INDEX

Explanations

references to lenses or visual perception

oai_token-act-pair · gpt-3.5-turbo

references to lenses in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

avez

-0.75

BY

-0.75

lly

-0.72

 Chatt

-0.70

anqu

-0.69

tes

-0.66

icted

-0.64

TY

-0.63

alky

-0.63

swick

-0.63

POSITIVE LOGITS

 lens

1.21

 Lens

1.14

 lenses

1.13

 flares

1.02

Lens

0.97

 flare

0.90

 Flare

0.88

linger

0.85

Cra

0.84

edIn

0.78

Activations Density 0.010%

references to lenses or visual perception

references to lenses in various contexts

No Comments

No Known Activations