© 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. 48302
    Prev
    Next
    INDEX
    Explanations

    audition, auditory, audacity

    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
    hood
    -0.09
    æ»ĵ
    -0.09
    lef
    -0.09
    æīĭä¸ŃçļĦ
    -0.09
    brew
    -0.09
    ade
    -0.09
    æīĭèĩĤ
    -0.09
    oj
    -0.09
    çľ¼çľĭ
    -0.08
    orts
    -0.08
    POSITIVE LOGITS
    Trail
    0.10
    пÑĢавлÑı
    0.09
    acity
    0.09
    çĨ
    0.09
    ERM
    0.09
    ÙĨÙĬÙĨ
    0.09
    trail
    0.09
    robe
    0.09
    itable
    0.09
    tube
    0.08
    Activations Density 0.013%

    No Known Activations