Super-squash branch 'main' using huggingface_hub

Co-authored-by: SFconvertbot <SFconvertbot@users.noreply.huggingface.co>
Co-authored-by: librarian-bot <librarian-bot@users.noreply.huggingface.co>
Co-authored-by: ArthurZ <ArthurZ@users.noreply.huggingface.co>
Co-authored-by: patrickvonplaten <patrickvonplaten@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,28 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zstandard filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+model.safetensors filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

README.md

@@ -0,0 +1,112 @@
+---
+license:
+- other
+- apache-2.0
+library_name: transformers
+tags:
+- generated_from_trainer
+- text-generation
+- OPT
+- non-commercial
+- dialogue
+- chatbot
+- ai-msgbot
+pipeline_tag: text-generation
+widget:
+- text: 'If you could live anywhere, where would it be? peter szemraj:'
+  example_title: live anywhere
+- text: 'What would you sing at Karaoke night? peter szemraj:'
+  example_title: Karaoke
+- text: 'If you could hire someone to help you, would it be with cleaning, cooking,
+    or yard work? peter szemraj:'
+  example_title: help
+- text: 'What form of public transportation do you prefer? (air, boat, train, bus,
+    car, etc.) peter szemraj:'
+  example_title: transportation
+- text: 'What''s your favorite zoo animal? peter szemraj:'
+  example_title: animal
+- text: 'Do you like or dislike surprises? Why or why not? peter szemraj:'
+  example_title: surprises
+- text: 'What celebrity would you like to meet at Starbucks for a cup of coffee? peter
+    szemraj:'
+  example_title: 'celebrity '
+base_model: facebook/opt-2.7b
+---
+
+# pszemraj/opt-peter-2.7B
+
+ <a href="https://colab.research.google.com/gist/pszemraj/4068382a40bbf7aab50638b062bd97a9/opt-peter-2-7b-example-csearch-generation.ipynb">
+  <img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/>
+</a>
+
+This model is a fine-tuned version of [facebook/opt-2.7b](https://huggingface.co/facebook/opt-2.7b) on about 80k WhatsApp/text messages (mine). Please use responsibly :)
+
+Test it out on Google Colab by clicking the button above.
+
+![chatdemo](https://i.imgur.com/1EgQYat.png)
+
+## Model description
+
+- Exploring to see how OPT does in terms of dialogue/conversational applications
+- Seems to do a lot better than GPT-Neo with similar training parameters 
+- you can create your own digital clone and deploy it leveraging [this repository I am working on](https://github.com/pszemraj/ai-msgbot).
+
+### sharded checkpoint
+
+As this model file is 10+ GB, it can impose some constraints with lower RAM runtimes and/or download speeds. To help with this issue, a sharded checkpoint of this model is available [here](https://huggingface.co/pszemraj/opt-peter-2.7B-sharded).
+
+The `pszemraj/opt-peter-2.7B-sharded` model can be used as a drop-in replacement for this one for all use cases.
+
+## Intended uses & limitations
+
+> The base model has a custom license that propagates to this one. **Most importantly, it cannot be used commercially**. Read more here: [facebook/opt-2.7b](https://huggingface.co/facebook/opt-2.7b) 
+
+- the model is probably too large to use via API here. Use in Python with GPU RAM / CPU RAM > 12 GB, Colab notebook linked above.
+  - alternatively, you can message [a bot on telegram](http://t.me/GPTPeter_bot) where I test LLMs for dialogue generation
+- **any statements or claims made by this model do not reflect actual claims/statements by me.** Keep in mind it is a _fine-tuned_ version of the model on my data, so things from pre-training are also present in outputs.
+
+## Training and evaluation data
+
+WhatsApp & iMessage data were parsed using [ai-msgbot](https://github.com/pszemraj/ai-msgbot) and then fed as a text dataset to the HF trainer.
+
+## Training procedure
+
+### Training hyperparameters
+
+**SESSION ONE**
+
+The following hyperparameters were used during training:
+- learning_rate: 4e-05
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 128
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.01
+- num_epochs: 3
+
+**SESSION TWO**
+
+The following hyperparameters were used during training:
+- learning_rate: 1e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- distributed_type: multi-GPU
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.05
+- num_epochs: 4
+
+
+### Framework versions
+
+- Transformers 4.19.2
+- Pytorch 1.10.0+cu113
+- Datasets 2.2.2
+- Tokenizers 0.12.1

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "pszemraj/opt-peter-2.7B",
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "architectures": [
+    "OPTForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 2,
+  "do_layer_norm_before": true,
+  "dropout": 0.1,
+  "eos_token_id": 2,
+  "ffn_dim": 10240,
+  "hidden_size": 2560,
+  "init_std": 0.02,
+  "layerdrop": 0.0,
+  "max_position_embeddings": 2048,
+  "model_type": "opt",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "pad_token_id": 1,
+  "prefix": "</s>",
+  "torch_dtype": "float32",
+  "transformers_version": "4.19.2",
+  "use_cache": false,
+  "vocab_size": 50265,
+  "word_embed_proj_dim": 2560,
+  "_remove_final_layer_norm": true
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "do_sample": true,
+  "eos_token_id": 2,
+  "max_length": 128,
+  "min_length": 16,
+  "pad_token_id": 2,
+  "penalty_alpha": 0.6,
+  "top_k": 4,
+  "transformers_version": "4.26.1"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f4a92f22de7d7d9712f1d8e976c9468151e11b351db1a63da87c84f398e89d3
+size 10606355864

opt-2pt7b-ps_DS-msgs_Ep-3_Bs-8_training_metadata.json

@@ -0,0 +1,101 @@
+{
+   "output_dir":"/content/drive/MyDrive/Programming/hf-trainer/opt-2pt7b-ps_DS-msgs_Ep-3_Bs-8",
+   "overwrite_output_dir":true,
+   "do_train":false,
+   "do_eval":false,
+   "do_predict":false,
+   "evaluation_strategy":"no",
+   "prediction_loss_only":false,
+   "per_device_train_batch_size":8,
+   "per_device_eval_batch_size":8,
+   "per_gpu_train_batch_size":"None",
+   "per_gpu_eval_batch_size":"None",
+   "gradient_accumulation_steps":16,
+   "eval_accumulation_steps":8,
+   "eval_delay":0,
+   "learning_rate":4e-05,
+   "weight_decay":0.1,
+   "adam_beta1":0.9,
+   "adam_beta2":0.999,
+   "adam_epsilon":1e-08,
+   "max_grad_norm":1,
+   "num_train_epochs":3,
+   "max_steps":-1,
+   "lr_scheduler_type":"cosine",
+   "warmup_ratio":0.01,
+   "warmup_steps":0,
+   "log_level":-1,
+   "log_level_replica":-1,
+   "log_on_each_node":true,
+   "logging_dir":"/content/drive/MyDrive/Programming/hf-trainer/opt-2pt7b-ps_DS-msgs_Ep-3_Bs-8/logs",
+   "logging_strategy":"steps",
+   "logging_first_step":false,
+   "logging_steps":5,
+   "logging_nan_inf_filter":true,
+   "save_strategy":"epoch",
+   "save_steps":500,
+   "save_total_limit":1,
+   "save_on_each_node":false,
+   "no_cuda":false,
+   "seed":42,
+   "data_seed":"None",
+   "bf16":true,
+   "fp16":false,
+   "fp16_opt_level":"O1",
+   "half_precision_backend":"amp",
+   "bf16_full_eval":true,
+   "fp16_full_eval":false,
+   "tf32":"None",
+   "local_rank":0,
+   "xpu_backend":"None",
+   "tpu_num_cores":"None",
+   "tpu_metrics_debug":false,
+   "debug":"[]",
+   "dataloader_drop_last":false,
+   "eval_steps":"None",
+   "dataloader_num_workers":0,
+   "past_index":-1,
+   "run_name":"/content/drive/MyDrive/Programming/hf-trainer/opt-2pt7b-ps_DS-msgs_Ep-3_Bs-8",
+   "disable_tqdm":false,
+   "remove_unused_columns":true,
+   "label_names":"None",
+   "load_best_model_at_end":false,
+   "metric_for_best_model":"None",
+   "greater_is_better":"None",
+   "ignore_data_skip":false,
+   "sharded_ddp":"[]",
+   "fsdp":"[]",
+   "fsdp_min_num_params":0,
+   "deepspeed":"ds_config_zero2_bf16.json",
+   "label_smoothing_factor":0.0,
+   "optim":"adamw_hf",
+   "adafactor":false,
+   "group_by_length":false,
+   "length_column_name":"length",
+   "report_to":"['tensorboard']",
+   "ddp_find_unused_parameters":"None",
+   "ddp_bucket_cap_mb":"None",
+   "dataloader_pin_memory":true,
+   "skip_memory_metrics":true,
+   "use_legacy_prediction_loop":false,
+   "push_to_hub":true,
+   "resume_from_checkpoint":"None",
+   "hub_model_id":"opt-2pt7b-ps_DS-msgs_Ep-3_Bs-8",
+   "hub_strategy":"end",
+   "hub_token":"<HUB_TOKEN>",
+   "hub_private_repo":false,
+   "gradient_checkpointing":true,
+   "include_inputs_for_metrics":false,
+   "fp16_backend":"auto",
+   "push_to_hub_model_id":"None",
+   "push_to_hub_organization":"None",
+   "push_to_hub_token":"<PUSH_TO_HUB_TOKEN>",
+   "_n_gpu":1,
+   "mp_parameters":"",
+   "auto_find_batch_size":false,
+   "full_determinism":false,
+   "train_batch_size":8,
+   "eval_batch_size":8,
+   "configs_src":"opt-2pt7b-ps_DS-msgs_Ep-3_Bs-8",
+   "data_tag":"text-file-input"
+}

opt-peter-2pt7B-ps_DS-msgs_Ep-4_Bs-16_training_metadata.json

@@ -0,0 +1,101 @@
+{
+   "output_dir":"/content/drive/MyDrive/Programming/hf-trainer/opt-peter-2pt7B-ps_DS-msgs_Ep-4_Bs-16",
+   "overwrite_output_dir":true,
+   "do_train":false,
+   "do_eval":false,
+   "do_predict":false,
+   "evaluation_strategy":"no",
+   "prediction_loss_only":false,
+   "per_device_train_batch_size":16,
+   "per_device_eval_batch_size":16,
+   "per_gpu_train_batch_size":"None",
+   "per_gpu_eval_batch_size":"None",
+   "gradient_accumulation_steps":4,
+   "eval_accumulation_steps":2,
+   "eval_delay":0,
+   "learning_rate":1e-05,
+   "weight_decay":0.1,
+   "adam_beta1":0.9,
+   "adam_beta2":0.999,
+   "adam_epsilon":1e-08,
+   "max_grad_norm":1,
+   "num_train_epochs":4,
+   "max_steps":-1,
+   "lr_scheduler_type":"cosine",
+   "warmup_ratio":0.05,
+   "warmup_steps":0,
+   "log_level":-1,
+   "log_level_replica":-1,
+   "log_on_each_node":true,
+   "logging_dir":"/content/drive/MyDrive/Programming/hf-trainer/opt-peter-2pt7B-ps_DS-msgs_Ep-4_Bs-16/logs",
+   "logging_strategy":"steps",
+   "logging_first_step":false,
+   "logging_steps":5,
+   "logging_nan_inf_filter":true,
+   "save_strategy":"epoch",
+   "save_steps":500,
+   "save_total_limit":1,
+   "save_on_each_node":false,
+   "no_cuda":false,
+   "seed":42,
+   "data_seed":"None",
+   "bf16":true,
+   "fp16":false,
+   "fp16_opt_level":"O1",
+   "half_precision_backend":"amp",
+   "bf16_full_eval":true,
+   "fp16_full_eval":false,
+   "tf32":"None",
+   "local_rank":0,
+   "xpu_backend":"None",
+   "tpu_num_cores":"None",
+   "tpu_metrics_debug":false,
+   "debug":"[]",
+   "dataloader_drop_last":false,
+   "eval_steps":"None",
+   "dataloader_num_workers":0,
+   "past_index":-1,
+   "run_name":"/content/drive/MyDrive/Programming/hf-trainer/opt-peter-2pt7B-ps_DS-msgs_Ep-4_Bs-16",
+   "disable_tqdm":false,
+   "remove_unused_columns":true,
+   "label_names":"None",
+   "load_best_model_at_end":false,
+   "metric_for_best_model":"None",
+   "greater_is_better":"None",
+   "ignore_data_skip":false,
+   "sharded_ddp":"[]",
+   "fsdp":"[]",
+   "fsdp_min_num_params":0,
+   "deepspeed":"ds_config_zero2_bf16.json",
+   "label_smoothing_factor":0.0,
+   "optim":"adamw_hf",
+   "adafactor":false,
+   "group_by_length":false,
+   "length_column_name":"length",
+   "report_to":"['tensorboard']",
+   "ddp_find_unused_parameters":"None",
+   "ddp_bucket_cap_mb":"None",
+   "dataloader_pin_memory":true,
+   "skip_memory_metrics":true,
+   "use_legacy_prediction_loop":false,
+   "push_to_hub":true,
+   "resume_from_checkpoint":"None",
+   "hub_model_id":"opt-peter-2pt7B-ps_DS-msgs_Ep-4_Bs-16",
+   "hub_strategy":"end",
+   "hub_token":"<HUB_TOKEN>",
+   "hub_private_repo":false,
+   "gradient_checkpointing":true,
+   "include_inputs_for_metrics":false,
+   "fp16_backend":"auto",
+   "push_to_hub_model_id":"None",
+   "push_to_hub_organization":"None",
+   "push_to_hub_token":"<PUSH_TO_HUB_TOKEN>",
+   "_n_gpu":1,
+   "mp_parameters":"",
+   "auto_find_batch_size":false,
+   "full_determinism":false,
+   "train_batch_size":16,
+   "eval_batch_size":16,
+   "configs_src":"opt-peter-2pt7B-ps_DS-msgs_Ep-4_Bs-16",
+   "data_tag":"text-file-input"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6d7c906a2eb9b7a88445e0f8bbedeb997ecfa7b70430a58f59f1269251cf80e
+size 10606359699

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "unk_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "pad_token": "</s>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "errors": "replace", "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_bos_token": true, "special_tokens_map_file": null, "name_or_path": "facebook/opt-2.7b", "model_max_length": 512}

trainer_state.json

@@ -0,0 +1,1498 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9992919995954284,
+  "global_step": 1235,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.0607287449392715e-08,
+      "loss": 1.0392,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.0121457489878543e-07,
+      "loss": 1.0115,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5182186234817813e-07,
+      "loss": 0.99,
+      "step": 15
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0242914979757086e-07,
+      "loss": 1.0143,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5303643724696356e-07,
+      "loss": 1.0301,
+      "step": 25
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0364372469635626e-07,
+      "loss": 0.9651,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 3.54251012145749e-07,
+      "loss": 0.9984,
+      "step": 35
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.048582995951417e-07,
+      "loss": 0.9842,
+      "step": 40
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.5546558704453447e-07,
+      "loss": 1.0033,
+      "step": 45
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 5.060728744939271e-07,
+      "loss": 0.9764,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 5.566801619433199e-07,
+      "loss": 1.018,
+      "step": 55
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 6.072874493927125e-07,
+      "loss": 0.9808,
+      "step": 60
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 6.578947368421053e-07,
+      "loss": 0.9508,
+      "step": 65
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 7.08502024291498e-07,
+      "loss": 0.9654,
+      "step": 70
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 7.591093117408907e-07,
+      "loss": 0.9801,
+      "step": 75
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 8.097165991902834e-07,
+      "loss": 0.9825,
+      "step": 80
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 8.603238866396761e-07,
+      "loss": 1.0046,
+      "step": 85
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.109311740890689e-07,
+      "loss": 0.9797,
+      "step": 90
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.615384615384617e-07,
+      "loss": 0.9904,
+      "step": 95
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.0121457489878542e-06,
+      "loss": 0.987,
+      "step": 100
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.062753036437247e-06,
+      "loss": 0.9782,
+      "step": 105
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.1133603238866398e-06,
+      "loss": 1.0131,
+      "step": 110
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.1639676113360325e-06,
+      "loss": 0.9269,
+      "step": 115
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.214574898785425e-06,
+      "loss": 0.9312,
+      "step": 120
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.265182186234818e-06,
+      "loss": 1.0038,
+      "step": 125
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.3157894736842106e-06,
+      "loss": 0.9896,
+      "step": 130
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.3663967611336035e-06,
+      "loss": 0.9544,
+      "step": 135
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.417004048582996e-06,
+      "loss": 0.9811,
+      "step": 140
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.4676113360323888e-06,
+      "loss": 1.0079,
+      "step": 145
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.5182186234817814e-06,
+      "loss": 0.9216,
+      "step": 150
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.5688259109311743e-06,
+      "loss": 0.9537,
+      "step": 155
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.6194331983805669e-06,
+      "loss": 0.9364,
+      "step": 160
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.6700404858299596e-06,
+      "loss": 0.9203,
+      "step": 165
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.7206477732793522e-06,
+      "loss": 0.9639,
+      "step": 170
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.7712550607287451e-06,
+      "loss": 0.9563,
+      "step": 175
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.8218623481781379e-06,
+      "loss": 0.9413,
+      "step": 180
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.8724696356275304e-06,
+      "loss": 0.9173,
+      "step": 185
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9230769230769234e-06,
+      "loss": 0.9208,
+      "step": 190
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.973684210526316e-06,
+      "loss": 0.9382,
+      "step": 195
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 2.0242914979757085e-06,
+      "loss": 0.8964,
+      "step": 200
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 2.0748987854251012e-06,
+      "loss": 0.9241,
+      "step": 205
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 2.125506072874494e-06,
+      "loss": 0.9458,
+      "step": 210
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 2.1761133603238867e-06,
+      "loss": 0.93,
+      "step": 215
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 2.2267206477732795e-06,
+      "loss": 0.9031,
+      "step": 220
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 2.2773279352226723e-06,
+      "loss": 0.9207,
+      "step": 225
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 2.327935222672065e-06,
+      "loss": 0.9123,
+      "step": 230
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 2.3785425101214578e-06,
+      "loss": 0.9057,
+      "step": 235
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 2.42914979757085e-06,
+      "loss": 0.8909,
+      "step": 240
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 2.4797570850202433e-06,
+      "loss": 0.9171,
+      "step": 245
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 2.530364372469636e-06,
+      "loss": 0.8959,
+      "step": 250
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 2.5809716599190288e-06,
+      "loss": 0.946,
+      "step": 255
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 2.631578947368421e-06,
+      "loss": 0.9071,
+      "step": 260
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 2.682186234817814e-06,
+      "loss": 0.8789,
+      "step": 265
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 2.732793522267207e-06,
+      "loss": 0.9864,
+      "step": 270
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 2.7834008097165994e-06,
+      "loss": 0.8932,
+      "step": 275
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 2.834008097165992e-06,
+      "loss": 0.9064,
+      "step": 280
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 2.8846153846153845e-06,
+      "loss": 0.9127,
+      "step": 285
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 2.9352226720647776e-06,
+      "loss": 0.914,
+      "step": 290
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 2.9858299595141704e-06,
+      "loss": 0.9254,
+      "step": 295
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 3.0364372469635627e-06,
+      "loss": 0.9396,
+      "step": 300
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 3.087044534412956e-06,
+      "loss": 0.9208,
+      "step": 305
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 3.1376518218623487e-06,
+      "loss": 0.9242,
+      "step": 310
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 3.188259109311741e-06,
+      "loss": 0.9057,
+      "step": 315
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 3.2388663967611337e-06,
+      "loss": 0.9632,
+      "step": 320
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 3.289473684210527e-06,
+      "loss": 0.9344,
+      "step": 325
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 3.3400809716599193e-06,
+      "loss": 0.9578,
+      "step": 330
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 3.390688259109312e-06,
+      "loss": 0.947,
+      "step": 335
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 3.4412955465587043e-06,
+      "loss": 0.9344,
+      "step": 340
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 3.4919028340080975e-06,
+      "loss": 0.9472,
+      "step": 345
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 3.5425101214574903e-06,
+      "loss": 1.0034,
+      "step": 350
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 3.5931174089068826e-06,
+      "loss": 0.9558,
+      "step": 355
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 3.6437246963562758e-06,
+      "loss": 0.9274,
+      "step": 360
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.6943319838056685e-06,
+      "loss": 0.968,
+      "step": 365
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.744939271255061e-06,
+      "loss": 0.9543,
+      "step": 370
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.7955465587044536e-06,
+      "loss": 1.0063,
+      "step": 375
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.846153846153847e-06,
+      "loss": 0.9491,
+      "step": 380
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.896761133603239e-06,
+      "loss": 0.8891,
+      "step": 385
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.947368421052632e-06,
+      "loss": 0.9015,
+      "step": 390
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.997975708502025e-06,
+      "loss": 0.9852,
+      "step": 395
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.048582995951417e-06,
+      "loss": 0.9636,
+      "step": 400
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.09919028340081e-06,
+      "loss": 0.9683,
+      "step": 405
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.1497975708502025e-06,
+      "loss": 0.9127,
+      "step": 410
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.200404858299596e-06,
+      "loss": 0.9229,
+      "step": 415
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.251012145748988e-06,
+      "loss": 0.9329,
+      "step": 420
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.30161943319838e-06,
+      "loss": 0.9864,
+      "step": 425
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.3522267206477735e-06,
+      "loss": 0.9208,
+      "step": 430
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.402834008097167e-06,
+      "loss": 0.959,
+      "step": 435
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.453441295546559e-06,
+      "loss": 0.9411,
+      "step": 440
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.504048582995952e-06,
+      "loss": 0.9695,
+      "step": 445
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.5546558704453445e-06,
+      "loss": 0.9808,
+      "step": 450
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.605263157894737e-06,
+      "loss": 0.8951,
+      "step": 455
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.65587044534413e-06,
+      "loss": 0.9815,
+      "step": 460
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.706477732793522e-06,
+      "loss": 0.9565,
+      "step": 465
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.7570850202429155e-06,
+      "loss": 0.9335,
+      "step": 470
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.807692307692308e-06,
+      "loss": 0.9409,
+      "step": 475
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.8582995951417e-06,
+      "loss": 0.9057,
+      "step": 480
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.908906882591093e-06,
+      "loss": 0.9125,
+      "step": 485
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.9595141700404865e-06,
+      "loss": 0.9485,
+      "step": 490
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.9999993758760865e-06,
+      "loss": 0.9827,
+      "step": 495
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.999977531571805e-06,
+      "loss": 0.9135,
+      "step": 500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.999924481383433e-06,
+      "loss": 0.9547,
+      "step": 505
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.9998402259731634e-06,
+      "loss": 0.9506,
+      "step": 510
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.999724766392715e-06,
+      "loss": 0.9281,
+      "step": 515
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.999578104083307e-06,
+      "loss": 0.925,
+      "step": 520
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.999400240875647e-06,
+      "loss": 0.9808,
+      "step": 525
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.999191178989905e-06,
+      "loss": 0.963,
+      "step": 530
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.998950921035691e-06,
+      "loss": 0.9125,
+      "step": 535
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.998679470012015e-06,
+      "loss": 0.9833,
+      "step": 540
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.998376829307255e-06,
+      "loss": 0.949,
+      "step": 545
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.998043002699114e-06,
+      "loss": 0.9539,
+      "step": 550
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.997677994354573e-06,
+      "loss": 0.9612,
+      "step": 555
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.997281808829833e-06,
+      "loss": 0.962,
+      "step": 560
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.996854451070267e-06,
+      "loss": 0.9467,
+      "step": 565
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.996395926410354e-06,
+      "loss": 0.9273,
+      "step": 570
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.995906240573615e-06,
+      "loss": 0.9213,
+      "step": 575
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.995385399672532e-06,
+      "loss": 0.9405,
+      "step": 580
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.994833410208487e-06,
+      "loss": 0.9448,
+      "step": 585
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.994250279071669e-06,
+      "loss": 0.9146,
+      "step": 590
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.9936360135409915e-06,
+      "loss": 0.9891,
+      "step": 595
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.992990621284004e-06,
+      "loss": 0.9444,
+      "step": 600
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.992314110356793e-06,
+      "loss": 0.9599,
+      "step": 605
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.991606489203883e-06,
+      "loss": 1.0091,
+      "step": 610
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.99086776665813e-06,
+      "loss": 0.9725,
+      "step": 615
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.9900979519406154e-06,
+      "loss": 0.9283,
+      "step": 620
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.9892970546605226e-06,
+      "loss": 0.9856,
+      "step": 625
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.988465084815026e-06,
+      "loss": 0.9866,
+      "step": 630
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.987602052789159e-06,
+      "loss": 0.8948,
+      "step": 635
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.986707969355692e-06,
+      "loss": 0.9727,
+      "step": 640
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.985782845674988e-06,
+      "loss": 0.9579,
+      "step": 645
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.9848266932948745e-06,
+      "loss": 0.9343,
+      "step": 650
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.983839524150489e-06,
+      "loss": 0.9872,
+      "step": 655
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.982821350564136e-06,
+      "loss": 0.9586,
+      "step": 660
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.981772185245135e-06,
+      "loss": 0.9687,
+      "step": 665
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.9806920412896555e-06,
+      "loss": 0.9365,
+      "step": 670
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.979580932180556e-06,
+      "loss": 0.9754,
+      "step": 675
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.978438871787219e-06,
+      "loss": 0.9456,
+      "step": 680
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.977265874365374e-06,
+      "loss": 0.9345,
+      "step": 685
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.976061954556921e-06,
+      "loss": 0.9384,
+      "step": 690
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.9748271273897495e-06,
+      "loss": 0.9121,
+      "step": 695
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.9735614082775455e-06,
+      "loss": 0.9196,
+      "step": 700
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.972264813019605e-06,
+      "loss": 0.9427,
+      "step": 705
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.970937357800635e-06,
+      "loss": 0.9248,
+      "step": 710
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.969579059190549e-06,
+      "loss": 0.924,
+      "step": 715
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.968189934144263e-06,
+      "loss": 0.9705,
+      "step": 720
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.966770000001483e-06,
+      "loss": 0.992,
+      "step": 725
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.965319274486488e-06,
+      "loss": 0.9164,
+      "step": 730
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.963837775707911e-06,
+      "loss": 0.9343,
+      "step": 735
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.962325522158509e-06,
+      "loss": 0.9485,
+      "step": 740
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.960782532714934e-06,
+      "loss": 0.9483,
+      "step": 745
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.959208826637502e-06,
+      "loss": 0.959,
+      "step": 750
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.957604423569942e-06,
+      "loss": 0.9819,
+      "step": 755
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.955969343539162e-06,
+      "loss": 0.9399,
+      "step": 760
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.954303606954993e-06,
+      "loss": 0.8887,
+      "step": 765
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.952607234609935e-06,
+      "loss": 0.989,
+      "step": 770
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.950880247678897e-06,
+      "loss": 0.9565,
+      "step": 775
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.949122667718935e-06,
+      "loss": 0.9373,
+      "step": 780
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.947334516668981e-06,
+      "loss": 0.8964,
+      "step": 785
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.945515816849567e-06,
+      "loss": 0.9158,
+      "step": 790
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.9436665909625555e-06,
+      "loss": 0.9352,
+      "step": 795
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.941786862090842e-06,
+      "loss": 0.9417,
+      "step": 800
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.9398766536980795e-06,
+      "loss": 1.0072,
+      "step": 805
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.937935989628377e-06,
+      "loss": 0.9596,
+      "step": 810
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.935964894106011e-06,
+      "loss": 0.9452,
+      "step": 815
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.933963391735111e-06,
+      "loss": 0.9791,
+      "step": 820
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.9319315074993626e-06,
+      "loss": 0.9491,
+      "step": 825
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.929869266761692e-06,
+      "loss": 0.9722,
+      "step": 830
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.9277766952639485e-06,
+      "loss": 0.9667,
+      "step": 835
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.9256538191265836e-06,
+      "loss": 0.9705,
+      "step": 840
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.923500664848327e-06,
+      "loss": 0.9772,
+      "step": 845
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.921317259305852e-06,
+      "loss": 1.0082,
+      "step": 850
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.9191036297534455e-06,
+      "loss": 0.957,
+      "step": 855
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.916859803822662e-06,
+      "loss": 0.9832,
+      "step": 860
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.914585809521982e-06,
+      "loss": 0.9521,
+      "step": 865
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.912281675236461e-06,
+      "loss": 0.9631,
+      "step": 870
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.909947429727378e-06,
+      "loss": 0.9403,
+      "step": 875
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.907583102131871e-06,
+      "loss": 1.0201,
+      "step": 880
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.905188721962579e-06,
+      "loss": 0.9873,
+      "step": 885
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.902764319107271e-06,
+      "loss": 0.9681,
+      "step": 890
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.900309923828474e-06,
+      "loss": 0.9125,
+      "step": 895
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.89782556676309e-06,
+      "loss": 0.9349,
+      "step": 900
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.895311278922023e-06,
+      "loss": 0.9779,
+      "step": 905
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.892767091689786e-06,
+      "loss": 0.9427,
+      "step": 910
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.890193036824107e-06,
+      "loss": 0.9845,
+      "step": 915
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.88758914645554e-06,
+      "loss": 0.9601,
+      "step": 920
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.884955453087056e-06,
+      "loss": 0.9894,
+      "step": 925
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.882291989593644e-06,
+      "loss": 1.0139,
+      "step": 930
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.879598789221893e-06,
+      "loss": 0.9599,
+      "step": 935
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.876875885589589e-06,
+      "loss": 0.9855,
+      "step": 940
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.87412331268528e-06,
+      "loss": 0.9763,
+      "step": 945
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.8713411048678635e-06,
+      "loss": 0.946,
+      "step": 950
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.868529296866156e-06,
+      "loss": 0.9446,
+      "step": 955
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.865687923778452e-06,
+      "loss": 0.9792,
+      "step": 960
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.862817021072096e-06,
+      "loss": 0.9644,
+      "step": 965
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.8599166245830306e-06,
+      "loss": 0.9269,
+      "step": 970
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.856986770515358e-06,
+      "loss": 0.9846,
+      "step": 975
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.854027495440881e-06,
+      "loss": 0.9325,
+      "step": 980
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.851038836298649e-06,
+      "loss": 0.968,
+      "step": 985
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.848020830394498e-06,
+      "loss": 1.0115,
+      "step": 990
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.844973515400584e-06,
+      "loss": 1.0004,
+      "step": 995
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.8418969293549106e-06,
+      "loss": 0.9422,
+      "step": 1000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.83879111066086e-06,
+      "loss": 0.9617,
+      "step": 1005
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.8356560980867064e-06,
+      "loss": 1.0238,
+      "step": 1010
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.832491930765137e-06,
+      "loss": 0.9467,
+      "step": 1015
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.829298648192763e-06,
+      "loss": 0.9567,
+      "step": 1020
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.826076290229625e-06,
+      "loss": 0.9723,
+      "step": 1025
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.822824897098697e-06,
+      "loss": 0.9852,
+      "step": 1030
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.819544509385381e-06,
+      "loss": 0.9636,
+      "step": 1035
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.8162351680370046e-06,
+      "loss": 0.9497,
+      "step": 1040
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.81289691436231e-06,
+      "loss": 1.0008,
+      "step": 1045
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.809529790030931e-06,
+      "loss": 1.0033,
+      "step": 1050
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.806133837072886e-06,
+      "loss": 1.0142,
+      "step": 1055
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.802709097878039e-06,
+      "loss": 0.9727,
+      "step": 1060
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.799255615195582e-06,
+      "loss": 0.9863,
+      "step": 1065
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.795773432133492e-06,
+      "loss": 0.984,
+      "step": 1070
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.792262592158002e-06,
+      "loss": 0.9521,
+      "step": 1075
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.788723139093051e-06,
+      "loss": 0.9497,
+      "step": 1080
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.785155117119742e-06,
+      "loss": 1.0392,
+      "step": 1085
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.781558570775787e-06,
+      "loss": 0.9197,
+      "step": 1090
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.777933544954951e-06,
+      "loss": 1.0343,
+      "step": 1095
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.774280084906498e-06,
+      "loss": 1.0113,
+      "step": 1100
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.770598236234617e-06,
+      "loss": 0.9948,
+      "step": 1105
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.766888044897856e-06,
+      "loss": 0.9312,
+      "step": 1110
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.763149557208554e-06,
+      "loss": 0.9207,
+      "step": 1115
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.759382819832256e-06,
+      "loss": 0.9621,
+      "step": 1120
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.755587879787131e-06,
+      "loss": 1.0067,
+      "step": 1125
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.75176478444339e-06,
+      "loss": 0.9758,
+      "step": 1130
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.747913581522689e-06,
+      "loss": 0.9287,
+      "step": 1135
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.744034319097536e-06,
+      "loss": 0.9803,
+      "step": 1140
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.740127045590692e-06,
+      "loss": 1.0133,
+      "step": 1145
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.736191809774567e-06,
+      "loss": 0.9588,
+      "step": 1150
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.7322286607706056e-06,
+      "loss": 0.9763,
+      "step": 1155
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.72823764804868e-06,
+      "loss": 0.9944,
+      "step": 1160
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.724218821426472e-06,
+      "loss": 0.9897,
+      "step": 1165
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.720172231068845e-06,
+      "loss": 0.9902,
+      "step": 1170
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.716097927487225e-06,
+      "loss": 0.969,
+      "step": 1175
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.711995961538969e-06,
+      "loss": 0.9458,
+      "step": 1180
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.7078663844267245e-06,
+      "loss": 0.9782,
+      "step": 1185
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.7037092476978e-06,
+      "loss": 0.9987,
+      "step": 1190
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.699524603243509e-06,
+      "loss": 1.0171,
+      "step": 1195
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.695312503298535e-06,
+      "loss": 0.9727,
+      "step": 1200
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.69107300044027e-06,
+      "loss": 0.969,
+      "step": 1205
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.686806147588166e-06,
+      "loss": 0.977,
+      "step": 1210
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.6825119980030664e-06,
+      "loss": 0.9552,
+      "step": 1215
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.678190605286546e-06,
+      "loss": 0.9912,
+      "step": 1220
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.673842023380243e-06,
+      "loss": 0.9702,
+      "step": 1225
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.669466306565181e-06,
+      "loss": 0.9792,
+      "step": 1230
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.665063509461098e-06,
+      "loss": 1.0204,
+      "step": 1235
+    }
+  ],
+  "max_steps": 4940,
+  "num_train_epochs": 4,
+  "total_flos": 6.117066158416855e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de2ba60d145cb1598356d13c34552cb02e8a823370db813704fe0b11ee852e92
+size 4271