<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
roberta-base-mnli-combined-multivalue
This model is a fine-tuned version of roberta-base on the GLUE MNLI dataset. It achieves the following results on the evaluation set:
- Loss: 0.3918
- Accuracy: 0.8569
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: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 5.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
1.1005 | 0.01 | 500 | 1.0983 | 0.3366 |
1.0983 | 0.02 | 1000 | 1.0925 | 0.3553 |
1.0308 | 0.03 | 1500 | 0.9173 | 0.5742 |
0.8148 | 0.04 | 2000 | 0.6755 | 0.7252 |
0.6979 | 0.05 | 2500 | 0.6042 | 0.7605 |
0.6274 | 0.06 | 3000 | 0.5757 | 0.7740 |
0.6115 | 0.07 | 3500 | 0.5600 | 0.7833 |
0.5549 | 0.08 | 4000 | 0.5196 | 0.8014 |
0.5681 | 0.09 | 4500 | 0.4972 | 0.8069 |
0.553 | 0.1 | 5000 | 0.4781 | 0.8109 |
0.5422 | 0.11 | 5500 | 0.5118 | 0.8064 |
0.5385 | 0.12 | 6000 | 0.4821 | 0.8136 |
0.5271 | 0.13 | 6500 | 0.4645 | 0.8179 |
0.5022 | 0.14 | 7000 | 0.4794 | 0.8182 |
0.4973 | 0.15 | 7500 | 0.5009 | 0.8149 |
0.503 | 0.16 | 8000 | 0.4590 | 0.8220 |
0.5054 | 0.17 | 8500 | 0.4745 | 0.8258 |
0.5007 | 0.18 | 9000 | 0.4605 | 0.8246 |
0.4988 | 0.19 | 9500 | 0.4377 | 0.8302 |
0.4853 | 0.2 | 10000 | 0.4459 | 0.8315 |
0.4877 | 0.21 | 10500 | 0.4546 | 0.8251 |
0.4729 | 0.22 | 11000 | 0.4619 | 0.8287 |
0.4848 | 0.23 | 11500 | 0.4518 | 0.8244 |
0.4849 | 0.24 | 12000 | 0.4914 | 0.8286 |
0.4793 | 0.25 | 12500 | 0.4583 | 0.8296 |
0.4846 | 0.26 | 13000 | 0.4247 | 0.8367 |
0.4736 | 0.28 | 13500 | 0.4730 | 0.8217 |
0.4758 | 0.29 | 14000 | 0.4596 | 0.8261 |
0.4951 | 0.3 | 14500 | 0.4421 | 0.8357 |
0.4662 | 0.31 | 15000 | 0.4583 | 0.8324 |
0.4779 | 0.32 | 15500 | 0.4264 | 0.8406 |
0.4672 | 0.33 | 16000 | 0.4496 | 0.8267 |
0.477 | 0.34 | 16500 | 0.4235 | 0.8388 |
0.4589 | 0.35 | 17000 | 0.4788 | 0.8235 |
0.4622 | 0.36 | 17500 | 0.4387 | 0.8284 |
0.4741 | 0.37 | 18000 | 0.4556 | 0.8387 |
0.4381 | 0.38 | 18500 | 0.4394 | 0.8372 |
0.4597 | 0.39 | 19000 | 0.4143 | 0.8412 |
0.443 | 0.4 | 19500 | 0.4318 | 0.8377 |
0.4437 | 0.41 | 20000 | 0.4359 | 0.8362 |
0.436 | 0.42 | 20500 | 0.4440 | 0.8332 |
0.4366 | 0.43 | 21000 | 0.4155 | 0.8453 |
0.4536 | 0.44 | 21500 | 0.4283 | 0.8461 |
0.4433 | 0.45 | 22000 | 0.4513 | 0.8344 |
0.4321 | 0.46 | 22500 | 0.4238 | 0.8460 |
0.4332 | 0.47 | 23000 | 0.4640 | 0.8368 |
0.434 | 0.48 | 23500 | 0.4230 | 0.8433 |
0.4355 | 0.49 | 24000 | 0.4378 | 0.8443 |
0.4379 | 0.5 | 24500 | 0.4087 | 0.8431 |
0.4336 | 0.51 | 25000 | 0.4328 | 0.8440 |
0.4205 | 0.52 | 25500 | 0.4387 | 0.8393 |
0.4161 | 0.53 | 26000 | 0.4205 | 0.8413 |
0.4217 | 0.54 | 26500 | 0.4266 | 0.8379 |
0.4145 | 0.55 | 27000 | 0.4220 | 0.8416 |
0.405 | 0.56 | 27500 | 0.4228 | 0.8454 |
0.428 | 0.57 | 28000 | 0.4255 | 0.8483 |
0.4113 | 0.58 | 28500 | 0.4181 | 0.8485 |
0.4112 | 0.59 | 29000 | 0.4191 | 0.8475 |
0.4276 | 0.6 | 29500 | 0.4102 | 0.8457 |
0.4197 | 0.61 | 30000 | 0.4098 | 0.8495 |
0.4132 | 0.62 | 30500 | 0.4300 | 0.8478 |
0.4164 | 0.63 | 31000 | 0.4296 | 0.8362 |
0.4062 | 0.64 | 31500 | 0.4368 | 0.8381 |
0.4123 | 0.65 | 32000 | 0.4221 | 0.8436 |
0.4022 | 0.66 | 32500 | 0.4062 | 0.8478 |
0.4063 | 0.67 | 33000 | 0.4414 | 0.8416 |
0.3973 | 0.68 | 33500 | 0.4106 | 0.8483 |
0.4095 | 0.69 | 34000 | 0.3999 | 0.8505 |
0.4006 | 0.7 | 34500 | 0.4306 | 0.8514 |
0.3982 | 0.71 | 35000 | 0.4110 | 0.8530 |
0.4193 | 0.72 | 35500 | 0.4158 | 0.8478 |
0.394 | 0.73 | 36000 | 0.4135 | 0.8477 |
0.4009 | 0.74 | 36500 | 0.4211 | 0.8559 |
0.3997 | 0.75 | 37000 | 0.4229 | 0.8404 |
0.4054 | 0.76 | 37500 | 0.4251 | 0.8500 |
0.3907 | 0.77 | 38000 | 0.4208 | 0.8513 |
0.3827 | 0.78 | 38500 | 0.4399 | 0.8427 |
0.3943 | 0.79 | 39000 | 0.4497 | 0.8399 |
0.3901 | 0.8 | 39500 | 0.4174 | 0.8496 |
0.3905 | 0.81 | 40000 | 0.4100 | 0.8504 |
0.4034 | 0.83 | 40500 | 0.4259 | 0.8505 |
0.3761 | 0.84 | 41000 | 0.4022 | 0.8492 |
0.3777 | 0.85 | 41500 | 0.3925 | 0.8561 |
0.3775 | 0.86 | 42000 | 0.4287 | 0.8550 |
0.3835 | 0.87 | 42500 | 0.4055 | 0.8493 |
0.384 | 0.88 | 43000 | 0.4081 | 0.8550 |
0.3719 | 0.89 | 43500 | 0.4131 | 0.8511 |
0.3799 | 0.9 | 44000 | 0.4096 | 0.8507 |
0.389 | 0.91 | 44500 | 0.4160 | 0.8532 |
0.3835 | 0.92 | 45000 | 0.3972 | 0.8556 |
0.3813 | 0.93 | 45500 | 0.4134 | 0.8538 |
0.371 | 0.94 | 46000 | 0.4574 | 0.8506 |
0.3762 | 0.95 | 46500 | 0.4067 | 0.8496 |
0.3813 | 0.96 | 47000 | 0.4100 | 0.8526 |
0.3694 | 0.97 | 47500 | 0.3969 | 0.8499 |
0.3971 | 0.98 | 48000 | 0.4142 | 0.8531 |
0.3661 | 0.99 | 48500 | 0.4106 | 0.8532 |
0.3787 | 1.0 | 49000 | 0.4162 | 0.8529 |
0.321 | 1.01 | 49500 | 0.4591 | 0.8494 |
0.314 | 1.02 | 50000 | 0.4458 | 0.8539 |
0.3079 | 1.03 | 50500 | 0.4392 | 0.8560 |
0.2967 | 1.04 | 51000 | 0.4482 | 0.8501 |
0.3062 | 1.05 | 51500 | 0.4355 | 0.8502 |
0.3037 | 1.06 | 52000 | 0.4790 | 0.8528 |
0.3143 | 1.07 | 52500 | 0.4456 | 0.8513 |
0.3055 | 1.08 | 53000 | 0.4474 | 0.8521 |
0.295 | 1.09 | 53500 | 0.4733 | 0.8495 |
0.3046 | 1.1 | 54000 | 0.4562 | 0.8594 |
0.3131 | 1.11 | 54500 | 0.4519 | 0.8533 |
0.3022 | 1.12 | 55000 | 0.4392 | 0.8576 |
0.2858 | 1.13 | 55500 | 0.4498 | 0.8534 |
0.3044 | 1.14 | 56000 | 0.4417 | 0.8496 |
0.2957 | 1.15 | 56500 | 0.4497 | 0.8593 |
0.3063 | 1.16 | 57000 | 0.4504 | 0.8551 |
0.3001 | 1.17 | 57500 | 0.4671 | 0.8526 |
0.3137 | 1.18 | 58000 | 0.4186 | 0.8559 |
0.2918 | 1.19 | 58500 | 0.4568 | 0.8583 |
0.302 | 1.2 | 59000 | 0.4533 | 0.8569 |
0.2944 | 1.21 | 59500 | 0.4400 | 0.8505 |
0.295 | 1.22 | 60000 | 0.4876 | 0.8576 |
0.3083 | 1.23 | 60500 | 0.4565 | 0.8496 |
0.3003 | 1.24 | 61000 | 0.4779 | 0.8516 |
0.3007 | 1.25 | 61500 | 0.4405 | 0.8603 |
0.2849 | 1.26 | 62000 | 0.4665 | 0.8543 |
0.2957 | 1.27 | 62500 | 0.4501 | 0.8509 |
0.2893 | 1.28 | 63000 | 0.4618 | 0.8525 |
0.2994 | 1.29 | 63500 | 0.4217 | 0.8538 |
0.2848 | 1.3 | 64000 | 0.4569 | 0.8608 |
0.3062 | 1.31 | 64500 | 0.4445 | 0.8517 |
0.3033 | 1.32 | 65000 | 0.4486 | 0.8527 |
0.2919 | 1.33 | 65500 | 0.4366 | 0.8550 |
0.294 | 1.34 | 66000 | 0.4745 | 0.8525 |
0.3094 | 1.35 | 66500 | 0.4593 | 0.8561 |
0.2928 | 1.36 | 67000 | 0.4660 | 0.8558 |
0.2934 | 1.38 | 67500 | 0.4348 | 0.8570 |
0.2903 | 1.39 | 68000 | 0.4804 | 0.8520 |
0.3052 | 1.4 | 68500 | 0.4554 | 0.8564 |
0.2944 | 1.41 | 69000 | 0.4287 | 0.8573 |
0.2867 | 1.42 | 69500 | 0.4141 | 0.8546 |
0.2953 | 1.43 | 70000 | 0.5205 | 0.8515 |
0.303 | 1.44 | 70500 | 0.4617 | 0.8500 |
0.2919 | 1.45 | 71000 | 0.4243 | 0.8555 |
0.2769 | 1.46 | 71500 | 0.4479 | 0.8506 |
0.2825 | 1.47 | 72000 | 0.4671 | 0.8573 |
0.2809 | 1.48 | 72500 | 0.4770 | 0.8572 |
0.2789 | 1.49 | 73000 | 0.4434 | 0.8528 |
0.2865 | 1.5 | 73500 | 0.4788 | 0.8536 |
0.2914 | 1.51 | 74000 | 0.4250 | 0.8587 |
0.2778 | 1.52 | 74500 | 0.4679 | 0.8583 |
0.2798 | 1.53 | 75000 | 0.4451 | 0.8542 |
0.2821 | 1.54 | 75500 | 0.4699 | 0.8575 |
0.2888 | 1.55 | 76000 | 0.4190 | 0.8588 |
0.297 | 1.56 | 76500 | 0.4692 | 0.8599 |
0.2927 | 1.57 | 77000 | 0.4465 | 0.8575 |
0.2788 | 1.58 | 77500 | 0.4531 | 0.8579 |
0.2871 | 1.59 | 78000 | 0.4216 | 0.8569 |
0.2724 | 1.6 | 78500 | 0.4848 | 0.8515 |
0.2775 | 1.61 | 79000 | 0.4835 | 0.8548 |
0.28 | 1.62 | 79500 | 0.4978 | 0.8557 |
0.2987 | 1.63 | 80000 | 0.4339 | 0.8579 |
0.2877 | 1.64 | 80500 | 0.4577 | 0.8558 |
0.2938 | 1.65 | 81000 | 0.4640 | 0.8582 |
0.2686 | 1.66 | 81500 | 0.4475 | 0.8606 |
0.2771 | 1.67 | 82000 | 0.4663 | 0.8524 |
0.2789 | 1.68 | 82500 | 0.4413 | 0.8577 |
0.2855 | 1.69 | 83000 | 0.4490 | 0.8522 |
0.2741 | 1.7 | 83500 | 0.4751 | 0.8547 |
0.2859 | 1.71 | 84000 | 0.4891 | 0.8520 |
0.2761 | 1.72 | 84500 | 0.4924 | 0.8545 |
0.2654 | 1.73 | 85000 | 0.4823 | 0.8521 |
0.2774 | 1.74 | 85500 | 0.4432 | 0.8561 |
0.2667 | 1.75 | 86000 | 0.4519 | 0.8500 |
0.2748 | 1.76 | 86500 | 0.4956 | 0.8500 |
0.2771 | 1.77 | 87000 | 0.4315 | 0.8556 |
0.2661 | 1.78 | 87500 | 0.5186 | 0.8529 |
0.2744 | 1.79 | 88000 | 0.5249 | 0.8504 |
0.277 | 1.8 | 88500 | 0.4780 | 0.8548 |
0.2782 | 1.81 | 89000 | 0.4763 | 0.8575 |
0.2736 | 1.82 | 89500 | 0.4585 | 0.8577 |
0.2662 | 1.83 | 90000 | 0.4679 | 0.8566 |
0.2594 | 1.84 | 90500 | 0.5044 | 0.8522 |
0.2684 | 1.85 | 91000 | 0.4698 | 0.8552 |
0.2639 | 1.86 | 91500 | 0.5030 | 0.8532 |
0.2707 | 1.87 | 92000 | 0.4859 | 0.8521 |
0.2605 | 1.88 | 92500 | 0.4907 | 0.8526 |
0.2649 | 1.89 | 93000 | 0.4844 | 0.8548 |
0.2627 | 1.9 | 93500 | 0.4541 | 0.8552 |
0.2636 | 1.91 | 94000 | 0.4497 | 0.8580 |
0.259 | 1.93 | 94500 | 0.4748 | 0.8505 |
0.2607 | 1.94 | 95000 | 0.4909 | 0.8559 |
0.2623 | 1.95 | 95500 | 0.4954 | 0.8571 |
0.2742 | 1.96 | 96000 | 0.4420 | 0.8585 |
0.2545 | 1.97 | 96500 | 0.4562 | 0.8564 |
0.2623 | 1.98 | 97000 | 0.5162 | 0.8490 |
0.2669 | 1.99 | 97500 | 0.4562 | 0.8526 |
0.2625 | 2.0 | 98000 | 0.4853 | 0.8583 |
0.2155 | 2.01 | 98500 | 0.6241 | 0.8550 |
0.2043 | 2.02 | 99000 | 0.6258 | 0.8565 |
0.1968 | 2.03 | 99500 | 0.6347 | 0.8522 |
0.212 | 2.04 | 100000 | 0.5714 | 0.8580 |
0.1921 | 2.05 | 100500 | 0.5958 | 0.8569 |
0.1981 | 2.06 | 101000 | 0.6816 | 0.8543 |
0.1998 | 2.07 | 101500 | 0.6267 | 0.8544 |
0.2105 | 2.08 | 102000 | 0.5886 | 0.8590 |
0.2004 | 2.09 | 102500 | 0.6216 | 0.8577 |
0.2045 | 2.1 | 103000 | 0.6211 | 0.8551 |
0.2129 | 2.11 | 103500 | 0.5851 | 0.8594 |
0.2001 | 2.12 | 104000 | 0.6792 | 0.8468 |
0.2222 | 2.13 | 104500 | 0.5870 | 0.8554 |
0.2081 | 2.14 | 105000 | 0.5643 | 0.8592 |
0.2073 | 2.15 | 105500 | 0.5265 | 0.8596 |
0.2118 | 2.16 | 106000 | 0.6301 | 0.8549 |
0.211 | 2.17 | 106500 | 0.5945 | 0.8622 |
0.2029 | 2.18 | 107000 | 0.6731 | 0.8509 |
0.2073 | 2.19 | 107500 | 0.6049 | 0.8552 |
0.2051 | 2.2 | 108000 | 0.5709 | 0.8616 |
0.1987 | 2.21 | 108500 | 0.5796 | 0.8601 |
0.208 | 2.22 | 109000 | 0.6507 | 0.8576 |
0.2081 | 2.23 | 109500 | 0.6255 | 0.8557 |
0.2063 | 2.24 | 110000 | 0.5992 | 0.8512 |
0.2112 | 2.25 | 110500 | 0.5834 | 0.8519 |
0.2114 | 2.26 | 111000 | 0.5814 | 0.8532 |
0.1993 | 2.27 | 111500 | 0.6537 | 0.8559 |
0.2153 | 2.28 | 112000 | 0.5545 | 0.8602 |
0.1983 | 2.29 | 112500 | 0.5710 | 0.8558 |
0.2132 | 2.3 | 113000 | 0.5757 | 0.8576 |
0.2058 | 2.31 | 113500 | 0.6018 | 0.8530 |
0.2051 | 2.32 | 114000 | 0.6388 | 0.8523 |
0.2005 | 2.33 | 114500 | 0.6415 | 0.8532 |
0.2187 | 2.34 | 115000 | 0.6010 | 0.8553 |
0.1978 | 2.35 | 115500 | 0.6147 | 0.8567 |
0.2136 | 2.36 | 116000 | 0.5904 | 0.8547 |
0.1997 | 2.37 | 116500 | 0.5887 | 0.8544 |
0.1919 | 2.38 | 117000 | 0.6138 | 0.8578 |
0.1959 | 2.39 | 117500 | 0.6519 | 0.8546 |
0.2079 | 2.4 | 118000 | 0.6237 | 0.8551 |
0.2045 | 2.41 | 118500 | 0.6382 | 0.8561 |
0.1963 | 2.42 | 119000 | 0.6502 | 0.8550 |
0.2039 | 2.43 | 119500 | 0.6160 | 0.8554 |
0.2086 | 2.44 | 120000 | 0.6367 | 0.8553 |
0.2007 | 2.45 | 120500 | 0.6417 | 0.8546 |
0.2113 | 2.46 | 121000 | 0.6231 | 0.8583 |
0.2037 | 2.48 | 121500 | 0.6063 | 0.8550 |
0.212 | 2.49 | 122000 | 0.5978 | 0.8544 |
0.2126 | 2.5 | 122500 | 0.5809 | 0.8567 |
0.1935 | 2.51 | 123000 | 0.6117 | 0.8534 |
0.205 | 2.52 | 123500 | 0.6242 | 0.8527 |
0.1931 | 2.53 | 124000 | 0.6220 | 0.8538 |
0.1973 | 2.54 | 124500 | 0.6151 | 0.8523 |
0.2002 | 2.55 | 125000 | 0.6011 | 0.8542 |
0.2041 | 2.56 | 125500 | 0.6344 | 0.8551 |
0.1919 | 2.57 | 126000 | 0.6695 | 0.8519 |
0.2025 | 2.58 | 126500 | 0.6053 | 0.8556 |
0.2047 | 2.59 | 127000 | 0.6209 | 0.8530 |
0.1956 | 2.6 | 127500 | 0.6641 | 0.8553 |
0.2004 | 2.61 | 128000 | 0.5823 | 0.8581 |
0.2035 | 2.62 | 128500 | 0.6495 | 0.8565 |
0.2038 | 2.63 | 129000 | 0.6086 | 0.8608 |
0.2081 | 2.64 | 129500 | 0.6034 | 0.8528 |
0.1965 | 2.65 | 130000 | 0.5650 | 0.8587 |
0.2072 | 2.66 | 130500 | 0.6245 | 0.8614 |
0.2001 | 2.67 | 131000 | 0.6267 | 0.8615 |
0.1984 | 2.68 | 131500 | 0.6314 | 0.8536 |
0.1991 | 2.69 | 132000 | 0.6229 | 0.8583 |
0.1904 | 2.7 | 132500 | 0.6191 | 0.8562 |
0.1945 | 2.71 | 133000 | 0.6019 | 0.8604 |
0.2044 | 2.72 | 133500 | 0.5939 | 0.8598 |
0.2017 | 2.73 | 134000 | 0.5907 | 0.8546 |
0.1952 | 2.74 | 134500 | 0.5762 | 0.8564 |
0.1921 | 2.75 | 135000 | 0.5756 | 0.8551 |
0.1986 | 2.76 | 135500 | 0.6371 | 0.8573 |
0.19 | 2.77 | 136000 | 0.6144 | 0.8567 |
0.1932 | 2.78 | 136500 | 0.7048 | 0.8524 |
0.1932 | 2.79 | 137000 | 0.6870 | 0.8494 |
0.1979 | 2.8 | 137500 | 0.6523 | 0.8525 |
0.2033 | 2.81 | 138000 | 0.6148 | 0.8562 |
0.1923 | 2.82 | 138500 | 0.5841 | 0.8559 |
0.1927 | 2.83 | 139000 | 0.6553 | 0.8558 |
0.1864 | 2.84 | 139500 | 0.6576 | 0.8554 |
0.1959 | 2.85 | 140000 | 0.6552 | 0.8543 |
0.1961 | 2.86 | 140500 | 0.6603 | 0.8521 |
0.1846 | 2.87 | 141000 | 0.6102 | 0.8607 |
0.1887 | 2.88 | 141500 | 0.6500 | 0.8536 |
0.1928 | 2.89 | 142000 | 0.6345 | 0.8569 |
0.19 | 2.9 | 142500 | 0.6240 | 0.8590 |
0.1857 | 2.91 | 143000 | 0.6417 | 0.8555 |
0.1879 | 2.92 | 143500 | 0.6619 | 0.8543 |
0.1958 | 2.93 | 144000 | 0.6151 | 0.8594 |
0.1918 | 2.94 | 144500 | 0.6122 | 0.8571 |
0.1828 | 2.95 | 145000 | 0.6402 | 0.8555 |
0.1955 | 2.96 | 145500 | 0.6392 | 0.8532 |
0.1951 | 2.97 | 146000 | 0.6329 | 0.8556 |
0.1812 | 2.98 | 146500 | 0.6222 | 0.8584 |
0.1806 | 2.99 | 147000 | 0.6225 | 0.8555 |
0.1652 | 3.0 | 147500 | 0.7286 | 0.8562 |
0.1463 | 3.01 | 148000 | 0.7288 | 0.8579 |
0.1485 | 3.03 | 148500 | 0.7197 | 0.8524 |
0.141 | 3.04 | 149000 | 0.7578 | 0.8538 |
0.1536 | 3.05 | 149500 | 0.7195 | 0.8570 |
0.1433 | 3.06 | 150000 | 0.7697 | 0.8550 |
0.1474 | 3.07 | 150500 | 0.7237 | 0.8571 |
0.152 | 3.08 | 151000 | 0.7408 | 0.8548 |
0.1408 | 3.09 | 151500 | 0.7297 | 0.8585 |
0.1153 | 3.1 | 152000 | 0.7630 | 0.8594 |
0.1441 | 3.11 | 152500 | 0.7307 | 0.8593 |
0.1452 | 3.12 | 153000 | 0.7505 | 0.8551 |
0.1397 | 3.13 | 153500 | 0.7455 | 0.8576 |
0.1321 | 3.14 | 154000 | 0.7695 | 0.8552 |
0.1519 | 3.15 | 154500 | 0.7413 | 0.8563 |
0.1436 | 3.16 | 155000 | 0.7371 | 0.8517 |
0.1282 | 3.17 | 155500 | 0.7739 | 0.8562 |
0.1439 | 3.18 | 156000 | 0.7858 | 0.8521 |
0.1309 | 3.19 | 156500 | 0.7487 | 0.8562 |
0.1355 | 3.2 | 157000 | 0.7628 | 0.8590 |
0.1245 | 3.21 | 157500 | 0.7772 | 0.8551 |
0.1313 | 3.22 | 158000 | 0.8121 | 0.8556 |
0.1396 | 3.23 | 158500 | 0.7780 | 0.8540 |
0.135 | 3.24 | 159000 | 0.8283 | 0.8474 |
0.1424 | 3.25 | 159500 | 0.7937 | 0.8530 |
0.1344 | 3.26 | 160000 | 0.8198 | 0.8497 |
0.1428 | 3.27 | 160500 | 0.7596 | 0.8532 |
0.1361 | 3.28 | 161000 | 0.7585 | 0.8556 |
0.1451 | 3.29 | 161500 | 0.7733 | 0.8569 |
0.1314 | 3.3 | 162000 | 0.7815 | 0.8559 |
0.1508 | 3.31 | 162500 | 0.7524 | 0.8529 |
0.133 | 3.32 | 163000 | 0.8037 | 0.8554 |
0.1562 | 3.33 | 163500 | 0.7041 | 0.8570 |
0.1373 | 3.34 | 164000 | 0.7814 | 0.8559 |
0.1453 | 3.35 | 164500 | 0.7847 | 0.8545 |
0.1493 | 3.36 | 165000 | 0.7332 | 0.8582 |
0.1238 | 3.37 | 165500 | 0.7789 | 0.8591 |
0.1455 | 3.38 | 166000 | 0.7825 | 0.8537 |
0.1437 | 3.39 | 166500 | 0.7808 | 0.8550 |
0.1429 | 3.4 | 167000 | 0.7747 | 0.8537 |
0.1464 | 3.41 | 167500 | 0.7245 | 0.8567 |
0.1303 | 3.42 | 168000 | 0.7983 | 0.8542 |
0.14 | 3.43 | 168500 | 0.7847 | 0.8583 |
0.1347 | 3.44 | 169000 | 0.7704 | 0.8601 |
0.1379 | 3.45 | 169500 | 0.7692 | 0.8572 |
0.129 | 3.46 | 170000 | 0.7920 | 0.8558 |
0.134 | 3.47 | 170500 | 0.7868 | 0.8549 |
0.1395 | 3.48 | 171000 | 0.7483 | 0.8561 |
0.1395 | 3.49 | 171500 | 0.7453 | 0.8564 |
0.1317 | 3.5 | 172000 | 0.7779 | 0.8566 |
0.1405 | 3.51 | 172500 | 0.7883 | 0.8537 |
0.1313 | 3.52 | 173000 | 0.7753 | 0.8579 |
0.1414 | 3.53 | 173500 | 0.7961 | 0.8530 |
0.1439 | 3.54 | 174000 | 0.7842 | 0.8556 |
0.1293 | 3.55 | 174500 | 0.7367 | 0.8595 |
0.1208 | 3.57 | 175000 | 0.8063 | 0.8558 |
0.1399 | 3.58 | 175500 | 0.7805 | 0.8592 |
0.1335 | 3.59 | 176000 | 0.7635 | 0.8590 |
0.1462 | 3.6 | 176500 | 0.7314 | 0.8569 |
0.1299 | 3.61 | 177000 | 0.7712 | 0.8569 |
0.1187 | 3.62 | 177500 | 0.7781 | 0.8592 |
0.1337 | 3.63 | 178000 | 0.7943 | 0.8559 |
0.1353 | 3.64 | 178500 | 0.7611 | 0.8589 |
0.1341 | 3.65 | 179000 | 0.7553 | 0.8567 |
0.1448 | 3.66 | 179500 | 0.7639 | 0.8554 |
0.1223 | 3.67 | 180000 | 0.7942 | 0.8561 |
0.1385 | 3.68 | 180500 | 0.7682 | 0.8563 |
0.132 | 3.69 | 181000 | 0.7679 | 0.8571 |
0.1177 | 3.7 | 181500 | 0.7988 | 0.8553 |
0.1313 | 3.71 | 182000 | 0.8045 | 0.8545 |
0.1287 | 3.72 | 182500 | 0.7976 | 0.8555 |
0.1221 | 3.73 | 183000 | 0.8561 | 0.8528 |
0.1373 | 3.74 | 183500 | 0.7870 | 0.8538 |
0.1236 | 3.75 | 184000 | 0.8089 | 0.8535 |
0.1258 | 3.76 | 184500 | 0.8221 | 0.8549 |
0.1358 | 3.77 | 185000 | 0.7897 | 0.8563 |
0.1432 | 3.78 | 185500 | 0.7578 | 0.8530 |
0.1326 | 3.79 | 186000 | 0.7752 | 0.8557 |
0.1286 | 3.8 | 186500 | 0.7791 | 0.8580 |
0.1259 | 3.81 | 187000 | 0.7960 | 0.8588 |
0.1195 | 3.82 | 187500 | 0.7853 | 0.8572 |
0.1332 | 3.83 | 188000 | 0.7800 | 0.8597 |
0.1266 | 3.84 | 188500 | 0.7868 | 0.8587 |
0.1293 | 3.85 | 189000 | 0.7858 | 0.8578 |
0.1165 | 3.86 | 189500 | 0.7846 | 0.8626 |
0.12 | 3.87 | 190000 | 0.8114 | 0.8599 |
0.1282 | 3.88 | 190500 | 0.7862 | 0.8604 |
0.1258 | 3.89 | 191000 | 0.7816 | 0.8584 |
0.1329 | 3.9 | 191500 | 0.7855 | 0.8588 |
0.1201 | 3.91 | 192000 | 0.8030 | 0.8571 |
0.1221 | 3.92 | 192500 | 0.8041 | 0.8587 |
0.121 | 3.93 | 193000 | 0.8183 | 0.8570 |
0.1205 | 3.94 | 193500 | 0.7720 | 0.8578 |
0.1086 | 3.95 | 194000 | 0.8209 | 0.8574 |
0.1207 | 3.96 | 194500 | 0.8044 | 0.8576 |
0.1097 | 3.97 | 195000 | 0.8197 | 0.8586 |
0.1315 | 3.98 | 195500 | 0.7841 | 0.8540 |
0.1285 | 3.99 | 196000 | 0.7809 | 0.8579 |
0.114 | 4.0 | 196500 | 0.8300 | 0.8555 |
0.0815 | 4.01 | 197000 | 0.8590 | 0.8579 |
0.0947 | 4.02 | 197500 | 0.8669 | 0.8557 |
0.0824 | 4.03 | 198000 | 0.8923 | 0.8549 |
0.0967 | 4.04 | 198500 | 0.8733 | 0.8540 |
0.074 | 4.05 | 199000 | 0.8776 | 0.8578 |
0.0857 | 4.06 | 199500 | 0.8710 | 0.8586 |
0.0711 | 4.07 | 200000 | 0.8935 | 0.8581 |
0.0838 | 4.08 | 200500 | 0.8909 | 0.8554 |
0.0975 | 4.09 | 201000 | 0.8622 | 0.8593 |
0.0942 | 4.1 | 201500 | 0.8574 | 0.8592 |
0.0814 | 4.12 | 202000 | 0.8987 | 0.8571 |
0.0786 | 4.13 | 202500 | 0.8932 | 0.8591 |
0.0822 | 4.14 | 203000 | 0.8760 | 0.8597 |
0.0852 | 4.15 | 203500 | 0.9043 | 0.8541 |
0.0784 | 4.16 | 204000 | 0.8749 | 0.8579 |
0.0962 | 4.17 | 204500 | 0.8760 | 0.8566 |
0.0929 | 4.18 | 205000 | 0.8875 | 0.8548 |
0.0796 | 4.19 | 205500 | 0.8650 | 0.8612 |
0.0827 | 4.2 | 206000 | 0.8913 | 0.8598 |
0.0865 | 4.21 | 206500 | 0.9193 | 0.8549 |
0.0795 | 4.22 | 207000 | 0.9041 | 0.8576 |
0.0738 | 4.23 | 207500 | 0.8887 | 0.8594 |
0.0977 | 4.24 | 208000 | 0.8578 | 0.8585 |
0.0872 | 4.25 | 208500 | 0.8666 | 0.8545 |
0.0791 | 4.26 | 209000 | 0.8585 | 0.8608 |
0.0761 | 4.27 | 209500 | 0.8992 | 0.8594 |
0.0789 | 4.28 | 210000 | 0.8969 | 0.8604 |
0.0935 | 4.29 | 210500 | 0.8766 | 0.8588 |
0.0837 | 4.3 | 211000 | 0.9028 | 0.8585 |
0.0854 | 4.31 | 211500 | 0.8875 | 0.8571 |
0.0798 | 4.32 | 212000 | 0.8960 | 0.8567 |
0.0797 | 4.33 | 212500 | 0.9093 | 0.8555 |
0.0764 | 4.34 | 213000 | 0.8981 | 0.8581 |
0.0868 | 4.35 | 213500 | 0.8687 | 0.8603 |
0.079 | 4.36 | 214000 | 0.8900 | 0.8583 |
0.0767 | 4.37 | 214500 | 0.8849 | 0.8618 |
0.0748 | 4.38 | 215000 | 0.9142 | 0.8571 |
0.0736 | 4.39 | 215500 | 0.9084 | 0.8579 |
0.0688 | 4.4 | 216000 | 0.9309 | 0.8565 |
0.0743 | 4.41 | 216500 | 0.9332 | 0.8568 |
0.0727 | 4.42 | 217000 | 0.9368 | 0.8586 |
0.0862 | 4.43 | 217500 | 0.9165 | 0.8581 |
0.0776 | 4.44 | 218000 | 0.9160 | 0.8580 |
0.0919 | 4.45 | 218500 | 0.9005 | 0.8563 |
0.0919 | 4.46 | 219000 | 0.8861 | 0.8570 |
0.0701 | 4.47 | 219500 | 0.8882 | 0.8596 |
0.0727 | 4.48 | 220000 | 0.9038 | 0.8592 |
0.0847 | 4.49 | 220500 | 0.8863 | 0.8596 |
0.075 | 4.5 | 221000 | 0.8946 | 0.8597 |
0.0805 | 4.51 | 221500 | 0.9081 | 0.8594 |
0.0897 | 4.52 | 222000 | 0.8750 | 0.8589 |
0.0825 | 4.53 | 222500 | 0.8962 | 0.8596 |
0.0803 | 4.54 | 223000 | 0.8732 | 0.8608 |
0.0795 | 4.55 | 223500 | 0.8846 | 0.8590 |
0.0775 | 4.56 | 224000 | 0.8976 | 0.8595 |
0.0687 | 4.57 | 224500 | 0.9001 | 0.8598 |
0.0769 | 4.58 | 225000 | 0.8921 | 0.8596 |
0.0707 | 4.59 | 225500 | 0.9036 | 0.8597 |
0.08 | 4.6 | 226000 | 0.9301 | 0.8557 |
0.076 | 4.61 | 226500 | 0.8969 | 0.8588 |
0.0793 | 4.62 | 227000 | 0.8928 | 0.8597 |
0.0826 | 4.63 | 227500 | 0.8890 | 0.8580 |
0.0893 | 4.64 | 228000 | 0.8843 | 0.8591 |
0.0715 | 4.65 | 228500 | 0.9125 | 0.8578 |
0.0669 | 4.67 | 229000 | 0.9085 | 0.8591 |
0.0724 | 4.68 | 229500 | 0.9056 | 0.8588 |
0.0741 | 4.69 | 230000 | 0.9138 | 0.8592 |
0.0662 | 4.7 | 230500 | 0.9181 | 0.8587 |
0.0703 | 4.71 | 231000 | 0.9096 | 0.8601 |
0.0696 | 4.72 | 231500 | 0.8982 | 0.8615 |
0.0704 | 4.73 | 232000 | 0.9190 | 0.8597 |
0.0754 | 4.74 | 232500 | 0.9082 | 0.8596 |
0.0718 | 4.75 | 233000 | 0.8955 | 0.8612 |
0.0686 | 4.76 | 233500 | 0.9087 | 0.8606 |
0.0647 | 4.77 | 234000 | 0.9036 | 0.8602 |
0.071 | 4.78 | 234500 | 0.9109 | 0.8589 |
0.0701 | 4.79 | 235000 | 0.9095 | 0.8602 |
0.0732 | 4.8 | 235500 | 0.9077 | 0.8601 |
0.0626 | 4.81 | 236000 | 0.9172 | 0.8615 |
0.083 | 4.82 | 236500 | 0.9144 | 0.8600 |
0.0749 | 4.83 | 237000 | 0.9153 | 0.8600 |
0.0651 | 4.84 | 237500 | 0.9186 | 0.8598 |
0.0609 | 4.85 | 238000 | 0.9195 | 0.8604 |
0.0733 | 4.86 | 238500 | 0.9185 | 0.8594 |
0.0776 | 4.87 | 239000 | 0.9074 | 0.8599 |
0.071 | 4.88 | 239500 | 0.9091 | 0.8596 |
0.0724 | 4.89 | 240000 | 0.9041 | 0.8601 |
0.0627 | 4.9 | 240500 | 0.9075 | 0.8594 |
0.0722 | 4.91 | 241000 | 0.9051 | 0.8604 |
0.0629 | 4.92 | 241500 | 0.9043 | 0.8601 |
0.0706 | 4.93 | 242000 | 0.9077 | 0.8591 |
0.0763 | 4.94 | 242500 | 0.9047 | 0.8593 |
0.0785 | 4.95 | 243000 | 0.9006 | 0.8600 |
0.0667 | 4.96 | 243500 | 0.9030 | 0.8594 |
0.0743 | 4.97 | 244000 | 0.9025 | 0.8594 |
0.0775 | 4.98 | 244500 | 0.9024 | 0.8594 |
0.0562 | 4.99 | 245000 | 0.9020 | 0.8596 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu102
- Datasets 2.6.1
- Tokenizers 0.11.6