© 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. 19050
    Prev
    Next
    INDEX
    Explanations

    negative consequences

    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
    å¤ĸæĿ¥
    -0.08
    åı°åįĹ
    -0.08
    æĺ¯æľĢ好çļĦ
    -0.08
    åıĹåĪ°
    -0.08
     Platform
    -0.08
    峨
    -0.08
    éĺ²
    -0.08
    ello
    -0.08
     BOOST
    -0.07
     technically
    -0.07
    POSITIVE LOGITS
    åIJİæŀľ
    0.18
     consequences
    0.13
    æ·±è¿ľ
    0.12
    ä¸įå°½
    0.11
     consequence
    0.10
    æĴ®
    0.10
    æİıåĩº
    0.10
    æ·±æ¸Ĭ
    0.10
    ious
    0.10
    æ¢Ń
    0.10
    Activations Density 0.046%

    No Known Activations