| ======================== | |
| START TIME: Thu Jul 4 22:42:58 UTC 2024 | |
| python3 version = Python 3.10.14 | |
| ======================== | |
| The token has not been saved to the git credentials helper. Pass `add_to_git_credential=True` in this function directly or `--add-to-git-credential` if using via `huggingface-cli` if you want to set the git credential as well. | |
| Token is valid (permission: write). | |
| Your token has been saved to /admin/home/ferdinand_mom/.cache/huggingface/token | |
| Login successful | |
| fatal: Unable to create '/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/.git/index.lock': File exists. | |
| Another git process seems to be running in this repository, e.g. | |
| an editor opened by 'git commit'. Please make sure all processes | |
| are terminated then try again. If it still fails, a git process | |
| may have crashed in this repository earlier: | |
| remove the file manually to continue. | |
| Job status: RUNNING | |
| [2024-07-04 22:43:06,816] torch.distributed.run: [WARNING] | |
| [2024-07-04 22:43:06,816] torch.distributed.run: [WARNING] ***************************************** | |
| [2024-07-04 22:43:06,816] torch.distributed.run: [WARNING] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. | |
| [2024-07-04 22:43:06,816] torch.distributed.run: [WARNING] ***************************************** | |
| [default0]:07/04/2024 22:43:25 [WARNING|DP=0|PP=0|TP=0|ip-26-0-171-230]: [Vocab Size Padding] Padded vocab (size: 50257) with 7 dummy tokens (new size: 50264) | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Config: | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Config(general=GeneralArgs(project='bench_cluster', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: run='%date_%jobid', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: seed=42, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: step=None, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: consumed_train_samples=None, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: benchmark_csv_path=None, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: ignore_sanity_checks=True), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: parallelism=ParallelismArgs(dp=1, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: pp=1, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: tp=8, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: pp_engine=<nanotron.parallel.pipeline_parallel.engine.OneForwardOneBackwardPipelineEngine object at 0x7fe0027d48e0>, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: tp_mode=<TensorParallelLinearMode.REDUCE_SCATTER: 2>, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: tp_linear_async_communication=False, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: expert_parallel_size=1), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: model=ModelArgs(model_config=LlamaConfig(bos_token_id=1, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: eos_token_id=2, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: hidden_act='silu', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: hidden_size=2048, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: initializer_range=0.02, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: intermediate_size=4096, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: is_llama_config=True, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: max_position_embeddings=4096, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: num_attention_heads=32, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: num_hidden_layers=24, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: num_key_value_heads=32, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: pad_token_id=None, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: pretraining_tp=1, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: rms_norm_eps=1e-05, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: rope_scaling=None, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: rope_theta=10000.0, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: tie_word_embeddings=True, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: use_cache=True, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: vocab_size=50264), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: init_method=RandomInit(std=0.025), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: dtype=torch.bfloat16, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: make_vocab_size_divisible_by=1, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: ddp_bucket_cap_mb=25), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: tokenizer=TokenizerArgs(tokenizer_name_or_path='openai-community/gpt2', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: tokenizer_revision=None, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: tokenizer_max_length=None), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: checkpoints=CheckpointsArgs(checkpoints_path=PosixPath('/dev/null'), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: checkpoint_interval=100000, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: save_initial_state=False, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: resume_checkpoint_path=None, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: checkpoints_path_is_shared_file_system=False), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: logging=LoggingArgs(log_level='info', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: log_level_replica='info', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: iteration_step_info_interval=1), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: tokens=TokensArgs(sequence_length=4096, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: train_steps=20, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: micro_batch_size=2, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: batch_accumulation_per_replica=512, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: val_check_interval=-1, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: limit_val_batches=0, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: limit_test_batches=0), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: optimizer=OptimizerArgs(optimizer_factory=AdamWOptimizerArgs(adam_eps=1e-08, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: adam_beta1=0.9, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: adam_beta2=0.95, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: torch_adam_is_fused=True, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: name='adamW'), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: zero_stage=1, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: weight_decay=0.01, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: clip_grad=1.0, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: accumulate_grad_in_fp32=True, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: learning_rate_scheduler=LRSchedulerArgs(learning_rate=0.0001, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: lr_warmup_steps=1, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: lr_warmup_style='linear', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: lr_decay_style='linear', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: lr_decay_steps=19, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: lr_decay_starting_step=None, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: min_decay_lr=1e-05)), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: data_stages=[DatasetStageArgs(name='Training Stage', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: start_training_step=1, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: data=DataArgs(dataset=PretrainDatasetsArgs(hf_dataset_or_datasets='roneneldan/TinyStories', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: hf_dataset_splits='train', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: hf_dataset_config_name=None, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: dataset_processing_num_proc_per_process=64, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: dataset_overwrite_cache=False, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: text_column_name='text'), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: seed=42, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: num_loading_workers=0))], | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: profiler=ProfilerArgs(profiler_export_path=PosixPath('/fsx/ferdinandmom/ferdinand-hf/bench_cluster/results/llama-1B/8_GPUS/dp-1_tp-8_pp-1_mbz-2')), | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: lighteval=None) | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Model Config: | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: LlamaConfig(bos_token_id=1, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: eos_token_id=2, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: hidden_act='silu', | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: hidden_size=2048, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: initializer_range=0.02, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: intermediate_size=4096, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: is_llama_config=True, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: max_position_embeddings=4096, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: num_attention_heads=32, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: num_hidden_layers=24, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: num_key_value_heads=32, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: pad_token_id=None, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: pretraining_tp=1, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: rms_norm_eps=1e-05, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: rope_scaling=None, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: rope_theta=10000.0, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: tie_word_embeddings=True, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: use_cache=True, | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: vocab_size=50264) | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Building model.. | |
| [default0]:07/04/2024 22:43:25 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Setting PP block ranks... | |
| [default3]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=3|ip-26-0-171-230]: Local number of parameters: 139M (264.73MiB) | |
| [default3]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=3|ip-26-0-171-230]: [After model building] Memory usage: 290.76MiB. Peak allocated: 317.33MiB Peak reserved: 324.00MiB | |
| [default3]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=3|ip-26-0-171-230]: No checkpoint path provided. | |
| [default1]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=1|ip-26-0-171-230]: Local number of parameters: 139M (264.73MiB) | |
| [default1]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=1|ip-26-0-171-230]: [After model building] Memory usage: 290.76MiB. Peak allocated: 317.33MiB Peak reserved: 324.00MiB | |
| [default1]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=1|ip-26-0-171-230]: No checkpoint path provided. | |
| [default0]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Total number of parameters: 1.11G (2117.88MiB) | |
| [default5]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=5|ip-26-0-171-230]: Local number of parameters: 139M (264.73MiB) | |
| [default5]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=5|ip-26-0-171-230]: [After model building] Memory usage: 290.76MiB. Peak allocated: 317.33MiB Peak reserved: 324.00MiB | |
| [default7]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=7|ip-26-0-171-230]: Local number of parameters: 139M (264.73MiB) | |
| [default7]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=7|ip-26-0-171-230]: [After model building] Memory usage: 290.76MiB. Peak allocated: 317.33MiB Peak reserved: 324.00MiB | |
| [default7]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=7|ip-26-0-171-230]: No checkpoint path provided. | |
| [default5]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=5|ip-26-0-171-230]: No checkpoint path provided. | |
| [default0]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Local number of parameters: 139M (264.73MiB) | |
| [default0]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: [After model building] Memory usage: 290.76MiB. Peak allocated: 317.33MiB Peak reserved: 324.00MiB | |
| [default0]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: No checkpoint path provided. | |
| [default0]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Parametrizing model parameters using StandardParametrizator | |
| [default0]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: [Optimizer Building] Using LearningRateForSP as learning rate | |
| [default0]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: [ZeRO sharding] Size of optimizer params per rank: | |
| [default0]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: [ZeRO sharding] DP Rank 0 has 139M out of 139M (100.00%) params' optimizer states | |
| [default4]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=4|ip-26-0-171-230]: Local number of parameters: 139M (264.73MiB) | |
| [default4]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=4|ip-26-0-171-230]: [After model building] Memory usage: 290.76MiB. Peak allocated: 317.33MiB Peak reserved: 324.00MiB | |
| [default4]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=4|ip-26-0-171-230]: No checkpoint path provided. | |
| [default6]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=6|ip-26-0-171-230]: Local number of parameters: 139M (264.73MiB) | |
| [default6]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=6|ip-26-0-171-230]: [After model building] Memory usage: 290.76MiB. Peak allocated: 317.33MiB Peak reserved: 324.00MiB | |
| [default6]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=6|ip-26-0-171-230]: No checkpoint path provided. | |
| [default2]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=2|ip-26-0-171-230]: Local number of parameters: 139M (264.73MiB) | |
| [default2]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=2|ip-26-0-171-230]: [After model building] Memory usage: 290.76MiB. Peak allocated: 317.33MiB Peak reserved: 324.00MiB | |
| [default2]:07/04/2024 22:43:40 [INFO|DP=0|PP=0|TP=2|ip-26-0-171-230]: No checkpoint path provided. | |
| [default0]:07/04/2024 22:43:42 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: [Training Plan] Stage Training Stage has 19 remaining training steps and has consumed 0 samples | |
| [default0]:07/04/2024 22:43:42 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Using `datasets` library | |
| [default0]:07/04/2024 22:43:42 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Loading tokenizer from openai-community/gpt2 and transformers/hf_hub versions ('4.41.2', '0.23.4') | |
| [default0]:07/04/2024 22:43:42 [WARNING|DP=0|PP=0|TP=0|ip-26-0-171-230]: Repo card metadata block was not found. Setting CardData to empty. | |
| [default0]:Repo card metadata block was not found. Setting CardData to empty. | |
| [default0]:07/04/2024 22:43:45 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: [Training Plan] There are 1 training stages | |
| [default0]:07/04/2024 22:43:45 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: [Stage Training Stage] start from step 1 | |
| [default0]:07/04/2024 22:43:45 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: | |
| [default0]:07/04/2024 22:43:45 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: [Start training] datetime: 2024-07-04 22:43:45.787051 | mbs: 2 | grad_accum: 512 | global_batch_size: 1024 | sequence_length: 4096 | train_steps: 20 | start_iteration_step: 0 | consumed_train_samples: 0 | |
| [default4]:Repo card metadata block was not found. Setting CardData to empty. | |
| [default7]:07/04/2024 22:43:45 [WARNING|DP=0|PP=0|TP=7|ip-26-0-171-230]: Repo card metadata block was not found. Setting CardData to empty. | |
| [default4]:07/04/2024 22:43:45 [WARNING|DP=0|PP=0|TP=4|ip-26-0-171-230]: Repo card metadata block was not found. Setting CardData to empty. | |
| [default6]:07/04/2024 22:43:45 [WARNING|DP=0|PP=0|TP=6|ip-26-0-171-230]: Repo card metadata block was not found. Setting CardData to empty. | |
| [default7]:Repo card metadata block was not found. Setting CardData to empty. | |
| [default6]:Repo card metadata block was not found. Setting CardData to empty. | |
| [default3]:07/04/2024 22:43:45 [WARNING|DP=0|PP=0|TP=3|ip-26-0-171-230]: Repo card metadata block was not found. Setting CardData to empty. | |
| [default3]:Repo card metadata block was not found. Setting CardData to empty. | |
| [default1]:07/04/2024 22:43:45 [WARNING|DP=0|PP=0|TP=1|ip-26-0-171-230]: Repo card metadata block was not found. Setting CardData to empty. | |
| [default5]:07/04/2024 22:43:45 [WARNING|DP=0|PP=0|TP=5|ip-26-0-171-230]: Repo card metadata block was not found. Setting CardData to empty. | |
| [default1]:Repo card metadata block was not found. Setting CardData to empty. | |
| [default2]:07/04/2024 22:43:46 [WARNING|DP=0|PP=0|TP=2|ip-26-0-171-230]: Repo card metadata block was not found. Setting CardData to empty. | |
| [default2]:Repo card metadata block was not found. Setting CardData to empty. | |
| [default5]:Repo card metadata block was not found. Setting CardData to empty. | |
| [default0]:07/04/2024 22:43:51 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Resuming training from stage Training Stage, it has trained for 0 samples and has 19 remaining train steps | |
| [default0]:07/04/2024 22:43:51 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Memory usage: 1350.75MiB. Peak allocated 1350.76MiB. Peak reserved: 1384.00MiB | |
| [default4]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/autograd/__init__.py:266: UserWarning: c10d::allreduce_: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at ../torch/csrc/autograd/autograd_not_implemented_fallback.cpp:63.) | |
| [default4]: Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass | |
| [default3]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/autograd/__init__.py:266: UserWarning: c10d::allreduce_: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at ../torch/csrc/autograd/autograd_not_implemented_fallback.cpp:63.) | |
| [default3]: Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass | |
| [default0]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/autograd/__init__.py:266: UserWarning: c10d::allreduce_: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at ../torch/csrc/autograd/autograd_not_implemented_fallback.cpp:63.) | |
| [default0]: Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass | |
| [default1]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/autograd/__init__.py:266: UserWarning: c10d::allreduce_: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at ../torch/csrc/autograd/autograd_not_implemented_fallback.cpp:63.) | |
| [default1]: Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass | |
| [default2]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/autograd/__init__.py:266: UserWarning: c10d::allreduce_: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at ../torch/csrc/autograd/autograd_not_implemented_fallback.cpp:63.) | |
| [default2]: Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass | |
| [default7]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/autograd/__init__.py:266: UserWarning: c10d::allreduce_: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at ../torch/csrc/autograd/autograd_not_implemented_fallback.cpp:63.) | |
| [default5]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/autograd/__init__.py:266: UserWarning: c10d::allreduce_: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at ../torch/csrc/autograd/autograd_not_implemented_fallback.cpp:63.) | |
| [default5]: Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass | |
| [default7]: Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass | |
| [default6]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/autograd/__init__.py:266: UserWarning: c10d::allreduce_: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at ../torch/csrc/autograd/autograd_not_implemented_fallback.cpp:63.) | |
| [default6]: Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass | |
| [default0]:07/04/2024 22:45:10 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Memory usage: 1427.15MiB. Peak allocated 5028.02MiB. Peak reserved: 5396.00MiB | |
| [default0]:07/04/2024 22:45:10 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: iteration: 1 / 20 | consumed_tokens: 4.19M | elapsed_time_per_iteration_ms: 78.6K | tokens_per_sec: 53.3K | tokens_per_sec_per_gpu: 6.67K | global_batch_size: 1.02K | lm_loss: 11.5 | lr: 0.0001 | model_tflops_per_gpu: 60.5 | hardware_tflops_per_gpu: 60.5 | grad_norm: 15.7 | cuda_memory_allocated: 2.61G | cuda_max_memory_reserved: 5.66G | hd_total_memory_tb: 312G | hd_used_memory_tb: 81.6G | hd_free_memory_tb: 231G | |
| [default0]:07/04/2024 22:45:10 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Memory usage: 2486.16MiB. Peak allocated 2486.16MiB. Peak reserved: 5396.00MiB | |
| [default0]:07/04/2024 22:46:21 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Memory usage: 2486.41MiB. Peak allocated 6087.28MiB. Peak reserved: 6248.00MiB | |
| [default0]:07/04/2024 22:46:22 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: iteration: 2 / 20 | consumed_tokens: 8.39M | elapsed_time_per_iteration_ms: 71.7K | tokens_per_sec: 58.5K | tokens_per_sec_per_gpu: 7.32K | global_batch_size: 1.02K | lm_loss: 11.5 | lr: 9.53e-05 | model_tflops_per_gpu: 66.4 | hardware_tflops_per_gpu: 66.4 | grad_norm: 16 | cuda_memory_allocated: 2.61G | cuda_max_memory_reserved: 6.55G | hd_total_memory_tb: 312G | hd_used_memory_tb: 81.6G | hd_free_memory_tb: 231G | |
| [default0]:07/04/2024 22:46:22 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Memory usage: 2486.16MiB. Peak allocated 2486.48MiB. Peak reserved: 6248.00MiB | |
| [default0]:07/04/2024 22:47:44 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Memory usage: 2486.41MiB. Peak allocated 6087.28MiB. Peak reserved: 6248.00MiB | |
| [default0]:STAGE:2024-07-04 22:47:44 2948075:2948075 ActivityProfilerController.cpp:314] Completed Stage: Warm Up | |
| [default0]:07/04/2024 22:47:44 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: iteration: 3 / 20 | consumed_tokens: 12.6M | elapsed_time_per_iteration_ms: 82.7K | tokens_per_sec: 50.7K | tokens_per_sec_per_gpu: 6.34K | global_batch_size: 1.02K | lm_loss: 12.8 | lr: 9.05e-05 | model_tflops_per_gpu: 57.5 | hardware_tflops_per_gpu: 57.5 | grad_norm: 137 | cuda_memory_allocated: 2.61G | cuda_max_memory_reserved: 6.55G | hd_total_memory_tb: 312G | hd_used_memory_tb: 81.6G | hd_free_memory_tb: 231G | |
| [default0]:07/04/2024 22:47:44 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Memory usage: 2486.16MiB. Peak allocated 2486.48MiB. Peak reserved: 6248.00MiB | |
| [default0]:07/04/2024 22:49:36 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Memory usage: 2486.41MiB. Peak allocated 6087.28MiB. Peak reserved: 6248.00MiB | |
| [default0]:07/04/2024 22:49:36 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: iteration: 4 / 20 | consumed_tokens: 16.8M | elapsed_time_per_iteration_ms: 111K | tokens_per_sec: 37.7K | tokens_per_sec_per_gpu: 4.71K | global_batch_size: 1.02K | lm_loss: 12.2 | lr: 8.58e-05 | model_tflops_per_gpu: 42.7 | hardware_tflops_per_gpu: 42.7 | grad_norm: 22.4 | cuda_memory_allocated: 2.61G | cuda_max_memory_reserved: 6.55G | hd_total_memory_tb: 312G | hd_used_memory_tb: 81.6G | hd_free_memory_tb: 231G | |
| [default0]:07/04/2024 22:49:36 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Memory usage: 2486.16MiB. Peak allocated 2486.48MiB. Peak reserved: 6248.00MiB | |
| [default0]:07/04/2024 22:51:29 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: iteration: 5 / 20 | consumed_tokens: 21M | elapsed_time_per_iteration_ms: 113K | tokens_per_sec: 37.2K | tokens_per_sec_per_gpu: 4.64K | global_batch_size: 1.02K | lm_loss: 12.4 | lr: 8.11e-05 | model_tflops_per_gpu: 42.1 | hardware_tflops_per_gpu: 42.1 | grad_norm: 42.8 | |
| [default0]:07/04/2024 22:51:29 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: Memory usage: 2486.16MiB. Peak allocated 6087.28MiB. Peak reserved: 6248.00MiB | |
| [default0]:07/04/2024 22:53:21 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-230]: iteration: 6 / 20 | consumed_tokens: 25.2M | elapsed_time_per_iteration_ms: 113K | tokens_per_sec: 37.2K | tokens_per_sec_per_gpu: 4.65K | global_batch_size: 1.02K | lm_loss: 11.1 | lr: 7.63e-05 | model_tflops_per_gpu: 42.2 | hardware_tflops_per_gpu: 42.2 | grad_norm: 24.8 | |
| [default0]:STAGE:2024-07-04 22:58:36 2948075:2948075 ActivityProfilerController.cpp:320] Completed Stage: Collection | |
| [default0]:STAGE:2024-07-04 22:59:24 2948075:2948075 ActivityProfilerController.cpp:324] Completed Stage: Post Processing | |
| [default3]:[rank3]:[E ProcessGroupNCCL.cpp:523] [Rank 3] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=26, OpType=ALLREDUCE, NumelIn=1, NumelOut=1, Timeout(ms)=1800000) ran for 1800394 milliseconds before timing out. | |
| [default7]:[rank7]:[E ProcessGroupNCCL.cpp:523] [Rank 7] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=26, OpType=ALLREDUCE, NumelIn=1, NumelOut=1, Timeout(ms)=1800000) ran for 1800363 milliseconds before timing out. | |
| [default4]:[rank4]:[E ProcessGroupNCCL.cpp:523] [Rank 4] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=26, OpType=ALLREDUCE, NumelIn=1, NumelOut=1, Timeout(ms)=1800000) ran for 1800361 milliseconds before timing out. | |
| [default2]:[rank2]:[E ProcessGroupNCCL.cpp:523] [Rank 2] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=26, OpType=ALLREDUCE, NumelIn=1, NumelOut=1, Timeout(ms)=1800000) ran for 1800379 milliseconds before timing out. | |
| [default6]:[rank6]:[E ProcessGroupNCCL.cpp:523] [Rank 6] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=26, OpType=ALLREDUCE, NumelIn=1, NumelOut=1, Timeout(ms)=1800000) ran for 1800356 milliseconds before timing out. | |
| [default5]:[rank5]:[E ProcessGroupNCCL.cpp:523] [Rank 5] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=26, OpType=ALLREDUCE, NumelIn=1, NumelOut=1, Timeout(ms)=1800000) ran for 1800379 milliseconds before timing out. | |
| [default1]:[rank1]:[E ProcessGroupNCCL.cpp:523] [Rank 1] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=26, OpType=ALLREDUCE, NumelIn=1, NumelOut=1, Timeout(ms)=1800000) ran for 1800378 milliseconds before timing out. | |
| [default5]:[rank5]:[E ProcessGroupNCCL.cpp:537] Some NCCL operations have failed or timed out. Due to the asynchronous nature of CUDA kernels, subsequent GPU operations might run on corrupted/incomplete data. | |
| [default5]:[rank5]:[E ProcessGroupNCCL.cpp:543] To avoid data inconsistency, we are taking the entire process down. | |
| [default5]:[rank5]:[E ProcessGroupNCCL.cpp:1182] [Rank 5] NCCL watchdog thread terminated with exception: [Rank 5] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=26, OpType=ALLREDUCE, NumelIn=1, NumelOut=1, Timeout(ms)=1800000) ran for 1800379 milliseconds before timing out. | |
| [default5]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:525 (most recent call first): | |
| [default5]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f9cfdf2cd87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) | |
| [default5]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1e6 (0x7f9cff0d46e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) | |
| [default5]:frame #2: c10d::ProcessGroupNCCL::workCleanupLoop() + 0x19d (0x7f9cff0d7c3d in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) | |
| [default5]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x119 (0x7f9cff0d8839 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) | |
| [default5]:frame #4: <unknown function> + 0xd3e95 (0x7f9d48ddce95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6) | |
| [default5]:frame #5: <unknown function> + 0x8609 (0x7f9d4dee4609 in /lib/x86_64-linux-gnu/libpthread.so.0) | |
| [default5]:frame #6: clone + 0x43 (0x7f9d4dcaf353 in /lib/x86_64-linux-gnu/libc.so.6) | |
| [default5]: | |
| [default5]:terminate called after throwing an instance of 'c10::DistBackendError' | |
| [default5]: what(): [Rank 5] NCCL watchdog thread terminated with exception: [Rank 5] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=26, OpType=ALLREDUCE, NumelIn=1, NumelOut=1, Timeout(ms)=1800000) ran for 1800379 milliseconds before timing out. | |
| [default5]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:525 (most recent call first): | |
| [default5]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f9cfdf2cd87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) | |
| [default5]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1e6 (0x7f9cff0d46e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) | |
| [default5]:frame #2: c10d::ProcessGroupNCCL::workCleanupLoop() + 0x19d (0x7f9cff0d7c3d in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) | |
| [default5]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x119 (0x7f9cff0d8839 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) | |
| [default5]:frame #4: <unknown function> + 0xd3e95 (0x7f9d48ddce95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6) | |
| [default5]:frame #5: <unknown function> + 0x8609 (0x7f9d4dee4609 in /lib/x86_64-linux-gnu/libpthread.so.0) | |
| [default5]:frame #6: clone + 0x43 (0x7f9d4dcaf353 in /lib/x86_64-linux-gnu/libc.so.6) | |
| [default5]: | |
| [default5]:Exception raised from ncclCommWatchdog at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1186 (most recent call first): | |
| [default5]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f9cfdf2cd87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) | |
| [default5]:frame #1: <unknown function> + 0xdf6b11 (0x7f9cfee2eb11 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) | |
| [default5]:frame #2: <unknown function> + 0xd3e95 (0x7f9d48ddce95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6) | |
| [default5]:frame #3: <unknown function> + 0x8609 (0x7f9d4dee4609 in /lib/x86_64-linux-gnu/libpthread.so.0) | |
| [default5]:frame #4: clone + 0x43 (0x7f9d4dcaf353 in /lib/x86_64-linux-gnu/libc.so.6) | |
| [default5]: | |
| [2024-07-04 23:23:24,554] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 2948075 closing signal SIGTERM | |
| [2024-07-04 23:23:24,555] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 2948076 closing signal SIGTERM | |
| [2024-07-04 23:23:24,555] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 2948077 closing signal SIGTERM | |
| [2024-07-04 23:23:24,556] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 2948078 closing signal SIGTERM | |
| [2024-07-04 23:23:24,556] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 2948079 closing signal SIGTERM | |
| [2024-07-04 23:23:24,556] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 2948081 closing signal SIGTERM | |
| [2024-07-04 23:23:24,558] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 2948082 closing signal SIGTERM | |
| [2024-07-04 23:23:36,683] torch.distributed.elastic.multiprocessing.api: [ERROR] failed (exitcode: -6) local_rank: 5 (pid: 2948080) of binary: /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10 | |
| Traceback (most recent call last): | |
| File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/torchrun", line 8, in <module> | |
| sys.exit(main()) | |
| File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 347, in wrapper | |
| return f(*args, **kwargs) | |
| File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/run.py", line 812, in main | |
| run(args) | |
| File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/run.py", line 803, in run | |
| elastic_launch( | |
| File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 135, in __call__ | |
| return launch_agent(self._config, self._entrypoint, list(args)) | |
| File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 268, in launch_agent | |
| raise ChildFailedError( | |
| torch.distributed.elastic.multiprocessing.errors.ChildFailedError: | |
| ============================================================ | |
| /fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py FAILED | |
| ------------------------------------------------------------ | |
| Failures: | |
| <NO_OTHER_FAILURES> | |
| ------------------------------------------------------------ | |
| Root Cause (first observed failure): | |
| [0]: | |
| time : 2024-07-04_23:23:24 | |
| host : ip-26-0-171-230.ec2.internal | |
| rank : 5 (local_rank: 5) | |
| exitcode : -6 (pid: 2948080) | |
| error_file: <N/A> | |
| traceback : Signal 6 (SIGABRT) received by PID 2948080 | |
| ============================================================ | |
| srun: error: ip-26-0-171-230: task 0: Exited with exit code 1 | |
| Consider using `hf_transfer` for faster uploads. This solution comes with some limitations. See https://huggingface.co/docs/huggingface_hub/hf_transfer for more details. | |