© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 11300
    Prev
    Next
    INDEX
    Explanations

    The neuron is primarily picking up on dash‐ or hyphen‐style punctuation (the “–”/“-” tokens that connect clauses or compound terms).

    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_24_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
    0
    0.75
    MASK
    0.70
    <0x0D>
    0.69
    Methyl
    0.69
    .
    0.68
    ↵
    0.68
    Meta
    0.66
    |\
    0.66
    Ther
    0.66
    Coord
    0.66
    POSITIVE LOGITS
     tourisme
    0.78
     creciendo
    0.71
     ONU
    0.70
     रक्षाबंधनाच्या
    0.70
     treaty
    0.69
     ባለ
    0.69
     constitutions
    0.69
     यादगार
    0.69
    偌
    0.68
     सच्चा
    0.68
    Activations Density 0.001%

    No Known Activations