© 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. 25843
    Prev
    Next
    INDEX
    Explanations

    specific days and times referenced in events or occurrences

    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
     Christmas
    -0.16
     Days
    -0.15
    δά
    -0.14
     nowhere
    -0.14
    .GetObject
    -0.14
    enda
    -0.14
     stil
    -0.14
     Pace
    -0.14
     conc
    -0.14
    aceous
    -0.13
    POSITIVE LOGITS
    duk
    0.16
    æĹ¥ãģ®
    0.15
     afternoon
    0.15
    ctor
    0.15
    itta
    0.15
     morning
    0.14
    ÅĤug
    0.14
    rupa
    0.14
    æijĺè¦ģ
    0.14
     marked
    0.14
    Activations Density 0.032%

    No Known Activations