© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 68523
    Prev
    Next
    INDEX
    Explanations

    Microsoft followed by dot

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

    The neuron activates on the token “Microsoft,” so it’s detecting occurrences of the Microsoft namespace or brand name in code.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     or
    -2.33
     Either
    -2.11
     anderen
    -1.94
     disseminate
    -1.91
    していますが
    -1.87
     graciosas
    -1.85
    u
    -1.84
    不管是
    -1.82
     abarca
    -1.76
     céle
    -1.75
    POSITIVE LOGITS
    Ꭳ
    2.13
    ”،
    2.13
     linke
    2.00
     asas
    2.00
    涐
    1.98
     sene
    1.95
     formen
    1.92
     vixion
    1.91
    蒌
    1.91
     mii
    1.90
    Activations Density 0.005%

    No Known Activations