© 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. 7166
    Prev
    Next
    INDEX
    Explanations

    version / версия

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

    The neuron strongly activates on title‐style or proper‐noun tokens—i.e. capitalized words and headings.

    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
    产业链
    0.71
    制造业
    0.71
    bankruptcy
    0.70
    娼
    0.70
    anedi
    0.67
    穢
    0.67
    oclonal
    0.66
    ophagy
    0.66
     autophagy
    0.64
    卤
    0.64
    POSITIVE LOGITS
     Deluxe
    1.01
     Sunrise
    0.88
     Premium
    0.86
     Moonlight
    0.84
     Royale
    0.84
     Classic
    0.84
     Platinum
    0.83
     Sapphire
    0.82
     Explorer
    0.81
     Chardonnay
    0.80
    Activations Density 1.079%

    No Known Activations