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

 relief

-0.71

å§

-0.67

 dstg

-0.66

ä¸ī

-0.64

Han

-0.63

camera

-0.63

docker

-0.62

Pak

-0.61

ggle

-0.61

Yan

-0.61

POSITIVE LOGITS

Nou

0.70

cci

0.69

iffs

0.69

intage

0.67

 disproportion

0.65

 eleph

0.65

ounge

0.65

akespeare

0.64

ortium

0.64

oqu

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.