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. 23579
    Prev
    Next
    INDEX
    Explanations

    terms related to international diplomacy and cooperation

    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
    agna
    -0.17
     oppos
    -0.16
    åĭŁ
    -0.15
     Kanun
    -0.15
    itol
    -0.15
     ä»¶
    -0.14
    ulen
    -0.14
     pil
    -0.14
    undra
    -0.14
     bip
    -0.14
    POSITIVE LOGITS
     countries
    0.32
     nations
    0.29
     Countries
    0.24
    countries
    0.23
     paÃŃses
    0.22
     ministers
    0.22
     member
    0.22
    Countries
    0.22
     meeting
    0.21
    -member
    0.20
    Activations Density 0.031%

    No Known Activations