© 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. 53168
    Prev
    Next
    INDEX
    Explanations

    many have tried

    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
    沧å·ŀ
    -0.09
    åįģä½³
    -0.09
    iatric
    -0.09
    æŀķ头
    -0.09
    åįķ纯çļĦ
    -0.08
    ä¸Ģç§į
    -0.08
    roat
    -0.08
    çīĮåŃIJ
    -0.08
    tube
    -0.08
    æµĩ
    -0.08
    POSITIVE LOGITS
     Ñĥзн
    0.10
    ç»ıå¼Ģ
    0.09
    åħ³äºİæĪij们
    0.09
    htag
    0.09
    åºŃ
    0.08
    å¿ĥæĥ³
    0.08
    ä¼IJ
    0.08
     DISCLAIMER
    0.08
    趾
    0.08
    SSID
    0.08
    Activations Density 0.026%

    No Known Activations