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

 sweaty

-0.68

sonian

-0.68

 itiner

-0.68

 Examiner

-0.66

 cough

-0.65

 panicked

-0.64

 scratched

-0.63

 sweat

-0.62

 kindred

-0.62

 Suns

-0.61

POSITIVE LOGITS

osuke

0.80

appa

0.79

ntil

0.77

zb

0.75

uten

0.75

quickShipAvailable

0.74

ategy

0.72

asel

0.71

MB

0.71

udeau

0.71

Activations Density 0.000%

No Known Activations

This feature has no known activations.