© 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. 42328
    Prev
    Next
    INDEX
    Explanations

    references to locations in Australia

    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
    acket
    -0.14
    /*č↵
    -0.14
    udu
    -0.13
    ãĤ¤ãĥ³ãĥĪ
    -0.13
    owell
    -0.13
    رÛĮÙħ
    -0.13
    oined
    -0.13
    ackets
    -0.13
    levard
    -0.13
    toJson
    -0.12
    POSITIVE LOGITS
     Australian
    0.82
     Australia
    0.82
    Australian
    0.71
    Australia
    0.70
     Australians
    0.69
     Aussie
    0.65
     australia
    0.62
    .au
    0.58
     Austral
    0.57
     Melbourne
    0.56
    Activations Density 0.057%

    No Known Activations