© 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. 21965
    Prev
    Next
    INDEX
    Explanations

    undulating, winding, rugged surface/shape

    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
    _hub
    -0.09
    åı·çłģ
    -0.09
     Hire
    -0.09
    éħ¸å¥¶
    -0.09
    æݥ纳
    -0.08
     há»Ļi
    -0.08
    İĺìĿ´
    -0.08
    å¥Ĺè£ħ
    -0.08
    ILE
    -0.08
     Hub
    -0.08
    POSITIVE LOGITS
    æĽ²æĬĺ
    0.12
    ynth
    0.10
     вокÑĢÑĥг
    0.10
    ç»ķ
    0.10
    æ²³æµģ
    0.09
    ä¸įå®ļ
    0.09
    èµ·ä¼ı
    0.09
    轨迹
    0.09
     khúc
    0.09
    å°ıé¾ĻèĻ¾
    0.09
    Activations Density 0.029%

    No Known Activations