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

    Neuronpedia

    INDEX
    Explanations

    suspect poem beginning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Recursive
    0.37
    Python
    0.37
    Dependencies
    0.37
    Debug
    0.34
     阅读全文
    0.34
    Kernel
    0.34
    <unused1465>
    0.33
    Payload
    0.33
    Ü
    0.32
    사용
    0.32
    POSITIVE LOGITS
     those
    0.53
     childhood
    0.50
     heartfelt
    0.50
     praise
    0.48
     revenge
    0.47
     plenty
    0.47
     camaraderie
    0.46
     widespread
    0.45
     heartbreaking
    0.45
     heartwarming
    0.45
    Act Density 9.907%

    No Known Activations