Hainingwang's workspace
Runs
1
Name
1 visualized
State
Notes
User
Tags
Created
Runtime
Sweep
_n_gpu
_name_or_path
adafactor
adam_beta1
adam_beta2
adam_epsilon
add_cross_attention
architectures
attention_probs_dropout_prob
bf16
bf16_full_eval
chunk_size_feed_forward
dataloader_drop_last
dataloader_num_workers
dataloader_pin_memory
ddp_bucket_cap_mb
ddp_find_unused_parameters
debug
deepspeed
disable_tqdm
diversity_penalty
do_eval
do_predict
do_sample
do_train
early_stopping
encoder_no_repeat_ngram_size
eval_accumulation_steps
eval_batch_size
eval_steps
evaluation_strategy
fp16
fp16_backend
fp16_full_eval
fp16_opt_level
gradient_accumulation_steps
gradient_checkpointing
greater_is_better
group_by_length
half_precision_backend
hidden_act
hidden_dropout_prob
hidden_size
hub_model_id
Finished
-
hainingwang
4h 5m 53s
-
4
uer/chinese_roberta_L-12_H-768
false
0.99
0.9999
1.0000e-8
false
["BertForMaskedLM"]
0.1
false
false
0
false
4
true
None
None
[]
None
false
0
true
false
false
false
false
0
None
24
None
epoch
false
auto
false
O1
1
false
false
false
auto
gelu
0.1
768
None
1-1
of 1