© 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. 7442
    Prev
    Next
    INDEX
    Explanations

    pip install

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

    The neuron detects personal proper names—especially surnames or name fragments.

    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
    נ
    0.57
     Technically
    0.54
     बारी
    0.54
    доб
    0.52
    윙
    0.50
    邪魔
    0.50
    fade
    0.49
     profiled
    0.49
     sient
    0.49
    듕
    0.48
    POSITIVE LOGITS
     préc
    0.54
    噠
    0.54
     outnumbered
    0.53
    ulating
    0.52
    一定
    0.52
     Geld
    0.52
     Million
    0.52
     Concert
    0.52
    iyi
    0.52
    ού
    0.52
    Activations Density 0.295%

    No Known Activations