© 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. 8447
    Prev
    Next
    INDEX
    Explanations

    numerical references and statistics

    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
    ãģĦãĤĭ
    -0.19
    iating
    -0.17
    ention
    -0.16
    ennen
    -0.15
    ooke
    -0.15
    otel
    -0.15
    aney
    -0.15
    /groups
    -0.15
    Johnny
    -0.15
    olson
    -0.15
    POSITIVE LOGITS
    719
    0.20
    ture
    0.19
    ãģ°
    0.17
    azzo
    0.17
    /player
    0.16
    datable
    0.15
                ↵
    0.15
    pered
    0.15
    stery
    0.14
    leys
    0.14
    Activations Density 0.065%

    No Known Activations