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

    Neuronpedia

    INDEX
    Explanations

    past tense verbs and forms of the verb "to be."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     armour
    -0.18
     Colour
    -0.18
     Initialise
    -0.17
     jewellery
    -0.17
    _colour
    -0.17
     recognised
    -0.16
     initialise
    -0.16
     authorised
    -0.16
     colourful
    -0.16
    Colour
    -0.16
    POSITIVE LOGITS
     prepared
    0.21
     famed
    0.18
     lumber
    0.18
    prepared
    0.16
     roster
    0.16
     holiday
    0.16
     minded
    0.16
     Ke
    0.16
     drip
    0.16
     resident
    0.15
    Act Density 0.273%

    No Known Activations