eleutherai

Preetham-gali's group workspace

Group: hSY45YakxtAmpHucKxDCZz_19ewbxaz

1-1

of 1

Tags

Notes

Author

stellaathena

State

Finished

Start time

August 17th, 2021 7:56:55 AM

Runtime

3h 52m 58s

Tracked hours

3h 52m 40s

Run path

eleutherai/distilling/1tikf8xm

Linux-5.8.0-59-generic-x86_64-with-glibc2.29

Python version

3.8.5

Git repository

git clone https://github.com/EleutherAI/gpt-neox.git

Git state

git checkout -b "shiv-0-0" ecc3c2327d2fd053cc67c2cc656e1d847db2f797

Command

pretrain_gpt2.py --local_rank=0 --num_gpus 4 --deepspeed_config "{\"train_batch_size\": 540, \"train_micro_batch_size_per_gpu\": 5, \"gradient_accumulation_steps\": 27, \"optimizer\": {\"type\": \"Adam\", \"params\": {\"lr\": 0.0003, \"betas\": [0.9, 0.999], \"eps\": 1e-08}}, \"fp16\": {\"fp16\": true, \"enabled\": true, \"loss_scale\": 0, \"loss_scale_window\": 1000, \"hysteresis\": 2, \"min_loss_scale\": 1}, \"gradient_clipping\": 1.0, \"zero_optimization\": {\"stage\": 1, \"allgather_partitions\": true, \"allgather_bucket_size\": 500000000, \"overlap_comm\": true, \"reduce_scatter\": true, \"reduce_bucket_size\": 500000000, \"contiguous_gradients\": true, \"cpu_offload\": false}, \"wall_clock_breakdown\": true}" --megatron_config "{\"num_gpus\": 4, \"train_batch_size\": 540, \"train_micro_batch_size_per_gpu\": 5, \"gradient_accumulation_steps\": 27, \"optimizer\": {\"type\": \"Adam\", \"params\": {\"lr\": 0.0003, \"betas\": [0.9, 0.999], \"eps\": 1e-08}}, \"fp16\": {\"fp16\": true, \"enabled\": true, \"loss_scale\": 0, \"loss_scale_window\": 1000, \"hysteresis\": 2, \"min_loss_scale\": 1}, \"gradient_clipping\": 1.0, \"zero_optimization\": {\"stage\": 1, \"allgather_partitions\": true, \"allgather_bucket_size\": 500000000, \"overlap_comm\": true, \"reduce_scatter\": true, \"reduce_bucket_size\": 500000000, \"contiguous_gradients\": true, \"cpu_offload\": false}, \"wall_clock_breakdown\": true, \"precision\": \"fp16\", \"num_layers\": 24, \"hidden_size\": 1024, \"num_attention_heads\": 8, \"seq_length\": 2048, \"max_position_embeddings\": 2048, \"pos_emb\": \"rotary\", \"no_weight_tying\": true, \"attention_config\": [\"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\", \"global\"], \"sparsity_config\": {}, \"lr_decay_style\": \"cosine\", \"lr_decay_iters\": 320000, \"optimizer_type\": \"Adam\", \"zero_stage\": 1, \"zero_reduce_scatter\": true, \"zero_contiguous_gradients\": true, \"zero_reduce_bucket_size\": 500000000, \"zero_allgather_bucket_size\": 500000000, \"lr\": 0.0003, \"data_path\": \"/mnt/ssd-1/data/pile/pile_text_document\", \"data_impl\": \"mmap\", \"save\": \"checkpoints/medium-training/1-1-0\", \"save_interval\": 10000, \"finetune\": true, \"batch_size\": 5, \"train_iters\": 320000, \"eval_iters\": 10, \"keep_last_n_checkpoints\": 4, \"split\": \"949,50,1\", \"vocab_file\": \"/mnt/ssd-1/data/gpt2-vocab.json\", \"merge_file\": \"/mnt/ssd-1/data/gpt2-merges.txt\", \"attention_dropout\": 0, \"hidden_dropout\": 0, \"weight_decay\": 0, \"checkpoint_activations\": true, \"synchronize_each_layer\": true, \"partition_activations\": true, \"gas\": 27, \"clip_grad\": 1.0, \"dynamic_loss_scale\": true, \"pipe_parallel_size\": 1, \"is_pipe_parallel\": true, \"use_wandb\": true, \"wandb_group\": \"hSY45YakxtAmpHucKxDCZz_19ewbxaz\", \"wandb_team\": \"eleutherai\", \"wandb_project\": \"distilling\", \"log_dir\": \"logs\", \"log_interval\": 100, \"user_script\": \"pretrain_gpt2.py\"}"

System Hardware

CPU count	128
GPU count	4
GPU type	NVIDIA A100-SXM4-40GB

W&B CLI Version

0.10.28

Group

hSY45YakxtAmpHucKxDCZz_19ewbxaz

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 186 keys
- activation:
  "gelu"
- adlr_autoresume:
  false
- adlr_autoresume_interval:
  1,000
- alpha_kld:
  0
- alpha_lm:
  0
- alpha_mse:
  0
- amp:
  null
- apply_query_key_layer_scaling:
  false
- apply_residual_connection_post_layernorm:
  false
- ▶
  attention_config:[] 24 items
- attention_dropout:
  0
- attention_softmax_in_fp32:
  false
- batch_size:
  5
- bias_dropout_fusion:
  false
- bias_gelu_fusion:
  false
- checkpoint_activations:
  true
- checkpoint_in_cpu:
  false
- checkpoint_num_layers:
  1
- checkpoint_validation_with_forward_pass:
  false
- clip_grad:
  1
- contiguous_checkpointing:
  false
- data_impl:
  "mmap"
- data_path:
  "/mnt/ssd-1/data/pile/pile_text_document"
- deepscale:
  false
- deepscale_config:
  null
- deepspeed:
  true
- deepspeed_activation_checkpointing:
  true
- deepspeed_mpi:
  false
- detect_nvlink_pairs:
  false
- distributed_backend:
  "nccl"
- do_distillation:
  false
- do_test:
  null
- do_train:
  null
- do_valid:
  null
- dump_state:
  false
- dynamic_loss_scale:
  true
- eod_mask_loss:
  false
- eval_interval:
  1,000
- eval_iters:
  10
- eval_results_prefix:
  ""
- eval_tasks:
  null
- exclude:
  null
- exit_interval:
  null
- finetune:
  true
- flops_profiler:
  null
- ▶
  fp16:{} 6 keys
- ▶
  zero_optimization:{} 8 keys
- zero_reduce_bucket_size:
  500,000,000
- zero_reduce_scatter:
  true
- zero_stage:
  1

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 29 keys
- runtime/flops_per_sec_per_gpu:
  109,629,937,622,472.44
- runtime/iteration_time:
  13.749365980625152
- runtime/samples_per_sec:
  39.27453824132242
- timers/_step_check_overflow:
  0.5195140838623047
- timers/_step_clipping:
  0.1118183135986328
- timers/_step_step:
  404.874324798584
- timers/_step_zero_grad:
  0.8051395416259766
- timers/backward:
  102,997.76983261108
- timers/backward_allreduce:
  4.068851470947266
- timers/backward_allreduce_microstep:
  11.81316375732422
- timers/backward_inner:
  102,967.50450134276
- timers/backward_inner_microstep:
  102,975.14224052428
- timers/backward_microstep:
  103,004.50539588928
- timers/batch_input:
  280.1547050476074
- timers/comms:
  359.0736389160156
- timers/forward:
  29,366.528511047363
- timers/pct_backward:
  74.8878157428386
- timers/pct_comms:
  0.2610759490516577
- timers/pct_fwd:
  21.351871790196995
- timers/pct_optimizer_step:
  0.2958599806630684
- timers/reduce_grads:
  307.31940269470215
- timers/reduce_tied_grads:
  0.4107952117919922
- timers/step:
  406.91423416137695
- timers/train_batch:
  0
- train/learning_rate:
  0.0000928125
- train/loss:
  3.6650736331939697
- train/loss_scale:
  32,768
- validation/loss:
  3.665481567382813
- validation/loss_ppl:
  39.07494885379973

This run produced these artifacts as outputs. Total: 1. Learn more

wandb-history

run-1tikf8xm-history:v0