© 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. 30697
    Prev
    Next
    INDEX
    Explanations

    country Luxembourg, Watts Field

    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
    onaut
    -0.09
    swer
    -0.09
    ed
    -0.09
    æİĮæı¡
    -0.08
    ally
    -0.08
    ëĪĮ
    -0.08
    CAD
    -0.08
    å¾Ĵ
    -0.08
    ëĭĪëĭ¤
    -0.08
    çĻ½è¡£
    -0.08
    POSITIVE LOGITS
    æī£
    0.09
    lam
    0.09
    å·ŀ
    0.08
    åľ°åĮºçļĦ
    0.08
     Hath
    0.08
    åĬŀ
    0.08
    TRACE
    0.08
    angkan
    0.08
     organism
    0.08
     straw
    0.08
    Activations Density 0.001%

    No Known Activations