© 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. 22-GEMMASCOPE-RES-131K
    7. 33339
    Prev
    Next
    INDEX
    Explanations

    division by 2

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

    The neuron strongly activates on numeric literals (digits and numbers) in code.

    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
    ↵
    -1.93
    {
    -1.66
    色々
    -1.65
    ほら
    -1.64
    うわ
    -1.64
    の地
    -1.63
    おわりに
    -1.63
    _
    -1.59
    <i>
    -1.59
    あー
    -1.57
    POSITIVE LOGITS
    <bos>
    2.53
    中でも
    1.69
    。　　
    1.68
    Кроме
    1.67
     Testamento
    1.66
    cję
    1.65
    と
    1.65
    🫃
    1.64
     μόνο
    1.63
    を
    1.63
    Activations Density 0.006%

    No Known Activations