INDEX

Explanations

mathematical symbols and specific numerical values

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

citations 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 @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 defaultstate

-0.58

 ModelExpression

-0.57

 AssemblyCompany

-0.56

Бахар

-0.53

tagHelperRunner

-0.45

 الحره

-0.45

 intptr

-0.45

 Weakness

-0.44

 يتيمه

-0.44

Gön

-0.43

POSITIVE LOGITS

 Савезне

0.50

 împ

0.43

Спасылкі

0.42

本日は

0.40

 ivelany

0.38

oredCriteria

0.38

Partager

0.38

 kautta

0.37

Бележки

0.37

 sesama

0.37

Activations Density 0.314%

mathematical symbols and specific numerical values

citations and identifiers

No Comments

No Known Activations