INDEX

Explanations

phrases indicating mechanical or computational processes and their descriptions

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

how something is described

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

OGND

-0.77

 ModelExpression

-0.63

новништво

-0.60

 arşivlendi

-0.59

 kohdetta

-0.59

 ExecuteAsync

-0.57

ItemBackground

-0.56

 estekak

-0.56

 يتيمه

-0.55

UIControlState

-0.55

POSITIVE LOGITS

bootstrapcdn

0.35

 kullanılan

0.31

usiai

0.31

 inoltre

0.30

 currently

0.29

 ajutor

0.28

 primarily

0.28

 mainly

0.28

 '\\;'

0.27

ืม

0.27

Activations Density 0.102%

phrases indicating mechanical or computational processes and their descriptions

how something is described

No Comments

No Known Activations