© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 12326
    Prev
    Next
    INDEX
    Explanations

    comparisons and evaluations related to performance and speed

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
    azon
    -0.18
     Funk
    -0.15
    ersions
    -0.15
    ุà¸Ķà¸Ĺ
    -0.14
    ifen
    -0.13
    ynchronized
    -0.13
    sg
    -0.13
    PÅĻi
    -0.13
    ÑijÑĢ
    -0.12
     ÙĩÙħÛĮÙĨ
    -0.12
    POSITIVE LOGITS
     by
    0.66
     oleh
    0.48
    by
    0.42
     تÙĪسط
    0.40
     far
    0.39
    _by
    0.39
     bợi
    0.37
     By
    0.37
    -by
    0.36
    far
    0.34
    Activations Density 0.043%

    No Known Activations