© 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. 21933
    Prev
    Next
    INDEX
    Explanations

    words related to adherence to rules or principles

    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
    yz
    -0.13
    ÏĢά
    -0.13
    ĴĪ
    -0.13
    idges
    -0.13
    parate
    -0.12
    ÏĢοÏĤ
    -0.12
     Ùħع
    -0.12
    aad
    -0.12
    FROM
    -0.11
     mám
    -0.11
    POSITIVE LOGITS
     by
    1.16
    _by
    0.79
     oleh
    0.78
    by
    0.77
     By
    0.70
    -by
    0.69
     تÙĪسط
    0.67
     bợi
    0.61
    .by
    0.61
    By
    0.60
    Activations Density 0.050%

    No Known Activations