© 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. 36325
    Prev
    Next
    INDEX
    Explanations

    chess problem

    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
     Hab
    -0.10
    uku
    -0.09
    鸾
    -0.09
    \Id
    -0.09
    ä¸įæĩĤ
    -0.09
    .ERR
    -0.09
    åºĻ
    -0.09
    é¹ı
    -0.09
    uated
    -0.09
    ahren
    -0.09
    POSITIVE LOGITS
    çĿ¦
    0.11
    åīįçŀ»
    0.10
    èĪŀåı°ä¸Ĭ
    0.09
    çķ¦
    0.09
    è¿Ļé¦ĸæŃĮ
    0.09
    çķĮ
    0.09
    çĶŁæķĪ
    0.09
    èĬ¥
    0.09
     sar
    0.08
    éĺı
    0.08
    Activations Density 0.068%

    No Known Activations