© 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. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 45016
    Prev
    Next
    INDEX
    Explanations

    instances of names being mentioned, particularly in relation to calls or communication

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    /browse
    -0.16
    lej
    -0.15
    eder
    -0.15
     beden
    -0.15
    byt
    -0.15
    POINTS
    -0.14
    vc
    -0.14
    inas
    -0.14
    uen
    -0.14
    adera
    -0.14
    POSITIVE LOGITS
     collect
    0.20
    DonaldTrump
    0.16
    ©
    0.16
    undef
    0.15
     names
    0.15
    è
    0.15
     directly
    0.15
     late
    0.15
     Collect
    0.14
     personally
    0.14
    Activations Density 0.036%

    No Known Activations