| [2024-08-20 21:56:15,604] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| FLYWHEEL_USE_CUDA_DEVICES = 0,1,2,3,4,5,6,7 | |
| CUDA_VISIBLE_DEVICES = 0,1,2,3,4,5,6,7 | |
| FLYWHEEL_USE_CUDA_DEVICES = 0,1,2,3,4,5,6,7 | |
| CUDA_VISIBLE_DEVICES = 0,1,2,3,4,5,6,7 | |
| FLYWHEEL_USE_CUDA_DEVICES = 0,1,2,3,4,5,6,7 | |
| CUDA_VISIBLE_DEVICES = 0,1,2,3,4,5,6,7 | |
| FLYWHEEL_USE_CUDA_DEVICES = 0,1,2,3,4,5,6,7 | |
| CUDA_VISIBLE_DEVICES = 0,1,2,3,4,5,6,7 | |
| FLYWHEEL_USE_CUDA_DEVICES = 0,1,2,3,4,5,6,7 | |
| CUDA_VISIBLE_DEVICES = 0,1,2,3,4,5,6,7 | |
| FLYWHEEL_USE_CUDA_DEVICES = 0,1,2,3,4,5,6,7 | |
| CUDA_VISIBLE_DEVICES = 0,1,2,3,4,5,6,7 | |
| FLYWHEEL_USE_CUDA_DEVICES = 0,1,2,3,4,5,6,7 | |
| CUDA_VISIBLE_DEVICES = 0,1,2,3,4,5,6,7 | |
| FLYWHEEL_USE_CUDA_DEVICES = 0,1,2,3,4,5,6,7 | |
| CUDA_VISIBLE_DEVICES = 0,1,2,3,4,5,6,7 | |
| local_rank = 0, device_ranks = ['0', '1', '2', '3', '4', '5', '6', '7'], set device_rank = 0 | |
| local_rank = 6, device_ranks = ['0', '1', '2', '3', '4', '5', '6', '7'], set device_rank = 6 | |
| local_rank = 1, device_ranks = ['0', '1', '2', '3', '4', '5', '6', '7'], set device_rank = 1 | |
| local_rank = 7, device_ranks = ['0', '1', '2', '3', '4', '5', '6', '7'], set device_rank = 7 | |
| local_rank = 3, device_ranks = ['0', '1', '2', '3', '4', '5', '6', '7'], set device_rank = 3 | |
| local_rank = 5, device_ranks = ['0', '1', '2', '3', '4', '5', '6', '7'], set device_rank = 5 | |
| local_rank = 4, device_ranks = ['0', '1', '2', '3', '4', '5', '6', '7'], set device_rank = 4 | |
| local_rank = 2, device_ranks = ['0', '1', '2', '3', '4', '5', '6', '7'], set device_rank = 2 | |
| Rank = 0 | |
| Rank = 2 | |
| Rank = 1 | |
| Rank = 4 | |
| Rank = 3 | |
| Rank = 5 | |
| Rank = 6 | |
| Rank = 7 | |
| [2024-08-20 21:56:26,507] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2024-08-20 21:56:26,508] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2024-08-20 21:56:26,508] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2024-08-20 21:56:26,508] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2024-08-20 21:56:26,509] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2024-08-20 21:56:26,510] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2024-08-20 21:56:26,511] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [2024-08-20 21:56:26,511] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| [93m [WARNING] [0m async_io requires the dev libaio .so object and headers but these were not found. | |
| [93m [WARNING] [0m async_io requires the dev libaio .so object and headers but these were not found. | |
| [93m [WARNING] [0m async_io requires the dev libaio .so object and headers but these were not found. | |
| [93m [WARNING] [0m async_io requires the dev libaio .so object and headers but these were not found. | |
| [93m [WARNING] [0m async_io requires the dev libaio .so object and headers but these were not found. | |
| [93m [WARNING] [0m async_io requires the dev libaio .so object and headers but these were not found. | |
| [93m [WARNING] [0m async_io requires the dev libaio .so object and headers but these were not found. | |
| [93m [WARNING] [0m async_io requires the dev libaio .so object and headers but these were not found. | |
| [93m [WARNING] [0m async_io: please install the libaio-dev package with apt | |
| [93m [WARNING] [0m If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found. | |
| [93m [WARNING] [0m async_io: please install the libaio-dev package with apt | |
| [93m [WARNING] [0m If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found. | |
| [93m [WARNING] [0m async_io: please install the libaio-dev package with apt | |
| [93m [WARNING] [0m async_io: please install the libaio-dev package with apt[93m [WARNING] [0m If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found. | |
| [93m [WARNING] [0m If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found. | |
| [93m [WARNING] [0m async_io: please install the libaio-dev package with apt | |
| [93m [WARNING] [0m If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found. | |
| [93m [WARNING] [0m async_io: please install the libaio-dev package with apt | |
| [93m [WARNING] [0m If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found. | |
| [93m [WARNING] [0m async_io: please install the libaio-dev package with apt[93m [WARNING] [0m async_io: please install the libaio-dev package with apt | |
| [93m [WARNING] [0m If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found. | |
| [93m [WARNING] [0m If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found. | |
| Installed CUDA version 12.5 does not match the version torch was compiled with 12.4 but since the APIs are compatible, accepting this combination | |
| Installed CUDA version 12.5 does not match the version torch was compiled with 12.4 but since the APIs are compatible, accepting this combination | |
| Installed CUDA version 12.5 does not match the version torch was compiled with 12.4 but since the APIs are compatible, accepting this combination | |
| Installed CUDA version 12.5 does not match the version torch was compiled with 12.4 but since the APIs are compatible, accepting this combination | |
| Installed CUDA version 12.5 does not match the version torch was compiled with 12.4 but since the APIs are compatible, accepting this combination | |
| Installed CUDA version 12.5 does not match the version torch was compiled with 12.4 but since the APIs are compatible, accepting this combination | |
| Installed CUDA version 12.5 does not match the version torch was compiled with 12.4 but since the APIs are compatible, accepting this combination | |
| Installed CUDA version 12.5 does not match the version torch was compiled with 12.4 but since the APIs are compatible, accepting this combination | |
| ninja: no work to do. | |
| Time to load cpu_adam op: 0.3381340503692627 seconds | |
| Time to load cpu_adam op: 0.3769395351409912 seconds | |
| Time to load cpu_adam op: 0.3873555660247803 seconds | |
| Time to load cpu_adam op: 0.395129919052124 seconds | |
| Time to load cpu_adam op: 0.39567041397094727 seconds | |
| Time to load cpu_adam op: 0.3963282108306885 seconds | |
| Time to load cpu_adam op: 0.39649391174316406 seconds | |
| Time to load cpu_adam op: 0.39492321014404297 seconds | |
| [2024-08-20 21:56:28,273] [INFO] [comm.py:637:init_distributed] cdb=None | |
| [2024-08-20 21:56:28,302] [INFO] [comm.py:637:init_distributed] cdb=None | |
| [2024-08-20 21:56:28,312] [INFO] [comm.py:637:init_distributed] cdb=None | |
| [2024-08-20 21:56:28,313] [INFO] [comm.py:637:init_distributed] cdb=None | |
| [2024-08-20 21:56:28,313] [INFO] [comm.py:637:init_distributed] cdb=None | |
| [2024-08-20 21:56:28,314] [INFO] [comm.py:637:init_distributed] cdb=None | |
| [2024-08-20 21:56:28,314] [INFO] [comm.py:637:init_distributed] cdb=None | |
| [2024-08-20 21:56:28,315] [INFO] [comm.py:637:init_distributed] cdb=None | |
| [2024-08-20 21:56:42,890] [INFO] [partition_parameters.py:345:__exit__] finished initializing model - num_params = 291, num_elems = 6.74B | |
| Is prompt masked: True | |
| Is prompt masked: True | |
| Is prompt masked: True | |
| Is prompt masked: True | |
| Is prompt masked: True | |
| Is prompt masked: True | |
| Is prompt masked: True | |
| Is prompt masked: True | |
| Adam Optimizer #0 is created with AVX512 arithmetic capability. | |
| Config: alpha=0.000020, betas=(0.900000, 0.999000), weight_decay=0.010000, adam_w=1 | |
| [2024-08-20 21:56:49,474] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed info: version=0.14.5, git-hash=unknown, git-branch=unknown | |
| Adam Optimizer #0 is created with AVX512 arithmetic capability. | |
| Config: alpha=0.000020, betas=(0.900000, 0.999000), weight_decay=0.010000, adam_w=1 | |
| Adam Optimizer #0 is created with AVX512 arithmetic capability. | |
| Config: alpha=0.000020, betas=(0.900000, 0.999000), weight_decay=0.010000, adam_w=1 | |
| Adam Optimizer #0 is created with AVX512 arithmetic capability. | |
| Config: alpha=0.000020, betas=(0.900000, 0.999000), weight_decay=0.010000, adam_w=1 | |
| Adam Optimizer #0 is created with AVX512 arithmetic capability. | |
| Config: alpha=0.000020, betas=(0.900000, 0.999000), weight_decay=0.010000, adam_w=1 | |
| Adam Optimizer #0 is created with AVX512 arithmetic capability. | |
| Config: alpha=0.000020, betas=(0.900000, 0.999000), weight_decay=0.010000, adam_w=1 | |
| Adam Optimizer #0 is created with AVX512 arithmetic capability. | |
| Config: alpha=0.000020, betas=(0.900000, 0.999000), weight_decay=0.010000, adam_w=1 | |
| Adam Optimizer #0 is created with AVX512 arithmetic capability. | |
| Config: alpha=0.000020, betas=(0.900000, 0.999000), weight_decay=0.010000, adam_w=1 | |
| [2024-08-20 21:56:49,547] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed Flops Profiler Enabled: False | |
| [2024-08-20 21:56:49,548] [INFO] [logging.py:96:log_dist] [Rank 0] Using client Optimizer as basic optimizer | |
| [2024-08-20 21:56:49,548] [INFO] [logging.py:96:log_dist] [Rank 0] Removing param_group that has no 'params' in the basic Optimizer | |
| [2024-08-20 21:56:49,556] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed Basic Optimizer = DeepSpeedCPUAdam | |
| [2024-08-20 21:56:49,556] [INFO] [utils.py:59:is_zero_supported_optimizer] Checking ZeRO support for optimizer=DeepSpeedCPUAdam type=<class 'deepspeed.ops.adam.cpu_adam.DeepSpeedCPUAdam'> | |
| [2024-08-20 21:56:49,556] [INFO] [logging.py:96:log_dist] [Rank 0] Creating fp16 ZeRO stage 3 optimizer, MiCS is enabled False, Hierarchical params gather False | |
| [2024-08-20 21:56:49,556] [INFO] [logging.py:96:log_dist] [Rank 0] Creating torch.bfloat16 ZeRO stage 3 optimizer | |
| [2024-08-20 21:56:49,742] [INFO] [utils.py:781:see_memory_usage] Stage 3 initialize beginning | |
| [2024-08-20 21:56:49,742] [INFO] [utils.py:782:see_memory_usage] MA 0.0 GB Max_MA 0.49 GB CA 0.49 GB Max_CA 0 GB | |
| [2024-08-20 21:56:49,742] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 44.84 GB, percent = 2.4% | |
| [2024-08-20 21:56:49,744] [INFO] [stage3.py:130:__init__] Reduce bucket size 16777216 | |
| [2024-08-20 21:56:49,744] [INFO] [stage3.py:131:__init__] Prefetch bucket size 15099494 | |
| [2024-08-20 21:56:49,879] [INFO] [utils.py:781:see_memory_usage] DeepSpeedZeRoOffload initialize [begin] | |
| [2024-08-20 21:56:49,879] [INFO] [utils.py:782:see_memory_usage] MA 0.0 GB Max_MA 0.0 GB CA 0.49 GB Max_CA 0 GB | |
| [2024-08-20 21:56:49,879] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 44.84 GB, percent = 2.4% | |
| Parameter Offload: Total persistent parameters: 266240 in 65 params | |
| [2024-08-20 21:56:50,008] [INFO] [utils.py:781:see_memory_usage] DeepSpeedZeRoOffload initialize [end] | |
| [2024-08-20 21:56:50,008] [INFO] [utils.py:782:see_memory_usage] MA 0.0 GB Max_MA 0.0 GB CA 0.49 GB Max_CA 0 GB | |
| [2024-08-20 21:56:50,008] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 44.84 GB, percent = 2.4% | |
| [2024-08-20 21:56:50,115] [INFO] [utils.py:781:see_memory_usage] Before creating fp16 partitions | |
| [2024-08-20 21:56:50,116] [INFO] [utils.py:782:see_memory_usage] MA 0.0 GB Max_MA 0.0 GB CA 0.49 GB Max_CA 0 GB | |
| [2024-08-20 21:56:50,116] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 44.83 GB, percent = 2.4% | |
| [2024-08-20 21:56:51,868] [INFO] [utils.py:781:see_memory_usage] After creating fp16 partitions: 1 | |
| [2024-08-20 21:56:51,868] [INFO] [utils.py:782:see_memory_usage] MA 0.0 GB Max_MA 0.0 GB CA 0.49 GB Max_CA 0 GB | |
| [2024-08-20 21:56:51,869] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 66.25 GB, percent = 3.5% | |
| [2024-08-20 21:56:52,107] [INFO] [utils.py:781:see_memory_usage] Before creating fp32 partitions | |
| [2024-08-20 21:56:52,108] [INFO] [utils.py:782:see_memory_usage] MA 0.0 GB Max_MA 0.0 GB CA 0.49 GB Max_CA 0 GB | |
| [2024-08-20 21:56:52,108] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 71.42 GB, percent = 3.8% | |
| [2024-08-20 21:56:53,699] [INFO] [utils.py:781:see_memory_usage] After creating fp32 partitions | |
| [2024-08-20 21:56:53,699] [INFO] [utils.py:782:see_memory_usage] MA 0.0 GB Max_MA 0.0 GB CA 0.49 GB Max_CA 0 GB | |
| [2024-08-20 21:56:53,700] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 86.24 GB, percent = 4.6% | |
| [2024-08-20 21:56:53,896] [INFO] [utils.py:781:see_memory_usage] Before initializing optimizer states | |
| [2024-08-20 21:56:53,897] [INFO] [utils.py:782:see_memory_usage] MA 0.0 GB Max_MA 0.0 GB CA 0.49 GB Max_CA 0 GB | |
| [2024-08-20 21:56:53,897] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 92.43 GB, percent = 4.9% | |
| [2024-08-20 21:56:57,740] [INFO] [utils.py:781:see_memory_usage] After initializing optimizer states | |
| [2024-08-20 21:56:57,740] [INFO] [utils.py:782:see_memory_usage] MA 0.0 GB Max_MA 0.0 GB CA 0.49 GB Max_CA 0 GB | |
| [2024-08-20 21:56:57,741] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 118.86 GB, percent = 6.4% | |
| [2024-08-20 21:56:57,746] [INFO] [stage3.py:485:_setup_for_real_optimizer] optimizer state initialized | |
| [2024-08-20 21:57:00,138] [INFO] [utils.py:781:see_memory_usage] After initializing ZeRO optimizer | |
| [2024-08-20 21:57:00,139] [INFO] [utils.py:782:see_memory_usage] MA 0.03 GB Max_MA 0.52 GB CA 0.74 GB Max_CA 1 GB | |
| [2024-08-20 21:57:00,139] [INFO] [utils.py:789:see_memory_usage] CPU Virtual Memory: used = 135.23 GB, percent = 7.2% | |
| [2024-08-20 21:57:00,139] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed Final Optimizer = DeepSpeedZeroOptimizer_Stage3 | |
| [2024-08-20 21:57:00,139] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed using configured LR scheduler = None | |
| [2024-08-20 21:57:00,139] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed LR Scheduler = None | |
| [2024-08-20 21:57:00,139] [INFO] [logging.py:96:log_dist] [Rank 0] step=0, skipped=0, lr=[0.0], mom=[(0.9, 0.999)] | |
| [2024-08-20 21:57:00,140] [INFO] [config.py:997:print] DeepSpeedEngine configuration: | |
| [2024-08-20 21:57:00,140] [INFO] [config.py:1001:print] activation_checkpointing_config { | |
| "partition_activations": false, | |
| "contiguous_memory_optimization": false, | |
| "cpu_checkpointing": false, | |
| "number_checkpoints": null, | |
| "synchronize_checkpoint_boundary": false, | |
| "profile": false | |
| } | |
| [2024-08-20 21:57:00,140] [INFO] [config.py:1001:print] aio_config ................... {'block_size': 1048576, 'queue_depth': 8, 'thread_count': 1, 'single_submit': False, 'overlap_events': True} | |
| [2024-08-20 21:57:00,140] [INFO] [config.py:1001:print] amp_enabled .................. False | |
| [2024-08-20 21:57:00,140] [INFO] [config.py:1001:print] amp_params ................... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001: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 | |
| } | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] bfloat16_enabled ............. True | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] bfloat16_immediate_grad_update False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] checkpoint_parallel_write_pipeline False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] checkpoint_tag_validation_enabled True | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] checkpoint_tag_validation_fail False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] comms_config ................. <deepspeed.comm.config.DeepSpeedCommsConfig object at 0x7fee949eee70> | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] communication_data_type ...... None | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001: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}} | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] curriculum_enabled_legacy .... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] curriculum_params_legacy ..... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] data_efficiency_config ....... {'enabled': False, 'seed': 1234, 'data_sampling': {'enabled': False, 'num_epochs': 1000, 'num_workers': 0, 'curriculum_learning': {'enabled': False}}, 'data_routing': {'enabled': False, 'random_ltd': {'enabled': False, 'layer_token_lr_schedule': {'enabled': False}}}} | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] data_efficiency_enabled ...... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] dataloader_drop_last ......... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] disable_allgather ............ False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] dump_state ................... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] dynamic_loss_scale_args ...... None | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] eigenvalue_enabled ........... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] eigenvalue_gas_boundary_resolution 1 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] eigenvalue_layer_name ........ bert.encoder.layer | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] eigenvalue_layer_num ......... 0 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] eigenvalue_max_iter .......... 100 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] eigenvalue_stability ......... 1e-06 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] eigenvalue_tol ............... 0.01 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] eigenvalue_verbose ........... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] elasticity_enabled ........... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001: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 | |
| } | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] fp16_auto_cast ............... None | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] fp16_enabled ................. False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] fp16_master_weights_and_gradients False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] global_rank .................. 0 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] grad_accum_dtype ............. None | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] gradient_accumulation_steps .. 1 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] gradient_clipping ............ 1.0 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] gradient_predivide_factor .... 1.0 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] graph_harvesting ............. False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001: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 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] initial_dynamic_scale ........ 1 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] load_universal_checkpoint .... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] loss_scale ................... 1.0 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] memory_breakdown ............. False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] mics_hierarchial_params_gather False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] mics_shard_size .............. -1 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001: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') enabled=False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001: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 | |
| } | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] optimizer_legacy_fusion ...... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] optimizer_name ............... None | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] optimizer_params ............. None | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] pipeline ..................... {'stages': 'auto', 'partition': 'best', 'seed_layers': False, 'activation_checkpoint_interval': 0, 'pipe_partitioned': True, 'grad_partitioned': True} | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] pld_enabled .................. False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] pld_params ................... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] prescale_gradients ........... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] scheduler_name ............... None | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] scheduler_params ............. None | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] seq_parallel_communication_data_type torch.float32 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] sparse_attention ............. None | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] sparse_gradients_enabled ..... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] steps_per_print .............. inf | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] timers_config ................ enabled=True synchronized=True | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] train_batch_size ............. 128 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] train_micro_batch_size_per_gpu 16 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] use_data_before_expert_parallel_ False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] use_node_local_storage ....... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] wall_clock_breakdown ......... False | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] weight_quantization_config ... None | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] world_size ................... 8 | |
| [2024-08-20 21:57:00,141] [INFO] [config.py:1001:print] zero_allow_untested_optimizer True | |
| [2024-08-20 21:57:00,142] [INFO] [config.py:1001:print] zero_config .................. stage=3 contiguous_gradients=True reduce_scatter=True reduce_bucket_size=16777216 use_multi_rank_bucket_allreduce=True allgather_partitions=True allgather_bucket_size=500,000,000 overlap_comm=True load_from_fp32_weights=True elastic_checkpoint=False offload_param=DeepSpeedZeroOffloadParamConfig(device='cpu', nvme_path=None, buffer_count=5, buffer_size=100,000,000, max_in_cpu=1,000,000,000, pin_memory=True) offload_optimizer=DeepSpeedZeroOffloadOptimizerConfig(device='cpu', nvme_path=None, buffer_count=4, pin_memory=True, pipeline=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=15099494 param_persistence_threshold=40960 model_persistence_threshold=sys.maxsize max_live_parameters=4000000000 max_reuse_distance=4000000000 gather_16bit_weights_on_model_save=True 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 mics_shard_size=-1 mics_hierarchical_params_gather=False memory_efficient_linear=True pipeline_loading_checkpoint=False override_module_apply=True | |
| [2024-08-20 21:57:00,142] [INFO] [config.py:1001:print] zero_enabled ................. True | |
| [2024-08-20 21:57:00,142] [INFO] [config.py:1001:print] zero_force_ds_cpu_optimizer .. True | |
| [2024-08-20 21:57:00,142] [INFO] [config.py:1001:print] zero_optimization_stage ...... 3 | |
| [2024-08-20 21:57:00,142] [INFO] [config.py:987:print_user_config] json = { | |
| "bf16": { | |
| "enabled": true | |
| }, | |
| "zero_optimization": { | |
| "stage": 3, | |
| "offload_optimizer": { | |
| "device": "cpu", | |
| "pin_memory": true | |
| }, | |
| "offload_param": { | |
| "device": "cpu", | |
| "pin_memory": true | |
| }, | |
| "overlap_comm": true, | |
| "contiguous_gradients": true, | |
| "sub_group_size": 1.000000e+09, | |
| "reduce_bucket_size": 1.677722e+07, | |
| "stage3_prefetch_bucket_size": 1.509949e+07, | |
| "stage3_param_persistence_threshold": 4.096000e+04, | |
| "stage3_max_live_parameters": 4.000000e+09, | |
| "stage3_max_reuse_distance": 4.000000e+09, | |
| "stage3_gather_16bit_weights_on_model_save": true | |
| }, | |
| "gradient_accumulation_steps": 1, | |
| "gradient_clipping": 1.0, | |
| "steps_per_print": inf, | |
| "train_batch_size": 128, | |
| "train_micro_batch_size_per_gpu": 16, | |
| "wall_clock_breakdown": false, | |
| "fp16": { | |
| "enabled": false | |
| }, | |
| "zero_allow_untested_optimizer": true | |
| } | |
| [93m [WARNING] [0m async_io requires the dev libaio .so object and headers but these were not found. | |
| [93m [WARNING] [0m async_io: please install the libaio-dev package with apt | |
| [93m [WARNING] [0m If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found. | |