INDEX

Explanations

actions and functions related to customization and creation

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

enables an action

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

 елның

-0.53

GEBURTSDATUM

-0.50

 ویکی‌پدی

-0.47

onsored

-0.45

 AssemblyTitle

-0.45

 ModelExpression

-0.44

 resourceCulture

-0.44

 calendriers

-0.43

Verweise

-0.43

 Вікі

-0.43

POSITIVE LOGITS

 safely

0.45

 comod

0.44

Litteratur

0.44

 easily

0.41

 CURIAM

0.40

 grecque

0.39

 Lohan

0.38

낼

0.38

 comfortably

0.38

 cómoda

0.37

Activations Density 0.056%

actions and functions related to customization and creation

enables an action

No Comments

No Known Activations