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

    Neuronpedia

    INDEX
    Explanations

    pronoun followed by punctuation/verb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     viable
    0.52
     predominant
    0.51
     poco
    0.47
    s
    0.47
     generalization
    0.47
     batch
    0.46
     Lagrangian
    0.45
     telescop
    0.45
     evolving
    0.43
     impactful
    0.43
    POSITIVE LOGITS
    <unused2217>
    0.67
    <unused2223>
    0.57
    ə
    0.57
    <unused2163>
    0.57
    <unused2218>
    0.57
    LoggerFactory
    0.56
    <unused2160>
    0.56
    inerary
    0.55
    <unused232>
    0.55
     wrześ
    0.55
    Act Density 1.525%

    No Known Activations