© 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. 3418
    Prev
    Next
    INDEX
    Explanations

    guide, RPG, East, Sales

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

    This neuron responds strongly to numeric tokens, especially floating‐point numbers (numbers containing a decimal).

    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
     sollic
    0.73
     global
    0.71
     cardinals
    0.70
    רת
    0.69
     international
    0.68
    Georges
    0.67
    멜
    0.66
     dl
    0.65
    Michel
    0.65
    ків
    0.65
    POSITIVE LOGITS
     данные
    0.88
     Результа
    0.85
    дной
    0.83
    ik
    0.81
    ियस
    0.80
     Соответ
    0.80
    еты
    0.79
     मार्गदर्शन
    0.79
    hta
    0.78
    ermodel
    0.78
    Activations Density 0.000%

    No Known Activations