© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7240
    Prev
    Next
    INDEX
    Explanations

    AP Style, Reuters, AP Physics

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

    The neuron detects the standard news‐article dateline byline marker, especially the “(AP) —” pattern.

    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_31_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
    /
    0.62
     সূত্রে
    0.58
    opinion
    0.57
    当該
    0.57
    concrete
    0.57
     конкре
    0.57
     প্রসঙ্গে
    0.55
    admin
    0.53
    ratio
    0.53
     remarqu
    0.53
    POSITIVE LOGITS
     AP
    0.91
     ___
    0.71
    AP
    0.67
    .___
    0.67
    ___
    0.67
    坛
    0.66
    kanen
    0.64
     निशाना
    0.61
     khắp
    0.60
     Associated
    0.60
    Activations Density 0.089%

    No Known Activations