© 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. 25892
    Prev
    Next
    INDEX
    Explanations

    phrases describing injuries and physical contact in sports 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
     curtains
    -0.17
    acha
    -0.17
    allon
    -0.16
    lr
    -0.16
    çĢ
    -0.16
    ohana
    -0.15
    é¹
    -0.15
     Yuan
    -0.15
     hoops
    -0.14
     Haj
    -0.14
    POSITIVE LOGITS
     pads
    0.29
     helmet
    0.28
     fac
    0.24
    hel
    0.23
     cle
    0.22
     pad
    0.22
     helmets
    0.21
    Helmet
    0.20
     padding
    0.20
     Helmet
    0.20
    Activations Density 0.064%

    No Known Activations