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

    Neuronpedia

    INDEX
    Explanations

    sequences that involve hyphens possibly used to emphasize or categorize information

    oai_token-act-pair · gpt-3.5-turbo

    negations or negative concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦
    -0.75
    ously
    -0.75
    .):
    -0.71
    HL
    -0.67
    :[
    -0.65
    âĢ¦âĢ¦
    -0.65
    âĢ¦)
    -0.65
    /(
    -0.65
    entials
    -0.62
    edo
    -0.62
    POSITIVE LOGITS
    _-
    1.80
    webkit
    0.82
    agine
    0.81
    =-=-=-=-=-=-=-=-
    0.80
     ie
    0.79
     [|
    0.74
    /-
    0.73
    named
    0.70
    especially
    0.70
    enough
    0.69
    Act Density 0.067%

    No Known Activations