INDEX

Explanations

information related to specific categories or classifications

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

categories and named entities

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

 deberes

-0.43

inquirer

-0.42

 arquitecto

-0.40

 dianteiro

-0.40

webElement

-0.40

 nemico

-0.40

 templado

-0.39

 Eindruck

-0.39

 executar

-0.39

 arquitectura

-0.38

POSITIVE LOGITS

 يتيمه

0.53

0.46

bootstrapcdn

0.45

 GoogleFonts

0.43

ècie

0.42

estacks

0.42

VYMaps

0.42

 surla

0.41

wieś

0.41

Autoritní

0.41

Activations Density 0.103%

information related to specific categories or classifications

categories and named entities

No Comments

No Known Activations