© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2467
    Prev
    Next
    INDEX
    Explanations

    numeric digits

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

    The neuron fires on mixed letter–digit tokens (e.g. “A1”, “N1”, “W8”)—it’s picking out alphanumeric codes.

    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
    _
    0.66
    ��������
    0.66
    ^^^^
    0.63
    AddNew
    0.61
    xxxx
    0.60
    bdb
    0.59
    ০০০
    0.58
    させ
    0.57
    ����
    0.57
    ════
    0.56
    POSITIVE LOGITS
    మ్‌
    0.62
     ایکس
    0.62
     schre
    0.62
     bergantung
    0.61
     นี่
    0.61
     সামরিক
    0.59
     말했다
    0.59
    এক্স
    0.59
    🇾
    0.59
     tưởng
    0.58
    Activations Density 0.211%

    No Known Activations