© 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. 3158
    Prev
    Next
    INDEX
    Explanations

    numbers and code contexts

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

    The neuron primarily detects numeric tokens representing decimal values (floating-point numbers).

    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
    ig
    0.86
    ل
    0.80
    ndrome
    0.78
    un
    0.70
    EVER
    0.70
    in
    0.68
    ून
    0.67
    ف
    0.67
    ndon
    0.64
     occurred
    0.64
    POSITIVE LOGITS
     funcionários
    0.89
     eigenlijk
    0.83
     อายุ
    0.79
     управления
    0.78
     presión
    0.77
     ilusión
    0.76
    его
    0.76
     ответственности
    0.75
     necessárias
    0.75
     gestão
    0.75
    Activations Density 0.000%

    No Known Activations