INDEX

Explanations

technical terminology and complex scientific concepts

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

onset, capacity, parallel, flashy

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

 Réponses

-0.55

NameInMap

-0.50



-0.48

AsUp

-0.47

évaluateur

-0.45

bcryptjs

-0.45

 noDo

-0.42

 Numerade

-0.41

httphttps

-0.39

liceerd

-0.39

POSITIVE LOGITS

 chi̍t

0.50

 دیکھیے

0.50

fjspx

0.45

kuuta

0.45

IonicModule

0.40

cshtml

0.39

раздо

0.39

튿

0.39

 NgModule

0.38

 Mild

0.36

Activations Density 0.395%

technical terminology and complex scientific concepts

onset, capacity, parallel, flashy

No Comments

No Known Activations