© 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. 19796
    Prev
    Next
    INDEX
    Explanations

    to or stubborn

    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
    çĹķ
    -0.09
     há»Ļ
    -0.09
    ä¸įä¸ĭ
    -0.08
    à¹Ģลย
    -0.08
    éĩĮç¨ĭ
    -0.08
    anker
    -0.08
    ãĤĦãģ£ãģ¦
    -0.08
    inger
    -0.08
    igers
    -0.08
    bows
    -0.08
    POSITIVE LOGITS
     kepada
    0.21
     Ø¥ÙĦÙĬÙĩا
    0.20
     Ø¥ÙĦÙĬÙĩ
    0.19
    åĪ°
    0.18
     to
    0.17
     Ø¥ÙĦÙī
    0.16
     اÙĦÙī
    0.14
    ToFile
    0.13
    åĪ°è¿ĻéĩĮ
    0.13
    æĸ¼
    0.13
    Activations Density 0.732%

    No Known Activations