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

    Neuronpedia

    INDEX
    Explanations

    prompts to continue

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    爆
    1.05
    新
    1.05
    各
    1.03
    爱
    1.02
    他
    1.00
    安
    0.98
    我
    0.96
    大
    0.96
    সি
    0.96
    考
    0.96
    POSITIVE LOGITS
     raczej
    1.22
     realmente
    1.22
     abbastanza
    1.21
     certainly
    1.19
     completamente
    1.18
     irgendwie
    1.15
     ತುಂಬಾ
    1.15
     véritable
    1.14
     estremamente
    1.14
     souvent
    1.10
    Act Density 0.006%

    No Known Activations