© 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. 62166
    Prev
    Next
    INDEX
    Explanations

    choice between

    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
     جدا
    -0.10
    éĺ²æĻĴ
    -0.09
    åľ¬
    -0.09
    OrElse
    -0.09
    è§ĦåĪĴ设计
    -0.09
    æľ½
    -0.09
    ushima
    -0.09
     ÐľÐ°Ðº
    -0.09
    acho
    -0.09
    лоÑĤ
    -0.09
    POSITIVE LOGITS
    çıŃ
    0.08
    å®ī举
    0.08
    ä»Ĭ天æĪij们
    0.08
    æį¢
    0.08
     Hob
    0.08
    æĮĸ
    0.08
    æĴĴ
    0.07
     Put
    0.07
    åĢ¾åIJ¬
    0.07
    群
    0.07
    Activations Density 0.021%

    No Known Activations