|
--- |
|
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: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# 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: 0.2987 |
|
|
|
## 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: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 0.2166 | 2.23 | 10000 | 0.2847 | |
|
| 0.2222 | 2.24 | 10050 | 0.2844 | |
|
| 0.2205 | 2.26 | 10100 | 0.2843 | |
|
| 0.2148 | 2.27 | 10150 | 0.2841 | |
|
| 0.211 | 2.28 | 10200 | 0.2851 | |
|
| 0.2213 | 2.29 | 10250 | 0.2839 | |
|
| 0.2035 | 2.3 | 10300 | 0.2835 | |
|
| 0.2268 | 2.31 | 10350 | 0.2843 | |
|
| 0.1978 | 2.32 | 10400 | 0.2869 | |
|
| 0.2085 | 2.33 | 10450 | 0.2858 | |
|
| 0.212 | 2.35 | 10500 | 0.2846 | |
|
| 0.2086 | 2.36 | 10550 | 0.2843 | |
|
| 0.2031 | 2.37 | 10600 | 0.2849 | |
|
| 0.2139 | 2.38 | 10650 | 0.2826 | |
|
| 0.182 | 2.39 | 10700 | 0.2832 | |
|
| 0.1991 | 2.4 | 10750 | 0.2826 | |
|
| 0.1905 | 2.41 | 10800 | 0.2869 | |
|
| 0.2095 | 2.42 | 10850 | 0.2818 | |
|
| 0.2211 | 2.43 | 10900 | 0.2800 | |
|
| 0.2235 | 2.45 | 10950 | 0.2811 | |
|
| 0.2376 | 2.46 | 11000 | 0.2820 | |
|
| 0.2517 | 2.47 | 11050 | 0.2824 | |
|
| 0.2099 | 2.48 | 11100 | 0.2780 | |
|
| 0.2106 | 2.49 | 11150 | 0.2800 | |
|
| 0.2222 | 2.5 | 11200 | 0.2781 | |
|
| 0.238 | 2.51 | 11250 | 0.2772 | |
|
| 0.2042 | 2.52 | 11300 | 0.2774 | |
|
| 0.2312 | 2.54 | 11350 | 0.2805 | |
|
| 0.2305 | 2.55 | 11400 | 0.2773 | |
|
| 0.2078 | 2.56 | 11450 | 0.2752 | |
|
| 0.2035 | 2.57 | 11500 | 0.2778 | |
|
| 0.2203 | 2.58 | 11550 | 0.2772 | |
|
| 0.2192 | 2.59 | 11600 | 0.2775 | |
|
| 0.2299 | 2.6 | 11650 | 0.2762 | |
|
| 0.2198 | 2.61 | 11700 | 0.2767 | |
|
| 0.1911 | 2.62 | 11750 | 0.2804 | |
|
| 0.1987 | 2.64 | 11800 | 0.2771 | |
|
| 0.2159 | 2.65 | 11850 | 0.2764 | |
|
| 0.2234 | 2.66 | 11900 | 0.2756 | |
|
| 0.2055 | 2.67 | 11950 | 0.2748 | |
|
| 0.2071 | 2.68 | 12000 | 0.2759 | |
|
| 0.225 | 2.69 | 12050 | 0.2745 | |
|
| 0.2416 | 2.7 | 12100 | 0.2770 | |
|
| 0.1886 | 2.71 | 12150 | 0.2767 | |
|
| 0.2027 | 2.73 | 12200 | 0.2747 | |
|
| 0.1961 | 2.74 | 12250 | 0.2779 | |
|
| 0.2249 | 2.75 | 12300 | 0.2718 | |
|
| 0.219 | 2.76 | 12350 | 0.2729 | |
|
| 0.2249 | 2.77 | 12400 | 0.2713 | |
|
| 0.2029 | 2.78 | 12450 | 0.2722 | |
|
| 0.2062 | 2.79 | 12500 | 0.2740 | |
|
| 0.195 | 2.8 | 12550 | 0.2734 | |
|
| 0.2151 | 2.81 | 12600 | 0.2731 | |
|
| 0.209 | 2.83 | 12650 | 0.2696 | |
|
| 0.1948 | 2.84 | 12700 | 0.2713 | |
|
| 0.2222 | 2.85 | 12750 | 0.2685 | |
|
| 0.1905 | 2.86 | 12800 | 0.2719 | |
|
| 0.224 | 2.87 | 12850 | 0.2720 | |
|
| 0.1984 | 2.88 | 12900 | 0.2703 | |
|
| 0.2171 | 2.89 | 12950 | 0.2692 | |
|
| 0.2118 | 2.9 | 13000 | 0.2687 | |
|
| 0.1976 | 2.91 | 13050 | 0.2669 | |
|
| 0.2155 | 2.93 | 13100 | 0.2687 | |
|
| 0.1784 | 2.94 | 13150 | 0.2693 | |
|
| 0.2089 | 2.95 | 13200 | 0.2697 | |
|
| 0.1918 | 2.96 | 13250 | 0.2671 | |
|
| 0.196 | 2.97 | 13300 | 0.2705 | |
|
| 0.1874 | 2.98 | 13350 | 0.2696 | |
|
| 0.227 | 2.99 | 13400 | 0.2668 | |
|
| 0.197 | 3.0 | 13450 | 0.2779 | |
|
| 0.1421 | 3.02 | 13500 | 0.2857 | |
|
| 0.162 | 3.03 | 13550 | 0.2859 | |
|
| 0.139 | 3.04 | 13600 | 0.2891 | |
|
| 0.1418 | 3.05 | 13650 | 0.2852 | |
|
| 0.1477 | 3.06 | 13700 | 0.2878 | |
|
| 0.143 | 3.07 | 13750 | 0.2885 | |
|
| 0.148 | 3.08 | 13800 | 0.2847 | |
|
| 0.1433 | 3.09 | 13850 | 0.2874 | |
|
| 0.1513 | 3.1 | 13900 | 0.2860 | |
|
| 0.1405 | 3.12 | 13950 | 0.2857 | |
|
| 0.141 | 3.13 | 14000 | 0.2907 | |
|
| 0.1554 | 3.14 | 14050 | 0.2859 | |
|
| 0.1546 | 3.15 | 14100 | 0.2856 | |
|
| 0.1494 | 3.16 | 14150 | 0.2865 | |
|
| 0.1485 | 3.17 | 14200 | 0.2853 | |
|
| 0.1365 | 3.18 | 14250 | 0.2866 | |
|
| 0.1537 | 3.19 | 14300 | 0.2869 | |
|
| 0.1599 | 3.21 | 14350 | 0.2824 | |
|
| 0.147 | 3.22 | 14400 | 0.2847 | |
|
| 0.1576 | 3.23 | 14450 | 0.2826 | |
|
| 0.1439 | 3.24 | 14500 | 0.2830 | |
|
| 0.1463 | 3.25 | 14550 | 0.2810 | |
|
| 0.1471 | 3.26 | 14600 | 0.2853 | |
|
| 0.1708 | 3.27 | 14650 | 0.2809 | |
|
| 0.1555 | 3.28 | 14700 | 0.2821 | |
|
| 0.1563 | 3.29 | 14750 | 0.2816 | |
|
| 0.1498 | 3.31 | 14800 | 0.2820 | |
|
| 0.1464 | 3.32 | 14850 | 0.2835 | |
|
| 0.159 | 3.33 | 14900 | 0.2821 | |
|
| 0.1477 | 3.34 | 14950 | 0.2836 | |
|
| 0.1531 | 3.35 | 15000 | 0.2849 | |
|
| 0.1413 | 3.36 | 15050 | 0.2843 | |
|
| 0.1509 | 3.37 | 15100 | 0.2830 | |
|
| 0.1501 | 3.38 | 15150 | 0.2810 | |
|
| 0.146 | 3.4 | 15200 | 0.2799 | |
|
| 0.1567 | 3.41 | 15250 | 0.2819 | |
|
| 0.1503 | 3.42 | 15300 | 0.2825 | |
|
| 0.1688 | 3.43 | 15350 | 0.2829 | |
|
| 0.1483 | 3.44 | 15400 | 0.2835 | |
|
| 0.1446 | 3.45 | 15450 | 0.2844 | |
|
| 0.144 | 3.46 | 15500 | 0.2809 | |
|
| 0.1377 | 3.47 | 15550 | 0.2823 | |
|
| 0.1554 | 3.48 | 15600 | 0.2800 | |
|
| 0.1453 | 3.5 | 15650 | 0.2817 | |
|
| 0.1448 | 3.51 | 15700 | 0.2814 | |
|
| 0.1519 | 3.52 | 15750 | 0.2815 | |
|
| 0.1372 | 3.53 | 15800 | 0.2813 | |
|
| 0.1843 | 3.54 | 15850 | 0.2757 | |
|
| 0.1433 | 3.55 | 15900 | 0.2789 | |
|
| 0.1664 | 3.56 | 15950 | 0.2794 | |
|
| 0.1495 | 3.57 | 16000 | 0.2779 | |
|
| 0.1548 | 3.58 | 16050 | 0.2781 | |
|
| 0.1459 | 3.6 | 16100 | 0.2798 | |
|
| 0.1476 | 3.61 | 16150 | 0.2798 | |
|
| 0.1509 | 3.62 | 16200 | 0.2784 | |
|
| 0.1368 | 3.63 | 16250 | 0.2814 | |
|
| 0.1386 | 3.64 | 16300 | 0.2788 | |
|
| 0.1463 | 3.65 | 16350 | 0.2779 | |
|
| 0.1427 | 3.66 | 16400 | 0.2769 | |
|
| 0.1444 | 3.67 | 16450 | 0.2808 | |
|
| 0.1401 | 3.69 | 16500 | 0.2754 | |
|
| 0.168 | 3.7 | 16550 | 0.2770 | |
|
| 0.158 | 3.71 | 16600 | 0.2774 | |
|
| 0.1661 | 3.72 | 16650 | 0.2791 | |
|
| 0.1528 | 3.73 | 16700 | 0.2780 | |
|
| 0.1616 | 3.74 | 16750 | 0.2758 | |
|
| 0.1591 | 3.75 | 16800 | 0.2748 | |
|
| 0.1483 | 3.76 | 16850 | 0.2742 | |
|
| 0.154 | 3.77 | 16900 | 0.2748 | |
|
| 0.1545 | 3.79 | 16950 | 0.2747 | |
|
| 0.1418 | 3.8 | 17000 | 0.2772 | |
|
| 0.1301 | 3.81 | 17050 | 0.2781 | |
|
| 0.1577 | 3.82 | 17100 | 0.2765 | |
|
| 0.1553 | 3.83 | 17150 | 0.2747 | |
|
| 0.159 | 3.84 | 17200 | 0.2752 | |
|
| 0.1477 | 3.85 | 17250 | 0.2766 | |
|
| 0.1458 | 3.86 | 17300 | 0.2746 | |
|
| 0.1531 | 3.88 | 17350 | 0.2762 | |
|
| 0.1461 | 3.89 | 17400 | 0.2738 | |
|
| 0.1417 | 3.9 | 17450 | 0.2763 | |
|
| 0.1471 | 3.91 | 17500 | 0.2753 | |
|
| 0.1445 | 3.92 | 17550 | 0.2736 | |
|
| 0.1505 | 3.93 | 17600 | 0.2738 | |
|
| 0.1447 | 3.94 | 17650 | 0.2725 | |
|
| 0.146 | 3.95 | 17700 | 0.2745 | |
|
| 0.138 | 3.96 | 17750 | 0.2741 | |
|
| 0.1514 | 3.98 | 17800 | 0.2723 | |
|
| 0.1469 | 3.99 | 17850 | 0.2738 | |
|
| 0.1344 | 4.0 | 17900 | 0.2752 | |
|
| 0.1128 | 4.01 | 17950 | 0.2935 | |
|
| 0.1037 | 4.02 | 18000 | 0.2976 | |
|
| 0.0909 | 4.03 | 18050 | 0.2982 | |
|
| 0.0912 | 4.04 | 18100 | 0.2959 | |
|
| 0.1141 | 4.05 | 18150 | 0.2938 | |
|
| 0.1047 | 4.07 | 18200 | 0.2974 | |
|
| 0.096 | 4.08 | 18250 | 0.2974 | |
|
| 0.1128 | 4.09 | 18300 | 0.2952 | |
|
| 0.1147 | 4.1 | 18350 | 0.2954 | |
|
| 0.1081 | 4.11 | 18400 | 0.2960 | |
|
| 0.1058 | 4.12 | 18450 | 0.2943 | |
|
| 0.1068 | 4.13 | 18500 | 0.2966 | |
|
| 0.0939 | 4.14 | 18550 | 0.2999 | |
|
| 0.0948 | 4.15 | 18600 | 0.2977 | |
|
| 0.0935 | 4.17 | 18650 | 0.2992 | |
|
| 0.11 | 4.18 | 18700 | 0.2968 | |
|
| 0.1039 | 4.19 | 18750 | 0.2972 | |
|
| 0.0915 | 4.2 | 18800 | 0.3043 | |
|
| 0.0932 | 4.21 | 18850 | 0.2985 | |
|
| 0.0896 | 4.22 | 18900 | 0.2995 | |
|
| 0.097 | 4.23 | 18950 | 0.2987 | |
|
| 0.0965 | 4.24 | 19000 | 0.2943 | |
|
| 0.1011 | 4.26 | 19050 | 0.2948 | |
|
| 0.1019 | 4.27 | 19100 | 0.2969 | |
|
| 0.1037 | 4.28 | 19150 | 0.2986 | |
|
| 0.1046 | 4.29 | 19200 | 0.2950 | |
|
| 0.1004 | 4.3 | 19250 | 0.2954 | |
|
| 0.0998 | 4.31 | 19300 | 0.2999 | |
|
| 0.0969 | 4.32 | 19350 | 0.2972 | |
|
| 0.0925 | 4.33 | 19400 | 0.2990 | |
|
| 0.0964 | 4.34 | 19450 | 0.3001 | |
|
| 0.098 | 4.36 | 19500 | 0.2993 | |
|
| 0.0915 | 4.37 | 19550 | 0.3003 | |
|
| 0.089 | 4.38 | 19600 | 0.2993 | |
|
| 0.0959 | 4.39 | 19650 | 0.2969 | |
|
| 0.0975 | 4.4 | 19700 | 0.2967 | |
|
| 0.0939 | 4.41 | 19750 | 0.2979 | |
|
| 0.0993 | 4.42 | 19800 | 0.2976 | |
|
| 0.0889 | 4.43 | 19850 | 0.2986 | |
|
| 0.0998 | 4.44 | 19900 | 0.3001 | |
|
| 0.0996 | 4.46 | 19950 | 0.2985 | |
|
| 0.1021 | 4.47 | 20000 | 0.3000 | |
|
| 0.1012 | 4.48 | 20050 | 0.2991 | |
|
| 0.0981 | 4.49 | 20100 | 0.2992 | |
|
| 0.1031 | 4.5 | 20150 | 0.2994 | |
|
| 0.0952 | 4.51 | 20200 | 0.3004 | |
|
| 0.1021 | 4.52 | 20250 | 0.2980 | |
|
| 0.0965 | 4.53 | 20300 | 0.2991 | |
|
| 0.0926 | 4.55 | 20350 | 0.2986 | |
|
| 0.0921 | 4.56 | 20400 | 0.2996 | |
|
| 0.0922 | 4.57 | 20450 | 0.2996 | |
|
| 0.0961 | 4.58 | 20500 | 0.2998 | |
|
| 0.0929 | 4.59 | 20550 | 0.3013 | |
|
| 0.1007 | 4.6 | 20600 | 0.2985 | |
|
| 0.0957 | 4.61 | 20650 | 0.2989 | |
|
| 0.0955 | 4.62 | 20700 | 0.2996 | |
|
| 0.1003 | 4.63 | 20750 | 0.3003 | |
|
| 0.09 | 4.65 | 20800 | 0.3001 | |
|
| 0.0975 | 4.66 | 20850 | 0.3000 | |
|
| 0.0976 | 4.67 | 20900 | 0.2987 | |
|
| 0.0911 | 4.68 | 20950 | 0.2982 | |
|
| 0.0939 | 4.69 | 21000 | 0.2991 | |
|
| 0.0956 | 4.7 | 21050 | 0.2988 | |
|
| 0.1091 | 4.71 | 21100 | 0.2971 | |
|
| 0.095 | 4.72 | 21150 | 0.2962 | |
|
| 0.0898 | 4.74 | 21200 | 0.2960 | |
|
| 0.0898 | 4.75 | 21250 | 0.2976 | |
|
| 0.0915 | 4.76 | 21300 | 0.2991 | |
|
| 0.0967 | 4.77 | 21350 | 0.2977 | |
|
| 0.0929 | 4.78 | 21400 | 0.2982 | |
|
| 0.0928 | 4.79 | 21450 | 0.2975 | |
|
| 0.0865 | 4.8 | 21500 | 0.2989 | |
|
| 0.0988 | 4.81 | 21550 | 0.2988 | |
|
| 0.0871 | 4.82 | 21600 | 0.2993 | |
|
| 0.0996 | 4.84 | 21650 | 0.2987 | |
|
| 0.0914 | 4.85 | 21700 | 0.2988 | |
|
| 0.0818 | 4.86 | 21750 | 0.2986 | |
|
| 0.0909 | 4.87 | 21800 | 0.2992 | |
|
| 0.0879 | 4.88 | 21850 | 0.2993 | |
|
| 0.0879 | 4.89 | 21900 | 0.2996 | |
|
| 0.09 | 4.9 | 21950 | 0.2993 | |
|
| 0.095 | 4.91 | 22000 | 0.2989 | |
|
| 0.0845 | 4.93 | 22050 | 0.2991 | |
|
| 0.0974 | 4.94 | 22100 | 0.2992 | |
|
| 0.0991 | 4.95 | 22150 | 0.2991 | |
|
| 0.0902 | 4.96 | 22200 | 0.2987 | |
|
| 0.0881 | 4.97 | 22250 | 0.2987 | |
|
| 0.0989 | 4.98 | 22300 | 0.2987 | |
|
| 0.093 | 4.99 | 22350 | 0.2987 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.10.0 |
|
- Transformers 4.39.2 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
