© 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. 37694
    Prev
    Next
    INDEX
    Explanations

    two and three types

    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
    主æĦı
    -0.10
    ç¼Ń
    -0.09
    好åIJĹ
    -0.09
    çļĦä¸ľè¥¿
    -0.09
    çħ§äº®
    -0.09
    èµĸ以
    -0.09
    ãĥ³ãĤ°
    -0.09
    碰
    -0.09
    æ¯ĹéĤ»
    -0.09
    åģļä¸ĢäºĽ
    -0.08
    POSITIVE LOGITS
    两ç§į
    0.17
    åĩłç§į
    0.16
     types
    0.15
    两类
    0.13
    åĽĽç§į
    0.13
     sexes
    0.13
    ä¸īç§į
    0.13
    åĽĽå¤§
    0.12
    两大
    0.12
     Types
    0.12
    Activations Density 0.149%

    No Known Activations