© 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. 25559
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, specifically periods

    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
    Ø¢Ùħ
    -0.14
    âĨIJ
    -0.14
    "use
    -0.13
     Uhr
    -0.13
    .Messaging
    -0.12
     sond
    -0.12
    incl
    -0.12
     âĢĮ
    -0.12
    лин
    -0.12
    #ad
    -0.11
    POSITIVE LOGITS
     âĢ¢
    0.15
    inel
    0.15
    0.15
     destinationViewController
    0.13
    Answer
    0.13
    861
    0.13
     def
    0.13
    HeaderCode
    0.13
    ifact
    0.13
    0.13
    Activations Density 0.575%

    No Known Activations