© 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. 26677
    Prev
    Next
    INDEX
    Explanations

    subjects followed by verbs/prepositions

    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
     sometimes
    -0.14
    æľīæŶ
    -0.13
    æľīæŶåĢĻ
    -0.12
    sometimes
    -0.11
     Sometimes
    -0.10
    æŁIJäºĽ
    -0.10
     иногда
    -0.10
    Sometimes
    -0.10
    ä¸Ģå®ļè¦ģ
    -0.10
    å¬ĸ
    -0.10
    POSITIVE LOGITS
    éĽĨä¸Ńåľ¨
    0.13
    éĥ½ä¼ļ
    0.12
    ä¹Łéĥ½
    0.11
    éĥ½æĺ¯
    0.11
    æĥħåĨµä¸ĭ
    0.10
    éĥ½åľ¨
    0.10
    人éĥ½
    0.10
    éĥ½è®¤ä¸º
    0.10
    éĥ½èĥ½
    0.10
    éĥ½è¦ģ
    0.10
    Activations Density 0.127%

    No Known Activations