INDEX

Explanations

the word "life"

oai_token-act-pair · gpt-3.5-turbo

references to life, living, or vitality-related themes

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

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

chall

-0.75

PASS

-0.65

 Sabha

-0.62

secret

-0.60

 Franken

-0.59

 Liberals

-0.59

Program

-0.59

Minnesota

-0.59

Marginal

-0.58

 fine

-0.58

POSITIVE LOGITS

ife

1.16

yip

0.97

mite

0.88

yer

0.78

terness

0.77

llor

0.75

zie

0.75

ternity

0.75

lette

0.72

etus

0.72

Activations Density 0.013%

the word "life"

references to life, living, or vitality-related themes

No Comments

No Known Activations