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

    Neuronpedia

    INDEX
    Explanations

    negation doesn't

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     tenha
    0.51
     Jangan
    0.50
     retreated
    0.48
    傳
    0.47
     choked
    0.46
     TBR
    0.46
    пова
    0.46
     камен
    0.45
     Sabemos
    0.45
    劇
    0.45
    POSITIVE LOGITS
    👁
    0.49
     itertools
    0.46
    ur
    0.45
    FBSDKInternal
    0.45
    Author
    0.44
    specification
    0.44
    itts
    0.43
    Specifications
    0.43
    Yves
    0.43
    Anders
    0.42
    Act Density 0.001%

    No Known Activations