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

    Neuronpedia

    INDEX
    Explanations

    transition words introducing conclusions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     isotherms
    0.22
     scler
    0.21
     decryption
    0.20
     restraints
    0.20
     axles
    0.20
     charms
    0.20
     prohibits
    0.19
     painkillers
    0.19
     anses
    0.19
     <
    0.19
    POSITIVE LOGITS
    с
    0.26
    ik
    0.25
     we
    0.24
    我們
    0.24
     it
    0.23
    我
    0.23
    什么
    0.22
    í
    0.22
    스
    0.22
    我的
    0.21
    Act Density 0.871%

    No Known Activations