© 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. 31689
    Prev
    Next
    INDEX
    Explanations

    file path identifier

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

    The neuron is responsive to file‐path indicators (tokens like “path” and “/”) in code or comments.

    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
     Attributes
    -0.77
     attrs
    -0.75
    сно
    -0.75
    设施
    -0.73
    setti
    -0.73
    Hometown
    -0.73
     Attribute
    -0.72
     久
    -0.71
    atihan
    -0.71
    סה
    -0.70
    POSITIVE LOGITS
    constructed
    0.92
    довые
    0.84
    ↞
    0.77
     Babe
    0.77
     نقط
    0.77
     comin
    0.76
     زي
    0.76
     affection
    0.76
     Babyl
    0.75
     path
    0.75
    Activations Density 0.005%

    No Known Activations