© 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. 32972
    Prev
    Next
    INDEX
    Explanations

    their + possessive 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
    æĥħ趣
    -0.10
    çļĦä¸Ģéĥ¨åĪĨ
    -0.09
    /or
    -0.09
    few
    -0.09
     окÑĢÑĥжа
    -0.09
    ournal
    -0.09
    Ìģ
    -0.09
     ведÑĮ
    -0.08
    /she
    -0.08
    -0.08
    POSITIVE LOGITS
     respective
    0.11
    缮çļĦæĺ¯
    0.10
     own
    0.10
    å¤ļãģı
    0.10
    åIJĦèĩªçļĦ
    0.10
    ç¾İ好çĶŁæ´»
    0.09
    <?=
    0.09
    etails
    0.09
     gonna
    0.09
    åij¼å£°
    0.09
    Activations Density 0.104%

    No Known Activations