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

arted

-0.73

market

-0.72

aid

-0.70

enge

-0.70

played

-0.70

eyed

-0.66

ester

-0.66

alach

-0.66

nell

-0.66

 unaccount

-0.65

POSITIVE LOGITS

ESCO

0.73

ãĤ´ãĥ³

0.72

é¾

0.68

Mp

0.65

IENCE

0.64

 Clicker

0.64

æĸ

0.63

 centimeters

0.62

issance

0.61

Compan

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.