Model save

.ipynb_checkpoints/llama3_lora_sft-checkpoint.yaml

@@ -0,0 +1,44 @@
+### model
+model_name_or_path: mistralai/Mistral-7B-Instruct-v0.3
+adapter_name_or_path: chchen/Mistral-7B-Instruct-v0.3-ORPO
+
+### method
+stage: sft
+do_train: true
+finetuning_type: lora
+lora_target: all
+
+### dataset
+dataset: bct_non_cot_sft_1000
+dataset_dir: data_private
+template: mistral
+cutoff_len: 1024
+# max_samples: 1000
+overwrite_cache: true
+preprocessing_num_workers: 16
+
+### output
+output_dir: saves/Mistral-7B-Instruct-v0.3/lora/orpo-sft
+logging_steps: 10
+save_steps: 500
+plot_loss: true
+overwrite_output_dir: true
+save_total_limit: 3
+load_best_model_at_end: true
+push_to_hub: true
+hub_model_id: chchen/Mistral-7B-Instruct-v0.3-ORPO-SFT
+
+### train
+per_device_train_batch_size: 2
+gradient_accumulation_steps: 8
+learning_rate: 0.000005
+num_train_epochs: 3.0
+lr_scheduler_type: cosine
+warmup_steps: 0.1
+fp16: true
+
+### eval
+val_size: 0.1
+per_device_eval_batch_size: 2
+evaluation_strategy: steps
+eval_steps: 500

README.md

@@ -0,0 +1,59 @@
+---
+license: apache-2.0
+library_name: peft
+tags:
+- llama-factory
+- generated_from_trainer
+base_model: mistralai/Mistral-7B-Instruct-v0.3
+model-index:
+- name: Mistral-7B-Instruct-v0.3-ORPO-SFT
+  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-Instruct-v0.3-ORPO-SFT
+
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on an unknown 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: 5e-06
+- train_batch_size: 2
+- eval_batch_size: 2
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 16
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 0.1
+- num_epochs: 3.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.40.1
+- Pytorch 2.3.0
+- Datasets 2.19.0
+- Tokenizers 0.19.1

adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mistral-7B-Instruct-v0.3",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "q_proj",
+    "o_proj",
+    "down_proj",
+    "gate_proj",
+    "up_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f7f015cf23166c18732c60a171adbfb36532160fe1a7d9dfc84ef0f50c230bd
+size 83945296

llama3_lora_sft.yaml

@@ -0,0 +1,44 @@
+### model
+model_name_or_path: mistralai/Mistral-7B-Instruct-v0.3
+adapter_name_or_path: chchen/Mistral-7B-Instruct-v0.3-ORPO
+
+### method
+stage: sft
+do_train: true
+finetuning_type: lora
+lora_target: all
+
+### dataset
+dataset: bct_non_cot_sft_1000
+dataset_dir: data_private
+template: mistral
+cutoff_len: 1024
+# max_samples: 1000
+overwrite_cache: true
+preprocessing_num_workers: 16
+
+### output
+output_dir: saves/Mistral-7B-Instruct-v0.3/lora/orpo-sft
+logging_steps: 10
+save_steps: 500
+plot_loss: true
+overwrite_output_dir: true
+save_total_limit: 3
+load_best_model_at_end: true
+push_to_hub: true
+hub_model_id: chchen/Mistral-7B-Instruct-v0.3-ORPO-SFT
+
+### train
+per_device_train_batch_size: 2
+gradient_accumulation_steps: 8
+learning_rate: 0.000005
+num_train_epochs: 3.0
+lr_scheduler_type: cosine
+warmup_steps: 0.1
+fp16: true
+
+### eval
+val_size: 0.1
+per_device_eval_batch_size: 2
+evaluation_strategy: steps
+eval_steps: 500

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

trainer_log.jsonl

@@ -0,0 +1,17 @@
+{"current_steps": 10, "total_steps": 168, "loss": 0.5704, "learning_rate": 4.97273712672844e-06, "epoch": 0.17777777777777778, "percentage": 5.95, "elapsed_time": "0:00:23", "remaining_time": "0:06:09"}
+{"current_steps": 20, "total_steps": 168, "loss": 0.2394, "learning_rate": 4.861084470200228e-06, "epoch": 0.35555555555555557, "percentage": 11.9, "elapsed_time": "0:00:44", "remaining_time": "0:05:27"}
+{"current_steps": 30, "total_steps": 168, "loss": 0.1046, "learning_rate": 4.667009949002349e-06, "epoch": 0.5333333333333333, "percentage": 17.86, "elapsed_time": "0:01:05", "remaining_time": "0:05:00"}
+{"current_steps": 40, "total_steps": 168, "loss": 0.0639, "learning_rate": 4.397288409237892e-06, "epoch": 0.7111111111111111, "percentage": 23.81, "elapsed_time": "0:01:26", "remaining_time": "0:04:36"}
+{"current_steps": 50, "total_steps": 168, "loss": 0.0841, "learning_rate": 4.061335419273658e-06, "epoch": 0.8888888888888888, "percentage": 29.76, "elapsed_time": "0:01:47", "remaining_time": "0:04:13"}
+{"current_steps": 60, "total_steps": 168, "loss": 0.1054, "learning_rate": 3.6708785865814186e-06, "epoch": 1.0666666666666667, "percentage": 35.71, "elapsed_time": "0:02:08", "remaining_time": "0:03:51"}
+{"current_steps": 70, "total_steps": 168, "loss": 0.0709, "learning_rate": 3.239548164813544e-06, "epoch": 1.2444444444444445, "percentage": 41.67, "elapsed_time": "0:02:29", "remaining_time": "0:03:28"}
+{"current_steps": 80, "total_steps": 168, "loss": 0.0684, "learning_rate": 2.782401242396799e-06, "epoch": 1.4222222222222223, "percentage": 47.62, "elapsed_time": "0:02:50", "remaining_time": "0:03:07"}
+{"current_steps": 90, "total_steps": 168, "loss": 0.082, "learning_rate": 2.3153961224961665e-06, "epoch": 1.6, "percentage": 53.57, "elapsed_time": "0:03:11", "remaining_time": "0:02:46"}
+{"current_steps": 100, "total_steps": 168, "loss": 0.0505, "learning_rate": 1.854835242944048e-06, "epoch": 1.7777777777777777, "percentage": 59.52, "elapsed_time": "0:03:32", "remaining_time": "0:02:24"}
+{"current_steps": 110, "total_steps": 168, "loss": 0.0843, "learning_rate": 1.4167960829520933e-06, "epoch": 1.9555555555555557, "percentage": 65.48, "elapsed_time": "0:03:54", "remaining_time": "0:02:03"}
+{"current_steps": 120, "total_steps": 168, "loss": 0.0642, "learning_rate": 1.0165699227860215e-06, "epoch": 2.1333333333333333, "percentage": 71.43, "elapsed_time": "0:04:14", "remaining_time": "0:01:41"}
+{"current_steps": 130, "total_steps": 168, "loss": 0.0684, "learning_rate": 6.681280484488576e-07, "epoch": 2.311111111111111, "percentage": 77.38, "elapsed_time": "0:04:35", "remaining_time": "0:01:20"}
+{"current_steps": 140, "total_steps": 168, "loss": 0.0837, "learning_rate": 3.8363403535449846e-07, "epoch": 2.488888888888889, "percentage": 83.33, "elapsed_time": "0:04:56", "remaining_time": "0:00:59"}
+{"current_steps": 150, "total_steps": 168, "loss": 0.0603, "learning_rate": 1.7301913642614382e-07, "epoch": 2.6666666666666665, "percentage": 89.29, "elapsed_time": "0:05:17", "remaining_time": "0:00:38"}
+{"current_steps": 160, "total_steps": 168, "loss": 0.06, "learning_rate": 4.3635597174694347e-08, "epoch": 2.8444444444444446, "percentage": 95.24, "elapsed_time": "0:05:38", "remaining_time": "0:00:16"}
+{"current_steps": 168, "total_steps": 168, "epoch": 2.986666666666667, "percentage": 100.0, "elapsed_time": "0:05:55", "remaining_time": "0:00:00"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3ad68ebce35a71fc610abb04de35579931dea0b4685aab6de3fdff3ec84e0c9
+size 5240