Training in progress, step 84

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

.ipynb_checkpoints/llama3_lora_sft-checkpoint.yaml

@@ -0,0 +1,43 @@
+### model
+model_name_or_path: mistralai/Mistral-Nemo-Instruct-2407
+
+### method
+stage: sft
+do_train: true
+finetuning_type: lora
+lora_target: all
+
+### dataset
+dataset: bct_non_cot_sft_500
+dataset_dir: data_private
+template: mistral
+cutoff_len: 1024
+# max_samples: 1000
+overwrite_cache: true
+preprocessing_num_workers: 16
+
+### output
+output_dir: saves/Mistral-Nemo-12B-Instruct/lora/sft-half
+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-Nemo-12B-Instruct-SFT-Half
+
+### 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_ratio: 0.1
+fp16: true
+
+### eval
+val_size: 0.1
+per_device_eval_batch_size: 2
+evaluation_strategy: steps
+eval_steps: 500

adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mistral-Nemo-Instruct-2407",
+  "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": [
+    "gate_proj",
+    "up_proj",
+    "q_proj",
+    "v_proj",
+    "down_proj",
+    "k_proj",
+    "o_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 sha256:29bc929133ea4f47a2abbe7e811c3aa8928ddb5225ced8ef0158ec0db292613a
+size 114106856

llama3_lora_sft.yaml

@@ -0,0 +1,43 @@
+### model
+model_name_or_path: mistralai/Mistral-Nemo-Instruct-2407
+
+### method
+stage: sft
+do_train: true
+finetuning_type: lora
+lora_target: all
+
+### dataset
+dataset: bct_non_cot_sft_500
+dataset_dir: data_private
+template: mistral
+cutoff_len: 1024
+# max_samples: 1000
+overwrite_cache: true
+preprocessing_num_workers: 16
+
+### output
+output_dir: saves/Mistral-Nemo-12B-Instruct/lora/sft-half
+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-Nemo-12B-Instruct-SFT-Half
+
+### 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_ratio: 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.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0240ce510f08e6c2041724e9043e33be9d251d1e4a4d94eb68cd47b954b61d2
+size 17078292

trainer_log.jsonl

@@ -0,0 +1,9 @@
+{"current_steps": 10, "total_steps": 84, "loss": 1.2892, "learning_rate": 4.444444444444444e-06, "epoch": 0.35555555555555557, "percentage": 11.9, "elapsed_time": "0:00:28", "remaining_time": "0:03:32"}
+{"current_steps": 20, "total_steps": 84, "loss": 1.0878, "learning_rate": 4.824441214720629e-06, "epoch": 0.7111111111111111, "percentage": 23.81, "elapsed_time": "0:00:56", "remaining_time": "0:03:00"}
+{"current_steps": 30, "total_steps": 84, "loss": 0.8957, "learning_rate": 4.249158351283414e-06, "epoch": 1.0666666666666667, "percentage": 35.71, "elapsed_time": "0:01:24", "remaining_time": "0:02:31"}
+{"current_steps": 40, "total_steps": 84, "loss": 0.7399, "learning_rate": 3.3714301183045382e-06, "epoch": 1.4222222222222223, "percentage": 47.62, "elapsed_time": "0:01:51", "remaining_time": "0:02:02"}
+{"current_steps": 50, "total_steps": 84, "loss": 0.6107, "learning_rate": 2.3430237011767166e-06, "epoch": 1.7777777777777777, "percentage": 59.52, "elapsed_time": "0:02:18", "remaining_time": "0:01:34"}
+{"current_steps": 60, "total_steps": 84, "loss": 0.4844, "learning_rate": 1.3417599122003464e-06, "epoch": 2.1333333333333333, "percentage": 71.43, "elapsed_time": "0:02:46", "remaining_time": "0:01:06"}
+{"current_steps": 70, "total_steps": 84, "loss": 0.4673, "learning_rate": 5.407663566854008e-07, "epoch": 2.488888888888889, "percentage": 83.33, "elapsed_time": "0:03:13", "remaining_time": "0:00:38"}
+{"current_steps": 80, "total_steps": 84, "loss": 0.4532, "learning_rate": 7.854209717842231e-08, "epoch": 2.8444444444444446, "percentage": 95.24, "elapsed_time": "0:03:40", "remaining_time": "0:00:11"}
+{"current_steps": 84, "total_steps": 84, "epoch": 2.986666666666667, "percentage": 100.0, "elapsed_time": "0:03:53", "remaining_time": "0:00:00"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ce690d1ec19158e125774208e79614df3257cada0d1352ea3aac5474ffd256f
+size 5496