© 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. 1157
    Prev
    Next
    INDEX
    Explanations

    lover, relationship, place

    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
    为主é¢ĺçļĦ
    -0.10
    æ¶Ī
    -0.09
    åĨ³
    -0.09
    ÑĢай
    -0.09
    ä¸Ńå¿ĥåŁİå¸Ĥ
    -0.09
    æĬ¹
    -0.08
    èĻ
    -0.08
    æľĢåIJİä¸Ģ次
    -0.08
    خب
    -0.08
    ationToken
    -0.08
    POSITIVE LOGITS
    è¿Ļç±»
    0.09
    æѤ类
    0.09
    çĽ²
    0.09
     Camb
    0.09
    æĪIJè´¥
    0.08
    undler
    0.08
    æ±¹
    0.08
     заÑĢ
    0.08
     szczeg
    0.08
     Blind
    0.08
    Activations Density 0.020%

    No Known Activations