Spaces:
Running
Running
| <html lang="ru"> | |
| <head> | |
| <meta charset="utf-8" /> | |
| <meta name="viewport" content="width=device-width, initial-scale=1.0" /> | |
| <title>Bench Local Model Result</title> | |
| <link rel="stylesheet" href="style.css" /> | |
| <style> | |
| .tab-radio { | |
| display: none; | |
| } | |
| .tab-nav { | |
| display: flex; | |
| gap: 8px; | |
| border-bottom: 2px solid #e0e0e0; | |
| margin-bottom: 20px; | |
| } | |
| .tab-label { | |
| padding: 10px 18px; | |
| cursor: pointer; | |
| font-weight: 600; | |
| border-radius: 6px 6px 0 0; | |
| background-color: #f1f3f5; | |
| color: #495057; | |
| transition: background-color 0.2s, color 0.2s; | |
| } | |
| .tab-panel { | |
| display: none; | |
| } | |
| #tab1:checked ~ .tab-nav label[for="tab1"], | |
| #tab2:checked ~ .tab-nav label[for="tab2"] { | |
| background-color: #0d6efd; | |
| color: #ffffff; | |
| } | |
| #tab1:checked ~ .tab-content #panel1, | |
| #tab2:checked ~ .tab-content #panel2 { | |
| display: block; | |
| } | |
| .gallery-vertical { | |
| display: flex; | |
| flex-direction: column; | |
| gap: 24px; | |
| } | |
| .gallery-vertical img { | |
| width: 100%; | |
| height: auto; | |
| border-radius: 8px; | |
| box-shadow: 0 2px 8px rgba(0,0,0,0.1); | |
| } | |
| </style> | |
| </head> | |
| <body> | |
| <div class="card"> | |
| <h1>Bench Local Model Result</h1> | |
| <p><strong>Model:</strong> gemma-4-E2B-it-Q8_0</p> | |
| <p><strong>Setting:</strong> Min_P = 1</p> | |
| <input type="radio" id="tab1" name="tab-group" class="tab-radio" checked /> | |
| <input type="radio" id="tab2" name="tab-group" class="tab-radio" /> | |
| <div class="tab-nav"> | |
| <label for="tab1" class="tab-label">Результаты тестов</label> | |
| <label for="tab2" class="tab-label">Дополнительные графики</label> | |
| </div> | |
| <div class="tab-content"> | |
| <div id="panel1" class="tab-panel"> | |
| <section class="bench-group"> | |
| <h2>Format & Constraints — Score: 95</h2> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/eAuJD0nht38lrDhx7CK84.png" alt="Format & Constraints graph" /> | |
| <ul> | |
| <li>Format Constraints: 80</li> | |
| <li>Ordering and Sorting: 100</li> | |
| <li>Multi-Domain: 100</li> | |
| <li>Precision Under Pressure: 93</li> | |
| <li>Adversarial: 100</li> | |
| </ul> | |
| </section> | |
| <section class="bench-group"> | |
| <h2>Information Extraction: 71</h2> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/BHxtJ2Zz1yDGxGeJL3z2J.png" alt="Information Extraction graph" /> | |
| <ul> | |
| <li>Clean Extraction: 77</li> | |
| <li>Noisy and Informal: 69</li> | |
| <li>Multi-Entity: 80</li> | |
| <li>Implicit and Missing: 69</li> | |
| <li>Complex Documents: 59</li> | |
| </ul> | |
| </section> | |
| <section class="bench-group"> | |
| <h2>Tool Selection — Score: 77</h2> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/4IVwXrKsD527aqIKaBN6T.png" alt="Tool Selection graph" /> | |
| <ul> | |
| <li>Tool Selection: 83</li> | |
| <li>Parameter Precision: 100</li> | |
| <li>Multi-Step Chains: 67</li> | |
| <li>Restraint & Refusal: 83</li> | |
| <li>Error Recovery: 50</li> | |
| </ul> | |
| </section> | |
| <section class="bench-group"> | |
| <h2>Reasoning & Logic — Score: 65</h2> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/gwEsB-9EyZLc_TWt2yu-I.png" alt="Reasoning graph" /> | |
| <ul> | |
| <li>Everyday Arithmetic: 77</li> | |
| <li>Logic Puzzles: 33</li> | |
| <li>Multi-Step Word Problems: 85</li> | |
| <li>Trick Questions and Traps: 77</li> | |
| <li>Applied Reasoning: 62</li> | |
| </ul> | |
| </section> | |
| <section class="bench-group"> | |
| <h2>Conflicts & Overrides — Score: 60</h2> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/Gth2Wt-_ah2aX-2tZkyWm.png" alt="Conflicts graph" /> | |
| <ul> | |
| <li>Hard System vs User (1/3): 33</li> | |
| <li>User Overrides Defaults (3/3): 100</li> | |
| <li>User vs Tool Data (3/3): 100</li> | |
| <li>Hard System vs Tool Data (1/3): 33</li> | |
| <li>Layered Conflicts (1/3): 33</li> | |
| </ul> | |
| </section> | |
| </div> | |
| <div id="panel2" class="tab-panel"> | |
| <h2>Графики и визуализации</h2> | |
| <div class="gallery-vertical"> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/FnATmzaORTRLkgQmWdw9o.png" alt="График 1" /> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/LH08pNK3VW181e0FhXKfY.png" alt="График 2" /> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/z7pwVNQI_NpdJkPw7qVdE.png" alt="График 3" /> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/9jVF7zuefKZcRuGnv0thh.png" alt="График 4" /> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/KwW95vryu8AlGDNzGbFGJ.png" alt="График 5" /> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/iOsoy4M1GC4bBepPyikrL.png" alt="График 6" /> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/SeAg-j_QQ90_d2kVhCYLn.png" alt="График 7" /> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/XOJwyLnzyMPIm3ES434Jz.png" alt="График 8" /> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/3zewAkQaNxRmwrgOfxOa0.png" alt="График 9" /> | |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/631a02620d7478ae006c64da/G2SmFz87zq--FmFciuXMy.png" alt="График 10" /> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </body> | |
| </html> |