© 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. 30546
    Prev
    Next
    INDEX
    Explanations

    references to pace and tempo 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
    indow
    -0.18
    ilter
    -0.16
    rooms
    -0.16
    ness
    -0.15
    à¹Ĩ
    -0.15
    horn
    -0.15
     Dude
    -0.15
    lod
    -0.14
    adle
    -0.14
    ties
    -0.14
    POSITIVE LOGITS
    -paced
    0.24
     paced
    0.18
    -setting
    0.18
    /temp
    0.17
    SSIP
    0.17
    yon
    0.16
     pace
    0.16
    ToFit
    0.16
    ometer
    0.16
    Setter
    0.15
    Activations Density 0.033%

    No Known Activations