© 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. 43696
    Prev
    Next
    INDEX
    Explanations

    references to sudden or temporary events

    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
    rej
    -0.18
    PTY
    -0.16
    é¬
    -0.16
    phetamine
    -0.14
     Duy
    -0.14
    east
    -0.14
     PUS
    -0.14
    lexport
    -0.14
    InBackground
    -0.14
    ãĤ¶ãĥ¼
    -0.14
    POSITIVE LOGITS
    backs
    0.26
     flash
    0.26
    point
    0.26
    flash
    0.24
    points
    0.24
    mob
    0.24
     mob
    0.23
    Flash
    0.23
     Flash
    0.23
    cards
    0.22
    Activations Density 0.009%

    No Known Activations