© 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. 31-GEMMASCOPE-2-RES-16K
    7. 6293
    Prev
    Next
    INDEX
    Explanations

    function arguments

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

    The neuron is picking out tokens that occur at the beginnings of structured lines—e.g. list items, script or shell commands, or clause‐style sentences.

    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_31_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
     pelaksanaan
    0.56
    
    0.54
     ―
    0.52
    ्‍यादा
    0.52
    亨
    0.51
    ⠀
    0.50
    ﾈ
    0.49
     станда
    0.48
    完璧
    0.48
     presidential
    0.48
    POSITIVE LOGITS
     <!--<
    0.83
    /*
    0.74
    <!--
    0.68
    <i>
    0.65
    //#
    0.63
    也不
    0.63
    //@
    0.61
     कूल
    0.61
    //
    0.59
     somehow
    0.59
    Activations Density 0.006%

    No Known Activations