© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 45224
    Prev
    Next
    INDEX
    Explanations

    references to reparations and justice for historical grievances

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
    igham
    -0.17
    rael
    -0.14
    à¸µà¸Ľ
    -0.14
    egie
    -0.14
     emergencies
    -0.14
    ernes
    -0.14
    arendra
    -0.14
    flen
    -0.14
     Patron
    -0.14
     crises
    -0.13
    POSITIVE LOGITS
     compensation
    0.58
     Compensation
    0.53
     compens
    0.51
     compensate
    0.48
     compensated
    0.42
     repar
    0.42
     restitution
    0.35
     apologies
    0.28
     apology
    0.28
     justice
    0.26
    Activations Density 0.028%

    No Known Activations