© 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. 45009
    Prev
    Next
    INDEX
    Explanations

    references to quantities and concepts related to measurement or classification

    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
     lost
    -0.19
     Lost
    -0.18
     taken
    -0.17
    Lost
    -0.16
     removed
    -0.16
    lost
    -0.16
    est
    -0.16
    bal
    -0.16
     satisfied
    -0.15
    en
    -0.15
    POSITIVE LOGITS
    icensed
    0.24
    aged
    0.20
    AGED
    0.20
    armed
    0.20
     plated
    0.20
     heavily
    0.20
    enced
    0.20
    (encoded
    0.19
    .Typed
    0.19
     gated
    0.19
    Activations Density 0.011%

    No Known Activations