© 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 EvalsExportsAPI 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. 28133
    Prev
    Next
    INDEX
    Explanations

    phrases related to regulatory actions and oversight processes

    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
    igaret
    -0.14
    strap
    -0.14
    æľĽ
    -0.13
    .actor
    -0.13
    kle
    -0.13
    _fwd
    -0.13
    ιÏĥÏĦή
    -0.13
     >č↵
    -0.12
    Forward
    -0.12
    istro
    -0.12
    POSITIVE LOGITS
     behind
    1.18
     Behind
    0.98
    Behind
    0.87
    beh
    0.71
     achter
    0.56
     underlying
    0.51
     backend
    0.50
     backstage
    0.50
     پشت
    0.49
     Backend
    0.43
    Activations Density 0.022%

    No Known Activations