© 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. 57944
    Prev
    Next
    INDEX
    Explanations

    accuracy and inaccuracy

    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
    allee
    -0.09
    è¿Ļä¹Ī大
    -0.09
    æľºæ¢°åĮĸ
    -0.08
    ä»Ĭå¹´
    -0.08
     StringComparison
    -0.08
    tring
    -0.08
    @show
    -0.08
     dönÃ¼ÅŁ
    -0.08
    Laugh
    -0.08
    Õ®
    -0.08
    POSITIVE LOGITS
    åĩĨç¡®
    0.19
    åĩĨç¡®æĢ§
    0.18
     accuracy
    0.17
    accuracy
    0.16
     Accuracy
    0.16
    Accuracy
    0.15
     accurate
    0.15
    äºĭå®ŀ
    0.14
     facts
    0.14
     inaccur
    0.14
    Activations Density 0.071%

    No Known Activations