© 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. 24460
    Prev
    Next
    INDEX
    Explanations

    references to potential problems or issues arising in various contexts

    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
    ìĶĢ
    -0.19
     Cuando
    -0.16
     ìĸ¸ìłľ
    -0.15
     WHEN
    -0.14
     Whenever
    -0.14
    ãģ¨ãģį
    -0.14
    iedy
    -0.14
    anneer
    -0.14
     Quando
    -0.14
    ForObject
    -0.14
    POSITIVE LOGITS
     if
    0.80
    if
    0.52
     If
    0.50
     nếu
    0.49
     еÑģли
    0.49
    	if
    0.47
    If
    0.45
     jika
    0.45
     kdyby
    0.44
     had
    0.42
    Activations Density 0.089%

    No Known Activations