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. 62
    Prev
    Next
    INDEX
    Explanations

    concepts related to blockchain technology and its applications

    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
    arend
    -0.16
     olmaz
    -0.15
    _INCLUDED
    -0.15
    _paragraph
    -0.15
    lest
    -0.15
    EEK
    -0.14
    uder
    -0.14
    anka
    -0.14
     adipiscing
    -0.14
    à¥Īद
    -0.14
    POSITIVE LOGITS
     holds
    0.47
     hold
    0.46
     could
    0.44
    holds
    0.39
     Holds
    0.39
     Hold
    0.38
     offers
    0.37
     promise
    0.37
     holding
    0.37
    hold
    0.36
    Activations Density 0.127%

    No Known Activations