Training in progress, step 168

.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_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-Nemo-12B-Instruct/lora/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-Nemo-12B-Instruct-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_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": [
+    "k_proj",
+    "down_proj",
+    "q_proj",
+    "o_proj",
+    "v_proj",
+    "gate_proj",
+    "up_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 sha256:41dba1a3813b753b664642d6a5bdebf06876732d2cc16c6bc18f22cdb1f9e654
+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_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-Nemo-12B-Instruct/lora/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-Nemo-12B-Instruct-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_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 +1,17 @@
+{"current_steps": 10, "total_steps": 168, "loss": 1.2612, "learning_rate": 2.647058823529412e-06, "epoch": 0.17777777777777778, "percentage": 5.95, "elapsed_time": "0:00:29", "remaining_time": "0:07:48"}
+{"current_steps": 20, "total_steps": 168, "loss": 1.2578, "learning_rate": 4.997836020254328e-06, "epoch": 0.35555555555555557, "percentage": 11.9, "elapsed_time": "0:00:57", "remaining_time": "0:07:06"}
+{"current_steps": 30, "total_steps": 168, "loss": 1.0594, "learning_rate": 4.922489359292928e-06, "epoch": 0.5333333333333333, "percentage": 17.86, "elapsed_time": "0:01:25", "remaining_time": "0:06:32"}
+{"current_steps": 40, "total_steps": 168, "loss": 0.7524, "learning_rate": 4.7426609101991605e-06, "epoch": 0.7111111111111111, "percentage": 23.81, "elapsed_time": "0:01:53", "remaining_time": "0:06:03"}
+{"current_steps": 50, "total_steps": 168, "loss": 0.5438, "learning_rate": 4.466106660773884e-06, "epoch": 0.8888888888888888, "percentage": 29.76, "elapsed_time": "0:02:21", "remaining_time": "0:05:34"}
+{"current_steps": 60, "total_steps": 168, "loss": 0.481, "learning_rate": 4.104754375481665e-06, "epoch": 1.0666666666666667, "percentage": 35.71, "elapsed_time": "0:02:49", "remaining_time": "0:05:05"}
+{"current_steps": 70, "total_steps": 168, "loss": 0.2657, "learning_rate": 3.674189151845515e-06, "epoch": 1.2444444444444445, "percentage": 41.67, "elapsed_time": "0:03:17", "remaining_time": "0:04:36"}
+{"current_steps": 80, "total_steps": 168, "loss": 0.1834, "learning_rate": 3.1929812363354766e-06, "epoch": 1.4222222222222223, "percentage": 47.62, "elapsed_time": "0:03:45", "remaining_time": "0:04:08"}
+{"current_steps": 90, "total_steps": 168, "loss": 0.164, "learning_rate": 2.68188509100236e-06, "epoch": 1.6, "percentage": 53.57, "elapsed_time": "0:04:13", "remaining_time": "0:03:40"}
+{"current_steps": 100, "total_steps": 168, "loss": 0.1039, "learning_rate": 2.1629442550539283e-06, "epoch": 1.7777777777777777, "percentage": 59.52, "elapsed_time": "0:04:41", "remaining_time": "0:03:11"}
+{"current_steps": 110, "total_steps": 168, "loss": 0.1299, "learning_rate": 1.6585406086279847e-06, "epoch": 1.9555555555555557, "percentage": 65.48, "elapsed_time": "0:05:10", "remaining_time": "0:02:43"}
+{"current_steps": 120, "total_steps": 168, "loss": 0.0871, "learning_rate": 1.1904290439459974e-06, "epoch": 2.1333333333333333, "percentage": 71.43, "elapsed_time": "0:05:38", "remaining_time": "0:02:15"}
+{"current_steps": 130, "total_steps": 168, "loss": 0.1068, "learning_rate": 7.787991784095e-07, "epoch": 2.311111111111111, "percentage": 77.38, "elapsed_time": "0:06:05", "remaining_time": "0:01:46"}
+{"current_steps": 140, "total_steps": 168, "loss": 0.1238, "learning_rate": 4.414045778845144e-07, "epoch": 2.488888888888889, "percentage": 83.33, "elapsed_time": "0:06:33", "remaining_time": "0:01:18"}
+{"current_steps": 150, "total_steps": 168, "loss": 0.1024, "learning_rate": 1.927970467097573e-07, "epoch": 2.6666666666666665, "percentage": 89.29, "elapsed_time": "0:07:01", "remaining_time": "0:00:50"}
+{"current_steps": 160, "total_steps": 168, "loss": 0.0923, "learning_rate": 4.369900944435734e-08, "epoch": 2.8444444444444446, "percentage": 95.24, "elapsed_time": "0:07:29", "remaining_time": "0:00:22"}
+{"current_steps": 168, "total_steps": 168, "epoch": 2.986666666666667, "percentage": 100.0, "elapsed_time": "0:07:53", "remaining_time": "0:00:00"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5607cec29c654bea9d011aef860bb72d887f8478ae20b5a2d11f46368b515c7b
+size 5432