llama-1B/8_GPUS/dp-1_tp-1_pp-8_mbz-1/log.out

========================
START TIME: Thu Jul  4 00:02:39 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
Already on 'bench_cluster'
M	examples/config_tiny_llama.py
M	examples/config_tiny_llama.yaml
M	examples/train_tiny_llama.sh
M	src/nanotron/models/llama.py
M	src/nanotron/trainer.py
Your branch is up to date with 'origin/bench_cluster'.
Job status: RUNNING
W0704 00:02:42.152000 140557240698688 torch/distributed/run.py:757] 
W0704 00:02:42.152000 140557240698688 torch/distributed/run.py:757] *****************************************
W0704 00:02:42.152000 140557240698688 torch/distributed/run.py:757] 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. 
W0704 00:02:42.152000 140557240698688 torch/distributed/run.py:757] *****************************************
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: Config:
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: Config(general=GeneralArgs(project='bench_cluster',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                            run='%date_%jobid',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                            seed=42,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                            step=None,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                            consumed_train_samples=None,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                            benchmark_csv_path=None,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                            ignore_sanity_checks=True),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:        parallelism=ParallelismArgs(dp=1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                    pp=8,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                    tp=1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                    pp_engine=<nanotron.parallel.pipeline_parallel.engine.OneForwardOneBackwardPipelineEngine object at 0x7fcd3b5cc700>,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                    tp_mode=<TensorParallelLinearMode.REDUCE_SCATTER: 2>,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                    tp_linear_async_communication=False,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                    expert_parallel_size=1),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:        model=ModelArgs(model_config=LlamaConfig(bos_token_id=1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 eos_token_id=2,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 hidden_act='silu',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 hidden_size=2048,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 initializer_range=0.02,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 intermediate_size=4096,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 is_llama_config=True,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 max_position_embeddings=4096,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 num_attention_heads=32,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 num_hidden_layers=24,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 num_key_value_heads=32,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 pad_token_id=None,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 pretraining_tp=1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 rms_norm_eps=1e-05,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 rope_scaling=None,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 rope_theta=10000.0,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 tie_word_embeddings=True,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 use_cache=True,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                 vocab_size=50257),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                        init_method=RandomInit(std=0.025),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                        dtype=torch.bfloat16,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                        make_vocab_size_divisible_by=1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                        ddp_bucket_cap_mb=25),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:        tokenizer=TokenizerArgs(tokenizer_name_or_path='openai-community/gpt2',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                tokenizer_revision=None,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                tokenizer_max_length=None),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:        checkpoints=CheckpointsArgs(checkpoints_path=Path('/dev/null'),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                    checkpoint_interval=100000,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                    save_initial_state=False,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                    resume_checkpoint_path=None,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                    checkpoints_path_is_shared_file_system=False),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:        logging=LoggingArgs(log_level='info',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                            log_level_replica='info',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                            iteration_step_info_interval=1),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:        tokens=TokensArgs(sequence_length=4096,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                          train_steps=20,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                          micro_batch_size=1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                          batch_accumulation_per_replica=1024,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                          val_check_interval=-1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                          limit_val_batches=0,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                          limit_test_batches=0),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:        optimizer=OptimizerArgs(optimizer_factory=AdamWOptimizerArgs(adam_eps=1e-08,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                     adam_beta1=0.9,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                     adam_beta2=0.95,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                     torch_adam_is_fused=True,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                     name='adamW'),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                zero_stage=1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                weight_decay=0.01,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                clip_grad=1.0,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                accumulate_grad_in_fp32=True,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                learning_rate_scheduler=LRSchedulerArgs(learning_rate=0.0001,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                        lr_warmup_steps=1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                        lr_warmup_style='linear',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                        lr_decay_style='linear',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                        lr_decay_steps=19,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                        lr_decay_starting_step=None,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                        min_decay_lr=1e-05)),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:        data_stages=[DatasetStageArgs(name='Training Stage',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                      start_training_step=1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                      data=DataArgs(dataset=PretrainDatasetsArgs(hf_dataset_or_datasets='roneneldan/TinyStories',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                                 hf_dataset_splits='train',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                                 hf_dataset_config_name=None,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                                 dataset_processing_num_proc_per_process=64,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                                 dataset_overwrite_cache=False,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                                                 text_column_name='text'),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                    seed=42,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:                                                    num_loading_workers=0))],
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:        profiler=ProfilerArgs(profiler_export_path=Path('/fsx/ferdinandmom/ferdinand-hf/bench_cluster/results/llama-1B/8_GPUS/dp-1_tp-1_pp-8_mbz-1')),
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:        lighteval=None)
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: Model Config:
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: LlamaConfig(bos_token_id=1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             eos_token_id=2,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             hidden_act='silu',
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             hidden_size=2048,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             initializer_range=0.02,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             intermediate_size=4096,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             is_llama_config=True,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             max_position_embeddings=4096,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             num_attention_heads=32,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             num_hidden_layers=24,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             num_key_value_heads=32,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             pad_token_id=None,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             pretraining_tp=1,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             rms_norm_eps=1e-05,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             rope_scaling=None,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             rope_theta=10000.0,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             tie_word_embeddings=True,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             use_cache=True,
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:             vocab_size=50257)
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: Building model..
[default0]:07/04/2024 00:02:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: Setting PP block ranks...
[default1]:07/04/2024 00:03:14 [INFO|DP=0|PP=1|TP=0|ip-26-0-169-139]: Local number of parameters: 126M (240.02MiB)
[default1]:07/04/2024 00:03:14 [INFO|DP=0|PP=1|TP=0|ip-26-0-169-139]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB
[default1]:07/04/2024 00:03:14 [INFO|DP=0|PP=1|TP=0|ip-26-0-169-139]: No checkpoint path provided.
[default2]:07/04/2024 00:03:14 [INFO|DP=0|PP=2|TP=0|ip-26-0-169-139]: Local number of parameters: 126M (240.02MiB)
[default2]:07/04/2024 00:03:14 [INFO|DP=0|PP=2|TP=0|ip-26-0-169-139]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB
[default2]:07/04/2024 00:03:14 [INFO|DP=0|PP=2|TP=0|ip-26-0-169-139]: No checkpoint path provided.
[default4]:07/04/2024 00:03:14 [INFO|DP=0|PP=4|TP=0|ip-26-0-169-139]: Local number of parameters: 126M (240.02MiB)
[default4]:07/04/2024 00:03:14 [INFO|DP=0|PP=4|TP=0|ip-26-0-169-139]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB
[default4]:07/04/2024 00:03:14 [INFO|DP=0|PP=4|TP=0|ip-26-0-169-139]: No checkpoint path provided.
[default0]:07/04/2024 00:03:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: Total number of parameters: 1.21G (2312.82MiB)
[default0]:07/04/2024 00:03:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: Local number of parameters: 271M (516.35MiB)
[default0]:07/04/2024 00:03:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: [After model building] Memory usage: 520.36MiB. Peak allocated: 522.39MiB Peak reserved: 534.00MiB
[default0]:07/04/2024 00:03:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: No checkpoint path provided.
[default0]:07/04/2024 00:03:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: Parametrizing model parameters using StandardParametrizator
[default3]:07/04/2024 00:03:14 [INFO|DP=0|PP=3|TP=0|ip-26-0-169-139]: Local number of parameters: 168M (320.03MiB)
[default3]:07/04/2024 00:03:14 [INFO|DP=0|PP=3|TP=0|ip-26-0-169-139]: [After model building] Memory usage: 324.04MiB. Peak allocated: 326.07MiB Peak reserved: 336.00MiB
[default3]:07/04/2024 00:03:14 [INFO|DP=0|PP=3|TP=0|ip-26-0-169-139]: No checkpoint path provided.
[default5]:07/04/2024 00:03:14 [INFO|DP=0|PP=5|TP=0|ip-26-0-169-139]: Local number of parameters: 126M (240.02MiB)
[default5]:07/04/2024 00:03:14 [INFO|DP=0|PP=5|TP=0|ip-26-0-169-139]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB
[default5]:07/04/2024 00:03:14 [INFO|DP=0|PP=5|TP=0|ip-26-0-169-139]: No checkpoint path provided.
[default7]:07/04/2024 00:03:14 [INFO|DP=0|PP=7|TP=0|ip-26-0-169-139]: Local number of parameters: 103M (196.32MiB)
[default7]:07/04/2024 00:03:14 [INFO|DP=0|PP=7|TP=0|ip-26-0-169-139]: [After model building] Memory usage: 196.33MiB. Peak allocated: 196.33MiB Peak reserved: 200.00MiB
[default7]:07/04/2024 00:03:14 [INFO|DP=0|PP=7|TP=0|ip-26-0-169-139]: No checkpoint path provided.
[default6]:07/04/2024 00:03:14 [INFO|DP=0|PP=6|TP=0|ip-26-0-169-139]: Local number of parameters: 168M (320.03MiB)
[default6]:07/04/2024 00:03:14 [INFO|DP=0|PP=6|TP=0|ip-26-0-169-139]: [After model building] Memory usage: 324.04MiB. Peak allocated: 326.07MiB Peak reserved: 336.00MiB
[default6]:07/04/2024 00:03:14 [INFO|DP=0|PP=6|TP=0|ip-26-0-169-139]: No checkpoint path provided.
[default0]:07/04/2024 00:03:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: [Optimizer Building] Using LearningRateForSP as learning rate
[default0]:07/04/2024 00:03:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: [ZeRO sharding] Size of optimizer params per rank:
[default0]:07/04/2024 00:03:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: [ZeRO sharding] DP Rank 0 has 271M out of 271M (100.00%) params' optimizer states
[default0]:07/04/2024 00:03:15 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: [Training Plan] Stage Training Stage has 19 remaining training steps and has consumed 0 samples
[default0]:07/04/2024 00:03:15 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: Using `datasets` library
[default0]:07/04/2024 00:03:15 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: Loading tokenizer from openai-community/gpt2 and transformers/hf_hub versions ('4.41.2', '0.23.4')
[default0]:07/04/2024 00:03:15 [WARNING|DP=0|PP=0|TP=0|ip-26-0-169-139]: 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 00:03:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: [Training Plan] There are 1 training stages 
[default0]:07/04/2024 00:03:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: [Stage Training Stage] start from step 1 
[default0]:07/04/2024 00:03:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: 
[default0]:07/04/2024 00:03:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: [Start training] datetime: 2024-07-04 00:03:16.525106 | mbs: 1 | grad_accum: 1024 | global_batch_size: 1024 | sequence_length: 4096 | train_steps: 20 | start_iteration_step: 0 | consumed_train_samples: 0
[default0]:07/04/2024 00:03:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]: Resuming training from stage Training Stage, it has trained for 0 samples and has 19 remaining train steps
[default0]:07/04/2024 00:03:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:  Memory usage: 2585.75MiB. Peak allocated 2585.75MiB. Peak reserved: 2602.00MiB
[default1]:07/04/2024 00:03:16 [WARNING|DP=0|PP=1|TP=0|ip-26-0-169-139]: Repo card metadata block was not found. Setting CardData to empty.
[default4]:07/04/2024 00:03:16 [WARNING|DP=0|PP=4|TP=0|ip-26-0-169-139]: Repo card metadata block was not found. Setting CardData to empty.
[default1]:Repo card metadata block was not found. Setting CardData to empty.
[default2]:Repo card metadata block was not found. Setting CardData to empty.
[default4]:Repo card metadata block was not found. Setting CardData to empty.
[default2]:07/04/2024 00:03:16 [WARNING|DP=0|PP=2|TP=0|ip-26-0-169-139]: 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.
[default7]:07/04/2024 00:03:16 [WARNING|DP=0|PP=7|TP=0|ip-26-0-169-139]: Repo card metadata block was not found. Setting CardData to empty.
[default6]:07/04/2024 00:03:16 [WARNING|DP=0|PP=6|TP=0|ip-26-0-169-139]: Repo card metadata block was not found. Setting CardData to empty.
[default3]:07/04/2024 00:03:16 [WARNING|DP=0|PP=3|TP=0|ip-26-0-169-139]: Repo card metadata block was not found. Setting CardData to empty.
[default3]:Repo card metadata block was not found. Setting CardData to empty.
[default5]:Repo card metadata block was not found. Setting CardData to empty.
[default5]:07/04/2024 00:03:20 [WARNING|DP=0|PP=5|TP=0|ip-26-0-169-139]: Repo card metadata block was not found. Setting CardData to empty.
[default7]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/autograd/graph.py:744: 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.)
[default7]:  return 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/graph.py:744: 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]:  return Variable._execution_engine.run_backward(  # Calls into the C++ engine to run the backward pass
[default5]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/autograd/graph.py:744: 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]:  return Variable._execution_engine.run_backward(  # Calls into the C++ engine to run the backward pass
[default4]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/autograd/graph.py:744: 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]:  return 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/graph.py:744: 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]:  return 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/graph.py:744: 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]:  return 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/graph.py:744: 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]:  return 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/graph.py:744: UserWarning: Attempting to run cuBLAS, but there was no current CUDA context! Attempting to set the primary context... (Triggered internally at ../aten/src/ATen/cuda/CublasHandlePool.cpp:135.)
[default0]:  return 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/graph.py:744: 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]:  return 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/distributed/distributed_c10d.py:2261: UserWarning: torch.distributed.all_reduce_coalesced will be deprecated. If you must use it, please revisit our documentation later at https://pytorch.org/docs/master/distributed.html#collective-functions
[default7]:  warnings.warn(
[default0]:07/04/2024 00:04:13 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:  Memory usage: 2651.78MiB. Peak allocated 12082.50MiB. Peak reserved: 12264.00MiB
[default0]:/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py:2261: UserWarning: torch.distributed.all_reduce_coalesced will be deprecated. If you must use it, please revisit our documentation later at https://pytorch.org/docs/master/distributed.html#collective-functions
[default0]:  warnings.warn(
[default0]:07/04/2024 00:04:13 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:  Memory usage: 4717.19MiB. Peak allocated 4717.19MiB. Peak reserved: 13052.00MiB
[default7]:07/04/2024 00:04:13 [INFO|DP=0|PP=7|TP=0|ip-26-0-169-139]: iteration: 1 / 20 | consumed_tokens: 4.19M | elapsed_time_per_iteration_ms: 55.2K | tokens_per_sec: 76K | tokens_per_sec_per_gpu: 9.49K | global_batch_size: 1.02K | lm_loss: 11.1 | lr: 0.0001 | model_tflops_per_gpu: 86.1 | hardware_tflops_per_gpu: 86.1 | grad_norm: 24.9 | cuda_memory_allocated: 1.92G | cuda_max_memory_reserved: 4.21G | hd_total_memory_tb: 312G | hd_used_memory_tb: 65.5G | hd_free_memory_tb: 247G
[default0]:07/04/2024 00:04:44 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:  Memory usage: 4717.19MiB. Peak allocated 14046.46MiB. Peak reserved: 14488.00MiB
[default7]:07/04/2024 00:04:44 [INFO|DP=0|PP=7|TP=0|ip-26-0-169-139]: iteration: 2 / 20 | consumed_tokens: 8.39M | elapsed_time_per_iteration_ms: 31.2K | tokens_per_sec: 134K | tokens_per_sec_per_gpu: 16.8K | global_batch_size: 1.02K | lm_loss: 11.1 | lr: 9.53e-05 | model_tflops_per_gpu: 152 | hardware_tflops_per_gpu: 152 | grad_norm: 25.1 | cuda_memory_allocated: 1.92G | cuda_max_memory_reserved: 4.21G | hd_total_memory_tb: 312G | hd_used_memory_tb: 65.5G | hd_free_memory_tb: 247G
[default0]:07/04/2024 00:04:45 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:  Memory usage: 4717.19MiB. Peak allocated 4717.20MiB. Peak reserved: 14488.00MiB
[default0]:07/04/2024 00:05:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:  Memory usage: 4717.19MiB. Peak allocated 14046.46MiB. Peak reserved: 14488.00MiB
[default0]:STAGE:2024-07-04 00:05:16 678752:678752 ActivityProfilerController.cpp:314] Completed Stage: Warm Up
[default7]:07/04/2024 00:05:16 [INFO|DP=0|PP=7|TP=0|ip-26-0-169-139]: iteration: 3 / 20 | consumed_tokens: 12.6M | elapsed_time_per_iteration_ms: 31.3K | tokens_per_sec: 134K | tokens_per_sec_per_gpu: 16.8K | global_batch_size: 1.02K | lm_loss: 9.49 | lr: 9.05e-05 | model_tflops_per_gpu: 152 | hardware_tflops_per_gpu: 152 | grad_norm: 21.6 | cuda_memory_allocated: 1.92G | cuda_max_memory_reserved: 4.21G | hd_total_memory_tb: 312G | hd_used_memory_tb: 65.5G | hd_free_memory_tb: 247G
[default0]:07/04/2024 00:05:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:  Memory usage: 4717.19MiB. Peak allocated 4717.20MiB. Peak reserved: 14488.00MiB
[default7]:07/04/2024 00:05:51 [INFO|DP=0|PP=7|TP=0|ip-26-0-169-139]: iteration: 4 / 20 | consumed_tokens: 16.8M | elapsed_time_per_iteration_ms: 35.3K | tokens_per_sec: 119K | tokens_per_sec_per_gpu: 14.8K | global_batch_size: 1.02K | lm_loss: 9.36 | lr: 8.58e-05 | model_tflops_per_gpu: 135 | hardware_tflops_per_gpu: 135 | grad_norm: 21.4 | cuda_memory_allocated: 1.92G | cuda_max_memory_reserved: 4.21G | hd_total_memory_tb: 312G | hd_used_memory_tb: 65.5G | hd_free_memory_tb: 247G
[default0]:07/04/2024 00:05:51 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:  Memory usage: 4717.19MiB. Peak allocated 14046.46MiB. Peak reserved: 14488.00MiB
[default0]:07/04/2024 00:05:51 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:  Memory usage: 4717.19MiB. Peak allocated 4717.20MiB. Peak reserved: 14488.00MiB
[default7]:07/04/2024 00:06:27 [INFO|DP=0|PP=7|TP=0|ip-26-0-169-139]: iteration: 5 / 20 | consumed_tokens: 21M | elapsed_time_per_iteration_ms: 35.6K | tokens_per_sec: 118K | tokens_per_sec_per_gpu: 14.7K | global_batch_size: 1.02K | lm_loss: 9.02 | lr: 8.11e-05 | model_tflops_per_gpu: 134 | hardware_tflops_per_gpu: 134 | grad_norm: 12.8
[default0]:07/04/2024 00:06:27 [INFO|DP=0|PP=0|TP=0|ip-26-0-169-139]:  Memory usage: 4717.19MiB. Peak allocated 14046.46MiB. Peak reserved: 14488.00MiB
[default7]:07/04/2024 00:07:03 [INFO|DP=0|PP=7|TP=0|ip-26-0-169-139]: iteration: 6 / 20 | consumed_tokens: 25.2M | elapsed_time_per_iteration_ms: 36.1K | tokens_per_sec: 116K | tokens_per_sec_per_gpu: 14.5K | global_batch_size: 1.02K | lm_loss: 10.3 | lr: 7.63e-05 | model_tflops_per_gpu: 132 | hardware_tflops_per_gpu: 132 | grad_norm: 47.1
[default0]:STAGE:2024-07-04 00:08:44 678752:678752 ActivityProfilerController.cpp:320] Completed Stage: Collection
[default0]:STAGE:2024-07-04 00:08:53 678752:678752 ActivityProfilerController.cpp:324] Completed Stage: Post Processing
[default6]:[rank6]:[E ProcessGroupNCCL.cpp:563] [Rank 6] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=92169, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600005 milliseconds before timing out.
[default2]:[rank2]:[E ProcessGroupNCCL.cpp:563] [Rank 2] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600030 milliseconds before timing out.
[default5]:[rank5]:[E ProcessGroupNCCL.cpp:563] [Rank 5] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600079 milliseconds before timing out.
[default3]:[rank3]:[E ProcessGroupNCCL.cpp:563] [Rank 3] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600004 milliseconds before timing out.
[default7]:[rank7]:[E ProcessGroupNCCL.cpp:563] [Rank 7] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=36873, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600068 milliseconds before timing out.
[default4]:[rank4]:[E ProcessGroupNCCL.cpp:563] [Rank 4] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600057 milliseconds before timing out.
[default1]:[rank1]:[E ProcessGroupNCCL.cpp:563] [Rank 1] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600057 milliseconds before timing out.
[default6]:[rank6]: Traceback (most recent call last):
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in <module>
[default6]:[rank6]:     trainer.train(dataloader)
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train
[default6]:[rank6]:     outputs, loss_avg = self.training_step(dataloader=self.current_dataloader)
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step
[default6]:[rank6]:     outputs = self.pipeline_engine.train_batch_iter(
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter
[default6]:[rank6]:     output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model)
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward
[default2]:[rank2]: Traceback (most recent call last):
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in <module>
[default6]:[rank6]:     output = model(**micro_batch)
[default6]:[rank6]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default2]:[rank2]:     trainer.train(dataloader)
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train
[default2]:[rank2]:     outputs, loss_avg = self.training_step(dataloader=self.current_dataloader)
[default6]:[rank6]:     return self._call_impl(*args, **kwargs)
[default6]:[rank6]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default6]:[rank6]:     return forward_call(*args, **kwargs)
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward
[default6]:[rank6]:     sharded_logits = self.model(
[default6]:[rank6]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step
[default2]:[rank2]:     outputs = self.pipeline_engine.train_batch_iter(
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter
[default2]:[rank2]:     output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model)
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward
[default2]:[rank2]:     output = model(**micro_batch)
[default2]:[rank2]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default6]:[rank6]:     return self._call_impl(*args, **kwargs)
[default2]:[rank2]:     return self._call_impl(*args, **kwargs)
[default6]:[rank6]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default2]:[rank2]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default6]:[rank6]:     return forward_call(*args, **kwargs)
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward
[default6]:[rank6]:     return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0]
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states
[default6]:[rank6]:     hidden_encoder_states = encoder_block(**hidden_encoder_states)
[default2]:[rank2]:     return forward_call(*args, **kwargs)
[default6]:[rank6]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default6]:[rank6]:     return self._call_impl(*args, **kwargs)
[default6]:[rank6]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default6]:[rank6]:     return forward_call(*args, **kwargs)
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward
[default6]:[rank6]:     new_kwargs[name] = recv_from_pipeline_state_buffer(
[default2]:[rank2]:     sharded_logits = self.model(
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer
[default2]:[rank2]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default6]:[rank6]:     pipeline_state.run_communication()
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication
[default2]:[rank2]:     return self._call_impl(*args, **kwargs)
[default6]:[rank6]:     recv_activation_tensor = recv_activation()
[default2]:[rank2]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__
[default6]:[rank6]:     return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0]
[default2]:[rank2]:     return forward_call(*args, **kwargs)
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors
[default6]:[rank6]:     buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag)
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors
[default6]:[rank6]:     meta = self._recv_meta(from_rank=from_rank, tag=tag)
[default6]:[rank6]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 267, in _recv_meta
[default6]:[rank6]:     self.second_metadata = torch.empty(second_metadata_num_elements, dtype=torch.long, device=self.device)
[default6]:[rank6]: torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate more than 1EB memory.
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward
[default2]:[rank2]:     return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0]
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states
[default2]:[rank2]:     hidden_encoder_states = encoder_block(**hidden_encoder_states)
[default2]:[rank2]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default2]:[rank2]:     return self._call_impl(*args, **kwargs)
[default2]:[rank2]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default2]:[rank2]:     return forward_call(*args, **kwargs)
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward
[default2]:[rank2]:     new_kwargs[name] = recv_from_pipeline_state_buffer(
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer
[default2]:[rank2]:     pipeline_state.run_communication()
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication
[default2]:[rank2]:     recv_activation_tensor = recv_activation()
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__
[default2]:[rank2]:     return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0]
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors
[default2]:[rank2]:     buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag)
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors
[default2]:[rank2]:     meta = self._recv_meta(from_rank=from_rank, tag=tag)
[default2]:[rank2]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 267, in _recv_meta
[default2]:[rank2]:     self.second_metadata = torch.empty(second_metadata_num_elements, dtype=torch.long, device=self.device)
[default2]:[rank2]: torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate more than 1EB memory.
[default5]:[rank5]: Traceback (most recent call last):
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in <module>
[default5]:[rank5]:     trainer.train(dataloader)
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train
[default5]:[rank5]:     outputs, loss_avg = self.training_step(dataloader=self.current_dataloader)
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step
[default5]:[rank5]:     outputs = self.pipeline_engine.train_batch_iter(
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter
[default5]:[rank5]:     output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model)
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward
[default5]:[rank5]:     output = model(**micro_batch)
[default5]:[rank5]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default5]:[rank5]:     return self._call_impl(*args, **kwargs)
[default5]:[rank5]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default5]:[rank5]:     return forward_call(*args, **kwargs)
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward
[default5]:[rank5]:     sharded_logits = self.model(
[default5]:[rank5]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default5]:[rank5]:     return self._call_impl(*args, **kwargs)
[default5]:[rank5]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default5]:[rank5]:     return forward_call(*args, **kwargs)
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward
[default5]:[rank5]:     return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0]
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states
[default5]:[rank5]:     hidden_encoder_states = encoder_block(**hidden_encoder_states)
[default5]:[rank5]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default5]:[rank5]:     return self._call_impl(*args, **kwargs)
[default5]:[rank5]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default5]:[rank5]:     return forward_call(*args, **kwargs)
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward
[default5]:[rank5]:     new_kwargs[name] = recv_from_pipeline_state_buffer(
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer
[default5]:[rank5]:     pipeline_state.run_communication()
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication
[default5]:[rank5]:     recv_activation_tensor = recv_activation()
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__
[default5]:[rank5]:     return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0]
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors
[default5]:[rank5]:     buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag)
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors
[default5]:[rank5]:     meta = self._recv_meta(from_rank=from_rank, tag=tag)
[default5]:[rank5]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 267, in _recv_meta
[default5]:[rank5]:     self.second_metadata = torch.empty(second_metadata_num_elements, dtype=torch.long, device=self.device)
[default5]:[rank5]: torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate more than 1EB memory.
[default3]:[rank3]: Traceback (most recent call last):
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in <module>
[default3]:[rank3]:     trainer.train(dataloader)
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train
[default3]:[rank3]:     outputs, loss_avg = self.training_step(dataloader=self.current_dataloader)
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step
[default3]:[rank3]:     outputs = self.pipeline_engine.train_batch_iter(
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter
[default3]:[rank3]:     output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model)
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward
[default3]:[rank3]:     output = model(**micro_batch)
[default3]:[rank3]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default3]:[rank3]:     return self._call_impl(*args, **kwargs)
[default3]:[rank3]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default3]:[rank3]:     return forward_call(*args, **kwargs)
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward
[default3]:[rank3]:     sharded_logits = self.model(
[default3]:[rank3]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default3]:[rank3]:     return self._call_impl(*args, **kwargs)
[default3]:[rank3]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default3]:[rank3]:     return forward_call(*args, **kwargs)
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward
[default3]:[rank3]:     return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0]
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states
[default3]:[rank3]:     hidden_encoder_states = encoder_block(**hidden_encoder_states)
[default3]:[rank3]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default3]:[rank3]:     return self._call_impl(*args, **kwargs)
[default3]:[rank3]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default3]:[rank3]:     return forward_call(*args, **kwargs)
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward
[default3]:[rank3]:     new_kwargs[name] = recv_from_pipeline_state_buffer(
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer
[default3]:[rank3]:     pipeline_state.run_communication()
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication
[default3]:[rank3]:     recv_activation_tensor = recv_activation()
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__
[default3]:[rank3]:     return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0]
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors
[default3]:[rank3]:     buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag)
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors
[default3]:[rank3]:     meta = self._recv_meta(from_rank=from_rank, tag=tag)
[default3]:[rank3]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 267, in _recv_meta
[default3]:[rank3]:     self.second_metadata = torch.empty(second_metadata_num_elements, dtype=torch.long, device=self.device)
[default3]:[rank3]: torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate more than 1EB memory.
[default4]:[rank4]: Traceback (most recent call last):
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in <module>
[default4]:[rank4]:     trainer.train(dataloader)
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train
[default4]:[rank4]:     outputs, loss_avg = self.training_step(dataloader=self.current_dataloader)
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step
[default4]:[rank4]:     outputs = self.pipeline_engine.train_batch_iter(
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter
[default4]:[rank4]:     output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model)
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward
[default4]:[rank4]:     output = model(**micro_batch)
[default4]:[rank4]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default4]:[rank4]:     return self._call_impl(*args, **kwargs)
[default4]:[rank4]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default4]:[rank4]:     return forward_call(*args, **kwargs)
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward
[default4]:[rank4]:     sharded_logits = self.model(
[default4]:[rank4]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default4]:[rank4]:     return self._call_impl(*args, **kwargs)
[default4]:[rank4]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default4]:[rank4]:     return forward_call(*args, **kwargs)
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward
[default4]:[rank4]:     return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0]
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states
[default4]:[rank4]:     hidden_encoder_states = encoder_block(**hidden_encoder_states)
[default4]:[rank4]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default4]:[rank4]:     return self._call_impl(*args, **kwargs)
[default4]:[rank4]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default4]:[rank4]:     return forward_call(*args, **kwargs)
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward
[default4]:[rank4]:     new_kwargs[name] = recv_from_pipeline_state_buffer(
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer
[default4]:[rank4]:     pipeline_state.run_communication()
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication
[default4]:[rank4]:     recv_activation_tensor = recv_activation()
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__
[default4]:[rank4]:     return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0]
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors
[default4]:[rank4]:     buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag)
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors
[default4]:[rank4]:     meta = self._recv_meta(from_rank=from_rank, tag=tag)
[default4]:[rank4]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 267, in _recv_meta
[default4]:[rank4]:     self.second_metadata = torch.empty(second_metadata_num_elements, dtype=torch.long, device=self.device)
[default4]:[rank4]: torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate more than 1EB memory.
[default1]:[rank1]: Traceback (most recent call last):
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in <module>
[default1]:[rank1]:     trainer.train(dataloader)
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train
[default1]:[rank1]:     outputs, loss_avg = self.training_step(dataloader=self.current_dataloader)
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step
[default1]:[rank1]:     outputs = self.pipeline_engine.train_batch_iter(
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter
[default1]:[rank1]:     output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model)
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward
[default1]:[rank1]:     output = model(**micro_batch)
[default1]:[rank1]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default1]:[rank1]:     return self._call_impl(*args, **kwargs)
[default1]:[rank1]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default1]:[rank1]:     return forward_call(*args, **kwargs)
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward
[default1]:[rank1]:     sharded_logits = self.model(
[default1]:[rank1]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default1]:[rank1]:     return self._call_impl(*args, **kwargs)
[default1]:[rank1]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default1]:[rank1]:     return forward_call(*args, **kwargs)
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward
[default1]:[rank1]:     return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0]
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states
[default1]:[rank1]:     hidden_encoder_states = encoder_block(**hidden_encoder_states)
[default1]:[rank1]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default1]:[rank1]:     return self._call_impl(*args, **kwargs)
[default1]:[rank1]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default1]:[rank1]:     return forward_call(*args, **kwargs)
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward
[default1]:[rank1]:     new_kwargs[name] = recv_from_pipeline_state_buffer(
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer
[default1]:[rank1]:     pipeline_state.run_communication()
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication
[default1]:[rank1]:     recv_activation_tensor = recv_activation()
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__
[default1]:[rank1]:     return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0]
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors
[default1]:[rank1]:     buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag)
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors
[default1]:[rank1]:     meta = self._recv_meta(from_rank=from_rank, tag=tag)
[default1]:[rank1]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 267, in _recv_meta
[default1]:[rank1]:     self.second_metadata = torch.empty(second_metadata_num_elements, dtype=torch.long, device=self.device)
[default1]:[rank1]: torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate more than 1EB memory.
[default7]:[rank7]: Traceback (most recent call last):
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in <module>
[default7]:[rank7]:     trainer.train(dataloader)
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train
[default7]:[rank7]:     outputs, loss_avg = self.training_step(dataloader=self.current_dataloader)
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step
[default7]:[rank7]:     outputs = self.pipeline_engine.train_batch_iter(
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 278, in train_batch_iter
[default7]:[rank7]:     output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model)
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward
[default7]:[rank7]:     output = model(**micro_batch)
[default7]:[rank7]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default7]:[rank7]:     return self._call_impl(*args, **kwargs)
[default7]:[rank7]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default7]:[rank7]:     return forward_call(*args, **kwargs)
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward
[default7]:[rank7]:     sharded_logits = self.model(
[default7]:[rank7]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default7]:[rank7]:     return self._call_impl(*args, **kwargs)
[default7]:[rank7]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default7]:[rank7]:     return forward_call(*args, **kwargs)
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward
[default7]:[rank7]:     return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0]
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 782, in forward_with_hidden_states
[default7]:[rank7]:     hidden_states = self.final_layer_norm(input=hidden_encoder_states["hidden_states"])["hidden_states"]
[default7]:[rank7]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default7]:[rank7]:     return self._call_impl(*args, **kwargs)
[default7]:[rank7]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default7]:[rank7]:     return forward_call(*args, **kwargs)
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward
[default7]:[rank7]:     new_kwargs[name] = recv_from_pipeline_state_buffer(
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer
[default7]:[rank7]:     pipeline_state.run_communication()
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication
[default7]:[rank7]:     recv_activation_tensor = recv_activation()
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__
[default7]:[rank7]:     return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0]
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors
[default7]:[rank7]:     buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag)
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors
[default7]:[rank7]:     meta = self._recv_meta(from_rank=from_rank, tag=tag)
[default7]:[rank7]:   File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 269, in _recv_meta
[default7]:[rank7]:     dist.recv(
[default7]:[rank7]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper
[default7]:[rank7]:     return func(*args, **kwargs)
[default7]:[rank7]:   File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv
[default7]:[rank7]:     pg.recv([tensor], group_src_rank, tag).wait()
[default7]:[rank7]: torch.distributed.DistBackendError: NCCL communicator was aborted on rank 1. 
[default5]:[rank5]:[E ProcessGroupNCCL.cpp:1537] [PG 3 Rank 5] Timeout at NCCL work: 110601, last enqueued NCCL work: 110601, last completed NCCL work: 110600.
[default5]:[rank5]:[E ProcessGroupNCCL.cpp:577] [Rank 5] 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:583] [Rank 5] To avoid data inconsistency, we are taking the entire process down.
[default5]:[rank5]:[E ProcessGroupNCCL.cpp:1414] [PG 3 Rank 5] Process group watchdog thread terminated with exception: [Rank 5] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600079 milliseconds before timing out.
[default5]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default5]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f77119f5897 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> > >) + 0x1d2 (0x7f7712ccec62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default5]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7f7712cd3a80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default5]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7f7712cd4dcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default5]:frame #4: <unknown function> + 0xd3e95 (0x7f775e76de95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default5]:frame #5: <unknown function> + 0x8609 (0x7f77637b4609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default5]:frame #6: clone + 0x43 (0x7f776357f353 in /lib/x86_64-linux-gnu/libc.so.6)
[default5]:
[default5]:terminate called after throwing an instance of 'c10::DistBackendError'
[default5]:  what():  [PG 3 Rank 5] Process group watchdog thread terminated with exception: [Rank 5] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600079 milliseconds before timing out.
[default5]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default5]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f77119f5897 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> > >) + 0x1d2 (0x7f7712ccec62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default5]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7f7712cd3a80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default5]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7f7712cd4dcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default5]:frame #4: <unknown function> + 0xd3e95 (0x7f775e76de95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default5]:frame #5: <unknown function> + 0x8609 (0x7f77637b4609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default5]:frame #6: clone + 0x43 (0x7f776357f353 in /lib/x86_64-linux-gnu/libc.so.6)
[default5]:
[default5]:Exception raised from ncclCommWatchdog at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1418 (most recent call first):
[default5]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f77119f5897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default5]:frame #1: <unknown function> + 0xe32119 (0x7f7712958119 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default5]:frame #2: <unknown function> + 0xd3e95 (0x7f775e76de95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default5]:frame #3: <unknown function> + 0x8609 (0x7f77637b4609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default5]:frame #4: clone + 0x43 (0x7f776357f353 in /lib/x86_64-linux-gnu/libc.so.6)
[default5]:
[default7]:[rank7]:[E ProcessGroupNCCL.cpp:1537] [PG 3 Rank 7] Timeout at NCCL work: 36873, last enqueued NCCL work: 36873, last completed NCCL work: 36872.
[default7]:[rank7]:[E ProcessGroupNCCL.cpp:577] [Rank 7] 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.
[default7]:[rank7]:[E ProcessGroupNCCL.cpp:583] [Rank 7] To avoid data inconsistency, we are taking the entire process down.
[default7]:[rank7]:[E ProcessGroupNCCL.cpp:1414] [PG 3 Rank 7] Process group watchdog thread terminated with exception: [Rank 7] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=36873, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600068 milliseconds before timing out.
[default7]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default7]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f58db8cf897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default7]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7f58dcba8c62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default7]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7f58dcbada80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default7]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7f58dcbaedcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default7]:frame #4: <unknown function> + 0xd3e95 (0x7f5928647e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default7]:frame #5: <unknown function> + 0x8609 (0x7f592d68e609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default7]:frame #6: clone + 0x43 (0x7f592d459353 in /lib/x86_64-linux-gnu/libc.so.6)
[default7]:
[default7]:terminate called after throwing an instance of 'c10::DistBackendError'
[default7]:  what():  [PG 3 Rank 7] Process group watchdog thread terminated with exception: [Rank 7] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=36873, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600068 milliseconds before timing out.
[default7]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default7]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f58db8cf897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default7]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7f58dcba8c62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default7]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7f58dcbada80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default7]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7f58dcbaedcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default7]:frame #4: <unknown function> + 0xd3e95 (0x7f5928647e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default7]:frame #5: <unknown function> + 0x8609 (0x7f592d68e609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default7]:frame #6: clone + 0x43 (0x7f592d459353 in /lib/x86_64-linux-gnu/libc.so.6)
[default7]:
[default7]:Exception raised from ncclCommWatchdog at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1418 (most recent call first):
[default7]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f58db8cf897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default7]:frame #1: <unknown function> + 0xe32119 (0x7f58dc832119 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default7]:frame #2: <unknown function> + 0xd3e95 (0x7f5928647e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default7]:frame #3: <unknown function> + 0x8609 (0x7f592d68e609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default7]:frame #4: clone + 0x43 (0x7f592d459353 in /lib/x86_64-linux-gnu/libc.so.6)
[default7]:
[default1]:[rank1]:[E ProcessGroupNCCL.cpp:1537] [PG 3 Rank 1] Timeout at NCCL work: 110601, last enqueued NCCL work: 110601, last completed NCCL work: 110600.
[default1]:[rank1]:[E ProcessGroupNCCL.cpp:577] [Rank 1] 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.
[default1]:[rank1]:[E ProcessGroupNCCL.cpp:583] [Rank 1] To avoid data inconsistency, we are taking the entire process down.
[default1]:[rank1]:[E ProcessGroupNCCL.cpp:1414] [PG 3 Rank 1] Process group watchdog thread terminated with exception: [Rank 1] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600057 milliseconds before timing out.
[default1]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default1]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f9fd8b2c897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default1]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7f9fd9e05c62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default1]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7f9fd9e0aa80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default1]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7f9fd9e0bdcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default1]:frame #4: <unknown function> + 0xd3e95 (0x7fa0258a4e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default1]:frame #5: <unknown function> + 0x8609 (0x7fa02a8eb609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default1]:frame #6: clone + 0x43 (0x7fa02a6b6353 in /lib/x86_64-linux-gnu/libc.so.6)
[default1]:
[default1]:terminate called after throwing an instance of 'c10::DistBackendError'
[default1]:  what():  [PG 3 Rank 1] Process group watchdog thread terminated with exception: [Rank 1] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600057 milliseconds before timing out.
[default1]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default1]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f9fd8b2c897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default1]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7f9fd9e05c62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default1]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7f9fd9e0aa80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default1]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7f9fd9e0bdcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default1]:frame #4: <unknown function> + 0xd3e95 (0x7fa0258a4e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default1]:frame #5: <unknown function> + 0x8609 (0x7fa02a8eb609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default1]:frame #6: clone + 0x43 (0x7fa02a6b6353 in /lib/x86_64-linux-gnu/libc.so.6)
[default1]:
[default1]:Exception raised from ncclCommWatchdog at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1418 (most recent call first):
[default1]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f9fd8b2c897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default1]:frame #1: <unknown function> + 0xe32119 (0x7f9fd9a8f119 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default1]:frame #2: <unknown function> + 0xd3e95 (0x7fa0258a4e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default1]:frame #3: <unknown function> + 0x8609 (0x7fa02a8eb609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default1]:frame #4: clone + 0x43 (0x7fa02a6b6353 in /lib/x86_64-linux-gnu/libc.so.6)
[default1]:
[default2]:[rank2]:[E ProcessGroupNCCL.cpp:1537] [PG 3 Rank 2] Timeout at NCCL work: 110601, last enqueued NCCL work: 110601, last completed NCCL work: 110600.
[default2]:[rank2]:[E ProcessGroupNCCL.cpp:577] [Rank 2] 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.
[default2]:[rank2]:[E ProcessGroupNCCL.cpp:583] [Rank 2] To avoid data inconsistency, we are taking the entire process down.
[default2]:[rank2]:[E ProcessGroupNCCL.cpp:1414] [PG 3 Rank 2] Process group watchdog thread terminated with exception: [Rank 2] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600030 milliseconds before timing out.
[default2]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default2]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7fe1575e1897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default2]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7fe1588bac62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default2]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7fe1588bfa80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default2]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7fe1588c0dcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default2]:frame #4: <unknown function> + 0xd3e95 (0x7fe1a4359e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default2]:frame #5: <unknown function> + 0x8609 (0x7fe1a93a0609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default2]:frame #6: clone + 0x43 (0x7fe1a916b353 in /lib/x86_64-linux-gnu/libc.so.6)
[default2]:
[default2]:terminate called after throwing an instance of 'c10::DistBackendError'
[default2]:  what():  [PG 3 Rank 2] Process group watchdog thread terminated with exception: [Rank 2] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600030 milliseconds before timing out.
[default2]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default2]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7fe1575e1897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default2]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7fe1588bac62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default2]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7fe1588bfa80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default2]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7fe1588c0dcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default2]:frame #4: <unknown function> + 0xd3e95 (0x7fe1a4359e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default2]:frame #5: <unknown function> + 0x8609 (0x7fe1a93a0609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default2]:frame #6: clone + 0x43 (0x7fe1a916b353 in /lib/x86_64-linux-gnu/libc.so.6)
[default2]:
[default2]:Exception raised from ncclCommWatchdog at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1418 (most recent call first):
[default2]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7fe1575e1897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default2]:frame #1: <unknown function> + 0xe32119 (0x7fe158544119 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default2]:frame #2: <unknown function> + 0xd3e95 (0x7fe1a4359e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default2]:frame #3: <unknown function> + 0x8609 (0x7fe1a93a0609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default2]:frame #4: clone + 0x43 (0x7fe1a916b353 in /lib/x86_64-linux-gnu/libc.so.6)
[default2]:
[default3]:[rank3]:[E ProcessGroupNCCL.cpp:1537] [PG 3 Rank 3] Timeout at NCCL work: 110601, last enqueued NCCL work: 110601, last completed NCCL work: 110600.
[default3]:[rank3]:[E ProcessGroupNCCL.cpp:577] [Rank 3] 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.
[default3]:[rank3]:[E ProcessGroupNCCL.cpp:583] [Rank 3] To avoid data inconsistency, we are taking the entire process down.
[default3]:[rank3]:[E ProcessGroupNCCL.cpp:1414] [PG 3 Rank 3] Process group watchdog thread terminated with exception: [Rank 3] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600004 milliseconds before timing out.
[default3]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default3]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f2025190897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default3]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7f2026469c62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default3]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7f202646ea80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default3]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7f202646fdcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default3]:frame #4: <unknown function> + 0xd3e95 (0x7f2071f08e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default3]:frame #5: <unknown function> + 0x8609 (0x7f2076f4f609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default3]:frame #6: clone + 0x43 (0x7f2076d1a353 in /lib/x86_64-linux-gnu/libc.so.6)
[default3]:
[default3]:terminate called after throwing an instance of 'c10::DistBackendError'
[default3]:  what():  [PG 3 Rank 3] Process group watchdog thread terminated with exception: [Rank 3] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600004 milliseconds before timing out.
[default3]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default3]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f2025190897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default3]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7f2026469c62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default3]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7f202646ea80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default3]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7f202646fdcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default3]:frame #4: <unknown function> + 0xd3e95 (0x7f2071f08e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default3]:frame #5: <unknown function> + 0x8609 (0x7f2076f4f609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default3]:frame #6: clone + 0x43 (0x7f2076d1a353 in /lib/x86_64-linux-gnu/libc.so.6)
[default3]:
[default3]:Exception raised from ncclCommWatchdog at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1418 (most recent call first):
[default3]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f2025190897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default3]:frame #1: <unknown function> + 0xe32119 (0x7f20260f3119 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default3]:frame #2: <unknown function> + 0xd3e95 (0x7f2071f08e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default3]:frame #3: <unknown function> + 0x8609 (0x7f2076f4f609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default3]:frame #4: clone + 0x43 (0x7f2076d1a353 in /lib/x86_64-linux-gnu/libc.so.6)
[default3]:
[default6]:[rank6]:[E ProcessGroupNCCL.cpp:1537] [PG 3 Rank 6] Timeout at NCCL work: 92169, last enqueued NCCL work: 92169, last completed NCCL work: 92168.
[default6]:[rank6]:[E ProcessGroupNCCL.cpp:577] [Rank 6] 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.
[default6]:[rank6]:[E ProcessGroupNCCL.cpp:583] [Rank 6] To avoid data inconsistency, we are taking the entire process down.
[default6]:[rank6]:[E ProcessGroupNCCL.cpp:1414] [PG 3 Rank 6] Process group watchdog thread terminated with exception: [Rank 6] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=92169, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600005 milliseconds before timing out.
[default6]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default6]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7fca98159897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default6]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7fca99432c62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default6]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7fca99437a80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default6]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7fca99438dcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default6]:frame #4: <unknown function> + 0xd3e95 (0x7fcae4ed1e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default6]:frame #5: <unknown function> + 0x8609 (0x7fcae9f18609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default6]:frame #6: clone + 0x43 (0x7fcae9ce3353 in /lib/x86_64-linux-gnu/libc.so.6)
[default6]:
[default6]:terminate called after throwing an instance of 'c10::DistBackendError'
[default6]:  what():  [PG 3 Rank 6] Process group watchdog thread terminated with exception: [Rank 6] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=92169, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600005 milliseconds before timing out.
[default6]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default6]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7fca98159897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default6]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7fca99432c62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default6]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7fca99437a80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default6]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7fca99438dcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default6]:frame #4: <unknown function> + 0xd3e95 (0x7fcae4ed1e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default6]:frame #5: <unknown function> + 0x8609 (0x7fcae9f18609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default6]:frame #6: clone + 0x43 (0x7fcae9ce3353 in /lib/x86_64-linux-gnu/libc.so.6)
[default6]:
[default6]:Exception raised from ncclCommWatchdog at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1418 (most recent call first):
[default6]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7fca98159897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default6]:frame #1: <unknown function> + 0xe32119 (0x7fca990bc119 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default6]:frame #2: <unknown function> + 0xd3e95 (0x7fcae4ed1e95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default6]:frame #3: <unknown function> + 0x8609 (0x7fcae9f18609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default6]:frame #4: clone + 0x43 (0x7fcae9ce3353 in /lib/x86_64-linux-gnu/libc.so.6)
[default6]:
[default4]:[rank4]:[E ProcessGroupNCCL.cpp:1537] [PG 3 Rank 4] Timeout at NCCL work: 110601, last enqueued NCCL work: 110601, last completed NCCL work: 110600.
[default4]:[rank4]:[E ProcessGroupNCCL.cpp:577] [Rank 4] 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.
[default4]:[rank4]:[E ProcessGroupNCCL.cpp:583] [Rank 4] To avoid data inconsistency, we are taking the entire process down.
[default4]:[rank4]:[E ProcessGroupNCCL.cpp:1414] [PG 3 Rank 4] Process group watchdog thread terminated with exception: [Rank 4] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600057 milliseconds before timing out.
[default4]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default4]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f3c1ff32897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default4]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7f3c2120bc62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default4]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7f3c21210a80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default4]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7f3c21211dcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default4]:frame #4: <unknown function> + 0xd3e95 (0x7f3c6ccaae95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default4]:frame #5: <unknown function> + 0x8609 (0x7f3c71cf1609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default4]:frame #6: clone + 0x43 (0x7f3c71abc353 in /lib/x86_64-linux-gnu/libc.so.6)
[default4]:
[default4]:terminate called after throwing an instance of 'c10::DistBackendError'
[default4]:  what():  [PG 3 Rank 4] Process group watchdog thread terminated with exception: [Rank 4] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=110601, OpType=RECV, NumelIn=7, NumelOut=7, Timeout(ms)=600000) ran for 600057 milliseconds before timing out.
[default4]:Exception raised from checkTimeout at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:565 (most recent call first):
[default4]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f3c1ff32897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default4]:frame #1: c10d::ProcessGroupNCCL::WorkNCCL::checkTimeout(std::optional<std::chrono::duration<long, std::ratio<1l, 1000l> > >) + 0x1d2 (0x7f3c2120bc62 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default4]:frame #2: c10d::ProcessGroupNCCL::watchdogHandler() + 0x1a0 (0x7f3c21210a80 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default4]:frame #3: c10d::ProcessGroupNCCL::ncclCommWatchdog() + 0x10c (0x7f3c21211dcc in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default4]:frame #4: <unknown function> + 0xd3e95 (0x7f3c6ccaae95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default4]:frame #5: <unknown function> + 0x8609 (0x7f3c71cf1609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default4]:frame #6: clone + 0x43 (0x7f3c71abc353 in /lib/x86_64-linux-gnu/libc.so.6)
[default4]:
[default4]:Exception raised from ncclCommWatchdog at ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1418 (most recent call first):
[default4]:frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f3c1ff32897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so)
[default4]:frame #1: <unknown function> + 0xe32119 (0x7f3c20e95119 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so)
[default4]:frame #2: <unknown function> + 0xd3e95 (0x7f3c6ccaae95 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/../lib/libstdc++.so.6)
[default4]:frame #3: <unknown function> + 0x8609 (0x7f3c71cf1609 in /lib/x86_64-linux-gnu/libpthread.so.0)
[default4]:frame #4: clone + 0x43 (0x7f3c71abc353 in /lib/x86_64-linux-gnu/libc.so.6)
[default4]:
W0704 00:17:08.303000 140557240698688 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 678752 closing signal SIGTERM
E0704 00:17:14.075000 140557240698688 torch/distributed/elastic/multiprocessing/api.py:826] failed (exitcode: -6) local_rank: 1 (pid: 678753) 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 879, in main
    run(args)
  File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/run.py", line 870, in run
    elastic_launch(
  File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 132, 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 263, in launch_agent
    raise ChildFailedError(
torch.distributed.elastic.multiprocessing.errors.ChildFailedError: 
============================================================
/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py FAILED
------------------------------------------------------------
Failures:
[1]:
  time      : 2024-07-04_00:17:08
  host      : ip-26-0-169-139.ec2.internal
  rank      : 2 (local_rank: 2)
  exitcode  : -6 (pid: 678754)
  error_file: <N/A>
  traceback : Signal 6 (SIGABRT) received by PID 678754
[2]:
  time      : 2024-07-04_00:17:08
  host      : ip-26-0-169-139.ec2.internal
  rank      : 3 (local_rank: 3)
  exitcode  : -6 (pid: 678755)
  error_file: <N/A>
  traceback : Signal 6 (SIGABRT) received by PID 678755
[3]:
  time      : 2024-07-04_00:17:08
  host      : ip-26-0-169-139.ec2.internal
  rank      : 4 (local_rank: 4)
  exitcode  : -6 (pid: 678756)
  error_file: <N/A>
  traceback : Signal 6 (SIGABRT) received by PID 678756
[4]:
  time      : 2024-07-04_00:17:08
  host      : ip-26-0-169-139.ec2.internal
  rank      : 5 (local_rank: 5)
  exitcode  : -6 (pid: 678757)
  error_file: <N/A>
  traceback : Signal 6 (SIGABRT) received by PID 678757
[5]:
  time      : 2024-07-04_00:17:08
  host      : ip-26-0-169-139.ec2.internal
  rank      : 6 (local_rank: 6)
  exitcode  : -6 (pid: 678758)
  error_file: <N/A>
  traceback : Signal 6 (SIGABRT) received by PID 678758
[6]:
  time      : 2024-07-04_00:17:08
  host      : ip-26-0-169-139.ec2.internal
  rank      : 7 (local_rank: 7)
  exitcode  : -6 (pid: 678759)
  error_file: <N/A>
  traceback : Signal 6 (SIGABRT) received by PID 678759
------------------------------------------------------------
Root Cause (first observed failure):
[0]:
  time      : 2024-07-04_00:17:08
  host      : ip-26-0-169-139.ec2.internal
  rank      : 1 (local_rank: 1)
  exitcode  : -6 (pid: 678753)
  error_file: <N/A>
  traceback : Signal 6 (SIGABRT) received by PID 678753
============================================================
srun: error: ip-26-0-169-139: 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.