ArabicNewSplits7_B_usingWellWrittenEssays_FineTuningAraBERT_run2_AugV5_k6_task5_organization

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6306
  • Qwk: 0.5818
  • Mse: 0.6306
  • Rmse: 0.7941

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: 2e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • 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 Qwk Mse Rmse
No log 0.0625 2 3.8318 -0.0046 3.8318 1.9575
No log 0.125 4 2.0455 0.0032 2.0455 1.4302
No log 0.1875 6 1.4902 0.0030 1.4902 1.2208
No log 0.25 8 1.2208 0.0380 1.2208 1.1049
No log 0.3125 10 1.0912 0.2611 1.0912 1.0446
No log 0.375 12 1.0446 0.1233 1.0446 1.0221
No log 0.4375 14 1.1563 0.0417 1.1563 1.0753
No log 0.5 16 1.1839 0.1727 1.1839 1.0881
No log 0.5625 18 1.3017 0.1370 1.3017 1.1409
No log 0.625 20 1.2762 0.2308 1.2762 1.1297
No log 0.6875 22 1.0551 0.2035 1.0551 1.0272
No log 0.75 24 1.0046 0.2643 1.0046 1.0023
No log 0.8125 26 1.0163 0.2674 1.0163 1.0081
No log 0.875 28 1.2096 0.2084 1.2096 1.0998
No log 0.9375 30 1.4982 0.2474 1.4982 1.2240
No log 1.0 32 1.4608 0.2070 1.4608 1.2086
No log 1.0625 34 1.2222 0.1552 1.2222 1.1055
No log 1.125 36 0.8781 0.3753 0.8781 0.9371
No log 1.1875 38 1.0700 0.1233 1.0700 1.0344
No log 1.25 40 1.1015 0.1233 1.1015 1.0495
No log 1.3125 42 0.8820 0.3625 0.8820 0.9391
No log 1.375 44 0.8125 0.4624 0.8125 0.9014
No log 1.4375 46 0.7701 0.5165 0.7701 0.8775
No log 1.5 48 0.8188 0.4861 0.8188 0.9049
No log 1.5625 50 1.1387 0.2395 1.1387 1.0671
No log 1.625 52 1.1312 0.2283 1.1312 1.0636
No log 1.6875 54 0.8064 0.4741 0.8064 0.8980
No log 1.75 56 0.7298 0.5333 0.7298 0.8543
No log 1.8125 58 0.9355 0.3792 0.9355 0.9672
No log 1.875 60 0.8526 0.4144 0.8526 0.9234
No log 1.9375 62 0.7065 0.5051 0.7065 0.8405
No log 2.0 64 0.8725 0.2758 0.8725 0.9341
No log 2.0625 66 0.9144 0.2758 0.9144 0.9563
No log 2.125 68 0.8326 0.3734 0.8326 0.9124
No log 2.1875 70 0.7555 0.5214 0.7555 0.8692
No log 2.25 72 0.7168 0.5214 0.7168 0.8466
No log 2.3125 74 0.7697 0.5573 0.7697 0.8773
No log 2.375 76 0.7773 0.5841 0.7773 0.8816
No log 2.4375 78 0.7589 0.5919 0.7589 0.8711
No log 2.5 80 0.7450 0.5830 0.7450 0.8632
No log 2.5625 82 0.7193 0.5133 0.7193 0.8481
No log 2.625 84 0.7566 0.5345 0.7566 0.8698
No log 2.6875 86 0.7407 0.5117 0.7407 0.8606
No log 2.75 88 0.6689 0.5809 0.6689 0.8178
No log 2.8125 90 0.8435 0.5946 0.8435 0.9184
No log 2.875 92 1.0919 0.4705 1.0919 1.0449
No log 2.9375 94 1.0126 0.4655 1.0126 1.0063
No log 3.0 96 0.7120 0.5585 0.7120 0.8438
No log 3.0625 98 0.6381 0.5404 0.6381 0.7988
No log 3.125 100 0.6655 0.5892 0.6655 0.8158
No log 3.1875 102 0.6773 0.5752 0.6773 0.8230
No log 3.25 104 0.9116 0.4917 0.9116 0.9548
No log 3.3125 106 1.1538 0.4694 1.1538 1.0742
No log 3.375 108 1.1146 0.4975 1.1146 1.0557
No log 3.4375 110 0.8457 0.5134 0.8457 0.9196
No log 3.5 112 0.6833 0.5002 0.6833 0.8266
No log 3.5625 114 0.6904 0.5246 0.6904 0.8309
No log 3.625 116 0.6860 0.5522 0.6860 0.8283
No log 3.6875 118 0.7847 0.5468 0.7847 0.8858
No log 3.75 120 1.0945 0.4531 1.0945 1.0462
No log 3.8125 122 1.2415 0.4387 1.2415 1.1142
No log 3.875 124 1.0859 0.4745 1.0859 1.0421
No log 3.9375 126 0.8932 0.5016 0.8932 0.9451
No log 4.0 128 0.7480 0.6142 0.7480 0.8649
No log 4.0625 130 0.7393 0.6231 0.7393 0.8598
No log 4.125 132 0.6974 0.5702 0.6974 0.8351
No log 4.1875 134 0.6850 0.5961 0.6850 0.8276
No log 4.25 136 0.7544 0.6004 0.7544 0.8686
No log 4.3125 138 0.8110 0.5966 0.8110 0.9005
No log 4.375 140 0.7358 0.6386 0.7358 0.8578
No log 4.4375 142 0.6444 0.5966 0.6444 0.8027
No log 4.5 144 0.6464 0.6606 0.6464 0.8040
No log 4.5625 146 0.6432 0.6144 0.6432 0.8020
No log 4.625 148 0.7604 0.6100 0.7604 0.8720
No log 4.6875 150 0.8825 0.5781 0.8825 0.9394
No log 4.75 152 0.8899 0.5520 0.8899 0.9433
No log 4.8125 154 0.7858 0.5632 0.7858 0.8865
No log 4.875 156 0.6552 0.6292 0.6552 0.8094
No log 4.9375 158 0.6596 0.5894 0.6596 0.8121
No log 5.0 160 0.6671 0.5774 0.6671 0.8168
No log 5.0625 162 0.6416 0.5274 0.6416 0.8010
No log 5.125 164 0.7081 0.5575 0.7081 0.8415
No log 5.1875 166 0.8788 0.4908 0.8788 0.9374
No log 5.25 168 0.9073 0.5357 0.9073 0.9525
No log 5.3125 170 0.7723 0.6204 0.7723 0.8788
No log 5.375 172 0.6209 0.5969 0.6209 0.7880
No log 5.4375 174 0.6003 0.5972 0.6003 0.7748
No log 5.5 176 0.6069 0.6277 0.6069 0.7790
No log 5.5625 178 0.5912 0.5607 0.5912 0.7689
No log 5.625 180 0.6044 0.5969 0.6044 0.7775
No log 5.6875 182 0.6659 0.6900 0.6659 0.8160
No log 5.75 184 0.6513 0.6748 0.6513 0.8070
No log 5.8125 186 0.6224 0.6649 0.6224 0.7889
No log 5.875 188 0.6481 0.6498 0.6481 0.8051
No log 5.9375 190 0.6905 0.6002 0.6905 0.8310
No log 6.0 192 0.8493 0.5627 0.8493 0.9216
No log 6.0625 194 0.8953 0.5541 0.8953 0.9462
No log 6.125 196 0.7158 0.6266 0.7158 0.8461
No log 6.1875 198 0.5801 0.7098 0.5801 0.7617
No log 6.25 200 0.5669 0.6830 0.5669 0.7529
No log 6.3125 202 0.5796 0.6729 0.5796 0.7613
No log 6.375 204 0.6823 0.6266 0.6823 0.8260
No log 6.4375 206 0.7487 0.5978 0.7487 0.8653
No log 6.5 208 0.7133 0.6184 0.7133 0.8446
No log 6.5625 210 0.6499 0.5386 0.6499 0.8062
No log 6.625 212 0.6442 0.6464 0.6442 0.8026
No log 6.6875 214 0.6421 0.6619 0.6421 0.8013
No log 6.75 216 0.6541 0.6383 0.6541 0.8088
No log 6.8125 218 0.7398 0.5246 0.7398 0.8601
No log 6.875 220 0.7497 0.5471 0.7497 0.8659
No log 6.9375 222 0.6822 0.5887 0.6822 0.8259
No log 7.0 224 0.7047 0.6004 0.7047 0.8395
No log 7.0625 226 0.7904 0.5759 0.7904 0.8890
No log 7.125 228 0.8548 0.5799 0.8548 0.9245
No log 7.1875 230 0.8531 0.5714 0.8531 0.9236
No log 7.25 232 0.6677 0.6747 0.6677 0.8172
No log 7.3125 234 0.5985 0.6565 0.5985 0.7737
No log 7.375 236 0.5960 0.6758 0.5960 0.7720
No log 7.4375 238 0.5739 0.7273 0.5739 0.7576
No log 7.5 240 0.5609 0.6830 0.5609 0.7489
No log 7.5625 242 0.6240 0.7156 0.6240 0.7899
No log 7.625 244 0.7439 0.5251 0.7439 0.8625
No log 7.6875 246 0.7105 0.6007 0.7105 0.8429
No log 7.75 248 0.6012 0.7191 0.6012 0.7754
No log 7.8125 250 0.6248 0.6245 0.6248 0.7904
No log 7.875 252 0.6437 0.6073 0.6437 0.8023
No log 7.9375 254 0.6007 0.6187 0.6007 0.7750
No log 8.0 256 0.5986 0.6919 0.5986 0.7737
No log 8.0625 258 0.6443 0.6166 0.6443 0.8027
No log 8.125 260 0.6177 0.6669 0.6177 0.7859
No log 8.1875 262 0.5811 0.6952 0.5811 0.7623
No log 8.25 264 0.5816 0.6952 0.5816 0.7627
No log 8.3125 266 0.5896 0.6445 0.5896 0.7679
No log 8.375 268 0.6182 0.6394 0.6182 0.7862
No log 8.4375 270 0.5960 0.6578 0.5960 0.7720
No log 8.5 272 0.6085 0.6485 0.6085 0.7801
No log 8.5625 274 0.6795 0.6064 0.6795 0.8243
No log 8.625 276 0.6731 0.6806 0.6731 0.8205
No log 8.6875 278 0.6331 0.6791 0.6331 0.7957
No log 8.75 280 0.6302 0.6585 0.6302 0.7939
No log 8.8125 282 0.6182 0.6352 0.6182 0.7863
No log 8.875 284 0.6094 0.6535 0.6094 0.7806
No log 8.9375 286 0.6075 0.6349 0.6075 0.7794
No log 9.0 288 0.6633 0.5845 0.6633 0.8144
No log 9.0625 290 0.7740 0.5658 0.7740 0.8798
No log 9.125 292 0.7432 0.6085 0.7432 0.8621
No log 9.1875 294 0.6428 0.6366 0.6428 0.8018
No log 9.25 296 0.6029 0.6433 0.6029 0.7764
No log 9.3125 298 0.5966 0.6578 0.5966 0.7724
No log 9.375 300 0.5771 0.6940 0.5771 0.7597
No log 9.4375 302 0.5859 0.6882 0.5859 0.7654
No log 9.5 304 0.6112 0.6452 0.6112 0.7818
No log 9.5625 306 0.6319 0.5932 0.6319 0.7949
No log 9.625 308 0.6117 0.6107 0.6117 0.7821
No log 9.6875 310 0.5979 0.6882 0.5979 0.7732
No log 9.75 312 0.5988 0.6882 0.5988 0.7738
No log 9.8125 314 0.6151 0.6195 0.6151 0.7843
No log 9.875 316 0.6288 0.6097 0.6288 0.7929
No log 9.9375 318 0.6262 0.6325 0.6262 0.7913
No log 10.0 320 0.6123 0.6442 0.6123 0.7825
No log 10.0625 322 0.6204 0.5835 0.6204 0.7876
No log 10.125 324 0.6185 0.5835 0.6185 0.7864
No log 10.1875 326 0.6210 0.5647 0.6210 0.7880
No log 10.25 328 0.7161 0.5245 0.7161 0.8462
No log 10.3125 330 0.7648 0.5258 0.7648 0.8746
No log 10.375 332 0.6841 0.5348 0.6841 0.8271
No log 10.4375 334 0.6151 0.5805 0.6151 0.7843
No log 10.5 336 0.5910 0.6442 0.5910 0.7688
No log 10.5625 338 0.5972 0.6619 0.5972 0.7728
No log 10.625 340 0.6052 0.5773 0.6052 0.7780
No log 10.6875 342 0.7275 0.6324 0.7275 0.8529
No log 10.75 344 0.8946 0.5429 0.8946 0.9459
No log 10.8125 346 0.8201 0.6046 0.8201 0.9056
No log 10.875 348 0.6603 0.6925 0.6603 0.8126
No log 10.9375 350 0.6390 0.6508 0.6390 0.7994
No log 11.0 352 0.6972 0.6657 0.6972 0.8350
No log 11.0625 354 0.8593 0.5611 0.8593 0.9270
No log 11.125 356 0.8350 0.5740 0.8350 0.9138
No log 11.1875 358 0.6823 0.6174 0.6823 0.8260
No log 11.25 360 0.5700 0.6778 0.5700 0.7550
No log 11.3125 362 0.5603 0.6672 0.5603 0.7485
No log 11.375 364 0.5552 0.6528 0.5552 0.7451
No log 11.4375 366 0.5510 0.6658 0.5510 0.7423
No log 11.5 368 0.5521 0.6555 0.5521 0.7431
No log 11.5625 370 0.5575 0.6499 0.5575 0.7467
No log 11.625 372 0.5604 0.6452 0.5604 0.7486
No log 11.6875 374 0.5643 0.6903 0.5643 0.7512
No log 11.75 376 0.5640 0.6729 0.5640 0.7510
No log 11.8125 378 0.5578 0.6835 0.5578 0.7469
No log 11.875 380 0.5623 0.7198 0.5623 0.7498
No log 11.9375 382 0.5489 0.6938 0.5489 0.7409
No log 12.0 384 0.5304 0.7040 0.5304 0.7283
No log 12.0625 386 0.5273 0.7040 0.5273 0.7261
No log 12.125 388 0.5803 0.6640 0.5803 0.7618
No log 12.1875 390 0.5886 0.6542 0.5886 0.7672
No log 12.25 392 0.5460 0.6884 0.5460 0.7389
No log 12.3125 394 0.5300 0.7040 0.5300 0.7280
No log 12.375 396 0.5394 0.7115 0.5394 0.7345
No log 12.4375 398 0.5542 0.7115 0.5542 0.7445
No log 12.5 400 0.5647 0.7070 0.5647 0.7515
No log 12.5625 402 0.5627 0.7041 0.5627 0.7501
No log 12.625 404 0.5581 0.7041 0.5581 0.7471
No log 12.6875 406 0.5481 0.6882 0.5481 0.7404
No log 12.75 408 0.5595 0.6119 0.5595 0.7480
No log 12.8125 410 0.5594 0.5783 0.5594 0.7480
No log 12.875 412 0.5659 0.5694 0.5659 0.7523
No log 12.9375 414 0.5917 0.5686 0.5917 0.7692
No log 13.0 416 0.5947 0.5770 0.5947 0.7712
No log 13.0625 418 0.5683 0.6828 0.5683 0.7539
No log 13.125 420 0.5425 0.6653 0.5425 0.7365
No log 13.1875 422 0.5299 0.6407 0.5299 0.7280
No log 13.25 424 0.5319 0.6407 0.5319 0.7293
No log 13.3125 426 0.5484 0.6787 0.5484 0.7405
No log 13.375 428 0.5710 0.5951 0.5710 0.7557
No log 13.4375 430 0.5871 0.5759 0.5871 0.7662
No log 13.5 432 0.5892 0.5852 0.5892 0.7676
No log 13.5625 434 0.5847 0.6292 0.5847 0.7647
No log 13.625 436 0.5912 0.6555 0.5912 0.7689
No log 13.6875 438 0.5960 0.6772 0.5960 0.7720
No log 13.75 440 0.6119 0.6721 0.6119 0.7822
No log 13.8125 442 0.6137 0.6154 0.6137 0.7834
No log 13.875 444 0.5933 0.6042 0.5933 0.7702
No log 13.9375 446 0.5953 0.6370 0.5953 0.7716
No log 14.0 448 0.6217 0.6259 0.6217 0.7885
No log 14.0625 450 0.6119 0.6371 0.6119 0.7823
No log 14.125 452 0.5633 0.6740 0.5633 0.7505
No log 14.1875 454 0.5844 0.6977 0.5844 0.7644
No log 14.25 456 0.7233 0.6092 0.7233 0.8505
No log 14.3125 458 0.8142 0.5821 0.8142 0.9023
No log 14.375 460 0.8414 0.5821 0.8414 0.9173
No log 14.4375 462 0.6995 0.6524 0.6995 0.8364
No log 14.5 464 0.5645 0.6479 0.5645 0.7513
No log 14.5625 466 0.5420 0.6813 0.5420 0.7362
No log 14.625 468 0.6271 0.6756 0.6271 0.7919
No log 14.6875 470 0.6733 0.6655 0.6733 0.8205
No log 14.75 472 0.6353 0.7239 0.6353 0.7971
No log 14.8125 474 0.6403 0.6642 0.6403 0.8002
No log 14.875 476 0.6557 0.6476 0.6557 0.8097
No log 14.9375 478 0.6354 0.6661 0.6354 0.7971
No log 15.0 480 0.5916 0.5843 0.5916 0.7691
No log 15.0625 482 0.5926 0.5402 0.5926 0.7698
No log 15.125 484 0.6016 0.5402 0.6016 0.7757
No log 15.1875 486 0.6071 0.5388 0.6071 0.7792
No log 15.25 488 0.5951 0.6065 0.5951 0.7714
No log 15.3125 490 0.5861 0.6046 0.5861 0.7656
No log 15.375 492 0.5947 0.6280 0.5947 0.7712
No log 15.4375 494 0.5938 0.7135 0.5938 0.7706
No log 15.5 496 0.5842 0.7089 0.5842 0.7643
No log 15.5625 498 0.5829 0.6775 0.5829 0.7634
0.3226 15.625 500 0.5709 0.6459 0.5709 0.7556
0.3226 15.6875 502 0.5675 0.6620 0.5675 0.7533
0.3226 15.75 504 0.5520 0.6147 0.5520 0.7430
0.3226 15.8125 506 0.5392 0.6207 0.5392 0.7343
0.3226 15.875 508 0.5425 0.6345 0.5425 0.7365
0.3226 15.9375 510 0.5532 0.6370 0.5532 0.7438
0.3226 16.0 512 0.5610 0.6469 0.5610 0.7490
0.3226 16.0625 514 0.5581 0.6062 0.5581 0.7471
0.3226 16.125 516 0.5463 0.6370 0.5463 0.7391
0.3226 16.1875 518 0.5460 0.5638 0.5460 0.7389
0.3226 16.25 520 0.5916 0.6129 0.5916 0.7691
0.3226 16.3125 522 0.6134 0.6525 0.6134 0.7832
0.3226 16.375 524 0.5781 0.6332 0.5781 0.7603
0.3226 16.4375 526 0.5482 0.5961 0.5482 0.7404
0.3226 16.5 528 0.5444 0.6619 0.5444 0.7378
0.3226 16.5625 530 0.5549 0.6720 0.5549 0.7449
0.3226 16.625 532 0.5505 0.6720 0.5505 0.7419
0.3226 16.6875 534 0.5547 0.6426 0.5547 0.7448
0.3226 16.75 536 0.5962 0.6470 0.5962 0.7721
0.3226 16.8125 538 0.6357 0.6371 0.6357 0.7973
0.3226 16.875 540 0.6274 0.6371 0.6274 0.7921
0.3226 16.9375 542 0.5772 0.6701 0.5772 0.7597
0.3226 17.0 544 0.5255 0.7165 0.5255 0.7249
0.3226 17.0625 546 0.5249 0.6143 0.5249 0.7245
0.3226 17.125 548 0.5290 0.6025 0.5290 0.7273
0.3226 17.1875 550 0.5186 0.6065 0.5186 0.7202
0.3226 17.25 552 0.5438 0.5704 0.5438 0.7374
0.3226 17.3125 554 0.6306 0.5818 0.6306 0.7941

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
-
Safetensors
Model size
0.1B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for MayBashendy/ArabicNewSplits7_B_usingWellWrittenEssays_FineTuningAraBERT_run2_AugV5_k6_task5_organization

Finetuned
(4033)
this model