© 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. 61631
    Prev
    Next
    INDEX
    Explanations

    D&D level and game mechanics

    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æĮģ
    -0.11
    uche
    -0.10
    æĬĵ
    -0.10
    çķ´
    -0.09
    .Keys
    -0.09
    缫
    -0.09
    ä¸»æľº
    -0.09
    åĪĩåħ¥
    -0.09
    太平æ´ĭ
    -0.09
    æ¡ij
    -0.09
    POSITIVE LOGITS
    æ±ĩ
    0.11
    ä¸Ģ次æĢ§
    0.09
    jer
    0.09
    å°¸
    0.09
    èħĭ
    0.08
    ä¸Ģç»ı
    0.08
    Interview
    0.08
     Powell
    0.08
    èݲ
    0.08
     Hawks
    0.08
    Activations Density 0.059%

    No Known Activations