INDEX

Explanations

programming keywords and types commonly used in coding

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

programming constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

tonode

-0.52

 TType

-0.46

LookAnd

-0.45

Personendaten

-0.45

GraphicsUnit

-0.45

EnableWeb

-0.43

 <=",

-0.43

تقاوى

-0.42

 learns

-0.41

 Roskov

-0.40

POSITIVE LOGITS

 Occupations

1.28

ConverterFactory

1.05

URLException

0.90

HtmlAttribute

0.84

SqlServer

0.82

Parcelize

0.75

ByteString

0.69

 للمعارف

0.67

makeConstraints

0.65

ceğine

0.59

Activations Density 0.001%

programming keywords and types commonly used in coding

programming constructs

No Known Activations