v0_mistral_lora_last_n
This model is a fine-tuned version of peiyi9979/math-shepherd-mistral-7b-prm on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3319
- Accuracy: 0.8850
- Precision: 0.9048
- Recall: 0.57
- F1: 0.6994
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
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- total_eval_batch_size: 32
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
---|---|---|---|---|---|---|---|
1.038 | 0.0054 | 5 | 0.6512 | 0.6291 | 0.3067 | 0.46 | 0.368 |
0.9593 | 0.0109 | 10 | 0.6474 | 0.6315 | 0.3087 | 0.46 | 0.3695 |
1.1635 | 0.0163 | 15 | 0.6440 | 0.6432 | 0.3219 | 0.47 | 0.3821 |
1.0183 | 0.0217 | 20 | 0.6377 | 0.6526 | 0.3310 | 0.47 | 0.3884 |
0.9214 | 0.0271 | 25 | 0.6317 | 0.6690 | 0.3435 | 0.45 | 0.3896 |
0.8285 | 0.0326 | 30 | 0.6184 | 0.6854 | 0.3396 | 0.36 | 0.3495 |
0.904 | 0.0380 | 35 | 0.6050 | 0.7113 | 0.3678 | 0.32 | 0.3422 |
0.7794 | 0.0434 | 40 | 0.5964 | 0.7277 | 0.3824 | 0.26 | 0.3095 |
0.7693 | 0.0488 | 45 | 0.5864 | 0.7394 | 0.4068 | 0.24 | 0.3019 |
0.738 | 0.0543 | 50 | 0.5789 | 0.7465 | 0.4231 | 0.22 | 0.2895 |
0.5718 | 0.0597 | 55 | 0.5729 | 0.7488 | 0.4103 | 0.16 | 0.2302 |
0.7026 | 0.0651 | 60 | 0.5632 | 0.7465 | 0.3824 | 0.13 | 0.1940 |
0.6761 | 0.0705 | 65 | 0.5578 | 0.7441 | 0.3548 | 0.11 | 0.1679 |
0.7159 | 0.0760 | 70 | 0.5407 | 0.7488 | 0.3793 | 0.11 | 0.1705 |
0.7457 | 0.0814 | 75 | 0.5320 | 0.7582 | 0.4545 | 0.15 | 0.2256 |
0.6426 | 0.0868 | 80 | 0.5254 | 0.7535 | 0.4286 | 0.15 | 0.2222 |
0.6202 | 0.0922 | 85 | 0.5227 | 0.7629 | 0.4815 | 0.13 | 0.2047 |
0.6266 | 0.0977 | 90 | 0.5178 | 0.7653 | 0.5 | 0.13 | 0.2063 |
0.6528 | 0.1031 | 95 | 0.5088 | 0.7817 | 0.8889 | 0.08 | 0.1468 |
0.5805 | 0.1085 | 100 | 0.5090 | 0.7746 | 1.0 | 0.04 | 0.0769 |
0.6823 | 0.1139 | 105 | 0.5058 | 0.7746 | 1.0 | 0.04 | 0.0769 |
0.5413 | 0.1194 | 110 | 0.5061 | 0.7770 | 0.8571 | 0.06 | 0.1121 |
0.5933 | 0.1248 | 115 | 0.5127 | 0.7723 | 0.6667 | 0.06 | 0.1101 |
0.4927 | 0.1302 | 120 | 0.5054 | 0.7746 | 0.8333 | 0.05 | 0.0943 |
0.5509 | 0.1356 | 125 | 0.5055 | 0.7723 | 1.0 | 0.03 | 0.0583 |
0.5958 | 0.1411 | 130 | 0.5016 | 0.7746 | 1.0 | 0.04 | 0.0769 |
0.6447 | 0.1465 | 135 | 0.5068 | 0.7723 | 0.7143 | 0.05 | 0.0935 |
0.561 | 0.1519 | 140 | 0.5127 | 0.7817 | 0.8889 | 0.08 | 0.1468 |
0.5959 | 0.1574 | 145 | 0.5026 | 0.7770 | 1.0 | 0.05 | 0.0952 |
0.6159 | 0.1628 | 150 | 0.5036 | 0.7746 | 0.8333 | 0.05 | 0.0943 |
0.5744 | 0.1682 | 155 | 0.4998 | 0.7746 | 0.8333 | 0.05 | 0.0943 |
0.6541 | 0.1736 | 160 | 0.4993 | 0.7770 | 0.8571 | 0.06 | 0.1121 |
0.6763 | 0.1791 | 165 | 0.4986 | 0.7817 | 0.8889 | 0.08 | 0.1468 |
0.6543 | 0.1845 | 170 | 0.4953 | 0.7817 | 0.8889 | 0.08 | 0.1468 |
0.5478 | 0.1899 | 175 | 0.4902 | 0.7840 | 0.9 | 0.09 | 0.1636 |
0.4365 | 0.1953 | 180 | 0.4891 | 0.7887 | 0.9167 | 0.11 | 0.1964 |
0.4885 | 0.2008 | 185 | 0.4829 | 0.7934 | 0.9286 | 0.13 | 0.2281 |
0.5827 | 0.2062 | 190 | 0.4835 | 0.7887 | 0.8125 | 0.13 | 0.2241 |
0.556 | 0.2116 | 195 | 0.4824 | 0.8005 | 0.8 | 0.2 | 0.32 |
0.499 | 0.2170 | 200 | 0.4755 | 0.7958 | 0.9333 | 0.14 | 0.2435 |
0.5283 | 0.2225 | 205 | 0.4751 | 0.7864 | 0.9091 | 0.1 | 0.1802 |
0.5419 | 0.2279 | 210 | 0.4674 | 0.8005 | 0.8571 | 0.18 | 0.2975 |
0.5653 | 0.2333 | 215 | 0.4716 | 0.8099 | 0.8065 | 0.25 | 0.3817 |
0.5264 | 0.2387 | 220 | 0.4746 | 0.8028 | 0.7857 | 0.22 | 0.3438 |
0.5869 | 0.2442 | 225 | 0.4668 | 0.7934 | 0.875 | 0.14 | 0.2414 |
0.6876 | 0.2496 | 230 | 0.4654 | 0.7911 | 0.8235 | 0.14 | 0.2393 |
0.5536 | 0.2550 | 235 | 0.4627 | 0.7981 | 0.85 | 0.17 | 0.2833 |
0.5298 | 0.2604 | 240 | 0.4613 | 0.8052 | 0.84 | 0.21 | 0.336 |
0.5933 | 0.2659 | 245 | 0.4610 | 0.8028 | 0.8333 | 0.2 | 0.3226 |
0.4468 | 0.2713 | 250 | 0.4613 | 0.8099 | 0.8519 | 0.23 | 0.3622 |
0.5832 | 0.2767 | 255 | 0.4573 | 0.8075 | 0.8462 | 0.22 | 0.3492 |
0.5867 | 0.2821 | 260 | 0.4527 | 0.8099 | 0.8519 | 0.23 | 0.3622 |
0.5597 | 0.2876 | 265 | 0.4496 | 0.8122 | 0.8571 | 0.24 | 0.375 |
0.5674 | 0.2930 | 270 | 0.4390 | 0.8052 | 0.8696 | 0.2 | 0.3252 |
0.4905 | 0.2984 | 275 | 0.4356 | 0.7981 | 0.85 | 0.17 | 0.2833 |
0.5892 | 0.3039 | 280 | 0.4336 | 0.8005 | 0.8571 | 0.18 | 0.2975 |
0.6111 | 0.3093 | 285 | 0.4320 | 0.8075 | 0.8462 | 0.22 | 0.3492 |
0.6202 | 0.3147 | 290 | 0.4303 | 0.8192 | 0.8485 | 0.28 | 0.4211 |
0.5541 | 0.3201 | 295 | 0.4305 | 0.8263 | 0.9062 | 0.29 | 0.4394 |
0.5864 | 0.3256 | 300 | 0.4263 | 0.8286 | 0.8649 | 0.32 | 0.4672 |
0.7254 | 0.3310 | 305 | 0.4277 | 0.8263 | 0.8095 | 0.34 | 0.4789 |
0.5439 | 0.3364 | 310 | 0.4279 | 0.8451 | 0.8036 | 0.45 | 0.5769 |
0.5388 | 0.3418 | 315 | 0.4156 | 0.8333 | 0.8718 | 0.34 | 0.4892 |
0.4984 | 0.3473 | 320 | 0.4128 | 0.8310 | 0.9118 | 0.31 | 0.4627 |
0.5593 | 0.3527 | 325 | 0.4099 | 0.8239 | 0.9032 | 0.28 | 0.4275 |
0.5564 | 0.3581 | 330 | 0.4053 | 0.8286 | 0.9091 | 0.3 | 0.4511 |
0.6122 | 0.3635 | 335 | 0.4005 | 0.8568 | 0.9333 | 0.42 | 0.5793 |
0.5366 | 0.3690 | 340 | 0.3929 | 0.8615 | 0.9020 | 0.46 | 0.6093 |
0.6113 | 0.3744 | 345 | 0.3915 | 0.8545 | 0.9130 | 0.42 | 0.5753 |
0.6386 | 0.3798 | 350 | 0.3866 | 0.8662 | 0.8909 | 0.49 | 0.6323 |
0.4795 | 0.3852 | 355 | 0.3879 | 0.8592 | 0.8125 | 0.52 | 0.6341 |
0.5393 | 0.3907 | 360 | 0.3800 | 0.8685 | 0.8929 | 0.5 | 0.6410 |
0.5117 | 0.3961 | 365 | 0.3788 | 0.8732 | 0.8966 | 0.52 | 0.6582 |
0.5432 | 0.4015 | 370 | 0.3788 | 0.8756 | 0.9123 | 0.52 | 0.6624 |
0.5301 | 0.4069 | 375 | 0.3817 | 0.8826 | 0.9310 | 0.54 | 0.6835 |
0.5486 | 0.4124 | 380 | 0.3813 | 0.8732 | 0.9259 | 0.5 | 0.6494 |
0.5887 | 0.4178 | 385 | 0.3821 | 0.8756 | 0.9273 | 0.51 | 0.6581 |
0.583 | 0.4232 | 390 | 0.3803 | 0.8662 | 0.9388 | 0.46 | 0.6174 |
0.5682 | 0.4286 | 395 | 0.3792 | 0.8685 | 0.9074 | 0.49 | 0.6364 |
0.5331 | 0.4341 | 400 | 0.3814 | 0.8732 | 0.8382 | 0.57 | 0.6786 |
0.5498 | 0.4395 | 405 | 0.3799 | 0.8685 | 0.8056 | 0.58 | 0.6744 |
0.578 | 0.4449 | 410 | 0.3704 | 0.8850 | 0.8923 | 0.58 | 0.7030 |
0.5605 | 0.4504 | 415 | 0.3672 | 0.8779 | 0.9138 | 0.53 | 0.6709 |
0.5768 | 0.4558 | 420 | 0.3656 | 0.8826 | 0.9310 | 0.54 | 0.6835 |
0.5379 | 0.4612 | 425 | 0.3685 | 0.8732 | 0.8485 | 0.56 | 0.6747 |
0.4722 | 0.4666 | 430 | 0.3728 | 0.8709 | 0.8261 | 0.57 | 0.6746 |
0.6306 | 0.4721 | 435 | 0.3643 | 0.8803 | 0.9298 | 0.53 | 0.6752 |
0.5539 | 0.4775 | 440 | 0.3684 | 0.8662 | 0.9216 | 0.47 | 0.6225 |
0.4614 | 0.4829 | 445 | 0.3703 | 0.8662 | 0.9216 | 0.47 | 0.6225 |
0.5376 | 0.4883 | 450 | 0.3710 | 0.8685 | 0.9231 | 0.48 | 0.6316 |
0.5177 | 0.4938 | 455 | 0.3717 | 0.8685 | 0.9231 | 0.48 | 0.6316 |
0.4773 | 0.4992 | 460 | 0.3704 | 0.8732 | 0.8710 | 0.54 | 0.6667 |
0.6133 | 0.5046 | 465 | 0.3715 | 0.8662 | 0.8028 | 0.57 | 0.6667 |
0.4302 | 0.5100 | 470 | 0.3586 | 0.8732 | 0.8710 | 0.54 | 0.6667 |
0.5382 | 0.5155 | 475 | 0.3582 | 0.8709 | 0.9245 | 0.49 | 0.6405 |
0.5394 | 0.5209 | 480 | 0.3574 | 0.8709 | 0.9412 | 0.48 | 0.6358 |
0.4772 | 0.5263 | 485 | 0.3469 | 0.8709 | 0.8571 | 0.54 | 0.6626 |
0.4767 | 0.5317 | 490 | 0.3490 | 0.8779 | 0.8429 | 0.59 | 0.6941 |
0.7296 | 0.5372 | 495 | 0.3502 | 0.8709 | 0.8358 | 0.56 | 0.6707 |
0.5884 | 0.5426 | 500 | 0.3540 | 0.8779 | 0.8529 | 0.58 | 0.6905 |
0.626 | 0.5480 | 505 | 0.3588 | 0.8803 | 0.8451 | 0.6 | 0.7018 |
0.4887 | 0.5534 | 510 | 0.3558 | 0.8803 | 0.8657 | 0.58 | 0.6946 |
0.647 | 0.5589 | 515 | 0.3495 | 0.8732 | 0.9107 | 0.51 | 0.6538 |
0.4802 | 0.5643 | 520 | 0.3582 | 0.8685 | 0.94 | 0.47 | 0.6267 |
0.6024 | 0.5697 | 525 | 0.3502 | 0.8662 | 0.9057 | 0.48 | 0.6275 |
0.5087 | 0.5751 | 530 | 0.3441 | 0.8803 | 0.8889 | 0.56 | 0.6871 |
0.5407 | 0.5806 | 535 | 0.3514 | 0.8873 | 0.8714 | 0.61 | 0.7176 |
0.5428 | 0.5860 | 540 | 0.3484 | 0.8873 | 0.8714 | 0.61 | 0.7176 |
0.5368 | 0.5914 | 545 | 0.3493 | 0.8897 | 0.8533 | 0.64 | 0.7314 |
0.5315 | 0.5969 | 550 | 0.3424 | 0.8850 | 0.8923 | 0.58 | 0.7030 |
0.4935 | 0.6023 | 555 | 0.3472 | 0.8779 | 0.9 | 0.54 | 0.675 |
0.5853 | 0.6077 | 560 | 0.3482 | 0.8779 | 0.9138 | 0.53 | 0.6709 |
0.562 | 0.6131 | 565 | 0.3461 | 0.8779 | 0.8871 | 0.55 | 0.6790 |
0.6008 | 0.6186 | 570 | 0.3493 | 0.8826 | 0.8571 | 0.6 | 0.7059 |
0.4707 | 0.6240 | 575 | 0.3449 | 0.8873 | 0.8714 | 0.61 | 0.7176 |
0.5917 | 0.6294 | 580 | 0.3403 | 0.8756 | 0.8730 | 0.55 | 0.6748 |
0.5038 | 0.6348 | 585 | 0.3427 | 0.8709 | 0.8814 | 0.52 | 0.6541 |
0.4744 | 0.6403 | 590 | 0.3440 | 0.8685 | 0.9231 | 0.48 | 0.6316 |
0.5818 | 0.6457 | 595 | 0.3419 | 0.8685 | 0.9074 | 0.49 | 0.6364 |
0.5183 | 0.6511 | 600 | 0.3377 | 0.8709 | 0.8947 | 0.51 | 0.6497 |
0.6047 | 0.6565 | 605 | 0.3359 | 0.8732 | 0.8833 | 0.53 | 0.6625 |
0.4523 | 0.6620 | 610 | 0.3370 | 0.8897 | 0.8841 | 0.61 | 0.7219 |
0.6272 | 0.6674 | 615 | 0.3412 | 0.8897 | 0.8732 | 0.62 | 0.7251 |
0.5166 | 0.6728 | 620 | 0.3408 | 0.8873 | 0.8714 | 0.61 | 0.7176 |
0.504 | 0.6782 | 625 | 0.3427 | 0.8779 | 0.8871 | 0.55 | 0.6790 |
0.5734 | 0.6837 | 630 | 0.3422 | 0.8685 | 0.8793 | 0.51 | 0.6456 |
0.4946 | 0.6891 | 635 | 0.3410 | 0.8732 | 0.8966 | 0.52 | 0.6582 |
0.617 | 0.6945 | 640 | 0.3391 | 0.8803 | 0.8769 | 0.57 | 0.6909 |
0.6055 | 0.6999 | 645 | 0.3425 | 0.8826 | 0.8472 | 0.61 | 0.7093 |
0.5427 | 0.7054 | 650 | 0.3412 | 0.8873 | 0.8611 | 0.62 | 0.7209 |
0.4839 | 0.7108 | 655 | 0.3384 | 0.8803 | 0.8657 | 0.58 | 0.6946 |
0.5573 | 0.7162 | 660 | 0.3379 | 0.8779 | 0.9138 | 0.53 | 0.6709 |
0.4199 | 0.7216 | 665 | 0.3351 | 0.8826 | 0.9167 | 0.55 | 0.6875 |
0.5563 | 0.7271 | 670 | 0.3351 | 0.8803 | 0.9153 | 0.54 | 0.6792 |
0.5772 | 0.7325 | 675 | 0.3363 | 0.8803 | 0.9298 | 0.53 | 0.6752 |
0.5363 | 0.7379 | 680 | 0.3369 | 0.8803 | 0.9153 | 0.54 | 0.6792 |
0.5554 | 0.7434 | 685 | 0.3350 | 0.8826 | 0.9167 | 0.55 | 0.6875 |
0.5154 | 0.7488 | 690 | 0.3338 | 0.8873 | 0.9194 | 0.57 | 0.7037 |
0.4925 | 0.7542 | 695 | 0.3340 | 0.8850 | 0.9180 | 0.56 | 0.6957 |
0.5371 | 0.7596 | 700 | 0.3327 | 0.8944 | 0.9231 | 0.6 | 0.7273 |
0.5402 | 0.7651 | 705 | 0.3348 | 0.8873 | 0.8714 | 0.61 | 0.7176 |
0.5634 | 0.7705 | 710 | 0.3347 | 0.8873 | 0.8514 | 0.63 | 0.7241 |
0.5088 | 0.7759 | 715 | 0.3339 | 0.8897 | 0.8732 | 0.62 | 0.7251 |
0.4872 | 0.7813 | 720 | 0.3316 | 0.8897 | 0.8955 | 0.6 | 0.7186 |
0.5487 | 0.7868 | 725 | 0.3297 | 0.8873 | 0.9062 | 0.58 | 0.7073 |
0.4821 | 0.7922 | 730 | 0.3289 | 0.8850 | 0.9048 | 0.57 | 0.6994 |
0.6054 | 0.7976 | 735 | 0.3299 | 0.8826 | 0.8788 | 0.58 | 0.6988 |
0.4619 | 0.8030 | 740 | 0.3298 | 0.8873 | 0.9062 | 0.58 | 0.7073 |
0.6107 | 0.8085 | 745 | 0.3309 | 0.8826 | 0.8788 | 0.58 | 0.6988 |
0.4162 | 0.8139 | 750 | 0.3305 | 0.8850 | 0.9048 | 0.57 | 0.6994 |
0.4735 | 0.8193 | 755 | 0.3307 | 0.8897 | 0.9206 | 0.58 | 0.7117 |
0.5067 | 0.8247 | 760 | 0.3308 | 0.8826 | 0.8906 | 0.57 | 0.6951 |
0.6646 | 0.8302 | 765 | 0.3304 | 0.8850 | 0.9048 | 0.57 | 0.6994 |
0.5315 | 0.8356 | 770 | 0.3312 | 0.8826 | 0.8906 | 0.57 | 0.6951 |
0.4793 | 0.8410 | 775 | 0.3303 | 0.8850 | 0.9048 | 0.57 | 0.6994 |
0.6197 | 0.8464 | 780 | 0.3310 | 0.8873 | 0.9194 | 0.57 | 0.7037 |
0.5175 | 0.8519 | 785 | 0.3300 | 0.8873 | 0.9194 | 0.57 | 0.7037 |
0.456 | 0.8573 | 790 | 0.3301 | 0.8850 | 0.9180 | 0.56 | 0.6957 |
0.5674 | 0.8627 | 795 | 0.3298 | 0.8850 | 0.9180 | 0.56 | 0.6957 |
0.4572 | 0.8681 | 800 | 0.3297 | 0.8873 | 0.9194 | 0.57 | 0.7037 |
0.5919 | 0.8736 | 805 | 0.3305 | 0.8850 | 0.9180 | 0.56 | 0.6957 |
0.6688 | 0.8790 | 810 | 0.3291 | 0.8873 | 0.9194 | 0.57 | 0.7037 |
0.6046 | 0.8844 | 815 | 0.3296 | 0.8873 | 0.9194 | 0.57 | 0.7037 |
0.5199 | 0.8899 | 820 | 0.3308 | 0.8873 | 0.9194 | 0.57 | 0.7037 |
0.5188 | 0.8953 | 825 | 0.3310 | 0.8826 | 0.9032 | 0.56 | 0.6914 |
0.6291 | 0.9007 | 830 | 0.3302 | 0.8873 | 0.9194 | 0.57 | 0.7037 |
0.5297 | 0.9061 | 835 | 0.3301 | 0.8873 | 0.9062 | 0.58 | 0.7073 |
0.4918 | 0.9116 | 840 | 0.3312 | 0.8850 | 0.9048 | 0.57 | 0.6994 |
0.6324 | 0.9170 | 845 | 0.3305 | 0.8826 | 0.8906 | 0.57 | 0.6951 |
0.5935 | 0.9224 | 850 | 0.3318 | 0.8873 | 0.9062 | 0.58 | 0.7073 |
0.5409 | 0.9278 | 855 | 0.3316 | 0.8873 | 0.9062 | 0.58 | 0.7073 |
0.5559 | 0.9333 | 860 | 0.3320 | 0.8873 | 0.9062 | 0.58 | 0.7073 |
0.5595 | 0.9387 | 865 | 0.3316 | 0.8873 | 0.9062 | 0.58 | 0.7073 |
0.5309 | 0.9441 | 870 | 0.3318 | 0.8897 | 0.9077 | 0.59 | 0.7152 |
0.5631 | 0.9495 | 875 | 0.3329 | 0.8897 | 0.9206 | 0.58 | 0.7117 |
0.494 | 0.9550 | 880 | 0.3326 | 0.8873 | 0.9062 | 0.58 | 0.7073 |
0.5215 | 0.9604 | 885 | 0.3322 | 0.8873 | 0.8939 | 0.59 | 0.7108 |
0.5443 | 0.9658 | 890 | 0.3313 | 0.8920 | 0.9219 | 0.59 | 0.7195 |
0.508 | 0.9712 | 895 | 0.3323 | 0.8873 | 0.9062 | 0.58 | 0.7073 |
0.4527 | 0.9767 | 900 | 0.3311 | 0.8850 | 0.8923 | 0.58 | 0.7030 |
0.575 | 0.9821 | 905 | 0.3318 | 0.8873 | 0.9062 | 0.58 | 0.7073 |
0.5813 | 0.9875 | 910 | 0.3336 | 0.8826 | 0.8906 | 0.57 | 0.6951 |
0.4968 | 0.9929 | 915 | 0.3311 | 0.8873 | 0.9062 | 0.58 | 0.7073 |
0.5967 | 0.9984 | 920 | 0.3319 | 0.8850 | 0.9048 | 0.57 | 0.6994 |
Framework versions
- PEFT 0.12.0
- Transformers 4.46.0
- Pytorch 2.4.0+cu118
- Datasets 3.0.0
- Tokenizers 0.20.1
- Downloads last month
- 2
Model tree for mtzig/v0_mistral_lora_last_n
Base model
peiyi9979/math-shepherd-mistral-7b-prm