INDEX

Explanations

programming keywords and variable management

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

variable assignments and code constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

Tembelea

-0.47

HasAnnotation

-0.47

AndEndTag

-0.47

rund

-0.44

Tikang

-0.43

 المعيارى

-0.43

CloseOperation

-0.43

ⓧ

-0.43

TypedDataSet

-0.42

 defaultstate

-0.42

POSITIVE LOGITS

LEGGI

0.43

 CURIAM

0.43

weilen

0.42

ographia

0.42

 ExecuteAsync

0.41

endphp

0.41

SharedDtor

0.41

 bowiem

0.40

IsString

0.40

skjaer

0.39

Activations Density 0.009%

programming keywords and variable management

variable assignments and code constructs

No Known Activations