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

    Neuronpedia

    INDEX
    Explanations

    phrases related to movement and location

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Valid
    -0.66
     pse
    -0.63
     Success
    -0.63
     Failure
    -0.61
     cracked
    -0.60
    ottest
    -0.59
     Breach
    -0.58
    opus
    -0.58
    ieu
    -0.57
    kov
    -0.57
    POSITIVE LOGITS
     cursor
    1.05
     closer
    1.04
     needle
    0.93
     focus
    0.89
     forward
    0.89
     toward
    0.87
     hither
    0.83
    orts
    0.83
    izont
    0.81
     horizontally
    0.80
    Act Density 0.159%

    No Known Activations