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

aff

-0.80

agging

-0.79

illy

-0.70

warts

-0.68

icking

-0.66

imes

-0.66

HP

-0.66

atism

-0.66

agraph

-0.65

Force

-0.64

POSITIVE LOGITS

EStream

0.79

eem

0.74

 UCHIJ

0.70

 indent

0.65

sonian

0.64

 partName

0.64

stein

0.63

bis

0.62

Sie

0.62

hene

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.