--- license: llama3 library_name: peft tags: - generated_from_trainer base_model: meta-llama/Meta-Llama-3-8B-Instruct model-index: - name: final_llama results: [] --- # final_llama This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.0641 ## 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: 4 - eval_batch_size: 4 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 100 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | 2.767 | 1.0 | 1 | 2.7479 | | 2.767 | 2.0 | 2 | 2.7247 | | 2.767 | 3.0 | 3 | 2.7010 | | 2.767 | 4.0 | 4 | 2.6766 | | 2.767 | 5.0 | 5 | 2.6526 | | 2.767 | 6.0 | 6 | 2.6286 | | 2.767 | 7.0 | 7 | 2.6044 | | 2.767 | 8.0 | 8 | 2.5795 | | 2.767 | 9.0 | 9 | 2.5543 | | 2.767 | 10.0 | 10 | 2.5293 | | 2.767 | 11.0 | 11 | 2.5038 | | 2.767 | 12.0 | 12 | 2.4779 | | 2.767 | 13.0 | 13 | 2.4519 | | 2.767 | 14.0 | 14 | 2.4254 | | 2.767 | 15.0 | 15 | 2.3991 | | 2.767 | 16.0 | 16 | 2.3726 | | 2.767 | 17.0 | 17 | 2.3458 | | 2.767 | 18.0 | 18 | 2.3199 | | 2.767 | 19.0 | 19 | 2.2934 | | 2.767 | 20.0 | 20 | 2.2677 | | 2.767 | 21.0 | 21 | 2.2425 | | 2.767 | 22.0 | 22 | 2.2178 | | 2.767 | 23.0 | 23 | 2.1940 | | 2.767 | 24.0 | 24 | 2.1701 | | 2.767 | 25.0 | 25 | 2.1468 | | 2.767 | 26.0 | 26 | 2.1236 | | 2.767 | 27.0 | 27 | 2.1001 | | 2.767 | 28.0 | 28 | 2.0772 | | 2.767 | 29.0 | 29 | 2.0543 | | 2.767 | 30.0 | 30 | 2.0314 | | 2.767 | 31.0 | 31 | 2.0088 | | 2.767 | 32.0 | 32 | 1.9860 | | 2.767 | 33.0 | 33 | 1.9644 | | 2.767 | 34.0 | 34 | 1.9425 | | 2.767 | 35.0 | 35 | 1.9207 | | 2.767 | 36.0 | 36 | 1.8995 | | 2.767 | 37.0 | 37 | 1.8785 | | 2.767 | 38.0 | 38 | 1.8575 | | 2.767 | 39.0 | 39 | 1.8370 | | 2.767 | 40.0 | 40 | 1.8163 | | 2.767 | 41.0 | 41 | 1.7959 | | 2.767 | 42.0 | 42 | 1.7752 | | 2.767 | 43.0 | 43 | 1.7550 | | 2.767 | 44.0 | 44 | 1.7349 | | 2.767 | 45.0 | 45 | 1.7146 | | 2.767 | 46.0 | 46 | 1.6944 | | 2.767 | 47.0 | 47 | 1.6746 | | 2.767 | 48.0 | 48 | 1.6544 | | 2.767 | 49.0 | 49 | 1.6346 | | 2.767 | 50.0 | 50 | 1.6150 | | 2.767 | 51.0 | 51 | 1.5955 | | 2.767 | 52.0 | 52 | 1.5760 | | 2.767 | 53.0 | 53 | 1.5566 | | 2.767 | 54.0 | 54 | 1.5377 | | 2.767 | 55.0 | 55 | 1.5191 | | 2.767 | 56.0 | 56 | 1.5005 | | 2.767 | 57.0 | 57 | 1.4819 | | 2.767 | 58.0 | 58 | 1.4646 | | 2.767 | 59.0 | 59 | 1.4469 | | 2.767 | 60.0 | 60 | 1.4297 | | 2.767 | 61.0 | 61 | 1.4130 | | 2.767 | 62.0 | 62 | 1.3963 | | 2.767 | 63.0 | 63 | 1.3803 | | 2.767 | 64.0 | 64 | 1.3645 | | 2.767 | 65.0 | 65 | 1.3488 | | 2.767 | 66.0 | 66 | 1.3336 | | 2.767 | 67.0 | 67 | 1.3183 | | 2.767 | 68.0 | 68 | 1.3041 | | 2.767 | 69.0 | 69 | 1.2896 | | 2.767 | 70.0 | 70 | 1.2756 | | 2.767 | 71.0 | 71 | 1.2623 | | 2.767 | 72.0 | 72 | 1.2494 | | 2.767 | 73.0 | 73 | 1.2368 | | 2.767 | 74.0 | 74 | 1.2244 | | 2.767 | 75.0 | 75 | 1.2128 | | 2.767 | 76.0 | 76 | 1.2019 | | 2.767 | 77.0 | 77 | 1.1909 | | 2.767 | 78.0 | 78 | 1.1804 | | 2.767 | 79.0 | 79 | 1.1706 | | 2.767 | 80.0 | 80 | 1.1607 | | 2.767 | 81.0 | 81 | 1.1516 | | 2.767 | 82.0 | 82 | 1.1430 | | 2.767 | 83.0 | 83 | 1.1347 | | 2.767 | 84.0 | 84 | 1.1268 | | 2.767 | 85.0 | 85 | 1.1196 | | 2.767 | 86.0 | 86 | 1.1125 | | 2.767 | 87.0 | 87 | 1.1058 | | 2.767 | 88.0 | 88 | 1.0998 | | 2.767 | 89.0 | 89 | 1.0939 | | 2.767 | 90.0 | 90 | 1.0889 | | 2.767 | 91.0 | 91 | 1.0843 | | 2.767 | 92.0 | 92 | 1.0799 | | 2.767 | 93.0 | 93 | 1.0766 | | 2.767 | 94.0 | 94 | 1.0734 | | 2.767 | 95.0 | 95 | 1.0707 | | 2.767 | 96.0 | 96 | 1.0682 | | 2.767 | 97.0 | 97 | 1.0669 | | 2.767 | 98.0 | 98 | 1.0655 | | 2.767 | 99.0 | 99 | 1.0646 | | 1.7143 | 100.0 | 100 | 1.0641 | ### Framework versions - PEFT 0.11.1 - Transformers 4.41.2 - Pytorch 2.1.2 - Datasets 2.19.2 - Tokenizers 0.19.1