Model save

README.md

@@ -0,0 +1,64 @@
+---
+tags:
+- generated_from_trainer
+model-index:
+- name: vicuna-adv-robust-u5-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. -->
+
+# vicuna-adv-robust-u5-sft-lora
+
+This model was trained from scratch on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.2913
+
+## 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: 3e-05
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 128
+- total_train_batch_size: 2048
+- total_eval_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- num_epochs: 5
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 1.4063        | 0.63  | 7    | 1.3767          |
+| 1.3987        | 1.61  | 14   | 1.3552          |
+| 1.3645        | 2.59  | 21   | 1.3310          |
+| 1.35          | 3.57  | 28   | 1.3090          |
+| 1.3197        | 4.64  | 36   | 1.2913          |
+
+
+### Framework versions
+
+- Transformers 4.35.0
+- Pytorch 2.1.0a0+32f93b1
+- 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": "/data/models/vicuna-7b-v1.3",
+  "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": [
+    "v_proj",
+    "o_proj",
+    "q_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fafc799dc70cf6f908e2f27c283f42c792981a11bb34365e853aaa1a02e266e6
+size 134252592

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 4.64,
+    "eval_loss": 1.291313886642456,
+    "eval_runtime": 1335.0386,
+    "eval_samples": 23344,
+    "eval_samples_per_second": 17.486,
+    "eval_steps_per_second": 0.547,
+    "train_loss": 1.366490102476544,
+    "train_runtime": 26909.2833,
+    "train_samples": 22883,
+    "train_samples_per_second": 4.252,
+    "train_steps_per_second": 0.002
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.64,
+    "eval_loss": 1.291313886642456,
+    "eval_runtime": 1335.0386,
+    "eval_samples": 23344,
+    "eval_samples_per_second": 17.486,
+    "eval_steps_per_second": 0.547
+}

runs/Dec31_18-42-46_f63ae9056bcd/events.out.tfevents.1704048198.f63ae9056bcd.761409.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:874347d561f2958ffd1b4c844e3952e12b274a9cc06182987951e388552cffd2
+size 7232

runs/Dec31_18-42-46_f63ae9056bcd/events.out.tfevents.1704076442.f63ae9056bcd.761409.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3abf473c2431a5f39e29dfbe8caad0bc96d69c12af2a3230f54db0fc1c33ef45
+size 306

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "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": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "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 +1,43 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "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: ' + message['content'] + ' '}}{% elif message['role'] == 'system' %}{{ message['content'] + ' '}}{% elif message['role'] == 'assistant' %}\n{{ 'ASSISTANT: '  + message['content'] + ' '}}{% 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": "<unk>",
+  "padding_side": "right",
+  "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": 4.64,
+    "train_loss": 1.366490102476544,
+    "train_runtime": 26909.2833,
+    "train_samples": 22883,
+    "train_samples_per_second": 4.252,
+    "train_steps_per_second": 0.002
+}

trainer_state.json

@@ -0,0 +1,116 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.64011180992313,
+  "eval_steps": 500,
+  "global_step": 36,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09,
+      "learning_rate": 2.9975536558892034e-05,
+      "loss": 1.4153,
+      "step": 1
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.939239460421746e-05,
+      "loss": 1.4063,
+      "step": 5
+    },
+    {
+      "epoch": 0.63,
+      "eval_loss": 1.3766872882843018,
+      "eval_runtime": 1336.0539,
+      "eval_samples_per_second": 17.472,
+      "eval_steps_per_second": 0.546,
+      "step": 7
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 2.7618802992467718e-05,
+      "loss": 1.3987,
+      "step": 10
+    },
+    {
+      "epoch": 1.61,
+      "eval_loss": 1.3552420139312744,
+      "eval_runtime": 1337.2341,
+      "eval_samples_per_second": 17.457,
+      "eval_steps_per_second": 0.546,
+      "step": 14
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 2.482291100917928e-05,
+      "loss": 1.3839,
+      "step": 15
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 2.12312251950283e-05,
+      "loss": 1.3645,
+      "step": 20
+    },
+    {
+      "epoch": 2.59,
+      "eval_loss": 1.3309670686721802,
+      "eval_runtime": 1334.4732,
+      "eval_samples_per_second": 17.493,
+      "eval_steps_per_second": 0.547,
+      "step": 21
+    },
+    {
+      "epoch": 3.3,
+      "learning_rate": 1.713472257409928e-05,
+      "loss": 1.35,
+      "step": 25
+    },
+    {
+      "epoch": 3.57,
+      "eval_loss": 1.3089555501937866,
+      "eval_runtime": 1335.8968,
+      "eval_samples_per_second": 17.474,
+      "eval_steps_per_second": 0.546,
+      "step": 28
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 1.2865277425900725e-05,
+      "loss": 1.3338,
+      "step": 30
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 8.768774804971705e-06,
+      "loss": 1.3197,
+      "step": 35
+    },
+    {
+      "epoch": 4.64,
+      "eval_loss": 1.291340708732605,
+      "eval_runtime": 1335.9273,
+      "eval_samples_per_second": 17.474,
+      "eval_steps_per_second": 0.546,
+      "step": 36
+    },
+    {
+      "epoch": 4.64,
+      "step": 36,
+      "total_flos": 4.623623575620813e+16,
+      "train_loss": 1.366490102476544,
+      "train_runtime": 26909.2833,
+      "train_samples_per_second": 4.252,
+      "train_steps_per_second": 0.002
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 55,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 4.623623575620813e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93d9d22b3fc76e5b26ca0c048f4d83849186eb7a94aceb92582e47e9a5c394db
+size 5688