© 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. 42790
    Prev
    Next
    INDEX
    Explanations

    comparative phrases indicating increase or decrease in attributes or quantities

    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
    oir
    -0.14
    tern
    -0.13
    edException
    -0.13
    BarButton
    -0.13
    ain
    -0.13
    egt
    -0.13
    íĭ´
    -0.13
    æł¡
    -0.13
     buflen
    -0.13
    /Input
    -0.12
    POSITIVE LOGITS
    -and
    0.43
     and
    0.35
    _and
    0.32
     &
    0.27
    åĴĮ
    0.26
     и
    0.26
     vÃł
    0.25
    and
    0.25
    à¹ģละ
    0.24
    &S
    0.23
    Activations Density 0.025%

    No Known Activations