© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 13894
    Prev
    Next
    INDEX
    Explanations

    concepts related to artificial intelligence and design

    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
    sj
    -0.14
    -devel
    -0.14
    .jquery
    -0.14
    agal
    -0.13
     âĨĴ↵↵
    -0.13
    _accessible
    -0.13
    etik
    -0.13
     sniper
    -0.13
    obody
    -0.13
    ;!
    -0.13
    POSITIVE LOGITS
     systems
    0.35
     machines
    0.28
     technology
    0.28
    systems
    0.27
     Systems
    0.26
     devices
    0.25
    Systems
    0.23
    ystems
    0.22
    technology
    0.22
     technologies
    0.22
    Activations Density 0.030%

    No Known Activations