© 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. 13246
    Prev
    Next
    INDEX
    Explanations

    concepts related to energy generation and entropy in the universe

    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
     Atlantis
    -0.15
    abal
    -0.14
     facial
    -0.14
    丶
    -0.14
    bler
    -0.14
     Leak
    -0.14
     vole
    -0.14
    λμ
    -0.13
     airborne
    -0.13
    ,buf
    -0.13
    POSITIVE LOGITS
     stars
    0.35
     star
    0.29
     Stars
    0.23
    stars
    0.23
    Stars
    0.23
     stellar
    0.22
    -stars
    0.21
    stellar
    0.20
     fusion
    0.20
    -star
    0.19
    Activations Density 0.056%

    No Known Activations