© 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. 33451
    Prev
    Next
    INDEX
    Explanations

    phrases related to self-sufficiency and the importance of individual actions in gaining recognition

    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
    ux
    -0.14
    úi
    -0.14
    leness
    -0.14
     cref
    -0.13
    sgi
    -0.13
     Sellers
    -0.13
    ÑĢаÑĤно
    -0.13
    ÑĥÑĢи
    -0.13
    //=
    -0.12
     даÑĤ
    -0.12
    POSITIVE LOGITS
     DIY
    0.56
     amateur
    0.50
     homemade
    0.46
     amateurs
    0.42
     diy
    0.42
     Amateur
    0.38
     self
    0.36
    DI
    0.36
     Homemade
    0.35
     hobby
    0.33
    Activations Density 0.027%

    No Known Activations