Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Qwen3-4B
    3. 23-TRANSCODER-HP
    4. 7526
    Prev
    Next
    INDEX
    Explanations

    People and relationships

    np_max-act · gemini-2.0-flash

    references to immediate family relationships, especially mentions of parents and their children.

    oai_token-act-pair · gpt-5Triggered by @liuliujiujiu05
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    gé
    -0.27
    usat
    -0.26
     panels
    -0.26
    ovation
    -0.25
    unan
    -0.25
    owi
    -0.25
    Amt
    -0.24
    ijľ
    -0.24
     Soph
    -0.24
    antom
    -0.24
    POSITIVE LOGITS
    æľ¬é¢Ĩ
    0.26
    è´¨éĩıåĴĮ
    0.25
    对æĪij们
    0.25
    ework
    0.24
    @author
    0.24
    æĸ¹æ¡Ī
    0.24
    hei
    0.24
    æľīæķο̧
    0.23
     wal
    0.23
     ald
    0.23
    Activations Density 0.001%

    No Known Activations