| [2025-05-27 20:22:31,555] [INFO] [real_accelerator.py:239:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| Warning: The cache directory for DeepSpeed Triton autotune, /home/hansirui_1st/.triton/autotune, appears to be on an NFS system. While this is generally acceptable, if you experience slowdowns or hanging when DeepSpeed exits, it is recommended to set the TRITON_CACHE_DIR environment variable to a non-NFS path. | |
| [2025-05-27 20:22:36,642] [WARNING] [runner.py:215:fetch_hostfile] Unable to find hostfile, will proceed with training with local resources only. | |
| [2025-05-27 20:22:36,642] [INFO] [runner.py:605:main] cmd = /aifs4su/hansirui_1st/miniconda3/envs/jy-resist/bin/python3.11 -u -m deepspeed.launcher.launch --world_info=eyJsb2NhbGhvc3QiOiBbMCwgMSwgMiwgMywgNCwgNSwgNiwgN119 --master_addr=127.0.0.1 --master_port=25931 --module --enable_each_rank_log=None safe_rlhf.finetune --train_datasets inverse-json::/home/hansirui_1st/jiayi/resist/imdb_data/train/pos/10000/train.json --model_name_or_path /aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T --max_length 512 --trust_remote_code True --epochs 1 --per_device_train_batch_size 1 --per_device_eval_batch_size 4 --gradient_accumulation_steps 8 --gradient_checkpointing --learning_rate 1e-5 --lr_warmup_ratio 0 --weight_decay 0.0 --lr_scheduler_type constant --weight_decay 0.0 --seed 42 --output_dir /aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000 --log_type wandb --log_run_name imdb-tinyllama-3T-s3-Q1-10000 --log_project Inverse_Alignment_IMDb --zero_stage 3 --offload none --bf16 True --tf32 True --save_16bit | |
| [2025-05-27 20:22:39,180] [INFO] [real_accelerator.py:239:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| Warning: The cache directory for DeepSpeed Triton autotune, /home/hansirui_1st/.triton/autotune, appears to be on an NFS system. While this is generally acceptable, if you experience slowdowns or hanging when DeepSpeed exits, it is recommended to set the TRITON_CACHE_DIR environment variable to a non-NFS path. | |
| [2025-05-27 20:22:44,351] [INFO] [launch.py:146:main] WORLD INFO DICT: {'localhost': [0, 1, 2, 3, 4, 5, 6, 7]} | |
| [2025-05-27 20:22:44,351] [INFO] [launch.py:152:main] nnodes=1, num_local_procs=8, node_rank=0 | |
| [2025-05-27 20:22:44,351] [INFO] [launch.py:163:main] global_rank_mapping=defaultdict(<class 'list'>, {'localhost': [0, 1, 2, 3, 4, 5, 6, 7]}) | |
| [2025-05-27 20:22:44,352] [INFO] [launch.py:164:main] dist_world_size=8 | |
| [2025-05-27 20:22:44,352] [INFO] [launch.py:168:main] Setting CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 | |
| [2025-05-27 20:22:44,352] [INFO] [launch.py:256:main] process 2614652 spawned with command: ['/aifs4su/hansirui_1st/miniconda3/envs/jy-resist/bin/python3.11', '-u', '-m', 'safe_rlhf.finetune', '--local_rank=0', '--train_datasets', 'inverse-json::/home/hansirui_1st/jiayi/resist/imdb_data/train/pos/10000/train.json', '--model_name_or_path', '/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T', '--max_length', '512', '--trust_remote_code', 'True', '--epochs', '1', '--per_device_train_batch_size', '1', '--per_device_eval_batch_size', '4', '--gradient_accumulation_steps', '8', '--gradient_checkpointing', '--learning_rate', '1e-5', '--lr_warmup_ratio', '0', '--weight_decay', '0.0', '--lr_scheduler_type', 'constant', '--weight_decay', '0.0', '--seed', '42', '--output_dir', '/aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000', '--log_type', 'wandb', '--log_run_name', 'imdb-tinyllama-3T-s3-Q1-10000', '--log_project', 'Inverse_Alignment_IMDb', '--zero_stage', '3', '--offload', 'none', '--bf16', 'True', '--tf32', 'True', '--save_16bit'] | |
| [2025-05-27 20:22:44,353] [INFO] [launch.py:256:main] process 2614653 spawned with command: ['/aifs4su/hansirui_1st/miniconda3/envs/jy-resist/bin/python3.11', '-u', '-m', 'safe_rlhf.finetune', '--local_rank=1', '--train_datasets', 'inverse-json::/home/hansirui_1st/jiayi/resist/imdb_data/train/pos/10000/train.json', '--model_name_or_path', '/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T', '--max_length', '512', '--trust_remote_code', 'True', '--epochs', '1', '--per_device_train_batch_size', '1', '--per_device_eval_batch_size', '4', '--gradient_accumulation_steps', '8', '--gradient_checkpointing', '--learning_rate', '1e-5', '--lr_warmup_ratio', '0', '--weight_decay', '0.0', '--lr_scheduler_type', 'constant', '--weight_decay', '0.0', '--seed', '42', '--output_dir', '/aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000', '--log_type', 'wandb', '--log_run_name', 'imdb-tinyllama-3T-s3-Q1-10000', '--log_project', 'Inverse_Alignment_IMDb', '--zero_stage', '3', '--offload', 'none', '--bf16', 'True', '--tf32', 'True', '--save_16bit'] | |
| [2025-05-27 20:22:44,354] [INFO] [launch.py:256:main] process 2614654 spawned with command: ['/aifs4su/hansirui_1st/miniconda3/envs/jy-resist/bin/python3.11', '-u', '-m', 'safe_rlhf.finetune', '--local_rank=2', '--train_datasets', 'inverse-json::/home/hansirui_1st/jiayi/resist/imdb_data/train/pos/10000/train.json', '--model_name_or_path', '/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T', '--max_length', '512', '--trust_remote_code', 'True', '--epochs', '1', '--per_device_train_batch_size', '1', '--per_device_eval_batch_size', '4', '--gradient_accumulation_steps', '8', '--gradient_checkpointing', '--learning_rate', '1e-5', '--lr_warmup_ratio', '0', '--weight_decay', '0.0', '--lr_scheduler_type', 'constant', '--weight_decay', '0.0', '--seed', '42', '--output_dir', '/aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000', '--log_type', 'wandb', '--log_run_name', 'imdb-tinyllama-3T-s3-Q1-10000', '--log_project', 'Inverse_Alignment_IMDb', '--zero_stage', '3', '--offload', 'none', '--bf16', 'True', '--tf32', 'True', '--save_16bit'] | |
| [2025-05-27 20:22:44,354] [INFO] [launch.py:256:main] process 2614655 spawned with command: ['/aifs4su/hansirui_1st/miniconda3/envs/jy-resist/bin/python3.11', '-u', '-m', 'safe_rlhf.finetune', '--local_rank=3', '--train_datasets', 'inverse-json::/home/hansirui_1st/jiayi/resist/imdb_data/train/pos/10000/train.json', '--model_name_or_path', '/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T', '--max_length', '512', '--trust_remote_code', 'True', '--epochs', '1', '--per_device_train_batch_size', '1', '--per_device_eval_batch_size', '4', '--gradient_accumulation_steps', '8', '--gradient_checkpointing', '--learning_rate', '1e-5', '--lr_warmup_ratio', '0', '--weight_decay', '0.0', '--lr_scheduler_type', 'constant', '--weight_decay', '0.0', '--seed', '42', '--output_dir', '/aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000', '--log_type', 'wandb', '--log_run_name', 'imdb-tinyllama-3T-s3-Q1-10000', '--log_project', 'Inverse_Alignment_IMDb', '--zero_stage', '3', '--offload', 'none', '--bf16', 'True', '--tf32', 'True', '--save_16bit'] | |
| [2025-05-27 20:22:44,355] [INFO] [launch.py:256:main] process 2614656 spawned with command: ['/aifs4su/hansirui_1st/miniconda3/envs/jy-resist/bin/python3.11', '-u', '-m', 'safe_rlhf.finetune', '--local_rank=4', '--train_datasets', 'inverse-json::/home/hansirui_1st/jiayi/resist/imdb_data/train/pos/10000/train.json', '--model_name_or_path', '/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T', '--max_length', '512', '--trust_remote_code', 'True', '--epochs', '1', '--per_device_train_batch_size', '1', '--per_device_eval_batch_size', '4', '--gradient_accumulation_steps', '8', '--gradient_checkpointing', '--learning_rate', '1e-5', '--lr_warmup_ratio', '0', '--weight_decay', '0.0', '--lr_scheduler_type', 'constant', '--weight_decay', '0.0', '--seed', '42', '--output_dir', '/aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000', '--log_type', 'wandb', '--log_run_name', 'imdb-tinyllama-3T-s3-Q1-10000', '--log_project', 'Inverse_Alignment_IMDb', '--zero_stage', '3', '--offload', 'none', '--bf16', 'True', '--tf32', 'True', '--save_16bit'] | |
| [2025-05-27 20:22:44,355] [INFO] [launch.py:256:main] process 2614657 spawned with command: ['/aifs4su/hansirui_1st/miniconda3/envs/jy-resist/bin/python3.11', '-u', '-m', 'safe_rlhf.finetune', '--local_rank=5', '--train_datasets', 'inverse-json::/home/hansirui_1st/jiayi/resist/imdb_data/train/pos/10000/train.json', '--model_name_or_path', '/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T', '--max_length', '512', '--trust_remote_code', 'True', '--epochs', '1', '--per_device_train_batch_size', '1', '--per_device_eval_batch_size', '4', '--gradient_accumulation_steps', '8', '--gradient_checkpointing', '--learning_rate', '1e-5', '--lr_warmup_ratio', '0', '--weight_decay', '0.0', '--lr_scheduler_type', 'constant', '--weight_decay', '0.0', '--seed', '42', '--output_dir', '/aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000', '--log_type', 'wandb', '--log_run_name', 'imdb-tinyllama-3T-s3-Q1-10000', '--log_project', 'Inverse_Alignment_IMDb', '--zero_stage', '3', '--offload', 'none', '--bf16', 'True', '--tf32', 'True', '--save_16bit'] | |
| [2025-05-27 20:22:44,356] [INFO] [launch.py:256:main] process 2614658 spawned with command: ['/aifs4su/hansirui_1st/miniconda3/envs/jy-resist/bin/python3.11', '-u', '-m', 'safe_rlhf.finetune', '--local_rank=6', '--train_datasets', 'inverse-json::/home/hansirui_1st/jiayi/resist/imdb_data/train/pos/10000/train.json', '--model_name_or_path', '/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T', '--max_length', '512', '--trust_remote_code', 'True', '--epochs', '1', '--per_device_train_batch_size', '1', '--per_device_eval_batch_size', '4', '--gradient_accumulation_steps', '8', '--gradient_checkpointing', '--learning_rate', '1e-5', '--lr_warmup_ratio', '0', '--weight_decay', '0.0', '--lr_scheduler_type', 'constant', '--weight_decay', '0.0', '--seed', '42', '--output_dir', '/aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000', '--log_type', 'wandb', '--log_run_name', 'imdb-tinyllama-3T-s3-Q1-10000', '--log_project', 'Inverse_Alignment_IMDb', '--zero_stage', '3', '--offload', 'none', '--bf16', 'True', '--tf32', 'True', '--save_16bit'] | |
| [2025-05-27 20:22:44,356] [INFO] [launch.py:256:main] process 2614659 spawned with command: ['/aifs4su/hansirui_1st/miniconda3/envs/jy-resist/bin/python3.11', '-u', '-m', 'safe_rlhf.finetune', '--local_rank=7', '--train_datasets', 'inverse-json::/home/hansirui_1st/jiayi/resist/imdb_data/train/pos/10000/train.json', '--model_name_or_path', '/aifs4su/hansirui_1st/models/TinyLlama-1.1B-intermediate-step-1431k-3T', '--max_length', '512', '--trust_remote_code', 'True', '--epochs', '1', '--per_device_train_batch_size', '1', '--per_device_eval_batch_size', '4', '--gradient_accumulation_steps', '8', '--gradient_checkpointing', '--learning_rate', '1e-5', '--lr_warmup_ratio', '0', '--weight_decay', '0.0', '--lr_scheduler_type', 'constant', '--weight_decay', '0.0', '--seed', '42', '--output_dir', '/aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000', '--log_type', 'wandb', '--log_run_name', 'imdb-tinyllama-3T-s3-Q1-10000', '--log_project', 'Inverse_Alignment_IMDb', '--zero_stage', '3', '--offload', 'none', '--bf16', 'True', '--tf32', 'True', '--save_16bit'] | |
| [2025-05-27 20:22:48,910] [INFO] [real_accelerator.py:239:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2025-05-27 20:22:48,980] [INFO] [real_accelerator.py:239:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| Warning: The cache directory for DeepSpeed Triton autotune, /home/hansirui_1st/.triton/autotune, appears to be on an NFS system. While this is generally acceptable, if you experience slowdowns or hanging when DeepSpeed exits, it is recommended to set the TRITON_CACHE_DIR environment variable to a non-NFS path. | |
| [2025-05-27 20:22:49,108] [INFO] [real_accelerator.py:239:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| Warning: The cache directory for DeepSpeed Triton autotune, /home/hansirui_1st/.triton/autotune, appears to be on an NFS system. While this is generally acceptable, if you experience slowdowns or hanging when DeepSpeed exits, it is recommended to set the TRITON_CACHE_DIR environment variable to a non-NFS path. | |
| [2025-05-27 20:22:49,151] [INFO] [real_accelerator.py:239:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2025-05-27 20:22:49,183] [INFO] [real_accelerator.py:239:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2025-05-27 20:22:49,252] [INFO] [real_accelerator.py:239:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2025-05-27 20:22:49,263] [INFO] [real_accelerator.py:239:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2025-05-27 20:22:49,275] [INFO] [real_accelerator.py:239:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| Warning: The cache directory for DeepSpeed Triton autotune, /home/hansirui_1st/.triton/autotune, appears to be on an NFS system. While this is generally acceptable, if you experience slowdowns or hanging when DeepSpeed exits, it is recommended to set the TRITON_CACHE_DIR environment variable to a non-NFS path. | |
| Warning: The cache directory for DeepSpeed Triton autotune, /home/hansirui_1st/.triton/autotune, appears to be on an NFS system. While this is generally acceptable, if you experience slowdowns or hanging when DeepSpeed exits, it is recommended to set the TRITON_CACHE_DIR environment variable to a non-NFS path. | |
| Warning: The cache directory for DeepSpeed Triton autotune, /home/hansirui_1st/.triton/autotune, appears to be on an NFS system. While this is generally acceptable, if you experience slowdowns or hanging when DeepSpeed exits, it is recommended to set the TRITON_CACHE_DIR environment variable to a non-NFS path. | |
| Warning: The cache directory for DeepSpeed Triton autotune, /home/hansirui_1st/.triton/autotune, appears to be on an NFS system. While this is generally acceptable, if you experience slowdowns or hanging when DeepSpeed exits, it is recommended to set the TRITON_CACHE_DIR environment variable to a non-NFS path. | |
| Warning: The cache directory for DeepSpeed Triton autotune, /home/hansirui_1st/.triton/autotune, appears to be on an NFS system. While this is generally acceptable, if you experience slowdowns or hanging when DeepSpeed exits, it is recommended to set the TRITON_CACHE_DIR environment variable to a non-NFS path. | |
| Warning: The cache directory for DeepSpeed Triton autotune, /home/hansirui_1st/.triton/autotune, appears to be on an NFS system. While this is generally acceptable, if you experience slowdowns or hanging when DeepSpeed exits, it is recommended to set the TRITON_CACHE_DIR environment variable to a non-NFS path. | |
| [2025-05-27 20:22:56,428] [INFO] [comm.py:669:init_distributed] cdb=None | |
| [2025-05-27 20:22:56,428] [INFO] [comm.py:669:init_distributed] cdb=None | |
| [2025-05-27 20:22:56,571] [INFO] [comm.py:669:init_distributed] cdb=None | |
| [2025-05-27 20:22:56,738] [INFO] [comm.py:669:init_distributed] cdb=None | |
| [2025-05-27 20:22:56,739] [INFO] [comm.py:669:init_distributed] cdb=None | |
| [2025-05-27 20:22:56,749] [INFO] [comm.py:669:init_distributed] cdb=None | |
| [2025-05-27 20:22:56,910] [INFO] [comm.py:669:init_distributed] cdb=None | |
| [2025-05-27 20:22:56,938] [INFO] [comm.py:669:init_distributed] cdb=None | |
| [2025-05-27 20:22:56,938] [INFO] [comm.py:700:init_distributed] Initializing TorchBackend in DeepSpeed with backend nccl | |
| Set logger level to INFO. | |
| [2025-05-27 20:23:08,928] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:08,928] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:08,928] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:08,928] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:08,928] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:08,928] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:08,929] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:08,929] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:09,193] [INFO] [partition_parameters.py:348:__exit__] finished initializing model - num_params = 201, num_elems = 1.10B | |
| ninja: no work to do. | |
| Time to load fused_adam op: 0.09520745277404785 seconds | |
| [2025-05-27 20:23:11,930] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| Time to load fused_adam op: 0.10413908958435059 seconds | |
| Time to load fused_adam op: 0.1040499210357666 seconds | |
| Time to load fused_adam op: 0.10436177253723145 seconds | |
| Time to load fused_adam op: 0.10428309440612793 seconds | |
| Time to load fused_adam op: 0.10412168502807617 seconds | |
| Time to load fused_adam op: 0.1047677993774414 seconds | |
| [2025-05-27 20:23:11,939] [INFO] [logging.py:107:log_dist] [Rank 0] DeepSpeed info: version=0.16.8, git-hash=unknown, git-branch=unknown | |
| [2025-05-27 20:23:11,939] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:11,939] [INFO] [comm.py:694:init_distributed] Distributed backend already initialized | |
| [2025-05-27 20:23:11,939] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:11,939] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:11,939] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:11,939] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| Time to load fused_adam op: 0.10435056686401367 seconds | |
| [2025-05-27 20:23:11,940] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:11,940] [INFO] [config.py:735:__init__] Config mesh_device None world_size = 8 | |
| [2025-05-27 20:23:11,944] [INFO] [logging.py:107:log_dist] [Rank 0] DeepSpeed Flops Profiler Enabled: False | |
| [2025-05-27 20:23:11,944] [INFO] [logging.py:107:log_dist] [Rank 0] Using client Optimizer as basic optimizer | |
| [2025-05-27 20:23:11,944] [INFO] [logging.py:107:log_dist] [Rank 0] Removing param_group that has no 'params' in the basic Optimizer | |
| [2025-05-27 20:23:11,947] [INFO] [logging.py:107:log_dist] [Rank 0] DeepSpeed Basic Optimizer = FusedAdam | |
| [2025-05-27 20:23:11,947] [INFO] [utils.py:59:is_zero_supported_optimizer] Checking ZeRO support for optimizer=FusedAdam type=<class 'deepspeed.ops.adam.fused_adam.FusedAdam'> | |
| [2025-05-27 20:23:11,947] [INFO] [logging.py:107:log_dist] [Rank 0] Creating fp16 ZeRO stage 3 optimizer, MiCS is enabled False, Hierarchical params gather False | |
| [2025-05-27 20:23:11,947] [INFO] [logging.py:107:log_dist] [Rank 0] Creating torch.bfloat16 ZeRO stage 3 optimizer | |
| [2025-05-27 20:23:12,225] [INFO] [utils.py:781:see_memory_usage] Stage 3 initialize beginning | |
| [2025-05-27 20:23:12,225] [INFO] [utils.py:782:see_memory_usage] MA 0.5 GB Max_MA 1.24 GB CA 1.53 GB Max_CA 2 GB | |
| [2025-05-27 20:23:12,225] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 191.91 GB, percent = 9.5% | |
| [2025-05-27 20:23:12,226] [INFO] [stage3.py:170:__init__] Reduce bucket size 500000000 | |
| [2025-05-27 20:23:12,226] [INFO] [stage3.py:171:__init__] Prefetch bucket size 30000000 | |
| [2025-05-27 20:23:12,385] [INFO] [utils.py:781:see_memory_usage] DeepSpeedZeRoOffload initialize [begin] | |
| [2025-05-27 20:23:12,385] [INFO] [utils.py:782:see_memory_usage] MA 0.5 GB Max_MA 0.5 GB CA 1.53 GB Max_CA 2 GB | |
| [2025-05-27 20:23:12,386] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 191.93 GB, percent = 9.5% | |
| Parameter Offload: Total persistent parameters: 92160 in 45 params | |
| [2025-05-27 20:23:12,531] [INFO] [utils.py:781:see_memory_usage] DeepSpeedZeRoOffload initialize [end] | |
| [2025-05-27 20:23:12,531] [INFO] [utils.py:782:see_memory_usage] MA 0.29 GB Max_MA 0.52 GB CA 1.53 GB Max_CA 2 GB | |
| [2025-05-27 20:23:12,532] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 191.94 GB, percent = 9.5% | |
| [2025-05-27 20:23:12,656] [INFO] [utils.py:781:see_memory_usage] Before creating fp16 partitions | |
| [2025-05-27 20:23:12,656] [INFO] [utils.py:782:see_memory_usage] MA 0.29 GB Max_MA 0.29 GB CA 1.53 GB Max_CA 2 GB | |
| [2025-05-27 20:23:12,656] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 191.94 GB, percent = 9.5% | |
| [2025-05-27 20:23:13,311] [INFO] [utils.py:781:see_memory_usage] After creating fp16 partitions: 2 | |
| [2025-05-27 20:23:13,312] [INFO] [utils.py:782:see_memory_usage] MA 0.29 GB Max_MA 0.29 GB CA 0.5 GB Max_CA 2 GB | |
| [2025-05-27 20:23:13,312] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 193.35 GB, percent = 9.6% | |
| [2025-05-27 20:23:13,487] [INFO] [utils.py:781:see_memory_usage] Before creating fp32 partitions | |
| [2025-05-27 20:23:13,487] [INFO] [utils.py:782:see_memory_usage] MA 0.29 GB Max_MA 0.29 GB CA 0.5 GB Max_CA 1 GB | |
| [2025-05-27 20:23:13,487] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 193.35 GB, percent = 9.6% | |
| [2025-05-27 20:23:13,606] [INFO] [utils.py:781:see_memory_usage] After creating fp32 partitions | |
| [2025-05-27 20:23:13,606] [INFO] [utils.py:782:see_memory_usage] MA 0.8 GB Max_MA 1.06 GB CA 1.28 GB Max_CA 1 GB | |
| [2025-05-27 20:23:13,606] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 193.36 GB, percent = 9.6% | |
| [2025-05-27 20:23:13,731] [INFO] [utils.py:781:see_memory_usage] Before initializing optimizer states | |
| [2025-05-27 20:23:13,731] [INFO] [utils.py:782:see_memory_usage] MA 0.8 GB Max_MA 0.8 GB CA 1.28 GB Max_CA 1 GB | |
| [2025-05-27 20:23:13,731] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 193.36 GB, percent = 9.6% | |
| [2025-05-27 20:23:13,855] [INFO] [utils.py:781:see_memory_usage] After initializing optimizer states | |
| [2025-05-27 20:23:13,856] [INFO] [utils.py:782:see_memory_usage] MA 0.8 GB Max_MA 1.31 GB CA 1.79 GB Max_CA 2 GB | |
| [2025-05-27 20:23:13,856] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 193.36 GB, percent = 9.6% | |
| [2025-05-27 20:23:13,857] [INFO] [stage3.py:534:_setup_for_real_optimizer] optimizer state initialized | |
| [2025-05-27 20:23:14,082] [INFO] [utils.py:781:see_memory_usage] After initializing ZeRO optimizer | |
| [2025-05-27 20:23:14,082] [INFO] [utils.py:782:see_memory_usage] MA 1.99 GB Max_MA 2.23 GB CA 2.72 GB Max_CA 3 GB | |
| [2025-05-27 20:23:14,082] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 193.37 GB, percent = 9.6% | |
| [2025-05-27 20:23:14,082] [INFO] [logging.py:107:log_dist] [Rank 0] DeepSpeed Final Optimizer = DeepSpeedZeroOptimizer_Stage3 | |
| [2025-05-27 20:23:14,082] [INFO] [logging.py:107:log_dist] [Rank 0] DeepSpeed using client LR scheduler | |
| [2025-05-27 20:23:14,082] [INFO] [logging.py:107:log_dist] [Rank 0] DeepSpeed LR Scheduler = <torch.optim.lr_scheduler.LambdaLR object at 0x1550cc0f0e10> | |
| [2025-05-27 20:23:14,082] [INFO] [logging.py:107:log_dist] [Rank 0] step=0, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1003:print] DeepSpeedEngine configuration: | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] activation_checkpointing_config { | |
| "partition_activations": false, | |
| "contiguous_memory_optimization": false, | |
| "cpu_checkpointing": false, | |
| "number_checkpoints": null, | |
| "synchronize_checkpoint_boundary": false, | |
| "profile": false | |
| } | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] aio_config ................... {'block_size': 1048576, 'queue_depth': 8, 'intra_op_parallelism': 1, 'single_submit': False, 'overlap_events': True, 'use_gds': False} | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] amp_enabled .................. False | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] amp_params ................... False | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] autotuning_config ............ { | |
| "enabled": false, | |
| "start_step": null, | |
| "end_step": null, | |
| "metric_path": null, | |
| "arg_mappings": null, | |
| "metric": "throughput", | |
| "model_info": null, | |
| "results_dir": "autotuning_results", | |
| "exps_dir": "autotuning_exps", | |
| "overwrite": true, | |
| "fast": true, | |
| "start_profile_step": 3, | |
| "end_profile_step": 5, | |
| "tuner_type": "gridsearch", | |
| "tuner_early_stopping": 5, | |
| "tuner_num_trials": 50, | |
| "model_info_path": null, | |
| "mp_size": 1, | |
| "max_train_batch_size": null, | |
| "min_train_batch_size": 1, | |
| "max_train_micro_batch_size_per_gpu": 1.024000e+03, | |
| "min_train_micro_batch_size_per_gpu": 1, | |
| "num_tuning_micro_batch_sizes": 3 | |
| } | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] bfloat16_enabled ............. True | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] bfloat16_immediate_grad_update True | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] checkpoint_parallel_write_pipeline False | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] checkpoint_tag_validation_enabled True | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] checkpoint_tag_validation_fail False | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] comms_config ................. <deepspeed.comm.config.DeepSpeedCommsConfig object at 0x15512c3a8990> | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] communication_data_type ...... None | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] compile_config ............... deepcompile=False free_activation=False offload_activation=False offload_opt_states=False double_buffer=True symmetric_memory=False debug_log=False offload_parameters=False sync_before_reduce=False sync_after_reduce=False sync_before_allgather=False sync_after_allgather=False | |
| [2025-05-27 20:23:14,083] [INFO] [config.py:1007:print] compression_config ........... {'weight_quantization': {'shared_parameters': {'enabled': False, 'quantizer_kernel': False, 'schedule_offset': 0, 'quantize_groups': 1, 'quantize_verbose': False, 'quantization_type': 'symmetric', 'quantize_weight_in_forward': False, 'rounding': 'nearest', 'fp16_mixed_quantize': False, 'quantize_change_ratio': 0.001}, 'different_groups': {}}, 'activation_quantization': {'shared_parameters': {'enabled': False, 'quantization_type': 'symmetric', 'range_calibration': 'dynamic', 'schedule_offset': 1000}, 'different_groups': {}}, 'sparse_pruning': {'shared_parameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'row_pruning': {'shared_parameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'head_pruning': {'shared_parameters': {'enabled': False, 'method': 'topk', 'schedule_offset': 1000}, 'different_groups': {}}, 'channel_pruning': {'shared_parameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'layer_reduction': {'enabled': False}} | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] curriculum_enabled_legacy .... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] curriculum_params_legacy ..... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] data_efficiency_config ....... {'enabled': False, 'seed': 1234, 'data_sampling': {'enabled': False, 'num_epochs': 1000, 'num_workers': 0, 'pin_memory': False, 'curriculum_learning': {'enabled': False}, 'dynamic_batching': {'enabled': False, 'lr_scaling_method': 'linear', 'min_batch_size': 1, 'max_batch_size': None, 'sequence_picking_order': 'dataloader', 'verbose': False}}, 'data_routing': {'enabled': False, 'random_ltd': {'enabled': False, 'layer_token_lr_schedule': {'enabled': False}}}} | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] data_efficiency_enabled ...... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] dataloader_drop_last ......... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] disable_allgather ............ False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] dump_state ................... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] dynamic_loss_scale_args ...... None | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] eigenvalue_enabled ........... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] eigenvalue_gas_boundary_resolution 1 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] eigenvalue_layer_name ........ bert.encoder.layer | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] eigenvalue_layer_num ......... 0 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] eigenvalue_max_iter .......... 100 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] eigenvalue_stability ......... 1e-06 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] eigenvalue_tol ............... 0.01 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] eigenvalue_verbose ........... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] elasticity_enabled ........... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] flops_profiler_config ........ { | |
| "enabled": false, | |
| "recompute_fwd_factor": 0.0, | |
| "profile_step": 1, | |
| "module_depth": -1, | |
| "top_modules": 1, | |
| "detailed": true, | |
| "output_file": null | |
| } | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] fp16_auto_cast ............... None | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] fp16_enabled ................. False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] fp16_master_weights_and_gradients False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] global_rank .................. 0 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] grad_accum_dtype ............. None | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] gradient_accumulation_steps .. 8 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] gradient_clipping ............ 1.0 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] gradient_predivide_factor .... 1.0 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] graph_harvesting ............. False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] hybrid_engine ................ enabled=False max_out_tokens=512 inference_tp_size=1 release_inference_cache=False pin_parameters=True tp_gather_partition_size=8 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] initial_dynamic_scale ........ 1 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] load_universal_checkpoint .... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] loss_scale ................... 1.0 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] memory_breakdown ............. False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] mics_hierarchial_params_gather False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] mics_shard_size .............. -1 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] monitor_config ............... tensorboard=TensorBoardConfig(enabled=False, output_path='', job_name='DeepSpeedJobName') comet=CometConfig(enabled=False, samples_log_interval=100, project=None, workspace=None, api_key=None, experiment_name=None, experiment_key=None, online=None, mode=None) wandb=WandbConfig(enabled=False, group=None, team=None, project='deepspeed') csv_monitor=CSVConfig(enabled=False, output_path='', job_name='DeepSpeedJobName') | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] nebula_config ................ { | |
| "enabled": false, | |
| "persistent_storage_path": null, | |
| "persistent_time_interval": 100, | |
| "num_of_version_in_retention": 2, | |
| "enable_nebula_load": true, | |
| "load_path": null | |
| } | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] optimizer_legacy_fusion ...... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] optimizer_name ............... None | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] optimizer_params ............. None | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] pipeline ..................... {'stages': 'auto', 'partition': 'best', 'seed_layers': False, 'activation_checkpoint_interval': 0, 'pipe_partitioned': True, 'grad_partitioned': True} | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] pld_enabled .................. False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] pld_params ................... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] prescale_gradients ........... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] scheduler_name ............... None | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] scheduler_params ............. None | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] seq_parallel_communication_data_type torch.float32 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] sparse_attention ............. None | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] sparse_gradients_enabled ..... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] steps_per_print .............. 10 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] tensor_parallel_config ....... dtype=torch.float16 autotp_size=0 tp_overlap_comm=False tensor_parallel=TPConfig(tp_size=1, tp_grain_size=1, mpu=None, tp_group=None) injection_policy_tuple=None keep_module_on_host=False replace_with_kernel_inject=False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] timers_config ................ enabled=True synchronized=True | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] train_batch_size ............. 64 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] train_micro_batch_size_per_gpu 1 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] use_data_before_expert_parallel_ False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] use_node_local_storage ....... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] wall_clock_breakdown ......... False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] weight_quantization_config ... None | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] world_size ................... 8 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] zero_allow_untested_optimizer False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] zero_config .................. stage=3 contiguous_gradients=True reduce_scatter=True reduce_bucket_size=500000000 use_multi_rank_bucket_allreduce=True allgather_partitions=True allgather_bucket_size=500000000 overlap_comm=True load_from_fp32_weights=True elastic_checkpoint=False offload_param=DeepSpeedZeroOffloadParamConfig(device='none', nvme_path=None, buffer_count=5, buffer_size=100000000, max_in_cpu=1000000000, pin_memory=False) offload_optimizer=DeepSpeedZeroOffloadOptimizerConfig(device='none', nvme_path=None, buffer_count=4, pin_memory=False, pipeline_read=False, pipeline_write=False, fast_init=False, ratio=1.0) sub_group_size=1000000000 cpu_offload_param=None cpu_offload_use_pin_memory=None cpu_offload=None prefetch_bucket_size=30000000 param_persistence_threshold=10000 model_persistence_threshold=9223372036854775807 max_live_parameters=30000000 max_reuse_distance=1000000000 gather_16bit_weights_on_model_save=True module_granularity_threshold=0 use_all_reduce_for_fetch_params=False stage3_gather_fp16_weights_on_model_save=False ignore_unused_parameters=True legacy_stage1=False round_robin_gradients=False zero_hpz_partition_size=1 zero_quantized_weights=False zero_quantized_nontrainable_weights=False zero_quantized_gradients=False zeropp_loco_param=None mics_shard_size=-1 mics_hierarchical_params_gather=False memory_efficient_linear=False pipeline_loading_checkpoint=False override_module_apply=True log_trace_cache_warnings=False | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] zero_enabled ................. True | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] zero_force_ds_cpu_optimizer .. True | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:1007:print] zero_optimization_stage ...... 3 | |
| [2025-05-27 20:23:14,084] [INFO] [config.py:993:print_user_config] json = { | |
| "train_batch_size": 64, | |
| "train_micro_batch_size_per_gpu": 1, | |
| "gradient_accumulation_steps": 8, | |
| "steps_per_print": 10, | |
| "zero_optimization": { | |
| "stage": 3, | |
| "offload_param": { | |
| "device": "none" | |
| }, | |
| "offload_optimizer": { | |
| "device": "none" | |
| }, | |
| "param_persistence_threshold": 1.000000e+04, | |
| "max_live_parameters": 3.000000e+07, | |
| "prefetch_bucket_size": 3.000000e+07, | |
| "memory_efficient_linear": false, | |
| "gather_16bit_weights_on_model_save": true | |
| }, | |
| "gradient_clipping": 1.0, | |
| "prescale_gradients": false, | |
| "wall_clock_breakdown": false, | |
| "hybrid_engine": { | |
| "enabled": false, | |
| "max_out_tokens": 512, | |
| "inference_tp_size": 1, | |
| "release_inference_cache": false, | |
| "pin_parameters": true, | |
| "tp_gather_partition_size": 8 | |
| }, | |
| "bf16": { | |
| "enabled": true | |
| } | |
| } | |
| ***** Running training ***** | |
| [2025-05-27 20:25:19,827] [INFO] [logging.py:107:log_dist] [Rank 0] step=10, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:25:19,828] [INFO] [timer.py:264:stop] epoch=0/micro_step=80/global_step=10, RunningAvgSamplesPerSec=5.578530206713796, CurrSamplesPerSec=5.638705899378041, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:27:14,911] [INFO] [logging.py:107:log_dist] [Rank 0] step=20, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:27:14,911] [INFO] [timer.py:264:stop] epoch=0/micro_step=160/global_step=20, RunningAvgSamplesPerSec=5.658473565902261, CurrSamplesPerSec=5.14326774247884, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:29:10,620] [INFO] [logging.py:107:log_dist] [Rank 0] step=30, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:29:10,621] [INFO] [timer.py:264:stop] epoch=0/micro_step=240/global_step=30, RunningAvgSamplesPerSec=5.669602441915102, CurrSamplesPerSec=5.6512849891428765, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:31:09,620] [INFO] [logging.py:107:log_dist] [Rank 0] step=40, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:31:09,621] [INFO] [timer.py:264:stop] epoch=0/micro_step=320/global_step=40, RunningAvgSamplesPerSec=5.634702172676114, CurrSamplesPerSec=6.240526063295889, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:33:05,632] [INFO] [logging.py:107:log_dist] [Rank 0] step=50, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:33:05,632] [INFO] [timer.py:264:stop] epoch=0/micro_step=400/global_step=50, RunningAvgSamplesPerSec=5.644377398319037, CurrSamplesPerSec=5.9442386000955, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:35:02,777] [INFO] [logging.py:107:log_dist] [Rank 0] step=60, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:35:02,777] [INFO] [timer.py:264:stop] epoch=0/micro_step=480/global_step=60, RunningAvgSamplesPerSec=5.641251107041514, CurrSamplesPerSec=5.834201607514314, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:36:57,230] [INFO] [logging.py:107:log_dist] [Rank 0] step=70, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:36:57,230] [INFO] [timer.py:264:stop] epoch=0/micro_step=560/global_step=70, RunningAvgSamplesPerSec=5.659102360784789, CurrSamplesPerSec=5.482655791936472, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:38:54,726] [INFO] [logging.py:107:log_dist] [Rank 0] step=80, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:38:54,727] [INFO] [timer.py:264:stop] epoch=0/micro_step=640/global_step=80, RunningAvgSamplesPerSec=5.654024541881246, CurrSamplesPerSec=4.870766103763893, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:40:56,702] [INFO] [logging.py:107:log_dist] [Rank 0] step=90, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:40:56,702] [INFO] [timer.py:264:stop] epoch=0/micro_step=720/global_step=90, RunningAvgSamplesPerSec=5.622414233224792, CurrSamplesPerSec=5.1381117740003415, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:42:52,731] [INFO] [logging.py:107:log_dist] [Rank 0] step=100, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:42:52,731] [INFO] [timer.py:264:stop] epoch=0/micro_step=800/global_step=100, RunningAvgSamplesPerSec=5.629401901449947, CurrSamplesPerSec=6.127899602862844, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:44:43,158] [INFO] [logging.py:107:log_dist] [Rank 0] step=110, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:44:43,158] [INFO] [timer.py:264:stop] epoch=0/micro_step=880/global_step=110, RunningAvgSamplesPerSec=5.660330617557043, CurrSamplesPerSec=6.106843238506613, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:46:36,145] [INFO] [logging.py:107:log_dist] [Rank 0] step=120, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:46:36,145] [INFO] [timer.py:264:stop] epoch=0/micro_step=960/global_step=120, RunningAvgSamplesPerSec=5.675055646435481, CurrSamplesPerSec=6.670197993086072, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:48:34,960] [INFO] [logging.py:107:log_dist] [Rank 0] step=130, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:48:34,961] [INFO] [timer.py:264:stop] epoch=0/micro_step=1040/global_step=130, RunningAvgSamplesPerSec=5.664345840776106, CurrSamplesPerSec=4.8056893397972305, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:50:32,299] [INFO] [logging.py:107:log_dist] [Rank 0] step=140, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:50:32,300] [INFO] [timer.py:264:stop] epoch=0/micro_step=1120/global_step=140, RunningAvgSamplesPerSec=5.66102714725188, CurrSamplesPerSec=5.546249766962067, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| [2025-05-27 20:52:32,781] [INFO] [logging.py:107:log_dist] [Rank 0] step=150, skipped=0, lr=[1e-05, 1e-05], mom=[(0.9, 0.95), (0.9, 0.95)] | |
| [2025-05-27 20:52:32,781] [INFO] [timer.py:264:stop] epoch=0/micro_step=1200/global_step=150, RunningAvgSamplesPerSec=5.647692420557184, CurrSamplesPerSec=5.0752502154306365, MemAllocated=3.04GB, MaxMemAllocated=4.24GB | |
| Saving model to "/aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000" ... | |
| Saving 16-bit model... | |
| [2025-05-27 20:53:44,783] [INFO] [torch_checkpoint_engine.py:33:commit] [Torch] Checkpoint global_step156 is ready now! | |
| [2025-05-27 20:53:44,784] [INFO] [torch_checkpoint_engine.py:33:commit] [Torch] Checkpoint global_step156 is ready now! | |
| [2025-05-27 20:53:44,784] [INFO] [torch_checkpoint_engine.py:33:commit] [Torch] Checkpoint global_step156 is ready now! | |
| [2025-05-27 20:53:44,785] [INFO] [torch_checkpoint_engine.py:33:commit] [Torch] Checkpoint global_step156 is ready now! | |
| [2025-05-27 20:53:44,785] [INFO] [torch_checkpoint_engine.py:33:commit] [Torch] Checkpoint global_step156 is ready now! | |
| [2025-05-27 20:53:44,786] [INFO] [logging.py:107:log_dist] [Rank 0] [Torch] Checkpoint global_step156 is about to be saved! | |
| [2025-05-27 20:53:44,787] [INFO] [torch_checkpoint_engine.py:33:commit] [Torch] Checkpoint global_step156 is ready now! | |
| [2025-05-27 20:53:44,787] [INFO] [torch_checkpoint_engine.py:33:commit] [Torch] Checkpoint global_step156 is ready now! | |
| [2025-05-27 20:53:44,788] [INFO] [engine.py:3860:save_16bit_model] Saving model weights to /aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000/pytorch_model.bin, tag: global_step156 | |
| [2025-05-27 20:53:44,788] [INFO] [torch_checkpoint_engine.py:21:save] [Torch] Saving /aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000/pytorch_model.bin... | |
| [2025-05-27 20:53:46,536] [INFO] [launch.py:351:main] Process 2614659 exits successfully. | |
| [2025-05-27 20:53:47,009] [INFO] [torch_checkpoint_engine.py:23:save] [Torch] Saved /aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000/pytorch_model.bin. | |
| [2025-05-27 20:53:47,009] [INFO] [torch_checkpoint_engine.py:33:commit] [Torch] Checkpoint global_step156 is ready now! | |
| Model saved! | |
| [2025-05-27 20:53:47,537] [INFO] [launch.py:351:main] Process 2614656 exits successfully. | |
| [2025-05-27 20:53:47,537] [INFO] [launch.py:351:main] Process 2614653 exits successfully. | |
| [2025-05-27 20:53:47,537] [INFO] [launch.py:351:main] Process 2614655 exits successfully. | |
| [2025-05-27 20:53:47,537] [INFO] [launch.py:351:main] Process 2614654 exits successfully. | |
| [2025-05-27 20:53:48,538] [INFO] [launch.py:351:main] Process 2614658 exits successfully. | |
| [2025-05-27 20:53:48,538] [INFO] [launch.py:351:main] Process 2614657 exits successfully. | |
| [1;34mwandb[0m: [1;34mwandb[0m: 🚀 View run [33mimdb-tinyllama-3T-s3-Q1-10000[0m at: [34mhttps://wandb.ai/xtom/Inverse_Alignment_IMDb/runs/j30d8uyq[0m[1;34mwandb[0m: Find logs at: [1;35m../../../../../../aifs4su/hansirui_1st/jiayi/setting3-imdb/tinyllama-3T/tinyllama-3T-s3-Q1-10000/wandb/run-20250527_202314-j30d8uyq/logs[0m[2025-05-27 20:53:50,538] [INFO] [launch.py:351:main] Process 2614652 exits successfully. | |