© 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. 42228
    Prev
    Next
    INDEX
    Explanations

    items and concepts that are significant or noteworthy in various contexts

    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
     ìĿ´ëŁ¬
    -0.15
    è¿Ļä¸Ģ
    -0.15
    thus
    -0.15
     here
    -0.15
    anka
    -0.14
    mando
    -0.14
    ellido
    -0.14
    importe
    -0.14
    ležit
    -0.14
    warf
    -0.14
    POSITIVE LOGITS
     alone
    0.29
    alone
    0.28
     Alone
    0.21
    -ÑĤо
    0.19
     plus
    0.18
     itself
    0.17
    ib
    0.16
    iane
    0.16
    ibe
    0.16
     thôi
    0.16
    Activations Density 0.091%

    No Known Activations