INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ga

-0.78

lev

-0.68

 regress

-0.68

 diarr

-0.65

 reuse

-0.64

esters

-0.64

 balloons

-0.63

 hairs

-0.62

 Crim

-0.62

 crime

-0.62

POSITIVE LOGITS

sonian

0.99

hetamine

0.82

assetsadobe

0.78

displayText

0.75

iterranean

0.75

VIDIA

0.73

pees

0.72

lyak

0.71

uties

0.71

illet

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.