© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 5640
    Prev
    Next
    INDEX
    Explanations

    about quantities or durations

    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
    ãģ¾ãģķ
    -0.11
    ç®Ģ缴æĺ¯
    -0.10
    ç®Ģ缴
    -0.10
    ÑĪка
    -0.10
    åį³ä½¿æĺ¯
    -0.09
    IVAL
    -0.09
    ä½ķåĨµ
    -0.09
    ç«ŁçĦ¶
    -0.09
    å°½éĩı
    -0.09
    åĩłä¹İ
    -0.09
    POSITIVE LOGITS
    -sized
    0.12
    /all
    0.10
    har
    0.10
     dozen
    0.09
    å½ĵä¹ĭ
    0.09
    place
    0.08
    ä¸įéĶĻçļĦ
    0.08
    大çļĦ
    0.08
    ä¼ļéĢīæĭ©
    0.08
     (?)
    0.08
    Activations Density 0.035%

    No Known Activations