© 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. 42744
    Prev
    Next
    INDEX
    Explanations

    instances of humor involving social commentary

    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
     dang
    -0.16
     darn
    -0.16
    Mom
    -0.15
     Refriger
    -0.15
    oren
    -0.15
    Damn
    -0.14
     Mom
    -0.14
    .generic
    -0.14
     refrigerator
    -0.14
     Color
    -0.14
    POSITIVE LOGITS
     mate
    0.37
     mates
    0.34
     proper
    0.30
     blo
    0.29
    mate
    0.27
    Oi
    0.26
     Proper
    0.25
    blo
    0.23
    proper
    0.23
     nonce
    0.22
    Activations Density 0.097%

    No Known Activations