© 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. 29963
    Prev
    Next
    INDEX
    Explanations

    bearingless type, conventional type, inverted type

    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
    åĪ»
    -0.09
    对ä¼ģä¸ļ
    -0.09
    iesta
    -0.08
    adan
    -0.08
    éķĸ
    -0.08
    èĦ¸
    -0.08
    äºĶéĩij
    -0.08
    è´§
    -0.08
    uttle
    -0.08
    æŶåĪ»
    -0.08
    POSITIVE LOGITS
    ç»ĵæŀĦ
    0.12
     designs
    0.11
    åİŁçIJĨ
    0.11
    æŀ¶æŀĦ
    0.10
    çµIJæ§ĭ
    0.10
    è¨Ńè¨Ī
    0.10
    设计
    0.10
    æŀĦéĢł
    0.09
    è¿IJä½ľ
    0.09
     Designs
    0.09
    Activations Density 1.580%

    No Known Activations