Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 15858
    Prev
    Next
    INDEX
    Explanations

    references to significant events or changes in media coverage

    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
     lign
    -0.15
    ritten
    -0.15
     Duis
    -0.15
     Sesso
    -0.15
    resher
    -0.14
    ahoma
    -0.14
    /***/
    -0.14
    eyse
    -0.14
    .IContainer
    -0.14
    .ISupportInitialize
    -0.14
    POSITIVE LOGITS
     network
    0.46
     networks
    0.44
     cable
    0.41
     Networks
    0.36
    network
    0.35
     Network
    0.34
    -network
    0.34
    Network
    0.33
     Cable
    0.32
     NETWORK
    0.32
    Activations Density 0.067%

    No Known Activations