INDEX

Explanations

concepts and terminology related to evaluation and comparison of quality and effectiveness

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

looks best

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

evos

-0.54

 AspNetCore

-0.50

الحياه

-0.49

----</

-0.46

 GenerationType

-0.45

ⓧ

-0.45

TagMode

-0.45

 pinulongan

-0.44

····

-0.44

arnings

-0.44

POSITIVE LOGITS

 AssemblyVersion

0.49

WebElementEntity

0.37

󠁣

0.36

確かに

0.36

exitRule

0.34

IVEREF

0.33

✭✭

0.32

endforeach

0.32

siblings

0.32

 persegu

0.30

Activations Density 0.084%

concepts and terminology related to evaluation and comparison of quality and effectiveness

looks best

No Comments

No Known Activations