© 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. 32576
    Prev
    Next
    INDEX
    Explanations

    visit the

    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
    away
    -0.10
    ä¸ĢåĽŀäºĭ
    -0.09
    äºĶçĻ¾
    -0.09
    нÑĮ
    -0.09
    _ACTIV
    -0.09
    biz
    -0.09
    èĢĥèĻijåĪ°
    -0.08
    åĻİ
    -0.08
    æĿ¥çļĦ
    -0.08
    对ä¼ģä¸ļ
    -0.08
    POSITIVE LOGITS
    s
    0.09
    åIJįæł¡
    0.08
     elkaar
    0.08
     бол
    0.08
    羡
    0.08
     slack
    0.08
    è¦ģåĬłå¼º
    0.08
    åij¨åĽ´çļĦ
    0.08
    iny
    0.08
     japan
    0.08
    Activations Density 0.020%

    No Known Activations