INDEX

Explanations

responses and conclusions in discussions or statements

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

testimony, reiterated, scientists

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

 Италијани

-0.70

Portály

-0.68

GEBURTSDATUM

-0.68

Personensuche

-0.62

__*/

-0.56

ValueStyle

-0.56

 Wikimedijinoj

-0.56



-0.54

 resourceCulture

-0.54

__::

-0.53

POSITIVE LOGITS

avras

0.35

CrossRef

0.34

 मा

0.32

Initializable

0.32

inescence

0.32

lah

0.31

策

0.31

ској

0.31

findpost

0.30

Pas

0.30

Activations Density 0.005%

responses and conclusions in discussions or statements

testimony, reiterated, scientists

No Comments

No Known Activations