© 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. 6459
    Prev
    Next
    INDEX
    Explanations

    command line instructions

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

    The neuron strongly activates on infrequent or unusually specific words (i.e. rare tokens) in the text.

    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
    京都
    0.80
     तरुणा
    0.77
    欖
    0.75
    UseCase
    0.74
     বাংলাদেশে
    0.71
    格兰
    0.71
     மறை
    0.70
    燾
    0.70
     Luxembourg
    0.70
    LoginPage
    0.69
    POSITIVE LOGITS
    ầng
    0.85
     Dafür
    0.83
     ocurrió
    0.82
     alegría
    0.79
     ocur
    0.77
     recordó
    0.77
     permita
    0.77
     diámetro
    0.77
     onderdeel
    0.74
     logró
    0.73
    Activations Density 0.000%

    No Known Activations