© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 22685
    Prev
    Next
    INDEX
    Explanations

    references to rights and legal entitlements

    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
    orus
    -0.14
    ëŀ
    -0.14
    ÏĢί
    -0.14
    orian
    -0.14
    jev
    -0.14
    eniz
    -0.14
    oulos
    -0.13
    _WEB
    -0.13
     rencont
    -0.13
    ori
    -0.13
    POSITIVE LOGITS
     right
    1.13
    right
    0.91
     RIGHT
    0.87
     Right
    0.83
    Right
    0.79
    -right
    0.75
    _right
    0.69
    .right
    0.68
    RIGHT
    0.65
    	right
    0.64
    Activations Density 0.049%

    No Known Activations