INDEX

Explanations

citations and digital object identifiers (DOIs)

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

doi links

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

новништво

-0.48

 ส์

-0.45

 évaluation

-0.43

gól

-0.43

 áudio

-0.43

 rescate

-0.42

 scrittura

-0.41

 utilising

-0.40

 sonidos

-0.40

 utilización

-0.40

POSITIVE LOGITS

LookAnd

0.81

AndroidJUnit

0.77

UseVisualStyle

0.69

bootstrapcdn

0.67

PreferredItem

0.67

AsUp

0.62

unmodifiable

0.61

Diwedd

0.60

 uLocal

0.60

 تضيفلها

0.59

Activations Density 0.008%

citations and digital object identifiers (DOIs)

doi links

No Comments

No Known Activations

citations and digital object identifiers (DOIs)

doi links

No Comments

No Known Activations