INDEX

Explanations

references to the company Warner Bros

oai_token-act-pair · gpt-3.5-turbo

mentions of the Warner Bros. company

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

xual

-0.79

ombo

-0.77

sembly

-0.76

oral

-0.76

obos

-0.75

omorphic

-0.72

DonaldTrump

-0.72

PsyNetMessage

-0.71

anwhile

-0.71

£ı

-0.71

POSITIVE LOGITS

 Bros

1.26

 Brothers

1.00

 Warner

0.92

 Cable

0.80

Film

0.78

riors

0.78

stown

0.77

sonian

0.73

ewater

0.72

izon

0.71

Activations Density 0.014%

references to the company Warner Bros

mentions of the Warner Bros. company

No Comments

No Known Activations