INDEX

Explanations

phrases related to men's issues or movements

oai_token-act-pair · gpt-3.5-turbo

references to gender-related topics, specifically focusing on women's issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SIGN

-0.73

OVER

-0.66

 snail

-0.63

ļéĨĴ

-0.60

 Leilan

-0.58

SET

-0.57

 commons

-0.56

 Heist

-0.55

EStream

-0.55

 Presidents

-0.54

POSITIVE LOGITS

atisf

1.02

kaya

0.93

outhern

0.91

lightly

0.90

pecially

0.90

wered

0.89

atellite

0.89

bestos

0.89

ween

0.87

omew

0.87

Activations Density 0.190%

phrases related to men's issues or movements

references to gender-related topics, specifically focusing on women's issues

No Comments

No Known Activations