INDEX

Explanations

keywords related to legal and organizational entities, as well as media and content production

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

types of entities or collections

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

 صوتيه

-0.48

 Variant

-0.46

 Decay

-0.46

 كومونز

-0.45

NameInMap

-0.45

parsedMessage

-0.44

getragen

-0.43

 goederen

-0.42

enderror

-0.42

 sexuelle

-0.41

POSITIVE LOGITS

 fandoms

0.44

 empires

0.41

 companies

0.41

 wineries

0.40

 corporations

0.39

 distanciation

0.38

 rivers

0.38

Попис

0.37

 repositories

0.37

hoeddwyd

0.37

Activations Density 0.077%

keywords related to legal and organizational entities, as well as media and content production

types of entities or collections

No Comments

No Known Activations