© 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. 11254
    Prev
    Next
    INDEX
    Explanations

    "great" to / "nice" if

    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
    é¢ĩæľī
    -0.10
    ä¹°ä¸ľè¥¿
    -0.09
    è·¸
    -0.09
    çļĦæĢģ度
    -0.08
    å¾Īæľī
    -0.08
    éĢļ常
    -0.08
    å·²ç»ıæĺ¯
    -0.08
    å·²æĪIJ为
    -0.08
    æģ¼
    -0.08
    iller
    -0.08
    POSITIVE LOGITS
    æľ«ç«¯
    0.10
    è¿Ľä¸ĢæŃ¥
    0.09
    ollen
    0.09
    æĭĶ
    0.09
    owi
    0.08
     someday
    0.08
    åĨį次
    0.08
     nữa
    0.08
    Uses
    0.08
    éĤ°
    0.08
    Activations Density 0.050%

    No Known Activations