INDEX

Explanations

instances of numerical scores or statistics in sports-related contexts

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

basketball game outcomes

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

 CanadaChoose

-0.50

contentLoaded

-0.44

 ModelAndView

-0.43

"](

-0.41

ernoo

-0.41

Personendaten

-0.40

 typelib

-0.40

)”.

-0.39

 estekak

-0.39

səhifə

-0.39

POSITIVE LOGITS

🏀

0.44

Basketball

0.41

irtschaft

0.40

basketball

0.40

 xuyên

0.39

 Basketball

0.39

 basketball

0.39

Leaders

0.38

 Otis

0.38

 suffoc

0.37

Activations Density 0.032%

instances of numerical scores or statistics in sports-related contexts

basketball game outcomes

No Comments

No Known Activations