© 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. 44965
    Prev
    Next
    INDEX
    Explanations

    references to player performance and injuries in sports

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     plá
    -0.15
    odium
    -0.15
     incentiv
    -0.14
    \Eloquent
    -0.14
     “[
    -0.14
    maal
    -0.14
    â
    -0.14
     loginUser
    -0.13
     Bylo
    -0.13
     whilst
    -0.13
    POSITIVE LOGITS
    .AP
    0.17
    AP
    0.15
    BITS
    0.15
    (strtolower
    0.15
    ustum
    0.15
    _invoke
    0.14
    zon
    0.14
    heed
    0.14
     Scal
    0.14
    .opens
    0.13
    Activations Density 0.017%

    No Known Activations