© 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. 6418
    Prev
    Next
    INDEX
    Explanations

    percentage calculations or specific values

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

    The neuron activates on numeric tokens that represent decimal or floating‐point values (numbers with a decimal point).

    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
     discretized
    0.57
     bias
    0.56
     asymptotically
    0.55
    omen
    0.55
    焐
    0.54
     amplitude
    0.52
     discretization
    0.52
    repetitions
    0.52
    <0x96>
    0.52
     seconds
    0.51
    POSITIVE LOGITS
    тную
    0.61
     болезни
    0.59
    itoare
    0.55
    acyjnego
    0.55
    stargo
    0.54
    ंचं
    0.54
     vragen
    0.54
    ListPage
    0.54
     новую
    0.54
     اللغة
    0.54
    Activations Density 0.002%

    No Known Activations