© 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. 106570
    Prev
    Next
    INDEX
    Explanations

    tickets for

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

    The neuron primarily responds to numeric tokens—i.e. any standalone numbers.

    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
    Pied
    -0.76
    vailing
    -0.74
     regalos
    -0.72
     Той
    -0.71
     Contain
    -0.70
    Most
    -0.70
     wspól
    -0.69
     paternal
    -0.69
    лә
    -0.69
    沿革
    -0.68
    POSITIVE LOGITS
    OfBirth
    0.91
     Malhotra
    0.74
     reverence
    0.73
     хиру
    0.72
     [
    0.71
    oudh
    0.71
    ایی
    0.70
    otonic
    0.69
    нению
    0.69
    банк
    0.69
    Activations Density 0.005%

    No Known Activations