© 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. 1747
    Prev
    Next
    INDEX
    Explanations

    "be" followed by "like" or "known"

    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
    à¹Į
    -0.13
    HING
    -0.11
    ï¸ı
    -0.10
    æĪijçļĦå¿ĥ
    -0.10
    еннÑĭй
    -0.10
    çļĦæĢĿæĥ³
    -0.10
    浪费
    -0.09
    monic
    -0.09
    зал
    -0.09
     камеÑĢ
    -0.09
    POSITIVE LOGITS
    ä¹Łåı¯èĥ½
    0.10
    ç¼´
    0.09
    veled
    0.08
    cu
    0.08
    elman
    0.08
    angers
    0.08
    boo
    0.08
    ç»Ń
    0.08
     fore
    0.08
    win
    0.08
    Activations Density 0.030%

    No Known Activations