| 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=<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=<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=<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=<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 *** | |