INDEX

Explanations

phrases related to crimes against humanity

oai_token-act-pair · gpt-3.5-turbo

phrases related to crimes against humanity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

NH

-0.57

abs

-0.56

iss

-0.54

 Potential

-0.51

ãĥĪ

-0.50

 Cosponsors

-0.50

inarily

-0.50

 Impossible

-0.49

itech

-0.49

ingen

-0.49

POSITIVE LOGITS

axies

0.58

bye

0.56

ults

0.54

gger

0.54

 Roses

0.53

 Lunch

0.50

ellery

0.48

elines

0.48

osexual

0.48

thing

0.48

Activations Density 0.488%

phrases related to crimes against humanity

phrases related to crimes against humanity

No Comments

No Known Activations