distily
/

distily_test_attn_linear

Generated from Trainer

Model card Files Files and versions

Metrics Training metrics Community

distily_test_attn_linear / logs

296 MB

1 contributor

History: 9 commits

lapp0's picture

Training in progress, step 61875

9664f42 verified over 1 year ago

attn_layer_mapper=all, attn_loss_fn=cos, attn_projector=linear
Training in progress, step 61875 over 1 year ago
attn_layer_mapper=all, attn_loss_fn=raw_mse, attn_projector=linear
Training in progress, step 61875 over 1 year ago
attn_layer_mapper=last, attn_loss_fn=cos, attn_projector=linear
Training in progress, step 61875 over 1 year ago
attn_layer_mapper=last, attn_loss_fn=raw_mse, attn_projector=linear
Training in progress, step 61875 over 1 year ago
attn_layer_mapper=last_k_2, attn_loss_fn=cos, attn_projector=linear
Training in progress, step 61875 over 1 year ago
attn_layer_mapper=last_k_2, attn_loss_fn=raw_mse, attn_projector=linear
Training in progress, step 61875 over 1 year ago
attn_layer_mapper=layer-2, attn_loss_fn=cos, attn_projector=linear
Training in progress, step 61875 over 1 year ago
attn_layer_mapper=layer-2, attn_loss_fn=raw_mse, attn_projector=linear
End of training over 1 year ago