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

    Neuronpedia

    INDEX
    Explanations

    I can statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     forgot
    0.41
     couldn
    0.39
     wrote
    0.38
     chose
    0.37
     swore
    0.33
     cannot
    0.33
     gave
    0.33
     haven
    0.32
     drank
    0.32
     drove
    0.31
    POSITIVE LOGITS
     நிச்சயம்
    0.27
     Scheduling
    0.27
     রয়েছে
    0.27
     Relevance
    0.26
     էին
    0.26
    Scheduling
    0.26
    Aware
    0.25
     Assistance
    0.25
    Recogn
    0.25
    PERT
    0.25
    Act Density 0.003%

    No Known Activations