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

    Neuronpedia

    INDEX
    Explanations

    expressing evaluation and strong states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     شائع
    0.36
    ("",
    0.36
     allowable
    0.34
     Attempts
    0.34
     isValid
    0.34
     solves
    0.33
     obeys
    0.33
    চিৎ
    0.33
     controllability
    0.33
     समझा
    0.32
    POSITIVE LOGITS
     intrigued
    1.05
     impressed
    0.87
     excited
    0.86
     astonished
    0.85
     thrilled
    0.85
     amazed
    0.83
     fascinated
    0.83
     annoyed
    0.82
     interested
    0.80
     skeptical
    0.78
    Act Density 0.019%

    No Known Activations