© 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. 12643
    Prev
    Next
    INDEX
    Explanations

    2 followed by numbers

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

    The neuron selectively activates on numeric tokens—especially multi‐digit numbers like dates, counts, or measurements.

    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
     elementos
    0.66
    ρα
    0.63
     statuses
    0.63
     gramos
    0.63
    re
    0.63
     Narod
    0.63
    喈
    0.63
    ారు
    0.62
     gering
    0.62
    กก
    0.61
    POSITIVE LOGITS
     Noon
    0.91
     noon
    0.88
     ราศี
    0.79
    ールド
    0.66
    <0xA2>
    0.64
    let
    0.62
     scrollView
    0.61
     Apostles
    0.61
     Aéroport
    0.61
    volt
    0.61
    Activations Density 0.145%

    No Known Activations