© 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. 63891
    Prev
    Next
    INDEX
    Explanations

    references to linguistic expertise and educational qualifications

    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
    çµĮ
    -0.17
    adero
    -0.15
    shal
    -0.15
    erli
    -0.14
    ADER
    -0.14
    locker
    -0.14
    atoi
    -0.14
    ieren
    -0.14
    Lens
    -0.13
    ABCDEFG
    -0.13
    POSITIVE LOGITS
     language
    0.39
     Language
    0.39
     languages
    0.35
    è¯Ńè¨Ģ
    0.33
     lingu
    0.32
     LANGUAGE
    0.32
    ling
    0.32
    Language
    0.31
     Languages
    0.31
    _language
    0.29
    Activations Density 0.016%

    No Known Activations