INDEX

Explanations

code-related terms or specific 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 @ 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

NamedQueries

-0.65

FieldNumber

-0.55

AndroidJUnit

-0.54

AsUp

-0.53

mybatisplus

-0.53

UnusedPrivate

-0.51

elemField

-0.50

migrationBuilder

-0.47

 considérons

-0.47

 anſ

-0.46

POSITIVE LOGITS

 ivelany

0.50

 عقد

0.45

iastes

0.44

Quoted

0.40

かた

0.39

pato

0.39

 McColl

0.38

장

0.38

ielo

0.38

 series

0.38

Activations Density 0.296%

code-related terms or specific programming constructs

No Comments

No Known Activations