Files
mistral-7b-indian-universit…/logs/training_log.txt
ModelHub XC 466afa54a8 初始化项目,由ModelHub XC社区提供模型
Model: millat/mistral-7b-indian-university-guidance
Source: Original Platform
2026-06-20 04:15:17 +08:00

277 lines
14 KiB
Plaintext

Training Log - Started at 2026-01-23T07:59:07.106112
================================================================================
Step 5 | Epoch 0.01 | Loss 1.9168 | GPU 4.8GB
Step 10 | Epoch 0.03 | Loss 1.0361 | GPU 4.8GB
Step 15 | Epoch 0.04 | Loss 0.7093 | GPU 4.8GB
Step 20 | Epoch 0.05 | Loss 0.6154 | GPU 4.8GB
Step 25 | Epoch 0.06 | Loss 0.5921 | GPU 4.8GB
Step 30 | Epoch 0.08 | Loss 0.5827 | GPU 4.8GB
Step 35 | Epoch 0.09 | Loss 0.5111 | GPU 4.8GB
Step 40 | Epoch 0.10 | Loss 0.5045 | GPU 4.8GB
Step 45 | Epoch 0.11 | Loss 0.5077 | GPU 4.8GB
Step 50 | Epoch 0.13 | Loss 0.5044 | GPU 4.8GB
Step 50 | Epoch 0.13 | Loss 0.0000 | Val 0.4850 | GPU 4.8GB
Step 55 | Epoch 0.14 | Loss 0.4785 | GPU 4.8GB
Step 60 | Epoch 0.15 | Loss 0.4746 | GPU 4.8GB
Step 65 | Epoch 0.16 | Loss 0.4666 | GPU 4.8GB
Step 70 | Epoch 0.18 | Loss 0.4631 | GPU 4.8GB
Step 75 | Epoch 0.19 | Loss 0.4890 | GPU 4.8GB
Step 80 | Epoch 0.20 | Loss 0.4868 | GPU 4.8GB
Step 85 | Epoch 0.21 | Loss 0.4278 | GPU 4.8GB
Step 90 | Epoch 0.23 | Loss 0.4620 | GPU 4.8GB
Step 95 | Epoch 0.24 | Loss 0.4813 | GPU 4.8GB
Step 100 | Epoch 0.25 | Loss 0.4653 | GPU 4.8GB
Step 100 | Epoch 0.25 | Loss 0.0000 | Val 0.4488 | GPU 4.8GB
Step 105 | Epoch 0.27 | Loss 0.4807 | GPU 4.8GB
Step 110 | Epoch 0.28 | Loss 0.4357 | GPU 4.8GB
Step 115 | Epoch 0.29 | Loss 0.4286 | GPU 4.8GB
Step 120 | Epoch 0.30 | Loss 0.4678 | GPU 4.8GB
Step 125 | Epoch 0.32 | Loss 0.4372 | GPU 4.8GB
Step 130 | Epoch 0.33 | Loss 0.4309 | GPU 4.8GB
Step 135 | Epoch 0.34 | Loss 0.4248 | GPU 4.8GB
Step 140 | Epoch 0.35 | Loss 0.4486 | GPU 4.8GB
Step 145 | Epoch 0.37 | Loss 0.4346 | GPU 4.8GB
Step 150 | Epoch 0.38 | Loss 0.4279 | GPU 4.8GB
Step 150 | Epoch 0.38 | Loss 0.0000 | Val 0.4310 | GPU 4.8GB
Step 155 | Epoch 0.39 | Loss 0.4378 | GPU 4.8GB
Step 160 | Epoch 0.40 | Loss 0.4076 | GPU 4.8GB
Step 165 | Epoch 0.42 | Loss 0.4501 | GPU 4.8GB
Step 170 | Epoch 0.43 | Loss 0.4491 | GPU 4.8GB
Step 175 | Epoch 0.44 | Loss 0.3968 | GPU 4.8GB
Step 180 | Epoch 0.45 | Loss 0.4237 | GPU 4.8GB
Step 185 | Epoch 0.47 | Loss 0.4219 | GPU 4.8GB
Step 190 | Epoch 0.48 | Loss 0.4165 | GPU 4.8GB
Step 195 | Epoch 0.49 | Loss 0.4052 | GPU 4.8GB
Step 200 | Epoch 0.50 | Loss 0.4385 | GPU 4.8GB
Step 200 | Epoch 0.50 | Loss 0.0000 | Val 0.4132 | GPU 4.8GB
Step 205 | Epoch 0.52 | Loss 0.4246 | GPU 4.8GB
Step 210 | Epoch 0.53 | Loss 0.3806 | GPU 4.8GB
Step 215 | Epoch 0.54 | Loss 0.4069 | GPU 4.8GB
Step 220 | Epoch 0.56 | Loss 0.4568 | GPU 4.8GB
Step 225 | Epoch 0.57 | Loss 0.4390 | GPU 4.8GB
Step 230 | Epoch 0.58 | Loss 0.3828 | GPU 4.8GB
Step 235 | Epoch 0.59 | Loss 0.4024 | GPU 4.8GB
Step 240 | Epoch 0.61 | Loss 0.4213 | GPU 4.8GB
Step 245 | Epoch 0.62 | Loss 0.4079 | GPU 4.8GB
Step 250 | Epoch 0.63 | Loss 0.4136 | GPU 4.8GB
Step 250 | Epoch 0.63 | Loss 0.0000 | Val 0.3971 | GPU 4.8GB
Step 255 | Epoch 0.64 | Loss 0.4068 | GPU 4.8GB
Step 260 | Epoch 0.66 | Loss 0.3866 | GPU 4.8GB
Step 265 | Epoch 0.67 | Loss 0.4004 | GPU 4.8GB
Step 270 | Epoch 0.68 | Loss 0.3792 | GPU 4.8GB
Step 275 | Epoch 0.69 | Loss 0.3888 | GPU 4.8GB
Step 280 | Epoch 0.71 | Loss 0.4107 | GPU 4.8GB
Step 285 | Epoch 0.72 | Loss 0.3998 | GPU 4.8GB
Step 290 | Epoch 0.73 | Loss 0.3698 | GPU 4.8GB
Step 295 | Epoch 0.74 | Loss 0.4095 | GPU 4.8GB
Step 300 | Epoch 0.76 | Loss 0.4059 | GPU 4.8GB
Step 300 | Epoch 0.76 | Loss 0.0000 | Val 0.3879 | GPU 4.8GB
Step 305 | Epoch 0.77 | Loss 0.3909 | GPU 4.8GB
Step 310 | Epoch 0.78 | Loss 0.4087 | GPU 4.8GB
Step 315 | Epoch 0.80 | Loss 0.3873 | GPU 4.8GB
Step 320 | Epoch 0.81 | Loss 0.3795 | GPU 4.8GB
Step 325 | Epoch 0.82 | Loss 0.4153 | GPU 4.8GB
Step 330 | Epoch 0.83 | Loss 0.3712 | GPU 4.8GB
Step 335 | Epoch 0.85 | Loss 0.3775 | GPU 4.8GB
Step 340 | Epoch 0.86 | Loss 0.3969 | GPU 4.8GB
Step 345 | Epoch 0.87 | Loss 0.3909 | GPU 4.8GB
Step 350 | Epoch 0.88 | Loss 0.3680 | GPU 4.8GB
Step 350 | Epoch 0.88 | Loss 0.0000 | Val 0.3820 | GPU 4.8GB
Step 355 | Epoch 0.90 | Loss 0.3800 | GPU 4.8GB
Step 360 | Epoch 0.91 | Loss 0.3959 | GPU 4.8GB
Step 365 | Epoch 0.92 | Loss 0.4114 | GPU 4.8GB
Step 370 | Epoch 0.93 | Loss 0.4359 | GPU 4.8GB
Step 375 | Epoch 0.95 | Loss 0.3860 | GPU 4.8GB
Step 380 | Epoch 0.96 | Loss 0.3927 | GPU 4.8GB
Step 385 | Epoch 0.97 | Loss 0.4058 | GPU 4.8GB
Step 390 | Epoch 0.98 | Loss 0.3936 | GPU 4.8GB
Step 395 | Epoch 1.00 | Loss 0.3589 | GPU 4.8GB
Step 397 | Epoch 1.00 | Loss 0.0000 | GPU 4.8GB
Step 397 | Epoch 1.00 | Loss 0.0000 | Val 0.3820 | GPU 4.6GB
Step 400 | Epoch 1.01 | Loss 0.3643 | GPU 4.8GB
Step 400 | Epoch 1.01 | Loss 0.0000 | Val 0.3832 | GPU 4.8GB
Step 405 | Epoch 1.02 | Loss 0.3570 | GPU 4.8GB
Step 410 | Epoch 1.03 | Loss 0.3536 | GPU 4.8GB
Step 415 | Epoch 1.05 | Loss 0.3230 | GPU 4.8GB
Step 420 | Epoch 1.06 | Loss 0.3551 | GPU 4.8GB
Step 425 | Epoch 1.07 | Loss 0.3472 | GPU 4.8GB
Step 430 | Epoch 1.08 | Loss 0.3598 | GPU 4.8GB
Step 435 | Epoch 1.10 | Loss 0.3664 | GPU 4.8GB
Step 440 | Epoch 1.11 | Loss 0.3455 | GPU 4.8GB
Step 445 | Epoch 1.12 | Loss 0.3546 | GPU 4.8GB
Step 450 | Epoch 1.13 | Loss 0.3654 | GPU 4.8GB
Step 450 | Epoch 1.13 | Loss 0.0000 | Val 0.4002 | GPU 4.8GB
Step 455 | Epoch 1.15 | Loss 0.3657 | GPU 4.8GB
Step 460 | Epoch 1.16 | Loss 0.3574 | GPU 4.8GB
Step 465 | Epoch 1.17 | Loss 0.3576 | GPU 4.8GB
Step 470 | Epoch 1.18 | Loss 0.3417 | GPU 4.8GB
Step 475 | Epoch 1.20 | Loss 0.3179 | GPU 4.8GB
Step 480 | Epoch 1.21 | Loss 0.3612 | GPU 4.8GB
Step 485 | Epoch 1.22 | Loss 0.3688 | GPU 4.8GB
Step 490 | Epoch 1.23 | Loss 0.3215 | GPU 4.8GB
Step 495 | Epoch 1.25 | Loss 0.3952 | GPU 4.8GB
Step 500 | Epoch 1.26 | Loss 0.3359 | GPU 4.8GB
Step 500 | Epoch 1.26 | Loss 0.0000 | Val 0.3917 | GPU 4.8GB
Step 500 | Epoch 1.26 | Loss 0.0000 | GPU 4.8GB
Step 500 | Epoch 1.26 | Loss 0.0000 | Val 0.3820 | GPU 4.6GB
Step 500 | Epoch 1.26 | Loss 0.0000 | GPU 4.7GB
Step 500 | Epoch 1.26 | Loss 0.0000 | Val 0.3820 | GPU 4.6GB
Step 505 | Epoch 1.27 | Loss 0.3206 | GPU 4.8GB
Step 510 | Epoch 1.29 | Loss 0.3052 | GPU 4.8GB
Step 515 | Epoch 1.30 | Loss 0.3186 | GPU 4.8GB
Step 520 | Epoch 1.31 | Loss 0.3023 | GPU 4.8GB
Step 525 | Epoch 1.32 | Loss 0.2954 | GPU 4.8GB
Step 530 | Epoch 1.34 | Loss 0.3072 | GPU 4.8GB
Step 535 | Epoch 1.35 | Loss 0.3076 | GPU 4.8GB
Step 540 | Epoch 1.36 | Loss 0.3134 | GPU 4.8GB
Step 545 | Epoch 1.37 | Loss 0.3121 | GPU 4.8GB
Step 550 | Epoch 1.39 | Loss 0.2900 | GPU 4.8GB
Step 550 | Epoch 1.39 | Loss 0.0000 | Val 0.4004 | GPU 4.8GB
Step 555 | Epoch 1.40 | Loss 0.3086 | GPU 4.8GB
Step 560 | Epoch 1.41 | Loss 0.3168 | GPU 4.8GB
Step 565 | Epoch 1.42 | Loss 0.3118 | GPU 4.8GB
Step 570 | Epoch 1.44 | Loss 0.2896 | GPU 4.8GB
Step 575 | Epoch 1.45 | Loss 0.3152 | GPU 4.8GB
Step 580 | Epoch 1.46 | Loss 0.3118 | GPU 4.8GB
Step 585 | Epoch 1.47 | Loss 0.2953 | GPU 4.8GB
Step 590 | Epoch 1.49 | Loss 0.3127 | GPU 4.8GB
Step 595 | Epoch 1.50 | Loss 0.3350 | GPU 4.8GB
Step 600 | Epoch 1.51 | Loss 0.3059 | GPU 4.8GB
Step 600 | Epoch 1.51 | Loss 0.0000 | Val 0.3939 | GPU 4.8GB
Step 605 | Epoch 1.52 | Loss 0.3096 | GPU 4.8GB
Step 610 | Epoch 1.54 | Loss 0.3248 | GPU 4.8GB
Step 615 | Epoch 1.55 | Loss 0.3303 | GPU 4.8GB
Step 620 | Epoch 1.56 | Loss 0.3266 | GPU 4.8GB
Step 625 | Epoch 1.58 | Loss 0.3235 | GPU 4.8GB
Step 630 | Epoch 1.59 | Loss 0.3059 | GPU 4.8GB
Step 635 | Epoch 1.60 | Loss 0.3180 | GPU 4.8GB
Step 640 | Epoch 1.61 | Loss 0.3287 | GPU 4.8GB
Step 645 | Epoch 1.63 | Loss 0.3167 | GPU 4.8GB
Step 650 | Epoch 1.64 | Loss 0.3293 | GPU 4.8GB
Step 650 | Epoch 1.64 | Loss 0.0000 | Val 0.3847 | GPU 4.8GB
Step 655 | Epoch 1.65 | Loss 0.3239 | GPU 4.8GB
Step 660 | Epoch 1.66 | Loss 0.3384 | GPU 4.8GB
Step 665 | Epoch 1.68 | Loss 0.3049 | GPU 4.8GB
Step 670 | Epoch 1.69 | Loss 0.3324 | GPU 4.8GB
Step 675 | Epoch 1.70 | Loss 0.3216 | GPU 4.8GB
Step 680 | Epoch 1.71 | Loss 0.3559 | GPU 4.8GB
Step 685 | Epoch 1.73 | Loss 0.3253 | GPU 4.8GB
Step 690 | Epoch 1.74 | Loss 0.3222 | GPU 4.8GB
Step 695 | Epoch 1.75 | Loss 0.3433 | GPU 4.8GB
Step 700 | Epoch 1.76 | Loss 0.3485 | GPU 4.8GB
Step 700 | Epoch 1.76 | Loss 0.0000 | Val 0.3790 | GPU 4.8GB
Step 705 | Epoch 1.78 | Loss 0.3362 | GPU 4.8GB
Step 710 | Epoch 1.79 | Loss 0.3697 | GPU 4.8GB
Step 705 | Epoch 1.78 | Loss 0.3362 | GPU 4.8GB
Step 710 | Epoch 1.79 | Loss 0.3697 | GPU 4.8GB
Step 715 | Epoch 1.80 | Loss 0.3305 | GPU 4.8GB
Step 720 | Epoch 1.82 | Loss 0.3573 | GPU 4.8GB
Step 725 | Epoch 1.83 | Loss 0.3588 | GPU 4.8GB
Step 730 | Epoch 1.84 | Loss 0.3293 | GPU 4.8GB
Step 735 | Epoch 1.85 | Loss 0.3286 | GPU 4.8GB
Step 740 | Epoch 1.87 | Loss 0.3739 | GPU 4.8GB
Step 745 | Epoch 1.88 | Loss 0.3495 | GPU 4.8GB
Step 750 | Epoch 1.89 | Loss 0.3455 | GPU 4.8GB
Step 750 | Epoch 1.89 | Loss 0.0000 | Val 0.3757 | GPU 4.8GB
Step 755 | Epoch 1.90 | Loss 0.3478 | GPU 4.8GB
Step 760 | Epoch 1.92 | Loss 0.3945 | GPU 4.8GB
Step 765 | Epoch 1.93 | Loss 0.3643 | GPU 4.8GB
Step 770 | Epoch 1.94 | Loss 0.3985 | GPU 4.8GB
Step 775 | Epoch 1.95 | Loss 0.3645 | GPU 4.8GB
Step 780 | Epoch 1.97 | Loss 0.3662 | GPU 4.8GB
Step 785 | Epoch 1.98 | Loss 0.3323 | GPU 4.8GB
Step 790 | Epoch 1.99 | Loss 0.2940 | GPU 4.8GB
Step 794 | Epoch 2.00 | Loss 0.0000 | GPU 4.8GB
Step 794 | Epoch 2.00 | Loss 0.0000 | Val 0.3757 | GPU 4.6GB
Step 795 | Epoch 2.00 | Loss 0.3075 | GPU 5.1GB
Step 800 | Epoch 2.02 | Loss 0.3087 | GPU 5.1GB
Step 800 | Epoch 2.02 | Loss 0.0000 | Val 0.3855 | GPU 5.1GB
Step 805 | Epoch 2.03 | Loss 0.3242 | GPU 5.1GB
Step 810 | Epoch 2.04 | Loss 0.3122 | GPU 5.1GB
Step 815 | Epoch 2.05 | Loss 0.3247 | GPU 5.1GB
Step 820 | Epoch 2.07 | Loss 0.3181 | GPU 5.1GB
Step 825 | Epoch 2.08 | Loss 0.3175 | GPU 5.1GB
Step 830 | Epoch 2.09 | Loss 0.3379 | GPU 5.1GB
Step 835 | Epoch 2.10 | Loss 0.3155 | GPU 5.1GB
Step 840 | Epoch 2.12 | Loss 0.3060 | GPU 5.1GB
Step 845 | Epoch 2.13 | Loss 0.2912 | GPU 5.1GB
Step 850 | Epoch 2.14 | Loss 0.3125 | GPU 5.1GB
Step 850 | Epoch 2.14 | Loss 0.0000 | Val 0.3904 | GPU 5.1GB
Step 855 | Epoch 2.15 | Loss 0.3081 | GPU 5.1GB
Step 860 | Epoch 2.17 | Loss 0.3101 | GPU 5.1GB
Step 865 | Epoch 2.18 | Loss 0.3210 | GPU 5.1GB
Step 870 | Epoch 2.19 | Loss 0.2998 | GPU 5.1GB
Step 875 | Epoch 2.20 | Loss 0.2875 | GPU 5.1GB
Step 880 | Epoch 2.22 | Loss 0.3301 | GPU 5.1GB
Step 885 | Epoch 2.23 | Loss 0.3124 | GPU 5.1GB
Step 890 | Epoch 2.24 | Loss 0.3235 | GPU 5.1GB
Step 895 | Epoch 2.25 | Loss 0.3139 | GPU 5.1GB
Step 900 | Epoch 2.27 | Loss 0.3078 | GPU 5.1GB
Step 900 | Epoch 2.27 | Loss 0.0000 | Val 0.3843 | GPU 5.1GB
Step 905 | Epoch 2.28 | Loss 0.2974 | GPU 5.1GB
Step 910 | Epoch 2.29 | Loss 0.3265 | GPU 5.1GB
Step 915 | Epoch 2.31 | Loss 0.2984 | GPU 5.1GB
Step 920 | Epoch 2.32 | Loss 0.2988 | GPU 5.1GB
Step 925 | Epoch 2.33 | Loss 0.2909 | GPU 5.1GB
Step 930 | Epoch 2.34 | Loss 0.3014 | GPU 5.1GB
Step 935 | Epoch 2.36 | Loss 0.3084 | GPU 5.1GB
Step 940 | Epoch 2.37 | Loss 0.2792 | GPU 5.1GB
Step 945 | Epoch 2.38 | Loss 0.3307 | GPU 5.1GB
Step 950 | Epoch 2.39 | Loss 0.2959 | GPU 5.1GB
Step 950 | Epoch 2.39 | Loss 0.0000 | Val 0.3808 | GPU 5.1GB
Step 955 | Epoch 2.41 | Loss 0.3184 | GPU 5.1GB
Step 960 | Epoch 2.42 | Loss 0.3087 | GPU 5.1GB
Step 965 | Epoch 2.43 | Loss 0.2931 | GPU 5.1GB
Step 970 | Epoch 2.44 | Loss 0.2797 | GPU 5.1GB
Step 975 | Epoch 2.46 | Loss 0.3237 | GPU 5.1GB
Step 980 | Epoch 2.47 | Loss 0.3137 | GPU 5.1GB
Step 985 | Epoch 2.48 | Loss 0.3107 | GPU 5.1GB
Step 990 | Epoch 2.49 | Loss 0.2910 | GPU 5.1GB
Step 995 | Epoch 2.51 | Loss 0.3042 | GPU 5.1GB
Step 1000 | Epoch 2.52 | Loss 0.2947 | GPU 5.1GB
Step 1000 | Epoch 2.52 | Loss 0.0000 | Val 0.3781 | GPU 5.1GB
Step 1005 | Epoch 2.53 | Loss 0.3105 | GPU 5.1GB
Step 1010 | Epoch 2.55 | Loss 0.3056 | GPU 5.1GB
Step 1015 | Epoch 2.56 | Loss 0.3169 | GPU 5.1GB
Step 1020 | Epoch 2.57 | Loss 0.3064 | GPU 5.1GB
Step 1025 | Epoch 2.58 | Loss 0.2946 | GPU 5.1GB
Step 1030 | Epoch 2.60 | Loss 0.3067 | GPU 5.1GB
Step 1035 | Epoch 2.61 | Loss 0.3105 | GPU 5.1GB
Step 1040 | Epoch 2.62 | Loss 0.3075 | GPU 5.1GB
Step 1045 | Epoch 2.63 | Loss 0.3018 | GPU 5.1GB
Step 1050 | Epoch 2.65 | Loss 0.3251 | GPU 5.1GB
Step 1050 | Epoch 2.65 | Loss 0.0000 | Val 0.3762 | GPU 5.1GB
Step 1055 | Epoch 2.66 | Loss 0.2916 | GPU 5.1GB
Step 1060 | Epoch 2.67 | Loss 0.3052 | GPU 5.1GB
Step 1065 | Epoch 2.68 | Loss 0.2960 | GPU 5.1GB
Step 1070 | Epoch 2.70 | Loss 0.3054 | GPU 5.1GB
Step 1075 | Epoch 2.71 | Loss 0.2839 | GPU 5.1GB
Step 1080 | Epoch 2.72 | Loss 0.2944 | GPU 5.1GB
Step 1085 | Epoch 2.73 | Loss 0.2868 | GPU 5.1GB
Step 1090 | Epoch 2.75 | Loss 0.2899 | GPU 5.1GB
Step 1095 | Epoch 2.76 | Loss 0.3230 | GPU 5.1GB
Step 1100 | Epoch 2.77 | Loss 0.2995 | GPU 5.1GB
Step 1100 | Epoch 2.77 | Loss 0.0000 | Val 0.3757 | GPU 5.1GB
Step 1105 | Epoch 2.78 | Loss 0.2799 | GPU 5.1GB
Step 1110 | Epoch 2.80 | Loss 0.2857 | GPU 5.1GB
Step 1115 | Epoch 2.81 | Loss 0.2893 | GPU 5.1GB
Step 1120 | Epoch 2.82 | Loss 0.3118 | GPU 5.1GB
Step 1125 | Epoch 2.84 | Loss 0.2831 | GPU 5.1GB
Step 1130 | Epoch 2.85 | Loss 0.3088 | GPU 5.1GB
Step 1135 | Epoch 2.86 | Loss 0.2960 | GPU 5.1GB
Step 1140 | Epoch 2.87 | Loss 0.2957 | GPU 5.1GB
Step 1145 | Epoch 2.89 | Loss 0.3266 | GPU 5.1GB
Step 1150 | Epoch 2.90 | Loss 0.2961 | GPU 5.1GB
Step 1150 | Epoch 2.90 | Loss 0.0000 | Val 0.3754 | GPU 5.1GB
Step 1155 | Epoch 2.91 | Loss 0.2960 | GPU 5.1GB
Step 1160 | Epoch 2.92 | Loss 0.2968 | GPU 5.1GB
Step 1165 | Epoch 2.94 | Loss 0.2791 | GPU 5.1GB
Step 1170 | Epoch 2.95 | Loss 0.2964 | GPU 5.1GB
Step 1175 | Epoch 2.96 | Loss 0.3137 | GPU 5.1GB
Step 1180 | Epoch 2.97 | Loss 0.2946 | GPU 5.1GB
Step 1185 | Epoch 2.99 | Loss 0.3130 | GPU 5.1GB
Step 1190 | Epoch 3.00 | Loss 0.2985 | GPU 5.1GB
Step 1191 | Epoch 3.00 | Loss 0.0000 | GPU 5.1GB
Step 1191 | Epoch 3.00 | Loss 0.0000 | Val 0.3754 | GPU 4.9GB