INDEX

Explanations

phrases or sentences indicating the beginning or initiation of something

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the beginning of events or actions

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

illard

-0.87

itsch

-0.76

aths

-0.75

ukong

-0.71

itte

-0.69

mented

-0.68

aud

-0.64

hold

-0.64

ugs

-0.64

abytes

-0.64

POSITIVE LOGITS

 anew

0.94

 raining

0.73

here

0.69

 airing

0.69

 BELOW

0.68

 dusk

0.68

Ń·

0.66

OPLE

0.65

 leaking

0.65

nings

0.64

Activations Density 0.066%

phrases or sentences indicating the beginning or initiation of something

phrases indicating the beginning of events or actions

No Comments

No Known Activations