© 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. 44394
    Prev
    Next
    INDEX
    Explanations

    discussions about the complexities and challenges of regulation and functionality within organizations or agencies

    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
    aign
    -0.20
    esub
    -0.16
    illac
    -0.15
    contri
    -0.14
    inish
    -0.14
    еÑģи
    -0.14
     mimo
    -0.14
    iphers
    -0.13
    undy
    -0.13
    igin
    -0.13
    POSITIVE LOGITS
     balance
    0.68
     Balance
    0.55
    balance
    0.54
     balances
    0.54
     balancing
    0.53
    Balance
    0.52
     balanced
    0.47
     Bal
    0.43
    _balance
    0.42
    balances
    0.40
    Activations Density 0.096%

    No Known Activations