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

achy

-0.70

ysis

-0.68

BuyableInstoreAndOnline

-0.68

chid

-0.66

Kus

-0.65

hao

-0.65

Oo

-0.65

 Deity

-0.64

chan

-0.64

listed

-0.62

POSITIVE LOGITS

etimes

0.80

maid

0.76

 budgets

0.66

thirst

0.65

ilib

0.63

kov

0.63

Reb

0.63

illard

0.62

versions

0.62

Relations

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.