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

    Neuronpedia

    INDEX
    Explanations

    `

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    λογ
    0.43
    OLDER
    0.42
     ..,
    0.39
    いますが
    0.39
     হয়েছিল
    0.38
    ुआ
    0.38
    <start_of_image>
    0.38
    いましたが
    0.37
     সহ্য
    0.36
    zen
    0.36
    POSITIVE LOGITS
    )</
    0.50
     politič
    0.49
    0.49
     полити
    0.47
    0.47
    、​
    0.46
    更
    0.46
    nın
    0.45
    Το
    0.45
    <unused2174>
    0.45
    Act Density 0.456%

    No Known Activations