© 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. 53920
    Prev
    Next
    INDEX
    Explanations

    the conjunctions and phrases that indicate continuous or connected actions and states

    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
    åĪĢ
    -0.15
    ohl
    -0.15
    wer
    -0.14
    itudes
    -0.14
    acz
    -0.14
    avian
    -0.14
    amil
    -0.14
    instein
    -0.13
    ãģĻãģĻ
    -0.13
    aub
    -0.13
    POSITIVE LOGITS
     week
    0.37
    week
    0.30
    -week
    0.23
     night
    0.23
     weeks
    0.23
     недели
    0.23
     Week
    0.23
     reck
    0.22
     month
    0.21
    .week
    0.21
    Activations Density 0.013%

    No Known Activations