© 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. 51985
    Prev
    Next
    INDEX
    Explanations

    recommendations and advice

    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
    åħ¬å¼Ģåıij
    -0.09
    (Expected
    -0.09
    åIJĪä½ľåįıè®®
    -0.09
    åŁ
    -0.09
    uzzi
    -0.09
    åķĸ
    -0.08
    __(*
    -0.08
    æĸĩç«ł
    -0.08
    å¿ĥ仪
    -0.08
    åħ±åIJĮåĬªåĬĽ
    -0.08
    POSITIVE LOGITS
    建议
    0.28
    æݨèįIJ
    0.24
    建èŃ°
    0.22
     recommend
    0.20
     recommends
    0.19
     recommendation
    0.18
    ãģĬåĭ§ãĤģ
    0.18
     khuyên
    0.18
    æݨèĸ¦
    0.18
     recommending
    0.17
    Activations Density 0.056%

    No Known Activations