© 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. 63435
    Prev
    Next
    INDEX
    Explanations

    phrases related to publication rights and restrictions

    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
    err
    -0.15
     nay
    -0.14
    iber
    -0.14
    ieri
    -0.14
    loe
    -0.14
    rous
    -0.13
    kd
    -0.13
    istical
    -0.13
     userAgent
    -0.13
    ernet
    -0.13
    POSITIVE LOGITS
     or
    0.16
    up
    0.15
    .clients
    0.15
     Heller
    0.14
    coma
    0.14
    enden
    0.14
    uilt
    0.13
     Fle
    0.13
    eor
    0.13
    à¸Ŀ
    0.13
    Activations Density 0.159%

    No Known Activations