INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ãĥĥãĥī

-0.78

ftime

-0.63

CSS

-0.62

ophen

-0.60

Pol

-0.60

opian

-0.60

odor

-0.59

ulhu

-0.59

 feeds

-0.58

azard

-0.57

POSITIVE LOGITS

amy

0.76

terness

0.72

owan

0.66

NRS

0.65

stru

0.65

 toget

0.62

é¾į

0.62

vana

0.62

amic

0.60

ARI

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.