© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 7705
    Prev
    Next
    INDEX
    Explanations

    code/command line terms

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

    The neuron strongly activates on numeric literals and number-like tokens (e.g. integers, floats, line 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
    loé
    0.68
     Duis
    0.61
    ƈ
    0.59
    rieron
    0.58
     తాగ
    0.58
     Bekasi
    0.57
     विरु
    0.57
     উহা
    0.57
     कॉल्ड
    0.56
    schod
    0.56
    POSITIVE LOGITS
    :
    0.63
     ಎಂಬುದ
    0.62
     என்பத
    0.57
     =
    0.57
    .
    0.55
    。
    0.54
    =
    0.54
    /#
    0.52
    //@
    0.52
    ::
    0.52
    Activations Density 0.013%

    No Known Activations