INDEX

Explanations

technical terms and elements relevant to programming and data structures

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

java.commons.math library

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

-------------</

-0.59

SourceChecksum

-0.52

 ویکی‌پدی

-0.44

UserScript

-0.42

contentLoaded

-0.40

();?>

-0.34

autoreleasepool

-0.34

цездатний

-0.34

 Photocase

-0.32

rzez

-0.32

POSITIVE LOGITS

 vooz

0.59

 Geiſt

0.58

 cappuccio

0.57

 パンチラ

0.56

abetes

0.54

⟬

0.54

majánló

0.53

 desmotivaciones

0.53

 صوتيه

0.52

<unused3>

0.52

Activations Density 0.421%

technical terms and elements relevant to programming and data structures

java.commons.math library

No Comments

No Known Activations