| 2025-05-27 14:45:36,910 INFO MainThread:1815479 [wandb_setup.py:_flush():70] Current SDK version is 0.19.11 |
| 2025-05-27 14:45:36,911 INFO MainThread:1815479 [wandb_setup.py:_flush():70] Configure stats pid to 1815479 |
| 2025-05-27 14:45:36,911 INFO MainThread:1815479 [wandb_setup.py:_flush():70] Loading settings from /home/hansirui_1st/.config/wandb/settings |
| 2025-05-27 14:45:36,911 INFO MainThread:1815479 [wandb_setup.py:_flush():70] Loading settings from /home/hansirui_1st/jiayi/resist/setting3/scripts/wandb/settings |
| 2025-05-27 14:45:36,911 INFO MainThread:1815479 [wandb_setup.py:_flush():70] Loading settings from environment variables |
| 2025-05-27 14:45:36,911 INFO MainThread:1815479 [wandb_init.py:setup_run_log_directory():724] Logging user logs to /aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-2T/tinyllama-2T-s3-Q1-2000/wandb/run-20250527_144536-79mv42w3/logs/debug.log |
| 2025-05-27 14:45:36,911 INFO MainThread:1815479 [wandb_init.py:setup_run_log_directory():725] Logging internal logs to /aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-2T/tinyllama-2T-s3-Q1-2000/wandb/run-20250527_144536-79mv42w3/logs/debug-internal.log |
| 2025-05-27 14:45:36,911 INFO MainThread:1815479 [wandb_init.py:init():852] calling init triggers |
| 2025-05-27 14:45:36,911 INFO MainThread:1815479 [wandb_init.py:init():857] wandb.init called with sweep_config: {} |
| config: {'model_name_or_path': '/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-955k-token-2T', 'max_length': 512, 'trust_remote_code': True, 'train_datasets': [('inverse-json', {'proportion': 1.0, 'path': '/home/hansirui_1st/jiayi/resist/imdb_data/train/pos/2000/train.json'})], 'eval_datasets': None, 'epochs': 1, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 4, 'gradient_accumulation_steps': 8, 'gradient_checkpointing': True, 'lr': 1e-05, 'lr_scheduler_type': <SchedulerType.CONSTANT: 'constant'>, 'lr_warmup_ratio': 0.0, 'weight_decay': 0.0, 'seed': 42, 'fp16': False, 'bf16': True, 'tf32': True, 'eval_strategy': 'epoch', 'eval_interval': 1000000, 'need_eval': False, 'eval_split_ratio': None, 'output_dir': '/aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-2T/tinyllama-2T-s3-Q1-2000', 'log_type': 'wandb', 'log_dir': '/aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-2T/tinyllama-2T-s3-Q1-2000', 'log_project': 'Inverse_Alignment_IMDb', 'log_run_name': 'imdb-tinyllama-2T-s3-Q1-2000', 'save_16bit': True, 'save_interval': 1000000, 'local_rank': 0, 'zero_stage': 3, 'offload': 'none', 'deepspeed': False, 'deepspeed_config': None, 'deepscale': False, 'deepscale_config': None, 'global_rank': 0, 'device': device(type='cuda', index=0), 'num_update_steps_per_epoch': 32, 'total_training_steps': 32, '_wandb': {}} |
| 2025-05-27 14:45:36,911 INFO MainThread:1815479 [wandb_init.py:init():893] starting backend |
| 2025-05-27 14:45:36,911 INFO MainThread:1815479 [wandb_init.py:init():897] sending inform_init request |
| 2025-05-27 14:45:36,915 INFO MainThread:1815479 [backend.py:_multiprocessing_setup():101] multiprocessing start_methods=fork,spawn,forkserver, using: spawn |
| 2025-05-27 14:45:36,915 INFO MainThread:1815479 [wandb_init.py:init():907] backend started and connected |
| 2025-05-27 14:45:36,917 INFO MainThread:1815479 [wandb_init.py:init():1005] updated telemetry |
| 2025-05-27 14:45:36,917 INFO MainThread:1815479 [wandb_init.py:init():1029] communicating run to backend with 90.0 second timeout |
| 2025-05-27 14:45:37,569 INFO MainThread:1815479 [wandb_init.py:init():1104] starting run threads in backend |
| 2025-05-27 14:45:37,782 INFO MainThread:1815479 [wandb_run.py:_console_start():2573] atexit reg |
| 2025-05-27 14:45:37,783 INFO MainThread:1815479 [wandb_run.py:_redirect():2421] redirect: wrap_raw |
| 2025-05-27 14:45:37,783 INFO MainThread:1815479 [wandb_run.py:_redirect():2490] Wrapping output streams. |
| 2025-05-27 14:45:37,783 INFO MainThread:1815479 [wandb_run.py:_redirect():2513] Redirects installed. |
| 2025-05-27 14:45:37,785 INFO MainThread:1815479 [wandb_init.py:init():1150] run started, returning control to user process |
|
|