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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 24888
    Prev
    Next
    INDEX
    Explanations

    references to music equipment and technology related to DJing and music production

    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
    _COD
    -0.15
    alace
    -0.15
     tiener
    -0.15
     CHIP
    -0.15
    ãĤ«ãĥ¼
    -0.14
    unday
    -0.14
     kola
    -0.14
    çĹ
    -0.14
    _OID
    -0.14
    odom
    -0.14
    POSITIVE LOGITS
     gear
    0.26
     pedals
    0.24
     pedal
    0.24
     Roland
    0.24
     synth
    0.22
     synthes
    0.22
     studio
    0.21
     syn
    0.21
    AES
    0.20
     analog
    0.20
    Activations Density 0.024%

    No Known Activations