.3757828810020877, + "grad_norm": 0.8945695161819458, + "learning_rate": 4.982621582632735e-06, + "loss": 0.8038, + "step": 240 + }, + { + "epoch": 0.37734864300626303, + "grad_norm": 0.9325360655784607, + "learning_rate": 4.9823727260730705e-06, + "loss": 0.8908, + "step": 241 + }, + { + "epoch": 0.37891440501043844, + "grad_norm": 0.987630307674408, + "learning_rate": 4.98212210666526e-06, + "loss": 0.8642, + "step": 242 + }, + { + "epoch": 0.3804801670146138, + "grad_norm": 0.9151357412338257, + "learning_rate": 4.98186972458728e-06, + "loss": 0.8437, + "step": 243 + }, + { + "epoch": 0.38204592901878914, + "grad_norm": 0.9084850549697876, + "learning_rate": 4.981615580018358e-06, + "loss": 0.8775, + "step": 244 + }, + { + "epoch": 0.3836116910229645, + "grad_norm": 0.9786385297775269, + "learning_rate": 4.981359673138974e-06, + "loss": 0.8251, + "step": 245 + }, + { + "epoch": 0.38517745302713985, + "grad_norm": 0.8947877883911133, + "learning_rate": 4.98110200413086e-06, + "loss": 0.7709, + "step": 246 + }, + { + "epoch": 0.38674321503131526, + "grad_norm": 0.967019259929657, + "learning_rate": 4.9808425731769985e-06, + "loss": 0.8954, + "step": 247 + }, + { + "epoch": 0.3883089770354906, + "grad_norm": 1.1066704988479614, + "learning_rate": 4.980581380461622e-06, + "loss": 0.8544, + "step": 248 + }, + { + "epoch": 0.38987473903966596, + "grad_norm": 0.9750981330871582, + "learning_rate": 4.980318426170218e-06, + "loss": 0.8354, + "step": 249 + }, + { + "epoch": 0.3914405010438413, + "grad_norm": 0.8727232217788696, + "learning_rate": 4.980053710489521e-06, + "loss": 0.8911, + "step": 250 + }, + { + "epoch": 0.3930062630480167, + "grad_norm": 0.9337188601493835, + "learning_rate": 4.979787233607518e-06, + "loss": 0.8567, + "step": 251 + }, + { + "epoch": 0.3945720250521921, + "grad_norm": 0.890701174736023, + "learning_rate": 4.979518995713448e-06, + "loss": 0.8327, + "step": 252 + }, + { + "epoch": 0.39613778705636743, + "grad_norm": 0.9973195195198059, + "learning_rate": 4.979248996997797e-06, + "loss": 0.9065, + "step": 253 + }, + { + "epoch": 0.3977035490605428, + "grad_norm": 0.9894055128097534, + "learning_rate": 4.978977237652306e-06, + "loss": 0.8729, + "step": 254 + }, + { + "epoch": 0.39926931106471814, + "grad_norm": 1.0398536920547485, + "learning_rate": 4.978703717869961e-06, + "loss": 0.861, + "step": 255 + }, + { + "epoch": 0.40083507306889354, + "grad_norm": 1.1713942289352417, + "learning_rate": 4.978428437845003e-06, + "loss": 0.8183, + "step": 256 + }, + { + "epoch": 0.4024008350730689, + "grad_norm": 0.9868776202201843, + "learning_rate": 4.9781513977729225e-06, + "loss": 0.941, + "step": 257 + }, + { + "epoch": 0.40396659707724425, + "grad_norm": 0.9260863065719604, + "learning_rate": 4.977872597850455e-06, + "loss": 0.8963, + "step": 258 + }, + { + "epoch": 0.4055323590814196, + "grad_norm": 0.9553239941596985, + "learning_rate": 4.977592038275592e-06, + "loss": 0.8102, + "step": 259 + }, + { + "epoch": 0.407098121085595, + "grad_norm": 0.9426907300949097, + "learning_rate": 4.977309719247571e-06, + "loss": 0.89, + "step": 260 + }, + { + "epoch": 0.40866388308977036, + "grad_norm": 0.9691527485847473, + "learning_rate": 4.97702564096688e-06, + "loss": 0.7967, + "step": 261 + }, + { + "epoch": 0.4102296450939457, + "grad_norm": 0.9917487502098083, + "learning_rate": 4.976739803635256e-06, + "loss": 0.8327, + "step": 262 + }, + { + "epoch": 0.41179540709812107, + "grad_norm": 0.992340087890625, + "learning_rate": 4.976452207455686e-06, + "loss": 0.9462, + "step": 263 + }, + { + "epoch": 0.4133611691022965, + "grad_norm": 0.9554483890533447, + "learning_rate": 4.9761628526324035e-06, + "loss": 0.8294, + "step": 264 + }, + { + "epoch": 0.41492693110647183, + "grad_norm": 0.9318789839744568, + "learning_rate": 4.975871739370895e-06, + "loss": 0.8079, + "step": 265 + }, + { + "epoch": 0.4164926931106472, + "grad_norm": 0.9158215522766113, + "learning_rate": 4.975578867877893e-06, + "loss": 0.8632, + "step": 266 + }, + { + "epoch": 0.41805845511482254, + "grad_norm": 0.9444370865821838, + "learning_rate": 4.975284238361378e-06, + "loss": 0.8897, + "step": 267 + }, + { + "epoch": 0.4196242171189979, + "grad_norm": 0.9949231743812561, + "learning_rate": 4.974987851030581e-06, + "loss": 0.7905, + "step": 268 + }, + { + "epoch": 0.4211899791231733, + "grad_norm": 0.9819440245628357, + "learning_rate": 4.974689706095981e-06, + "loss": 0.9183, + "step": 269 + }, + { + "epoch": 0.42275574112734865, + "grad_norm": 1.0787206888198853, + "learning_rate": 4.974389803769304e-06, + "loss": 0.7739, + "step": 270 + }, + { + "epoch": 0.424321503131524, + "grad_norm": 0.8871361017227173, + "learning_rate": 4.9740881442635235e-06, + "loss": 0.9454, + "step": 271 + }, + { + "epoch": 0.42588726513569936, + "grad_norm": 0.9125893115997314, + "learning_rate": 4.973784727792863e-06, + "loss": 0.8017, + "step": 272 + }, + { + "epoch": 0.42745302713987476, + "grad_norm": 0.9102838635444641, + "learning_rate": 4.973479554572793e-06, + "loss": 0.8307, + "step": 273 + }, + { + "epoch": 0.4290187891440501, + "grad_norm": 0.8780474662780762, + "learning_rate": 4.973172624820031e-06, + "loss": 0.9402, + "step": 274 + }, + { + "epoch": 0.43058455114822547, + "grad_norm": 1.0059378147125244, + "learning_rate": 4.972863938752541e-06, + "loss": 0.7438, + "step": 275 + }, + { + "epoch": 0.4321503131524008, + "grad_norm": 0.9684615135192871, + "learning_rate": 4.972553496589537e-06, + "loss": 0.8897, + "step": 276 + }, + { + "epoch": 0.4337160751565762, + "grad_norm": 0.9615057110786438, + "learning_rate": 4.972241298551477e-06, + "loss": 0.8157, + "step": 277 + }, + { + "epoch": 0.4352818371607516, + "grad_norm": 0.8589702844619751, + "learning_rate": 4.971927344860068e-06, + "loss": 0.7813, + "step": 278 + }, + { + "epoch": 0.43684759916492694, + "grad_norm": 0.900324285030365, + "learning_rate": 4.971611635738263e-06, + "loss": 0.7534, + "step": 279 + }, + { + "epoch": 0.4384133611691023, + "grad_norm": 0.8881833553314209, + "learning_rate": 4.971294171410261e-06, + "loss": 0.8255, + "step": 280 + }, + { + "epoch": 0.43997912317327764, + "grad_norm": 0.9723751544952393, + "learning_rate": 4.9709749521015085e-06, + "loss": 0.8796, + "step": 281 + }, + { + "epoch": 0.44154488517745305, + "grad_norm": 0.9341713190078735, + "learning_rate": 4.9706539780386995e-06, + "loss": 0.7429, + "step": 282 + }, + { + "epoch": 0.4431106471816284, + "grad_norm": 0.9354205131530762, + "learning_rate": 4.97033124944977e-06, + "loss": 0.8545, + "step": 283 + }, + { + "epoch": 0.44467640918580376, + "grad_norm": 0.9294624328613281, + "learning_rate": 4.970006766563906e-06, + "loss": 0.7783, + "step": 284 + }, + { + "epoch": 0.4462421711899791, + "grad_norm": 0.9584887027740479, + "learning_rate": 4.969680529611538e-06, + "loss": 0.8068, + "step": 285 + }, + { + "epoch": 0.44780793319415446, + "grad_norm": 0.9525735378265381, + "learning_rate": 4.9693525388243424e-06, + "loss": 0.8055, + "step": 286 + }, + { + "epoch": 0.44937369519832987, + "grad_norm": 0.8845317959785461, + "learning_rate": 4.96902279443524e-06, + "loss": 0.8318, + "step": 287 + }, + { + "epoch": 0.4509394572025052, + "grad_norm": 0.8932276964187622, + "learning_rate": 4.9686912966783965e-06, + "loss": 0.8185, + "step": 288 + }, + { + "epoch": 0.4525052192066806, + "grad_norm": 0.9881951808929443, + "learning_rate": 4.968358045789225e-06, + "loss": 0.9001, + "step": 289 + }, + { + "epoch": 0.45407098121085593, + "grad_norm": 0.995212197303772, + "learning_rate": 4.968023042004384e-06, + "loss": 0.75, + "step": 290 + }, + { + "epoch": 0.45563674321503134, + "grad_norm": 0.9997846484184265, + "learning_rate": 4.967686285561773e-06, + "loss": 0.8993, + "step": 291 + }, + { + "epoch": 0.4572025052192067, + "grad_norm": 0.9438545107841492, + "learning_rate": 4.967347776700538e-06, + "loss": 0.8964, + "step": 292 + }, + { + "epoch": 0.45876826722338204, + "grad_norm": 0.9164395332336426, + "learning_rate": 4.967007515661072e-06, + "loss": 0.8671, + "step": 293 + }, + { + "epoch": 0.4603340292275574, + "grad_norm": 0.895302951335907, + "learning_rate": 4.966665502685009e-06, + "loss": 0.8785, + "step": 294 + }, + { + "epoch": 0.4618997912317328, + "grad_norm": 0.9122863411903381, + "learning_rate": 4.966321738015227e-06, + "loss": 0.8499, + "step": 295 + }, + { + "epoch": 0.46346555323590816, + "grad_norm": 1.0097336769104004, + "learning_rate": 4.965976221895852e-06, + "loss": 0.83, + "step": 296 + }, + { + "epoch": 0.4650313152400835, + "grad_norm": 1.0664499998092651, + "learning_rate": 4.965628954572249e-06, + "loss": 0.7799, + "step": 297 + }, + { + "epoch": 0.46659707724425886, + "grad_norm": 0.9602341055870056, + "learning_rate": 4.96527993629103e-06, + "loss": 0.8061, + "step": 298 + }, + { + "epoch": 0.4681628392484342, + "grad_norm": 1.0133620500564575, + "learning_rate": 4.964929167300048e-06, + "loss": 0.9148, + "step": 299 + }, + { + "epoch": 0.4697286012526096, + "grad_norm": 0.9351868033409119, + "learning_rate": 4.964576647848401e-06, + "loss": 0.83, + "step": 300 + }, + { + "epoch": 0.471294363256785, + "grad_norm": 1.0078860521316528, + "learning_rate": 4.964222378186429e-06, + "loss": 0.8391, + "step": 301 + }, + { + "epoch": 0.47286012526096033, + "grad_norm": 0.8966493606567383, + "learning_rate": 4.963866358565716e-06, + "loss": 0.8677, + "step": 302 + }, + { + "epoch": 0.4744258872651357, + "grad_norm": 0.9283657670021057, + "learning_rate": 4.963508589239087e-06, + "loss": 0.8787, + "step": 303 + }, + { + "epoch": 0.4759916492693111, + "grad_norm": 0.9540023803710938, + "learning_rate": 4.963149070460611e-06, + "loss": 0.8667, + "step": 304 + }, + { + "epoch": 0.47755741127348644, + "grad_norm": 0.9580283164978027, + "learning_rate": 4.9627878024855995e-06, + "loss": 0.8442, + "step": 305 + }, + { + "epoch": 0.4791231732776618, + "grad_norm": 0.9008278846740723, + "learning_rate": 4.962424785570605e-06, + "loss": 0.8605, + "step": 306 + }, + { + "epoch": 0.48068893528183715, + "grad_norm": 0.9681883454322815, + "learning_rate": 4.962060019973423e-06, + "loss": 0.7865, + "step": 307 + }, + { + "epoch": 0.4822546972860125, + "grad_norm": 0.8492597341537476, + "learning_rate": 4.9616935059530915e-06, + "loss": 0.8578, + "step": 308 + }, + { + "epoch": 0.4838204592901879, + "grad_norm": 0.9323164820671082, + "learning_rate": 4.9613252437698865e-06, + "loss": 0.8491, + "step": 309 + }, + { + "epoch": 0.48538622129436326, + "grad_norm": 0.9417833685874939, + "learning_rate": 4.960955233685331e-06, + "loss": 0.8108, + "step": 310 + }, + { + "epoch": 0.4869519832985386, + "grad_norm": 0.901364803314209, + "learning_rate": 4.960583475962185e-06, + "loss": 0.8958, + "step": 311 + }, + { + "epoch": 0.48851774530271397, + "grad_norm": 0.8929979205131531, + "learning_rate": 4.96020997086445e-06, + "loss": 0.8237, + "step": 312 + }, + { + "epoch": 0.4900835073068894, + "grad_norm": 0.9422337412834167, + "learning_rate": 4.9598347186573716e-06, + "loss": 0.8024, + "step": 313 + }, + { + "epoch": 0.49164926931106473, + "grad_norm": 0.9165504574775696, + "learning_rate": 4.959457719607432e-06, + "loss": 0.8124, + "step": 314 + }, + { + "epoch": 0.4932150313152401, + "grad_norm": 1.0233370065689087, + "learning_rate": 4.959078973982356e-06, + "loss": 0.788, + "step": 315 + }, + { + "epoch": 0.49478079331941544, + "grad_norm": 0.9375972151756287, + "learning_rate": 4.958698482051109e-06, + "loss": 0.839, + "step": 316 + }, + { + "epoch": 0.4963465553235908, + "grad_norm": 0.8805437684059143, + "learning_rate": 4.958316244083895e-06, + "loss": 0.7445, + "step": 317 + }, + { + "epoch": 0.4979123173277662, + "grad_norm": 1.0231893062591553, + "learning_rate": 4.95793226035216e-06, + "loss": 0.8257, + "step": 318 + }, + { + "epoch": 0.49947807933194155, + "grad_norm": 0.9125928282737732, + "learning_rate": 4.957546531128589e-06, + "loss": 0.7699, + "step": 319 + }, + { + "epoch": 0.5010438413361169, + "grad_norm": 0.8844016790390015, + "learning_rate": 4.957159056687104e-06, + "loss": 0.8784, + "step": 320 + }, + { + "epoch": 0.5026096033402923, + "grad_norm": 0.9770572781562805, + "learning_rate": 4.956769837302871e-06, + "loss": 0.8611, + "step": 321 + }, + { + "epoch": 0.5041753653444676, + "grad_norm": 0.9413725733757019, + "learning_rate": 4.956378873252293e-06, + "loss": 0.8056, + "step": 322 + }, + { + "epoch": 0.505741127348643, + "grad_norm": 0.9058850407600403, + "learning_rate": 4.955986164813008e-06, + "loss": 0.8162, + "step": 323 + }, + { + "epoch": 0.5073068893528184, + "grad_norm": 0.893100917339325, + "learning_rate": 4.955591712263901e-06, + "loss": 0.7554, + "step": 324 + }, + { + "epoch": 0.5088726513569938, + "grad_norm": 0.8967075943946838, + "learning_rate": 4.955195515885089e-06, + "loss": 0.8143, + "step": 325 + }, + { + "epoch": 0.5104384133611691, + "grad_norm": 0.9542785286903381, + "learning_rate": 4.954797575957929e-06, + "loss": 0.8102, + "step": 326 + }, + { + "epoch": 0.5120041753653445, + "grad_norm": 1.006451964378357, + "learning_rate": 4.954397892765018e-06, + "loss": 0.86, + "step": 327 + }, + { + "epoch": 0.5135699373695198, + "grad_norm": 0.9964715838432312, + "learning_rate": 4.953996466590189e-06, + "loss": 0.7166, + "step": 328 + }, + { + "epoch": 0.5151356993736952, + "grad_norm": 0.9568808674812317, + "learning_rate": 4.953593297718512e-06, + "loss": 0.8753, + "step": 329 + }, + { + "epoch": 0.5167014613778705, + "grad_norm": 0.9435825347900391, + "learning_rate": 4.953188386436298e-06, + "loss": 0.8961, + "step": 330 + }, + { + "epoch": 0.5182672233820459, + "grad_norm": 0.9386457800865173, + "learning_rate": 4.952781733031092e-06, + "loss": 0.8468, + "step": 331 + }, + { + "epoch": 0.5198329853862212, + "grad_norm": 0.9916554689407349, + "learning_rate": 4.952373337791678e-06, + "loss": 0.8274, + "step": 332 + }, + { + "epoch": 0.5213987473903967, + "grad_norm": 0.9351572394371033, + "learning_rate": 4.9519632010080765e-06, + "loss": 0.7964, + "step": 333 + }, + { + "epoch": 0.5229645093945721, + "grad_norm": 0.9936269521713257, + "learning_rate": 4.951551322971545e-06, + "loss": 0.8354, + "step": 334 + }, + { + "epoch": 0.5245302713987474, + "grad_norm": 0.9678221940994263, + "learning_rate": 4.951137703974576e-06, + "loss": 0.9115, + "step": 335 + }, + { + "epoch": 0.5260960334029228, + "grad_norm": 0.9453133940696716, + "learning_rate": 4.9507223443109005e-06, + "loss": 0.7725, + "step": 336 + }, + { + "epoch": 0.5276617954070981, + "grad_norm": 0.9577226638793945, + "learning_rate": 4.950305244275484e-06, + "loss": 0.8679, + "step": 337 + }, + { + "epoch": 0.5292275574112735, + "grad_norm": 0.998489260673523, + "learning_rate": 4.94988640416453e-06, + "loss": 0.7505, + "step": 338 + }, + { + "epoch": 0.5307933194154488, + "grad_norm": 0.976026713848114, + "learning_rate": 4.949465824275475e-06, + "loss": 0.9083, + "step": 339 + }, + { + "epoch": 0.5323590814196242, + "grad_norm": 1.0041502714157104, + "learning_rate": 4.9490435049069925e-06, + "loss": 0.8815, + "step": 340 + }, + { + "epoch": 0.5339248434237995, + "grad_norm": 0.9839801788330078, + "learning_rate": 4.948619446358991e-06, + "loss": 0.8517, + "step": 341 + }, + { + "epoch": 0.535490605427975, + "grad_norm": 0.9995140433311462, + "learning_rate": 4.948193648932616e-06, + "loss": 0.8157, + "step": 342 + }, + { + "epoch": 0.5370563674321504, + "grad_norm": 1.0782225131988525, + "learning_rate": 4.947766112930243e-06, + "loss": 0.9206, + "step": 343 + }, + { + "epoch": 0.5386221294363257, + "grad_norm": 0.9981802105903625, + "learning_rate": 4.947336838655487e-06, + "loss": 0.8389, + "step": 344 + }, + { + "epoch": 0.5401878914405011, + "grad_norm": 0.9261236190795898, + "learning_rate": 4.946905826413195e-06, + "loss": 0.8035, + "step": 345 + }, + { + "epoch": 0.5417536534446764, + "grad_norm": 0.9195284247398376, + "learning_rate": 4.946473076509449e-06, + "loss": 0.8265, + "step": 346 + }, + { + "epoch": 0.5433194154488518, + "grad_norm": 0.9535123109817505, + "learning_rate": 4.946038589251566e-06, + "loss": 0.7528, + "step": 347 + }, + { + "epoch": 0.5448851774530271, + "grad_norm": 0.9969532489776611, + "learning_rate": 4.9456023649480935e-06, + "loss": 0.7787, + "step": 348 + }, + { + "epoch": 0.5464509394572025, + "grad_norm": 0.9612399935722351, + "learning_rate": 4.945164403908816e-06, + "loss": 0.8298, + "step": 349 + }, + { + "epoch": 0.5480167014613778, + "grad_norm": 0.9541627764701843, + "learning_rate": 4.944724706444749e-06, + "loss": 0.8039, + "step": 350 + }, + { + "epoch": 0.5495824634655533, + "grad_norm": 0.9916557669639587, + "learning_rate": 4.944283272868143e-06, + "loss": 0.9012, + "step": 351 + }, + { + "epoch": 0.5511482254697286, + "grad_norm": 0.8807661533355713, + "learning_rate": 4.94384010349248e-06, + "loss": 0.9005, + "step": 352 + }, + { + "epoch": 0.552713987473904, + "grad_norm": 1.0179708003997803, + "learning_rate": 4.943395198632476e-06, + "loss": 0.8845, + "step": 353 + }, + { + "epoch": 0.5542797494780793, + "grad_norm": 0.923014223575592, + "learning_rate": 4.942948558604077e-06, + "loss": 0.8173, + "step": 354 + }, + { + "epoch": 0.5558455114822547, + "grad_norm": 0.9318749308586121, + "learning_rate": 4.942500183724463e-06, + "loss": 0.8828, + "step": 355 + }, + { + "epoch": 0.55741127348643, + "grad_norm": 1.0907000303268433, + "learning_rate": 4.942050074312048e-06, + "loss": 0.7951, + "step": 356 + }, + { + "epoch": 0.5589770354906054, + "grad_norm": 0.9811620712280273, + "learning_rate": 4.9415982306864744e-06, + "loss": 0.8826, + "step": 357 + }, + { + "epoch": 0.5605427974947808, + "grad_norm": 0.9888150691986084, + "learning_rate": 4.9411446531686156e-06, + "loss": 0.8338, + "step": 358 + }, + { + "epoch": 0.5621085594989561, + "grad_norm": 0.9389257431030273, + "learning_rate": 4.9406893420805804e-06, + "loss": 0.8316, + "step": 359 + }, + { + "epoch": 0.5636743215031316, + "grad_norm": 0.9500603675842285, + "learning_rate": 4.940232297745705e-06, + "loss": 0.8764, + "step": 360 + }, + { + "epoch": 0.5652400835073069, + "grad_norm": 0.9748769998550415, + "learning_rate": 4.939773520488559e-06, + "loss": 0.8308, + "step": 361 + }, + { + "epoch": 0.5668058455114823, + "grad_norm": 0.9179681539535522, + "learning_rate": 4.93931301063494e-06, + "loss": 0.789, + "step": 362 + }, + { + "epoch": 0.5683716075156576, + "grad_norm": 0.8609325289726257, + "learning_rate": 4.938850768511878e-06, + "loss": 0.7368, + "step": 363 + }, + { + "epoch": 0.569937369519833, + "grad_norm": 0.9375531673431396, + "learning_rate": 4.9383867944476325e-06, + "loss": 0.8164, + "step": 364 + }, + { + "epoch": 0.5715031315240083, + "grad_norm": 0.9386453032493591, + "learning_rate": 4.937921088771694e-06, + "loss": 0.7898, + "step": 365 + }, + { + "epoch": 0.5730688935281837, + "grad_norm": 1.025040864944458, + "learning_rate": 4.937453651814781e-06, + "loss": 0.8465, + "step": 366 + }, + { + "epoch": 0.574634655532359, + "grad_norm": 1.1037245988845825, + "learning_rate": 4.936984483908842e-06, + "loss": 0.7935, + "step": 367 + }, + { + "epoch": 0.5762004175365344, + "grad_norm": 1.230821132659912, + "learning_rate": 4.936513585387055e-06, + "loss": 0.8043, + "step": 368 + }, + { + "epoch": 0.5777661795407099, + "grad_norm": 1.090235948562622, + "learning_rate": 4.9360409565838265e-06, + "loss": 0.8726, + "step": 369 + }, + { + "epoch": 0.5793319415448852, + "grad_norm": 0.8977810144424438, + "learning_rate": 4.935566597834793e-06, + "loss": 0.8667, + "step": 370 + }, + { + "epoch": 0.5808977035490606, + "grad_norm": 0.96401447057724, + "learning_rate": 4.935090509476818e-06, + "loss": 0.8494, + "step": 371 + }, + { + "epoch": 0.5824634655532359, + "grad_norm": 0.9290149211883545, + "learning_rate": 4.934612691847995e-06, + "loss": 0.7678, + "step": 372 + }, + { + "epoch": 0.5840292275574113, + "grad_norm": 0.9765584468841553, + "learning_rate": 4.934133145287644e-06, + "loss": 0.7947, + "step": 373 + }, + { + "epoch": 0.5855949895615866, + "grad_norm": 0.8939430117607117, + "learning_rate": 4.933651870136313e-06, + "loss": 0.885, + "step": 374 + }, + { + "epoch": 0.587160751565762, + "grad_norm": 0.9512760043144226, + "learning_rate": 4.933168866735777e-06, + "loss": 0.8863, + "step": 375 + }, + { + "epoch": 0.5887265135699373, + "grad_norm": 1.0334441661834717, + "learning_rate": 4.932684135429041e-06, + "loss": 0.8362, + "step": 376 + }, + { + "epoch": 0.5902922755741128, + "grad_norm": 0.9402525424957275, + "learning_rate": 4.932197676560334e-06, + "loss": 0.8758, + "step": 377 + }, + { + "epoch": 0.5918580375782881, + "grad_norm": 0.9615777730941772, + "learning_rate": 4.931709490475113e-06, + "loss": 0.8324, + "step": 378 + }, + { + "epoch": 0.5934237995824635, + "grad_norm": 0.9407138228416443, + "learning_rate": 4.931219577520061e-06, + "loss": 0.8454, + "step": 379 + }, + { + "epoch": 0.5949895615866388, + "grad_norm": 1.0415815114974976, + "learning_rate": 4.930727938043091e-06, + "loss": 0.8649, + "step": 380 + }, + { + "epoch": 0.5965553235908142, + "grad_norm": 1.0946165323257446, + "learning_rate": 4.9302345723933344e-06, + "loss": 0.7794, + "step": 381 + }, + { + "epoch": 0.5981210855949896, + "grad_norm": 1.0532523393630981, + "learning_rate": 4.929739480921157e-06, + "loss": 0.8555, + "step": 382 + }, + { + "epoch": 0.5996868475991649, + "grad_norm": 0.9441399574279785, + "learning_rate": 4.929242663978144e-06, + "loss": 0.7613, + "step": 383 + }, + { + "epoch": 0.6012526096033403, + "grad_norm": 0.9970793128013611, + "learning_rate": 4.92874412191711e-06, + "loss": 0.9037, + "step": 384 + }, + { + "epoch": 0.6028183716075156, + "grad_norm": 0.9777354001998901, + "learning_rate": 4.928243855092091e-06, + "loss": 0.9044, + "step": 385 + }, + { + "epoch": 0.6043841336116911, + "grad_norm": 0.9507855772972107, + "learning_rate": 4.92774186385835e-06, + "loss": 0.8359, + "step": 386 + }, + { + "epoch": 0.6059498956158664, + "grad_norm": 0.9827553033828735, + "learning_rate": 4.9272381485723765e-06, + "loss": 0.7938, + "step": 387 + }, + { + "epoch": 0.6075156576200418, + "grad_norm": 0.8896475434303284, + "learning_rate": 4.926732709591879e-06, + "loss": 0.8545, + "step": 388 + }, + { + "epoch": 0.6090814196242171, + "grad_norm": 0.9817705750465393, + "learning_rate": 4.926225547275795e-06, + "loss": 0.8456, + "step": 389 + }, + { + "epoch": 0.6106471816283925, + "grad_norm": 0.9646008610725403, + "learning_rate": 4.925716661984283e-06, + "loss": 0.7906, + "step": 390 + }, + { + "epoch": 0.6122129436325678, + "grad_norm": 0.8940444588661194, + "learning_rate": 4.9252060540787275e-06, + "loss": 0.8371, + "step": 391 + }, + { + "epoch": 0.6137787056367432, + "grad_norm": 1.0812394618988037, + "learning_rate": 4.924693723921734e-06, + "loss": 0.8056, + "step": 392 + }, + { + "epoch": 0.6153444676409185, + "grad_norm": 0.9953856468200684, + "learning_rate": 4.924179671877132e-06, + "loss": 0.8651, + "step": 393 + }, + { + "epoch": 0.6169102296450939, + "grad_norm": 0.9760703444480896, + "learning_rate": 4.9236638983099735e-06, + "loss": 0.845, + "step": 394 + }, + { + "epoch": 0.6184759916492694, + "grad_norm": 1.0127615928649902, + "learning_rate": 4.923146403586533e-06, + "loss": 0.7987, + "step": 395 + }, + { + "epoch": 0.6200417536534447, + "grad_norm": 0.8771396279335022, + "learning_rate": 4.9226271880743086e-06, + "loss": 0.8092, + "step": 396 + }, + { + "epoch": 0.6216075156576201, + "grad_norm": 0.8906831741333008, + "learning_rate": 4.9221062521420196e-06, + "loss": 0.8039, + "step": 397 + }, + { + "epoch": 0.6231732776617954, + "grad_norm": 1.0218392610549927, + "learning_rate": 4.9215835961596045e-06, + "loss": 0.8354, + "step": 398 + }, + { + "epoch": 0.6247390396659708, + "grad_norm": 1.0382038354873657, + "learning_rate": 4.921059220498227e-06, + "loss": 0.8021, + "step": 399 + }, + { + "epoch": 0.6263048016701461, + "grad_norm": 0.8997588157653809, + "learning_rate": 4.920533125530272e-06, + "loss": 0.8716, + "step": 400 + }, + { + "epoch": 0.6278705636743215, + "grad_norm": 0.8725724220275879, + "learning_rate": 4.920005311629342e-06, + "loss": 0.7934, + "step": 401 + }, + { + "epoch": 0.6294363256784968, + "grad_norm": 0.9289800524711609, + "learning_rate": 4.9194757791702625e-06, + "loss": 0.812, + "step": 402 + }, + { + "epoch": 0.6310020876826722, + "grad_norm": 0.9969078898429871, + "learning_rate": 4.918944528529079e-06, + "loss": 0.866, + "step": 403 + }, + { + "epoch": 0.6325678496868476, + "grad_norm": 0.9544941186904907, + "learning_rate": 4.918411560083058e-06, + "loss": 0.8035, + "step": 404 + }, + { + "epoch": 0.634133611691023, + "grad_norm": 0.9678053855895996, + "learning_rate": 4.917876874210686e-06, + "loss": 0.8254, + "step": 405 + }, + { + "epoch": 0.6356993736951984, + "grad_norm": 0.9782965779304504, + "learning_rate": 4.917340471291666e-06, + "loss": 0.7954, + "step": 406 + }, + { + "epoch": 0.6372651356993737, + "grad_norm": 0.9380600452423096, + "learning_rate": 4.916802351706923e-06, + "loss": 0.8287, + "step": 407 + }, + { + "epoch": 0.6388308977035491, + "grad_norm": 0.9455891847610474, + "learning_rate": 4.916262515838603e-06, + "loss": 0.8714, + "step": 408 + }, + { + "epoch": 0.6403966597077244, + "grad_norm": 0.8937395811080933, + "learning_rate": 4.915720964070065e-06, + "loss": 0.7618, + "step": 409 + }, + { + "epoch": 0.6419624217118998, + "grad_norm": 1.0304689407348633, + "learning_rate": 4.915177696785894e-06, + "loss": 0.8436, + "step": 410 + }, + { + "epoch": 0.6435281837160751, + "grad_norm": 0.9623110294342041, + "learning_rate": 4.914632714371885e-06, + "loss": 0.865, + "step": 411 + }, + { + "epoch": 0.6450939457202505, + "grad_norm": 0.9212867021560669, + "learning_rate": 4.914086017215059e-06, + "loss": 0.8232, + "step": 412 + }, + { + "epoch": 0.6466597077244259, + "grad_norm": 0.9817205667495728, + "learning_rate": 4.9135376057036485e-06, + "loss": 0.7926, + "step": 413 + }, + { + "epoch": 0.6482254697286013, + "grad_norm": 0.9576719403266907, + "learning_rate": 4.912987480227108e-06, + "loss": 0.8812, + "step": 414 + }, + { + "epoch": 0.6497912317327766, + "grad_norm": 0.9415407180786133, + "learning_rate": 4.912435641176106e-06, + "loss": 0.8985, + "step": 415 + }, + { + "epoch": 0.651356993736952, + "grad_norm": 0.9084275364875793, + "learning_rate": 4.91188208894253e-06, + "loss": 0.8085, + "step": 416 + }, + { + "epoch": 0.6529227557411273, + "grad_norm": 0.8890969753265381, + "learning_rate": 4.911326823919482e-06, + "loss": 0.8127, + "step": 417 + }, + { + "epoch": 0.6544885177453027, + "grad_norm": 0.8727678656578064, + "learning_rate": 4.9107698465012815e-06, + "loss": 0.8656, + "step": 418 + }, + { + "epoch": 0.656054279749478, + "grad_norm": 1.0174667835235596, + "learning_rate": 4.910211157083465e-06, + "loss": 0.8358, + "step": 419 + }, + { + "epoch": 0.6576200417536534, + "grad_norm": 1.045121431350708, + "learning_rate": 4.909650756062782e-06, + "loss": 0.8239, + "step": 420 + }, + { + "epoch": 0.6591858037578288, + "grad_norm": 1.0688676834106445, + "learning_rate": 4.909088643837202e-06, + "loss": 0.9246, + "step": 421 + }, + { + "epoch": 0.6607515657620042, + "grad_norm": 0.9371877908706665, + "learning_rate": 4.908524820805905e-06, + "loss": 0.7862, + "step": 422 + }, + { + "epoch": 0.6623173277661796, + "grad_norm": 0.9567264318466187, + "learning_rate": 4.907959287369288e-06, + "loss": 0.8164, + "step": 423 + }, + { + "epoch": 0.6638830897703549, + "grad_norm": 0.9109166860580444, + "learning_rate": 4.907392043928964e-06, + "loss": 0.7016, + "step": 424 + }, + { + "epoch": 0.6654488517745303, + "grad_norm": 1.0378203392028809, + "learning_rate": 4.906823090887757e-06, + "loss": 0.8986, + "step": 425 + }, + { + "epoch": 0.6670146137787056, + "grad_norm": 0.9871951341629028, + "learning_rate": 4.906252428649708e-06, + "loss": 0.8366, + "step": 426 + }, + { + "epoch": 0.668580375782881, + "grad_norm": 0.8682780265808105, + "learning_rate": 4.905680057620072e-06, + "loss": 0.738, + "step": 427 + }, + { + "epoch": 0.6701461377870563, + "grad_norm": 0.8834647536277771, + "learning_rate": 4.9051059782053125e-06, + "loss": 0.8304, + "step": 428 + }, + { + "epoch": 0.6717118997912317, + "grad_norm": 1.0133378505706787, + "learning_rate": 4.904530190813113e-06, + "loss": 0.9369, + "step": 429 + }, + { + "epoch": 0.673277661795407, + "grad_norm": 0.9967373013496399, + "learning_rate": 4.9039526958523676e-06, + "loss": 0.782, + "step": 430 + }, + { + "epoch": 0.6748434237995825, + "grad_norm": 1.0258032083511353, + "learning_rate": 4.90337349373318e-06, + "loss": 0.8214, + "step": 431 + }, + { + "epoch": 0.6764091858037579, + "grad_norm": 0.9814366102218628, + "learning_rate": 4.90279258486687e-06, + "loss": 0.8173, + "step": 432 + }, + { + "epoch": 0.6779749478079332, + "grad_norm": 0.9351440668106079, + "learning_rate": 4.902209969665966e-06, + "loss": 0.8815, + "step": 433 + }, + { + "epoch": 0.6795407098121086, + "grad_norm": 0.9490398168563843, + "learning_rate": 4.901625648544212e-06, + "loss": 0.8295, + "step": 434 + }, + { + "epoch": 0.6811064718162839, + "grad_norm": 1.0183534622192383, + "learning_rate": 4.9010396219165614e-06, + "loss": 0.8538, + "step": 435 + }, + { + "epoch": 0.6826722338204593, + "grad_norm": 0.9546058773994446, + "learning_rate": 4.900451890199179e-06, + "loss": 0.8385, + "step": 436 + }, + { + "epoch": 0.6842379958246346, + "grad_norm": 1.076120138168335, + "learning_rate": 4.8998624538094394e-06, + "loss": 0.8958, + "step": 437 + }, + { + "epoch": 0.68580375782881, + "grad_norm": 0.8935732841491699, + "learning_rate": 4.899271313165929e-06, + "loss": 0.8674, + "step": 438 + }, + { + "epoch": 0.6873695198329853, + "grad_norm": 0.9574604630470276, + "learning_rate": 4.898678468688445e-06, + "loss": 0.8245, + "step": 439 + }, + { + "epoch": 0.6889352818371608, + "grad_norm": 0.931751549243927, + "learning_rate": 4.898083920797993e-06, + "loss": 0.8409, + "step": 440 + }, + { + "epoch": 0.6905010438413361, + "grad_norm": 0.937874436378479, + "learning_rate": 4.89748766991679e-06, + "loss": 0.8222, + "step": 441 + }, + { + "epoch": 0.6920668058455115, + "grad_norm": 0.9606375694274902, + "learning_rate": 4.896889716468259e-06, + "loss": 0.8123, + "step": 442 + }, + { + "epoch": 0.6936325678496869, + "grad_norm": 0.8692249059677124, + "learning_rate": 4.896290060877038e-06, + "loss": 0.8532, + "step": 443 + }, + { + "epoch": 0.6951983298538622, + "grad_norm": 0.9597892165184021, + "learning_rate": 4.895688703568968e-06, + "loss": 0.8365, + "step": 444 + }, + { + "epoch": 0.6967640918580376, + "grad_norm": 0.958342432975769, + "learning_rate": 4.895085644971102e-06, + "loss": 0.8535, + "step": 445 + }, + { + "epoch": 0.6983298538622129, + "grad_norm": 0.9603601098060608, + "learning_rate": 4.894480885511699e-06, + "loss": 0.8227, + "step": 446 + }, + { + "epoch": 0.6998956158663883, + "grad_norm": 1.0010526180267334, + "learning_rate": 4.893874425620226e-06, + "loss": 0.8785, + "step": 447 + }, + { + "epoch": 0.7014613778705637, + "grad_norm": 1.0061219930648804, + "learning_rate": 4.893266265727361e-06, + "loss": 0.8336, + "step": 448 + }, + { + "epoch": 0.7030271398747391, + "grad_norm": 0.8967173099517822, + "learning_rate": 4.892656406264984e-06, + "loss": 0.8305, + "step": 449 + }, + { + "epoch": 0.7045929018789144, + "grad_norm": 0.962729275226593, + "learning_rate": 4.892044847666186e-06, + "loss": 0.7843, + "step": 450 + }, + { + "epoch": 0.7061586638830898, + "grad_norm": 1.0067342519760132, + "learning_rate": 4.891431590365262e-06, + "loss": 0.8111, + "step": 451 + }, + { + "epoch": 0.7077244258872651, + "grad_norm": 1.0252268314361572, + "learning_rate": 4.890816634797716e-06, + "loss": 0.8253, + "step": 452 + }, + { + "epoch": 0.7092901878914405, + "grad_norm": 0.9098564386367798, + "learning_rate": 4.890199981400255e-06, + "loss": 0.8526, + "step": 453 + }, + { + "epoch": 0.7108559498956158, + "grad_norm": 0.893135130405426, + "learning_rate": 4.889581630610793e-06, + "loss": 0.8217, + "step": 454 + }, + { + "epoch": 0.7124217118997912, + "grad_norm": 0.9225817322731018, + "learning_rate": 4.888961582868453e-06, + "loss": 0.8296, + "step": 455 + }, + { + "epoch": 0.7139874739039666, + "grad_norm": 0.973386824131012, + "learning_rate": 4.888339838613555e-06, + "loss": 0.7429, + "step": 456 + }, + { + "epoch": 0.715553235908142, + "grad_norm": 0.9829993844032288, + "learning_rate": 4.8877163982876306e-06, + "loss": 0.7405, + "step": 457 + }, + { + "epoch": 0.7171189979123174, + "grad_norm": 0.927728533744812, + "learning_rate": 4.887091262333414e-06, + "loss": 0.8257, + "step": 458 + }, + { + "epoch": 0.7186847599164927, + "grad_norm": 0.9419897794723511, + "learning_rate": 4.886464431194844e-06, + "loss": 0.7762, + "step": 459 + }, + { + "epoch": 0.7202505219206681, + "grad_norm": 0.9358823299407959, + "learning_rate": 4.885835905317061e-06, + "loss": 0.7794, + "step": 460 + }, + { + "epoch": 0.7218162839248434, + "grad_norm": 1.0290783643722534, + "learning_rate": 4.88520568514641e-06, + "loss": 0.8723, + "step": 461 + }, + { + "epoch": 0.7233820459290188, + "grad_norm": 0.9909068942070007, + "learning_rate": 4.884573771130442e-06, + "loss": 0.8302, + "step": 462 + }, + { + "epoch": 0.7249478079331941, + "grad_norm": 0.9972653985023499, + "learning_rate": 4.883940163717906e-06, + "loss": 0.7961, + "step": 463 + }, + { + "epoch": 0.7265135699373695, + "grad_norm": 0.9245638251304626, + "learning_rate": 4.883304863358757e-06, + "loss": 0.8057, + "step": 464 + }, + { + "epoch": 0.7280793319415448, + "grad_norm": 1.0097148418426514, + "learning_rate": 4.882667870504153e-06, + "loss": 0.7296, + "step": 465 + }, + { + "epoch": 0.7296450939457203, + "grad_norm": 0.9792670607566833, + "learning_rate": 4.882029185606448e-06, + "loss": 0.8006, + "step": 466 + }, + { + "epoch": 0.7312108559498957, + "grad_norm": 0.9308120608329773, + "learning_rate": 4.881388809119206e-06, + "loss": 0.7957, + "step": 467 + }, + { + "epoch": 0.732776617954071, + "grad_norm": 0.9472301602363586, + "learning_rate": 4.880746741497187e-06, + "loss": 0.8984, + "step": 468 + }, + { + "epoch": 0.7343423799582464, + "grad_norm": 0.9898937344551086, + "learning_rate": 4.880102983196351e-06, + "loss": 0.8365, + "step": 469 + }, + { + "epoch": 0.7359081419624217, + "grad_norm": 0.9376612901687622, + "learning_rate": 4.879457534673863e-06, + "loss": 0.7838, + "step": 470 + }, + { + "epoch": 0.7374739039665971, + "grad_norm": 1.030616283416748, + "learning_rate": 4.878810396388085e-06, + "loss": 0.7698, + "step": 471 + }, + { + "epoch": 0.7390396659707724, + "grad_norm": 0.9583387970924377, + "learning_rate": 4.87816156879858e-06, + "loss": 0.7381, + "step": 472 + }, + { + "epoch": 0.7406054279749478, + "grad_norm": 1.00180184841156, + "learning_rate": 4.87751105236611e-06, + "loss": 0.8497, + "step": 473 + }, + { + "epoch": 0.7421711899791231, + "grad_norm": 0.9382562637329102, + "learning_rate": 4.876858847552638e-06, + "loss": 0.8698, + "step": 474 + }, + { + "epoch": 0.7437369519832986, + "grad_norm": 1.0104063749313354, + "learning_rate": 4.8762049548213255e-06, + "loss": 0.8444, + "step": 475 + }, + { + "epoch": 0.7453027139874739, + "grad_norm": 0.9321765303611755, + "learning_rate": 4.87554937463653e-06, + "loss": 0.8027, + "step": 476 + }, + { + "epoch": 0.7468684759916493, + "grad_norm": 0.9527666568756104, + "learning_rate": 4.874892107463811e-06, + "loss": 0.8069, + "step": 477 + }, + { + "epoch": 0.7484342379958246, + "grad_norm": 0.9576907157897949, + "learning_rate": 4.874233153769924e-06, + "loss": 0.774, + "step": 478 + }, + { + "epoch": 0.75, + "grad_norm": 1.0516161918640137, + "learning_rate": 4.873572514022822e-06, + "loss": 0.8176, + "step": 479 + }, + { + "epoch": 0.7515657620041754, + "grad_norm": 1.026928186416626, + "learning_rate": 4.872910188691657e-06, + "loss": 0.8797, + "step": 480 + }, + { + "epoch": 0.7531315240083507, + "grad_norm": 0.9312179088592529, + "learning_rate": 4.872246178246776e-06, + "loss": 0.7747, + "step": 481 + }, + { + "epoch": 0.7546972860125261, + "grad_norm": 1.0676255226135254, + "learning_rate": 4.871580483159725e-06, + "loss": 0.7801, + "step": 482 + }, + { + "epoch": 0.7562630480167014, + "grad_norm": 1.066056489944458, + "learning_rate": 4.870913103903245e-06, + "loss": 0.7837, + "step": 483 + }, + { + "epoch": 0.7578288100208769, + "grad_norm": 0.9388124346733093, + "learning_rate": 4.87024404095127e-06, + "loss": 0.7801, + "step": 484 + }, + { + "epoch": 0.7593945720250522, + "grad_norm": 0.9879791140556335, + "learning_rate": 4.8695732947789356e-06, + "loss": 0.7849, + "step": 485 + }, + { + "epoch": 0.7609603340292276, + "grad_norm": 1.0002318620681763, + "learning_rate": 4.868900865862569e-06, + "loss": 0.8219, + "step": 486 + }, + { + "epoch": 0.7625260960334029, + "grad_norm": 0.9599125385284424, + "learning_rate": 4.8682267546796945e-06, + "loss": 0.8888, + "step": 487 + }, + { + "epoch": 0.7640918580375783, + "grad_norm": 0.9099973440170288, + "learning_rate": 4.867550961709027e-06, + "loss": 0.7585, + "step": 488 + }, + { + "epoch": 0.7656576200417536, + "grad_norm": 0.9715955257415771, + "learning_rate": 4.866873487430481e-06, + "loss": 0.8237, + "step": 489 + }, + { + "epoch": 0.767223382045929, + "grad_norm": 1.0194706916809082, + "learning_rate": 4.86619433232516e-06, + "loss": 0.7708, + "step": 490 + }, + { + "epoch": 0.7687891440501043, + "grad_norm": 1.044005036354065, + "learning_rate": 4.865513496875367e-06, + "loss": 0.8, + "step": 491 + }, + { + "epoch": 0.7703549060542797, + "grad_norm": 0.9776361584663391, + "learning_rate": 4.8648309815645915e-06, + "loss": 0.7696, + "step": 492 + }, + { + "epoch": 0.7719206680584552, + "grad_norm": 0.9213030934333801, + "learning_rate": 4.864146786877521e-06, + "loss": 0.8558, + "step": 493 + }, + { + "epoch": 0.7734864300626305, + "grad_norm": 0.9361440539360046, + "learning_rate": 4.8634609133000336e-06, + "loss": 0.7851, + "step": 494 + }, + { + "epoch": 0.7750521920668059, + "grad_norm": 1.0199540853500366, + "learning_rate": 4.8627733613192e-06, + "loss": 0.8229, + "step": 495 + }, + { + "epoch": 0.7766179540709812, + "grad_norm": 1.0244604349136353, + "learning_rate": 4.862084131423283e-06, + "loss": 0.772, + "step": 496 + }, + { + "epoch": 0.7781837160751566, + "grad_norm": 0.9414368271827698, + "learning_rate": 4.861393224101736e-06, + "loss": 0.7579, + "step": 497 + }, + { + "epoch": 0.7797494780793319, + "grad_norm": 0.9198965430259705, + "learning_rate": 4.860700639845205e-06, + "loss": 0.855, + "step": 498 + }, + { + "epoch": 0.7813152400835073, + "grad_norm": 0.9398712515830994, + "learning_rate": 4.8600063791455275e-06, + "loss": 0.7824, + "step": 499 + }, + { + "epoch": 0.7828810020876826, + "grad_norm": 0.9544969201087952, + "learning_rate": 4.8593104424957275e-06, + "loss": 0.8104, + "step": 500 + }, + { + "epoch": 0.784446764091858, + "grad_norm": 0.9522370100021362, + "learning_rate": 4.858612830390024e-06, + "loss": 0.813, + "step": 501 + }, + { + "epoch": 0.7860125260960334, + "grad_norm": 0.9830234050750732, + "learning_rate": 4.857913543323823e-06, + "loss": 0.8656, + "step": 502 + }, + { + "epoch": 0.7875782881002088, + "grad_norm": 1.0527008771896362, + "learning_rate": 4.8572125817937195e-06, + "loss": 0.8417, + "step": 503 + }, + { + "epoch": 0.7891440501043842, + "grad_norm": 0.9772782921791077, + "learning_rate": 4.856509946297501e-06, + "loss": 0.7983, + "step": 504 + }, + { + "epoch": 0.7907098121085595, + "grad_norm": 0.9250578880310059, + "learning_rate": 4.855805637334141e-06, + "loss": 0.7391, + "step": 505 + }, + { + "epoch": 0.7922755741127349, + "grad_norm": 0.9792859554290771, + "learning_rate": 4.855099655403802e-06, + "loss": 0.7753, + "step": 506 + }, + { + "epoch": 0.7938413361169102, + "grad_norm": 0.8876479864120483, + "learning_rate": 4.854392001007834e-06, + "loss": 0.8249, + "step": 507 + }, + { + "epoch": 0.7954070981210856, + "grad_norm": 0.9336771368980408, + "learning_rate": 4.853682674648775e-06, + "loss": 0.8462, + "step": 508 + }, + { + "epoch": 0.7969728601252609, + "grad_norm": 0.9732388854026794, + "learning_rate": 4.852971676830352e-06, + "loss": 0.8791, + "step": 509 + }, + { + "epoch": 0.7985386221294363, + "grad_norm": 0.9135870337486267, + "learning_rate": 4.852259008057476e-06, + "loss": 0.7279, + "step": 510 + }, + { + "epoch": 0.8001043841336117, + "grad_norm": 0.9782464504241943, + "learning_rate": 4.8515446688362476e-06, + "loss": 0.7797, + "step": 511 + }, + { + "epoch": 0.8016701461377871, + "grad_norm": 0.9571824073791504, + "learning_rate": 4.85082865967395e-06, + "loss": 0.8349, + "step": 512 + }, + { + "epoch": 0.8032359081419624, + "grad_norm": 1.011786937713623, + "learning_rate": 4.850110981079057e-06, + "loss": 0.8057, + "step": 513 + }, + { + "epoch": 0.8048016701461378, + "grad_norm": 0.9418787956237793, + "learning_rate": 4.8493916335612255e-06, + "loss": 0.8592, + "step": 514 + }, + { + "epoch": 0.8063674321503131, + "grad_norm": 0.9292604923248291, + "learning_rate": 4.848670617631296e-06, + "loss": 0.7471, + "step": 515 + }, + { + "epoch": 0.8079331941544885, + "grad_norm": 0.9303188323974609, + "learning_rate": 4.847947933801296e-06, + "loss": 0.8768, + "step": 516 + }, + { + "epoch": 0.8094989561586639, + "grad_norm": 0.9897359609603882, + "learning_rate": 4.847223582584437e-06, + "loss": 0.8343, + "step": 517 + }, + { + "epoch": 0.8110647181628392, + "grad_norm": 0.9695891737937927, + "learning_rate": 4.846497564495114e-06, + "loss": 0.8318, + "step": 518 + }, + { + "epoch": 0.8126304801670147, + "grad_norm": 1.0175273418426514, + "learning_rate": 4.845769880048906e-06, + "loss": 0.9137, + "step": 519 + }, + { + "epoch": 0.81419624217119, + "grad_norm": 0.9497922658920288, + "learning_rate": 4.845040529762577e-06, + "loss": 0.6549, + "step": 520 + }, + { + "epoch": 0.8157620041753654, + "grad_norm": 1.0647574663162231, + "learning_rate": 4.844309514154071e-06, + "loss": 0.8243, + "step": 521 + }, + { + "epoch": 0.8173277661795407, + "grad_norm": 0.9710965156555176, + "learning_rate": 4.843576833742517e-06, + "loss": 0.7864, + "step": 522 + }, + { + "epoch": 0.8188935281837161, + "grad_norm": 0.9881441593170166, + "learning_rate": 4.842842489048225e-06, + "loss": 0.7799, + "step": 523 + }, + { + "epoch": 0.8204592901878914, + "grad_norm": 0.9455597400665283, + "learning_rate": 4.842106480592687e-06, + "loss": 0.7943, + "step": 524 + }, + { + "epoch": 0.8220250521920668, + "grad_norm": 0.9948442578315735, + "learning_rate": 4.841368808898577e-06, + "loss": 0.8587, + "step": 525 + }, + { + "epoch": 0.8235908141962421, + "grad_norm": 0.9959186315536499, + "learning_rate": 4.84062947448975e-06, + "loss": 0.8477, + "step": 526 + }, + { + "epoch": 0.8251565762004175, + "grad_norm": 0.9791848659515381, + "learning_rate": 4.839888477891243e-06, + "loss": 0.7568, + "step": 527 + }, + { + "epoch": 0.826722338204593, + "grad_norm": 1.1499426364898682, + "learning_rate": 4.839145819629269e-06, + "loss": 0.8296, + "step": 528 + }, + { + "epoch": 0.8282881002087683, + "grad_norm": 0.9269506335258484, + "learning_rate": 4.8384015002312265e-06, + "loss": 0.7935, + "step": 529 + }, + { + "epoch": 0.8298538622129437, + "grad_norm": 0.970565140247345, + "learning_rate": 4.837655520225692e-06, + "loss": 0.8138, + "step": 530 + }, + { + "epoch": 0.831419624217119, + "grad_norm": 0.9852119088172913, + "learning_rate": 4.836907880142418e-06, + "loss": 0.7909, + "step": 531 + }, + { + "epoch": 0.8329853862212944, + "grad_norm": 0.927129328250885, + "learning_rate": 4.836158580512339e-06, + "loss": 0.8105, + "step": 532 + }, + { + "epoch": 0.8345511482254697, + "grad_norm": 1.0711781978607178, + "learning_rate": 4.835407621867569e-06, + "loss": 0.8241, + "step": 533 + }, + { + "epoch": 0.8361169102296451, + "grad_norm": 1.0472421646118164, + "learning_rate": 4.834655004741397e-06, + "loss": 0.7923, + "step": 534 + }, + { + "epoch": 0.8376826722338204, + "grad_norm": 0.9761579632759094, + "learning_rate": 4.833900729668293e-06, + "loss": 0.8203, + "step": 535 + }, + { + "epoch": 0.8392484342379958, + "grad_norm": 1.0559637546539307, + "learning_rate": 4.833144797183902e-06, + "loss": 0.8915, + "step": 536 + }, + { + "epoch": 0.8408141962421712, + "grad_norm": 0.9951052069664001, + "learning_rate": 4.832387207825045e-06, + "loss": 0.8253, + "step": 537 + }, + { + "epoch": 0.8423799582463466, + "grad_norm": 0.894079864025116, + "learning_rate": 4.831627962129725e-06, + "loss": 0.7479, + "step": 538 + }, + { + "epoch": 0.843945720250522, + "grad_norm": 0.9832531213760376, + "learning_rate": 4.830867060637115e-06, + "loss": 0.8484, + "step": 539 + }, + { + "epoch": 0.8455114822546973, + "grad_norm": 1.0477591753005981, + "learning_rate": 4.8301045038875665e-06, + "loss": 0.8052, + "step": 540 + }, + { + "epoch": 0.8470772442588727, + "grad_norm": 0.9388875365257263, + "learning_rate": 4.829340292422607e-06, + "loss": 0.7998, + "step": 541 + }, + { + "epoch": 0.848643006263048, + "grad_norm": 0.8975532054901123, + "learning_rate": 4.8285744267849395e-06, + "loss": 0.848, + "step": 542 + }, + { + "epoch": 0.8502087682672234, + "grad_norm": 0.9211076498031616, + "learning_rate": 4.8278069075184396e-06, + "loss": 0.8337, + "step": 543 + }, + { + "epoch": 0.8517745302713987, + "grad_norm": 0.970134437084198, + "learning_rate": 4.827037735168159e-06, + "loss": 0.7843, + "step": 544 + }, + { + "epoch": 0.8533402922755741, + "grad_norm": 0.9714353084564209, + "learning_rate": 4.826266910280322e-06, + "loss": 0.6991, + "step": 545 + }, + { + "epoch": 0.8549060542797495, + "grad_norm": 0.8607689142227173, + "learning_rate": 4.8254944334023265e-06, + "loss": 0.7314, + "step": 546 + }, + { + "epoch": 0.8564718162839249, + "grad_norm": 0.9843030571937561, + "learning_rate": 4.824720305082747e-06, + "loss": 0.7497, + "step": 547 + }, + { + "epoch": 0.8580375782881002, + "grad_norm": 0.9638474583625793, + "learning_rate": 4.823944525871324e-06, + "loss": 0.776, + "step": 548 + }, + { + "epoch": 0.8596033402922756, + "grad_norm": 0.9342365860939026, + "learning_rate": 4.823167096318979e-06, + "loss": 0.77, + "step": 549 + }, + { + "epoch": 0.8611691022964509, + "grad_norm": 0.9608579874038696, + "learning_rate": 4.822388016977797e-06, + "loss": 0.8281, + "step": 550 + }, + { + "epoch": 0.8627348643006263, + "grad_norm": 0.8996383547782898, + "learning_rate": 4.8216072884010395e-06, + "loss": 0.7519, + "step": 551 + }, + { + "epoch": 0.8643006263048016, + "grad_norm": 0.9922810196876526, + "learning_rate": 4.820824911143139e-06, + "loss": 0.8714, + "step": 552 + }, + { + "epoch": 0.865866388308977, + "grad_norm": 0.9861842393875122, + "learning_rate": 4.820040885759697e-06, + "loss": 0.7998, + "step": 553 + }, + { + "epoch": 0.8674321503131524, + "grad_norm": 1.013168215751648, + "learning_rate": 4.819255212807486e-06, + "loss": 0.8291, + "step": 554 + }, + { + "epoch": 0.8689979123173278, + "grad_norm": 0.9695484638214111, + "learning_rate": 4.818467892844451e-06, + "loss": 0.8257, + "step": 555 + }, + { + "epoch": 0.8705636743215032, + "grad_norm": 1.0018948316574097, + "learning_rate": 4.817678926429702e-06, + "loss": 0.7551, + "step": 556 + }, + { + "epoch": 0.8721294363256785, + "grad_norm": 1.0259422063827515, + "learning_rate": 4.816888314123521e-06, + "loss": 0.8582, + "step": 557 + }, + { + "epoch": 0.8736951983298539, + "grad_norm": 0.9045076966285706, + "learning_rate": 4.81609605648736e-06, + "loss": 0.811, + "step": 558 + }, + { + "epoch": 0.8752609603340292, + "grad_norm": 0.9112316370010376, + "learning_rate": 4.815302154083837e-06, + "loss": 0.818, + "step": 559 + }, + { + "epoch": 0.8768267223382046, + "grad_norm": 1.0060032606124878, + "learning_rate": 4.814506607476739e-06, + "loss": 0.766, + "step": 560 + }, + { + "epoch": 0.8783924843423799, + "grad_norm": 0.9126970171928406, + "learning_rate": 4.813709417231021e-06, + "loss": 0.8903, + "step": 561 + }, + { + "epoch": 0.8799582463465553, + "grad_norm": 1.0825906991958618, + "learning_rate": 4.812910583912804e-06, + "loss": 0.8466, + "step": 562 + }, + { + "epoch": 0.8815240083507306, + "grad_norm": 0.9530975818634033, + "learning_rate": 4.812110108089377e-06, + "loss": 0.8163, + "step": 563 + }, + { + "epoch": 0.8830897703549061, + "grad_norm": 0.9566405415534973, + "learning_rate": 4.8113079903291955e-06, + "loss": 0.8244, + "step": 564 + }, + { + "epoch": 0.8846555323590815, + "grad_norm": 0.919980525970459, + "learning_rate": 4.81050423120188e-06, + "loss": 0.8335, + "step": 565 + }, + { + "epoch": 0.8862212943632568, + "grad_norm": 0.9123868346214294, + "learning_rate": 4.809698831278217e-06, + "loss": 0.7712, + "step": 566 + }, + { + "epoch": 0.8877870563674322, + "grad_norm": 0.9589663743972778, + "learning_rate": 4.8088917911301595e-06, + "loss": 0.8467, + "step": 567 + }, + { + "epoch": 0.8893528183716075, + "grad_norm": 0.9052798748016357, + "learning_rate": 4.808083111330823e-06, + "loss": 0.8078, + "step": 568 + }, + { + "epoch": 0.8909185803757829, + "grad_norm": 0.9250494241714478, + "learning_rate": 4.807272792454489e-06, + "loss": 0.8536, + "step": 569 + }, + { + "epoch": 0.8924843423799582, + "grad_norm": 0.8451298475265503, + "learning_rate": 4.806460835076603e-06, + "loss": 0.7752, + "step": 570 + }, + { + "epoch": 0.8940501043841336, + "grad_norm": 1.0324240922927856, + "learning_rate": 4.805647239773773e-06, + "loss": 0.7871, + "step": 571 + }, + { + "epoch": 0.8956158663883089, + "grad_norm": 0.9338528513908386, + "learning_rate": 4.804832007123771e-06, + "loss": 0.8082, + "step": 572 + }, + { + "epoch": 0.8971816283924844, + "grad_norm": 1.0080739259719849, + "learning_rate": 4.804015137705531e-06, + "loss": 0.7636, + "step": 573 + }, + { + "epoch": 0.8987473903966597, + "grad_norm": 0.9817036986351013, + "learning_rate": 4.803196632099152e-06, + "loss": 0.8599, + "step": 574 + }, + { + "epoch": 0.9003131524008351, + "grad_norm": 0.9284219145774841, + "learning_rate": 4.80237649088589e-06, + "loss": 0.8152, + "step": 575 + }, + { + "epoch": 0.9018789144050104, + "grad_norm": 0.8922317028045654, + "learning_rate": 4.801554714648166e-06, + "loss": 0.8146, + "step": 576 + }, + { + "epoch": 0.9034446764091858, + "grad_norm": 0.9465014338493347, + "learning_rate": 4.800731303969565e-06, + "loss": 0.7663, + "step": 577 + }, + { + "epoch": 0.9050104384133612, + "grad_norm": 1.0325525999069214, + "learning_rate": 4.799906259434824e-06, + "loss": 0.8458, + "step": 578 + }, + { + "epoch": 0.9065762004175365, + "grad_norm": 0.9366307258605957, + "learning_rate": 4.7990795816298485e-06, + "loss": 0.7992, + "step": 579 + }, + { + "epoch": 0.9081419624217119, + "grad_norm": 0.9923847317695618, + "learning_rate": 4.7982512711416995e-06, + "loss": 0.7256, + "step": 580 + }, + { + "epoch": 0.9097077244258872, + "grad_norm": 1.0387133359909058, + "learning_rate": 4.7974213285586e-06, + "loss": 0.7766, + "step": 581 + }, + { + "epoch": 0.9112734864300627, + "grad_norm": 0.9617043733596802, + "learning_rate": 4.79658975446993e-06, + "loss": 0.7602, + "step": 582 + }, + { + "epoch": 0.912839248434238, + "grad_norm": 0.9530353546142578, + "learning_rate": 4.795756549466229e-06, + "loss": 0.7922, + "step": 583 + }, + { + "epoch": 0.9144050104384134, + "grad_norm": 0.9305918216705322, + "learning_rate": 4.794921714139194e-06, + "loss": 0.8377, + "step": 584 + }, + { + "epoch": 0.9159707724425887, + "grad_norm": 0.899860143661499, + "learning_rate": 4.794085249081682e-06, + "loss": 0.8343, + "step": 585 + }, + { + "epoch": 0.9175365344467641, + "grad_norm": 0.9024940729141235, + "learning_rate": 4.793247154887703e-06, + "loss": 0.8271, + "step": 586 + }, + { + "epoch": 0.9191022964509394, + "grad_norm": 0.980869472026825, + "learning_rate": 4.7924074321524286e-06, + "loss": 0.7468, + "step": 587 + }, + { + "epoch": 0.9206680584551148, + "grad_norm": 1.0506541728973389, + "learning_rate": 4.791566081472185e-06, + "loss": 0.9207, + "step": 588 + }, + { + "epoch": 0.9222338204592901, + "grad_norm": 1.0723917484283447, + "learning_rate": 4.790723103444454e-06, + "loss": 0.7615, + "step": 589 + }, + { + "epoch": 0.9237995824634656, + "grad_norm": 1.0455081462860107, + "learning_rate": 4.789878498667873e-06, + "loss": 0.8115, + "step": 590 + }, + { + "epoch": 0.925365344467641, + "grad_norm": 0.9729639291763306, + "learning_rate": 4.789032267742234e-06, + "loss": 0.8622, + "step": 591 + }, + { + "epoch": 0.9269311064718163, + "grad_norm": 0.9695239067077637, + "learning_rate": 4.788184411268488e-06, + "loss": 0.8165, + "step": 592 + }, + { + "epoch": 0.9284968684759917, + "grad_norm": 1.0099278688430786, + "learning_rate": 4.7873349298487345e-06, + "loss": 0.759, + "step": 593 + }, + { + "epoch": 0.930062630480167, + "grad_norm": 0.9377149939537048, + "learning_rate": 4.786483824086231e-06, + "loss": 0.8323, + "step": 594 + }, + { + "epoch": 0.9316283924843424, + "grad_norm": 0.8885690569877625, + "learning_rate": 4.785631094585387e-06, + "loss": 0.8448, + "step": 595 + }, + { + "epoch": 0.9331941544885177, + "grad_norm": 0.9133430123329163, + "learning_rate": 4.784776741951766e-06, + "loss": 0.8135, + "step": 596 + }, + { + "epoch": 0.9347599164926931, + "grad_norm": 1.015851378440857, + "learning_rate": 4.783920766792082e-06, + "loss": 0.8464, + "step": 597 + }, + { + "epoch": 0.9363256784968684, + "grad_norm": 0.8883825540542603, + "learning_rate": 4.783063169714203e-06, + "loss": 0.7445, + "step": 598 + }, + { + "epoch": 0.9378914405010439, + "grad_norm": 0.9872002005577087, + "learning_rate": 4.782203951327149e-06, + "loss": 0.8293, + "step": 599 + }, + { + "epoch": 0.9394572025052192, + "grad_norm": 1.0305813550949097, + "learning_rate": 4.781343112241091e-06, + "loss": 0.8606, + "step": 600 + }, + { + "epoch": 0.9410229645093946, + "grad_norm": 0.9667104482650757, + "learning_rate": 4.780480653067351e-06, + "loss": 0.8057, + "step": 601 + }, + { + "epoch": 0.94258872651357, + "grad_norm": 1.067185640335083, + "learning_rate": 4.7796165744184e-06, + "loss": 0.849, + "step": 602 + }, + { + "epoch": 0.9441544885177453, + "grad_norm": 0.9017406105995178, + "learning_rate": 4.778750876907862e-06, + "loss": 0.7927, + "step": 603 + }, + { + "epoch": 0.9457202505219207, + "grad_norm": 0.9827666878700256, + "learning_rate": 4.777883561150508e-06, + "loss": 0.8087, + "step": 604 + }, + { + "epoch": 0.947286012526096, + "grad_norm": 0.9627811908721924, + "learning_rate": 4.77701462776226e-06, + "loss": 0.8488, + "step": 605 + }, + { + "epoch": 0.9488517745302714, + "grad_norm": 1.0321338176727295, + "learning_rate": 4.776144077360186e-06, + "loss": 0.8289, + "step": 606 + }, + { + "epoch": 0.9504175365344467, + "grad_norm": 1.0012633800506592, + "learning_rate": 4.775271910562508e-06, + "loss": 0.7184, + "step": 607 + }, + { + "epoch": 0.9519832985386222, + "grad_norm": 0.9186056852340698, + "learning_rate": 4.7743981279885875e-06, + "loss": 0.8727, + "step": 608 + }, + { + "epoch": 0.9535490605427975, + "grad_norm": 0.910044252872467, + "learning_rate": 4.773522730258942e-06, + "loss": 0.8455, + "step": 609 + }, + { + "epoch": 0.9551148225469729, + "grad_norm": 0.9342944025993347, + "learning_rate": 4.77264571799523e-06, + "loss": 0.7548, + "step": 610 + }, + { + "epoch": 0.9566805845511482, + "grad_norm": 0.9609628319740295, + "learning_rate": 4.771767091820259e-06, + "loss": 0.8294, + "step": 611 + }, + { + "epoch": 0.9582463465553236, + "grad_norm": 0.9708508253097534, + "learning_rate": 4.770886852357983e-06, + "loss": 0.7857, + "step": 612 + }, + { + "epoch": 0.9598121085594989, + "grad_norm": 1.0910807847976685, + "learning_rate": 4.7700050002334995e-06, + "loss": 0.8162, + "step": 613 + }, + { + "epoch": 0.9613778705636743, + "grad_norm": 0.9703771471977234, + "learning_rate": 4.769121536073054e-06, + "loss": 0.7205, + "step": 614 + }, + { + "epoch": 0.9629436325678496, + "grad_norm": 1.027938961982727, + "learning_rate": 4.768236460504035e-06, + "loss": 0.7641, + "step": 615 + }, + { + "epoch": 0.964509394572025, + "grad_norm": 1.0003989934921265, + "learning_rate": 4.767349774154974e-06, + "loss": 0.7588, + "step": 616 + }, + { + "epoch": 0.9660751565762005, + "grad_norm": 0.9572570323944092, + "learning_rate": 4.766461477655552e-06, + "loss": 0.8274, + "step": 617 + }, + { + "epoch": 0.9676409185803758, + "grad_norm": 0.977287769317627, + "learning_rate": 4.765571571636586e-06, + "loss": 0.7856, + "step": 618 + }, + { + "epoch": 0.9692066805845512, + "grad_norm": 0.9582562446594238, + "learning_rate": 4.764680056730042e-06, + "loss": 0.8102, + "step": 619 + }, + { + "epoch": 0.9707724425887265, + "grad_norm": 0.9842986464500427, + "learning_rate": 4.763786933569025e-06, + "loss": 0.7515, + "step": 620 + }, + { + "epoch": 0.9723382045929019, + "grad_norm": 0.9955743551254272, + "learning_rate": 4.762892202787782e-06, + "loss": 0.7398, + "step": 621 + }, + { + "epoch": 0.9739039665970772, + "grad_norm": 1.0011107921600342, + "learning_rate": 4.761995865021706e-06, + "loss": 0.8182, + "step": 622 + }, + { + "epoch": 0.9754697286012526, + "grad_norm": 0.9586448669433594, + "learning_rate": 4.761097920907325e-06, + "loss": 0.8434, + "step": 623 + }, + { + "epoch": 0.9770354906054279, + "grad_norm": 0.979123055934906, + "learning_rate": 4.760198371082312e-06, + "loss": 0.7954, + "step": 624 + }, + { + "epoch": 0.9786012526096033, + "grad_norm": 1.0081250667572021, + "learning_rate": 4.75929721618548e-06, + "loss": 0.7931, + "step": 625 + }, + { + "epoch": 0.9801670146137788, + "grad_norm": 1.027185320854187, + "learning_rate": 4.75839445685678e-06, + "loss": 0.7339, + "step": 626 + }, + { + "epoch": 0.9817327766179541, + "grad_norm": 1.0553215742111206, + "learning_rate": 4.757490093737303e-06, + "loss": 0.7639, + "step": 627 + }, + { + "epoch": 0.9832985386221295, + "grad_norm": 0.8578981757164001, + "learning_rate": 4.75658412746928e-06, + "loss": 0.8128, + "step": 628 + }, + { + "epoch": 0.9848643006263048, + "grad_norm": 1.0143071413040161, + "learning_rate": 4.75567655869608e-06, + "loss": 0.8169, + "step": 629 + }, + { + "epoch": 0.9864300626304802, + "grad_norm": 0.9743554592132568, + "learning_rate": 4.754767388062208e-06, + "loss": 0.8179, + "step": 630 + }, + { + "epoch": 0.9879958246346555, + "grad_norm": 1.0276602506637573, + "learning_rate": 4.7538566162133105e-06, + "loss": 0.8009, + "step": 631 + }, + { + "epoch": 0.9895615866388309, + "grad_norm": 0.9296634793281555, + "learning_rate": 4.752944243796167e-06, + "loss": 0.8262, + "step": 632 + }, + { + "epoch": 0.9911273486430062, + "grad_norm": 0.9216791987419128, + "learning_rate": 4.752030271458696e-06, + "loss": 0.8929, + "step": 633 + }, + { + "epoch": 0.9926931106471816, + "grad_norm": 0.9272832274436951, + "learning_rate": 4.751114699849954e-06, + "loss": 0.7396, + "step": 634 + }, + { + "epoch": 0.994258872651357, + "grad_norm": 0.9121203422546387, + "learning_rate": 4.750197529620127e-06, + "loss": 0.8311, + "step": 635 + }, + { + "epoch": 0.9958246346555324, + "grad_norm": 0.999600887298584, + "learning_rate": 4.7492787614205425e-06, + "loss": 0.8125, + "step": 636 + }, + { + "epoch": 0.9973903966597077, + "grad_norm": 1.012978196144104, + "learning_rate": 4.748358395903661e-06, + "loss": 0.7966, + "step": 637 + }, + { + "epoch": 0.9989561586638831, + "grad_norm": 0.9767918586730957, + "learning_rate": 4.747436433723075e-06, + "loss": 0.8024, + "step": 638 + } + ], + "logging_steps": 1, + "max_steps": 3828, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 638, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.6116114125330842e+18, + "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..07c8c9aae2a66d9fb6a8b65759b1cd86d2f7d08b --- /dev/null +++ b/checkpoint-638/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81aeca374ba1e1a5384aca380c5dd9da735cc6d6164e0b5b0681e227cf61f41c +size 7992 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/config.json b/config.json new file mode 100644 index 0000000000000000000000000000000000000000..9ac0c7ce54228ed0dad420ba8128a782afb571de --- /dev/null +++ b/config.json @@ -0,0 +1,51 @@ +{ + "_attn_implementation_autoset": true, + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128009, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "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": "4.50.0", + "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..b480d149dc274c0a23db079bfd38032d90082e49 --- /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": "{%- if messages[0]['role'] == 'system' -%}{%- set system_message = messages[0]['content'] | trim -%}{%- set messages = messages[1:] -%}{%- else -%}{%- set system_message = '' -%}{%- endif -%}{%- if tools is not none -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message -}} {{- '\n\n' if system_message else '' -}} {{- '[' -}} {% for t in tools %}{{- (t.function if t.function is defined else t) | tojson() -}}{{- ', ' if not loop.last else '' -}}{%- endfor -%} {{- ']' -}} {{- '<|eot_id|>' -}}{%- else -%}{{- '<|begin_of_text|><|start_header_id|>system<|end_header_id|>' + '\n\n' + system_message + '<|eot_id|>' -}}{%- endif -%}{%- for message in messages -%}{%- if (message['role'] in ['user', 'tool']) != (loop.index0 % 2 == 0) -%}{{- raise_exception('Conversation roles must alternate between user/tool and assistant') -}}{%- elif message['role'] == 'user' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- elif message['role'] == 'tool' -%}{%- set tool_response = '[' + message['content'] | trim + ']' -%}{{- '<|start_header_id|>user<|end_header_id|>' + '\n\n' + tool_response + '<|eot_id|>' -}}{%- elif message['role'] == 'assistant' and message.get('tool_calls') is not none -%}{%- set tool_calls = message['tool_calls'] -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + '[' -}}{%- for tool_call in tool_calls -%}{{ '{' + '\"name\": \"' + tool_call.function.name + '\", \"arguments\": ' + tool_call.function.arguments | tojson + '}' }}{%- if not loop.last -%}{{ ', ' }}{%- else -%}{{ ']' + '<|eot_id|>' }}{%- endif -%}{%- endfor -%}{%- elif message['role'] == 'assistant' -%}{{- '<|start_header_id|>assistant<|end_header_id|>' + '\n\n' + message['content'] | trim + '<|eot_id|>' -}}{%- endif -%}{%- 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" +}