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

    Neuronpedia

    INDEX
    Explanations

    words related to avoiding actions or situations

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to the act of avoiding negative situations or outcomes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    raq
    -0.86
    soon
    -0.82
    DragonMagazine
    -0.78
    uid
    -0.77
    tek
    -0.76
    Haunted
    -0.76
    geist
    -0.72
    lease
    -0.70
    bleacher
    -0.70
    elf
    -0.70
    POSITIVE LOGITS
     pitfalls
    1.29
     detection
    1.15
     paying
    1.05
     harming
    1.03
     mentioning
    1.03
     wasting
    1.02
     confrontation
    1.01
     offending
    0.98
     answering
    0.98
     disclosing
    0.97
    Act Density 0.066%

    No Known Activations