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

    Neuronpedia

    INDEX
    Explanations

    temperature ranges and descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     recrystall
    0.43
     thermal
    0.42
    thermal
    0.41
     nấu
    0.39
     horn
    0.39
     varnish
    0.39
     fumar
    0.39
     torches
    0.38
     object
    0.38
     methanol
    0.38
    POSITIVE LOGITS
     seventy
    0.59
     eighty
    0.55
     Seventy
    0.54
     Jersey
    0.48
    jersey
    0.48
     sixty
    0.47
     warm
    0.46
    乐
    0.46
     shorts
    0.45
     прият
    0.44
    Act Density 0.026%

    No Known Activations