© 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. 56318
    Prev
    Next
    INDEX
    Explanations

    very adjective + noun

    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
    æĽ´å¥½çļĦ
    -0.12
    å·²ä¹ħçļĦ
    -0.11
    è¶³å¤ŁçļĦ
    -0.10
    èĢĮä¸įæĺ¯
    -0.10
    æĽ´å¥½åľ°
    -0.10
    ÑħваÑĤ
    -0.10
    æĽ´å¥½
    -0.09
    æĹłä¸į
    -0.09
    æĽ´å¤ļ
    -0.08
    ntl
    -0.08
    POSITIVE LOGITS
     indeed
    0.11
    å¼Ģå±Ģ
    0.09
    ä¹ŁéĿŀ常
    0.09
    åıĮæīĭ
    0.09
    æĭ¥
    0.09
    HING
    0.08
    oud
    0.08
    éĤ¦
    0.08
    è¿Ļå¥Ĺ
    0.08
    alls
    0.08
    Activations Density 0.166%

    No Known Activations