© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37171
    Prev
    Next
    INDEX
    Explanations

    ICYMI, FYI, IIUC, BTW

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    бо
    -0.10
    mixed
    -0.09
    [email
    -0.08
    got
    -0.08
    pray
    -0.08
    æĬij
    -0.08
    zig
    -0.08
    Mixed
    -0.08
    欢è¿İæĤ¨
    -0.08
     powering
    -0.08
    POSITIVE LOGITS
    åIJĪãģĦ
    0.09
    ãĤ¤ãĥĪ
    0.09
    å®ļäºĨ
    0.09
    duk
    0.09
    äºŃ
    0.09
    è¾į
    0.08
    å¢Ļä¸Ĭ
    0.08
    æķ´ä¸ªäºº
    0.08
    _cmp
    0.08
    èĬĻ
    0.08
    Activations Density 0.024%

    No Known Activations