© 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. 9924
    Prev
    Next
    INDEX
    Explanations

    desire to be

    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
    aben
    -0.11
    pped
    -0.09
    ä¸Ģä¸ĭ
    -0.09
    æ¡Ģ
    -0.09
    avail
    -0.09
    éŁ
    -0.09
    åijĪ
    -0.08
    羿
    -0.08
    áb
    -0.08
    æ´¾åĩº
    -0.08
    POSITIVE LOGITS
    éķ¿å¤§äºĨ
    0.11
    åĶ¿
    0.10
    æł·
    0.10
    ëĶ©
    0.09
    éķ¿å¤§
    0.09
    鸿
    0.09
    åĨįä¹Łä¸į
    0.08
    çļĦèģĮä¸ļ
    0.08
    å¯IJ
    0.08
    åĮ»åĬ¡
    0.08
    Activations Density 0.036%

    No Known Activations