INDEX

Explanations

specific details related to automotive engineering and specifications

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

engine cubic inch

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

国民

-0.37

钥

-0.35

业

-0.34

toBeDefined

-0.34

 quedado

-0.33

 rembour

-0.33

 concentrado

-0.32

uyện

-0.32

ionato

-0.32

 spectateurs

-0.31

POSITIVE LOGITS

RectangleBorder

0.57

 surla

0.57

 jadx

0.55

RefNanny

0.49

forName

0.48

 jdbcTemplate

0.47

 TextStyle

0.47

 BoxFit

0.46

 SafeMath

0.46

rungsseite

0.45

Activations Density 0.017%

specific details related to automotive engineering and specifications

engine cubic inch

No Comments

No Known Activations