INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ãĤº

-0.72

peg

-0.68

stocks

-0.67

awatts

-0.66

 gigs

-0.66

rss

-0.63

tumblr

-0.61

cffffcc

-0.61

 cous

-0.61

ãĥ¼ãĥ³

-0.60

POSITIVE LOGITS

Fran

0.69

rition

0.67

iva

0.66

 Painter

0.63

 Butcher

0.63

Stew

0.62

resso

0.62

obo

0.61

 Tayyip

0.61

 attm

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.