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

    Neuronpedia

    INDEX
    Explanations

    promises made

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     detect
    0.39
     adimensional
    0.38
     homogen
    0.37
    ገና
    0.37
     tiring
    0.35
     oxidize
    0.35
    tml
    0.34
    缸
    0.34
     subprocess
    0.34
     приго
    0.34
    POSITIVE LOGITS
     promise
    2.36
     promises
    2.36
     обеща
    2.27
    承诺
    2.22
    promise
    2.14
    promises
    2.14
     promised
    2.09
     Promises
    2.08
     pledges
    2.05
     pledge
    2.03
    Act Density 0.027%

    No Known Activations