INDEX

Explanations

highlights from product reviews and experiences

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

review and tested

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 cherchés

-0.34

hi

-0.34

Pet

-0.34

pro

-0.33

xy

-0.33

 composition

-0.33

fog

-0.32

EndGlobalSection

-0.31

bew

-0.31

VersionUID

-0.31

POSITIVE LOGITS

 étoient

0.56

 avoient

0.55

脚注の使い方

0.55

Personensuche

0.54

 éditoriale

0.52

ⓘ

0.51

 redak

0.51

 Journalists

0.49

 Serikat

0.49

 journaliste

0.49

Activations Density 0.158%

highlights from product reviews and experiences

review and tested

No Comments

No Known Activations

highlights from product reviews and experiences

review and tested

No Comments

No Known Activations