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

    Neuronpedia

    INDEX
    Explanations

    asking for refinement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     clearer
    0.84
     annat
    0.80
     further
    0.80
     farther
    0.80
    の方が
    0.76
    又
    0.76
    again
    0.76
    क्राम
    0.75
     again
    0.74
    >>>
    0.74
    POSITIVE LOGITS
    ﻥ
    0.86
     మరియు
    0.81
    এবং
    0.80
    образова
    0.78
     einf
    0.76
    および
    0.73
     codebase
    0.73
     pojav
    0.73
     Bitcoin
    0.73
     extraordinary
    0.73
    Act Density 0.011%

    No Known Activations