© 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. 32129
    Prev
    Next
    INDEX
    Explanations

    occasional mistake, type of defect, odd impossible, types of

    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
    家éķ¿ä»¬
    -0.12
     times
    -0.09
     spaces
    -0.09
    管家
    -0.08
     Schultz
    -0.08
    æľĭåıĭ们
    -0.08
     sides
    -0.08
     roles
    -0.08
    çª
    -0.08
     levels
    -0.07
    POSITIVE LOGITS
    THING
    0.10
    èıģ
    0.10
    enefit
    0.10
    rotein
    0.10
    htar
    0.09
    大大å°ıå°ı
    0.09
    æѤ类
    0.08
    ä¼İ
    0.08
    å¿ĹæĦ¿èĢħ
    0.08
    çĽİ
    0.08
    Activations Density 0.430%

    No Known Activations