initial commit

README.md

@@ -1,3 +1,72 @@
 ---
 license: apache-2.0
+datasets:
+- LumiOpen/instruction-collection-fin
+language:
+- fi
+- en
 ---
+<div align="center">
+<img src="./poro-logo.png" width="200px">
+</div>
+
+# Poro 34B Chat
+
+Poro 34b chat is a chat-tuned version of [Poro 34B](https://huggingface.co/LumiOpen/Poro-34B) trained to follow instructions in both Finnish and English.
+
+Because of the limited amount of instruction tuning available for Finnish, documents from the English datasets were machine-translated by the Poro 34B base model into Finnish, then used to train this chat version.  We selected only datasets that are available for commercial use and only contain synthetic data if it was gathered in ToS-compliant fashion.
+
+More information about the data selection and translation process for our Finnish dataset are available on the [LumiOpen/instruction-collection-fin](https://huggingface.co/datasets/LumiOpen/instruction-collection-fin) page.
+
+Poro was created in a collaboration between [SiloGen](https://www.silo.ai/silogen) from [Silo AI](https://www.silo.ai/), the [TurkuNLP group](https://turkunlp.org/) of the University of Turku, and [High Performance Language Technologies](https://hplt-project.org/) (HPLT).  Training was conducted on the [LUMI supercomputer](https://www.lumi-supercomputer.eu/), using compute resources generously provided by [CSC](https://csc.fi/) - IT Center for Science, Finland.
+
+This project is part of an ongoing effort to create open source large language models for non-English and especially low resource languages like Finnish.  Through the combination of English and Finnish training data we get a model that outperforms previous Finnish only models, while also being fluent in English and code, and capable of basic translation between English and Finnish.
+
+
+## Fine Tuning
+Zephyr--??? TODO
+
+
+## Datasets
+
+#### Finnish and Cross-lingual
+- [LumiOpen/instruction-collection-fin](https://huggingface.co/datasets/LumiOpen/instruction-collection-fin)
+
+#### English
+- [databricks-dolly-15k](https://huggingface.co/datasets/databricks/databricks-dolly-15k)
+- [Curated OASST2](https://huggingface.co/datasets/sablo/oasst2_curated)
+- [Argilla/10k_prompts_ranked_mistral_large_responses](https://huggingface.co/datasets/argilla/10k_prompts_ranked_mistral_large_responses)
+
+
+
+## Evaluations
+
+We relied on the popular MTBench benchmark to evaluate multi-turn performance.
+
+Since MTBench is an English only benchmark, we also release this fork of [MTBench Finnish](https://github.com/LumiOpen/FastChat/tree/main/fastchat/llm_judge) with multilingual support and machine translated Finnish prompts.  Our scores for both benchmarks follow.
+
+| Eval | Score  |
+| :-------------- | :----: |
+| MTBench         | 5.93   |
+| MTBench Finnish | 5.90   |
+
+
+## License
+
+Poro 34B chat is released under the Apache 2.0 license.
+
+
+## Citation
+
+```
+@misc{luukkonen2024poro,
+      title={Poro 34B and the Blessing of Multilinguality},
+      author={Risto Luukkonen and Jonathan Burdge and Elaine Zosa and Aarne
+Talman and Ville Komulainen and Väinö Hatanpää and Peter Sarlin and Sampo
+Pyysalo},
+      year={2024},
+      eprint={2404.01856},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}   
+```

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 1.8350725173950195,
+    "eval_runtime": 157.2695,
+    "eval_samples": 1680,
+    "eval_samples_per_second": 10.682,
+    "eval_steps_per_second": 0.114,
+    "train_loss": 0.8629729963555334,
+    "train_runtime": 46452.2696,
+    "train_samples": 30684,
+    "train_samples_per_second": 1.982,
+    "train_steps_per_second": 0.041
+}

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "/scratch/project_462000319/zosaelai2/models/Poro-34b-Chat",
+  "apply_residual_connection_post_layernorm": false,
+  "architectures": [
+    "BloomForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "attention_softmax_in_fp32": true,
+  "bos_token_id": 1,
+  "eos_token_id": 5,
+  "hidden_dropout": 0.0,
+  "hidden_size": 7168,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "masked_softmax_fusion": true,
+  "model_type": "bloom",
+  "n_head": 56,
+  "n_layer": 54,
+  "pad_token_id": 3,
+  "pretraining_tp": 2,
+  "slow_but_exact": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.35.1",
+  "use_cache": true,
+  "vocab_size": 128000
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 1.8350725173950195,
+    "eval_runtime": 157.2695,
+    "eval_samples": 1680,
+    "eval_samples_per_second": 10.682,
+    "eval_steps_per_second": 0.114
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 5,
+  "pad_token_id": 3,
+  "transformers_version": "4.35.1",
+  "use_cache": false
+}

model-00001-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38de9c1e4633685421dcaa683fd2a8cb7c70bd5cbb0c682af921fd2f615bcf46
+size 4712820784

model-00002-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e7a6c0f5177d3c3a2d9b069e6b23759f9a09a8f88e664485cbdfe8b3e2569cb
+size 4933252680

model-00003-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ada11489dda3f478d6ff1029575da42caf868bba11265f130de21668cdfdfac
+size 4933252648

model-00004-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec56cb3be444b4fa904691379e681f50f7d78631cf291df82e4cd1bb99b8583e
+size 4933252728

model-00005-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f373d669efe898bf2be116dbba6d16adaf67c2d27b9da180e90925c1d423174
+size 4933252728

model-00006-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:512f2eb21b8da914743a058d9eafca045bf892d9e1d76ed6383f1985a115351e
+size 4933252728

model-00007-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4742908ec02fe0bd16bb25cbacffbfbdb11f841742fc7c3ef05d93ac462dec18
+size 4933252728

model-00008-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:553770591bd517202774133836f44f184b2c17fe479a6db02901a4c20540bca9
+size 4933252728

model-00009-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9859f40805916de43ffb88038c97c391baa7c6a02329bc7dab721ac29cce52c
+size 4933252728

model-00010-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f23589843bf8f5b8c816ef0a6d4f83b8799e08683875f68428ac12e231a40183
+size 4933252728

model-00011-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55617a0c812b930e4e66203e86428fdb3d72c9512717b1034ecc66fa5becaaa9
+size 4933252728

model-00012-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9bb9fbca6395d7cb37c72337ec75fb8d093eb4897b7c349ee6aabec6746630a9
+size 4933252728

model-00013-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01908d02c716c09520e3b187a11ce824baacd7d09bde52ecac25a8edd7be54ea
+size 4933252728

model-00014-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6b5d6327d01ab793387803ca6bf85a8a8cd6a9cc0549a786b44eb301e381801
+size 4522124144

model.safetensors.index.json

@@ -0,0 +1,660 @@
+{
+  "metadata": {
+    "total_size": 68433899520
+  },
+  "weight_map": {
+    "transformer.h.0.input_layernorm.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.input_layernorm.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.0.mlp.dense_4h_to_h.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.mlp.dense_4h_to_h.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.0.mlp.dense_h_to_4h.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.mlp.dense_h_to_4h.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.0.post_attention_layernorm.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.0.self_attention.dense.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.self_attention.dense.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.0.self_attention.query_key_value.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.self_attention.query_key_value.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.input_layernorm.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.input_layernorm.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.mlp.dense_4h_to_h.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.mlp.dense_4h_to_h.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.mlp.dense_h_to_4h.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.mlp.dense_h_to_4h.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.post_attention_layernorm.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.self_attention.dense.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.self_attention.dense.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.self_attention.query_key_value.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.self_attention.query_key_value.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.10.input_layernorm.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.10.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.10.mlp.dense_4h_to_h.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.10.mlp.dense_4h_to_h.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.10.mlp.dense_h_to_4h.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.10.mlp.dense_h_to_4h.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.10.post_attention_layernorm.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.10.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.10.self_attention.dense.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.10.self_attention.dense.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.10.self_attention.query_key_value.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.10.self_attention.query_key_value.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.11.input_layernorm.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.11.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.11.mlp.dense_4h_to_h.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.11.mlp.dense_4h_to_h.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.11.mlp.dense_h_to_4h.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.11.mlp.dense_h_to_4h.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.11.post_attention_layernorm.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.11.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.11.self_attention.dense.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.11.self_attention.dense.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.11.self_attention.query_key_value.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.11.self_attention.query_key_value.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.12.input_layernorm.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.12.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.12.mlp.dense_4h_to_h.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.12.mlp.dense_4h_to_h.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.12.mlp.dense_h_to_4h.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.12.mlp.dense_h_to_4h.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.12.post_attention_layernorm.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.12.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.12.self_attention.dense.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.12.self_attention.dense.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.12.self_attention.query_key_value.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.12.self_attention.query_key_value.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.13.input_layernorm.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.13.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.13.mlp.dense_4h_to_h.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.13.mlp.dense_4h_to_h.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.13.mlp.dense_h_to_4h.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.13.mlp.dense_h_to_4h.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.13.post_attention_layernorm.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.13.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.13.self_attention.dense.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.13.self_attention.dense.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.13.self_attention.query_key_value.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.13.self_attention.query_key_value.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.14.input_layernorm.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.14.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.14.mlp.dense_4h_to_h.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.14.mlp.dense_4h_to_h.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.14.mlp.dense_h_to_4h.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.14.mlp.dense_h_to_4h.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.14.post_attention_layernorm.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.14.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.14.self_attention.dense.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.14.self_attention.dense.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.14.self_attention.query_key_value.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.14.self_attention.query_key_value.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.15.input_layernorm.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.15.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.15.mlp.dense_4h_to_h.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.15.mlp.dense_4h_to_h.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.15.mlp.dense_h_to_4h.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.15.mlp.dense_h_to_4h.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.15.post_attention_layernorm.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.15.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.15.self_attention.dense.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.15.self_attention.dense.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.15.self_attention.query_key_value.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.15.self_attention.query_key_value.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.16.input_layernorm.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.16.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.16.mlp.dense_4h_to_h.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.16.mlp.dense_4h_to_h.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.16.mlp.dense_h_to_4h.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.16.mlp.dense_h_to_4h.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.16.post_attention_layernorm.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.16.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.16.self_attention.dense.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.16.self_attention.dense.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.16.self_attention.query_key_value.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.16.self_attention.query_key_value.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.17.input_layernorm.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.17.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.17.mlp.dense_4h_to_h.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.17.mlp.dense_4h_to_h.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.17.mlp.dense_h_to_4h.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.17.mlp.dense_h_to_4h.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.17.post_attention_layernorm.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.17.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.17.self_attention.dense.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.17.self_attention.dense.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.17.self_attention.query_key_value.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.17.self_attention.query_key_value.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.18.input_layernorm.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.18.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.18.mlp.dense_4h_to_h.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.18.mlp.dense_4h_to_h.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.18.mlp.dense_h_to_4h.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.18.mlp.dense_h_to_4h.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.18.post_attention_layernorm.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.18.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.18.self_attention.dense.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.18.self_attention.dense.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.18.self_attention.query_key_value.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.18.self_attention.query_key_value.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.19.input_layernorm.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.19.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.19.mlp.dense_4h_to_h.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.19.mlp.dense_4h_to_h.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.19.mlp.dense_h_to_4h.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.19.mlp.dense_h_to_4h.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.19.post_attention_layernorm.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.19.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.19.self_attention.dense.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.19.self_attention.dense.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.19.self_attention.query_key_value.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.19.self_attention.query_key_value.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.2.input_layernorm.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.2.input_layernorm.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.2.mlp.dense_4h_to_h.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.2.mlp.dense_4h_to_h.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.2.mlp.dense_h_to_4h.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.2.mlp.dense_h_to_4h.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.2.post_attention_layernorm.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.2.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.2.self_attention.dense.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.2.self_attention.dense.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.2.self_attention.query_key_value.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.2.self_attention.query_key_value.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.20.input_layernorm.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.20.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.20.mlp.dense_4h_to_h.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.20.mlp.dense_4h_to_h.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.20.mlp.dense_h_to_4h.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.20.mlp.dense_h_to_4h.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.20.post_attention_layernorm.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.20.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.20.self_attention.dense.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.20.self_attention.dense.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.20.self_attention.query_key_value.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.20.self_attention.query_key_value.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.21.input_layernorm.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.21.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.21.mlp.dense_4h_to_h.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.21.mlp.dense_4h_to_h.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.21.mlp.dense_h_to_4h.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.21.mlp.dense_h_to_4h.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.21.post_attention_layernorm.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.21.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.21.self_attention.dense.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.21.self_attention.dense.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.21.self_attention.query_key_value.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.21.self_attention.query_key_value.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.22.input_layernorm.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.22.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.22.mlp.dense_4h_to_h.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.22.mlp.dense_4h_to_h.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.22.mlp.dense_h_to_4h.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.22.mlp.dense_h_to_4h.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.22.post_attention_layernorm.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.22.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.22.self_attention.dense.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.22.self_attention.dense.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.22.self_attention.query_key_value.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.22.self_attention.query_key_value.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.23.input_layernorm.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.23.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.23.mlp.dense_4h_to_h.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.23.mlp.dense_4h_to_h.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.23.mlp.dense_h_to_4h.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.23.mlp.dense_h_to_4h.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.23.post_attention_layernorm.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.23.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.23.self_attention.dense.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.23.self_attention.dense.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.23.self_attention.query_key_value.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.23.self_attention.query_key_value.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.24.input_layernorm.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.24.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.24.mlp.dense_4h_to_h.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.24.mlp.dense_4h_to_h.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.24.mlp.dense_h_to_4h.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.24.mlp.dense_h_to_4h.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.24.post_attention_layernorm.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.24.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.24.self_attention.dense.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.24.self_attention.dense.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.24.self_attention.query_key_value.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.24.self_attention.query_key_value.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.25.input_layernorm.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.25.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.25.mlp.dense_4h_to_h.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.25.mlp.dense_4h_to_h.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.25.mlp.dense_h_to_4h.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.25.mlp.dense_h_to_4h.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.25.post_attention_layernorm.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.25.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.25.self_attention.dense.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.25.self_attention.dense.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.25.self_attention.query_key_value.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.25.self_attention.query_key_value.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.26.input_layernorm.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.26.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.26.mlp.dense_4h_to_h.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.26.mlp.dense_4h_to_h.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.26.mlp.dense_h_to_4h.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.26.mlp.dense_h_to_4h.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.26.post_attention_layernorm.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.26.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.26.self_attention.dense.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.26.self_attention.dense.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.26.self_attention.query_key_value.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.26.self_attention.query_key_value.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.27.input_layernorm.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.27.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.27.mlp.dense_4h_to_h.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.27.mlp.dense_4h_to_h.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.27.mlp.dense_h_to_4h.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.27.mlp.dense_h_to_4h.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.27.post_attention_layernorm.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.27.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.27.self_attention.dense.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.27.self_attention.dense.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.27.self_attention.query_key_value.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.27.self_attention.query_key_value.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.28.input_layernorm.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.28.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.28.mlp.dense_4h_to_h.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.28.mlp.dense_4h_to_h.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.28.mlp.dense_h_to_4h.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.28.mlp.dense_h_to_4h.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.28.post_attention_layernorm.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.28.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.28.self_attention.dense.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.28.self_attention.dense.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.28.self_attention.query_key_value.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.28.self_attention.query_key_value.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.29.input_layernorm.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.29.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.29.mlp.dense_4h_to_h.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.29.mlp.dense_4h_to_h.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.29.mlp.dense_h_to_4h.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.29.mlp.dense_h_to_4h.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.29.post_attention_layernorm.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.29.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.29.self_attention.dense.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.29.self_attention.dense.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.29.self_attention.query_key_value.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.29.self_attention.query_key_value.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.3.input_layernorm.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.3.mlp.dense_4h_to_h.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.mlp.dense_4h_to_h.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.3.mlp.dense_h_to_4h.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.mlp.dense_h_to_4h.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.3.post_attention_layernorm.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.3.self_attention.dense.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.self_attention.dense.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.3.self_attention.query_key_value.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.self_attention.query_key_value.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.30.input_layernorm.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.30.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.30.mlp.dense_4h_to_h.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.30.mlp.dense_4h_to_h.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.30.mlp.dense_h_to_4h.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.30.mlp.dense_h_to_4h.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.30.post_attention_layernorm.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.30.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.30.self_attention.dense.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.30.self_attention.dense.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.30.self_attention.query_key_value.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.30.self_attention.query_key_value.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.31.input_layernorm.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.31.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.31.mlp.dense_4h_to_h.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.31.mlp.dense_4h_to_h.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.31.mlp.dense_h_to_4h.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.31.mlp.dense_h_to_4h.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.31.post_attention_layernorm.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.31.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.31.self_attention.dense.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.31.self_attention.dense.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.31.self_attention.query_key_value.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.31.self_attention.query_key_value.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.32.input_layernorm.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.32.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.32.mlp.dense_4h_to_h.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.32.mlp.dense_4h_to_h.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.32.mlp.dense_h_to_4h.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.32.mlp.dense_h_to_4h.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.32.post_attention_layernorm.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.32.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.32.self_attention.dense.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.32.self_attention.dense.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.32.self_attention.query_key_value.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.32.self_attention.query_key_value.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.33.input_layernorm.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.33.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.33.mlp.dense_4h_to_h.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.33.mlp.dense_4h_to_h.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.33.mlp.dense_h_to_4h.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.33.mlp.dense_h_to_4h.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.33.post_attention_layernorm.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.33.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.33.self_attention.dense.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.33.self_attention.dense.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.33.self_attention.query_key_value.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.33.self_attention.query_key_value.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.34.input_layernorm.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.34.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.34.mlp.dense_4h_to_h.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.34.mlp.dense_4h_to_h.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.34.mlp.dense_h_to_4h.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.34.mlp.dense_h_to_4h.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.34.post_attention_layernorm.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.34.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.34.self_attention.dense.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.34.self_attention.dense.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.34.self_attention.query_key_value.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.34.self_attention.query_key_value.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.35.input_layernorm.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.35.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.35.mlp.dense_4h_to_h.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.35.mlp.dense_4h_to_h.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.35.mlp.dense_h_to_4h.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.35.mlp.dense_h_to_4h.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.35.post_attention_layernorm.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.35.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.35.self_attention.dense.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.35.self_attention.dense.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.35.self_attention.query_key_value.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.35.self_attention.query_key_value.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.36.input_layernorm.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.36.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.36.mlp.dense_4h_to_h.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.36.mlp.dense_4h_to_h.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.36.mlp.dense_h_to_4h.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.36.mlp.dense_h_to_4h.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.36.post_attention_layernorm.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.36.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.36.self_attention.dense.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.36.self_attention.dense.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.36.self_attention.query_key_value.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.36.self_attention.query_key_value.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.37.input_layernorm.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.37.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.37.mlp.dense_4h_to_h.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.37.mlp.dense_4h_to_h.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.37.mlp.dense_h_to_4h.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.37.mlp.dense_h_to_4h.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.37.post_attention_layernorm.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.37.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.37.self_attention.dense.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.37.self_attention.dense.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.37.self_attention.query_key_value.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.37.self_attention.query_key_value.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.38.input_layernorm.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.38.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.38.mlp.dense_4h_to_h.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.38.mlp.dense_4h_to_h.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.38.mlp.dense_h_to_4h.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.38.mlp.dense_h_to_4h.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.38.post_attention_layernorm.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.38.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.38.self_attention.dense.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.38.self_attention.dense.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.38.self_attention.query_key_value.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.38.self_attention.query_key_value.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.39.input_layernorm.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.39.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.39.mlp.dense_4h_to_h.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.39.mlp.dense_4h_to_h.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.39.mlp.dense_h_to_4h.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.39.mlp.dense_h_to_4h.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.39.post_attention_layernorm.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.39.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.39.self_attention.dense.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.39.self_attention.dense.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.39.self_attention.query_key_value.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.39.self_attention.query_key_value.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.4.input_layernorm.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.4.mlp.dense_4h_to_h.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.mlp.dense_4h_to_h.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.4.mlp.dense_h_to_4h.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.mlp.dense_h_to_4h.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.4.post_attention_layernorm.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.4.self_attention.dense.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.self_attention.dense.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.4.self_attention.query_key_value.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.self_attention.query_key_value.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.40.input_layernorm.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.40.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.40.mlp.dense_4h_to_h.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.40.mlp.dense_4h_to_h.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.40.mlp.dense_h_to_4h.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.40.mlp.dense_h_to_4h.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.40.post_attention_layernorm.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.40.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.40.self_attention.dense.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.40.self_attention.dense.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.40.self_attention.query_key_value.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.40.self_attention.query_key_value.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.41.input_layernorm.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.41.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.41.mlp.dense_4h_to_h.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.41.mlp.dense_4h_to_h.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.41.mlp.dense_h_to_4h.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.41.mlp.dense_h_to_4h.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.41.post_attention_layernorm.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.41.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.41.self_attention.dense.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.41.self_attention.dense.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.41.self_attention.query_key_value.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.41.self_attention.query_key_value.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.42.input_layernorm.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.42.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.42.mlp.dense_4h_to_h.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.42.mlp.dense_4h_to_h.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.42.mlp.dense_h_to_4h.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.42.mlp.dense_h_to_4h.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.42.post_attention_layernorm.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.42.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.42.self_attention.dense.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.42.self_attention.dense.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.42.self_attention.query_key_value.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.42.self_attention.query_key_value.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.43.input_layernorm.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.43.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.43.mlp.dense_4h_to_h.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.43.mlp.dense_4h_to_h.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.43.mlp.dense_h_to_4h.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.43.mlp.dense_h_to_4h.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.43.post_attention_layernorm.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.43.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.43.self_attention.dense.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.43.self_attention.dense.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.43.self_attention.query_key_value.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.43.self_attention.query_key_value.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.44.input_layernorm.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.44.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.44.mlp.dense_4h_to_h.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.44.mlp.dense_4h_to_h.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.44.mlp.dense_h_to_4h.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.44.mlp.dense_h_to_4h.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.44.post_attention_layernorm.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.44.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.44.self_attention.dense.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.44.self_attention.dense.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.44.self_attention.query_key_value.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.44.self_attention.query_key_value.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.45.input_layernorm.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.45.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.45.mlp.dense_4h_to_h.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.45.mlp.dense_4h_to_h.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.45.mlp.dense_h_to_4h.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.45.mlp.dense_h_to_4h.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.45.post_attention_layernorm.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.45.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.45.self_attention.dense.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.45.self_attention.dense.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.45.self_attention.query_key_value.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.45.self_attention.query_key_value.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.46.input_layernorm.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.46.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.46.mlp.dense_4h_to_h.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.46.mlp.dense_4h_to_h.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.46.mlp.dense_h_to_4h.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.46.mlp.dense_h_to_4h.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.46.post_attention_layernorm.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.46.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.46.self_attention.dense.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.46.self_attention.dense.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.46.self_attention.query_key_value.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.46.self_attention.query_key_value.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.47.input_layernorm.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.47.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.47.mlp.dense_4h_to_h.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.47.mlp.dense_4h_to_h.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.47.mlp.dense_h_to_4h.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.47.mlp.dense_h_to_4h.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.47.post_attention_layernorm.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.47.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.47.self_attention.dense.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.47.self_attention.dense.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.47.self_attention.query_key_value.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.47.self_attention.query_key_value.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.48.input_layernorm.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.48.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.48.mlp.dense_4h_to_h.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.48.mlp.dense_4h_to_h.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.48.mlp.dense_h_to_4h.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.48.mlp.dense_h_to_4h.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.48.post_attention_layernorm.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.48.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.48.self_attention.dense.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.48.self_attention.dense.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.48.self_attention.query_key_value.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.48.self_attention.query_key_value.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.49.input_layernorm.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.49.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.49.mlp.dense_4h_to_h.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.49.mlp.dense_4h_to_h.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.49.mlp.dense_h_to_4h.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.49.mlp.dense_h_to_4h.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.49.post_attention_layernorm.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.49.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.49.self_attention.dense.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.49.self_attention.dense.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.49.self_attention.query_key_value.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.49.self_attention.query_key_value.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.5.input_layernorm.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.5.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.5.mlp.dense_4h_to_h.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.5.mlp.dense_4h_to_h.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.5.mlp.dense_h_to_4h.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.5.mlp.dense_h_to_4h.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.5.post_attention_layernorm.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.5.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.5.self_attention.dense.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.5.self_attention.dense.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.5.self_attention.query_key_value.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.5.self_attention.query_key_value.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.50.input_layernorm.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.50.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.50.mlp.dense_4h_to_h.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.50.mlp.dense_4h_to_h.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.50.mlp.dense_h_to_4h.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.50.mlp.dense_h_to_4h.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.50.post_attention_layernorm.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.50.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.50.self_attention.dense.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.50.self_attention.dense.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.50.self_attention.query_key_value.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.50.self_attention.query_key_value.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.51.input_layernorm.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.51.input_layernorm.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.51.mlp.dense_4h_to_h.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.51.mlp.dense_4h_to_h.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.51.mlp.dense_h_to_4h.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.51.mlp.dense_h_to_4h.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.51.post_attention_layernorm.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.51.post_attention_layernorm.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.51.self_attention.dense.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.51.self_attention.dense.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.51.self_attention.query_key_value.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.51.self_attention.query_key_value.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.52.input_layernorm.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.52.input_layernorm.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.52.mlp.dense_4h_to_h.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.52.mlp.dense_4h_to_h.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.52.mlp.dense_h_to_4h.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.52.mlp.dense_h_to_4h.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.52.post_attention_layernorm.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.52.post_attention_layernorm.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.52.self_attention.dense.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.52.self_attention.dense.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.52.self_attention.query_key_value.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.52.self_attention.query_key_value.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.53.input_layernorm.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.53.input_layernorm.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.53.mlp.dense_4h_to_h.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.53.mlp.dense_4h_to_h.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.53.mlp.dense_h_to_4h.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.53.mlp.dense_h_to_4h.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.53.post_attention_layernorm.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.53.post_attention_layernorm.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.53.self_attention.dense.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.53.self_attention.dense.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.53.self_attention.query_key_value.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.53.self_attention.query_key_value.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.6.input_layernorm.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.6.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.6.mlp.dense_4h_to_h.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.6.mlp.dense_4h_to_h.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.6.mlp.dense_h_to_4h.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.6.mlp.dense_h_to_4h.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.6.post_attention_layernorm.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.6.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.6.self_attention.dense.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.6.self_attention.dense.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.6.self_attention.query_key_value.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.6.self_attention.query_key_value.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.7.input_layernorm.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.7.mlp.dense_4h_to_h.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.mlp.dense_4h_to_h.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.7.mlp.dense_h_to_4h.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.mlp.dense_h_to_4h.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.7.post_attention_layernorm.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.7.self_attention.dense.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.self_attention.dense.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.7.self_attention.query_key_value.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.self_attention.query_key_value.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.8.input_layernorm.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.8.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.8.mlp.dense_4h_to_h.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.8.mlp.dense_4h_to_h.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.8.mlp.dense_h_to_4h.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.8.mlp.dense_h_to_4h.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.8.post_attention_layernorm.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.8.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.8.self_attention.dense.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.8.self_attention.dense.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.8.self_attention.query_key_value.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.8.self_attention.query_key_value.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.9.input_layernorm.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.9.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.9.mlp.dense_4h_to_h.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.9.mlp.dense_4h_to_h.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.9.mlp.dense_h_to_4h.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.9.mlp.dense_h_to_4h.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.9.post_attention_layernorm.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.9.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.9.self_attention.dense.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.9.self_attention.dense.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.9.self_attention.query_key_value.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.9.self_attention.query_key_value.weight": "model-00003-of-00014.safetensors",
+    "transformer.ln_f.bias": "model-00014-of-00014.safetensors",
+    "transformer.ln_f.weight": "model-00014-of-00014.safetensors",
+    "transformer.word_embeddings.weight": "model-00001-of-00014.safetensors",
+    "transformer.word_embeddings_layernorm.bias": "model-00001-of-00014.safetensors",
+    "transformer.word_embeddings_layernorm.weight": "model-00001-of-00014.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<fim_prefix>",
+    "<fim_middle>",
+    "<fim_suffix>",
+    "<fim_pad>",
+    "<filename>",
+    "<gh_stars>",
+    "<issue_start>",
+    "<issue_comment>",
+    "<issue_closed>",
+    "<jupyter_start>",
+    "<jupyter_text>",
+    "<jupyter_code>",
+    "<jupyter_output>",
+    "<empty_output>",
+    "<commit_before>",
+    "<commit_msg>",
+    "<commit_after>",
+    "<reponame>"
+  ],
+  "bos_token": "<s>",
+  "eos_token": "<|im_end|>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,228 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "<fim_prefix>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "<fim_middle>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "<fim_suffix>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "<fim_pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<filename>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "<gh_stars>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "<issue_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "<issue_comment>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<issue_closed>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<jupyter_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<jupyter_text>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "17": {
+      "content": "<jupyter_code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "18": {
+      "content": "<jupyter_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "19": {
+      "content": "<empty_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "20": {
+      "content": "<commit_before>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "21": {
+      "content": "<commit_msg>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "22": {
+      "content": "<commit_after>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "23": {
+      "content": "<reponame>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<fim_prefix>",
+    "<fim_middle>",
+    "<fim_suffix>",
+    "<fim_pad>",
+    "<filename>",
+    "<gh_stars>",
+    "<issue_start>",
+    "<issue_comment>",
+    "<issue_closed>",
+    "<jupyter_start>",
+    "<jupyter_text>",
+    "<jupyter_code>",
+    "<jupyter_output>",
+    "<empty_output>",
+    "<commit_before>",
+    "<commit_msg>",
+    "<commit_after>",
+    "<reponame>"
+  ],
+  "bos_token": "<s>",
+  "chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "model_max_length": 2048,
+  "pad_token": "<pad>",
+  "padding_side": "left",
+  "tokenizer_class": "BloomTokenizer",
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.8629729963555334,
+    "train_runtime": 46452.2696,
+    "train_samples": 30684,
+    "train_samples_per_second": 1.982,
+    "train_steps_per_second": 0.041
+}

trainer_state.json

@@ -0,0 +1,2404 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.9976544175136826,
+  "eval_steps": 200,
+  "global_step": 1917,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.0416666666666667e-07,
+      "loss": 3.061,
+      "step": 1
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.208333333333334e-07,
+      "loss": 3.2765,
+      "step": 5
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.0416666666666667e-06,
+      "loss": 3.0777,
+      "step": 10
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5625e-06,
+      "loss": 2.7423,
+      "step": 15
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 2.0833333333333334e-06,
+      "loss": 2.6377,
+      "step": 20
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 2.604166666666667e-06,
+      "loss": 2.2237,
+      "step": 25
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 3.125e-06,
+      "loss": 1.9237,
+      "step": 30
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 3.6458333333333333e-06,
+      "loss": 1.7644,
+      "step": 35
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.166666666666667e-06,
+      "loss": 1.7165,
+      "step": 40
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.6875000000000004e-06,
+      "loss": 1.7481,
+      "step": 45
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 5.208333333333334e-06,
+      "loss": 1.6687,
+      "step": 50
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 5.729166666666667e-06,
+      "loss": 1.6279,
+      "step": 55
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 6.25e-06,
+      "loss": 1.6304,
+      "step": 60
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 6.770833333333334e-06,
+      "loss": 1.5538,
+      "step": 65
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 7.291666666666667e-06,
+      "loss": 1.63,
+      "step": 70
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 7.8125e-06,
+      "loss": 1.5741,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 1.5827,
+      "step": 80
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 8.854166666666667e-06,
+      "loss": 1.5819,
+      "step": 85
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.375000000000001e-06,
+      "loss": 1.5961,
+      "step": 90
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.895833333333334e-06,
+      "loss": 1.5914,
+      "step": 95
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.0416666666666668e-05,
+      "loss": 1.6306,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.0937500000000002e-05,
+      "loss": 1.5785,
+      "step": 105
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.1458333333333333e-05,
+      "loss": 1.5712,
+      "step": 110
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.1979166666666669e-05,
+      "loss": 1.5669,
+      "step": 115
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.25e-05,
+      "loss": 1.5288,
+      "step": 120
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.3020833333333334e-05,
+      "loss": 1.5003,
+      "step": 125
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.3541666666666668e-05,
+      "loss": 1.563,
+      "step": 130
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.4062500000000001e-05,
+      "loss": 1.5413,
+      "step": 135
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.4583333333333333e-05,
+      "loss": 1.5475,
+      "step": 140
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.5104166666666668e-05,
+      "loss": 1.501,
+      "step": 145
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.5625e-05,
+      "loss": 1.5094,
+      "step": 150
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.6145833333333334e-05,
+      "loss": 1.4577,
+      "step": 155
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 1.5209,
+      "step": 160
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.71875e-05,
+      "loss": 1.5535,
+      "step": 165
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.7708333333333335e-05,
+      "loss": 1.5203,
+      "step": 170
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.8229166666666668e-05,
+      "loss": 1.5073,
+      "step": 175
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 1.4751,
+      "step": 180
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9270833333333335e-05,
+      "loss": 1.529,
+      "step": 185
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.979166666666667e-05,
+      "loss": 1.5237,
+      "step": 190
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9999850743593963e-05,
+      "loss": 1.4995,
+      "step": 195
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9998938637247684e-05,
+      "loss": 1.4602,
+      "step": 200
+    },
+    {
+      "epoch": 0.31,
+      "eval_loss": 1.5713448524475098,
+      "eval_runtime": 158.4513,
+      "eval_samples_per_second": 10.603,
+      "eval_steps_per_second": 0.114,
+      "step": 200
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.999719742032076e-05,
+      "loss": 1.4455,
+      "step": 205
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9994627237194654e-05,
+      "loss": 1.5262,
+      "step": 210
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9991228300988586e-05,
+      "loss": 1.447,
+      "step": 215
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9987000893541863e-05,
+      "loss": 1.5364,
+      "step": 220
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9981945365390517e-05,
+      "loss": 1.5138,
+      "step": 225
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.997606213573823e-05,
+      "loss": 1.5337,
+      "step": 230
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.996935169242157e-05,
+      "loss": 1.3919,
+      "step": 235
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9961814591869558e-05,
+      "loss": 1.5198,
+      "step": 240
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.99534514590575e-05,
+      "loss": 1.5039,
+      "step": 245
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9944262987455188e-05,
+      "loss": 1.4199,
+      "step": 250
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9934249938969396e-05,
+      "loss": 1.4122,
+      "step": 255
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.992341314388069e-05,
+      "loss": 1.5318,
+      "step": 260
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9911753500774592e-05,
+      "loss": 1.4192,
+      "step": 265
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9899271976467058e-05,
+      "loss": 1.4537,
+      "step": 270
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.988596960592432e-05,
+      "loss": 1.3926,
+      "step": 275
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9871847492177066e-05,
+      "loss": 1.4431,
+      "step": 280
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9856906806228988e-05,
+      "loss": 1.459,
+      "step": 285
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.984114878695964e-05,
+      "loss": 1.3678,
+      "step": 290
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9824574741021765e-05,
+      "loss": 1.416,
+      "step": 295
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9807186042732908e-05,
+      "loss": 1.335,
+      "step": 300
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9788984133961482e-05,
+      "loss": 1.3805,
+      "step": 305
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.976997052400719e-05,
+      "loss": 1.3934,
+      "step": 310
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9750146789475885e-05,
+      "loss": 1.3793,
+      "step": 315
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9729514574148842e-05,
+      "loss": 1.4001,
+      "step": 320
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.970807558884644e-05,
+      "loss": 1.3879,
+      "step": 325
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.9685831611286312e-05,
+      "loss": 1.3787,
+      "step": 330
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.9662784485935945e-05,
+      "loss": 1.3915,
+      "step": 335
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.9638936123859715e-05,
+      "loss": 1.3017,
+      "step": 340
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.961428850256044e-05,
+      "loss": 1.3724,
+      "step": 345
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.9588843665815404e-05,
+      "loss": 1.4637,
+      "step": 350
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.9562603723506873e-05,
+      "loss": 1.3966,
+      "step": 355
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.9535570851447166e-05,
+      "loss": 1.3669,
+      "step": 360
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.9507747291198213e-05,
+      "loss": 1.32,
+      "step": 365
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.9479135349885706e-05,
+      "loss": 1.3099,
+      "step": 370
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.944973740000778e-05,
+      "loss": 1.3427,
+      "step": 375
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9419555879238295e-05,
+      "loss": 1.3112,
+      "step": 380
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.9388593290224686e-05,
+      "loss": 1.3634,
+      "step": 385
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.9356852200380466e-05,
+      "loss": 1.3982,
+      "step": 390
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.9324335241672327e-05,
+      "loss": 1.3739,
+      "step": 395
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.9291045110401897e-05,
+      "loss": 1.2899,
+      "step": 400
+    },
+    {
+      "epoch": 0.63,
+      "eval_loss": 1.5519251823425293,
+      "eval_runtime": 157.2629,
+      "eval_samples_per_second": 10.683,
+      "eval_steps_per_second": 0.114,
+      "step": 400
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.925698456698216e-05,
+      "loss": 1.3417,
+      "step": 405
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.922215643570857e-05,
+      "loss": 1.3533,
+      "step": 410
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.918656360452486e-05,
+      "loss": 1.3617,
+      "step": 415
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.9150209024783564e-05,
+      "loss": 1.3763,
+      "step": 420
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.9113095711001307e-05,
+      "loss": 1.223,
+      "step": 425
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.9075226740608825e-05,
+      "loss": 1.3439,
+      "step": 430
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.9036605253695804e-05,
+      "loss": 1.2973,
+      "step": 435
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.899723445275049e-05,
+      "loss": 1.2989,
+      "step": 440
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.895711760239413e-05,
+      "loss": 1.375,
+      "step": 445
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.8916258029110305e-05,
+      "loss": 1.3077,
+      "step": 450
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.887465912096906e-05,
+      "loss": 1.2874,
+      "step": 455
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.883232432734599e-05,
+      "loss": 1.2875,
+      "step": 460
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.8789257158636202e-05,
+      "loss": 1.2862,
+      "step": 465
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.8745461185963253e-05,
+      "loss": 1.3208,
+      "step": 470
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.8700940040883016e-05,
+      "loss": 1.2789,
+      "step": 475
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.8655697415082556e-05,
+      "loss": 1.3243,
+      "step": 480
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.8609737060074016e-05,
+      "loss": 1.2808,
+      "step": 485
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8563062786883553e-05,
+      "loss": 1.3107,
+      "step": 490
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.851567846573531e-05,
+      "loss": 1.3436,
+      "step": 495
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8467588025730506e-05,
+      "loss": 1.2388,
+      "step": 500
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.8418795454521644e-05,
+      "loss": 1.2619,
+      "step": 505
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8369304797981843e-05,
+      "loss": 1.3432,
+      "step": 510
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8319120159869354e-05,
+      "loss": 1.3219,
+      "step": 515
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.826824570148729e-05,
+      "loss": 1.2347,
+      "step": 520
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.8216685641338562e-05,
+      "loss": 1.2329,
+      "step": 525
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8164444254776078e-05,
+      "loss": 1.2509,
+      "step": 530
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8111525873648236e-05,
+      "loss": 1.2106,
+      "step": 535
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8057934885939734e-05,
+      "loss": 1.313,
+      "step": 540
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.800367573540771e-05,
+      "loss": 1.2888,
+      "step": 545
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.7948752921213266e-05,
+      "loss": 1.3146,
+      "step": 550
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.789317099754841e-05,
+      "loss": 1.2901,
+      "step": 555
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.78369345732584e-05,
+      "loss": 1.1714,
+      "step": 560
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.7780048311459606e-05,
+      "loss": 1.2368,
+      "step": 565
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.7722516929152828e-05,
+      "loss": 1.2104,
+      "step": 570
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.7664345196832164e-05,
+      "loss": 1.199,
+      "step": 575
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.760553793808945e-05,
+      "loss": 1.2896,
+      "step": 580
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.7546100029214286e-05,
+      "loss": 1.2615,
+      "step": 585
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.748603639878968e-05,
+      "loss": 1.2779,
+      "step": 590
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.74253520272834e-05,
+      "loss": 1.1917,
+      "step": 595
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.7364051946634953e-05,
+      "loss": 1.1959,
+      "step": 600
+    },
+    {
+      "epoch": 0.94,
+      "eval_loss": 1.5434802770614624,
+      "eval_runtime": 157.2464,
+      "eval_samples_per_second": 10.684,
+      "eval_steps_per_second": 0.114,
+      "step": 600
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.7302141239838382e-05,
+      "loss": 1.2505,
+      "step": 605
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.7239625040520764e-05,
+      "loss": 1.1324,
+      "step": 610
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.717650853251651e-05,
+      "loss": 1.2398,
+      "step": 615
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.711279694943757e-05,
+      "loss": 1.2225,
+      "step": 620
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.704849557423941e-05,
+      "loss": 1.2447,
+      "step": 625
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.6983609738782993e-05,
+      "loss": 1.2089,
+      "step": 630
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.6918144823392634e-05,
+      "loss": 1.2168,
+      "step": 635
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.685210625640988e-05,
+      "loss": 1.2238,
+      "step": 640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6785499513743383e-05,
+      "loss": 1.1675,
+      "step": 645
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.671833011841485e-05,
+      "loss": 1.1623,
+      "step": 650
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6650603640101058e-05,
+      "loss": 1.1398,
+      "step": 655
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.6582325694672032e-05,
+      "loss": 1.0914,
+      "step": 660
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.6513501943725366e-05,
+      "loss": 1.1294,
+      "step": 665
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.6444138094116783e-05,
+      "loss": 1.1254,
+      "step": 670
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.63742398974869e-05,
+      "loss": 1.1714,
+      "step": 675
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.6303813149784313e-05,
+      "loss": 1.0935,
+      "step": 680
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.623286369078501e-05,
+      "loss": 1.1127,
+      "step": 685
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.616139740360811e-05,
+      "loss": 1.0169,
+      "step": 690
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.6089420214228063e-05,
+      "loss": 1.0674,
+      "step": 695
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.6016938090983254e-05,
+      "loss": 0.9835,
+      "step": 700
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.5943957044081117e-05,
+      "loss": 0.9599,
+      "step": 705
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.5870483125099753e-05,
+      "loss": 0.9643,
+      "step": 710
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.579652242648616e-05,
+      "loss": 0.9597,
+      "step": 715
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.5722081081051032e-05,
+      "loss": 0.9906,
+      "step": 720
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.564716526146022e-05,
+      "loss": 0.969,
+      "step": 725
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.5571781179722912e-05,
+      "loss": 0.9246,
+      "step": 730
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.5495935086676533e-05,
+      "loss": 0.9187,
+      "step": 735
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.541963327146841e-05,
+      "loss": 0.9398,
+      "step": 740
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.534288206103429e-05,
+      "loss": 0.8638,
+      "step": 745
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.526568781957371e-05,
+      "loss": 0.9261,
+      "step": 750
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.5188056948022286e-05,
+      "loss": 0.8823,
+      "step": 755
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.510999588352092e-05,
+      "loss": 0.8369,
+      "step": 760
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.503151109888207e-05,
+      "loss": 0.8108,
+      "step": 765
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.4952609102052997e-05,
+      "loss": 0.8346,
+      "step": 770
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.4873296435576145e-05,
+      "loss": 0.8182,
+      "step": 775
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.479357967604663e-05,
+      "loss": 0.8425,
+      "step": 780
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.4713465433566892e-05,
+      "loss": 0.8076,
+      "step": 785
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.463296035119862e-05,
+      "loss": 0.7666,
+      "step": 790
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.4552071104411874e-05,
+      "loss": 0.7442,
+      "step": 795
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.4470804400531589e-05,
+      "loss": 0.7682,
+      "step": 800
+    },
+    {
+      "epoch": 1.25,
+      "eval_loss": 1.6022878885269165,
+      "eval_runtime": 157.2873,
+      "eval_samples_per_second": 10.681,
+      "eval_steps_per_second": 0.114,
+      "step": 800
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.4389166978181374e-05,
+      "loss": 0.7653,
+      "step": 805
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.4307165606724777e-05,
+      "loss": 0.7913,
+      "step": 810
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.4224807085703943e-05,
+      "loss": 0.7481,
+      "step": 815
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.4142098244275809e-05,
+      "loss": 0.787,
+      "step": 820
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.4059045940645834e-05,
+      "loss": 0.7183,
+      "step": 825
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.3975657061499316e-05,
+      "loss": 0.7405,
+      "step": 830
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.3891938521430349e-05,
+      "loss": 0.7065,
+      "step": 835
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.3807897262368453e-05,
+      "loss": 0.6948,
+      "step": 840
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.372354025300297e-05,
+      "loss": 0.6679,
+      "step": 845
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.3638874488205213e-05,
+      "loss": 0.726,
+      "step": 850
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.3553906988448451e-05,
+      "loss": 0.738,
+      "step": 855
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.346864479922577e-05,
+      "loss": 0.7827,
+      "step": 860
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.338309499046587e-05,
+      "loss": 0.7376,
+      "step": 865
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.3297264655946816e-05,
+      "loss": 0.7404,
+      "step": 870
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.3211160912707837e-05,
+      "loss": 0.6552,
+      "step": 875
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.3124790900459169e-05,
+      "loss": 0.733,
+      "step": 880
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.3038161780990036e-05,
+      "loss": 0.7305,
+      "step": 885
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.2951280737574799e-05,
+      "loss": 0.6972,
+      "step": 890
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.286415497437732e-05,
+      "loss": 0.6935,
+      "step": 895
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.2776791715853585e-05,
+      "loss": 0.7121,
+      "step": 900
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.2689198206152657e-05,
+      "loss": 0.7106,
+      "step": 905
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.2601381708516007e-05,
+      "loss": 0.6916,
+      "step": 910
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.251334950467522e-05,
+      "loss": 0.6595,
+      "step": 915
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.2425108894248216e-05,
+      "loss": 0.7074,
+      "step": 920
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.2336667194133956e-05,
+      "loss": 0.751,
+      "step": 925
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.2248031737905732e-05,
+      "loss": 0.6674,
+      "step": 930
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.215920987520307e-05,
+      "loss": 0.6802,
+      "step": 935
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.2070208971122282e-05,
+      "loss": 0.6438,
+      "step": 940
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.1981036405605782e-05,
+      "loss": 0.6899,
+      "step": 945
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.189169957283011e-05,
+      "loss": 0.6966,
+      "step": 950
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.1802205880592839e-05,
+      "loss": 0.6552,
+      "step": 955
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.171256274969829e-05,
+      "loss": 0.7066,
+      "step": 960
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.1622777613342236e-05,
+      "loss": 0.6635,
+      "step": 965
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.1532857916495504e-05,
+      "loss": 0.6861,
+      "step": 970
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.144281111528667e-05,
+      "loss": 0.6957,
+      "step": 975
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.1352644676383782e-05,
+      "loss": 0.6338,
+      "step": 980
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.1262366076375224e-05,
+      "loss": 0.6755,
+      "step": 985
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.1171982801149774e-05,
+      "loss": 0.7302,
+      "step": 990
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.1081502345275857e-05,
+      "loss": 0.6579,
+      "step": 995
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.0990932211380099e-05,
+      "loss": 0.6793,
+      "step": 1000
+    },
+    {
+      "epoch": 1.56,
+      "eval_loss": 1.6625279188156128,
+      "eval_runtime": 157.3283,
+      "eval_samples_per_second": 10.678,
+      "eval_steps_per_second": 0.114,
+      "step": 1000
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.0900279909525226e-05,
+      "loss": 0.6442,
+      "step": 1005
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.080955295658731e-05,
+      "loss": 0.6029,
+      "step": 1010
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.0718758875632486e-05,
+      "loss": 0.6658,
+      "step": 1015
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.0627905195293135e-05,
+      "loss": 0.6443,
+      "step": 1020
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.0536999449143606e-05,
+      "loss": 0.6974,
+      "step": 1025
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.0446049175075544e-05,
+      "loss": 0.701,
+      "step": 1030
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.0355061914672831e-05,
+      "loss": 0.6889,
+      "step": 1035
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.0264045212586266e-05,
+      "loss": 0.6602,
+      "step": 1040
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.0173006615907936e-05,
+      "loss": 0.6654,
+      "step": 1045
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.0081953673545432e-05,
+      "loss": 0.6857,
+      "step": 1050
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.990893935595885e-06,
+      "loss": 0.6869,
+      "step": 1055
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 9.899834952719914e-06,
+      "loss": 0.7023,
+      "step": 1060
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 9.808784275515526e-06,
+      "loss": 0.5785,
+      "step": 1065
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 9.717749453892026e-06,
+      "loss": 0.6653,
+      "step": 1070
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 9.626738036443978e-06,
+      "loss": 0.6585,
+      "step": 1075
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 9.535757569825266e-06,
+      "loss": 0.6425,
+      "step": 1080
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.444815598123346e-06,
+      "loss": 0.6873,
+      "step": 1085
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.353919662233676e-06,
+      "loss": 0.6441,
+      "step": 1090
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.263077299234433e-06,
+      "loss": 0.6318,
+      "step": 1095
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.172296041761534e-06,
+      "loss": 0.618,
+      "step": 1100
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 9.08158341738404e-06,
+      "loss": 0.6278,
+      "step": 1105
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 8.990946947979955e-06,
+      "loss": 0.6535,
+      "step": 1110
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 8.900394149112543e-06,
+      "loss": 0.6229,
+      "step": 1115
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 8.80993252940709e-06,
+      "loss": 0.6632,
+      "step": 1120
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 8.719569589928353e-06,
+      "loss": 0.6442,
+      "step": 1125
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 8.629312823558505e-06,
+      "loss": 0.654,
+      "step": 1130
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 8.539169714375885e-06,
+      "loss": 0.6886,
+      "step": 1135
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 8.44914773703438e-06,
+      "loss": 0.6309,
+      "step": 1140
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 8.35925435614365e-06,
+      "loss": 0.6128,
+      "step": 1145
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 8.269497025650146e-06,
+      "loss": 0.6741,
+      "step": 1150
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 8.179883188219052e-06,
+      "loss": 0.6782,
+      "step": 1155
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 8.090420274617115e-06,
+      "loss": 0.6333,
+      "step": 1160
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 8.001115703096508e-06,
+      "loss": 0.5792,
+      "step": 1165
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.911976878779696e-06,
+      "loss": 0.62,
+      "step": 1170
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 7.823011193045413e-06,
+      "loss": 0.6014,
+      "step": 1175
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 7.73422602291576e-06,
+      "loss": 0.6273,
+      "step": 1180
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 7.645628730444524e-06,
+      "loss": 0.6791,
+      "step": 1185
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 7.557226662106679e-06,
+      "loss": 0.6503,
+      "step": 1190
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 7.469027148189268e-06,
+      "loss": 0.6116,
+      "step": 1195
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 7.3810375021835275e-06,
+      "loss": 0.5961,
+      "step": 1200
+    },
+    {
+      "epoch": 1.88,
+      "eval_loss": 1.6919642686843872,
+      "eval_runtime": 157.2369,
+      "eval_samples_per_second": 10.685,
+      "eval_steps_per_second": 0.114,
+      "step": 1200
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 7.293265020178495e-06,
+      "loss": 0.6073,
+      "step": 1205
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 7.205716980255985e-06,
+      "loss": 0.5941,
+      "step": 1210
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 7.118400641887116e-06,
+      "loss": 0.593,
+      "step": 1215
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 7.031323245330337e-06,
+      "loss": 0.6307,
+      "step": 1220
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.944492011031086e-06,
+      "loss": 0.6334,
+      "step": 1225
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.857914139023058e-06,
+      "loss": 0.6494,
+      "step": 1230
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 6.771596808331175e-06,
+      "loss": 0.5883,
+      "step": 1235
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 6.685547176376321e-06,
+      "loss": 0.6022,
+      "step": 1240
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 6.59977237838183e-06,
+      "loss": 0.6323,
+      "step": 1245
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 6.5142795267818505e-06,
+      "loss": 0.5627,
+      "step": 1250
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 6.4290757106315674e-06,
+      "loss": 0.5976,
+      "step": 1255
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 6.344167995019395e-06,
+      "loss": 0.6036,
+      "step": 1260
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 6.2595634204811215e-06,
+      "loss": 0.6299,
+      "step": 1265
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 6.17526900241614e-06,
+      "loss": 0.6476,
+      "step": 1270
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 6.091291730505684e-06,
+      "loss": 0.6248,
+      "step": 1275
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 6.007638568133303e-06,
+      "loss": 0.5561,
+      "step": 1280
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.924316451807407e-06,
+      "loss": 0.6042,
+      "step": 1285
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 5.841332290586126e-06,
+      "loss": 0.58,
+      "step": 1290
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 5.758692965504391e-06,
+      "loss": 0.5884,
+      "step": 1295
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 5.676405329003387e-06,
+      "loss": 0.5633,
+      "step": 1300
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 5.594476204362303e-06,
+      "loss": 0.5597,
+      "step": 1305
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 5.512912385132605e-06,
+      "loss": 0.5794,
+      "step": 1310
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 5.431720634574663e-06,
+      "loss": 0.5783,
+      "step": 1315
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 5.350907685096983e-06,
+      "loss": 0.5751,
+      "step": 1320
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 5.270480237697927e-06,
+      "loss": 0.5635,
+      "step": 1325
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 5.190444961410082e-06,
+      "loss": 0.4816,
+      "step": 1330
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 5.110808492747258e-06,
+      "loss": 0.5327,
+      "step": 1335
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 5.031577435154207e-06,
+      "loss": 0.5105,
+      "step": 1340
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 4.952758358459037e-06,
+      "loss": 0.4521,
+      "step": 1345
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 4.874357798328464e-06,
+      "loss": 0.4812,
+      "step": 1350
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.796382255725866e-06,
+      "loss": 0.4729,
+      "step": 1355
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 4.7188381963722385e-06,
+      "loss": 0.4983,
+      "step": 1360
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 4.641732050210032e-06,
+      "loss": 0.4844,
+      "step": 1365
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 4.565070210870015e-06,
+      "loss": 0.4523,
+      "step": 1370
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 4.488859035141071e-06,
+      "loss": 0.4526,
+      "step": 1375
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 4.413104842443149e-06,
+      "loss": 0.4514,
+      "step": 1380
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 4.3378139143032095e-06,
+      "loss": 0.4341,
+      "step": 1385
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 4.262992493834386e-06,
+      "loss": 0.4427,
+      "step": 1390
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 4.188646785218316e-06,
+      "loss": 0.4423,
+      "step": 1395
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 4.114782953190648e-06,
+      "loss": 0.4087,
+      "step": 1400
+    },
+    {
+      "epoch": 2.19,
+      "eval_loss": 1.7767020463943481,
+      "eval_runtime": 157.2878,
+      "eval_samples_per_second": 10.681,
+      "eval_steps_per_second": 0.114,
+      "step": 1400
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 4.0414071225299235e-06,
+      "loss": 0.4089,
+      "step": 1405
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 3.968525377549657e-06,
+      "loss": 0.3964,
+      "step": 1410
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 3.896143761593862e-06,
+      "loss": 0.3975,
+      "step": 1415
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 3.824268276535917e-06,
+      "loss": 0.4327,
+      "step": 1420
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 3.752904882280899e-06,
+      "loss": 0.3902,
+      "step": 1425
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 3.682059496271383e-06,
+      "loss": 0.3924,
+      "step": 1430
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 3.6117379929967857e-06,
+      "loss": 0.3883,
+      "step": 1435
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 3.5419462035062313e-06,
+      "loss": 0.3864,
+      "step": 1440
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 3.472689914925056e-06,
+      "loss": 0.3894,
+      "step": 1445
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 3.4039748699749287e-06,
+      "loss": 0.3984,
+      "step": 1450
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 3.3358067664976866e-06,
+      "loss": 0.399,
+      "step": 1455
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 3.2681912569828445e-06,
+      "loss": 0.393,
+      "step": 1460
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 3.2011339480989136e-06,
+      "loss": 0.3652,
+      "step": 1465
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 3.134640400228479e-06,
+      "loss": 0.3633,
+      "step": 1470
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 3.0687161270071465e-06,
+      "loss": 0.3651,
+      "step": 1475
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 3.003366594866345e-06,
+      "loss": 0.3589,
+      "step": 1480
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 2.938597222580063e-06,
+      "loss": 0.3561,
+      "step": 1485
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 2.874413380815506e-06,
+      "loss": 0.3556,
+      "step": 1490
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 2.8108203916877753e-06,
+      "loss": 0.3639,
+      "step": 1495
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 2.7478235283185504e-06,
+      "loss": 0.4166,
+      "step": 1500
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 2.685428014398859e-06,
+      "loss": 0.3924,
+      "step": 1505
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 2.623639023755902e-06,
+      "loss": 0.3776,
+      "step": 1510
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 2.562461679924072e-06,
+      "loss": 0.3471,
+      "step": 1515
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 2.501901055720074e-06,
+      "loss": 0.3657,
+      "step": 1520
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 2.4419621728223297e-06,
+      "loss": 0.3692,
+      "step": 1525
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 2.382650001354543e-06,
+      "loss": 0.364,
+      "step": 1530
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 2.3239694594736074e-06,
+      "loss": 0.3574,
+      "step": 1535
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 2.2659254129617712e-06,
+      "loss": 0.3645,
+      "step": 1540
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 2.2085226748231792e-06,
+      "loss": 0.3739,
+      "step": 1545
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 2.1517660048847765e-06,
+      "loss": 0.345,
+      "step": 1550
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 2.0956601094016217e-06,
+      "loss": 0.3299,
+      "step": 1555
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 2.0402096406666562e-06,
+      "loss": 0.386,
+      "step": 1560
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.9854191966249183e-06,
+      "loss": 0.3803,
+      "step": 1565
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 1.9312933204922947e-06,
+      "loss": 0.3339,
+      "step": 1570
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 1.8778365003787934e-06,
+      "loss": 0.3577,
+      "step": 1575
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 1.8250531689163997e-06,
+      "loss": 0.3306,
+      "step": 1580
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.7729477028915064e-06,
+      "loss": 0.3393,
+      "step": 1585
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 1.7215244228820027e-06,
+      "loss": 0.3596,
+      "step": 1590
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 1.6707875928990059e-06,
+      "loss": 0.3422,
+      "step": 1595
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 1.6207414200333027e-06,
+      "loss": 0.3497,
+      "step": 1600
+    },
+    {
+      "epoch": 2.5,
+      "eval_loss": 1.8313430547714233,
+      "eval_runtime": 157.2838,
+      "eval_samples_per_second": 10.681,
+      "eval_steps_per_second": 0.114,
+      "step": 1600
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.571390054106482e-06,
+      "loss": 0.3431,
+      "step": 1605
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 1.5227375873268445e-06,
+      "loss": 0.3821,
+      "step": 1610
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 1.4747880539500637e-06,
+      "loss": 0.346,
+      "step": 1615
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 1.4275454299446834e-06,
+      "loss": 0.3205,
+      "step": 1620
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 1.381013632662418e-06,
+      "loss": 0.357,
+      "step": 1625
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.335196520513332e-06,
+      "loss": 0.3662,
+      "step": 1630
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 1.2900978926458985e-06,
+      "loss": 0.3426,
+      "step": 1635
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 1.2457214886319757e-06,
+      "loss": 0.37,
+      "step": 1640
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 1.2020709881567184e-06,
+      "loss": 0.3317,
+      "step": 1645
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.15915001071347e-06,
+      "loss": 0.317,
+      "step": 1650
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 1.1169621153036204e-06,
+      "loss": 0.331,
+      "step": 1655
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 1.0755108001415038e-06,
+      "loss": 0.3533,
+      "step": 1660
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 1.0347995023643198e-06,
+      "loss": 0.3694,
+      "step": 1665
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 9.948315977471412e-07,
+      "loss": 0.3615,
+      "step": 1670
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 9.556104004229717e-07,
+      "loss": 0.3767,
+      "step": 1675
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 9.171391626079629e-07,
+      "loss": 0.3812,
+      "step": 1680
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 8.794210743317233e-07,
+      "loss": 0.3459,
+      "step": 1685
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 8.42459263172809e-07,
+      "loss": 0.3811,
+      "step": 1690
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 8.062567939993838e-07,
+      "loss": 0.3827,
+      "step": 1695
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 7.708166687150842e-07,
+      "loss": 0.3673,
+      "step": 1700
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 7.361418260100984e-07,
+      "loss": 0.3082,
+      "step": 1705
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 7.022351411174866e-07,
+      "loss": 0.366,
+      "step": 1710
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 6.690994255747774e-07,
+      "loss": 0.3622,
+      "step": 1715
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 6.367374269908233e-07,
+      "loss": 0.3569,
+      "step": 1720
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 6.051518288179847e-07,
+      "loss": 0.3445,
+      "step": 1725
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 5.74345250129601e-07,
+      "loss": 0.375,
+      "step": 1730
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 5.443202454028306e-07,
+      "loss": 0.3433,
+      "step": 1735
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 5.150793043068269e-07,
+      "loss": 0.3107,
+      "step": 1740
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 4.866248514963001e-07,
+      "loss": 0.3458,
+      "step": 1745
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 4.589592464104609e-07,
+      "loss": 0.3468,
+      "step": 1750
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 4.3208478307738e-07,
+      "loss": 0.3494,
+      "step": 1755
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 4.0600368992376074e-07,
+      "loss": 0.3597,
+      "step": 1760
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 3.8071812959016965e-07,
+      "loss": 0.364,
+      "step": 1765
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 3.5623019875169916e-07,
+      "loss": 0.3462,
+      "step": 1770
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 3.325419279441211e-07,
+      "loss": 0.3932,
+      "step": 1775
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 3.096552813955045e-07,
+      "loss": 0.3658,
+      "step": 1780
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 2.875721568633527e-07,
+      "loss": 0.3525,
+      "step": 1785
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 2.6629438547723484e-07,
+      "loss": 0.3846,
+      "step": 1790
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 2.458237315869527e-07,
+      "loss": 0.3767,
+      "step": 1795
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 2.2616189261623568e-07,
+      "loss": 0.3415,
+      "step": 1800
+    },
+    {
+      "epoch": 2.81,
+      "eval_loss": 1.8276134729385376,
+      "eval_runtime": 157.3051,
+      "eval_samples_per_second": 10.68,
+      "eval_steps_per_second": 0.114,
+      "step": 1800
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 2.0731049892199873e-07,
+      "loss": 0.3214,
+      "step": 1805
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 1.8927111365914297e-07,
+      "loss": 0.3368,
+      "step": 1810
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.7204523265095252e-07,
+      "loss": 0.354,
+      "step": 1815
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 1.55634284265046e-07,
+      "loss": 0.3343,
+      "step": 1820
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 1.400396292949513e-07,
+      "loss": 0.3874,
+      "step": 1825
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 1.2526256084725351e-07,
+      "loss": 0.3613,
+      "step": 1830
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 1.1130430423438642e-07,
+      "loss": 0.328,
+      "step": 1835
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 9.816601687301585e-08,
+      "loss": 0.3326,
+      "step": 1840
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 8.584878818807652e-08,
+      "loss": 0.3314,
+      "step": 1845
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 7.435363952243313e-08,
+      "loss": 0.3516,
+      "step": 1850
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 6.368152405219042e-08,
+      "loss": 0.3337,
+      "step": 1855
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 5.3833326707656284e-08,
+      "loss": 0.3457,
+      "step": 1860
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 4.4809864099965014e-08,
+      "loss": 0.3555,
+      "step": 1865
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 3.661188445336361e-08,
+      "loss": 0.3706,
+      "step": 1870
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 2.9240067543165883e-08,
+      "loss": 0.3241,
+      "step": 1875
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 2.2695024639390884e-08,
+      "loss": 0.3393,
+      "step": 1880
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 1.697729845607121e-08,
+      "loss": 0.3962,
+      "step": 1885
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 1.20873631062568e-08,
+      "loss": 0.3253,
+      "step": 1890
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 8.025624062695247e-09,
+      "loss": 0.3559,
+      "step": 1895
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 4.792418124214271e-09,
+      "loss": 0.3506,
+      "step": 1900
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 2.3880133877962617e-09,
+      "loss": 0.3776,
+      "step": 1905
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 8.126092263405217e-10,
+      "loss": 0.3922,
+      "step": 1910
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 6.633627213870064e-11,
+      "loss": 0.3804,
+      "step": 1915
+    },
+    {
+      "epoch": 3.0,
+      "step": 1917,
+      "total_flos": 889554929385472.0,
+      "train_loss": 0.8629729963555334,
+      "train_runtime": 46452.2696,
+      "train_samples_per_second": 1.982,
+      "train_steps_per_second": 0.041
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1917,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "total_flos": 889554929385472.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d60de374e019fe6a750376ddfd64309edcd4820af015671f6d8a9b32cc7c1f6c
+size 5752