Upload folder using huggingface_hub

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,58 @@
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+base_model: abacusai/Smaug-72B-v0.1
+model-index:
+- name: train_2024-02-17-03-49-55
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# train_2024-02-17-03-49-55
+
+This model is a fine-tuned version of [abacusai/Smaug-72B-v0.1](https://huggingface.co/abacusai/Smaug-72B-v0.1) on the Neural-DPO 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: 0.001
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- num_epochs: 3.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.8.2
+- Transformers 4.37.2
+- Pytorch 2.2.0+cu121
+- Datasets 2.17.0
+- Tokenizers 0.15.2

README.md

@@ -0,0 +1,58 @@
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+base_model: abacusai/Smaug-72B-v0.1
+model-index:
+- name: train_2024-02-17-03-49-55
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# train_2024-02-17-03-49-55
+
+This model is a fine-tuned version of [abacusai/Smaug-72B-v0.1](https://huggingface.co/abacusai/Smaug-72B-v0.1) on the Neural-DPO 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: 0.001
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- num_epochs: 3.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.8.2
+- Transformers 4.37.2
+- Pytorch 2.2.0+cu121
+- Datasets 2.17.0
+- Tokenizers 0.15.2

adapter_config.json

@@ -0,0 +1,27 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "abacusai/Smaug-72B-v0.1",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6440043dbeb9fe7771c6e93f272b86d815d11c916af2fa4a3a87d6c19346e807
+size 83928872

all_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.98,
+    "train_loss": 1.0966757915236733,
+    "train_runtime": 4531.1008,
+    "train_samples_per_second": 0.703,
+    "train_steps_per_second": 0.087
+}

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|im_end|>",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,1680 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|extra_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|extra_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|extra_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|extra_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|extra_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|extra_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|extra_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|extra_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|extra_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|extra_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|extra_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<|extra_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151658": {
+      "content": "<|extra_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151659": {
+      "content": "<|extra_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151660": {
+      "content": "<|extra_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151661": {
+      "content": "<|extra_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151662": {
+      "content": "<|extra_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151663": {
+      "content": "<|extra_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151664": {
+      "content": "<|extra_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151665": {
+      "content": "<|extra_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151666": {
+      "content": "<|extra_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151667": {
+      "content": "<|extra_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151668": {
+      "content": "<|extra_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151669": {
+      "content": "<|extra_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|extra_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<|extra_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151672": {
+      "content": "<|extra_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151673": {
+      "content": "<|extra_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151674": {
+      "content": "<|extra_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151675": {
+      "content": "<|extra_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151676": {
+      "content": "<|extra_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151677": {
+      "content": "<|extra_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151678": {
+      "content": "<|extra_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151679": {
+      "content": "<|extra_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151680": {
+      "content": "<|extra_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151681": {
+      "content": "<|extra_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151682": {
+      "content": "<|extra_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151683": {
+      "content": "<|extra_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151684": {
+      "content": "<|extra_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151685": {
+      "content": "<|extra_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151686": {
+      "content": "<|extra_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151687": {
+      "content": "<|extra_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151688": {
+      "content": "<|extra_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151689": {
+      "content": "<|extra_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151690": {
+      "content": "<|extra_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151691": {
+      "content": "<|extra_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151692": {
+      "content": "<|extra_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151693": {
+      "content": "<|extra_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151694": {
+      "content": "<|extra_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151695": {
+      "content": "<|extra_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151696": {
+      "content": "<|extra_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151697": {
+      "content": "<|extra_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151698": {
+      "content": "<|extra_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151699": {
+      "content": "<|extra_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151700": {
+      "content": "<|extra_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151701": {
+      "content": "<|extra_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151702": {
+      "content": "<|extra_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151703": {
+      "content": "<|extra_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151704": {
+      "content": "<|extra_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151705": {
+      "content": "<|extra_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151706": {
+      "content": "<|extra_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151707": {
+      "content": "<|extra_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151708": {
+      "content": "<|extra_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151709": {
+      "content": "<|extra_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151710": {
+      "content": "<|extra_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151711": {
+      "content": "<|extra_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151712": {
+      "content": "<|extra_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151713": {
+      "content": "<|extra_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151714": {
+      "content": "<|extra_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151715": {
+      "content": "<|extra_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151716": {
+      "content": "<|extra_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151717": {
+      "content": "<|extra_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151718": {
+      "content": "<|extra_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151719": {
+      "content": "<|extra_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151720": {
+      "content": "<|extra_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151721": {
+      "content": "<|extra_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151722": {
+      "content": "<|extra_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151723": {
+      "content": "<|extra_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151724": {
+      "content": "<|extra_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151725": {
+      "content": "<|extra_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151726": {
+      "content": "<|extra_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151727": {
+      "content": "<|extra_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151728": {
+      "content": "<|extra_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151729": {
+      "content": "<|extra_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151730": {
+      "content": "<|extra_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151731": {
+      "content": "<|extra_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151732": {
+      "content": "<|extra_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151733": {
+      "content": "<|extra_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151734": {
+      "content": "<|extra_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151735": {
+      "content": "<|extra_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151736": {
+      "content": "<|extra_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151737": {
+      "content": "<|extra_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151738": {
+      "content": "<|extra_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151739": {
+      "content": "<|extra_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151740": {
+      "content": "<|extra_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151741": {
+      "content": "<|extra_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151742": {
+      "content": "<|extra_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151743": {
+      "content": "<|extra_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151744": {
+      "content": "<|extra_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151745": {
+      "content": "<|extra_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151746": {
+      "content": "<|extra_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151747": {
+      "content": "<|extra_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151748": {
+      "content": "<|extra_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151749": {
+      "content": "<|extra_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151750": {
+      "content": "<|extra_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151751": {
+      "content": "<|extra_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151752": {
+      "content": "<|extra_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151753": {
+      "content": "<|extra_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151754": {
+      "content": "<|extra_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151755": {
+      "content": "<|extra_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151756": {
+      "content": "<|extra_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151757": {
+      "content": "<|extra_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151758": {
+      "content": "<|extra_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151759": {
+      "content": "<|extra_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151760": {
+      "content": "<|extra_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151761": {
+      "content": "<|extra_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151762": {
+      "content": "<|extra_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151763": {
+      "content": "<|extra_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151764": {
+      "content": "<|extra_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151765": {
+      "content": "<|extra_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151766": {
+      "content": "<|extra_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151767": {
+      "content": "<|extra_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151768": {
+      "content": "<|extra_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151769": {
+      "content": "<|extra_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151770": {
+      "content": "<|extra_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151771": {
+      "content": "<|extra_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151772": {
+      "content": "<|extra_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151773": {
+      "content": "<|extra_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151774": {
+      "content": "<|extra_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151775": {
+      "content": "<|extra_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151776": {
+      "content": "<|extra_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151777": {
+      "content": "<|extra_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151778": {
+      "content": "<|extra_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151779": {
+      "content": "<|extra_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151780": {
+      "content": "<|extra_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151781": {
+      "content": "<|extra_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151782": {
+      "content": "<|extra_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151783": {
+      "content": "<|extra_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151784": {
+      "content": "<|extra_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151785": {
+      "content": "<|extra_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151786": {
+      "content": "<|extra_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151787": {
+      "content": "<|extra_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151788": {
+      "content": "<|extra_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151789": {
+      "content": "<|extra_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151790": {
+      "content": "<|extra_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151791": {
+      "content": "<|extra_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151792": {
+      "content": "<|extra_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151793": {
+      "content": "<|extra_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151794": {
+      "content": "<|extra_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151795": {
+      "content": "<|extra_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151796": {
+      "content": "<|extra_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151797": {
+      "content": "<|extra_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151798": {
+      "content": "<|extra_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151799": {
+      "content": "<|extra_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151800": {
+      "content": "<|extra_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151801": {
+      "content": "<|extra_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151802": {
+      "content": "<|extra_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151803": {
+      "content": "<|extra_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151804": {
+      "content": "<|extra_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151805": {
+      "content": "<|extra_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151806": {
+      "content": "<|extra_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151807": {
+      "content": "<|extra_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151808": {
+      "content": "<|extra_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151809": {
+      "content": "<|extra_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151810": {
+      "content": "<|extra_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151811": {
+      "content": "<|extra_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151812": {
+      "content": "<|extra_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151813": {
+      "content": "<|extra_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151814": {
+      "content": "<|extra_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151815": {
+      "content": "<|extra_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151816": {
+      "content": "<|extra_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151817": {
+      "content": "<|extra_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151818": {
+      "content": "<|extra_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151819": {
+      "content": "<|extra_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151820": {
+      "content": "<|extra_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151821": {
+      "content": "<|extra_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151822": {
+      "content": "<|extra_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151823": {
+      "content": "<|extra_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151824": {
+      "content": "<|extra_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151825": {
+      "content": "<|extra_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151826": {
+      "content": "<|extra_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151827": {
+      "content": "<|extra_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151828": {
+      "content": "<|extra_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151829": {
+      "content": "<|extra_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151830": {
+      "content": "<|extra_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151831": {
+      "content": "<|extra_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151832": {
+      "content": "<|extra_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151833": {
+      "content": "<|extra_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151834": {
+      "content": "<|extra_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151835": {
+      "content": "<|extra_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151836": {
+      "content": "<|extra_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151837": {
+      "content": "<|extra_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151838": {
+      "content": "<|extra_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151839": {
+      "content": "<|extra_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151840": {
+      "content": "<|extra_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151841": {
+      "content": "<|extra_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151842": {
+      "content": "<|extra_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151843": {
+      "content": "<|extra_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151844": {
+      "content": "<|extra_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151845": {
+      "content": "<|extra_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151846": {
+      "content": "<|extra_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151847": {
+      "content": "<|extra_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151848": {
+      "content": "<|extra_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151849": {
+      "content": "<|extra_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151850": {
+      "content": "<|extra_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 4096,
+  "pad_token": "<|im_end|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.98,
+    "train_loss": 1.0966757915236733,
+    "train_runtime": 4531.1008,
+    "train_samples_per_second": 0.703,
+    "train_steps_per_second": 0.087
+}

trainer_log.jsonl

@@ -0,0 +1,80 @@
+{"current_steps": 5, "total_steps": 396, "loss": 2.8429, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009996066923030483, "epoch": 0.04, "percentage": 1.26, "elapsed_time": "0:00:51", "remaining_time": "1:07:21"}
+{"current_steps": 10, "total_steps": 396, "loss": 1.7586, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009984273879759713, "epoch": 0.08, "percentage": 2.53, "elapsed_time": "0:01:40", "remaining_time": "1:04:35"}
+{"current_steps": 15, "total_steps": 396, "loss": 1.7084, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009964639423366442, "epoch": 0.11, "percentage": 3.79, "elapsed_time": "0:02:29", "remaining_time": "1:03:07"}
+{"current_steps": 20, "total_steps": 396, "loss": 1.7022, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009937194443381972, "epoch": 0.15, "percentage": 5.05, "elapsed_time": "0:03:17", "remaining_time": "1:01:57"}
+{"current_steps": 25, "total_steps": 396, "loss": 1.436, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009901982117093786, "epoch": 0.19, "percentage": 6.31, "elapsed_time": "0:04:06", "remaining_time": "1:00:55"}
+{"current_steps": 30, "total_steps": 396, "loss": 1.5345, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.000985905784161771, "epoch": 0.23, "percentage": 7.58, "elapsed_time": "0:04:55", "remaining_time": "1:00:00"}
+{"current_steps": 35, "total_steps": 396, "loss": 1.5542, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009808489146745465, "epoch": 0.26, "percentage": 8.84, "elapsed_time": "0:05:43", "remaining_time": "0:59:03"}
+{"current_steps": 40, "total_steps": 396, "loss": 1.5173, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009750355588704727, "epoch": 0.3, "percentage": 10.1, "elapsed_time": "0:06:32", "remaining_time": "0:58:08"}
+{"current_steps": 45, "total_steps": 396, "loss": 1.4094, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.000968474862499881, "epoch": 0.34, "percentage": 11.36, "elapsed_time": "0:07:20", "remaining_time": "0:57:18"}
+{"current_steps": 50, "total_steps": 396, "loss": 1.6779, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009611771470522907, "epoch": 0.38, "percentage": 12.63, "elapsed_time": "0:08:08", "remaining_time": "0:56:23"}
+{"current_steps": 55, "total_steps": 396, "loss": 1.593, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009531538935183251, "epoch": 0.41, "percentage": 13.89, "elapsed_time": "0:08:57", "remaining_time": "0:55:30"}
+{"current_steps": 60, "total_steps": 396, "loss": 1.8026, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009444177243274617, "epoch": 0.45, "percentage": 15.15, "elapsed_time": "0:09:45", "remaining_time": "0:54:39"}
+{"current_steps": 65, "total_steps": 396, "loss": 1.1646, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009349823834900395, "epoch": 0.49, "percentage": 16.41, "elapsed_time": "0:10:36", "remaining_time": "0:54:00"}
+{"current_steps": 70, "total_steps": 396, "loss": 1.4566, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009248627149747573, "epoch": 0.53, "percentage": 17.68, "elapsed_time": "0:11:24", "remaining_time": "0:53:08"}
+{"current_steps": 75, "total_steps": 396, "loss": 1.3848, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009140746393556853, "epoch": 0.56, "percentage": 18.94, "elapsed_time": "0:12:13", "remaining_time": "0:52:19"}
+{"current_steps": 80, "total_steps": 396, "loss": 1.6641, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0009026351287655293, "epoch": 0.6, "percentage": 20.2, "elapsed_time": "0:13:01", "remaining_time": "0:51:28"}
+{"current_steps": 85, "total_steps": 396, "loss": 1.2664, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0008905621801945467, "epoch": 0.64, "percentage": 21.46, "elapsed_time": "0:13:50", "remaining_time": "0:50:40"}
+{"current_steps": 90, "total_steps": 396, "loss": 1.4103, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0008778747871771292, "epoch": 0.68, "percentage": 22.73, "elapsed_time": "0:14:40", "remaining_time": "0:49:52"}
+{"current_steps": 95, "total_steps": 396, "loss": 1.3391, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0008645929099105886, "epoch": 0.72, "percentage": 23.99, "elapsed_time": "0:15:28", "remaining_time": "0:49:03"}
+{"current_steps": 100, "total_steps": 396, "loss": 1.3329, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0008507374438531607, "epoch": 0.75, "percentage": 25.25, "elapsed_time": "0:16:17", "remaining_time": "0:48:13"}
+{"current_steps": 105, "total_steps": 396, "loss": 1.4708, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0008363301868506264, "epoch": 0.79, "percentage": 26.52, "elapsed_time": "0:17:07", "remaining_time": "0:47:26"}
+{"current_steps": 110, "total_steps": 396, "loss": 1.5326, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0008213938048432696, "epoch": 0.83, "percentage": 27.78, "elapsed_time": "0:17:56", "remaining_time": "0:46:37"}
+{"current_steps": 115, "total_steps": 396, "loss": 1.5317, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0008059517962071233, "epoch": 0.87, "percentage": 29.04, "elapsed_time": "0:18:45", "remaining_time": "0:45:49"}
+{"current_steps": 120, "total_steps": 396, "loss": 1.2481, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0007900284547855992, "epoch": 0.9, "percentage": 30.3, "elapsed_time": "0:19:34", "remaining_time": "0:45:00"}
+{"current_steps": 125, "total_steps": 396, "loss": 1.4773, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0007736488316696662, "epoch": 0.94, "percentage": 31.57, "elapsed_time": "0:20:23", "remaining_time": "0:44:11"}
+{"current_steps": 130, "total_steps": 396, "loss": 1.2638, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0007568386957867032, "epoch": 0.98, "percentage": 32.83, "elapsed_time": "0:21:11", "remaining_time": "0:43:22"}
+{"current_steps": 135, "total_steps": 396, "loss": 1.2867, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0007396244933600284, "epoch": 1.02, "percentage": 34.09, "elapsed_time": "0:22:00", "remaining_time": "0:42:32"}
+{"current_steps": 140, "total_steps": 396, "loss": 1.0248, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0007220333063028871, "epoch": 1.05, "percentage": 35.35, "elapsed_time": "0:22:48", "remaining_time": "0:41:43"}
+{"current_steps": 145, "total_steps": 396, "loss": 1.0408, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0007040928096123516, "epoch": 1.09, "percentage": 36.62, "elapsed_time": "0:23:37", "remaining_time": "0:40:54"}
+{"current_steps": 150, "total_steps": 396, "loss": 1.1847, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0006858312278301637, "epoch": 1.13, "percentage": 37.88, "elapsed_time": "0:24:26", "remaining_time": "0:40:05"}
+{"current_steps": 155, "total_steps": 396, "loss": 0.9884, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0006672772906390176, "epoch": 1.17, "percentage": 39.14, "elapsed_time": "0:25:15", "remaining_time": "0:39:15"}
+{"current_steps": 160, "total_steps": 396, "loss": 1.0086, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0006484601876641375, "epoch": 1.21, "percentage": 40.4, "elapsed_time": "0:26:03", "remaining_time": "0:38:26"}
+{"current_steps": 165, "total_steps": 396, "loss": 1.0438, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0006294095225512603, "epoch": 1.24, "percentage": 41.67, "elapsed_time": "0:26:52", "remaining_time": "0:37:37"}
+{"current_steps": 170, "total_steps": 396, "loss": 0.9305, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0006101552663932703, "epoch": 1.28, "percentage": 42.93, "elapsed_time": "0:27:40", "remaining_time": "0:36:48"}
+{"current_steps": 175, "total_steps": 396, "loss": 1.0619, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0005907277105787513, "epoch": 1.32, "percentage": 44.19, "elapsed_time": "0:28:30", "remaining_time": "0:35:59"}
+{"current_steps": 180, "total_steps": 396, "loss": 1.5694, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0005711574191366427, "epoch": 1.36, "percentage": 45.45, "elapsed_time": "0:29:18", "remaining_time": "0:35:10"}
+{"current_steps": 185, "total_steps": 396, "loss": 1.0704, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0005514751806519673, "epoch": 1.39, "percentage": 46.72, "elapsed_time": "0:30:07", "remaining_time": "0:34:22"}
+{"current_steps": 190, "total_steps": 396, "loss": 1.1666, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0005317119598282822, "epoch": 1.43, "percentage": 47.98, "elapsed_time": "0:30:56", "remaining_time": "0:33:33"}
+{"current_steps": 195, "total_steps": 396, "loss": 0.9914, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0005118988487730537, "epoch": 1.47, "percentage": 49.24, "elapsed_time": "0:31:45", "remaining_time": "0:32:44"}
+{"current_steps": 200, "total_steps": 396, "loss": 1.0523, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.000492067018082596, "epoch": 1.51, "percentage": 50.51, "elapsed_time": "0:32:34", "remaining_time": "0:31:55"}
+{"current_steps": 205, "total_steps": 396, "loss": 0.9491, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00047224766780353, "epoch": 1.54, "percentage": 51.77, "elapsed_time": "0:33:23", "remaining_time": "0:31:06"}
+{"current_steps": 210, "total_steps": 396, "loss": 1.041, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004524719783479088, "epoch": 1.58, "percentage": 53.03, "elapsed_time": "0:34:12", "remaining_time": "0:30:17"}
+{"current_steps": 215, "total_steps": 396, "loss": 1.237, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004327710614392341, "epoch": 1.62, "percentage": 54.29, "elapsed_time": "0:35:02", "remaining_time": "0:29:29"}
+{"current_steps": 220, "total_steps": 396, "loss": 1.0301, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00041317591116653486, "epoch": 1.66, "percentage": 55.56, "elapsed_time": "0:35:51", "remaining_time": "0:28:41"}
+{"current_steps": 225, "total_steps": 396, "loss": 1.0687, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00039371735522351166, "epoch": 1.69, "percentage": 56.82, "elapsed_time": "0:36:40", "remaining_time": "0:27:52"}
+{"current_steps": 230, "total_steps": 396, "loss": 0.9959, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00037442600640946044, "epoch": 1.73, "percentage": 58.08, "elapsed_time": "0:37:29", "remaining_time": "0:27:03"}
+{"current_steps": 235, "total_steps": 396, "loss": 0.9951, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0003553322144682737, "epoch": 1.77, "percentage": 59.34, "elapsed_time": "0:38:18", "remaining_time": "0:26:14"}
+{"current_steps": 240, "total_steps": 396, "loss": 1.1158, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0003364660183412892, "epoch": 1.81, "percentage": 60.61, "elapsed_time": "0:39:06", "remaining_time": "0:25:25"}
+{"current_steps": 245, "total_steps": 396, "loss": 1.2224, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0003178570989091028, "epoch": 1.85, "percentage": 61.87, "elapsed_time": "0:39:55", "remaining_time": "0:24:36"}
+{"current_steps": 250, "total_steps": 396, "loss": 1.0617, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00029953473229669324, "epoch": 1.88, "percentage": 63.13, "elapsed_time": "0:40:43", "remaining_time": "0:23:47"}
+{"current_steps": 255, "total_steps": 396, "loss": 0.9955, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0002815277438153203, "epoch": 1.92, "percentage": 64.39, "elapsed_time": "0:41:32", "remaining_time": "0:22:58"}
+{"current_steps": 260, "total_steps": 396, "loss": 0.9129, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0002638644626136587, "epoch": 1.96, "percentage": 65.66, "elapsed_time": "0:42:21", "remaining_time": "0:22:09"}
+{"current_steps": 265, "total_steps": 396, "loss": 1.0105, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00024657267710950857, "epoch": 2.0, "percentage": 66.92, "elapsed_time": "0:43:10", "remaining_time": "0:21:20"}
+{"current_steps": 270, "total_steps": 396, "loss": 0.7617, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0002296795912722014, "epoch": 2.03, "percentage": 68.18, "elapsed_time": "0:43:59", "remaining_time": "0:20:31"}
+{"current_steps": 275, "total_steps": 396, "loss": 0.629, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00021321178182447708, "epoch": 2.07, "percentage": 69.44, "elapsed_time": "0:44:48", "remaining_time": "0:19:42"}
+{"current_steps": 280, "total_steps": 396, "loss": 0.7202, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00019719515643116677, "epoch": 2.11, "percentage": 70.71, "elapsed_time": "0:45:37", "remaining_time": "0:18:54"}
+{"current_steps": 285, "total_steps": 396, "loss": 0.7211, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00018165491294045593, "epoch": 2.15, "percentage": 71.97, "elapsed_time": "0:46:26", "remaining_time": "0:18:05"}
+{"current_steps": 290, "total_steps": 396, "loss": 0.6464, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00016661549974185424, "epoch": 2.18, "percentage": 73.23, "elapsed_time": "0:47:15", "remaining_time": "0:17:16"}
+{"current_steps": 295, "total_steps": 396, "loss": 0.6799, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00015210057730323618, "epoch": 2.22, "percentage": 74.49, "elapsed_time": "0:48:05", "remaining_time": "0:16:27"}
+{"current_steps": 300, "total_steps": 396, "loss": 0.6218, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0001381329809474649, "epoch": 2.26, "percentage": 75.76, "elapsed_time": "0:48:54", "remaining_time": "0:15:38"}
+{"current_steps": 305, "total_steps": 396, "loss": 0.7277, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00012473468492715895, "epoch": 2.3, "percentage": 77.02, "elapsed_time": "0:49:43", "remaining_time": "0:14:50"}
+{"current_steps": 310, "total_steps": 396, "loss": 0.6244, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00011192676785412154, "epoch": 2.34, "percentage": 78.28, "elapsed_time": "0:50:32", "remaining_time": "0:14:01"}
+{"current_steps": 315, "total_steps": 396, "loss": 0.6352, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.972937953781985e-05, "epoch": 2.37, "percentage": 79.55, "elapsed_time": "0:51:21", "remaining_time": "0:13:12"}
+{"current_steps": 320, "total_steps": 396, "loss": 0.6462, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.816170928508365e-05, "epoch": 2.41, "percentage": 80.81, "elapsed_time": "0:52:09", "remaining_time": "0:12:23"}
+{"current_steps": 325, "total_steps": 396, "loss": 0.6377, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.724195571089787e-05, "epoch": 2.45, "percentage": 82.07, "elapsed_time": "0:52:58", "remaining_time": "0:11:34"}
+{"current_steps": 330, "total_steps": 396, "loss": 0.7165, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.698729810778065e-05, "epoch": 2.49, "percentage": 83.33, "elapsed_time": "0:53:46", "remaining_time": "0:10:45"}
+{"current_steps": 335, "total_steps": 396, "loss": 0.6948, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.741386941879179e-05, "epoch": 2.52, "percentage": 84.6, "elapsed_time": "0:54:34", "remaining_time": "0:09:56"}
+{"current_steps": 340, "total_steps": 396, "loss": 0.6849, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.853673085668947e-05, "epoch": 2.56, "percentage": 85.86, "elapsed_time": "0:55:22", "remaining_time": "0:09:07"}
+{"current_steps": 345, "total_steps": 396, "loss": 0.6742, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.036984820916722e-05, "epoch": 2.6, "percentage": 87.12, "elapsed_time": "0:56:11", "remaining_time": "0:08:18"}
+{"current_steps": 350, "total_steps": 396, "loss": 0.7078, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.292606986744667e-05, "epoch": 2.64, "percentage": 88.38, "elapsed_time": "0:57:00", "remaining_time": "0:07:29"}
+{"current_steps": 355, "total_steps": 396, "loss": 0.6499, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6217106612792528e-05, "epoch": 2.67, "percentage": 89.65, "elapsed_time": "0:57:49", "remaining_time": "0:06:40"}
+{"current_steps": 360, "total_steps": 396, "loss": 0.6878, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.025351319275137e-05, "epoch": 2.71, "percentage": 90.91, "elapsed_time": "0:58:37", "remaining_time": "0:05:51"}
+{"current_steps": 365, "total_steps": 396, "loss": 0.5469, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5044671716097413e-05, "epoch": 2.75, "percentage": 92.17, "elapsed_time": "0:59:26", "remaining_time": "0:05:02"}
+{"current_steps": 370, "total_steps": 396, "loss": 0.7233, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0598776892610684e-05, "epoch": 2.79, "percentage": 93.43, "elapsed_time": "1:00:15", "remaining_time": "0:04:14"}
+{"current_steps": 375, "total_steps": 396, "loss": 0.7482, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.9228231409067535e-06, "epoch": 2.82, "percentage": 94.7, "elapsed_time": "1:01:03", "remaining_time": "0:03:25"}
+{"current_steps": 380, "total_steps": 396, "loss": 0.6865, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.02259358460233e-06, "epoch": 2.86, "percentage": 95.96, "elapsed_time": "1:01:52", "remaining_time": "0:02:36"}
+{"current_steps": 385, "total_steps": 396, "loss": 0.6953, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9026509541272275e-06, "epoch": 2.9, "percentage": 97.22, "elapsed_time": "1:02:41", "remaining_time": "0:01:47"}
+{"current_steps": 390, "total_steps": 396, "loss": 0.671, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.663304084960185e-07, "epoch": 2.94, "percentage": 98.48, "elapsed_time": "1:03:30", "remaining_time": "0:00:58"}
+{"current_steps": 395, "total_steps": 396, "loss": 0.6152, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.573428833345769e-08, "epoch": 2.98, "percentage": 99.75, "elapsed_time": "1:04:19", "remaining_time": "0:00:09"}
+{"current_steps": 396, "total_steps": 396, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 2.98, "percentage": 100.0, "elapsed_time": "1:04:28", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,504 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.983050847457627,
+  "eval_steps": 500,
+  "global_step": 396,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.0009996066923030483,
+      "loss": 2.8429,
+      "step": 5
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0009984273879759713,
+      "loss": 1.7586,
+      "step": 10
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0009964639423366442,
+      "loss": 1.7084,
+      "step": 15
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0009937194443381972,
+      "loss": 1.7022,
+      "step": 20
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0009901982117093786,
+      "loss": 1.436,
+      "step": 25
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.000985905784161771,
+      "loss": 1.5345,
+      "step": 30
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0009808489146745465,
+      "loss": 1.5542,
+      "step": 35
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0009750355588704727,
+      "loss": 1.5173,
+      "step": 40
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.000968474862499881,
+      "loss": 1.4094,
+      "step": 45
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.0009611771470522907,
+      "loss": 1.6779,
+      "step": 50
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0009531538935183251,
+      "loss": 1.593,
+      "step": 55
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.0009444177243274617,
+      "loss": 1.8026,
+      "step": 60
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0009349823834900395,
+      "loss": 1.1646,
+      "step": 65
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0009248627149747573,
+      "loss": 1.4566,
+      "step": 70
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0009140746393556853,
+      "loss": 1.3848,
+      "step": 75
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0009026351287655293,
+      "loss": 1.6641,
+      "step": 80
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0008905621801945467,
+      "loss": 1.2664,
+      "step": 85
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.0008778747871771292,
+      "loss": 1.4103,
+      "step": 90
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0008645929099105886,
+      "loss": 1.3391,
+      "step": 95
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.0008507374438531607,
+      "loss": 1.3329,
+      "step": 100
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0008363301868506264,
+      "loss": 1.4708,
+      "step": 105
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.0008213938048432696,
+      "loss": 1.5326,
+      "step": 110
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0008059517962071233,
+      "loss": 1.5317,
+      "step": 115
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.0007900284547855992,
+      "loss": 1.2481,
+      "step": 120
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.0007736488316696662,
+      "loss": 1.4773,
+      "step": 125
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.0007568386957867032,
+      "loss": 1.2638,
+      "step": 130
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.0007396244933600284,
+      "loss": 1.2867,
+      "step": 135
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.0007220333063028871,
+      "loss": 1.0248,
+      "step": 140
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.0007040928096123516,
+      "loss": 1.0408,
+      "step": 145
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.0006858312278301637,
+      "loss": 1.1847,
+      "step": 150
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.0006672772906390176,
+      "loss": 0.9884,
+      "step": 155
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.0006484601876641375,
+      "loss": 1.0086,
+      "step": 160
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.0006294095225512603,
+      "loss": 1.0438,
+      "step": 165
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.0006101552663932703,
+      "loss": 0.9305,
+      "step": 170
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.0005907277105787513,
+      "loss": 1.0619,
+      "step": 175
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.0005711574191366427,
+      "loss": 1.5694,
+      "step": 180
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.0005514751806519673,
+      "loss": 1.0704,
+      "step": 185
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.0005317119598282822,
+      "loss": 1.1666,
+      "step": 190
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.0005118988487730537,
+      "loss": 0.9914,
+      "step": 195
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 0.000492067018082596,
+      "loss": 1.0523,
+      "step": 200
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.00047224766780353,
+      "loss": 0.9491,
+      "step": 205
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 0.0004524719783479088,
+      "loss": 1.041,
+      "step": 210
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 0.0004327710614392341,
+      "loss": 1.237,
+      "step": 215
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 0.00041317591116653486,
+      "loss": 1.0301,
+      "step": 220
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.00039371735522351166,
+      "loss": 1.0687,
+      "step": 225
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 0.00037442600640946044,
+      "loss": 0.9959,
+      "step": 230
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.0003553322144682737,
+      "loss": 0.9951,
+      "step": 235
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.0003364660183412892,
+      "loss": 1.1158,
+      "step": 240
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.0003178570989091028,
+      "loss": 1.2224,
+      "step": 245
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 0.00029953473229669324,
+      "loss": 1.0617,
+      "step": 250
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.0002815277438153203,
+      "loss": 0.9955,
+      "step": 255
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 0.0002638644626136587,
+      "loss": 0.9129,
+      "step": 260
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00024657267710950857,
+      "loss": 1.0105,
+      "step": 265
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 0.0002296795912722014,
+      "loss": 0.7617,
+      "step": 270
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 0.00021321178182447708,
+      "loss": 0.629,
+      "step": 275
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 0.00019719515643116677,
+      "loss": 0.7202,
+      "step": 280
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 0.00018165491294045593,
+      "loss": 0.7211,
+      "step": 285
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 0.00016661549974185424,
+      "loss": 0.6464,
+      "step": 290
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 0.00015210057730323618,
+      "loss": 0.6799,
+      "step": 295
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 0.0001381329809474649,
+      "loss": 0.6218,
+      "step": 300
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 0.00012473468492715895,
+      "loss": 0.7277,
+      "step": 305
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 0.00011192676785412154,
+      "loss": 0.6244,
+      "step": 310
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 9.972937953781985e-05,
+      "loss": 0.6352,
+      "step": 315
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 8.816170928508365e-05,
+      "loss": 0.6462,
+      "step": 320
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 7.724195571089787e-05,
+      "loss": 0.6377,
+      "step": 325
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 6.698729810778065e-05,
+      "loss": 0.7165,
+      "step": 330
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 5.741386941879179e-05,
+      "loss": 0.6948,
+      "step": 335
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 4.853673085668947e-05,
+      "loss": 0.6849,
+      "step": 340
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 4.036984820916722e-05,
+      "loss": 0.6742,
+      "step": 345
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 3.292606986744667e-05,
+      "loss": 0.7078,
+      "step": 350
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 2.6217106612792528e-05,
+      "loss": 0.6499,
+      "step": 355
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 2.025351319275137e-05,
+      "loss": 0.6878,
+      "step": 360
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 1.5044671716097413e-05,
+      "loss": 0.5469,
+      "step": 365
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 1.0598776892610684e-05,
+      "loss": 0.7233,
+      "step": 370
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 6.9228231409067535e-06,
+      "loss": 0.7482,
+      "step": 375
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 4.02259358460233e-06,
+      "loss": 0.6865,
+      "step": 380
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 1.9026509541272275e-06,
+      "loss": 0.6953,
+      "step": 385
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 5.663304084960185e-07,
+      "loss": 0.671,
+      "step": 390
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 1.573428833345769e-08,
+      "loss": 0.6152,
+      "step": 395
+    },
+    {
+      "epoch": 2.98,
+      "step": 396,
+      "total_flos": 1.0914041201688576e+17,
+      "train_loss": 1.0966757915236733,
+      "train_runtime": 4531.1008,
+      "train_samples_per_second": 0.703,
+      "train_steps_per_second": 0.087
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 396,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "total_flos": 1.0914041201688576e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60c825abcb49ce02abce1bb425e5a98b9ba846e1ef0638277e96552fc964127a
+size 4920