© 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. 25284
    Prev
    Next
    INDEX
    Explanations

    demonstrative pronouns and their contexts

    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
     Solve
    -0.20
    istrovstvÃŃ
    -0.16
    eyim
    -0.14
    bÄĽh
    -0.14
    ooth
    -0.13
    auce
    -0.13
    rette
    -0.12
    λοι
    -0.12
     ìĿ´ë¥¼
    -0.12
    thon
    -0.12
    POSITIVE LOGITS
     kind
    1.13
     kinds
    0.99
     type
    0.94
    kind
    0.85
     sort
    0.84
     sorts
    0.78
     types
    0.76
     KIND
    0.71
    type
    0.68
     kinda
    0.68
    Activations Density 0.061%

    No Known Activations