INDEX

Explanations

instances of technical formatting or notation, particularly related to mathematical expressions

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

\nonumber \\ formatting

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

PreExecute

-0.61

 ModelExpression

-0.61

 defaultstate

-0.57

InjectAttribute

-0.56

TestingModule

-0.54

 Obrador

-0.50

ьаж

-0.49

expandindo

-0.49

ISupport

-0.48

Parcelize

-0.47

POSITIVE LOGITS

downe

0.52

 autonomía

0.39

 Springsteen

0.38

wamy

0.35

 consentimiento

0.35

 Vázquez

0.35

Conteúdo

0.34

pperware

0.34

Vì

0.34

seks

0.34

Activations Density 0.053%

instances of technical formatting or notation, particularly related to mathematical expressions

\nonumber \\ formatting

No Comments

No Known Activations