© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19508
    Prev
    Next
    INDEX
    Explanations

    Persian

    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
    åĻª
    -0.09
    asel
    -0.09
    ç¥IJ
    -0.09
     ÑĤомÑĥ
    -0.08
    åIJİç»Ń
    -0.08
    iciel
    -0.08
    ument
    -0.08
    å§Ķä¼ļ
    -0.08
    ä½ij
    -0.08
    éĵĽ
    -0.08
    POSITIVE LOGITS
    ate
    0.09
    ariat
    0.09
    flu
    0.09
    ä¸į解
    0.09
    ating
    0.08
    endor
    0.08
    lob
    0.08
    fall
    0.08
    à¸ķะ
    0.08
    说çļĦè¯Ŀ
    0.08
    Activations Density 0.015%

    No Known Activations