© 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. 29772
    Prev
    Next
    INDEX
    Explanations

    significant moments and events related to time

    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
    uka
    -0.17
    zb
    -0.14
    mez
    -0.13
    ominated
    -0.13
     prob
    -0.13
    æĬ±
    -0.13
    instr
    -0.13
     Verfügung
    -0.13
    uku
    -0.12
    ekt
    -0.12
    POSITIVE LOGITS
     marks
    0.74
     mark
    0.67
    mark
    0.62
    marks
    0.60
     marking
    0.58
     Marks
    0.56
     marked
    0.53
    Marks
    0.50
    marked
    0.47
    _marks
    0.46
    Activations Density 0.039%

    No Known Activations