© 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. 420
    Prev
    Next
    Layer 20Layer 22
    INDEX
    Compare
    Explanations
    No Explanations Found
    Auto-Interp Tools
    Top Features by Cosine Similarity
    Embeds
    IFrame
    Link
    Negative Logits
     overhead
    -0.10
    ilarity
    -0.09
    d
    -0.09
     Country
    -0.09
    _exchange
    -0.09
     terms
    -0.09
     Ted
    -0.09
    Src
    -0.09
     concess
    -0.09
     exchanges
    -0.09
    ыш
    -0.09
    ethe
    -0.09
    tone
    -0.09
     Carrier
    -0.09
    diag
    -0.09
     oxid
    -0.09
    elic
    -0.08
    �
    -0.08
    .opend
    -0.08
     scoring
    -0.08
    Positive Logits
    ァ
    0.11
    urn
    0.11
    adows
    0.10
     〉
    0.10
     prose
    0.10
    街道
    0.10
     Academ
    0.10
    dden
    0.10
    _IOC
    0.10
    休
    0.10
    annes
    0.10
     Watkins
    0.09
    umi
    0.09
    pike
    0.09
    sbin
    0.09
     ｉ
    0.09
    _RW
    0.09
     Furn
    0.09
     Nile
    0.09
    aware
    0.09
    Activations Density 57.931%
    Loading dashboard...