| { | |
| "best_metric": 1.0, | |
| "best_model_checkpoint": "./swin-soiral/checkpoint-72", | |
| "epoch": 100.0, | |
| "eval_steps": 500, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.7685546875, | |
| "eval_runtime": 0.2552, | |
| "eval_samples_per_second": 47.013, | |
| "eval_steps_per_second": 7.835, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.7776, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.7459920048713684, | |
| "eval_runtime": 0.2163, | |
| "eval_samples_per_second": 55.485, | |
| "eval_steps_per_second": 9.247, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 2.25e-06, | |
| "loss": 0.7839, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.71142578125, | |
| "eval_runtime": 0.2223, | |
| "eval_samples_per_second": 53.982, | |
| "eval_steps_per_second": 8.997, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.6965, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.6708170771598816, | |
| "eval_runtime": 0.1987, | |
| "eval_samples_per_second": 60.404, | |
| "eval_steps_per_second": 10.067, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.75e-06, | |
| "loss": 0.6778, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.62982177734375, | |
| "eval_runtime": 0.1998, | |
| "eval_samples_per_second": 60.045, | |
| "eval_steps_per_second": 10.008, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.5903117060661316, | |
| "eval_runtime": 0.2941, | |
| "eval_samples_per_second": 40.798, | |
| "eval_steps_per_second": 6.8, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 6e-06, | |
| "loss": 0.6297, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.5383097529411316, | |
| "eval_runtime": 0.2066, | |
| "eval_samples_per_second": 58.089, | |
| "eval_steps_per_second": 9.682, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 7.25e-06, | |
| "loss": 0.5812, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.4876708984375, | |
| "eval_runtime": 0.2154, | |
| "eval_samples_per_second": 55.709, | |
| "eval_steps_per_second": 9.285, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "learning_rate": 8.5e-06, | |
| "loss": 0.5141, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.42340087890625, | |
| "eval_runtime": 0.2045, | |
| "eval_samples_per_second": 58.685, | |
| "eval_steps_per_second": 9.781, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 9.75e-06, | |
| "loss": 0.4186, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.3495737612247467, | |
| "eval_runtime": 0.2143, | |
| "eval_samples_per_second": 55.994, | |
| "eval_steps_per_second": 9.332, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8333333333333334, | |
| "eval_loss": 0.2956085205078125, | |
| "eval_runtime": 0.204, | |
| "eval_samples_per_second": 58.832, | |
| "eval_steps_per_second": 9.805, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 9.88888888888889e-06, | |
| "loss": 0.3791, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.26324209570884705, | |
| "eval_runtime": 0.2276, | |
| "eval_samples_per_second": 52.73, | |
| "eval_steps_per_second": 8.788, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 9.75e-06, | |
| "loss": 0.2906, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.21923191845417023, | |
| "eval_runtime": 0.2866, | |
| "eval_samples_per_second": 41.866, | |
| "eval_steps_per_second": 6.978, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "learning_rate": 9.611111111111112e-06, | |
| "loss": 0.2247, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.18271827697753906, | |
| "eval_runtime": 0.2073, | |
| "eval_samples_per_second": 57.879, | |
| "eval_steps_per_second": 9.647, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 9.472222222222223e-06, | |
| "loss": 0.1724, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.1536579132080078, | |
| "eval_runtime": 0.3096, | |
| "eval_samples_per_second": 38.763, | |
| "eval_steps_per_second": 6.461, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.11898771673440933, | |
| "eval_runtime": 0.207, | |
| "eval_samples_per_second": 57.977, | |
| "eval_steps_per_second": 9.663, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.1077, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.12817128002643585, | |
| "eval_runtime": 0.2101, | |
| "eval_samples_per_second": 57.105, | |
| "eval_steps_per_second": 9.517, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "learning_rate": 9.194444444444445e-06, | |
| "loss": 0.1336, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0798807144165039, | |
| "eval_runtime": 0.3334, | |
| "eval_samples_per_second": 35.996, | |
| "eval_steps_per_second": 5.999, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "learning_rate": 9.083333333333333e-06, | |
| "loss": 0.0656, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.05580584332346916, | |
| "eval_runtime": 0.206, | |
| "eval_samples_per_second": 58.24, | |
| "eval_steps_per_second": 9.707, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 8.944444444444446e-06, | |
| "loss": 0.0564, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.046213943511247635, | |
| "eval_runtime": 0.3162, | |
| "eval_samples_per_second": 37.95, | |
| "eval_steps_per_second": 6.325, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.05256287381052971, | |
| "eval_runtime": 0.2134, | |
| "eval_samples_per_second": 56.23, | |
| "eval_steps_per_second": 9.372, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "learning_rate": 8.805555555555557e-06, | |
| "loss": 0.0703, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.03809436038136482, | |
| "eval_runtime": 0.2, | |
| "eval_samples_per_second": 60.004, | |
| "eval_steps_per_second": 10.001, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "learning_rate": 8.666666666666668e-06, | |
| "loss": 0.044, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.01835465431213379, | |
| "eval_runtime": 0.1962, | |
| "eval_samples_per_second": 61.174, | |
| "eval_steps_per_second": 10.196, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 23.75, | |
| "learning_rate": 8.527777777777779e-06, | |
| "loss": 0.0239, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.020169338211417198, | |
| "eval_runtime": 0.2151, | |
| "eval_samples_per_second": 55.779, | |
| "eval_steps_per_second": 9.297, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 8.38888888888889e-06, | |
| "loss": 0.0214, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.024330079555511475, | |
| "eval_runtime": 0.2292, | |
| "eval_samples_per_second": 52.363, | |
| "eval_steps_per_second": 8.727, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.011685073375701904, | |
| "eval_runtime": 0.2099, | |
| "eval_samples_per_second": 57.169, | |
| "eval_steps_per_second": 9.528, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 26.25, | |
| "learning_rate": 8.25e-06, | |
| "loss": 0.031, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.008951862342655659, | |
| "eval_runtime": 0.2024, | |
| "eval_samples_per_second": 59.277, | |
| "eval_steps_per_second": 9.88, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "learning_rate": 8.111111111111112e-06, | |
| "loss": 0.0334, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.03853602334856987, | |
| "eval_runtime": 0.2061, | |
| "eval_samples_per_second": 58.221, | |
| "eval_steps_per_second": 9.703, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 28.75, | |
| "learning_rate": 7.972222222222224e-06, | |
| "loss": 0.0046, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.03924691677093506, | |
| "eval_runtime": 0.3171, | |
| "eval_samples_per_second": 37.847, | |
| "eval_steps_per_second": 6.308, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 7.833333333333333e-06, | |
| "loss": 0.0051, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0031117696780711412, | |
| "eval_runtime": 0.217, | |
| "eval_samples_per_second": 55.303, | |
| "eval_steps_per_second": 9.217, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0025899012107402086, | |
| "eval_runtime": 0.318, | |
| "eval_samples_per_second": 37.734, | |
| "eval_steps_per_second": 6.289, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 31.25, | |
| "learning_rate": 7.694444444444446e-06, | |
| "loss": 0.0045, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.002510249614715576, | |
| "eval_runtime": 0.2034, | |
| "eval_samples_per_second": 58.986, | |
| "eval_steps_per_second": 9.831, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 32.5, | |
| "learning_rate": 7.555555555555556e-06, | |
| "loss": 0.0133, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.07476559281349182, | |
| "eval_runtime": 0.2934, | |
| "eval_samples_per_second": 40.893, | |
| "eval_steps_per_second": 6.816, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 33.75, | |
| "learning_rate": 7.416666666666668e-06, | |
| "loss": 0.0014, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.16315531730651855, | |
| "eval_runtime": 0.2053, | |
| "eval_samples_per_second": 58.463, | |
| "eval_steps_per_second": 9.744, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "learning_rate": 7.277777777777778e-06, | |
| "loss": 0.0134, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.006152550224214792, | |
| "eval_runtime": 0.2135, | |
| "eval_samples_per_second": 56.219, | |
| "eval_steps_per_second": 9.37, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.005291670560836792, | |
| "eval_runtime": 0.2054, | |
| "eval_samples_per_second": 58.422, | |
| "eval_steps_per_second": 9.737, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 36.25, | |
| "learning_rate": 7.13888888888889e-06, | |
| "loss": 0.0365, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0365130789577961, | |
| "eval_runtime": 0.2268, | |
| "eval_samples_per_second": 52.911, | |
| "eval_steps_per_second": 8.819, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 37.5, | |
| "learning_rate": 7e-06, | |
| "loss": 0.001, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0017459988594055176, | |
| "eval_runtime": 0.2143, | |
| "eval_samples_per_second": 56.005, | |
| "eval_steps_per_second": 9.334, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 38.75, | |
| "learning_rate": 6.88888888888889e-06, | |
| "loss": 0.0503, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.05090367794036865, | |
| "eval_runtime": 0.208, | |
| "eval_samples_per_second": 57.695, | |
| "eval_steps_per_second": 9.616, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 6.750000000000001e-06, | |
| "loss": 0.0094, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.07808921486139297, | |
| "eval_runtime": 0.2086, | |
| "eval_samples_per_second": 57.523, | |
| "eval_steps_per_second": 9.587, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.22892427444458008, | |
| "eval_runtime": 0.2158, | |
| "eval_samples_per_second": 55.599, | |
| "eval_steps_per_second": 9.267, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 41.25, | |
| "learning_rate": 6.6111111111111115e-06, | |
| "loss": 0.0231, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.16741518676280975, | |
| "eval_runtime": 0.2199, | |
| "eval_samples_per_second": 54.565, | |
| "eval_steps_per_second": 9.094, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 42.5, | |
| "learning_rate": 6.472222222222223e-06, | |
| "loss": 0.0002, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.25974369049072266, | |
| "eval_runtime": 0.297, | |
| "eval_samples_per_second": 40.411, | |
| "eval_steps_per_second": 6.735, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 43.75, | |
| "learning_rate": 6.333333333333333e-06, | |
| "loss": 0.0785, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.2814592123031616, | |
| "eval_runtime": 0.1943, | |
| "eval_samples_per_second": 61.754, | |
| "eval_steps_per_second": 10.292, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "learning_rate": 6.194444444444445e-06, | |
| "loss": 0.0009, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.2058378905057907, | |
| "eval_runtime": 0.3052, | |
| "eval_samples_per_second": 39.313, | |
| "eval_steps_per_second": 6.552, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.000926971435546875, | |
| "eval_runtime": 0.2098, | |
| "eval_samples_per_second": 57.192, | |
| "eval_steps_per_second": 9.532, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 46.25, | |
| "learning_rate": 6.055555555555555e-06, | |
| "loss": 0.0395, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0010882416972890496, | |
| "eval_runtime": 0.3081, | |
| "eval_samples_per_second": 38.948, | |
| "eval_steps_per_second": 6.491, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 47.5, | |
| "learning_rate": 5.916666666666667e-06, | |
| "loss": 0.0158, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.000389377266401425, | |
| "eval_runtime": 0.2087, | |
| "eval_samples_per_second": 57.493, | |
| "eval_steps_per_second": 9.582, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 48.75, | |
| "learning_rate": 5.777777777777778e-06, | |
| "loss": 0.001, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0034777026157826185, | |
| "eval_runtime": 0.2113, | |
| "eval_samples_per_second": 56.801, | |
| "eval_steps_per_second": 9.467, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 5.638888888888889e-06, | |
| "loss": 0.0027, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0018774966010823846, | |
| "eval_runtime": 0.1971, | |
| "eval_samples_per_second": 60.883, | |
| "eval_steps_per_second": 10.147, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0005501409177668393, | |
| "eval_runtime": 0.2093, | |
| "eval_samples_per_second": 57.332, | |
| "eval_steps_per_second": 9.555, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 51.25, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.0005, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0013666549930348992, | |
| "eval_runtime": 0.2027, | |
| "eval_samples_per_second": 59.215, | |
| "eval_steps_per_second": 9.869, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 52.5, | |
| "learning_rate": 5.361111111111112e-06, | |
| "loss": 0.003, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0018307765712961555, | |
| "eval_runtime": 0.2153, | |
| "eval_samples_per_second": 55.739, | |
| "eval_steps_per_second": 9.29, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 53.75, | |
| "learning_rate": 5.2222222222222226e-06, | |
| "loss": 0.0039, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.01105162501335144, | |
| "eval_runtime": 0.2005, | |
| "eval_samples_per_second": 59.849, | |
| "eval_steps_per_second": 9.975, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "learning_rate": 5.0833333333333335e-06, | |
| "loss": 0.0081, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.021254947409033775, | |
| "eval_runtime": 0.2174, | |
| "eval_samples_per_second": 55.192, | |
| "eval_steps_per_second": 9.199, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.014328837394714355, | |
| "eval_runtime": 0.2399, | |
| "eval_samples_per_second": 50.017, | |
| "eval_steps_per_second": 8.336, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 56.25, | |
| "learning_rate": 4.944444444444445e-06, | |
| "loss": 0.0001, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.00964482594281435, | |
| "eval_runtime": 0.2188, | |
| "eval_samples_per_second": 54.838, | |
| "eval_steps_per_second": 9.14, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 57.5, | |
| "learning_rate": 4.805555555555556e-06, | |
| "loss": 0.0002, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0017709335079416633, | |
| "eval_runtime": 0.2916, | |
| "eval_samples_per_second": 41.158, | |
| "eval_steps_per_second": 6.86, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 58.75, | |
| "learning_rate": 4.666666666666667e-06, | |
| "loss": 0.0035, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0005045731668360531, | |
| "eval_runtime": 0.2036, | |
| "eval_samples_per_second": 58.93, | |
| "eval_steps_per_second": 9.822, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "learning_rate": 4.527777777777778e-06, | |
| "loss": 0.0049, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0014440218219533563, | |
| "eval_runtime": 0.3276, | |
| "eval_samples_per_second": 36.631, | |
| "eval_steps_per_second": 6.105, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0020179252605885267, | |
| "eval_runtime": 0.2179, | |
| "eval_samples_per_second": 55.076, | |
| "eval_steps_per_second": 9.179, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 61.25, | |
| "learning_rate": 4.388888888888889e-06, | |
| "loss": 0.0006, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0003785987792070955, | |
| "eval_runtime": 0.2267, | |
| "eval_samples_per_second": 52.937, | |
| "eval_steps_per_second": 8.823, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 62.5, | |
| "learning_rate": 4.25e-06, | |
| "loss": 0.0001, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.00021861989807803184, | |
| "eval_runtime": 0.2108, | |
| "eval_samples_per_second": 56.934, | |
| "eval_steps_per_second": 9.489, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 63.75, | |
| "learning_rate": 4.111111111111111e-06, | |
| "loss": 0.0272, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0002175172121496871, | |
| "eval_runtime": 0.2182, | |
| "eval_samples_per_second": 55.001, | |
| "eval_steps_per_second": 9.167, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "learning_rate": 3.972222222222223e-06, | |
| "loss": 0.0016, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0003321866097394377, | |
| "eval_runtime": 0.1986, | |
| "eval_samples_per_second": 60.421, | |
| "eval_steps_per_second": 10.07, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0012783011188730597, | |
| "eval_runtime": 0.2135, | |
| "eval_samples_per_second": 56.201, | |
| "eval_steps_per_second": 9.367, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 66.25, | |
| "learning_rate": 3.833333333333334e-06, | |
| "loss": 0.0002, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.004446456674486399, | |
| "eval_runtime": 0.2127, | |
| "eval_samples_per_second": 56.422, | |
| "eval_steps_per_second": 9.404, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 67.5, | |
| "learning_rate": 3.694444444444445e-06, | |
| "loss": 0.0001, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.013481984846293926, | |
| "eval_runtime": 0.2043, | |
| "eval_samples_per_second": 58.732, | |
| "eval_steps_per_second": 9.789, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 68.75, | |
| "learning_rate": 3.555555555555556e-06, | |
| "loss": 0.0002, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.012796193361282349, | |
| "eval_runtime": 0.2261, | |
| "eval_samples_per_second": 53.079, | |
| "eval_steps_per_second": 8.846, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "learning_rate": 3.416666666666667e-06, | |
| "loss": 0.0002, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.01079349685460329, | |
| "eval_runtime": 0.1991, | |
| "eval_samples_per_second": 60.277, | |
| "eval_steps_per_second": 10.046, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.008993417024612427, | |
| "eval_runtime": 0.2933, | |
| "eval_samples_per_second": 40.92, | |
| "eval_steps_per_second": 6.82, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 71.25, | |
| "learning_rate": 3.277777777777778e-06, | |
| "loss": 0.0001, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.006850639823824167, | |
| "eval_runtime": 0.2074, | |
| "eval_samples_per_second": 57.85, | |
| "eval_steps_per_second": 9.642, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 72.5, | |
| "learning_rate": 3.138888888888889e-06, | |
| "loss": 0.0001, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0049311816692352295, | |
| "eval_runtime": 0.2813, | |
| "eval_samples_per_second": 42.663, | |
| "eval_steps_per_second": 7.111, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 73.75, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0002, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0031823813915252686, | |
| "eval_runtime": 0.2076, | |
| "eval_samples_per_second": 57.79, | |
| "eval_steps_per_second": 9.632, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 2.861111111111111e-06, | |
| "loss": 0.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0018174747237935662, | |
| "eval_runtime": 0.1974, | |
| "eval_samples_per_second": 60.778, | |
| "eval_steps_per_second": 10.13, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0033631324768066406, | |
| "eval_runtime": 0.2244, | |
| "eval_samples_per_second": 53.472, | |
| "eval_steps_per_second": 8.912, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 76.25, | |
| "learning_rate": 2.7222222222222224e-06, | |
| "loss": 0.0003, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.005034655332565308, | |
| "eval_runtime": 0.2054, | |
| "eval_samples_per_second": 58.436, | |
| "eval_steps_per_second": 9.739, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 77.5, | |
| "learning_rate": 2.5833333333333337e-06, | |
| "loss": 0.0001, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0060021779499948025, | |
| "eval_runtime": 0.2165, | |
| "eval_samples_per_second": 55.438, | |
| "eval_steps_per_second": 9.24, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 78.75, | |
| "learning_rate": 2.4444444444444447e-06, | |
| "loss": 0.0, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.006388425827026367, | |
| "eval_runtime": 0.2076, | |
| "eval_samples_per_second": 57.793, | |
| "eval_steps_per_second": 9.632, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "learning_rate": 2.305555555555556e-06, | |
| "loss": 0.0001, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0061742267571389675, | |
| "eval_runtime": 0.2188, | |
| "eval_samples_per_second": 54.852, | |
| "eval_steps_per_second": 9.142, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.005842983722686768, | |
| "eval_runtime": 0.2181, | |
| "eval_samples_per_second": 55.023, | |
| "eval_steps_per_second": 9.171, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 81.25, | |
| "learning_rate": 2.166666666666667e-06, | |
| "loss": 0.0001, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0059954822063446045, | |
| "eval_runtime": 0.2026, | |
| "eval_samples_per_second": 59.232, | |
| "eval_steps_per_second": 9.872, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 82.5, | |
| "learning_rate": 2.027777777777778e-06, | |
| "loss": 0.0001, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.006137927528470755, | |
| "eval_runtime": 0.2087, | |
| "eval_samples_per_second": 57.509, | |
| "eval_steps_per_second": 9.585, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 83.75, | |
| "learning_rate": 1.888888888888889e-06, | |
| "loss": 0.0, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.007910847663879395, | |
| "eval_runtime": 0.2136, | |
| "eval_samples_per_second": 56.177, | |
| "eval_steps_per_second": 9.363, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "learning_rate": 1.75e-06, | |
| "loss": 0.0005, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.012745578773319721, | |
| "eval_runtime": 0.2199, | |
| "eval_samples_per_second": 54.578, | |
| "eval_steps_per_second": 9.096, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.021893838420510292, | |
| "eval_runtime": 0.2149, | |
| "eval_samples_per_second": 55.845, | |
| "eval_steps_per_second": 9.307, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 86.25, | |
| "learning_rate": 1.6111111111111113e-06, | |
| "loss": 0.0002, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.031531721353530884, | |
| "eval_runtime": 0.2233, | |
| "eval_samples_per_second": 53.743, | |
| "eval_steps_per_second": 8.957, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 87.5, | |
| "learning_rate": 1.4722222222222225e-06, | |
| "loss": 0.0001, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.03811268135905266, | |
| "eval_runtime": 0.2096, | |
| "eval_samples_per_second": 57.255, | |
| "eval_steps_per_second": 9.542, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 88.75, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 0.0001, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.04021235182881355, | |
| "eval_runtime": 0.2105, | |
| "eval_samples_per_second": 57.008, | |
| "eval_steps_per_second": 9.501, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "learning_rate": 1.1944444444444446e-06, | |
| "loss": 0.0, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.040413498878479004, | |
| "eval_runtime": 0.2044, | |
| "eval_samples_per_second": 58.696, | |
| "eval_steps_per_second": 9.783, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0374186746776104, | |
| "eval_runtime": 0.316, | |
| "eval_samples_per_second": 37.973, | |
| "eval_steps_per_second": 6.329, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 91.25, | |
| "learning_rate": 1.0555555555555557e-06, | |
| "loss": 0.0, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.03140836954116821, | |
| "eval_runtime": 0.2142, | |
| "eval_samples_per_second": 56.034, | |
| "eval_steps_per_second": 9.339, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 92.5, | |
| "learning_rate": 9.166666666666666e-07, | |
| "loss": 0.0003, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.01769857667386532, | |
| "eval_runtime": 0.2018, | |
| "eval_samples_per_second": 59.478, | |
| "eval_steps_per_second": 9.913, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 93.75, | |
| "learning_rate": 7.777777777777779e-07, | |
| "loss": 0.0001, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.011687318794429302, | |
| "eval_runtime": 0.2027, | |
| "eval_samples_per_second": 59.211, | |
| "eval_steps_per_second": 9.869, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "learning_rate": 6.388888888888889e-07, | |
| "loss": 0.0001, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0088284807279706, | |
| "eval_runtime": 0.2065, | |
| "eval_samples_per_second": 58.106, | |
| "eval_steps_per_second": 9.684, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.007670491933822632, | |
| "eval_runtime": 0.2024, | |
| "eval_samples_per_second": 59.303, | |
| "eval_steps_per_second": 9.884, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 96.25, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 0.0, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.00695762038230896, | |
| "eval_runtime": 0.2135, | |
| "eval_samples_per_second": 56.218, | |
| "eval_steps_per_second": 9.37, | |
| "step": 388 | |
| }, | |
| { | |
| "epoch": 97.5, | |
| "learning_rate": 3.611111111111111e-07, | |
| "loss": 0.0001, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0064276158809661865, | |
| "eval_runtime": 0.2059, | |
| "eval_samples_per_second": 58.292, | |
| "eval_steps_per_second": 9.715, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 98.75, | |
| "learning_rate": 2.2222222222222224e-07, | |
| "loss": 0.0001, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.006080567836761475, | |
| "eval_runtime": 0.2091, | |
| "eval_samples_per_second": 57.386, | |
| "eval_steps_per_second": 9.564, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 8.333333333333334e-08, | |
| "loss": 0.0001, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.005917658563703299, | |
| "eval_runtime": 0.2205, | |
| "eval_samples_per_second": 54.429, | |
| "eval_steps_per_second": 9.072, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 400, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "total_flos": 4.70073333768192e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |