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

    Neuronpedia

    INDEX
    Explanations

    future situations and past feelings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Unable
    0.30
    Unable
    0.29
     Consider
    0.29
    choose
    0.29
    providing
    0.28
     Expect
    0.28
     synchronized
    0.27
     Associ
    0.27
    Choose
    0.27
     fournit
    0.27
    POSITIVE LOGITS
     really
    0.41
     worked
    0.40
     hurts
    0.37
     mattered
    0.37
     feels
    0.36
     happened
    0.36
     REALLY
    0.36
     creep
    0.34
     resonate
    0.34
     works
    0.33
    Act Density 0.063%

    No Known Activations