INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

dayName

-0.77

 aliases

-0.69

 tyres

-0.69

 cler

-0.66

 paints

-0.63

 Journals

-0.63

 Marcel

-0.62

 sheds

-0.61

Cyr

-0.61

 genders

-0.61

POSITIVE LOGITS

achus

0.80

uin

0.77

Į

0.77

INGTON

0.73

Ban

0.73

é¾įå

0.72

MAL

0.72

ĺħ

0.71

uku

0.71

INTON

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.