INDEX

Explanations

references to gender identity and related topics like restrooms, facilities, and locker rooms

oai_token-act-pair · gpt-3.5-turbo

references to gender identity and related policies

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

 thunder

-0.73

Byr

-0.69

uncture

-0.67

Exc

-0.67

 Winc

-0.66

baugh

-0.66

 Prices

-0.65

Money

-0.65

UF

-0.64

Ley

-0.63

POSITIVE LOGITS

iannopoulos

0.88

 pronouns

0.84

 disorder

0.79

aki

0.78

 identity

0.75

 dysph

0.74

 transitioned

0.73

 Identity

0.72

 Disorders

0.71

ference

0.70

Activations Density 0.091%

references to gender identity and related topics like restrooms, facilities, and locker rooms

references to gender identity and related policies

No Comments

No Known Activations