INDEX

Explanations

descriptions related to cartoons

oai_token-act-pair · gpt-3.5-turbo

references to cartoons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

forces

-0.71

alez

-0.71

govern

-0.69

acia

-0.66

utherford

-0.66

FUL

-0.65

CI

-0.64

forced

-0.64

ttp

-0.62

vae

-0.62

POSITIVE LOGITS

 cartoons

1.18

ishly

1.06

 cartoon

1.00

 caric

0.95

 frog

0.94

 Cartoon

0.89

 caricature

0.86

 sketches

0.86

 Hebdo

0.85

eers

0.83

Activations Density 0.012%

descriptions related to cartoons

references to cartoons

No Comments

No Known Activations