INDEX

Explanations

statements or phrases that express existence or presence within a context

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

specific noun phrases

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

tagHelperRunner

-0.65

queryInterface

-0.53

 ModelExpression

-0.48

şört

-0.47

 Paglinawan

-0.46

.*")]

-0.46

oproject

-0.44

 intStringLen

-0.44

IsContent

-0.43

 Vorgang

-0.43

POSITIVE LOGITS

 autorytatywna

0.43

#![

0.42

GTCX

0.39

setVerticalGroup

0.38

पया

0.35

 Teich

0.35

 Mast

0.34

 Motivation

0.34

 motivation

0.33

INTRODUCTION

0.32

Activations Density 0.030%

statements or phrases that express existence or presence within a context

specific noun phrases

No Comments

No Known Activations