upload model checkpoint

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,3 +1,76 @@
----
-license: gemma
----
+---
+library_name: peft
+tags:
+- alignment-handbook
+- generated_from_trainer
+datasets:
+- llama-duo/synth_summarize_dataset_dedup
+base_model: google/gemma-7b
+model-index:
+- name: gemma7b-summarize-claude3sonnet-32k
+  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. -->
+
+# gemma7b-summarize-claude3sonnet-32k
+
+This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the llama-duo/synth_summarize_dataset_dedup dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.5524
+
+## 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: 0.0002
+- train_batch_size: 4
+- eval_batch_size: 2
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- total_eval_batch_size: 16
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 10
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 1.6901        | 1.0   | 76   | 2.9966          |
+| 1.1272        | 2.0   | 152  | 2.6070          |
+| 1.0337        | 3.0   | 228  | 2.5657          |
+| 0.9638        | 4.0   | 304  | 2.5379          |
+| 0.9419        | 5.0   | 380  | 2.5376          |
+| 0.9117        | 6.0   | 456  | 2.5333          |
+| 0.8944        | 7.0   | 532  | 2.5417          |
+| 0.8824        | 8.0   | 608  | 2.5474          |
+| 0.8759        | 9.0   | 684  | 2.5541          |
+| 0.8735        | 10.0  | 760  | 2.5524          |
+
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.40.0
+- Pytorch 2.1.2+cu121
+- Datasets 2.18.0
+- Tokenizers 0.19.1

adapter_config.json

@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-7b",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6246c0c074e47dbab1c260de365a1dd3cd5f6dee4b8edc86131bd452b7492d8f
+size 12860096

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 10.0,
+    "eval_loss": 2.5523786544799805,
+    "eval_runtime": 0.2402,
+    "eval_samples": 25,
+    "eval_samples_per_second": 41.632,
+    "eval_steps_per_second": 4.163,
+    "total_flos": 2.318334860390826e+18,
+    "train_loss": 2.2621336485210217,
+    "train_runtime": 1856.2046,
+    "train_samples": 31747,
+    "train_samples_per_second": 26.156,
+    "train_steps_per_second": 0.409
+}

config.json

@@ -0,0 +1,44 @@
+{
+  "_name_or_path": "google/gemma-7b",
+  "architectures": [
+    "GemmaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 2,
+  "eos_token_id": 1,
+  "head_dim": 256,
+  "hidden_act": "gelu",
+  "hidden_activation": null,
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 24576,
+  "max_position_embeddings": 8192,
+  "model_type": "gemma",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 16,
+  "pad_token_id": 0,
+  "quantization_config": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
+    "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_quant_storage": "uint8",
+    "bnb_4bit_quant_type": "nf4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.0",
+  "use_cache": true,
+  "vocab_size": 256000
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "eval_loss": 2.5523786544799805,
+    "eval_runtime": 0.2402,
+    "eval_samples": 25,
+    "eval_samples_per_second": 41.632,
+    "eval_steps_per_second": 4.163
+}

runs/Jun13_05-58-32_gpu1-1/events.out.tfevents.1718229608.gpu1-1.440205.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:496d8947cbc337a0c498ea8911e5c7c0774ac3699b879cee1d7140ea5726b3cf
+size 40721

runs/Jun13_05-58-32_gpu1-1/events.out.tfevents.1718231465.gpu1-1.440205.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b23bca1b61bda1a02121cfeb428edc004551a46a8dde1ba53809071c871f1e94
+size 359

special_tokens_map.json

@@ -0,0 +1,34 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:322a5f52ab5cab196761ab397a022d6fa3a2e1418585e532bb6efb2fedd2ae94
+size 17477501

tokenizer_config.json

@@ -0,0 +1,70 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "106": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{% if messages[0]['role'] == 'user' or messages[0]['role'] == 'system' %}{{ bos_token }}{% endif %}{% for message in messages %}{{ '<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n' }}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% elif messages[-1]['role'] == 'assistant' %}{{ eos_token }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "legacy": null,
+  "model_max_length": 2048,
+  "pad_token": "<pad>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 2.318334860390826e+18,
+    "train_loss": 2.2621336485210217,
+    "train_runtime": 1856.2046,
+    "train_samples": 31747,
+    "train_samples_per_second": 26.156,
+    "train_steps_per_second": 0.409
+}

trainer_state.json

@@ -0,0 +1,1181 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 760,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.013157894736842105,
+      "grad_norm": 251.0,
+      "learning_rate": 2.631578947368421e-06,
+      "loss": 35.372,
+      "step": 1
+    },
+    {
+      "epoch": 0.06578947368421052,
+      "grad_norm": 239.0,
+      "learning_rate": 1.3157894736842106e-05,
+      "loss": 35.3482,
+      "step": 5
+    },
+    {
+      "epoch": 0.13157894736842105,
+      "grad_norm": 83.5,
+      "learning_rate": 2.6315789473684212e-05,
+      "loss": 30.5543,
+      "step": 10
+    },
+    {
+      "epoch": 0.19736842105263158,
+      "grad_norm": 37.5,
+      "learning_rate": 3.9473684210526316e-05,
+      "loss": 23.0935,
+      "step": 15
+    },
+    {
+      "epoch": 0.2631578947368421,
+      "grad_norm": 17.875,
+      "learning_rate": 5.2631578947368424e-05,
+      "loss": 18.9215,
+      "step": 20
+    },
+    {
+      "epoch": 0.32894736842105265,
+      "grad_norm": 11.125,
+      "learning_rate": 6.578947368421054e-05,
+      "loss": 16.7379,
+      "step": 25
+    },
+    {
+      "epoch": 0.39473684210526316,
+      "grad_norm": 3.671875,
+      "learning_rate": 7.894736842105263e-05,
+      "loss": 14.9012,
+      "step": 30
+    },
+    {
+      "epoch": 0.4605263157894737,
+      "grad_norm": 3.09375,
+      "learning_rate": 9.210526315789474e-05,
+      "loss": 14.2412,
+      "step": 35
+    },
+    {
+      "epoch": 0.5263157894736842,
+      "grad_norm": 4.0,
+      "learning_rate": 0.00010526315789473685,
+      "loss": 13.5992,
+      "step": 40
+    },
+    {
+      "epoch": 0.5921052631578947,
+      "grad_norm": 6.15625,
+      "learning_rate": 0.00011842105263157894,
+      "loss": 12.7389,
+      "step": 45
+    },
+    {
+      "epoch": 0.6578947368421053,
+      "grad_norm": 11.875,
+      "learning_rate": 0.00013157894736842108,
+      "loss": 11.4753,
+      "step": 50
+    },
+    {
+      "epoch": 0.7236842105263158,
+      "grad_norm": 19.25,
+      "learning_rate": 0.00014473684210526317,
+      "loss": 8.7885,
+      "step": 55
+    },
+    {
+      "epoch": 0.7894736842105263,
+      "grad_norm": 20.75,
+      "learning_rate": 0.00015789473684210527,
+      "loss": 4.8341,
+      "step": 60
+    },
+    {
+      "epoch": 0.8552631578947368,
+      "grad_norm": 5.90625,
+      "learning_rate": 0.00017105263157894739,
+      "loss": 2.389,
+      "step": 65
+    },
+    {
+      "epoch": 0.9210526315789473,
+      "grad_norm": 3.953125,
+      "learning_rate": 0.00018421052631578948,
+      "loss": 1.9682,
+      "step": 70
+    },
+    {
+      "epoch": 0.9868421052631579,
+      "grad_norm": 1.078125,
+      "learning_rate": 0.00019736842105263157,
+      "loss": 1.6901,
+      "step": 75
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.9966416358947754,
+      "eval_runtime": 0.2381,
+      "eval_samples_per_second": 41.997,
+      "eval_steps_per_second": 4.2,
+      "step": 76
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 1.0234375,
+      "learning_rate": 0.00019998312416333227,
+      "loss": 1.565,
+      "step": 80
+    },
+    {
+      "epoch": 1.118421052631579,
+      "grad_norm": 0.765625,
+      "learning_rate": 0.0001999145758387301,
+      "loss": 1.4481,
+      "step": 85
+    },
+    {
+      "epoch": 1.1842105263157894,
+      "grad_norm": 0.97265625,
+      "learning_rate": 0.00019979333640833947,
+      "loss": 1.3767,
+      "step": 90
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 1.078125,
+      "learning_rate": 0.00019961946980917456,
+      "loss": 1.3246,
+      "step": 95
+    },
+    {
+      "epoch": 1.3157894736842106,
+      "grad_norm": 0.58203125,
+      "learning_rate": 0.00019939306773179497,
+      "loss": 1.2875,
+      "step": 100
+    },
+    {
+      "epoch": 1.381578947368421,
+      "grad_norm": 0.46484375,
+      "learning_rate": 0.00019911424957195158,
+      "loss": 1.244,
+      "step": 105
+    },
+    {
+      "epoch": 1.4473684210526316,
+      "grad_norm": 0.859375,
+      "learning_rate": 0.00019878316236762196,
+      "loss": 1.2335,
+      "step": 110
+    },
+    {
+      "epoch": 1.513157894736842,
+      "grad_norm": 0.78515625,
+      "learning_rate": 0.000198399980721468,
+      "loss": 1.2052,
+      "step": 115
+    },
+    {
+      "epoch": 1.5789473684210527,
+      "grad_norm": 0.59765625,
+      "learning_rate": 0.0001979649067087574,
+      "loss": 1.1933,
+      "step": 120
+    },
+    {
+      "epoch": 1.6447368421052633,
+      "grad_norm": 1.0703125,
+      "learning_rate": 0.00019747816977079671,
+      "loss": 1.1718,
+      "step": 125
+    },
+    {
+      "epoch": 1.7105263157894737,
+      "grad_norm": 1.7734375,
+      "learning_rate": 0.00019694002659393305,
+      "loss": 1.1603,
+      "step": 130
+    },
+    {
+      "epoch": 1.776315789473684,
+      "grad_norm": 0.5703125,
+      "learning_rate": 0.00019635076097418734,
+      "loss": 1.1664,
+      "step": 135
+    },
+    {
+      "epoch": 1.8421052631578947,
+      "grad_norm": 0.765625,
+      "learning_rate": 0.00019571068366759143,
+      "loss": 1.1523,
+      "step": 140
+    },
+    {
+      "epoch": 1.9078947368421053,
+      "grad_norm": 0.51953125,
+      "learning_rate": 0.00019502013222630712,
+      "loss": 1.1372,
+      "step": 145
+    },
+    {
+      "epoch": 1.973684210526316,
+      "grad_norm": 0.703125,
+      "learning_rate": 0.00019427947082061432,
+      "loss": 1.1272,
+      "step": 150
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.6070284843444824,
+      "eval_runtime": 0.2379,
+      "eval_samples_per_second": 42.033,
+      "eval_steps_per_second": 4.203,
+      "step": 152
+    },
+    {
+      "epoch": 2.039473684210526,
+      "grad_norm": 0.6875,
+      "learning_rate": 0.00019348909004686152,
+      "loss": 1.1013,
+      "step": 155
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 1.15625,
+      "learning_rate": 0.00019264940672148018,
+      "loss": 1.1011,
+      "step": 160
+    },
+    {
+      "epoch": 2.1710526315789473,
+      "grad_norm": 0.59765625,
+      "learning_rate": 0.00019176086366117211,
+      "loss": 1.0822,
+      "step": 165
+    },
+    {
+      "epoch": 2.236842105263158,
+      "grad_norm": 1.21875,
+      "learning_rate": 0.00019082392944938466,
+      "loss": 1.0853,
+      "step": 170
+    },
+    {
+      "epoch": 2.3026315789473686,
+      "grad_norm": 0.859375,
+      "learning_rate": 0.0001898390981891979,
+      "loss": 1.0778,
+      "step": 175
+    },
+    {
+      "epoch": 2.3684210526315788,
+      "grad_norm": 0.58203125,
+      "learning_rate": 0.00018880688924275378,
+      "loss": 1.0601,
+      "step": 180
+    },
+    {
+      "epoch": 2.4342105263157894,
+      "grad_norm": 0.8359375,
+      "learning_rate": 0.0001877278469573643,
+      "loss": 1.0645,
+      "step": 185
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 1.0078125,
+      "learning_rate": 0.00018660254037844388,
+      "loss": 1.0663,
+      "step": 190
+    },
+    {
+      "epoch": 2.5657894736842106,
+      "grad_norm": 2.015625,
+      "learning_rate": 0.0001854315629494165,
+      "loss": 1.0582,
+      "step": 195
+    },
+    {
+      "epoch": 2.6315789473684212,
+      "grad_norm": 1.015625,
+      "learning_rate": 0.00018421553219875658,
+      "loss": 1.05,
+      "step": 200
+    },
+    {
+      "epoch": 2.6973684210526314,
+      "grad_norm": 2.375,
+      "learning_rate": 0.00018295508941432815,
+      "loss": 1.0463,
+      "step": 205
+    },
+    {
+      "epoch": 2.763157894736842,
+      "grad_norm": 0.7890625,
+      "learning_rate": 0.0001816508993051943,
+      "loss": 1.0336,
+      "step": 210
+    },
+    {
+      "epoch": 2.8289473684210527,
+      "grad_norm": 0.84375,
+      "learning_rate": 0.0001803036496510752,
+      "loss": 1.0233,
+      "step": 215
+    },
+    {
+      "epoch": 2.8947368421052633,
+      "grad_norm": 0.76171875,
+      "learning_rate": 0.00017891405093963938,
+      "loss": 1.0287,
+      "step": 220
+    },
+    {
+      "epoch": 2.9605263157894735,
+      "grad_norm": 0.75390625,
+      "learning_rate": 0.00017748283599182014,
+      "loss": 1.0337,
+      "step": 225
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.565699338912964,
+      "eval_runtime": 0.2367,
+      "eval_samples_per_second": 42.254,
+      "eval_steps_per_second": 4.225,
+      "step": 228
+    },
+    {
+      "epoch": 3.026315789473684,
+      "grad_norm": 0.9375,
+      "learning_rate": 0.00017601075957535364,
+      "loss": 1.0152,
+      "step": 230
+    },
+    {
+      "epoch": 3.0921052631578947,
+      "grad_norm": 1.109375,
+      "learning_rate": 0.00017449859800674371,
+      "loss": 0.9987,
+      "step": 235
+    },
+    {
+      "epoch": 3.1578947368421053,
+      "grad_norm": 2.046875,
+      "learning_rate": 0.0001729471487418621,
+      "loss": 0.9947,
+      "step": 240
+    },
+    {
+      "epoch": 3.223684210526316,
+      "grad_norm": 0.94921875,
+      "learning_rate": 0.00017135722995540107,
+      "loss": 0.9919,
+      "step": 245
+    },
+    {
+      "epoch": 3.2894736842105265,
+      "grad_norm": 0.99609375,
+      "learning_rate": 0.00016972968010939954,
+      "loss": 0.9945,
+      "step": 250
+    },
+    {
+      "epoch": 3.3552631578947367,
+      "grad_norm": 0.890625,
+      "learning_rate": 0.00016806535751107037,
+      "loss": 0.9904,
+      "step": 255
+    },
+    {
+      "epoch": 3.4210526315789473,
+      "grad_norm": 1.1171875,
+      "learning_rate": 0.00016636513986016213,
+      "loss": 0.9886,
+      "step": 260
+    },
+    {
+      "epoch": 3.486842105263158,
+      "grad_norm": 1.1171875,
+      "learning_rate": 0.00016462992378609407,
+      "loss": 0.99,
+      "step": 265
+    },
+    {
+      "epoch": 3.5526315789473686,
+      "grad_norm": 0.78515625,
+      "learning_rate": 0.0001628606243751082,
+      "loss": 0.9947,
+      "step": 270
+    },
+    {
+      "epoch": 3.6184210526315788,
+      "grad_norm": 0.76953125,
+      "learning_rate": 0.00016105817468768798,
+      "loss": 0.991,
+      "step": 275
+    },
+    {
+      "epoch": 3.6842105263157894,
+      "grad_norm": 1.03125,
+      "learning_rate": 0.00015922352526649803,
+      "loss": 0.982,
+      "step": 280
+    },
+    {
+      "epoch": 3.75,
+      "grad_norm": 1.1171875,
+      "learning_rate": 0.0001573576436351046,
+      "loss": 0.9797,
+      "step": 285
+    },
+    {
+      "epoch": 3.8157894736842106,
+      "grad_norm": 1.703125,
+      "learning_rate": 0.00015546151378774086,
+      "loss": 0.9724,
+      "step": 290
+    },
+    {
+      "epoch": 3.8815789473684212,
+      "grad_norm": 0.6796875,
+      "learning_rate": 0.00015353613567038607,
+      "loss": 0.9695,
+      "step": 295
+    },
+    {
+      "epoch": 3.9473684210526314,
+      "grad_norm": 0.87109375,
+      "learning_rate": 0.00015158252465343242,
+      "loss": 0.9638,
+      "step": 300
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.537924289703369,
+      "eval_runtime": 0.2365,
+      "eval_samples_per_second": 42.288,
+      "eval_steps_per_second": 4.229,
+      "step": 304
+    },
+    {
+      "epoch": 4.0131578947368425,
+      "grad_norm": 0.7734375,
+      "learning_rate": 0.00014960171099621795,
+      "loss": 0.9669,
+      "step": 305
+    },
+    {
+      "epoch": 4.078947368421052,
+      "grad_norm": 0.7109375,
+      "learning_rate": 0.00014759473930370736,
+      "loss": 0.9414,
+      "step": 310
+    },
+    {
+      "epoch": 4.144736842105263,
+      "grad_norm": 0.921875,
+      "learning_rate": 0.00014556266797560732,
+      "loss": 0.9589,
+      "step": 315
+    },
+    {
+      "epoch": 4.2105263157894735,
+      "grad_norm": 0.66796875,
+      "learning_rate": 0.00014350656864820733,
+      "loss": 0.9489,
+      "step": 320
+    },
+    {
+      "epoch": 4.276315789473684,
+      "grad_norm": 0.79296875,
+      "learning_rate": 0.00014142752562923988,
+      "loss": 0.9502,
+      "step": 325
+    },
+    {
+      "epoch": 4.342105263157895,
+      "grad_norm": 1.2578125,
+      "learning_rate": 0.0001393266353260583,
+      "loss": 0.9626,
+      "step": 330
+    },
+    {
+      "epoch": 4.407894736842105,
+      "grad_norm": 0.59765625,
+      "learning_rate": 0.00013720500566743362,
+      "loss": 0.9511,
+      "step": 335
+    },
+    {
+      "epoch": 4.473684210526316,
+      "grad_norm": 1.65625,
+      "learning_rate": 0.00013506375551927547,
+      "loss": 0.9456,
+      "step": 340
+    },
+    {
+      "epoch": 4.5394736842105265,
+      "grad_norm": 1.4765625,
+      "learning_rate": 0.00013290401409458532,
+      "loss": 0.9461,
+      "step": 345
+    },
+    {
+      "epoch": 4.605263157894737,
+      "grad_norm": 1.875,
+      "learning_rate": 0.00013072692035795305,
+      "loss": 0.9418,
+      "step": 350
+    },
+    {
+      "epoch": 4.671052631578947,
+      "grad_norm": 0.98828125,
+      "learning_rate": 0.00012853362242491053,
+      "loss": 0.9481,
+      "step": 355
+    },
+    {
+      "epoch": 4.7368421052631575,
+      "grad_norm": 0.87890625,
+      "learning_rate": 0.00012632527695645993,
+      "loss": 0.9461,
+      "step": 360
+    },
+    {
+      "epoch": 4.802631578947368,
+      "grad_norm": 1.140625,
+      "learning_rate": 0.00012410304854909495,
+      "loss": 0.9493,
+      "step": 365
+    },
+    {
+      "epoch": 4.868421052631579,
+      "grad_norm": 0.87890625,
+      "learning_rate": 0.0001218681091206376,
+      "loss": 0.9283,
+      "step": 370
+    },
+    {
+      "epoch": 4.934210526315789,
+      "grad_norm": 0.62109375,
+      "learning_rate": 0.0001196216372922136,
+      "loss": 0.9395,
+      "step": 375
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.6484375,
+      "learning_rate": 0.00011736481776669306,
+      "loss": 0.9419,
+      "step": 380
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.5376482009887695,
+      "eval_runtime": 0.2349,
+      "eval_samples_per_second": 42.567,
+      "eval_steps_per_second": 4.257,
+      "step": 380
+    },
+    {
+      "epoch": 5.065789473684211,
+      "grad_norm": 0.95703125,
+      "learning_rate": 0.00011509884070392369,
+      "loss": 0.9265,
+      "step": 385
+    },
+    {
+      "epoch": 5.131578947368421,
+      "grad_norm": 0.60546875,
+      "learning_rate": 0.00011282490109308633,
+      "loss": 0.9248,
+      "step": 390
+    },
+    {
+      "epoch": 5.197368421052632,
+      "grad_norm": 0.6484375,
+      "learning_rate": 0.00011054419812250338,
+      "loss": 0.915,
+      "step": 395
+    },
+    {
+      "epoch": 5.2631578947368425,
+      "grad_norm": 0.61328125,
+      "learning_rate": 0.00010825793454723325,
+      "loss": 0.9219,
+      "step": 400
+    },
+    {
+      "epoch": 5.328947368421053,
+      "grad_norm": 0.73046875,
+      "learning_rate": 0.0001059673160547834,
+      "loss": 0.9223,
+      "step": 405
+    },
+    {
+      "epoch": 5.394736842105263,
+      "grad_norm": 0.64453125,
+      "learning_rate": 0.00010367355062927726,
+      "loss": 0.9236,
+      "step": 410
+    },
+    {
+      "epoch": 5.4605263157894735,
+      "grad_norm": 0.6640625,
+      "learning_rate": 0.00010137784791440965,
+      "loss": 0.9252,
+      "step": 415
+    },
+    {
+      "epoch": 5.526315789473684,
+      "grad_norm": 0.7890625,
+      "learning_rate": 9.908141857552737e-05,
+      "loss": 0.917,
+      "step": 420
+    },
+    {
+      "epoch": 5.592105263157895,
+      "grad_norm": 0.671875,
+      "learning_rate": 9.678547366117083e-05,
+      "loss": 0.9213,
+      "step": 425
+    },
+    {
+      "epoch": 5.657894736842105,
+      "grad_norm": 0.84375,
+      "learning_rate": 9.449122396441345e-05,
+      "loss": 0.9098,
+      "step": 430
+    },
+    {
+      "epoch": 5.723684210526316,
+      "grad_norm": 0.8203125,
+      "learning_rate": 9.219987938433621e-05,
+      "loss": 0.9233,
+      "step": 435
+    },
+    {
+      "epoch": 5.7894736842105265,
+      "grad_norm": 0.77734375,
+      "learning_rate": 8.991264828797319e-05,
+      "loss": 0.9113,
+      "step": 440
+    },
+    {
+      "epoch": 5.855263157894737,
+      "grad_norm": 0.55859375,
+      "learning_rate": 8.763073687306524e-05,
+      "loss": 0.9198,
+      "step": 445
+    },
+    {
+      "epoch": 5.921052631578947,
+      "grad_norm": 1.1328125,
+      "learning_rate": 8.535534853195786e-05,
+      "loss": 0.9186,
+      "step": 450
+    },
+    {
+      "epoch": 5.9868421052631575,
+      "grad_norm": 0.6796875,
+      "learning_rate": 8.308768321697815e-05,
+      "loss": 0.9117,
+      "step": 455
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.5333404541015625,
+      "eval_runtime": 0.2353,
+      "eval_samples_per_second": 42.503,
+      "eval_steps_per_second": 4.25,
+      "step": 456
+    },
+    {
+      "epoch": 6.052631578947368,
+      "grad_norm": 0.53515625,
+      "learning_rate": 8.082893680762619e-05,
+      "loss": 0.9067,
+      "step": 460
+    },
+    {
+      "epoch": 6.118421052631579,
+      "grad_norm": 0.73828125,
+      "learning_rate": 7.858030047991411e-05,
+      "loss": 0.9008,
+      "step": 465
+    },
+    {
+      "epoch": 6.184210526315789,
+      "grad_norm": 0.5546875,
+      "learning_rate": 7.634296007818576e-05,
+      "loss": 0.8891,
+      "step": 470
+    },
+    {
+      "epoch": 6.25,
+      "grad_norm": 0.59375,
+      "learning_rate": 7.411809548974792e-05,
+      "loss": 0.9044,
+      "step": 475
+    },
+    {
+      "epoch": 6.315789473684211,
+      "grad_norm": 0.53515625,
+      "learning_rate": 7.190688002264308e-05,
+      "loss": 0.8954,
+      "step": 480
+    },
+    {
+      "epoch": 6.381578947368421,
+      "grad_norm": 0.55859375,
+      "learning_rate": 6.971047978689189e-05,
+      "loss": 0.895,
+      "step": 485
+    },
+    {
+      "epoch": 6.447368421052632,
+      "grad_norm": 0.54296875,
+      "learning_rate": 6.753005307953167e-05,
+      "loss": 0.9084,
+      "step": 490
+    },
+    {
+      "epoch": 6.5131578947368425,
+      "grad_norm": 0.55859375,
+      "learning_rate": 6.536674977377496e-05,
+      "loss": 0.8972,
+      "step": 495
+    },
+    {
+      "epoch": 6.578947368421053,
+      "grad_norm": 0.66796875,
+      "learning_rate": 6.322171071261071e-05,
+      "loss": 0.8986,
+      "step": 500
+    },
+    {
+      "epoch": 6.644736842105263,
+      "grad_norm": 0.5234375,
+      "learning_rate": 6.109606710716741e-05,
+      "loss": 0.8889,
+      "step": 505
+    },
+    {
+      "epoch": 6.7105263157894735,
+      "grad_norm": 0.859375,
+      "learning_rate": 5.8990939940156e-05,
+      "loss": 0.9018,
+      "step": 510
+    },
+    {
+      "epoch": 6.776315789473684,
+      "grad_norm": 0.53515625,
+      "learning_rate": 5.690743937470657e-05,
+      "loss": 0.9085,
+      "step": 515
+    },
+    {
+      "epoch": 6.842105263157895,
+      "grad_norm": 0.640625,
+      "learning_rate": 5.484666416891109e-05,
+      "loss": 0.9082,
+      "step": 520
+    },
+    {
+      "epoch": 6.907894736842105,
+      "grad_norm": 0.54296875,
+      "learning_rate": 5.280970109638047e-05,
+      "loss": 0.8921,
+      "step": 525
+    },
+    {
+      "epoch": 6.973684210526316,
+      "grad_norm": 0.58203125,
+      "learning_rate": 5.079762437312219e-05,
+      "loss": 0.8944,
+      "step": 530
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.5417115688323975,
+      "eval_runtime": 0.2355,
+      "eval_samples_per_second": 42.467,
+      "eval_steps_per_second": 4.247,
+      "step": 532
+    },
+    {
+      "epoch": 7.0394736842105265,
+      "grad_norm": 0.5078125,
+      "learning_rate": 4.8811495091039926e-05,
+      "loss": 0.8912,
+      "step": 535
+    },
+    {
+      "epoch": 7.105263157894737,
+      "grad_norm": 0.58984375,
+      "learning_rate": 4.685236065835443e-05,
+      "loss": 0.8893,
+      "step": 540
+    },
+    {
+      "epoch": 7.171052631578948,
+      "grad_norm": 0.5078125,
+      "learning_rate": 4.492125424724086e-05,
+      "loss": 0.8881,
+      "step": 545
+    },
+    {
+      "epoch": 7.2368421052631575,
+      "grad_norm": 0.5234375,
+      "learning_rate": 4.301919424897338e-05,
+      "loss": 0.8913,
+      "step": 550
+    },
+    {
+      "epoch": 7.302631578947368,
+      "grad_norm": 0.498046875,
+      "learning_rate": 4.114718373686481e-05,
+      "loss": 0.8871,
+      "step": 555
+    },
+    {
+      "epoch": 7.368421052631579,
+      "grad_norm": 0.55859375,
+      "learning_rate": 3.9306209937284346e-05,
+      "loss": 0.8854,
+      "step": 560
+    },
+    {
+      "epoch": 7.434210526315789,
+      "grad_norm": 0.6328125,
+      "learning_rate": 3.749724370903216e-05,
+      "loss": 0.8908,
+      "step": 565
+    },
+    {
+      "epoch": 7.5,
+      "grad_norm": 0.51171875,
+      "learning_rate": 3.5721239031346066e-05,
+      "loss": 0.8887,
+      "step": 570
+    },
+    {
+      "epoch": 7.565789473684211,
+      "grad_norm": 0.494140625,
+      "learning_rate": 3.3979132500809405e-05,
+      "loss": 0.8858,
+      "step": 575
+    },
+    {
+      "epoch": 7.631578947368421,
+      "grad_norm": 0.49609375,
+      "learning_rate": 3.227184283742591e-05,
+      "loss": 0.8847,
+      "step": 580
+    },
+    {
+      "epoch": 7.697368421052632,
+      "grad_norm": 0.48046875,
+      "learning_rate": 3.0600270400122335e-05,
+      "loss": 0.8808,
+      "step": 585
+    },
+    {
+      "epoch": 7.7631578947368425,
+      "grad_norm": 0.498046875,
+      "learning_rate": 2.89652967119336e-05,
+      "loss": 0.8825,
+      "step": 590
+    },
+    {
+      "epoch": 7.828947368421053,
+      "grad_norm": 0.470703125,
+      "learning_rate": 2.73677839951215e-05,
+      "loss": 0.8923,
+      "step": 595
+    },
+    {
+      "epoch": 7.894736842105263,
+      "grad_norm": 0.447265625,
+      "learning_rate": 2.5808574716471856e-05,
+      "loss": 0.8809,
+      "step": 600
+    },
+    {
+      "epoch": 7.9605263157894735,
+      "grad_norm": 0.490234375,
+      "learning_rate": 2.4288491143009795e-05,
+      "loss": 0.8824,
+      "step": 605
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.547354221343994,
+      "eval_runtime": 0.2365,
+      "eval_samples_per_second": 42.286,
+      "eval_steps_per_second": 4.229,
+      "step": 608
+    },
+    {
+      "epoch": 8.026315789473685,
+      "grad_norm": 0.478515625,
+      "learning_rate": 2.2808334908367914e-05,
+      "loss": 0.8926,
+      "step": 610
+    },
+    {
+      "epoch": 8.092105263157896,
+      "grad_norm": 0.48828125,
+      "learning_rate": 2.1368886590035443e-05,
+      "loss": 0.8774,
+      "step": 615
+    },
+    {
+      "epoch": 8.157894736842104,
+      "grad_norm": 0.4921875,
+      "learning_rate": 1.9970905297711606e-05,
+      "loss": 0.8834,
+      "step": 620
+    },
+    {
+      "epoch": 8.223684210526315,
+      "grad_norm": 0.53125,
+      "learning_rate": 1.861512827298051e-05,
+      "loss": 0.8845,
+      "step": 625
+    },
+    {
+      "epoch": 8.289473684210526,
+      "grad_norm": 0.5859375,
+      "learning_rate": 1.7302270500518182e-05,
+      "loss": 0.8801,
+      "step": 630
+    },
+    {
+      "epoch": 8.355263157894736,
+      "grad_norm": 0.49609375,
+      "learning_rate": 1.6033024331037138e-05,
+      "loss": 0.873,
+      "step": 635
+    },
+    {
+      "epoch": 8.421052631578947,
+      "grad_norm": 0.5078125,
+      "learning_rate": 1.4808059116167305e-05,
+      "loss": 0.8795,
+      "step": 640
+    },
+    {
+      "epoch": 8.486842105263158,
+      "grad_norm": 0.466796875,
+      "learning_rate": 1.3628020855465572e-05,
+      "loss": 0.8807,
+      "step": 645
+    },
+    {
+      "epoch": 8.552631578947368,
+      "grad_norm": 0.46484375,
+      "learning_rate": 1.2493531855740625e-05,
+      "loss": 0.8821,
+      "step": 650
+    },
+    {
+      "epoch": 8.618421052631579,
+      "grad_norm": 0.47265625,
+      "learning_rate": 1.1405190402872202e-05,
+      "loss": 0.8731,
+      "step": 655
+    },
+    {
+      "epoch": 8.68421052631579,
+      "grad_norm": 0.466796875,
+      "learning_rate": 1.0363570446297999e-05,
+      "loss": 0.8878,
+      "step": 660
+    },
+    {
+      "epoch": 8.75,
+      "grad_norm": 0.53125,
+      "learning_rate": 9.369221296335006e-06,
+      "loss": 0.8827,
+      "step": 665
+    },
+    {
+      "epoch": 8.81578947368421,
+      "grad_norm": 0.46484375,
+      "learning_rate": 8.422667334494249e-06,
+      "loss": 0.8847,
+      "step": 670
+    },
+    {
+      "epoch": 8.881578947368421,
+      "grad_norm": 0.48046875,
+      "learning_rate": 7.524407736942174e-06,
+      "loss": 0.8844,
+      "step": 675
+    },
+    {
+      "epoch": 8.947368421052632,
+      "grad_norm": 0.53125,
+      "learning_rate": 6.674916211254289e-06,
+      "loss": 0.8759,
+      "step": 680
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 2.5541369915008545,
+      "eval_runtime": 0.2345,
+      "eval_samples_per_second": 42.638,
+      "eval_steps_per_second": 4.264,
+      "step": 684
+    },
+    {
+      "epoch": 9.013157894736842,
+      "grad_norm": 0.455078125,
+      "learning_rate": 5.8746407466000464e-06,
+      "loss": 0.8768,
+      "step": 685
+    },
+    {
+      "epoch": 9.078947368421053,
+      "grad_norm": 0.4765625,
+      "learning_rate": 5.124003377490582e-06,
+      "loss": 0.8787,
+      "step": 690
+    },
+    {
+      "epoch": 9.144736842105264,
+      "grad_norm": 0.515625,
+      "learning_rate": 4.423399961213892e-06,
+      "loss": 0.8874,
+      "step": 695
+    },
+    {
+      "epoch": 9.210526315789474,
+      "grad_norm": 0.462890625,
+      "learning_rate": 3.7731999690749585e-06,
+      "loss": 0.875,
+      "step": 700
+    },
+    {
+      "epoch": 9.276315789473685,
+      "grad_norm": 0.46875,
+      "learning_rate": 3.1737462915508277e-06,
+      "loss": 0.8777,
+      "step": 705
+    },
+    {
+      "epoch": 9.342105263157896,
+      "grad_norm": 0.46875,
+      "learning_rate": 2.6253550574632303e-06,
+      "loss": 0.8728,
+      "step": 710
+    },
+    {
+      "epoch": 9.407894736842104,
+      "grad_norm": 0.470703125,
+      "learning_rate": 2.128315467264552e-06,
+      "loss": 0.8791,
+      "step": 715
+    },
+    {
+      "epoch": 9.473684210526315,
+      "grad_norm": 0.4765625,
+      "learning_rate": 1.6828896405244988e-06,
+      "loss": 0.8809,
+      "step": 720
+    },
+    {
+      "epoch": 9.539473684210526,
+      "grad_norm": 0.458984375,
+      "learning_rate": 1.28931247769839e-06,
+      "loss": 0.8836,
+      "step": 725
+    },
+    {
+      "epoch": 9.605263157894736,
+      "grad_norm": 0.455078125,
+      "learning_rate": 9.477915362496758e-07,
+      "loss": 0.8819,
+      "step": 730
+    },
+    {
+      "epoch": 9.671052631578947,
+      "grad_norm": 0.4609375,
+      "learning_rate": 6.585069211921035e-07,
+      "loss": 0.874,
+      "step": 735
+    },
+    {
+      "epoch": 9.736842105263158,
+      "grad_norm": 0.458984375,
+      "learning_rate": 4.216111901092501e-07,
+      "loss": 0.8871,
+      "step": 740
+    },
+    {
+      "epoch": 9.802631578947368,
+      "grad_norm": 0.490234375,
+      "learning_rate": 2.372292727015557e-07,
+      "loss": 0.8811,
+      "step": 745
+    },
+    {
+      "epoch": 9.868421052631579,
+      "grad_norm": 0.45703125,
+      "learning_rate": 1.0545840490313596e-07,
+      "loss": 0.8768,
+      "step": 750
+    },
+    {
+      "epoch": 9.93421052631579,
+      "grad_norm": 0.46875,
+      "learning_rate": 2.6368077603367015e-08,
+      "loss": 0.8774,
+      "step": 755
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.47265625,
+      "learning_rate": 0.0,
+      "loss": 0.8735,
+      "step": 760
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.5523786544799805,
+      "eval_runtime": 0.2353,
+      "eval_samples_per_second": 42.494,
+      "eval_steps_per_second": 4.249,
+      "step": 760
+    },
+    {
+      "epoch": 10.0,
+      "step": 760,
+      "total_flos": 2.318334860390826e+18,
+      "train_loss": 2.2621336485210217,
+      "train_runtime": 1856.2046,
+      "train_samples_per_second": 26.156,
+      "train_steps_per_second": 0.409
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 760,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 100,
+  "total_flos": 2.318334860390826e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e72d22e36be57e700734a4d42088c5a29b0893c81375cd69c80ac04f2224952
+size 5176