End of training
fca8285
verified
-
gradient_accumulation_steps=2, per_device_eval_batch_size=8, per_device_train_batch_size=4, run_name=baseline
Training in progress, step 5000
-
gradient_accumulation_steps=2, per_device_train_batch_size=4, run_name=baseline
End of training
-
per_device_eval_batch_size=8, per_device_train_batch_size=8, run_name=baseline
Training in progress, step 5000
-
per_device_train_batch_size=4, run_name=baseline
Training in progress, step 40000