© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2566
    Prev
    Next
    INDEX
    Explanations

    swift and kotlin code

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is detecting programming‐language code tokens—especially API method names with their colons and capitalized type identifiers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     عمان
    0.75
    Colombo
    0.74
     Colombo
    0.72
    Florida
    0.69
    Roof
    0.68
    Usu
    0.68
    合い
    0.68
    \[
    0.66
     Ceylon
    0.66
    Miami
    0.64
    POSITIVE LOGITS
     DispatchQueue
    1.06
     Xcode
    1.06
     SwiftUI
    0.95
     UserDefaults
    0.95
     ViewController
    0.95
     collectionView
    0.94
    collectionView
    0.86
     SKAction
    0.86
     VStack
    0.86
     CGFloat
    0.84
    Activations Density 0.000%

    No Known Activations