© 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. 11226
    Prev
    Next
    INDEX
    Explanations

    HON'BLE, HON'MY

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

    The neuron is strongly triggered by capitalized words and phrases—particularly proper names, titles, and other headline‐style or menu‐style tokens.

    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
    ay
    0.75
    ak
    0.75
    +
    0.75
    u
    0.73
    am
    0.70
     (
    0.70
    k
    0.70
    ad
    0.69
    ac
    0.69
    em
    0.68
    POSITIVE LOGITS
    ฯ
    0.93
     本当
    0.93
     或者
    0.92
     های
    0.91
     이라고
    0.91
     တော့
    0.90
     라는
    0.90
     тобто
    0.89
     って
    0.89
     tiến
    0.88
    Activations Density 0.727%

    No Known Activations