Pszemraj's workspace
Runs
11
Name
11 visualized
1-11
of 11Panel Section
1
Parameter importance with respect toeval/loss
eval/loss
1-10
of 23n_layer
n_layer
n_head
n_head
n_embd
n_embd
eval_batch_size
eval_batch_size
per_device_eval_batch_size
per_device_eval_batch_size
weight_decay
weight_decay
gradient_accumulation_steps
gradient_accumulation_steps
num_train_epochs
num_train_epochs
hidden_size
hidden_size
dataloader_num_workers
dataloader_num_workers
Loading...
eval
4
train
9