© 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. 42659
    Prev
    Next
    INDEX
    Explanations

    references to the act of watching or viewing content

    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
    .DropDown
    -0.18
    æĮ¯
    -0.17
    isle
    -0.16
    ures
    -0.16
    ander
    -0.16
    enaire
    -0.16
    .scalablytyped
    -0.15
    лÑıÑħ
    -0.15
    echan
    -0.15
    uder
    -0.15
    POSITIVE LOGITS
    tower
    0.34
    /list
    0.29
    ful
    0.28
    ?v
    0.25
    fulness
    0.24
     closely
    0.24
    dog
    0.22
    amac
    0.22
    s
    0.20
    t
    0.20
    Activations Density 0.030%

    No Known Activations