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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 32
    Prev
    Next
    INDEX
    Explanations

    phrases related to prestigious appointments and professional recognition

    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
    oucher
    -0.18
    าà¸ĸ
    -0.15
    izo
    -0.14
    ledged
    -0.14
    portal
    -0.14
    straction
    -0.13
    atron
    -0.13
    raham
    -0.13
    ostel
    -0.13
    .sul
    -0.12
    POSITIVE LOGITS
     position
    0.83
     positions
    0.77
    position
    0.65
     Position
    0.65
     Positions
    0.63
    positions
    0.60
    Position
    0.57
     posts
    0.55
     POSITION
    0.53
    Positions
    0.53
    Activations Density 0.089%

    No Known Activations