INDEX

Explanations

structures indicating data or code organization and parameters

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

argument identifiers

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

はじめに

-0.29

scriptcase

-0.29

contra

-0.29

CONTRIBUT

-0.28

 ThemeData

-0.27

 exciting

-0.27

DIPSETTING

-0.27

 meilleures

-0.26

AutoModerator

-0.26

 endIndex

-0.26

POSITIVE LOGITS

RTLR

0.70

BagConstraints

0.61

 дописавши

0.59

Cyfeiriadau

0.57

verwijspagina

0.56

__':

0.56

脚注の使い方

0.56

WriteAttribute

0.56

 initComponents

0.55

HasIndex

0.54

Activations Density 0.012%

structures indicating data or code organization and parameters

argument identifiers

No Comments

No Known Activations