© 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. 63865
    Prev
    Next
    INDEX
    Explanations

    elements describing musical techniques and styles

    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
    mani
    -0.15
    ampil
    -0.15
    avec
    -0.14
    stor
    -0.14
    enate
    -0.14
    quier
    -0.14
     orchestra
    -0.13
    tering
    -0.13
     loud
    -0.13
    KeyEvent
    -0.13
    POSITIVE LOGITS
     mel
    0.29
     patterns
    0.27
     melodies
    0.27
     Patterns
    0.26
    patterns
    0.24
    mel
    0.24
    Patterns
    0.23
     melody
    0.23
     Mel
    0.23
     phrases
    0.22
    Activations Density 0.026%

    No Known Activations