Initial upload

README.md

@@ -0,0 +1,34 @@
+---
+license: mit
+language:
+- en
+pipeline_tag: text-generation
+arxiv:
+- https://arxiv.org/abs/2508.06595
+library_name: transformers
+---
+## Model Details
+
+Best [Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) checkpoint unlearned using [RR](https://arxiv.org/abs/2406.04313) with the Keyword-Bio forget set. For more details, please check [our paper](https://arxiv.org/abs/2508.06595). 
+
+### sources
+- Base model: [Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3)
+- Repository: [https://github.com/xyzhu123/Synthetic_Textbook)
+### Performance
+|                                             |  WMDP-Bio |  tinyMMLU  |  GSM8k  | TriviaQA |
+|---------------------------------------------|:---------:|:----------:|:-------:|:--------:| 
+| Mistral-7B-Instruct-v0.3                    |   67.48  |   64.20    |  50.19  |  56.81   |
+| Mistral-7B-Instruct-v0.3_RR_Keyword-Bio   |   49.18  |   63.95   |  47.38  |  57.48  |
+## Citation
+If you find this useful in your research, please consider citing our paper:
+```
+@misc{zhu2025llmunlearningexpertcurated,
+      title={LLM Unlearning Without an Expert Curated Dataset}, 
+      author={Xiaoyuan Zhu and Muru Zhang and Ollie Liu and Robin Jia and Willie Neiswanger},
+      year={2025},
+      eprint={2508.06595},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL},
+      url={https://arxiv.org/abs/2508.06595}, 
+}
+```

checkpoint-200/adapter_config.json

@@ -0,0 +1,59 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mistral-7B-Instruct-v0.3",
+  "bias": "none",
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": [
+    0,
+    1,
+    2,
+    3,
+    4,
+    5,
+    6,
+    7,
+    8,
+    9,
+    10,
+    11,
+    12,
+    13,
+    14,
+    15,
+    16,
+    17,
+    18,
+    19,
+    20
+  ],
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj",
+    "gate_proj",
+    "up_proj",
+    "k_proj",
+    "o_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-200/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ce27d54d6a3e768ea101a32b0b95f0584cf70d6f8dbeb1d34e909e7c9bae34b
+size 55089616

checkpoint-200/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-200/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37f00374dea48658ee8f5d0f21895b9bc55cb0103939607c8185bfd1c6ca1f89
+size 587404

checkpoint-200/trainer_state.json

@@ -0,0 +1,1433 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.16,
+  "eval_steps": 1000,
+  "global_step": 200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0008,
+      "grad_norm": 0.0,
+      "learning_rate": 5e-05,
+      "loss": 9.9524,
+      "step": 1
+    },
+    {
+      "epoch": 0.0016,
+      "grad_norm": 1.0841511539183557e-05,
+      "learning_rate": 5e-05,
+      "loss": 9.9048,
+      "step": 2
+    },
+    {
+      "epoch": 0.0024,
+      "grad_norm": 38.986961364746094,
+      "learning_rate": 5e-05,
+      "loss": 9.8427,
+      "step": 3
+    },
+    {
+      "epoch": 0.0032,
+      "grad_norm": 30.884456634521484,
+      "learning_rate": 5e-05,
+      "loss": 9.7653,
+      "step": 4
+    },
+    {
+      "epoch": 0.004,
+      "grad_norm": 153.22972106933594,
+      "learning_rate": 5e-05,
+      "loss": 9.5434,
+      "step": 5
+    },
+    {
+      "epoch": 0.0048,
+      "grad_norm": 173.10813903808594,
+      "learning_rate": 5e-05,
+      "loss": 9.1256,
+      "step": 6
+    },
+    {
+      "epoch": 0.0056,
+      "grad_norm": 270.6604919433594,
+      "learning_rate": 5e-05,
+      "loss": 8.1729,
+      "step": 7
+    },
+    {
+      "epoch": 0.0064,
+      "grad_norm": 244.15834045410156,
+      "learning_rate": 5e-05,
+      "loss": 7.3432,
+      "step": 8
+    },
+    {
+      "epoch": 0.0072,
+      "grad_norm": 435.95330810546875,
+      "learning_rate": 5e-05,
+      "loss": 6.7985,
+      "step": 9
+    },
+    {
+      "epoch": 0.008,
+      "grad_norm": 210.4911346435547,
+      "learning_rate": 5e-05,
+      "loss": 6.3699,
+      "step": 10
+    },
+    {
+      "epoch": 0.0088,
+      "grad_norm": 150.57620239257812,
+      "learning_rate": 5e-05,
+      "loss": 4.8502,
+      "step": 11
+    },
+    {
+      "epoch": 0.0096,
+      "grad_norm": 227.07766723632812,
+      "learning_rate": 5e-05,
+      "loss": 5.0107,
+      "step": 12
+    },
+    {
+      "epoch": 0.0104,
+      "grad_norm": 151.02149963378906,
+      "learning_rate": 5e-05,
+      "loss": 2.7742,
+      "step": 13
+    },
+    {
+      "epoch": 0.0112,
+      "grad_norm": 34.4033317565918,
+      "learning_rate": 5e-05,
+      "loss": 2.5491,
+      "step": 14
+    },
+    {
+      "epoch": 0.012,
+      "grad_norm": 23.278841018676758,
+      "learning_rate": 5e-05,
+      "loss": 2.3394,
+      "step": 15
+    },
+    {
+      "epoch": 0.0128,
+      "grad_norm": 5.93737268447876,
+      "learning_rate": 5e-05,
+      "loss": 2.1644,
+      "step": 16
+    },
+    {
+      "epoch": 0.0136,
+      "grad_norm": 63.415889739990234,
+      "learning_rate": 5e-05,
+      "loss": 2.5026,
+      "step": 17
+    },
+    {
+      "epoch": 0.0144,
+      "grad_norm": 63.13791275024414,
+      "learning_rate": 5e-05,
+      "loss": 2.9117,
+      "step": 18
+    },
+    {
+      "epoch": 0.0152,
+      "grad_norm": 91.84156036376953,
+      "learning_rate": 5e-05,
+      "loss": 2.9422,
+      "step": 19
+    },
+    {
+      "epoch": 0.016,
+      "grad_norm": 84.56642150878906,
+      "learning_rate": 5e-05,
+      "loss": 2.884,
+      "step": 20
+    },
+    {
+      "epoch": 0.0168,
+      "grad_norm": 100.07738494873047,
+      "learning_rate": 5e-05,
+      "loss": 2.8033,
+      "step": 21
+    },
+    {
+      "epoch": 0.0176,
+      "grad_norm": 149.86077880859375,
+      "learning_rate": 5e-05,
+      "loss": 4.9988,
+      "step": 22
+    },
+    {
+      "epoch": 0.0184,
+      "grad_norm": 65.63778686523438,
+      "learning_rate": 5e-05,
+      "loss": 2.8649,
+      "step": 23
+    },
+    {
+      "epoch": 0.0192,
+      "grad_norm": 93.3624267578125,
+      "learning_rate": 5e-05,
+      "loss": 2.9432,
+      "step": 24
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 124.06131744384766,
+      "learning_rate": 5e-05,
+      "loss": 3.6676,
+      "step": 25
+    },
+    {
+      "epoch": 0.0208,
+      "grad_norm": 83.38105773925781,
+      "learning_rate": 5e-05,
+      "loss": 2.5666,
+      "step": 26
+    },
+    {
+      "epoch": 0.0216,
+      "grad_norm": 130.11135864257812,
+      "learning_rate": 5e-05,
+      "loss": 3.0716,
+      "step": 27
+    },
+    {
+      "epoch": 0.0224,
+      "grad_norm": 94.01095581054688,
+      "learning_rate": 5e-05,
+      "loss": 2.5299,
+      "step": 28
+    },
+    {
+      "epoch": 0.0232,
+      "grad_norm": 111.0047836303711,
+      "learning_rate": 5e-05,
+      "loss": 3.3937,
+      "step": 29
+    },
+    {
+      "epoch": 0.024,
+      "grad_norm": 73.18660736083984,
+      "learning_rate": 5e-05,
+      "loss": 2.9916,
+      "step": 30
+    },
+    {
+      "epoch": 0.0248,
+      "grad_norm": 38.29869842529297,
+      "learning_rate": 5e-05,
+      "loss": 1.5562,
+      "step": 31
+    },
+    {
+      "epoch": 0.0256,
+      "grad_norm": 84.11820983886719,
+      "learning_rate": 5e-05,
+      "loss": 3.0164,
+      "step": 32
+    },
+    {
+      "epoch": 0.0264,
+      "grad_norm": 32.736000061035156,
+      "learning_rate": 5e-05,
+      "loss": 2.1702,
+      "step": 33
+    },
+    {
+      "epoch": 0.0272,
+      "grad_norm": 11.174713134765625,
+      "learning_rate": 5e-05,
+      "loss": 2.5096,
+      "step": 34
+    },
+    {
+      "epoch": 0.028,
+      "grad_norm": 90.99018859863281,
+      "learning_rate": 5e-05,
+      "loss": 3.0305,
+      "step": 35
+    },
+    {
+      "epoch": 0.0288,
+      "grad_norm": 100.95272064208984,
+      "learning_rate": 5e-05,
+      "loss": 1.2811,
+      "step": 36
+    },
+    {
+      "epoch": 0.0296,
+      "grad_norm": 53.98057556152344,
+      "learning_rate": 5e-05,
+      "loss": 2.7442,
+      "step": 37
+    },
+    {
+      "epoch": 0.0304,
+      "grad_norm": 54.6403923034668,
+      "learning_rate": 5e-05,
+      "loss": 1.6351,
+      "step": 38
+    },
+    {
+      "epoch": 0.0312,
+      "grad_norm": 18.330358505249023,
+      "learning_rate": 5e-05,
+      "loss": 1.4583,
+      "step": 39
+    },
+    {
+      "epoch": 0.032,
+      "grad_norm": 31.618942260742188,
+      "learning_rate": 5e-05,
+      "loss": 1.0834,
+      "step": 40
+    },
+    {
+      "epoch": 0.0328,
+      "grad_norm": 30.400362014770508,
+      "learning_rate": 5e-05,
+      "loss": 1.1342,
+      "step": 41
+    },
+    {
+      "epoch": 0.0336,
+      "grad_norm": 245.12649536132812,
+      "learning_rate": 5e-05,
+      "loss": 2.4042,
+      "step": 42
+    },
+    {
+      "epoch": 0.0344,
+      "grad_norm": 77.76103210449219,
+      "learning_rate": 5e-05,
+      "loss": 1.6741,
+      "step": 43
+    },
+    {
+      "epoch": 0.0352,
+      "grad_norm": 469.37353515625,
+      "learning_rate": 5e-05,
+      "loss": 2.1188,
+      "step": 44
+    },
+    {
+      "epoch": 0.036,
+      "grad_norm": 19.93143653869629,
+      "learning_rate": 5e-05,
+      "loss": 1.5071,
+      "step": 45
+    },
+    {
+      "epoch": 0.0368,
+      "grad_norm": 90.47894287109375,
+      "learning_rate": 5e-05,
+      "loss": 1.4904,
+      "step": 46
+    },
+    {
+      "epoch": 0.0376,
+      "grad_norm": 110.54138946533203,
+      "learning_rate": 5e-05,
+      "loss": 1.3893,
+      "step": 47
+    },
+    {
+      "epoch": 0.0384,
+      "grad_norm": 414.5013122558594,
+      "learning_rate": 5e-05,
+      "loss": 1.4519,
+      "step": 48
+    },
+    {
+      "epoch": 0.0392,
+      "grad_norm": 36.63255310058594,
+      "learning_rate": 5e-05,
+      "loss": 0.5828,
+      "step": 49
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 82.29105377197266,
+      "learning_rate": 5e-05,
+      "loss": 1.1791,
+      "step": 50
+    },
+    {
+      "epoch": 0.0408,
+      "grad_norm": 80.6322250366211,
+      "learning_rate": 5e-05,
+      "loss": 1.4122,
+      "step": 51
+    },
+    {
+      "epoch": 0.0416,
+      "grad_norm": 26.663949966430664,
+      "learning_rate": 5e-05,
+      "loss": 0.8776,
+      "step": 52
+    },
+    {
+      "epoch": 0.0424,
+      "grad_norm": 442.86920166015625,
+      "learning_rate": 5e-05,
+      "loss": 2.4989,
+      "step": 53
+    },
+    {
+      "epoch": 0.0432,
+      "grad_norm": 289.60760498046875,
+      "learning_rate": 5e-05,
+      "loss": 1.8916,
+      "step": 54
+    },
+    {
+      "epoch": 0.044,
+      "grad_norm": 247.45352172851562,
+      "learning_rate": 5e-05,
+      "loss": 2.4255,
+      "step": 55
+    },
+    {
+      "epoch": 0.0448,
+      "grad_norm": 137.6370391845703,
+      "learning_rate": 5e-05,
+      "loss": 1.3678,
+      "step": 56
+    },
+    {
+      "epoch": 0.0456,
+      "grad_norm": 31.924896240234375,
+      "learning_rate": 5e-05,
+      "loss": 1.0117,
+      "step": 57
+    },
+    {
+      "epoch": 0.0464,
+      "grad_norm": 70.21894836425781,
+      "learning_rate": 5e-05,
+      "loss": 1.6799,
+      "step": 58
+    },
+    {
+      "epoch": 0.0472,
+      "grad_norm": 23.50946807861328,
+      "learning_rate": 5e-05,
+      "loss": 1.2422,
+      "step": 59
+    },
+    {
+      "epoch": 0.048,
+      "grad_norm": 89.09075927734375,
+      "learning_rate": 5e-05,
+      "loss": 1.6662,
+      "step": 60
+    },
+    {
+      "epoch": 0.0488,
+      "grad_norm": 43.365318298339844,
+      "learning_rate": 5e-05,
+      "loss": 1.0336,
+      "step": 61
+    },
+    {
+      "epoch": 0.0496,
+      "grad_norm": 12.232425689697266,
+      "learning_rate": 5e-05,
+      "loss": 1.1569,
+      "step": 62
+    },
+    {
+      "epoch": 0.0504,
+      "grad_norm": 31.25607681274414,
+      "learning_rate": 5e-05,
+      "loss": 1.1013,
+      "step": 63
+    },
+    {
+      "epoch": 0.0512,
+      "grad_norm": 13.216445922851562,
+      "learning_rate": 5e-05,
+      "loss": 1.046,
+      "step": 64
+    },
+    {
+      "epoch": 0.052,
+      "grad_norm": 23.276023864746094,
+      "learning_rate": 5e-05,
+      "loss": 1.3445,
+      "step": 65
+    },
+    {
+      "epoch": 0.0528,
+      "grad_norm": 22.61781883239746,
+      "learning_rate": 5e-05,
+      "loss": 0.9096,
+      "step": 66
+    },
+    {
+      "epoch": 0.0536,
+      "grad_norm": 4.904455184936523,
+      "learning_rate": 5e-05,
+      "loss": 0.5843,
+      "step": 67
+    },
+    {
+      "epoch": 0.0544,
+      "grad_norm": 33.0333366394043,
+      "learning_rate": 5e-05,
+      "loss": 1.0526,
+      "step": 68
+    },
+    {
+      "epoch": 0.0552,
+      "grad_norm": 13.542631149291992,
+      "learning_rate": 5e-05,
+      "loss": 0.8788,
+      "step": 69
+    },
+    {
+      "epoch": 0.056,
+      "grad_norm": 97.36151885986328,
+      "learning_rate": 5e-05,
+      "loss": 1.5369,
+      "step": 70
+    },
+    {
+      "epoch": 0.0568,
+      "grad_norm": 17.800167083740234,
+      "learning_rate": 5e-05,
+      "loss": 0.8806,
+      "step": 71
+    },
+    {
+      "epoch": 0.0576,
+      "grad_norm": 86.41162109375,
+      "learning_rate": 5e-05,
+      "loss": 1.1313,
+      "step": 72
+    },
+    {
+      "epoch": 0.0584,
+      "grad_norm": 29.609695434570312,
+      "learning_rate": 5e-05,
+      "loss": 1.1085,
+      "step": 73
+    },
+    {
+      "epoch": 0.0592,
+      "grad_norm": 13.794703483581543,
+      "learning_rate": 5e-05,
+      "loss": 1.0289,
+      "step": 74
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 49.20547103881836,
+      "learning_rate": 5e-05,
+      "loss": 0.7551,
+      "step": 75
+    },
+    {
+      "epoch": 0.0608,
+      "grad_norm": 23.741147994995117,
+      "learning_rate": 5e-05,
+      "loss": 1.0409,
+      "step": 76
+    },
+    {
+      "epoch": 0.0616,
+      "grad_norm": 24.001245498657227,
+      "learning_rate": 5e-05,
+      "loss": 0.85,
+      "step": 77
+    },
+    {
+      "epoch": 0.0624,
+      "grad_norm": 79.71287536621094,
+      "learning_rate": 5e-05,
+      "loss": 0.9315,
+      "step": 78
+    },
+    {
+      "epoch": 0.0632,
+      "grad_norm": 12.45206356048584,
+      "learning_rate": 5e-05,
+      "loss": 0.691,
+      "step": 79
+    },
+    {
+      "epoch": 0.064,
+      "grad_norm": 24.522785186767578,
+      "learning_rate": 5e-05,
+      "loss": 0.8791,
+      "step": 80
+    },
+    {
+      "epoch": 0.0648,
+      "grad_norm": 51.09123229980469,
+      "learning_rate": 5e-05,
+      "loss": 1.2123,
+      "step": 81
+    },
+    {
+      "epoch": 0.0656,
+      "grad_norm": 90.79096221923828,
+      "learning_rate": 5e-05,
+      "loss": 1.0276,
+      "step": 82
+    },
+    {
+      "epoch": 0.0664,
+      "grad_norm": 8.028018951416016,
+      "learning_rate": 5e-05,
+      "loss": 0.6102,
+      "step": 83
+    },
+    {
+      "epoch": 0.0672,
+      "grad_norm": 10.042688369750977,
+      "learning_rate": 5e-05,
+      "loss": 0.7536,
+      "step": 84
+    },
+    {
+      "epoch": 0.068,
+      "grad_norm": 27.800851821899414,
+      "learning_rate": 5e-05,
+      "loss": 0.8081,
+      "step": 85
+    },
+    {
+      "epoch": 0.0688,
+      "grad_norm": 11.373533248901367,
+      "learning_rate": 5e-05,
+      "loss": 0.7412,
+      "step": 86
+    },
+    {
+      "epoch": 0.0696,
+      "grad_norm": 9.913511276245117,
+      "learning_rate": 5e-05,
+      "loss": 0.7042,
+      "step": 87
+    },
+    {
+      "epoch": 0.0704,
+      "grad_norm": 6.745258331298828,
+      "learning_rate": 5e-05,
+      "loss": 0.661,
+      "step": 88
+    },
+    {
+      "epoch": 0.0712,
+      "grad_norm": 38.00721740722656,
+      "learning_rate": 5e-05,
+      "loss": 0.591,
+      "step": 89
+    },
+    {
+      "epoch": 0.072,
+      "grad_norm": 76.44242858886719,
+      "learning_rate": 5e-05,
+      "loss": 1.0196,
+      "step": 90
+    },
+    {
+      "epoch": 0.0728,
+      "grad_norm": 32.63544464111328,
+      "learning_rate": 5e-05,
+      "loss": 1.0174,
+      "step": 91
+    },
+    {
+      "epoch": 0.0736,
+      "grad_norm": 44.3709831237793,
+      "learning_rate": 5e-05,
+      "loss": 0.8848,
+      "step": 92
+    },
+    {
+      "epoch": 0.0744,
+      "grad_norm": 10.414836883544922,
+      "learning_rate": 5e-05,
+      "loss": 0.7761,
+      "step": 93
+    },
+    {
+      "epoch": 0.0752,
+      "grad_norm": 46.201881408691406,
+      "learning_rate": 5e-05,
+      "loss": 0.7125,
+      "step": 94
+    },
+    {
+      "epoch": 0.076,
+      "grad_norm": 12.817488670349121,
+      "learning_rate": 5e-05,
+      "loss": 1.0058,
+      "step": 95
+    },
+    {
+      "epoch": 0.0768,
+      "grad_norm": 15.410782814025879,
+      "learning_rate": 5e-05,
+      "loss": 0.6288,
+      "step": 96
+    },
+    {
+      "epoch": 0.0776,
+      "grad_norm": 49.45286178588867,
+      "learning_rate": 5e-05,
+      "loss": 0.8551,
+      "step": 97
+    },
+    {
+      "epoch": 0.0784,
+      "grad_norm": 61.248252868652344,
+      "learning_rate": 5e-05,
+      "loss": 0.9115,
+      "step": 98
+    },
+    {
+      "epoch": 0.0792,
+      "grad_norm": 20.97413444519043,
+      "learning_rate": 5e-05,
+      "loss": 0.7247,
+      "step": 99
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 41.232147216796875,
+      "learning_rate": 5e-05,
+      "loss": 1.0657,
+      "step": 100
+    },
+    {
+      "epoch": 0.0808,
+      "grad_norm": 8.603179931640625,
+      "learning_rate": 5e-05,
+      "loss": 0.535,
+      "step": 101
+    },
+    {
+      "epoch": 0.0816,
+      "grad_norm": 19.60603141784668,
+      "learning_rate": 5e-05,
+      "loss": 1.2771,
+      "step": 102
+    },
+    {
+      "epoch": 0.0824,
+      "grad_norm": 23.732683181762695,
+      "learning_rate": 5e-05,
+      "loss": 0.6206,
+      "step": 103
+    },
+    {
+      "epoch": 0.0832,
+      "grad_norm": 50.34906768798828,
+      "learning_rate": 5e-05,
+      "loss": 1.1698,
+      "step": 104
+    },
+    {
+      "epoch": 0.084,
+      "grad_norm": 59.62549591064453,
+      "learning_rate": 5e-05,
+      "loss": 0.9733,
+      "step": 105
+    },
+    {
+      "epoch": 0.0848,
+      "grad_norm": 13.478074073791504,
+      "learning_rate": 5e-05,
+      "loss": 0.5164,
+      "step": 106
+    },
+    {
+      "epoch": 0.0856,
+      "grad_norm": 31.350929260253906,
+      "learning_rate": 5e-05,
+      "loss": 0.894,
+      "step": 107
+    },
+    {
+      "epoch": 0.0864,
+      "grad_norm": 10.55340576171875,
+      "learning_rate": 5e-05,
+      "loss": 0.7449,
+      "step": 108
+    },
+    {
+      "epoch": 0.0872,
+      "grad_norm": 170.94677734375,
+      "learning_rate": 5e-05,
+      "loss": 0.8123,
+      "step": 109
+    },
+    {
+      "epoch": 0.088,
+      "grad_norm": 15.990272521972656,
+      "learning_rate": 5e-05,
+      "loss": 0.7861,
+      "step": 110
+    },
+    {
+      "epoch": 0.0888,
+      "grad_norm": 12.844974517822266,
+      "learning_rate": 5e-05,
+      "loss": 0.632,
+      "step": 111
+    },
+    {
+      "epoch": 0.0896,
+      "grad_norm": 127.54450225830078,
+      "learning_rate": 5e-05,
+      "loss": 0.7552,
+      "step": 112
+    },
+    {
+      "epoch": 0.0904,
+      "grad_norm": 11.704280853271484,
+      "learning_rate": 5e-05,
+      "loss": 0.5294,
+      "step": 113
+    },
+    {
+      "epoch": 0.0912,
+      "grad_norm": 17.201644897460938,
+      "learning_rate": 5e-05,
+      "loss": 0.6632,
+      "step": 114
+    },
+    {
+      "epoch": 0.092,
+      "grad_norm": 167.67779541015625,
+      "learning_rate": 5e-05,
+      "loss": 1.4338,
+      "step": 115
+    },
+    {
+      "epoch": 0.0928,
+      "grad_norm": 28.528623580932617,
+      "learning_rate": 5e-05,
+      "loss": 0.7563,
+      "step": 116
+    },
+    {
+      "epoch": 0.0936,
+      "grad_norm": 112.38704681396484,
+      "learning_rate": 5e-05,
+      "loss": 0.7817,
+      "step": 117
+    },
+    {
+      "epoch": 0.0944,
+      "grad_norm": 40.8664665222168,
+      "learning_rate": 5e-05,
+      "loss": 1.1821,
+      "step": 118
+    },
+    {
+      "epoch": 0.0952,
+      "grad_norm": 9.281485557556152,
+      "learning_rate": 5e-05,
+      "loss": 0.2872,
+      "step": 119
+    },
+    {
+      "epoch": 0.096,
+      "grad_norm": 18.671119689941406,
+      "learning_rate": 5e-05,
+      "loss": 1.2851,
+      "step": 120
+    },
+    {
+      "epoch": 0.0968,
+      "grad_norm": 11.585042953491211,
+      "learning_rate": 5e-05,
+      "loss": 0.8585,
+      "step": 121
+    },
+    {
+      "epoch": 0.0976,
+      "grad_norm": 56.25844192504883,
+      "learning_rate": 5e-05,
+      "loss": 0.6552,
+      "step": 122
+    },
+    {
+      "epoch": 0.0984,
+      "grad_norm": 18.261459350585938,
+      "learning_rate": 5e-05,
+      "loss": 0.7763,
+      "step": 123
+    },
+    {
+      "epoch": 0.0992,
+      "grad_norm": 11.38892650604248,
+      "learning_rate": 5e-05,
+      "loss": 0.9474,
+      "step": 124
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 20.487577438354492,
+      "learning_rate": 5e-05,
+      "loss": 1.1939,
+      "step": 125
+    },
+    {
+      "epoch": 0.1008,
+      "grad_norm": 8.532489776611328,
+      "learning_rate": 5e-05,
+      "loss": 0.4274,
+      "step": 126
+    },
+    {
+      "epoch": 0.1016,
+      "grad_norm": 15.358688354492188,
+      "learning_rate": 5e-05,
+      "loss": 0.663,
+      "step": 127
+    },
+    {
+      "epoch": 0.1024,
+      "grad_norm": 20.530738830566406,
+      "learning_rate": 5e-05,
+      "loss": 0.5538,
+      "step": 128
+    },
+    {
+      "epoch": 0.1032,
+      "grad_norm": 18.50092315673828,
+      "learning_rate": 5e-05,
+      "loss": 0.9013,
+      "step": 129
+    },
+    {
+      "epoch": 0.104,
+      "grad_norm": 19.617862701416016,
+      "learning_rate": 5e-05,
+      "loss": 0.7365,
+      "step": 130
+    },
+    {
+      "epoch": 0.1048,
+      "grad_norm": 34.48674392700195,
+      "learning_rate": 5e-05,
+      "loss": 0.5827,
+      "step": 131
+    },
+    {
+      "epoch": 0.1056,
+      "grad_norm": 66.11063385009766,
+      "learning_rate": 5e-05,
+      "loss": 0.5861,
+      "step": 132
+    },
+    {
+      "epoch": 0.1064,
+      "grad_norm": 22.789804458618164,
+      "learning_rate": 5e-05,
+      "loss": 0.4679,
+      "step": 133
+    },
+    {
+      "epoch": 0.1072,
+      "grad_norm": 57.6861457824707,
+      "learning_rate": 5e-05,
+      "loss": 0.5005,
+      "step": 134
+    },
+    {
+      "epoch": 0.108,
+      "grad_norm": 85.26701354980469,
+      "learning_rate": 5e-05,
+      "loss": 1.0884,
+      "step": 135
+    },
+    {
+      "epoch": 0.1088,
+      "grad_norm": 65.12076568603516,
+      "learning_rate": 5e-05,
+      "loss": 0.9267,
+      "step": 136
+    },
+    {
+      "epoch": 0.1096,
+      "grad_norm": 305.76849365234375,
+      "learning_rate": 5e-05,
+      "loss": 2.0525,
+      "step": 137
+    },
+    {
+      "epoch": 0.1104,
+      "grad_norm": 11.639004707336426,
+      "learning_rate": 5e-05,
+      "loss": 0.7315,
+      "step": 138
+    },
+    {
+      "epoch": 0.1112,
+      "grad_norm": 18.41367530822754,
+      "learning_rate": 5e-05,
+      "loss": 0.6855,
+      "step": 139
+    },
+    {
+      "epoch": 0.112,
+      "grad_norm": 17.532852172851562,
+      "learning_rate": 5e-05,
+      "loss": 1.1566,
+      "step": 140
+    },
+    {
+      "epoch": 0.1128,
+      "grad_norm": 32.43601608276367,
+      "learning_rate": 5e-05,
+      "loss": 0.6385,
+      "step": 141
+    },
+    {
+      "epoch": 0.1136,
+      "grad_norm": 8.654618263244629,
+      "learning_rate": 5e-05,
+      "loss": 0.4523,
+      "step": 142
+    },
+    {
+      "epoch": 0.1144,
+      "grad_norm": 36.675498962402344,
+      "learning_rate": 5e-05,
+      "loss": 1.0112,
+      "step": 143
+    },
+    {
+      "epoch": 0.1152,
+      "grad_norm": 24.551057815551758,
+      "learning_rate": 5e-05,
+      "loss": 0.4609,
+      "step": 144
+    },
+    {
+      "epoch": 0.116,
+      "grad_norm": 22.59528350830078,
+      "learning_rate": 5e-05,
+      "loss": 0.6426,
+      "step": 145
+    },
+    {
+      "epoch": 0.1168,
+      "grad_norm": 67.65180206298828,
+      "learning_rate": 5e-05,
+      "loss": 1.1627,
+      "step": 146
+    },
+    {
+      "epoch": 0.1176,
+      "grad_norm": 11.300127983093262,
+      "learning_rate": 5e-05,
+      "loss": 0.3845,
+      "step": 147
+    },
+    {
+      "epoch": 0.1184,
+      "grad_norm": 21.384666442871094,
+      "learning_rate": 5e-05,
+      "loss": 0.7098,
+      "step": 148
+    },
+    {
+      "epoch": 0.1192,
+      "grad_norm": 62.568607330322266,
+      "learning_rate": 5e-05,
+      "loss": 1.1119,
+      "step": 149
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 24.714191436767578,
+      "learning_rate": 5e-05,
+      "loss": 0.826,
+      "step": 150
+    },
+    {
+      "epoch": 0.1208,
+      "grad_norm": 34.44570541381836,
+      "learning_rate": 5e-05,
+      "loss": 0.7143,
+      "step": 151
+    },
+    {
+      "epoch": 0.1216,
+      "grad_norm": 94.24315643310547,
+      "learning_rate": 5e-05,
+      "loss": 1.1708,
+      "step": 152
+    },
+    {
+      "epoch": 0.1224,
+      "grad_norm": 9.813197135925293,
+      "learning_rate": 5e-05,
+      "loss": 0.6275,
+      "step": 153
+    },
+    {
+      "epoch": 0.1232,
+      "grad_norm": 16.591598510742188,
+      "learning_rate": 5e-05,
+      "loss": 0.7265,
+      "step": 154
+    },
+    {
+      "epoch": 0.124,
+      "grad_norm": 13.579938888549805,
+      "learning_rate": 5e-05,
+      "loss": 0.425,
+      "step": 155
+    },
+    {
+      "epoch": 0.1248,
+      "grad_norm": 32.443199157714844,
+      "learning_rate": 5e-05,
+      "loss": 1.5429,
+      "step": 156
+    },
+    {
+      "epoch": 0.1256,
+      "grad_norm": 12.89716911315918,
+      "learning_rate": 5e-05,
+      "loss": 0.5597,
+      "step": 157
+    },
+    {
+      "epoch": 0.1264,
+      "grad_norm": 67.0173110961914,
+      "learning_rate": 5e-05,
+      "loss": 0.5709,
+      "step": 158
+    },
+    {
+      "epoch": 0.1272,
+      "grad_norm": 82.58885955810547,
+      "learning_rate": 5e-05,
+      "loss": 0.8661,
+      "step": 159
+    },
+    {
+      "epoch": 0.128,
+      "grad_norm": 21.63603401184082,
+      "learning_rate": 5e-05,
+      "loss": 0.8811,
+      "step": 160
+    },
+    {
+      "epoch": 0.1288,
+      "grad_norm": 23.74791717529297,
+      "learning_rate": 5e-05,
+      "loss": 0.2251,
+      "step": 161
+    },
+    {
+      "epoch": 0.1296,
+      "grad_norm": 15.014976501464844,
+      "learning_rate": 5e-05,
+      "loss": 0.4781,
+      "step": 162
+    },
+    {
+      "epoch": 0.1304,
+      "grad_norm": 28.59008026123047,
+      "learning_rate": 5e-05,
+      "loss": 1.0019,
+      "step": 163
+    },
+    {
+      "epoch": 0.1312,
+      "grad_norm": 27.027162551879883,
+      "learning_rate": 5e-05,
+      "loss": 0.3676,
+      "step": 164
+    },
+    {
+      "epoch": 0.132,
+      "grad_norm": 45.08534622192383,
+      "learning_rate": 5e-05,
+      "loss": 0.8913,
+      "step": 165
+    },
+    {
+      "epoch": 0.1328,
+      "grad_norm": 213.11904907226562,
+      "learning_rate": 5e-05,
+      "loss": 0.5001,
+      "step": 166
+    },
+    {
+      "epoch": 0.1336,
+      "grad_norm": 31.432199478149414,
+      "learning_rate": 5e-05,
+      "loss": 0.3927,
+      "step": 167
+    },
+    {
+      "epoch": 0.1344,
+      "grad_norm": 30.716411590576172,
+      "learning_rate": 5e-05,
+      "loss": 0.9168,
+      "step": 168
+    },
+    {
+      "epoch": 0.1352,
+      "grad_norm": 12.920553207397461,
+      "learning_rate": 5e-05,
+      "loss": 0.46,
+      "step": 169
+    },
+    {
+      "epoch": 0.136,
+      "grad_norm": 32.00464630126953,
+      "learning_rate": 5e-05,
+      "loss": 0.2163,
+      "step": 170
+    },
+    {
+      "epoch": 0.1368,
+      "grad_norm": 38.64021301269531,
+      "learning_rate": 5e-05,
+      "loss": 0.4853,
+      "step": 171
+    },
+    {
+      "epoch": 0.1376,
+      "grad_norm": 39.885597229003906,
+      "learning_rate": 5e-05,
+      "loss": 0.3095,
+      "step": 172
+    },
+    {
+      "epoch": 0.1384,
+      "grad_norm": 9.49242877960205,
+      "learning_rate": 5e-05,
+      "loss": 0.2723,
+      "step": 173
+    },
+    {
+      "epoch": 0.1392,
+      "grad_norm": 25.025390625,
+      "learning_rate": 5e-05,
+      "loss": 0.5781,
+      "step": 174
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 68.46289825439453,
+      "learning_rate": 5e-05,
+      "loss": 1.01,
+      "step": 175
+    },
+    {
+      "epoch": 0.1408,
+      "grad_norm": 26.592926025390625,
+      "learning_rate": 5e-05,
+      "loss": 0.6925,
+      "step": 176
+    },
+    {
+      "epoch": 0.1416,
+      "grad_norm": 66.05874633789062,
+      "learning_rate": 5e-05,
+      "loss": 0.9902,
+      "step": 177
+    },
+    {
+      "epoch": 0.1424,
+      "grad_norm": 76.02796173095703,
+      "learning_rate": 5e-05,
+      "loss": 0.7473,
+      "step": 178
+    },
+    {
+      "epoch": 0.1432,
+      "grad_norm": 34.00507354736328,
+      "learning_rate": 5e-05,
+      "loss": 1.0052,
+      "step": 179
+    },
+    {
+      "epoch": 0.144,
+      "grad_norm": 26.019983291625977,
+      "learning_rate": 5e-05,
+      "loss": 0.8738,
+      "step": 180
+    },
+    {
+      "epoch": 0.1448,
+      "grad_norm": 17.291433334350586,
+      "learning_rate": 5e-05,
+      "loss": 0.193,
+      "step": 181
+    },
+    {
+      "epoch": 0.1456,
+      "grad_norm": 28.545869827270508,
+      "learning_rate": 5e-05,
+      "loss": 0.3898,
+      "step": 182
+    },
+    {
+      "epoch": 0.1464,
+      "grad_norm": 90.70147705078125,
+      "learning_rate": 5e-05,
+      "loss": 0.8012,
+      "step": 183
+    },
+    {
+      "epoch": 0.1472,
+      "grad_norm": 42.486366271972656,
+      "learning_rate": 5e-05,
+      "loss": 0.8143,
+      "step": 184
+    },
+    {
+      "epoch": 0.148,
+      "grad_norm": 55.433929443359375,
+      "learning_rate": 5e-05,
+      "loss": 0.6451,
+      "step": 185
+    },
+    {
+      "epoch": 0.1488,
+      "grad_norm": 63.269493103027344,
+      "learning_rate": 5e-05,
+      "loss": 1.4339,
+      "step": 186
+    },
+    {
+      "epoch": 0.1496,
+      "grad_norm": 31.41185188293457,
+      "learning_rate": 5e-05,
+      "loss": 0.683,
+      "step": 187
+    },
+    {
+      "epoch": 0.1504,
+      "grad_norm": 16.21335792541504,
+      "learning_rate": 5e-05,
+      "loss": 0.3486,
+      "step": 188
+    },
+    {
+      "epoch": 0.1512,
+      "grad_norm": 23.199403762817383,
+      "learning_rate": 5e-05,
+      "loss": 0.619,
+      "step": 189
+    },
+    {
+      "epoch": 0.152,
+      "grad_norm": 34.67924118041992,
+      "learning_rate": 5e-05,
+      "loss": 0.8388,
+      "step": 190
+    },
+    {
+      "epoch": 0.1528,
+      "grad_norm": 22.459701538085938,
+      "learning_rate": 5e-05,
+      "loss": 0.5954,
+      "step": 191
+    },
+    {
+      "epoch": 0.1536,
+      "grad_norm": 30.882579803466797,
+      "learning_rate": 5e-05,
+      "loss": 0.4792,
+      "step": 192
+    },
+    {
+      "epoch": 0.1544,
+      "grad_norm": 40.165061950683594,
+      "learning_rate": 5e-05,
+      "loss": 0.32,
+      "step": 193
+    },
+    {
+      "epoch": 0.1552,
+      "grad_norm": 410.47467041015625,
+      "learning_rate": 5e-05,
+      "loss": 1.519,
+      "step": 194
+    },
+    {
+      "epoch": 0.156,
+      "grad_norm": 150.8389892578125,
+      "learning_rate": 5e-05,
+      "loss": 1.6528,
+      "step": 195
+    },
+    {
+      "epoch": 0.1568,
+      "grad_norm": 49.18366622924805,
+      "learning_rate": 5e-05,
+      "loss": 0.3058,
+      "step": 196
+    },
+    {
+      "epoch": 0.1576,
+      "grad_norm": 55.75705337524414,
+      "learning_rate": 5e-05,
+      "loss": 0.6507,
+      "step": 197
+    },
+    {
+      "epoch": 0.1584,
+      "grad_norm": 58.94297790527344,
+      "learning_rate": 5e-05,
+      "loss": 0.9967,
+      "step": 198
+    },
+    {
+      "epoch": 0.1592,
+      "grad_norm": 10.483887672424316,
+      "learning_rate": 5e-05,
+      "loss": 0.3913,
+      "step": 199
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 42.88235855102539,
+      "learning_rate": 5e-05,
+      "loss": 0.7079,
+      "step": 200
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 200,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 7.02021397118976e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-200/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ebcdb71d423afb91b29c6b8e7f2ffb998bd98942b05b44f31625a192b74fe9e
+size 7288

checkpoint-200/zero_to_fp32.py

@@ -0,0 +1,760 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device, weights_only=False)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
+
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    torch_state_dict = {}
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:  # shared tensors
+            shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
+            torch_state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                torch_state_dict[name] = tensor.contiguous()
+    return torch_state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in list(shard_state_dict.keys()):
+            del state_dict[tensor_name]
+            del shard_state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "mistralai/Mistral-7B-Instruct-v0.3",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.49.0",
+  "use_cache": true,
+  "vocab_size": 32768
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.49.0"
+}

lorra_config.json

@@ -0,0 +1,8 @@
+{
+  "target_layers": "10,20",
+  "transform_layers": "-1",
+  "lorra_alpha": 10.0,
+  "trainsets": null,
+  "valsets": null,
+  "full_layers": false
+}

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecb40ec3402c656fb629c59092bf1e17188459809519b70044901546c466361e
+size 4949453792

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68a92eabaf4998ad8b4927fb7551bc02f835ba0f1c434d3094785c1af7c1b591
+size 4999819336

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad1c303be23a5a85ac593d5cf1777cb68c31aba20f6579722b252c56d44c0278
+size 4915916184

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:834eac485fc701f874020061e9cae400392decd22b8ecb232e85fdbb0c9005a1
+size 4429367168

model.safetensors.index.json

@@ -0,0 +1,397 @@
+{
+  "metadata": {
+    "total_size": 19294511104
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.42.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.42.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.42.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.42.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.42.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.42.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.42.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.42.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.42.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.norm.weight": "model-00004-of-00004.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37f00374dea48658ee8f5d0f21895b9bc55cb0103939607c8185bfd1c6ca1f89
+size 587404