© 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. 33221
    Prev
    Next
    INDEX
    Explanations

    repetition of the word "Union" in various contexts

    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
    reve
    -0.16
    RACT
    -0.16
    ervice
    -0.15
    phis
    -0.15
    ulla
    -0.14
    plays
    -0.14
    'gc
    -0.14
    ollo
    -0.14
     Äijá»Ļng
    -0.14
    445
    -0.14
    POSITIVE LOGITS
    ization
    0.24
     Carb
    0.23
    ized
    0.22
     Jack
    0.22
    izable
    0.22
    dale
    0.21
    izing
    0.20
     jack
    0.20
    Jack
    0.19
     Station
    0.19
    Activations Density 0.008%

    No Known Activations