INDEX

Explanations

key terms related to mythology, legal terms, programming directives, and data management

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

link, directories

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

expandindo

-0.66

abestanden

-0.66

 ModelExpression

-0.55

addCriterion

-0.55

السكان

-0.53

 surla

-0.52

 createSprite

-0.51

angliski

-0.51

 protoimpl

-0.51

GEBURTSDATUM

-0.50

POSITIVE LOGITS

 auroit

0.51

 pouvoit

0.50

 muerta

0.47

 étoit

0.46

 habet

0.45

 Partager

0.44

 enfans

0.42

 feroit

0.42

Kesimpulan

0.42

/**

0.41

Activations Density 0.210%

key terms related to mythology, legal terms, programming directives, and data management

link, directories

No Comments

No Known Activations