© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 8536
    Prev
    Next
    INDEX
    Explanations

    times or twice

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

    The neuron detects numerical multipliers or repetition expressions (e.g. “twice,” “three times,” “ten thousand times,” etc.).

    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
    ⅜
    0.66
    BUUID
    0.63
     Supplied
    0.63
     Curb
    0.62
     Muitos
    0.61
     Ocak
    0.61
     Aralık
    0.61
     febbraio
    0.59
     Polarization
    0.59
    և
    0.59
    POSITIVE LOGITS
     ಹೀ
    0.65
     normal
    0.64
    normal
    0.64
     normales
    0.61
    zos
    0.58
     convictions
    0.58
    oos
    0.57
     కంటే
    0.57
     spind
    0.56
    倍
    0.55
    Activations Density 0.107%

    No Known Activations