© 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. 6586
    Prev
    Next
    INDEX
    Explanations

    tried and tested

    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
    äºĨä¸Ģ份
    -0.09
    urd
    -0.09
    ì·¨
    -0.08
    TERS
    -0.08
    ariat
    -0.08
    yat
    -0.08
    èĿĩ
    -0.08
     squ
    -0.08
    ä¸Ģ份
    -0.08
    åĵŃ
    -0.08
    POSITIVE LOGITS
    纡
    0.10
    åĬŀæ³ķ
    0.10
    èģ¯åIJĪ
    0.09
    ä¸İåIJ¦
    0.09
    sci
    0.09
    妥
    0.08
    çļĦåĬŀæ³ķ
    0.08
    辦æ³ķ
    0.08
    dden
    0.08
    好è¯Ħ
    0.08
    Activations Density 0.007%

    No Known Activations