© 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. 35800
    Prev
    Next
    INDEX
    Explanations

    operating systems and platforms

    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
    lect
    -0.10
    好象
    -0.09
    æĪĹ
    -0.09
    Encoder
    -0.09
    设计åĻ¨
    -0.09
    æĬ¹
    -0.09
    èĭķ
    -0.09
    .Raise
    -0.08
    æĿ
    -0.08
    åIJī祥
    -0.08
    POSITIVE LOGITS
    çıį
    0.09
    åī¯
    0.09
    æĭĨ
    0.08
    SystemService
    0.08
    สà¸Ļ
    0.08
    ÑĤол
    0.08
    estre
    0.08
    çĽIJ
    0.08
     Joi
    0.08
    ogene
    0.08
    Activations Density 0.091%

    No Known Activations