© 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. 48455
    Prev
    Next
    INDEX
    Explanations

    Vander 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
    letic
    -0.09
    çĪ±ä¸Ĭ
    -0.09
    çijķ
    -0.08
    åĢ¼å¾Ĺ
    -0.08
    omo
    -0.08
    iris
    -0.08
    ice
    -0.08
    ész
    -0.08
    æѤåĪ»
    -0.08
    çĿ¬
    -0.08
    POSITIVE LOGITS
    ocking
    0.10
    ees
    0.10
    以å¾Ģ
    0.09
    ÇĴ
    0.09
    äu
    0.09
    loff
    0.09
    lust
    0.09
    æĢľ
    0.09
    çİ°åĩº
    0.09
    OrUpdate
    0.09
    Activations Density 0.009%

    No Known Activations