--- license: apache-2.0 base_model: mistralai/Mistral-7B-v0.1 tags: - generated_from_trainer model-index: - name: Mistral_Sparse_refined_web_70p_2024-02-28 results: [] --- # Mistral_Sparse_refined_web_70p_2024-02-28 This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset. It achieves the following results on the evaluation set: - Loss: 2.1297 ## 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: 1e-05 - train_batch_size: 1 - eval_batch_size: 1 - seed: 0 - distributed_type: multi-GPU - num_devices: 4 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - total_eval_batch_size: 4 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - training_steps: 3750 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | 2.7195 | 0.0 | 25 | 2.8171 | | 2.4252 | 0.01 | 50 | 2.6962 | | 2.4132 | 0.01 | 75 | 2.6180 | | 2.3614 | 0.02 | 100 | 2.5748 | | 2.3289 | 0.02 | 125 | 2.5433 | | 2.4043 | 0.02 | 150 | 2.5219 | | 2.3267 | 0.03 | 175 | 2.5078 | | 2.3059 | 0.03 | 200 | 2.4980 | | 2.3307 | 0.04 | 225 | 2.4895 | | 2.3628 | 0.04 | 250 | 2.4807 | | 2.2833 | 0.04 | 275 | 2.4741 | | 2.294 | 0.05 | 300 | 2.4727 | | 2.3001 | 0.05 | 325 | 2.4657 | | 2.1377 | 0.06 | 350 | 2.4617 | | 2.2947 | 0.06 | 375 | 2.4605 | | 2.3341 | 0.06 | 400 | 2.4532 | | 2.3318 | 0.07 | 425 | 2.4520 | | 2.2793 | 0.07 | 450 | 2.4498 | | 2.2009 | 0.08 | 475 | 2.4445 | | 2.2665 | 0.08 | 500 | 2.4443 | | 2.3313 | 0.08 | 525 | 2.4395 | | 2.2272 | 0.09 | 550 | 2.4414 | | 2.3625 | 0.09 | 575 | 2.4408 | | 2.1607 | 0.1 | 600 | 2.4390 | | 2.2494 | 0.1 | 625 | 2.4408 | | 2.3305 | 0.1 | 650 | 2.4363 | | 2.32 | 0.11 | 675 | 2.4367 | | 2.2791 | 0.11 | 700 | 2.4322 | | 2.2702 | 0.12 | 725 | 2.4273 | | 2.1792 | 0.12 | 750 | 2.4286 | | 2.2145 | 0.12 | 775 | 2.4304 | | 2.3076 | 0.13 | 800 | 2.4273 | | 2.3154 | 0.13 | 825 | 2.4287 | | 2.2979 | 0.14 | 850 | 2.4295 | | 2.1638 | 0.14 | 875 | 2.4261 | | 2.2316 | 0.14 | 900 | 2.4287 | | 2.1786 | 0.15 | 925 | 2.4270 | | 2.3116 | 0.15 | 950 | 2.4246 | | 2.2439 | 0.16 | 975 | 2.4232 | | 2.2867 | 0.16 | 1000 | 2.4209 | | 2.291 | 0.16 | 1025 | 2.4231 | | 2.3092 | 0.17 | 1050 | 2.4261 | | 2.3367 | 0.17 | 1075 | 2.4226 | | 2.2603 | 0.18 | 1100 | 2.4185 | | 2.2899 | 0.18 | 1125 | 2.4241 | | 2.2575 | 0.18 | 1150 | 2.4241 | | 2.3044 | 0.19 | 1175 | 2.4180 | | 2.1659 | 0.19 | 1200 | 2.4153 | | 2.2555 | 0.2 | 1225 | 2.4175 | | 2.2407 | 0.2 | 1250 | 2.4174 | | 2.2809 | 0.2 | 1275 | 2.4172 | | 2.3096 | 0.21 | 1300 | 2.4159 | | 2.2855 | 0.21 | 1325 | 2.4168 | | 2.1743 | 0.22 | 1350 | 2.4171 | | 2.2595 | 0.22 | 1375 | 2.4145 | | 2.2657 | 0.22 | 1400 | 2.4205 | | 2.2053 | 0.23 | 1425 | 2.4166 | | 2.1895 | 0.23 | 1450 | 2.4209 | | 2.1552 | 0.24 | 1475 | 2.4197 | | 2.2134 | 0.24 | 1500 | 2.4188 | | 2.1482 | 0.24 | 1525 | 2.4186 | | 2.2142 | 0.25 | 1550 | 2.4222 | | 2.3327 | 0.25 | 1575 | 2.4174 | | 2.2899 | 0.26 | 1600 | 2.4167 | | 2.2577 | 0.26 | 1625 | 2.4172 | | 2.3402 | 0.26 | 1650 | 2.4211 | | 2.2345 | 0.27 | 1675 | 2.4183 | | 2.2808 | 0.27 | 1700 | 2.4164 | | 2.2678 | 0.28 | 1725 | 2.4181 | | 2.2512 | 0.28 | 1750 | 2.4161 | | 2.231 | 0.28 | 1775 | 2.4152 | | 2.1873 | 0.29 | 1800 | 2.4146 | | 2.2662 | 0.29 | 1825 | 2.4135 | | 2.262 | 0.3 | 1850 | 2.4188 | | 2.2492 | 0.3 | 1875 | 2.4194 | | 2.274 | 0.3 | 1900 | 2.4138 | | 2.3257 | 0.31 | 1925 | 2.4152 | | 2.2678 | 0.31 | 1950 | 2.4122 | | 2.2877 | 0.32 | 1975 | 2.4153 | | 2.2117 | 0.32 | 2000 | 2.4151 | | 2.2312 | 0.32 | 2025 | 2.4163 | | 2.2182 | 0.33 | 2050 | 2.4159 | | 2.2126 | 0.33 | 2075 | 2.4121 | | 2.3101 | 0.34 | 2100 | 2.4180 | | 2.3467 | 0.34 | 2125 | 2.4173 | | 2.2572 | 0.34 | 2150 | 2.4132 | | 2.2068 | 0.35 | 2175 | 2.4127 | | 2.2434 | 0.35 | 2200 | 2.4138 | | 2.1937 | 0.36 | 2225 | 2.4165 | | 2.1858 | 0.36 | 2250 | 2.4143 | | 2.3188 | 0.36 | 2275 | 2.4138 | | 2.2943 | 0.37 | 2300 | 2.4120 | | 2.1959 | 0.37 | 2325 | 2.4095 | | 2.2399 | 0.38 | 2350 | 2.4080 | | 2.2694 | 0.38 | 2375 | 2.4090 | | 2.2795 | 0.38 | 2400 | 2.4066 | | 2.1325 | 0.39 | 2425 | 2.4071 | | 2.1264 | 0.39 | 2450 | 2.4077 | | 2.2166 | 0.4 | 2475 | 2.4082 | | 2.2965 | 0.4 | 2500 | 2.4113 | | 2.3058 | 0.4 | 2525 | 2.4066 | | 2.1596 | 0.41 | 2550 | 2.4134 | | 2.1838 | 0.41 | 2575 | 2.4110 | | 2.2609 | 0.42 | 2600 | 2.4074 | | 2.2689 | 0.42 | 2625 | 2.4092 | | 2.3092 | 0.42 | 2650 | 2.4101 | | 2.2789 | 0.43 | 2675 | 2.4096 | | 2.3045 | 0.43 | 2700 | 2.4111 | | 2.2974 | 0.44 | 2725 | 2.4134 | | 2.2473 | 0.44 | 2750 | 2.4113 | | 2.2701 | 0.44 | 2775 | 2.4087 | | 2.3412 | 0.45 | 2800 | 2.4115 | | 2.2427 | 0.45 | 2825 | 2.4100 | | 2.2928 | 0.46 | 2850 | 2.4067 | | 2.1879 | 0.46 | 2875 | 2.4104 | | 2.2206 | 0.46 | 2900 | 2.4088 | | 2.2282 | 0.47 | 2925 | 2.4065 | | 2.3002 | 0.47 | 2950 | 2.4075 | | 2.2273 | 0.48 | 2975 | 2.4045 | | 2.2871 | 0.48 | 3000 | 2.4064 | | 2.3817 | 0.48 | 3025 | 2.4054 | | 2.2496 | 0.49 | 3050 | 2.4069 | | 2.2985 | 0.49 | 3075 | 2.4083 | | 2.3125 | 0.5 | 3100 | 2.4041 | | 2.2139 | 0.5 | 3125 | 2.4064 | | 2.2997 | 0.5 | 3150 | 2.4062 | | 2.187 | 0.51 | 3175 | 2.4055 | | 2.2584 | 0.51 | 3200 | 2.4062 | | 2.1848 | 0.52 | 3225 | 2.4075 | | 2.1779 | 0.52 | 3250 | 2.4066 | | 2.2542 | 0.52 | 3275 | 2.4041 | | 2.2406 | 0.53 | 3300 | 2.4066 | | 2.1247 | 0.53 | 3325 | 2.4023 | | 2.2576 | 0.54 | 3350 | 2.4041 | | 2.1636 | 0.54 | 3375 | 2.4023 | | 2.1781 | 0.54 | 3400 | 2.4056 | | 2.1949 | 0.55 | 3425 | 2.4047 | | 2.1119 | 0.55 | 3450 | 2.4070 | | 2.2437 | 0.56 | 3475 | 2.4096 | | 2.281 | 0.56 | 3500 | 2.4040 | | 2.2499 | 0.56 | 3525 | 2.4063 | | 2.2129 | 0.57 | 3550 | 2.4052 | | 2.2115 | 0.57 | 3575 | 2.4050 | | 2.375 | 0.58 | 3600 | 2.4050 | | 2.1891 | 0.58 | 3625 | 2.4082 | | 2.3929 | 0.58 | 3650 | 2.4038 | | 2.1928 | 0.59 | 3675 | 2.4079 | | 2.3194 | 0.59 | 3700 | 2.4067 | | 2.2286 | 0.6 | 3725 | 2.4086 | | 2.1629 | 0.6 | 3750 | 2.4058 | ### Framework versions - Transformers 4.36.2 - Pytorch 2.1.2+cu121 - Datasets 2.15.0 - Tokenizers 0.15.0