© 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. 43987
    Prev
    Next
    INDEX
    Explanations

    phrases related to communication and organization efforts

    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
     Minds
    -0.21
     Marilyn
    -0.20
     Milton
    -0.19
     Mund
    -0.18
     Manning
    -0.18
    apture
    -0.17
     Molly
    -0.17
     McCoy
    -0.16
     Mold
    -0.16
     Martial
    -0.16
    POSITIVE LOGITS
     message
    0.76
     messages
    0.63
    message
    0.60
     Message
    0.56
    -message
    0.54
    _message
    0.53
    Message
    0.52
     mess
    0.52
    .message
    0.51
     mes
    0.50
    Activations Density 0.026%

    No Known Activations