Model save

README.md

@@ -0,0 +1,63 @@
+---
+license: apache-2.0
+library_name: peft
+tags:
+- trl
+- sft
+- generated_from_trainer
+base_model: mistralai/Mistral-7B-Instruct-v0.1
+datasets:
+- generator
+model-index:
+- name: Mistral-7B-gsm8k
+  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. -->
+
+# Mistral-7B-gsm8k
+
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) on the generator dataset.
+
+## 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: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 3
+- total_train_batch_size: 6
+- total_eval_batch_size: 24
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.7.1
+- Transformers 4.36.2
+- Pytorch 2.1.2
+- Datasets 2.14.6
+- Tokenizers 0.15.2

adapter_config.json

@@ -0,0 +1,31 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mistral-7B-Instruct-v0.1",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 512,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 512,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "v_proj",
+    "down_proj",
+    "up_proj",
+    "o_proj",
+    "gate_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:062fe9d0c8f4c0d2bf2b831ac11a4edae6d44e9f7e01fb3f8d91f1e4e40969dc
+size 2684416656

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 0.8479397050605333,
+    "train_runtime": 621.2868,
+    "train_samples": 7473,
+    "train_samples_per_second": 1.167,
+    "train_steps_per_second": 0.195
+}

runs/Oct08_06-20-25_gpu4-119-5/events.out.tfevents.1728328885.gpu4-119-5.1531074.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a645f310e3cc81ee81ee49a81af788341325050c97cfbd5754eda34b95d93cf8
+size 4445

runs/Oct08_06-30-18_gpu4-119-5/events.out.tfevents.1728329491.gpu4-119-5.1539054.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0509d4ce98a988e9424fc9f9b3c25e99c5477e4cf19afbb703482826c0a40ed6
+size 8643

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:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443

tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "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": "{%- if messages[0]['role'] == 'system' %}\n    {%- set system_message = messages[0]['content'] %}\n    {%- set loop_messages = messages[1:] %}\n{%- else %}\n    {%- set loop_messages = messages %}\n{%- endif %}\n\n{{- bos_token }}\n{%- for message in loop_messages %}\n    {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}\n        {{- raise_exception('After the optional system message, conversation roles must alternate user/assistant/user/assistant/...') }}\n    {%- endif %}\n    {%- if message['role'] == 'user' %}\n        {%- if loop.first and system_message is defined %}\n            {{- ' [INST] ' + system_message + '\\n\\n' + message['content'] + ' [/INST]' }}\n        {%- else %}\n            {{- ' [INST] ' + message['content'] + ' [/INST]' }}\n        {%- endif %}\n    {%- elif message['role'] == 'assistant' %}\n        {{- ' ' + message['content'] + eos_token}}\n    {%- else %}\n        {{- raise_exception('Only user and assistant roles are supported, with the exception of an initial optional system message!') }}\n    {%- endif %}\n{%- endfor %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 0.8479397050605333,
+    "train_runtime": 621.2868,
+    "train_samples": 7473,
+    "train_samples_per_second": 1.167,
+    "train_steps_per_second": 0.195
+}

trainer_state.json

@@ -0,0 +1,180 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 121,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5384615384615387e-06,
+      "loss": 1.1746,
+      "step": 1
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 7.692307692307694e-06,
+      "loss": 1.1302,
+      "step": 5
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.5384615384615387e-05,
+      "loss": 0.9861,
+      "step": 10
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9983081582712684e-05,
+      "loss": 0.9154,
+      "step": 15
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9793406217655516e-05,
+      "loss": 0.8791,
+      "step": 20
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9396926207859085e-05,
+      "loss": 0.8825,
+      "step": 25
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.880201391180111e-05,
+      "loss": 0.8663,
+      "step": 30
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.802123192755044e-05,
+      "loss": 0.8349,
+      "step": 35
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.7071067811865477e-05,
+      "loss": 0.8493,
+      "step": 40
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.5971585917027864e-05,
+      "loss": 0.8408,
+      "step": 45
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.4746003697476406e-05,
+      "loss": 0.8537,
+      "step": 50
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.342020143325669e-05,
+      "loss": 0.8231,
+      "step": 55
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.2022175723320382e-05,
+      "loss": 0.8292,
+      "step": 60
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.0581448289104759e-05,
+      "loss": 0.81,
+      "step": 65
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 9.128442572523418e-06,
+      "loss": 0.8043,
+      "step": 70
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 7.6938412925756e-06,
+      "loss": 0.8119,
+      "step": 75
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 6.3079385268731575e-06,
+      "loss": 0.8052,
+      "step": 80
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 5.000000000000003e-06,
+      "loss": 0.7873,
+      "step": 85
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 3.797645087317401e-06,
+      "loss": 0.8023,
+      "step": 90
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 2.726263584269513e-06,
+      "loss": 0.8073,
+      "step": 95
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.808479557110081e-06,
+      "loss": 0.8061,
+      "step": 100
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.0636735967658785e-06,
+      "loss": 0.7997,
+      "step": 105
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 5.075735642696611e-07,
+      "loss": 0.7987,
+      "step": 110
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.519224698779198e-07,
+      "loss": 0.7965,
+      "step": 115
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.230499177994007e-09,
+      "loss": 0.8183,
+      "step": 120
+    },
+    {
+      "epoch": 1.0,
+      "step": 121,
+      "total_flos": 2375150469120.0,
+      "train_loss": 0.8479397050605333,
+      "train_runtime": 621.2868,
+      "train_samples_per_second": 1.167,
+      "train_steps_per_second": 0.195
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 121,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "total_flos": 2375150469120.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82f13e84b55cb0285e7494979cebe30b66616a43d816bbf5ca13854ed4954140
+size 5688