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

    Neuronpedia

    INDEX
    Explanations

    words related to investigative activities and searches

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    天
    -0.76
     Cancel
    -0.75
    NYSE
    -0.74
     resignation
    -0.72
     Reserved
    -0.71
     cancellation
    -0.71
    versions
    -0.70
     precedence
    -0.70
     detrim
    -0.68
     veto
    -0.68
    POSITIVE LOGITS
     uncover
    1.47
     excav
    1.40
     investigating
    1.37
     searched
    1.36
     surve
    1.35
     probing
    1.33
     interviewing
    1.33
     examining
    1.31
     analys
    1.31
     researching
    1.30
    Act Density 0.886%

    No Known Activations