© 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. 43838
    Prev
    Next
    INDEX
    Explanations

    instances of the word "back" in various contexts

    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
     backgrounds
    -0.19
    -background
    -0.19
     background
    -0.18
    background
    -0.18
    Background
    -0.17
    èĥĮæĻ¯
    -0.17
     backed
    -0.17
    regor
    -0.17
     backstory
    -0.17
     backing
    -0.17
    POSITIVE LOGITS
    wards
    0.31
    slash
    0.26
    side
    0.25
    slashes
    0.23
    door
    0.22
    dated
    0.22
    ronym
    0.22
    kk
    0.20
    gam
    0.20
    ero
    0.20
    Activations Density 0.044%

    No Known Activations