Model save

README.md

@@ -0,0 +1,62 @@
+---
+license: apache-2.0
+base_model: mistralai/Mistral-7B-v0.1
+tags:
+- generated_from_trainer
+model-index:
+- name: zephyr-7b-sft-lora
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# zephyr-7b-sft-lora
+
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.0667
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 32
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 512
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 1.063         | 0.67  | 272  | 1.0667          |
+
+
+### Framework versions
+
+- Transformers 4.35.0
+- Pytorch 2.1.1+cu121
+- Datasets 2.14.6
+- Tokenizers 0.14.1

adapter_config.json

@@ -0,0 +1,25 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mistral-7B-v0.1",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj",
+    "o_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38c4cc3db797cde9c3e4214f38f663a87bbca856e02bb4d2a95bb006aca3741f
+size 109086672

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 0.67,
+    "eval_loss": 1.0666559934616089,
+    "eval_runtime": 2404.6287,
+    "eval_samples": 23110,
+    "eval_samples_per_second": 9.611,
+    "eval_steps_per_second": 0.151,
+    "train_loss": 1.1599171240540112,
+    "train_runtime": 73159.9281,
+    "train_samples": 207865,
+    "train_samples_per_second": 2.841,
+    "train_steps_per_second": 0.006
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.67,
+    "eval_loss": 1.0666559934616089,
+    "eval_runtime": 2404.6287,
+    "eval_samples": 23110,
+    "eval_samples_per_second": 9.611,
+    "eval_steps_per_second": 0.151
+}

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_config.json

@@ -0,0 +1,41 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n'  + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 2048,
+  "pad_token": "</s>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.67,
+    "train_loss": 1.1599171240540112,
+    "train_runtime": 73159.9281,
+    "train_samples": 207865,
+    "train_samples_per_second": 2.841,
+    "train_steps_per_second": 0.006
+}

trainer_state.json

@@ -0,0 +1,366 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.6699507389162561,
+  "eval_steps": 500,
+  "global_step": 272,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.9999700625010444e-05,
+      "loss": 1.6931,
+      "step": 1
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.999251652147735e-05,
+      "loss": 1.6625,
+      "step": 5
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.997007728639956e-05,
+      "loss": 1.5879,
+      "step": 10
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9932715879473385e-05,
+      "loss": 1.5023,
+      "step": 15
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9880488219356086e-05,
+      "loss": 1.4524,
+      "step": 20
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.981347247496222e-05,
+      "loss": 1.3857,
+      "step": 25
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.973176894846855e-05,
+      "loss": 1.348,
+      "step": 30
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.963549992519223e-05,
+      "loss": 1.3286,
+      "step": 35
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9524809490566878e-05,
+      "loss": 1.2921,
+      "step": 40
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.939986331449053e-05,
+      "loss": 1.2736,
+      "step": 45
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.926084840336821e-05,
+      "loss": 1.2586,
+      "step": 50
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.910797282022027e-05,
+      "loss": 1.2334,
+      "step": 55
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.894146537327533e-05,
+      "loss": 1.2086,
+      "step": 60
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8761575273514005e-05,
+      "loss": 1.189,
+      "step": 65
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.8568571761675893e-05,
+      "loss": 1.1707,
+      "step": 70
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.8362743705288127e-05,
+      "loss": 1.1654,
+      "step": 75
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.814439916631857e-05,
+      "loss": 1.1668,
+      "step": 80
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.791386494010081e-05,
+      "loss": 1.1569,
+      "step": 85
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.7671486066220965e-05,
+      "loss": 1.1376,
+      "step": 90
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.7417625312098453e-05,
+      "loss": 1.1222,
+      "step": 95
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.7152662630033506e-05,
+      "loss": 1.1304,
+      "step": 100
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.6876994588534234e-05,
+      "loss": 1.1425,
+      "step": 105
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.659103377877423e-05,
+      "loss": 1.1127,
+      "step": 110
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.629520819706912e-05,
+      "loss": 1.1069,
+      "step": 115
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.598996060429634e-05,
+      "loss": 1.1078,
+      "step": 120
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.56757478632168e-05,
+      "loss": 1.103,
+      "step": 125
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.5353040254690396e-05,
+      "loss": 1.1107,
+      "step": 130
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.5022320773808612e-05,
+      "loss": 1.1062,
+      "step": 135
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.4684084406997903e-05,
+      "loss": 1.0909,
+      "step": 140
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.4338837391175582e-05,
+      "loss": 1.0942,
+      "step": 145
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.3987096456067236e-05,
+      "loss": 1.0887,
+      "step": 150
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.3629388050819547e-05,
+      "loss": 1.0813,
+      "step": 155
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.3266247556066122e-05,
+      "loss": 1.0699,
+      "step": 160
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.2898218482625606e-05,
+      "loss": 1.0875,
+      "step": 165
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.252585165803135e-05,
+      "loss": 1.0872,
+      "step": 170
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.2149704402110243e-05,
+      "loss": 1.0709,
+      "step": 175
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.1770339692844484e-05,
+      "loss": 1.0772,
+      "step": 180
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.1388325323764889e-05,
+      "loss": 1.0847,
+      "step": 185
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.1004233054136726e-05,
+      "loss": 1.0778,
+      "step": 190
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.0618637753210086e-05,
+      "loss": 1.0798,
+      "step": 195
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.0232116539815558e-05,
+      "loss": 1.0671,
+      "step": 200
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.845247918592937e-06,
+      "loss": 1.0715,
+      "step": 205
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 9.458610914145826e-06,
+      "loss": 1.07,
+      "step": 210
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 9.072784204417995e-06,
+      "loss": 1.0795,
+      "step": 215
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 8.688345254588579e-06,
+      "loss": 1.0686,
+      "step": 220
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 8.305869452782446e-06,
+      "loss": 1.0716,
+      "step": 225
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 7.92592924888925e-06,
+      "loss": 1.0707,
+      "step": 230
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 7.549093297780133e-06,
+      "loss": 1.067,
+      "step": 235
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 7.175925608204428e-06,
+      "loss": 1.0732,
+      "step": 240
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 6.806984698640202e-06,
+      "loss": 1.0746,
+      "step": 245
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 6.442822761362015e-06,
+      "loss": 1.0622,
+      "step": 250
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 6.083984835977154e-06,
+      "loss": 1.0552,
+      "step": 255
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 5.731007993667155e-06,
+      "loss": 1.0697,
+      "step": 260
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 5.38442053335571e-06,
+      "loss": 1.0599,
+      "step": 265
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 5.044741191005908e-06,
+      "loss": 1.063,
+      "step": 270
+    },
+    {
+      "epoch": 0.67,
+      "eval_loss": 1.066710114479065,
+      "eval_runtime": 2407.1881,
+      "eval_samples_per_second": 9.6,
+      "eval_steps_per_second": 0.15,
+      "step": 272
+    },
+    {
+      "epoch": 0.67,
+      "step": 272,
+      "total_flos": 4.932776606721638e+16,
+      "train_loss": 1.1599171240540112,
+      "train_runtime": 73159.9281,
+      "train_samples_per_second": 2.841,
+      "train_steps_per_second": 0.006
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 406,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "total_flos": 4.932776606721638e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:894ab50c9898c2dee04f953d44234befab17dc2499229944290efadb69ed352d
+size 5624