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

    Neuronpedia

    INDEX
    Explanations

    phrases related to errors or issues

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "error" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    electric
    -0.84
    tsky
    -0.79
    apeake
    -0.78
    amen
    -0.77
    apy
    -0.76
    atos
    -0.75
    edom
    -0.74
    nai
    -0.74
    Electric
    -0.74
    arov
    -0.69
    POSITIVE LOGITS
    ously
    0.87
    uracy
    0.85
     margin
    0.85
    gered
    0.79
     guiActiveUn
    0.77
     error
    0.72
     prone
    0.72
    fully
    0.72
     deceive
    0.71
     mishand
    0.71
    Act Density 0.028%

    No Known Activations