© 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. 53952
    Prev
    Next
    INDEX
    Explanations

    references to success and achievement within a competitive context

    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
    ItemAt
    -0.17
    mür
    -0.16
     ucwords
    -0.15
    avicon
    -0.14
     Marvin
    -0.14
    )}</
    -0.14
    redient
    -0.14
     '''
    -0.14
    ."&
    -0.14
    /********
    -0.14
    POSITIVE LOGITS
    ï¼½
    0.23
    }\"
    0.22
    ï½£
    0.21
    \",
    0.18
    ãĢķ
    0.18
     *@
    0.18
    #endregion
    0.17
     <<↵
    0.17
     ãĢį
    0.17
    \",\
    0.16
    Activations Density 0.014%

    No Known Activations