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

    Neuronpedia

    INDEX
    Explanations

    phrases related to personal experiences or sentiments

    oai_token-act-pair · gpt-3.5-turbo

    references to musical albums and songs

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     immersion
    -0.86
     unnecessarily
    -0.84
     anx
    -0.76
     barriers
    -0.76
     redundancy
    -0.76
     deleg
    -0.75
     disadvantages
    -0.75
     breaching
    -0.74
     deterrence
    -0.73
     caution
    -0.73
    POSITIVE LOGITS
     Lonely
    1.26
     Gone
    1.19
     Wrong
    1.17
     Tonight
    1.16
     Like
    1.15
     Alone
    1.14
     Called
    1.12
     Changed
    1.12
     Again
    1.12
     Alright
    1.11
    Act Density 0.207%

    No Known Activations