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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 54381
    Prev
    Next
    INDEX
    Explanations

    references to image hosting websites and URLs

    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
    ร
    -0.16
    enne
    -0.15
    arty
    -0.15
    fully
    -0.14
    htdocs
    -0.14
     Äijoạn
    -0.14
    uyá»ĩt
    -0.14
    arness
    -0.14
    rine
    -0.14
    Rated
    -0.13
    POSITIVE LOGITS
    ur
    0.24
    redients
    0.20
     src
    0.19
    orthand
    0.18
    urs
    0.17
    unks
    0.17
    aes
    0.17
    agers
    0.17
    ix
    0.16
    mers
    0.16
    Activations Density 0.003%

    No Known Activations