INDEX

Explanations

references to descriptions or definitions

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

the word description

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

ftagPool

-0.73

 sumpay

-0.65

 تضيفلها

-0.62

 الاطلاع

-0.60

tagHelperRunner

-0.59

تقاوى

-0.59

balleur

-0.59

InjectAttribute

-0.56

دانشنامهٔ

-0.56

 typelib

-0.55

POSITIVE LOGITS

Description

0.69

 description

0.63

description

0.62

 Description

0.54

DESCRIPTION

0.54

 DESCRIPTION

0.50

 وصف

0.42

 descriptions

0.41

 Purpose

0.41

Descrizione

0.41

Activations Density 0.002%

references to descriptions or definitions

the word description

No Comments

No Known Activations