© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51396
    Prev
    Next
    INDEX
    Explanations

    wood

    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
    ä¸į容æĺĵ
    -0.10
    çļĦæĵįä½ľ
    -0.09
    ä½İä½į
    -0.09
    _bh
    -0.09
    åºķ
    -0.09
    ivor
    -0.09
     lord
    -0.08
    æĸĩåĮĸ产ä¸ļ
    -0.08
    Lord
    -0.08
    ibli
    -0.08
    POSITIVE LOGITS
    èį«
    0.09
    涵
    0.09
    eries
    0.09
    æªIJ
    0.09
    ubi
    0.09
    å··
    0.08
    èĪĨ
    0.08
    ç»İ
    0.08
    çŀ©
    0.08
    ennes
    0.08
    Activations Density 0.025%

    No Known Activations