© 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. 63464
    Prev
    Next
    INDEX
    Explanations

    instances of societal conflict and violence

    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
    ãģ¡ãģ¯
    -0.18
    ÑĮомÑĥ
    -0.14
    uel
    -0.14
    uckle
    -0.13
    inton
    -0.13
    ãģ£ãģį
    -0.13
    òn
    -0.13
    look
    -0.13
    zin
    -0.13
    ãĢĤãģĤ
    -0.13
    POSITIVE LOGITS
     when
    0.46
     upon
    0.40
    when
    0.38
     When
    0.36
    When
    0.34
     Upon
    0.34
     cuando
    0.33
    Upon
    0.32
     quando
    0.31
    upon
    0.31
    Activations Density 0.197%

    No Known Activations