INDEX

Explanations

instances of the word "women"

oai_token-act-pair · gpt-3.5-turbo

references to women and their roles or experiences in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 constitu

-0.79

 heterogeneity

-0.68

arta

-0.66

ologically

-0.64

 Keynes

-0.64

 statist

-0.63

illin

-0.63

tics

-0.62

kson

-0.62

HOME

-0.61

POSITIVE LOGITS

volent

0.96

wagen

0.85

uthor

0.74

istries

0.74

pher

0.71

ibly

0.70

ager

0.68

geoning

0.68

 IMAGES

0.67

SUV

0.64

Activations Density 0.128%

instances of the word "women"

references to women and their roles or experiences in various contexts

No Comments

No Known Activations