© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 25709
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate competition or racing 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
    OutOfBoundsException
    -0.14
     SWAT
    -0.14
    ertoire
    -0.14
     Coh
    -0.14
     Eins
    -0.14
     supplement
    -0.13
     Cohen
    -0.13
     synergy
    -0.13
    semblies
    -0.13
    xit
    -0.13
    POSITIVE LOGITS
     race
    0.91
    race
    0.77
     Race
    0.74
    Race
    0.70
    _race
    0.62
     races
    0.61
     Races
    0.49
     raced
    0.45
     racing
    0.42
     competition
    0.39
    Activations Density 0.051%

    No Known Activations