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

    Neuronpedia

    INDEX
    Explanations

    words related to justifications or reasons for actions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
    icer
    -0.78
    sung
    -0.73
    apes
    -0.69
    quer
    -0.69
     Wend
    -0.68
    ascus
    -0.68
    estern
    -0.67
    resent
    -0.66
    chrom
    -0.66
     Patriarch
    -0.65
    POSITIVE LOGITS
     basis
    1.03
    tenance
    0.81
     grounds
    0.77
     premise
    0.76
     footing
    0.76
    uation
    0.73
    plates
    0.72
     foundation
    0.72
     grounding
    0.70
    ifiers
    0.70
    Act Density 0.028%

    No Known Activations