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. 23549
    Prev
    Next
    INDEX
    Explanations

    descriptions of surreal or extraordinary experiences

    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
    .drawString
    -0.17
    WithEmail
    -0.15
    /Dk
    -0.14
    SYS
    -0.13
    upal
    -0.13
     Sexe
    -0.13
    eriod
    -0.13
    upert
    -0.13
    ioxide
    -0.13
    ÑĢозÑĥм
    -0.13
    POSITIVE LOGITS
     scene
    0.62
     sight
    0.60
     scenes
    0.53
     sights
    0.49
     Scene
    0.48
    scene
    0.47
    Scene
    0.43
     Sight
    0.41
     Scenes
    0.40
    scenes
    0.39
    Activations Density 0.031%

    No Known Activations