End of training
Browse files- README.md +243 -0
- adapter_config.json +26 -0
- adapter_model.safetensors +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +42 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,243 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: TheBloke/Mistral-7B-v0.1-GPTQ
|
| 9 |
+
model-index:
|
| 10 |
+
- name: mistral-rand
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# mistral-rand
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [TheBloke/Mistral-7B-v0.1-GPTQ](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GPTQ) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.4527
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 2e-05
|
| 41 |
+
- train_batch_size: 4
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
+
- seed: 42
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: cosine
|
| 46 |
+
- num_epochs: 15
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 53 |
+
| 1.7543 | 0.03 | 50 | 0.9190 |
|
| 54 |
+
| 0.8445 | 0.05 | 100 | 0.7860 |
|
| 55 |
+
| 0.7819 | 0.07 | 150 | 0.7460 |
|
| 56 |
+
| 0.7231 | 0.1 | 200 | 0.7147 |
|
| 57 |
+
| 0.6985 | 0.12 | 250 | 0.6924 |
|
| 58 |
+
| 0.6887 | 0.15 | 300 | 0.6823 |
|
| 59 |
+
| 0.6836 | 0.17 | 350 | 0.6702 |
|
| 60 |
+
| 0.6624 | 0.2 | 400 | 0.6574 |
|
| 61 |
+
| 0.6712 | 0.23 | 450 | 0.6507 |
|
| 62 |
+
| 0.6354 | 0.25 | 500 | 0.6417 |
|
| 63 |
+
| 0.6089 | 0.28 | 550 | 0.6373 |
|
| 64 |
+
| 0.6236 | 0.3 | 600 | 0.6284 |
|
| 65 |
+
| 0.6161 | 0.33 | 650 | 0.6228 |
|
| 66 |
+
| 0.6367 | 0.35 | 700 | 0.6152 |
|
| 67 |
+
| 0.6329 | 0.38 | 750 | 0.6097 |
|
| 68 |
+
| 0.5944 | 0.4 | 800 | 0.6076 |
|
| 69 |
+
| 0.6036 | 0.42 | 850 | 0.6030 |
|
| 70 |
+
| 0.5767 | 0.45 | 900 | 0.5989 |
|
| 71 |
+
| 0.6079 | 0.47 | 950 | 0.5954 |
|
| 72 |
+
| 0.5915 | 0.5 | 1000 | 0.5916 |
|
| 73 |
+
| 0.5911 | 0.53 | 1050 | 0.5859 |
|
| 74 |
+
| 0.5752 | 0.55 | 1100 | 0.5847 |
|
| 75 |
+
| 0.5698 | 0.57 | 1150 | 0.5802 |
|
| 76 |
+
| 0.5813 | 0.6 | 1200 | 0.5754 |
|
| 77 |
+
| 0.5918 | 0.62 | 1250 | 0.5735 |
|
| 78 |
+
| 0.5587 | 0.65 | 1300 | 0.5677 |
|
| 79 |
+
| 0.5933 | 0.68 | 1350 | 0.5620 |
|
| 80 |
+
| 0.5262 | 0.7 | 1400 | 0.5522 |
|
| 81 |
+
| 0.5455 | 0.72 | 1450 | 0.5457 |
|
| 82 |
+
| 0.5472 | 0.75 | 1500 | 0.5416 |
|
| 83 |
+
| 0.536 | 0.78 | 1550 | 0.5400 |
|
| 84 |
+
| 0.527 | 0.8 | 1600 | 0.5393 |
|
| 85 |
+
| 0.5516 | 0.82 | 1650 | 0.5350 |
|
| 86 |
+
| 0.5578 | 0.85 | 1700 | 0.5356 |
|
| 87 |
+
| 0.5501 | 0.88 | 1750 | 0.5297 |
|
| 88 |
+
| 0.5316 | 0.9 | 1800 | 0.5288 |
|
| 89 |
+
| 0.5436 | 0.93 | 1850 | 0.5268 |
|
| 90 |
+
| 0.514 | 0.95 | 1900 | 0.5295 |
|
| 91 |
+
| 0.5249 | 0.97 | 1950 | 0.5246 |
|
| 92 |
+
| 0.538 | 1.0 | 2000 | 0.5226 |
|
| 93 |
+
| 0.4967 | 1.02 | 2050 | 0.5237 |
|
| 94 |
+
| 0.4991 | 1.05 | 2100 | 0.5261 |
|
| 95 |
+
| 0.5142 | 1.07 | 2150 | 0.5203 |
|
| 96 |
+
| 0.4891 | 1.1 | 2200 | 0.5174 |
|
| 97 |
+
| 0.5058 | 1.12 | 2250 | 0.5173 |
|
| 98 |
+
| 0.4895 | 1.15 | 2300 | 0.5182 |
|
| 99 |
+
| 0.4918 | 1.18 | 2350 | 0.5139 |
|
| 100 |
+
| 0.485 | 1.2 | 2400 | 0.5091 |
|
| 101 |
+
| 0.5173 | 1.23 | 2450 | 0.5121 |
|
| 102 |
+
| 0.5021 | 1.25 | 2500 | 0.5116 |
|
| 103 |
+
| 0.4834 | 1.27 | 2550 | 0.5097 |
|
| 104 |
+
| 0.4754 | 1.3 | 2600 | 0.5137 |
|
| 105 |
+
| 0.4907 | 1.32 | 2650 | 0.5059 |
|
| 106 |
+
| 0.5155 | 1.35 | 2700 | 0.5051 |
|
| 107 |
+
| 0.4965 | 1.38 | 2750 | 0.5050 |
|
| 108 |
+
| 0.5148 | 1.4 | 2800 | 0.5043 |
|
| 109 |
+
| 0.4709 | 1.43 | 2850 | 0.5032 |
|
| 110 |
+
| 0.4864 | 1.45 | 2900 | 0.5037 |
|
| 111 |
+
| 0.4794 | 1.48 | 2950 | 0.5029 |
|
| 112 |
+
| 0.4803 | 1.5 | 3000 | 0.5012 |
|
| 113 |
+
| 0.4843 | 1.52 | 3050 | 0.5017 |
|
| 114 |
+
| 0.4726 | 1.55 | 3100 | 0.4984 |
|
| 115 |
+
| 0.4773 | 1.57 | 3150 | 0.4968 |
|
| 116 |
+
| 0.4673 | 1.6 | 3200 | 0.4995 |
|
| 117 |
+
| 0.4803 | 1.62 | 3250 | 0.4990 |
|
| 118 |
+
| 0.4926 | 1.65 | 3300 | 0.4965 |
|
| 119 |
+
| 0.4814 | 1.68 | 3350 | 0.4973 |
|
| 120 |
+
| 0.4714 | 1.7 | 3400 | 0.4930 |
|
| 121 |
+
| 0.4797 | 1.73 | 3450 | 0.4903 |
|
| 122 |
+
| 0.4807 | 1.75 | 3500 | 0.4932 |
|
| 123 |
+
| 0.4815 | 1.77 | 3550 | 0.4888 |
|
| 124 |
+
| 0.4852 | 1.8 | 3600 | 0.4874 |
|
| 125 |
+
| 0.4802 | 1.82 | 3650 | 0.4887 |
|
| 126 |
+
| 0.4701 | 1.85 | 3700 | 0.4897 |
|
| 127 |
+
| 0.4572 | 1.88 | 3750 | 0.4873 |
|
| 128 |
+
| 0.4469 | 1.9 | 3800 | 0.4878 |
|
| 129 |
+
| 0.478 | 1.93 | 3850 | 0.4885 |
|
| 130 |
+
| 0.4449 | 1.95 | 3900 | 0.4866 |
|
| 131 |
+
| 0.4634 | 1.98 | 3950 | 0.4843 |
|
| 132 |
+
| 0.4718 | 2.0 | 4000 | 0.4838 |
|
| 133 |
+
| 0.4458 | 2.02 | 4050 | 0.4822 |
|
| 134 |
+
| 0.461 | 2.05 | 4100 | 0.4801 |
|
| 135 |
+
| 0.4247 | 2.08 | 4150 | 0.4856 |
|
| 136 |
+
| 0.4325 | 2.1 | 4200 | 0.4830 |
|
| 137 |
+
| 0.4354 | 2.12 | 4250 | 0.4827 |
|
| 138 |
+
| 0.4313 | 2.15 | 4300 | 0.4807 |
|
| 139 |
+
| 0.4753 | 2.17 | 4350 | 0.4812 |
|
| 140 |
+
| 0.4442 | 2.2 | 4400 | 0.4833 |
|
| 141 |
+
| 0.4431 | 2.23 | 4450 | 0.4851 |
|
| 142 |
+
| 0.4485 | 2.25 | 4500 | 0.4815 |
|
| 143 |
+
| 0.4416 | 2.27 | 4550 | 0.4813 |
|
| 144 |
+
| 0.4613 | 2.3 | 4600 | 0.4777 |
|
| 145 |
+
| 0.4121 | 2.33 | 4650 | 0.4775 |
|
| 146 |
+
| 0.4311 | 2.35 | 4700 | 0.4768 |
|
| 147 |
+
| 0.4532 | 2.38 | 4750 | 0.4765 |
|
| 148 |
+
| 0.4342 | 2.4 | 4800 | 0.4781 |
|
| 149 |
+
| 0.4189 | 2.42 | 4850 | 0.4743 |
|
| 150 |
+
| 0.443 | 2.45 | 4900 | 0.4742 |
|
| 151 |
+
| 0.4596 | 2.48 | 4950 | 0.4734 |
|
| 152 |
+
| 0.4193 | 2.5 | 5000 | 0.4719 |
|
| 153 |
+
| 0.4321 | 2.52 | 5050 | 0.4723 |
|
| 154 |
+
| 0.4456 | 2.55 | 5100 | 0.4713 |
|
| 155 |
+
| 0.4464 | 2.58 | 5150 | 0.4694 |
|
| 156 |
+
| 0.4273 | 2.6 | 5200 | 0.4700 |
|
| 157 |
+
| 0.4239 | 2.62 | 5250 | 0.4701 |
|
| 158 |
+
| 0.4282 | 2.65 | 5300 | 0.4687 |
|
| 159 |
+
| 0.4303 | 2.67 | 5350 | 0.4686 |
|
| 160 |
+
| 0.4559 | 2.7 | 5400 | 0.4695 |
|
| 161 |
+
| 0.4542 | 2.73 | 5450 | 0.4692 |
|
| 162 |
+
| 0.4532 | 2.75 | 5500 | 0.4685 |
|
| 163 |
+
| 0.4505 | 2.77 | 5550 | 0.4663 |
|
| 164 |
+
| 0.4533 | 2.8 | 5600 | 0.4660 |
|
| 165 |
+
| 0.4351 | 2.83 | 5650 | 0.4640 |
|
| 166 |
+
| 0.4354 | 2.85 | 5700 | 0.4651 |
|
| 167 |
+
| 0.4374 | 2.88 | 5750 | 0.4664 |
|
| 168 |
+
| 0.4571 | 2.9 | 5800 | 0.4662 |
|
| 169 |
+
| 0.4663 | 2.92 | 5850 | 0.4636 |
|
| 170 |
+
| 0.4211 | 2.95 | 5900 | 0.4645 |
|
| 171 |
+
| 0.4349 | 2.98 | 5950 | 0.4622 |
|
| 172 |
+
| 0.4167 | 3.0 | 6000 | 0.4634 |
|
| 173 |
+
| 0.4176 | 3.02 | 6050 | 0.4621 |
|
| 174 |
+
| 0.4387 | 3.05 | 6100 | 0.4607 |
|
| 175 |
+
| 0.395 | 3.08 | 6150 | 0.4638 |
|
| 176 |
+
| 0.4186 | 3.1 | 6200 | 0.4623 |
|
| 177 |
+
| 0.3993 | 3.12 | 6250 | 0.4622 |
|
| 178 |
+
| 0.4009 | 3.15 | 6300 | 0.4631 |
|
| 179 |
+
| 0.4033 | 3.17 | 6350 | 0.4640 |
|
| 180 |
+
| 0.389 | 3.2 | 6400 | 0.4662 |
|
| 181 |
+
| 0.4037 | 3.23 | 6450 | 0.4618 |
|
| 182 |
+
| 0.4287 | 3.25 | 6500 | 0.4617 |
|
| 183 |
+
| 0.3917 | 3.27 | 6550 | 0.4611 |
|
| 184 |
+
| 0.3944 | 3.3 | 6600 | 0.4626 |
|
| 185 |
+
| 0.4088 | 3.33 | 6650 | 0.4622 |
|
| 186 |
+
| 0.4205 | 3.35 | 6700 | 0.4604 |
|
| 187 |
+
| 0.4273 | 3.38 | 6750 | 0.4608 |
|
| 188 |
+
| 0.4139 | 3.4 | 6800 | 0.4607 |
|
| 189 |
+
| 0.3888 | 3.42 | 6850 | 0.4603 |
|
| 190 |
+
| 0.4353 | 3.45 | 6900 | 0.4573 |
|
| 191 |
+
| 0.4222 | 3.48 | 6950 | 0.4577 |
|
| 192 |
+
| 0.4083 | 3.5 | 7000 | 0.4571 |
|
| 193 |
+
| 0.4161 | 3.52 | 7050 | 0.4560 |
|
| 194 |
+
| 0.3879 | 3.55 | 7100 | 0.4540 |
|
| 195 |
+
| 0.3819 | 3.58 | 7150 | 0.4570 |
|
| 196 |
+
| 0.4345 | 3.6 | 7200 | 0.4551 |
|
| 197 |
+
| 0.4101 | 3.62 | 7250 | 0.4569 |
|
| 198 |
+
| 0.4194 | 3.65 | 7300 | 0.4543 |
|
| 199 |
+
| 0.4066 | 3.67 | 7350 | 0.4563 |
|
| 200 |
+
| 0.4144 | 3.7 | 7400 | 0.4553 |
|
| 201 |
+
| 0.4134 | 3.73 | 7450 | 0.4566 |
|
| 202 |
+
| 0.3906 | 3.75 | 7500 | 0.4550 |
|
| 203 |
+
| 0.4128 | 3.77 | 7550 | 0.4546 |
|
| 204 |
+
| 0.4227 | 3.8 | 7600 | 0.4535 |
|
| 205 |
+
| 0.4069 | 3.83 | 7650 | 0.4517 |
|
| 206 |
+
| 0.3927 | 3.85 | 7700 | 0.4548 |
|
| 207 |
+
| 0.3977 | 3.88 | 7750 | 0.4521 |
|
| 208 |
+
| 0.4184 | 3.9 | 7800 | 0.4516 |
|
| 209 |
+
| 0.3854 | 3.92 | 7850 | 0.4513 |
|
| 210 |
+
| 0.4129 | 3.95 | 7900 | 0.4524 |
|
| 211 |
+
| 0.3998 | 3.98 | 7950 | 0.4548 |
|
| 212 |
+
| 0.4227 | 4.0 | 8000 | 0.4534 |
|
| 213 |
+
| 0.3788 | 4.03 | 8050 | 0.4520 |
|
| 214 |
+
| 0.3732 | 4.05 | 8100 | 0.4501 |
|
| 215 |
+
| 0.375 | 4.08 | 8150 | 0.4565 |
|
| 216 |
+
| 0.3845 | 4.1 | 8200 | 0.4515 |
|
| 217 |
+
| 0.378 | 4.12 | 8250 | 0.4492 |
|
| 218 |
+
| 0.3874 | 4.15 | 8300 | 0.4508 |
|
| 219 |
+
| 0.3802 | 4.17 | 8350 | 0.4510 |
|
| 220 |
+
| 0.3596 | 4.2 | 8400 | 0.4524 |
|
| 221 |
+
| 0.4009 | 4.22 | 8450 | 0.4549 |
|
| 222 |
+
| 0.4105 | 4.25 | 8500 | 0.4515 |
|
| 223 |
+
| 0.3716 | 4.28 | 8550 | 0.4508 |
|
| 224 |
+
| 0.3673 | 4.3 | 8600 | 0.4497 |
|
| 225 |
+
| 0.3882 | 4.33 | 8650 | 0.4513 |
|
| 226 |
+
| 0.375 | 4.35 | 8700 | 0.4524 |
|
| 227 |
+
| 0.3654 | 4.38 | 8750 | 0.4503 |
|
| 228 |
+
| 0.3983 | 4.4 | 8800 | 0.4509 |
|
| 229 |
+
| 0.4067 | 4.42 | 8850 | 0.4487 |
|
| 230 |
+
| 0.3966 | 4.45 | 8900 | 0.4519 |
|
| 231 |
+
| 0.378 | 4.47 | 8950 | 0.4505 |
|
| 232 |
+
| 0.3755 | 4.5 | 9000 | 0.4508 |
|
| 233 |
+
| 0.3855 | 4.53 | 9050 | 0.4500 |
|
| 234 |
+
| 0.3938 | 4.55 | 9100 | 0.4527 |
|
| 235 |
+
|
| 236 |
+
|
| 237 |
+
### Framework versions
|
| 238 |
+
|
| 239 |
+
- PEFT 0.7.1
|
| 240 |
+
- Transformers 4.36.2
|
| 241 |
+
- Pytorch 2.1.2+cu121
|
| 242 |
+
- Datasets 2.16.1
|
| 243 |
+
- Tokenizers 0.15.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "TheBloke/Mistral-7B-v0.1-GPTQ",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
+
"lora_alpha": 64,
|
| 13 |
+
"lora_dropout": 0.05,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 64,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"q_proj",
|
| 23 |
+
"v_proj"
|
| 24 |
+
],
|
| 25 |
+
"task_type": "CAUSAL_LM"
|
| 26 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:833a38f2dac78e02757fe5510f275a7217a04b933e6b739b46ab9dae1f4fcb6b
|
| 3 |
+
size 109069176
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": true,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
| 3 |
+
size 493443
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<unk>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<s>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": true,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "</s>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": true,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
}
|
| 29 |
+
},
|
| 30 |
+
"additional_special_tokens": [],
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"clean_up_tokenization_spaces": false,
|
| 33 |
+
"eos_token": "</s>",
|
| 34 |
+
"legacy": true,
|
| 35 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 36 |
+
"pad_token": "</s>",
|
| 37 |
+
"sp_model_kwargs": {},
|
| 38 |
+
"spaces_between_special_tokens": false,
|
| 39 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 40 |
+
"unk_token": "<unk>",
|
| 41 |
+
"use_default_system_prompt": true
|
| 42 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c999347cd7233feb58c6450844581530c5797ee1474d5259832723b88117d3ca
|
| 3 |
+
size 4664
|