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

    Neuronpedia

    INDEX
    Explanations

    website links and phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     emphasizes
    0.48
     যথার্থ
    0.47
     emphasized
    0.47
     penalized
    0.46
     catalyze
    0.45
    সমূ
    0.44
     underscores
    0.44
     catalyzes
    0.43
     IMHO
    0.43
    avorable
    0.43
    POSITIVE LOGITS
     £
    0.75
     rubbish
    0.69
     civilisation
    0.68
     maths
    0.66
     haemorrh
    0.66
     recognisable
    0.66
     TikTok
    0.66
     bosses
    0.65
     (£
    0.65
     cheeky
    0.64
    Act Density 0.001%

    No Known Activations