run.log

[2022-12-18 08:40:52,091] [WARNING] [runner.py:179:fetch_hostfile] Unable to find hostfile, will proceed with training with local resources only.
[2022-12-18 08:40:52,100] [INFO] [runner.py:508:main] cmd = /usr/bin/python -u -m deepspeed.launcher.launch --world_info=eyJsb2NhbGhvc3QiOiBbMF19 --master_addr=127.0.0.1 --master_port=29500 run_speech_recognition_seq2seq_streaming.py --deepspeed=ds_config.json --model_name_or_path=openai/whisper-small --dataset_name=mozilla-foundation/common_voice_11_0 --dataset_config_name=ro --language=romanian --train_split_name=train+validation --eval_split_name=test --model_index_name=Whisper Small Romanian CV11 --max_steps=5000 --output_dir=./ --per_device_train_batch_size=64 --per_device_eval_batch_size=32 --logging_steps=25 --learning_rate=1e-5 --warmup_steps=500 --evaluation_strategy=steps --eval_steps=1000 --save_strategy=steps --save_steps=1000 --generation_max_length=225 --length_column_name=input_length --max_duration_in_seconds=30 --text_column_name=sentence --freeze_feature_encoder=False --report_to=tensorboard --metric_for_best_model=wer --greater_is_better=False --load_best_model_at_end --gradient_checkpointing --fp16 --overwrite_output_dir --do_train --do_eval --predict_with_generate --do_normalize_eval --streaming --use_auth_token --push_to_hub
[2022-12-18 08:40:55,346] [INFO] [launch.py:135:main] 0 NV_LIBNCCL_DEV_PACKAGE=libnccl-dev=2.13.4-1+cuda11.7
[2022-12-18 08:40:55,346] [INFO] [launch.py:135:main] 0 NV_LIBNCCL_DEV_PACKAGE_VERSION=2.13.4-1
[2022-12-18 08:40:55,346] [INFO] [launch.py:135:main] 0 NCCL_VERSION=2.13.4-1
[2022-12-18 08:40:55,347] [INFO] [launch.py:135:main] 0 NV_LIBNCCL_DEV_PACKAGE_NAME=libnccl-dev
[2022-12-18 08:40:55,347] [INFO] [launch.py:135:main] 0 NV_LIBNCCL_PACKAGE=libnccl2=2.13.4-1+cuda11.7
[2022-12-18 08:40:55,347] [INFO] [launch.py:135:main] 0 NV_LIBNCCL_PACKAGE_NAME=libnccl2
[2022-12-18 08:40:55,347] [INFO] [launch.py:135:main] 0 NV_LIBNCCL_PACKAGE_VERSION=2.13.4-1
[2022-12-18 08:40:55,347] [INFO] [launch.py:142:main] WORLD INFO DICT: {'localhost': [0]}
[2022-12-18 08:40:55,347] [INFO] [launch.py:148:main] nnodes=1, num_local_procs=1, node_rank=0
[2022-12-18 08:40:55,347] [INFO] [launch.py:161:main] global_rank_mapping=defaultdict(<class 'list'>, {'localhost': [0]})
[2022-12-18 08:40:55,347] [INFO] [launch.py:162:main] dist_world_size=1
[2022-12-18 08:40:55,347] [INFO] [launch.py:164:main] Setting CUDA_VISIBLE_DEVICES=0
[2022-12-18 08:41:04,141] [INFO] [comm.py:654:init_distributed] Initializing TorchBackend in DeepSpeed with backend nccl
12/18/2022 08:41:04 - WARNING - __main__ - Process rank: 0, device: cuda:0, n_gpu: 1distributed training: True, 16-bits training: True
12/18/2022 08:41:04 - INFO - __main__ - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
bf16_full_eval=False,
data_seed=None,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=None,
ddp_timeout=1800,
debug=[],
deepspeed=ds_config.json,
disable_tqdm=False,
do_eval=True,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_delay=0,
eval_steps=1000,
evaluation_strategy=steps,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
fsdp=[],
fsdp_min_num_params=0,
fsdp_transformer_layer_cls_to_wrap=None,
full_determinism=False,
generation_max_length=225,
generation_num_beams=None,
gradient_accumulation_steps=1,
gradient_checkpointing=True,
greater_is_better=False,
group_by_length=False,
half_precision_backend=auto,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
include_inputs_for_metrics=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=1e-05,
length_column_name=input_length,
load_best_model_at_end=True,
local_rank=0,
log_level=passive,
log_level_replica=passive,
log_on_each_node=True,
logging_dir=./runs/Dec18_08-41-04_fe2747a042f0,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=25,
logging_strategy=steps,
lr_scheduler_type=linear,
max_grad_norm=1.0,
max_steps=5000,
metric_for_best_model=wer,
mp_parameters=,
no_cuda=False,
num_train_epochs=3.0,
optim=adamw_hf,
optim_args=None,
output_dir=./,
overwrite_output_dir=True,
past_index=-1,
per_device_eval_batch_size=32,
per_device_train_batch_size=64,
predict_with_generate=True,
prediction_loss_only=False,
push_to_hub=True,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
ray_scope=last,
remove_unused_columns=True,
report_to=['tensorboard'],
resume_from_checkpoint=None,
run_name=./,
save_on_each_node=False,
save_steps=1000,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=False,
tf32=None,
torch_compile=False,
torch_compile_backend=None,
torch_compile_mode=None,
torchdynamo=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=500,
weight_decay=0.0,
xpu_backend=None,
)
12/18/2022 08:41:04 - INFO - __main__ - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
bf16_full_eval=False,
data_seed=None,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=None,
ddp_timeout=1800,
debug=[],
deepspeed=ds_config.json,
disable_tqdm=False,
do_eval=True,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_delay=0,
eval_steps=1000,
evaluation_strategy=steps,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
fsdp=[],
fsdp_min_num_params=0,
fsdp_transformer_layer_cls_to_wrap=None,
full_determinism=False,
generation_max_length=225,
generation_num_beams=None,
gradient_accumulation_steps=1,
gradient_checkpointing=True,
greater_is_better=False,
group_by_length=False,
half_precision_backend=auto,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
include_inputs_for_metrics=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=1e-05,
length_column_name=input_length,
load_best_model_at_end=True,
local_rank=0,
log_level=passive,
log_level_replica=passive,
log_on_each_node=True,
logging_dir=./runs/Dec18_08-41-04_fe2747a042f0,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=25,
logging_strategy=steps,
lr_scheduler_type=linear,
max_grad_norm=1.0,
max_steps=5000,
metric_for_best_model=wer,
mp_parameters=,
no_cuda=False,
num_train_epochs=3.0,
optim=adamw_hf,
optim_args=None,
output_dir=./,
overwrite_output_dir=True,
past_index=-1,
per_device_eval_batch_size=32,
per_device_train_batch_size=64,
predict_with_generate=True,
prediction_loss_only=False,
push_to_hub=True,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
ray_scope=last,
remove_unused_columns=True,
report_to=['tensorboard'],
resume_from_checkpoint=None,
run_name=./,
save_on_each_node=False,
save_steps=1000,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=False,
tf32=None,
torch_compile=False,
torch_compile_backend=None,
torch_compile_mode=None,
torchdynamo=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=500,
weight_decay=0.0,
xpu_backend=None,
)
12/18/2022 08:41:07 - INFO - datasets.info - Loading Dataset Infos from /root/.cache/huggingface/modules/datasets_modules/datasets/mozilla-foundation--common_voice_11_0/f8e47235d9b4e68fa24ed71d63266a02018ccf7194b2a8c9c598a5f3ab304d9f
12/18/2022 08:41:11 - INFO - datasets.info - Loading Dataset Infos from /root/.cache/huggingface/modules/datasets_modules/datasets/mozilla-foundation--common_voice_11_0/f8e47235d9b4e68fa24ed71d63266a02018ccf7194b2a8c9c598a5f3ab304d9f
12/18/2022 08:41:14 - INFO - datasets.info - Loading Dataset Infos from /root/.cache/huggingface/modules/datasets_modules/datasets/mozilla-foundation--common_voice_11_0/f8e47235d9b4e68fa24ed71d63266a02018ccf7194b2a8c9c598a5f3ab304d9f
12/18/2022 08:41:59 - WARNING - huggingface_hub.repository - /usr/src/app/models/whisper-small-ro-cv11/./ is already a clone of https://huggingface.co/mikr/whisper-small-ro-cv11. Make sure you pull the latest changes with `repo.git_pull()`.
[2022-12-18 08:42:04,348] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed info: version=0.8.0+a25c31b6, git-hash=a25c31b6, git-branch=master
[2022-12-18 08:42:04,669] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed Flops Profiler Enabled: False
Adam Optimizer #0 is created with AVX2 arithmetic capability.
Config: alpha=0.000010, betas=(0.900000, 0.999000), weight_decay=0.000000, adam_w=1
[2022-12-18 08:42:07,543] [INFO] [logging.py:68:log_dist] [Rank 0] Using DeepSpeed Optimizer param name adamw as basic optimizer
[2022-12-18 08:42:07,597] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed Basic Optimizer = DeepSpeedCPUAdam
[2022-12-18 08:42:07,597] [INFO] [utils.py:52:is_zero_supported_optimizer] Checking ZeRO support for optimizer=DeepSpeedCPUAdam type=<class 'deepspeed.ops.adam.cpu_adam.DeepSpeedCPUAdam'>
[2022-12-18 08:42:07,598] [INFO] [logging.py:68:log_dist] [Rank 0] Creating fp16 ZeRO stage 2 optimizer
[2022-12-18 08:42:07,598] [INFO] [stage_1_and_2.py:141:__init__] Reduce bucket size 200000000
[2022-12-18 08:42:07,598] [INFO] [stage_1_and_2.py:142:__init__] Allgather bucket size 200000000
[2022-12-18 08:42:07,598] [INFO] [stage_1_and_2.py:143:__init__] CPU Offload: True
[2022-12-18 08:42:07,598] [INFO] [stage_1_and_2.py:144:__init__] Round robin gradient partitioning: False
Rank: 0 partition count [1] and sizes[(241734912, False)] 
[2022-12-18 08:42:08,957] [INFO] [utils.py:831:see_memory_usage] Before initializing optimizer states
[2022-12-18 08:42:08,958] [INFO] [utils.py:832:see_memory_usage] MA 0.53 GB         Max_MA 0.53 GB         CA 0.53 GB         Max_CA 1 GB 
[2022-12-18 08:42:08,958] [INFO] [utils.py:840:see_memory_usage] CPU Virtual Memory:  used = 379.95 GB, percent = 75.4%
[2022-12-18 08:42:10,038] [INFO] [utils.py:831:see_memory_usage] After initializing optimizer states
[2022-12-18 08:42:10,039] [INFO] [utils.py:832:see_memory_usage] MA 0.53 GB         Max_MA 0.53 GB         CA 0.53 GB         Max_CA 1 GB 
[2022-12-18 08:42:10,039] [INFO] [utils.py:840:see_memory_usage] CPU Virtual Memory:  used = 382.79 GB, percent = 76.0%
[2022-12-18 08:42:10,039] [INFO] [stage_1_and_2.py:527:__init__] optimizer state initialized
[2022-12-18 08:42:10,147] [INFO] [utils.py:831:see_memory_usage] After initializing ZeRO optimizer
[2022-12-18 08:42:10,148] [INFO] [utils.py:832:see_memory_usage] MA 0.53 GB         Max_MA 0.53 GB         CA 0.53 GB         Max_CA 1 GB 
[2022-12-18 08:42:10,148] [INFO] [utils.py:840:see_memory_usage] CPU Virtual Memory:  used = 382.83 GB, percent = 76.0%
[2022-12-18 08:42:10,170] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed Final Optimizer = adamw
[2022-12-18 08:42:10,170] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed using configured LR scheduler = WarmupDecayLR
[2022-12-18 08:42:10,170] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed LR Scheduler = <deepspeed.runtime.lr_schedules.WarmupDecayLR object at 0x7ff2cab02f10>
[2022-12-18 08:42:10,170] [INFO] [logging.py:68:log_dist] [Rank 0] step=0, skipped=0, lr=[1e-05], mom=[[0.9, 0.999]]
[2022-12-18 08:42:10,172] [INFO] [config.py:1008:print] DeepSpeedEngine configuration:
[2022-12-18 08:42:10,172] [INFO] [config.py:1012:print]   activation_checkpointing_config  {
    "partition_activations": false, 
    "contiguous_memory_optimization": false, 
    "cpu_checkpointing": false, 
    "number_checkpoints": null, 
    "synchronize_checkpoint_boundary": false, 
    "profile": false
}
[2022-12-18 08:42:10,172] [INFO] [config.py:1012:print]   aio_config ................... {'block_size': 1048576, 'queue_depth': 8, 'thread_count': 1, 'single_submit': False, 'overlap_events': True}
[2022-12-18 08:42:10,173] [INFO] [config.py:1012:print]   amp_enabled .................. False
[2022-12-18 08:42:10,173] [INFO] [config.py:1012:print]   amp_params ................... False
[2022-12-18 08:42:10,173] [INFO] [config.py:1012: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
}
[2022-12-18 08:42:10,173] [INFO] [config.py:1012:print]   bfloat16_enabled ............. False
[2022-12-18 08:42:10,173] [INFO] [config.py:1012:print]   checkpoint_parallel_write_pipeline  False
[2022-12-18 08:42:10,173] [INFO] [config.py:1012:print]   checkpoint_tag_validation_enabled  True
[2022-12-18 08:42:10,173] [INFO] [config.py:1012:print]   checkpoint_tag_validation_fail  False
[2022-12-18 08:42:10,173] [INFO] [config.py:1012:print]   comms_config ................. <deepspeed.comm.config.DeepSpeedCommsConfig object at 0x7ff2cde355b0>
[2022-12-18 08:42:10,173] [INFO] [config.py:1012:print]   communication_data_type ...... None
[2022-12-18 08:42:10,173] [INFO] [config.py:1012: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}}
[2022-12-18 08:42:10,173] [INFO] [config.py:1012:print]   curriculum_enabled_legacy .... False
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   curriculum_params_legacy ..... False
[2022-12-18 08:42:10,174] [INFO] [config.py:1012: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}}}}
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   data_efficiency_enabled ...... False
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   dataloader_drop_last ......... False
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   disable_allgather ............ False
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   dump_state ................... False
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   dynamic_loss_scale_args ...... {'init_scale': 65536, 'scale_window': 1000, 'delayed_shift': 2, 'min_scale': 1}
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   eigenvalue_enabled ........... False
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   eigenvalue_gas_boundary_resolution  1
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   eigenvalue_layer_name ........ bert.encoder.layer
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   eigenvalue_layer_num ......... 0
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   eigenvalue_max_iter .......... 100
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   eigenvalue_stability ......... 1e-06
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   eigenvalue_tol ............... 0.01
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   eigenvalue_verbose ........... False
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   elasticity_enabled ........... False
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   flops_profiler_config ........ {
    "enabled": false, 
    "profile_step": 1, 
    "module_depth": -1, 
    "top_modules": 1, 
    "detailed": true, 
    "output_file": null
}
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   fp16_auto_cast ............... False
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   fp16_enabled ................. True
[2022-12-18 08:42:10,174] [INFO] [config.py:1012:print]   fp16_master_weights_and_gradients  False
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   global_rank .................. 0
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   grad_accum_dtype ............. None
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   gradient_accumulation_steps .. 1
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   gradient_clipping ............ 1.0
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   gradient_predivide_factor .... 1.0
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   initial_dynamic_scale ........ 65536
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   load_universal_checkpoint .... False
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   loss_scale ................... 0
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   memory_breakdown ............. False
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   monitor_config ............... <deepspeed.monitor.config.DeepSpeedMonitorConfig object at 0x7ff2cde359d0>
[2022-12-18 08:42:10,175] [INFO] [config.py:1012: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
}
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   optimizer_legacy_fusion ...... False
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   optimizer_name ............... adamw
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   optimizer_params ............. {'lr': 1e-05, 'betas': [0.9, 0.999], 'eps': 1e-08, 'weight_decay': 0.0}
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   pipeline ..................... {'stages': 'auto', 'partition': 'best', 'seed_layers': False, 'activation_checkpoint_interval': 0}
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   pld_enabled .................. False
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   pld_params ................... False
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   prescale_gradients ........... False
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   scheduler_name ............... WarmupDecayLR
[2022-12-18 08:42:10,175] [INFO] [config.py:1012:print]   scheduler_params ............. {'last_batch_iteration': -1, 'total_num_steps': 5000, 'warmup_min_lr': 0, 'warmup_max_lr': 1e-05, 'warmup_num_steps': 500}
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   sparse_attention ............. None
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   sparse_gradients_enabled ..... False
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   steps_per_print .............. 10
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   train_batch_size ............. 64
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   train_micro_batch_size_per_gpu  64
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   use_node_local_storage ....... False
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   wall_clock_breakdown ......... False
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   world_size ................... 1
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   zero_allow_untested_optimizer  False
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   zero_config .................. stage=2 contiguous_gradients=True reduce_scatter=True reduce_bucket_size=200000000 allgather_partitions=True allgather_bucket_size=200000000 overlap_comm=True load_from_fp32_weights=True elastic_checkpoint=False offload_param=None 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) sub_group_size=1,000,000,000 cpu_offload_param=None cpu_offload_use_pin_memory=None cpu_offload=None prefetch_bucket_size=50,000,000 param_persistence_threshold=100,000 model_persistence_threshold=sys.maxsize max_live_parameters=1,000,000,000 max_reuse_distance=1,000,000,000 gather_16bit_weights_on_model_save=False stage3_gather_fp16_weights_on_model_save=False ignore_unused_parameters=True legacy_stage1=False round_robin_gradients=False
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   zero_enabled ................. True
[2022-12-18 08:42:10,176] [INFO] [config.py:1012:print]   zero_optimization_stage ...... 2
[2022-12-18 08:42:10,176] [INFO] [config.py:997:print_user_config]   json = {
    "fp16": {
        "enabled": true, 
        "loss_scale": 0, 
        "loss_scale_window": 1000, 
        "initial_scale_power": 16, 
        "hysteresis": 2, 
        "min_loss_scale": 1
    }, 
    "optimizer": {
        "type": "AdamW", 
        "params": {
            "lr": 1e-05, 
            "betas": [0.9, 0.999], 
            "eps": 1e-08, 
            "weight_decay": 0.0
        }
    }, 
    "scheduler": {
        "type": "WarmupDecayLR", 
        "params": {
            "last_batch_iteration": -1, 
            "total_num_steps": 5.000000e+03, 
            "warmup_min_lr": 0, 
            "warmup_max_lr": 1e-05, 
            "warmup_num_steps": 500
        }
    }, 
    "zero_optimization": {
        "stage": 2, 
        "offload_optimizer": {
            "device": "cpu", 
            "pin_memory": true
        }, 
        "allgather_partitions": true, 
        "allgather_bucket_size": 2.000000e+08, 
        "overlap_comm": true, 
        "reduce_scatter": true, 
        "reduce_bucket_size": 2.000000e+08, 
        "contiguous_gradients": true
    }, 
    "gradient_accumulation_steps": 1, 
    "gradient_clipping": 1.0, 
    "train_batch_size": 64, 
    "train_micro_batch_size_per_gpu": 64
}
[2022-12-18 08:44:27,389] [INFO] [stage_1_and_2.py:1767:step] [deepspeed] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 65536, reducing to 65536
[2022-12-18 08:44:43,482] [INFO] [stage_1_and_2.py:1767:step] [deepspeed] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 65536, reducing to 32768.0
[2022-12-18 08:45:01,180] [INFO] [stage_1_and_2.py:1767:step] [deepspeed] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 32768.0, reducing to 16384.0
[2022-12-18 08:45:17,756] [INFO] [stage_1_and_2.py:1767:step] [deepspeed] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 16384.0, reducing to 8192.0
[2022-12-18 08:47:03,107] [INFO] [logging.py:68:log_dist] [Rank 0] step=10, skipped=4, lr=[2.883141528559073e-06], mom=[[0.9, 0.999]]
[2022-12-18 08:47:03,108] [INFO] [timer.py:196:stop] epoch=0/micro_step=10/global_step=10, RunningAvgSamplesPerSec=17.81398816266197, CurrSamplesPerSec=17.536521056985904, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 08:49:58,788] [INFO] [logging.py:68:log_dist] [Rank 0] step=20, skipped=4, lr=[4.461405575910259e-06], mom=[[0.9, 0.999]]
[2022-12-18 08:49:58,790] [INFO] [timer.py:196:stop] epoch=0/micro_step=20/global_step=20, RunningAvgSamplesPerSec=17.707913433080698, CurrSamplesPerSec=17.67266334257317, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.817, 'learning_rate': 4.898977360288234e-06, 'epoch': 0.01}
[2022-12-18 08:52:59,924] [INFO] [logging.py:68:log_dist] [Rank 0] step=30, skipped=4, lr=[5.242641991936178e-06], mom=[[0.9, 0.999]]
[2022-12-18 08:52:59,926] [INFO] [timer.py:196:stop] epoch=0/micro_step=30/global_step=30, RunningAvgSamplesPerSec=17.624193801450975, CurrSamplesPerSec=17.42146956646065, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 08:56:04,927] [INFO] [logging.py:68:log_dist] [Rank 0] step=40, skipped=4, lr=[5.766283057118146e-06], mom=[[0.9, 0.999]]
[2022-12-18 08:56:04,928] [INFO] [timer.py:196:stop] epoch=0/micro_step=40/global_step=40, RunningAvgSamplesPerSec=17.611825644201268, CurrSamplesPerSec=17.199435261553543, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 08:59:02,160] [INFO] [logging.py:68:log_dist] [Rank 0] step=50, skipped=4, lr=[6.160712527409633e-06], mom=[[0.9, 0.999]]
[2022-12-18 08:59:02,163] [INFO] [timer.py:196:stop] epoch=0/micro_step=50/global_step=50, RunningAvgSamplesPerSec=17.57224681838786, CurrSamplesPerSec=17.471532518337153, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.3452, 'learning_rate': 6.160712527409633e-06, 'epoch': 0.01}
[2022-12-18 09:01:57,588] [INFO] [logging.py:68:log_dist] [Rank 0] step=60, skipped=4, lr=[6.4772414076394205e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:01:57,590] [INFO] [timer.py:196:stop] epoch=0/micro_step=60/global_step=60, RunningAvgSamplesPerSec=17.556437081890596, CurrSamplesPerSec=17.58886056667178, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:04:51,502] [INFO] [logging.py:68:log_dist] [Rank 0] step=70, skipped=4, lr=[6.741623406776245e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:04:51,503] [INFO] [timer.py:196:stop] epoch=0/micro_step=70/global_step=70, RunningAvgSamplesPerSec=17.550034157580424, CurrSamplesPerSec=17.404004920974277, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.3043, 'learning_rate': 6.85912902234906e-06, 'epoch': 0.01}
[2022-12-18 09:07:42,313] [INFO] [logging.py:68:log_dist] [Rank 0] step=80, skipped=4, lr=[6.968634661590082e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:07:42,314] [INFO] [timer.py:196:stop] epoch=0/micro_step=80/global_step=80, RunningAvgSamplesPerSec=17.54554576845597, CurrSamplesPerSec=17.300649307375817, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:10:37,574] [INFO] [logging.py:68:log_dist] [Rank 0] step=90, skipped=4, lr=[7.1675433522258775e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:10:37,575] [INFO] [timer.py:196:stop] epoch=0/micro_step=90/global_step=90, RunningAvgSamplesPerSec=17.54364083829183, CurrSamplesPerSec=17.543068524409126, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:13:33,056] [INFO] [logging.py:68:log_dist] [Rank 0] step=100, skipped=4, lr=[7.344547104469332e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:13:33,058] [INFO] [timer.py:196:stop] epoch=0/micro_step=100/global_step=100, RunningAvgSamplesPerSec=17.544745558767882, CurrSamplesPerSec=17.311613610419286, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.2484, 'learning_rate': 7.344547104469332e-06, 'epoch': 0.02}
[2022-12-18 09:16:00,166] [INFO] [logging.py:68:log_dist] [Rank 0] step=110, skipped=4, lr=[7.503995457567235e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:16:00,167] [INFO] [timer.py:196:stop] epoch=0/micro_step=110/global_step=110, RunningAvgSamplesPerSec=17.544294890083517, CurrSamplesPerSec=17.83244775214949, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:19:12,626] [INFO] [logging.py:68:log_dist] [Rank 0] step=120, skipped=4, lr=[7.649058662787184e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:19:12,627] [INFO] [timer.py:196:stop] epoch=0/micro_step=120/global_step=120, RunningAvgSamplesPerSec=17.573285029996587, CurrSamplesPerSec=17.949146033725896, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.1897, 'learning_rate': 7.716963756434345e-06, 'epoch': 1.0}
[2022-12-18 09:22:01,411] [INFO] [logging.py:68:log_dist] [Rank 0] step=130, skipped=4, lr=[7.782118888847307e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:22:01,412] [INFO] [timer.py:196:stop] epoch=0/micro_step=130/global_step=130, RunningAvgSamplesPerSec=17.55456949498131, CurrSamplesPerSec=17.13766063397266, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:24:51,224] [INFO] [logging.py:68:log_dist] [Rank 0] step=140, skipped=4, lr=[7.905011559752758e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:24:51,225] [INFO] [timer.py:196:stop] epoch=0/micro_step=140/global_step=140, RunningAvgSamplesPerSec=17.55073845541743, CurrSamplesPerSec=17.72275019960219, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:27:38,350] [INFO] [logging.py:68:log_dist] [Rank 0] step=150, skipped=4, lr=[8.019180844200955e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:27:38,352] [INFO] [timer.py:196:stop] epoch=0/micro_step=150/global_step=150, RunningAvgSamplesPerSec=17.54153626249399, CurrSamplesPerSec=17.28127515360052, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.1751, 'learning_rate': 8.019180844200955e-06, 'epoch': 1.01}
[2022-12-18 09:30:26,932] [INFO] [logging.py:68:log_dist] [Rank 0] step=160, skipped=4, lr=[8.125783520495252e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:30:26,933] [INFO] [timer.py:196:stop] epoch=0/micro_step=160/global_step=160, RunningAvgSamplesPerSec=17.54237608474746, CurrSamplesPerSec=17.425431147803835, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:33:14,947] [INFO] [logging.py:68:log_dist] [Rank 0] step=170, skipped=4, lr=[8.225760510392298e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:33:14,948] [INFO] [timer.py:196:stop] epoch=0/micro_step=170/global_step=170, RunningAvgSamplesPerSec=17.5394079030006, CurrSamplesPerSec=17.734454226542, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.1499, 'learning_rate': 8.27351214279797e-06, 'epoch': 1.01}
[2022-12-18 09:36:02,185] [INFO] [logging.py:68:log_dist] [Rank 0] step=180, skipped=4, lr=[8.31988745412743e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:36:02,186] [INFO] [timer.py:196:stop] epoch=0/micro_step=180/global_step=180, RunningAvgSamplesPerSec=17.537096713384305, CurrSamplesPerSec=17.369822224845038, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:38:51,989] [INFO] [logging.py:68:log_dist] [Rank 0] step=190, skipped=4, lr=[8.408811289387583e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:38:51,991] [INFO] [timer.py:196:stop] epoch=0/micro_step=190/global_step=190, RunningAvgSamplesPerSec=17.532972451853638, CurrSamplesPerSec=17.369981828636313, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:41:41,517] [INFO] [logging.py:68:log_dist] [Rank 0] step=200, skipped=4, lr=[8.49307723936858e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:41:41,519] [INFO] [timer.py:196:stop] epoch=0/micro_step=200/global_step=200, RunningAvgSamplesPerSec=17.521639611761227, CurrSamplesPerSec=17.05773687963696, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.145, 'learning_rate': 8.49307723936858e-06, 'epoch': 1.02}
[2022-12-18 09:44:30,225] [INFO] [logging.py:68:log_dist] [Rank 0] step=210, skipped=4, lr=[8.573149077803088e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:44:30,227] [INFO] [timer.py:196:stop] epoch=0/micro_step=210/global_step=210, RunningAvgSamplesPerSec=17.518351230897235, CurrSamplesPerSec=17.491189976865027, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:47:20,015] [INFO] [logging.py:68:log_dist] [Rank 0] step=220, skipped=4, lr=[8.64942458567722e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:47:20,016] [INFO] [timer.py:196:stop] epoch=0/micro_step=220/global_step=220, RunningAvgSamplesPerSec=17.52044478044767, CurrSamplesPerSec=17.551188301729674, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.1039, 'learning_rate': 8.686247975778677e-06, 'epoch': 1.02}
[2022-12-18 09:48:47,089] [INFO] [logging.py:68:log_dist] [Rank 0] step=230, skipped=4, lr=[8.722247506883805e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:48:47,090] [INFO] [timer.py:196:stop] epoch=0/micro_step=230/global_step=230, RunningAvgSamplesPerSec=17.52583657733623, CurrSamplesPerSec=17.795484760153986, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:52:53,020] [INFO] [logging.py:68:log_dist] [Rank 0] step=240, skipped=4, lr=[8.79191691333329e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:52:53,022] [INFO] [timer.py:196:stop] epoch=0/micro_step=240/global_step=240, RunningAvgSamplesPerSec=17.540278044164523, CurrSamplesPerSec=17.340279812152833, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 09:55:42,935] [INFO] [logging.py:68:log_dist] [Rank 0] step=250, skipped=4, lr=[8.858694625217149e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:55:42,936] [INFO] [timer.py:196:stop] epoch=0/micro_step=250/global_step=250, RunningAvgSamplesPerSec=17.528756039040783, CurrSamplesPerSec=17.753428194487324, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0958, 'learning_rate': 8.858694625217149e-06, 'epoch': 2.0}
[2022-12-18 09:58:31,035] [INFO] [logging.py:68:log_dist] [Rank 0] step=260, skipped=4, lr=[8.922811151820517e-06], mom=[[0.9, 0.999]]
[2022-12-18 09:58:31,036] [INFO] [timer.py:196:stop] epoch=0/micro_step=260/global_step=260, RunningAvgSamplesPerSec=17.51445195778724, CurrSamplesPerSec=17.76546084196235, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:01:19,374] [INFO] [logging.py:68:log_dist] [Rank 0] step=270, skipped=4, lr=[8.984470493319244e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:01:19,375] [INFO] [timer.py:196:stop] epoch=0/micro_step=270/global_step=270, RunningAvgSamplesPerSec=17.514860492931785, CurrSamplesPerSec=17.65513873928837, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.086, 'learning_rate': 9.014436199608479e-06, 'epoch': 2.01}
[2022-12-18 10:04:09,796] [INFO] [logging.py:68:log_dist] [Rank 0] step=280, skipped=4, lr=[9.043854055968706e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:04:09,797] [INFO] [timer.py:196:stop] epoch=0/micro_step=280/global_step=280, RunningAvgSamplesPerSec=17.507487376679528, CurrSamplesPerSec=17.419261679194896, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:06:58,165] [INFO] [logging.py:68:log_dist] [Rank 0] step=290, skipped=4, lr=[9.10112387015335e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:06:58,167] [INFO] [timer.py:196:stop] epoch=0/micro_step=290/global_step=290, RunningAvgSamplesPerSec=17.506782737249797, CurrSamplesPerSec=17.55021064009011, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:09:48,667] [INFO] [logging.py:68:log_dist] [Rank 0] step=300, skipped=4, lr=[9.156425255148058e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:09:48,669] [INFO] [timer.py:196:stop] epoch=0/micro_step=300/global_step=300, RunningAvgSamplesPerSec=17.505229793083824, CurrSamplesPerSec=17.57551569533738, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0686, 'learning_rate': 9.156425255148058e-06, 'epoch': 2.01}
[2022-12-18 10:12:41,600] [INFO] [logging.py:68:log_dist] [Rank 0] step=310, skipped=4, lr=[9.209889040960644e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:12:41,601] [INFO] [timer.py:196:stop] epoch=0/micro_step=310/global_step=310, RunningAvgSamplesPerSec=17.506193858975763, CurrSamplesPerSec=17.60106809421937, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:15:34,448] [INFO] [logging.py:68:log_dist] [Rank 0] step=320, skipped=4, lr=[9.261633432763397e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:15:34,449] [INFO] [timer.py:196:stop] epoch=0/micro_step=320/global_step=320, RunningAvgSamplesPerSec=17.502672734107193, CurrSamplesPerSec=17.71995293610675, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0684, 'learning_rate': 9.28689473531776e-06, 'epoch': 2.02}
[2022-12-18 10:18:25,400] [INFO] [logging.py:68:log_dist] [Rank 0] step=330, skipped=4, lr=[9.311765584761373e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:18:25,402] [INFO] [timer.py:196:stop] epoch=0/micro_step=330/global_step=330, RunningAvgSamplesPerSec=17.502310309234087, CurrSamplesPerSec=17.587128551283918, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:21:18,016] [INFO] [logging.py:68:log_dist] [Rank 0] step=340, skipped=4, lr=[9.360382936198493e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:21:18,018] [INFO] [timer.py:196:stop] epoch=0/micro_step=340/global_step=340, RunningAvgSamplesPerSec=17.500643734444402, CurrSamplesPerSec=17.005072277531735, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:24:04,872] [INFO] [logging.py:68:log_dist] [Rank 0] step=350, skipped=4, lr=[9.407574351377137e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:24:04,874] [INFO] [timer.py:196:stop] epoch=0/micro_step=350/global_step=350, RunningAvgSamplesPerSec=17.51464932561927, CurrSamplesPerSec=17.630059089676152, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0482, 'learning_rate': 9.407574351377137e-06, 'epoch': 3.0}
[2022-12-18 10:26:57,641] [INFO] [logging.py:68:log_dist] [Rank 0] step=360, skipped=4, lr=[9.45342109721062e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:26:57,643] [INFO] [timer.py:196:stop] epoch=0/micro_step=360/global_step=360, RunningAvgSamplesPerSec=17.505842284973642, CurrSamplesPerSec=17.025209457256768, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:29:51,036] [INFO] [logging.py:68:log_dist] [Rank 0] step=370, skipped=4, lr=[9.497997685324628e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:29:51,038] [INFO] [timer.py:196:stop] epoch=0/micro_step=370/global_step=370, RunningAvgSamplesPerSec=17.505074655498845, CurrSamplesPerSec=17.17602944826478, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0504, 'learning_rate': 9.519831289296397e-06, 'epoch': 3.01}
[2022-12-18 10:32:41,738] [INFO] [logging.py:68:log_dist] [Rank 0] step=380, skipped=4, lr=[9.541372600623587e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:32:41,739] [INFO] [timer.py:196:stop] epoch=0/micro_step=380/global_step=380, RunningAvgSamplesPerSec=17.50077202769039, CurrSamplesPerSec=17.66240728131972, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:35:32,596] [INFO] [logging.py:68:log_dist] [Rank 0] step=390, skipped=4, lr=[9.583608934209288e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:35:32,597] [INFO] [timer.py:196:stop] epoch=0/micro_step=390/global_step=390, RunningAvgSamplesPerSec=17.502615176175013, CurrSamplesPerSec=17.671288199732384, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:38:24,713] [INFO] [logging.py:68:log_dist] [Rank 0] step=400, skipped=4, lr=[9.624764935335318e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:38:24,714] [INFO] [timer.py:196:stop] epoch=0/micro_step=400/global_step=400, RunningAvgSamplesPerSec=17.501141402908466, CurrSamplesPerSec=17.413306679702867, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0446, 'learning_rate': 9.624764935335318e-06, 'epoch': 3.01}
[2022-12-18 10:41:14,089] [INFO] [logging.py:68:log_dist] [Rank 0] step=410, skipped=4, lr=[9.664894494516345e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:41:14,090] [INFO] [timer.py:196:stop] epoch=0/micro_step=410/global_step=410, RunningAvgSamplesPerSec=17.50299955137452, CurrSamplesPerSec=16.856797996893206, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:44:03,121] [INFO] [logging.py:68:log_dist] [Rank 0] step=420, skipped=4, lr=[9.704047567846437e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:44:03,123] [INFO] [timer.py:196:stop] epoch=0/micro_step=420/global_step=420, RunningAvgSamplesPerSec=17.49955598397826, CurrSamplesPerSec=17.39210969006482, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0396, 'learning_rate': 9.723272550712454e-06, 'epoch': 3.02}
[2022-12-18 10:46:53,231] [INFO] [logging.py:68:log_dist] [Rank 0] step=430, skipped=4, lr=[9.742270550908135e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:46:53,233] [INFO] [timer.py:196:stop] epoch=0/micro_step=430/global_step=430, RunningAvgSamplesPerSec=17.502169263783284, CurrSamplesPerSec=17.82420181661996, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:49:44,733] [INFO] [logging.py:68:log_dist] [Rank 0] step=440, skipped=4, lr=[9.779606609292176e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:49:44,734] [INFO] [timer.py:196:stop] epoch=0/micro_step=440/global_step=440, RunningAvgSamplesPerSec=17.502468898132452, CurrSamplesPerSec=17.01432443514231, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:52:34,347] [INFO] [logging.py:68:log_dist] [Rank 0] step=450, skipped=4, lr=[9.816095971633122e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:52:34,348] [INFO] [timer.py:196:stop] epoch=0/micro_step=450/global_step=450, RunningAvgSamplesPerSec=17.50661530301726, CurrSamplesPerSec=17.68532190940408, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0309, 'learning_rate': 9.816095971633122e-06, 'epoch': 3.02}
[2022-12-18 10:54:31,089] [INFO] [logging.py:68:log_dist] [Rank 0] step=460, skipped=4, lr=[9.851776190149156e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:54:31,091] [INFO] [timer.py:196:stop] epoch=0/micro_step=460/global_step=460, RunningAvgSamplesPerSec=17.505343508354425, CurrSamplesPerSec=17.622945563320123, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 10:58:09,980] [INFO] [logging.py:68:log_dist] [Rank 0] step=470, skipped=4, lr=[9.886682372916766e-06], mom=[[0.9, 0.999]]
[2022-12-18 10:58:09,982] [INFO] [timer.py:196:stop] epoch=0/micro_step=470/global_step=470, RunningAvgSamplesPerSec=17.51432072204074, CurrSamplesPerSec=17.560018580808034, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0269, 'learning_rate': 9.90385555539545e-06, 'epoch': 4.0}
[2022-12-18 11:01:02,577] [INFO] [logging.py:68:log_dist] [Rank 0] step=480, skipped=4, lr=[9.92084739148192e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:01:02,579] [INFO] [timer.py:196:stop] epoch=0/micro_step=480/global_step=480, RunningAvgSamplesPerSec=17.51471041450659, CurrSamplesPerSec=17.770770916559375, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:03:53,720] [INFO] [logging.py:68:log_dist] [Rank 0] step=490, skipped=4, lr=[9.954302066885107e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:03:53,722] [INFO] [timer.py:196:stop] epoch=0/micro_step=490/global_step=490, RunningAvgSamplesPerSec=17.514707806076828, CurrSamplesPerSec=17.474965145825077, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:06:40,711] [INFO] [logging.py:68:log_dist] [Rank 0] step=500, skipped=4, lr=[9.987075336738768e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:06:40,713] [INFO] [timer.py:196:stop] epoch=0/micro_step=500/global_step=500, RunningAvgSamplesPerSec=17.51673358493395, CurrSamplesPerSec=17.728490220205128, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.03, 'learning_rate': 9.987075336738768e-06, 'epoch': 4.01}
[2022-12-18 11:09:32,253] [INFO] [logging.py:68:log_dist] [Rank 0] step=510, skipped=4, lr=[9.98888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:09:32,254] [INFO] [timer.py:196:stop] epoch=0/micro_step=510/global_step=510, RunningAvgSamplesPerSec=17.515181954535795, CurrSamplesPerSec=17.44658815742413, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:12:24,243] [INFO] [logging.py:68:log_dist] [Rank 0] step=520, skipped=4, lr=[9.966666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:12:24,244] [INFO] [timer.py:196:stop] epoch=0/micro_step=520/global_step=520, RunningAvgSamplesPerSec=17.516449396915945, CurrSamplesPerSec=17.71918211824117, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.025, 'learning_rate': 9.955555555555556e-06, 'epoch': 4.01}
[2022-12-18 11:15:10,174] [INFO] [logging.py:68:log_dist] [Rank 0] step=530, skipped=4, lr=[9.944444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:15:10,175] [INFO] [timer.py:196:stop] epoch=0/micro_step=530/global_step=530, RunningAvgSamplesPerSec=17.515811116445363, CurrSamplesPerSec=17.44589309419759, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:17:59,140] [INFO] [logging.py:68:log_dist] [Rank 0] step=540, skipped=4, lr=[9.922222222222222e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:17:59,141] [INFO] [timer.py:196:stop] epoch=0/micro_step=540/global_step=540, RunningAvgSamplesPerSec=17.51307013237252, CurrSamplesPerSec=17.709312456169634, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:20:51,468] [INFO] [logging.py:68:log_dist] [Rank 0] step=550, skipped=4, lr=[9.9e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:20:51,469] [INFO] [timer.py:196:stop] epoch=0/micro_step=550/global_step=550, RunningAvgSamplesPerSec=17.51360962534701, CurrSamplesPerSec=17.514523814355172, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0214, 'learning_rate': 9.9e-06, 'epoch': 4.02}
[2022-12-18 11:23:41,540] [INFO] [logging.py:68:log_dist] [Rank 0] step=560, skipped=4, lr=[9.877777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:23:41,542] [INFO] [timer.py:196:stop] epoch=0/micro_step=560/global_step=560, RunningAvgSamplesPerSec=17.515173467375842, CurrSamplesPerSec=17.700679243083353, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:26:28,412] [INFO] [logging.py:68:log_dist] [Rank 0] step=570, skipped=4, lr=[9.855555555555555e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:26:28,413] [INFO] [timer.py:196:stop] epoch=0/micro_step=570/global_step=570, RunningAvgSamplesPerSec=17.515792660700566, CurrSamplesPerSec=17.636638364608388, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0176, 'learning_rate': 9.844444444444446e-06, 'epoch': 4.02}
[2022-12-18 11:27:30,542] [INFO] [logging.py:68:log_dist] [Rank 0] step=580, skipped=4, lr=[9.833333333333333e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:27:30,544] [INFO] [timer.py:196:stop] epoch=0/micro_step=580/global_step=580, RunningAvgSamplesPerSec=17.525312485669467, CurrSamplesPerSec=22.15033486404057, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:31:57,731] [INFO] [logging.py:68:log_dist] [Rank 0] step=590, skipped=4, lr=[9.811111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:31:57,733] [INFO] [timer.py:196:stop] epoch=0/micro_step=590/global_step=590, RunningAvgSamplesPerSec=17.52578604906633, CurrSamplesPerSec=17.542104377157184, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:34:45,506] [INFO] [logging.py:68:log_dist] [Rank 0] step=600, skipped=4, lr=[9.78888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:34:45,507] [INFO] [timer.py:196:stop] epoch=0/micro_step=600/global_step=600, RunningAvgSamplesPerSec=17.525721346081685, CurrSamplesPerSec=17.728407089792164, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0186, 'learning_rate': 9.78888888888889e-06, 'epoch': 5.0}
[2022-12-18 11:37:32,907] [INFO] [logging.py:68:log_dist] [Rank 0] step=610, skipped=4, lr=[9.766666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:37:32,909] [INFO] [timer.py:196:stop] epoch=0/micro_step=610/global_step=610, RunningAvgSamplesPerSec=17.52620539215051, CurrSamplesPerSec=17.665249180375554, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:40:23,836] [INFO] [logging.py:68:log_dist] [Rank 0] step=620, skipped=4, lr=[9.744444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:40:23,837] [INFO] [timer.py:196:stop] epoch=0/micro_step=620/global_step=620, RunningAvgSamplesPerSec=17.524553828947862, CurrSamplesPerSec=17.647862254407272, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0179, 'learning_rate': 9.733333333333334e-06, 'epoch': 5.01}
[2022-12-18 11:43:10,955] [INFO] [logging.py:68:log_dist] [Rank 0] step=630, skipped=4, lr=[9.722222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:43:10,957] [INFO] [timer.py:196:stop] epoch=0/micro_step=630/global_step=630, RunningAvgSamplesPerSec=17.524283076579064, CurrSamplesPerSec=17.71436339529198, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:45:59,599] [INFO] [logging.py:68:log_dist] [Rank 0] step=640, skipped=4, lr=[9.7e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:45:59,600] [INFO] [timer.py:196:stop] epoch=0/micro_step=640/global_step=640, RunningAvgSamplesPerSec=17.5239493520232, CurrSamplesPerSec=17.29191637538188, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:48:51,158] [INFO] [logging.py:68:log_dist] [Rank 0] step=650, skipped=4, lr=[9.677777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:48:51,159] [INFO] [timer.py:196:stop] epoch=0/micro_step=650/global_step=650, RunningAvgSamplesPerSec=17.520107118857844, CurrSamplesPerSec=17.428385128642994, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0141, 'learning_rate': 9.677777777777778e-06, 'epoch': 5.01}
[2022-12-18 11:51:44,941] [INFO] [logging.py:68:log_dist] [Rank 0] step=660, skipped=4, lr=[9.655555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:51:44,942] [INFO] [timer.py:196:stop] epoch=0/micro_step=660/global_step=660, RunningAvgSamplesPerSec=17.519444727080156, CurrSamplesPerSec=16.994613216593315, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:54:37,926] [INFO] [logging.py:68:log_dist] [Rank 0] step=670, skipped=4, lr=[9.633333333333335e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:54:37,928] [INFO] [timer.py:196:stop] epoch=0/micro_step=670/global_step=670, RunningAvgSamplesPerSec=17.519984376324164, CurrSamplesPerSec=17.66502830473203, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0131, 'learning_rate': 9.622222222222222e-06, 'epoch': 5.02}
[2022-12-18 11:57:26,593] [INFO] [logging.py:68:log_dist] [Rank 0] step=680, skipped=4, lr=[9.611111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:57:26,595] [INFO] [timer.py:196:stop] epoch=0/micro_step=680/global_step=680, RunningAvgSamplesPerSec=17.51917469369641, CurrSamplesPerSec=17.81082610549716, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 11:59:46,615] [INFO] [logging.py:68:log_dist] [Rank 0] step=690, skipped=4, lr=[9.58888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 11:59:46,616] [INFO] [timer.py:196:stop] epoch=0/micro_step=690/global_step=690, RunningAvgSamplesPerSec=17.520953040279323, CurrSamplesPerSec=17.893703708073588, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:03:02,532] [INFO] [logging.py:68:log_dist] [Rank 0] step=700, skipped=4, lr=[9.566666666666668e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:03:02,534] [INFO] [timer.py:196:stop] epoch=0/micro_step=700/global_step=700, RunningAvgSamplesPerSec=17.526881403928282, CurrSamplesPerSec=17.811006916180688, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0111, 'learning_rate': 9.566666666666668e-06, 'epoch': 6.0}
[2022-12-18 12:05:50,396] [INFO] [logging.py:68:log_dist] [Rank 0] step=710, skipped=4, lr=[9.544444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:05:50,397] [INFO] [timer.py:196:stop] epoch=0/micro_step=710/global_step=710, RunningAvgSamplesPerSec=17.525663669476064, CurrSamplesPerSec=17.627963556175473, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:08:40,501] [INFO] [logging.py:68:log_dist] [Rank 0] step=720, skipped=4, lr=[9.522222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:08:40,503] [INFO] [timer.py:196:stop] epoch=0/micro_step=720/global_step=720, RunningAvgSamplesPerSec=17.522315207058163, CurrSamplesPerSec=17.08808021934021, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0115, 'learning_rate': 9.511111111111112e-06, 'epoch': 6.01}
[2022-12-18 12:11:29,479] [INFO] [logging.py:68:log_dist] [Rank 0] step=730, skipped=4, lr=[9.5e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:11:29,481] [INFO] [timer.py:196:stop] epoch=0/micro_step=730/global_step=730, RunningAvgSamplesPerSec=17.522397647878112, CurrSamplesPerSec=17.69114963575808, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:14:18,585] [INFO] [logging.py:68:log_dist] [Rank 0] step=740, skipped=4, lr=[9.47777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:14:18,587] [INFO] [timer.py:196:stop] epoch=0/micro_step=740/global_step=740, RunningAvgSamplesPerSec=17.522899649689435, CurrSamplesPerSec=17.835842357826117, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:17:07,842] [INFO] [logging.py:68:log_dist] [Rank 0] step=750, skipped=4, lr=[9.455555555555557e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:17:07,843] [INFO] [timer.py:196:stop] epoch=0/micro_step=750/global_step=750, RunningAvgSamplesPerSec=17.52216964269384, CurrSamplesPerSec=17.715532463583024, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0097, 'learning_rate': 9.455555555555557e-06, 'epoch': 6.01}
[2022-12-18 12:19:56,259] [INFO] [logging.py:68:log_dist] [Rank 0] step=760, skipped=4, lr=[9.433333333333335e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:19:56,260] [INFO] [timer.py:196:stop] epoch=0/micro_step=760/global_step=760, RunningAvgSamplesPerSec=17.52350085830269, CurrSamplesPerSec=17.825603229647047, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:22:43,814] [INFO] [logging.py:68:log_dist] [Rank 0] step=770, skipped=4, lr=[9.411111111111113e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:22:43,816] [INFO] [timer.py:196:stop] epoch=0/micro_step=770/global_step=770, RunningAvgSamplesPerSec=17.523872671951036, CurrSamplesPerSec=17.464620175491046, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0091, 'learning_rate': 9.4e-06, 'epoch': 6.02}
[2022-12-18 12:25:34,399] [INFO] [logging.py:68:log_dist] [Rank 0] step=780, skipped=4, lr=[9.38888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:25:34,401] [INFO] [timer.py:196:stop] epoch=0/micro_step=780/global_step=780, RunningAvgSamplesPerSec=17.523888473835193, CurrSamplesPerSec=17.62105645548852, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:28:22,069] [INFO] [logging.py:68:log_dist] [Rank 0] step=790, skipped=4, lr=[9.366666666666668e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:28:22,070] [INFO] [timer.py:196:stop] epoch=0/micro_step=790/global_step=790, RunningAvgSamplesPerSec=17.52396646475176, CurrSamplesPerSec=17.7942001353745, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:31:10,611] [INFO] [logging.py:68:log_dist] [Rank 0] step=800, skipped=4, lr=[9.344444444444446e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:31:10,613] [INFO] [timer.py:196:stop] epoch=0/micro_step=800/global_step=800, RunningAvgSamplesPerSec=17.52362785335887, CurrSamplesPerSec=17.645739287742696, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0099, 'learning_rate': 9.344444444444446e-06, 'epoch': 6.02}
[2022-12-18 12:32:39,816] [INFO] [logging.py:68:log_dist] [Rank 0] step=810, skipped=4, lr=[9.322222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:32:39,818] [INFO] [timer.py:196:stop] epoch=0/micro_step=810/global_step=810, RunningAvgSamplesPerSec=17.525217387924933, CurrSamplesPerSec=17.6307387978646, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:36:44,201] [INFO] [logging.py:68:log_dist] [Rank 0] step=820, skipped=4, lr=[9.3e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:36:44,203] [INFO] [timer.py:196:stop] epoch=0/micro_step=820/global_step=820, RunningAvgSamplesPerSec=17.527549173776915, CurrSamplesPerSec=17.24311195262599, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0073, 'learning_rate': 9.28888888888889e-06, 'epoch': 7.0}
[2022-12-18 12:39:35,915] [INFO] [logging.py:68:log_dist] [Rank 0] step=830, skipped=4, lr=[9.277777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:39:35,917] [INFO] [timer.py:196:stop] epoch=0/micro_step=830/global_step=830, RunningAvgSamplesPerSec=17.527139118776905, CurrSamplesPerSec=17.824774664911605, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:42:25,214] [INFO] [logging.py:68:log_dist] [Rank 0] step=840, skipped=4, lr=[9.255555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:42:25,215] [INFO] [timer.py:196:stop] epoch=0/micro_step=840/global_step=840, RunningAvgSamplesPerSec=17.529876938383726, CurrSamplesPerSec=17.677469879502205, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:45:16,974] [INFO] [logging.py:68:log_dist] [Rank 0] step=850, skipped=4, lr=[9.233333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:45:16,976] [INFO] [timer.py:196:stop] epoch=0/micro_step=850/global_step=850, RunningAvgSamplesPerSec=17.533049828024673, CurrSamplesPerSec=17.934881195101724, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0062, 'learning_rate': 9.233333333333334e-06, 'epoch': 7.01}
[2022-12-18 12:48:07,233] [INFO] [logging.py:68:log_dist] [Rank 0] step=860, skipped=4, lr=[9.211111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:48:07,234] [INFO] [timer.py:196:stop] epoch=0/micro_step=860/global_step=860, RunningAvgSamplesPerSec=17.536686576379214, CurrSamplesPerSec=17.72774792834313, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:50:56,254] [INFO] [logging.py:68:log_dist] [Rank 0] step=870, skipped=4, lr=[9.188888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:50:56,255] [INFO] [timer.py:196:stop] epoch=0/micro_step=870/global_step=870, RunningAvgSamplesPerSec=17.53939897716571, CurrSamplesPerSec=17.863981354170473, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.006, 'learning_rate': 9.17777777777778e-06, 'epoch': 7.01}
[2022-12-18 12:53:49,512] [INFO] [logging.py:68:log_dist] [Rank 0] step=880, skipped=4, lr=[9.166666666666666e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:53:49,514] [INFO] [timer.py:196:stop] epoch=0/micro_step=880/global_step=880, RunningAvgSamplesPerSec=17.54251129449292, CurrSamplesPerSec=17.672755259073455, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:56:39,481] [INFO] [logging.py:68:log_dist] [Rank 0] step=890, skipped=4, lr=[9.144444444444444e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:56:39,482] [INFO] [timer.py:196:stop] epoch=0/micro_step=890/global_step=890, RunningAvgSamplesPerSec=17.545431248601155, CurrSamplesPerSec=17.820363263008623, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 12:59:28,307] [INFO] [logging.py:68:log_dist] [Rank 0] step=900, skipped=4, lr=[9.122222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 12:59:28,309] [INFO] [timer.py:196:stop] epoch=0/micro_step=900/global_step=900, RunningAvgSamplesPerSec=17.5474477902634, CurrSamplesPerSec=17.84197843798558, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0046, 'learning_rate': 9.122222222222223e-06, 'epoch': 7.02}
[2022-12-18 13:02:23,366] [INFO] [logging.py:68:log_dist] [Rank 0] step=910, skipped=4, lr=[9.100000000000001e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:02:23,368] [INFO] [timer.py:196:stop] epoch=0/micro_step=910/global_step=910, RunningAvgSamplesPerSec=17.54704328541452, CurrSamplesPerSec=17.599693685140846, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 13:05:10,345] [INFO] [logging.py:68:log_dist] [Rank 0] step=920, skipped=4, lr=[9.077777777777779e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:05:10,346] [INFO] [timer.py:196:stop] epoch=0/micro_step=920/global_step=920, RunningAvgSamplesPerSec=17.548757345088557, CurrSamplesPerSec=17.80023677023281, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0045, 'learning_rate': 9.066666666666667e-06, 'epoch': 7.02}
[2022-12-18 13:07:52,456] [INFO] [logging.py:68:log_dist] [Rank 0] step=930, skipped=4, lr=[9.055555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:07:52,457] [INFO] [timer.py:196:stop] epoch=0/micro_step=930/global_step=930, RunningAvgSamplesPerSec=17.555835577865636, CurrSamplesPerSec=16.947261645741815, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 13:10:39,445] [INFO] [logging.py:68:log_dist] [Rank 0] step=940, skipped=4, lr=[9.033333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:10:39,447] [INFO] [timer.py:196:stop] epoch=0/micro_step=940/global_step=940, RunningAvgSamplesPerSec=17.558626670023333, CurrSamplesPerSec=17.837192265183017, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 13:13:28,825] [INFO] [logging.py:68:log_dist] [Rank 0] step=950, skipped=4, lr=[9.011111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:13:28,827] [INFO] [timer.py:196:stop] epoch=0/micro_step=950/global_step=950, RunningAvgSamplesPerSec=17.561116636037408, CurrSamplesPerSec=17.875913197667035, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0053, 'learning_rate': 9.011111111111111e-06, 'epoch': 8.0}
[2022-12-18 13:16:15,728] [INFO] [logging.py:68:log_dist] [Rank 0] step=960, skipped=4, lr=[8.988888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:16:15,729] [INFO] [timer.py:196:stop] epoch=0/micro_step=960/global_step=960, RunningAvgSamplesPerSec=17.56308806750036, CurrSamplesPerSec=17.781870571307575, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 13:19:02,146] [INFO] [logging.py:68:log_dist] [Rank 0] step=970, skipped=4, lr=[8.966666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:19:02,147] [INFO] [timer.py:196:stop] epoch=0/micro_step=970/global_step=970, RunningAvgSamplesPerSec=17.564578542499007, CurrSamplesPerSec=17.74904968508855, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0039, 'learning_rate': 8.955555555555555e-06, 'epoch': 8.01}
[2022-12-18 13:21:50,887] [INFO] [logging.py:68:log_dist] [Rank 0] step=980, skipped=4, lr=[8.944444444444446e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:21:50,889] [INFO] [timer.py:196:stop] epoch=0/micro_step=980/global_step=980, RunningAvgSamplesPerSec=17.566081411767524, CurrSamplesPerSec=18.11684822516133, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 13:24:37,286] [INFO] [logging.py:68:log_dist] [Rank 0] step=990, skipped=4, lr=[8.922222222222224e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:24:37,287] [INFO] [timer.py:196:stop] epoch=0/micro_step=990/global_step=990, RunningAvgSamplesPerSec=17.56737818663697, CurrSamplesPerSec=17.898017842980305, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 13:27:24,217] [INFO] [logging.py:68:log_dist] [Rank 0] step=1000, skipped=4, lr=[8.900000000000001e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:27:24,219] [INFO] [timer.py:196:stop] epoch=0/micro_step=1000/global_step=1000, RunningAvgSamplesPerSec=17.569257091709765, CurrSamplesPerSec=17.374920879792757, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0046, 'learning_rate': 8.900000000000001e-06, 'epoch': 8.01}
{'eval_loss': 0.28076171875, 'eval_wer': 17.571297148114077, 'eval_runtime': 1237.4696, 'eval_samples_per_second': 3.118, 'eval_steps_per_second': 0.098, 'epoch': 8.01}
[2022-12-18 13:48:02,674] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step1000 is begin to save!
[2022-12-18 13:48:02,684] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: ./checkpoint-1000/global_step1000/mp_rank_00_model_states.pt
[2022-12-18 13:48:02,684] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving ./checkpoint-1000/global_step1000/mp_rank_00_model_states.pt...
[2022-12-18 13:48:03,680] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved ./checkpoint-1000/global_step1000/mp_rank_00_model_states.pt.
[2022-12-18 13:48:03,682] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving ./checkpoint-1000/global_step1000/zero_pp_rank_0_mp_rank_00_optim_states.pt...
[2022-12-18 13:48:08,206] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved ./checkpoint-1000/global_step1000/zero_pp_rank_0_mp_rank_00_optim_states.pt.
[2022-12-18 13:48:08,208] [INFO] [engine.py:3394:_save_zero_checkpoint] zero checkpoint saved ./checkpoint-1000/global_step1000/zero_pp_rank_0_mp_rank_00_optim_states.pt
[2022-12-18 13:48:08,208] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now!
[2022-12-18 13:51:32,738] [INFO] [logging.py:68:log_dist] [Rank 0] step=1010, skipped=4, lr=[8.877777777777779e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:51:32,740] [INFO] [timer.py:196:stop] epoch=0/micro_step=1010/global_step=1010, RunningAvgSamplesPerSec=17.570497186974222, CurrSamplesPerSec=17.771050915545032, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 13:54:24,427] [INFO] [logging.py:68:log_dist] [Rank 0] step=1020, skipped=4, lr=[8.855555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:54:24,429] [INFO] [timer.py:196:stop] epoch=0/micro_step=1020/global_step=1020, RunningAvgSamplesPerSec=17.572388134802974, CurrSamplesPerSec=17.892674397538443, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0037, 'learning_rate': 8.844444444444445e-06, 'epoch': 8.02}
[2022-12-18 13:57:13,310] [INFO] [logging.py:68:log_dist] [Rank 0] step=1030, skipped=4, lr=[8.833333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:57:13,312] [INFO] [timer.py:196:stop] epoch=0/micro_step=1030/global_step=1030, RunningAvgSamplesPerSec=17.574865744089255, CurrSamplesPerSec=17.830042096987665, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 13:59:11,204] [INFO] [logging.py:68:log_dist] [Rank 0] step=1040, skipped=4, lr=[8.811111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 13:59:11,205] [INFO] [timer.py:196:stop] epoch=0/micro_step=1040/global_step=1040, RunningAvgSamplesPerSec=17.578049897136907, CurrSamplesPerSec=17.867588966876575, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:02:49,022] [INFO] [logging.py:68:log_dist] [Rank 0] step=1050, skipped=4, lr=[8.788888888888891e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:02:49,024] [INFO] [timer.py:196:stop] epoch=0/micro_step=1050/global_step=1050, RunningAvgSamplesPerSec=17.580728468924875, CurrSamplesPerSec=16.936598922699496, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0035, 'learning_rate': 8.788888888888891e-06, 'epoch': 9.0}
[2022-12-18 14:05:33,620] [INFO] [logging.py:68:log_dist] [Rank 0] step=1060, skipped=4, lr=[8.766666666666669e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:05:33,622] [INFO] [timer.py:196:stop] epoch=0/micro_step=1060/global_step=1060, RunningAvgSamplesPerSec=17.58019104202516, CurrSamplesPerSec=17.78021928381013, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:08:19,806] [INFO] [logging.py:68:log_dist] [Rank 0] step=1070, skipped=4, lr=[8.744444444444446e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:08:19,808] [INFO] [timer.py:196:stop] epoch=0/micro_step=1070/global_step=1070, RunningAvgSamplesPerSec=17.58085065511869, CurrSamplesPerSec=17.769715705552866, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0038, 'learning_rate': 8.733333333333333e-06, 'epoch': 9.01}
[2022-12-18 14:11:06,335] [INFO] [logging.py:68:log_dist] [Rank 0] step=1080, skipped=4, lr=[8.722222222222224e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:11:06,337] [INFO] [timer.py:196:stop] epoch=0/micro_step=1080/global_step=1080, RunningAvgSamplesPerSec=17.579904119591816, CurrSamplesPerSec=17.725968554296802, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:13:51,062] [INFO] [logging.py:68:log_dist] [Rank 0] step=1090, skipped=4, lr=[8.700000000000001e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:13:51,064] [INFO] [timer.py:196:stop] epoch=0/micro_step=1090/global_step=1090, RunningAvgSamplesPerSec=17.57849146162872, CurrSamplesPerSec=17.506514413151734, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:16:41,006] [INFO] [logging.py:68:log_dist] [Rank 0] step=1100, skipped=4, lr=[8.677777777777779e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:16:41,008] [INFO] [timer.py:196:stop] epoch=0/micro_step=1100/global_step=1100, RunningAvgSamplesPerSec=17.577930456746696, CurrSamplesPerSec=17.35344135867515, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.003, 'learning_rate': 8.677777777777779e-06, 'epoch': 9.01}
[2022-12-18 14:19:28,216] [INFO] [logging.py:68:log_dist] [Rank 0] step=1110, skipped=4, lr=[8.655555555555557e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:19:28,217] [INFO] [timer.py:196:stop] epoch=0/micro_step=1110/global_step=1110, RunningAvgSamplesPerSec=17.57769973626041, CurrSamplesPerSec=17.8288981283458, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:22:17,888] [INFO] [logging.py:68:log_dist] [Rank 0] step=1120, skipped=4, lr=[8.633333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:22:17,890] [INFO] [timer.py:196:stop] epoch=0/micro_step=1120/global_step=1120, RunningAvgSamplesPerSec=17.578493365622567, CurrSamplesPerSec=17.76609576686019, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0037, 'learning_rate': 8.622222222222223e-06, 'epoch': 9.02}
[2022-12-18 14:25:16,389] [INFO] [logging.py:68:log_dist] [Rank 0] step=1130, skipped=4, lr=[8.611111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:25:16,391] [INFO] [timer.py:196:stop] epoch=0/micro_step=1130/global_step=1130, RunningAvgSamplesPerSec=17.57871703334657, CurrSamplesPerSec=17.785122219131072, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:28:10,876] [INFO] [logging.py:68:log_dist] [Rank 0] step=1140, skipped=4, lr=[8.58888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:28:10,877] [INFO] [timer.py:196:stop] epoch=0/micro_step=1140/global_step=1140, RunningAvgSamplesPerSec=17.579459740076313, CurrSamplesPerSec=17.701639889108993, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:31:05,041] [INFO] [logging.py:68:log_dist] [Rank 0] step=1150, skipped=4, lr=[8.566666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:31:05,043] [INFO] [timer.py:196:stop] epoch=0/micro_step=1150/global_step=1150, RunningAvgSamplesPerSec=17.580003764189616, CurrSamplesPerSec=17.80629521962796, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0036, 'learning_rate': 8.566666666666667e-06, 'epoch': 9.02}
[2022-12-18 14:32:07,420] [INFO] [logging.py:68:log_dist] [Rank 0] step=1160, skipped=4, lr=[8.544444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:32:07,422] [INFO] [timer.py:196:stop] epoch=0/micro_step=1160/global_step=1160, RunningAvgSamplesPerSec=17.585442663890536, CurrSamplesPerSec=23.566298323625077, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:36:43,234] [INFO] [logging.py:68:log_dist] [Rank 0] step=1170, skipped=4, lr=[8.522222222222222e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:36:43,236] [INFO] [timer.py:196:stop] epoch=0/micro_step=1170/global_step=1170, RunningAvgSamplesPerSec=17.586206519395216, CurrSamplesPerSec=17.676859898043542, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0034, 'learning_rate': 8.511111111111113e-06, 'epoch': 10.0}
[2022-12-18 14:39:41,206] [INFO] [logging.py:68:log_dist] [Rank 0] step=1180, skipped=4, lr=[8.5e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:39:41,207] [INFO] [timer.py:196:stop] epoch=0/micro_step=1180/global_step=1180, RunningAvgSamplesPerSec=17.58715483416817, CurrSamplesPerSec=17.49722916739182, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:42:39,474] [INFO] [logging.py:68:log_dist] [Rank 0] step=1190, skipped=4, lr=[8.477777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:42:39,476] [INFO] [timer.py:196:stop] epoch=0/micro_step=1190/global_step=1190, RunningAvgSamplesPerSec=17.587668127524644, CurrSamplesPerSec=17.847793588674662, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:45:31,663] [INFO] [logging.py:68:log_dist] [Rank 0] step=1200, skipped=4, lr=[8.455555555555555e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:45:31,664] [INFO] [timer.py:196:stop] epoch=0/micro_step=1200/global_step=1200, RunningAvgSamplesPerSec=17.588774465446917, CurrSamplesPerSec=17.72776783125711, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0032, 'learning_rate': 8.455555555555555e-06, 'epoch': 10.01}
[2022-12-18 14:48:31,335] [INFO] [logging.py:68:log_dist] [Rank 0] step=1210, skipped=4, lr=[8.433333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:48:31,336] [INFO] [timer.py:196:stop] epoch=0/micro_step=1210/global_step=1210, RunningAvgSamplesPerSec=17.59026563335073, CurrSamplesPerSec=17.859623009470603, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:51:22,847] [INFO] [logging.py:68:log_dist] [Rank 0] step=1220, skipped=4, lr=[8.411111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:51:22,848] [INFO] [timer.py:196:stop] epoch=0/micro_step=1220/global_step=1220, RunningAvgSamplesPerSec=17.591299866434216, CurrSamplesPerSec=17.783575181860076, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0028, 'learning_rate': 8.400000000000001e-06, 'epoch': 10.01}
[2022-12-18 14:54:19,800] [INFO] [logging.py:68:log_dist] [Rank 0] step=1230, skipped=4, lr=[8.38888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:54:19,802] [INFO] [timer.py:196:stop] epoch=0/micro_step=1230/global_step=1230, RunningAvgSamplesPerSec=17.592045413292322, CurrSamplesPerSec=17.752746037986785, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 14:57:16,096] [INFO] [logging.py:68:log_dist] [Rank 0] step=1240, skipped=4, lr=[8.366666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 14:57:16,098] [INFO] [timer.py:196:stop] epoch=0/micro_step=1240/global_step=1240, RunningAvgSamplesPerSec=17.592574591618792, CurrSamplesPerSec=17.73023145463939, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:00:08,985] [INFO] [logging.py:68:log_dist] [Rank 0] step=1250, skipped=4, lr=[8.344444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:00:08,987] [INFO] [timer.py:196:stop] epoch=0/micro_step=1250/global_step=1250, RunningAvgSamplesPerSec=17.592765978430887, CurrSamplesPerSec=17.581207935360524, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0026, 'learning_rate': 8.344444444444445e-06, 'epoch': 10.02}
[2022-12-18 15:02:58,611] [INFO] [logging.py:68:log_dist] [Rank 0] step=1260, skipped=4, lr=[8.322222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:02:58,612] [INFO] [timer.py:196:stop] epoch=0/micro_step=1260/global_step=1260, RunningAvgSamplesPerSec=17.591578207832033, CurrSamplesPerSec=17.7094327944215, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:05:18,634] [INFO] [logging.py:68:log_dist] [Rank 0] step=1270, skipped=4, lr=[8.3e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:05:18,636] [INFO] [timer.py:196:stop] epoch=0/micro_step=1270/global_step=1270, RunningAvgSamplesPerSec=17.59139456418053, CurrSamplesPerSec=17.86228150125289, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0022, 'learning_rate': 8.288888888888889e-06, 'epoch': 10.02}
[2022-12-18 15:08:24,605] [INFO] [logging.py:68:log_dist] [Rank 0] step=1280, skipped=4, lr=[8.277777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:08:24,607] [INFO] [timer.py:196:stop] epoch=0/micro_step=1280/global_step=1280, RunningAvgSamplesPerSec=17.595103033188895, CurrSamplesPerSec=17.542366899024405, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:11:11,263] [INFO] [logging.py:68:log_dist] [Rank 0] step=1290, skipped=4, lr=[8.255555555555557e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:11:11,265] [INFO] [timer.py:196:stop] epoch=0/micro_step=1290/global_step=1290, RunningAvgSamplesPerSec=17.5945939842215, CurrSamplesPerSec=17.651455059768324, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:13:59,569] [INFO] [logging.py:68:log_dist] [Rank 0] step=1300, skipped=4, lr=[8.233333333333335e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:13:59,571] [INFO] [timer.py:196:stop] epoch=0/micro_step=1300/global_step=1300, RunningAvgSamplesPerSec=17.593519986136247, CurrSamplesPerSec=17.543569555967117, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.002, 'learning_rate': 8.233333333333335e-06, 'epoch': 11.0}
[2022-12-18 15:16:46,768] [INFO] [logging.py:68:log_dist] [Rank 0] step=1310, skipped=4, lr=[8.211111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:16:46,770] [INFO] [timer.py:196:stop] epoch=0/micro_step=1310/global_step=1310, RunningAvgSamplesPerSec=17.593159287186108, CurrSamplesPerSec=17.657518333371485, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:19:35,218] [INFO] [logging.py:68:log_dist] [Rank 0] step=1320, skipped=4, lr=[8.18888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:19:35,220] [INFO] [timer.py:196:stop] epoch=0/micro_step=1320/global_step=1320, RunningAvgSamplesPerSec=17.592123863305886, CurrSamplesPerSec=17.2624771803989, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0019, 'learning_rate': 8.177777777777779e-06, 'epoch': 11.01}
[2022-12-18 15:22:23,739] [INFO] [logging.py:68:log_dist] [Rank 0] step=1330, skipped=4, lr=[8.166666666666668e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:22:23,741] [INFO] [timer.py:196:stop] epoch=0/micro_step=1330/global_step=1330, RunningAvgSamplesPerSec=17.59144315331231, CurrSamplesPerSec=17.624369892176606, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:25:12,977] [INFO] [logging.py:68:log_dist] [Rank 0] step=1340, skipped=4, lr=[8.144444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:25:12,979] [INFO] [timer.py:196:stop] epoch=0/micro_step=1340/global_step=1340, RunningAvgSamplesPerSec=17.591467485388325, CurrSamplesPerSec=17.800980423298352, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:27:59,991] [INFO] [logging.py:68:log_dist] [Rank 0] step=1350, skipped=4, lr=[8.122222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:27:59,993] [INFO] [timer.py:196:stop] epoch=0/micro_step=1350/global_step=1350, RunningAvgSamplesPerSec=17.58933135085906, CurrSamplesPerSec=17.65995201259033, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0015, 'learning_rate': 8.122222222222223e-06, 'epoch': 11.01}
[2022-12-18 15:30:49,368] [INFO] [logging.py:68:log_dist] [Rank 0] step=1360, skipped=4, lr=[8.1e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:30:49,369] [INFO] [timer.py:196:stop] epoch=0/micro_step=1360/global_step=1360, RunningAvgSamplesPerSec=17.589196844316973, CurrSamplesPerSec=17.74898631234037, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:33:39,600] [INFO] [logging.py:68:log_dist] [Rank 0] step=1370, skipped=4, lr=[8.077777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:33:39,602] [INFO] [timer.py:196:stop] epoch=0/micro_step=1370/global_step=1370, RunningAvgSamplesPerSec=17.588471414233023, CurrSamplesPerSec=17.547551309022158, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0013, 'learning_rate': 8.066666666666667e-06, 'epoch': 11.02}
[2022-12-18 15:36:28,193] [INFO] [logging.py:68:log_dist] [Rank 0] step=1380, skipped=4, lr=[8.055555555555557e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:36:28,195] [INFO] [timer.py:196:stop] epoch=0/micro_step=1380/global_step=1380, RunningAvgSamplesPerSec=17.588757839650413, CurrSamplesPerSec=17.470279456311328, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:37:56,456] [INFO] [logging.py:68:log_dist] [Rank 0] step=1390, skipped=4, lr=[8.033333333333335e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:37:56,457] [INFO] [timer.py:196:stop] epoch=0/micro_step=1390/global_step=1390, RunningAvgSamplesPerSec=17.589444637240174, CurrSamplesPerSec=17.796452184981753, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:41:58,134] [INFO] [logging.py:68:log_dist] [Rank 0] step=1400, skipped=4, lr=[8.011111111111113e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:41:58,136] [INFO] [timer.py:196:stop] epoch=0/micro_step=1400/global_step=1400, RunningAvgSamplesPerSec=17.59123118643266, CurrSamplesPerSec=17.351121701395353, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0011, 'learning_rate': 8.011111111111113e-06, 'epoch': 12.0}
[2022-12-18 15:44:46,847] [INFO] [logging.py:68:log_dist] [Rank 0] step=1410, skipped=4, lr=[7.98888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:44:46,849] [INFO] [timer.py:196:stop] epoch=0/micro_step=1410/global_step=1410, RunningAvgSamplesPerSec=17.589641567546884, CurrSamplesPerSec=16.81193579966427, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:47:37,871] [INFO] [logging.py:68:log_dist] [Rank 0] step=1420, skipped=4, lr=[7.966666666666668e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:47:37,873] [INFO] [timer.py:196:stop] epoch=0/micro_step=1420/global_step=1420, RunningAvgSamplesPerSec=17.588067850747443, CurrSamplesPerSec=17.428626152497475, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0012, 'learning_rate': 7.955555555555557e-06, 'epoch': 12.01}
[2022-12-18 15:50:26,270] [INFO] [logging.py:68:log_dist] [Rank 0] step=1430, skipped=4, lr=[7.944444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:50:26,271] [INFO] [timer.py:196:stop] epoch=0/micro_step=1430/global_step=1430, RunningAvgSamplesPerSec=17.58741915847445, CurrSamplesPerSec=17.76162991584455, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:53:13,708] [INFO] [logging.py:68:log_dist] [Rank 0] step=1440, skipped=4, lr=[7.922222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:53:13,710] [INFO] [timer.py:196:stop] epoch=0/micro_step=1440/global_step=1440, RunningAvgSamplesPerSec=17.585312149367923, CurrSamplesPerSec=17.475912824942153, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 15:56:04,224] [INFO] [logging.py:68:log_dist] [Rank 0] step=1450, skipped=4, lr=[7.9e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:56:04,226] [INFO] [timer.py:196:stop] epoch=0/micro_step=1450/global_step=1450, RunningAvgSamplesPerSec=17.584736964932496, CurrSamplesPerSec=17.084766353718567, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0011, 'learning_rate': 7.9e-06, 'epoch': 12.01}
[2022-12-18 15:58:51,836] [INFO] [logging.py:68:log_dist] [Rank 0] step=1460, skipped=4, lr=[7.877777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 15:58:51,838] [INFO] [timer.py:196:stop] epoch=0/micro_step=1460/global_step=1460, RunningAvgSamplesPerSec=17.58396256582276, CurrSamplesPerSec=17.408034199285936, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:01:39,733] [INFO] [logging.py:68:log_dist] [Rank 0] step=1470, skipped=4, lr=[7.855555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:01:39,735] [INFO] [timer.py:196:stop] epoch=0/micro_step=1470/global_step=1470, RunningAvgSamplesPerSec=17.584820238828023, CurrSamplesPerSec=17.65451172025162, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0012, 'learning_rate': 7.844444444444446e-06, 'epoch': 12.02}
[2022-12-18 16:04:29,437] [INFO] [logging.py:68:log_dist] [Rank 0] step=1480, skipped=4, lr=[7.833333333333333e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:04:29,438] [INFO] [timer.py:196:stop] epoch=0/micro_step=1480/global_step=1480, RunningAvgSamplesPerSec=17.58523223228052, CurrSamplesPerSec=17.692160554263566, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:07:19,577] [INFO] [logging.py:68:log_dist] [Rank 0] step=1490, skipped=4, lr=[7.811111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:07:19,579] [INFO] [timer.py:196:stop] epoch=0/micro_step=1490/global_step=1490, RunningAvgSamplesPerSec=17.585349491276787, CurrSamplesPerSec=17.528617434648247, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:10:04,717] [INFO] [logging.py:68:log_dist] [Rank 0] step=1500, skipped=4, lr=[7.788888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:10:04,720] [INFO] [timer.py:196:stop] epoch=0/micro_step=1500/global_step=1500, RunningAvgSamplesPerSec=17.58553346224858, CurrSamplesPerSec=17.30400841724948, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.001, 'learning_rate': 7.788888888888889e-06, 'epoch': 12.02}
[2022-12-18 16:12:51,306] [INFO] [logging.py:68:log_dist] [Rank 0] step=1510, skipped=4, lr=[7.766666666666666e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:12:51,307] [INFO] [timer.py:196:stop] epoch=0/micro_step=1510/global_step=1510, RunningAvgSamplesPerSec=17.58739109597355, CurrSamplesPerSec=17.62212763286849, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:15:39,256] [INFO] [logging.py:68:log_dist] [Rank 0] step=1520, skipped=4, lr=[7.744444444444446e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:15:39,258] [INFO] [timer.py:196:stop] epoch=0/micro_step=1520/global_step=1520, RunningAvgSamplesPerSec=17.58624163007446, CurrSamplesPerSec=17.223591201370315, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0009, 'learning_rate': 7.733333333333334e-06, 'epoch': 13.0}
[2022-12-18 16:18:30,572] [INFO] [logging.py:68:log_dist] [Rank 0] step=1530, skipped=4, lr=[7.722222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:18:30,573] [INFO] [timer.py:196:stop] epoch=0/micro_step=1530/global_step=1530, RunningAvgSamplesPerSec=17.584710397614575, CurrSamplesPerSec=17.58900462856031, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:21:29,403] [INFO] [logging.py:68:log_dist] [Rank 0] step=1540, skipped=4, lr=[7.7e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:21:29,405] [INFO] [timer.py:196:stop] epoch=0/micro_step=1540/global_step=1540, RunningAvgSamplesPerSec=17.584223073868095, CurrSamplesPerSec=17.580700146010567, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:24:23,678] [INFO] [logging.py:68:log_dist] [Rank 0] step=1550, skipped=4, lr=[7.677777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:24:23,680] [INFO] [timer.py:196:stop] epoch=0/micro_step=1550/global_step=1550, RunningAvgSamplesPerSec=17.58439155673778, CurrSamplesPerSec=17.668182700157885, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0009, 'learning_rate': 7.677777777777778e-06, 'epoch': 13.01}
[2022-12-18 16:27:18,302] [INFO] [logging.py:68:log_dist] [Rank 0] step=1560, skipped=4, lr=[7.655555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:27:18,304] [INFO] [timer.py:196:stop] epoch=0/micro_step=1560/global_step=1560, RunningAvgSamplesPerSec=17.584671210437033, CurrSamplesPerSec=17.72357047648451, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:30:12,136] [INFO] [logging.py:68:log_dist] [Rank 0] step=1570, skipped=4, lr=[7.633333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:30:12,138] [INFO] [timer.py:196:stop] epoch=0/micro_step=1570/global_step=1570, RunningAvgSamplesPerSec=17.584434792027235, CurrSamplesPerSec=17.67633958453577, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0012, 'learning_rate': 7.622222222222223e-06, 'epoch': 13.01}
[2022-12-18 16:33:09,808] [INFO] [logging.py:68:log_dist] [Rank 0] step=1580, skipped=4, lr=[7.611111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:33:09,810] [INFO] [timer.py:196:stop] epoch=0/micro_step=1580/global_step=1580, RunningAvgSamplesPerSec=17.584395220473127, CurrSamplesPerSec=17.75029258656175, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:36:05,926] [INFO] [logging.py:68:log_dist] [Rank 0] step=1590, skipped=4, lr=[7.588888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:36:05,928] [INFO] [timer.py:196:stop] epoch=0/micro_step=1590/global_step=1590, RunningAvgSamplesPerSec=17.58474933416415, CurrSamplesPerSec=17.61421024415819, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:39:02,292] [INFO] [logging.py:68:log_dist] [Rank 0] step=1600, skipped=4, lr=[7.566666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:39:02,294] [INFO] [timer.py:196:stop] epoch=0/micro_step=1600/global_step=1600, RunningAvgSamplesPerSec=17.584912959793066, CurrSamplesPerSec=17.653429636601118, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0009, 'learning_rate': 7.566666666666667e-06, 'epoch': 13.02}
[2022-12-18 16:42:00,495] [INFO] [logging.py:68:log_dist] [Rank 0] step=1610, skipped=4, lr=[7.544444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:42:00,497] [INFO] [timer.py:196:stop] epoch=0/micro_step=1610/global_step=1610, RunningAvgSamplesPerSec=17.58463914064396, CurrSamplesPerSec=17.55525847226904, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:44:04,575] [INFO] [logging.py:68:log_dist] [Rank 0] step=1620, skipped=4, lr=[7.5222222222222226e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:44:04,578] [INFO] [timer.py:196:stop] epoch=0/micro_step=1620/global_step=1620, RunningAvgSamplesPerSec=17.585002091371685, CurrSamplesPerSec=17.664836496154194, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0008, 'learning_rate': 7.511111111111111e-06, 'epoch': 14.0}
[2022-12-18 16:47:55,627] [INFO] [logging.py:68:log_dist] [Rank 0] step=1630, skipped=4, lr=[7.500000000000001e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:47:55,629] [INFO] [timer.py:196:stop] epoch=0/micro_step=1630/global_step=1630, RunningAvgSamplesPerSec=17.587178112703374, CurrSamplesPerSec=17.63657926828507, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:50:51,334] [INFO] [logging.py:68:log_dist] [Rank 0] step=1640, skipped=4, lr=[7.477777777777779e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:50:51,337] [INFO] [timer.py:196:stop] epoch=0/micro_step=1640/global_step=1640, RunningAvgSamplesPerSec=17.58712243334102, CurrSamplesPerSec=17.347132175313174, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:53:53,203] [INFO] [logging.py:68:log_dist] [Rank 0] step=1650, skipped=4, lr=[7.455555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:53:53,205] [INFO] [timer.py:196:stop] epoch=0/micro_step=1650/global_step=1650, RunningAvgSamplesPerSec=17.587439830952103, CurrSamplesPerSec=17.622363633296064, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0008, 'learning_rate': 7.455555555555556e-06, 'epoch': 14.01}
[2022-12-18 16:56:55,071] [INFO] [logging.py:68:log_dist] [Rank 0] step=1660, skipped=4, lr=[7.433333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:56:55,073] [INFO] [timer.py:196:stop] epoch=0/micro_step=1660/global_step=1660, RunningAvgSamplesPerSec=17.58774677231902, CurrSamplesPerSec=17.723122298678486, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 16:59:53,412] [INFO] [logging.py:68:log_dist] [Rank 0] step=1670, skipped=4, lr=[7.411111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 16:59:53,415] [INFO] [timer.py:196:stop] epoch=0/micro_step=1670/global_step=1670, RunningAvgSamplesPerSec=17.586889141711087, CurrSamplesPerSec=17.5252632408432, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0008, 'learning_rate': 7.4e-06, 'epoch': 14.01}
[2022-12-18 17:02:55,081] [INFO] [logging.py:68:log_dist] [Rank 0] step=1680, skipped=4, lr=[7.38888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:02:55,082] [INFO] [timer.py:196:stop] epoch=0/micro_step=1680/global_step=1680, RunningAvgSamplesPerSec=17.58688935579313, CurrSamplesPerSec=17.783240595339233, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:05:57,804] [INFO] [logging.py:68:log_dist] [Rank 0] step=1690, skipped=4, lr=[7.3666666666666676e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:05:57,806] [INFO] [timer.py:196:stop] epoch=0/micro_step=1690/global_step=1690, RunningAvgSamplesPerSec=17.587471464153204, CurrSamplesPerSec=17.783433805737804, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:08:55,981] [INFO] [logging.py:68:log_dist] [Rank 0] step=1700, skipped=4, lr=[7.344444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:08:55,982] [INFO] [timer.py:196:stop] epoch=0/micro_step=1700/global_step=1700, RunningAvgSamplesPerSec=17.587408570215107, CurrSamplesPerSec=17.451048991964477, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0008, 'learning_rate': 7.344444444444445e-06, 'epoch': 14.02}
[2022-12-18 17:11:56,277] [INFO] [logging.py:68:log_dist] [Rank 0] step=1710, skipped=4, lr=[7.322222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:11:56,280] [INFO] [timer.py:196:stop] epoch=0/micro_step=1710/global_step=1710, RunningAvgSamplesPerSec=17.5876806833596, CurrSamplesPerSec=17.729723217717872, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:14:57,436] [INFO] [logging.py:68:log_dist] [Rank 0] step=1720, skipped=4, lr=[7.3e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:14:57,438] [INFO] [timer.py:196:stop] epoch=0/micro_step=1720/global_step=1720, RunningAvgSamplesPerSec=17.5872673472387, CurrSamplesPerSec=17.7992642130134, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0007, 'learning_rate': 7.28888888888889e-06, 'epoch': 14.02}
[2022-12-18 17:17:53,606] [INFO] [logging.py:68:log_dist] [Rank 0] step=1730, skipped=4, lr=[7.277777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:17:53,608] [INFO] [timer.py:196:stop] epoch=0/micro_step=1730/global_step=1730, RunningAvgSamplesPerSec=17.586756975707022, CurrSamplesPerSec=17.563738888918333, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:18:56,785] [INFO] [logging.py:68:log_dist] [Rank 0] step=1740, skipped=4, lr=[7.255555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:18:56,787] [INFO] [timer.py:196:stop] epoch=0/micro_step=1740/global_step=1740, RunningAvgSamplesPerSec=17.58932825421085, CurrSamplesPerSec=23.433776461872853, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:23:37,623] [INFO] [logging.py:68:log_dist] [Rank 0] step=1750, skipped=4, lr=[7.233333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:23:37,624] [INFO] [timer.py:196:stop] epoch=0/micro_step=1750/global_step=1750, RunningAvgSamplesPerSec=17.58929052224577, CurrSamplesPerSec=17.722104328870888, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0007, 'learning_rate': 7.233333333333334e-06, 'epoch': 15.0}
[2022-12-18 17:26:29,765] [INFO] [logging.py:68:log_dist] [Rank 0] step=1760, skipped=4, lr=[7.211111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:26:29,767] [INFO] [timer.py:196:stop] epoch=0/micro_step=1760/global_step=1760, RunningAvgSamplesPerSec=17.589265190884376, CurrSamplesPerSec=17.29575795734955, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:29:23,777] [INFO] [logging.py:68:log_dist] [Rank 0] step=1770, skipped=4, lr=[7.188888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:29:23,778] [INFO] [timer.py:196:stop] epoch=0/micro_step=1770/global_step=1770, RunningAvgSamplesPerSec=17.589199628233683, CurrSamplesPerSec=17.432689482722747, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0007, 'learning_rate': 7.177777777777778e-06, 'epoch': 15.01}
[2022-12-18 17:32:22,814] [INFO] [logging.py:68:log_dist] [Rank 0] step=1780, skipped=4, lr=[7.166666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:32:22,815] [INFO] [timer.py:196:stop] epoch=0/micro_step=1780/global_step=1780, RunningAvgSamplesPerSec=17.589156096057113, CurrSamplesPerSec=17.679121929983722, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:35:21,130] [INFO] [logging.py:68:log_dist] [Rank 0] step=1790, skipped=4, lr=[7.1444444444444446e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:35:21,132] [INFO] [timer.py:196:stop] epoch=0/micro_step=1790/global_step=1790, RunningAvgSamplesPerSec=17.589510541041726, CurrSamplesPerSec=17.75061537037282, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:38:22,077] [INFO] [logging.py:68:log_dist] [Rank 0] step=1800, skipped=4, lr=[7.122222222222222e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:38:22,079] [INFO] [timer.py:196:stop] epoch=0/micro_step=1800/global_step=1800, RunningAvgSamplesPerSec=17.589853715416716, CurrSamplesPerSec=17.70494516775891, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0007, 'learning_rate': 7.122222222222222e-06, 'epoch': 15.01}
[2022-12-18 17:41:22,983] [INFO] [logging.py:68:log_dist] [Rank 0] step=1810, skipped=4, lr=[7.100000000000001e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:41:22,985] [INFO] [timer.py:196:stop] epoch=0/micro_step=1810/global_step=1810, RunningAvgSamplesPerSec=17.589705689066374, CurrSamplesPerSec=17.502633426748798, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:44:18,958] [INFO] [logging.py:68:log_dist] [Rank 0] step=1820, skipped=4, lr=[7.077777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:44:18,960] [INFO] [timer.py:196:stop] epoch=0/micro_step=1820/global_step=1820, RunningAvgSamplesPerSec=17.58889554956267, CurrSamplesPerSec=17.227935399411322, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0007, 'learning_rate': 7.066666666666667e-06, 'epoch': 15.02}
[2022-12-18 17:47:21,773] [INFO] [logging.py:68:log_dist] [Rank 0] step=1830, skipped=4, lr=[7.055555555555557e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:47:21,775] [INFO] [timer.py:196:stop] epoch=0/micro_step=1830/global_step=1830, RunningAvgSamplesPerSec=17.588036262189785, CurrSamplesPerSec=17.291913033681215, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:50:27,447] [INFO] [logging.py:68:log_dist] [Rank 0] step=1840, skipped=4, lr=[7.033333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:50:27,448] [INFO] [timer.py:196:stop] epoch=0/micro_step=1840/global_step=1840, RunningAvgSamplesPerSec=17.587960597286486, CurrSamplesPerSec=17.724241030252486, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:52:52,644] [INFO] [logging.py:68:log_dist] [Rank 0] step=1850, skipped=4, lr=[7.011111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:52:52,646] [INFO] [timer.py:196:stop] epoch=0/micro_step=1850/global_step=1850, RunningAvgSamplesPerSec=17.587975214747647, CurrSamplesPerSec=17.64940665653735, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0006, 'learning_rate': 7.011111111111112e-06, 'epoch': 15.02}
[2022-12-18 17:56:08,287] [INFO] [logging.py:68:log_dist] [Rank 0] step=1860, skipped=4, lr=[6.9888888888888895e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:56:08,288] [INFO] [timer.py:196:stop] epoch=0/micro_step=1860/global_step=1860, RunningAvgSamplesPerSec=17.589801958416484, CurrSamplesPerSec=17.619758723894087, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 17:59:00,804] [INFO] [logging.py:68:log_dist] [Rank 0] step=1870, skipped=4, lr=[6.966666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 17:59:00,805] [INFO] [timer.py:196:stop] epoch=0/micro_step=1870/global_step=1870, RunningAvgSamplesPerSec=17.59022362418036, CurrSamplesPerSec=17.588560925502627, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0006, 'learning_rate': 6.955555555555557e-06, 'epoch': 16.0}
[2022-12-18 18:01:58,641] [INFO] [logging.py:68:log_dist] [Rank 0] step=1880, skipped=4, lr=[6.944444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:01:58,643] [INFO] [timer.py:196:stop] epoch=0/micro_step=1880/global_step=1880, RunningAvgSamplesPerSec=17.59061098794626, CurrSamplesPerSec=17.788324368998808, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 18:04:58,587] [INFO] [logging.py:68:log_dist] [Rank 0] step=1890, skipped=4, lr=[6.922222222222222e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:04:58,588] [INFO] [timer.py:196:stop] epoch=0/micro_step=1890/global_step=1890, RunningAvgSamplesPerSec=17.5910300704761, CurrSamplesPerSec=17.873249456898765, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 18:07:53,171] [INFO] [logging.py:68:log_dist] [Rank 0] step=1900, skipped=4, lr=[6.9e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:07:53,173] [INFO] [timer.py:196:stop] epoch=0/micro_step=1900/global_step=1900, RunningAvgSamplesPerSec=17.590747818093448, CurrSamplesPerSec=17.480120031668438, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0006, 'learning_rate': 6.9e-06, 'epoch': 16.01}
[2022-12-18 18:10:47,689] [INFO] [logging.py:68:log_dist] [Rank 0] step=1910, skipped=4, lr=[6.8777777777777785e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:10:47,691] [INFO] [timer.py:196:stop] epoch=0/micro_step=1910/global_step=1910, RunningAvgSamplesPerSec=17.59104725904212, CurrSamplesPerSec=17.710297408588108, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 18:13:42,983] [INFO] [logging.py:68:log_dist] [Rank 0] step=1920, skipped=4, lr=[6.855555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:13:42,985] [INFO] [timer.py:196:stop] epoch=0/micro_step=1920/global_step=1920, RunningAvgSamplesPerSec=17.590833049552817, CurrSamplesPerSec=17.4918464812311, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0006, 'learning_rate': 6.844444444444445e-06, 'epoch': 16.01}
[2022-12-18 18:16:38,362] [INFO] [logging.py:68:log_dist] [Rank 0] step=1930, skipped=4, lr=[6.833333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:16:38,364] [INFO] [timer.py:196:stop] epoch=0/micro_step=1930/global_step=1930, RunningAvgSamplesPerSec=17.591234933505756, CurrSamplesPerSec=17.439830364828275, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 18:19:28,894] [INFO] [logging.py:68:log_dist] [Rank 0] step=1940, skipped=4, lr=[6.811111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:19:28,896] [INFO] [timer.py:196:stop] epoch=0/micro_step=1940/global_step=1940, RunningAvgSamplesPerSec=17.590774338771933, CurrSamplesPerSec=17.599662529740584, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 18:22:22,602] [INFO] [logging.py:68:log_dist] [Rank 0] step=1950, skipped=4, lr=[6.788888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:22:22,604] [INFO] [timer.py:196:stop] epoch=0/micro_step=1950/global_step=1950, RunningAvgSamplesPerSec=17.589976661895054, CurrSamplesPerSec=17.679976600177262, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0006, 'learning_rate': 6.788888888888889e-06, 'epoch': 16.02}
[2022-12-18 18:25:14,881] [INFO] [logging.py:68:log_dist] [Rank 0] step=1960, skipped=4, lr=[6.7666666666666665e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:25:14,882] [INFO] [timer.py:196:stop] epoch=0/micro_step=1960/global_step=1960, RunningAvgSamplesPerSec=17.59061118827806, CurrSamplesPerSec=17.789994847282397, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 18:26:44,140] [INFO] [logging.py:68:log_dist] [Rank 0] step=1970, skipped=4, lr=[6.744444444444444e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:26:44,142] [INFO] [timer.py:196:stop] epoch=0/micro_step=1970/global_step=1970, RunningAvgSamplesPerSec=17.590670368534774, CurrSamplesPerSec=17.667350100092985, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0005, 'learning_rate': 6.733333333333334e-06, 'epoch': 17.0}
[2022-12-18 18:30:57,744] [INFO] [logging.py:68:log_dist] [Rank 0] step=1980, skipped=4, lr=[6.7222222222222235e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:30:57,745] [INFO] [timer.py:196:stop] epoch=0/micro_step=1980/global_step=1980, RunningAvgSamplesPerSec=17.591780103374816, CurrSamplesPerSec=17.696074726111295, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 18:33:47,038] [INFO] [logging.py:68:log_dist] [Rank 0] step=1990, skipped=4, lr=[6.700000000000001e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:33:47,039] [INFO] [timer.py:196:stop] epoch=0/micro_step=1990/global_step=1990, RunningAvgSamplesPerSec=17.591581591137096, CurrSamplesPerSec=17.78166561596026, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 18:36:39,443] [INFO] [logging.py:68:log_dist] [Rank 0] step=2000, skipped=4, lr=[6.677777777777779e-06], mom=[[0.9, 0.999]]
[2022-12-18 18:36:39,444] [INFO] [timer.py:196:stop] epoch=0/micro_step=2000/global_step=2000, RunningAvgSamplesPerSec=17.589825563330376, CurrSamplesPerSec=16.981787098279423, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0006, 'learning_rate': 6.677777777777779e-06, 'epoch': 17.01}
{'eval_loss': 0.30908203125, 'eval_wer': 17.69160002830656, 'eval_runtime': 1241.2535, 'eval_samples_per_second': 3.109, 'eval_steps_per_second': 0.097, 'epoch': 17.01}
[2022-12-18 18:57:21,750] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step2000 is begin to save!
[2022-12-18 18:57:21,758] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: ./checkpoint-2000/global_step2000/mp_rank_00_model_states.pt
[2022-12-18 18:57:21,758] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving ./checkpoint-2000/global_step2000/mp_rank_00_model_states.pt...
[2022-12-18 18:57:22,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved ./checkpoint-2000/global_step2000/mp_rank_00_model_states.pt.
[2022-12-18 18:57:22,823] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving ./checkpoint-2000/global_step2000/zero_pp_rank_0_mp_rank_00_optim_states.pt...
[2022-12-18 18:57:27,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved ./checkpoint-2000/global_step2000/zero_pp_rank_0_mp_rank_00_optim_states.pt.
[2022-12-18 18:57:27,460] [INFO] [engine.py:3394:_save_zero_checkpoint] zero checkpoint saved ./checkpoint-2000/global_step2000/zero_pp_rank_0_mp_rank_00_optim_states.pt
[2022-12-18 18:57:27,460] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now!
[2022-12-18 19:01:20,583] [INFO] [logging.py:68:log_dist] [Rank 0] step=2010, skipped=4, lr=[6.655555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:01:20,585] [INFO] [timer.py:196:stop] epoch=0/micro_step=2010/global_step=2010, RunningAvgSamplesPerSec=17.589107959120653, CurrSamplesPerSec=17.13468078137867, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:04:14,471] [INFO] [logging.py:68:log_dist] [Rank 0] step=2020, skipped=4, lr=[6.633333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:04:14,473] [INFO] [timer.py:196:stop] epoch=0/micro_step=2020/global_step=2020, RunningAvgSamplesPerSec=17.58855576233867, CurrSamplesPerSec=17.829921298335687, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0006, 'learning_rate': 6.6222222222222236e-06, 'epoch': 17.01}
[2022-12-18 19:07:02,768] [INFO] [logging.py:68:log_dist] [Rank 0] step=2030, skipped=4, lr=[6.6111111111111115e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:07:02,770] [INFO] [timer.py:196:stop] epoch=0/micro_step=2030/global_step=2030, RunningAvgSamplesPerSec=17.587637001272515, CurrSamplesPerSec=17.38889651837875, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:09:49,384] [INFO] [logging.py:68:log_dist] [Rank 0] step=2040, skipped=4, lr=[6.588888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:09:49,386] [INFO] [timer.py:196:stop] epoch=0/micro_step=2040/global_step=2040, RunningAvgSamplesPerSec=17.58716148344212, CurrSamplesPerSec=17.66067353280504, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:12:41,032] [INFO] [logging.py:68:log_dist] [Rank 0] step=2050, skipped=4, lr=[6.566666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:12:41,034] [INFO] [timer.py:196:stop] epoch=0/micro_step=2050/global_step=2050, RunningAvgSamplesPerSec=17.587390323827243, CurrSamplesPerSec=17.532843188322214, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0005, 'learning_rate': 6.566666666666667e-06, 'epoch': 17.02}
[2022-12-18 19:15:25,861] [INFO] [logging.py:68:log_dist] [Rank 0] step=2060, skipped=4, lr=[6.544444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:15:25,862] [INFO] [timer.py:196:stop] epoch=0/micro_step=2060/global_step=2060, RunningAvgSamplesPerSec=17.58717548663508, CurrSamplesPerSec=17.512568762330247, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:18:16,412] [INFO] [logging.py:68:log_dist] [Rank 0] step=2070, skipped=4, lr=[6.522222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:18:16,414] [INFO] [timer.py:196:stop] epoch=0/micro_step=2070/global_step=2070, RunningAvgSamplesPerSec=17.587141379364912, CurrSamplesPerSec=17.701924717934272, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0005, 'learning_rate': 6.511111111111112e-06, 'epoch': 17.02}
[2022-12-18 19:21:04,018] [INFO] [logging.py:68:log_dist] [Rank 0] step=2080, skipped=4, lr=[6.5000000000000004e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:21:04,020] [INFO] [timer.py:196:stop] epoch=0/micro_step=2080/global_step=2080, RunningAvgSamplesPerSec=17.58672549676673, CurrSamplesPerSec=17.32818066644779, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:23:49,078] [INFO] [logging.py:68:log_dist] [Rank 0] step=2090, skipped=4, lr=[6.477777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:23:49,079] [INFO] [timer.py:196:stop] epoch=0/micro_step=2090/global_step=2090, RunningAvgSamplesPerSec=17.58885423352245, CurrSamplesPerSec=17.887499821748353, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:26:33,814] [INFO] [logging.py:68:log_dist] [Rank 0] step=2100, skipped=4, lr=[6.455555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:26:33,816] [INFO] [timer.py:196:stop] epoch=0/micro_step=2100/global_step=2100, RunningAvgSamplesPerSec=17.5885968178544, CurrSamplesPerSec=17.73180553447264, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0005, 'learning_rate': 6.455555555555556e-06, 'epoch': 18.0}
[2022-12-18 19:29:21,465] [INFO] [logging.py:68:log_dist] [Rank 0] step=2110, skipped=4, lr=[6.433333333333333e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:29:21,467] [INFO] [timer.py:196:stop] epoch=0/micro_step=2110/global_step=2110, RunningAvgSamplesPerSec=17.588656612423332, CurrSamplesPerSec=17.617084056780946, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:32:07,991] [INFO] [logging.py:68:log_dist] [Rank 0] step=2120, skipped=4, lr=[6.411111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:32:07,992] [INFO] [timer.py:196:stop] epoch=0/micro_step=2120/global_step=2120, RunningAvgSamplesPerSec=17.5878081283693, CurrSamplesPerSec=17.254413772911384, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0005, 'learning_rate': 6.4000000000000006e-06, 'epoch': 18.01}
[2022-12-18 19:34:55,224] [INFO] [logging.py:68:log_dist] [Rank 0] step=2130, skipped=4, lr=[6.3888888888888885e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:34:55,226] [INFO] [timer.py:196:stop] epoch=0/micro_step=2130/global_step=2130, RunningAvgSamplesPerSec=17.587863612597186, CurrSamplesPerSec=17.59388799325281, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:37:42,843] [INFO] [logging.py:68:log_dist] [Rank 0] step=2140, skipped=4, lr=[6.366666666666668e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:37:42,845] [INFO] [timer.py:196:stop] epoch=0/micro_step=2140/global_step=2140, RunningAvgSamplesPerSec=17.58793523889989, CurrSamplesPerSec=17.737803418278745, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:40:33,196] [INFO] [logging.py:68:log_dist] [Rank 0] step=2150, skipped=4, lr=[6.3444444444444454e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:40:33,197] [INFO] [timer.py:196:stop] epoch=0/micro_step=2150/global_step=2150, RunningAvgSamplesPerSec=17.587421562465057, CurrSamplesPerSec=17.55252186092314, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0005, 'learning_rate': 6.3444444444444454e-06, 'epoch': 18.01}
[2022-12-18 19:43:19,596] [INFO] [logging.py:68:log_dist] [Rank 0] step=2160, skipped=4, lr=[6.322222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:43:19,597] [INFO] [timer.py:196:stop] epoch=0/micro_step=2160/global_step=2160, RunningAvgSamplesPerSec=17.586757230140798, CurrSamplesPerSec=16.966508150211634, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:46:07,555] [INFO] [logging.py:68:log_dist] [Rank 0] step=2170, skipped=4, lr=[6.300000000000001e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:46:07,556] [INFO] [timer.py:196:stop] epoch=0/micro_step=2170/global_step=2170, RunningAvgSamplesPerSec=17.58629169678668, CurrSamplesPerSec=17.659162009765595, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0005, 'learning_rate': 6.28888888888889e-06, 'epoch': 18.02}
[2022-12-18 19:48:58,381] [INFO] [logging.py:68:log_dist] [Rank 0] step=2180, skipped=4, lr=[6.277777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:48:58,383] [INFO] [timer.py:196:stop] epoch=0/micro_step=2180/global_step=2180, RunningAvgSamplesPerSec=17.58560190399963, CurrSamplesPerSec=17.75335539731907, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:51:46,973] [INFO] [logging.py:68:log_dist] [Rank 0] step=2190, skipped=4, lr=[6.255555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:51:46,974] [INFO] [timer.py:196:stop] epoch=0/micro_step=2190/global_step=2190, RunningAvgSamplesPerSec=17.585425818924406, CurrSamplesPerSec=17.477545622479145, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:53:39,660] [INFO] [logging.py:68:log_dist] [Rank 0] step=2200, skipped=4, lr=[6.2333333333333335e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:53:39,661] [INFO] [timer.py:196:stop] epoch=0/micro_step=2200/global_step=2200, RunningAvgSamplesPerSec=17.584854508426705, CurrSamplesPerSec=17.806048361446347, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0005, 'learning_rate': 6.2333333333333335e-06, 'epoch': 18.02}
[2022-12-18 19:57:14,000] [INFO] [logging.py:68:log_dist] [Rank 0] step=2210, skipped=4, lr=[6.211111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:57:14,001] [INFO] [timer.py:196:stop] epoch=0/micro_step=2210/global_step=2210, RunningAvgSamplesPerSec=17.58662087490927, CurrSamplesPerSec=17.174206367282927, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 19:59:59,224] [INFO] [logging.py:68:log_dist] [Rank 0] step=2220, skipped=4, lr=[6.18888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 19:59:59,226] [INFO] [timer.py:196:stop] epoch=0/micro_step=2220/global_step=2220, RunningAvgSamplesPerSec=17.586615200728613, CurrSamplesPerSec=17.67247369509411, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0005, 'learning_rate': 6.177777777777778e-06, 'epoch': 19.0}
[2022-12-18 20:02:48,766] [INFO] [logging.py:68:log_dist] [Rank 0] step=2230, skipped=4, lr=[6.166666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:02:48,768] [INFO] [timer.py:196:stop] epoch=0/micro_step=2230/global_step=2230, RunningAvgSamplesPerSec=17.586462619583337, CurrSamplesPerSec=17.578105236247605, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:05:35,691] [INFO] [logging.py:68:log_dist] [Rank 0] step=2240, skipped=4, lr=[6.144444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:05:35,692] [INFO] [timer.py:196:stop] epoch=0/micro_step=2240/global_step=2240, RunningAvgSamplesPerSec=17.585460173061477, CurrSamplesPerSec=17.385654131413393, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:08:21,835] [INFO] [logging.py:68:log_dist] [Rank 0] step=2250, skipped=4, lr=[6.1222222222222224e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:08:21,836] [INFO] [timer.py:196:stop] epoch=0/micro_step=2250/global_step=2250, RunningAvgSamplesPerSec=17.584986379905548, CurrSamplesPerSec=17.7243685933451, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0005, 'learning_rate': 6.1222222222222224e-06, 'epoch': 19.01}
[2022-12-18 20:11:07,430] [INFO] [logging.py:68:log_dist] [Rank 0] step=2260, skipped=4, lr=[6.1e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:11:07,431] [INFO] [timer.py:196:stop] epoch=0/micro_step=2260/global_step=2260, RunningAvgSamplesPerSec=17.584351972746852, CurrSamplesPerSec=17.710278713334784, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:13:55,437] [INFO] [logging.py:68:log_dist] [Rank 0] step=2270, skipped=4, lr=[6.077777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:13:55,439] [INFO] [timer.py:196:stop] epoch=0/micro_step=2270/global_step=2270, RunningAvgSamplesPerSec=17.583491407484367, CurrSamplesPerSec=17.62198765520925, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0005, 'learning_rate': 6.066666666666667e-06, 'epoch': 19.01}
[2022-12-18 20:16:41,129] [INFO] [logging.py:68:log_dist] [Rank 0] step=2280, skipped=4, lr=[6.055555555555555e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:16:41,131] [INFO] [timer.py:196:stop] epoch=0/micro_step=2280/global_step=2280, RunningAvgSamplesPerSec=17.583247003955186, CurrSamplesPerSec=16.80276775889636, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:19:29,886] [INFO] [logging.py:68:log_dist] [Rank 0] step=2290, skipped=4, lr=[6.033333333333335e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:19:29,888] [INFO] [timer.py:196:stop] epoch=0/micro_step=2290/global_step=2290, RunningAvgSamplesPerSec=17.582735062830277, CurrSamplesPerSec=17.198687024215424, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:22:16,948] [INFO] [logging.py:68:log_dist] [Rank 0] step=2300, skipped=4, lr=[6.011111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:22:16,950] [INFO] [timer.py:196:stop] epoch=0/micro_step=2300/global_step=2300, RunningAvgSamplesPerSec=17.5824598544171, CurrSamplesPerSec=17.64900167446127, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 6.011111111111112e-06, 'epoch': 19.02}
[2022-12-18 20:25:02,174] [INFO] [logging.py:68:log_dist] [Rank 0] step=2310, skipped=4, lr=[5.98888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:25:02,176] [INFO] [timer.py:196:stop] epoch=0/micro_step=2310/global_step=2310, RunningAvgSamplesPerSec=17.582553383094712, CurrSamplesPerSec=17.58853211438977, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:26:02,835] [INFO] [logging.py:68:log_dist] [Rank 0] step=2320, skipped=4, lr=[5.966666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:26:02,837] [INFO] [timer.py:196:stop] epoch=0/micro_step=2320/global_step=2320, RunningAvgSamplesPerSec=17.584496499885955, CurrSamplesPerSec=23.15811660693858, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.955555555555555e-06, 'epoch': 20.0}
[2022-12-18 20:30:31,400] [INFO] [logging.py:68:log_dist] [Rank 0] step=2330, skipped=4, lr=[5.944444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:30:31,401] [INFO] [timer.py:196:stop] epoch=0/micro_step=2330/global_step=2330, RunningAvgSamplesPerSec=17.584324916352653, CurrSamplesPerSec=17.849754179376177, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:33:19,566] [INFO] [logging.py:68:log_dist] [Rank 0] step=2340, skipped=4, lr=[5.922222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:33:19,568] [INFO] [timer.py:196:stop] epoch=0/micro_step=2340/global_step=2340, RunningAvgSamplesPerSec=17.584191831302526, CurrSamplesPerSec=17.293479318951103, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:36:08,224] [INFO] [logging.py:68:log_dist] [Rank 0] step=2350, skipped=4, lr=[5.9e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:36:08,226] [INFO] [timer.py:196:stop] epoch=0/micro_step=2350/global_step=2350, RunningAvgSamplesPerSec=17.5823554817332, CurrSamplesPerSec=17.549198668476794, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.9e-06, 'epoch': 20.01}
[2022-12-18 20:38:58,745] [INFO] [logging.py:68:log_dist] [Rank 0] step=2360, skipped=4, lr=[5.877777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:38:58,746] [INFO] [timer.py:196:stop] epoch=0/micro_step=2360/global_step=2360, RunningAvgSamplesPerSec=17.58123845801545, CurrSamplesPerSec=17.46357146881375, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:41:44,829] [INFO] [logging.py:68:log_dist] [Rank 0] step=2370, skipped=4, lr=[5.855555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:41:44,831] [INFO] [timer.py:196:stop] epoch=0/micro_step=2370/global_step=2370, RunningAvgSamplesPerSec=17.580745278930927, CurrSamplesPerSec=17.309718103592974, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.844444444444445e-06, 'epoch': 20.01}
[2022-12-18 20:44:30,833] [INFO] [logging.py:68:log_dist] [Rank 0] step=2380, skipped=4, lr=[5.833333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:44:30,835] [INFO] [timer.py:196:stop] epoch=0/micro_step=2380/global_step=2380, RunningAvgSamplesPerSec=17.580249296661712, CurrSamplesPerSec=16.98673678148918, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:47:19,173] [INFO] [logging.py:68:log_dist] [Rank 0] step=2390, skipped=4, lr=[5.8111111111111116e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:47:19,175] [INFO] [timer.py:196:stop] epoch=0/micro_step=2390/global_step=2390, RunningAvgSamplesPerSec=17.57983100210714, CurrSamplesPerSec=17.739075223701562, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:50:05,789] [INFO] [logging.py:68:log_dist] [Rank 0] step=2400, skipped=4, lr=[5.788888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:50:05,791] [INFO] [timer.py:196:stop] epoch=0/micro_step=2400/global_step=2400, RunningAvgSamplesPerSec=17.57936842818589, CurrSamplesPerSec=17.275238450785686, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.788888888888889e-06, 'epoch': 20.02}
[2022-12-18 20:52:58,867] [INFO] [logging.py:68:log_dist] [Rank 0] step=2410, skipped=4, lr=[5.766666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:52:58,869] [INFO] [timer.py:196:stop] epoch=0/micro_step=2410/global_step=2410, RunningAvgSamplesPerSec=17.579356999335438, CurrSamplesPerSec=17.78818291735644, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 20:55:55,624] [INFO] [logging.py:68:log_dist] [Rank 0] step=2420, skipped=4, lr=[5.744444444444444e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:55:55,626] [INFO] [timer.py:196:stop] epoch=0/micro_step=2420/global_step=2420, RunningAvgSamplesPerSec=17.579407925005107, CurrSamplesPerSec=17.430881689361808, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.733333333333334e-06, 'epoch': 20.02}
[2022-12-18 20:58:21,758] [INFO] [logging.py:68:log_dist] [Rank 0] step=2430, skipped=4, lr=[5.722222222222222e-06], mom=[[0.9, 0.999]]
[2022-12-18 20:58:21,759] [INFO] [timer.py:196:stop] epoch=0/micro_step=2430/global_step=2430, RunningAvgSamplesPerSec=17.579820294426774, CurrSamplesPerSec=17.848416611063957, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:01:36,112] [INFO] [logging.py:68:log_dist] [Rank 0] step=2440, skipped=4, lr=[5.7e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:01:36,114] [INFO] [timer.py:196:stop] epoch=0/micro_step=2440/global_step=2440, RunningAvgSamplesPerSec=17.58164190734309, CurrSamplesPerSec=17.721394159419724, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:04:29,479] [INFO] [logging.py:68:log_dist] [Rank 0] step=2450, skipped=4, lr=[5.677777777777779e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:04:29,480] [INFO] [timer.py:196:stop] epoch=0/micro_step=2450/global_step=2450, RunningAvgSamplesPerSec=17.58093222817337, CurrSamplesPerSec=17.640215008709855, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.677777777777779e-06, 'epoch': 21.0}
[2022-12-18 21:07:24,660] [INFO] [logging.py:68:log_dist] [Rank 0] step=2460, skipped=4, lr=[5.6555555555555566e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:07:24,662] [INFO] [timer.py:196:stop] epoch=0/micro_step=2460/global_step=2460, RunningAvgSamplesPerSec=17.581578510835662, CurrSamplesPerSec=17.685990737296397, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:10:17,643] [INFO] [logging.py:68:log_dist] [Rank 0] step=2470, skipped=4, lr=[5.633333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:10:17,645] [INFO] [timer.py:196:stop] epoch=0/micro_step=2470/global_step=2470, RunningAvgSamplesPerSec=17.581778337812683, CurrSamplesPerSec=17.34931845056666, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.622222222222222e-06, 'epoch': 21.01}
[2022-12-18 21:13:11,448] [INFO] [logging.py:68:log_dist] [Rank 0] step=2480, skipped=4, lr=[5.611111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:13:11,449] [INFO] [timer.py:196:stop] epoch=0/micro_step=2480/global_step=2480, RunningAvgSamplesPerSec=17.582006033901134, CurrSamplesPerSec=17.839142228697103, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:16:06,157] [INFO] [logging.py:68:log_dist] [Rank 0] step=2490, skipped=4, lr=[5.588888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:16:06,158] [INFO] [timer.py:196:stop] epoch=0/micro_step=2490/global_step=2490, RunningAvgSamplesPerSec=17.582427750603138, CurrSamplesPerSec=17.880589160865014, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:18:59,748] [INFO] [logging.py:68:log_dist] [Rank 0] step=2500, skipped=4, lr=[5.566666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:18:59,750] [INFO] [timer.py:196:stop] epoch=0/micro_step=2500/global_step=2500, RunningAvgSamplesPerSec=17.582743281644415, CurrSamplesPerSec=17.667207077532932, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.566666666666667e-06, 'epoch': 21.01}
[2022-12-18 21:21:55,675] [INFO] [logging.py:68:log_dist] [Rank 0] step=2510, skipped=4, lr=[5.544444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:21:55,676] [INFO] [timer.py:196:stop] epoch=0/micro_step=2510/global_step=2510, RunningAvgSamplesPerSec=17.583098376516066, CurrSamplesPerSec=17.81767938785372, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:24:52,558] [INFO] [logging.py:68:log_dist] [Rank 0] step=2520, skipped=4, lr=[5.522222222222222e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:24:52,560] [INFO] [timer.py:196:stop] epoch=0/micro_step=2520/global_step=2520, RunningAvgSamplesPerSec=17.582840937522246, CurrSamplesPerSec=17.70485992253366, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.511111111111112e-06, 'epoch': 21.02}
[2022-12-18 21:27:45,179] [INFO] [logging.py:68:log_dist] [Rank 0] step=2530, skipped=4, lr=[5.500000000000001e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:27:45,181] [INFO] [timer.py:196:stop] epoch=0/micro_step=2530/global_step=2530, RunningAvgSamplesPerSec=17.582970515129492, CurrSamplesPerSec=17.72798676626049, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:30:39,563] [INFO] [logging.py:68:log_dist] [Rank 0] step=2540, skipped=4, lr=[5.477777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:30:39,564] [INFO] [timer.py:196:stop] epoch=0/micro_step=2540/global_step=2540, RunningAvgSamplesPerSec=17.58328289403546, CurrSamplesPerSec=17.50187911710901, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:32:08,189] [INFO] [logging.py:68:log_dist] [Rank 0] step=2550, skipped=4, lr=[5.455555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:32:08,190] [INFO] [timer.py:196:stop] epoch=0/micro_step=2550/global_step=2550, RunningAvgSamplesPerSec=17.583618137138338, CurrSamplesPerSec=17.842301007185828, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.455555555555556e-06, 'epoch': 21.02}
[2022-12-18 21:36:17,169] [INFO] [logging.py:68:log_dist] [Rank 0] step=2560, skipped=4, lr=[5.4333333333333335e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:36:17,171] [INFO] [timer.py:196:stop] epoch=0/micro_step=2560/global_step=2560, RunningAvgSamplesPerSec=17.585649872292382, CurrSamplesPerSec=17.771402691482262, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:39:12,494] [INFO] [logging.py:68:log_dist] [Rank 0] step=2570, skipped=4, lr=[5.411111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:39:12,495] [INFO] [timer.py:196:stop] epoch=0/micro_step=2570/global_step=2570, RunningAvgSamplesPerSec=17.585777243664516, CurrSamplesPerSec=17.80074433622836, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.400000000000001e-06, 'epoch': 22.0}
[2022-12-18 21:42:06,025] [INFO] [logging.py:68:log_dist] [Rank 0] step=2580, skipped=4, lr=[5.388888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:42:06,027] [INFO] [timer.py:196:stop] epoch=0/micro_step=2580/global_step=2580, RunningAvgSamplesPerSec=17.58598090142047, CurrSamplesPerSec=17.732698105679262, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:45:00,930] [INFO] [logging.py:68:log_dist] [Rank 0] step=2590, skipped=4, lr=[5.366666666666666e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:45:00,931] [INFO] [timer.py:196:stop] epoch=0/micro_step=2590/global_step=2590, RunningAvgSamplesPerSec=17.58598904882969, CurrSamplesPerSec=17.75426775336124, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:47:53,472] [INFO] [logging.py:68:log_dist] [Rank 0] step=2600, skipped=4, lr=[5.344444444444446e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:47:53,473] [INFO] [timer.py:196:stop] epoch=0/micro_step=2600/global_step=2600, RunningAvgSamplesPerSec=17.586400031524665, CurrSamplesPerSec=17.76383611121313, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.344444444444446e-06, 'epoch': 22.01}
[2022-12-18 21:50:44,851] [INFO] [logging.py:68:log_dist] [Rank 0] step=2610, skipped=4, lr=[5.322222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:50:44,853] [INFO] [timer.py:196:stop] epoch=0/micro_step=2610/global_step=2610, RunningAvgSamplesPerSec=17.58646570100972, CurrSamplesPerSec=17.753315476544778, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:53:40,008] [INFO] [logging.py:68:log_dist] [Rank 0] step=2620, skipped=4, lr=[5.300000000000001e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:53:40,010] [INFO] [timer.py:196:stop] epoch=0/micro_step=2620/global_step=2620, RunningAvgSamplesPerSec=17.58683076347057, CurrSamplesPerSec=17.54458890529755, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.288888888888889e-06, 'epoch': 22.01}
[2022-12-18 21:56:35,777] [INFO] [logging.py:68:log_dist] [Rank 0] step=2630, skipped=4, lr=[5.2777777777777785e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:56:35,778] [INFO] [timer.py:196:stop] epoch=0/micro_step=2630/global_step=2630, RunningAvgSamplesPerSec=17.58673553198896, CurrSamplesPerSec=17.64228330921313, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 21:59:28,125] [INFO] [logging.py:68:log_dist] [Rank 0] step=2640, skipped=4, lr=[5.255555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 21:59:28,126] [INFO] [timer.py:196:stop] epoch=0/micro_step=2640/global_step=2640, RunningAvgSamplesPerSec=17.586800854228247, CurrSamplesPerSec=17.89417725454518, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:02:23,325] [INFO] [logging.py:68:log_dist] [Rank 0] step=2650, skipped=4, lr=[5.233333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:02:23,327] [INFO] [timer.py:196:stop] epoch=0/micro_step=2650/global_step=2650, RunningAvgSamplesPerSec=17.586772365655552, CurrSamplesPerSec=17.715312666966547, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.233333333333334e-06, 'epoch': 22.02}
[2022-12-18 22:05:15,888] [INFO] [logging.py:68:log_dist] [Rank 0] step=2660, skipped=4, lr=[5.211111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:05:15,890] [INFO] [timer.py:196:stop] epoch=0/micro_step=2660/global_step=2660, RunningAvgSamplesPerSec=17.586544231611196, CurrSamplesPerSec=17.743018393391534, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:08:01,767] [INFO] [logging.py:68:log_dist] [Rank 0] step=2670, skipped=4, lr=[5.188888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:08:01,768] [INFO] [timer.py:196:stop] epoch=0/micro_step=2670/global_step=2670, RunningAvgSamplesPerSec=17.588496695424183, CurrSamplesPerSec=17.52287111528657, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 5.177777777777779e-06, 'epoch': 23.0}
[2022-12-18 22:10:50,950] [INFO] [logging.py:68:log_dist] [Rank 0] step=2680, skipped=4, lr=[5.1666666666666675e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:10:50,952] [INFO] [timer.py:196:stop] epoch=0/micro_step=2680/global_step=2680, RunningAvgSamplesPerSec=17.58858867635249, CurrSamplesPerSec=17.248777082513122, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:13:51,347] [INFO] [logging.py:68:log_dist] [Rank 0] step=2690, skipped=4, lr=[5.144444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:13:51,348] [INFO] [timer.py:196:stop] epoch=0/micro_step=2690/global_step=2690, RunningAvgSamplesPerSec=17.588844419882182, CurrSamplesPerSec=17.658542836374163, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:16:44,879] [INFO] [logging.py:68:log_dist] [Rank 0] step=2700, skipped=4, lr=[5.122222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:16:44,882] [INFO] [timer.py:196:stop] epoch=0/micro_step=2700/global_step=2700, RunningAvgSamplesPerSec=17.58864397892246, CurrSamplesPerSec=17.456245398750664, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0004, 'learning_rate': 5.122222222222223e-06, 'epoch': 23.01}
[2022-12-18 22:19:38,471] [INFO] [logging.py:68:log_dist] [Rank 0] step=2710, skipped=4, lr=[5.1e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:19:38,473] [INFO] [timer.py:196:stop] epoch=0/micro_step=2710/global_step=2710, RunningAvgSamplesPerSec=17.588406670968315, CurrSamplesPerSec=17.600804966548136, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:22:34,792] [INFO] [logging.py:68:log_dist] [Rank 0] step=2720, skipped=4, lr=[5.077777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:22:34,794] [INFO] [timer.py:196:stop] epoch=0/micro_step=2720/global_step=2720, RunningAvgSamplesPerSec=17.588412191213653, CurrSamplesPerSec=17.538050615681325, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 5.0666666666666676e-06, 'epoch': 23.01}
[2022-12-18 22:25:28,224] [INFO] [logging.py:68:log_dist] [Rank 0] step=2730, skipped=4, lr=[5.0555555555555555e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:25:28,226] [INFO] [timer.py:196:stop] epoch=0/micro_step=2730/global_step=2730, RunningAvgSamplesPerSec=17.588617205030992, CurrSamplesPerSec=17.88840098418525, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:28:23,156] [INFO] [logging.py:68:log_dist] [Rank 0] step=2740, skipped=4, lr=[5.033333333333333e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:28:23,158] [INFO] [timer.py:196:stop] epoch=0/micro_step=2740/global_step=2740, RunningAvgSamplesPerSec=17.588461109394714, CurrSamplesPerSec=17.725434811734722, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:31:20,932] [INFO] [logging.py:68:log_dist] [Rank 0] step=2750, skipped=4, lr=[5.011111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:31:20,934] [INFO] [timer.py:196:stop] epoch=0/micro_step=2750/global_step=2750, RunningAvgSamplesPerSec=17.58873040259882, CurrSamplesPerSec=17.56283566739676, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 5.011111111111111e-06, 'epoch': 23.02}
[2022-12-18 22:34:16,935] [INFO] [logging.py:68:log_dist] [Rank 0] step=2760, skipped=4, lr=[4.988888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:34:16,937] [INFO] [timer.py:196:stop] epoch=0/micro_step=2760/global_step=2760, RunningAvgSamplesPerSec=17.588860264969593, CurrSamplesPerSec=17.791513521785795, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:37:12,149] [INFO] [logging.py:68:log_dist] [Rank 0] step=2770, skipped=4, lr=[4.966666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:37:12,151] [INFO] [timer.py:196:stop] epoch=0/micro_step=2770/global_step=2770, RunningAvgSamplesPerSec=17.58794510976748, CurrSamplesPerSec=17.61953551482156, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.9555555555555565e-06, 'epoch': 23.02}
[2022-12-18 22:39:12,736] [INFO] [logging.py:68:log_dist] [Rank 0] step=2780, skipped=4, lr=[4.944444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:39:12,737] [INFO] [timer.py:196:stop] epoch=0/micro_step=2780/global_step=2780, RunningAvgSamplesPerSec=17.587929523283982, CurrSamplesPerSec=17.720306201301174, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:42:55,052] [INFO] [logging.py:68:log_dist] [Rank 0] step=2790, skipped=4, lr=[4.922222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:42:55,054] [INFO] [timer.py:196:stop] epoch=0/micro_step=2790/global_step=2790, RunningAvgSamplesPerSec=17.589726417045362, CurrSamplesPerSec=17.559128376382404, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:45:49,838] [INFO] [logging.py:68:log_dist] [Rank 0] step=2800, skipped=4, lr=[4.9000000000000005e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:45:49,840] [INFO] [timer.py:196:stop] epoch=0/micro_step=2800/global_step=2800, RunningAvgSamplesPerSec=17.589809944824765, CurrSamplesPerSec=17.945423836116547, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.9000000000000005e-06, 'epoch': 24.0}
[2022-12-18 22:48:44,938] [INFO] [logging.py:68:log_dist] [Rank 0] step=2810, skipped=4, lr=[4.877777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:48:44,939] [INFO] [timer.py:196:stop] epoch=0/micro_step=2810/global_step=2810, RunningAvgSamplesPerSec=17.5902174761403, CurrSamplesPerSec=17.771930969802167, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:51:38,599] [INFO] [logging.py:68:log_dist] [Rank 0] step=2820, skipped=4, lr=[4.855555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:51:38,601] [INFO] [timer.py:196:stop] epoch=0/micro_step=2820/global_step=2820, RunningAvgSamplesPerSec=17.590656004828077, CurrSamplesPerSec=17.86646514749002, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.8444444444444446e-06, 'epoch': 24.01}
[2022-12-18 22:54:31,219] [INFO] [logging.py:68:log_dist] [Rank 0] step=2830, skipped=4, lr=[4.833333333333333e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:54:31,221] [INFO] [timer.py:196:stop] epoch=0/micro_step=2830/global_step=2830, RunningAvgSamplesPerSec=17.590752786947785, CurrSamplesPerSec=17.617874924359466, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 22:57:25,167] [INFO] [logging.py:68:log_dist] [Rank 0] step=2840, skipped=4, lr=[4.811111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-18 22:57:25,170] [INFO] [timer.py:196:stop] epoch=0/micro_step=2840/global_step=2840, RunningAvgSamplesPerSec=17.590555766000215, CurrSamplesPerSec=17.295558482869577, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 23:00:19,469] [INFO] [logging.py:68:log_dist] [Rank 0] step=2850, skipped=4, lr=[4.7888888888888894e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:00:19,471] [INFO] [timer.py:196:stop] epoch=0/micro_step=2850/global_step=2850, RunningAvgSamplesPerSec=17.591164207263002, CurrSamplesPerSec=17.847159930519627, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.7888888888888894e-06, 'epoch': 24.01}
[2022-12-18 23:03:13,584] [INFO] [logging.py:68:log_dist] [Rank 0] step=2860, skipped=4, lr=[4.766666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:03:13,586] [INFO] [timer.py:196:stop] epoch=0/micro_step=2860/global_step=2860, RunningAvgSamplesPerSec=17.590997354809485, CurrSamplesPerSec=17.796866322148755, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 23:06:04,529] [INFO] [logging.py:68:log_dist] [Rank 0] step=2870, skipped=4, lr=[4.744444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:06:04,530] [INFO] [timer.py:196:stop] epoch=0/micro_step=2870/global_step=2870, RunningAvgSamplesPerSec=17.5904808921738, CurrSamplesPerSec=17.66057360861417, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.7333333333333335e-06, 'epoch': 24.02}
[2022-12-18 23:08:58,225] [INFO] [logging.py:68:log_dist] [Rank 0] step=2880, skipped=4, lr=[4.722222222222222e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:08:58,227] [INFO] [timer.py:196:stop] epoch=0/micro_step=2880/global_step=2880, RunningAvgSamplesPerSec=17.590600070219743, CurrSamplesPerSec=17.506408233864114, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 23:11:51,964] [INFO] [logging.py:68:log_dist] [Rank 0] step=2890, skipped=4, lr=[4.7e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:11:51,965] [INFO] [timer.py:196:stop] epoch=0/micro_step=2890/global_step=2890, RunningAvgSamplesPerSec=17.590550965074637, CurrSamplesPerSec=17.614375526548752, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 23:12:54,939] [INFO] [logging.py:68:log_dist] [Rank 0] step=2900, skipped=4, lr=[4.677777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:12:54,941] [INFO] [timer.py:196:stop] epoch=0/micro_step=2900/global_step=2900, RunningAvgSamplesPerSec=17.592575141800577, CurrSamplesPerSec=23.402610615692446, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.677777777777778e-06, 'epoch': 24.02}
[2022-12-18 23:17:35,222] [INFO] [logging.py:68:log_dist] [Rank 0] step=2910, skipped=4, lr=[4.655555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:17:35,224] [INFO] [timer.py:196:stop] epoch=0/micro_step=2910/global_step=2910, RunningAvgSamplesPerSec=17.591926084578816, CurrSamplesPerSec=17.573272043622048, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 23:20:26,235] [INFO] [logging.py:68:log_dist] [Rank 0] step=2920, skipped=4, lr=[4.633333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:20:26,237] [INFO] [timer.py:196:stop] epoch=0/micro_step=2920/global_step=2920, RunningAvgSamplesPerSec=17.591872803186433, CurrSamplesPerSec=17.695904407273744, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.622222222222222e-06, 'epoch': 25.0}
[2022-12-18 23:23:20,314] [INFO] [logging.py:68:log_dist] [Rank 0] step=2930, skipped=4, lr=[4.611111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:23:20,315] [INFO] [timer.py:196:stop] epoch=0/micro_step=2930/global_step=2930, RunningAvgSamplesPerSec=17.59107076752098, CurrSamplesPerSec=17.480459245563846, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 23:26:13,930] [INFO] [logging.py:68:log_dist] [Rank 0] step=2940, skipped=4, lr=[4.58888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:26:13,931] [INFO] [timer.py:196:stop] epoch=0/micro_step=2940/global_step=2940, RunningAvgSamplesPerSec=17.59113715229797, CurrSamplesPerSec=17.790508903722518, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 23:29:07,857] [INFO] [logging.py:68:log_dist] [Rank 0] step=2950, skipped=4, lr=[4.566666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:29:07,858] [INFO] [timer.py:196:stop] epoch=0/micro_step=2950/global_step=2950, RunningAvgSamplesPerSec=17.591430627483753, CurrSamplesPerSec=17.717198652535526, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.566666666666667e-06, 'epoch': 25.01}
[2022-12-18 23:31:55,211] [INFO] [logging.py:68:log_dist] [Rank 0] step=2960, skipped=4, lr=[4.544444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:31:55,213] [INFO] [timer.py:196:stop] epoch=0/micro_step=2960/global_step=2960, RunningAvgSamplesPerSec=17.590838770130556, CurrSamplesPerSec=16.62445389317561, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 23:34:41,515] [INFO] [logging.py:68:log_dist] [Rank 0] step=2970, skipped=4, lr=[4.5222222222222225e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:34:41,517] [INFO] [timer.py:196:stop] epoch=0/micro_step=2970/global_step=2970, RunningAvgSamplesPerSec=17.590421144313158, CurrSamplesPerSec=17.376665340500757, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.511111111111111e-06, 'epoch': 25.02}
[2022-12-18 23:37:30,188] [INFO] [logging.py:68:log_dist] [Rank 0] step=2980, skipped=4, lr=[4.5e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:37:30,190] [INFO] [timer.py:196:stop] epoch=0/micro_step=2980/global_step=2980, RunningAvgSamplesPerSec=17.590793742559544, CurrSamplesPerSec=17.757612686575012, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 23:40:23,212] [INFO] [logging.py:68:log_dist] [Rank 0] step=2990, skipped=4, lr=[4.477777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:40:23,213] [INFO] [timer.py:196:stop] epoch=0/micro_step=2990/global_step=2990, RunningAvgSamplesPerSec=17.590394204062903, CurrSamplesPerSec=17.59256082394663, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-18 23:43:19,061] [INFO] [logging.py:68:log_dist] [Rank 0] step=3000, skipped=4, lr=[4.455555555555555e-06], mom=[[0.9, 0.999]]
[2022-12-18 23:43:19,062] [INFO] [timer.py:196:stop] epoch=0/micro_step=3000/global_step=3000, RunningAvgSamplesPerSec=17.59020976726772, CurrSamplesPerSec=17.293852551111918, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.455555555555555e-06, 'epoch': 25.02}
{'eval_loss': 0.32666015625, 'eval_wer': 17.688061708300896, 'eval_runtime': 1253.1843, 'eval_samples_per_second': 3.079, 'eval_steps_per_second': 0.097, 'epoch': 25.02}
[2022-12-19 00:04:13,336] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step3000 is begin to save!
[2022-12-19 00:04:13,344] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: ./checkpoint-3000/global_step3000/mp_rank_00_model_states.pt
[2022-12-19 00:04:13,344] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving ./checkpoint-3000/global_step3000/mp_rank_00_model_states.pt...
[2022-12-19 00:04:14,304] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved ./checkpoint-3000/global_step3000/mp_rank_00_model_states.pt.
[2022-12-19 00:04:14,306] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving ./checkpoint-3000/global_step3000/zero_pp_rank_0_mp_rank_00_optim_states.pt...
[2022-12-19 00:04:18,796] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved ./checkpoint-3000/global_step3000/zero_pp_rank_0_mp_rank_00_optim_states.pt.
[2022-12-19 00:04:18,797] [INFO] [engine.py:3394:_save_zero_checkpoint] zero checkpoint saved ./checkpoint-3000/global_step3000/zero_pp_rank_0_mp_rank_00_optim_states.pt
[2022-12-19 00:04:18,797] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now!
[2022-12-19 00:06:55,197] [INFO] [stage_1_and_2.py:1767:step] [deepspeed] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 65536.0, reducing to 65536.0
[2022-12-19 00:07:13,410] [INFO] [stage_1_and_2.py:1767:step] [deepspeed] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 65536.0, reducing to 32768.0
[2022-12-19 00:07:53,402] [INFO] [logging.py:68:log_dist] [Rank 0] step=3010, skipped=6, lr=[4.437777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:07:53,403] [INFO] [timer.py:196:stop] epoch=0/micro_step=3010/global_step=3010, RunningAvgSamplesPerSec=17.591068861221267, CurrSamplesPerSec=17.764448583346613, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 00:11:06,978] [INFO] [logging.py:68:log_dist] [Rank 0] step=3020, skipped=6, lr=[4.415555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:11:06,980] [INFO] [timer.py:196:stop] epoch=0/micro_step=3020/global_step=3020, RunningAvgSamplesPerSec=17.592794094238695, CurrSamplesPerSec=17.57114513059809, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.404444444444445e-06, 'epoch': 26.0}
[2022-12-19 00:14:04,963] [INFO] [logging.py:68:log_dist] [Rank 0] step=3030, skipped=6, lr=[4.393333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:14:04,964] [INFO] [timer.py:196:stop] epoch=0/micro_step=3030/global_step=3030, RunningAvgSamplesPerSec=17.593074428336145, CurrSamplesPerSec=17.549051816080418, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 00:16:56,281] [INFO] [logging.py:68:log_dist] [Rank 0] step=3040, skipped=6, lr=[4.371111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:16:56,282] [INFO] [timer.py:196:stop] epoch=0/micro_step=3040/global_step=3040, RunningAvgSamplesPerSec=17.593342880204634, CurrSamplesPerSec=17.68683907958883, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 00:19:47,359] [INFO] [logging.py:68:log_dist] [Rank 0] step=3050, skipped=6, lr=[4.348888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:19:47,360] [INFO] [timer.py:196:stop] epoch=0/micro_step=3050/global_step=3050, RunningAvgSamplesPerSec=17.593460947035926, CurrSamplesPerSec=17.70367941895282, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.348888888888889e-06, 'epoch': 26.01}
[2022-12-19 00:22:42,032] [INFO] [logging.py:68:log_dist] [Rank 0] step=3060, skipped=6, lr=[4.326666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:22:42,034] [INFO] [timer.py:196:stop] epoch=0/micro_step=3060/global_step=3060, RunningAvgSamplesPerSec=17.59375044318361, CurrSamplesPerSec=17.384547334188284, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 00:25:34,669] [INFO] [logging.py:68:log_dist] [Rank 0] step=3070, skipped=6, lr=[4.304444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:25:34,671] [INFO] [timer.py:196:stop] epoch=0/micro_step=3070/global_step=3070, RunningAvgSamplesPerSec=17.5937122892984, CurrSamplesPerSec=17.613768735779285, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.2933333333333334e-06, 'epoch': 26.01}
[2022-12-19 00:28:25,299] [INFO] [logging.py:68:log_dist] [Rank 0] step=3080, skipped=6, lr=[4.282222222222222e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:28:25,300] [INFO] [timer.py:196:stop] epoch=0/micro_step=3080/global_step=3080, RunningAvgSamplesPerSec=17.593914468788185, CurrSamplesPerSec=17.718895564207596, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 00:31:21,700] [INFO] [logging.py:68:log_dist] [Rank 0] step=3090, skipped=6, lr=[4.26e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:31:21,701] [INFO] [timer.py:196:stop] epoch=0/micro_step=3090/global_step=3090, RunningAvgSamplesPerSec=17.593445388805577, CurrSamplesPerSec=17.325692189061847, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 00:34:16,335] [INFO] [logging.py:68:log_dist] [Rank 0] step=3100, skipped=6, lr=[4.2377777777777775e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:34:16,336] [INFO] [timer.py:196:stop] epoch=0/micro_step=3100/global_step=3100, RunningAvgSamplesPerSec=17.59348166098487, CurrSamplesPerSec=17.3108812570425, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.2377777777777775e-06, 'epoch': 26.02}
[2022-12-19 00:37:12,056] [INFO] [logging.py:68:log_dist] [Rank 0] step=3110, skipped=6, lr=[4.215555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:37:12,058] [INFO] [timer.py:196:stop] epoch=0/micro_step=3110/global_step=3110, RunningAvgSamplesPerSec=17.593623928588435, CurrSamplesPerSec=17.863522480875613, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 00:40:07,743] [INFO] [logging.py:68:log_dist] [Rank 0] step=3120, skipped=6, lr=[4.1933333333333336e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:40:07,744] [INFO] [timer.py:196:stop] epoch=0/micro_step=3120/global_step=3120, RunningAvgSamplesPerSec=17.593697302631174, CurrSamplesPerSec=17.08051252585759, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.182222222222222e-06, 'epoch': 26.02}
[2022-12-19 00:41:38,634] [INFO] [logging.py:68:log_dist] [Rank 0] step=3130, skipped=6, lr=[4.171111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:41:38,636] [INFO] [timer.py:196:stop] epoch=0/micro_step=3130/global_step=3130, RunningAvgSamplesPerSec=17.593999749820743, CurrSamplesPerSec=17.741973511374713, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 00:45:51,863] [INFO] [logging.py:68:log_dist] [Rank 0] step=3140, skipped=6, lr=[4.148888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:45:51,865] [INFO] [timer.py:196:stop] epoch=0/micro_step=3140/global_step=3140, RunningAvgSamplesPerSec=17.595327756966622, CurrSamplesPerSec=17.445340938254706, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 00:48:49,392] [INFO] [logging.py:68:log_dist] [Rank 0] step=3150, skipped=6, lr=[4.126666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:48:49,394] [INFO] [timer.py:196:stop] epoch=0/micro_step=3150/global_step=3150, RunningAvgSamplesPerSec=17.595551909670867, CurrSamplesPerSec=17.525080176118156, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.126666666666667e-06, 'epoch': 27.0}
[2022-12-19 00:51:44,912] [INFO] [logging.py:68:log_dist] [Rank 0] step=3160, skipped=6, lr=[4.104444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:51:44,914] [INFO] [timer.py:196:stop] epoch=0/micro_step=3160/global_step=3160, RunningAvgSamplesPerSec=17.59533705410767, CurrSamplesPerSec=17.492849569178176, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 00:54:39,253] [INFO] [logging.py:68:log_dist] [Rank 0] step=3170, skipped=6, lr=[4.0822222222222225e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:54:39,254] [INFO] [timer.py:196:stop] epoch=0/micro_step=3170/global_step=3170, RunningAvgSamplesPerSec=17.595644102704394, CurrSamplesPerSec=17.775633328909876, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.071111111111111e-06, 'epoch': 27.01}
[2022-12-19 00:57:32,029] [INFO] [logging.py:68:log_dist] [Rank 0] step=3180, skipped=6, lr=[4.060000000000001e-06], mom=[[0.9, 0.999]]
[2022-12-19 00:57:32,031] [INFO] [timer.py:196:stop] epoch=0/micro_step=3180/global_step=3180, RunningAvgSamplesPerSec=17.59573223021548, CurrSamplesPerSec=17.551840136671807, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:00:26,835] [INFO] [logging.py:68:log_dist] [Rank 0] step=3190, skipped=6, lr=[4.0377777777777786e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:00:26,836] [INFO] [timer.py:196:stop] epoch=0/micro_step=3190/global_step=3190, RunningAvgSamplesPerSec=17.59571717237336, CurrSamplesPerSec=17.490995086617858, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:03:19,686] [INFO] [logging.py:68:log_dist] [Rank 0] step=3200, skipped=6, lr=[4.015555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:03:19,688] [INFO] [timer.py:196:stop] epoch=0/micro_step=3200/global_step=3200, RunningAvgSamplesPerSec=17.59580399188835, CurrSamplesPerSec=17.34269068867683, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 4.015555555555556e-06, 'epoch': 27.01}
[2022-12-19 01:06:14,654] [INFO] [logging.py:68:log_dist] [Rank 0] step=3210, skipped=6, lr=[3.993333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:06:14,655] [INFO] [timer.py:196:stop] epoch=0/micro_step=3210/global_step=3210, RunningAvgSamplesPerSec=17.596046138360002, CurrSamplesPerSec=17.725228814430604, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:09:07,511] [INFO] [logging.py:68:log_dist] [Rank 0] step=3220, skipped=6, lr=[3.971111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:09:07,512] [INFO] [timer.py:196:stop] epoch=0/micro_step=3220/global_step=3220, RunningAvgSamplesPerSec=17.59647348251674, CurrSamplesPerSec=17.602937912133203, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 3.96e-06, 'epoch': 27.02}
[2022-12-19 01:12:01,533] [INFO] [logging.py:68:log_dist] [Rank 0] step=3230, skipped=6, lr=[3.948888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:12:01,535] [INFO] [timer.py:196:stop] epoch=0/micro_step=3230/global_step=3230, RunningAvgSamplesPerSec=17.596052986386066, CurrSamplesPerSec=17.566659481255172, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:14:55,505] [INFO] [logging.py:68:log_dist] [Rank 0] step=3240, skipped=6, lr=[3.926666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:14:55,507] [INFO] [timer.py:196:stop] epoch=0/micro_step=3240/global_step=3240, RunningAvgSamplesPerSec=17.596036037139132, CurrSamplesPerSec=17.498455300383174, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:17:42,447] [INFO] [logging.py:68:log_dist] [Rank 0] step=3250, skipped=6, lr=[3.904444444444444e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:17:42,449] [INFO] [timer.py:196:stop] epoch=0/micro_step=3250/global_step=3250, RunningAvgSamplesPerSec=17.597616914434525, CurrSamplesPerSec=17.628235600063963, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 3.904444444444444e-06, 'epoch': 28.0}
[2022-12-19 01:20:40,342] [INFO] [logging.py:68:log_dist] [Rank 0] step=3260, skipped=6, lr=[3.882222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:20:40,344] [INFO] [timer.py:196:stop] epoch=0/micro_step=3260/global_step=3260, RunningAvgSamplesPerSec=17.59767515749172, CurrSamplesPerSec=17.64395546265293, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:23:41,372] [INFO] [logging.py:68:log_dist] [Rank 0] step=3270, skipped=6, lr=[3.86e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:23:41,373] [INFO] [timer.py:196:stop] epoch=0/micro_step=3270/global_step=3270, RunningAvgSamplesPerSec=17.597920364006136, CurrSamplesPerSec=17.849487124433125, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 3.848888888888889e-06, 'epoch': 28.01}
[2022-12-19 01:26:34,140] [INFO] [logging.py:68:log_dist] [Rank 0] step=3280, skipped=6, lr=[3.837777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:26:34,141] [INFO] [timer.py:196:stop] epoch=0/micro_step=3280/global_step=3280, RunningAvgSamplesPerSec=17.597884174195094, CurrSamplesPerSec=17.131544523208934, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:29:24,859] [INFO] [logging.py:68:log_dist] [Rank 0] step=3290, skipped=6, lr=[3.8155555555555555e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:29:24,861] [INFO] [timer.py:196:stop] epoch=0/micro_step=3290/global_step=3290, RunningAvgSamplesPerSec=17.597922188226125, CurrSamplesPerSec=17.589932443660675, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:32:24,242] [INFO] [logging.py:68:log_dist] [Rank 0] step=3300, skipped=6, lr=[3.793333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:32:24,244] [INFO] [timer.py:196:stop] epoch=0/micro_step=3300/global_step=3300, RunningAvgSamplesPerSec=17.59799470754608, CurrSamplesPerSec=17.711194827166782, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 3.793333333333334e-06, 'epoch': 28.01}
[2022-12-19 01:35:16,557] [INFO] [logging.py:68:log_dist] [Rank 0] step=3310, skipped=6, lr=[3.7711111111111116e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:35:16,558] [INFO] [timer.py:196:stop] epoch=0/micro_step=3310/global_step=3310, RunningAvgSamplesPerSec=17.598131006457837, CurrSamplesPerSec=17.669246821453136, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:38:09,745] [INFO] [logging.py:68:log_dist] [Rank 0] step=3320, skipped=6, lr=[3.7488888888888892e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:38:09,747] [INFO] [timer.py:196:stop] epoch=0/micro_step=3320/global_step=3320, RunningAvgSamplesPerSec=17.59810478665201, CurrSamplesPerSec=17.712917471280523, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 3.737777777777778e-06, 'epoch': 28.02}
[2022-12-19 01:40:59,612] [INFO] [logging.py:68:log_dist] [Rank 0] step=3330, skipped=6, lr=[3.726666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:40:59,613] [INFO] [timer.py:196:stop] epoch=0/micro_step=3330/global_step=3330, RunningAvgSamplesPerSec=17.5981314985267, CurrSamplesPerSec=17.433335941926863, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:43:50,657] [INFO] [logging.py:68:log_dist] [Rank 0] step=3340, skipped=6, lr=[3.704444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:43:50,659] [INFO] [timer.py:196:stop] epoch=0/micro_step=3340/global_step=3340, RunningAvgSamplesPerSec=17.59849399704335, CurrSamplesPerSec=17.73495453276351, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:46:42,496] [INFO] [logging.py:68:log_dist] [Rank 0] step=3350, skipped=6, lr=[3.6822222222222225e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:46:42,497] [INFO] [timer.py:196:stop] epoch=0/micro_step=3350/global_step=3350, RunningAvgSamplesPerSec=17.598006416218745, CurrSamplesPerSec=16.950657243750207, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 3.6822222222222225e-06, 'epoch': 28.02}
[2022-12-19 01:48:39,809] [INFO] [logging.py:68:log_dist] [Rank 0] step=3360, skipped=6, lr=[3.66e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:48:39,811] [INFO] [timer.py:196:stop] epoch=0/micro_step=3360/global_step=3360, RunningAvgSamplesPerSec=17.59821946036748, CurrSamplesPerSec=17.671893143144626, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:52:23,796] [INFO] [logging.py:68:log_dist] [Rank 0] step=3370, skipped=6, lr=[3.6377777777777777e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:52:23,797] [INFO] [timer.py:196:stop] epoch=0/micro_step=3370/global_step=3370, RunningAvgSamplesPerSec=17.599458275412836, CurrSamplesPerSec=17.74689643878602, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 3.6266666666666674e-06, 'epoch': 29.0}
[2022-12-19 01:55:17,361] [INFO] [logging.py:68:log_dist] [Rank 0] step=3380, skipped=6, lr=[3.615555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:55:17,362] [INFO] [timer.py:196:stop] epoch=0/micro_step=3380/global_step=3380, RunningAvgSamplesPerSec=17.599580206175794, CurrSamplesPerSec=17.84562699614934, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 01:58:09,807] [INFO] [logging.py:68:log_dist] [Rank 0] step=3390, skipped=6, lr=[3.593333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-19 01:58:09,809] [INFO] [timer.py:196:stop] epoch=0/micro_step=3390/global_step=3390, RunningAvgSamplesPerSec=17.599584006196714, CurrSamplesPerSec=17.690764887441887, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:01:00,280] [INFO] [logging.py:68:log_dist] [Rank 0] step=3400, skipped=6, lr=[3.5711111111111114e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:01:00,282] [INFO] [timer.py:196:stop] epoch=0/micro_step=3400/global_step=3400, RunningAvgSamplesPerSec=17.599061452055388, CurrSamplesPerSec=17.68211014301976, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 3.5711111111111114e-06, 'epoch': 29.01}
[2022-12-19 02:03:54,109] [INFO] [logging.py:68:log_dist] [Rank 0] step=3410, skipped=6, lr=[3.548888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:03:54,110] [INFO] [timer.py:196:stop] epoch=0/micro_step=3410/global_step=3410, RunningAvgSamplesPerSec=17.599020297195587, CurrSamplesPerSec=17.79227531486809, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:06:47,910] [INFO] [logging.py:68:log_dist] [Rank 0] step=3420, skipped=6, lr=[3.526666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:06:47,911] [INFO] [timer.py:196:stop] epoch=0/micro_step=3420/global_step=3420, RunningAvgSamplesPerSec=17.598935188193543, CurrSamplesPerSec=17.586873905797386, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 3.515555555555556e-06, 'epoch': 29.01}
[2022-12-19 02:09:44,739] [INFO] [logging.py:68:log_dist] [Rank 0] step=3430, skipped=6, lr=[3.5044444444444447e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:09:44,740] [INFO] [timer.py:196:stop] epoch=0/micro_step=3430/global_step=3430, RunningAvgSamplesPerSec=17.598879764126053, CurrSamplesPerSec=17.206409451147653, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:12:43,002] [INFO] [logging.py:68:log_dist] [Rank 0] step=3440, skipped=6, lr=[3.4822222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:12:43,004] [INFO] [timer.py:196:stop] epoch=0/micro_step=3440/global_step=3440, RunningAvgSamplesPerSec=17.59877386122984, CurrSamplesPerSec=17.647455022848266, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:15:49,935] [INFO] [logging.py:68:log_dist] [Rank 0] step=3450, skipped=6, lr=[3.46e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:15:49,936] [INFO] [timer.py:196:stop] epoch=0/micro_step=3450/global_step=3450, RunningAvgSamplesPerSec=17.59846592973248, CurrSamplesPerSec=17.705098144125525, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 3.46e-06, 'epoch': 29.02}
[2022-12-19 02:19:02,277] [INFO] [logging.py:68:log_dist] [Rank 0] step=3460, skipped=6, lr=[3.4377777777777784e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:19:02,279] [INFO] [timer.py:196:stop] epoch=0/micro_step=3460/global_step=3460, RunningAvgSamplesPerSec=17.59887474281011, CurrSamplesPerSec=17.339971779112254, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:21:59,258] [INFO] [logging.py:68:log_dist] [Rank 0] step=3470, skipped=6, lr=[3.415555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:21:59,259] [INFO] [timer.py:196:stop] epoch=0/micro_step=3470/global_step=3470, RunningAvgSamplesPerSec=17.598957745973937, CurrSamplesPerSec=17.47932554863768, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 3.404444444444445e-06, 'epoch': 29.02}
[2022-12-19 02:23:02,302] [INFO] [logging.py:68:log_dist] [Rank 0] step=3480, skipped=6, lr=[3.3933333333333336e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:23:02,304] [INFO] [timer.py:196:stop] epoch=0/micro_step=3480/global_step=3480, RunningAvgSamplesPerSec=17.600449447233288, CurrSamplesPerSec=23.367460582411866, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:27:40,487] [INFO] [logging.py:68:log_dist] [Rank 0] step=3490, skipped=6, lr=[3.371111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:27:40,488] [INFO] [timer.py:196:stop] epoch=0/micro_step=3490/global_step=3490, RunningAvgSamplesPerSec=17.600039649295883, CurrSamplesPerSec=17.403154157900058, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:30:34,939] [INFO] [logging.py:68:log_dist] [Rank 0] step=3500, skipped=6, lr=[3.3488888888888892e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:30:34,941] [INFO] [timer.py:196:stop] epoch=0/micro_step=3500/global_step=3500, RunningAvgSamplesPerSec=17.600239248768926, CurrSamplesPerSec=17.787628920083662, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0003, 'learning_rate': 3.3488888888888892e-06, 'epoch': 30.0}
[2022-12-19 02:33:30,948] [INFO] [logging.py:68:log_dist] [Rank 0] step=3510, skipped=6, lr=[3.326666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:33:30,950] [INFO] [timer.py:196:stop] epoch=0/micro_step=3510/global_step=3510, RunningAvgSamplesPerSec=17.60011342639632, CurrSamplesPerSec=17.535298747587174, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:36:27,492] [INFO] [logging.py:68:log_dist] [Rank 0] step=3520, skipped=6, lr=[3.3044444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:36:27,493] [INFO] [timer.py:196:stop] epoch=0/micro_step=3520/global_step=3520, RunningAvgSamplesPerSec=17.60047227468804, CurrSamplesPerSec=17.78531311367044, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 3.2933333333333333e-06, 'epoch': 30.01}
[2022-12-19 02:39:27,522] [INFO] [logging.py:68:log_dist] [Rank 0] step=3530, skipped=6, lr=[3.282222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:39:27,523] [INFO] [timer.py:196:stop] epoch=0/micro_step=3530/global_step=3530, RunningAvgSamplesPerSec=17.600453018305693, CurrSamplesPerSec=17.688962622702615, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:42:17,713] [INFO] [logging.py:68:log_dist] [Rank 0] step=3540, skipped=6, lr=[3.2600000000000006e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:42:17,714] [INFO] [timer.py:196:stop] epoch=0/micro_step=3540/global_step=3540, RunningAvgSamplesPerSec=17.600259041781936, CurrSamplesPerSec=17.62578056332801, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:45:09,551] [INFO] [logging.py:68:log_dist] [Rank 0] step=3550, skipped=6, lr=[3.237777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:45:09,553] [INFO] [timer.py:196:stop] epoch=0/micro_step=3550/global_step=3550, RunningAvgSamplesPerSec=17.60029625206999, CurrSamplesPerSec=17.671377775279126, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 3.237777777777778e-06, 'epoch': 30.01}
[2022-12-19 02:48:06,078] [INFO] [logging.py:68:log_dist] [Rank 0] step=3560, skipped=6, lr=[3.2155555555555558e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:48:06,079] [INFO] [timer.py:196:stop] epoch=0/micro_step=3560/global_step=3560, RunningAvgSamplesPerSec=17.600329628889238, CurrSamplesPerSec=17.71231088469093, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:51:00,756] [INFO] [logging.py:68:log_dist] [Rank 0] step=3570, skipped=6, lr=[3.193333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:51:00,758] [INFO] [timer.py:196:stop] epoch=0/micro_step=3570/global_step=3570, RunningAvgSamplesPerSec=17.60022112966504, CurrSamplesPerSec=17.219200593227026, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 3.1822222222222226e-06, 'epoch': 30.02}
[2022-12-19 02:53:51,858] [INFO] [logging.py:68:log_dist] [Rank 0] step=3580, skipped=6, lr=[3.1711111111111114e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:53:51,859] [INFO] [timer.py:196:stop] epoch=0/micro_step=3580/global_step=3580, RunningAvgSamplesPerSec=17.60064739187315, CurrSamplesPerSec=17.62833168565665, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:56:13,648] [INFO] [logging.py:68:log_dist] [Rank 0] step=3590, skipped=6, lr=[3.148888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:56:13,649] [INFO] [timer.py:196:stop] epoch=0/micro_step=3590/global_step=3590, RunningAvgSamplesPerSec=17.600825925087175, CurrSamplesPerSec=17.855607656356614, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 02:59:27,493] [INFO] [logging.py:68:log_dist] [Rank 0] step=3600, skipped=6, lr=[3.1266666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 02:59:27,494] [INFO] [timer.py:196:stop] epoch=0/micro_step=3600/global_step=3600, RunningAvgSamplesPerSec=17.601838241347792, CurrSamplesPerSec=17.581900003903662, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 3.1266666666666667e-06, 'epoch': 31.0}
[2022-12-19 03:02:16,659] [INFO] [logging.py:68:log_dist] [Rank 0] step=3610, skipped=6, lr=[3.104444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:02:16,661] [INFO] [timer.py:196:stop] epoch=0/micro_step=3610/global_step=3610, RunningAvgSamplesPerSec=17.601819994166515, CurrSamplesPerSec=17.664054192793216, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:05:08,497] [INFO] [logging.py:68:log_dist] [Rank 0] step=3620, skipped=6, lr=[3.0822222222222227e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:05:08,498] [INFO] [timer.py:196:stop] epoch=0/micro_step=3620/global_step=3620, RunningAvgSamplesPerSec=17.60192426766313, CurrSamplesPerSec=17.875462043633274, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 3.0711111111111115e-06, 'epoch': 31.01}
[2022-12-19 03:07:58,739] [INFO] [logging.py:68:log_dist] [Rank 0] step=3630, skipped=6, lr=[3.0600000000000003e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:07:58,740] [INFO] [timer.py:196:stop] epoch=0/micro_step=3630/global_step=3630, RunningAvgSamplesPerSec=17.60179920974838, CurrSamplesPerSec=17.7938144302562, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:10:48,446] [INFO] [logging.py:68:log_dist] [Rank 0] step=3640, skipped=6, lr=[3.037777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:10:48,447] [INFO] [timer.py:196:stop] epoch=0/micro_step=3640/global_step=3640, RunningAvgSamplesPerSec=17.6017878320369, CurrSamplesPerSec=17.83189928619381, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:13:33,096] [INFO] [logging.py:68:log_dist] [Rank 0] step=3650, skipped=6, lr=[3.015555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:13:33,098] [INFO] [timer.py:196:stop] epoch=0/micro_step=3650/global_step=3650, RunningAvgSamplesPerSec=17.601565935375557, CurrSamplesPerSec=17.83511830383586, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 3.015555555555556e-06, 'epoch': 31.01}
[2022-12-19 03:16:20,755] [INFO] [logging.py:68:log_dist] [Rank 0] step=3660, skipped=6, lr=[2.9933333333333336e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:16:20,756] [INFO] [timer.py:196:stop] epoch=0/micro_step=3660/global_step=3660, RunningAvgSamplesPerSec=17.601526408905308, CurrSamplesPerSec=17.55972566487229, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:19:08,899] [INFO] [logging.py:68:log_dist] [Rank 0] step=3670, skipped=6, lr=[2.9711111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:19:08,901] [INFO] [timer.py:196:stop] epoch=0/micro_step=3670/global_step=3670, RunningAvgSamplesPerSec=17.601866020071387, CurrSamplesPerSec=17.7394257348548, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.96e-06, 'epoch': 31.02}
[2022-12-19 03:22:05,005] [INFO] [logging.py:68:log_dist] [Rank 0] step=3680, skipped=6, lr=[2.948888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:22:05,006] [INFO] [timer.py:196:stop] epoch=0/micro_step=3680/global_step=3680, RunningAvgSamplesPerSec=17.602148084364135, CurrSamplesPerSec=17.71423246925165, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:25:01,391] [INFO] [logging.py:68:log_dist] [Rank 0] step=3690, skipped=6, lr=[2.9266666666666673e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:25:01,392] [INFO] [timer.py:196:stop] epoch=0/micro_step=3690/global_step=3690, RunningAvgSamplesPerSec=17.602265105171185, CurrSamplesPerSec=17.839620005142503, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:27:58,636] [INFO] [logging.py:68:log_dist] [Rank 0] step=3700, skipped=6, lr=[2.904444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:27:58,637] [INFO] [timer.py:196:stop] epoch=0/micro_step=3700/global_step=3700, RunningAvgSamplesPerSec=17.602024317328592, CurrSamplesPerSec=17.13545189933907, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.904444444444445e-06, 'epoch': 31.02}
[2022-12-19 03:29:28,476] [INFO] [logging.py:68:log_dist] [Rank 0] step=3710, skipped=6, lr=[2.8822222222222225e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:29:28,478] [INFO] [timer.py:196:stop] epoch=0/micro_step=3710/global_step=3710, RunningAvgSamplesPerSec=17.60249366241964, CurrSamplesPerSec=17.683989065307227, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:33:40,636] [INFO] [logging.py:68:log_dist] [Rank 0] step=3720, skipped=6, lr=[2.86e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:33:40,637] [INFO] [timer.py:196:stop] epoch=0/micro_step=3720/global_step=3720, RunningAvgSamplesPerSec=17.603601659594958, CurrSamplesPerSec=17.493026261358214, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.8488888888888894e-06, 'epoch': 32.0}
[2022-12-19 03:36:44,989] [INFO] [logging.py:68:log_dist] [Rank 0] step=3730, skipped=6, lr=[2.837777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:36:44,990] [INFO] [timer.py:196:stop] epoch=0/micro_step=3730/global_step=3730, RunningAvgSamplesPerSec=17.604015035004593, CurrSamplesPerSec=17.839234699560244, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:39:39,022] [INFO] [logging.py:68:log_dist] [Rank 0] step=3740, skipped=6, lr=[2.815555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:39:39,024] [INFO] [timer.py:196:stop] epoch=0/micro_step=3740/global_step=3740, RunningAvgSamplesPerSec=17.60423435088669, CurrSamplesPerSec=17.872316502218464, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:42:36,477] [INFO] [logging.py:68:log_dist] [Rank 0] step=3750, skipped=6, lr=[2.7933333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:42:36,479] [INFO] [timer.py:196:stop] epoch=0/micro_step=3750/global_step=3750, RunningAvgSamplesPerSec=17.60421286134035, CurrSamplesPerSec=17.6653445073596, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.7933333333333334e-06, 'epoch': 32.01}
[2022-12-19 03:47:44,486] [INFO] [logging.py:68:log_dist] [Rank 0] step=3760, skipped=6, lr=[2.771111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:47:44,487] [INFO] [timer.py:196:stop] epoch=0/micro_step=3760/global_step=3760, RunningAvgSamplesPerSec=17.60421484650127, CurrSamplesPerSec=17.76200482461907, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:50:36,617] [INFO] [logging.py:68:log_dist] [Rank 0] step=3770, skipped=6, lr=[2.748888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:50:36,619] [INFO] [timer.py:196:stop] epoch=0/micro_step=3770/global_step=3770, RunningAvgSamplesPerSec=17.603986155075713, CurrSamplesPerSec=17.835418109567822, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.7377777777777783e-06, 'epoch': 32.01}
[2022-12-19 03:53:34,204] [INFO] [logging.py:68:log_dist] [Rank 0] step=3780, skipped=6, lr=[2.726666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:53:34,206] [INFO] [timer.py:196:stop] epoch=0/micro_step=3780/global_step=3780, RunningAvgSamplesPerSec=17.604210162616486, CurrSamplesPerSec=17.369442334662356, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:56:31,105] [INFO] [logging.py:68:log_dist] [Rank 0] step=3790, skipped=6, lr=[2.7044444444444447e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:56:31,106] [INFO] [timer.py:196:stop] epoch=0/micro_step=3790/global_step=3790, RunningAvgSamplesPerSec=17.60431255398128, CurrSamplesPerSec=17.567898813891517, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 03:59:23,749] [INFO] [logging.py:68:log_dist] [Rank 0] step=3800, skipped=6, lr=[2.6822222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-19 03:59:23,750] [INFO] [timer.py:196:stop] epoch=0/micro_step=3800/global_step=3800, RunningAvgSamplesPerSec=17.60431104909582, CurrSamplesPerSec=17.70104341253724, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.6822222222222223e-06, 'epoch': 32.02}
[2022-12-19 04:02:19,616] [INFO] [logging.py:68:log_dist] [Rank 0] step=3810, skipped=6, lr=[2.6600000000000004e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:02:19,617] [INFO] [timer.py:196:stop] epoch=0/micro_step=3810/global_step=3810, RunningAvgSamplesPerSec=17.604126410600333, CurrSamplesPerSec=17.600326048215376, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:05:13,803] [INFO] [logging.py:68:log_dist] [Rank 0] step=3820, skipped=6, lr=[2.637777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:05:13,805] [INFO] [timer.py:196:stop] epoch=0/micro_step=3820/global_step=3820, RunningAvgSamplesPerSec=17.60442226776567, CurrSamplesPerSec=17.521556925430364, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.6266666666666668e-06, 'epoch': 32.02}
[2022-12-19 04:08:01,470] [INFO] [logging.py:68:log_dist] [Rank 0] step=3830, skipped=6, lr=[2.6155555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:08:01,472] [INFO] [timer.py:196:stop] epoch=0/micro_step=3830/global_step=3830, RunningAvgSamplesPerSec=17.605964489961178, CurrSamplesPerSec=17.66188200789337, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:10:57,833] [INFO] [logging.py:68:log_dist] [Rank 0] step=3840, skipped=6, lr=[2.5933333333333336e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:10:57,835] [INFO] [timer.py:196:stop] epoch=0/micro_step=3840/global_step=3840, RunningAvgSamplesPerSec=17.60612382198526, CurrSamplesPerSec=17.728355572744416, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:13:53,027] [INFO] [logging.py:68:log_dist] [Rank 0] step=3850, skipped=6, lr=[2.5711111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:13:53,028] [INFO] [timer.py:196:stop] epoch=0/micro_step=3850/global_step=3850, RunningAvgSamplesPerSec=17.606310071771578, CurrSamplesPerSec=17.81534865653578, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.5711111111111112e-06, 'epoch': 33.0}
[2022-12-19 04:16:48,115] [INFO] [logging.py:68:log_dist] [Rank 0] step=3860, skipped=6, lr=[2.5488888888888893e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:16:48,116] [INFO] [timer.py:196:stop] epoch=0/micro_step=3860/global_step=3860, RunningAvgSamplesPerSec=17.60637394058431, CurrSamplesPerSec=17.27665271608909, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:19:44,284] [INFO] [logging.py:68:log_dist] [Rank 0] step=3870, skipped=6, lr=[2.526666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:19:44,286] [INFO] [timer.py:196:stop] epoch=0/micro_step=3870/global_step=3870, RunningAvgSamplesPerSec=17.606653162319823, CurrSamplesPerSec=17.38970420612655, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.5155555555555557e-06, 'epoch': 33.01}
[2022-12-19 04:22:40,524] [INFO] [logging.py:68:log_dist] [Rank 0] step=3880, skipped=6, lr=[2.504444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:22:40,525] [INFO] [timer.py:196:stop] epoch=0/micro_step=3880/global_step=3880, RunningAvgSamplesPerSec=17.606719113033467, CurrSamplesPerSec=17.757985076265896, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:25:38,514] [INFO] [logging.py:68:log_dist] [Rank 0] step=3890, skipped=6, lr=[2.4822222222222225e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:25:38,516] [INFO] [timer.py:196:stop] epoch=0/micro_step=3890/global_step=3890, RunningAvgSamplesPerSec=17.60689324457354, CurrSamplesPerSec=17.80211727021623, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:28:36,061] [INFO] [logging.py:68:log_dist] [Rank 0] step=3900, skipped=6, lr=[2.46e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:28:36,063] [INFO] [timer.py:196:stop] epoch=0/micro_step=3900/global_step=3900, RunningAvgSamplesPerSec=17.607112140420526, CurrSamplesPerSec=17.612398120420085, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.46e-06, 'epoch': 33.01}
[2022-12-19 04:31:31,396] [INFO] [logging.py:68:log_dist] [Rank 0] step=3910, skipped=6, lr=[2.437777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:31:31,398] [INFO] [timer.py:196:stop] epoch=0/micro_step=3910/global_step=3910, RunningAvgSamplesPerSec=17.607239439571316, CurrSamplesPerSec=17.700039648390714, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:34:29,911] [INFO] [logging.py:68:log_dist] [Rank 0] step=3920, skipped=6, lr=[2.415555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:34:29,912] [INFO] [timer.py:196:stop] epoch=0/micro_step=3920/global_step=3920, RunningAvgSamplesPerSec=17.60749252768418, CurrSamplesPerSec=17.79277653010592, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.4044444444444446e-06, 'epoch': 33.02}
[2022-12-19 04:37:31,484] [INFO] [logging.py:68:log_dist] [Rank 0] step=3930, skipped=6, lr=[2.3933333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:37:31,485] [INFO] [timer.py:196:stop] epoch=0/micro_step=3930/global_step=3930, RunningAvgSamplesPerSec=17.607728287544596, CurrSamplesPerSec=17.50504171864284, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:39:30,333] [INFO] [logging.py:68:log_dist] [Rank 0] step=3940, skipped=6, lr=[2.371111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:39:30,335] [INFO] [timer.py:196:stop] epoch=0/micro_step=3940/global_step=3940, RunningAvgSamplesPerSec=17.60803244226165, CurrSamplesPerSec=17.897954595339627, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:43:16,008] [INFO] [logging.py:68:log_dist] [Rank 0] step=3950, skipped=6, lr=[2.348888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:43:16,010] [INFO] [timer.py:196:stop] epoch=0/micro_step=3950/global_step=3950, RunningAvgSamplesPerSec=17.609317404387298, CurrSamplesPerSec=17.338718477525717, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.348888888888889e-06, 'epoch': 34.0}
[2022-12-19 04:46:17,542] [INFO] [logging.py:68:log_dist] [Rank 0] step=3960, skipped=6, lr=[2.3266666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:46:17,543] [INFO] [timer.py:196:stop] epoch=0/micro_step=3960/global_step=3960, RunningAvgSamplesPerSec=17.609609085156343, CurrSamplesPerSec=17.51949968202784, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:49:16,425] [INFO] [logging.py:68:log_dist] [Rank 0] step=3970, skipped=6, lr=[2.3044444444444447e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:49:16,426] [INFO] [timer.py:196:stop] epoch=0/micro_step=3970/global_step=3970, RunningAvgSamplesPerSec=17.609706216869625, CurrSamplesPerSec=17.619688175337636, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.2933333333333335e-06, 'epoch': 34.01}
[2022-12-19 04:52:13,918] [INFO] [logging.py:68:log_dist] [Rank 0] step=3980, skipped=6, lr=[2.2822222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:52:13,919] [INFO] [timer.py:196:stop] epoch=0/micro_step=3980/global_step=3980, RunningAvgSamplesPerSec=17.60986454647832, CurrSamplesPerSec=17.457695134212692, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:55:16,822] [INFO] [logging.py:68:log_dist] [Rank 0] step=3990, skipped=6, lr=[2.2600000000000004e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:55:16,825] [INFO] [timer.py:196:stop] epoch=0/micro_step=3990/global_step=3990, RunningAvgSamplesPerSec=17.610149442844655, CurrSamplesPerSec=17.861005041208294, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 04:58:14,067] [INFO] [logging.py:68:log_dist] [Rank 0] step=4000, skipped=6, lr=[2.237777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 04:58:14,068] [INFO] [timer.py:196:stop] epoch=0/micro_step=4000/global_step=4000, RunningAvgSamplesPerSec=17.61035876229219, CurrSamplesPerSec=17.883168119082438, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.237777777777778e-06, 'epoch': 34.01}
{'eval_loss': 0.338134765625, 'eval_wer': 17.7977496284764, 'eval_runtime': 1277.681, 'eval_samples_per_second': 3.02, 'eval_steps_per_second': 0.095, 'epoch': 34.01}
[2022-12-19 05:19:32,876] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step4000 is begin to save!
[2022-12-19 05:19:32,885] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: ./checkpoint-4000/global_step4000/mp_rank_00_model_states.pt
[2022-12-19 05:19:32,885] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving ./checkpoint-4000/global_step4000/mp_rank_00_model_states.pt...
[2022-12-19 05:19:33,987] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved ./checkpoint-4000/global_step4000/mp_rank_00_model_states.pt.
[2022-12-19 05:19:33,988] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving ./checkpoint-4000/global_step4000/zero_pp_rank_0_mp_rank_00_optim_states.pt...
[2022-12-19 05:19:38,827] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved ./checkpoint-4000/global_step4000/zero_pp_rank_0_mp_rank_00_optim_states.pt.
[2022-12-19 05:19:38,828] [INFO] [engine.py:3394:_save_zero_checkpoint] zero checkpoint saved ./checkpoint-4000/global_step4000/zero_pp_rank_0_mp_rank_00_optim_states.pt
[2022-12-19 05:19:38,828] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now!
[2022-12-19 05:22:56,912] [INFO] [stage_1_and_2.py:1767:step] [deepspeed] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 65536.0, reducing to 65536.0
[2022-12-19 05:23:13,569] [INFO] [stage_1_and_2.py:1767:step] [deepspeed] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 65536.0, reducing to 32768.0
[2022-12-19 05:23:46,881] [INFO] [logging.py:68:log_dist] [Rank 0] step=4010, skipped=8, lr=[2.2200000000000003e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:23:46,883] [INFO] [timer.py:196:stop] epoch=0/micro_step=4010/global_step=4010, RunningAvgSamplesPerSec=17.611133683583187, CurrSamplesPerSec=17.462168466721625, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 05:26:38,748] [INFO] [logging.py:68:log_dist] [Rank 0] step=4020, skipped=8, lr=[2.197777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:26:38,750] [INFO] [timer.py:196:stop] epoch=0/micro_step=4020/global_step=4020, RunningAvgSamplesPerSec=17.611368779669082, CurrSamplesPerSec=17.378475405939998, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.1866666666666668e-06, 'epoch': 34.02}
[2022-12-19 05:29:37,420] [INFO] [logging.py:68:log_dist] [Rank 0] step=4030, skipped=8, lr=[2.1755555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:29:37,422] [INFO] [timer.py:196:stop] epoch=0/micro_step=4030/global_step=4030, RunningAvgSamplesPerSec=17.61113977607008, CurrSamplesPerSec=17.737646360030492, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 05:32:38,897] [INFO] [logging.py:68:log_dist] [Rank 0] step=4040, skipped=8, lr=[2.153333333333333e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:32:38,898] [INFO] [timer.py:196:stop] epoch=0/micro_step=4040/global_step=4040, RunningAvgSamplesPerSec=17.610970462806403, CurrSamplesPerSec=17.397608183728696, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 05:35:39,166] [INFO] [logging.py:68:log_dist] [Rank 0] step=4050, skipped=8, lr=[2.1311111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:35:39,168] [INFO] [timer.py:196:stop] epoch=0/micro_step=4050/global_step=4050, RunningAvgSamplesPerSec=17.611253470031247, CurrSamplesPerSec=17.833383657490664, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.1311111111111112e-06, 'epoch': 34.02}
[2022-12-19 05:36:43,776] [INFO] [logging.py:68:log_dist] [Rank 0] step=4060, skipped=8, lr=[2.108888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:36:43,777] [INFO] [timer.py:196:stop] epoch=0/micro_step=4060/global_step=4060, RunningAvgSamplesPerSec=17.612337722470095, CurrSamplesPerSec=23.447783642928687, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 05:41:45,291] [INFO] [logging.py:68:log_dist] [Rank 0] step=4070, skipped=8, lr=[2.086666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:41:45,292] [INFO] [timer.py:196:stop] epoch=0/micro_step=4070/global_step=4070, RunningAvgSamplesPerSec=17.612470816100952, CurrSamplesPerSec=17.877452530444263, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.0755555555555557e-06, 'epoch': 35.0}
[2022-12-19 05:44:43,364] [INFO] [logging.py:68:log_dist] [Rank 0] step=4080, skipped=8, lr=[2.064444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:44:43,365] [INFO] [timer.py:196:stop] epoch=0/micro_step=4080/global_step=4080, RunningAvgSamplesPerSec=17.612796189774304, CurrSamplesPerSec=17.736692349537687, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 05:47:41,092] [INFO] [logging.py:68:log_dist] [Rank 0] step=4090, skipped=8, lr=[2.0422222222222225e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:47:41,093] [INFO] [timer.py:196:stop] epoch=0/micro_step=4090/global_step=4090, RunningAvgSamplesPerSec=17.612917765334718, CurrSamplesPerSec=17.74363646795449, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 05:50:36,586] [INFO] [logging.py:68:log_dist] [Rank 0] step=4100, skipped=8, lr=[2.02e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:50:36,587] [INFO] [timer.py:196:stop] epoch=0/micro_step=4100/global_step=4100, RunningAvgSamplesPerSec=17.61301264236775, CurrSamplesPerSec=17.761625214894558, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.02e-06, 'epoch': 35.01}
[2022-12-19 05:53:34,916] [INFO] [logging.py:68:log_dist] [Rank 0] step=4110, skipped=8, lr=[1.9977777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:53:34,917] [INFO] [timer.py:196:stop] epoch=0/micro_step=4110/global_step=4110, RunningAvgSamplesPerSec=17.61307623650117, CurrSamplesPerSec=17.558381822907993, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 05:56:32,454] [INFO] [logging.py:68:log_dist] [Rank 0] step=4120, skipped=8, lr=[1.975555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:56:32,456] [INFO] [timer.py:196:stop] epoch=0/micro_step=4120/global_step=4120, RunningAvgSamplesPerSec=17.61328460598005, CurrSamplesPerSec=17.849423032435844, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.9644444444444446e-06, 'epoch': 35.01}
[2022-12-19 05:59:28,958] [INFO] [logging.py:68:log_dist] [Rank 0] step=4130, skipped=8, lr=[1.9533333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-19 05:59:28,959] [INFO] [timer.py:196:stop] epoch=0/micro_step=4130/global_step=4130, RunningAvgSamplesPerSec=17.61368146124737, CurrSamplesPerSec=17.80531019001806, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:02:37,403] [INFO] [logging.py:68:log_dist] [Rank 0] step=4140, skipped=8, lr=[1.9311111111111114e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:02:37,404] [INFO] [timer.py:196:stop] epoch=0/micro_step=4140/global_step=4140, RunningAvgSamplesPerSec=17.613849828726913, CurrSamplesPerSec=17.741122217014265, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:05:39,451] [INFO] [logging.py:68:log_dist] [Rank 0] step=4150, skipped=8, lr=[1.908888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:05:39,452] [INFO] [timer.py:196:stop] epoch=0/micro_step=4150/global_step=4150, RunningAvgSamplesPerSec=17.61420113756101, CurrSamplesPerSec=17.687389147202236, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.908888888888889e-06, 'epoch': 35.02}
[2022-12-19 06:08:34,312] [INFO] [logging.py:68:log_dist] [Rank 0] step=4160, skipped=8, lr=[1.8866666666666669e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:08:34,313] [INFO] [timer.py:196:stop] epoch=0/micro_step=4160/global_step=4160, RunningAvgSamplesPerSec=17.614407249990197, CurrSamplesPerSec=17.847925309782504, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:11:07,038] [INFO] [logging.py:68:log_dist] [Rank 0] step=4170, skipped=8, lr=[1.8644444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:11:07,039] [INFO] [timer.py:196:stop] epoch=0/micro_step=4170/global_step=4170, RunningAvgSamplesPerSec=17.614796295687643, CurrSamplesPerSec=17.6957562558827, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.8533333333333333e-06, 'epoch': 35.02}
[2022-12-19 06:14:26,808] [INFO] [logging.py:68:log_dist] [Rank 0] step=4180, skipped=8, lr=[1.8422222222222225e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:14:26,809] [INFO] [timer.py:196:stop] epoch=0/micro_step=4180/global_step=4180, RunningAvgSamplesPerSec=17.615978855666164, CurrSamplesPerSec=17.517680699168547, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:17:15,114] [INFO] [logging.py:68:log_dist] [Rank 0] step=4190, skipped=8, lr=[1.8200000000000002e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:17:15,116] [INFO] [timer.py:196:stop] epoch=0/micro_step=4190/global_step=4190, RunningAvgSamplesPerSec=17.616242542847015, CurrSamplesPerSec=17.71552661785773, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:20:05,755] [INFO] [logging.py:68:log_dist] [Rank 0] step=4200, skipped=8, lr=[1.797777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:20:05,756] [INFO] [timer.py:196:stop] epoch=0/micro_step=4200/global_step=4200, RunningAvgSamplesPerSec=17.61609844770078, CurrSamplesPerSec=17.366084738146398, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.797777777777778e-06, 'epoch': 36.0}
[2022-12-19 06:22:59,727] [INFO] [logging.py:68:log_dist] [Rank 0] step=4210, skipped=8, lr=[1.7755555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:22:59,728] [INFO] [timer.py:196:stop] epoch=0/micro_step=4210/global_step=4210, RunningAvgSamplesPerSec=17.616157473369498, CurrSamplesPerSec=17.633430354752342, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:25:55,318] [INFO] [logging.py:68:log_dist] [Rank 0] step=4220, skipped=8, lr=[1.7533333333333336e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:25:55,320] [INFO] [timer.py:196:stop] epoch=0/micro_step=4220/global_step=4220, RunningAvgSamplesPerSec=17.616198217116857, CurrSamplesPerSec=17.707843993918647, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.7422222222222224e-06, 'epoch': 36.01}
[2022-12-19 06:28:57,297] [INFO] [logging.py:68:log_dist] [Rank 0] step=4230, skipped=8, lr=[1.7311111111111112e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:28:57,298] [INFO] [timer.py:196:stop] epoch=0/micro_step=4230/global_step=4230, RunningAvgSamplesPerSec=17.61652951646244, CurrSamplesPerSec=17.71409803827759, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:31:59,467] [INFO] [logging.py:68:log_dist] [Rank 0] step=4240, skipped=8, lr=[1.708888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:31:59,468] [INFO] [timer.py:196:stop] epoch=0/micro_step=4240/global_step=4240, RunningAvgSamplesPerSec=17.617018924776456, CurrSamplesPerSec=17.79433342574028, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:34:52,886] [INFO] [logging.py:68:log_dist] [Rank 0] step=4250, skipped=8, lr=[1.6866666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:34:52,888] [INFO] [timer.py:196:stop] epoch=0/micro_step=4250/global_step=4250, RunningAvgSamplesPerSec=17.617133665736407, CurrSamplesPerSec=17.89488941065793, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.6866666666666667e-06, 'epoch': 36.01}
[2022-12-19 06:37:46,872] [INFO] [logging.py:68:log_dist] [Rank 0] step=4260, skipped=8, lr=[1.6644444444444447e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:37:46,874] [INFO] [timer.py:196:stop] epoch=0/micro_step=4260/global_step=4260, RunningAvgSamplesPerSec=17.61709970360881, CurrSamplesPerSec=17.72886490216432, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:40:46,545] [INFO] [logging.py:68:log_dist] [Rank 0] step=4270, skipped=8, lr=[1.6422222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:40:46,546] [INFO] [timer.py:196:stop] epoch=0/micro_step=4270/global_step=4270, RunningAvgSamplesPerSec=17.61723647083081, CurrSamplesPerSec=17.50934176766855, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.6311111111111114e-06, 'epoch': 36.02}
[2022-12-19 06:43:41,739] [INFO] [logging.py:68:log_dist] [Rank 0] step=4280, skipped=8, lr=[1.6200000000000002e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:43:41,741] [INFO] [timer.py:196:stop] epoch=0/micro_step=4280/global_step=4280, RunningAvgSamplesPerSec=17.617458572115385, CurrSamplesPerSec=17.8233958672181, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:45:13,558] [INFO] [logging.py:68:log_dist] [Rank 0] step=4290, skipped=8, lr=[1.5977777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:45:13,560] [INFO] [timer.py:196:stop] epoch=0/micro_step=4290/global_step=4290, RunningAvgSamplesPerSec=17.61748707733689, CurrSamplesPerSec=17.694014788106895, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:49:29,081] [INFO] [logging.py:68:log_dist] [Rank 0] step=4300, skipped=8, lr=[1.5755555555555558e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:49:29,083] [INFO] [timer.py:196:stop] epoch=0/micro_step=4300/global_step=4300, RunningAvgSamplesPerSec=17.618675777869846, CurrSamplesPerSec=17.604989397739672, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.5755555555555558e-06, 'epoch': 37.0}
[2022-12-19 06:52:25,502] [INFO] [logging.py:68:log_dist] [Rank 0] step=4310, skipped=8, lr=[1.5533333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:52:25,503] [INFO] [timer.py:196:stop] epoch=0/micro_step=4310/global_step=4310, RunningAvgSamplesPerSec=17.618917658378976, CurrSamplesPerSec=17.888580989149183, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 06:55:21,753] [INFO] [logging.py:68:log_dist] [Rank 0] step=4320, skipped=8, lr=[1.5311111111111113e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:55:21,754] [INFO] [timer.py:196:stop] epoch=0/micro_step=4320/global_step=4320, RunningAvgSamplesPerSec=17.619314539730464, CurrSamplesPerSec=17.772010979056823, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.52e-06, 'epoch': 37.01}
[2022-12-19 06:58:18,142] [INFO] [logging.py:68:log_dist] [Rank 0] step=4330, skipped=8, lr=[1.5088888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 06:58:18,144] [INFO] [timer.py:196:stop] epoch=0/micro_step=4330/global_step=4330, RunningAvgSamplesPerSec=17.619612314250787, CurrSamplesPerSec=17.79300532950759, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:01:11,886] [INFO] [logging.py:68:log_dist] [Rank 0] step=4340, skipped=8, lr=[1.486666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:01:11,888] [INFO] [timer.py:196:stop] epoch=0/micro_step=4340/global_step=4340, RunningAvgSamplesPerSec=17.619731554882748, CurrSamplesPerSec=17.831246620133083, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:04:10,104] [INFO] [logging.py:68:log_dist] [Rank 0] step=4350, skipped=8, lr=[1.4644444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:04:10,105] [INFO] [timer.py:196:stop] epoch=0/micro_step=4350/global_step=4350, RunningAvgSamplesPerSec=17.619778835395326, CurrSamplesPerSec=17.918863768337292, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.4644444444444445e-06, 'epoch': 37.01}
[2022-12-19 07:07:18,279] [INFO] [logging.py:68:log_dist] [Rank 0] step=4360, skipped=8, lr=[1.4422222222222223e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:07:18,280] [INFO] [timer.py:196:stop] epoch=0/micro_step=4360/global_step=4360, RunningAvgSamplesPerSec=17.61996490248731, CurrSamplesPerSec=17.26081994397919, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:10:24,446] [INFO] [logging.py:68:log_dist] [Rank 0] step=4370, skipped=8, lr=[1.42e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:10:24,448] [INFO] [timer.py:196:stop] epoch=0/micro_step=4370/global_step=4370, RunningAvgSamplesPerSec=17.62001193317469, CurrSamplesPerSec=17.401952624968608, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.4088888888888892e-06, 'epoch': 37.02}
[2022-12-19 07:13:17,536] [INFO] [logging.py:68:log_dist] [Rank 0] step=4380, skipped=8, lr=[1.397777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:13:17,537] [INFO] [timer.py:196:stop] epoch=0/micro_step=4380/global_step=4380, RunningAvgSamplesPerSec=17.619935746572583, CurrSamplesPerSec=17.670992723198474, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:16:09,586] [INFO] [logging.py:68:log_dist] [Rank 0] step=4390, skipped=8, lr=[1.3755555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:16:09,588] [INFO] [timer.py:196:stop] epoch=0/micro_step=4390/global_step=4390, RunningAvgSamplesPerSec=17.62012413538634, CurrSamplesPerSec=17.81988415153239, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:19:00,421] [INFO] [logging.py:68:log_dist] [Rank 0] step=4400, skipped=8, lr=[1.3533333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:19:00,423] [INFO] [timer.py:196:stop] epoch=0/micro_step=4400/global_step=4400, RunningAvgSamplesPerSec=17.620059714852356, CurrSamplesPerSec=17.51520378570747, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.3533333333333334e-06, 'epoch': 37.02}
[2022-12-19 07:21:47,475] [INFO] [logging.py:68:log_dist] [Rank 0] step=4410, skipped=8, lr=[1.3311111111111113e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:21:47,477] [INFO] [timer.py:196:stop] epoch=0/micro_step=4410/global_step=4410, RunningAvgSamplesPerSec=17.621140582831963, CurrSamplesPerSec=17.79351248317827, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:24:37,405] [INFO] [logging.py:68:log_dist] [Rank 0] step=4420, skipped=8, lr=[1.308888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:24:37,407] [INFO] [timer.py:196:stop] epoch=0/micro_step=4420/global_step=4420, RunningAvgSamplesPerSec=17.621243226652567, CurrSamplesPerSec=17.81804129124176, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.2977777777777779e-06, 'epoch': 38.0}
[2022-12-19 07:27:30,942] [INFO] [logging.py:68:log_dist] [Rank 0] step=4430, skipped=8, lr=[1.286666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:27:30,943] [INFO] [timer.py:196:stop] epoch=0/micro_step=4430/global_step=4430, RunningAvgSamplesPerSec=17.62120393290399, CurrSamplesPerSec=17.173806418163036, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:30:23,275] [INFO] [logging.py:68:log_dist] [Rank 0] step=4440, skipped=8, lr=[1.2644444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:30:23,277] [INFO] [timer.py:196:stop] epoch=0/micro_step=4440/global_step=4440, RunningAvgSamplesPerSec=17.621216006629336, CurrSamplesPerSec=17.66344978593792, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:33:14,028] [INFO] [logging.py:68:log_dist] [Rank 0] step=4450, skipped=8, lr=[1.2422222222222224e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:33:14,029] [INFO] [timer.py:196:stop] epoch=0/micro_step=4450/global_step=4450, RunningAvgSamplesPerSec=17.621087033427646, CurrSamplesPerSec=17.209054639288738, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.2422222222222224e-06, 'epoch': 38.01}
[2022-12-19 07:36:06,709] [INFO] [logging.py:68:log_dist] [Rank 0] step=4460, skipped=8, lr=[1.2200000000000002e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:36:06,710] [INFO] [timer.py:196:stop] epoch=0/micro_step=4460/global_step=4460, RunningAvgSamplesPerSec=17.620917138316685, CurrSamplesPerSec=17.78626293383821, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:39:01,972] [INFO] [logging.py:68:log_dist] [Rank 0] step=4470, skipped=8, lr=[1.1977777777777778e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:39:01,973] [INFO] [timer.py:196:stop] epoch=0/micro_step=4470/global_step=4470, RunningAvgSamplesPerSec=17.620983810960013, CurrSamplesPerSec=17.796430947697868, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.1866666666666668e-06, 'epoch': 38.01}
[2022-12-19 07:42:03,726] [INFO] [logging.py:68:log_dist] [Rank 0] step=4480, skipped=8, lr=[1.1755555555555556e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:42:03,728] [INFO] [timer.py:196:stop] epoch=0/micro_step=4480/global_step=4480, RunningAvgSamplesPerSec=17.62096470995377, CurrSamplesPerSec=17.768253674908266, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:44:54,699] [INFO] [logging.py:68:log_dist] [Rank 0] step=4490, skipped=8, lr=[1.1533333333333334e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:44:54,700] [INFO] [timer.py:196:stop] epoch=0/micro_step=4490/global_step=4490, RunningAvgSamplesPerSec=17.620810807024053, CurrSamplesPerSec=17.12529074539211, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:47:45,410] [INFO] [logging.py:68:log_dist] [Rank 0] step=4500, skipped=8, lr=[1.131111111111111e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:47:45,411] [INFO] [timer.py:196:stop] epoch=0/micro_step=4500/global_step=4500, RunningAvgSamplesPerSec=17.62074008642063, CurrSamplesPerSec=17.679443294912847, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.131111111111111e-06, 'epoch': 38.02}
[2022-12-19 07:50:35,915] [INFO] [logging.py:68:log_dist] [Rank 0] step=4510, skipped=8, lr=[1.1088888888888889e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:50:35,917] [INFO] [timer.py:196:stop] epoch=0/micro_step=4510/global_step=4510, RunningAvgSamplesPerSec=17.620564657864747, CurrSamplesPerSec=17.742456651398978, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:52:31,359] [INFO] [logging.py:68:log_dist] [Rank 0] step=4520, skipped=8, lr=[1.0866666666666667e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:52:31,361] [INFO] [timer.py:196:stop] epoch=0/micro_step=4520/global_step=4520, RunningAvgSamplesPerSec=17.62064058371775, CurrSamplesPerSec=17.814565970324615, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.0755555555555557e-06, 'epoch': 39.0}
[2022-12-19 07:56:08,715] [INFO] [logging.py:68:log_dist] [Rank 0] step=4530, skipped=8, lr=[1.0644444444444445e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:56:08,716] [INFO] [timer.py:196:stop] epoch=0/micro_step=4530/global_step=4530, RunningAvgSamplesPerSec=17.62153945771267, CurrSamplesPerSec=17.538607510040734, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 07:58:58,490] [INFO] [logging.py:68:log_dist] [Rank 0] step=4540, skipped=8, lr=[1.0422222222222221e-06], mom=[[0.9, 0.999]]
[2022-12-19 07:58:58,492] [INFO] [timer.py:196:stop] epoch=0/micro_step=4540/global_step=4540, RunningAvgSamplesPerSec=17.621602807771676, CurrSamplesPerSec=17.587700090174195, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:01:48,409] [INFO] [logging.py:68:log_dist] [Rank 0] step=4550, skipped=8, lr=[1.02e-06], mom=[[0.9, 0.999]]
[2022-12-19 08:01:48,411] [INFO] [timer.py:196:stop] epoch=0/micro_step=4550/global_step=4550, RunningAvgSamplesPerSec=17.62183130814374, CurrSamplesPerSec=17.59054796755746, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.02e-06, 'epoch': 39.01}
[2022-12-19 08:04:41,762] [INFO] [logging.py:68:log_dist] [Rank 0] step=4560, skipped=8, lr=[9.97777777777778e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:04:41,763] [INFO] [timer.py:196:stop] epoch=0/micro_step=4560/global_step=4560, RunningAvgSamplesPerSec=17.621998562878264, CurrSamplesPerSec=17.616155687347188, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:07:38,563] [INFO] [logging.py:68:log_dist] [Rank 0] step=4570, skipped=8, lr=[9.755555555555556e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:07:38,564] [INFO] [timer.py:196:stop] epoch=0/micro_step=4570/global_step=4570, RunningAvgSamplesPerSec=17.621957179552908, CurrSamplesPerSec=17.34689564215522, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 9.644444444444444e-07, 'epoch': 39.01}
[2022-12-19 08:10:35,179] [INFO] [logging.py:68:log_dist] [Rank 0] step=4580, skipped=8, lr=[9.533333333333335e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:10:35,180] [INFO] [timer.py:196:stop] epoch=0/micro_step=4580/global_step=4580, RunningAvgSamplesPerSec=17.622211087890516, CurrSamplesPerSec=17.696736200907107, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:13:29,992] [INFO] [logging.py:68:log_dist] [Rank 0] step=4590, skipped=8, lr=[9.311111111111113e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:13:29,993] [INFO] [timer.py:196:stop] epoch=0/micro_step=4590/global_step=4590, RunningAvgSamplesPerSec=17.62224562573623, CurrSamplesPerSec=17.885243737876667, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:16:25,912] [INFO] [logging.py:68:log_dist] [Rank 0] step=4600, skipped=8, lr=[9.08888888888889e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:16:25,914] [INFO] [timer.py:196:stop] epoch=0/micro_step=4600/global_step=4600, RunningAvgSamplesPerSec=17.62219766153009, CurrSamplesPerSec=17.437179457948137, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 9.08888888888889e-07, 'epoch': 39.02}
[2022-12-19 08:19:22,584] [INFO] [logging.py:68:log_dist] [Rank 0] step=4610, skipped=8, lr=[8.866666666666668e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:19:22,585] [INFO] [timer.py:196:stop] epoch=0/micro_step=4610/global_step=4610, RunningAvgSamplesPerSec=17.622381934268187, CurrSamplesPerSec=17.858060608665216, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:22:18,644] [INFO] [logging.py:68:log_dist] [Rank 0] step=4620, skipped=8, lr=[8.644444444444445e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:22:18,645] [INFO] [timer.py:196:stop] epoch=0/micro_step=4620/global_step=4620, RunningAvgSamplesPerSec=17.622457478635866, CurrSamplesPerSec=17.62142545329462, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 8.533333333333334e-07, 'epoch': 39.02}
[2022-12-19 08:25:19,257] [INFO] [logging.py:68:log_dist] [Rank 0] step=4630, skipped=8, lr=[8.422222222222224e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:25:19,259] [INFO] [timer.py:196:stop] epoch=0/micro_step=4630/global_step=4630, RunningAvgSamplesPerSec=17.62274117333296, CurrSamplesPerSec=17.796734173722378, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:26:22,516] [INFO] [logging.py:68:log_dist] [Rank 0] step=4640, skipped=8, lr=[8.200000000000001e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:26:22,517] [INFO] [timer.py:196:stop] epoch=0/micro_step=4640/global_step=4640, RunningAvgSamplesPerSec=17.624083237549367, CurrSamplesPerSec=23.52461524938626, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:31:08,975] [INFO] [logging.py:68:log_dist] [Rank 0] step=4650, skipped=8, lr=[7.977777777777779e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:31:08,977] [INFO] [timer.py:196:stop] epoch=0/micro_step=4650/global_step=4650, RunningAvgSamplesPerSec=17.624288170024037, CurrSamplesPerSec=17.88550113865089, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 7.977777777777779e-07, 'epoch': 40.0}
[2022-12-19 08:34:15,146] [INFO] [logging.py:68:log_dist] [Rank 0] step=4660, skipped=8, lr=[7.755555555555556e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:34:15,147] [INFO] [timer.py:196:stop] epoch=0/micro_step=4660/global_step=4660, RunningAvgSamplesPerSec=17.62444471234681, CurrSamplesPerSec=17.69476125609619, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:37:03,411] [INFO] [logging.py:68:log_dist] [Rank 0] step=4670, skipped=8, lr=[7.533333333333335e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:37:03,413] [INFO] [timer.py:196:stop] epoch=0/micro_step=4670/global_step=4670, RunningAvgSamplesPerSec=17.62448466929025, CurrSamplesPerSec=17.483528705282534, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 7.422222222222223e-07, 'epoch': 40.01}
[2022-12-19 08:39:55,845] [INFO] [logging.py:68:log_dist] [Rank 0] step=4680, skipped=8, lr=[7.311111111111112e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:39:55,846] [INFO] [timer.py:196:stop] epoch=0/micro_step=4680/global_step=4680, RunningAvgSamplesPerSec=17.620159783831323, CurrSamplesPerSec=17.60117311667079, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:42:42,018] [INFO] [logging.py:68:log_dist] [Rank 0] step=4690, skipped=8, lr=[7.08888888888889e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:42:42,019] [INFO] [timer.py:196:stop] epoch=0/micro_step=4690/global_step=4690, RunningAvgSamplesPerSec=17.620279729725365, CurrSamplesPerSec=17.93593334380126, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:45:27,450] [INFO] [logging.py:68:log_dist] [Rank 0] step=4700, skipped=8, lr=[6.866666666666667e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:45:27,452] [INFO] [timer.py:196:stop] epoch=0/micro_step=4700/global_step=4700, RunningAvgSamplesPerSec=17.620331377429462, CurrSamplesPerSec=17.754938283281913, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 6.866666666666667e-07, 'epoch': 40.01}
[2022-12-19 08:48:13,020] [INFO] [logging.py:68:log_dist] [Rank 0] step=4710, skipped=8, lr=[6.644444444444446e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:48:13,022] [INFO] [timer.py:196:stop] epoch=0/micro_step=4710/global_step=4710, RunningAvgSamplesPerSec=17.620347520092913, CurrSamplesPerSec=17.775257844597622, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:50:59,205] [INFO] [logging.py:68:log_dist] [Rank 0] step=4720, skipped=8, lr=[6.422222222222223e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:50:59,207] [INFO] [timer.py:196:stop] epoch=0/micro_step=4720/global_step=4720, RunningAvgSamplesPerSec=17.62027305945099, CurrSamplesPerSec=17.454496273463086, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 6.311111111111112e-07, 'epoch': 40.02}
[2022-12-19 08:53:45,256] [INFO] [logging.py:68:log_dist] [Rank 0] step=4730, skipped=8, lr=[6.200000000000001e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:53:45,258] [INFO] [timer.py:196:stop] epoch=0/micro_step=4730/global_step=4730, RunningAvgSamplesPerSec=17.620074181001, CurrSamplesPerSec=17.436431911964664, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:56:31,906] [INFO] [logging.py:68:log_dist] [Rank 0] step=4740, skipped=8, lr=[5.977777777777778e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:56:31,908] [INFO] [timer.py:196:stop] epoch=0/micro_step=4740/global_step=4740, RunningAvgSamplesPerSec=17.620138380065285, CurrSamplesPerSec=17.718109635002346, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 08:58:49,423] [INFO] [logging.py:68:log_dist] [Rank 0] step=4750, skipped=8, lr=[5.755555555555555e-07], mom=[[0.9, 0.999]]
[2022-12-19 08:58:49,424] [INFO] [timer.py:196:stop] epoch=0/micro_step=4750/global_step=4750, RunningAvgSamplesPerSec=17.620258476386315, CurrSamplesPerSec=17.843816764218595, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 5.755555555555555e-07, 'epoch': 40.02}
[2022-12-19 09:02:02,128] [INFO] [logging.py:68:log_dist] [Rank 0] step=4760, skipped=8, lr=[5.533333333333334e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:02:02,129] [INFO] [timer.py:196:stop] epoch=0/micro_step=4760/global_step=4760, RunningAvgSamplesPerSec=17.620845234938383, CurrSamplesPerSec=17.56852084543164, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:04:49,552] [INFO] [logging.py:68:log_dist] [Rank 0] step=4770, skipped=8, lr=[5.311111111111111e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:04:49,553] [INFO] [timer.py:196:stop] epoch=0/micro_step=4770/global_step=4770, RunningAvgSamplesPerSec=17.620781538403822, CurrSamplesPerSec=17.993237206011308, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 5.2e-07, 'epoch': 41.0}
[2022-12-19 09:07:39,410] [INFO] [logging.py:68:log_dist] [Rank 0] step=4780, skipped=8, lr=[5.088888888888889e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:07:39,411] [INFO] [timer.py:196:stop] epoch=0/micro_step=4780/global_step=4780, RunningAvgSamplesPerSec=17.620741159717415, CurrSamplesPerSec=17.62931691282607, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:10:31,165] [INFO] [logging.py:68:log_dist] [Rank 0] step=4790, skipped=8, lr=[4.866666666666666e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:10:31,166] [INFO] [timer.py:196:stop] epoch=0/micro_step=4790/global_step=4790, RunningAvgSamplesPerSec=17.621122717656146, CurrSamplesPerSec=17.9595660728393, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:13:22,998] [INFO] [logging.py:68:log_dist] [Rank 0] step=4800, skipped=8, lr=[4.6444444444444446e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:13:23,000] [INFO] [timer.py:196:stop] epoch=0/micro_step=4800/global_step=4800, RunningAvgSamplesPerSec=17.62146184710049, CurrSamplesPerSec=17.68574021221271, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 4.6444444444444446e-07, 'epoch': 41.01}
[2022-12-19 09:16:20,425] [INFO] [logging.py:68:log_dist] [Rank 0] step=4810, skipped=8, lr=[4.422222222222223e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:16:20,426] [INFO] [timer.py:196:stop] epoch=0/micro_step=4810/global_step=4810, RunningAvgSamplesPerSec=17.62207009255359, CurrSamplesPerSec=17.919497743569675, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:19:17,092] [INFO] [logging.py:68:log_dist] [Rank 0] step=4820, skipped=8, lr=[4.2000000000000006e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:19:17,093] [INFO] [timer.py:196:stop] epoch=0/micro_step=4820/global_step=4820, RunningAvgSamplesPerSec=17.62281812490437, CurrSamplesPerSec=18.063956802967372, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 4.0888888888888897e-07, 'epoch': 41.01}
[2022-12-19 09:22:11,760] [INFO] [logging.py:68:log_dist] [Rank 0] step=4830, skipped=8, lr=[3.9777777777777783e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:22:11,762] [INFO] [timer.py:196:stop] epoch=0/micro_step=4830/global_step=4830, RunningAvgSamplesPerSec=17.623217566897882, CurrSamplesPerSec=17.682097330909283, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:25:06,256] [INFO] [logging.py:68:log_dist] [Rank 0] step=4840, skipped=8, lr=[3.755555555555556e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:25:06,258] [INFO] [timer.py:196:stop] epoch=0/micro_step=4840/global_step=4840, RunningAvgSamplesPerSec=17.6239823084841, CurrSamplesPerSec=17.937164202870864, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:28:01,375] [INFO] [logging.py:68:log_dist] [Rank 0] step=4850, skipped=8, lr=[3.533333333333334e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:28:01,376] [INFO] [timer.py:196:stop] epoch=0/micro_step=4850/global_step=4850, RunningAvgSamplesPerSec=17.624502174178428, CurrSamplesPerSec=17.980924250155837, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 3.533333333333334e-07, 'epoch': 41.02}
[2022-12-19 09:30:55,163] [INFO] [logging.py:68:log_dist] [Rank 0] step=4860, skipped=8, lr=[3.3111111111111115e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:30:55,164] [INFO] [timer.py:196:stop] epoch=0/micro_step=4860/global_step=4860, RunningAvgSamplesPerSec=17.62510883090857, CurrSamplesPerSec=17.68973547651362, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:32:25,739] [INFO] [logging.py:68:log_dist] [Rank 0] step=4870, skipped=8, lr=[3.088888888888889e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:32:25,740] [INFO] [timer.py:196:stop] epoch=0/micro_step=4870/global_step=4870, RunningAvgSamplesPerSec=17.6258822208131, CurrSamplesPerSec=18.0027099754836, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.977777777777778e-07, 'epoch': 42.0}
[2022-12-19 09:36:36,623] [INFO] [logging.py:68:log_dist] [Rank 0] step=4880, skipped=8, lr=[2.866666666666667e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:36:36,625] [INFO] [timer.py:196:stop] epoch=0/micro_step=4880/global_step=4880, RunningAvgSamplesPerSec=17.627590163724953, CurrSamplesPerSec=17.970300168807466, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:39:29,570] [INFO] [logging.py:68:log_dist] [Rank 0] step=4890, skipped=8, lr=[2.6444444444444447e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:39:29,571] [INFO] [timer.py:196:stop] epoch=0/micro_step=4890/global_step=4890, RunningAvgSamplesPerSec=17.628267118175295, CurrSamplesPerSec=17.953751119369517, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:42:20,232] [INFO] [logging.py:68:log_dist] [Rank 0] step=4900, skipped=8, lr=[2.4222222222222224e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:42:20,233] [INFO] [timer.py:196:stop] epoch=0/micro_step=4900/global_step=4900, RunningAvgSamplesPerSec=17.628906943630938, CurrSamplesPerSec=17.925920168070032, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2.4222222222222224e-07, 'epoch': 42.01}
[2022-12-19 09:45:10,151] [INFO] [logging.py:68:log_dist] [Rank 0] step=4910, skipped=8, lr=[2.2e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:45:10,152] [INFO] [timer.py:196:stop] epoch=0/micro_step=4910/global_step=4910, RunningAvgSamplesPerSec=17.629482874220002, CurrSamplesPerSec=17.81305045026094, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:47:55,909] [INFO] [logging.py:68:log_dist] [Rank 0] step=4920, skipped=8, lr=[1.9777777777777778e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:47:55,910] [INFO] [timer.py:196:stop] epoch=0/micro_step=4920/global_step=4920, RunningAvgSamplesPerSec=17.629964973280497, CurrSamplesPerSec=17.82848486728267, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.866666666666667e-07, 'epoch': 42.01}
[2022-12-19 09:50:45,106] [INFO] [logging.py:68:log_dist] [Rank 0] step=4930, skipped=8, lr=[1.7555555555555558e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:50:45,108] [INFO] [timer.py:196:stop] epoch=0/micro_step=4930/global_step=4930, RunningAvgSamplesPerSec=17.630395440620926, CurrSamplesPerSec=17.707026340076307, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:53:36,465] [INFO] [logging.py:68:log_dist] [Rank 0] step=4940, skipped=8, lr=[1.5333333333333333e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:53:36,466] [INFO] [timer.py:196:stop] epoch=0/micro_step=4940/global_step=4940, RunningAvgSamplesPerSec=17.631168427205882, CurrSamplesPerSec=18.003925866621273, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 09:56:31,916] [INFO] [logging.py:68:log_dist] [Rank 0] step=4950, skipped=8, lr=[1.3111111111111113e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:56:31,917] [INFO] [timer.py:196:stop] epoch=0/micro_step=4950/global_step=4950, RunningAvgSamplesPerSec=17.63176081163297, CurrSamplesPerSec=18.076746529204566, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 1.3111111111111113e-07, 'epoch': 42.02}
[2022-12-19 09:59:24,350] [INFO] [logging.py:68:log_dist] [Rank 0] step=4960, skipped=8, lr=[1.088888888888889e-07], mom=[[0.9, 0.999]]
[2022-12-19 09:59:24,352] [INFO] [timer.py:196:stop] epoch=0/micro_step=4960/global_step=4960, RunningAvgSamplesPerSec=17.632263561334792, CurrSamplesPerSec=17.887386586858412, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 10:02:21,261] [INFO] [logging.py:68:log_dist] [Rank 0] step=4970, skipped=8, lr=[8.666666666666668e-08], mom=[[0.9, 0.999]]
[2022-12-19 10:02:21,262] [INFO] [timer.py:196:stop] epoch=0/micro_step=4970/global_step=4970, RunningAvgSamplesPerSec=17.632836729122953, CurrSamplesPerSec=18.04230886853236, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 7.555555555555556e-08, 'epoch': 42.02}
[2022-12-19 10:05:14,832] [INFO] [logging.py:68:log_dist] [Rank 0] step=4980, skipped=8, lr=[6.444444444444445e-08], mom=[[0.9, 0.999]]
[2022-12-19 10:05:14,834] [INFO] [timer.py:196:stop] epoch=0/micro_step=4980/global_step=4980, RunningAvgSamplesPerSec=17.633470780183274, CurrSamplesPerSec=17.935921359635877, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 10:08:03,799] [INFO] [logging.py:68:log_dist] [Rank 0] step=4990, skipped=8, lr=[4.222222222222222e-08], mom=[[0.9, 0.999]]
[2022-12-19 10:08:03,800] [INFO] [timer.py:196:stop] epoch=0/micro_step=4990/global_step=4990, RunningAvgSamplesPerSec=17.635077496959436, CurrSamplesPerSec=18.20354202004553, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
[2022-12-19 10:10:57,893] [INFO] [logging.py:68:log_dist] [Rank 0] step=5000, skipped=8, lr=[2e-08], mom=[[0.9, 0.999]]
[2022-12-19 10:10:57,895] [INFO] [timer.py:196:stop] epoch=0/micro_step=5000/global_step=5000, RunningAvgSamplesPerSec=17.635593279087654, CurrSamplesPerSec=18.094030560717506, MemAllocated=0.53GB, MaxMemAllocated=17.47GB
{'loss': 0.0002, 'learning_rate': 2e-08, 'epoch': 43.0}
{'eval_loss': 0.3427734375, 'eval_wer': 17.804826268487723, 'eval_runtime': 1211.7822, 'eval_samples_per_second': 3.185, 'eval_steps_per_second': 0.1, 'epoch': 43.0}
[2022-12-19 10:31:10,745] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step5000 is begin to save!
[2022-12-19 10:31:10,753] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: ./checkpoint-5000/global_step5000/mp_rank_00_model_states.pt
[2022-12-19 10:31:10,753] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving ./checkpoint-5000/global_step5000/mp_rank_00_model_states.pt...
[2022-12-19 10:31:11,749] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved ./checkpoint-5000/global_step5000/mp_rank_00_model_states.pt.
[2022-12-19 10:31:11,750] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving ./checkpoint-5000/global_step5000/zero_pp_rank_0_mp_rank_00_optim_states.pt...
[2022-12-19 10:31:15,938] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved ./checkpoint-5000/global_step5000/zero_pp_rank_0_mp_rank_00_optim_states.pt.
[2022-12-19 10:31:15,939] [INFO] [engine.py:3394:_save_zero_checkpoint] zero checkpoint saved ./checkpoint-5000/global_step5000/zero_pp_rank_0_mp_rank_00_optim_states.pt
[2022-12-19 10:31:15,939] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now!
[2022-12-19 10:32:38,091] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed info: version=0.8.0+a25c31b6, git-hash=a25c31b6, git-branch=master
[2022-12-19 10:32:38,141] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed Flops Profiler Enabled: False
Adam Optimizer #1 is created with AVX2 arithmetic capability.
Config: alpha=0.000010, betas=(0.900000, 0.999000), weight_decay=0.000000, adam_w=1
[2022-12-19 10:32:40,431] [INFO] [logging.py:68:log_dist] [Rank 0] Using DeepSpeed Optimizer param name adamw as basic optimizer
[2022-12-19 10:32:40,476] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed Basic Optimizer = DeepSpeedCPUAdam
[2022-12-19 10:32:40,476] [INFO] [utils.py:52:is_zero_supported_optimizer] Checking ZeRO support for optimizer=DeepSpeedCPUAdam type=<class 'deepspeed.ops.adam.cpu_adam.DeepSpeedCPUAdam'>
[2022-12-19 10:32:40,476] [INFO] [logging.py:68:log_dist] [Rank 0] Creating fp16 ZeRO stage 2 optimizer
[2022-12-19 10:32:40,477] [INFO] [stage_1_and_2.py:141:__init__] Reduce bucket size 200000000
[2022-12-19 10:32:40,477] [INFO] [stage_1_and_2.py:142:__init__] Allgather bucket size 200000000
[2022-12-19 10:32:40,477] [INFO] [stage_1_and_2.py:143:__init__] CPU Offload: True
[2022-12-19 10:32:40,477] [INFO] [stage_1_and_2.py:144:__init__] Round robin gradient partitioning: False
Rank: 0 partition count [1] and sizes[(241734912, False)] 
[2022-12-19 10:32:42,063] [INFO] [utils.py:831:see_memory_usage] Before initializing optimizer states
[2022-12-19 10:32:42,065] [INFO] [utils.py:832:see_memory_usage] MA 1.05 GB         Max_MA 17.47 GB         CA 1.05 GB         Max_CA 22 GB 
[2022-12-19 10:32:42,065] [INFO] [utils.py:840:see_memory_usage] CPU Virtual Memory:  used = 192.83 GB, percent = 38.3%
[2022-12-19 10:32:43,144] [INFO] [utils.py:831:see_memory_usage] After initializing optimizer states
[2022-12-19 10:32:43,145] [INFO] [utils.py:832:see_memory_usage] MA 1.05 GB         Max_MA 1.05 GB         CA 1.05 GB         Max_CA 1 GB 
[2022-12-19 10:32:43,146] [INFO] [utils.py:840:see_memory_usage] CPU Virtual Memory:  used = 195.68 GB, percent = 38.8%
[2022-12-19 10:32:43,146] [INFO] [stage_1_and_2.py:527:__init__] optimizer state initialized
[2022-12-19 10:32:43,225] [INFO] [utils.py:831:see_memory_usage] After initializing ZeRO optimizer
[2022-12-19 10:32:43,226] [INFO] [utils.py:832:see_memory_usage] MA 1.05 GB         Max_MA 1.05 GB         CA 1.05 GB         Max_CA 1 GB 
[2022-12-19 10:32:43,226] [INFO] [utils.py:840:see_memory_usage] CPU Virtual Memory:  used = 195.68 GB, percent = 38.8%
[2022-12-19 10:32:43,250] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed Final Optimizer = adamw
[2022-12-19 10:32:43,250] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed using configured LR scheduler = WarmupDecayLR
[2022-12-19 10:32:43,251] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed LR Scheduler = <deepspeed.runtime.lr_schedules.WarmupDecayLR object at 0x7ff11fd26280>
[2022-12-19 10:32:43,251] [INFO] [logging.py:68:log_dist] [Rank 0] step=0, skipped=0, lr=[1e-05], mom=[[0.9, 0.999]]
[2022-12-19 10:32:43,252] [INFO] [config.py:1008:print] DeepSpeedEngine configuration:
[2022-12-19 10:32:43,252] [INFO] [config.py:1012:print]   activation_checkpointing_config  {
    "partition_activations": false, 
    "contiguous_memory_optimization": false, 
    "cpu_checkpointing": false, 
    "number_checkpoints": null, 
    "synchronize_checkpoint_boundary": false, 
    "profile": false
}
[2022-12-19 10:32:43,252] [INFO] [config.py:1012:print]   aio_config ................... {'block_size': 1048576, 'queue_depth': 8, 'thread_count': 1, 'single_submit': False, 'overlap_events': True}
[2022-12-19 10:32:43,252] [INFO] [config.py:1012:print]   amp_enabled .................. False
[2022-12-19 10:32:43,252] [INFO] [config.py:1012:print]   amp_params ................... False
[2022-12-19 10:32:43,253] [INFO] [config.py:1012: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
}
[2022-12-19 10:32:43,253] [INFO] [config.py:1012:print]   bfloat16_enabled ............. False
[2022-12-19 10:32:43,253] [INFO] [config.py:1012:print]   checkpoint_parallel_write_pipeline  False
[2022-12-19 10:32:43,253] [INFO] [config.py:1012:print]   checkpoint_tag_validation_enabled  True
[2022-12-19 10:32:43,253] [INFO] [config.py:1012:print]   checkpoint_tag_validation_fail  False
[2022-12-19 10:32:43,253] [INFO] [config.py:1012:print]   comms_config ................. <deepspeed.comm.config.DeepSpeedCommsConfig object at 0x7ff04cccfca0>
[2022-12-19 10:32:43,253] [INFO] [config.py:1012:print]   communication_data_type ...... None
[2022-12-19 10:32:43,253] [INFO] [config.py:1012: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}}
[2022-12-19 10:32:43,253] [INFO] [config.py:1012:print]   curriculum_enabled_legacy .... False
[2022-12-19 10:32:43,253] [INFO] [config.py:1012:print]   curriculum_params_legacy ..... False
[2022-12-19 10:32:43,253] [INFO] [config.py:1012: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}}}}
[2022-12-19 10:32:43,253] [INFO] [config.py:1012:print]   data_efficiency_enabled ...... False
[2022-12-19 10:32:43,253] [INFO] [config.py:1012:print]   dataloader_drop_last ......... False
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   disable_allgather ............ False
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   dump_state ................... False
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   dynamic_loss_scale_args ...... {'init_scale': 65536, 'scale_window': 1000, 'delayed_shift': 2, 'min_scale': 1}
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   eigenvalue_enabled ........... False
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   eigenvalue_gas_boundary_resolution  1
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   eigenvalue_layer_name ........ bert.encoder.layer
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   eigenvalue_layer_num ......... 0
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   eigenvalue_max_iter .......... 100
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   eigenvalue_stability ......... 1e-06
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   eigenvalue_tol ............... 0.01
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   eigenvalue_verbose ........... False
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   elasticity_enabled ........... False
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   flops_profiler_config ........ {
    "enabled": false, 
    "profile_step": 1, 
    "module_depth": -1, 
    "top_modules": 1, 
    "detailed": true, 
    "output_file": null
}
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   fp16_auto_cast ............... False
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   fp16_enabled ................. True
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   fp16_master_weights_and_gradients  False
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   global_rank .................. 0
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   grad_accum_dtype ............. None
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   gradient_accumulation_steps .. 1
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   gradient_clipping ............ 1.0
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   gradient_predivide_factor .... 1.0
[2022-12-19 10:32:43,254] [INFO] [config.py:1012:print]   initial_dynamic_scale ........ 65536
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   load_universal_checkpoint .... False
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   loss_scale ................... 0
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   memory_breakdown ............. False
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   monitor_config ............... <deepspeed.monitor.config.DeepSpeedMonitorConfig object at 0x7ff10e7c26a0>
[2022-12-19 10:32:43,255] [INFO] [config.py:1012: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
}
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   optimizer_legacy_fusion ...... False
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   optimizer_name ............... adamw
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   optimizer_params ............. {'lr': 1e-05, 'betas': [0.9, 0.999], 'eps': 1e-08, 'weight_decay': 0.0}
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   pipeline ..................... {'stages': 'auto', 'partition': 'best', 'seed_layers': False, 'activation_checkpoint_interval': 0}
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   pld_enabled .................. False
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   pld_params ................... False
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   prescale_gradients ........... False
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   scheduler_name ............... WarmupDecayLR
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   scheduler_params ............. {'last_batch_iteration': -1, 'total_num_steps': 5000, 'warmup_min_lr': 0, 'warmup_max_lr': 1e-05, 'warmup_num_steps': 500}
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   sparse_attention ............. None
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   sparse_gradients_enabled ..... False
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   steps_per_print .............. 10
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   train_batch_size ............. 64
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   train_micro_batch_size_per_gpu  64
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   use_node_local_storage ....... False
[2022-12-19 10:32:43,255] [INFO] [config.py:1012:print]   wall_clock_breakdown ......... False
[2022-12-19 10:32:43,256] [INFO] [config.py:1012:print]   world_size ................... 1
[2022-12-19 10:32:43,256] [INFO] [config.py:1012:print]   zero_allow_untested_optimizer  False
[2022-12-19 10:32:43,256] [INFO] [config.py:1012:print]   zero_config .................. stage=2 contiguous_gradients=True reduce_scatter=True reduce_bucket_size=200000000 allgather_partitions=True allgather_bucket_size=200000000 overlap_comm=True load_from_fp32_weights=True elastic_checkpoint=False offload_param=None 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) sub_group_size=1,000,000,000 cpu_offload_param=None cpu_offload_use_pin_memory=None cpu_offload=None prefetch_bucket_size=50,000,000 param_persistence_threshold=100,000 model_persistence_threshold=sys.maxsize max_live_parameters=1,000,000,000 max_reuse_distance=1,000,000,000 gather_16bit_weights_on_model_save=False stage3_gather_fp16_weights_on_model_save=False ignore_unused_parameters=True legacy_stage1=False round_robin_gradients=False
[2022-12-19 10:32:43,256] [INFO] [config.py:1012:print]   zero_enabled ................. True
[2022-12-19 10:32:43,256] [INFO] [config.py:1012:print]   zero_optimization_stage ...... 2
[2022-12-19 10:32:43,256] [INFO] [config.py:997:print_user_config]   json = {
    "fp16": {
        "enabled": true, 
        "loss_scale": 0, 
        "loss_scale_window": 1000, 
        "initial_scale_power": 16, 
        "hysteresis": 2, 
        "min_loss_scale": 1
    }, 
    "optimizer": {
        "type": "AdamW", 
        "params": {
            "lr": 1e-05, 
            "betas": [0.9, 0.999], 
            "eps": 1e-08, 
            "weight_decay": 0.0
        }
    }, 
    "scheduler": {
        "type": "WarmupDecayLR", 
        "params": {
            "last_batch_iteration": -1, 
            "total_num_steps": 5.000000e+03, 
            "warmup_min_lr": 0, 
            "warmup_max_lr": 1e-05, 
            "warmup_num_steps": 500
        }
    }, 
    "zero_optimization": {
        "stage": 2, 
        "offload_optimizer": {
            "device": "cpu", 
            "pin_memory": true
        }, 
        "allgather_partitions": true, 
        "allgather_bucket_size": 2.000000e+08, 
        "overlap_comm": true, 
        "reduce_scatter": true, 
        "reduce_bucket_size": 2.000000e+08, 
        "contiguous_gradients": true
    }, 
    "gradient_accumulation_steps": 1, 
    "gradient_clipping": 1.0, 
    "train_batch_size": 64, 
    "train_micro_batch_size_per_gpu": 64
}
[2022-12-19 10:32:43,257] [INFO] [torch_checkpoint_engine.py:21:load] [Torch] Loading checkpoint from ./checkpoint-1000/global_step1000/mp_rank_00_model_states.pt...
[2022-12-19 10:32:44,286] [INFO] [torch_checkpoint_engine.py:23:load] [Torch] Loaded checkpoint from ./checkpoint-1000/global_step1000/mp_rank_00_model_states.pt.
[2022-12-19 10:32:44,325] [INFO] [torch_checkpoint_engine.py:21:load] [Torch] Loading checkpoint from ./checkpoint-1000/global_step1000/mp_rank_00_model_states.pt...
[2022-12-19 10:32:44,719] [INFO] [torch_checkpoint_engine.py:23:load] [Torch] Loaded checkpoint from ./checkpoint-1000/global_step1000/mp_rank_00_model_states.pt.
[2022-12-19 10:32:44,876] [INFO] [torch_checkpoint_engine.py:21:load] [Torch] Loading checkpoint from ./checkpoint-1000/global_step1000/zero_pp_rank_0_mp_rank_00_optim_states.pt...
[2022-12-19 10:32:51,167] [INFO] [torch_checkpoint_engine.py:23:load] [Torch] Loaded checkpoint from ./checkpoint-1000/global_step1000/zero_pp_rank_0_mp_rank_00_optim_states.pt.
[2022-12-19 10:32:51,167] [INFO] [engine.py:3020:_get_all_zero_checkpoint_state_dicts] successfully read 1 ZeRO state_dicts for rank 0
[2022-12-19 10:32:51,519] [INFO] [engine.py:2960:_load_zero_checkpoint] loading 1 zero partition checkpoints for rank 0
{'train_runtime': 93041.5511, 'train_samples_per_second': 3.439, 'train_steps_per_second': 0.054, 'train_loss': 0.01692909588813782, 'epoch': 43.0}
12/19/2022 10:34:29 - WARNING - huggingface_hub.repository - Several commits (2) will be pushed upstream.
12/19/2022 10:34:29 - WARNING - huggingface_hub.repository - The progress bars may be unreliable.
12/19/2022 10:34:38 - WARNING - huggingface_hub.repository - remote: Scanning LFS files for validity, may be slow...        
remote: LFS file scan complete.        
To https://huggingface.co/mikr/whisper-small-ro-cv11
   9101f9e..50c3b81  main -> main

12/19/2022 10:35:49 - WARNING - huggingface_hub.repository - To https://huggingface.co/mikr/whisper-small-ro-cv11
   50c3b81..dfadda3  main -> main

***** train metrics *****
  epoch                    =              43.0
  train_loss               =            0.0169
  train_runtime            = 1 day, 1:50:41.55
  train_samples_per_second =             3.439
  train_steps_per_second   =             0.054
12/19/2022 10:35:53 - INFO - __main__ - *** Evaluate ***
***** eval metrics *****
  epoch                   =       43.0
  eval_loss               =     0.2808
  eval_runtime            = 0:20:39.78
  eval_samples_per_second =      3.113
  eval_steps_per_second   =      0.098
  eval_wer                =    17.5713
12/19/2022 10:57:50 - WARNING - huggingface_hub.repository - remote: Scanning LFS files for validity, may be slow...        
remote: LFS file scan complete.        
To https://huggingface.co/mikr/whisper-small-ro-cv11
   dfadda3..0d47cec  main -> main