© 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. 37996
    Prev
    Next
    INDEX
    Explanations

    capacity or capability

    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
    客æĪ·æıIJä¾Ľ
    -0.12
     McB
    -0.11
    ä¸ĢåĪ»
    -0.11
    ä¸Ģ件äºĭ
    -0.10
    .Dropout
    -0.10
    ä¼ģä¸ļæıIJä¾Ľ
    -0.10
    äºŃ
    -0.10
    åŃ¦ä¹łæĪIJ绩
    -0.09
     Airways
    -0.09
    çĨł
    -0.09
    POSITIVE LOGITS
    èĥ½åĬĽ
    0.21
     capacity
    0.21
     capability
    0.18
    èĥ½åĬĽåĴĮ
    0.18
     capabilities
    0.17
    capacity
    0.16
     Capacity
    0.16
     capacities
    0.16
    Capacity
    0.16
     capacidad
    0.15
    Activations Density 0.052%

    No Known Activations