INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

gypt

-0.69

ophob

-0.65

 overwhel

-0.64

 euth

-0.64

 Chaser

-0.63

 bully

-0.63

liest

-0.63

 repe

-0.62

#$

-0.62

tical

-0.61

POSITIVE LOGITS

Others

0.80

Products

0.79

UTC

0.73

Interstitial

0.73

window

0.72

Registration

0.72

interstitial

0.69

gren

0.69

Companies

0.68

Ĭ±

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.