© 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. 24-GEMMASCOPE-2-RES-16K
    4. 6844
    Prev
    Next
    INDEX
    Explanations

    tokens followed by punctuation or delimiters

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

    The neuron strongly fires on metadata elements—proper titles, names, and date‐ or anniversary‐style references.

    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
     якщо
    0.79
    有名な
    0.69
    👎
    0.67
     sneezing
    0.64
     nếu
    0.64
    😰
    0.64
     если
    0.63
     когда
    0.63
    😫
    0.62
    если
    0.61
    POSITIVE LOGITS
     reimag
    0.95
     Projekt
    0.74
     Project
    0.71
     premiered
    0.71
     Fondation
    0.71
     pioneering
    0.71
     Powered
    0.71
     Thrive
    0.70
     Innovation
    0.69
     innovative
    0.69
    Activations Density 0.111%

    No Known Activations