INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Fairfax

-0.73

espie

-0.71

hedral

-0.67

 Monstrous

-0.63

MI

-0.61

 mosaic

-0.61

 background

-0.61

 Curious

-0.60

 faint

-0.59

 Bomber

-0.59

POSITIVE LOGITS

atar

0.83

opped

0.78

target

0.72

amn

0.71

chuk

0.70

tec

0.70

ARCH

0.69

aterial

0.69

unci

0.69

lik

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.