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

çīĪ

-0.80

asley

-0.76

prison

-0.75

DragonMagazine

-0.74

trial

-0.71

edience

-0.70

punk

-0.69

give

-0.69

bledon

-0.69

brance

-0.68

POSITIVE LOGITS

 Enchant

0.75

uras

0.73

 Telesc

0.70

 concess

0.69

ezvous

0.69

Cth

0.68

 enchant

0.68

 Desk

0.66

 Ceres

0.65

 desks

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.