INDEX

Explanations

comments and documentation-related annotations in the code

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

comments and TODO notes in code

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

mathematical expressions and code separators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 initComponents

-0.71

Cordialement

-0.66

 neutre

-0.64

 myſelf

-0.64

 onBind

-0.64

 raiſ

-0.62

ActionMode

-0.61

nachweis

-0.58

 chofe

-0.58

Conclusie

-0.57

POSITIVE LOGITS

uxxxx

0.54

脚注の使い方

0.53

 thiệu

0.52

Fifth

0.51

/*

0.51

 autorytatywna

0.50

 Blow

0.49

izzi

0.48

+//

0.48

Kon

0.48

Activations Density 0.045%

comments and documentation-related annotations in the code

comments and TODO notes in code

mathematical expressions and code separators

No Comments

No Known Activations