© 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. 15696
    Prev
    Next
    INDEX
    Explanations

    reconcile

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

    This neuron fires strongly on the first prominent content word (a topic‐signaling noun, adjective, or number) at the start of each document or section.

    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
    촛
    0.83
     breached
    0.77
    ท่อง
    0.76
     เร
    0.73
    Οι
    0.73
     hiked
    0.73
    latt
    0.71
     ospiti
    0.70
    벅
    0.70
    Ну
    0.70
    POSITIVE LOGITS
    外出
    0.73
    和
    0.72
    á
    0.69
    ”（
    0.68
     goodies
    0.67
    反而
    0.67
    和其他
    0.67
    accharide
    0.66
     посы
    0.65
    жное
    0.65
    Activations Density 0.000%

    No Known Activations