INDEX

Explanations

sections that contain administrative or procedural information

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

following section headers like "Overview" or "Description"

np_acts-logits-general · gemini-2.0-flash

section titles and percentages

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

Revenir

-0.66

 تضيفلها

-0.65

MemoryWarning

-0.62

XmlAccessorType

-0.58

ReusableCell

-0.55

发表于

-0.55

st

-0.54

KommentareTeilen

-0.53

 ProgressDialog

-0.53

 referrerpolicy

-0.51

POSITIVE LOGITS

Overview

1.00

 Overview

0.98

OVERVIEW

0.92

overview

0.91

Key

0.90

 overview

0.89

 objectifs

0.83

 avantages

0.81

 Highlights

0.81

About

0.81

Activations Density 0.425%

sections that contain administrative or procedural information

following section headers like "Overview" or "Description"

section titles and percentages

No Comments

No Known Activations