© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14840
    Prev
    Next
    INDEX
    Explanations

    classified as

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

    The main thing this neuron does is detect standalone numeric tokens (sequences of digits).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    rú
    -0.96
     概要
    -0.95
     ricca
    -0.89
     perfettamente
    -0.88
     estremamente
    -0.88
     particolarmente
    -0.87
     there
    -0.86
     abbastanza
    -0.85
    Influ
    -0.85
     literally
    -0.85
    POSITIVE LOGITS
    เป็น
    1.37
     является
    1.18
     menjadi
    1.16
    considered
    1.13
     merupakan
    1.13
    成为
    1.11
     is
    1.02
    作为
    1.02
     เป็น
    0.99
    称为
    0.99
    Activations Density 0.187%

    No Known Activations