© 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. 27741
    Prev
    Next
    INDEX
    Explanations

    audience responses and expectations

    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
    ish
    -0.10
    èĢģ人家
    -0.10
    itud
    -0.10
    огÑĢаÑĦ
    -0.09
    æł·çļĦ
    -0.09
    adic
    -0.09
    ints
    -0.09
    ìį¨
    -0.09
    主æĦı
    -0.08
    flix
    -0.08
    POSITIVE LOGITS
    hip
    0.13
    eer
    0.11
    æľĭåıĭ
    0.10
    /or
    0.10
    å¨ĺ
    0.10
    пÑĢавлÑı
    0.09
    пÑĢав
    0.09
    ymmetric
    0.09
    ize
    0.09
    atical
    0.09
    Activations Density 0.008%

    No Known Activations