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. 23491
    Prev
    Next
    INDEX
    Explanations

    technical terms and instructions related to version control and software 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
    izont
    -0.15
    енÑĤи
    -0.14
    ÑĸÑģно
    -0.14
    erti
    -0.14
    ẩm
    -0.14
    -runtime
    -0.13
    ζα
    -0.13
    ÅĻeh
    -0.13
    Äįan
    -0.13
    nelly
    -0.13
    POSITIVE LOGITS
     Git
    0.43
     repository
    0.41
     git
    0.41
     commit
    0.41
     commits
    0.39
     repositories
    0.37
    git
    0.36
     Repository
    0.35
     committing
    0.35
     branch
    0.35
    Activations Density 0.016%

    No Known Activations