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

    Neuronpedia

    INDEX
    Explanations

    phrases related to declining to comment

    oai_token-act-pair · gpt-3.5-turbo

    instances of refusal to comment or provide information

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    isf
    -0.84
    viks
    -0.75
    oise
    -0.75
    aptic
    -0.73
    atures
    -0.73
    oyal
    -0.73
    bourne
    -0.70
    è¦ļéĨĴ
    -0.70
    obyl
    -0.68
    onz
    -0.67
    POSITIVE LOGITS
     requests
    0.94
     invitations
    0.93
     comment
    0.87
     answering
    0.82
     invitation
    0.81
     extradition
    0.80
     suggestions
    0.76
     admission
    0.75
     reimbursement
    0.75
     Admission
    0.72
    Act Density 0.036%

    No Known Activations