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. 23363
    Prev
    Next
    INDEX
    Explanations

    phrases indicating time, specifically referring to events or occasions

    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
    erosis
    -0.17
    adele
    -0.15
    ÑħÑĥ
    -0.15
     çı
    -0.15
    _accessible
    -0.14
    atk
    -0.14
    urma
    -0.14
    ÑĭÑģ
    -0.14
    EdgeInsets
    -0.14
    iaÅĤa
    -0.14
    POSITIVE LOGITS
     purposes
    0.31
     reasons
    0.28
    aging
    0.25
     sake
    0.24
     Reasons
    0.20
    king
    0.19
    nic
    0.18
     whom
    0.18
    -profit
    0.18
     what
    0.17
    Activations Density 0.065%

    No Known Activations