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

    Neuronpedia

    INDEX
    Explanations

    action words referring to observing or experiencing entertainment content

    oai_token-act-pair · gpt-3.5-turbo

    references to watching and listening activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Berm
    -0.78
     ransom
    -0.72
     deployed
    -0.66
     dispatch
    -0.65
     exoner
    -0.65
     operational
    -0.65
    staking
    -0.63
    nel
    -0.63
     sworn
    -0.62
     regener
    -0.62
    POSITIVE LOGITS
    ables
    0.79
     enjoyment
    0.79
    artifacts
    0.76
     Trails
    0.74
     Watching
    0.72
    ences
    0.70
    é¾įå
    0.69
     habits
    0.69
    ERS
    0.69
    estyles
    0.69
    Act Density 0.177%

    No Known Activations