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

    Neuronpedia

    INDEX
    Explanations

    strong descriptors of violence and hardship

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ationship
    -0.14
    lder
    -0.14
    ucked
    -0.14
    ipt
    -0.14
    orris
    -0.14
    479
    -0.14
    ãĥ¼ãĤ¹ãĥĪ
    -0.14
    oria
    -0.14
    lds
    -0.14
    855
    -0.14
    POSITIVE LOGITS
    ly
    0.22
    lest
    0.19
     treatment
    0.17
     reality
    0.16
    -force
    0.16
     Treatment
    0.16
    PEND
    0.16
     winters
    0.15
     honesty
    0.15
    ities
    0.15
    Act Density 0.052%

    No Known Activations