© 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
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 8839
    Prev
    Next
    INDEX
    Explanations

    encoding and decoding

    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
    .gc
    -0.10
     jun
    -0.09
    auce
    -0.09
     اÙĦÙħÙĤ
    -0.09
    æ¼ı
    -0.09
     plag
    -0.09
    .cz
    -0.09
    quo
    -0.09
    attery
    -0.09
    stre
    -0.08
    POSITIVE LOGITS
    åύ
    0.11
    å¤±æľĽ
    0.10
    URIComponent
    0.10
    åĥ®
    0.10
    æĹŃ
    0.10
    åĨ²åĪº
    0.10
     Äijáo
    0.09
    _into
    0.09
     encodeURIComponent
    0.09
    /Object
    0.09
    Activations Density 0.135%

    No Known Activations