INDEX

Explanations

elements of structured programming or code constructs

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

code or calculations

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

DAV

-0.41

Privat

-0.41

PreInfinity

-0.39

avigation

-0.38

 ***!

-0.37

儡

-0.36

IBarButtonItem

-0.35

RetentionPolicy

-0.35

Giới

-0.35

 nacimiento

-0.35

POSITIVE LOGITS

endregion

0.59

 jadx

0.55

RegressionTest

0.55

الحياه

0.51

astéroïdes

0.49

دانشنامهٔ

0.48

 متعلقه

0.47

+:+

0.46

++]

0.45

TokenNameLBRACE

0.45

Activations Density 0.013%

elements of structured programming or code constructs

code or calculations

No Comments

No Known Activations