© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 30209
    Prev
    Next
    INDEX
    Explanations

    political terms related to extremism and anti-immigrant sentiment

    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
    ,eg
    -0.17
    jos
    -0.16
    ¯¼
    -0.15
    lessly
    -0.15
    ligt
    -0.14
    ertas
    -0.14
    надлеж
    -0.14
    ̣
    -0.14
    tti
    -0.14
    ÄĽlÃŃ
    -0.14
    POSITIVE LOGITS
    sonian
    0.16
    HING
    0.15
    ŀæĢ§
    0.14
    ceb
    0.14
    inction
    0.14
    apons
    0.13
    pherd
    0.13
    ilibrium
    0.13
    fbe
    0.13
    thesis
    0.13
    Activations Density 0.055%

    No Known Activations