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

    Neuronpedia

    INDEX
    Explanations

    Correctness validations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    汋
    0.44
     skater
    0.42
     wides
    0.42
     anden
    0.40
     persuade
    0.40
     spieg
    0.39
     questioned
    0.39
     bans
    0.38
    軹
    0.38
     persuasive
    0.38
    POSITIVE LOGITS
    正确
    0.56
    Correct
    0.54
     Correct
    0.54
     correct
    0.52
    correcto
    0.51
    correct
    0.51
     прави
    0.51
    ถูกต้อง
    0.50
     Прави
    0.47
     সঠিক
    0.46
    Act Density 0.000%

    No Known Activations