Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 23614
    Prev
    Next
    INDEX
    Explanations

    phrases related to making impactful plays in a sports context

    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
    egg
    -0.15
     Shoot
    -0.15
    ãĤ§
    -0.15
    TestFixture
    -0.14
    erek
    -0.14
    è¼ķ
    -0.13
     Millet
    -0.13
    iversite
    -0.13
    .Persistent
    -0.13
    539
    -0.13
    POSITIVE LOGITS
     plays
    0.40
    plays
    0.33
     Plays
    0.32
     play
    0.28
     highlight
    0.22
    play
    0.21
     moment
    0.20
    Play
    0.19
     Moment
    0.19
     moments
    0.19
    Activations Density 0.044%

    No Known Activations