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

    Neuronpedia

    INDEX
    Explanations

    concepts related to rewards and revealing information

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tvguidetime
    -0.47
    anguages
    -0.47
    Personendaten
    -0.47
     HasFactory
    -0.47
     kasarigan
    -0.46
    olkien
    -0.46
     Ojo
    -0.45
    Puissance
    -0.45
    ույ
    -0.44
    <%
    -0.44
    POSITIVE LOGITS
     reward
    0.75
     reveal
    0.66
     Reward
    0.62
    reward
    0.60
     tight
    0.59
     reveals
    0.57
     Wall
    0.56
    Reward
    0.56
     cop
    0.55
     Reveal
    0.54
    Act Density 0.170%

    No Known Activations