End of training
1553098
verified
-
attn_projector=mlp, per_device_train_batch_size=16, run_name=baseline
Training in progress, step 5000
-
attn_projector=mlp, per_device_train_batch_size=2, run_name=bs2
End of training
-
attn_projector=mlp, per_device_train_batch_size=2, run_name=bs2_liger, student_model_use_liger=True
Training in progress, step 5000
-
attn_projector=mlp, per_device_train_batch_size=4, run_name=bs4
Training in progress, step 5000
-
attn_projector=mlp, per_device_train_batch_size=8, run_name=bs8
Training in progress, step 5000
-
attn_weight=0.0, per_device_train_batch_size=4, run_name=bs4_NO_liger_baseline, student_model_use_liger=False
End of training
-
attn_weight=0.0, per_device_train_batch_size=4, run_name=bs4_NO_liger_baseline, student_model_use_liger=True
Training in progress, step 5000
-
attn_weight=0.0, per_device_train_batch_size=4, run_name=logits_only_bs4_liger, student_model_use_liger=True
Training in progress, step 5000
