© 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. 62383
    Prev
    Next
    INDEX
    Explanations

    election campaign practices, personal safety skills, regional production, fossils, alcohol expectancies

    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
    约å®ļ
    -0.10
     ÑĢаÑģÑģ
    -0.10
    æłijæŀĿ
    -0.09
    æ©«
    -0.09
    ichten
    -0.09
    æ³ķå¾ĭè§Ħå®ļ
    -0.09
    举
    -0.09
    еÑģÑĤво
    -0.08
    ije
    -0.08
    ä¸įä¸ĭ
    -0.08
    POSITIVE LOGITS
    è¿Ļä¸ĢçĤ¹
    0.11
    åīįèĢħ
    0.09
    als
    0.08
    ations
    0.08
    wer
    0.08
    ä¸ĢèĪ¬äºº
    0.08
    andom
    0.08
     Ji
    0.08
    çļĦåħ·ä½ĵ
    0.08
    æµģä¼ł
    0.08
    Activations Density 0.246%

    No Known Activations