© 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. 44627
    Prev
    Next
    INDEX
    Explanations

    references to traditional practices or sources of authority in regard to mental health treatment

    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
    opsy
    -0.15
    strand
    -0.14
    Assembler
    -0.14
     ============================================================================↵
    -0.14
    ourg
    -0.13
    صات
    -0.13
    uyu
    -0.13
    hausen
    -0.13
    iasm
    -0.13
    ocop
    -0.13
    POSITIVE LOGITS
     written
    0.55
    written
    0.44
     Written
    0.43
    Written
    0.41
     writing
    0.40
    -written
    0.38
    _written
    0.36
     escrit
    0.34
     пиÑģÑĮмен
    0.34
     écrit
    0.32
    Activations Density 0.020%

    No Known Activations