© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52230
    Prev
    Next
    INDEX
    Explanations

    camel

    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
    æ±Łæ·®
    -0.11
    lass
    -0.09
    iant
    -0.09
    uant
    -0.09
    enko
    -0.09
    sic
    -0.09
    磶
    -0.09
    ening
    -0.09
    rent
    -0.08
    edly
    -0.08
    POSITIVE LOGITS
    opard
    0.11
    aspers
    0.09
    PLIC
    0.09
    ouflage
    0.09
    berra
    0.09
    piece
    0.09
    oad
    0.08
    pie
    0.08
    ous
    0.08
    etas
    0.08
    Activations Density 0.007%

    No Known Activations