INDEX

Explanations

alphanumeric identifiers or codes related to programming or data structures

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

codes and identifiers

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 routinely

-0.36

AccessorTable

-0.35

`),

-0.35

Selfer

-0.34

 gainera

-0.34

setObjectName

-0.33

')}

-0.33

 multiple

-0.32

 createSprite

-0.31

')]

-0.31

POSITIVE LOGITS

HtmlAttribute

0.50

 Савезне

0.48

 INTERESAR

0.47

<?

0.47

XmlType

0.47

ècie

0.44

+#+

0.43

 ModelExpression

0.43

xhttp

0.43

 distancias

0.42

Activations Density 0.032%

alphanumeric identifiers or codes related to programming or data structures

codes and identifiers

No Comments

No Known Activations