INDEX

Explanations

phrases indicating acceptance and approval

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

foreign language phrases

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

Hentet

-0.73

 للمعارف

-0.71

httphttps

-0.69

ConstraintMaker

-0.69

الحياه

-0.69

tvguidetime

-0.69

rungsseite

-0.68

 تضيفلها

-0.66

GOTREF

-0.66

Personensuche

-0.66

POSITIVE LOGITS

 espada

0.40

("/:

0.35

 découver

0.35

('/:

0.34

 humo

0.33

 desco

0.30

 asti

0.30

 secuen

0.30

 Verstär

0.30

 silvestre

0.30

Activations Density 0.028%

phrases indicating acceptance and approval

foreign language phrases

No Comments

No Known Activations