© 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. 22018
    Prev
    Next
    INDEX
    Explanations

    learned from

    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
    é¸Ń
    -0.10
    OCK
    -0.10
    èĢĹ
    -0.09
    æħ¢
    -0.09
    åī©ä½Ļ
    -0.09
    åıĪæľī
    -0.09
    arel
    -0.08
    ä¸Ģä¼ļåĦ¿
    -0.08
    åľ¨è¿Ļ个
    -0.08
    线æĿ¡
    -0.08
    POSITIVE LOGITS
    åĩºèĩª
    0.13
    åIJ¬è¿ĩ
    0.12
    åĩºå¤Ħ
    0.11
    è¿Ļæĺ¯æĪij
    0.10
    åIJ¬è¯´è¿ĩ
    0.10
    æĿ¥åİĨ
    0.10
    æºIJäºİ
    0.09
    มาà¸Īาà¸ģ
    0.09
    åİŁåĪĽ
    0.09
    éĽı
    0.09
    Activations Density 0.119%

    No Known Activations