INDEX

Explanations

actions and verbs related to connection and engagement

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

verb followed by preposition or article

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

Palmarès

-0.32

 kullanılır

-0.31

 aastal

-0.30

 suficientemente

-0.29

CrossRef

-0.28

 vandaag

-0.28

PSL

-0.28

อื่น

-0.28

 kirj

-0.28

 сегодня

-0.27

POSITIVE LOGITS

 كومونز

0.78

 تكبرها

0.69

 ſind

0.68

 Numerade

0.67

UnknownFieldSet

0.61

ьажоргаш

0.57

GEBURTS

0.57

 autorytatywna

0.54

 Paglinawan

0.54

orianCalendar

0.54

Activations Density 0.051%

actions and verbs related to connection and engagement

verb followed by preposition or article

No Comments

No Known Activations