© 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 EvalsExportsAPI 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. 42911
    Prev
    Next
    INDEX
    Explanations

    actions related to fastening or securing objects, particularly tying knots

    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
    ól
    -0.17
    alled
    -0.15
    rieg
    -0.15
    lined
    -0.14
     voks
    -0.14
    geh
    -0.14
    bakan
    -0.14
    leh
    -0.14
    295
    -0.13
    виÑĩ
    -0.13
    POSITIVE LOGITS
     knot
    0.57
     knots
    0.56
     tie
    0.51
     Knot
    0.50
     tying
    0.48
     tied
    0.47
     ties
    0.46
     Tie
    0.42
    tie
    0.37
    .tie
    0.31
    Activations Density 0.010%

    No Known Activations