© 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. 35964
    Prev
    Next
    INDEX
    Explanations

    intended for

    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
     ÑĢеж
    -0.09
    辩论
    -0.08
    âĢ¾
    -0.08
    ä¼Ĺ
    -0.08
    axes
    -0.08
    çĬį
    -0.08
    æ³®
    -0.08
     FK
    -0.08
    FRAME
    -0.08
    è½®
    -0.08
    POSITIVE LOGITS
    èµĦè´¨
    0.12
    头çĸ¼
    0.10
    è±ģ
    0.10
    pred
    0.09
    åľ¨æĪijåĽ½
    0.09
    å®Įæ¯ķ
    0.09
    主ä½ĵ责任
    0.09
    âķĹ
    0.09
    æ³ķå¾ĭ责任
    0.09
    ä¸ļä½Ļ
    0.09
    Activations Density 0.117%

    No Known Activations