© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 23036
    Prev
    Next
    INDEX
    Explanations

    phrases related to the debate on intelligence, race, and ethical considerations in tax practices

    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
     inval
    -0.16
    omik
    -0.15
    Barrier
    -0.14
     Nah
    -0.14
     Dame
    -0.14
    udades
    -0.14
    ereo
    -0.13
    :UI
    -0.13
    uhan
    -0.13
    inati
    -0.13
    POSITIVE LOGITS
     contested
    0.45
     disputed
    0.44
     contest
    0.41
     hot
    0.39
     controversial
    0.35
     debated
    0.34
     disput
    0.34
     dispute
    0.33
     Hot
    0.33
     controversy
    0.33
    Activations Density 0.064%

    No Known Activations