© 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. 30915
    Prev
    Next
    INDEX
    Explanations

    oy, hoy, doy, soy names and places

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

    This neuron activates on the two-letter sequence “oy” (in any case or surrounding context).

    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
    باء
    -0.82
    QUENCE
    -0.79
    ficiencia
    -0.77
    kmal
    -0.75
     Advocates
    -0.74
    ներ
    -0.74
    UserInput
    -0.73
     constrained
    -0.73
     Many
    -0.71
     distancias
    -0.71
    POSITIVE LOGITS
    ster
    0.98
    sters
    0.93
    ston
    0.88
    alty
    0.87
    Hoy
    0.86
    STER
    0.85
    STON
    0.84
    ggable
    0.83
    oy
    0.82
    Æ
    0.80
    Activations Density 0.036%

    No Known Activations