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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 23850
    Prev
    Next
    INDEX
    Explanations

    technical specifications and requirements for gaming hardware

    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
    Æ¡
    -0.14
    apult
    -0.14
    ÙĪÛĮÛĮ
    -0.14
    .ir
    -0.14
    heed
    -0.14
    /utility
    -0.13
    xbe
    -0.13
    edo
    -0.13
    preneur
    -0.13
    xda
    -0.13
    POSITIVE LOGITS
     hardware
    0.39
     powerful
    0.37
     Hardware
    0.30
     specs
    0.29
    hardware
    0.29
     Powerful
    0.29
     specifications
    0.28
    Hardware
    0.27
     spec
    0.26
     specification
    0.26
    Activations Density 0.067%

    No Known Activations