INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

oro

-0.68

gh

-0.66

reve

-0.63

tou

-0.62

 knockout

-0.61

 MÃ©

-0.61

Ã¨re

-0.61

abol

-0.61

GH

-0.60

rosis

-0.60

POSITIVE LOGITS

imum

0.71

aukee

0.69

criptions

0.67

entimes

0.67

 Painter

0.66

ãģ®éŃĶ

0.65

thood

0.65

ilies

0.62

igers

0.62

imes

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.