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

    Neuronpedia

    INDEX
    Explanations

    student fine-tuning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     debuts
    0.42
    結果
    0.38
     response
    0.37
     output
    0.37
     result
    0.36
     ensure
    0.36
     preventable
    0.36
    getKey
    0.35
     outcome
    0.35
     hiçbir
    0.35
    POSITIVE LOGITS
     हों
    0.46
    вес
    0.45
    אל
    0.42
    hedron
    0.42
    prism
    0.41
    iern
    0.41
    Optionals
    0.41
    ird
    0.41
    穸
    0.40
    脛
    0.40
    Act Density 0.001%

    No Known Activations