© 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. 23113
    Prev
    Next
    INDEX
    Explanations

    quotations and phrases that suggest arguments or critiques regarding government and economic policies

    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
    'gc
    -0.15
    หล
    -0.15
    enny
    -0.14
    inha
    -0.14
    agers
    -0.13
    logan
    -0.13
    spot
    -0.13
    ille
    -0.13
    iasi
    -0.13
    ynom
    -0.13
    POSITIVE LOGITS
    ainment
    0.16
    .pem
    0.15
    gebn
    0.14
    ги
    0.14
     Carry
    0.14
    anik
    0.13
    éŁ
    0.13
     Nim
    0.13
     Vict
    0.13
     carry
    0.13
    Activations Density 0.014%

    No Known Activations