INDEX

Explanations

parts of the document indicating abstract sections or segment headers

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

abstracts and summaries

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

bootstrapcdn

-0.87

تقاوى

-0.82

 sumpay

-0.74

 AssemblyCompany

-0.72

 WaitForSeconds

-0.72

 MainAxisSize

-0.72

ейчас

-0.71

kloped

-0.70

 queſta

-0.69

uxxxx

-0.69

POSITIVE LOGITS

 purpose

0.37

cí

0.35

 subject

0.28

abstract

0.28

Overview

0.28

 singkat

0.28

 sujet

0.28

 goal

0.27

Abstract

0.27

 pré

0.27

Activations Density 0.023%

parts of the document indicating abstract sections or segment headers

abstracts and summaries

No Comments

No Known Activations