| { | |
| "best_metric": 0.9090909090909091, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-85-fold3/checkpoint-40", | |
| "epoch": 100.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3409090909090909, | |
| "eval_loss": 0.7891632318496704, | |
| "eval_runtime": 0.5322, | |
| "eval_samples_per_second": 82.681, | |
| "eval_steps_per_second": 3.758, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.554598331451416, | |
| "eval_runtime": 0.5484, | |
| "eval_samples_per_second": 80.239, | |
| "eval_steps_per_second": 3.647, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.6493009328842163, | |
| "eval_runtime": 0.5368, | |
| "eval_samples_per_second": 81.962, | |
| "eval_steps_per_second": 3.726, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.6647928953170776, | |
| "eval_runtime": 0.5414, | |
| "eval_samples_per_second": 81.274, | |
| "eval_steps_per_second": 3.694, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.9686079025268555, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6939, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.5186769366264343, | |
| "eval_runtime": 0.5423, | |
| "eval_samples_per_second": 81.132, | |
| "eval_steps_per_second": 3.688, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.49025213718414307, | |
| "eval_runtime": 0.5491, | |
| "eval_samples_per_second": 80.13, | |
| "eval_steps_per_second": 3.642, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.50870680809021, | |
| "eval_runtime": 0.555, | |
| "eval_samples_per_second": 79.279, | |
| "eval_steps_per_second": 3.604, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.5789340138435364, | |
| "eval_runtime": 0.5825, | |
| "eval_samples_per_second": 75.532, | |
| "eval_steps_per_second": 3.433, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4918547570705414, | |
| "eval_runtime": 0.5965, | |
| "eval_samples_per_second": 73.76, | |
| "eval_steps_per_second": 3.353, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.733920574188232, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4553, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.47069433331489563, | |
| "eval_runtime": 0.5732, | |
| "eval_samples_per_second": 76.757, | |
| "eval_steps_per_second": 3.489, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.5120064616203308, | |
| "eval_runtime": 0.5865, | |
| "eval_samples_per_second": 75.02, | |
| "eval_steps_per_second": 3.41, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.4733530282974243, | |
| "eval_runtime": 0.5854, | |
| "eval_samples_per_second": 75.165, | |
| "eval_steps_per_second": 3.417, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.42547714710235596, | |
| "eval_runtime": 0.6068, | |
| "eval_samples_per_second": 72.508, | |
| "eval_steps_per_second": 3.296, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3694608509540558, | |
| "eval_runtime": 0.6153, | |
| "eval_samples_per_second": 71.504, | |
| "eval_steps_per_second": 3.25, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 4.073104381561279, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.3658, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.38475659489631653, | |
| "eval_runtime": 0.6041, | |
| "eval_samples_per_second": 72.839, | |
| "eval_steps_per_second": 3.311, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.35861828923225403, | |
| "eval_runtime": 0.5924, | |
| "eval_samples_per_second": 74.28, | |
| "eval_steps_per_second": 3.376, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4961988627910614, | |
| "eval_runtime": 0.5831, | |
| "eval_samples_per_second": 75.461, | |
| "eval_steps_per_second": 3.43, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.36452916264533997, | |
| "eval_runtime": 0.5806, | |
| "eval_samples_per_second": 75.785, | |
| "eval_steps_per_second": 3.445, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3454984128475189, | |
| "eval_runtime": 0.5878, | |
| "eval_samples_per_second": 74.853, | |
| "eval_steps_per_second": 3.402, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 2.5263137817382812, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.2667, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.3476676046848297, | |
| "eval_runtime": 0.5913, | |
| "eval_samples_per_second": 74.418, | |
| "eval_steps_per_second": 3.383, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.32747599482536316, | |
| "eval_runtime": 0.5926, | |
| "eval_samples_per_second": 74.247, | |
| "eval_steps_per_second": 3.375, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3400273621082306, | |
| "eval_runtime": 0.5906, | |
| "eval_samples_per_second": 74.496, | |
| "eval_steps_per_second": 3.386, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.37799492478370667, | |
| "eval_runtime": 0.6042, | |
| "eval_samples_per_second": 72.822, | |
| "eval_steps_per_second": 3.31, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.42426249384880066, | |
| "eval_runtime": 0.5866, | |
| "eval_samples_per_second": 75.009, | |
| "eval_steps_per_second": 3.409, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 4.296473979949951, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.1794, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4429081976413727, | |
| "eval_runtime": 0.5876, | |
| "eval_samples_per_second": 74.879, | |
| "eval_steps_per_second": 3.404, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.502648651599884, | |
| "eval_runtime": 0.5924, | |
| "eval_samples_per_second": 74.28, | |
| "eval_steps_per_second": 3.376, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4810771048069, | |
| "eval_runtime": 0.5905, | |
| "eval_samples_per_second": 74.514, | |
| "eval_steps_per_second": 3.387, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.4732614755630493, | |
| "eval_runtime": 0.6054, | |
| "eval_samples_per_second": 72.684, | |
| "eval_steps_per_second": 3.304, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.43838709592819214, | |
| "eval_runtime": 0.594, | |
| "eval_samples_per_second": 74.074, | |
| "eval_steps_per_second": 3.367, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 4.2467732429504395, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.1861, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.4354090690612793, | |
| "eval_runtime": 0.5843, | |
| "eval_samples_per_second": 75.303, | |
| "eval_steps_per_second": 3.423, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.4511108100414276, | |
| "eval_runtime": 0.62, | |
| "eval_samples_per_second": 70.969, | |
| "eval_steps_per_second": 3.226, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3315487205982208, | |
| "eval_runtime": 0.6096, | |
| "eval_samples_per_second": 72.174, | |
| "eval_steps_per_second": 3.281, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.30998334288597107, | |
| "eval_runtime": 0.5972, | |
| "eval_samples_per_second": 73.674, | |
| "eval_steps_per_second": 3.349, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.3593769967556, | |
| "eval_runtime": 0.5949, | |
| "eval_samples_per_second": 73.956, | |
| "eval_steps_per_second": 3.362, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 1.2741260528564453, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.1521, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.40515002608299255, | |
| "eval_runtime": 0.5865, | |
| "eval_samples_per_second": 75.023, | |
| "eval_steps_per_second": 3.41, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.38777944445610046, | |
| "eval_runtime": 0.5929, | |
| "eval_samples_per_second": 74.207, | |
| "eval_steps_per_second": 3.373, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.3904772102832794, | |
| "eval_runtime": 0.5914, | |
| "eval_samples_per_second": 74.403, | |
| "eval_steps_per_second": 3.382, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.41729357838630676, | |
| "eval_runtime": 0.583, | |
| "eval_samples_per_second": 75.469, | |
| "eval_steps_per_second": 3.43, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.477445513010025, | |
| "eval_runtime": 0.5847, | |
| "eval_samples_per_second": 75.251, | |
| "eval_steps_per_second": 3.42, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 2.6318907737731934, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1333, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.5655959844589233, | |
| "eval_runtime": 0.5879, | |
| "eval_samples_per_second": 74.846, | |
| "eval_steps_per_second": 3.402, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.5146284103393555, | |
| "eval_runtime": 0.5847, | |
| "eval_samples_per_second": 75.25, | |
| "eval_steps_per_second": 3.42, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.415780633687973, | |
| "eval_runtime": 0.6067, | |
| "eval_samples_per_second": 72.521, | |
| "eval_steps_per_second": 3.296, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.40665361285209656, | |
| "eval_runtime": 0.591, | |
| "eval_samples_per_second": 74.446, | |
| "eval_steps_per_second": 3.384, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.4412151575088501, | |
| "eval_runtime": 0.5973, | |
| "eval_samples_per_second": 73.669, | |
| "eval_steps_per_second": 3.349, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 2.619199514389038, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1297, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.47330716252326965, | |
| "eval_runtime": 0.5794, | |
| "eval_samples_per_second": 75.942, | |
| "eval_steps_per_second": 3.452, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.42430323362350464, | |
| "eval_runtime": 0.5889, | |
| "eval_samples_per_second": 74.713, | |
| "eval_steps_per_second": 3.396, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.4278627634048462, | |
| "eval_runtime": 0.588, | |
| "eval_samples_per_second": 74.833, | |
| "eval_steps_per_second": 3.401, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.4019567370414734, | |
| "eval_runtime": 0.5932, | |
| "eval_samples_per_second": 74.176, | |
| "eval_steps_per_second": 3.372, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.38422030210494995, | |
| "eval_runtime": 0.5938, | |
| "eval_samples_per_second": 74.103, | |
| "eval_steps_per_second": 3.368, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 1.5373425483703613, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1038, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.38110530376434326, | |
| "eval_runtime": 0.5844, | |
| "eval_samples_per_second": 75.29, | |
| "eval_steps_per_second": 3.422, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.39466771483421326, | |
| "eval_runtime": 0.59, | |
| "eval_samples_per_second": 74.571, | |
| "eval_steps_per_second": 3.39, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.4587290585041046, | |
| "eval_runtime": 0.5871, | |
| "eval_samples_per_second": 74.944, | |
| "eval_steps_per_second": 3.407, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.4299503564834595, | |
| "eval_runtime": 0.5915, | |
| "eval_samples_per_second": 74.385, | |
| "eval_steps_per_second": 3.381, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.38039132952690125, | |
| "eval_runtime": 0.6082, | |
| "eval_samples_per_second": 72.349, | |
| "eval_steps_per_second": 3.289, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 1.628954529762268, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1101, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.42160940170288086, | |
| "eval_runtime": 0.5977, | |
| "eval_samples_per_second": 73.612, | |
| "eval_steps_per_second": 3.346, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.39656153321266174, | |
| "eval_runtime": 0.5978, | |
| "eval_samples_per_second": 73.608, | |
| "eval_steps_per_second": 3.346, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.42155352234840393, | |
| "eval_runtime": 0.5919, | |
| "eval_samples_per_second": 74.336, | |
| "eval_steps_per_second": 3.379, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.45689570903778076, | |
| "eval_runtime": 0.5906, | |
| "eval_samples_per_second": 74.506, | |
| "eval_steps_per_second": 3.387, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.43915289640426636, | |
| "eval_runtime": 0.6137, | |
| "eval_samples_per_second": 71.695, | |
| "eval_steps_per_second": 3.259, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 0.9616934061050415, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1085, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.44787395000457764, | |
| "eval_runtime": 0.6245, | |
| "eval_samples_per_second": 70.46, | |
| "eval_steps_per_second": 3.203, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.4657154083251953, | |
| "eval_runtime": 0.5987, | |
| "eval_samples_per_second": 73.496, | |
| "eval_steps_per_second": 3.341, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.5241695642471313, | |
| "eval_runtime": 0.6067, | |
| "eval_samples_per_second": 72.528, | |
| "eval_steps_per_second": 3.297, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.562565803527832, | |
| "eval_runtime": 0.5986, | |
| "eval_samples_per_second": 73.499, | |
| "eval_steps_per_second": 3.341, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.5569871068000793, | |
| "eval_runtime": 0.5953, | |
| "eval_samples_per_second": 73.907, | |
| "eval_steps_per_second": 3.359, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 2.443108320236206, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.105, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.5034798979759216, | |
| "eval_runtime": 0.5917, | |
| "eval_samples_per_second": 74.364, | |
| "eval_steps_per_second": 3.38, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.4489854574203491, | |
| "eval_runtime": 0.5955, | |
| "eval_samples_per_second": 73.885, | |
| "eval_steps_per_second": 3.358, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.43662911653518677, | |
| "eval_runtime": 0.5904, | |
| "eval_samples_per_second": 74.523, | |
| "eval_steps_per_second": 3.387, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.4415828585624695, | |
| "eval_runtime": 0.5951, | |
| "eval_samples_per_second": 73.942, | |
| "eval_steps_per_second": 3.361, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.45971566438674927, | |
| "eval_runtime": 0.5966, | |
| "eval_samples_per_second": 73.756, | |
| "eval_steps_per_second": 3.353, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 2.1702709197998047, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0918, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.47954508662223816, | |
| "eval_runtime": 0.5895, | |
| "eval_samples_per_second": 74.635, | |
| "eval_steps_per_second": 3.392, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.49216070771217346, | |
| "eval_runtime": 0.5928, | |
| "eval_samples_per_second": 74.221, | |
| "eval_steps_per_second": 3.374, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.5078458786010742, | |
| "eval_runtime": 0.6172, | |
| "eval_samples_per_second": 71.294, | |
| "eval_steps_per_second": 3.241, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.5089394450187683, | |
| "eval_runtime": 0.5889, | |
| "eval_samples_per_second": 74.717, | |
| "eval_steps_per_second": 3.396, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.5108653903007507, | |
| "eval_runtime": 0.5862, | |
| "eval_samples_per_second": 75.058, | |
| "eval_steps_per_second": 3.412, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 2.610532760620117, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.1072, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.5125123262405396, | |
| "eval_runtime": 0.5854, | |
| "eval_samples_per_second": 75.159, | |
| "eval_steps_per_second": 3.416, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.5267102718353271, | |
| "eval_runtime": 0.6066, | |
| "eval_samples_per_second": 72.535, | |
| "eval_steps_per_second": 3.297, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.5346480011940002, | |
| "eval_runtime": 0.5929, | |
| "eval_samples_per_second": 74.215, | |
| "eval_steps_per_second": 3.373, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.5291395783424377, | |
| "eval_runtime": 0.59, | |
| "eval_samples_per_second": 74.571, | |
| "eval_steps_per_second": 3.39, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.5187559723854065, | |
| "eval_runtime": 0.5911, | |
| "eval_samples_per_second": 74.443, | |
| "eval_steps_per_second": 3.384, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 1.9993984699249268, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0895, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.5222446918487549, | |
| "eval_runtime": 0.5872, | |
| "eval_samples_per_second": 74.937, | |
| "eval_steps_per_second": 3.406, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.5318968296051025, | |
| "eval_runtime": 0.5856, | |
| "eval_samples_per_second": 75.136, | |
| "eval_steps_per_second": 3.415, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.5474943518638611, | |
| "eval_runtime": 0.5881, | |
| "eval_samples_per_second": 74.821, | |
| "eval_steps_per_second": 3.401, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.5575869083404541, | |
| "eval_runtime": 0.5924, | |
| "eval_samples_per_second": 74.277, | |
| "eval_steps_per_second": 3.376, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.5440621972084045, | |
| "eval_runtime": 0.5934, | |
| "eval_samples_per_second": 74.15, | |
| "eval_steps_per_second": 3.37, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 2.033994436264038, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0836, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.5265628695487976, | |
| "eval_runtime": 0.5914, | |
| "eval_samples_per_second": 74.399, | |
| "eval_steps_per_second": 3.382, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.5046513676643372, | |
| "eval_runtime": 0.5949, | |
| "eval_samples_per_second": 73.962, | |
| "eval_steps_per_second": 3.362, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.4887816607952118, | |
| "eval_runtime": 0.5949, | |
| "eval_samples_per_second": 73.962, | |
| "eval_steps_per_second": 3.362, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.4824266731739044, | |
| "eval_runtime": 0.5995, | |
| "eval_samples_per_second": 73.396, | |
| "eval_steps_per_second": 3.336, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.4814087748527527, | |
| "eval_runtime": 0.5895, | |
| "eval_samples_per_second": 74.638, | |
| "eval_steps_per_second": 3.393, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 2.045288562774658, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0996, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.4823065996170044, | |
| "eval_runtime": 0.5936, | |
| "eval_samples_per_second": 74.122, | |
| "eval_steps_per_second": 3.369, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.48261451721191406, | |
| "eval_runtime": 0.6014, | |
| "eval_samples_per_second": 73.158, | |
| "eval_steps_per_second": 3.325, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.4841221868991852, | |
| "eval_runtime": 0.5979, | |
| "eval_samples_per_second": 73.595, | |
| "eval_steps_per_second": 3.345, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.48796969652175903, | |
| "eval_runtime": 0.5894, | |
| "eval_samples_per_second": 74.647, | |
| "eval_steps_per_second": 3.393, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.4878678619861603, | |
| "eval_runtime": 0.594, | |
| "eval_samples_per_second": 74.08, | |
| "eval_steps_per_second": 3.367, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 2.5146257877349854, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.086, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.48287439346313477, | |
| "eval_runtime": 0.5999, | |
| "eval_samples_per_second": 73.348, | |
| "eval_steps_per_second": 3.334, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.47976768016815186, | |
| "eval_runtime": 0.5906, | |
| "eval_samples_per_second": 74.506, | |
| "eval_steps_per_second": 3.387, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.4811286926269531, | |
| "eval_runtime": 0.6031, | |
| "eval_samples_per_second": 72.955, | |
| "eval_steps_per_second": 3.316, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.4819415211677551, | |
| "eval_runtime": 0.5867, | |
| "eval_samples_per_second": 74.992, | |
| "eval_steps_per_second": 3.409, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.48155146837234497, | |
| "eval_runtime": 0.5913, | |
| "eval_samples_per_second": 74.418, | |
| "eval_steps_per_second": 3.383, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 1.6985409259796143, | |
| "learning_rate": 0.0, | |
| "loss": 0.0745, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.48162177205085754, | |
| "eval_runtime": 0.5936, | |
| "eval_samples_per_second": 74.118, | |
| "eval_steps_per_second": 3.369, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.883089058199552e+18, | |
| "train_loss": 0.18109044224023818, | |
| "train_runtime": 1108.1238, | |
| "train_samples_per_second": 21.929, | |
| "train_steps_per_second": 0.18 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.3476676046848297, | |
| "eval_runtime": 0.6558, | |
| "eval_samples_per_second": 67.094, | |
| "eval_steps_per_second": 3.05, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.883089058199552e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |