© 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. 21758
    Prev
    Next
    INDEX
    Explanations

    references to significant works, contributions, or legacies in various fields

    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
    ìĽĶë¶ĢíĦ°
    -0.15
    trap
    -0.14
    ábado
    -0.14
    /*č↵
    -0.14
    eeper
    -0.14
    felt
    -0.13
    illac
    -0.13
    appen
    -0.13
    noticed
    -0.13
    åł¡
    -0.13
    POSITIVE LOGITS
     still
    0.39
     remains
    0.36
     remain
    0.31
    still
    0.31
     continues
    0.31
    Still
    0.30
     continue
    0.30
     Still
    0.29
     became
    0.29
     STILL
    0.29
    Activations Density 0.252%

    No Known Activations