© 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. 58979
    Prev
    Next
    INDEX
    Explanations

    loss of control

    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
     recess
    -0.09
    ä¸įå®ľ
    -0.09
    èĻ¹
    -0.08
    ç«Ļ
    -0.08
    ä¸įæŃ»
    -0.08
    åĩºéŨ
    -0.08
    鼷æĸ¯
    -0.08
    æĴ¤éĶĢ
    -0.08
     Fritz
    -0.08
    brero
    -0.08
    POSITIVE LOGITS
    失æݧ
    0.20
    æݧåĪ¶
    0.18
    åı¯æݧ
    0.18
     control
    0.17
    æİĮæݧ
    0.16
     Control
    0.15
    (control
    0.15
    -control
    0.15
    _Control
    0.15
    .control
    0.14
    Activations Density 0.058%

    No Known Activations