Upload sft_args.json

sft_args.json

@@ -7,7 +7,7 @@
   "additional_trainable_parameters": [],
   "tuner_backend": "peft",
   "template_type": "phi3-vl",
-  "output_dir": "D:\\_____NEW_NN\\LLM\\MiniCPM-V\\finetune\\output\\phi3-vision-128k-instruct\\v0-20240531-071942",
+  "output_dir": "D:\\_____NEW_NN\\LLM\\MiniCPM-V\\finetune\\output\\phi3-vision-128k-instruct\\v0-20240629-080216",
   "add_output_dir_suffix": true,
   "ddp_backend": null,
   "ddp_find_unused_parameters": null,
@@ -22,7 +22,7 @@
   ],
   "val_dataset": [],
   "dataset_seed": 42,
-  "dataset_test_ratio": 0.05,
+  "dataset_test_ratio": 0.07,
   "use_loss_scale": false,
   "system": null,
   "max_length": 2048,
@@ -45,21 +45,21 @@
   "bnb_4bit_use_double_quant": true,
   "bnb_4bit_quant_storage": null,
   "lora_target_modules": [
-    "v_proj",
-    "q_proj",
-    "out_proj",
-    "gate_up_proj",
+    "fc1",
     "img_projection.0",
-    "qkv_proj",
     "img_projection.2",
-    "fc1",
     "k_proj",
+    "out_proj",
     "fc2",
+    "qkv_proj",
     "o_proj",
-    "down_proj"
+    "q_proj",
+    "down_proj",
+    "v_proj",
+    "gate_up_proj"
   ],
-  "lora_rank": 64,
-  "lora_alpha": 64,
+  "lora_rank": 128,
+  "lora_alpha": 128,
   "lora_dropout_p": 0.05,
   "lora_bias_trainable": "none",
   "lora_modules_to_save": [],
@@ -115,26 +115,26 @@
   "lisa_step_interval": 20,
   "gradient_checkpointing": true,
   "deepspeed": null,
-  "batch_size": 1,
-  "eval_batch_size": 1,
+  "batch_size": 2,
+  "eval_batch_size": 2,
   "num_train_epochs": 4,
   "max_steps": -1,
   "optim": "adamw_torch",
   "adam_beta1": 0.9,
-  "adam_beta2": 0.999,
-  "learning_rate": 0.00015,
+  "adam_beta2": 0.95,
+  "learning_rate": 0.00014,
   "weight_decay": 0.1,
   "gradient_accumulation_steps": 2,
   "max_grad_norm": 0.5,
   "predict_with_generate": false,
-  "lr_scheduler_type": "linear",
+  "lr_scheduler_type": "cosine",
   "warmup_ratio": 0.05,
   "eval_steps": 50,
-  "save_steps": 369,
+  "save_steps": 300,
   "save_only_model": false,
-  "save_total_limit": 2,
+  "save_total_limit": 8,
   "logging_steps": 5,
-  "dataloader_num_workers": 1,
+  "dataloader_num_workers": 0,
   "dataloader_pin_memory": true,
   "dataloader_drop_last": false,
   "push_to_hub": false,
@@ -149,7 +149,7 @@
   "use_flash_attn": null,
   "ignore_args_error": false,
   "check_model_is_latest": true,
-  "logging_dir": "D:\\_____NEW_NN\\LLM\\MiniCPM-V\\finetune\\output\\phi3-vision-128k-instruct\\v0-20240531-071942/runs",
+  "logging_dir": "D:\\_____NEW_NN\\LLM\\MiniCPM-V\\finetune\\output\\phi3-vision-128k-instruct\\v0-20240629-080216/runs",
   "report_to": [
     "tensorboard"
   ],
@@ -179,6 +179,7 @@
   "fsdp_num": 1,
   "per_device_train_batch_size": null,
   "per_device_eval_batch_size": null,
+  "eval_strategy": null,
   "self_cognition_sample": 0,
   "train_dataset_mix_ratio": 0.0,
   "train_dataset_mix_ds": [
@@ -205,5 +206,5 @@
   "load_in_4bit": false,
   "load_in_8bit": false,
   "train_sampler_random": true,
-  "training_args": "Seq2SeqTrainingArguments(output_dir='D:\\\\_____NEW_NN\\\\LLM\\\\MiniCPM-V\\\\finetune\\\\output\\\\phi3-vision-128k-instruct\\\\v0-20240531-071942', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, evaluation_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=1, per_device_eval_batch_size=1, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=2, eval_accumulation_steps=None, eval_delay=0, learning_rate=0.00015, weight_decay=0.1, adam_beta1=0.9, adam_beta2=0.999, adam_epsilon=1e-08, max_grad_norm=0.5, num_train_epochs=4, max_steps=-1, lr_scheduler_type=<SchedulerType.LINEAR: 'linear'>, lr_scheduler_kwargs={}, warmup_ratio=0.05, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='D:\\\\_____NEW_NN\\\\LLM\\\\MiniCPM-V\\\\finetune\\\\output\\\\phi3-vision-128k-instruct\\\\v0-20240531-071942/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=5, logging_nan_inf_filter=True, save_strategy=<IntervalStrategy.STEPS: 'steps'>, save_steps=369, save_total_limit=2, save_safetensors=True, save_on_each_node=True, save_only_model=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=None, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=50, dataloader_num_workers=1, dataloader_prefetch_factor=None, past_index=-1, run_name='D:\\\\_____NEW_NN\\\\LLM\\\\MiniCPM-V\\\\finetune\\\\output\\\\phi3-vision-128k-instruct\\\\v0-20240531-071942', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=None, even_batches=True, use_seedable_sampler=True, gradient_accumulation_kwargs=None), deepspeed=None, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['tensorboard'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint=None, hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=False, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs=None, include_inputs_for_metrics=False, eval_do_concat_batches=True, fp16_backend='auto', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=1800, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=False, include_num_input_tokens_seen=False, neftune_noise_alpha=None, optim_target_modules=None, sortish_sampler=True, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=GenerationConfig {\n  \"do_sample\": true,\n  \"eos_token_id\": 32000,\n  \"max_new_tokens\": 2048,\n  \"pad_token_id\": 32000,\n  \"temperature\": 0.3,\n  \"top_k\": 20,\n  \"top_p\": 0.7\n}\n, train_sampler_random=True, push_hub_strategy='push_best', acc_strategy='token', additional_saved_files=[], metric_warmup_step=0, train_dataset_sample=738)"
+  "training_args": "Seq2SeqTrainingArguments(output_dir='D:\\\\_____NEW_NN\\\\LLM\\\\MiniCPM-V\\\\finetune\\\\output\\\\phi3-vision-128k-instruct\\\\v0-20240629-080216', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=2, per_device_eval_batch_size=2, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=2, eval_accumulation_steps=None, eval_delay=0, learning_rate=0.00014, weight_decay=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=0.5, num_train_epochs=4, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs={}, warmup_ratio=0.05, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='D:\\\\_____NEW_NN\\\\LLM\\\\MiniCPM-V\\\\finetune\\\\output\\\\phi3-vision-128k-instruct\\\\v0-20240629-080216/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=5, logging_nan_inf_filter=True, save_strategy=<IntervalStrategy.STEPS: 'steps'>, save_steps=300, save_total_limit=8, save_safetensors=True, save_on_each_node=True, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=None, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=50, dataloader_num_workers=0, dataloader_prefetch_factor=None, past_index=-1, run_name='D:\\\\_____NEW_NN\\\\LLM\\\\MiniCPM-V\\\\finetune\\\\output\\\\phi3-vision-128k-instruct\\\\v0-20240629-080216', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=None, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None), deepspeed=None, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['tensorboard'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint=None, hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=False, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs=None, include_inputs_for_metrics=False, eval_do_concat_batches=True, fp16_backend='auto', evaluation_strategy=None, push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=1800, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=False, include_num_input_tokens_seen=False, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, sortish_sampler=True, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=GenerationConfig {\n  \"do_sample\": true,\n  \"eos_token_id\": 32000,\n  \"max_new_tokens\": 2048,\n  \"pad_token_id\": 32000,\n  \"temperature\": 0.3,\n  \"top_k\": 20,\n  \"top_p\": 0.7\n}\n, train_sampler_random=True, push_hub_strategy='push_best', acc_strategy='token', additional_saved_files=[], metric_warmup_step=0, train_dataset_sample=1021)"
 }