© 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. 8766
    Prev
    Next
    INDEX
    Explanations

    write examples, list details

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

    The neuron strongly activates on numeric tokens (including multi‐digit numbers, decimals, time stamps) and adjacent punctuation, i.e. it’s detecting raw numerical data in the text.

    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
     કેટલાક
    0.61
     কয়েকটি
    0.56
    蕖
    0.54
     بیشتری
    0.53
     더욱
    0.51
     अधिक
    0.50
     কয়েকটি
    0.49
    िसोदिया
    0.49
     വിവിധ
    0.49
    Requirements
    0.49
    POSITIVE LOGITS
     kematian
    0.63
     hasn
    0.61
     geen
    0.60
     zegt
    0.59
    :
    0.59
     +
    0.57
     heeft
    0.57
     starb
    0.57
     sagt
    0.57
     teda
    0.56
    Activations Density 0.029%

    No Known Activations