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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 4296
    Prev
    Next
    INDEX
    Explanations

    costing more or less

    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
    ëįķ
    -0.10
    ÑģоÑģÑĤоÑı
    -0.09
    kah
    -0.09
    antha
    -0.08
    åĻľ
    -0.08
     Hairst
    -0.08
    бл
    -0.08
    人士
    -0.08
    ERA
    -0.08
    æĮĩ导ä¸ĭ
    -0.08
    POSITIVE LOGITS
    大家
    0.11
    ntl
    0.10
    leet
    0.09
    å¤ļå°ij
    0.09
    è¿Ļæł·æīįèĥ½
    0.09
    ander
    0.09
    afa
    0.09
    æ²Ļé¾Ļ
    0.09
    ä½ł
    0.09
    do
    0.08
    Activations Density 0.011%

    No Known Activations