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. 23548
    Prev
    Next
    INDEX
    Explanations

    mention of cartoons and comics

    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
     Bale
    -0.15
    _ENSURE
    -0.15
     Ale
    -0.15
     PROM
    -0.15
    sei
    -0.15
    æĸĻ
    -0.15
    :System
    -0.14
    rote
    -0.14
    eners
    -0.14
    foy
    -0.14
    POSITIVE LOGITS
     strip
    0.34
     strips
    0.32
    -strip
    0.30
     synd
    0.29
    strip
    0.28
    _strip
    0.27
     Strip
    0.27
     Synd
    0.25
    Strip
    0.24
    .strip
    0.23
    Activations Density 0.035%

    No Known Activations