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

    Neuronpedia

    INDEX
    Explanations

    applications, situations, tasks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     предложи
    0.68
     название
    0.68
    Separate
    0.67
     افس
    0.65
    CIR
    0.64
     Separate
    0.64
     tentativo
    0.64
    aissez
    0.63
    说法
    0.63
     seperate
    0.63
    POSITIVE LOGITS
    这类
    1.00
     involving
    0.93
     situations
    0.89
     nowadays
    0.84
     involve
    0.80
     scenarios
    0.76
     geometries
    0.75
     setups
    0.75
     workloads
    0.74
    prone
    0.72
    Act Density 0.513%

    No Known Activations