INDEX

Explanations

instances where the word "helps" is mentioned

oai_token-act-pair · gpt-3.5-turbo

phrases indicating assistance or benefits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Chern

-0.72

gran

-0.66

chal

-0.65

iche

-0.65

 Mortal

-0.64

 Phant

-0.64

 Viet

-0.63

War

-0.63

 Lifetime

-0.63

parts

-0.63

POSITIVE LOGITS

heet

1.05

soDeliveryDate

0.89

hift

0.88

olulu

0.83

 facilitate

0.82

erest

0.81

ettings

0.76

odcast

0.75

":[{"

0.74

 helping

0.74

Activations Density 0.007%

instances where the word "helps" is mentioned

phrases indicating assistance or benefits

No Comments

No Known Activations