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

BOOK

-0.69

flower

-0.66

 Prophe

-0.66

 forecast

-0.64

 outlook

-0.64

 projected

-0.63

 conjecture

-0.61

wcs

-0.60

 Maps

-0.60

reci

-0.60

POSITIVE LOGITS

 Cosponsors

0.76

fen

0.75

orus

0.75

 Freddie

0.74

utor

0.73

dl

0.73

idy

0.72

elsen

0.70

ppel

0.69

 Freddy

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.