© 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. 22045
    Prev
    Next
    INDEX
    Explanations

    statements related to international defense and cooperation

    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
    aret
    -0.16
    ué
    -0.16
    ascript
    -0.15
    :CGRect
    -0.15
     polit
    -0.14
    udeau
    -0.14
    stdin
    -0.14
     Union
    -0.14
     ebp
    -0.14
    é¤Ĭ
    -0.14
    POSITIVE LOGITS
     diplom
    0.17
     isol
    0.16
     deter
    0.16
     ourselves
    0.16
    rip
    0.15
     princip
    0.15
    UGH
    0.15
    imb
    0.15
     ours
    0.15
     partners
    0.15
    Activations Density 0.105%

    No Known Activations