© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 24465
    Prev
    Next
    INDEX
    Explanations

    terms related to life, death, and bodily harm

    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
    ksen
    -0.16
    ugins
    -0.15
    bum
    -0.14
    à¹Ģà¸ĭà¸Ńร
    -0.14
     Barrett
    -0.14
    .sponge
    -0.14
    BUM
    -0.14
    .tc
    -0.13
     Boris
    -0.13
    ubat
    -0.13
    POSITIVE LOGITS
     blood
    0.77
     Blood
    0.68
    Blood
    0.61
    blood
    0.61
    -blood
    0.57
    è¡Ģ
    0.55
    LOOD
    0.49
    lood
    0.48
     кÑĢови
    0.47
     máu
    0.44
    Activations Density 0.021%

    No Known Activations