INDEX

Explanations

syntax and structure indicators in programming code

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

code or formatting elements

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

 <<<<<<<<<<<<<<

-0.59

 שוליים

-0.56

әрмәләр

-0.56

oa̍t

-0.49

 Chwiliwch

-0.49

 وتسجيلات

-0.48

Cyfeiriadau

-0.47

xane

-0.46

 Unito

-0.46

esgue

-0.46

POSITIVE LOGITS

UnusedPrivate

0.56

EndContext

0.43

 AssemblyCompany

0.42

 pareil

0.40

 المعيارى

0.37

 GoogleFonts

0.36

setVerticalGroup

0.35

Vorte

0.35

InSection

0.35

здрав

0.35

Activations Density 0.101%

syntax and structure indicators in programming code

code or formatting elements

No Comments

No Known Activations