INDEX

Explanations

terms related to gender

oai_token-act-pair · gpt-3.5-turbo

mentions of gender and related disparities or 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

BLIC

-0.76

 Warrant

-0.72

 GOODMAN

-0.69

 Gerr

-0.69

 Showtime

-0.68

steen

-0.68

 Emanuel

-0.66

 Brees

-0.66

amina

-0.66

 Parish

-0.65

POSITIVE LOGITS

 dysph

1.11

 pronouns

0.89

bender

0.89

 equality

0.89

endered

0.86

 Equality

0.86

fuck

0.84

 genders

0.84

puter

0.81

 imbalance

0.81

Activations Density 0.016%

terms related to gender

mentions of gender and related disparities or issues

No Comments

No Known Activations