© 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. 48447
    Prev
    Next
    INDEX
    Explanations

    at least X times

    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
    erre
    -0.09
    çļĦæľĢä½³
    -0.08
    èĩªæĦ¿
    -0.08
    æĺ¯æľĢ好çļĦ
    -0.08
    åĨĻçļĦ
    -0.08
    æŃ£å¥½
    -0.08
    ever
    -0.08
    æľĢåIJĪéĢĤ
    -0.08
    imenti
    -0.08
    lose
    -0.08
    POSITIVE LOGITS
    åħ¶ä»ĸçļĦ
    0.11
    èµ·çłģ
    0.10
    éĤ£äºĽ
    0.10
    çľĭä¸Ģä¸ĭ
    0.09
    é«ĺä¸ī
    0.09
    åĽłä¸ºä½ł
    0.09
    ä¸įçĦ¶
    0.09
    èĩ³å°ij
    0.09
     ÑħоÑĤÑı
    0.09
    åĪ«çļĦ
    0.09
    Activations Density 0.058%

    No Known Activations