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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

doms

-0.88

cules

-0.83

cia

-0.80

ruits

-0.78

jug

-0.77

eday

-0.76

dates

-0.76

acia

-0.73

isons

-0.72

etheless

-0.72

POSITIVE LOGITS

 CLSID

0.72

wcsstore

0.68

CAT

0.66

 unfor

0.64

 overheard

0.64

 paraly

0.62

 bother

0.61

laughter

0.61

BUS

0.59

 worse

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.