--- license: apache-2.0 library_name: peft tags: - generated_from_trainer base_model: TheBloke/Marcoroni-7B-v3-GPTQ model-index: - name: 20231215-144935 results: [] --- # 20231215-144935 This model is a fine-tuned version of [TheBloke/Marcoroni-7B-v3-GPTQ](https://huggingface.co/TheBloke/Marcoroni-7B-v3-GPTQ) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.2792 ## 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: 2.5e-05 - train_batch_size: 2 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 2 - training_steps: 100 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | 2.1951 | 0.02 | 1 | 2.0427 | | 1.4965 | 0.03 | 2 | 2.0427 | | 2.2716 | 0.05 | 3 | 1.9661 | | 1.9397 | 0.06 | 4 | 1.9661 | | 1.8659 | 0.08 | 5 | 1.9661 | | 1.9903 | 0.09 | 6 | 1.8551 | | 1.9031 | 0.11 | 7 | 1.7846 | | 1.929 | 0.12 | 8 | 1.7253 | | 0.9994 | 0.14 | 9 | 1.6749 | | 1.4887 | 0.15 | 10 | 1.6266 | | 1.2112 | 0.17 | 11 | 1.5763 | | 1.3437 | 0.18 | 12 | 1.5326 | | 1.1061 | 0.2 | 13 | 1.5014 | | 1.4769 | 0.22 | 14 | 1.4676 | | 1.452 | 0.23 | 15 | 1.4313 | | 1.2393 | 0.25 | 16 | 1.3963 | | 1.1998 | 0.26 | 17 | 1.3662 | | 1.529 | 0.28 | 18 | 1.3353 | | 0.949 | 0.29 | 19 | 1.3088 | | 1.1857 | 0.31 | 20 | 1.2798 | | 1.4852 | 0.32 | 21 | 1.2484 | | 1.2845 | 0.34 | 22 | 1.2146 | | 1.332 | 0.35 | 23 | 1.1795 | | 0.8718 | 0.37 | 24 | 1.1435 | | 1.2038 | 0.38 | 25 | 1.1069 | | 1.2105 | 0.4 | 26 | 1.0683 | | 0.9161 | 0.42 | 27 | 1.0277 | | 1.0415 | 0.43 | 28 | 0.9849 | | 1.0367 | 0.45 | 29 | 0.9400 | | 0.9279 | 0.46 | 30 | 0.8924 | | 0.82 | 0.48 | 31 | 0.8467 | | 0.8286 | 0.49 | 32 | 0.8077 | | 0.7855 | 0.51 | 33 | 0.7645 | | 0.7919 | 0.52 | 34 | 0.7164 | | 0.6294 | 0.54 | 35 | 0.6712 | | 0.6205 | 0.55 | 36 | 0.6302 | | 0.5754 | 0.57 | 37 | 0.5926 | | 0.6331 | 0.58 | 38 | 0.5578 | | 0.5068 | 0.6 | 39 | 0.5251 | | 0.5252 | 0.62 | 40 | 0.5009 | | 0.442 | 0.63 | 41 | 0.4800 | | 0.4451 | 0.65 | 42 | 0.4628 | | 0.3742 | 0.66 | 43 | 0.4491 | | 0.8263 | 0.68 | 44 | 0.4383 | | 0.4392 | 0.69 | 45 | 0.4277 | | 0.5359 | 0.71 | 46 | 0.4174 | | 0.3903 | 0.72 | 47 | 0.4051 | | 0.4177 | 0.74 | 48 | 0.3921 | | 0.4606 | 0.75 | 49 | 0.3792 | | 0.5233 | 0.77 | 50 | 0.3700 | | 0.3838 | 0.78 | 51 | 0.3638 | | 0.353 | 0.8 | 52 | 0.3587 | | 0.3108 | 0.82 | 53 | 0.3530 | | 0.4029 | 0.83 | 54 | 0.3470 | | 0.3425 | 0.85 | 55 | 0.3411 | | 0.2976 | 0.86 | 56 | 0.3345 | | 0.5347 | 0.88 | 57 | 0.3291 | | 0.3208 | 0.89 | 58 | 0.3239 | | 0.3319 | 0.91 | 59 | 0.3194 | | 0.3333 | 0.92 | 60 | 0.3160 | | 0.3102 | 0.94 | 61 | 0.3135 | | 0.4149 | 0.95 | 62 | 0.3118 | | 0.3044 | 0.97 | 63 | 0.3098 | | 0.2688 | 0.98 | 64 | 0.3076 | | 0.3926 | 1.0 | 65 | 0.3060 | | 0.2969 | 1.02 | 66 | 0.3045 | | 0.3799 | 1.03 | 67 | 0.3029 | | 0.3045 | 1.05 | 68 | 0.3014 | | 0.2295 | 1.06 | 69 | 0.2997 | | 0.4113 | 1.08 | 70 | 0.2983 | | 0.3093 | 1.09 | 71 | 0.2969 | | 0.2542 | 1.11 | 72 | 0.2957 | | 0.2445 | 1.12 | 73 | 0.2947 | | 0.2592 | 1.14 | 74 | 0.2935 | | 0.6259 | 1.15 | 75 | 0.2925 | | 0.3217 | 1.17 | 76 | 0.2916 | | 0.2969 | 1.18 | 77 | 0.2907 | | 0.287 | 1.2 | 78 | 0.2898 | | 0.2642 | 1.22 | 79 | 0.2890 | | 0.2735 | 1.23 | 80 | 0.2880 | | 0.2472 | 1.25 | 81 | 0.2870 | | 0.3068 | 1.26 | 82 | 0.2862 | | 0.2779 | 1.28 | 83 | 0.2855 | | 0.2915 | 1.29 | 84 | 0.2848 | | 0.3926 | 1.31 | 85 | 0.2841 | | 0.3057 | 1.32 | 86 | 0.2836 | | 0.2413 | 1.34 | 87 | 0.2830 | | 0.2851 | 1.35 | 88 | 0.2826 | | 0.2295 | 1.37 | 89 | 0.2821 | | 0.2611 | 1.38 | 90 | 0.2817 | | 0.6346 | 1.4 | 91 | 0.2814 | | 0.2297 | 1.42 | 92 | 0.2811 | | 0.4868 | 1.43 | 93 | 0.2808 | | 0.2819 | 1.45 | 94 | 0.2805 | | 0.2589 | 1.46 | 95 | 0.2803 | | 0.2171 | 1.48 | 96 | 0.2801 | | 0.2581 | 1.49 | 97 | 0.2798 | | 0.2495 | 1.51 | 98 | 0.2796 | | 0.2702 | 1.52 | 99 | 0.2793 | | 0.271 | 1.54 | 100 | 0.2792 | ### Framework versions - PEFT 0.7.2.dev0 - Transformers 4.37.0.dev0 - Pytorch 2.1.0+cu121 - Datasets 2.15.0 - Tokenizers 0.15.0