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. 50
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to legal violations and traffic offenses

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    words related to legal issues and law enforcement, particularly traffic violations and criminal charges.

    oai_token-act-pair · gpt-3.5-turboTriggered by @noracai0625
    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
     robber
    -0.14
    ola
    -0.14
     intelligence
    -0.14
     warped
    -0.14
    rello
    -0.14
     coincidence
    -0.14
     Plat
    -0.13
     pud
    -0.13
    लà¤Ĺ
    -0.13
    hurst
    -0.13
    POSITIVE LOGITS
     litter
    0.24
     Speed
    0.20
     Disorder
    0.17
     SPEED
    0.17
     speeding
    0.17
     Grade
    0.17
     disorder
    0.17
    -Speed
    0.17
    Grade
    0.16
    -speed
    0.16
    Activations Density 0.042%

    No Known Activations