INDEX

Explanations

references to a person named "Jo."

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Jo."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

IFIED

-0.74

enegger

-0.73

flies

-0.72

 amplification

-0.70

CLASS

-0.70

ashtra

-0.68

 IMAGES

-0.66

 interests

-0.66

intendent

-0.65

oÄŁ

-0.64

POSITIVE LOGITS

aquin

1.33

Jo

1.29

jo

1.13

ining

1.03

aqu

0.97

zeb

0.96

anne

0.95

anna

0.95

areth

0.91

zy

0.91

Activations Density 0.011%

references to a person named "Jo."

mentions of the name "Jo."

No Comments

No Known Activations