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

    Neuronpedia

    INDEX
    Explanations

    such content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     idea
    0.56
     incredible
    0.55
     phenomenon
    0.52
     fantastic
    0.52
     concept
    0.52
     amazing
    0.52
     beautiful
    0.51
     poetic
    0.50
     wonderful
    0.49
     strange
    0.48
    POSITIVE LOGITS
    something
    0.49
    risi
    0.46
    those
    0.43
    某种
    0.43
    (<
    0.43
    .):
    0.42
    Something
    0.40
    '):
    0.40
     qualcosa
    0.39
    would
    0.39
    Act Density 0.008%

    No Known Activations