© 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. 12590
    Prev
    Next
    INDEX
    Explanations

    attributes related to product quality and design elements

    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
    APON
    -0.15
    ì´Ŀ
    -0.15
    ë°į
    -0.14
    .webdriver
    -0.14
    å±±å¸Ĥ
    -0.14
     Harness
    -0.14
    _LD
    -0.14
    lexical
    -0.13
     LENG
    -0.13
    fastcall
    -0.13
    POSITIVE LOGITS
     functionality
    0.47
     functional
    0.47
     function
    0.46
     Function
    0.42
     Functional
    0.41
    function
    0.39
     FUNCTION
    0.39
    Function
    0.38
    functional
    0.36
    -functional
    0.35
    Activations Density 0.070%

    No Known Activations