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

    Neuronpedia

    INDEX
    Explanations

    use, prioritize, evaluates, data

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ve
    0.48
     b
    0.47
     x
    0.43
     K
    0.42
     an
    0.42
     ocean
    0.42
     jit
    0.41
     hello
    0.41
     z
    0.41
     olives
    0.40
    POSITIVE LOGITS
    憸
    0.45
     चाहिँ
    0.44
     जास्त
    0.43
    Logistic
    0.42
    媟
    0.41
    went
    0.41
    addEnemy
    0.41
    recycl
    0.41
     intérêts
    0.41
    পশ্চিম
    0.40
    Act Density 0.000%

    No Known Activations