INDEX

Explanations

terms associated with news programs and broadcasting

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

SF, Channel, Theme, carrier

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

Datuak

-0.63

 CanadaChoose

-0.56

CppMethod

-0.55

脚注の使い方

-0.52

ьаж

-0.52

ésultats

-0.49

 >=",

-0.48

 ویکی‌آمباردا

-0.48

 تانيه

-0.47

 mierda

-0.47

POSITIVE LOGITS

bon

0.45

bon

0.42

Á

0.42

Che

0.40

Sp

0.40

Col

0.39

Che

0.39

á

0.39

Url

0.38

juh

0.38

Activations Density 0.057%

terms associated with news programs and broadcasting

SF, Channel, Theme, carrier

No Comments

No Known Activations