© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 5312
    Prev
    Next
    INDEX
    Explanations

    public

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

    The neuron fires strongly on prominent section or category labels—single‐word headings or tags such as “public,” “business,” “static,” “span,” “part,” etc.

    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
     무엇
    0.61
    Ნ
    0.58
    ん
    0.55
    就可以
    0.55
    ।
    0.55
    瀛
    0.54
    ziak
    0.53
    ။
    0.53
    你可以
    0.52
    겠죠
    0.51
    POSITIVE LOGITS
    кор
    0.63
    اعة
    0.56
    监听页面
    0.55
    altungs
    0.52
     ساک
    0.52
    ровать
    0.51
    ্যাপ
    0.51
     bipartisan
    0.51
    äl
    0.50
     مصرى
    0.50
    Activations Density 0.002%

    No Known Activations