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. 31351
    Prev
    Next
    INDEX
    Explanations

    references to American values and ideals, particularly regarding ethics, democracy, and inclusion

    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
     lá»Ĺ
    -0.15
    γη
    -0.13
    ÑĶм
    -0.13
    алеж
    -0.13
     addCriterion
    -0.13
    weetalert
    -0.13
    æ±Ĥè´Ń
    -0.13
     RGBA
    -0.12
    سÙĦ
    -0.12
    _skb
    -0.12
    POSITIVE LOGITS
     values
    0.76
     Values
    0.64
    values
    0.61
     principles
    0.56
    Values
    0.56
    -values
    0.54
     VALUES
    0.53
    _values
    0.51
     Principles
    0.47
     valeurs
    0.45
    Activations Density 0.136%

    No Known Activations