© 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. 9823
    Prev
    Next
    INDEX
    Explanations

    date and time formats

    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
    å«ī
    -0.10
    éĢij
    -0.09
    å´Ł
    -0.08
    ilik
    -0.08
     rough
    -0.08
    ạnh
    -0.08
     ethic
    -0.08
    æºIJæºIJ
    -0.08
    ä½İä¿Ŀ
    -0.08
    пло
    -0.08
    POSITIVE LOGITS
    éĹ°
    0.14
     calendars
    0.12
     calendar
    0.10
    åĩĨæŶ
    0.10
     clocks
    0.10
    Calendar
    0.10
     Clock
    0.09
    _calendar
    0.09
    clock
    0.09
     fixed
    0.09
    Activations Density 0.489%

    No Known Activations