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

    Neuronpedia

    Jacobian LensNEW
    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. 14185
    Prev
    Next
    INDEX
    Explanations

    statements and claims made by various individuals or entities

    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
     himself
    -0.31
     seinen
    -0.18
     Himself
    -0.17
     jego
    -0.17
    his
    -0.17
    ä»ĸçļĦ
    -0.17
     jeho
    -0.17
     “âĢ¦
    -0.16
     его
    -0.16
     his
    -0.16
    POSITIVE LOGITS
     they
    0.43
    they
    0.30
     themselves
    0.30
     há»į
    0.28
     that
    0.27
     They
    0.27
     THEY
    0.25
    They
    0.23
     они
    0.21
    ï¼Įä»ĸ们
    0.21
    Activations Density 0.083%

    No Known Activations