© 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. 30319
    Prev
    Next
    INDEX
    Explanations

    phrases related to promotional strategies and media practices

    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
    atar
    -0.14
    μβ
    -0.14
    .scalablytyped
    -0.14
     flush
    -0.13
    asti
    -0.13
    à¹īาà¸Ļà¸Ķ
    -0.13
     or
    -0.13
    quared
    -0.13
    anas
    -0.13
    uty
    -0.12
    POSITIVE LOGITS
     random
    0.34
    random
    0.25
     randomly
    0.24
     RANDOM
    0.24
    Random
    0.23
     Random
    0.23
     arbitrary
    0.23
    _random
    0.22
    -random
    0.21
     stuff
    0.21
    Activations Density 0.115%

    No Known Activations