© 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. 6044
    Prev
    Next
    INDEX
    Explanations

    measure of prediction or classification

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

    This neuron fires strongly on bibliographic or citation metadata—years, author names, publication types (e.g. “thesis,” “MBA essays”), legal‐case abbreviations (“Ins.,” “Co.”), and similar reference 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
    prescription
    0.73
    住在
    0.72
    سته
    0.71
    yd
    0.70
    땠
    0.69
    栏
    0.68
    riente
    0.66
    你在
    0.66
     Prescription
    0.66
    column
    0.66
    POSITIVE LOGITS
     당
    0.76
    CDR
    0.76
    পাতি
    0.74
     príp
    0.73
     farther
    0.73
    Naive
    0.73
    ில்
    0.71
    CLN
    0.71
     국가
    0.70
     Ceci
    0.70
    Activations Density 0.001%

    No Known Activations