Model save
Browse files- README.md +163 -132
- config.json +1 -1
- generation_config.json +1 -1
- model-00001-of-00003.safetensors +1 -1
- model-00002-of-00003.safetensors +1 -1
- model-00003-of-00003.safetensors +1 -1
- tokenizer.json +1 -6
- training_args.bin +2 -2
README.md
CHANGED
|
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [deepseek-ai/deepseek-math-7b-base](https://huggingface.co/deepseek-ai/deepseek-math-7b-base) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
@@ -36,151 +36,182 @@ More information needed
|
|
| 36 |
### Training hyperparameters
|
| 37 |
|
| 38 |
The following hyperparameters were used during training:
|
| 39 |
-
- learning_rate:
|
| 40 |
-
- train_batch_size:
|
| 41 |
- eval_batch_size: 8
|
| 42 |
- seed: 42
|
| 43 |
- gradient_accumulation_steps: 4
|
| 44 |
-
- total_train_batch_size:
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
-
- lr_scheduler_type:
|
| 47 |
-
- lr_scheduler_warmup_steps:
|
| 48 |
- num_epochs: 1
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.
|
| 76 |
-
| 0.
|
| 77 |
-
| 0.
|
| 78 |
-
| 0.
|
| 79 |
-
| 0.
|
| 80 |
-
| 0.
|
| 81 |
-
| 0.
|
| 82 |
-
| 0.
|
| 83 |
-
| 0.
|
| 84 |
-
| 0.
|
| 85 |
-
| 0.
|
| 86 |
-
| 0.
|
| 87 |
-
| 0.
|
| 88 |
-
| 0.
|
| 89 |
-
| 0.
|
| 90 |
-
| 0.
|
| 91 |
-
| 0.
|
| 92 |
-
| 0.
|
| 93 |
-
| 0.
|
| 94 |
-
| 0.
|
| 95 |
-
| 0.
|
| 96 |
-
| 0.
|
| 97 |
-
| 0.
|
| 98 |
-
| 0.
|
| 99 |
-
| 0.
|
| 100 |
-
| 0.
|
| 101 |
-
| 0.
|
| 102 |
-
| 0.
|
| 103 |
-
| 0.
|
| 104 |
-
| 0.
|
| 105 |
-
| 0.
|
| 106 |
-
| 0.
|
| 107 |
-
| 0.
|
| 108 |
-
| 0.
|
| 109 |
-
| 0.
|
| 110 |
-
| 0.
|
| 111 |
-
| 0.
|
| 112 |
-
| 0.
|
| 113 |
-
| 0.
|
| 114 |
-
| 0.
|
| 115 |
-
| 0.
|
| 116 |
-
| 0.
|
| 117 |
-
| 0.
|
| 118 |
-
| 0.
|
| 119 |
-
| 0.
|
| 120 |
-
| 0.
|
| 121 |
-
| 0.
|
| 122 |
-
| 0.
|
| 123 |
-
| 0.
|
| 124 |
-
| 0.
|
| 125 |
-
| 0.
|
| 126 |
-
| 0.
|
| 127 |
-
| 0.
|
| 128 |
-
| 0.
|
| 129 |
-
| 0.
|
| 130 |
-
| 0.
|
| 131 |
-
| 0.
|
| 132 |
-
| 0.
|
| 133 |
-
| 0.
|
| 134 |
-
| 0.
|
| 135 |
-
| 0.
|
| 136 |
-
| 0.
|
| 137 |
-
| 0.
|
| 138 |
-
| 0.
|
| 139 |
-
| 0.
|
| 140 |
-
| 0.
|
| 141 |
-
| 0.
|
| 142 |
-
| 0.
|
| 143 |
-
| 0.
|
| 144 |
-
| 0.
|
| 145 |
-
| 0.
|
| 146 |
-
| 0.
|
| 147 |
-
| 0.
|
| 148 |
-
| 0.
|
| 149 |
-
| 0.
|
| 150 |
-
| 0.
|
| 151 |
-
| 0.
|
| 152 |
-
| 0.
|
| 153 |
-
| 0.
|
| 154 |
-
| 0.
|
| 155 |
-
| 0.
|
| 156 |
-
| 0.
|
| 157 |
-
| 0.
|
| 158 |
-
| 0.
|
| 159 |
-
| 0.
|
| 160 |
-
| 0.
|
| 161 |
-
| 0.
|
| 162 |
-
| 0.
|
| 163 |
-
| 0.
|
| 164 |
-
| 0.
|
| 165 |
-
| 0.
|
| 166 |
-
| 0.
|
| 167 |
-
| 0.
|
| 168 |
-
| 0.
|
| 169 |
-
| 0.
|
| 170 |
-
| 0.
|
| 171 |
-
| 0.
|
| 172 |
-
| 0.
|
| 173 |
-
| 0.
|
| 174 |
-
| 0.
|
| 175 |
-
| 0.
|
| 176 |
-
| 0.
|
| 177 |
-
| 0.
|
| 178 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 179 |
|
| 180 |
|
| 181 |
### Framework versions
|
| 182 |
|
| 183 |
-
- Transformers 4.
|
| 184 |
- Pytorch 2.1.0a0+32f93b1
|
| 185 |
- Datasets 2.17.1
|
| 186 |
- Tokenizers 0.15.2
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [deepseek-ai/deepseek-math-7b-base](https://huggingface.co/deepseek-ai/deepseek-math-7b-base) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.0475
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 36 |
### Training hyperparameters
|
| 37 |
|
| 38 |
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 3e-06
|
| 40 |
+
- train_batch_size: 16
|
| 41 |
- eval_batch_size: 8
|
| 42 |
- seed: 42
|
| 43 |
- gradient_accumulation_steps: 4
|
| 44 |
+
- total_train_batch_size: 64
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: constant_with_warmup
|
| 47 |
+
- lr_scheduler_warmup_steps: 1
|
| 48 |
- num_epochs: 1
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 54 |
+
| 0.8425 | 0.01 | 1 | 0.4687 |
|
| 55 |
+
| 0.8496 | 0.01 | 2 | 0.4640 |
|
| 56 |
+
| 0.819 | 0.02 | 3 | 0.4579 |
|
| 57 |
+
| 0.8173 | 0.03 | 4 | 0.4532 |
|
| 58 |
+
| 0.8297 | 0.03 | 5 | 0.4495 |
|
| 59 |
+
| 0.8014 | 0.04 | 6 | 0.4447 |
|
| 60 |
+
| 0.7981 | 0.04 | 7 | 0.4399 |
|
| 61 |
+
| 0.7716 | 0.05 | 8 | 0.4375 |
|
| 62 |
+
| 0.6996 | 0.06 | 9 | 0.4332 |
|
| 63 |
+
| 0.6631 | 0.06 | 10 | 0.4302 |
|
| 64 |
+
| 0.6596 | 0.07 | 11 | 0.4247 |
|
| 65 |
+
| 0.6887 | 0.08 | 12 | 0.4221 |
|
| 66 |
+
| 0.6373 | 0.08 | 13 | 0.4180 |
|
| 67 |
+
| 0.6536 | 0.09 | 14 | 0.4161 |
|
| 68 |
+
| 0.6112 | 0.1 | 15 | 0.4102 |
|
| 69 |
+
| 0.6113 | 0.1 | 16 | 0.4072 |
|
| 70 |
+
| 0.5427 | 0.11 | 17 | 0.4041 |
|
| 71 |
+
| 0.5434 | 0.12 | 18 | 0.4005 |
|
| 72 |
+
| 0.5098 | 0.12 | 19 | 0.3955 |
|
| 73 |
+
| 0.5567 | 0.13 | 20 | 0.3914 |
|
| 74 |
+
| 0.4984 | 0.13 | 21 | 0.3889 |
|
| 75 |
+
| 0.5414 | 0.14 | 22 | 0.3852 |
|
| 76 |
+
| 0.4851 | 0.15 | 23 | 0.3818 |
|
| 77 |
+
| 0.4599 | 0.15 | 24 | 0.3769 |
|
| 78 |
+
| 0.4423 | 0.16 | 25 | 0.3736 |
|
| 79 |
+
| 0.4562 | 0.17 | 26 | 0.3711 |
|
| 80 |
+
| 0.4359 | 0.17 | 27 | 0.3677 |
|
| 81 |
+
| 0.4182 | 0.18 | 28 | 0.3631 |
|
| 82 |
+
| 0.4302 | 0.19 | 29 | 0.3612 |
|
| 83 |
+
| 0.4294 | 0.19 | 30 | 0.3556 |
|
| 84 |
+
| 0.4169 | 0.2 | 31 | 0.3527 |
|
| 85 |
+
| 0.3764 | 0.2 | 32 | 0.3483 |
|
| 86 |
+
| 0.3773 | 0.21 | 33 | 0.3451 |
|
| 87 |
+
| 0.3611 | 0.22 | 34 | 0.3423 |
|
| 88 |
+
| 0.3524 | 0.22 | 35 | 0.3380 |
|
| 89 |
+
| 0.3528 | 0.23 | 36 | 0.3337 |
|
| 90 |
+
| 0.3666 | 0.24 | 37 | 0.3306 |
|
| 91 |
+
| 0.3536 | 0.24 | 38 | 0.3266 |
|
| 92 |
+
| 0.3812 | 0.25 | 39 | 0.3223 |
|
| 93 |
+
| 0.2976 | 0.26 | 40 | 0.3191 |
|
| 94 |
+
| 0.2805 | 0.26 | 41 | 0.3157 |
|
| 95 |
+
| 0.2739 | 0.27 | 42 | 0.3115 |
|
| 96 |
+
| 0.2994 | 0.28 | 43 | 0.3089 |
|
| 97 |
+
| 0.2864 | 0.28 | 44 | 0.3031 |
|
| 98 |
+
| 0.2745 | 0.29 | 45 | 0.3003 |
|
| 99 |
+
| 0.2666 | 0.29 | 46 | 0.2968 |
|
| 100 |
+
| 0.3109 | 0.3 | 47 | 0.2914 |
|
| 101 |
+
| 0.2603 | 0.31 | 48 | 0.2873 |
|
| 102 |
+
| 0.2761 | 0.31 | 49 | 0.2821 |
|
| 103 |
+
| 0.2566 | 0.32 | 50 | 0.2792 |
|
| 104 |
+
| 0.2637 | 0.33 | 51 | 0.2758 |
|
| 105 |
+
| 0.2388 | 0.33 | 52 | 0.2706 |
|
| 106 |
+
| 0.2474 | 0.34 | 53 | 0.2651 |
|
| 107 |
+
| 0.2462 | 0.35 | 54 | 0.2607 |
|
| 108 |
+
| 0.2617 | 0.35 | 55 | 0.2567 |
|
| 109 |
+
| 0.2387 | 0.36 | 56 | 0.2513 |
|
| 110 |
+
| 0.2171 | 0.36 | 57 | 0.2482 |
|
| 111 |
+
| 0.2324 | 0.37 | 58 | 0.2428 |
|
| 112 |
+
| 0.227 | 0.38 | 59 | 0.2384 |
|
| 113 |
+
| 0.2366 | 0.38 | 60 | 0.2332 |
|
| 114 |
+
| 0.2174 | 0.39 | 61 | 0.2288 |
|
| 115 |
+
| 0.2083 | 0.4 | 62 | 0.2237 |
|
| 116 |
+
| 0.1935 | 0.4 | 63 | 0.2198 |
|
| 117 |
+
| 0.2141 | 0.41 | 64 | 0.2152 |
|
| 118 |
+
| 0.2044 | 0.42 | 65 | 0.2111 |
|
| 119 |
+
| 0.2061 | 0.42 | 66 | 0.2064 |
|
| 120 |
+
| 0.2005 | 0.43 | 67 | 0.2011 |
|
| 121 |
+
| 0.1895 | 0.44 | 68 | 0.1944 |
|
| 122 |
+
| 0.1823 | 0.44 | 69 | 0.1902 |
|
| 123 |
+
| 0.1763 | 0.45 | 70 | 0.1874 |
|
| 124 |
+
| 0.1897 | 0.45 | 71 | 0.1814 |
|
| 125 |
+
| 0.1853 | 0.46 | 72 | 0.1764 |
|
| 126 |
+
| 0.1674 | 0.47 | 73 | 0.1706 |
|
| 127 |
+
| 0.1642 | 0.47 | 74 | 0.1648 |
|
| 128 |
+
| 0.1677 | 0.48 | 75 | 0.1604 |
|
| 129 |
+
| 0.1551 | 0.49 | 76 | 0.1544 |
|
| 130 |
+
| 0.1457 | 0.49 | 77 | 0.1486 |
|
| 131 |
+
| 0.1497 | 0.5 | 78 | 0.1434 |
|
| 132 |
+
| 0.1389 | 0.51 | 79 | 0.1374 |
|
| 133 |
+
| 0.1396 | 0.51 | 80 | 0.1325 |
|
| 134 |
+
| 0.1297 | 0.52 | 81 | 0.1266 |
|
| 135 |
+
| 0.1298 | 0.52 | 82 | 0.1211 |
|
| 136 |
+
| 0.1162 | 0.53 | 83 | 0.1164 |
|
| 137 |
+
| 0.13 | 0.54 | 84 | 0.1127 |
|
| 138 |
+
| 0.1207 | 0.54 | 85 | 0.1085 |
|
| 139 |
+
| 0.1165 | 0.55 | 86 | 0.1039 |
|
| 140 |
+
| 0.1088 | 0.56 | 87 | 0.1012 |
|
| 141 |
+
| 0.1082 | 0.56 | 88 | 0.0969 |
|
| 142 |
+
| 0.109 | 0.57 | 89 | 0.0952 |
|
| 143 |
+
| 0.1102 | 0.58 | 90 | 0.0924 |
|
| 144 |
+
| 0.1026 | 0.58 | 91 | 0.0891 |
|
| 145 |
+
| 0.1058 | 0.59 | 92 | 0.0875 |
|
| 146 |
+
| 0.0973 | 0.6 | 93 | 0.0861 |
|
| 147 |
+
| 0.0963 | 0.6 | 94 | 0.0859 |
|
| 148 |
+
| 0.0941 | 0.61 | 95 | 0.0838 |
|
| 149 |
+
| 0.0983 | 0.61 | 96 | 0.0824 |
|
| 150 |
+
| 0.0884 | 0.62 | 97 | 0.0812 |
|
| 151 |
+
| 0.0855 | 0.63 | 98 | 0.0798 |
|
| 152 |
+
| 0.0875 | 0.63 | 99 | 0.0781 |
|
| 153 |
+
| 0.0833 | 0.64 | 100 | 0.0766 |
|
| 154 |
+
| 0.0836 | 0.65 | 101 | 0.0757 |
|
| 155 |
+
| 0.0809 | 0.65 | 102 | 0.0745 |
|
| 156 |
+
| 0.0839 | 0.66 | 103 | 0.0731 |
|
| 157 |
+
| 0.0748 | 0.67 | 104 | 0.0717 |
|
| 158 |
+
| 0.0779 | 0.67 | 105 | 0.0710 |
|
| 159 |
+
| 0.0768 | 0.68 | 106 | 0.0705 |
|
| 160 |
+
| 0.0794 | 0.68 | 107 | 0.0693 |
|
| 161 |
+
| 0.079 | 0.69 | 108 | 0.0679 |
|
| 162 |
+
| 0.0808 | 0.7 | 109 | 0.0667 |
|
| 163 |
+
| 0.0785 | 0.7 | 110 | 0.0658 |
|
| 164 |
+
| 0.0669 | 0.71 | 111 | 0.0649 |
|
| 165 |
+
| 0.0715 | 0.72 | 112 | 0.0640 |
|
| 166 |
+
| 0.0751 | 0.72 | 113 | 0.0632 |
|
| 167 |
+
| 0.0727 | 0.73 | 114 | 0.0626 |
|
| 168 |
+
| 0.0725 | 0.74 | 115 | 0.0625 |
|
| 169 |
+
| 0.0665 | 0.74 | 116 | 0.0614 |
|
| 170 |
+
| 0.0627 | 0.75 | 117 | 0.0605 |
|
| 171 |
+
| 0.0681 | 0.76 | 118 | 0.0597 |
|
| 172 |
+
| 0.0673 | 0.76 | 119 | 0.0593 |
|
| 173 |
+
| 0.0741 | 0.77 | 120 | 0.0592 |
|
| 174 |
+
| 0.0686 | 0.77 | 121 | 0.0584 |
|
| 175 |
+
| 0.0618 | 0.78 | 122 | 0.0584 |
|
| 176 |
+
| 0.065 | 0.79 | 123 | 0.0574 |
|
| 177 |
+
| 0.061 | 0.79 | 124 | 0.0572 |
|
| 178 |
+
| 0.0685 | 0.8 | 125 | 0.0571 |
|
| 179 |
+
| 0.0621 | 0.81 | 126 | 0.0560 |
|
| 180 |
+
| 0.0636 | 0.81 | 127 | 0.0558 |
|
| 181 |
+
| 0.0581 | 0.82 | 128 | 0.0550 |
|
| 182 |
+
| 0.0576 | 0.83 | 129 | 0.0547 |
|
| 183 |
+
| 0.0628 | 0.83 | 130 | 0.0544 |
|
| 184 |
+
| 0.056 | 0.84 | 131 | 0.0542 |
|
| 185 |
+
| 0.0572 | 0.84 | 132 | 0.0536 |
|
| 186 |
+
| 0.0605 | 0.85 | 133 | 0.0529 |
|
| 187 |
+
| 0.0626 | 0.86 | 134 | 0.0520 |
|
| 188 |
+
| 0.0566 | 0.86 | 135 | 0.0517 |
|
| 189 |
+
| 0.0575 | 0.87 | 136 | 0.0519 |
|
| 190 |
+
| 0.0571 | 0.88 | 137 | 0.0514 |
|
| 191 |
+
| 0.0594 | 0.88 | 138 | 0.0510 |
|
| 192 |
+
| 0.0528 | 0.89 | 139 | 0.0513 |
|
| 193 |
+
| 0.0507 | 0.9 | 140 | 0.0508 |
|
| 194 |
+
| 0.0587 | 0.9 | 141 | 0.0503 |
|
| 195 |
+
| 0.0558 | 0.91 | 142 | 0.0504 |
|
| 196 |
+
| 0.0538 | 0.92 | 143 | 0.0500 |
|
| 197 |
+
| 0.0509 | 0.92 | 144 | 0.0509 |
|
| 198 |
+
| 0.0538 | 0.93 | 145 | 0.0504 |
|
| 199 |
+
| 0.0524 | 0.93 | 146 | 0.0498 |
|
| 200 |
+
| 0.059 | 0.94 | 147 | 0.0496 |
|
| 201 |
+
| 0.0508 | 0.95 | 148 | 0.0494 |
|
| 202 |
+
| 0.0563 | 0.95 | 149 | 0.0491 |
|
| 203 |
+
| 0.0472 | 0.96 | 150 | 0.0484 |
|
| 204 |
+
| 0.0526 | 0.97 | 151 | 0.0482 |
|
| 205 |
+
| 0.0525 | 0.97 | 152 | 0.0482 |
|
| 206 |
+
| 0.0483 | 0.98 | 153 | 0.0478 |
|
| 207 |
+
| 0.0541 | 0.99 | 154 | 0.0483 |
|
| 208 |
+
| 0.0521 | 0.99 | 155 | 0.0474 |
|
| 209 |
+
| 0.0556 | 1.0 | 156 | 0.0475 |
|
| 210 |
|
| 211 |
|
| 212 |
### Framework versions
|
| 213 |
|
| 214 |
+
- Transformers 4.37.2
|
| 215 |
- Pytorch 2.1.0a0+32f93b1
|
| 216 |
- Datasets 2.17.1
|
| 217 |
- Tokenizers 0.15.2
|
config.json
CHANGED
|
@@ -22,7 +22,7 @@
|
|
| 22 |
"rope_theta": 10000.0,
|
| 23 |
"tie_word_embeddings": false,
|
| 24 |
"torch_dtype": "bfloat16",
|
| 25 |
-
"transformers_version": "4.
|
| 26 |
"use_cache": false,
|
| 27 |
"vocab_size": 102400
|
| 28 |
}
|
|
|
|
| 22 |
"rope_theta": 10000.0,
|
| 23 |
"tie_word_embeddings": false,
|
| 24 |
"torch_dtype": "bfloat16",
|
| 25 |
+
"transformers_version": "4.37.2",
|
| 26 |
"use_cache": false,
|
| 27 |
"vocab_size": 102400
|
| 28 |
}
|
generation_config.json
CHANGED
|
@@ -2,5 +2,5 @@
|
|
| 2 |
"_from_model_config": true,
|
| 3 |
"bos_token_id": 100000,
|
| 4 |
"eos_token_id": 100001,
|
| 5 |
-
"transformers_version": "4.
|
| 6 |
}
|
|
|
|
| 2 |
"_from_model_config": true,
|
| 3 |
"bos_token_id": 100000,
|
| 4 |
"eos_token_id": 100001,
|
| 5 |
+
"transformers_version": "4.37.2"
|
| 6 |
}
|
model-00001-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4987202208
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f91dc19225ec73e87754e34b76609afca36af9c2dd55d3d72cf1b8c7eef94b04
|
| 3 |
size 4987202208
|
model-00002-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4980945440
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d22277bbf5dc25eee82e9e2462c9cfc1145b2ce7569768ea593fd692a11a270
|
| 3 |
size 4980945440
|
model-00003-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3852615520
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:92cc01c4570a98e323bc663759b694e7830453f13d8020fd8dcef38fab31651f
|
| 3 |
size 3852615520
|
tokenizer.json
CHANGED
|
@@ -1,11 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"version": "1.0",
|
| 3 |
-
"truncation":
|
| 4 |
-
"direction": "Right",
|
| 5 |
-
"max_length": 1500,
|
| 6 |
-
"strategy": "LongestFirst",
|
| 7 |
-
"stride": 0
|
| 8 |
-
},
|
| 9 |
"padding": null,
|
| 10 |
"added_tokens": [
|
| 11 |
{
|
|
|
|
| 1 |
{
|
| 2 |
"version": "1.0",
|
| 3 |
+
"truncation": null,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
"padding": null,
|
| 5 |
"added_tokens": [
|
| 6 |
{
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b0ff3d1d65f57060aa7c8d146100b0a1c115fbecf69a47a8a6da62026c873448
|
| 3 |
+
size 4728
|