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

    Neuronpedia

    INDEX
    Explanations

    reach conclusions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     upon
    0.38
    Upon
    0.38
     Upon
    0.36
     Regular
    0.35
    adoff
    0.34
    ൾ
    0.34
    iled
    0.34
    Prosec
    0.34
     Reflex
    0.33
     സംസാ
    0.33
    POSITIVE LOGITS
    ソッド
    0.41
    ガネ
    0.41
    业主
    0.39
    lavery
    0.38
     pasando
    0.38
    ;');
    0.38
     สอน
    0.38
    ন্নত
    0.38
     agents
    0.38
     prohibits
    0.37
    Act Density 0.006%

    No Known Activations