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

 Pistol

-0.71

 UNCLASSIFIED

-0.69

peria

-0.68

 bluff

-0.68

 Arrow

-0.67

Pod

-0.67

 bracelet

-0.66

 pistol

-0.65

phrine

-0.64

draft

-0.64

POSITIVE LOGITS

enium

0.82

endez

0.81

ando

0.79

iddles

0.77

inus

0.72

ictionary

0.72

ented

0.71

iscovery

0.71

ĸļ

0.70

urations

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.