INDEX

Explanations

beginning comments in code that outline functions or classes

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

code comments (/**)

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

 Италијани

-0.69

 Autorizaciones

-0.65

parsedMessage

-0.65

oredCriteria

-0.63

windowFixed

-0.61

AndEndTag

-0.60

 XNUMX

-0.59

 "..\..\..\

-0.57

 arşivlendi

-0.57

 SqlCommand

-0.57

POSITIVE LOGITS

/**

0.54

0.43

Autowired

0.40

/*

0.38

 Whis

0.35

/**

0.34

lytic

0.33

 advantage

0.33

0.32

 contract

0.32

Activations Density 0.002%

beginning comments in code that outline functions or classes

code comments (/**)

No Comments

No Known Activations