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

    Neuronpedia

    INDEX
    Explanations

    months or references

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Sum
    -0.70
     Sh
    -0.68
     sum
    -0.65
     
    -0.57
     Tri
    -0.57
     her
    -0.55
     in
    -0.54
     the
    -0.53
     no
    -0.52
     het
    -0.49
    POSITIVE LOGITS
     Majefty
    1.01
     myſelf
    0.98
     виправивши
    0.96
     itſelf
    0.93
     Jefus
    0.89
     ſeveral
    0.89
    ſelf
    0.88
     purpoſe
    0.87
    ſelves
    0.87
     ―――――
    0.87
    Act Density 0.484%

    No Known Activations