© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Neuronpedia · MLP Neurons
    5. Llama3.3-70B-IT
    7. MLP Neurons - Layer 21
    9. 346
    Prev
    Next
    Layer 20Layer 22
    INDEX
    Compare
    Explanations
    No Explanations Found
    Auto-Interp Tools
    Top Features by Cosine Similarity
    Embeds
    IFrame
    Link
    Negative Logits
    /*
↵
    -0.11
    amodel
    -0.11
    504
    -0.11
    483
    -0.10
     Opr
    -0.10
    =%.
    -0.10
    pillar
    -0.10
    cala
    -0.10
     fisse
    -0.10
    mia
    -0.10
     Tome
    -0.10
    ennes
    -0.09
     opr
    -0.09
    adders
    -0.09
     Fountain
    -0.09
     paramet
    -0.09
    rior
    -0.09
    aminer
    -0.09
    цо
    -0.09
    _DC
    -0.09
    Positive Logits
     memor
    0.11
    Type
    0.10
     Rin
    0.10
     ul
    0.10
    locker
    0.09
    亡
    0.09
    ayas
    0.09
    tern
    0.09
     sacred
    0.09
     Stern
    0.09
     cous
    0.09
     voc
    0.09
    uler
    0.09
     dumb
    0.09
    败
    0.09
     Gong
    0.09
    yi
    0.09
     locker
    0.09
    odore
    0.09
     lev
    0.09
    Activations Density 52.839%
    Loading dashboard...