© 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. 22147
    Prev
    Next
    INDEX
    Explanations

    conjunctions and auxiliary verbs indicating relationships or actions in context

    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
    âĢ¦.↵↵
    -0.13
     hete
    -0.11
    âĢ¦âĢ¦âĢ¦âĢ¦
    -0.10
    ......↵↵
    -0.10
    .DropDown
    -0.10
    âĢ¦âĢ¦↵↵
    -0.10
    âĢ¦"↵↵
    -0.10
     fitte
    -0.10
    алÑİ
    -0.10
    ----</
    -0.10
    POSITIVE LOGITS
    ĶåĽŀ
    0.10
    UILTIN
    0.10
    _HERSHEY
    0.09
    WebHost
    0.09
    ÙĪاء
    0.08
    ableOpacity
    0.08
    ¢åįķ
    0.08
    ocommerce
    0.08
    -cols
    0.08
    UpInside
    0.08
    Activations Density 0.000%

    No Known Activations