INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

yrights

-0.86

ãĤ¼ãĤ¦ãĤ¹

-0.68

Newsletter

-0.68

 Horizons

-0.65

 Masquerade

-0.65

CRIPTION

-0.65

 Dome

-0.64

 Confederation

-0.63

 Basin

-0.63

 Keynes

-0.61

POSITIVE LOGITS

lov

0.69

rr

0.68

ranged

0.66

boarding

0.65

indal

0.65

achev

0.65

handled

0.63

cens

0.62

regor

0.62

 disarm

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.