© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36927
    Prev
    Next
    INDEX
    Explanations

    some sort of

    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
    ä¸Ģç§į
    -0.09
    THING
    -0.09
    ä¸Ģéĥ¨åĪĨ
    -0.09
     Ø´ÙĬØ¡
    -0.09
     coron
    -0.09
    ä¸įå®Į
    -0.08
     rigor
    -0.08
    æ±½
    -0.08
    ting
    -0.08
    æ¯ıä¸Ģ
    -0.08
    POSITIVE LOGITS
     вн
    0.09
    缮çļĦåľ¨äºİ
    0.09
    次è¦ģ
    0.09
    ç½®
    0.08
    åŁ¹åħ»åѦçĶŁ
    0.08
     lắm
    0.08
    å¾ĪæŃ£å¸¸
    0.08
    éªij
    0.08
    HECK
    0.08
    ors
    0.08
    Activations Density 0.167%

    No Known Activations