© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 41908
    Prev
    Next
    INDEX
    Explanations

    statements about societal norms and expectations regarding behavior and discrimination

    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
    enaire
    -0.19
    ittest
    -0.16
    reste
    -0.14
    /*č↵
    -0.14
    abay
    -0.13
     interven
    -0.13
     Hur
    -0.13
    meld
    -0.13
    .Bounds
    -0.13
    .nr
    -0.13
    POSITIVE LOGITS
     led
    0.23
     guided
    0.20
     haunted
    0.17
     drawn
    0.17
     able
    0.17
     sensitive
    0.17
     influenced
    0.17
     unable
    0.17
     inclined
    0.17
     forced
    0.16
    Activations Density 0.191%

    No Known Activations