INDEX

Explanations

components and attributes related to entities in a database schema or structure

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

organization names and places

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

WebElementEntity

-0.38

ربای

-0.37

AsUp

-0.33

階

-0.31

ühl

-0.31

 migrants

-0.30

 offensive

-0.30

ımı

-0.30

 taskId

-0.29

 dust

-0.29

POSITIVE LOGITS

Lähteet

0.63

 AssemblyCompany

0.55

 Majefty

0.54

 дописавши

0.54

 createStore

0.53

 providers

0.52

belec

0.52

机构

0.51

LookAnd

0.50

 vendors

0.49

Activations Density 0.055%

components and attributes related to entities in a database schema or structure

organization names and places

No Comments

No Known Activations