© 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. 15561
    Prev
    Next
    INDEX
    Explanations

    phrases indicating poor performance or negative outcomes

    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
    çݲ
    -0.16
    ÐIJÑĢÑħÑĸв
    -0.16
    रल
    -0.16
    pios
    -0.16
    rej
    -0.15
    emade
    -0.15
    æĮĻ
    -0.14
    ladu
    -0.14
    ÑĨик
    -0.14
    Å¡ek
    -0.14
    POSITIVE LOGITS
     start
    0.44
     slow
    0.37
     starts
    0.35
    start
    0.33
     started
    0.32
     Start
    0.31
    Start
    0.31
    -start
    0.31
     slower
    0.30
    Slow
    0.29
    Activations Density 0.010%

    No Known Activations