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. 31324
    Prev
    Next
    INDEX
    Explanations

    instances of the word "was."

    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
    Å
    -0.15
    ¼åIJĪ
    -0.14
    ERCHANT
    -0.13
    ade
    -0.13
    ãĥ¼ãĥ³
    -0.13
    keleton
    -0.13
    @Spring
    -0.13
    umer
    -0.13
    marine
    -0.13
    oken
    -0.12
    POSITIVE LOGITS
    ij
    0.17
    yun
    0.14
    .MixedReality
    0.14
    Äħd
    0.14
    tainment
    0.14
    ossal
    0.14
    yd
    0.13
    .defaults
    0.13
    ildo
    0.13
    äd
    0.13
    Activations Density 0.000%

    No Known Activations