Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 54656
    Prev
    Next
    INDEX
    Explanations

    awards and honors related to sports players

    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
    pra
    -0.15
    ohana
    -0.15
    nesc
    -0.14
    инок
    -0.14
    sert
    -0.14
    OMBRE
    -0.14
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    -0.14
    ãĤ¯ãĤ»
    -0.14
    //*
    -0.14
    esub
    -0.14
    POSITIVE LOGITS
    OF
    0.17
    undy
    0.15
     OF
    0.15
     guard
    0.14
     guards
    0.14
     RW
    0.14
     –
    0.14
     POS
    0.13
    заб
    0.13
     III
    0.13
    Activations Density 0.071%

    No Known Activations