INDEX

Explanations

references to user engagement and interaction

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

addressing you directly

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

 ویکی‌پدیای

-0.52

Bioaccumulative

-0.47

 plegable

-0.46

 Grüsse

-0.46

 maternidade

-0.45

 mijne

-0.44

 Grüße

-0.44

 própria

-0.43

 malheureusement

-0.43

 antiga

-0.42

POSITIVE LOGITS

RTEX

0.47

ſelf

0.47

parsedMessage

0.45

 purpoſe

0.44

DockStyle

0.43

CAC

0.43

saraba

0.42

 MessageBoxIcon

0.42

YOU

0.42

ollectionView

0.42

Activations Density 0.021%

references to user engagement and interaction

addressing you directly

No Comments

No Known Activations