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

    Neuronpedia

    INDEX
    Explanations

    past tense verbs

    oai_token-act-pair · gpt-3.5-turbo

    expressions of actions or tasks that have not been completed

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    accompanied
    -0.59
     Manit
    -0.54
    abet
    -0.53
    ran
    -0.52
    bos
    -0.51
    inki
    -0.51
    hoe
    -0.51
     Tycoon
    -0.51
    verty
    -0.49
    asions
    -0.49
    POSITIVE LOGITS
    yet
    1.38
     nor
    1.37
     yet
    1.23
     since
    1.23
     anywhere
    1.13
     lately
    1.04
     anything
    1.04
     anymore
    1.01
     anybody
    1.00
    since
    0.98
    Act Density 0.293%

    No Known Activations