© 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. 5007
    Prev
    Next
    INDEX
    Explanations

    long period, low intensity, slow wavelength

    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
    çŁŃæļĤ
    -0.10
    çŁŃ
    -0.10
    short
    -0.10
    éĶIJ
    -0.09
    çģ¿
    -0.09
    çĭŃ
    -0.09
     shortest
    -0.09
    Short
    -0.09
     brief
    -0.09
    ç¹ģå¿Ļ
    -0.09
    POSITIVE LOGITS
    ç¼ĵæħ¢
    0.19
    ç¼ĵç¼ĵ
    0.16
    æħ¢
    0.14
    ãĤĨãģ£ãģıãĤĬ
    0.13
     slow
    0.12
    æĮģä¹ħ
    0.12
    éķ¿æķĪ
    0.12
    Slow
    0.12
    æĤłæĤł
    0.12
     Slow
    0.12
    Activations Density 0.057%

    No Known Activations