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

    Neuronpedia

    INDEX
    Explanations

    tax deductions and losses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    实验
    0.55
    镇
    0.52
    阍
    0.49
    难以
    0.49
     Predictions
    0.48
    树
    0.48
    杨
    0.48
    酱
    0.48
     சீ
    0.47
    朝鲜
    0.47
    POSITIVE LOGITS
     deducted
    0.82
     deductible
    0.77
     deducting
    0.75
     dollars
    0.75
     creditors
    0.74
     deduct
    0.73
     deductions
    0.72
    計算
    0.71
     dollar
    0.69
     subtracting
    0.69
    Act Density 0.047%

    No Known Activations