huggingartists

README.md

@@ -0,0 +1,97 @@
+---
+language: en
+datasets:
+- huggingartists/pyrokinesis
+tags:
+- huggingartists
+- lyrics
+- lm-head
+- causal-lm
+widget:
+- text: "I am"
+---
+
+<div class="inline-flex flex-col" style="line-height: 1.5;">
+    <div class="flex">
+        <div
+			style="display:DISPLAY_1; margin-left: auto; margin-right: auto; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url(&#39;https://images.genius.com/e701c222dfb8725065dd99c8a43988da.1000x1000x1.jpg&#39;)">
+        </div>
+    </div>
+    <div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 HuggingArtists Model 🤖</div>
+    <div style="text-align: center; font-size: 16px; font-weight: 800">​​pyrokinesis</div>
+    <a href="https://genius.com/artists/pyrokinesis">
+    	<div style="text-align: center; font-size: 14px;">@pyrokinesis</div>
+    </a>
+</div>
+
+I was made with [huggingartists](https://github.com/AlekseyKorshuk/huggingartists).
+
+Create your own bot based on your favorite artist with [the demo](https://colab.research.google.com/github/AlekseyKorshuk/huggingartists/blob/master/huggingartists-demo.ipynb)!
+
+## How does it work?
+
+To understand how the model was developed, check the [W&B report](https://wandb.ai/huggingartists/huggingartists/reportlist).
+
+## Training data
+
+The model was trained on lyrics from ​​pyrokinesis.
+
+Dataset is available [here](https://huggingface.co/datasets/huggingartists/pyrokinesis).
+And can be used with:
+
+```python
+from datasets import load_dataset
+
+dataset = load_dataset("huggingartists/pyrokinesis")
+```
+
+[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/1s8696f3/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
+
+## Training procedure
+
+The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on ​​pyrokinesis's lyrics.
+
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/22hm2utc) for full transparency and reproducibility.
+
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/22hm2utc/artifacts) is logged and versioned.
+
+## How to use
+
+You can use this model directly with a pipeline for text generation:
+
+```python
+from transformers import pipeline
+generator = pipeline('text-generation',
+                     model='huggingartists/pyrokinesis')
+generator("I am", num_return_sequences=5)
+```
+
+Or with Transformers library:
+
+```python
+from transformers import AutoTokenizer, AutoModelWithLMHead
+  
+tokenizer = AutoTokenizer.from_pretrained("huggingartists/pyrokinesis")
+
+model = AutoModelWithLMHead.from_pretrained("huggingartists/pyrokinesis")
+```
+
+## Limitations and bias
+
+The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias).
+
+In addition, the data present in the user's tweets further affects the text generated by the model.
+
+## About
+
+*Built by Aleksey Korshuk*
+
+[![Follow](https://img.shields.io/github/followers/AlekseyKorshuk?style=social)](https://github.com/AlekseyKorshuk)
+
+[![Follow](https://img.shields.io/twitter/follow/alekseykorshuk?style=social)](https://twitter.com/intent/follow?screen_name=alekseykorshuk)
+
+[![Follow](https://img.shields.io/badge/dynamic/json?color=blue&label=Telegram%20Channel&query=%24.result&url=https%3A%2F%2Fapi.telegram.org%2Fbot1929545866%3AAAFGhV-KKnegEcLiyYJxsc4zV6C-bdPEBtQ%2FgetChatMemberCount%3Fchat_id%3D-1001253621662&style=social&logo=telegram)](https://t.me/joinchat/_CQ04KjcJ-4yZTky)
+
+For more details, visit the project repository.
+
+[![GitHub stars](https://img.shields.io/github/stars/AlekseyKorshuk/huggingartists?style=social)](https://github.com/AlekseyKorshuk/huggingartists)

config.json

@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "gpt2",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "gradient_checkpointing": false,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "resid_pdrop": 0.1,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 200,
+      "min_length": 100,
+      "temperature": 1.0,
+      "top_p": 0.95
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.10.1",
+  "use_cache": true,
+  "vocab_size": 50257
+}

evaluation.txt

@@ -0,0 +1 @@
+{"eval_loss": 1.4917341470718384, "eval_runtime": 11.5522, "eval_samples_per_second": 21.295, "eval_steps_per_second": 2.683, "epoch": 10.0}

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d9730bbb999065cfa430a8bc81673e210a85938666d7a794b012e19cf4402c2
+size 497764120

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d909f6ea5314c280b899bff45b5a900c0495bf0cad83ae36d6c427d26d57163f
+size 995604017

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd5a80fca145e5a5c2a0a19c7827a1bc8c9bdb82c12f7dc7f530cdc4502f436c
+size 510403817

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7e966aef487c29db07b1b2126be8ff2cd2e3d8419aad00c32907e3ec5ce0c8d
+size 14503

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ae78cbe0368ef751528f9df25bdcc580eaf09b6340910f9e0e6aae733d99451
+size 623

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -0,0 +1,2194 @@
+{
+  "best_metric": 1.4917341470718384,
+  "best_model_checkpoint": "output/pyrokinesis/checkpoint-1755",
+  "epoch": 9.0,
+  "global_step": 1755,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.0001369775513380068,
+      "loss": 2.5666,
+      "step": 5
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00013631164801696083,
+      "loss": 2.2799,
+      "step": 10
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00013520660867542716,
+      "loss": 2.1886,
+      "step": 15
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00013366959991757417,
+      "loss": 2.1192,
+      "step": 20
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00013171058983499535,
+      "loss": 2.0709,
+      "step": 25
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00012934228335981018,
+      "loss": 2.0524,
+      "step": 30
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00012658003986830432,
+      "loss": 2.1101,
+      "step": 35
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0001234417735694802,
+      "loss": 1.9895,
+      "step": 40
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00011994783732453754,
+      "loss": 2.1651,
+      "step": 45
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00011612089065075853,
+      "loss": 2.0113,
+      "step": 50
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00011198575276584287,
+      "loss": 2.0278,
+      "step": 55
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00010756924162575728,
+      "loss": 2.0199,
+      "step": 60
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00010290000000000001,
+      "loss": 1.9674,
+      "step": 65
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.800830971224951e-05,
+      "loss": 1.9432,
+      "step": 70
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 9.292589525111794e-05,
+      "loss": 1.8305,
+      "step": 75
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 8.768571802466866e-05,
+      "loss": 1.9299,
+      "step": 80
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 8.232176259303665e-05,
+      "loss": 1.9406,
+      "step": 85
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 7.686881626551515e-05,
+      "loss": 1.9433,
+      "step": 90
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 7.136224349150589e-05,
+      "loss": 1.8757,
+      "step": 95
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 6.583775650849414e-05,
+      "loss": 1.8865,
+      "step": 100
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 6.033118373448485e-05,
+      "loss": 1.9857,
+      "step": 105
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 5.487823740696336e-05,
+      "loss": 1.9976,
+      "step": 110
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.951428197533135e-05,
+      "loss": 1.9144,
+      "step": 115
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.427410474888205e-05,
+      "loss": 1.8636,
+      "step": 120
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.9191690287750474e-05,
+      "loss": 1.9044,
+      "step": 125
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 3.4300000000000014e-05,
+      "loss": 1.8643,
+      "step": 130
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.963075837424272e-05,
+      "loss": 1.924,
+      "step": 135
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 2.5214247234157127e-05,
+      "loss": 1.8417,
+      "step": 140
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 2.1079109349241483e-05,
+      "loss": 1.8783,
+      "step": 145
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.725216267546246e-05,
+      "loss": 1.8383,
+      "step": 150
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.3758226430519834e-05,
+      "loss": 1.854,
+      "step": 155
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.061996013169569e-05,
+      "loss": 1.779,
+      "step": 160
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 7.857716640189809e-06,
+      "loss": 1.8699,
+      "step": 165
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 5.4894101650046585e-06,
+      "loss": 1.863,
+      "step": 170
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.53040008242582e-06,
+      "loss": 1.9041,
+      "step": 175
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.993391324572832e-06,
+      "loss": 1.7998,
+      "step": 180
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 8.883519830391788e-07,
+      "loss": 1.9326,
+      "step": 185
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.2244866199319883e-07,
+      "loss": 1.8503,
+      "step": 190
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0,
+      "loss": 1.7822,
+      "step": 195
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.7307456731796265,
+      "eval_runtime": 11.5169,
+      "eval_samples_per_second": 21.36,
+      "eval_steps_per_second": 2.692,
+      "step": 195
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 2.2244866199319123e-07,
+      "loss": 1.8197,
+      "step": 200
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 8.883519830391788e-07,
+      "loss": 1.7915,
+      "step": 205
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.9933913245728244e-06,
+      "loss": 1.8277,
+      "step": 210
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 3.53040008242582e-06,
+      "loss": 1.8665,
+      "step": 215
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 5.4894101650046585e-06,
+      "loss": 1.8632,
+      "step": 220
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 7.857716640189785e-06,
+      "loss": 1.8499,
+      "step": 225
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.0619960131695675e-05,
+      "loss": 1.851,
+      "step": 230
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3758226430519804e-05,
+      "loss": 1.8915,
+      "step": 235
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7252162675462473e-05,
+      "loss": 1.896,
+      "step": 240
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 2.1079109349241466e-05,
+      "loss": 1.9118,
+      "step": 245
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 2.5214247234157134e-05,
+      "loss": 1.8666,
+      "step": 250
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 2.9630758374242703e-05,
+      "loss": 1.8693,
+      "step": 255
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 3.429999999999997e-05,
+      "loss": 1.8738,
+      "step": 260
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 3.919169028775049e-05,
+      "loss": 1.7909,
+      "step": 265
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 4.427410474888204e-05,
+      "loss": 1.7761,
+      "step": 270
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 4.9514281975331363e-05,
+      "loss": 1.8239,
+      "step": 275
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 5.487823740696335e-05,
+      "loss": 1.845,
+      "step": 280
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 6.03311837344848e-05,
+      "loss": 1.851,
+      "step": 285
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 6.583775650849411e-05,
+      "loss": 1.7638,
+      "step": 290
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 7.136224349150585e-05,
+      "loss": 1.9163,
+      "step": 295
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 7.686881626551516e-05,
+      "loss": 1.7542,
+      "step": 300
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 8.232176259303664e-05,
+      "loss": 1.92,
+      "step": 305
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 8.768571802466861e-05,
+      "loss": 1.8303,
+      "step": 310
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.292589525111794e-05,
+      "loss": 1.7917,
+      "step": 315
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.800830971224949e-05,
+      "loss": 1.7499,
+      "step": 320
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.00010290000000000001,
+      "loss": 1.7705,
+      "step": 325
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.00010756924162575728,
+      "loss": 1.8492,
+      "step": 330
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.00011198575276584284,
+      "loss": 1.8068,
+      "step": 335
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 0.00011612089065075851,
+      "loss": 1.7533,
+      "step": 340
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.0001199478373245375,
+      "loss": 1.8224,
+      "step": 345
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.0001234417735694802,
+      "loss": 1.738,
+      "step": 350
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 0.00012658003986830432,
+      "loss": 1.8536,
+      "step": 355
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.0001293422833598102,
+      "loss": 1.7585,
+      "step": 360
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 0.00013171058983499532,
+      "loss": 1.7811,
+      "step": 365
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.00013366959991757417,
+      "loss": 1.801,
+      "step": 370
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.00013520660867542716,
+      "loss": 1.7785,
+      "step": 375
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.00013631164801696083,
+      "loss": 1.756,
+      "step": 380
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.0001369775513380068,
+      "loss": 1.8216,
+      "step": 385
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.0001372,
+      "loss": 1.6845,
+      "step": 390
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.6902183294296265,
+      "eval_runtime": 11.5295,
+      "eval_samples_per_second": 21.337,
+      "eval_steps_per_second": 2.689,
+      "step": 390
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 0.0001369775513380068,
+      "loss": 1.8284,
+      "step": 395
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 0.00013631164801696085,
+      "loss": 1.7341,
+      "step": 400
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 0.00013520660867542716,
+      "loss": 1.7068,
+      "step": 405
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 0.00013366959991757417,
+      "loss": 1.8545,
+      "step": 410
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 0.00013171058983499535,
+      "loss": 1.7491,
+      "step": 415
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 0.00012934228335981023,
+      "loss": 1.7728,
+      "step": 420
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 0.00012658003986830435,
+      "loss": 1.8086,
+      "step": 425
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 0.00012344177356948016,
+      "loss": 1.7479,
+      "step": 430
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.00011994783732453754,
+      "loss": 1.8108,
+      "step": 435
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 0.00011612089065075854,
+      "loss": 1.7602,
+      "step": 440
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.00011198575276584292,
+      "loss": 1.6977,
+      "step": 445
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 0.00010756924162575734,
+      "loss": 1.7494,
+      "step": 450
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.00010289999999999998,
+      "loss": 1.6501,
+      "step": 455
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 9.800830971224951e-05,
+      "loss": 1.6887,
+      "step": 460
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 9.292589525111797e-05,
+      "loss": 1.68,
+      "step": 465
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 8.76857180246687e-05,
+      "loss": 1.6036,
+      "step": 470
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 8.232176259303673e-05,
+      "loss": 1.6931,
+      "step": 475
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 7.686881626551514e-05,
+      "loss": 1.7376,
+      "step": 480
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 7.136224349150589e-05,
+      "loss": 1.68,
+      "step": 485
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 6.583775650849415e-05,
+      "loss": 1.7219,
+      "step": 490
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 6.03311837344849e-05,
+      "loss": 1.7081,
+      "step": 495
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 5.4878237406963316e-05,
+      "loss": 1.6974,
+      "step": 500
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 4.951428197533134e-05,
+      "loss": 1.6991,
+      "step": 505
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 4.427410474888207e-05,
+      "loss": 1.6574,
+      "step": 510
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 3.919169028775052e-05,
+      "loss": 1.6158,
+      "step": 515
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 3.4300000000000054e-05,
+      "loss": 1.7649,
+      "step": 520
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 2.9630758374242683e-05,
+      "loss": 1.6256,
+      "step": 525
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 2.521424723415712e-05,
+      "loss": 1.6033,
+      "step": 530
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 2.107910934924149e-05,
+      "loss": 1.7139,
+      "step": 535
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.7252162675462497e-05,
+      "loss": 1.6601,
+      "step": 540
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 1.3758226430519865e-05,
+      "loss": 1.7914,
+      "step": 545
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 1.0619960131695668e-05,
+      "loss": 1.6809,
+      "step": 550
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 7.8577166401898e-06,
+      "loss": 1.653,
+      "step": 555
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 5.489410165004666e-06,
+      "loss": 1.7064,
+      "step": 560
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 3.5304000824258354e-06,
+      "loss": 1.7002,
+      "step": 565
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 1.9933913245728472e-06,
+      "loss": 1.5901,
+      "step": 570
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 8.883519830391712e-07,
+      "loss": 1.6791,
+      "step": 575
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 2.2244866199319123e-07,
+      "loss": 1.6187,
+      "step": 580
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.0,
+      "loss": 1.6345,
+      "step": 585
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.5962461233139038,
+      "eval_runtime": 11.5211,
+      "eval_samples_per_second": 21.352,
+      "eval_steps_per_second": 2.691,
+      "step": 585
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 2.2244866199319123e-07,
+      "loss": 1.6608,
+      "step": 590
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 8.883519830391636e-07,
+      "loss": 1.659,
+      "step": 595
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 1.9933913245728396e-06,
+      "loss": 1.6294,
+      "step": 600
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 3.5304000824258126e-06,
+      "loss": 1.6214,
+      "step": 605
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 5.489410165004651e-06,
+      "loss": 1.6499,
+      "step": 610
+    },
+    {
+      "epoch": 3.15,
+      "learning_rate": 7.857716640189778e-06,
+      "loss": 1.6205,
+      "step": 615
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 1.0619960131695638e-05,
+      "loss": 1.6322,
+      "step": 620
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 1.3758226430519834e-05,
+      "loss": 1.6142,
+      "step": 625
+    },
+    {
+      "epoch": 3.23,
+      "learning_rate": 1.725216267546246e-05,
+      "loss": 1.7086,
+      "step": 630
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 2.107910934924146e-05,
+      "loss": 1.6773,
+      "step": 635
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 2.5214247234157073e-05,
+      "loss": 1.737,
+      "step": 640
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 2.9630758374242642e-05,
+      "loss": 1.5679,
+      "step": 645
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.4300000000000014e-05,
+      "loss": 1.6389,
+      "step": 650
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 3.9191690287750474e-05,
+      "loss": 1.653,
+      "step": 655
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 4.427410474888202e-05,
+      "loss": 1.6666,
+      "step": 660
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 4.951428197533129e-05,
+      "loss": 1.5802,
+      "step": 665
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 5.487823740696327e-05,
+      "loss": 1.5621,
+      "step": 670
+    },
+    {
+      "epoch": 3.46,
+      "learning_rate": 6.033118373448485e-05,
+      "loss": 1.5473,
+      "step": 675
+    },
+    {
+      "epoch": 3.49,
+      "learning_rate": 6.58377565084941e-05,
+      "loss": 1.6707,
+      "step": 680
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 7.136224349150583e-05,
+      "loss": 1.639,
+      "step": 685
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 7.68688162655151e-05,
+      "loss": 1.6476,
+      "step": 690
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 8.232176259303666e-05,
+      "loss": 1.704,
+      "step": 695
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 8.768571802466866e-05,
+      "loss": 1.5819,
+      "step": 700
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 9.292589525111793e-05,
+      "loss": 1.5765,
+      "step": 705
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 9.800830971224949e-05,
+      "loss": 1.6459,
+      "step": 710
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 0.00010289999999999994,
+      "loss": 1.6142,
+      "step": 715
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 0.00010756924162575731,
+      "loss": 1.6049,
+      "step": 720
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 0.00011198575276584287,
+      "loss": 1.6587,
+      "step": 725
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 0.0001161208906507585,
+      "loss": 1.559,
+      "step": 730
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 0.0001199478373245375,
+      "loss": 1.695,
+      "step": 735
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 0.00012344177356948013,
+      "loss": 1.6336,
+      "step": 740
+    },
+    {
+      "epoch": 3.82,
+      "learning_rate": 0.00012658003986830432,
+      "loss": 1.5868,
+      "step": 745
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 0.00012934228335981018,
+      "loss": 1.5476,
+      "step": 750
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 0.00013171058983499532,
+      "loss": 1.6785,
+      "step": 755
+    },
+    {
+      "epoch": 3.9,
+      "learning_rate": 0.00013366959991757417,
+      "loss": 1.8145,
+      "step": 760
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 0.00013520660867542716,
+      "loss": 1.6926,
+      "step": 765
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 0.00013631164801696083,
+      "loss": 1.6759,
+      "step": 770
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 0.00013697755133800678,
+      "loss": 1.6476,
+      "step": 775
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.0001372,
+      "loss": 1.697,
+      "step": 780
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.5882556438446045,
+      "eval_runtime": 11.5211,
+      "eval_samples_per_second": 21.352,
+      "eval_steps_per_second": 2.691,
+      "step": 780
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 0.0001369775513380068,
+      "loss": 1.6285,
+      "step": 785
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 0.00013631164801696085,
+      "loss": 1.5054,
+      "step": 790
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 0.0001352066086754272,
+      "loss": 1.612,
+      "step": 795
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 0.00013366959991757425,
+      "loss": 1.6493,
+      "step": 800
+    },
+    {
+      "epoch": 4.13,
+      "learning_rate": 0.0001317105898349953,
+      "loss": 1.5501,
+      "step": 805
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 0.00012934228335981015,
+      "loss": 1.5358,
+      "step": 810
+    },
+    {
+      "epoch": 4.18,
+      "learning_rate": 0.0001265800398683043,
+      "loss": 1.6262,
+      "step": 815
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 0.0001234417735694802,
+      "loss": 1.59,
+      "step": 820
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 0.00011994783732453755,
+      "loss": 1.5257,
+      "step": 825
+    },
+    {
+      "epoch": 4.26,
+      "learning_rate": 0.00011612089065075855,
+      "loss": 1.5782,
+      "step": 830
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 0.00011198575276584294,
+      "loss": 1.6528,
+      "step": 835
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 0.00010756924162575738,
+      "loss": 1.64,
+      "step": 840
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 0.00010290000000000009,
+      "loss": 1.5902,
+      "step": 845
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 9.800830971224965e-05,
+      "loss": 1.5726,
+      "step": 850
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 9.292589525111788e-05,
+      "loss": 1.5698,
+      "step": 855
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 8.76857180246686e-05,
+      "loss": 1.5357,
+      "step": 860
+    },
+    {
+      "epoch": 4.44,
+      "learning_rate": 8.232176259303662e-05,
+      "loss": 1.6191,
+      "step": 865
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 7.686881626551516e-05,
+      "loss": 1.5824,
+      "step": 870
+    },
+    {
+      "epoch": 4.49,
+      "learning_rate": 7.13622434915059e-05,
+      "loss": 1.5826,
+      "step": 875
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 6.583775650849417e-05,
+      "loss": 1.5749,
+      "step": 880
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 6.033118373448492e-05,
+      "loss": 1.5323,
+      "step": 885
+    },
+    {
+      "epoch": 4.56,
+      "learning_rate": 5.487823740696345e-05,
+      "loss": 1.5293,
+      "step": 890
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 4.951428197533147e-05,
+      "loss": 1.6373,
+      "step": 895
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 4.42741047488822e-05,
+      "loss": 1.5182,
+      "step": 900
+    },
+    {
+      "epoch": 4.64,
+      "learning_rate": 3.919169028775043e-05,
+      "loss": 1.5626,
+      "step": 905
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 3.429999999999996e-05,
+      "loss": 1.6378,
+      "step": 910
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 2.9630758374242696e-05,
+      "loss": 1.5834,
+      "step": 915
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 2.5214247234157127e-05,
+      "loss": 1.6246,
+      "step": 920
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 2.1079109349241507e-05,
+      "loss": 1.5866,
+      "step": 925
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 1.7252162675462504e-05,
+      "loss": 1.6128,
+      "step": 930
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 1.3758226430519873e-05,
+      "loss": 1.4767,
+      "step": 935
+    },
+    {
+      "epoch": 4.82,
+      "learning_rate": 1.0619960131695736e-05,
+      "loss": 1.5603,
+      "step": 940
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 7.857716640189861e-06,
+      "loss": 1.5784,
+      "step": 945
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 5.4894101650047195e-06,
+      "loss": 1.6181,
+      "step": 950
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 3.5304000824257973e-06,
+      "loss": 1.5659,
+      "step": 955
+    },
+    {
+      "epoch": 4.92,
+      "learning_rate": 1.9933913245728244e-06,
+      "loss": 1.523,
+      "step": 960
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 8.883519830391788e-07,
+      "loss": 1.5859,
+      "step": 965
+    },
+    {
+      "epoch": 4.97,
+      "learning_rate": 2.2244866199319883e-07,
+      "loss": 1.5622,
+      "step": 970
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.0,
+      "loss": 1.5034,
+      "step": 975
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 1.5320487022399902,
+      "eval_runtime": 11.5033,
+      "eval_samples_per_second": 21.385,
+      "eval_steps_per_second": 2.695,
+      "step": 975
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 2.2244866199319123e-07,
+      "loss": 1.549,
+      "step": 980
+    },
+    {
+      "epoch": 5.05,
+      "learning_rate": 8.883519830391636e-07,
+      "loss": 1.5255,
+      "step": 985
+    },
+    {
+      "epoch": 5.08,
+      "learning_rate": 1.9933913245728015e-06,
+      "loss": 1.5572,
+      "step": 990
+    },
+    {
+      "epoch": 5.1,
+      "learning_rate": 3.5304000824257744e-06,
+      "loss": 1.5562,
+      "step": 995
+    },
+    {
+      "epoch": 5.13,
+      "learning_rate": 5.489410165004689e-06,
+      "loss": 1.5428,
+      "step": 1000
+    },
+    {
+      "epoch": 5.15,
+      "learning_rate": 7.857716640189824e-06,
+      "loss": 1.458,
+      "step": 1005
+    },
+    {
+      "epoch": 5.18,
+      "learning_rate": 1.061996013169569e-05,
+      "loss": 1.5417,
+      "step": 1010
+    },
+    {
+      "epoch": 5.21,
+      "learning_rate": 1.375822643051982e-05,
+      "loss": 1.5147,
+      "step": 1015
+    },
+    {
+      "epoch": 5.23,
+      "learning_rate": 1.725216267546245e-05,
+      "loss": 1.4622,
+      "step": 1020
+    },
+    {
+      "epoch": 5.26,
+      "learning_rate": 2.1079109349241446e-05,
+      "loss": 1.5768,
+      "step": 1025
+    },
+    {
+      "epoch": 5.28,
+      "learning_rate": 2.5214247234157066e-05,
+      "loss": 1.48,
+      "step": 1030
+    },
+    {
+      "epoch": 5.31,
+      "learning_rate": 2.963075837424263e-05,
+      "loss": 1.4608,
+      "step": 1035
+    },
+    {
+      "epoch": 5.33,
+      "learning_rate": 3.429999999999989e-05,
+      "loss": 1.511,
+      "step": 1040
+    },
+    {
+      "epoch": 5.36,
+      "learning_rate": 3.919169028775035e-05,
+      "loss": 1.4951,
+      "step": 1045
+    },
+    {
+      "epoch": 5.38,
+      "learning_rate": 4.4274104748882125e-05,
+      "loss": 1.4858,
+      "step": 1050
+    },
+    {
+      "epoch": 5.41,
+      "learning_rate": 4.951428197533139e-05,
+      "loss": 1.5487,
+      "step": 1055
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 5.487823740696337e-05,
+      "loss": 1.4313,
+      "step": 1060
+    },
+    {
+      "epoch": 5.46,
+      "learning_rate": 6.033118373448483e-05,
+      "loss": 1.58,
+      "step": 1065
+    },
+    {
+      "epoch": 5.49,
+      "learning_rate": 6.583775650849409e-05,
+      "loss": 1.5329,
+      "step": 1070
+    },
+    {
+      "epoch": 5.51,
+      "learning_rate": 7.136224349150582e-05,
+      "loss": 1.5095,
+      "step": 1075
+    },
+    {
+      "epoch": 5.54,
+      "learning_rate": 7.686881626551508e-05,
+      "loss": 1.4715,
+      "step": 1080
+    },
+    {
+      "epoch": 5.56,
+      "learning_rate": 8.232176259303654e-05,
+      "loss": 1.5381,
+      "step": 1085
+    },
+    {
+      "epoch": 5.59,
+      "learning_rate": 8.768571802466852e-05,
+      "loss": 1.4487,
+      "step": 1090
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 9.292589525111778e-05,
+      "loss": 1.4744,
+      "step": 1095
+    },
+    {
+      "epoch": 5.64,
+      "learning_rate": 9.800830971224957e-05,
+      "loss": 1.533,
+      "step": 1100
+    },
+    {
+      "epoch": 5.67,
+      "learning_rate": 0.00010290000000000003,
+      "loss": 1.4971,
+      "step": 1105
+    },
+    {
+      "epoch": 5.69,
+      "learning_rate": 0.0001075692416257573,
+      "loss": 1.5636,
+      "step": 1110
+    },
+    {
+      "epoch": 5.72,
+      "learning_rate": 0.00011198575276584287,
+      "loss": 1.6137,
+      "step": 1115
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 0.0001161208906507585,
+      "loss": 1.6421,
+      "step": 1120
+    },
+    {
+      "epoch": 5.77,
+      "learning_rate": 0.00011994783732453749,
+      "loss": 1.5246,
+      "step": 1125
+    },
+    {
+      "epoch": 5.79,
+      "learning_rate": 0.00012344177356948013,
+      "loss": 1.5616,
+      "step": 1130
+    },
+    {
+      "epoch": 5.82,
+      "learning_rate": 0.00012658003986830427,
+      "loss": 1.5668,
+      "step": 1135
+    },
+    {
+      "epoch": 5.85,
+      "learning_rate": 0.00012934228335981013,
+      "loss": 1.4651,
+      "step": 1140
+    },
+    {
+      "epoch": 5.87,
+      "learning_rate": 0.00013171058983499527,
+      "loss": 1.5195,
+      "step": 1145
+    },
+    {
+      "epoch": 5.9,
+      "learning_rate": 0.0001336695999175742,
+      "loss": 1.571,
+      "step": 1150
+    },
+    {
+      "epoch": 5.92,
+      "learning_rate": 0.00013520660867542716,
+      "loss": 1.5089,
+      "step": 1155
+    },
+    {
+      "epoch": 5.95,
+      "learning_rate": 0.00013631164801696083,
+      "loss": 1.5889,
+      "step": 1160
+    },
+    {
+      "epoch": 5.97,
+      "learning_rate": 0.00013697755133800678,
+      "loss": 1.4905,
+      "step": 1165
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.0001372,
+      "loss": 1.4218,
+      "step": 1170
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.549699306488037,
+      "eval_runtime": 11.5077,
+      "eval_samples_per_second": 21.377,
+      "eval_steps_per_second": 2.694,
+      "step": 1170
+    },
+    {
+      "epoch": 6.03,
+      "learning_rate": 0.0001369775513380068,
+      "loss": 1.5342,
+      "step": 1175
+    },
+    {
+      "epoch": 6.05,
+      "learning_rate": 0.00013631164801696085,
+      "loss": 1.5204,
+      "step": 1180
+    },
+    {
+      "epoch": 6.08,
+      "learning_rate": 0.0001352066086754272,
+      "loss": 1.4333,
+      "step": 1185
+    },
+    {
+      "epoch": 6.1,
+      "learning_rate": 0.00013366959991757425,
+      "loss": 1.4939,
+      "step": 1190
+    },
+    {
+      "epoch": 6.13,
+      "learning_rate": 0.00013171058983499532,
+      "loss": 1.5237,
+      "step": 1195
+    },
+    {
+      "epoch": 6.15,
+      "learning_rate": 0.00012934228335981018,
+      "loss": 1.5111,
+      "step": 1200
+    },
+    {
+      "epoch": 6.18,
+      "learning_rate": 0.00012658003986830432,
+      "loss": 1.5271,
+      "step": 1205
+    },
+    {
+      "epoch": 6.21,
+      "learning_rate": 0.0001234417735694802,
+      "loss": 1.4337,
+      "step": 1210
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 0.00011994783732453757,
+      "loss": 1.5122,
+      "step": 1215
+    },
+    {
+      "epoch": 6.26,
+      "learning_rate": 0.00011612089065075855,
+      "loss": 1.5887,
+      "step": 1220
+    },
+    {
+      "epoch": 6.28,
+      "learning_rate": 0.00011198575276584294,
+      "loss": 1.4314,
+      "step": 1225
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 0.00010756924162575738,
+      "loss": 1.3917,
+      "step": 1230
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 0.00010290000000000012,
+      "loss": 1.4896,
+      "step": 1235
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 9.800830971224966e-05,
+      "loss": 1.4996,
+      "step": 1240
+    },
+    {
+      "epoch": 6.38,
+      "learning_rate": 9.292589525111789e-05,
+      "loss": 1.4866,
+      "step": 1245
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 8.768571802466861e-05,
+      "loss": 1.4984,
+      "step": 1250
+    },
+    {
+      "epoch": 6.44,
+      "learning_rate": 8.232176259303664e-05,
+      "loss": 1.47,
+      "step": 1255
+    },
+    {
+      "epoch": 6.46,
+      "learning_rate": 7.686881626551518e-05,
+      "loss": 1.5694,
+      "step": 1260
+    },
+    {
+      "epoch": 6.49,
+      "learning_rate": 7.136224349150593e-05,
+      "loss": 1.4892,
+      "step": 1265
+    },
+    {
+      "epoch": 6.51,
+      "learning_rate": 6.583775650849418e-05,
+      "loss": 1.5118,
+      "step": 1270
+    },
+    {
+      "epoch": 6.54,
+      "learning_rate": 6.033118373448493e-05,
+      "loss": 1.4075,
+      "step": 1275
+    },
+    {
+      "epoch": 6.56,
+      "learning_rate": 5.487823740696347e-05,
+      "loss": 1.3875,
+      "step": 1280
+    },
+    {
+      "epoch": 6.59,
+      "learning_rate": 4.9514281975331485e-05,
+      "loss": 1.4728,
+      "step": 1285
+    },
+    {
+      "epoch": 6.62,
+      "learning_rate": 4.427410474888221e-05,
+      "loss": 1.484,
+      "step": 1290
+    },
+    {
+      "epoch": 6.64,
+      "learning_rate": 3.919169028775044e-05,
+      "loss": 1.5624,
+      "step": 1295
+    },
+    {
+      "epoch": 6.67,
+      "learning_rate": 3.429999999999998e-05,
+      "loss": 1.4664,
+      "step": 1300
+    },
+    {
+      "epoch": 6.69,
+      "learning_rate": 2.9630758374242713e-05,
+      "loss": 1.4719,
+      "step": 1305
+    },
+    {
+      "epoch": 6.72,
+      "learning_rate": 2.521424723415714e-05,
+      "loss": 1.5392,
+      "step": 1310
+    },
+    {
+      "epoch": 6.74,
+      "learning_rate": 2.107910934924152e-05,
+      "loss": 1.3993,
+      "step": 1315
+    },
+    {
+      "epoch": 6.77,
+      "learning_rate": 1.725216267546251e-05,
+      "loss": 1.446,
+      "step": 1320
+    },
+    {
+      "epoch": 6.79,
+      "learning_rate": 1.375822643051988e-05,
+      "loss": 1.3712,
+      "step": 1325
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 1.0619960131695743e-05,
+      "loss": 1.4234,
+      "step": 1330
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 7.857716640189868e-06,
+      "loss": 1.4638,
+      "step": 1335
+    },
+    {
+      "epoch": 6.87,
+      "learning_rate": 5.489410165004727e-06,
+      "loss": 1.4316,
+      "step": 1340
+    },
+    {
+      "epoch": 6.9,
+      "learning_rate": 3.530400082425805e-06,
+      "loss": 1.4005,
+      "step": 1345
+    },
+    {
+      "epoch": 6.92,
+      "learning_rate": 1.9933913245728244e-06,
+      "loss": 1.4936,
+      "step": 1350
+    },
+    {
+      "epoch": 6.95,
+      "learning_rate": 8.883519830391788e-07,
+      "loss": 1.4579,
+      "step": 1355
+    },
+    {
+      "epoch": 6.97,
+      "learning_rate": 2.2244866199319883e-07,
+      "loss": 1.4931,
+      "step": 1360
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.0,
+      "loss": 1.5549,
+      "step": 1365
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 1.5031161308288574,
+      "eval_runtime": 11.524,
+      "eval_samples_per_second": 21.347,
+      "eval_steps_per_second": 2.69,
+      "step": 1365
+    },
+    {
+      "epoch": 7.03,
+      "learning_rate": 2.2244866199319123e-07,
+      "loss": 1.3518,
+      "step": 1370
+    },
+    {
+      "epoch": 7.05,
+      "learning_rate": 8.883519830391636e-07,
+      "loss": 1.331,
+      "step": 1375
+    },
+    {
+      "epoch": 7.08,
+      "learning_rate": 1.9933913245728015e-06,
+      "loss": 1.3751,
+      "step": 1380
+    },
+    {
+      "epoch": 7.1,
+      "learning_rate": 3.530400082425767e-06,
+      "loss": 1.3551,
+      "step": 1385
+    },
+    {
+      "epoch": 7.13,
+      "learning_rate": 5.489410165004681e-06,
+      "loss": 1.516,
+      "step": 1390
+    },
+    {
+      "epoch": 7.15,
+      "learning_rate": 7.857716640189815e-06,
+      "loss": 1.3005,
+      "step": 1395
+    },
+    {
+      "epoch": 7.18,
+      "learning_rate": 1.0619960131695684e-05,
+      "loss": 1.36,
+      "step": 1400
+    },
+    {
+      "epoch": 7.21,
+      "learning_rate": 1.3758226430519812e-05,
+      "loss": 1.4466,
+      "step": 1405
+    },
+    {
+      "epoch": 7.23,
+      "learning_rate": 1.7252162675462436e-05,
+      "loss": 1.3912,
+      "step": 1410
+    },
+    {
+      "epoch": 7.26,
+      "learning_rate": 2.1079109349241436e-05,
+      "loss": 1.4354,
+      "step": 1415
+    },
+    {
+      "epoch": 7.28,
+      "learning_rate": 2.521424723415705e-05,
+      "loss": 1.3196,
+      "step": 1420
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 2.963075837424261e-05,
+      "loss": 1.4088,
+      "step": 1425
+    },
+    {
+      "epoch": 7.33,
+      "learning_rate": 3.429999999999988e-05,
+      "loss": 1.3855,
+      "step": 1430
+    },
+    {
+      "epoch": 7.36,
+      "learning_rate": 3.919169028775034e-05,
+      "loss": 1.4227,
+      "step": 1435
+    },
+    {
+      "epoch": 7.38,
+      "learning_rate": 4.4274104748882104e-05,
+      "loss": 1.3923,
+      "step": 1440
+    },
+    {
+      "epoch": 7.41,
+      "learning_rate": 4.951428197533137e-05,
+      "loss": 1.509,
+      "step": 1445
+    },
+    {
+      "epoch": 7.44,
+      "learning_rate": 5.4878237406963356e-05,
+      "loss": 1.4401,
+      "step": 1450
+    },
+    {
+      "epoch": 7.46,
+      "learning_rate": 6.0331183734484816e-05,
+      "loss": 1.4038,
+      "step": 1455
+    },
+    {
+      "epoch": 7.49,
+      "learning_rate": 6.583775650849407e-05,
+      "loss": 1.4624,
+      "step": 1460
+    },
+    {
+      "epoch": 7.51,
+      "learning_rate": 7.13622434915058e-05,
+      "loss": 1.4675,
+      "step": 1465
+    },
+    {
+      "epoch": 7.54,
+      "learning_rate": 7.686881626551506e-05,
+      "loss": 1.3629,
+      "step": 1470
+    },
+    {
+      "epoch": 7.56,
+      "learning_rate": 8.232176259303652e-05,
+      "loss": 1.4762,
+      "step": 1475
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 8.76857180246685e-05,
+      "loss": 1.4092,
+      "step": 1480
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 9.292589525111778e-05,
+      "loss": 1.4516,
+      "step": 1485
+    },
+    {
+      "epoch": 7.64,
+      "learning_rate": 9.800830971224955e-05,
+      "loss": 1.4641,
+      "step": 1490
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 0.00010290000000000002,
+      "loss": 1.4513,
+      "step": 1495
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 0.00010756924162575728,
+      "loss": 1.4179,
+      "step": 1500
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 0.00011198575276584284,
+      "loss": 1.4565,
+      "step": 1505
+    },
+    {
+      "epoch": 7.74,
+      "learning_rate": 0.00011612089065075847,
+      "loss": 1.4936,
+      "step": 1510
+    },
+    {
+      "epoch": 7.77,
+      "learning_rate": 0.00011994783732453747,
+      "loss": 1.4875,
+      "step": 1515
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 0.0001234417735694801,
+      "loss": 1.4635,
+      "step": 1520
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 0.00012658003986830424,
+      "loss": 1.4118,
+      "step": 1525
+    },
+    {
+      "epoch": 7.85,
+      "learning_rate": 0.00012934228335981013,
+      "loss": 1.3592,
+      "step": 1530
+    },
+    {
+      "epoch": 7.87,
+      "learning_rate": 0.00013171058983499527,
+      "loss": 1.4221,
+      "step": 1535
+    },
+    {
+      "epoch": 7.9,
+      "learning_rate": 0.0001336695999175742,
+      "loss": 1.4614,
+      "step": 1540
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 0.00013520660867542716,
+      "loss": 1.4663,
+      "step": 1545
+    },
+    {
+      "epoch": 7.95,
+      "learning_rate": 0.00013631164801696083,
+      "loss": 1.5087,
+      "step": 1550
+    },
+    {
+      "epoch": 7.97,
+      "learning_rate": 0.00013697755133800678,
+      "loss": 1.449,
+      "step": 1555
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.0001372,
+      "loss": 1.4238,
+      "step": 1560
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 1.5225884914398193,
+      "eval_runtime": 11.5201,
+      "eval_samples_per_second": 21.354,
+      "eval_steps_per_second": 2.691,
+      "step": 1560
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 0.0001369775513380068,
+      "loss": 1.4416,
+      "step": 1565
+    },
+    {
+      "epoch": 8.05,
+      "learning_rate": 0.00013631164801696085,
+      "loss": 1.4836,
+      "step": 1570
+    },
+    {
+      "epoch": 8.08,
+      "learning_rate": 0.0001352066086754272,
+      "loss": 1.3901,
+      "step": 1575
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 0.00013366959991757425,
+      "loss": 1.2685,
+      "step": 1580
+    },
+    {
+      "epoch": 8.13,
+      "learning_rate": 0.00013171058983499543,
+      "loss": 1.4028,
+      "step": 1585
+    },
+    {
+      "epoch": 8.15,
+      "learning_rate": 0.0001293422833598103,
+      "loss": 1.4551,
+      "step": 1590
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 0.00012658003986830446,
+      "loss": 1.4609,
+      "step": 1595
+    },
+    {
+      "epoch": 8.21,
+      "learning_rate": 0.00012344177356948035,
+      "loss": 1.3414,
+      "step": 1600
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 0.00011994783732453773,
+      "loss": 1.3587,
+      "step": 1605
+    },
+    {
+      "epoch": 8.26,
+      "learning_rate": 0.0001161208906507584,
+      "loss": 1.3815,
+      "step": 1610
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 0.00011198575276584276,
+      "loss": 1.434,
+      "step": 1615
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 0.00010756924162575719,
+      "loss": 1.3752,
+      "step": 1620
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 0.00010289999999999993,
+      "loss": 1.4958,
+      "step": 1625
+    },
+    {
+      "epoch": 8.36,
+      "learning_rate": 9.800830971224945e-05,
+      "loss": 1.4,
+      "step": 1630
+    },
+    {
+      "epoch": 8.38,
+      "learning_rate": 9.29258952511179e-05,
+      "loss": 1.3682,
+      "step": 1635
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 8.768571802466864e-05,
+      "loss": 1.4391,
+      "step": 1640
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 8.232176259303665e-05,
+      "loss": 1.3929,
+      "step": 1645
+    },
+    {
+      "epoch": 8.46,
+      "learning_rate": 7.68688162655152e-05,
+      "loss": 1.3586,
+      "step": 1650
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 7.136224349150594e-05,
+      "loss": 1.5063,
+      "step": 1655
+    },
+    {
+      "epoch": 8.51,
+      "learning_rate": 6.583775650849421e-05,
+      "loss": 1.3837,
+      "step": 1660
+    },
+    {
+      "epoch": 8.54,
+      "learning_rate": 6.0331183734484945e-05,
+      "loss": 1.3909,
+      "step": 1665
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 5.4878237406963485e-05,
+      "loss": 1.4117,
+      "step": 1670
+    },
+    {
+      "epoch": 8.59,
+      "learning_rate": 4.95142819753315e-05,
+      "loss": 1.3474,
+      "step": 1675
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 4.4274104748882226e-05,
+      "loss": 1.3716,
+      "step": 1680
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 3.919169028775067e-05,
+      "loss": 1.417,
+      "step": 1685
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 3.4300000000000204e-05,
+      "loss": 1.4424,
+      "step": 1690
+    },
+    {
+      "epoch": 8.69,
+      "learning_rate": 2.9630758374242923e-05,
+      "loss": 1.457,
+      "step": 1695
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 2.521424723415734e-05,
+      "loss": 1.41,
+      "step": 1700
+    },
+    {
+      "epoch": 8.74,
+      "learning_rate": 2.1079109349241704e-05,
+      "loss": 1.3425,
+      "step": 1705
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 1.7252162675462368e-05,
+      "loss": 1.3511,
+      "step": 1710
+    },
+    {
+      "epoch": 8.79,
+      "learning_rate": 1.3758226430519743e-05,
+      "loss": 1.3591,
+      "step": 1715
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 1.061996013169563e-05,
+      "loss": 1.3275,
+      "step": 1720
+    },
+    {
+      "epoch": 8.85,
+      "learning_rate": 7.85771664018977e-06,
+      "loss": 1.454,
+      "step": 1725
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 5.489410165004643e-06,
+      "loss": 1.3359,
+      "step": 1730
+    },
+    {
+      "epoch": 8.9,
+      "learning_rate": 3.5304000824258126e-06,
+      "loss": 1.4167,
+      "step": 1735
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 1.993391324572832e-06,
+      "loss": 1.2789,
+      "step": 1740
+    },
+    {
+      "epoch": 8.95,
+      "learning_rate": 8.883519830391788e-07,
+      "loss": 1.3723,
+      "step": 1745
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 2.2244866199319883e-07,
+      "loss": 1.3274,
+      "step": 1750
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.0,
+      "loss": 1.3058,
+      "step": 1755
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 1.4917341470718384,
+      "eval_runtime": 11.5163,
+      "eval_samples_per_second": 21.361,
+      "eval_steps_per_second": 2.692,
+      "step": 1755
+    }
+  ],
+  "max_steps": 1950,
+  "num_train_epochs": 10,
+  "total_flos": 1834270064640000.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:998a9f66eac7ad3e593dcd5b7ffd120fd9628e28379c2c878569fc65ecf51e10
+size 2671