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

    Neuronpedia

    INDEX
    Explanations

    negative words depicting failures or problems

    oai_token-act-pair · gpt-3.5-turbo

    negative descriptors related to failure and disappointment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ancies
    -0.98
    Sources
    -0.79
    icts
    -0.77
    rams
    -0.77
    anooga
    -0.75
    ensions
    -0.72
     profiles
    -0.72
    eways
    -0.72
     Roots
    -0.71
    fixes
    -0.71
    POSITIVE LOGITS
     unto
    1.09
     compared
    0.87
     affair
    0.83
     worthy
    0.79
     akin
    0.78
     breaker
    0.78
     nonetheless
    0.76
     reel
    0.75
     worth
    0.73
     unworthy
    0.73
    Act Density 0.267%

    No Known Activations