© 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. 21865
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal or ethical violations in professional conduct

    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
    äs
    -0.15
    rani
    -0.14
    pron
    -0.14
     mano
    -0.14
    complexType
    -0.14
    lobs
    -0.14
    ümÃ¼ÅŁ
    -0.14
     asn
    -0.14
    æģ¯
    -0.13
    Å¡ÃŃ
    -0.13
    POSITIVE LOGITS
     reasonably
    0.23
    []
    0.16
     or
    0.16
    лиÑĤ
    0.15
     expectancy
    0.15
     nor
    0.14
     Covered
    0.14
    [],
    0.14
     è©
    0.13
    commerce
    0.13
    Activations Density 0.495%

    No Known Activations