INDEX

Explanations

occurrences of the keyword "void" in the context of programming

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

programming keywords: void, boolean

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

Negative Logits

Personendaten

-0.77

 surla

-0.75

PreInfinity

-0.73

 ModelExpression

-0.73

TagMode

-0.71

 NSCoder

-0.71

sizeCache

-0.70

 nakalista

-0.69

 createState

-0.68

Diweddarwch

-0.68

POSITIVE LOGITS

NOPQRST

0.46

 zelve

0.42

 själva

0.39

 zoude

0.39

 händ

0.39

ainte

0.37

 conventional

0.36

üst

0.36

our

0.36

 tanggung

0.35

Activations Density 0.002%

occurrences of the keyword "void" in the context of programming

programming keywords: void, boolean

No Known Activations