INDEX

Explanations

mentions of locations or government-related terms

oai_token-act-pair · gpt-3.5-turbo

the use of the word "de" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ilial

-0.83

ippy

-0.75

uyomi

-0.74

mble

-0.73

atically

-0.71

krit

-0.70

ambers

-0.69

NetMessage

-0.69

DragonMagazine

-0.69

imum

-0.68

POSITIVE LOGITS

la

1.11

los

1.07

cember

1.05

utsche

1.02

las

0.95

arest

0.90

bris

0.89

uces

0.88

utsch

0.87

 jour

0.84

Activations Density 0.022%

mentions of locations or government-related terms

the use of the word "de" in various contexts

No Comments

No Known Activations