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

lon

-0.63

dragon

-0.59

 Katy

-0.59

"/>

-0.58

Limit

-0.58

 Balloon

-0.58

 Document

-0.58

LOS

-0.58

audi

-0.58

fast

-0.57

POSITIVE LOGITS

senal

0.72

aturdays

0.70

BILITY

0.70

acebook

0.68

 sidx

0.66

Äĩ

0.64

INAL

0.64

BILITIES

0.64

unci

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.