© 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. 48114
    Prev
    Next
    INDEX
    Explanations

    unwanted or continuing states

    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
    ipi
    -0.09
     pep
    -0.09
    å±±ä¸Ĭ
    -0.09
    عÙħ
    -0.08
    常æĢģåĮĸ
    -0.08
     omitted
    -0.08
    ä¿Ń
    -0.08
     pale
    -0.08
    åħ»æĬ¤
    -0.08
    åı²ä¸ĬæľĢ
    -0.08
    POSITIVE LOGITS
    ä¸įæŃ»
    0.12
    ä¸įæĺĵ
    0.11
    顽强
    0.11
     persistence
    0.11
    ä¸į容æĺĵ
    0.10
    æĮģä¹ħ
    0.10
    éĴī
    0.10
     Persistence
    0.10
    ä¸ĭæĿ¥çļĦ
    0.10
     unwanted
    0.09
    Activations Density 0.056%

    No Known Activations