© 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. 34-GEMMASCOPE-RES-131K
    7. 59219
    Prev
    Next
    INDEX
    Explanations

    class, unit, ALL

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

    The neuron activates on numeric tokens and digit‐heavy identifiers (e.g. version numbers, section markers, patent or class IDs).

    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
     丸
    -0.88
    fileType
    -0.82
    꺾
    -0.79
     vase
    -0.77
    ñador
    -0.76
    otechnical
    -0.73
    ｖ
    -0.73
     afla
    -0.73
     commissioner
    -0.71
     Mavericks
    -0.70
    POSITIVE LOGITS
    yscy
    0.71
    單位
    0.69
     kõik
    0.65
     ALL
    0.63
     दि
    0.61
     Gegen
    0.60
    әне
    0.60
     Did
    0.60
    单位
    0.59
    ForAll
    0.59
    Activations Density 0.043%

    No Known Activations