eleutherai

Chilli's group workspace

Group: BBtbK5KrG6G23TABQGZXM5eUaezasHviLSJrcy5hiUMp

1-1

of 1

Tags

Notes

Author

shivanshupurohit

State

Crashed

Start time

April 27th, 2021 6:18:22 PM

Runtime

11m 2s

Tracked hours

10m 52s

Run path

eleutherai/neox/1kfcju34

Linux-5.4.0-54-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 "neox-visual-grounding-0-0" 9992042ab113428022e5e91421c04917577b8e00

Command

pretrain_gpt2.py --local_rank=0 --num_gpus 6 --deepspeed_config "{\"train_batch_size\": 96, \"train_micro_batch_size_per_gpu\": 16, \"optimizer\": {\"type\": \"Adam\", \"params\": {\"lr\": 0.00025, \"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\": 6, \"train_batch_size\": 96, \"train_micro_batch_size_per_gpu\": 16, \"optimizer\": {\"type\": \"Adam\", \"params\": {\"lr\": 0.00025, \"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\": 1536, \"num_attention_heads\": 16, \"seq_length\": 2048, \"max_position_embeddings\": 2048, \"pos_emb\": \"rotary\", \"no_weight_tying\": true, \"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\": 0.00025, \"data_path\": \"/mnt/ssd-cluster/data/enron/enron_text_document\", \"data_impl\": \"mmap\", \"save\": \"/mnt/ssd-cluster/checkpoints\", \"load\": \"/mnt/ssd-cluster/checkpoints\", \"save_interval\": 10000, \"batch_size\": 16, \"train_iters\": 320000, \"eval_iters\": 10, \"keep_last_n_checkpoints\": 4, \"split\": \"900,99,1\", \"vocab_file\": \"/mnt/ssd-cluster/data/gpt2-vocab.json\", \"merge_file\": \"/mnt/ssd-cluster/data/gpt2-merges.txt\", \"attention_dropout\": 0, \"hidden_dropout\": 0, \"weight_decay\": 0, \"checkpoint_activations\": true, \"synchronize_each_layer\": true, \"partition_activations\": true, \"gas\": 1, \"clip_grad\": 1.0, \"dynamic_loss_scale\": true, \"pipe_parallel_size\": 1, \"world_size\": 1, \"wandb_group\": \"BBtbK5KrG6G23TABQGZXM5\", \"log_dir\": \"/mnt/ssd-cluster/logs\", \"tensorboard_dir\": \"/mnt/ssd-cluster/tensorboard\", \"log_interval\": 100, \"local_rank\": 0, \"rank\": 0, \"user_script\": \"pretrain_gpt2.py\"}"

System Hardware

CPU count	112
GPU count	6
GPU type	A100-PCIE-40GB

W&B CLI Version

0.10.25

Group

BBtbK5KrG6G23TABQGZXM5eUaezasHviLSJrcy5hiUMp

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 162 keys
- adlr_autoresume:
  false
- adlr_autoresume_interval:
  1,000
- amp:
  null
- apply_query_key_layer_scaling:
  false
- apply_residual_connection_post_layernorm:
  false
- attention_dropout:
  0
- attention_softmax_in_fp32:
  false
- batch_size:
  16
- bias_dropout_fusion:
  false
- bias_gelu_fusion:
  false
- checkpoint_activations:
  true
- checkpoint_in_cpu:
  false
- checkpoint_num_layers:
  1
- clip_grad:
  1
- contiguous_checkpointing:
  false
- data_impl:
  "mmap"
- data_path:
  "/mnt/ssd-cluster/data/enron/enron_text_document"
- deepscale:
  false
- deepscale_config:
  null
- deepspeed:
  true
- deepspeed_activation_checkpointing:
  false
- deepspeed_mpi:
  false
- detect_nvlink_pairs:
  false
- distribute_checkpointed_activations:
  false
- distributed_backend:
  "nccl"
- 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
- exclude:
  null
- exit_interval:
  null
- finetune:
  false
- flops_profiler:
  null
- ▶
  fp16:{} 6 keys
- fp16_lm_cross_entropy:
  false
- fp32_allreduce:
  false
- gas:
  1
- geglu:
  false
- genfile:
  null
- git_hash:
  "9992042"
- gradient_accumulation_steps:
  1
- gradient_clipping:
  1
- ▶
  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:{} 3 keys
- train/learning_rate:
  0.00000390625
- train/lm_loss:
  9.11441421508789
- train/loss_scale:
  16,384

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

wandb-history

run-1kfcju34-history:v0