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

    Neuronpedia

    INDEX
    Explanations

    start with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     followed
    0.88
    EME
    0.71
    となる
    0.71
     stages
    0.71
     ഇറ
    0.70
     stage
    0.69
     phase
    0.68
     protocol
    0.67
    CP
    0.66
     dage
    0.65
    POSITIVE LOGITS
     by
    0.97
    by
    0.96
    By
    0.84
     By
    0.83
     oleh
    0.81
     зараз
    0.79
     langfrist
    0.77
     anci
    0.77
    함으로써
    0.77
     এখানেই
    0.76
    Act Density 0.040%

    No Known Activations