© 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. 45086
    Prev
    Next
    INDEX
    Explanations

    references to French literature or liberal thought

    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
    gateway
    -0.16
    anga
    -0.15
    plugins
    -0.15
    Gear
    -0.14
    .cloudflare
    -0.14
    .scalajs
    -0.14
    gear
    -0.14
     Gauge
    -0.14
    _identity
    -0.14
     liners
    -0.14
    POSITIVE LOGITS
     gold
    1.06
     Gold
    0.98
    Gold
    0.90
    gold
    0.88
     GOLD
    0.87
    _gold
    0.75
    .gold
    0.71
    éĩij
    0.71
     golden
    0.70
     éĩij
    0.61
    Activations Density 0.009%

    No Known Activations