© 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. 10-GEMMASCOPE-RES-131K
    7. 58899
    Prev
    Next
    INDEX
    Explanations

    wil names and places

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

    The neuron is keyed to the three‐letter sequence “Wil,” strongly activating on tokens that start with “Wil” (as in Wilmer, Wiltshire, Wilmington, 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
    تیجه
    -3.83
    鬟
    -3.64
    ᡵ
    -3.17
     великолеп
    -3.03
    gucig
    -2.94
     ochran
    -2.89
    ৹
    -2.83
    椙
    -2.73
     bruge
    -2.72
    왘
    -2.66
    POSITIVE LOGITS
    1
    5.41
    2
    3.98
    9
    3.80
    7
    3.66
    5
    3.66
    *
    3.47
    3
    3.39
    4
    3.22
    8
    3.20
    that
    3.17
    Activations Density 0.007%

    No Known Activations