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. 31343
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to connections and relationships between entities or individuals

    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
    Africa
    -0.24
    Asia
    -0.19
     africa
    -0.19
     Africa
    -0.18
     Arabia
    -0.18
    öy
    -0.18
    frica
    -0.17
     Blacks
    -0.16
     asia
    -0.16
     Mongolia
    -0.16
    POSITIVE LOGITS
     American
    0.53
     British
    0.52
     Canadian
    0.48
     Australian
    0.42
     French
    0.42
    American
    0.41
     Italian
    0.41
    British
    0.41
     German
    0.40
    Canadian
    0.40
    Activations Density 0.114%

    No Known Activations