© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 1522
    Prev
    Next
    INDEX
    Explanations

    one followed by numerals or common words

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

    The neuron strongly activates on occurrences of the word “One,” especially when it’s used to begin a clause (i.e., enumeration like “One …”).

    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
     Beginn
    0.57
    ش
    0.53
    مكان
    0.52
    ComboBox
    0.52
     tiež
    0.52
    Cuando
    0.52
    دين
    0.50
     Cualquier
    0.49
     Accesat
    0.49
    د
    0.49
    POSITIVE LOGITS
     hundred
    0.93
     thing
    0.86
     lakh
    0.78
    hundred
    0.74
     thousand
    0.73
    жды
    0.73
     heck
    0.71
     crore
    0.69
    onta
    0.65
    надца
    0.64
    Activations Density 0.513%

    No Known Activations