INDEX

Explanations

No Explanations Found

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

awan

-0.65

legal

-0.64

cra

-0.63

 fetish

-0.63

loo

-0.61

 ring

-0.61

cano

-0.60

 Monkey

-0.60

 Clicker

-0.60

wow

-0.59

POSITIVE LOGITS

ļéĨĴ

0.82

externalActionCode

0.78

etsk

0.76

earchers

0.73

hemer

0.73

RTX

0.72

ministic

0.70

utenberg

0.69

 Uran

0.69

iHUD

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.