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

    Neuronpedia

    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. 55229
    Prev
    Next
    INDEX
    Explanations

    terms and concepts related to memory allocation in programming, specifically in the context of game development

    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
     slow
    -0.16
     Slow
    -0.15
    assic
    -0.14
    Slow
    -0.13
    ansom
    -0.13
    ieren
    -0.13
    ipo
    -0.13
    omain
    -0.12
    اÙĪÙĨد
    -0.12
    fon
    -0.12
    POSITIVE LOGITS
     temporary
    1.02
     Temporary
    0.89
    temporary
    0.82
    Temporary
    0.81
     temporarily
    0.72
     TEMP
    0.71
     temp
    0.70
     transient
    0.70
     tempor
    0.65
     Temp
    0.65
    Activations Density 0.015%

    No Known Activations