© 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. 50217
    Prev
    Next
    INDEX
    Explanations

    right thing to do

    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
    WARE
    -0.09
    工夫
    -0.09
    æĴħ
    -0.08
    spo
    -0.08
    odia
    -0.08
     fant
    -0.08
    ä¸Ģåij³
    -0.08
    çľĭäºĨä¸Ģ
    -0.08
    lag
    -0.08
    è¿ĩ度
    -0.08
    POSITIVE LOGITS
    idor
    0.11
    _descriptor
    0.10
    没éĶĻ
    0.10
    使åij½æĦŁ
    0.10
    æŃ£ç¡®
    0.10
    å½³
    0.10
    éĥ½æľīçĿĢ
    0.10
    æŃ£ç¡®çļĦ
    0.09
    æľĢåIJĪéĢĤ
    0.09
    lish
    0.09
    Activations Density 0.036%

    No Known Activations