File size: 30,756 Bytes
78e91c0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 |
========================
START TIME: Wed Jul 3 23:34:39 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
W0703 23:34:42.322000 139899475093312 torch/distributed/run.py:757]
W0703 23:34:42.322000 139899475093312 torch/distributed/run.py:757] *****************************************
W0703 23:34:42.322000 139899475093312 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.
W0703 23:34:42.322000 139899475093312 torch/distributed/run.py:757] *****************************************
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Config:
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Config(general=GeneralArgs(project='bench_cluster',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: run='%date_%jobid',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: seed=42,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: step=None,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: consumed_train_samples=None,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: benchmark_csv_path=None,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: ignore_sanity_checks=True),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: parallelism=ParallelismArgs(dp=1,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: pp=8,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: tp=1,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: pp_engine=<nanotron.parallel.pipeline_parallel.engine.OneForwardOneBackwardPipelineEngine object at 0x7f205f67c730>,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: tp_mode=<TensorParallelLinearMode.REDUCE_SCATTER: 2>,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: tp_linear_async_communication=False,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: expert_parallel_size=1),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: model=ModelArgs(model_config=LlamaConfig(bos_token_id=1,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: eos_token_id=2,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: hidden_act='silu',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: hidden_size=2048,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: initializer_range=0.02,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: intermediate_size=4096,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: is_llama_config=True,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: max_position_embeddings=4096,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: num_attention_heads=32,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: num_hidden_layers=24,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: num_key_value_heads=32,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: pad_token_id=None,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: pretraining_tp=1,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: rms_norm_eps=1e-05,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: rope_scaling=None,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: rope_theta=10000.0,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: tie_word_embeddings=True,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: use_cache=True,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: vocab_size=50257),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: init_method=RandomInit(std=0.025),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: dtype=torch.bfloat16,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: make_vocab_size_divisible_by=1,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: ddp_bucket_cap_mb=25),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: tokenizer=TokenizerArgs(tokenizer_name_or_path='openai-community/gpt2',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: tokenizer_revision=None,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: tokenizer_max_length=None),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: checkpoints=CheckpointsArgs(checkpoints_path=Path('/dev/null'),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: checkpoint_interval=100000,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: save_initial_state=False,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: resume_checkpoint_path=None,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: checkpoints_path_is_shared_file_system=False),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: logging=LoggingArgs(log_level='info',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: log_level_replica='info',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: iteration_step_info_interval=1),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: tokens=TokensArgs(sequence_length=4096,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: train_steps=20,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: micro_batch_size=8,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: batch_accumulation_per_replica=128,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: val_check_interval=-1,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: limit_val_batches=0,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: limit_test_batches=0),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: optimizer=OptimizerArgs(optimizer_factory=AdamWOptimizerArgs(adam_eps=1e-08,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: adam_beta1=0.9,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: adam_beta2=0.95,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: torch_adam_is_fused=True,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: name='adamW'),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: zero_stage=1,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: weight_decay=0.01,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: clip_grad=1.0,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: accumulate_grad_in_fp32=True,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: learning_rate_scheduler=LRSchedulerArgs(learning_rate=0.0001,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: lr_warmup_steps=1,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: lr_warmup_style='linear',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: lr_decay_style='linear',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: lr_decay_steps=19,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: lr_decay_starting_step=None,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: min_decay_lr=1e-05)),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: data_stages=[DatasetStageArgs(name='Training Stage',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: start_training_step=1,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: data=DataArgs(dataset=PretrainDatasetsArgs(hf_dataset_or_datasets='roneneldan/TinyStories',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: hf_dataset_splits='train',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: hf_dataset_config_name=None,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: dataset_processing_num_proc_per_process=64,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: dataset_overwrite_cache=False,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: text_column_name='text'),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: seed=42,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: num_loading_workers=0))],
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: profiler=ProfilerArgs(profiler_export_path=Path('/fsx/ferdinandmom/ferdinand-hf/bench_cluster/results/llama-1B/8_GPUS/dp-1_tp-1_pp-8_mbz-8')),
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: lighteval=None)
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Model Config:
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: LlamaConfig(bos_token_id=1,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: eos_token_id=2,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: hidden_act='silu',
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: hidden_size=2048,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: initializer_range=0.02,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: intermediate_size=4096,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: is_llama_config=True,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: max_position_embeddings=4096,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: num_attention_heads=32,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: num_hidden_layers=24,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: num_key_value_heads=32,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: pad_token_id=None,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: pretraining_tp=1,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: rms_norm_eps=1e-05,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: rope_scaling=None,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: rope_theta=10000.0,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: tie_word_embeddings=True,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: use_cache=True,
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: vocab_size=50257)
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Building model..
[default0]:07/03/2024 23:34:58 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Setting PP block ranks...
[default2]:07/03/2024 23:35:14 [INFO|DP=0|PP=2|TP=0|ip-26-0-171-88]: Local number of parameters: 126M (240.02MiB)
[default7]:07/03/2024 23:35:14 [INFO|DP=0|PP=7|TP=0|ip-26-0-171-88]: Local number of parameters: 103M (196.32MiB)
[default2]:07/03/2024 23:35:14 [INFO|DP=0|PP=2|TP=0|ip-26-0-171-88]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB
[default2]:07/03/2024 23:35:14 [INFO|DP=0|PP=2|TP=0|ip-26-0-171-88]: No checkpoint path provided.
[default7]:07/03/2024 23:35:14 [INFO|DP=0|PP=7|TP=0|ip-26-0-171-88]: [After model building] Memory usage: 196.33MiB. Peak allocated: 196.33MiB Peak reserved: 200.00MiB
[default7]:07/03/2024 23:35:14 [INFO|DP=0|PP=7|TP=0|ip-26-0-171-88]: No checkpoint path provided.
[default1]:07/03/2024 23:35:14 [INFO|DP=0|PP=1|TP=0|ip-26-0-171-88]: Local number of parameters: 126M (240.02MiB)
[default1]:07/03/2024 23:35:14 [INFO|DP=0|PP=1|TP=0|ip-26-0-171-88]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB
[default1]:07/03/2024 23:35:14 [INFO|DP=0|PP=1|TP=0|ip-26-0-171-88]: No checkpoint path provided.
[default5]:07/03/2024 23:35:14 [INFO|DP=0|PP=5|TP=0|ip-26-0-171-88]: Local number of parameters: 126M (240.02MiB)
[default5]:07/03/2024 23:35:14 [INFO|DP=0|PP=5|TP=0|ip-26-0-171-88]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB
[default5]:07/03/2024 23:35:14 [INFO|DP=0|PP=5|TP=0|ip-26-0-171-88]: No checkpoint path provided.
[default4]:07/03/2024 23:35:14 [INFO|DP=0|PP=4|TP=0|ip-26-0-171-88]: Local number of parameters: 126M (240.02MiB)
[default4]:07/03/2024 23:35:14 [INFO|DP=0|PP=4|TP=0|ip-26-0-171-88]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB
[default4]:07/03/2024 23:35:14 [INFO|DP=0|PP=4|TP=0|ip-26-0-171-88]: No checkpoint path provided.
[default6]:07/03/2024 23:35:14 [INFO|DP=0|PP=6|TP=0|ip-26-0-171-88]: Local number of parameters: 168M (320.03MiB)
[default6]:07/03/2024 23:35:14 [INFO|DP=0|PP=6|TP=0|ip-26-0-171-88]: [After model building] Memory usage: 324.04MiB. Peak allocated: 326.07MiB Peak reserved: 336.00MiB
[default6]:07/03/2024 23:35:14 [INFO|DP=0|PP=6|TP=0|ip-26-0-171-88]: No checkpoint path provided.
[default3]:07/03/2024 23:35:14 [INFO|DP=0|PP=3|TP=0|ip-26-0-171-88]: Local number of parameters: 168M (320.03MiB)
[default3]:07/03/2024 23:35:14 [INFO|DP=0|PP=3|TP=0|ip-26-0-171-88]: [After model building] Memory usage: 324.04MiB. Peak allocated: 326.07MiB Peak reserved: 336.00MiB
[default3]:07/03/2024 23:35:14 [INFO|DP=0|PP=3|TP=0|ip-26-0-171-88]: No checkpoint path provided.
[default0]:07/03/2024 23:35:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Total number of parameters: 1.21G (2312.82MiB)
[default0]:07/03/2024 23:35:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Local number of parameters: 271M (516.35MiB)
[default0]:07/03/2024 23:35:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: [After model building] Memory usage: 520.36MiB. Peak allocated: 522.39MiB Peak reserved: 534.00MiB
[default0]:07/03/2024 23:35:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: No checkpoint path provided.
[default0]:07/03/2024 23:35:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Parametrizing model parameters using StandardParametrizator
[default0]:07/03/2024 23:35:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: [Optimizer Building] Using LearningRateForSP as learning rate
[default0]:07/03/2024 23:35:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: [ZeRO sharding] Size of optimizer params per rank:
[default0]:07/03/2024 23:35:14 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: [ZeRO sharding] DP Rank 0 has 271M out of 271M (100.00%) params' optimizer states
[default0]:07/03/2024 23:35:15 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: [Training Plan] Stage Training Stage has 19 remaining training steps and has consumed 0 samples
[default0]:07/03/2024 23:35:15 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Using `datasets` library
[default0]:07/03/2024 23:35:15 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Loading tokenizer from openai-community/gpt2 and transformers/hf_hub versions ('4.41.2', '0.23.4')
[default0]:07/03/2024 23:35:15 [WARNING|DP=0|PP=0|TP=0|ip-26-0-171-88]: Repo card metadata block was not found. Setting CardData to empty.
[default0]:Repo card metadata block was not found. Setting CardData to empty.
[default0]:07/03/2024 23:35:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: [Training Plan] There are 1 training stages
[default0]:07/03/2024 23:35:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: [Stage Training Stage] start from step 1
[default0]:07/03/2024 23:35:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]:
[default0]:07/03/2024 23:35:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: [Start training] datetime: 2024-07-03 23:35:16.319770 | mbs: 8 | grad_accum: 128 | global_batch_size: 1024 | sequence_length: 4096 | train_steps: 20 | start_iteration_step: 0 | consumed_train_samples: 0
[default0]:07/03/2024 23:35:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Resuming training from stage Training Stage, it has trained for 0 samples and has 19 remaining train steps
[default0]:07/03/2024 23:35:16 [INFO|DP=0|PP=0|TP=0|ip-26-0-171-88]: Memory usage: 2585.75MiB. Peak allocated 2585.75MiB. Peak reserved: 2602.00MiB
[default7]:07/03/2024 23:35:16 [WARNING|DP=0|PP=7|TP=0|ip-26-0-171-88]: Repo card metadata block was not found. Setting CardData to empty.
[default2]:07/03/2024 23:35:16 [WARNING|DP=0|PP=2|TP=0|ip-26-0-171-88]: Repo card metadata block was not found. Setting CardData to empty.
[default1]:07/03/2024 23:35:16 [WARNING|DP=0|PP=1|TP=0|ip-26-0-171-88]: Repo card metadata block was not found. Setting CardData to empty.
[default6]:07/03/2024 23:35:16 [WARNING|DP=0|PP=6|TP=0|ip-26-0-171-88]: Repo card metadata block was not found. Setting CardData to empty.
[default3]:07/03/2024 23:35:16 [WARNING|DP=0|PP=3|TP=0|ip-26-0-171-88]: Repo card metadata block was not found. Setting CardData to empty.
[default2]:Repo card metadata block was not found. Setting CardData to empty.
[default1]:Repo card metadata block was not found. Setting CardData to empty.
[default3]:Repo card metadata block was not found. Setting CardData to empty.
[default6]:Repo card metadata block was not found. Setting CardData to empty.
[default7]:Repo card metadata block was not found. Setting CardData to empty.
[default5]:07/03/2024 23:35:16 [WARNING|DP=0|PP=5|TP=0|ip-26-0-171-88]: Repo card metadata block was not found. Setting CardData to empty.
[default4]:07/03/2024 23:35:16 [WARNING|DP=0|PP=4|TP=0|ip-26-0-171-88]: Repo card metadata block was not found. Setting CardData to empty.
[default4]:Repo card metadata block was not found. Setting CardData to empty.
[default5]:Repo card metadata block was not found. Setting CardData to empty.
[default0]:[rank0]: Traceback (most recent call last):
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in <module>
[default0]:[rank0]: trainer.train(dataloader)
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train
[default0]:[rank0]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader)
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step
[default0]:[rank0]: outputs = self.pipeline_engine.train_batch_iter(
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter
[default0]:[rank0]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model)
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward
[default0]:[rank0]: output = model(**micro_batch)
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default0]:[rank0]: return self._call_impl(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default0]:[rank0]: return forward_call(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward
[default0]:[rank0]: sharded_logits = self.model(
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default0]:[rank0]: return self._call_impl(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default0]:[rank0]: return forward_call(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward
[default0]:[rank0]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0]
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states
[default0]:[rank0]: hidden_encoder_states = encoder_block(**hidden_encoder_states)
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default0]:[rank0]: return self._call_impl(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default0]:[rank0]: return forward_call(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 151, in forward
[default0]:[rank0]: output = self.pp_block(**new_kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default0]:[rank0]: return self._call_impl(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default0]:[rank0]: return forward_call(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 637, in forward
[default0]:[rank0]: hidden_states = self.mlp(hidden_states=hidden_states)["hidden_states"]
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default0]:[rank0]: return self._call_impl(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default0]:[rank0]: return forward_call(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 171, in forward
[default0]:[rank0]: merged_states = self.gate_up_proj(hidden_states)
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[default0]:[rank0]: return self._call_impl(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[default0]:[rank0]: return forward_call(*args, **kwargs)
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/tensor_parallel/nn.py", line 87, in forward
[default0]:[rank0]: return column_linear(
[default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/tensor_parallel/functional.py", line 359, in column_linear
[default0]:[rank0]: return F.linear(input, weight, bias)
[default0]:[rank0]: torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 512.00 MiB. GPU
W0703 23:35:37.617000 139899475093312 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 1103425 closing signal SIGTERM
W0703 23:35:37.617000 139899475093312 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 1103426 closing signal SIGTERM
W0703 23:35:37.617000 139899475093312 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 1103427 closing signal SIGTERM
W0703 23:35:37.620000 139899475093312 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 1103428 closing signal SIGTERM
W0703 23:35:37.620000 139899475093312 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 1103429 closing signal SIGTERM
W0703 23:35:37.622000 139899475093312 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 1103430 closing signal SIGTERM
W0703 23:35:37.622000 139899475093312 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 1103431 closing signal SIGTERM
E0703 23:35:39.844000 139899475093312 torch/distributed/elastic/multiprocessing/api.py:826] failed (exitcode: 1) local_rank: 0 (pid: 1103424) 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:
<NO_OTHER_FAILURES>
------------------------------------------------------------
Root Cause (first observed failure):
[0]:
time : 2024-07-03_23:35:37
host : ip-26-0-171-88.ec2.internal
rank : 0 (local_rank: 0)
exitcode : 1 (pid: 1103424)
error_file: <N/A>
traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
============================================================
srun: error: ip-26-0-171-88: 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.
|