© 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. 34-GEMMASCOPE-RES-131K
    7. 99052
    Prev
    Next
    INDEX
    Explanations

    cas words

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

    This neuron detects code‐like tokens and symbols—especially import paths, punctuation, and identifiers (e.g. slash-separated module names) typical of source 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_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
     Eğ
    -0.83
    єн
    -0.81
     席
    -0.79
    桌面
    -0.79
    śle
    -0.77
    getNumberOf
    -0.75
     rigid
    -0.75
     рослин
    -0.73
     inusual
    -0.71
     unfortunate
    -0.71
    POSITIVE LOGITS
    曽
    0.81
    Cas
    0.78
     Fakten
    0.73
     カウンター
    0.73
    くださった
    0.70
    а
    0.69
    bul
    0.69
     ô
    0.68
    न्ध
    0.68
    TRL
    0.68
    Activations Density 0.050%

    No Known Activations