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. 23180
    Prev
    Next
    INDEX
    Explanations

    phrases indicating roles and responsibilities within a structured organization or event

    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
    ots
    -0.14
    iselect
    -0.14
    _fault
    -0.14
    ÏĢει
    -0.14
    owers
    -0.13
    OTS
    -0.13
    -*-
    -0.13
    uien
    -0.13
    utura
    -0.13
    ients
    -0.13
    POSITIVE LOGITS
     leadership
    0.36
     roles
    0.35
     role
    0.34
     leading
    0.33
     starring
    0.29
     Leadership
    0.29
    leading
    0.29
     lead
    0.29
     Leading
    0.29
     key
    0.28
    Activations Density 0.058%

    No Known Activations