Model save

README.md

@@ -0,0 +1,61 @@
+---
+license: apache-2.0
+base_model: princeton-nlp/Sheared-LLaMA-1.3B
+tags:
+- generated_from_trainer
+model-index:
+- name: Sheared-LLaMA-1.3B-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. -->
+
+# Sheared-LLaMA-1.3B-sft-lora
+
+This model is a fine-tuned version of [princeton-nlp/Sheared-LLaMA-1.3B](https://huggingface.co/princeton-nlp/Sheared-LLaMA-1.3B) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.3605
+
+## 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: 1e-05
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- gradient_accumulation_steps: 128
+- total_train_batch_size: 512
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- training_steps: 272
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 1.3556        | 0.67  | 272  | 1.3605          |
+
+
+### Framework versions
+
+- Transformers 4.35.0
+- Pytorch 2.1.0+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": "princeton-nlp/Sheared-LLaMA-1.3B",
+  "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": [
+    "k_proj",
+    "q_proj",
+    "o_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e87f2eb451fd2f04e4d543f86120c35d1924ca7a41054223379afb13dc382ee
+size 100689136

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 0.67,
+    "eval_loss": 1.3605483770370483,
+    "eval_runtime": 2549.7249,
+    "eval_samples": 23110,
+    "eval_samples_per_second": 9.064,
+    "eval_steps_per_second": 1.133,
+    "train_loss": 1.3824449540937649,
+    "train_runtime": 69463.2349,
+    "train_samples": 207865,
+    "train_samples_per_second": 2.005,
+    "train_steps_per_second": 0.004
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.67,
+    "eval_loss": 1.3605483770370483,
+    "eval_runtime": 2549.7249,
+    "eval_samples": 23110,
+    "eval_samples_per_second": 9.064,
+    "eval_steps_per_second": 1.133
+}

runs/Nov20_16-54-52_sebastian-dell/events.out.tfevents.1700495704.sebastian-dell.450708.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecd418e025e4ca5e890df5ee82f36a2dbf7559d73be3d9a3a696d4d8ebc60f25
+size 4678

runs/Nov20_17-19-26_sebastian-dell/events.out.tfevents.1700497179.sebastian-dell.452266.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c12ab2c8303b4388f00c7e6295270659d7d990ad1b025ce7e1090016fa93cf71
+size 13552

runs/Nov20_17-19-26_sebastian-dell/events.out.tfevents.1700569192.sebastian-dell.452266.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16e80c4274c11d607955bc030bb315da10591d05ca3669199c8d3fc3a6249984
+size 359

special_tokens_map.json

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

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,38 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "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>",
+  "model_max_length": 2048,
+  "pad_token": "</s>",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.67,
+    "train_loss": 1.3824449540937649,
+    "train_runtime": 69463.2349,
+    "train_samples": 207865,
+    "train_samples_per_second": 2.005,
+    "train_steps_per_second": 0.004
+}

trainer_state.json

@@ -0,0 +1,366 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.6699636307656782,
+  "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": 3.5714285714285716e-07,
+      "loss": 1.4263,
+      "step": 1
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7857142857142859e-06,
+      "loss": 1.4275,
+      "step": 5
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5714285714285718e-06,
+      "loss": 1.4298,
+      "step": 10
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 5.357142857142857e-06,
+      "loss": 1.424,
+      "step": 15
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 7.1428571428571436e-06,
+      "loss": 1.4395,
+      "step": 20
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 8.92857142857143e-06,
+      "loss": 1.4306,
+      "step": 25
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.998342337571566e-06,
+      "loss": 1.4306,
+      "step": 30
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.97970625586178e-06,
+      "loss": 1.4257,
+      "step": 35
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.940439480455386e-06,
+      "loss": 1.4302,
+      "step": 40
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.880704691794608e-06,
+      "loss": 1.4312,
+      "step": 45
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.80074936835801e-06,
+      "loss": 1.4136,
+      "step": 50
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.70090476136855e-06,
+      "loss": 1.4217,
+      "step": 55
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.581584522435025e-06,
+      "loss": 1.4138,
+      "step": 60
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.443282989812494e-06,
+      "loss": 1.4091,
+      "step": 65
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.286573140381663e-06,
+      "loss": 1.4004,
+      "step": 70
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.112104215832047e-06,
+      "loss": 1.3973,
+      "step": 75
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 8.920599032883553e-06,
+      "loss": 1.3976,
+      "step": 80
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 8.712850988690094e-06,
+      "loss": 1.3982,
+      "step": 85
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 8.489720773831717e-06,
+      "loss": 1.3978,
+      "step": 90
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 8.25213280651317e-06,
+      "loss": 1.3858,
+      "step": 95
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.001071402741843e-06,
+      "loss": 1.379,
+      "step": 100
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 7.737576698351878e-06,
+      "loss": 1.386,
+      "step": 105
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 7.462740339769323e-06,
+      "loss": 1.4033,
+      "step": 110
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 7.177700961371239e-06,
+      "loss": 1.3782,
+      "step": 115
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 6.883639468175926e-06,
+      "loss": 1.3731,
+      "step": 120
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 6.58177414340781e-06,
+      "loss": 1.3741,
+      "step": 125
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 6.273355601206143e-06,
+      "loss": 1.3714,
+      "step": 130
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 5.959661605388229e-06,
+      "loss": 1.3783,
+      "step": 135
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 5.641991775732756e-06,
+      "loss": 1.3813,
+      "step": 140
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 5.321662203714909e-06,
+      "loss": 1.3619,
+      "step": 145
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 5e-06,
+      "loss": 1.3739,
+      "step": 150
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.678337796285093e-06,
+      "loss": 1.364,
+      "step": 155
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.358008224267245e-06,
+      "loss": 1.3588,
+      "step": 160
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.040338394611772e-06,
+      "loss": 1.3539,
+      "step": 165
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.726644398793857e-06,
+      "loss": 1.3588,
+      "step": 170
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.4182258565921933e-06,
+      "loss": 1.3632,
+      "step": 175
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 3.116360531824074e-06,
+      "loss": 1.3584,
+      "step": 180
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.822299038628762e-06,
+      "loss": 1.3546,
+      "step": 185
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.537259660230679e-06,
+      "loss": 1.3628,
+      "step": 190
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.2624233016481224e-06,
+      "loss": 1.3627,
+      "step": 195
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9989285972581595e-06,
+      "loss": 1.3644,
+      "step": 200
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7478671934868302e-06,
+      "loss": 1.356,
+      "step": 205
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.5102792261682813e-06,
+      "loss": 1.3556,
+      "step": 210
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.2871490113099066e-06,
+      "loss": 1.3545,
+      "step": 215
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.0794009671164484e-06,
+      "loss": 1.3502,
+      "step": 220
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 8.878957841679542e-07,
+      "loss": 1.3656,
+      "step": 225
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 7.13426859618338e-07,
+      "loss": 1.356,
+      "step": 230
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 5.567170101875074e-07,
+      "loss": 1.3602,
+      "step": 235
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.184154775649768e-07,
+      "loss": 1.3582,
+      "step": 240
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 2.990952386314505e-07,
+      "loss": 1.3545,
+      "step": 245
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.992506316419912e-07,
+      "loss": 1.3565,
+      "step": 250
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.192953082053927e-07,
+      "loss": 1.3637,
+      "step": 255
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 5.9560519544614725e-08,
+      "loss": 1.3604,
+      "step": 260
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2.0293744138219495e-08,
+      "loss": 1.347,
+      "step": 265
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.657662428434792e-09,
+      "loss": 1.3556,
+      "step": 270
+    },
+    {
+      "epoch": 0.67,
+      "eval_loss": 1.3605483770370483,
+      "eval_runtime": 2578.3151,
+      "eval_samples_per_second": 8.963,
+      "eval_steps_per_second": 1.12,
+      "step": 272
+    },
+    {
+      "epoch": 0.67,
+      "step": 272,
+      "total_flos": 2.233306234264486e+18,
+      "train_loss": 1.3824449540937649,
+      "train_runtime": 69463.2349,
+      "train_samples_per_second": 2.005,
+      "train_steps_per_second": 0.004
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 272,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "total_flos": 2.233306234264486e+18,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f37e038d92775abb991e0fc4e4d2831c4ef90f42f1523b65feb97c2dd836154
+size 4664