© 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. 985
    Prev
    Next
    INDEX
    Explanations

    preposition followed by noun

    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
    ulan
    -0.09
    æĦŁæĥħ
    -0.09
    rovers
    -0.08
    äºĮæīĭ
    -0.08
    åĿļ
    -0.08
     suspected
    -0.08
    éģĵ士
    -0.08
    大èĩ£
    -0.08
    iete
    -0.08
    éľĨ
    -0.08
    POSITIVE LOGITS
    engl
    0.09
    ä¸Ģæī¹
    0.09
    è¿Ļ家
    0.09
    ä¸Ģéĥ¨åĪĨ
    0.09
    èĭĵ
    0.09
    ç¼Ļ
    0.09
    é¢Īæ¤İ
    0.09
    è¿Ļåı¥è¯Ŀ
    0.08
    ä¸Ģ大æī¹
    0.08
     [/
    0.08
    Activations Density 0.012%

    No Known Activations