© 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. 28714
    Prev
    Next
    INDEX
    Explanations

    elements related to shelter and outdoor settings

    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
    år
    -0.15
    oki
    -0.14
    anity
    -0.14
    arena
    -0.14
    mund
    -0.13
    หว
    -0.13
    isser
    -0.13
    åŃĿ
    -0.13
    ean
    -0.12
    AYOUT
    -0.12
    POSITIVE LOGITS
     shade
    0.84
     shadow
    0.83
     Shade
    0.73
    shadow
    0.71
     Shadow
    0.70
     shadows
    0.70
    Shadow
    0.67
    shade
    0.64
     shading
    0.60
    -shadow
    0.59
    Activations Density 0.014%

    No Known Activations