INDEX

Explanations

programming constructs and syntax elements in code

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

code snippets

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

ęki

-0.40

 виправивши

-0.39

 Вікі

-0.37

Largeur

-0.37

 आय

-0.36

Jeografia

-0.36

 Fotografía

-0.35

 parties

-0.35

Rohy

-0.35

chufe

-0.34

POSITIVE LOGITS

wapV

0.66

SharedDtor

0.57

GraphicsUnit

0.54

fjspx

0.51

crollView

0.49

verifyException

0.48

 onBackPressed

0.47

enterOuterAlt

0.47

 <<<<<<<<<<<<<<

0.47

 للاسماء

0.46

Activations Density 0.025%

programming constructs and syntax elements in code

code snippets

No Comments

No Known Activations