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

    Neuronpedia

    INDEX
    Explanations

    references to utility or strategic abilities in a combat context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    user turn start

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     representation
    -0.47
     commercial
    -0.46
     sociali
    -0.44
     Native
    -0.44
     modern
    -0.44
     betweenstory
    -0.43
     representations
    -0.43
     Crusoe
    -0.42
    mercial
    -0.42
     digital
    -0.42
    POSITIVE LOGITS
     reposition
    0.53
     flanks
    0.48
     spamming
    0.48
     teammates
    0.47
     cooldown
    0.46
     retreating
    0.46
    Personendaten
    0.46
     cepat
    0.45
     aprovechar
    0.45
     rápidos
    0.45
    Act Density 0.067%

    No Known Activations