INDEX

Explanations

instances of the word "autistic" or variations of it

oai_token-act-pair · gpt-3.5-turbo

terms related to autism

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mary

-0.78

liam

-0.70

 Soda

-0.69

Else

-0.68

ORK

-0.66

 FIRE

-0.65

 Beir

-0.64

BE

-0.63

ç«

-0.63

Water

-0.60

POSITIVE LOGITS

umn

1.05

ilus

1.02

emort

0.99

opsy

0.95

onomous

0.95

aut

0.91

obil

0.91

ica

0.86

iliary

0.86

olog

0.86

Activations Density 0.005%

instances of the word "autistic" or variations of it

terms related to autism

No Comments

No Known Activations