© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 40995
    Prev
    Next
    INDEX
    Explanations

    references to favoritism in legal or social contexts regarding treatment of individuals based on their status or beliefs

    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
    ackle
    -0.14
     Reputation
    -0.13
     reput
    -0.13
    quared
    -0.13
    ึ
    -0.13
    498
    -0.13
    uÄį
    -0.13
    esin
    -0.13
    erli
    -0.13
    rak
    -0.12
    POSITIVE LOGITS
     special
    0.51
     privileges
    0.45
     privileged
    0.43
     privilege
    0.41
     prefer
    0.41
    special
    0.39
    ç®Ĭ
    0.37
     SPECIAL
    0.35
     treatment
    0.34
     Priv
    0.34
    Activations Density 0.052%

    No Known Activations