ArabicNewSplits8_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k6_task1_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: 0.8484
  • Qwk: 0.6230
  • Mse: 0.8484
  • Rmse: 0.9211

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 5.5693 -0.0574 5.5693 2.3599
No log 0.125 4 3.1522 0.0589 3.1522 1.7754
No log 0.1875 6 2.0267 0.0392 2.0267 1.4236
No log 0.25 8 1.7690 0.0731 1.7690 1.3301
No log 0.3125 10 1.9208 -0.0056 1.9208 1.3859
No log 0.375 12 1.4947 0.0212 1.4947 1.2226
No log 0.4375 14 1.2500 0.1336 1.2500 1.1181
No log 0.5 16 1.2345 0.1534 1.2345 1.1111
No log 0.5625 18 1.2575 0.1190 1.2575 1.1214
No log 0.625 20 1.2375 0.1679 1.2375 1.1124
No log 0.6875 22 1.1810 0.1322 1.1810 1.0867
No log 0.75 24 1.1789 0.1410 1.1789 1.0858
No log 0.8125 26 1.2999 0.1622 1.2999 1.1401
No log 0.875 28 1.3099 0.1690 1.3099 1.1445
No log 0.9375 30 1.3816 0.0593 1.3816 1.1754
No log 1.0 32 1.3504 0.0994 1.3504 1.1621
No log 1.0625 34 1.3179 0.1525 1.3179 1.1480
No log 1.125 36 1.2825 0.1692 1.2825 1.1325
No log 1.1875 38 1.2055 0.3212 1.2055 1.0979
No log 1.25 40 1.1277 0.2866 1.1277 1.0619
No log 1.3125 42 1.0920 0.2856 1.0920 1.0450
No log 1.375 44 1.1180 0.3347 1.1180 1.0573
No log 1.4375 46 1.1308 0.3266 1.1308 1.0634
No log 1.5 48 1.1557 0.2784 1.1557 1.0750
No log 1.5625 50 1.2161 0.3312 1.2161 1.1028
No log 1.625 52 1.2636 0.2848 1.2636 1.1241
No log 1.6875 54 1.1928 0.3670 1.1928 1.0921
No log 1.75 56 1.1257 0.3615 1.1257 1.0610
No log 1.8125 58 1.1075 0.3355 1.1075 1.0524
No log 1.875 60 1.0643 0.3132 1.0643 1.0317
No log 1.9375 62 1.0532 0.2939 1.0532 1.0262
No log 2.0 64 1.0834 0.2606 1.0834 1.0409
No log 2.0625 66 1.0776 0.3334 1.0776 1.0381
No log 2.125 68 1.1740 0.3007 1.1740 1.0835
No log 2.1875 70 1.5527 0.1613 1.5527 1.2461
No log 2.25 72 1.7535 0.1329 1.7535 1.3242
No log 2.3125 74 1.5904 0.1703 1.5904 1.2611
No log 2.375 76 1.6078 0.1703 1.6078 1.2680
No log 2.4375 78 1.2504 0.3112 1.2504 1.1182
No log 2.5 80 1.0210 0.3108 1.0210 1.0105
No log 2.5625 82 1.0137 0.2991 1.0137 1.0068
No log 2.625 84 1.0546 0.2079 1.0546 1.0269
No log 2.6875 86 1.1540 0.2883 1.1540 1.0742
No log 2.75 88 1.1893 0.2953 1.1893 1.0905
No log 2.8125 90 1.1320 0.3011 1.1320 1.0639
No log 2.875 92 1.0345 0.3753 1.0345 1.0171
No log 2.9375 94 1.0399 0.4591 1.0399 1.0197
No log 3.0 96 1.0364 0.4937 1.0364 1.0180
No log 3.0625 98 0.9659 0.5454 0.9659 0.9828
No log 3.125 100 0.9923 0.5140 0.9923 0.9961
No log 3.1875 102 1.0857 0.3863 1.0857 1.0420
No log 3.25 104 1.1231 0.3539 1.1231 1.0598
No log 3.3125 106 1.1679 0.2094 1.1679 1.0807
No log 3.375 108 1.1276 0.2970 1.1276 1.0619
No log 3.4375 110 1.0032 0.3654 1.0032 1.0016
No log 3.5 112 0.8871 0.5386 0.8871 0.9419
No log 3.5625 114 0.8270 0.6296 0.8270 0.9094
No log 3.625 116 0.8159 0.5504 0.8159 0.9033
No log 3.6875 118 0.8776 0.5664 0.8776 0.9368
No log 3.75 120 1.1822 0.4443 1.1822 1.0873
No log 3.8125 122 1.2879 0.4171 1.2879 1.1349
No log 3.875 124 1.1442 0.4550 1.1442 1.0697
No log 3.9375 126 1.1873 0.4351 1.1873 1.0896
No log 4.0 128 1.4428 0.4152 1.4428 1.2012
No log 4.0625 130 1.7703 0.3909 1.7703 1.3305
No log 4.125 132 1.6316 0.4182 1.6316 1.2773
No log 4.1875 134 1.5729 0.4618 1.5729 1.2542
No log 4.25 136 1.4272 0.4578 1.4272 1.1947
No log 4.3125 138 1.4307 0.4538 1.4307 1.1961
No log 4.375 140 1.4773 0.4442 1.4773 1.2154
No log 4.4375 142 1.4803 0.4281 1.4803 1.2167
No log 4.5 144 1.5892 0.4045 1.5892 1.2606
No log 4.5625 146 1.5541 0.4045 1.5541 1.2467
No log 4.625 148 1.3206 0.4365 1.3206 1.1492
No log 4.6875 150 1.1154 0.5102 1.1154 1.0561
No log 4.75 152 1.0172 0.5800 1.0172 1.0086
No log 4.8125 154 0.9380 0.5974 0.9380 0.9685
No log 4.875 156 0.9844 0.5752 0.9844 0.9922
No log 4.9375 158 1.0386 0.5508 1.0386 1.0191
No log 5.0 160 1.0758 0.5442 1.0758 1.0372
No log 5.0625 162 0.9445 0.6180 0.9445 0.9719
No log 5.125 164 0.8276 0.6509 0.8276 0.9097
No log 5.1875 166 0.8684 0.6514 0.8684 0.9319
No log 5.25 168 0.9756 0.6003 0.9756 0.9877
No log 5.3125 170 0.9514 0.5874 0.9514 0.9754
No log 5.375 172 0.8481 0.6194 0.8481 0.9209
No log 5.4375 174 0.7930 0.6716 0.7930 0.8905
No log 5.5 176 0.8014 0.6824 0.8014 0.8952
No log 5.5625 178 0.8812 0.6857 0.8812 0.9387
No log 5.625 180 0.8198 0.6997 0.8198 0.9054
No log 5.6875 182 0.9362 0.6879 0.9362 0.9676
No log 5.75 184 1.1096 0.5898 1.1096 1.0534
No log 5.8125 186 1.1072 0.5567 1.1072 1.0522
No log 5.875 188 0.9187 0.6425 0.9187 0.9585
No log 5.9375 190 1.0096 0.5926 1.0096 1.0048
No log 6.0 192 1.1972 0.5828 1.1972 1.0942
No log 6.0625 194 1.3883 0.5241 1.3883 1.1783
No log 6.125 196 1.1252 0.5377 1.1252 1.0607
No log 6.1875 198 0.8426 0.6446 0.8426 0.9179
No log 6.25 200 0.7296 0.6777 0.7296 0.8542
No log 6.3125 202 0.7404 0.6717 0.7404 0.8605
No log 6.375 204 0.8368 0.6036 0.8368 0.9148
No log 6.4375 206 0.9086 0.5953 0.9086 0.9532
No log 6.5 208 0.8781 0.6201 0.8781 0.9371
No log 6.5625 210 0.7986 0.6568 0.7986 0.8936
No log 6.625 212 0.8477 0.6640 0.8477 0.9207
No log 6.6875 214 1.0639 0.5782 1.0639 1.0315
No log 6.75 216 1.1497 0.5862 1.1497 1.0723
No log 6.8125 218 1.0035 0.5881 1.0035 1.0018
No log 6.875 220 0.8190 0.6862 0.8190 0.9050
No log 6.9375 222 0.7579 0.6882 0.7579 0.8706
No log 7.0 224 0.8329 0.6239 0.8329 0.9126
No log 7.0625 226 1.0100 0.5756 1.0100 1.0050
No log 7.125 228 1.1559 0.4968 1.1559 1.0751
No log 7.1875 230 1.1285 0.5133 1.1285 1.0623
No log 7.25 232 1.1015 0.5274 1.1015 1.0495
No log 7.3125 234 1.2965 0.5156 1.2965 1.1386
No log 7.375 236 1.4001 0.5489 1.4001 1.1832
No log 7.4375 238 1.1557 0.5903 1.1557 1.0750
No log 7.5 240 0.8524 0.6703 0.8524 0.9233
No log 7.5625 242 0.7009 0.7208 0.7009 0.8372
No log 7.625 244 0.7428 0.6736 0.7428 0.8619
No log 7.6875 246 0.9742 0.6136 0.9742 0.9870
No log 7.75 248 1.1487 0.5501 1.1487 1.0718
No log 7.8125 250 1.1754 0.5444 1.1754 1.0842
No log 7.875 252 0.9976 0.5475 0.9976 0.9988
No log 7.9375 254 0.7710 0.6209 0.7710 0.8781
No log 8.0 256 0.7458 0.6200 0.7458 0.8636
No log 8.0625 258 0.7486 0.6418 0.7486 0.8652
No log 8.125 260 0.8261 0.6054 0.8261 0.9089
No log 8.1875 262 0.9973 0.5426 0.9973 0.9986
No log 8.25 264 1.0382 0.5407 1.0382 1.0189
No log 8.3125 266 0.9152 0.5754 0.9152 0.9567
No log 8.375 268 0.7955 0.6496 0.7955 0.8919
No log 8.4375 270 0.7569 0.6463 0.7569 0.8700
No log 8.5 272 0.8259 0.6088 0.8259 0.9088
No log 8.5625 274 0.9145 0.5756 0.9145 0.9563
No log 8.625 276 0.9414 0.5743 0.9414 0.9702
No log 8.6875 278 0.9471 0.5694 0.9471 0.9732
No log 8.75 280 0.8357 0.5756 0.8357 0.9142
No log 8.8125 282 0.7813 0.5756 0.7813 0.8839
No log 8.875 284 0.8513 0.5941 0.8513 0.9227
No log 8.9375 286 0.9928 0.5489 0.9928 0.9964
No log 9.0 288 0.9476 0.5491 0.9476 0.9734
No log 9.0625 290 0.8095 0.6181 0.8095 0.8997
No log 9.125 292 0.6982 0.6731 0.6982 0.8356
No log 9.1875 294 0.6730 0.7082 0.6730 0.8204
No log 9.25 296 0.7394 0.6991 0.7394 0.8599
No log 9.3125 298 0.9712 0.6167 0.9712 0.9855
No log 9.375 300 1.0892 0.5741 1.0892 1.0436
No log 9.4375 302 0.9766 0.5629 0.9766 0.9882
No log 9.5 304 0.7748 0.6724 0.7748 0.8802
No log 9.5625 306 0.7248 0.6910 0.7248 0.8514
No log 9.625 308 0.8067 0.6346 0.8067 0.8982
No log 9.6875 310 1.0620 0.5655 1.0620 1.0305
No log 9.75 312 1.1711 0.5514 1.1711 1.0822
No log 9.8125 314 1.0107 0.5618 1.0107 1.0053
No log 9.875 316 0.8040 0.6543 0.8040 0.8967
No log 9.9375 318 0.7435 0.6758 0.7435 0.8623
No log 10.0 320 0.7241 0.6353 0.7241 0.8509
No log 10.0625 322 0.7382 0.5563 0.7382 0.8592
No log 10.125 324 0.8029 0.6190 0.8029 0.8960
No log 10.1875 326 0.9877 0.5462 0.9877 0.9938
No log 10.25 328 1.1184 0.5172 1.1184 1.0575
No log 10.3125 330 1.0162 0.5453 1.0162 1.0080
No log 10.375 332 0.8084 0.6440 0.8084 0.8991
No log 10.4375 334 0.7450 0.6706 0.7450 0.8631
No log 10.5 336 0.8021 0.6590 0.8021 0.8956
No log 10.5625 338 0.9470 0.5956 0.9470 0.9731
No log 10.625 340 1.1524 0.5416 1.1524 1.0735
No log 10.6875 342 1.1894 0.5535 1.1894 1.0906
No log 10.75 344 1.0526 0.5398 1.0526 1.0260
No log 10.8125 346 0.8922 0.6050 0.8922 0.9445
No log 10.875 348 0.8840 0.6066 0.8840 0.9402
No log 10.9375 350 0.9697 0.5807 0.9697 0.9847
No log 11.0 352 0.9519 0.5845 0.9519 0.9757
No log 11.0625 354 0.9433 0.5845 0.9433 0.9712
No log 11.125 356 0.9010 0.5957 0.9010 0.9492
No log 11.1875 358 0.7918 0.6178 0.7918 0.8898
No log 11.25 360 0.8116 0.6077 0.8116 0.9009
No log 11.3125 362 0.8730 0.5783 0.8730 0.9343
No log 11.375 364 0.8219 0.5770 0.8219 0.9066
No log 11.4375 366 0.7988 0.5837 0.7988 0.8937
No log 11.5 368 0.8381 0.6039 0.8381 0.9155
No log 11.5625 370 0.8847 0.5869 0.8847 0.9406
No log 11.625 372 0.8397 0.6032 0.8397 0.9164
No log 11.6875 374 0.8410 0.6165 0.8410 0.9171
No log 11.75 376 0.9570 0.5814 0.9570 0.9783
No log 11.8125 378 1.1253 0.5225 1.1253 1.0608
No log 11.875 380 1.1483 0.5307 1.1483 1.0716
No log 11.9375 382 1.0764 0.5290 1.0764 1.0375
No log 12.0 384 1.0271 0.5307 1.0271 1.0134
No log 12.0625 386 1.0046 0.5568 1.0046 1.0023
No log 12.125 388 0.8325 0.6619 0.8325 0.9124
No log 12.1875 390 0.7901 0.6808 0.7901 0.8889
No log 12.25 392 0.8089 0.6667 0.8089 0.8994
No log 12.3125 394 0.9111 0.6341 0.9111 0.9545
No log 12.375 396 0.9388 0.6282 0.9388 0.9689
No log 12.4375 398 0.9848 0.6394 0.9848 0.9923
No log 12.5 400 1.0328 0.6340 1.0328 1.0163
No log 12.5625 402 0.9876 0.6067 0.9876 0.9938
No log 12.625 404 0.9530 0.5823 0.9530 0.9762
No log 12.6875 406 0.9004 0.6094 0.9004 0.9489
No log 12.75 408 0.9656 0.5512 0.9656 0.9826
No log 12.8125 410 1.0588 0.5388 1.0588 1.0290
No log 12.875 412 1.0902 0.5248 1.0902 1.0441
No log 12.9375 414 0.9417 0.5522 0.9417 0.9704
No log 13.0 416 0.9329 0.5657 0.9329 0.9658
No log 13.0625 418 1.1008 0.5304 1.1008 1.0492
No log 13.125 420 1.1286 0.5303 1.1286 1.0624
No log 13.1875 422 1.0153 0.6065 1.0153 1.0076
No log 13.25 424 1.0073 0.6065 1.0073 1.0036
No log 13.3125 426 0.9017 0.6207 0.9017 0.9496
No log 13.375 428 0.9072 0.6009 0.9072 0.9525
No log 13.4375 430 0.8011 0.6419 0.8011 0.8950
No log 13.5 432 0.8016 0.6367 0.8016 0.8953
No log 13.5625 434 0.9447 0.5981 0.9447 0.9720
No log 13.625 436 0.9710 0.6084 0.9710 0.9854
No log 13.6875 438 0.9619 0.6084 0.9619 0.9808
No log 13.75 440 0.8377 0.6279 0.8377 0.9153
No log 13.8125 442 0.7907 0.6418 0.7907 0.8892
No log 13.875 444 0.9034 0.6067 0.9034 0.9505
No log 13.9375 446 1.1370 0.5591 1.1370 1.0663
No log 14.0 448 1.2169 0.5401 1.2169 1.1031
No log 14.0625 450 1.0105 0.5649 1.0105 1.0052
No log 14.125 452 0.7444 0.6586 0.7444 0.8628
No log 14.1875 454 0.6487 0.6611 0.6487 0.8054
No log 14.25 456 0.6589 0.6534 0.6589 0.8117
No log 14.3125 458 0.7250 0.6612 0.7250 0.8514
No log 14.375 460 0.7909 0.6351 0.7909 0.8893
No log 14.4375 462 0.9304 0.5693 0.9304 0.9646
No log 14.5 464 0.9139 0.5693 0.9139 0.9560
No log 14.5625 466 0.8165 0.6552 0.8165 0.9036
No log 14.625 468 0.8697 0.6516 0.8697 0.9326
No log 14.6875 470 1.0260 0.5421 1.0260 1.0129
No log 14.75 472 1.3079 0.5210 1.3079 1.1437
No log 14.8125 474 1.3855 0.4624 1.3855 1.1771
No log 14.875 476 1.2268 0.4770 1.2268 1.1076
No log 14.9375 478 0.9937 0.5630 0.9937 0.9968
No log 15.0 480 0.8324 0.6083 0.8324 0.9124
No log 15.0625 482 0.8056 0.6367 0.8056 0.8976
No log 15.125 484 0.9219 0.5795 0.9219 0.9602
No log 15.1875 486 1.1031 0.5174 1.1031 1.0503
No log 15.25 488 1.0989 0.5331 1.0989 1.0483
No log 15.3125 490 0.9529 0.5841 0.9529 0.9761
No log 15.375 492 0.8717 0.6023 0.8717 0.9337
No log 15.4375 494 0.8342 0.6466 0.8342 0.9133
No log 15.5 496 0.8485 0.6351 0.8485 0.9212
No log 15.5625 498 0.8589 0.6376 0.8589 0.9268
0.4639 15.625 500 0.8230 0.6676 0.8230 0.9072
0.4639 15.6875 502 0.7317 0.6910 0.7317 0.8554
0.4639 15.75 504 0.6553 0.7146 0.6553 0.8095
0.4639 15.8125 506 0.6849 0.6998 0.6849 0.8276
0.4639 15.875 508 0.7312 0.6986 0.7312 0.8551
0.4639 15.9375 510 0.7915 0.7015 0.7915 0.8896
0.4639 16.0 512 0.8870 0.6590 0.8870 0.9418
0.4639 16.0625 514 0.9779 0.6403 0.9779 0.9889
0.4639 16.125 516 0.8985 0.6017 0.8985 0.9479
0.4639 16.1875 518 0.8250 0.6539 0.8250 0.9083
0.4639 16.25 520 0.8484 0.6230 0.8484 0.9211

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
7
Safetensors
Model size
135M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and the model is not deployed on the HF Inference API.

Model tree for MayBashendy/ArabicNewSplits8_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k6_task1_organization

Finetuned
(4222)
this model