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

    Neuronpedia

    INDEX
    Explanations

    ability to do things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Allow
    0.63
     Allowing
    0.62
     Attempts
    0.58
    Allow
    0.57
     Attempt
    0.53
    allow
    0.53
     Ability
    0.53
     võimal
    0.52
     einzigen
    0.51
    เพื่อ
    0.51
    POSITIVE LOGITS
     navigate
    0.97
     understand
    0.86
     communicate
    0.79
     anticipate
    0.76
     overcome
    0.74
     organize
    0.71
     innovate
    0.70
     maintain
    0.68
     comprehend
    0.68
     manage
    0.67
    Act Density 0.800%

    No Known Activations