gpt2-cs00
This model is a fine-tuned version of gpt2 on the gpt2-cs00 dataset. It achieves the following results on the evaluation set:
- Loss: 1.3143
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: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
11.0601 | 0.02 | 200 | 1.8848 |
1.8408 | 0.05 | 400 | 1.7685 |
1.7219 | 0.07 | 600 | 1.6998 |
1.7133 | 0.09 | 800 | 1.6720 |
1.6776 | 0.12 | 1000 | 1.6420 |
1.6309 | 0.14 | 1200 | 1.7187 |
1.6157 | 0.16 | 1400 | 1.6025 |
1.5546 | 0.18 | 1600 | 1.5661 |
1.4834 | 0.21 | 1800 | 1.5589 |
1.5641 | 0.23 | 2000 | 1.5451 |
1.5133 | 0.25 | 2200 | 1.5195 |
1.5373 | 0.28 | 2400 | 1.5099 |
1.498 | 0.3 | 2600 | 1.5026 |
1.4382 | 0.32 | 2800 | 1.4915 |
1.4585 | 0.35 | 3000 | 1.4937 |
1.4493 | 0.37 | 3200 | 1.4737 |
1.403 | 0.39 | 3400 | 1.4713 |
1.4216 | 0.42 | 3600 | 1.4573 |
1.4204 | 0.44 | 3800 | 1.4684 |
1.5143 | 0.46 | 4000 | 1.4458 |
1.5003 | 0.48 | 4200 | 1.4115 |
1.4828 | 0.51 | 4400 | 1.4446 |
1.4098 | 0.53 | 4600 | 1.4133 |
1.4208 | 0.55 | 4800 | 1.4178 |
1.401 | 0.58 | 5000 | 1.3915 |
1.3639 | 0.6 | 5200 | 1.4326 |
1.3752 | 0.62 | 5400 | 1.3989 |
1.4016 | 0.65 | 5600 | 1.3873 |
1.4157 | 0.67 | 5800 | 1.3792 |
1.4421 | 0.69 | 6000 | 1.3809 |
1.4024 | 0.72 | 6200 | 1.3780 |
1.4031 | 0.74 | 6400 | 1.4014 |
1.4033 | 0.76 | 6600 | 1.4148 |
1.4009 | 0.78 | 6800 | 1.3824 |
1.4519 | 0.81 | 7000 | 1.3795 |
1.377 | 0.83 | 7200 | 1.3762 |
1.4153 | 0.85 | 7400 | 1.3608 |
1.4112 | 0.88 | 7600 | 1.3853 |
1.409 | 0.9 | 7800 | 1.3728 |
1.4125 | 0.92 | 8000 | 1.3661 |
1.3637 | 0.95 | 8200 | 1.3609 |
1.3902 | 0.97 | 8400 | 1.3591 |
1.4463 | 0.99 | 8600 | 1.3665 |
1.3782 | 1.02 | 8800 | 1.3634 |
1.3468 | 1.04 | 9000 | 1.3728 |
1.3339 | 1.06 | 9200 | 1.3712 |
1.3171 | 1.09 | 9400 | 1.3557 |
1.357 | 1.11 | 9600 | 1.3723 |
1.3791 | 1.13 | 9800 | 1.3617 |
1.3888 | 1.15 | 10000 | 1.3477 |
1.3923 | 1.18 | 10200 | 1.3512 |
1.342 | 1.2 | 10400 | 1.3538 |
1.3485 | 1.22 | 10600 | 1.3595 |
1.3523 | 1.25 | 10800 | 1.3623 |
1.3881 | 1.27 | 11000 | 1.3416 |
1.3741 | 1.29 | 11200 | 1.3523 |
1.3869 | 1.32 | 11400 | 1.3442 |
1.3545 | 1.34 | 11600 | 1.3490 |
1.3571 | 1.36 | 11800 | 1.3491 |
1.3396 | 1.39 | 12000 | 1.3510 |
1.3713 | 1.41 | 12200 | 1.3341 |
1.3165 | 1.43 | 12400 | 1.3376 |
1.3236 | 1.45 | 12600 | 1.3364 |
1.3028 | 1.48 | 12800 | 1.3322 |
1.3671 | 1.5 | 13000 | 1.3403 |
1.3295 | 1.52 | 13200 | 1.3377 |
1.3807 | 1.55 | 13400 | 1.3264 |
1.3714 | 1.57 | 13600 | 1.3271 |
1.3249 | 1.59 | 13800 | 1.3388 |
1.3656 | 1.62 | 14000 | 1.3319 |
1.2864 | 1.64 | 14200 | 1.3321 |
1.352 | 1.66 | 14400 | 1.3497 |
1.3599 | 1.69 | 14600 | 1.3268 |
1.3191 | 1.71 | 14800 | 1.3339 |
1.3136 | 1.73 | 15000 | 1.3336 |
1.3338 | 1.75 | 15200 | 1.3265 |
1.3528 | 1.78 | 15400 | 1.3363 |
1.3538 | 1.8 | 15600 | 1.3196 |
1.2879 | 1.82 | 15800 | 1.3335 |
1.3217 | 1.85 | 16000 | 1.3376 |
1.3657 | 1.87 | 16200 | 1.3257 |
1.3351 | 1.89 | 16400 | 1.3262 |
1.3469 | 1.92 | 16600 | 1.3299 |
1.3053 | 1.94 | 16800 | 1.3329 |
1.3332 | 1.96 | 17000 | 1.3212 |
1.3466 | 1.99 | 17200 | 1.3317 |
1.3743 | 2.01 | 17400 | 1.3302 |
1.3227 | 2.03 | 17600 | 1.3332 |
1.2728 | 2.05 | 17800 | 1.3450 |
1.3239 | 2.08 | 18000 | 1.3414 |
1.3661 | 2.1 | 18200 | 1.3243 |
1.298 | 2.12 | 18400 | 1.3315 |
1.2974 | 2.15 | 18600 | 1.3310 |
1.3174 | 2.17 | 18800 | 1.3224 |
1.3121 | 2.19 | 19000 | 1.3233 |
1.3527 | 2.22 | 19200 | 1.3211 |
1.3712 | 2.24 | 19400 | 1.3143 |
1.2873 | 2.26 | 19600 | 1.3302 |
1.306 | 2.29 | 19800 | 1.3211 |
1.3161 | 2.31 | 20000 | 1.3242 |
1.308 | 2.33 | 20200 | 1.3176 |
1.3403 | 2.35 | 20400 | 1.3143 |
1.3688 | 2.38 | 20600 | 1.3195 |
1.2743 | 2.4 | 20800 | 1.3230 |
1.2892 | 2.42 | 21000 | 1.3287 |
1.3782 | 2.45 | 21200 | 1.3137 |
1.3331 | 2.47 | 21400 | 1.3148 |
1.3182 | 2.49 | 21600 | 1.3220 |
1.2542 | 2.52 | 21800 | 1.3332 |
1.2879 | 2.54 | 22000 | 1.3229 |
1.316 | 2.56 | 22200 | 1.3181 |
1.2989 | 2.59 | 22400 | 1.3155 |
1.3095 | 2.61 | 22600 | 1.3218 |
1.2457 | 2.63 | 22800 | 1.3185 |
1.3053 | 2.65 | 23000 | 1.3168 |
1.3036 | 2.68 | 23200 | 1.3180 |
1.2861 | 2.7 | 23400 | 1.3117 |
1.3 | 2.72 | 23600 | 1.3208 |
1.3026 | 2.75 | 23800 | 1.3147 |
1.3006 | 2.77 | 24000 | 1.3211 |
1.3477 | 2.79 | 24200 | 1.3140 |
1.2851 | 2.82 | 24400 | 1.3208 |
1.2859 | 2.84 | 24600 | 1.3172 |
1.3286 | 2.86 | 24800 | 1.3151 |
1.3237 | 2.89 | 25000 | 1.3148 |
1.3503 | 2.91 | 25200 | 1.3133 |
1.27 | 2.93 | 25400 | 1.3138 |
1.2998 | 2.96 | 25600 | 1.3151 |
1.3461 | 2.98 | 25800 | 1.3143 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.1+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0
- Downloads last month
- 9