© 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. 13065
    Prev
    Next
    INDEX
    Explanations

    acronyms and abbreviations associated with organizations or programs

    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
    ed
    -0.20
    a
    -0.19
    i
    -0.18
    e
    -0.18
    ième
    -0.17
    edReader
    -0.16
    ing
    -0.16
    edBy
    -0.16
    aft
    -0.16
    ity
    -0.16
    POSITIVE LOGITS
    Ùij
    0.16
    tings
    0.15
    mana
    0.14
    ä¸Ķ
    0.14
    xor
    0.14
    prm
    0.14
    rib
    0.14
    usb
    0.14
    indi
    0.14
    代
    0.14
    Activations Density 0.015%

    No Known Activations