INDEX

Explanations

mentions of specific body parts or items related to racing

oai_token-act-pair · gpt-3.5-turbo

references to laps 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Flavoring

-1.09

ãĥ¼ãĥĨãĤ£

-0.82

xual

-0.79

NESS

-0.75

 Marketable

-0.71

merce

-0.67

é¾įå¥ĳå£«

-0.67

Ø³

-0.65

clave

-0.65

omes

-0.64

POSITIVE LOGITS

dogs

1.00

ocobo

1.00

dog

0.97

ipop

0.93

Lap

0.89

robe

0.86

alm

0.85

antry

0.82

roach

0.80

apan

0.80

Activations Density 0.026%

mentions of specific body parts or items related to racing

references to laps in various contexts

No Comments

No Known Activations