© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 60732
    Prev
    Next
    INDEX
    Explanations

    places and entities

    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
    Negative Logits
    æīĭæĦŁ
    -0.10
    ç»Ī身
    -0.09
    ä¸Ģ次
    -0.09
    被èªī
    -0.09
    éļį
    -0.09
    åŃ½
    -0.09
    对åѦçĶŁ
    -0.09
    åĩºèº«
    -0.09
    azer
    -0.09
    ä¸Ģ次次
    -0.09
    POSITIVE LOGITS
    ä¹ĭåľ°
    0.13
    åľºæīĢ
    0.13
    æīĢåľ¨çļĦ
    0.12
    çĶŁäº§åŁºåľ°
    0.12
    çļĦä¸ĸçķĮ
    0.11
    ä¹ĭ乡
    0.11
     bombed
    0.10
     меÑģÑĤа
    0.10
    æīĢåľ¨åľ°
    0.10
     place
    0.10
    Activations Density 0.794%

    No Known Activations