© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37779
    Prev
    Next
    INDEX
    Explanations

    shredded or shuffled

    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
    pheres
    -0.10
     thuyá»ģn
    -0.09
    à¹ĥà¸Ī
    -0.09
     Hancock
    -0.09
    Ùĥر
    -0.09
    asher
    -0.09
    shore
    -0.08
     shoe
    -0.08
    æıĸ
    -0.08
    oons
    -0.08
    POSITIVE LOGITS
    踯
    0.09
    ç¼ķ
    0.09
    æŀ
    0.09
    æĮĩ导æĦıè§ģ
    0.09
    otel
    0.08
    à¸ĩาà¸Ļ
    0.08
     tear
    0.08
    irectory
    0.08
    æ¿ij
    0.08
    åīIJ
    0.08
    Activations Density 0.002%

    No Known Activations