INDEX

Explanations

programming-related terminology, specifically regarding data types and structures

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

code context

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

 Administrativna

-0.40

 femininas

-0.40

 super

-0.37

 aikaa

-0.36

 pessoais

-0.36

 мәкал

-0.35

AddHtmlAttribute

-0.35

 crossorigin

-0.35

sum

-0.35

ścią

-0.35

POSITIVE LOGITS

ViewFeatures

0.67

 defaultstate

0.63

AutoScale

0.57

richTextPanel

0.51

ſelf

0.48

endregion

0.47

/**

0.47

UseVisualStyle

0.46

 Profe

0.46

parsedMessage

0.44

Activations Density 0.000%

programming-related terminology, specifically regarding data types and structures

code context

No Comments

No Known Activations