© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 9191
    Prev
    Next
    INDEX
    Explanations

    mentions of indie artists and their work

    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
     Legendary
    -0.17
    Legendary
    -0.16
     legendary
    -0.16
    DVD
    -0.15
    oppins
    -0.14
    ocus
    -0.14
     legacy
    -0.14
    legend
    -0.14
     immortal
    -0.14
    qua
    -0.14
    POSITIVE LOGITS
     electro
    0.20
     folk
    0.20
     trap
    0.19
     electronic
    0.18
     breakout
    0.18
     NPR
    0.18
     indie
    0.17
    alternative
    0.17
     alternative
    0.17
     independently
    0.16
    Activations Density 0.036%

    No Known Activations