INDEX

Explanations

references to playing cards or related activities

oai_token-act-pair · gpt-3.5-turbo

references to playing cards and card-related activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Siem

-0.68

¿½

-0.67

Kut

-0.66

merce

-0.66

 mosqu

-0.64

 Fiction

-0.64

 Amend

-0.62

PID

-0.62

 Cummings

-0.62

Pie

-0.62

POSITIVE LOGITS

inals

1.38

iologist

1.31

igan

1.27

igans

1.26

iovascular

1.14

holder

1.07

INAL

1.03

iology

1.01

ozo

0.97

holders

0.97

Activations Density 0.050%

references to playing cards or related activities

references to playing cards and card-related activities

No Comments

No Known Activations