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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 13861
    Prev
    Next
    INDEX
    Explanations

    references to regular time intervals for content, such as monthly and weekly updates

    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
     annual
    -0.28
     daily
    -0.28
     annually
    -0.26
     yearly
    -0.24
    annual
    -0.24
     Annual
    -0.24
     Daily
    -0.23
    daily
    -0.23
    Daily
    -0.22
    Annual
    -0.21
    POSITIVE LOGITS
    /month
    0.32
     dose
    0.22
     doses
    0.18
    mente
    0.18
    /year
    0.18
     newsletter
    0.18
     grind
    0.17
     basis
    0.16
    ized
    0.16
    orgetown
    0.16
    Activations Density 0.022%

    No Known Activations