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

 mosqu

-0.73

 etiquette

-0.71

 boarding

-0.70

lyn

-0.68

wired

-0.67

aida

-0.67

 rook

-0.65

 therap

-0.65

abulary

-0.65

 textual

-0.65

POSITIVE LOGITS

 Petty

0.75

 Qiao

0.74

Bah

0.73

Fah

0.73

ICO

0.72

 Foster

0.70

UAE

0.69

 Chevron

0.69

Boh

0.69

 Grizz

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.