Add skip_verbose_naming in add_hook to give an option for skipping the naming#635

Open

verlocks wants to merge 21 commits intoTransformerLensOrg:mainfrom verlocks:verlock-dev

+13,508-88

Commits on Jun 11, 2024

Add skip_verbose_naming in add_hook
verlocks
authored

Commits on Jul 5, 2024

Merge branch 'dev' into verlock-dev
bryce13950
committed

Commits on Oct 7, 2024

add transformer diagram
akozlo
committed

Commits on Oct 9, 2024

Commits on Oct 15, 2024

Add support for Mistral-Nemo-Base-2407 model (TransformerLensOrg#751 )

ryanhoangt
and
bryce13950
authored

Commits on Oct 17, 2024

added template for reporting model compatibility (TransformerLensOrg#759 )
bryce13950
authored

Commits on Oct 22, 2024

Add configurations for Llama 3.1 models(Llama-3.1-8B and Llama-3.1-70B) (TransformerLensOrg#761 )

vatsalrathod16
and
bryce13950
authored

Commits on Oct 25, 2024

added logit comparator (TransformerLensOrg#765 )

authored

Commits on Oct 26, 2024

Add support for NTK-by-Part Rotary Embedding & set correct rotary base for Llama-3.1 series (TransformerLensOrg#764 )

Hzfinfdu
and
bryce13950
authored

Commits on Nov 15, 2024

Commits on Nov 16, 2024

Fix that if use_past_kv_cache is set to True models from the Bloom family produce weird outputs. (TransformerLensOrg#777 )

authored

Commits on Nov 19, 2024

added typeguard dependency (TransformerLensOrg#786 )
bryce13950
authored

Commits on Nov 21, 2024

remove einsum in forward pass in AbstractAttention (TransformerLensOrg#783 )

authored

Commits on Nov 25, 2024