--- base_model: meta-llama/Llama-2-7b-hf library_name: peft license: llama2 tags: - generated_from_trainer model-index: - name: legal_llama results: [] --- # legal_llama This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.8815 - Law Precision: 0.4949 - Law Recall: 0.6622 - Law F1: 0.5665 - Law Number: 74 - Violated by Precision: 0.5116 - Violated by Recall: 0.6197 - Violated by F1: 0.5605 - Violated by Number: 71 - Violated on Precision: 0.2373 - Violated on Recall: 0.28 - Violated on F1: 0.2569 - Violated on Number: 50 - Violation Precision: 0.2670 - Violation Recall: 0.4405 - Violation F1: 0.3325 - Violation Number: 597 - Overall Precision: 0.3011 - Overall Recall: 0.4672 - Overall F1: 0.3662 - Overall Accuracy: 0.9060 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results | Training Loss | Epoch | Step | Validation Loss | Law Precision | Law Recall | Law F1 | Law Number | Violated by Precision | Violated by Recall | Violated by F1 | Violated by Number | Violated on Precision | Violated on Recall | Violated on F1 | Violated on Number | Violation Precision | Violation Recall | Violation F1 | Violation Number | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy | |:-------------:|:-----:|:----:|:---------------:|:-------------:|:----------:|:------:|:----------:|:---------------------:|:------------------:|:--------------:|:------------------:|:---------------------:|:------------------:|:--------------:|:------------------:|:-------------------:|:----------------:|:------------:|:----------------:|:-----------------:|:--------------:|:----------:|:----------------:| | No log | 1.0 | 45 | 0.6132 | 0.0 | 0.0 | 0.0 | 74 | 0.0 | 0.0 | 0.0 | 71 | 0.0 | 0.0 | 0.0 | 50 | 0.0046 | 0.0084 | 0.0059 | 597 | 0.0046 | 0.0063 | 0.0053 | 0.7722 | | No log | 2.0 | 90 | 0.5228 | 0.0170 | 0.0405 | 0.0240 | 74 | 0.0 | 0.0 | 0.0 | 71 | 0.0 | 0.0 | 0.0 | 50 | 0.0242 | 0.0519 | 0.0330 | 597 | 0.0231 | 0.0429 | 0.0301 | 0.8191 | | No log | 3.0 | 135 | 0.4312 | 0.1231 | 0.1081 | 0.1151 | 74 | 0.0526 | 0.0423 | 0.0469 | 71 | 0.0 | 0.0 | 0.0 | 50 | 0.0549 | 0.1240 | 0.0761 | 597 | 0.0575 | 0.1073 | 0.0749 | 0.8525 | | No log | 4.0 | 180 | 0.4691 | 0.2281 | 0.1757 | 0.1985 | 74 | 0.3333 | 0.0986 | 0.1522 | 71 | 0.0 | 0.0 | 0.0 | 50 | 0.0651 | 0.1106 | 0.0819 | 597 | 0.0787 | 0.1086 | 0.0912 | 0.8512 | | No log | 5.0 | 225 | 0.3555 | 0.2388 | 0.2162 | 0.2270 | 74 | 0.3226 | 0.1408 | 0.1961 | 71 | 0.0 | 0.0 | 0.0 | 50 | 0.1525 | 0.3166 | 0.2059 | 597 | 0.1603 | 0.2715 | 0.2016 | 0.8840 | | No log | 6.0 | 270 | 0.3772 | 0.2842 | 0.3649 | 0.3195 | 74 | 0.5818 | 0.4507 | 0.5079 | 71 | 0.1875 | 0.06 | 0.0909 | 50 | 0.1148 | 0.2010 | 0.1462 | 597 | 0.1503 | 0.2298 | 0.1817 | 0.8852 | | No log | 7.0 | 315 | 0.3975 | 0.4342 | 0.4459 | 0.44 | 74 | 0.5 | 0.5634 | 0.5298 | 71 | 0.0952 | 0.12 | 0.1062 | 50 | 0.1262 | 0.2278 | 0.1624 | 597 | 0.1658 | 0.2715 | 0.2058 | 0.8885 | | No log | 8.0 | 360 | 0.3774 | 0.4146 | 0.6892 | 0.5178 | 74 | 0.3636 | 0.6761 | 0.4729 | 71 | 0.125 | 0.16 | 0.1404 | 50 | 0.1783 | 0.3199 | 0.2290 | 597 | 0.2144 | 0.3763 | 0.2731 | 0.8927 | | No log | 9.0 | 405 | 0.4564 | 0.4071 | 0.6216 | 0.4920 | 74 | 0.2740 | 0.5634 | 0.3687 | 71 | 0.12 | 0.18 | 0.144 | 50 | 0.2046 | 0.3400 | 0.2555 | 597 | 0.2247 | 0.3763 | 0.2814 | 0.8942 | | No log | 10.0 | 450 | 0.4914 | 0.51 | 0.6892 | 0.5862 | 74 | 0.4018 | 0.6338 | 0.4918 | 71 | 0.1443 | 0.28 | 0.1905 | 50 | 0.1878 | 0.3250 | 0.2380 | 597 | 0.2265 | 0.3838 | 0.2849 | 0.8925 | | No log | 11.0 | 495 | 0.5342 | 0.3684 | 0.5676 | 0.4468 | 74 | 0.4737 | 0.6338 | 0.5422 | 71 | 0.1633 | 0.16 | 0.1616 | 50 | 0.1859 | 0.3266 | 0.2369 | 597 | 0.2219 | 0.3662 | 0.2763 | 0.8937 | | 0.3436 | 12.0 | 540 | 0.5091 | 0.4545 | 0.6757 | 0.5435 | 74 | 0.3133 | 0.6620 | 0.4253 | 71 | 0.1494 | 0.26 | 0.1898 | 50 | 0.2422 | 0.4405 | 0.3125 | 597 | 0.2603 | 0.4710 | 0.3353 | 0.8975 | | 0.3436 | 13.0 | 585 | 0.5243 | 0.4630 | 0.6757 | 0.5495 | 74 | 0.4034 | 0.6761 | 0.5053 | 71 | 0.1558 | 0.24 | 0.1890 | 50 | 0.2190 | 0.4372 | 0.2918 | 597 | 0.2480 | 0.4684 | 0.3243 | 0.8950 | | 0.3436 | 14.0 | 630 | 0.5990 | 0.3836 | 0.7568 | 0.5091 | 74 | 0.3511 | 0.6479 | 0.4554 | 71 | 0.1647 | 0.28 | 0.2074 | 50 | 0.1954 | 0.3384 | 0.2477 | 597 | 0.2278 | 0.4015 | 0.2907 | 0.8948 | | 0.3436 | 15.0 | 675 | 0.5916 | 0.4483 | 0.7027 | 0.5474 | 74 | 0.3701 | 0.6620 | 0.4747 | 71 | 0.2209 | 0.38 | 0.2794 | 50 | 0.2165 | 0.3869 | 0.2776 | 597 | 0.25 | 0.4407 | 0.3190 | 0.8966 | | 0.3436 | 16.0 | 720 | 0.5755 | 0.4490 | 0.5946 | 0.5116 | 74 | 0.4713 | 0.5775 | 0.5190 | 71 | 0.1970 | 0.26 | 0.2241 | 50 | 0.2228 | 0.4020 | 0.2867 | 597 | 0.2545 | 0.4268 | 0.3189 | 0.8999 | | 0.3436 | 17.0 | 765 | 0.6197 | 0.4815 | 0.7027 | 0.5714 | 74 | 0.4608 | 0.6620 | 0.5434 | 71 | 0.2727 | 0.42 | 0.3307 | 50 | 0.2051 | 0.3484 | 0.2582 | 597 | 0.2521 | 0.4141 | 0.3134 | 0.8956 | | 0.3436 | 18.0 | 810 | 0.6345 | 0.5402 | 0.6351 | 0.5839 | 74 | 0.5152 | 0.4789 | 0.4964 | 71 | 0.22 | 0.22 | 0.22 | 50 | 0.2322 | 0.4037 | 0.2948 | 597 | 0.2683 | 0.4205 | 0.3276 | 0.9011 | | 0.3436 | 19.0 | 855 | 0.6967 | 0.3590 | 0.5676 | 0.4398 | 74 | 0.4167 | 0.5634 | 0.4790 | 71 | 0.2206 | 0.3 | 0.2542 | 50 | 0.1922 | 0.3300 | 0.2429 | 597 | 0.2251 | 0.3712 | 0.2803 | 0.8940 | | 0.3436 | 20.0 | 900 | 0.6172 | 0.4486 | 0.6486 | 0.5304 | 74 | 0.4831 | 0.6056 | 0.5375 | 71 | 0.2286 | 0.32 | 0.2667 | 50 | 0.2313 | 0.4305 | 0.3009 | 597 | 0.2643 | 0.4596 | 0.3356 | 0.8989 | | 0.3436 | 21.0 | 945 | 0.6819 | 0.4717 | 0.6757 | 0.5556 | 74 | 0.4059 | 0.5775 | 0.4767 | 71 | 0.2586 | 0.3 | 0.2778 | 50 | 0.2106 | 0.3601 | 0.2658 | 597 | 0.2496 | 0.4053 | 0.3090 | 0.9004 | | 0.3436 | 22.0 | 990 | 0.6565 | 0.4444 | 0.6486 | 0.5275 | 74 | 0.4725 | 0.6056 | 0.5309 | 71 | 0.2807 | 0.32 | 0.2991 | 50 | 0.2531 | 0.4439 | 0.3224 | 597 | 0.2855 | 0.4697 | 0.3551 | 0.9044 | | 0.0088 | 23.0 | 1035 | 0.6754 | 0.4949 | 0.6622 | 0.5665 | 74 | 0.5176 | 0.6197 | 0.5641 | 71 | 0.25 | 0.28 | 0.2642 | 50 | 0.2547 | 0.4322 | 0.3205 | 597 | 0.2913 | 0.4609 | 0.3570 | 0.9049 | | 0.0088 | 24.0 | 1080 | 0.6782 | 0.48 | 0.6486 | 0.5517 | 74 | 0.5119 | 0.6056 | 0.5548 | 71 | 0.2281 | 0.26 | 0.2430 | 50 | 0.2673 | 0.4523 | 0.3360 | 597 | 0.2990 | 0.4722 | 0.3661 | 0.9062 | | 0.0088 | 25.0 | 1125 | 0.6931 | 0.4902 | 0.6757 | 0.5682 | 74 | 0.5 | 0.6197 | 0.5535 | 71 | 0.2632 | 0.3 | 0.2804 | 50 | 0.2667 | 0.4472 | 0.3342 | 597 | 0.3013 | 0.4747 | 0.3686 | 0.9051 | | 0.0088 | 26.0 | 1170 | 0.6949 | 0.4554 | 0.6216 | 0.5257 | 74 | 0.5422 | 0.6338 | 0.5844 | 71 | 0.2407 | 0.26 | 0.25 | 50 | 0.2689 | 0.4523 | 0.3373 | 597 | 0.3011 | 0.4722 | 0.3677 | 0.9052 | | 0.0088 | 27.0 | 1215 | 0.7091 | 0.4752 | 0.6486 | 0.5486 | 74 | 0.5181 | 0.6056 | 0.5584 | 71 | 0.2414 | 0.28 | 0.2593 | 50 | 0.2604 | 0.4405 | 0.3273 | 597 | 0.2939 | 0.4646 | 0.3601 | 0.9055 | | 0.0088 | 28.0 | 1260 | 0.7128 | 0.5 | 0.6892 | 0.5795 | 74 | 0.4944 | 0.6197 | 0.55 | 71 | 0.2586 | 0.3 | 0.2778 | 50 | 0.2657 | 0.4456 | 0.3329 | 597 | 0.3008 | 0.4747 | 0.3683 | 0.9055 | | 0.0088 | 29.0 | 1305 | 0.7147 | 0.4653 | 0.6351 | 0.5371 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2456 | 0.28 | 0.2617 | 50 | 0.2659 | 0.4489 | 0.3340 | 597 | 0.2977 | 0.4710 | 0.3648 | 0.9058 | | 0.0088 | 30.0 | 1350 | 0.7223 | 0.4653 | 0.6351 | 0.5371 | 74 | 0.5 | 0.6056 | 0.5478 | 71 | 0.2414 | 0.28 | 0.2593 | 50 | 0.2639 | 0.4456 | 0.3315 | 597 | 0.2953 | 0.4672 | 0.3619 | 0.9056 | | 0.0088 | 31.0 | 1395 | 0.7284 | 0.4752 | 0.6486 | 0.5486 | 74 | 0.4945 | 0.6338 | 0.5556 | 71 | 0.2069 | 0.24 | 0.2222 | 50 | 0.2696 | 0.4489 | 0.3369 | 597 | 0.2998 | 0.4710 | 0.3664 | 0.9059 | | 0.0088 | 32.0 | 1440 | 0.7287 | 0.49 | 0.6622 | 0.5632 | 74 | 0.5287 | 0.6479 | 0.5823 | 71 | 0.2667 | 0.32 | 0.2909 | 50 | 0.2654 | 0.4472 | 0.3331 | 597 | 0.3017 | 0.4773 | 0.3697 | 0.9055 | | 0.0088 | 33.0 | 1485 | 0.7377 | 0.48 | 0.6486 | 0.5517 | 74 | 0.5357 | 0.6338 | 0.5806 | 71 | 0.2069 | 0.24 | 0.2222 | 50 | 0.2676 | 0.4456 | 0.3344 | 597 | 0.3002 | 0.4684 | 0.3659 | 0.9059 | | 0.0003 | 34.0 | 1530 | 0.7451 | 0.4747 | 0.6351 | 0.5434 | 74 | 0.5357 | 0.6338 | 0.5806 | 71 | 0.2143 | 0.24 | 0.2264 | 50 | 0.2694 | 0.4472 | 0.3363 | 597 | 0.3016 | 0.4684 | 0.3670 | 0.9062 | | 0.0003 | 35.0 | 1575 | 0.7493 | 0.48 | 0.6486 | 0.5517 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.1754 | 0.2 | 0.1869 | 50 | 0.2739 | 0.4506 | 0.3407 | 597 | 0.3029 | 0.4684 | 0.3679 | 0.9058 | | 0.0003 | 36.0 | 1620 | 0.7565 | 0.48 | 0.6486 | 0.5517 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2069 | 0.24 | 0.2222 | 50 | 0.2660 | 0.4389 | 0.3312 | 597 | 0.2978 | 0.4621 | 0.3622 | 0.9060 | | 0.0003 | 37.0 | 1665 | 0.7559 | 0.48 | 0.6486 | 0.5517 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2069 | 0.24 | 0.2222 | 50 | 0.2691 | 0.4422 | 0.3346 | 597 | 0.3004 | 0.4646 | 0.3649 | 0.9058 | | 0.0003 | 38.0 | 1710 | 0.7632 | 0.5 | 0.6486 | 0.5647 | 74 | 0.5357 | 0.6338 | 0.5806 | 71 | 0.2069 | 0.24 | 0.2222 | 50 | 0.2676 | 0.4389 | 0.3325 | 597 | 0.3016 | 0.4634 | 0.3654 | 0.9062 | | 0.0003 | 39.0 | 1755 | 0.7659 | 0.4747 | 0.6351 | 0.5434 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.1905 | 0.24 | 0.2124 | 50 | 0.2621 | 0.4338 | 0.3268 | 597 | 0.2926 | 0.4571 | 0.3568 | 0.9057 | | 0.0003 | 40.0 | 1800 | 0.7705 | 0.4898 | 0.6486 | 0.5581 | 74 | 0.5 | 0.6197 | 0.5535 | 71 | 0.2034 | 0.24 | 0.2202 | 50 | 0.2642 | 0.4355 | 0.3289 | 597 | 0.2962 | 0.4596 | 0.3602 | 0.9058 | | 0.0003 | 41.0 | 1845 | 0.7740 | 0.4948 | 0.6486 | 0.5614 | 74 | 0.5244 | 0.6056 | 0.5621 | 71 | 0.2069 | 0.24 | 0.2222 | 50 | 0.2704 | 0.4439 | 0.3361 | 597 | 0.3024 | 0.4646 | 0.3664 | 0.9062 | | 0.0003 | 42.0 | 1890 | 0.7735 | 0.4752 | 0.6486 | 0.5486 | 74 | 0.4944 | 0.6197 | 0.55 | 71 | 0.2034 | 0.24 | 0.2202 | 50 | 0.2666 | 0.4439 | 0.3331 | 597 | 0.2969 | 0.4659 | 0.3627 | 0.9059 | | 0.0003 | 43.0 | 1935 | 0.7785 | 0.49 | 0.6622 | 0.5632 | 74 | 0.4944 | 0.6197 | 0.55 | 71 | 0.2 | 0.24 | 0.2182 | 50 | 0.2654 | 0.4405 | 0.3312 | 597 | 0.2968 | 0.4646 | 0.3622 | 0.9060 | | 0.0003 | 44.0 | 1980 | 0.7821 | 0.4898 | 0.6486 | 0.5581 | 74 | 0.5294 | 0.6338 | 0.5769 | 71 | 0.2 | 0.24 | 0.2182 | 50 | 0.2652 | 0.4372 | 0.3302 | 597 | 0.2983 | 0.4621 | 0.3626 | 0.9062 | | 0.0001 | 45.0 | 2025 | 0.7853 | 0.4848 | 0.6486 | 0.5549 | 74 | 0.5238 | 0.6197 | 0.5677 | 71 | 0.2034 | 0.24 | 0.2202 | 50 | 0.2667 | 0.4422 | 0.3327 | 597 | 0.2987 | 0.4646 | 0.3636 | 0.9065 | | 0.0001 | 46.0 | 2070 | 0.7897 | 0.4898 | 0.6486 | 0.5581 | 74 | 0.5294 | 0.6338 | 0.5769 | 71 | 0.2203 | 0.26 | 0.2385 | 50 | 0.2640 | 0.4355 | 0.3287 | 597 | 0.2983 | 0.4621 | 0.3626 | 0.9062 | | 0.0001 | 47.0 | 2115 | 0.7899 | 0.48 | 0.6486 | 0.5517 | 74 | 0.5172 | 0.6338 | 0.5696 | 71 | 0.2097 | 0.26 | 0.2321 | 50 | 0.2670 | 0.4405 | 0.3325 | 597 | 0.2990 | 0.4659 | 0.3643 | 0.9065 | | 0.0001 | 48.0 | 2160 | 0.7959 | 0.48 | 0.6486 | 0.5517 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2203 | 0.26 | 0.2385 | 50 | 0.2682 | 0.4389 | 0.3329 | 597 | 0.3003 | 0.4634 | 0.3644 | 0.9064 | | 0.0001 | 49.0 | 2205 | 0.7945 | 0.49 | 0.6622 | 0.5632 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2295 | 0.28 | 0.2523 | 50 | 0.2685 | 0.4439 | 0.3346 | 597 | 0.3012 | 0.4697 | 0.3670 | 0.9066 | | 0.0001 | 50.0 | 2250 | 0.7989 | 0.48 | 0.6486 | 0.5517 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.25 | 0.3 | 0.2727 | 50 | 0.2672 | 0.4422 | 0.3331 | 597 | 0.3006 | 0.4684 | 0.3662 | 0.9061 | | 0.0001 | 51.0 | 2295 | 0.8055 | 0.48 | 0.6486 | 0.5517 | 74 | 0.5238 | 0.6197 | 0.5677 | 71 | 0.2414 | 0.28 | 0.2593 | 50 | 0.2627 | 0.4338 | 0.3272 | 597 | 0.2972 | 0.4609 | 0.3614 | 0.9060 | | 0.0001 | 52.0 | 2340 | 0.8055 | 0.49 | 0.6622 | 0.5632 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2333 | 0.28 | 0.2545 | 50 | 0.2636 | 0.4389 | 0.3294 | 597 | 0.2976 | 0.4659 | 0.3632 | 0.9064 | | 0.0001 | 53.0 | 2385 | 0.8095 | 0.4898 | 0.6486 | 0.5581 | 74 | 0.5233 | 0.6338 | 0.5732 | 71 | 0.2241 | 0.26 | 0.2407 | 50 | 0.2665 | 0.4405 | 0.3321 | 597 | 0.3002 | 0.4659 | 0.3652 | 0.9066 | | 0.0001 | 54.0 | 2430 | 0.8113 | 0.49 | 0.6622 | 0.5632 | 74 | 0.5233 | 0.6338 | 0.5732 | 71 | 0.2 | 0.24 | 0.2182 | 50 | 0.2684 | 0.4405 | 0.3335 | 597 | 0.3010 | 0.4659 | 0.3657 | 0.9065 | | 0.0001 | 55.0 | 2475 | 0.8160 | 0.4706 | 0.6486 | 0.5455 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2069 | 0.24 | 0.2222 | 50 | 0.2636 | 0.4372 | 0.3289 | 597 | 0.2953 | 0.4609 | 0.3600 | 0.9062 | | 0.0001 | 56.0 | 2520 | 0.8221 | 0.4949 | 0.6622 | 0.5665 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2167 | 0.26 | 0.2364 | 50 | 0.2673 | 0.4405 | 0.3327 | 597 | 0.3002 | 0.4659 | 0.3652 | 0.9058 | | 0.0001 | 57.0 | 2565 | 0.8214 | 0.5102 | 0.6757 | 0.5814 | 74 | 0.5233 | 0.6338 | 0.5732 | 71 | 0.2295 | 0.28 | 0.2523 | 50 | 0.2685 | 0.4439 | 0.3346 | 597 | 0.3036 | 0.4722 | 0.3696 | 0.9061 | | 0.0001 | 58.0 | 2610 | 0.8212 | 0.49 | 0.6622 | 0.5632 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2295 | 0.28 | 0.2523 | 50 | 0.2672 | 0.4422 | 0.3331 | 597 | 0.3004 | 0.4684 | 0.3661 | 0.9060 | | 0.0001 | 59.0 | 2655 | 0.8258 | 0.4848 | 0.6486 | 0.5549 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2281 | 0.26 | 0.2430 | 50 | 0.2681 | 0.4456 | 0.3348 | 597 | 0.3006 | 0.4684 | 0.3662 | 0.9062 | | 0.0001 | 60.0 | 2700 | 0.8291 | 0.49 | 0.6622 | 0.5632 | 74 | 0.5233 | 0.6338 | 0.5732 | 71 | 0.2203 | 0.26 | 0.2385 | 50 | 0.2662 | 0.4405 | 0.3319 | 597 | 0.3001 | 0.4672 | 0.3654 | 0.9060 | | 0.0001 | 61.0 | 2745 | 0.8327 | 0.4848 | 0.6486 | 0.5549 | 74 | 0.5172 | 0.6338 | 0.5696 | 71 | 0.2333 | 0.28 | 0.2545 | 50 | 0.2678 | 0.4405 | 0.3331 | 597 | 0.3013 | 0.4672 | 0.3663 | 0.9063 | | 0.0001 | 62.0 | 2790 | 0.8328 | 0.4851 | 0.6622 | 0.5600 | 74 | 0.5294 | 0.6338 | 0.5769 | 71 | 0.2414 | 0.28 | 0.2593 | 50 | 0.2683 | 0.4422 | 0.3340 | 597 | 0.3029 | 0.4697 | 0.3683 | 0.9064 | | 0.0001 | 63.0 | 2835 | 0.8346 | 0.49 | 0.6622 | 0.5632 | 74 | 0.5233 | 0.6338 | 0.5732 | 71 | 0.2241 | 0.26 | 0.2407 | 50 | 0.2701 | 0.4439 | 0.3359 | 597 | 0.3037 | 0.4697 | 0.3689 | 0.9060 | | 0.0001 | 64.0 | 2880 | 0.8394 | 0.4653 | 0.6351 | 0.5371 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2414 | 0.28 | 0.2593 | 50 | 0.2706 | 0.4456 | 0.3367 | 597 | 0.3021 | 0.4684 | 0.3673 | 0.9063 | | 0.0001 | 65.0 | 2925 | 0.8417 | 0.4752 | 0.6486 | 0.5486 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2069 | 0.24 | 0.2222 | 50 | 0.2667 | 0.4405 | 0.3323 | 597 | 0.2981 | 0.4634 | 0.3628 | 0.9060 | | 0.0001 | 66.0 | 2970 | 0.8386 | 0.5 | 0.6757 | 0.5747 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2241 | 0.26 | 0.2407 | 50 | 0.2695 | 0.4456 | 0.3359 | 597 | 0.3030 | 0.4710 | 0.3688 | 0.9061 | | 0.0001 | 67.0 | 3015 | 0.8435 | 0.5102 | 0.6757 | 0.5814 | 74 | 0.4828 | 0.5915 | 0.5316 | 71 | 0.2105 | 0.24 | 0.2243 | 50 | 0.2682 | 0.4439 | 0.3344 | 597 | 0.3 | 0.4659 | 0.3650 | 0.9061 | | 0.0001 | 68.0 | 3060 | 0.8453 | 0.5 | 0.6622 | 0.5698 | 74 | 0.5233 | 0.6338 | 0.5732 | 71 | 0.2542 | 0.3 | 0.2752 | 50 | 0.2663 | 0.4389 | 0.3314 | 597 | 0.3024 | 0.4684 | 0.3675 | 0.9061 | | 0.0001 | 69.0 | 3105 | 0.8487 | 0.4898 | 0.6486 | 0.5581 | 74 | 0.4886 | 0.6056 | 0.5409 | 71 | 0.2333 | 0.28 | 0.2545 | 50 | 0.2704 | 0.4439 | 0.3361 | 597 | 0.3018 | 0.4672 | 0.3667 | 0.9061 | | 0.0001 | 70.0 | 3150 | 0.8510 | 0.5 | 0.6622 | 0.5698 | 74 | 0.4886 | 0.6056 | 0.5409 | 71 | 0.2333 | 0.28 | 0.2545 | 50 | 0.2696 | 0.4439 | 0.3354 | 597 | 0.3019 | 0.4684 | 0.3671 | 0.9062 | | 0.0001 | 71.0 | 3195 | 0.8522 | 0.4949 | 0.6622 | 0.5665 | 74 | 0.4943 | 0.6056 | 0.5443 | 71 | 0.2241 | 0.26 | 0.2407 | 50 | 0.2657 | 0.4405 | 0.3314 | 597 | 0.2982 | 0.4646 | 0.3633 | 0.9058 | | 0.0001 | 72.0 | 3240 | 0.8543 | 0.5155 | 0.6757 | 0.5848 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2105 | 0.24 | 0.2243 | 50 | 0.2686 | 0.4422 | 0.3342 | 597 | 0.3023 | 0.4672 | 0.3671 | 0.9062 | | 0.0001 | 73.0 | 3285 | 0.8577 | 0.4653 | 0.6351 | 0.5371 | 74 | 0.5 | 0.6197 | 0.5535 | 71 | 0.2295 | 0.28 | 0.2523 | 50 | 0.2693 | 0.4439 | 0.3352 | 597 | 0.2998 | 0.4672 | 0.3653 | 0.9059 | | 0.0001 | 74.0 | 3330 | 0.8584 | 0.5051 | 0.6757 | 0.5780 | 74 | 0.5 | 0.6197 | 0.5535 | 71 | 0.25 | 0.3 | 0.2727 | 50 | 0.2678 | 0.4405 | 0.3331 | 597 | 0.3027 | 0.4697 | 0.3681 | 0.9060 | | 0.0001 | 75.0 | 3375 | 0.8609 | 0.5053 | 0.6486 | 0.5680 | 74 | 0.4886 | 0.6056 | 0.5409 | 71 | 0.2281 | 0.26 | 0.2430 | 50 | 0.2703 | 0.4456 | 0.3365 | 597 | 0.3023 | 0.4672 | 0.3671 | 0.9063 | | 0.0001 | 76.0 | 3420 | 0.8586 | 0.5102 | 0.6757 | 0.5814 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2333 | 0.28 | 0.2545 | 50 | 0.2675 | 0.4422 | 0.3333 | 597 | 0.3019 | 0.4697 | 0.3676 | 0.9061 | | 0.0001 | 77.0 | 3465 | 0.8621 | 0.5052 | 0.6622 | 0.5731 | 74 | 0.4831 | 0.6056 | 0.5375 | 71 | 0.2295 | 0.28 | 0.2523 | 50 | 0.2685 | 0.4439 | 0.3346 | 597 | 0.3006 | 0.4684 | 0.3662 | 0.9061 | | 0.0 | 78.0 | 3510 | 0.8638 | 0.5 | 0.6622 | 0.5698 | 74 | 0.5 | 0.6197 | 0.5535 | 71 | 0.2542 | 0.3 | 0.2752 | 50 | 0.2675 | 0.4422 | 0.3333 | 597 | 0.3019 | 0.4697 | 0.3676 | 0.9062 | | 0.0 | 79.0 | 3555 | 0.8642 | 0.4949 | 0.6622 | 0.5665 | 74 | 0.4944 | 0.6197 | 0.55 | 71 | 0.25 | 0.3 | 0.2727 | 50 | 0.2706 | 0.4456 | 0.3367 | 597 | 0.3038 | 0.4722 | 0.3697 | 0.9060 | | 0.0 | 80.0 | 3600 | 0.8647 | 0.48 | 0.6486 | 0.5517 | 74 | 0.5056 | 0.6338 | 0.5625 | 71 | 0.2295 | 0.28 | 0.2523 | 50 | 0.2677 | 0.4422 | 0.3335 | 597 | 0.3002 | 0.4684 | 0.3659 | 0.9059 | | 0.0 | 81.0 | 3645 | 0.8701 | 0.4747 | 0.6351 | 0.5434 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2373 | 0.28 | 0.2569 | 50 | 0.2691 | 0.4422 | 0.3346 | 597 | 0.3010 | 0.4659 | 0.3657 | 0.9063 | | 0.0 | 82.0 | 3690 | 0.8683 | 0.5052 | 0.6622 | 0.5731 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.25 | 0.3 | 0.2727 | 50 | 0.2680 | 0.4422 | 0.3338 | 597 | 0.3029 | 0.4697 | 0.3683 | 0.9063 | | 0.0 | 83.0 | 3735 | 0.8703 | 0.5052 | 0.6622 | 0.5731 | 74 | 0.5 | 0.6197 | 0.5535 | 71 | 0.2542 | 0.3 | 0.2752 | 50 | 0.2688 | 0.4422 | 0.3344 | 597 | 0.3034 | 0.4697 | 0.3687 | 0.9060 | | 0.0 | 84.0 | 3780 | 0.8720 | 0.5155 | 0.6757 | 0.5848 | 74 | 0.5 | 0.6197 | 0.5535 | 71 | 0.2586 | 0.3 | 0.2778 | 50 | 0.2657 | 0.4389 | 0.3310 | 597 | 0.3019 | 0.4684 | 0.3671 | 0.9062 | | 0.0 | 85.0 | 3825 | 0.8745 | 0.49 | 0.6622 | 0.5632 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2373 | 0.28 | 0.2569 | 50 | 0.2660 | 0.4389 | 0.3312 | 597 | 0.2998 | 0.4659 | 0.3648 | 0.9056 | | 0.0 | 86.0 | 3870 | 0.8757 | 0.5 | 0.6622 | 0.5698 | 74 | 0.5 | 0.6197 | 0.5535 | 71 | 0.25 | 0.3 | 0.2727 | 50 | 0.2649 | 0.4389 | 0.3304 | 597 | 0.2996 | 0.4672 | 0.3651 | 0.9058 | | 0.0 | 87.0 | 3915 | 0.8754 | 0.49 | 0.6622 | 0.5632 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2586 | 0.3 | 0.2778 | 50 | 0.2678 | 0.4405 | 0.3331 | 597 | 0.3024 | 0.4684 | 0.3675 | 0.9059 | | 0.0 | 88.0 | 3960 | 0.8744 | 0.5 | 0.6622 | 0.5698 | 74 | 0.4944 | 0.6197 | 0.55 | 71 | 0.2333 | 0.28 | 0.2545 | 50 | 0.2670 | 0.4405 | 0.3325 | 597 | 0.3003 | 0.4672 | 0.3656 | 0.9058 | | 0.0 | 89.0 | 4005 | 0.8789 | 0.5052 | 0.6622 | 0.5731 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2333 | 0.28 | 0.2545 | 50 | 0.2696 | 0.4439 | 0.3354 | 597 | 0.3032 | 0.4697 | 0.3685 | 0.9058 | | 0.0 | 90.0 | 4050 | 0.8788 | 0.4949 | 0.6622 | 0.5665 | 74 | 0.5 | 0.6197 | 0.5535 | 71 | 0.2542 | 0.3 | 0.2752 | 50 | 0.2677 | 0.4422 | 0.3335 | 597 | 0.3019 | 0.4697 | 0.3676 | 0.9058 | | 0.0 | 91.0 | 4095 | 0.8795 | 0.5 | 0.6757 | 0.5747 | 74 | 0.4889 | 0.6197 | 0.5466 | 71 | 0.25 | 0.3 | 0.2727 | 50 | 0.2674 | 0.4439 | 0.3338 | 597 | 0.3014 | 0.4722 | 0.3679 | 0.9058 | | 0.0 | 92.0 | 4140 | 0.8818 | 0.5052 | 0.6622 | 0.5731 | 74 | 0.4943 | 0.6056 | 0.5443 | 71 | 0.2333 | 0.28 | 0.2545 | 50 | 0.2688 | 0.4439 | 0.3348 | 597 | 0.3016 | 0.4684 | 0.3670 | 0.9059 | | 0.0 | 93.0 | 4185 | 0.8807 | 0.5 | 0.6622 | 0.5698 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2373 | 0.28 | 0.2569 | 50 | 0.2696 | 0.4439 | 0.3354 | 597 | 0.3032 | 0.4697 | 0.3685 | 0.9062 | | 0.0 | 94.0 | 4230 | 0.8810 | 0.5 | 0.6622 | 0.5698 | 74 | 0.4886 | 0.6056 | 0.5409 | 71 | 0.2167 | 0.26 | 0.2364 | 50 | 0.2677 | 0.4422 | 0.3335 | 597 | 0.2995 | 0.4659 | 0.3646 | 0.9059 | | 0.0 | 95.0 | 4275 | 0.8803 | 0.5 | 0.6622 | 0.5698 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2373 | 0.28 | 0.2569 | 50 | 0.2675 | 0.4422 | 0.3333 | 597 | 0.3014 | 0.4684 | 0.3668 | 0.9059 | | 0.0 | 96.0 | 4320 | 0.8808 | 0.5102 | 0.6757 | 0.5814 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2333 | 0.28 | 0.2545 | 50 | 0.2662 | 0.4405 | 0.3319 | 597 | 0.3009 | 0.4684 | 0.3664 | 0.9059 | | 0.0 | 97.0 | 4365 | 0.8830 | 0.5 | 0.6622 | 0.5698 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2414 | 0.28 | 0.2593 | 50 | 0.2686 | 0.4422 | 0.3342 | 597 | 0.3026 | 0.4684 | 0.3677 | 0.9059 | | 0.0 | 98.0 | 4410 | 0.8807 | 0.5052 | 0.6622 | 0.5731 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2373 | 0.28 | 0.2569 | 50 | 0.2669 | 0.4422 | 0.3329 | 597 | 0.3011 | 0.4684 | 0.3666 | 0.9059 | | 0.0 | 99.0 | 4455 | 0.8822 | 0.5052 | 0.6622 | 0.5731 | 74 | 0.5057 | 0.6197 | 0.5570 | 71 | 0.2414 | 0.28 | 0.2593 | 50 | 0.2682 | 0.4439 | 0.3344 | 597 | 0.3024 | 0.4697 | 0.3680 | 0.9058 | | 0.0 | 100.0 | 4500 | 0.8815 | 0.4949 | 0.6622 | 0.5665 | 74 | 0.5116 | 0.6197 | 0.5605 | 71 | 0.2373 | 0.28 | 0.2569 | 50 | 0.2670 | 0.4405 | 0.3325 | 597 | 0.3011 | 0.4672 | 0.3662 | 0.9060 | ### Framework versions - PEFT 0.12.0 - Transformers 4.44.0 - Pytorch 2.4.0+cu121 - Datasets 2.20.0 - Tokenizers 0.19.1