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

 unfocusedRange

-0.78

 millenn

-0.71

 informant

-0.67

 tiss

-0.66

©¶æ

-0.63

inelli

-0.63

retty

-0.61

ÃĥÃĤ

-0.61

 tyres

-0.60

 duct

-0.59

POSITIVE LOGITS

advertisement

0.76

Zh

0.75

urga

0.71

atl

0.71

Thumbnail

0.69

gat

0.67

0.66

BIL

0.65

ption

0.64

Iw

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.