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. 31297
    Prev
    Next
    INDEX
    Explanations

    phrases related to winning or achieving success in competitive contexts

    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
     anymore
    -0.16
    ube
    -0.15
    IMO
    -0.15
    unless
    -0.15
    aim
    -0.14
    à¸Ńà¸ĩ
    -0.14
    wig
    -0.14
     nhằm
    -0.14
     yerde
    -0.14
    cloak
    -0.14
    POSITIVE LOGITS
     thanks
    0.59
    thanks
    0.45
     Thanks
    0.43
    Thanks
    0.40
     gracias
    0.38
     nhá»Ŀ
    0.35
     grâce
    0.34
     благодаÑĢÑı
    0.32
     despite
    0.31
     dÃŃky
    0.29
    Activations Density 0.056%

    No Known Activations