© 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. 23795
    Prev
    Next
    INDEX
    Explanations

    that are

    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
     itself
    -0.15
    was
    -0.12
    ãģĦãģ¾ãģĽãĤĵ
    -0.11
    ãģĮãģĤãĤĬãģ¾ãģĹãģŁ
    -0.11
    íĸĪìĬµëĭĪëĭ¤
    -0.10
    åĴĮå®ĮåĸĦ
    -0.10
    ãģ¯ãģĤãĤĬãģ¾ãģĽãĤĵ
    -0.10
    's
    -0.10
    æŃ¤äºº
    -0.10
    ’s
    -0.10
    POSITIVE LOGITS
     themselves
    0.24
     are
    0.14
    ä¸Ģ个个
    0.14
     thems
    0.13
     aren
    0.13
     ÑıвлÑıÑİÑĤÑģÑı
    0.13
     коÑĤоÑĢÑĭе
    0.13
    uite
    0.12
    _are
    0.12
    mith
    0.12
    Activations Density 0.177%

    No Known Activations