© 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. 25851
    Prev
    Next
    INDEX
    Explanations

    discussions about reforms and overhauls in policies, particularly related to taxes and immigration

    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
    anh
    -0.15
    اÙĨÙĪ
    -0.14
    bÄĽ
    -0.14
     ><?
    -0.14
    æľī人
    -0.14
    ados
    -0.14
    berman
    -0.14
    eyh
    -0.14
    ä¸Ģ人
    -0.13
    елеÑĦон
    -0.13
    POSITIVE LOGITS
     existing
    0.28
     how
    0.27
     way
    0.24
     Existing
    0.22
     rules
    0.22
     outdated
    0.21
     laws
    0.21
    how
    0.21
    rules
    0.19
     ways
    0.19
    Activations Density 0.062%

    No Known Activations