INDEX

Explanations

references to clowns, particularly in relation to controversial or mocking contexts

oai_token-act-pair · gpt-3.5-turbo

references to clowns and associated themes

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

ECD

-0.75

WARD

-0.74

ASED

-0.68

dated

-0.66

equality

-0.65

neau

-0.65

galitarian

-0.65

EMP

-0.64

CE

-0.63

RA

-0.62

POSITIVE LOGITS

 Clown

1.03

fish

0.96

 clown

0.93

amn

0.84

ey

0.81

erey

0.79

obyl

0.78

tail

0.77

opard

0.77

 mascot

0.76

Activations Density 0.015%

references to clowns, particularly in relation to controversial or mocking contexts

references to clowns and associated themes

No Comments

No Known Activations