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

Wolf

-0.67

 Balloon

-0.66

TNT

-0.64

 Vict

-0.64

Column

-0.63

 Credits

-0.62

 Friendly

-0.62

Season

-0.62

 Garn

-0.62

coord

-0.62

POSITIVE LOGITS

fman

0.92

alian

0.88

chy

0.87

ijn

0.75

 pains

0.73

 metast

0.70

ierrez

0.70

resso

0.70

umblr

0.69

atic

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.