2024-12-26 19:18:51.609 | INFO | __main__:setup_everything:71 - train_args:TrainingArguments( _n_gpu=1, accelerator_config={'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, adafactor=False, adam_beta1=0.9, adam_beta2=0.999, adam_epsilon=1e-08, auto_find_batch_size=False, average_tokens_across_devices=False, batch_eval_metrics=False, bf16=False, bf16_full_eval=False, data_seed=None, dataloader_drop_last=False, dataloader_num_workers=0, dataloader_persistent_workers=False, dataloader_pin_memory=True, dataloader_prefetch_factor=None, ddp_backend=None, ddp_broadcast_buffers=None, ddp_bucket_cap_mb=None, ddp_find_unused_parameters=None, ddp_timeout=1800, debug=[], deepspeed=./train_args/ds_z3_config.json, disable_tqdm=False, dispatch_batches=None, do_eval=False, do_predict=False, do_train=False, eval_accumulation_steps=None, eval_delay=0, eval_do_concat_batches=True, eval_on_start=False, eval_steps=None, eval_strategy=no, eval_use_gather_object=False, evaluation_strategy=None, fp16=True, fp16_backend=auto, fp16_full_eval=False, fp16_opt_level=O1, fsdp=[], fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_min_num_params=0, fsdp_transformer_layer_cls_to_wrap=None, full_determinism=False, gradient_accumulation_steps=16, gradient_checkpointing=False, gradient_checkpointing_kwargs=None, greater_is_better=None, group_by_length=False, half_precision_backend=auto, hub_always_push=False, hub_model_id=None, hub_private_repo=False, hub_strategy=every_save, hub_token=, ignore_data_skip=False, include_for_metrics=[], include_inputs_for_metrics=False, include_num_input_tokens_seen=False, include_tokens_per_second=False, jit_mode_eval=False, label_names=None, label_smoothing_factor=0.0, learning_rate=1e-05, length_column_name=length, load_best_model_at_end=False, local_rank=0, log_level=passive, log_level_replica=warning, log_on_each_node=True, logging_dir=output/firefly-qwen25-15b-sft-full-all-M-3/runs/Dec26_19-18-51_h800-2, logging_first_step=False, logging_nan_inf_filter=True, logging_steps=200, logging_strategy=steps, lr_scheduler_kwargs={}, lr_scheduler_type=cosine, max_grad_norm=1.0, max_steps=-1, metric_for_best_model=None, mp_parameters=, neftune_noise_alpha=None, no_cuda=False, num_train_epochs=3, optim=adamw_hf, optim_args=None, optim_target_modules=None, output_dir=output/firefly-qwen25-15b-sft-full-all-M-3, overwrite_output_dir=False, past_index=-1, per_device_eval_batch_size=8, per_device_train_batch_size=4, prediction_loss_only=False, push_to_hub=False, push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=, ray_scope=last, remove_unused_columns=False, report_to=['tensorboard'], restore_callback_states_from_checkpoint=False, resume_from_checkpoint=None, run_name=output/firefly-qwen25-15b-sft-full-all-M-3, save_on_each_node=False, save_only_model=False, save_safetensors=True, save_steps=200, save_strategy=steps, save_total_limit=1, seed=42, skip_memory_metrics=True, split_batches=None, tf32=None, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, torch_empty_cache_steps=None, torchdynamo=None, tpu_metrics_debug=False, tpu_num_cores=None, use_cpu=False, use_ipex=False, use_legacy_prediction_loop=False, use_liger_kernel=False, use_mps_device=False, warmup_ratio=0.0, warmup_steps=50, weight_decay=0, ) 2024-12-26 19:18:51.612 | INFO | __main__:init_components:383 - Initializing components... 2024-12-26 19:18:51.784 | INFO | __main__:setup_everything:71 - train_args:TrainingArguments( _n_gpu=1, accelerator_config={'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, adafactor=False, adam_beta1=0.9, adam_beta2=0.999, adam_epsilon=1e-08, auto_find_batch_size=False, average_tokens_across_devices=False, batch_eval_metrics=False, bf16=False, bf16_full_eval=False, data_seed=None, dataloader_drop_last=False, dataloader_num_workers=0, dataloader_persistent_workers=False, dataloader_pin_memory=True, dataloader_prefetch_factor=None, ddp_backend=None, ddp_broadcast_buffers=None, ddp_bucket_cap_mb=None, ddp_find_unused_parameters=None, ddp_timeout=1800, debug=[], deepspeed=./train_args/ds_z3_config.json, disable_tqdm=False, dispatch_batches=None, do_eval=False, do_predict=False, do_train=False, eval_accumulation_steps=None, eval_delay=0, eval_do_concat_batches=True, eval_on_start=False, eval_steps=None, eval_strategy=no, eval_use_gather_object=False, evaluation_strategy=None, fp16=True, fp16_backend=auto, fp16_full_eval=False, fp16_opt_level=O1, fsdp=[], fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_min_num_params=0, fsdp_transformer_layer_cls_to_wrap=None, full_determinism=False, gradient_accumulation_steps=16, gradient_checkpointing=False, gradient_checkpointing_kwargs=None, greater_is_better=None, group_by_length=False, half_precision_backend=auto, hub_always_push=False, hub_model_id=None, hub_private_repo=False, hub_strategy=every_save, hub_token=, ignore_data_skip=False, include_for_metrics=[], include_inputs_for_metrics=False, include_num_input_tokens_seen=False, include_tokens_per_second=False, jit_mode_eval=False, label_names=None, label_smoothing_factor=0.0, learning_rate=1e-05, length_column_name=length, load_best_model_at_end=False, local_rank=1, log_level=passive, log_level_replica=warning, log_on_each_node=True, logging_dir=output/firefly-qwen25-15b-sft-full-all-M-3/runs/Dec26_19-18-51_h800-2, logging_first_step=False, logging_nan_inf_filter=True, logging_steps=200, logging_strategy=steps, lr_scheduler_kwargs={}, lr_scheduler_type=cosine, max_grad_norm=1.0, max_steps=-1, metric_for_best_model=None, mp_parameters=, neftune_noise_alpha=None, no_cuda=False, num_train_epochs=3, optim=adamw_hf, optim_args=None, optim_target_modules=None, output_dir=output/firefly-qwen25-15b-sft-full-all-M-3, overwrite_output_dir=False, past_index=-1, per_device_eval_batch_size=8, per_device_train_batch_size=4, prediction_loss_only=False, push_to_hub=False, push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=, ray_scope=last, remove_unused_columns=False, report_to=['tensorboard'], restore_callback_states_from_checkpoint=False, resume_from_checkpoint=None, run_name=output/firefly-qwen25-15b-sft-full-all-M-3, save_on_each_node=False, save_only_model=False, save_safetensors=True, save_steps=200, save_strategy=steps, save_total_limit=1, seed=42, skip_memory_metrics=True, split_batches=None, tf32=None, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, torch_empty_cache_steps=None, torchdynamo=None, tpu_metrics_debug=False, tpu_num_cores=None, use_cpu=False, use_ipex=False, use_legacy_prediction_loop=False, use_liger_kernel=False, use_mps_device=False, warmup_ratio=0.0, warmup_steps=50, weight_decay=0, ) 2024-12-26 19:18:51.787 | INFO | __main__:init_components:383 - Initializing components... 2024-12-26 19:18:51.917 | INFO | __main__:load_tokenizer:231 - vocab_size of tokenizer: 151643 2024-12-26 19:18:51.917 | INFO | __main__:load_model:270 - Loading model from base model: /d1/zhaojihao/models/Qwen2.5-1.5B-Instruct 2024-12-26 19:18:51.917 | INFO | __main__:load_model:271 - Train model with full 2024-12-26 19:18:52.125 | INFO | __main__:load_tokenizer:231 - vocab_size of tokenizer: 151643 2024-12-26 19:18:52.126 | INFO | __main__:load_model:270 - Loading model from base model: /d1/zhaojihao/models/Qwen2.5-1.5B-Instruct 2024-12-26 19:18:52.126 | INFO | __main__:load_model:271 - Train model with full 2024-12-26 19:18:56.469 | INFO | __main__:load_model:345 - Total model params: 0.00M 2024-12-26 19:18:56.470 | INFO | __main__:init_components:402 - Train model with sft task 2024-12-26 19:18:56.470 | INFO | __main__:load_sft_dataset:365 - Loading data with UnifiedSFTDataset 2024-12-26 19:18:56.470 | INFO | component.dataset:__init__:19 - Loading data: /d1/zhaojihao/chunk2/train_GPT/conversations_20000_MASK.jsonl 2024-12-26 19:18:56.557 | INFO | __main__:load_model:345 - Total model params: 0.00M 2024-12-26 19:18:56.557 | INFO | __main__:init_components:402 - Train model with sft task 2024-12-26 19:18:56.557 | INFO | __main__:load_sft_dataset:365 - Loading data with UnifiedSFTDataset 2024-12-26 19:18:56.557 | INFO | component.dataset:__init__:19 - Loading data: /d1/zhaojihao/chunk2/train_GPT/conversations_20000_MASK.jsonl 2024-12-26 19:18:56.672 | INFO | component.dataset:__init__:22 - Use template "qwen" for training 2024-12-26 19:18:56.673 | INFO | component.dataset:__init__:23 - There are 20000 data in dataset 2024-12-26 19:18:56.729 | INFO | __main__:main:440 - *** starting training *** 2024-12-26 19:18:56.766 | INFO | component.dataset:__init__:22 - Use template "qwen" for training 2024-12-26 19:18:56.766 | INFO | component.dataset:__init__:23 - There are 20000 data in dataset 2024-12-26 19:18:56.819 | INFO | __main__:main:440 - *** starting training ***