© 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. 3658
    Prev
    Next
    INDEX
    Explanations

    common and acceptable phrases

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

    The neuron activates on modal expressions of possibility, especially tokens like “possible” (as in “it is possible to …”).

    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
     sashimi
    0.80
     granola
    0.78
     cappuccino
    0.74
     headband
    0.71
     gyr
    0.71
    냔
    0.70
     می‌تواند
    0.70
     carbohyd
    0.69
    जेस्टिव
    0.69
     红
    0.68
    POSITIVE LOGITS
     to
    0.92
     أنّ
    0.76
     that
    0.75
     أن
    0.75
    ว่า
    0.73
     bahwa
    0.71
    to
    0.67
    that
    0.66
    کہ
    0.66
    قاء
    0.65
    Activations Density 0.247%

    No Known Activations