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

    Neuronpedia

    INDEX
    Explanations

    prefixes and suffixes forming words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sedative
    0.46
     समाप्ति
    0.45
     headwinds
    0.43
     permissive
    0.42
     guesswork
    0.41
     disappearance
    0.41
     playtime
    0.41
     mishaps
    0.41
     reliance
    0.40
     meeste
    0.40
    POSITIVE LOGITS
     creates
    0.50
    ifferentiate
    0.49
    ifies
    0.47
     anticipates
    0.46
     provide
    0.46
     encrypt
    0.45
     pursues
    0.44
     temporarily
    0.44
    ilize
    0.43
     forbade
    0.43
    Act Density 0.148%

    No Known Activations