© 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. 14433
    Prev
    Next
    INDEX
    Explanations

    references to sports defenses and performance metrics

    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
    leftright
    -0.15
    ationToken
    -0.14
    abilia
    -0.14
    lamaz
    -0.14
    大åħ¨
    -0.14
    DECL
    -0.14
     peÅŁ
    -0.13
    μά
    -0.13
     hurd
    -0.13
    UnderTest
    -0.13
    POSITIVE LOGITS
     defense
    0.28
     defence
    0.24
     Defence
    0.23
     sting
    0.23
     shutting
    0.22
     shutdown
    0.22
     stout
    0.22
     shut
    0.22
     Defense
    0.21
     leak
    0.21
    Activations Density 0.050%

    No Known Activations