© 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. 50851
    Prev
    Next
    INDEX
    Explanations

    spare, reserve, unused

    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
    主线
    -0.09
    åİĮ
    -0.09
    ãĤ¡
    -0.09
    à¹Ģà¸Ńà¸ģ
    -0.09
    anko
    -0.09
    人çĶŁçļĦ
    -0.09
    iska
    -0.08
    çľŁäºº
    -0.08
    ä¸Ģ个å¤ļ
    -0.08
    éĻĮçĶŁäºº
    -0.08
    POSITIVE LOGITS
    å¤ĩçĶ¨
    0.17
    åĤ¨å¤ĩ
    0.16
    åIJİå¤ĩ
    0.16
     reserve
    0.13
    reserve
    0.13
    åĢĻ
    0.12
     reserves
    0.12
    åĢĻéĢī
    0.11
     unused
    0.11
     spare
    0.10
    Activations Density 0.107%

    No Known Activations