© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 15534
    Prev
    Next
    INDEX
    Explanations

    references to celestial objects and aviation-related themes

    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
    .scalablytyped
    -0.15
    adows
    -0.15
    uner
    -0.15
    ož
    -0.14
    dings
    -0.14
    éro
    -0.14
    imitive
    -0.14
    NECT
    -0.14
     Blonde
    -0.14
    icator
    -0.14
    POSITIVE LOGITS
     sky
    0.90
     skies
    0.73
    sky
    0.71
     Sky
    0.70
    Sky
    0.65
    天
    0.59
     heavens
    0.57
    .sky
    0.52
     天
    0.47
     heaven
    0.45
    Activations Density 0.030%

    No Known Activations