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

    Neuronpedia

    INDEX
    Explanations

    adding 'too', 'anymore', 'either'

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    /
    0.43
    ו
    0.43
    و
    0.36
    ो
    0.36
    </b>
    0.36
    A
    0.36
    </i>
    0.35
    ↵
    0.34
    \"
    0.34
    P
    0.33
    POSITIVE LOGITS
    يل
    0.50
     这里
    0.43
     về
    0.40
     Сред
    0.38
    на
    0.38
    льного
    0.37
     ένα
    0.37
     那
    0.37
    ánea
    0.36
    주고
    0.36
    Act Density 0.572%

    No Known Activations