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. 23674
    Prev
    Next
    INDEX
    Explanations

    statements about the complexity and challenges of various subjects

    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
     itself
    -0.35
    æĺ¯ä¸Ģ个
    -0.15
    çļĦä¸Ģ个
    -0.15
     uveden
    -0.15
     one
    -0.15
     perv
    -0.14
    839
    -0.14
     Ñıке
    -0.14
    wiÄħ
    -0.14
    uto
    -0.14
    POSITIVE LOGITS
     ones
    0.43
     themselves
    0.40
     those
    0.36
    those
    0.32
    ones
    0.31
     Ones
    0.30
     Those
    0.30
    éĤ£äºĽ
    0.29
    Those
    0.29
     denen
    0.27
    Activations Density 0.141%

    No Known Activations