INDEX

Explanations

references to procedures and operations related to system functions and setups

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

initial or first

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

AndEndTag

-0.56

잖

-0.42

 noticia

-0.42

 creatividad

-0.40

 незавершена

-0.38

mapStateToProps

-0.38

findpost

-0.38

wakilan

-0.38

 ویکی‌پدیا

-0.38

setVerticalGroup

-0.37

POSITIVE LOGITS

windowFixed

0.52

 debut

0.48

 initial

0.48

IConfiguration

0.48

Debut

0.47

basicConfig

0.45

 Initial

0.45

debut

0.44

initial

0.44

 INITIAL

0.44

Activations Density 0.345%

references to procedures and operations related to system functions and setups

initial or first

No Comments

No Known Activations