© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 11501
    Prev
    Next
    INDEX
    Explanations

    on how to

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

    The neuron fires on numeric tokens (standalone numbers) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    
    -2.02
     this
    -1.86
     some
    -1.77
     ライブ
    -1.70
     motocicletas
    -1.69
     semacam
    -1.67
     некоторы
    -1.65
    お疲れ
    -1.59
     Kiedy
    -1.57
    Inggris
    -1.56
    POSITIVE LOGITS
     of
    2.72
     is
    2.00
    再也
    1.98
    as
    1.89
     zrobić
    1.84
    corrhi
    1.81
     powiedzieć
    1.75
     he
    1.74
    which
    1.73
    vegli
    1.73
    Activations Density 0.038%

    No Known Activations