INDEX

Explanations

pronominal references, especially related to gendered pronouns and their antecedents

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hom

-0.43

 arşivlendi

-0.42

num

-0.41

きれい

-0.41

ev

-0.39

ex

-0.39

 haya

-0.39

 paci

-0.39

Sco

-0.38

POSITIVE LOGITS

#+#

0.81

 >=",

0.80

 Signalez

0.78

 beginnetje

0.77

++];

0.76

tvguidetime

0.73

 Paglinawan

0.72

RTDA

0.72

 invokingState

0.71

%)$

0.71

Activations Density 0.267%

pronominal references, especially related to gendered pronouns and their antecedents

No Comments

No Known Activations