© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10128
    Prev
    Next
    INDEX
    Explanations

    citations with bracketed numbers and names

    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
    psilon
    -0.09
    .getAs
    -0.08
    alu
    -0.08
    é«ĵ
    -0.08
     metav
    -0.08
     Fet
    -0.08
    Ùħض
    -0.08
    erview
    -0.08
     Antar
    -0.08
     supervised
    -0.08
    POSITIVE LOGITS
    ãĢĬ
    0.11
    詳ãģĹãģı
    0.09
    Bur
    0.08
     подÑĢоб
    0.08
    åıĮéĩį
    0.08
    åĴĴ
    0.08
    å¤ļåĬŁèĥ½
    0.08
    éĢļãĤĬ
    0.08
    å°ı鸣
    0.08
     capita
    0.08
    Activations Density 0.256%

    No Known Activations