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. 23345
    Prev
    Next
    INDEX
    Explanations

    references to economic concerns and crises

    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
    ISCO
    -0.16
    ami
    -0.15
    åįĥ
    -0.14
    ếp
    -0.14
    .heroku
    -0.14
     ÐĴÑģ
    -0.13
    itzer
    -0.13
    iez
    -0.13
    kol
    -0.13
    Ñ
    -0.13
    POSITIVE LOGITS
     Martial
    0.19
    911
    0.19
     Amerika
    0.17
    ilder
    0.16
     freedoms
    0.16
    uese
    0.15
     awakening
    0.15
     puppet
    0.15
    hf
    0.15
     Puppet
    0.15
    Activations Density 0.033%

    No Known Activations