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

    Neuronpedia

    INDEX
    Explanations

    words related to negative connotations or consequences

    oai_token-act-pair · gpt-3.5-turbo

    instances of negative sentiment or connotation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    conservancy
    -0.84
    dropping
    -0.82
    hower
    -0.80
    DOM
    -0.80
    abiding
    -0.80
    adr
    -0.79
    iterator
    -0.79
    arers
    -0.79
    raltar
    -0.78
    cffff
    -0.77
    POSITIVE LOGITS
     reinforcement
    1.08
     impact
    0.94
     spiral
    0.92
     publicity
    0.92
     consequences
    0.88
     feedback
    0.88
     effects
    0.87
     gearing
    0.87
     consequence
    0.86
     karma
    0.86
    Act Density 0.025%

    No Known Activations