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

    Neuronpedia

    INDEX
    Explanations

    sentences starting with a verb and a pronoun as the subject

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     histor
    -0.77
     biography
    -0.76
    Lenin
    -0.74
     synthesis
    -0.73
    æĪ¦
    -0.73
    Born
    -0.72
     reconstruct
    -0.72
     ensemble
    -0.71
     reconstruction
    -0.71
     greatness
    -0.70
    POSITIVE LOGITS
     deterrent
    1.10
     discourage
    1.02
     loophole
    1.02
     workaround
    1.02
     FTC
    1.00
     enforcement
    1.00
     discriminatory
    1.00
     Customers
    0.97
     Consumers
    0.97
     loopholes
    0.94
    Act Density 1.021%

    No Known Activations