© 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. 54813
    Prev
    Next
    INDEX
    Explanations

    at first

    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
    缮åīį为æŃ¢
    -0.12
    缮åīį
    -0.12
    ä¸Ģç«Ļ
    -0.11
    缮åīįæŃ£åľ¨
    -0.11
    åİĨæĿ¥
    -0.11
    æѤæŶ
    -0.11
    currently
    -0.10
    æĪªèĩ³çĽ®åīį
    -0.10
    缮åīįå·²
    -0.10
    æĪªæŃ¢
    -0.10
    POSITIVE LOGITS
    /AP
    0.09
    æĺ¯åĽłä¸º
    0.09
    åıªæĺ¯
    0.09
    _NS
    0.09
    mare
    0.08
    innie
    0.08
     outright
    0.08
    .binding
    0.08
    PropertyValue
    0.08
     simply
    0.07
    Activations Density 0.061%

    No Known Activations