INDEX

Explanations

sections or formatting in articles or documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 informée

-0.79

verwijspagina

-0.78

LookAnd

-0.74

 betweenstory

-0.72

rungsseite

-0.72

 AssemblyCulture

-0.71

 Италијани

-0.71

 jspb

-0.65

出版年

-0.65

tagHelperRunner

-0.65

POSITIVE LOGITS

<u>

0.43

key

0.36

Key

0.33

 absolutely

0.33

<b>

0.33

Key

0.33

 fact

0.33

 absolut

0.32

<strong>

0.31

Radians

0.31

Activations Density 0.407%

sections or formatting in articles or documents

No Comments

No Known Activations