INDEX

Explanations

coding terminology and structures related to data handling and proxy functionality

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

interface settings

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

ziasztok

-0.31

 erwähnten

-0.31

 Bestandteil

-0.31

 fidèle

-0.30

 lavorare

-0.30

 enamorado

-0.30

 Kindheit

-0.29

insegna

-0.29

 Gesells

-0.28

 Stoffe

-0.28

POSITIVE LOGITS

下载附件

0.63

 kasarigan

0.61

Kjelder

0.60

Халык

0.59

出版年

0.58

RenderAtEndOf

0.58

허

0.57

ResumeLayout

0.56

 lenker

0.56

featureID

0.56

Activations Density 0.126%

coding terminology and structures related to data handling and proxy functionality

interface settings

No Comments

No Known Activations