INDEX

Explanations

specific programming languages and their related concepts or terms

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

medical acronyms

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

 EconPapers

-0.63

 صوتيه

-0.61

DeleteBehavior

-0.60

 незавершена

-0.57

adaptiveStyles

-0.57

 gynhyrchwyd

-0.57

InjectMocks

-0.56

 estekak

-0.56

 resourceCulture

-0.55

tvguidetime

-0.54

POSITIVE LOGITS

include

0.34

tel

0.32

marke

0.31

 endwhile

0.31

ModelAdmin

0.30

zed

0.29

ket

0.29

mar

0.29

льня

0.28

iro

0.28

Activations Density 0.059%

specific programming languages and their related concepts or terms

medical acronyms

No Comments

No Known Activations