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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 23128
    Prev
    Next
    INDEX
    Explanations

    references to legal disputes and enforcement actions related to copyright issues

    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
    ÙģÙĪ
    -0.14
    presso
    -0.14
    ousing
    -0.14
    _PROC
    -0.14
    ãģİ
    -0.13
    Ø·ØŃ
    -0.13
    κÏģα
    -0.13
    que
    -0.13
    ÏĢον
    -0.13
     Disqus
    -0.13
    POSITIVE LOGITS
     piracy
    0.36
     pirate
    0.34
     pir
    0.33
     Pirate
    0.33
    pir
    0.30
     pirates
    0.30
     copyright
    0.30
     Pir
    0.29
     Pirates
    0.28
     infr
    0.27
    Activations Density 0.049%

    No Known Activations