| { | |
| "best_metric": 0.8354430379746836, | |
| "best_model_checkpoint": "beit-base-patch16-224-fold3/checkpoint-189", | |
| "epoch": 85.71428571428571, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8571428571428571, | |
| "eval_accuracy": 0.4810126582278481, | |
| "eval_loss": 0.841671347618103, | |
| "eval_runtime": 1.3079, | |
| "eval_samples_per_second": 60.4, | |
| "eval_steps_per_second": 2.294, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5822784810126582, | |
| "eval_loss": 0.6764000654220581, | |
| "eval_runtime": 1.2582, | |
| "eval_samples_per_second": 62.79, | |
| "eval_steps_per_second": 2.384, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 6.912475109100342, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.71, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.5316455696202531, | |
| "eval_loss": 0.7271922826766968, | |
| "eval_runtime": 1.2257, | |
| "eval_samples_per_second": 64.453, | |
| "eval_steps_per_second": 2.448, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6835443037974683, | |
| "eval_loss": 0.6122164726257324, | |
| "eval_runtime": 1.2293, | |
| "eval_samples_per_second": 64.264, | |
| "eval_steps_per_second": 2.44, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "eval_accuracy": 0.5949367088607594, | |
| "eval_loss": 0.6287635564804077, | |
| "eval_runtime": 1.2672, | |
| "eval_samples_per_second": 62.343, | |
| "eval_steps_per_second": 2.367, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 5.590318202972412, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6227, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5949367088607594, | |
| "eval_loss": 0.6549976468086243, | |
| "eval_runtime": 1.2556, | |
| "eval_samples_per_second": 62.919, | |
| "eval_steps_per_second": 2.389, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.6329113924050633, | |
| "eval_loss": 0.6240243315696716, | |
| "eval_runtime": 1.2458, | |
| "eval_samples_per_second": 63.411, | |
| "eval_steps_per_second": 2.408, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6708860759493671, | |
| "eval_loss": 0.5877071022987366, | |
| "eval_runtime": 1.2286, | |
| "eval_samples_per_second": 64.302, | |
| "eval_steps_per_second": 2.442, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 7.712191581726074, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5472, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.857142857142858, | |
| "eval_accuracy": 0.5822784810126582, | |
| "eval_loss": 0.7285463809967041, | |
| "eval_runtime": 1.2382, | |
| "eval_samples_per_second": 63.8, | |
| "eval_steps_per_second": 2.423, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5822784810126582, | |
| "eval_loss": 0.8305177092552185, | |
| "eval_runtime": 1.2577, | |
| "eval_samples_per_second": 62.812, | |
| "eval_steps_per_second": 2.385, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.5101767182350159, | |
| "eval_runtime": 1.2416, | |
| "eval_samples_per_second": 63.629, | |
| "eval_steps_per_second": 2.416, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 8.292058944702148, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.4766, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 0.5352110862731934, | |
| "eval_runtime": 1.2476, | |
| "eval_samples_per_second": 63.323, | |
| "eval_steps_per_second": 2.405, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 12.857142857142858, | |
| "eval_accuracy": 0.6962025316455697, | |
| "eval_loss": 0.5357041954994202, | |
| "eval_runtime": 1.2502, | |
| "eval_samples_per_second": 63.191, | |
| "eval_steps_per_second": 2.4, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6329113924050633, | |
| "eval_loss": 0.7417603731155396, | |
| "eval_runtime": 1.263, | |
| "eval_samples_per_second": 62.55, | |
| "eval_steps_per_second": 2.375, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "grad_norm": 13.762903213500977, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.408, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.6835443037974683, | |
| "eval_loss": 0.6150128841400146, | |
| "eval_runtime": 1.2653, | |
| "eval_samples_per_second": 62.435, | |
| "eval_steps_per_second": 2.371, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.48701009154319763, | |
| "eval_runtime": 1.2472, | |
| "eval_samples_per_second": 63.341, | |
| "eval_steps_per_second": 2.405, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 16.857142857142858, | |
| "eval_accuracy": 0.6962025316455697, | |
| "eval_loss": 0.642661452293396, | |
| "eval_runtime": 1.2332, | |
| "eval_samples_per_second": 64.061, | |
| "eval_steps_per_second": 2.433, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 7.023232936859131, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.4078, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.4822278916835785, | |
| "eval_runtime": 1.2496, | |
| "eval_samples_per_second": 63.22, | |
| "eval_steps_per_second": 2.401, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.4947206377983093, | |
| "eval_runtime": 1.2702, | |
| "eval_samples_per_second": 62.193, | |
| "eval_steps_per_second": 2.362, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.348341941833496, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.3478, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7088607594936709, | |
| "eval_loss": 0.6846950650215149, | |
| "eval_runtime": 1.248, | |
| "eval_samples_per_second": 63.3, | |
| "eval_steps_per_second": 2.404, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.857142857142858, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.6153857707977295, | |
| "eval_runtime": 1.2398, | |
| "eval_samples_per_second": 63.721, | |
| "eval_steps_per_second": 2.42, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.5383514761924744, | |
| "eval_runtime": 1.2457, | |
| "eval_samples_per_second": 63.418, | |
| "eval_steps_per_second": 2.408, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 8.417755126953125, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.3006, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.5938844084739685, | |
| "eval_runtime": 1.2794, | |
| "eval_samples_per_second": 61.747, | |
| "eval_steps_per_second": 2.345, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.5214145183563232, | |
| "eval_runtime": 1.2515, | |
| "eval_samples_per_second": 63.125, | |
| "eval_steps_per_second": 2.397, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 24.857142857142858, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.54517662525177, | |
| "eval_runtime": 1.2373, | |
| "eval_samples_per_second": 63.847, | |
| "eval_steps_per_second": 2.425, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "grad_norm": 7.407799243927002, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2977, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 0.6152894496917725, | |
| "eval_runtime": 1.2508, | |
| "eval_samples_per_second": 63.158, | |
| "eval_steps_per_second": 2.398, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.47301825881004333, | |
| "eval_runtime": 1.2451, | |
| "eval_samples_per_second": 63.449, | |
| "eval_steps_per_second": 2.409, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.48605620861053467, | |
| "eval_runtime": 1.2609, | |
| "eval_samples_per_second": 62.655, | |
| "eval_steps_per_second": 2.379, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 4.174748420715332, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2768, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 28.857142857142858, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.6705207228660583, | |
| "eval_runtime": 1.2466, | |
| "eval_samples_per_second": 63.37, | |
| "eval_steps_per_second": 2.406, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.6362075805664062, | |
| "eval_runtime": 1.2368, | |
| "eval_samples_per_second": 63.872, | |
| "eval_steps_per_second": 2.426, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.654819130897522, | |
| "eval_runtime": 1.2542, | |
| "eval_samples_per_second": 62.988, | |
| "eval_steps_per_second": 2.392, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 31.428571428571427, | |
| "grad_norm": 6.274244785308838, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.2348, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.5099911689758301, | |
| "eval_runtime": 1.2687, | |
| "eval_samples_per_second": 62.268, | |
| "eval_steps_per_second": 2.365, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 32.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7156453132629395, | |
| "eval_runtime": 1.2466, | |
| "eval_samples_per_second": 63.371, | |
| "eval_steps_per_second": 2.407, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.48589497804641724, | |
| "eval_runtime": 1.256, | |
| "eval_samples_per_second": 62.898, | |
| "eval_steps_per_second": 2.389, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 10.481765747070312, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.2199, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.8490024209022522, | |
| "eval_runtime": 1.2289, | |
| "eval_samples_per_second": 64.284, | |
| "eval_steps_per_second": 2.441, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.6094659566879272, | |
| "eval_runtime": 1.2637, | |
| "eval_samples_per_second": 62.513, | |
| "eval_steps_per_second": 2.374, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 36.857142857142854, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.6426916718482971, | |
| "eval_runtime": 1.2388, | |
| "eval_samples_per_second": 63.771, | |
| "eval_steps_per_second": 2.422, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 37.142857142857146, | |
| "grad_norm": 5.726836204528809, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.201, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.6283360123634338, | |
| "eval_runtime": 1.2579, | |
| "eval_samples_per_second": 62.804, | |
| "eval_steps_per_second": 2.385, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.8882798552513123, | |
| "eval_runtime": 1.2878, | |
| "eval_samples_per_second": 61.347, | |
| "eval_steps_per_second": 2.33, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 7.843417644500732, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.1868, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7146226167678833, | |
| "eval_runtime": 1.2507, | |
| "eval_samples_per_second": 63.163, | |
| "eval_steps_per_second": 2.399, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 40.857142857142854, | |
| "eval_accuracy": 0.6962025316455697, | |
| "eval_loss": 1.3800476789474487, | |
| "eval_runtime": 1.2652, | |
| "eval_samples_per_second": 62.441, | |
| "eval_steps_per_second": 2.371, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.5907676815986633, | |
| "eval_runtime": 1.2473, | |
| "eval_samples_per_second": 63.338, | |
| "eval_steps_per_second": 2.405, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "grad_norm": 5.149726390838623, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2011, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.6157855987548828, | |
| "eval_runtime": 1.2767, | |
| "eval_samples_per_second": 61.879, | |
| "eval_steps_per_second": 2.35, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.5477380156517029, | |
| "eval_runtime": 1.279, | |
| "eval_samples_per_second": 61.766, | |
| "eval_steps_per_second": 2.346, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 44.857142857142854, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.8354409337043762, | |
| "eval_runtime": 1.2731, | |
| "eval_samples_per_second": 62.053, | |
| "eval_steps_per_second": 2.356, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 9.848665237426758, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.1807, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.783032238483429, | |
| "eval_runtime": 1.2407, | |
| "eval_samples_per_second": 63.673, | |
| "eval_steps_per_second": 2.418, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.6327481269836426, | |
| "eval_runtime": 1.2476, | |
| "eval_samples_per_second": 63.321, | |
| "eval_steps_per_second": 2.405, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7857685685157776, | |
| "eval_runtime": 1.2776, | |
| "eval_samples_per_second": 61.837, | |
| "eval_steps_per_second": 2.348, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "grad_norm": 7.769134998321533, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.1579, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 48.857142857142854, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.8322492241859436, | |
| "eval_runtime": 1.2614, | |
| "eval_samples_per_second": 62.63, | |
| "eval_steps_per_second": 2.378, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7500829696655273, | |
| "eval_runtime": 1.2618, | |
| "eval_samples_per_second": 62.608, | |
| "eval_steps_per_second": 2.378, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.8303415775299072, | |
| "eval_runtime": 1.2565, | |
| "eval_samples_per_second": 62.873, | |
| "eval_steps_per_second": 2.388, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 4.354771614074707, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.2066, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.6830740571022034, | |
| "eval_runtime": 1.2586, | |
| "eval_samples_per_second": 62.767, | |
| "eval_steps_per_second": 2.384, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 52.857142857142854, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.7836630344390869, | |
| "eval_runtime": 1.2491, | |
| "eval_samples_per_second": 63.246, | |
| "eval_steps_per_second": 2.402, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.8354430379746836, | |
| "eval_loss": 0.5596601963043213, | |
| "eval_runtime": 1.2455, | |
| "eval_samples_per_second": 63.429, | |
| "eval_steps_per_second": 2.409, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 54.285714285714285, | |
| "grad_norm": 3.7188525199890137, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.1647, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.8354430379746836, | |
| "eval_loss": 0.5483526587486267, | |
| "eval_runtime": 1.254, | |
| "eval_samples_per_second": 62.999, | |
| "eval_steps_per_second": 2.392, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 1.0046799182891846, | |
| "eval_runtime": 1.2306, | |
| "eval_samples_per_second": 64.194, | |
| "eval_steps_per_second": 2.438, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 56.857142857142854, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.7815341949462891, | |
| "eval_runtime": 1.2396, | |
| "eval_samples_per_second": 63.732, | |
| "eval_steps_per_second": 2.42, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 3.9917469024658203, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1404, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6807761192321777, | |
| "eval_runtime": 1.2498, | |
| "eval_samples_per_second": 63.212, | |
| "eval_steps_per_second": 2.4, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 58.857142857142854, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 1.0068491697311401, | |
| "eval_runtime": 1.2712, | |
| "eval_samples_per_second": 62.144, | |
| "eval_steps_per_second": 2.36, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 6.757404327392578, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1451, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.7697803378105164, | |
| "eval_runtime": 1.2386, | |
| "eval_samples_per_second": 63.779, | |
| "eval_steps_per_second": 2.422, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 60.857142857142854, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.6494631171226501, | |
| "eval_runtime": 1.2538, | |
| "eval_samples_per_second": 63.011, | |
| "eval_steps_per_second": 2.393, | |
| "step": 213 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.8354430379746836, | |
| "eval_loss": 0.7065846920013428, | |
| "eval_runtime": 1.259, | |
| "eval_samples_per_second": 62.75, | |
| "eval_steps_per_second": 2.383, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 62.857142857142854, | |
| "grad_norm": 7.866526126861572, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.1341, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 62.857142857142854, | |
| "eval_accuracy": 0.8354430379746836, | |
| "eval_loss": 0.6249613761901855, | |
| "eval_runtime": 1.2401, | |
| "eval_samples_per_second": 63.705, | |
| "eval_steps_per_second": 2.419, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.5573354959487915, | |
| "eval_runtime": 1.2328, | |
| "eval_samples_per_second": 64.084, | |
| "eval_steps_per_second": 2.434, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 64.85714285714286, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.6051273941993713, | |
| "eval_runtime": 1.2693, | |
| "eval_samples_per_second": 62.24, | |
| "eval_steps_per_second": 2.364, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 65.71428571428571, | |
| "grad_norm": 3.092421531677246, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.127, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.7575756907463074, | |
| "eval_runtime": 1.2716, | |
| "eval_samples_per_second": 62.128, | |
| "eval_steps_per_second": 2.359, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 66.85714285714286, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.8296853303909302, | |
| "eval_runtime": 1.2431, | |
| "eval_samples_per_second": 63.552, | |
| "eval_steps_per_second": 2.413, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 1.0732265710830688, | |
| "eval_runtime": 1.2413, | |
| "eval_samples_per_second": 63.643, | |
| "eval_steps_per_second": 2.417, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 68.57142857142857, | |
| "grad_norm": 3.272094249725342, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1129, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 68.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 1.0503491163253784, | |
| "eval_runtime": 1.2344, | |
| "eval_samples_per_second": 64.0, | |
| "eval_steps_per_second": 2.43, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.7519514560699463, | |
| "eval_runtime": 1.2545, | |
| "eval_samples_per_second": 62.975, | |
| "eval_steps_per_second": 2.391, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 70.85714285714286, | |
| "eval_accuracy": 0.8354430379746836, | |
| "eval_loss": 0.682456374168396, | |
| "eval_runtime": 1.253, | |
| "eval_samples_per_second": 63.049, | |
| "eval_steps_per_second": 2.394, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 71.42857142857143, | |
| "grad_norm": 3.541659355163574, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.1205, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7002361416816711, | |
| "eval_runtime": 1.2614, | |
| "eval_samples_per_second": 62.63, | |
| "eval_steps_per_second": 2.378, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 72.85714285714286, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.742997944355011, | |
| "eval_runtime": 1.2676, | |
| "eval_samples_per_second": 62.322, | |
| "eval_steps_per_second": 2.367, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.760958731174469, | |
| "eval_runtime": 1.2386, | |
| "eval_samples_per_second": 63.781, | |
| "eval_steps_per_second": 2.422, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 74.28571428571429, | |
| "grad_norm": 5.87249755859375, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.1199, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 74.85714285714286, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.6854208707809448, | |
| "eval_runtime": 1.2531, | |
| "eval_samples_per_second": 63.044, | |
| "eval_steps_per_second": 2.394, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8354430379746836, | |
| "eval_loss": 0.6767454743385315, | |
| "eval_runtime": 1.2777, | |
| "eval_samples_per_second": 61.832, | |
| "eval_steps_per_second": 2.348, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 76.85714285714286, | |
| "eval_accuracy": 0.8354430379746836, | |
| "eval_loss": 0.6684638857841492, | |
| "eval_runtime": 1.2649, | |
| "eval_samples_per_second": 62.458, | |
| "eval_steps_per_second": 2.372, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 77.14285714285714, | |
| "grad_norm": 5.196723461151123, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1165, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7134199738502502, | |
| "eval_runtime": 1.2592, | |
| "eval_samples_per_second": 62.737, | |
| "eval_steps_per_second": 2.382, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 78.85714285714286, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.734351634979248, | |
| "eval_runtime": 1.2491, | |
| "eval_samples_per_second": 63.245, | |
| "eval_steps_per_second": 2.402, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 2.7902886867523193, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.1213, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.7403169274330139, | |
| "eval_runtime": 1.2505, | |
| "eval_samples_per_second": 63.176, | |
| "eval_steps_per_second": 2.399, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 80.85714285714286, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7817804217338562, | |
| "eval_runtime": 1.2417, | |
| "eval_samples_per_second": 63.625, | |
| "eval_steps_per_second": 2.416, | |
| "step": 283 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.76201993227005, | |
| "eval_runtime": 1.2665, | |
| "eval_samples_per_second": 62.375, | |
| "eval_steps_per_second": 2.369, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 82.85714285714286, | |
| "grad_norm": 3.400395393371582, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.1024, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 82.85714285714286, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.7539292573928833, | |
| "eval_runtime": 1.2393, | |
| "eval_samples_per_second": 63.745, | |
| "eval_steps_per_second": 2.421, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.7659382224082947, | |
| "eval_runtime": 1.2353, | |
| "eval_samples_per_second": 63.95, | |
| "eval_steps_per_second": 2.428, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 84.85714285714286, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7686355113983154, | |
| "eval_runtime": 1.2942, | |
| "eval_samples_per_second": 61.042, | |
| "eval_steps_per_second": 2.318, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "grad_norm": 5.54146146774292, | |
| "learning_rate": 0.0, | |
| "loss": 0.1109, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7686384916305542, | |
| "eval_runtime": 1.2604, | |
| "eval_samples_per_second": 62.677, | |
| "eval_steps_per_second": 2.38, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "step": 300, | |
| "total_flos": 2.9349165326823014e+18, | |
| "train_loss": 0.24998331944147745, | |
| "train_runtime": 1789.229, | |
| "train_samples_per_second": 24.703, | |
| "train_steps_per_second": 0.168 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "eval_accuracy": 0.8354430379746836, | |
| "eval_loss": 0.5596601963043213, | |
| "eval_runtime": 1.2452, | |
| "eval_samples_per_second": 63.445, | |
| "eval_steps_per_second": 2.409, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "total_flos": 2.9349165326823014e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |