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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 31326
    Prev
    Next
    INDEX
    Explanations

    references to memorialization and remembrance of historical events, specifically related to Indian soldiers in World War I

    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
    .SIG
    -0.16
    oyer
    -0.16
    ç§Ģ
    -0.15
    amiliar
    -0.13
    ylko
    -0.13
    ('_',
    -0.13
    atrix
    -0.13
    õi
    -0.13
    ancia
    -0.13
    632
    -0.13
    POSITIVE LOGITS
     memorial
    0.36
     commem
    0.32
     Memorial
    0.31
     memory
    0.30
     Memory
    0.28
     memor
    0.28
     monument
    0.28
     memoria
    0.28
     museum
    0.27
     monuments
    0.26
    Activations Density 0.048%

    No Known Activations