© 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. 13993
    Prev
    Next
    INDEX
    Explanations

    terms related to imitation and references to quotes or citations

    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
    ddit
    -0.12
    .Undef
    -0.12
    SharedPointer
    -0.12
    .AutoComplete
    -0.11
     epit
    -0.11
     /**<
    -0.11
    (er
    -0.11
     Ä
    -0.11
     pioneered
    -0.11
    guest
    -0.11
    POSITIVE LOGITS
    ÌĨ
    0.15
     volunte
    0.14
    ãĤĩ
    0.14
    Ā
    0.14
     tô
    0.13
    637
    0.13
    ãĤĵãģ©
    0.13
    _BLE
    0.13
    SSERT
    0.13
    IRMWARE
    0.13
    Activations Density 0.000%

    No Known Activations