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

ãĥ¼ãĥ³

-0.83

ãĥīãĥ©

-0.83

 Mahjong

-0.79

ãĥ«

-0.76

NetMessage

-0.73

ãĥĲ

-0.73

Ĭ±

-0.72

 Qiao

-0.70

================================================================

-0.69

catentry

-0.68

POSITIVE LOGITS

naire

0.77

ylum

0.72

stones

0.66

zie

0.65

 roots

0.63

 Nether

0.62

quest

0.62

pard

0.61

eals

0.60

velt

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.