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

    Neuronpedia

    INDEX
    Explanations

    Reinforcement Learning from Human Feedback

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     अना
    0.46
    idents
    0.42
    यॉर्क
    0.41
    getcwd
    0.41
     чуде
    0.40
     namespaces
    0.40
     SQLAlchemy
    0.39
     तूफान
    0.39
    <0x1C>
    0.39
     offensive
    0.39
    POSITIVE LOGITS
    Feedback
    0.82
     Reward
    0.77
     Feedback
    0.75
     feedback
    0.74
     reward
    0.74
    Reward
    0.73
    feedback
    0.70
     Rein
    0.70
    Rein
    0.69
     reinforcement
    0.68
    Act Density 0.018%

    No Known Activations