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

    Neuronpedia

    INDEX
    Explanations

    references to examples and instances in discussions or arguments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    introducing an example

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ſta
    -0.68
     houſe
    -0.61
     faſt
    -0.57
     Houſe
    -0.55
     purpoſe
    -0.54
    ſelf
    -0.53
     Majefty
    -0.53
     myſelf
    -0.52
     ſch
    -0.52
     pleaſure
    -0.52
    POSITIVE LOGITS
     egregious
    0.47
     instances
    0.43
     actitudes
    0.42
     irritating
    0.39
     infuriating
    0.39
     incidents
    0.39
     situation
    0.38
    fatto
    0.38
     situations
    0.37
    たとえば
    0.37
    Act Density 0.100%

    No Known Activations