© 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. 11633
    Prev
    Next
    INDEX
    Explanations

    To find all

    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
    奢ä¾Ī
    -0.09
    åŃIJ宫
    -0.09
    çĸ£
    -0.08
    .usage
    -0.08
    phony
    -0.08
    èĥļèĥİ
    -0.08
    -unused
    -0.08
    ken
    -0.08
    ä¸įæ¸ħ
    -0.08
    å´ĩæĭľ
    -0.08
    POSITIVE LOGITS
    ãģĿãĤĮãģ§ãĤĤ
    0.11
    转移åĪ°
    0.11
     nonetheless
    0.11
    æĤ¨åı¯ä»¥
    0.11
    éĻ¤å¤ĸ
    0.10
    ä¾ĿçĦ¶
    0.10
     nevertheless
    0.10
    inia
    0.10
    ä¸į代表
    0.10
    æŃ¤é¡¹
    0.09
    Activations Density 0.083%

    No Known Activations