Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    for

    np_max-act · gemini-2.0-flash

    This neuron detects phrases where the author reports having discovered or found the solution themselves (e.g. "I found the solution for …", "I think I found a solution").

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    stringValue
    -0.07
    UserInfo
    -0.07
     CES
    -0.07
    VALID
    -0.07
    _ALL
    -0.07
     migrationBuilder
    -0.06
     ženy
    -0.06
     Codable
    -0.06
     территории
    -0.06
    人民共和国
    -0.06
    POSITIVE LOGITS
     коли
    0.06
     mücadel
    0.06
     hroz
    0.06
     spécial
    0.06
     fakt
    0.06
     nonlinear
    0.06
     det
    0.06
     Tru
    0.06
     skulle
    0.06
     enabling
    0.06
    Act Density 0.091%

    No Known Activations