© 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. 19260
    Prev
    Next
    INDEX
    Explanations

    had/was followed by adjective/participle

    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
    éĵ²
    -0.10
    éĻ¤åİ»
    -0.09
    æ¡¢
    -0.09
    该åİ¿
    -0.09
    åģļåĪ°äºĨ
    -0.09
    indre
    -0.09
    åĹĵåŃIJ
    -0.09
    ikk
    -0.08
    auce
    -0.08
    çļĦçľ¼
    -0.08
    POSITIVE LOGITS
     Giuliani
    0.10
     Äijáng
    0.09
     ÐļÑĥ
    0.08
    mek
    0.08
    tle
    0.08
    SPAN
    0.08
     spies
    0.08
    _CTX
    0.08
    hoa
    0.08
    sap
    0.08
    Activations Density 0.099%

    No Known Activations