© 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. 23740
    Prev
    Next
    INDEX
    Explanations

    is a comparison

    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
    æīĭä¸ŃçļĦ
    -0.10
    çķĻä¸ĭäºĨ
    -0.09
    sWith
    -0.09
    é£Ļ
    -0.09
    åı®
    -0.09
    åŁ¹è®ŃçıŃ
    -0.08
    身为
    -0.08
    æ´½
    -0.08
    åĩī
    -0.08
    åĻĻ
    -0.08
    POSITIVE LOGITS
    criptors
    0.10
    .ArgumentParser
    0.10
    äºķ
    0.10
    éĿ¢ä¸Ĭ
    0.10
    fare
    0.09
    éĽĨè£ħ
    0.09
    鹤
    0.09
    âĹĢ
    0.09
    ê¼Ń
    0.09
    igue
    0.09
    Activations Density 0.430%

    No Known Activations