INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ouver

-0.82

 horns

-0.79

natureconservancy

-0.69

ergic

-0.67

hers

-0.67

zza

-0.67

oust

-0.66

ence

-0.66

å¥³

-0.66

zar

-0.66

POSITIVE LOGITS

intendent

0.99

ebin

0.82

 Pastebin

0.72

 Konami

0.71

opter

0.71

swer

0.70

Bank

0.69

ISC

0.62

 Genius

0.60

IU

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.