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

 GOODMAN

-0.84

kai

-0.77

 iPhones

-0.71

 Accessed

-0.70

IRC

-0.69

NW

-0.66

 Buddh

-0.65

 Coinbase

-0.64

neau

-0.64

 æľ

-0.63

POSITIVE LOGITS

 occupant

0.74

 consolation

0.74

enei

0.73

 obscured

0.71

Ĥİ

0.68

enhagen

0.67

 physique

0.65

 occupants

0.64

yles

0.63

»Ĵ

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.