| { | |
| "best_metric": 0.35846731066703796, | |
| "best_model_checkpoint": "finetune-tomato-leaf/checkpoint-875", | |
| "epoch": 5.0, | |
| "eval_steps": 100, | |
| "global_step": 875, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05714285714285714, | |
| "grad_norm": 66266.859375, | |
| "learning_rate": 1.9771428571428574e-05, | |
| "loss": 2.3535, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11428571428571428, | |
| "grad_norm": 82785.25, | |
| "learning_rate": 1.9542857142857143e-05, | |
| "loss": 2.2735, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17142857142857143, | |
| "grad_norm": 76933.2109375, | |
| "learning_rate": 1.9314285714285718e-05, | |
| "loss": 2.1921, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.22857142857142856, | |
| "grad_norm": 88895.1796875, | |
| "learning_rate": 1.9085714285714287e-05, | |
| "loss": 2.1107, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 85212.125, | |
| "learning_rate": 1.885714285714286e-05, | |
| "loss": 1.9926, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.34285714285714286, | |
| "grad_norm": 90773.2109375, | |
| "learning_rate": 1.8628571428571428e-05, | |
| "loss": 1.8996, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 85021.5234375, | |
| "learning_rate": 1.8400000000000003e-05, | |
| "loss": 1.7758, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.45714285714285713, | |
| "grad_norm": 93720.375, | |
| "learning_rate": 1.8171428571428572e-05, | |
| "loss": 1.6423, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5142857142857142, | |
| "grad_norm": 94810.765625, | |
| "learning_rate": 1.7942857142857144e-05, | |
| "loss": 1.5665, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 97387.890625, | |
| "learning_rate": 1.7714285714285717e-05, | |
| "loss": 1.4395, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6285714285714286, | |
| "grad_norm": 155846.140625, | |
| "learning_rate": 1.748571428571429e-05, | |
| "loss": 1.4227, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6857142857142857, | |
| "grad_norm": 86622.5859375, | |
| "learning_rate": 1.7257142857142857e-05, | |
| "loss": 1.3027, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.7428571428571429, | |
| "grad_norm": 89916.921875, | |
| "learning_rate": 1.702857142857143e-05, | |
| "loss": 1.2367, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 117233.484375, | |
| "learning_rate": 1.6800000000000002e-05, | |
| "loss": 1.2059, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 110025.3203125, | |
| "learning_rate": 1.6571428571428574e-05, | |
| "loss": 1.1757, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9142857142857143, | |
| "grad_norm": 104192.2734375, | |
| "learning_rate": 1.6342857142857146e-05, | |
| "loss": 1.0949, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.9714285714285714, | |
| "grad_norm": 125921.359375, | |
| "learning_rate": 1.6114285714285715e-05, | |
| "loss": 1.0819, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9018311291963378, | |
| "eval_loss": 0.9894934892654419, | |
| "eval_runtime": 21.9196, | |
| "eval_samples_per_second": 89.692, | |
| "eval_steps_per_second": 1.414, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.0285714285714285, | |
| "grad_norm": 120913.9296875, | |
| "learning_rate": 1.5885714285714287e-05, | |
| "loss": 1.014, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0857142857142856, | |
| "grad_norm": 108828.4609375, | |
| "learning_rate": 1.5657142857142856e-05, | |
| "loss": 1.0122, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 114912.9296875, | |
| "learning_rate": 1.542857142857143e-05, | |
| "loss": 0.9622, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 126387.203125, | |
| "learning_rate": 1.5200000000000002e-05, | |
| "loss": 0.9023, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.2571428571428571, | |
| "grad_norm": 145092.53125, | |
| "learning_rate": 1.4971428571428572e-05, | |
| "loss": 0.8747, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.3142857142857143, | |
| "grad_norm": 148149.9375, | |
| "learning_rate": 1.4742857142857143e-05, | |
| "loss": 0.8614, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.3714285714285714, | |
| "grad_norm": 118238.6875, | |
| "learning_rate": 1.4514285714285717e-05, | |
| "loss": 0.835, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 117691.6953125, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.8084, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.4857142857142858, | |
| "grad_norm": 161266.21875, | |
| "learning_rate": 1.4057142857142858e-05, | |
| "loss": 0.7982, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.5428571428571427, | |
| "grad_norm": 101406.703125, | |
| "learning_rate": 1.382857142857143e-05, | |
| "loss": 0.8023, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 125123.390625, | |
| "learning_rate": 1.3600000000000002e-05, | |
| "loss": 0.714, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.657142857142857, | |
| "grad_norm": 101578.0703125, | |
| "learning_rate": 1.3371428571428572e-05, | |
| "loss": 0.7169, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 165960.28125, | |
| "learning_rate": 1.3142857142857145e-05, | |
| "loss": 0.6705, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.7714285714285714, | |
| "grad_norm": 161066.1875, | |
| "learning_rate": 1.2914285714285715e-05, | |
| "loss": 0.7121, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.8285714285714287, | |
| "grad_norm": 308346.46875, | |
| "learning_rate": 1.2685714285714286e-05, | |
| "loss": 0.6873, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.8857142857142857, | |
| "grad_norm": 135629.84375, | |
| "learning_rate": 1.245714285714286e-05, | |
| "loss": 0.6417, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.9428571428571428, | |
| "grad_norm": 134140.0, | |
| "learning_rate": 1.222857142857143e-05, | |
| "loss": 0.6366, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 365796.21875, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.6466, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9593082400813835, | |
| "eval_loss": 0.5825250148773193, | |
| "eval_runtime": 21.8245, | |
| "eval_samples_per_second": 90.082, | |
| "eval_steps_per_second": 1.42, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.057142857142857, | |
| "grad_norm": 110862.375, | |
| "learning_rate": 1.177142857142857e-05, | |
| "loss": 0.6206, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.1142857142857143, | |
| "grad_norm": 99193.8984375, | |
| "learning_rate": 1.1542857142857145e-05, | |
| "loss": 0.5719, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.1714285714285713, | |
| "grad_norm": 133490.75, | |
| "learning_rate": 1.1314285714285715e-05, | |
| "loss": 0.5888, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.2285714285714286, | |
| "grad_norm": 134935.203125, | |
| "learning_rate": 1.1085714285714286e-05, | |
| "loss": 0.6034, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 129637.0703125, | |
| "learning_rate": 1.0857142857142858e-05, | |
| "loss": 0.5734, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.342857142857143, | |
| "grad_norm": 174524.84375, | |
| "learning_rate": 1.062857142857143e-05, | |
| "loss": 0.5673, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 88518.9921875, | |
| "learning_rate": 1.04e-05, | |
| "loss": 0.5314, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.4571428571428573, | |
| "grad_norm": 137873.953125, | |
| "learning_rate": 1.0171428571428573e-05, | |
| "loss": 0.5602, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.5142857142857142, | |
| "grad_norm": 156413.375, | |
| "learning_rate": 9.942857142857145e-06, | |
| "loss": 0.5819, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 137907.96875, | |
| "learning_rate": 9.714285714285715e-06, | |
| "loss": 0.524, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.6285714285714286, | |
| "grad_norm": 161790.703125, | |
| "learning_rate": 9.485714285714287e-06, | |
| "loss": 0.4902, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.685714285714286, | |
| "grad_norm": 118504.765625, | |
| "learning_rate": 9.257142857142858e-06, | |
| "loss": 0.5091, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.742857142857143, | |
| "grad_norm": 135314.796875, | |
| "learning_rate": 9.028571428571428e-06, | |
| "loss": 0.4945, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 135834.125, | |
| "learning_rate": 8.8e-06, | |
| "loss": 0.4792, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 161810.28125, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.4867, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.914285714285714, | |
| "grad_norm": 120542.2890625, | |
| "learning_rate": 8.342857142857143e-06, | |
| "loss": 0.4809, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.9714285714285715, | |
| "grad_norm": 132218.109375, | |
| "learning_rate": 8.114285714285715e-06, | |
| "loss": 0.4737, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9608341810783316, | |
| "eval_loss": 0.4544333815574646, | |
| "eval_runtime": 21.6981, | |
| "eval_samples_per_second": 90.607, | |
| "eval_steps_per_second": 1.429, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 3.0285714285714285, | |
| "grad_norm": 175208.0625, | |
| "learning_rate": 7.885714285714286e-06, | |
| "loss": 0.4208, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.085714285714286, | |
| "grad_norm": 152119.046875, | |
| "learning_rate": 7.657142857142858e-06, | |
| "loss": 0.4432, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 147346.734375, | |
| "learning_rate": 7.428571428571429e-06, | |
| "loss": 0.4572, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 124751.4609375, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 0.4745, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.257142857142857, | |
| "grad_norm": 147328.890625, | |
| "learning_rate": 6.971428571428573e-06, | |
| "loss": 0.466, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.314285714285714, | |
| "grad_norm": 226765.96875, | |
| "learning_rate": 6.742857142857143e-06, | |
| "loss": 0.467, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.3714285714285714, | |
| "grad_norm": 126136.609375, | |
| "learning_rate": 6.514285714285715e-06, | |
| "loss": 0.4399, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 133291.390625, | |
| "learning_rate": 6.285714285714286e-06, | |
| "loss": 0.4362, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.4857142857142858, | |
| "grad_norm": 210495.234375, | |
| "learning_rate": 6.057142857142858e-06, | |
| "loss": 0.4548, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.5428571428571427, | |
| "grad_norm": 143375.640625, | |
| "learning_rate": 5.828571428571429e-06, | |
| "loss": 0.4342, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 164240.890625, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.4189, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.657142857142857, | |
| "grad_norm": 200663.953125, | |
| "learning_rate": 5.371428571428572e-06, | |
| "loss": 0.4396, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 100999.9453125, | |
| "learning_rate": 5.142857142857142e-06, | |
| "loss": 0.4061, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.7714285714285714, | |
| "grad_norm": 117663.09375, | |
| "learning_rate": 4.9142857142857145e-06, | |
| "loss": 0.3983, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.8285714285714287, | |
| "grad_norm": 124319.609375, | |
| "learning_rate": 4.685714285714286e-06, | |
| "loss": 0.448, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 3.8857142857142857, | |
| "grad_norm": 187423.625, | |
| "learning_rate": 4.457142857142858e-06, | |
| "loss": 0.3981, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 3.942857142857143, | |
| "grad_norm": 179787.3125, | |
| "learning_rate": 4.228571428571429e-06, | |
| "loss": 0.4109, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 115739.109375, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.3983, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9715157680569685, | |
| "eval_loss": 0.3787924349308014, | |
| "eval_runtime": 21.6761, | |
| "eval_samples_per_second": 90.699, | |
| "eval_steps_per_second": 1.43, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.057142857142857, | |
| "grad_norm": 100285.75, | |
| "learning_rate": 3.771428571428572e-06, | |
| "loss": 0.3988, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.114285714285714, | |
| "grad_norm": 119137.453125, | |
| "learning_rate": 3.542857142857143e-06, | |
| "loss": 0.3963, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.171428571428572, | |
| "grad_norm": 101105.1015625, | |
| "learning_rate": 3.314285714285714e-06, | |
| "loss": 0.4028, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 4.228571428571429, | |
| "grad_norm": 231164.015625, | |
| "learning_rate": 3.085714285714286e-06, | |
| "loss": 0.4107, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 106757.0859375, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.3895, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.3428571428571425, | |
| "grad_norm": 144770.203125, | |
| "learning_rate": 2.6285714285714286e-06, | |
| "loss": 0.4163, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 164616.515625, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.4121, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 4.457142857142857, | |
| "grad_norm": 92120.4140625, | |
| "learning_rate": 2.1714285714285716e-06, | |
| "loss": 0.379, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 4.514285714285714, | |
| "grad_norm": 111853.0234375, | |
| "learning_rate": 1.942857142857143e-06, | |
| "loss": 0.4009, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 181065.359375, | |
| "learning_rate": 1.7142857142857145e-06, | |
| "loss": 0.3857, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.628571428571428, | |
| "grad_norm": 165270.3125, | |
| "learning_rate": 1.4857142857142858e-06, | |
| "loss": 0.4084, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 4.685714285714286, | |
| "grad_norm": 195606.078125, | |
| "learning_rate": 1.2571428571428573e-06, | |
| "loss": 0.391, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 4.742857142857143, | |
| "grad_norm": 146015.765625, | |
| "learning_rate": 1.0285714285714286e-06, | |
| "loss": 0.4011, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 165905.390625, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 0.3932, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 185854.59375, | |
| "learning_rate": 5.714285714285715e-07, | |
| "loss": 0.3896, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.914285714285715, | |
| "grad_norm": 212310.71875, | |
| "learning_rate": 3.428571428571429e-07, | |
| "loss": 0.3811, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 4.9714285714285715, | |
| "grad_norm": 198072.859375, | |
| "learning_rate": 1.142857142857143e-07, | |
| "loss": 0.3876, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9720244150559512, | |
| "eval_loss": 0.35846731066703796, | |
| "eval_runtime": 21.7929, | |
| "eval_samples_per_second": 90.213, | |
| "eval_steps_per_second": 1.422, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 875, | |
| "total_flos": 4.315877010121421e+18, | |
| "train_loss": 0.7534563433783394, | |
| "train_runtime": 1829.0504, | |
| "train_samples_per_second": 30.447, | |
| "train_steps_per_second": 0.478 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 875, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.315877010121421e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |