© 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. 61222
    Prev
    Next
    INDEX
    Explanations

    positive and negative effects

    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
    裳
    -0.10
    è¿°
    -0.10
    usc
    -0.09
    ORT
    -0.09
    æ½ľåľ¨
    -0.09
    MMC
    -0.08
    è¿Ļæł·æīįèĥ½
    -0.08
    è¯Ŀ说
    -0.08
    ãģĴ
    -0.08
    bane
    -0.08
    POSITIVE LOGITS
    ly
    0.14
    mente
    0.10
    cono
    0.10
    ities
    0.10
    å½¼
    0.09
    ity
    0.09
     dvd
    0.08
    çļĦåģļæ³ķ
    0.08
    oop
    0.08
    çļĦå·¥ä½ľ
    0.08
    Activations Density 0.025%

    No Known Activations