© 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. 21786
    Prev
    Next
    INDEX
    Explanations

    connections between individuals that indicate strong personal bonds or relationships

    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
    kü
    -0.15
     ancestor
    -0.15
    ancestor
    -0.15
    ichick
    -0.14
    ãĥĶãĥ¼
    -0.14
     colleague
    -0.14
     descendant
    -0.14
    apter
    -0.14
    ddit
    -0.14
    Expired
    -0.14
    POSITIVE LOGITS
     close
    0.87
     Close
    0.73
    close
    0.72
    Close
    0.71
    -close
    0.67
     CLOSE
    0.66
     closest
    0.65
    _close
    0.59
     closer
    0.58
    (close
    0.52
    Activations Density 0.029%

    No Known Activations