© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 48702
    Prev
    Next
    INDEX
    Explanations

    three count

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

    The main thing this neuron does is detect numeric tokens (digits or 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_34/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
    ⑨
    -0.77
     venido
    -0.77
     Confederation
    -0.75
     Untold
    -0.75
    منع
    -0.72
    Pratique
    -0.71
    ベビー
    -0.70
    Prijs
    -0.68
    trump
    -0.67
     fifth
    -0.66
    POSITIVE LOGITS
    ３
    0.91
    THREE
    0.88
     THREE
    0.86
     three
    0.84
     २
    0.81
    ２
    0.78
     ২
    0.75
     ۲
    0.74
    three
    0.74
    噎
    0.73
    Activations Density 0.036%

    No Known Activations