© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 15556
    Prev
    Next
    INDEX
    Explanations

    references to discussions, feedback, and strategic actions in a context of governance or decision-making

    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
    Ïħγ
    -0.16
    ceed
    -0.14
    iard
    -0.14
    gu
    -0.14
     Everest
    -0.14
    åħ¥ãĤĬ
    -0.13
    gua
    -0.13
     fick
    -0.13
    acom
    -0.13
    dling
    -0.13
    POSITIVE LOGITS
    ÄĻ
    0.18
     already
    0.16
    Already
    0.15
    _ALREADY
    0.15
    Ù
    0.14
    ValueCollection
    0.14
    /tiny
    0.14
    _gener
    0.14
    ElementException
    0.14
     first
    0.14
    Activations Density 0.046%

    No Known Activations