add: add all test results

results/lora_sft_Baichuan-7B_4_gpu_50_step_20240905050501/lora_sft_Baichuan-7B_4_gpu_50_step_20240905050501.yml

@@ -0,0 +1,31 @@
+bf16: true
+cutoff_len: 1024
+dataset: belle_1m
+ddp_timeout: 180000000
+do_train: true
+eval_steps: 500
+eval_strategy: steps
+finetuning_type: lora
+gradient_accumulation_steps: 8
+include_num_input_tokens_seen: true
+include_tokens_per_second: true
+learning_rate: 0.0001
+logging_steps: 3
+lora_target: all
+lr_scheduler_type: cosine
+max_samples: 10000
+max_steps: '50'
+model_name_or_path: ''
+num_train_epochs: 10.0
+output_dir: ./results/lora_sft_Baichuan-7B_4_gpu_50_step_20240905050501
+overwrite_cache: true
+overwrite_output_dir: true
+per_device_eval_batch_size: 2
+per_device_train_batch_size: 2
+plot_loss: true
+preprocessing_num_workers: 16
+save_steps: 500
+stage: sft
+template: ''
+val_size: 0.1
+warmup_ratio: 0.1

results/lora_sft_Baichuan-7B_4_gpu_50_step_20240905050958/lora_sft_Baichuan-7B_4_gpu_50_step_20240905050958.yaml

@@ -0,0 +1,31 @@
+bf16: true
+cutoff_len: 1024
+dataset: belle_1m
+ddp_timeout: 180000000
+do_train: true
+eval_steps: 500
+eval_strategy: steps
+finetuning_type: lora
+gradient_accumulation_steps: 8
+include_num_input_tokens_seen: true
+include_tokens_per_second: true
+learning_rate: 0.0001
+logging_steps: 3
+lora_target: all
+lr_scheduler_type: cosine
+max_samples: 10000
+max_steps: '50'
+model_name_or_path: ''
+num_train_epochs: 10.0
+output_dir: ./results/lora_sft_Baichuan-7B_4_gpu_50_step_20240905050958
+overwrite_cache: true
+overwrite_output_dir: true
+per_device_eval_batch_size: 2
+per_device_train_batch_size: 2
+plot_loss: true
+preprocessing_num_workers: 16
+save_steps: 500
+stage: sft
+template: ''
+val_size: 0.1
+warmup_ratio: 0.1

results/lora_sft_Baichuan-7B_4_gpu_50_step_20240905051039/lora_sft_Baichuan-7B_4_gpu_50_step_20240905051039.yaml

@@ -0,0 +1,31 @@
+bf16: true
+cutoff_len: 1024
+dataset: belle_1m
+ddp_timeout: 180000000
+do_train: true
+eval_steps: 500
+eval_strategy: steps
+finetuning_type: lora
+gradient_accumulation_steps: 8
+include_num_input_tokens_seen: true
+include_tokens_per_second: true
+learning_rate: 0.0001
+logging_steps: 3
+lora_target: all
+lr_scheduler_type: cosine
+max_samples: 10000
+max_steps: '50'
+model_name_or_path: ''
+num_train_epochs: 10.0
+output_dir: ./results/lora_sft_Baichuan-7B_4_gpu_50_step_20240905051039
+overwrite_cache: true
+overwrite_output_dir: true
+per_device_eval_batch_size: 2
+per_device_train_batch_size: 2
+plot_loss: true
+preprocessing_num_workers: 16
+save_steps: 500
+stage: sft
+template: ''
+val_size: 0.1
+warmup_ratio: 0.1

results/lora_sft_Qwen-7B_4_gpu_50_step_20240905052241/lora_sft_Qwen-7B_4_gpu_50_step_20240905052241.yaml

@@ -0,0 +1,31 @@
+bf16: true
+cutoff_len: 1024
+dataset: belle_1m
+ddp_timeout: 180000000
+do_train: true
+eval_steps: 500
+eval_strategy: steps
+finetuning_type: lora
+gradient_accumulation_steps: 8
+include_num_input_tokens_seen: true
+include_tokens_per_second: true
+learning_rate: 0.0001
+logging_steps: 3
+lora_target: all
+lr_scheduler_type: cosine
+max_samples: 10000
+max_steps: '50'
+model_name_or_path: ../../models/Qwen-7B
+num_train_epochs: 10.0
+output_dir: ./results/lora_sft_Qwen-7B_4_gpu_50_step_20240905052241
+overwrite_cache: true
+overwrite_output_dir: true
+per_device_eval_batch_size: 2
+per_device_train_batch_size: 2
+plot_loss: true
+preprocessing_num_workers: 16
+save_steps: 500
+stage: sft
+template: qwen
+val_size: 0.1
+warmup_ratio: 0.1

results/lora_sft_Qwen-7B_4_gpu_50_step_20240905053758/lora_sft_Qwen-7B_4_gpu_50_step_20240905053758.yaml

@@ -0,0 +1,31 @@
+bf16: true
+cutoff_len: 1024
+dataset: belle_1m
+ddp_timeout: 180000000
+do_train: true
+eval_steps: 500
+eval_strategy: steps
+finetuning_type: lora
+gradient_accumulation_steps: 8
+include_num_input_tokens_seen: true
+include_tokens_per_second: true
+learning_rate: 0.0001
+logging_steps: 3
+lora_target: all
+lr_scheduler_type: cosine
+max_samples: 10000
+max_steps: '50'
+model_name_or_path: ../../models/Qwen-7B
+num_train_epochs: 10.0
+output_dir: ./results/lora_sft_Qwen-7B_4_gpu_50_step_20240905053758
+overwrite_cache: true
+overwrite_output_dir: true
+per_device_eval_batch_size: 2
+per_device_train_batch_size: 2
+plot_loss: true
+preprocessing_num_workers: 16
+save_steps: 500
+stage: sft
+template: qwen
+val_size: 0.1
+warmup_ratio: 0.1

results/lora_sft_Qwen-7B_4_gpu_50_step_20240905064243/lora_sft_Qwen-7B_4_gpu_50_step_20240905064243.yaml

@@ -0,0 +1,31 @@
+bf16: true
+cutoff_len: 1024
+dataset: belle_1m
+ddp_timeout: 180000000
+do_train: true
+eval_steps: 500
+eval_strategy: steps
+finetuning_type: lora
+gradient_accumulation_steps: 8
+include_num_input_tokens_seen: true
+include_tokens_per_second: true
+learning_rate: 0.0001
+logging_steps: 3
+lora_target: all
+lr_scheduler_type: cosine
+max_samples: 10000
+max_steps: '50'
+model_name_or_path: ../../models/Qwen-7B
+num_train_epochs: 10.0
+output_dir: ./results/lora_sft_Qwen-7B_4_gpu_50_step_20240905064243
+overwrite_cache: true
+overwrite_output_dir: true
+per_device_eval_batch_size: 2
+per_device_train_batch_size: 2
+plot_loss: true
+preprocessing_num_workers: 16
+save_steps: 500
+stage: sft
+template: qwen
+val_size: 0.1
+warmup_ratio: 0.1

results/lora_sft_Qwen-7B_4_gpu_50_step_20240905064736/lora_sft_Qwen-7B_4_gpu_50_step_20240905064736.yaml

@@ -0,0 +1,31 @@
+bf16: true
+cutoff_len: 1024
+dataset: belle_1m
+ddp_timeout: 180000000
+do_train: true
+eval_steps: 500
+eval_strategy: steps
+finetuning_type: lora
+gradient_accumulation_steps: 8
+include_num_input_tokens_seen: true
+include_tokens_per_second: true
+learning_rate: 0.0001
+logging_steps: 3
+lora_target: all
+lr_scheduler_type: cosine
+max_samples: 10000
+max_steps: '50'
+model_name_or_path: ../../models/Qwen-7B
+num_train_epochs: 10.0
+output_dir: ./results/lora_sft_Qwen-7B_4_gpu_50_step_20240905064736
+overwrite_cache: true
+overwrite_output_dir: true
+per_device_eval_batch_size: 2
+per_device_train_batch_size: 2
+plot_loss: true
+preprocessing_num_workers: 16
+save_steps: 500
+stage: sft
+template: qwen
+val_size: 0.1
+warmup_ratio: 0.1