© 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 EvalsExportsAPI 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. 14967
    Prev
    Next
    INDEX
    Explanations

    references to notable individuals and their contributions or actions

    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
    ekk
    -0.17
    iry
    -0.17
    uez
    -0.16
    jej
    -0.15
    ameleon
    -0.14
    omor
    -0.14
    ftar
    -0.14
    omm
    -0.14
    ulus
    -0.13
    urement
    -0.13
    POSITIVE LOGITS
     popular
    0.59
    popular
    0.48
     Popular
    0.47
    Popular
    0.43
     invent
    0.40
     coin
    0.34
     Invent
    0.32
     invention
    0.31
    coin
    0.30
     first
    0.29
    Activations Density 0.077%

    No Known Activations