© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 127767
    Prev
    Next
    INDEX
    Explanations

    centuries, numbers, or Roman numerals after punctuation

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

    The neuron fires on numeric tokens and related notation—digits (including multi‐digit and decimal numbers), fraction slashes, and ordinal suffixes (e.g. “st,” “nd,” “th”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     також
    -1.93
    ladung
    -1.87
     寶
    -1.85
    芣
    -1.76
     MÁS
    -1.73
     بنات
    -1.71
    iapkan
    -1.69
    obtain
    -1.68
    蹓
    -1.67
    羮
    -1.66
    POSITIVE LOGITS
    '
    2.66
     That
    1.78
    有很多
    1.71
     molta
    1.66
     While
    1.59
     importantly
    1.57
     These
    1.55
    you
    1.53
     ah
    1.52
     soprattutto
    1.52
    Activations Density 0.024%

    No Known Activations