© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 24015
    Prev
    Next
    INDEX
    Explanations

    words related to the functionality and performance of programming languages and components

    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
    opsis
    -0.15
    umann
    -0.15
    ALES
    -0.15
    ritte
    -0.14
    apolis
    -0.14
    rees
    -0.14
    arel
    -0.14
    otts
    -0.13
    QRST
    -0.13
    elop
    -0.13
    POSITIVE LOGITS
     fine
    0.27
     correctly
    0.23
    fine
    0.21
     with
    0.20
     properly
    0.20
    orrect
    0.20
     without
    0.20
     correct
    0.20
     transparent
    0.18
    correct
    0.17
    Activations Density 0.055%

    No Known Activations