© 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. 6078
    Prev
    Next
    INDEX
    Explanations

    def followed by function/variable/module

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

    The neuron fires on occurrences of the token fragment “def” (in any casing), flagging places where definitions or function‐definition keywords appear.

    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
     effort
    0.88
    effort
    0.79
     esfuerzo
    0.77
     esfor
    0.71
     노력
    0.68
     sfor
    0.66
     প্রচেষ্টা
    0.64
    {
    0.63
     esforços
    0.61
     savun
    0.61
    POSITIVE LOGITS
     Damages
    0.69
    誌
    0.62
     അതേ
    0.59
    Sx
    0.57
    গ্রহায়ণ
    0.57
    ẩy
    0.56
    igny
    0.55
    小物
    0.55
    📶
    0.55
     Damage
    0.55
    Activations Density 0.000%

    No Known Activations