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. 23505
    Prev
    Next
    INDEX
    Explanations

    instances of the word "naive" and its variations

    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
    uing
    -0.16
    .opend
    -0.15
    olf
    -0.15
    esar
    -0.15
    eree
    -0.15
    ricks
    -0.15
    Ìģ
    -0.14
     sund
    -0.14
    auled
    -0.14
    .bd
    -0.14
    POSITIVE LOGITS
    ive
    0.19
    IVE
    0.18
    yer
    0.18
    ivé
    0.17
    ve
    0.17
     Zot
    0.16
    ively
    0.16
    ivet
    0.16
    shire
    0.16
    creativecommons
    0.16
    Activations Density 0.001%

    No Known Activations