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. 23226
    Prev
    Next
    INDEX
    Explanations

    information related to nuclear energy policies and actions taken by countries

    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
    ahir
    -0.16
    εÏĤ
    -0.15
    uez
    -0.15
    elik
    -0.15
    annya
    -0.15
    thren
    -0.15
     кад
    -0.14
     ><?
    -0.14
    readcr
    -0.14
    imas
    -0.14
    POSITIVE LOGITS
     Feed
    0.25
     feed
    0.25
     FIT
    0.25
    Feed
    0.23
    feed
    0.21
     Energ
    0.20
     Fukushima
    0.20
     EEG
    0.20
     RET
    0.19
     energy
    0.19
    Activations Density 0.018%

    No Known Activations