| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 88.88888888888889, | |
| "eval_steps": 500, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8888888888888888, | |
| "eval_accuracy": 0.11578947368421053, | |
| "eval_loss": 1.6652907133102417, | |
| "eval_runtime": 1.8033, | |
| "eval_samples_per_second": 52.682, | |
| "eval_steps_per_second": 1.664, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4814814814814814, | |
| "grad_norm": 5.346019268035889, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.6915, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.925925925925926, | |
| "eval_accuracy": 0.23157894736842105, | |
| "eval_loss": 1.5941598415374756, | |
| "eval_runtime": 1.5205, | |
| "eval_samples_per_second": 62.481, | |
| "eval_steps_per_second": 1.973, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 8.961028099060059, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.5895, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "eval_accuracy": 0.3894736842105263, | |
| "eval_loss": 1.4918299913406372, | |
| "eval_runtime": 1.5278, | |
| "eval_samples_per_second": 62.179, | |
| "eval_steps_per_second": 1.964, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5052631578947369, | |
| "eval_loss": 1.363681674003601, | |
| "eval_runtime": 1.5355, | |
| "eval_samples_per_second": 61.868, | |
| "eval_steps_per_second": 1.954, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 12.23503303527832, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.433, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.888888888888889, | |
| "eval_accuracy": 0.6736842105263158, | |
| "eval_loss": 1.2241352796554565, | |
| "eval_runtime": 1.595, | |
| "eval_samples_per_second": 59.559, | |
| "eval_steps_per_second": 1.881, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 5.925925925925926, | |
| "grad_norm": 29.141326904296875, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.2158, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.925925925925926, | |
| "eval_accuracy": 0.7473684210526316, | |
| "eval_loss": 1.0278701782226562, | |
| "eval_runtime": 1.5753, | |
| "eval_samples_per_second": 60.307, | |
| "eval_steps_per_second": 1.904, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.962962962962963, | |
| "eval_accuracy": 0.8105263157894737, | |
| "eval_loss": 0.8302664756774902, | |
| "eval_runtime": 1.6713, | |
| "eval_samples_per_second": 56.842, | |
| "eval_steps_per_second": 1.795, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 7.407407407407407, | |
| "grad_norm": 28.541772842407227, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.9438, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8210526315789474, | |
| "eval_loss": 0.6755695343017578, | |
| "eval_runtime": 1.5959, | |
| "eval_samples_per_second": 59.527, | |
| "eval_steps_per_second": 1.88, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 14.663153648376465, | |
| "learning_rate": 5e-05, | |
| "loss": 0.727, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_loss": 0.5751134157180786, | |
| "eval_runtime": 1.7131, | |
| "eval_samples_per_second": 55.455, | |
| "eval_steps_per_second": 1.751, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 9.925925925925926, | |
| "eval_accuracy": 0.8631578947368421, | |
| "eval_loss": 0.4678628146648407, | |
| "eval_runtime": 1.5414, | |
| "eval_samples_per_second": 61.633, | |
| "eval_steps_per_second": 1.946, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 10.37037037037037, | |
| "grad_norm": 36.71305847167969, | |
| "learning_rate": 4.9074074074074075e-05, | |
| "loss": 0.5516, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 10.962962962962964, | |
| "eval_accuracy": 0.8526315789473684, | |
| "eval_loss": 0.4432494640350342, | |
| "eval_runtime": 1.5533, | |
| "eval_samples_per_second": 61.159, | |
| "eval_steps_per_second": 1.931, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 11.851851851851851, | |
| "grad_norm": 25.495798110961914, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.4337, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8526315789473684, | |
| "eval_loss": 0.38154730200767517, | |
| "eval_runtime": 1.7941, | |
| "eval_samples_per_second": 52.95, | |
| "eval_steps_per_second": 1.672, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 12.88888888888889, | |
| "eval_accuracy": 0.8842105263157894, | |
| "eval_loss": 0.36410775780677795, | |
| "eval_runtime": 1.5444, | |
| "eval_samples_per_second": 61.514, | |
| "eval_steps_per_second": 1.943, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 20.939546585083008, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.3757, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 13.925925925925926, | |
| "eval_accuracy": 0.8736842105263158, | |
| "eval_loss": 0.3233407735824585, | |
| "eval_runtime": 1.5363, | |
| "eval_samples_per_second": 61.836, | |
| "eval_steps_per_second": 1.953, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 14.814814814814815, | |
| "grad_norm": 87.37303924560547, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.3017, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 14.962962962962964, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.32304683327674866, | |
| "eval_runtime": 1.5993, | |
| "eval_samples_per_second": 59.403, | |
| "eval_steps_per_second": 1.876, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8842105263157894, | |
| "eval_loss": 0.30177825689315796, | |
| "eval_runtime": 1.5164, | |
| "eval_samples_per_second": 62.65, | |
| "eval_steps_per_second": 1.978, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 16.296296296296298, | |
| "grad_norm": 25.734861373901367, | |
| "learning_rate": 4.5370370370370374e-05, | |
| "loss": 0.2495, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 16.88888888888889, | |
| "eval_accuracy": 0.9052631578947369, | |
| "eval_loss": 0.34445926547050476, | |
| "eval_runtime": 1.5124, | |
| "eval_samples_per_second": 62.812, | |
| "eval_steps_per_second": 1.984, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 17.77777777777778, | |
| "grad_norm": 22.723478317260742, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.2177, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 17.925925925925927, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_loss": 0.29870516061782837, | |
| "eval_runtime": 1.6176, | |
| "eval_samples_per_second": 58.728, | |
| "eval_steps_per_second": 1.855, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 18.962962962962962, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_loss": 0.27269816398620605, | |
| "eval_runtime": 1.5486, | |
| "eval_samples_per_second": 61.345, | |
| "eval_steps_per_second": 1.937, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 19.25925925925926, | |
| "grad_norm": 21.11945343017578, | |
| "learning_rate": 4.351851851851852e-05, | |
| "loss": 0.1738, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8842105263157894, | |
| "eval_loss": 0.28645142912864685, | |
| "eval_runtime": 1.7081, | |
| "eval_samples_per_second": 55.618, | |
| "eval_steps_per_second": 1.756, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 20.74074074074074, | |
| "grad_norm": 55.72514343261719, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.1572, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 20.88888888888889, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.2645561993122101, | |
| "eval_runtime": 1.8502, | |
| "eval_samples_per_second": 51.345, | |
| "eval_steps_per_second": 1.621, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 21.925925925925927, | |
| "eval_accuracy": 0.9052631578947369, | |
| "eval_loss": 0.31001701951026917, | |
| "eval_runtime": 1.5502, | |
| "eval_samples_per_second": 61.282, | |
| "eval_steps_per_second": 1.935, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 22.22222222222222, | |
| "grad_norm": 11.173165321350098, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.1165, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 22.962962962962962, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.30389800667762756, | |
| "eval_runtime": 1.5599, | |
| "eval_samples_per_second": 60.903, | |
| "eval_steps_per_second": 1.923, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 23.703703703703702, | |
| "grad_norm": 13.963062286376953, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.1057, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9052631578947369, | |
| "eval_loss": 0.3022925555706024, | |
| "eval_runtime": 1.682, | |
| "eval_samples_per_second": 56.482, | |
| "eval_steps_per_second": 1.784, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 24.88888888888889, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.2254202663898468, | |
| "eval_runtime": 1.5031, | |
| "eval_samples_per_second": 63.201, | |
| "eval_steps_per_second": 1.996, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 25.185185185185187, | |
| "grad_norm": 22.49197769165039, | |
| "learning_rate": 3.981481481481482e-05, | |
| "loss": 0.0825, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 25.925925925925927, | |
| "eval_accuracy": 0.8736842105263158, | |
| "eval_loss": 0.3308357298374176, | |
| "eval_runtime": 1.5096, | |
| "eval_samples_per_second": 62.929, | |
| "eval_steps_per_second": 1.987, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 26.666666666666668, | |
| "grad_norm": 17.941728591918945, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.0795, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 26.962962962962962, | |
| "eval_accuracy": 0.9368421052631579, | |
| "eval_loss": 0.20397017896175385, | |
| "eval_runtime": 1.5389, | |
| "eval_samples_per_second": 61.731, | |
| "eval_steps_per_second": 1.949, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.21477854251861572, | |
| "eval_runtime": 1.5127, | |
| "eval_samples_per_second": 62.8, | |
| "eval_steps_per_second": 1.983, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 28.14814814814815, | |
| "grad_norm": 11.859210014343262, | |
| "learning_rate": 3.7962962962962964e-05, | |
| "loss": 0.072, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 28.88888888888889, | |
| "eval_accuracy": 0.8631578947368421, | |
| "eval_loss": 0.3449535667896271, | |
| "eval_runtime": 1.5393, | |
| "eval_samples_per_second": 61.716, | |
| "eval_steps_per_second": 1.949, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 29.62962962962963, | |
| "grad_norm": 17.80461883544922, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.0701, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 29.925925925925927, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.24177835881710052, | |
| "eval_runtime": 1.7581, | |
| "eval_samples_per_second": 54.036, | |
| "eval_steps_per_second": 1.706, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 30.962962962962962, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.24954235553741455, | |
| "eval_runtime": 1.5375, | |
| "eval_samples_per_second": 61.787, | |
| "eval_steps_per_second": 1.951, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 31.11111111111111, | |
| "grad_norm": 11.903740882873535, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.0635, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_loss": 0.3266756236553192, | |
| "eval_runtime": 1.5287, | |
| "eval_samples_per_second": 62.145, | |
| "eval_steps_per_second": 1.962, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 32.592592592592595, | |
| "grad_norm": 9.013089179992676, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.0537, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 32.888888888888886, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.3727841377258301, | |
| "eval_runtime": 1.7008, | |
| "eval_samples_per_second": 55.857, | |
| "eval_steps_per_second": 1.764, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 33.925925925925924, | |
| "eval_accuracy": 0.9052631578947369, | |
| "eval_loss": 0.28518009185791016, | |
| "eval_runtime": 1.5355, | |
| "eval_samples_per_second": 61.868, | |
| "eval_steps_per_second": 1.954, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 34.074074074074076, | |
| "grad_norm": 37.26246643066406, | |
| "learning_rate": 3.425925925925926e-05, | |
| "loss": 0.0607, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 34.96296296296296, | |
| "eval_accuracy": 0.9473684210526315, | |
| "eval_loss": 0.23858819901943207, | |
| "eval_runtime": 1.5204, | |
| "eval_samples_per_second": 62.485, | |
| "eval_steps_per_second": 1.973, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 35.55555555555556, | |
| "grad_norm": 14.775269508361816, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.052, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.20699043571949005, | |
| "eval_runtime": 1.6355, | |
| "eval_samples_per_second": 58.086, | |
| "eval_steps_per_second": 1.834, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 36.888888888888886, | |
| "eval_accuracy": 0.9473684210526315, | |
| "eval_loss": 0.18596884608268738, | |
| "eval_runtime": 1.5409, | |
| "eval_samples_per_second": 61.654, | |
| "eval_steps_per_second": 1.947, | |
| "step": 249 | |
| }, | |
| { | |
| "epoch": 37.03703703703704, | |
| "grad_norm": 7.913994312286377, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.049, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 37.925925925925924, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_loss": 0.3068939745426178, | |
| "eval_runtime": 1.5267, | |
| "eval_samples_per_second": 62.227, | |
| "eval_steps_per_second": 1.965, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 38.51851851851852, | |
| "grad_norm": 28.389638900756836, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.0578, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 38.96296296296296, | |
| "eval_accuracy": 0.8736842105263158, | |
| "eval_loss": 0.4477124810218811, | |
| "eval_runtime": 1.6335, | |
| "eval_samples_per_second": 58.158, | |
| "eval_steps_per_second": 1.837, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 66.57148742675781, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.0533, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_loss": 0.26121658086776733, | |
| "eval_runtime": 1.5219, | |
| "eval_samples_per_second": 62.422, | |
| "eval_steps_per_second": 1.971, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 40.888888888888886, | |
| "eval_accuracy": 0.8842105263157894, | |
| "eval_loss": 0.264914333820343, | |
| "eval_runtime": 1.509, | |
| "eval_samples_per_second": 62.957, | |
| "eval_steps_per_second": 1.988, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 41.48148148148148, | |
| "grad_norm": 5.600070953369141, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.0505, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 41.925925925925924, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.19498933851718903, | |
| "eval_runtime": 1.6324, | |
| "eval_samples_per_second": 58.195, | |
| "eval_steps_per_second": 1.838, | |
| "step": 283 | |
| }, | |
| { | |
| "epoch": 42.96296296296296, | |
| "grad_norm": 5.095949172973633, | |
| "learning_rate": 2.8703703703703706e-05, | |
| "loss": 0.0433, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 42.96296296296296, | |
| "eval_accuracy": 0.8842105263157894, | |
| "eval_loss": 0.29025542736053467, | |
| "eval_runtime": 1.5094, | |
| "eval_samples_per_second": 62.938, | |
| "eval_steps_per_second": 1.988, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.9368421052631579, | |
| "eval_loss": 0.2526479661464691, | |
| "eval_runtime": 1.523, | |
| "eval_samples_per_second": 62.376, | |
| "eval_steps_per_second": 1.97, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 44.44444444444444, | |
| "grad_norm": 9.107303619384766, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0395, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 44.888888888888886, | |
| "eval_accuracy": 0.8842105263157894, | |
| "eval_loss": 0.30155882239341736, | |
| "eval_runtime": 1.7275, | |
| "eval_samples_per_second": 54.991, | |
| "eval_steps_per_second": 1.737, | |
| "step": 303 | |
| }, | |
| { | |
| "epoch": 45.925925925925924, | |
| "grad_norm": 10.148141860961914, | |
| "learning_rate": 2.6851851851851855e-05, | |
| "loss": 0.035, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 45.925925925925924, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_loss": 0.3509025275707245, | |
| "eval_runtime": 1.5108, | |
| "eval_samples_per_second": 62.88, | |
| "eval_steps_per_second": 1.986, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 46.96296296296296, | |
| "eval_accuracy": 0.8842105263157894, | |
| "eval_loss": 0.29430651664733887, | |
| "eval_runtime": 1.5062, | |
| "eval_samples_per_second": 63.071, | |
| "eval_steps_per_second": 1.992, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 47.407407407407405, | |
| "grad_norm": 4.681646347045898, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.0335, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8842105263157894, | |
| "eval_loss": 0.2613106667995453, | |
| "eval_runtime": 1.6847, | |
| "eval_samples_per_second": 56.39, | |
| "eval_steps_per_second": 1.781, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 48.888888888888886, | |
| "grad_norm": 10.262737274169922, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0408, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 48.888888888888886, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.21650759875774384, | |
| "eval_runtime": 1.5245, | |
| "eval_samples_per_second": 62.315, | |
| "eval_steps_per_second": 1.968, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 49.925925925925924, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.28715887665748596, | |
| "eval_runtime": 1.5241, | |
| "eval_samples_per_second": 62.331, | |
| "eval_steps_per_second": 1.968, | |
| "step": 337 | |
| }, | |
| { | |
| "epoch": 50.370370370370374, | |
| "grad_norm": 9.619718551635742, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.0244, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 50.96296296296296, | |
| "eval_accuracy": 0.8842105263157894, | |
| "eval_loss": 0.31339362263679504, | |
| "eval_runtime": 1.6247, | |
| "eval_samples_per_second": 58.472, | |
| "eval_steps_per_second": 1.846, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 51.851851851851855, | |
| "grad_norm": 6.175246715545654, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.0323, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.3006380796432495, | |
| "eval_runtime": 1.5132, | |
| "eval_samples_per_second": 62.78, | |
| "eval_steps_per_second": 1.983, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 52.888888888888886, | |
| "eval_accuracy": 0.8736842105263158, | |
| "eval_loss": 0.37583690881729126, | |
| "eval_runtime": 1.5262, | |
| "eval_samples_per_second": 62.247, | |
| "eval_steps_per_second": 1.966, | |
| "step": 357 | |
| }, | |
| { | |
| "epoch": 53.333333333333336, | |
| "grad_norm": 11.230114936828613, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0241, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 53.925925925925924, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.3033463954925537, | |
| "eval_runtime": 1.6639, | |
| "eval_samples_per_second": 57.093, | |
| "eval_steps_per_second": 1.803, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 54.81481481481482, | |
| "grad_norm": 6.497807502746582, | |
| "learning_rate": 2.1296296296296296e-05, | |
| "loss": 0.0193, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 54.96296296296296, | |
| "eval_accuracy": 0.9368421052631579, | |
| "eval_loss": 0.27406617999076843, | |
| "eval_runtime": 1.5407, | |
| "eval_samples_per_second": 61.662, | |
| "eval_steps_per_second": 1.947, | |
| "step": 371 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.9368421052631579, | |
| "eval_loss": 0.1684454083442688, | |
| "eval_runtime": 1.5222, | |
| "eval_samples_per_second": 62.409, | |
| "eval_steps_per_second": 1.971, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 56.2962962962963, | |
| "grad_norm": 1.369285225868225, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.0273, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 56.888888888888886, | |
| "eval_accuracy": 0.9473684210526315, | |
| "eval_loss": 0.240325465798378, | |
| "eval_runtime": 1.6193, | |
| "eval_samples_per_second": 58.666, | |
| "eval_steps_per_second": 1.853, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 57.77777777777778, | |
| "grad_norm": 16.219879150390625, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.0244, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 57.925925925925924, | |
| "eval_accuracy": 0.9473684210526315, | |
| "eval_loss": 0.14995306730270386, | |
| "eval_runtime": 1.509, | |
| "eval_samples_per_second": 62.958, | |
| "eval_steps_per_second": 1.988, | |
| "step": 391 | |
| }, | |
| { | |
| "epoch": 58.96296296296296, | |
| "eval_accuracy": 0.9368421052631579, | |
| "eval_loss": 0.13768525421619415, | |
| "eval_runtime": 1.5002, | |
| "eval_samples_per_second": 63.323, | |
| "eval_steps_per_second": 2.0, | |
| "step": 398 | |
| }, | |
| { | |
| "epoch": 59.25925925925926, | |
| "grad_norm": 17.92609214782715, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.0268, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.18984580039978027, | |
| "eval_runtime": 1.5627, | |
| "eval_samples_per_second": 60.793, | |
| "eval_steps_per_second": 1.92, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 60.74074074074074, | |
| "grad_norm": 4.754770278930664, | |
| "learning_rate": 1.7592592592592595e-05, | |
| "loss": 0.0405, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 60.888888888888886, | |
| "eval_accuracy": 0.9052631578947369, | |
| "eval_loss": 0.17558255791664124, | |
| "eval_runtime": 1.5023, | |
| "eval_samples_per_second": 63.237, | |
| "eval_steps_per_second": 1.997, | |
| "step": 411 | |
| }, | |
| { | |
| "epoch": 61.925925925925924, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.1907454878091812, | |
| "eval_runtime": 1.5123, | |
| "eval_samples_per_second": 62.817, | |
| "eval_steps_per_second": 1.984, | |
| "step": 418 | |
| }, | |
| { | |
| "epoch": 62.22222222222222, | |
| "grad_norm": 6.852460861206055, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0219, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 62.96296296296296, | |
| "eval_accuracy": 0.9052631578947369, | |
| "eval_loss": 0.17904597520828247, | |
| "eval_runtime": 1.5634, | |
| "eval_samples_per_second": 60.763, | |
| "eval_steps_per_second": 1.919, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 63.7037037037037, | |
| "grad_norm": 17.433446884155273, | |
| "learning_rate": 1.574074074074074e-05, | |
| "loss": 0.0329, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.9368421052631579, | |
| "eval_loss": 0.18854853510856628, | |
| "eval_runtime": 1.5466, | |
| "eval_samples_per_second": 61.427, | |
| "eval_steps_per_second": 1.94, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 64.88888888888889, | |
| "eval_accuracy": 0.9368421052631579, | |
| "eval_loss": 0.15500715374946594, | |
| "eval_runtime": 1.5091, | |
| "eval_samples_per_second": 62.951, | |
| "eval_steps_per_second": 1.988, | |
| "step": 438 | |
| }, | |
| { | |
| "epoch": 65.18518518518519, | |
| "grad_norm": 12.199813842773438, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.019, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 65.92592592592592, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.18106068670749664, | |
| "eval_runtime": 1.5434, | |
| "eval_samples_per_second": 61.553, | |
| "eval_steps_per_second": 1.944, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 66.66666666666667, | |
| "grad_norm": 5.085737228393555, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0205, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 66.96296296296296, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.2165663242340088, | |
| "eval_runtime": 1.5356, | |
| "eval_samples_per_second": 61.864, | |
| "eval_steps_per_second": 1.954, | |
| "step": 452 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.9052631578947369, | |
| "eval_loss": 0.17012225091457367, | |
| "eval_runtime": 1.505, | |
| "eval_samples_per_second": 63.122, | |
| "eval_steps_per_second": 1.993, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 68.14814814814815, | |
| "grad_norm": 4.042739391326904, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.0232, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 68.88888888888889, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.21532098948955536, | |
| "eval_runtime": 1.4982, | |
| "eval_samples_per_second": 63.41, | |
| "eval_steps_per_second": 2.002, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 69.62962962962963, | |
| "grad_norm": 3.4432425498962402, | |
| "learning_rate": 1.2037037037037037e-05, | |
| "loss": 0.0269, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 69.92592592592592, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.22287000715732574, | |
| "eval_runtime": 1.5307, | |
| "eval_samples_per_second": 62.063, | |
| "eval_steps_per_second": 1.96, | |
| "step": 472 | |
| }, | |
| { | |
| "epoch": 70.96296296296296, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.2237367182970047, | |
| "eval_runtime": 1.5031, | |
| "eval_samples_per_second": 63.204, | |
| "eval_steps_per_second": 1.996, | |
| "step": 479 | |
| }, | |
| { | |
| "epoch": 71.11111111111111, | |
| "grad_norm": 9.640032768249512, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0306, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.9368421052631579, | |
| "eval_loss": 0.18282432854175568, | |
| "eval_runtime": 1.5, | |
| "eval_samples_per_second": 63.334, | |
| "eval_steps_per_second": 2.0, | |
| "step": 486 | |
| }, | |
| { | |
| "epoch": 72.5925925925926, | |
| "grad_norm": 9.28294849395752, | |
| "learning_rate": 1.0185185185185185e-05, | |
| "loss": 0.0298, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 72.88888888888889, | |
| "eval_accuracy": 0.9368421052631579, | |
| "eval_loss": 0.14476627111434937, | |
| "eval_runtime": 1.5255, | |
| "eval_samples_per_second": 62.276, | |
| "eval_steps_per_second": 1.967, | |
| "step": 492 | |
| }, | |
| { | |
| "epoch": 73.92592592592592, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.19477160274982452, | |
| "eval_runtime": 1.5341, | |
| "eval_samples_per_second": 61.925, | |
| "eval_steps_per_second": 1.956, | |
| "step": 499 | |
| }, | |
| { | |
| "epoch": 74.07407407407408, | |
| "grad_norm": 0.5367471575737, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.0154, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 74.96296296296296, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.25699615478515625, | |
| "eval_runtime": 1.5191, | |
| "eval_samples_per_second": 62.538, | |
| "eval_steps_per_second": 1.975, | |
| "step": 506 | |
| }, | |
| { | |
| "epoch": 75.55555555555556, | |
| "grad_norm": 16.879281997680664, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0193, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.24617379903793335, | |
| "eval_runtime": 1.5485, | |
| "eval_samples_per_second": 61.351, | |
| "eval_steps_per_second": 1.937, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 76.88888888888889, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.21939770877361298, | |
| "eval_runtime": 1.5513, | |
| "eval_samples_per_second": 61.24, | |
| "eval_steps_per_second": 1.934, | |
| "step": 519 | |
| }, | |
| { | |
| "epoch": 77.03703703703704, | |
| "grad_norm": 9.813831329345703, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.0188, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 77.92592592592592, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.22538267076015472, | |
| "eval_runtime": 1.5394, | |
| "eval_samples_per_second": 61.71, | |
| "eval_steps_per_second": 1.949, | |
| "step": 526 | |
| }, | |
| { | |
| "epoch": 78.51851851851852, | |
| "grad_norm": 6.672292709350586, | |
| "learning_rate": 6.481481481481481e-06, | |
| "loss": 0.0198, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 78.96296296296296, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.192366823554039, | |
| "eval_runtime": 1.511, | |
| "eval_samples_per_second": 62.873, | |
| "eval_steps_per_second": 1.985, | |
| "step": 533 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 12.444893836975098, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0147, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.9368421052631579, | |
| "eval_loss": 0.15250863134860992, | |
| "eval_runtime": 1.5101, | |
| "eval_samples_per_second": 62.91, | |
| "eval_steps_per_second": 1.987, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 80.88888888888889, | |
| "eval_accuracy": 0.9473684210526315, | |
| "eval_loss": 0.13136789202690125, | |
| "eval_runtime": 1.5108, | |
| "eval_samples_per_second": 62.88, | |
| "eval_steps_per_second": 1.986, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 81.48148148148148, | |
| "grad_norm": 5.455392837524414, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.0282, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 81.92592592592592, | |
| "eval_accuracy": 0.9368421052631579, | |
| "eval_loss": 0.13807313144207, | |
| "eval_runtime": 1.5108, | |
| "eval_samples_per_second": 62.881, | |
| "eval_steps_per_second": 1.986, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 82.96296296296296, | |
| "grad_norm": 0.557128369808197, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0168, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 82.96296296296296, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.14955918490886688, | |
| "eval_runtime": 1.5176, | |
| "eval_samples_per_second": 62.599, | |
| "eval_steps_per_second": 1.977, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.1806280016899109, | |
| "eval_runtime": 1.5082, | |
| "eval_samples_per_second": 62.991, | |
| "eval_steps_per_second": 1.989, | |
| "step": 567 | |
| }, | |
| { | |
| "epoch": 84.44444444444444, | |
| "grad_norm": 6.701883316040039, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.018, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 84.88888888888889, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.20803439617156982, | |
| "eval_runtime": 1.5233, | |
| "eval_samples_per_second": 62.366, | |
| "eval_steps_per_second": 1.969, | |
| "step": 573 | |
| }, | |
| { | |
| "epoch": 85.92592592592592, | |
| "grad_norm": 5.107237339019775, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.0172, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 85.92592592592592, | |
| "eval_accuracy": 0.9157894736842105, | |
| "eval_loss": 0.21987786889076233, | |
| "eval_runtime": 1.5032, | |
| "eval_samples_per_second": 63.198, | |
| "eval_steps_per_second": 1.996, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 86.96296296296296, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.19394682347774506, | |
| "eval_runtime": 1.5704, | |
| "eval_samples_per_second": 60.494, | |
| "eval_steps_per_second": 1.91, | |
| "step": 587 | |
| }, | |
| { | |
| "epoch": 87.4074074074074, | |
| "grad_norm": 13.376716613769531, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 0.0117, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.18152017891407013, | |
| "eval_runtime": 1.5369, | |
| "eval_samples_per_second": 61.812, | |
| "eval_steps_per_second": 1.952, | |
| "step": 594 | |
| }, | |
| { | |
| "epoch": 88.88888888888889, | |
| "grad_norm": 8.48166561126709, | |
| "learning_rate": 0.0, | |
| "loss": 0.0149, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 88.88888888888889, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_loss": 0.1811211258172989, | |
| "eval_runtime": 1.529, | |
| "eval_samples_per_second": 62.13, | |
| "eval_steps_per_second": 1.962, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 88.88888888888889, | |
| "step": 600, | |
| "total_flos": 1.9132429834630103e+18, | |
| "train_loss": 0.19732174752900997, | |
| "train_runtime": 1871.4873, | |
| "train_samples_per_second": 45.686, | |
| "train_steps_per_second": 0.321 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "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": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.9132429834630103e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |