© 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. 22538
    Prev
    Next
    INDEX
    Explanations

    themes related to social manners and acceptable behavior

    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
    127
    -0.13
     Toy
    -0.13
    arel
    -0.13
    olla
    -0.13
    argins
    -0.13
     fis
    -0.12
    andon
    -0.12
    ัà¸Ļà¸ĺ
    -0.12
    605
    -0.12
    angu
    -0.12
    POSITIVE LOGITS
     courtesy
    0.54
     Courtesy
    0.47
    Courtesy
    0.43
     civ
    0.42
     etiquette
    0.42
     manners
    0.41
    cour
    0.40
     courteous
    0.40
     civil
    0.39
     polit
    0.39
    Activations Density 0.044%

    No Known Activations