© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37338
    Prev
    Next
    INDEX
    Explanations

    I/we followed by modal verb

    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
    æī¿åıĹ
    -0.08
    æĺ¯ä¸ª
    -0.08
    å®ĥæĺ¯
    -0.08
    çī¹çĤ¹æĺ¯
    -0.08
    åĩºçİ°åľ¨
    -0.08
    à¸Ļà¸Ńà¸Ļ
    -0.07
    è¿ĻéĩĮæĺ¯
    -0.07
    æĪIJ为
    -0.07
    jee
    -0.07
    jej
    -0.07
    POSITIVE LOGITS
     will
    0.26
     sẽ
    0.25
    à¸Īะ
    0.23
    å°Ĩä¼ļ
    0.22
     akan
    0.22
    å°±ä¼ļ
    0.22
    ä¼ļæľī
    0.21
    ’ll
    0.21
    ä¼ļåľ¨
    0.21
    will
    0.21
    Activations Density 0.245%

    No Known Activations