© 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. 22212
    Prev
    Next
    INDEX
    Explanations

    locations or groups

    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
    апÑĢ
    -0.11
    åı¤äºº
    -0.10
    nor
    -0.10
    æĢİä¹Ī说
    -0.10
    avit
    -0.10
    éĻIJæŶ
    -0.09
    GST
    -0.09
    æĵĴ
    -0.09
    èĢģ人家
    -0.09
    ROME
    -0.09
    POSITIVE LOGITS
    æĪijæł¡
    0.20
    æĪijéĻ¢
    0.15
     ours
    0.13
    ç¬ĶèĢħ
    0.13
    æĪij家
    0.12
    å°ıç¼ĸ
    0.11
    æĪijçľģ
    0.11
    åľ¨æĪij
    0.11
    æĪijåĸľæ¬¢
    0.10
    ç§ģãĤĤ
    0.10
    Activations Density 0.514%

    No Known Activations