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

 Pinball

-0.92

onite

-0.79

atan

-0.79

onga

-0.75

mill

-0.74

ersion

-0.74

hattan

-0.73

ellen

-0.73

hest

-0.72

xit

-0.72

POSITIVE LOGITS

 fools

0.70

 Synd

0.69

×Ļ×

0.66

CW

0.65

âĺ

0.61

åĳ

0.61

 shin

0.60

 Canter

0.60

 overlook

0.60

Uz

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.