© 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. 10-GEMMASCOPE-RES-131K
    7. 61383
    Prev
    Next
    INDEX
    Explanations

    pan followed by word parts

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

    This neuron responds to occurrences of the subword “pan” (e.g. as a standalone token or prefix in words like “pan-cancer,” “pantoprazole,” “pancreatic,” “Panache,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    縂
    -2.77
    であるが
    -2.70
    ではある
    -2.56
    踽
    -2.56
    鬩
    -2.50
    лением
    -2.48
    躥
    -2.47
    椟
    -2.45
    㩒
    -2.44
     verbess
    -2.42
    POSITIVE LOGITS
    All
    2.91
    c
    2.81
    {
    2.58
    ｉ
    2.52
    Other
    2.44
    .(
    2.38
     resolute
    2.38
    amp
    2.33
    他和
    2.27
    景象
    2.27
    Activations Density 0.016%

    No Known Activations