© 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. 21709
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, specifically periods and commas

    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
    "
    -0.13
    'n
    -0.13
    -m
    -0.12
     WithEvents
    -0.12
    lok
    -0.12
    ``↵
    -0.11
    ist
    -0.11
    æĸ¹éĿ¢
    -0.11
     ìĹĨìĿ´
    -0.11
     kep
    -0.11
    POSITIVE LOGITS
    .↵↵
    0.16
    ;↵↵
    0.15
    GenerationStrategy
    0.14
    ↵↵
    0.13
    ](
    0.13
    GuidId
    0.13
    undos
    0.12
     stag
    0.12
    .Invariant
    0.12
    racat
    0.12
    Activations Density 1.354%

    No Known Activations