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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating an inability to control a strong urge or desire

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a sense of needing assistance or support

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    theless
    -0.74
    andom
    -0.69
    pool
    -0.68
    bread
    -0.67
    wood
    -0.65
    ires
    -0.61
    isk
    -0.60
    prints
    -0.60
    bon
    -0.60
    oven
    -0.60
    POSITIVE LOGITS
     Desk
    0.73
     anybody
    0.72
     them
    0.65
     alleviate
    0.63
    ãĤ®
    0.61
    sth
    0.60
     anyone
    0.59
     financially
    0.59
    fully
    0.59
     anymore
    0.58
    Act Density 0.033%

    No Known Activations