© 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. 22971
    Prev
    Next
    INDEX
    Explanations

    phrases indicating conditions or reasons in a statement

    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
     eldre
    -0.17
    yg
    -0.15
    .bits
    -0.14
     NÄĽk
    -0.14
    Ïģιν
    -0.14
    goo
    -0.14
    alers
    -0.14
    ala
    -0.14
    aux
    -0.13
    _supp
    -0.13
    POSITIVE LOGITS
     several
    0.94
     Several
    0.70
    Several
    0.66
     two
    0.58
     plusieurs
    0.55
     three
    0.54
    veral
    0.49
     various
    0.47
    åĩłä¸ª
    0.47
     many
    0.47
    Activations Density 0.119%

    No Known Activations