INDEX

Explanations

references to coding or programming constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mauva

-0.59

 plufieurs

-0.56

 saveiro

-0.55

 queſta

-0.54

iſen

-0.54

<unused41>

-0.51

--}}

-0.51

<unused47>

-0.50

<unused43>

-0.50

<unused23>

-0.50

POSITIVE LOGITS

TagMode

0.59

 noDo

0.52

帖最后由

0.51

Grüsse

0.43

Tikang

0.40

IsContent

0.39

ReusableCell

0.38

SourceChecksum

0.37

 CreateTagHelper

0.36

FormTagHelper

0.36

Activations Density 0.346%

references to coding or programming constructs

No Comments

No Known Activations

references to coding or programming constructs

No Comments

No Known Activations