© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 10774
    Prev
    Next
    INDEX
    Explanations

    report writing, establishing descent, password protection

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

    The neuron fires on numeric tokens—especially decimal numbers or reference-style figures.

    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_24_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
    时长
    0.89
     divisor
    0.89
    恂
    0.83
    dynamics
    0.82
     xhr
    0.80
     MinMaxScaler
    0.80
    >+</
    0.78
     toolPath
    0.77
    ㅛ
    0.77
     renderEncoder
    0.76
    POSITIVE LOGITS
    '
    0.73
     Да
    0.72
    สวัสดี
    0.69
    این
    0.68
    ここでは
    0.68
    Yeah
    0.67
     プ
    0.67
     പക്ഷേ
    0.67
    iamo
    0.66
     ін
    0.66
    Activations Density 0.001%

    No Known Activations