© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19870
    Prev
    Next
    INDEX
    Explanations

    disruptive behavior and discipline

    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
    俸
    -0.09
    ä¾ĽåºĶ
    -0.08
    ossal
    -0.08
    éĻ·
    -0.08
     deb
    -0.08
     доб
    -0.08
     Seymour
    -0.08
    RID
    -0.08
     recycl
    -0.08
    åĬ¿åĬĽ
    -0.08
    POSITIVE LOGITS
    æī°ä¹±
    0.12
    è¡Į为
    0.12
    纪å¾ĭ
    0.12
    è¿Ŀ纪
    0.11
    课åłĤ
    0.11
    è°ĥçļ®
    0.11
    æīĵæŀ¶
    0.11
    è¡ĮçĤº
    0.10
    æļ´åĬĽ
    0.10
     behavior
    0.10
    Activations Density 0.199%

    No Known Activations