© 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. 8969
    Prev
    Next
    INDEX
    Explanations

    names and collaborators

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

    The neuron fires on sequences of author names (and connecting “&”) in bibliography‐style citations.

    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
    ЗА
    0.62
    СТ
    0.61
    ஜ
    0.61
    НА
    0.61
    ᾽
    0.59
    ЕР
    0.57
    ಝ
    0.55
     ОТ
    0.55
    ЕЛЬ
    0.54
    Стра
    0.54
    POSITIVE LOGITS
     collaborators
    1.09
     coworkers
    1.04
     partner
    1.02
     Associates
    1.02
     colleagues
    0.97
     partners
    0.96
     partenaire
    0.95
     co
    0.94
     Partners
    0.94
     Partner
    0.93
    Activations Density 0.186%

    No Known Activations