--- license: llama2 library_name: peft tags: - trl - sft - generated_from_trainer base_model: codellama/CodeLlama-7b-Instruct-hf model-index: - name: Codellama-7b-lora-rps-adapter results: [] --- # Codellama-7b-lora-rps-adapter This model is a fine-tuned version of [codellama/CodeLlama-7b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.2781 ## 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: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.03 - num_epochs: 4 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:-----:|:---------------:| | 0.2411 | 1.79 | 8000 | 0.2641 | | 0.2531 | 1.8 | 8050 | 0.2630 | | 0.2648 | 1.81 | 8100 | 0.2649 | | 0.243 | 1.82 | 8150 | 0.2639 | | 0.236 | 1.83 | 8200 | 0.2633 | | 0.2836 | 1.84 | 8250 | 0.2635 | | 0.2512 | 1.85 | 8300 | 0.2616 | | 0.2416 | 1.87 | 8350 | 0.2609 | | 0.2565 | 1.88 | 8400 | 0.2608 | | 0.2646 | 1.89 | 8450 | 0.2612 | | 0.2292 | 1.9 | 8500 | 0.2614 | | 0.2697 | 1.91 | 8550 | 0.2620 | | 0.2509 | 1.92 | 8600 | 0.2607 | | 0.2541 | 1.93 | 8650 | 0.2588 | | 0.261 | 1.94 | 8700 | 0.2585 | | 0.2653 | 1.95 | 8750 | 0.2565 | | 0.2161 | 1.97 | 8800 | 0.2574 | | 0.2283 | 1.98 | 8850 | 0.2568 | | 0.2355 | 1.99 | 8900 | 0.2571 | | 0.2255 | 2.0 | 8950 | 0.2564 | | 0.1783 | 2.01 | 9000 | 0.2682 | | 0.1631 | 2.02 | 9050 | 0.2701 | | 0.1741 | 2.03 | 9100 | 0.2702 | | 0.1785 | 2.04 | 9150 | 0.2695 | | 0.1796 | 2.05 | 9200 | 0.2682 | | 0.1858 | 2.07 | 9250 | 0.2735 | | 0.197 | 2.08 | 9300 | 0.2744 | | 0.1838 | 2.09 | 9350 | 0.2704 | | 0.1812 | 2.1 | 9400 | 0.2701 | | 0.1771 | 2.11 | 9450 | 0.2687 | | 0.1877 | 2.12 | 9500 | 0.2690 | | 0.1713 | 2.13 | 9550 | 0.2709 | | 0.2012 | 2.14 | 9600 | 0.2696 | | 0.1886 | 2.16 | 9650 | 0.2668 | | 0.1803 | 2.17 | 9700 | 0.2695 | | 0.1736 | 2.18 | 9750 | 0.2691 | | 0.172 | 2.19 | 9800 | 0.2699 | | 0.1847 | 2.2 | 9850 | 0.2713 | | 0.1813 | 2.21 | 9900 | 0.2675 | | 0.162 | 2.22 | 9950 | 0.2681 | | 0.1759 | 2.23 | 10000 | 0.2688 | | 0.1785 | 2.24 | 10050 | 0.2675 | | 0.1794 | 2.26 | 10100 | 0.2690 | | 0.1724 | 2.27 | 10150 | 0.2687 | | 0.179 | 2.28 | 10200 | 0.2674 | | 0.1839 | 2.29 | 10250 | 0.2646 | | 0.1654 | 2.3 | 10300 | 0.2689 | | 0.1845 | 2.31 | 10350 | 0.2671 | | 0.1632 | 2.32 | 10400 | 0.2693 | | 0.1679 | 2.33 | 10450 | 0.2702 | | 0.1676 | 2.35 | 10500 | 0.2680 | | 0.1747 | 2.36 | 10550 | 0.2698 | | 0.1702 | 2.37 | 10600 | 0.2656 | | 0.1706 | 2.38 | 10650 | 0.2678 | | 0.1535 | 2.39 | 10700 | 0.2666 | | 0.162 | 2.4 | 10750 | 0.2640 | | 0.1557 | 2.41 | 10800 | 0.2664 | | 0.1729 | 2.42 | 10850 | 0.2658 | | 0.1778 | 2.43 | 10900 | 0.2672 | | 0.1815 | 2.45 | 10950 | 0.2651 | | 0.1898 | 2.46 | 11000 | 0.2637 | | 0.2043 | 2.47 | 11050 | 0.2636 | | 0.171 | 2.48 | 11100 | 0.2647 | | 0.1747 | 2.49 | 11150 | 0.2619 | | 0.1767 | 2.5 | 11200 | 0.2615 | | 0.192 | 2.51 | 11250 | 0.2626 | | 0.1636 | 2.52 | 11300 | 0.2638 | | 0.1823 | 2.54 | 11350 | 0.2649 | | 0.1913 | 2.55 | 11400 | 0.2608 | | 0.1719 | 2.56 | 11450 | 0.2628 | | 0.1721 | 2.57 | 11500 | 0.2624 | | 0.1721 | 2.58 | 11550 | 0.2638 | | 0.1788 | 2.59 | 11600 | 0.2617 | | 0.1837 | 2.6 | 11650 | 0.2615 | | 0.1857 | 2.61 | 11700 | 0.2606 | | 0.158 | 2.62 | 11750 | 0.2640 | | 0.1593 | 2.64 | 11800 | 0.2612 | | 0.1738 | 2.65 | 11850 | 0.2606 | | 0.1767 | 2.66 | 11900 | 0.2604 | | 0.1685 | 2.67 | 11950 | 0.2612 | | 0.1724 | 2.68 | 12000 | 0.2596 | | 0.1889 | 2.69 | 12050 | 0.2580 | | 0.1967 | 2.7 | 12100 | 0.2607 | | 0.1557 | 2.71 | 12150 | 0.2604 | | 0.1643 | 2.73 | 12200 | 0.2593 | | 0.1618 | 2.74 | 12250 | 0.2606 | | 0.1847 | 2.75 | 12300 | 0.2573 | | 0.1761 | 2.76 | 12350 | 0.2584 | | 0.1802 | 2.77 | 12400 | 0.2578 | | 0.1651 | 2.78 | 12450 | 0.2582 | | 0.1698 | 2.79 | 12500 | 0.2579 | | 0.1621 | 2.8 | 12550 | 0.2570 | | 0.1768 | 2.81 | 12600 | 0.2582 | | 0.1629 | 2.83 | 12650 | 0.2596 | | 0.1592 | 2.84 | 12700 | 0.2592 | | 0.179 | 2.85 | 12750 | 0.2574 | | 0.1539 | 2.86 | 12800 | 0.2577 | | 0.1752 | 2.87 | 12850 | 0.2590 | | 0.1615 | 2.88 | 12900 | 0.2570 | | 0.1711 | 2.89 | 12950 | 0.2579 | | 0.1718 | 2.9 | 13000 | 0.2570 | | 0.1626 | 2.91 | 13050 | 0.2570 | | 0.1595 | 2.93 | 13100 | 0.2583 | | 0.1537 | 2.94 | 13150 | 0.2568 | | 0.164 | 2.95 | 13200 | 0.2571 | | 0.1591 | 2.96 | 13250 | 0.2562 | | 0.1661 | 2.97 | 13300 | 0.2575 | | 0.16 | 2.98 | 13350 | 0.2570 | | 0.1803 | 2.99 | 13400 | 0.2568 | | 0.16 | 3.0 | 13450 | 0.2644 | | 0.1143 | 3.02 | 13500 | 0.2766 | | 0.1218 | 3.03 | 13550 | 0.2799 | | 0.1106 | 3.04 | 13600 | 0.2765 | | 0.1174 | 3.05 | 13650 | 0.2776 | | 0.1167 | 3.06 | 13700 | 0.2783 | | 0.1175 | 3.07 | 13750 | 0.2834 | | 0.1165 | 3.08 | 13800 | 0.2797 | | 0.1117 | 3.09 | 13850 | 0.2810 | | 0.1178 | 3.1 | 13900 | 0.2821 | | 0.1089 | 3.12 | 13950 | 0.2784 | | 0.1108 | 3.13 | 14000 | 0.2824 | | 0.1174 | 3.14 | 14050 | 0.2820 | | 0.1202 | 3.15 | 14100 | 0.2808 | | 0.1198 | 3.16 | 14150 | 0.2817 | | 0.1178 | 3.17 | 14200 | 0.2799 | | 0.1047 | 3.18 | 14250 | 0.2802 | | 0.1159 | 3.19 | 14300 | 0.2815 | | 0.1263 | 3.21 | 14350 | 0.2785 | | 0.1148 | 3.22 | 14400 | 0.2792 | | 0.1242 | 3.23 | 14450 | 0.2779 | | 0.1148 | 3.24 | 14500 | 0.2775 | | 0.1178 | 3.25 | 14550 | 0.2775 | | 0.1189 | 3.26 | 14600 | 0.2789 | | 0.1251 | 3.27 | 14650 | 0.2783 | | 0.1177 | 3.28 | 14700 | 0.2802 | | 0.1195 | 3.29 | 14750 | 0.2792 | | 0.1191 | 3.31 | 14800 | 0.2787 | | 0.1194 | 3.32 | 14850 | 0.2776 | | 0.1239 | 3.33 | 14900 | 0.2800 | | 0.1124 | 3.34 | 14950 | 0.2806 | | 0.1132 | 3.35 | 15000 | 0.2789 | | 0.1124 | 3.36 | 15050 | 0.2815 | | 0.1155 | 3.37 | 15100 | 0.2781 | | 0.1124 | 3.38 | 15150 | 0.2805 | | 0.1149 | 3.4 | 15200 | 0.2787 | | 0.1236 | 3.41 | 15250 | 0.2796 | | 0.1151 | 3.42 | 15300 | 0.2795 | | 0.1355 | 3.43 | 15350 | 0.2794 | | 0.1142 | 3.44 | 15400 | 0.2779 | | 0.112 | 3.45 | 15450 | 0.2798 | | 0.1124 | 3.46 | 15500 | 0.2805 | | 0.1117 | 3.47 | 15550 | 0.2793 | | 0.1195 | 3.48 | 15600 | 0.2788 | | 0.1078 | 3.5 | 15650 | 0.2817 | | 0.1085 | 3.51 | 15700 | 0.2802 | | 0.1137 | 3.52 | 15750 | 0.2808 | | 0.1094 | 3.53 | 15800 | 0.2803 | | 0.139 | 3.54 | 15850 | 0.2773 | | 0.107 | 3.55 | 15900 | 0.2766 | | 0.1161 | 3.56 | 15950 | 0.2781 | | 0.1202 | 3.57 | 16000 | 0.2777 | | 0.1132 | 3.58 | 16050 | 0.2783 | | 0.113 | 3.6 | 16100 | 0.2776 | | 0.1109 | 3.61 | 16150 | 0.2790 | | 0.1125 | 3.62 | 16200 | 0.2783 | | 0.1096 | 3.63 | 16250 | 0.2784 | | 0.1093 | 3.64 | 16300 | 0.2774 | | 0.1082 | 3.65 | 16350 | 0.2768 | | 0.1204 | 3.66 | 16400 | 0.2764 | | 0.1059 | 3.67 | 16450 | 0.2783 | | 0.1072 | 3.69 | 16500 | 0.2775 | | 0.1248 | 3.7 | 16550 | 0.2771 | | 0.1171 | 3.71 | 16600 | 0.2766 | | 0.1297 | 3.72 | 16650 | 0.2767 | | 0.118 | 3.73 | 16700 | 0.2770 | | 0.1217 | 3.74 | 16750 | 0.2764 | | 0.1208 | 3.75 | 16800 | 0.2781 | | 0.1117 | 3.76 | 16850 | 0.2775 | | 0.1098 | 3.77 | 16900 | 0.2789 | | 0.1124 | 3.79 | 16950 | 0.2804 | | 0.1065 | 3.8 | 17000 | 0.2799 | | 0.1041 | 3.81 | 17050 | 0.2786 | | 0.1112 | 3.82 | 17100 | 0.2776 | | 0.1086 | 3.83 | 17150 | 0.2775 | | 0.1229 | 3.84 | 17200 | 0.2777 | | 0.1099 | 3.85 | 17250 | 0.2778 | | 0.1121 | 3.86 | 17300 | 0.2780 | | 0.1175 | 3.88 | 17350 | 0.2784 | | 0.1131 | 3.89 | 17400 | 0.2780 | | 0.1031 | 3.9 | 17450 | 0.2781 | | 0.1123 | 3.91 | 17500 | 0.2782 | | 0.1113 | 3.92 | 17550 | 0.2783 | | 0.1126 | 3.93 | 17600 | 0.2781 | | 0.1068 | 3.94 | 17650 | 0.2779 | | 0.1095 | 3.95 | 17700 | 0.2782 | | 0.1058 | 3.96 | 17750 | 0.2782 | | 0.1105 | 3.98 | 17800 | 0.2781 | | 0.1108 | 3.99 | 17850 | 0.2781 | | 0.1071 | 4.0 | 17900 | 0.2781 | ### Framework versions - PEFT 0.10.0 - Transformers 4.39.3 - Pytorch 2.2.1+cu121 - Datasets 2.18.0 - Tokenizers 0.15.2