© 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. 36336
    Prev
    Next
    INDEX
    Explanations

    want

    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
    íħIJ
    -0.09
    åĩºå¤Ħ
    -0.09
    onet
    -0.08
     пÑĢоÑĨ
    -0.08
    æı¡æīĭ
    -0.08
    ä»Ģä¹ĪæĦıæĢĿ
    -0.08
    å¼ķé¢Ĩ
    -0.08
    олов
    -0.08
    cone
    -0.08
    Recognizer
    -0.08
    POSITIVE LOGITS
    è¿Ļæł·åģļ
    0.14
    çļĦç¼ĺæķħ
    0.12
    è¿Ļä¹Īåģļ
    0.12
    æľīæĽ´å¥½çļĦ
    0.10
    want
    0.10
    å¸ĮæľĽéĢļè¿ĩ
    0.09
    太å¤ļäºĨ
    0.09
     Ñĥдоб
    0.09
     muá»ijn
    0.09
    æĸ¹ä¾¿
    0.09
    Activations Density 0.122%

    No Known Activations