INDEX

Explanations

No Explanations Found

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Kahn

-0.79

ochem

-0.63

pas

-0.63

acons

-0.62

 Mats

-0.60

croft

-0.60

Buk

-0.59

prints

-0.59

igslist

-0.59

 Layout

-0.59

POSITIVE LOGITS

rities

0.78

ãĥ¼ãĥĨãĤ£

0.77

ãĤ¡

0.72

 fuss

0.71

EF

0.71

å°Ĩ

0.70

é¾

0.65

 Â·

0.65

ildo

0.64

 fool

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.