INDEX

Explanations

references to gameplay elements and strategies in video games

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

location descriptors

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

ategorias

-0.40

DockStyle

-0.39

 Tochter

-0.39

isateur

-0.37

ergies

-0.37

 professionelle

-0.37

wój

-0.36

 pulseras

-0.36

 hijas

-0.36

 strijd

-0.36

POSITIVE LOGITS

webElementXpaths

0.68

帖最后由

0.56

contentLoaded

0.49

 areas

0.43

areas

0.42

 swears

0.42

 area

0.42

 around

0.41

domada

0.41

tagPool

0.40

Activations Density 0.013%

references to gameplay elements and strategies in video games

location descriptors

No Comments

No Known Activations