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

    Neuronpedia

    INDEX
    Explanations

    create a plan

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     symmetric
    0.96
     harmonize
    0.95
     symmetrical
    0.93
     geï
    0.90
     GPU
    0.89
     concise
    0.88
     gluing
    0.88
     stiffness
    0.88
     apply
    0.86
    GPU
    0.86
    POSITIVE LOGITS
    ifter
    0.92
    ances
    0.86
    anil
    0.81
    idades
    0.80
    anks
    0.80
    ança
    0.80
    undet
    0.80
    वय
    0.77
    ogram
    0.77
     لأنه
    0.76
    Act Density 0.000%

    No Known Activations