INDEX

Explanations

elements related to introductions or opening statements in texts

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

Tokens after "Introduction" or beginning sections

np_acts-logits-general · gemini-2.0-flash

introduction of

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

-0.89

RenderAtEndOf

-0.86

EndGlobalSection

-0.83

 estekak

-0.83

Diweddarwch

-0.83

msgTypes

-0.73

原始内容存档于

-0.73

PostInfinity

-0.71

 Италијани

-0.71

elemField

-0.70

POSITIVE LOGITS

 also

0.31

 therefore

0.27

Sor

0.26

 working

0.25

 cliquant

0.24

 really

0.23

 donc

0.23

 accepted

0.23

bra

0.22

 anders

0.22

Activations Density 1.221%

elements related to introductions or opening statements in texts

Tokens after "Introduction" or beginning sections

introduction of

No Comments

No Known Activations