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

    Neuronpedia

    INDEX
    Explanations

    subject or topic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     runway
    0.87
     running
    0.82
     burning
    0.82
     beginning
    0.80
     mas
    0.78
     promoting
    0.78
     avoid
    0.78
     ninth
    0.75
     dodging
    0.75
     begin
    0.75
    POSITIVE LOGITS
    科目
    0.86
    Namespace
    0.84
    ből
    0.83
    sigmoid
    0.83
    首先
    0.83
    NPC
    0.82
    CRUD
    0.82
    لول
    0.81
    واج
    0.81
    SUBJECT
    0.80
    Act Density 0.000%

    No Known Activations