INDEX

Explanations

terms related to broadcasting and broadcast media

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

broadcast television journalism

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

QV

-0.47

eV

-0.45

 wattpad

-0.45

 Etsy

-0.43

eyed

-0.42

Fancy

-0.41

KTM

-0.41

 Yorkers

-0.41

 ItemType

-0.40

 الأمر

-0.40

POSITIVE LOGITS

 broadcast

1.27

 Broadcast

1.22

broadcast

1.13

Broadcast

1.12

 broadcasting

1.05

 broadcasts

0.97

 Broadcasting

0.93

 broadcaster

0.78

 BROAD

0.77

 broadcasters

0.77

Activations Density 0.010%

terms related to broadcasting and broadcast media

broadcast television journalism

No Comments

No Known Activations