© 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
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 1720
    Prev
    Next
    INDEX
    Explanations

    team of dentiststeam of big thinkersteam of experienced staffteam of three Romanianteam of student leadersteam of doctorsteam of eight researchersteam of four individualsteam of 10 membersThe neuron detects "team of" followed by a number or descriptive word

    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
    è¿Ļç¬Ķ
    -0.10
    åģļä¸Ģ个
    -0.09
    ä¸Ģç¬Ķ
    -0.09
    å±ĢéĿ¢
    -0.09
    产ä¸ļéĵ¾
    -0.09
    大ä¼Ĺ
    -0.09
    产ä¸ļç»ĵæŀĦ
    -0.08
    ä¸Ģ个æĸ°çļĦ
    -0.08
     ç¬
    -0.08
    åħ¬åı¸çļĦ
    -0.08
    POSITIVE LOGITS
    allis
    0.10
    .Direction
    0.09
    åĦª
    0.09
    orta
    0.09
    resse
    0.09
    виз
    0.09
    orte
    0.09
    apeutics
    0.09
    ä»Ħ
    0.08
    QtCore
    0.08
    Activations Density 0.020%

    No Known Activations