© 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 EvalsExportsinterp-engineNEWAPI 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. 13120
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to legal actions and complaints

    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
    _REQ
    -0.14
    ollo
    -0.14
    ümÃ¼ÅŁ
    -0.14
    otron
    -0.14
     rol
    -0.14
     Jacket
    -0.13
     moderated
    -0.13
     Legion
    -0.13
    acman
    -0.13
     Auditor
    -0.13
    POSITIVE LOGITS
     FIR
    0.26
     complaint
    0.22
     registered
    0.20
     charge
    0.20
     station
    0.19
    registered
    0.18
     Registered
    0.18
     Station
    0.18
     stations
    0.18
     register
    0.18
    Activations Density 0.032%

    No Known Activations