2021-07-26 00:12:33,307 INFO MainThread:3000110 [wandb_setup.py:_flush():69] setting env: {} 2021-07-26 00:12:33,307 INFO MainThread:3000110 [wandb_setup.py:_flush():69] setting login settings: {} 2021-07-26 00:12:33,307 INFO MainThread:3000110 [wandb_init.py:_log_setup():337] Logging user logs to /var/hf/experiment-base-exp-512seq-stepwise/wandb/run-20210726_001233-17u6inbn/logs/debug.log 2021-07-26 00:12:33,307 INFO MainThread:3000110 [wandb_init.py:_log_setup():338] Logging internal logs to /var/hf/experiment-base-exp-512seq-stepwise/wandb/run-20210726_001233-17u6inbn/logs/debug-internal.log 2021-07-26 00:12:33,307 INFO MainThread:3000110 [wandb_init.py:init():370] calling init triggers 2021-07-26 00:12:33,307 INFO MainThread:3000110 [wandb_init.py:init():375] wandb.init called with sweep_config: {} config: {} 2021-07-26 00:12:33,307 INFO MainThread:3000110 [wandb_init.py:init():419] starting backend 2021-07-26 00:12:33,307 INFO MainThread:3000110 [backend.py:_multiprocessing_setup():70] multiprocessing start_methods=fork,spawn,forkserver, using: spawn 2021-07-26 00:12:33,351 INFO MainThread:3000110 [backend.py:ensure_launched():135] starting backend process... 2021-07-26 00:12:33,394 INFO MainThread:3000110 [backend.py:ensure_launched():139] started backend process with pid: 3001431 2021-07-26 00:12:33,396 INFO MainThread:3000110 [wandb_init.py:init():424] backend started and connected 2021-07-26 00:12:33,399 INFO MainThread:3000110 [wandb_init.py:init():472] updated telemetry 2021-07-26 00:12:33,400 INFO MainThread:3000110 [wandb_init.py:init():491] communicating current version 2021-07-26 00:12:34,050 INFO MainThread:3000110 [wandb_init.py:init():496] got version response upgrade_message: "wandb version 0.11.0 is available! To upgrade, please run:\n $ pip install wandb --upgrade" 2021-07-26 00:12:34,050 INFO MainThread:3000110 [wandb_init.py:init():504] communicating run to backend with 30 second timeout 2021-07-26 00:12:34,261 INFO MainThread:3000110 [wandb_init.py:init():529] starting run threads in backend 2021-07-26 00:12:35,502 INFO MainThread:3000110 [wandb_run.py:_console_start():1623] atexit reg 2021-07-26 00:12:35,502 INFO MainThread:3000110 [wandb_run.py:_redirect():1497] redirect: SettingsConsole.REDIRECT 2021-07-26 00:12:35,503 INFO MainThread:3000110 [wandb_run.py:_redirect():1502] Redirecting console. 2021-07-26 00:12:35,505 INFO MainThread:3000110 [wandb_run.py:_redirect():1558] Redirects installed. 2021-07-26 00:12:35,505 INFO MainThread:3000110 [wandb_init.py:init():554] run started, returning control to user process 2021-07-26 00:12:35,506 INFO MainThread:3000110 [wandb_run.py:_config_callback():872] config_cb None None {'output_dir': './outputs', 'overwrite_output_dir': True, 'do_train': False, 'do_eval': False, 'do_predict': False, 'evaluation_strategy': 'IntervalStrategy.NO', 'prediction_loss_only': False, 'per_device_train_batch_size': 48, 'per_device_eval_batch_size': 48, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 1, 'eval_accumulation_steps': None, 'learning_rate': 0.0006, 'weight_decay': 0.01, 'adam_beta1': 0.9, 'adam_beta2': 0.98, 'adam_epsilon': 1e-06, 'max_grad_norm': 1.0, 'num_train_epochs': 3.0, 'max_steps': -1, 'lr_scheduler_type': 'SchedulerType.LINEAR', 'warmup_ratio': 0.0, 'warmup_steps': 500, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './outputs/runs/Jul26_00-12-25_tablespoon', 'logging_strategy': 'IntervalStrategy.STEPS', 'logging_first_step': False, 'logging_steps': 500, 'save_strategy': 'IntervalStrategy.STEPS', 'save_steps': 1000, 'save_total_limit': 5, 'save_on_each_node': False, 'no_cuda': False, 'seed': 42, 'fp16': False, 'fp16_opt_level': 'O1', 'fp16_backend': 'auto', 'fp16_full_eval': False, 'local_rank': -1, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 1000, 'dataloader_num_workers': 0, 'past_index': -1, 'run_name': './outputs', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'sharded_ddp': [], 'deepspeed': None, 'label_smoothing_factor': 0.0, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['tensorboard', 'wandb'], 'ddp_find_unused_parameters': None, 'dataloader_pin_memory': True, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': None, 'push_to_hub_model_id': 'outputs', 'push_to_hub_organization': None, 'push_to_hub_token': None, 'mp_parameters': '', '_n_gpu': 0, '__cached__setup_devices': 'cpu'} 2021-07-26 00:12:35,507 INFO MainThread:3000110 [wandb_run.py:_config_callback():872] config_cb None None {'model_name_or_path': 'bertin-project/bertin-base-stepwise', 'model_type': 'roberta', 'config_name': './configs/base', 'tokenizer_name': './configs/base', 'cache_dir': None, 'use_fast_tokenizer': True, 'dtype': 'bfloat16'} 2021-07-26 00:12:35,508 INFO MainThread:3000110 [wandb_run.py:_config_callback():872] config_cb None None {'dataset_name': 'bertin-project/mc4-es-sampled', 'dataset_config_name': 'stepwise', 'train_file': None, 'validation_file': None, 'train_ref_file': None, 'validation_ref_file': None, 'overwrite_cache': False, 'validation_split_percentage': 5, 'max_seq_length': 512, 'preprocessing_num_workers': None, 'mlm_probability': 0.15, 'pad_to_max_length': True, 'line_by_line': False, 'text_column_name': 'text', 'shuffle_buffer_size': 10000, 'num_train_steps': 50000, 'num_eval_samples': 50000} 2021-07-26 00:12:35,587 INFO MainThread:3000110 [wandb_run.py:_tensorboard_callback():943] tensorboard callback: outputs, None