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

    Neuronpedia

    INDEX
    Explanations

    time units with ago or later

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    屑
    0.36
     използ
    0.34
    혜
    0.34
     উপজেলায়
    0.34
    每个
    0.34
    హిత
    0.34
    每個
    0.34
    ኝነት
    0.34
     нәрсә
    0.33
    пер
    0.33
    POSITIVE LOGITS
     ago
    1.04
     geleden
    0.80
     later
    0.75
    Ago
    0.70
     Ago
    0.70
     później
    0.70
     später
    0.67
     away
    0.62
    ago
    0.60
     позже
    0.56
    Act Density 0.014%

    No Known Activations