INDEX

Explanations

mentions of nuclear weapons or related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of nuclear weapons

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lus

-0.80

juries

-0.77

tem

-0.77

é¾įåĸļå£«

-0.76

ItemThumbnailImage

-0.76

DEM

-0.73

MAL

-0.73

 Proud

-0.72

é»Ĵ

-0.71

LOC

-0.70

POSITIVE LOGITS

 submarines

0.98

 deton

0.92

 dismant

0.92

 warheads

0.89

 stockpile

0.88

 installations

0.87

 proliferation

0.87

 submarine

0.87

 dictatorship

0.86

eers

0.86

Activations Density 0.016%

mentions of nuclear weapons or related terms

mentions of nuclear weapons

No Comments

No Known Activations