INDEX

Explanations

technical terms and code syntax related to programming and computer science

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

docker image, LINUX config, proto definition

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

 propOrder

-0.62

sizeCache

-0.59

verwijspagina

-0.56

RepeatedField

-0.51

ynchronously

-0.51

 betweenstory

-0.51

ModelBuilder

-0.50

exitRule

-0.50

 indisponible

-0.49

 wikipagina

-0.49

POSITIVE LOGITS

 történ

0.38

原始内容存档

0.37

 fútbol

0.37

DataSetChanged

0.35

řev

0.34

wählen

0.33

diakan

0.32

DNEY

0.32

voerd

0.31

 gérer

0.31

Activations Density 0.227%

technical terms and code syntax related to programming and computer science

docker image, LINUX config, proto definition

No Comments

No Known Activations