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

    Neuronpedia

    INDEX
    Explanations

    passive voice constructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝙃
    0.50
     éviter
    0.46
    𝙋
    0.46
     vorhand
    0.46
    𝐇
    0.44
     migli
    0.44
     voisines
    0.44
     thei
    0.43
     lepší
    0.43
     prévoir
    0.43
    POSITIVE LOGITS
    7
    0.48
    3
    0.47
    2
    0.43
     accused
    0.41
    6
    0.40
    ১০
    0.40
     placed
    0.40
    9
    0.40
    8
    0.39
     sent
    0.38
    Act Density 0.040%

    No Known Activations