© 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. 3830
    Prev
    Next
    INDEX
    Explanations

    given reasons for

    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
    ä¸Ģè¶Ł
    -0.09
     zag
    -0.09
    txn
    -0.09
    è¦ģ注æĦı
    -0.09
    æķ·
    -0.08
    еÑģÑĤв
    -0.08
    stor
    -0.08
    irl
    -0.08
    ç»ĦæĪIJ
    -0.08
    第ä¸ĢèĬĤ
    -0.08
    POSITIVE LOGITS
    çIJĨçĶ±
    0.15
    为åķ¥
    0.11
    为ä»Ģä¹Īè¦ģ
    0.11
    çļĦçIJĨçĶ±
    0.11
     motives
    0.10
    èĶij
    0.10
    htt
    0.10
    ç¼ĺ
    0.09
    æį®è¯´
    0.09
    为ä»Ģä¹Īä¸į
    0.09
    Activations Density 0.040%

    No Known Activations