<!-- 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. -->
headlines
This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.4046
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: 0.0005
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 2000
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.3722 | 0.01 | 1000 | 6.1947 |
| 5.8723 | 0.03 | 2000 | 5.7025 |
| 5.5296 | 0.04 | 3000 | 5.4065 |
| 5.3667 | 0.05 | 4000 | 5.2681 |
| 5.2477 | 0.07 | 5000 | 5.1903 |
| 5.1877 | 0.08 | 6000 | 5.0935 |
| 5.1422 | 0.1 | 7000 | 5.0452 |
| 5.0818 | 0.11 | 8000 | 4.9876 |
| 5.104 | 0.12 | 9000 | 4.9677 |
| 5.0345 | 0.14 | 10000 | 4.9417 |
| 5.0253 | 0.15 | 11000 | 4.8983 |
| 4.9818 | 0.16 | 12000 | 4.8814 |
| 4.9996 | 0.18 | 13000 | 4.8422 |
| 4.9835 | 0.19 | 14000 | 4.8159 |
| 4.9212 | 0.21 | 15000 | 4.7951 |
| 4.9698 | 0.22 | 16000 | 4.7970 |
| 4.9405 | 0.23 | 17000 | 4.7749 |
| 4.9429 | 0.25 | 18000 | 4.7551 |
| 4.915 | 0.26 | 19000 | 4.7575 |
| 4.9308 | 0.27 | 20000 | 4.7387 |
| 4.8789 | 0.29 | 21000 | 4.7353 |
| 4.9302 | 0.3 | 22000 | 4.7149 |
| 4.8546 | 0.32 | 23000 | 4.7028 |
| 4.8979 | 0.33 | 24000 | 4.6912 |
| 4.8402 | 0.34 | 25000 | 4.6899 |
| 4.8194 | 0.36 | 26000 | 4.6762 |
| 4.8176 | 0.37 | 27000 | 4.6629 |
| 4.881 | 0.38 | 28000 | 4.6611 |
| 4.7831 | 0.4 | 29000 | 4.6705 |
| 4.8144 | 0.41 | 30000 | 4.6440 |
| 4.7651 | 0.43 | 31000 | 4.6262 |
| 4.8063 | 0.44 | 32000 | 4.6144 |
| 4.8004 | 0.45 | 33000 | 4.5894 |
| 4.7853 | 0.47 | 34000 | 4.6104 |
| 4.7983 | 0.48 | 35000 | 4.5856 |
| 4.7683 | 0.49 | 36000 | 4.5874 |
| 4.7828 | 0.51 | 37000 | 4.5970 |
| 4.7491 | 0.52 | 38000 | 4.5622 |
| 4.6805 | 0.54 | 39000 | 4.5814 |
| 4.7669 | 0.55 | 40000 | 4.5596 |
| 4.7515 | 0.56 | 41000 | 4.5585 |
| 4.7032 | 0.58 | 42000 | 4.5397 |
| 4.7043 | 0.59 | 43000 | 4.5224 |
| 4.6694 | 0.6 | 44000 | 4.5010 |
| 4.6543 | 0.62 | 45000 | 4.5089 |
| 4.6597 | 0.63 | 46000 | 4.4671 |
| 4.6814 | 0.65 | 47000 | 4.4496 |
| 4.6546 | 0.66 | 48000 | 4.4511 |
| 4.6576 | 0.67 | 49000 | 4.4552 |
| 4.6602 | 0.69 | 50000 | 4.4423 |
| 4.656 | 0.7 | 51000 | 4.4215 |
| 4.609 | 0.71 | 52000 | 4.4343 |
| 4.6102 | 0.73 | 53000 | 4.4076 |
| 4.6264 | 0.74 | 54000 | 4.3941 |
| 4.6187 | 0.76 | 55000 | 4.3950 |
| 4.5811 | 0.77 | 56000 | 4.3798 |
| 4.5749 | 0.78 | 57000 | 4.3555 |
| 4.4991 | 0.8 | 58000 | 4.3596 |
| 4.596 | 0.81 | 59000 | 4.3433 |
| 4.5383 | 0.82 | 60000 | 4.3221 |
| 4.565 | 0.84 | 61000 | 4.3264 |
| 4.4996 | 0.85 | 62000 | 4.3128 |
| 4.5016 | 0.87 | 63000 | 4.2906 |
| 4.5139 | 0.88 | 64000 | 4.2749 |
| 4.4796 | 0.89 | 65000 | 4.2676 |
| 4.497 | 0.91 | 66000 | 4.2679 |
| 4.459 | 0.92 | 67000 | 4.2367 |
| 4.5178 | 0.93 | 68000 | 4.2294 |
| 4.3981 | 0.95 | 69000 | 4.2066 |
| 4.4179 | 0.96 | 70000 | 4.1949 |
| 4.4269 | 0.98 | 71000 | 4.1782 |
| 4.3541 | 0.99 | 72000 | 4.1718 |
| 4.4032 | 1.0 | 73000 | 4.1599 |
| 4.3656 | 1.02 | 74000 | 4.1415 |
| 4.3256 | 1.03 | 75000 | 4.1283 |
| 4.3469 | 1.04 | 76000 | 4.1057 |
| 4.2923 | 1.06 | 77000 | 4.1011 |
| 4.3402 | 1.07 | 78000 | 4.0753 |
| 4.3233 | 1.09 | 79000 | 4.0664 |
| 4.2872 | 1.1 | 80000 | 4.0397 |
| 4.2772 | 1.11 | 81000 | 4.0319 |
| 4.2364 | 1.13 | 82000 | 4.0200 |
| 4.3073 | 1.14 | 83000 | 4.0171 |
| 4.1852 | 1.15 | 84000 | 4.0025 |
| 4.2935 | 1.17 | 85000 | 3.9770 |
| 4.3258 | 1.18 | 86000 | 3.9670 |
| 4.2046 | 1.19 | 87000 | 3.9425 |
| 4.2298 | 1.21 | 88000 | 3.9377 |
| 4.2161 | 1.22 | 89000 | 3.9316 |
| 4.1894 | 1.24 | 90000 | 3.9136 |
| 4.1873 | 1.25 | 91000 | 3.8801 |
| 4.2103 | 1.26 | 92000 | 3.8825 |
| 4.1437 | 1.28 | 93000 | 3.8588 |
| 4.1136 | 1.29 | 94000 | 3.8515 |
| 4.1303 | 1.3 | 95000 | 3.8311 |
| 4.2183 | 1.32 | 96000 | 3.8139 |
| 4.1434 | 1.33 | 97000 | 3.7916 |
| 4.1247 | 1.35 | 98000 | 3.7898 |
| 4.0946 | 1.36 | 99000 | 3.7764 |
| 4.0452 | 1.37 | 100000 | 3.7577 |
| 4.0954 | 1.39 | 101000 | 3.7483 |
| 4.1195 | 1.4 | 102000 | 3.7259 |
| 4.0884 | 1.41 | 103000 | 3.7151 |
| 4.0441 | 1.43 | 104000 | 3.7042 |
| 4.0575 | 1.44 | 105000 | 3.6830 |
| 4.0668 | 1.46 | 106000 | 3.6778 |
| 4.0006 | 1.47 | 107000 | 3.6596 |
| 4.0275 | 1.48 | 108000 | 3.6575 |
| 4.0107 | 1.5 | 109000 | 3.6429 |
| 4.0156 | 1.51 | 110000 | 3.6217 |
| 3.9957 | 1.52 | 111000 | 3.6097 |
| 3.988 | 1.54 | 112000 | 3.5979 |
| 3.9985 | 1.55 | 113000 | 3.5856 |
| 3.9697 | 1.57 | 114000 | 3.5750 |
| 3.922 | 1.58 | 115000 | 3.5601 |
| 3.9144 | 1.59 | 116000 | 3.5500 |
| 3.9287 | 1.61 | 117000 | 3.5421 |
| 3.9738 | 1.62 | 118000 | 3.5266 |
| 3.9523 | 1.63 | 119000 | 3.5256 |
| 3.9573 | 1.65 | 120000 | 3.5076 |
| 3.9187 | 1.66 | 121000 | 3.4987 |
| 3.8782 | 1.68 | 122000 | 3.4914 |
| 3.8946 | 1.69 | 123000 | 3.4825 |
| 3.9306 | 1.7 | 124000 | 3.4740 |
| 3.852 | 1.72 | 125000 | 3.4690 |
| 3.8554 | 1.73 | 126000 | 3.4605 |
| 3.8673 | 1.74 | 127000 | 3.4530 |
| 3.8667 | 1.76 | 128000 | 3.4457 |
| 3.8303 | 1.77 | 129000 | 3.4427 |
| 3.8967 | 1.79 | 130000 | 3.4384 |
| 3.8384 | 1.8 | 131000 | 3.4299 |
| 3.8537 | 1.81 | 132000 | 3.4254 |
| 3.8981 | 1.83 | 133000 | 3.4217 |
| 3.8538 | 1.84 | 134000 | 3.4184 |
| 3.9361 | 1.85 | 135000 | 3.4149 |
| 3.9096 | 1.87 | 136000 | 3.4111 |
| 3.8507 | 1.88 | 137000 | 3.4106 |
| 3.8434 | 1.9 | 138000 | 3.4089 |
| 3.8476 | 1.91 | 139000 | 3.4070 |
| 3.8788 | 1.92 | 140000 | 3.4059 |
| 3.8648 | 1.94 | 141000 | 3.4059 |
| 3.8596 | 1.95 | 142000 | 3.4051 |
| 3.8782 | 1.96 | 143000 | 3.4047 |
| 3.7519 | 1.98 | 144000 | 3.4046 |
| 3.8754 | 1.99 | 145000 | 3.4046 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3
