|
{ |
|
"best_metric": 0.40983906388282776, |
|
"best_model_checkpoint": "mikhail-panzo/fil_b64_le5_s8000/checkpoint-5500", |
|
"epoch": 326.0869565217391, |
|
"eval_steps": 500, |
|
"global_step": 7500, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 2.1739130434782608, |
|
"grad_norm": 4.140745639801025, |
|
"learning_rate": 2.4500000000000004e-07, |
|
"loss": 0.8272, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 4.3478260869565215, |
|
"grad_norm": 1.78262197971344, |
|
"learning_rate": 4.95e-07, |
|
"loss": 0.7463, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 6.521739130434782, |
|
"grad_norm": 4.001489639282227, |
|
"learning_rate": 7.450000000000001e-07, |
|
"loss": 0.7504, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 8.695652173913043, |
|
"grad_norm": 1.0013818740844727, |
|
"learning_rate": 9.950000000000002e-07, |
|
"loss": 0.7295, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 10.869565217391305, |
|
"grad_norm": 1.9014511108398438, |
|
"learning_rate": 1.2450000000000002e-06, |
|
"loss": 0.7182, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 13.043478260869565, |
|
"grad_norm": 13.358899116516113, |
|
"learning_rate": 1.495e-06, |
|
"loss": 0.7037, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 15.217391304347826, |
|
"grad_norm": 1.4039109945297241, |
|
"learning_rate": 1.745e-06, |
|
"loss": 0.6966, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 17.391304347826086, |
|
"grad_norm": 3.0226972103118896, |
|
"learning_rate": 1.9950000000000004e-06, |
|
"loss": 0.6702, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 19.565217391304348, |
|
"grad_norm": 1.716300129890442, |
|
"learning_rate": 2.24e-06, |
|
"loss": 0.6487, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 21.73913043478261, |
|
"grad_norm": 2.9351744651794434, |
|
"learning_rate": 2.4900000000000003e-06, |
|
"loss": 0.6071, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 21.73913043478261, |
|
"eval_loss": 0.521294116973877, |
|
"eval_runtime": 6.6451, |
|
"eval_samples_per_second": 23.927, |
|
"eval_steps_per_second": 3.01, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 23.91304347826087, |
|
"grad_norm": 1.9992552995681763, |
|
"learning_rate": 2.7400000000000004e-06, |
|
"loss": 0.5759, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 26.08695652173913, |
|
"grad_norm": 1.2594470977783203, |
|
"learning_rate": 2.99e-06, |
|
"loss": 0.56, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 28.26086956521739, |
|
"grad_norm": 1.0602363348007202, |
|
"learning_rate": 3.2400000000000003e-06, |
|
"loss": 0.5709, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 30.434782608695652, |
|
"grad_norm": 1.116231083869934, |
|
"learning_rate": 3.49e-06, |
|
"loss": 0.5322, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 32.608695652173914, |
|
"grad_norm": 1.5203620195388794, |
|
"learning_rate": 3.74e-06, |
|
"loss": 0.5264, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 34.78260869565217, |
|
"grad_norm": 1.6139357089996338, |
|
"learning_rate": 3.990000000000001e-06, |
|
"loss": 0.5191, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 36.95652173913044, |
|
"grad_norm": 1.436273217201233, |
|
"learning_rate": 4.24e-06, |
|
"loss": 0.5272, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 39.130434782608695, |
|
"grad_norm": 1.757323145866394, |
|
"learning_rate": 4.49e-06, |
|
"loss": 0.5016, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 41.30434782608695, |
|
"grad_norm": 0.9879368543624878, |
|
"learning_rate": 4.74e-06, |
|
"loss": 0.4975, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 43.47826086956522, |
|
"grad_norm": 1.3402222394943237, |
|
"learning_rate": 4.9900000000000005e-06, |
|
"loss": 0.5126, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 43.47826086956522, |
|
"eval_loss": 0.4506620466709137, |
|
"eval_runtime": 6.4644, |
|
"eval_samples_per_second": 24.596, |
|
"eval_steps_per_second": 3.094, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 45.65217391304348, |
|
"grad_norm": 1.9366440773010254, |
|
"learning_rate": 5.240000000000001e-06, |
|
"loss": 0.5043, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 47.82608695652174, |
|
"grad_norm": 1.5358505249023438, |
|
"learning_rate": 5.490000000000001e-06, |
|
"loss": 0.5034, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 50.0, |
|
"grad_norm": 1.6836323738098145, |
|
"learning_rate": 5.74e-06, |
|
"loss": 0.499, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 52.17391304347826, |
|
"grad_norm": 1.5092076063156128, |
|
"learning_rate": 5.99e-06, |
|
"loss": 0.487, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 54.34782608695652, |
|
"grad_norm": 1.0751235485076904, |
|
"learning_rate": 6.24e-06, |
|
"loss": 0.4871, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 56.52173913043478, |
|
"grad_norm": 1.327528476715088, |
|
"learning_rate": 6.4900000000000005e-06, |
|
"loss": 0.4911, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 58.69565217391305, |
|
"grad_norm": 1.7827174663543701, |
|
"learning_rate": 6.740000000000001e-06, |
|
"loss": 0.4772, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 60.869565217391305, |
|
"grad_norm": 1.2707558870315552, |
|
"learning_rate": 6.99e-06, |
|
"loss": 0.4786, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 63.04347826086956, |
|
"grad_norm": 1.0794321298599243, |
|
"learning_rate": 7.24e-06, |
|
"loss": 0.4741, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 65.21739130434783, |
|
"grad_norm": 2.0180580615997314, |
|
"learning_rate": 7.49e-06, |
|
"loss": 0.4749, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 65.21739130434783, |
|
"eval_loss": 0.4311259388923645, |
|
"eval_runtime": 6.4494, |
|
"eval_samples_per_second": 24.653, |
|
"eval_steps_per_second": 3.101, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 67.3913043478261, |
|
"grad_norm": 1.2938276529312134, |
|
"learning_rate": 7.74e-06, |
|
"loss": 0.4684, |
|
"step": 1550 |
|
}, |
|
{ |
|
"epoch": 69.56521739130434, |
|
"grad_norm": 1.8100048303604126, |
|
"learning_rate": 7.990000000000001e-06, |
|
"loss": 0.4797, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 71.73913043478261, |
|
"grad_norm": 1.8357532024383545, |
|
"learning_rate": 8.24e-06, |
|
"loss": 0.4813, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 73.91304347826087, |
|
"grad_norm": 1.4251928329467773, |
|
"learning_rate": 8.49e-06, |
|
"loss": 0.4622, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 76.08695652173913, |
|
"grad_norm": 1.1764353513717651, |
|
"learning_rate": 8.740000000000001e-06, |
|
"loss": 0.4642, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 78.26086956521739, |
|
"grad_norm": 1.6363924741744995, |
|
"learning_rate": 8.99e-06, |
|
"loss": 0.4669, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 80.43478260869566, |
|
"grad_norm": 1.485127568244934, |
|
"learning_rate": 9.240000000000001e-06, |
|
"loss": 0.4595, |
|
"step": 1850 |
|
}, |
|
{ |
|
"epoch": 82.6086956521739, |
|
"grad_norm": 1.2844778299331665, |
|
"learning_rate": 9.49e-06, |
|
"loss": 0.463, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 84.78260869565217, |
|
"grad_norm": 1.0181550979614258, |
|
"learning_rate": 9.74e-06, |
|
"loss": 0.4529, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 86.95652173913044, |
|
"grad_norm": 1.3658286333084106, |
|
"learning_rate": 9.990000000000001e-06, |
|
"loss": 0.454, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 86.95652173913044, |
|
"eval_loss": 0.4230838716030121, |
|
"eval_runtime": 6.4713, |
|
"eval_samples_per_second": 24.57, |
|
"eval_steps_per_second": 3.091, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 89.1304347826087, |
|
"grad_norm": 2.8028907775878906, |
|
"learning_rate": 9.920000000000002e-06, |
|
"loss": 0.4694, |
|
"step": 2050 |
|
}, |
|
{ |
|
"epoch": 91.30434782608695, |
|
"grad_norm": 0.8982108235359192, |
|
"learning_rate": 9.836666666666668e-06, |
|
"loss": 0.4495, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 93.47826086956522, |
|
"grad_norm": 2.500758409500122, |
|
"learning_rate": 9.753333333333335e-06, |
|
"loss": 0.4456, |
|
"step": 2150 |
|
}, |
|
{ |
|
"epoch": 95.65217391304348, |
|
"grad_norm": 1.7864420413970947, |
|
"learning_rate": 9.67e-06, |
|
"loss": 0.4816, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 97.82608695652173, |
|
"grad_norm": 1.3289756774902344, |
|
"learning_rate": 9.586666666666667e-06, |
|
"loss": 0.4512, |
|
"step": 2250 |
|
}, |
|
{ |
|
"epoch": 100.0, |
|
"grad_norm": 1.5230562686920166, |
|
"learning_rate": 9.503333333333334e-06, |
|
"loss": 0.4564, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 102.17391304347827, |
|
"grad_norm": 2.709244728088379, |
|
"learning_rate": 9.42e-06, |
|
"loss": 0.4543, |
|
"step": 2350 |
|
}, |
|
{ |
|
"epoch": 104.34782608695652, |
|
"grad_norm": 2.3624541759490967, |
|
"learning_rate": 9.336666666666666e-06, |
|
"loss": 0.4505, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 106.52173913043478, |
|
"grad_norm": 1.4635056257247925, |
|
"learning_rate": 9.253333333333333e-06, |
|
"loss": 0.4558, |
|
"step": 2450 |
|
}, |
|
{ |
|
"epoch": 108.69565217391305, |
|
"grad_norm": 1.9029183387756348, |
|
"learning_rate": 9.17e-06, |
|
"loss": 0.443, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 108.69565217391305, |
|
"eval_loss": 0.41728243231773376, |
|
"eval_runtime": 6.4572, |
|
"eval_samples_per_second": 24.624, |
|
"eval_steps_per_second": 3.097, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 110.8695652173913, |
|
"grad_norm": 1.4863485097885132, |
|
"learning_rate": 9.088333333333334e-06, |
|
"loss": 0.4562, |
|
"step": 2550 |
|
}, |
|
{ |
|
"epoch": 113.04347826086956, |
|
"grad_norm": 1.1933757066726685, |
|
"learning_rate": 9.005000000000001e-06, |
|
"loss": 0.4438, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 115.21739130434783, |
|
"grad_norm": 2.2257957458496094, |
|
"learning_rate": 8.921666666666668e-06, |
|
"loss": 0.4511, |
|
"step": 2650 |
|
}, |
|
{ |
|
"epoch": 117.3913043478261, |
|
"grad_norm": 1.382228970527649, |
|
"learning_rate": 8.838333333333335e-06, |
|
"loss": 0.4442, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 119.56521739130434, |
|
"grad_norm": 1.1577272415161133, |
|
"learning_rate": 8.755e-06, |
|
"loss": 0.4376, |
|
"step": 2750 |
|
}, |
|
{ |
|
"epoch": 121.73913043478261, |
|
"grad_norm": 1.16239333152771, |
|
"learning_rate": 8.671666666666667e-06, |
|
"loss": 0.4378, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 123.91304347826087, |
|
"grad_norm": 1.3845179080963135, |
|
"learning_rate": 8.588333333333334e-06, |
|
"loss": 0.4368, |
|
"step": 2850 |
|
}, |
|
{ |
|
"epoch": 126.08695652173913, |
|
"grad_norm": 1.4513936042785645, |
|
"learning_rate": 8.505e-06, |
|
"loss": 0.4362, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 128.2608695652174, |
|
"grad_norm": 2.2269773483276367, |
|
"learning_rate": 8.421666666666668e-06, |
|
"loss": 0.4498, |
|
"step": 2950 |
|
}, |
|
{ |
|
"epoch": 130.43478260869566, |
|
"grad_norm": 2.3055992126464844, |
|
"learning_rate": 8.338333333333335e-06, |
|
"loss": 0.4376, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 130.43478260869566, |
|
"eval_loss": 0.4168698191642761, |
|
"eval_runtime": 6.612, |
|
"eval_samples_per_second": 24.047, |
|
"eval_steps_per_second": 3.025, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 132.6086956521739, |
|
"grad_norm": 1.1443853378295898, |
|
"learning_rate": 8.255000000000001e-06, |
|
"loss": 0.4369, |
|
"step": 3050 |
|
}, |
|
{ |
|
"epoch": 134.7826086956522, |
|
"grad_norm": 1.3519920110702515, |
|
"learning_rate": 8.171666666666668e-06, |
|
"loss": 0.4322, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 136.95652173913044, |
|
"grad_norm": 1.4229977130889893, |
|
"learning_rate": 8.088333333333334e-06, |
|
"loss": 0.436, |
|
"step": 3150 |
|
}, |
|
{ |
|
"epoch": 139.1304347826087, |
|
"grad_norm": 3.4992916584014893, |
|
"learning_rate": 8.005e-06, |
|
"loss": 0.4275, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 141.30434782608697, |
|
"grad_norm": 0.9590508937835693, |
|
"learning_rate": 7.921666666666667e-06, |
|
"loss": 0.4366, |
|
"step": 3250 |
|
}, |
|
{ |
|
"epoch": 143.47826086956522, |
|
"grad_norm": 4.028496265411377, |
|
"learning_rate": 7.838333333333334e-06, |
|
"loss": 0.437, |
|
"step": 3300 |
|
}, |
|
{ |
|
"epoch": 145.65217391304347, |
|
"grad_norm": 0.9438303709030151, |
|
"learning_rate": 7.755000000000001e-06, |
|
"loss": 0.4317, |
|
"step": 3350 |
|
}, |
|
{ |
|
"epoch": 147.82608695652175, |
|
"grad_norm": 1.5847684144973755, |
|
"learning_rate": 7.671666666666668e-06, |
|
"loss": 0.4303, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 150.0, |
|
"grad_norm": 1.676293969154358, |
|
"learning_rate": 7.588333333333334e-06, |
|
"loss": 0.4312, |
|
"step": 3450 |
|
}, |
|
{ |
|
"epoch": 152.17391304347825, |
|
"grad_norm": 0.9531082510948181, |
|
"learning_rate": 7.505e-06, |
|
"loss": 0.4287, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 152.17391304347825, |
|
"eval_loss": 0.41326722502708435, |
|
"eval_runtime": 6.485, |
|
"eval_samples_per_second": 24.518, |
|
"eval_steps_per_second": 3.084, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 154.34782608695653, |
|
"grad_norm": 2.30401611328125, |
|
"learning_rate": 7.421666666666667e-06, |
|
"loss": 0.4248, |
|
"step": 3550 |
|
}, |
|
{ |
|
"epoch": 156.52173913043478, |
|
"grad_norm": 0.9738525152206421, |
|
"learning_rate": 7.338333333333334e-06, |
|
"loss": 0.4354, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 158.69565217391303, |
|
"grad_norm": 2.276857376098633, |
|
"learning_rate": 7.255000000000001e-06, |
|
"loss": 0.4345, |
|
"step": 3650 |
|
}, |
|
{ |
|
"epoch": 160.8695652173913, |
|
"grad_norm": 1.118604063987732, |
|
"learning_rate": 7.171666666666667e-06, |
|
"loss": 0.4243, |
|
"step": 3700 |
|
}, |
|
{ |
|
"epoch": 163.04347826086956, |
|
"grad_norm": 1.1995043754577637, |
|
"learning_rate": 7.088333333333334e-06, |
|
"loss": 0.4229, |
|
"step": 3750 |
|
}, |
|
{ |
|
"epoch": 165.2173913043478, |
|
"grad_norm": 1.1676712036132812, |
|
"learning_rate": 7.005000000000001e-06, |
|
"loss": 0.4225, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 167.3913043478261, |
|
"grad_norm": 1.7886515855789185, |
|
"learning_rate": 6.921666666666668e-06, |
|
"loss": 0.4243, |
|
"step": 3850 |
|
}, |
|
{ |
|
"epoch": 169.56521739130434, |
|
"grad_norm": 1.0336685180664062, |
|
"learning_rate": 6.838333333333334e-06, |
|
"loss": 0.4288, |
|
"step": 3900 |
|
}, |
|
{ |
|
"epoch": 171.7391304347826, |
|
"grad_norm": 3.572209596633911, |
|
"learning_rate": 6.7550000000000005e-06, |
|
"loss": 0.4262, |
|
"step": 3950 |
|
}, |
|
{ |
|
"epoch": 173.91304347826087, |
|
"grad_norm": 3.631232261657715, |
|
"learning_rate": 6.6716666666666674e-06, |
|
"loss": 0.4264, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 173.91304347826087, |
|
"eval_loss": 0.4149872660636902, |
|
"eval_runtime": 6.8222, |
|
"eval_samples_per_second": 23.306, |
|
"eval_steps_per_second": 2.932, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 176.08695652173913, |
|
"grad_norm": 1.0449178218841553, |
|
"learning_rate": 6.588333333333334e-06, |
|
"loss": 0.4232, |
|
"step": 4050 |
|
}, |
|
{ |
|
"epoch": 178.2608695652174, |
|
"grad_norm": 1.121854305267334, |
|
"learning_rate": 6.505e-06, |
|
"loss": 0.4214, |
|
"step": 4100 |
|
}, |
|
{ |
|
"epoch": 180.43478260869566, |
|
"grad_norm": 1.9070690870285034, |
|
"learning_rate": 6.421666666666667e-06, |
|
"loss": 0.4349, |
|
"step": 4150 |
|
}, |
|
{ |
|
"epoch": 182.6086956521739, |
|
"grad_norm": 1.6927332878112793, |
|
"learning_rate": 6.338333333333334e-06, |
|
"loss": 0.4216, |
|
"step": 4200 |
|
}, |
|
{ |
|
"epoch": 184.7826086956522, |
|
"grad_norm": 1.3572489023208618, |
|
"learning_rate": 6.255e-06, |
|
"loss": 0.4218, |
|
"step": 4250 |
|
}, |
|
{ |
|
"epoch": 186.95652173913044, |
|
"grad_norm": 1.9043023586273193, |
|
"learning_rate": 6.171666666666667e-06, |
|
"loss": 0.4204, |
|
"step": 4300 |
|
}, |
|
{ |
|
"epoch": 189.1304347826087, |
|
"grad_norm": 1.420276165008545, |
|
"learning_rate": 6.088333333333334e-06, |
|
"loss": 0.4259, |
|
"step": 4350 |
|
}, |
|
{ |
|
"epoch": 191.30434782608697, |
|
"grad_norm": 0.8243815302848816, |
|
"learning_rate": 6.005000000000001e-06, |
|
"loss": 0.4178, |
|
"step": 4400 |
|
}, |
|
{ |
|
"epoch": 193.47826086956522, |
|
"grad_norm": 0.995266854763031, |
|
"learning_rate": 5.921666666666667e-06, |
|
"loss": 0.4302, |
|
"step": 4450 |
|
}, |
|
{ |
|
"epoch": 195.65217391304347, |
|
"grad_norm": 1.3305209875106812, |
|
"learning_rate": 5.838333333333334e-06, |
|
"loss": 0.423, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 195.65217391304347, |
|
"eval_loss": 0.41338014602661133, |
|
"eval_runtime": 6.5844, |
|
"eval_samples_per_second": 24.148, |
|
"eval_steps_per_second": 3.037, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 197.82608695652175, |
|
"grad_norm": 1.1709681749343872, |
|
"learning_rate": 5.755000000000001e-06, |
|
"loss": 0.4178, |
|
"step": 4550 |
|
}, |
|
{ |
|
"epoch": 200.0, |
|
"grad_norm": 3.5350754261016846, |
|
"learning_rate": 5.671666666666668e-06, |
|
"loss": 0.4211, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 202.17391304347825, |
|
"grad_norm": 2.050933361053467, |
|
"learning_rate": 5.588333333333334e-06, |
|
"loss": 0.4266, |
|
"step": 4650 |
|
}, |
|
{ |
|
"epoch": 204.34782608695653, |
|
"grad_norm": 1.973078966140747, |
|
"learning_rate": 5.505000000000001e-06, |
|
"loss": 0.4187, |
|
"step": 4700 |
|
}, |
|
{ |
|
"epoch": 206.52173913043478, |
|
"grad_norm": 1.4094629287719727, |
|
"learning_rate": 5.4216666666666676e-06, |
|
"loss": 0.4294, |
|
"step": 4750 |
|
}, |
|
{ |
|
"epoch": 208.69565217391303, |
|
"grad_norm": 0.9872909188270569, |
|
"learning_rate": 5.3383333333333345e-06, |
|
"loss": 0.4239, |
|
"step": 4800 |
|
}, |
|
{ |
|
"epoch": 210.8695652173913, |
|
"grad_norm": 1.4357950687408447, |
|
"learning_rate": 5.2550000000000005e-06, |
|
"loss": 0.4117, |
|
"step": 4850 |
|
}, |
|
{ |
|
"epoch": 213.04347826086956, |
|
"grad_norm": 1.4760295152664185, |
|
"learning_rate": 5.171666666666667e-06, |
|
"loss": 0.4239, |
|
"step": 4900 |
|
}, |
|
{ |
|
"epoch": 215.2173913043478, |
|
"grad_norm": 0.9877930879592896, |
|
"learning_rate": 5.088333333333334e-06, |
|
"loss": 0.4151, |
|
"step": 4950 |
|
}, |
|
{ |
|
"epoch": 217.3913043478261, |
|
"grad_norm": 1.3829108476638794, |
|
"learning_rate": 5.0049999999999995e-06, |
|
"loss": 0.4223, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 217.3913043478261, |
|
"eval_loss": 0.41126948595046997, |
|
"eval_runtime": 6.7795, |
|
"eval_samples_per_second": 23.453, |
|
"eval_steps_per_second": 2.95, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 219.56521739130434, |
|
"grad_norm": 1.438590168952942, |
|
"learning_rate": 4.921666666666666e-06, |
|
"loss": 0.4114, |
|
"step": 5050 |
|
}, |
|
{ |
|
"epoch": 221.7391304347826, |
|
"grad_norm": 0.9894844889640808, |
|
"learning_rate": 4.838333333333334e-06, |
|
"loss": 0.413, |
|
"step": 5100 |
|
}, |
|
{ |
|
"epoch": 223.91304347826087, |
|
"grad_norm": 0.9712705612182617, |
|
"learning_rate": 4.755e-06, |
|
"loss": 0.4108, |
|
"step": 5150 |
|
}, |
|
{ |
|
"epoch": 226.08695652173913, |
|
"grad_norm": 0.9972001314163208, |
|
"learning_rate": 4.671666666666667e-06, |
|
"loss": 0.4173, |
|
"step": 5200 |
|
}, |
|
{ |
|
"epoch": 228.2608695652174, |
|
"grad_norm": 0.9916852116584778, |
|
"learning_rate": 4.588333333333333e-06, |
|
"loss": 0.426, |
|
"step": 5250 |
|
}, |
|
{ |
|
"epoch": 230.43478260869566, |
|
"grad_norm": 3.0285840034484863, |
|
"learning_rate": 4.505e-06, |
|
"loss": 0.4211, |
|
"step": 5300 |
|
}, |
|
{ |
|
"epoch": 232.6086956521739, |
|
"grad_norm": 0.8678073883056641, |
|
"learning_rate": 4.421666666666667e-06, |
|
"loss": 0.4165, |
|
"step": 5350 |
|
}, |
|
{ |
|
"epoch": 234.7826086956522, |
|
"grad_norm": 1.2311697006225586, |
|
"learning_rate": 4.338333333333334e-06, |
|
"loss": 0.4192, |
|
"step": 5400 |
|
}, |
|
{ |
|
"epoch": 236.95652173913044, |
|
"grad_norm": 1.1257262229919434, |
|
"learning_rate": 4.255e-06, |
|
"loss": 0.431, |
|
"step": 5450 |
|
}, |
|
{ |
|
"epoch": 239.1304347826087, |
|
"grad_norm": 0.9861388206481934, |
|
"learning_rate": 4.171666666666667e-06, |
|
"loss": 0.4104, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 239.1304347826087, |
|
"eval_loss": 0.40983906388282776, |
|
"eval_runtime": 6.76, |
|
"eval_samples_per_second": 23.521, |
|
"eval_steps_per_second": 2.959, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 241.30434782608697, |
|
"grad_norm": 0.9192537665367126, |
|
"learning_rate": 4.088333333333334e-06, |
|
"loss": 0.4105, |
|
"step": 5550 |
|
}, |
|
{ |
|
"epoch": 243.47826086956522, |
|
"grad_norm": 1.0816397666931152, |
|
"learning_rate": 4.005000000000001e-06, |
|
"loss": 0.4116, |
|
"step": 5600 |
|
}, |
|
{ |
|
"epoch": 245.65217391304347, |
|
"grad_norm": 1.2673571109771729, |
|
"learning_rate": 3.921666666666667e-06, |
|
"loss": 0.4122, |
|
"step": 5650 |
|
}, |
|
{ |
|
"epoch": 247.82608695652175, |
|
"grad_norm": 1.131993055343628, |
|
"learning_rate": 3.8383333333333336e-06, |
|
"loss": 0.4097, |
|
"step": 5700 |
|
}, |
|
{ |
|
"epoch": 250.0, |
|
"grad_norm": 2.051192283630371, |
|
"learning_rate": 3.7550000000000005e-06, |
|
"loss": 0.4075, |
|
"step": 5750 |
|
}, |
|
{ |
|
"epoch": 252.17391304347825, |
|
"grad_norm": 0.8863391280174255, |
|
"learning_rate": 3.6716666666666665e-06, |
|
"loss": 0.4077, |
|
"step": 5800 |
|
}, |
|
{ |
|
"epoch": 254.34782608695653, |
|
"grad_norm": 0.8587177395820618, |
|
"learning_rate": 3.588333333333334e-06, |
|
"loss": 0.4047, |
|
"step": 5850 |
|
}, |
|
{ |
|
"epoch": 256.5217391304348, |
|
"grad_norm": 1.0080711841583252, |
|
"learning_rate": 3.505e-06, |
|
"loss": 0.4093, |
|
"step": 5900 |
|
}, |
|
{ |
|
"epoch": 258.69565217391306, |
|
"grad_norm": 1.022581934928894, |
|
"learning_rate": 3.4216666666666672e-06, |
|
"loss": 0.4069, |
|
"step": 5950 |
|
}, |
|
{ |
|
"epoch": 260.8695652173913, |
|
"grad_norm": 1.0036355257034302, |
|
"learning_rate": 3.3383333333333333e-06, |
|
"loss": 0.4192, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 260.8695652173913, |
|
"eval_loss": 0.41063958406448364, |
|
"eval_runtime": 6.6166, |
|
"eval_samples_per_second": 24.031, |
|
"eval_steps_per_second": 3.023, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 263.04347826086956, |
|
"grad_norm": 1.1398719549179077, |
|
"learning_rate": 3.255e-06, |
|
"loss": 0.4187, |
|
"step": 6050 |
|
}, |
|
{ |
|
"epoch": 265.2173913043478, |
|
"grad_norm": 0.8705397248268127, |
|
"learning_rate": 3.1716666666666667e-06, |
|
"loss": 0.4189, |
|
"step": 6100 |
|
}, |
|
{ |
|
"epoch": 267.39130434782606, |
|
"grad_norm": 1.1828796863555908, |
|
"learning_rate": 3.0883333333333336e-06, |
|
"loss": 0.412, |
|
"step": 6150 |
|
}, |
|
{ |
|
"epoch": 269.5652173913044, |
|
"grad_norm": 1.2808082103729248, |
|
"learning_rate": 3.005e-06, |
|
"loss": 0.4135, |
|
"step": 6200 |
|
}, |
|
{ |
|
"epoch": 271.7391304347826, |
|
"grad_norm": 1.2082769870758057, |
|
"learning_rate": 2.921666666666667e-06, |
|
"loss": 0.4163, |
|
"step": 6250 |
|
}, |
|
{ |
|
"epoch": 273.9130434782609, |
|
"grad_norm": 1.2765779495239258, |
|
"learning_rate": 2.8383333333333334e-06, |
|
"loss": 0.409, |
|
"step": 6300 |
|
}, |
|
{ |
|
"epoch": 276.0869565217391, |
|
"grad_norm": 1.0085750818252563, |
|
"learning_rate": 2.7550000000000003e-06, |
|
"loss": 0.4168, |
|
"step": 6350 |
|
}, |
|
{ |
|
"epoch": 278.2608695652174, |
|
"grad_norm": 1.2481540441513062, |
|
"learning_rate": 2.6716666666666668e-06, |
|
"loss": 0.4163, |
|
"step": 6400 |
|
}, |
|
{ |
|
"epoch": 280.4347826086956, |
|
"grad_norm": 1.0520100593566895, |
|
"learning_rate": 2.5883333333333337e-06, |
|
"loss": 0.4165, |
|
"step": 6450 |
|
}, |
|
{ |
|
"epoch": 282.60869565217394, |
|
"grad_norm": 1.2364658117294312, |
|
"learning_rate": 2.505e-06, |
|
"loss": 0.4089, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 282.60869565217394, |
|
"eval_loss": 0.4121515154838562, |
|
"eval_runtime": 6.5878, |
|
"eval_samples_per_second": 24.136, |
|
"eval_steps_per_second": 3.036, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 284.7826086956522, |
|
"grad_norm": 1.1955006122589111, |
|
"learning_rate": 2.421666666666667e-06, |
|
"loss": 0.4065, |
|
"step": 6550 |
|
}, |
|
{ |
|
"epoch": 286.95652173913044, |
|
"grad_norm": 1.8908520936965942, |
|
"learning_rate": 2.3383333333333335e-06, |
|
"loss": 0.4162, |
|
"step": 6600 |
|
}, |
|
{ |
|
"epoch": 289.1304347826087, |
|
"grad_norm": 1.1614066362380981, |
|
"learning_rate": 2.2550000000000004e-06, |
|
"loss": 0.4099, |
|
"step": 6650 |
|
}, |
|
{ |
|
"epoch": 291.30434782608694, |
|
"grad_norm": 0.8983772993087769, |
|
"learning_rate": 2.171666666666667e-06, |
|
"loss": 0.4121, |
|
"step": 6700 |
|
}, |
|
{ |
|
"epoch": 293.4782608695652, |
|
"grad_norm": 1.2100547552108765, |
|
"learning_rate": 2.088333333333334e-06, |
|
"loss": 0.4041, |
|
"step": 6750 |
|
}, |
|
{ |
|
"epoch": 295.6521739130435, |
|
"grad_norm": 1.1354186534881592, |
|
"learning_rate": 2.006666666666667e-06, |
|
"loss": 0.4127, |
|
"step": 6800 |
|
}, |
|
{ |
|
"epoch": 297.82608695652175, |
|
"grad_norm": 1.1679037809371948, |
|
"learning_rate": 1.9233333333333333e-06, |
|
"loss": 0.4068, |
|
"step": 6850 |
|
}, |
|
{ |
|
"epoch": 300.0, |
|
"grad_norm": 1.502341389656067, |
|
"learning_rate": 1.8400000000000002e-06, |
|
"loss": 0.4147, |
|
"step": 6900 |
|
}, |
|
{ |
|
"epoch": 302.17391304347825, |
|
"grad_norm": 0.9498834013938904, |
|
"learning_rate": 1.7566666666666669e-06, |
|
"loss": 0.4056, |
|
"step": 6950 |
|
}, |
|
{ |
|
"epoch": 304.3478260869565, |
|
"grad_norm": 0.8643518686294556, |
|
"learning_rate": 1.6733333333333335e-06, |
|
"loss": 0.4146, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 304.3478260869565, |
|
"eval_loss": 0.41151124238967896, |
|
"eval_runtime": 6.5375, |
|
"eval_samples_per_second": 24.321, |
|
"eval_steps_per_second": 3.059, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 306.5217391304348, |
|
"grad_norm": 0.9846628308296204, |
|
"learning_rate": 1.5900000000000002e-06, |
|
"loss": 0.4115, |
|
"step": 7050 |
|
}, |
|
{ |
|
"epoch": 308.69565217391306, |
|
"grad_norm": 1.0331532955169678, |
|
"learning_rate": 1.506666666666667e-06, |
|
"loss": 0.4143, |
|
"step": 7100 |
|
}, |
|
{ |
|
"epoch": 310.8695652173913, |
|
"grad_norm": 0.9437756538391113, |
|
"learning_rate": 1.4233333333333336e-06, |
|
"loss": 0.4109, |
|
"step": 7150 |
|
}, |
|
{ |
|
"epoch": 313.04347826086956, |
|
"grad_norm": 1.0904934406280518, |
|
"learning_rate": 1.34e-06, |
|
"loss": 0.4109, |
|
"step": 7200 |
|
}, |
|
{ |
|
"epoch": 315.2173913043478, |
|
"grad_norm": 0.9227349162101746, |
|
"learning_rate": 1.2566666666666668e-06, |
|
"loss": 0.406, |
|
"step": 7250 |
|
}, |
|
{ |
|
"epoch": 317.39130434782606, |
|
"grad_norm": 0.991308867931366, |
|
"learning_rate": 1.1733333333333335e-06, |
|
"loss": 0.4161, |
|
"step": 7300 |
|
}, |
|
{ |
|
"epoch": 319.5652173913044, |
|
"grad_norm": 0.9621168375015259, |
|
"learning_rate": 1.0900000000000002e-06, |
|
"loss": 0.4123, |
|
"step": 7350 |
|
}, |
|
{ |
|
"epoch": 321.7391304347826, |
|
"grad_norm": 1.1078619956970215, |
|
"learning_rate": 1.0066666666666668e-06, |
|
"loss": 0.4109, |
|
"step": 7400 |
|
}, |
|
{ |
|
"epoch": 323.9130434782609, |
|
"grad_norm": 0.8910597562789917, |
|
"learning_rate": 9.233333333333334e-07, |
|
"loss": 0.4052, |
|
"step": 7450 |
|
}, |
|
{ |
|
"epoch": 326.0869565217391, |
|
"grad_norm": 0.760857880115509, |
|
"learning_rate": 8.400000000000001e-07, |
|
"loss": 0.4116, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 326.0869565217391, |
|
"eval_loss": 0.4111355245113373, |
|
"eval_runtime": 6.8924, |
|
"eval_samples_per_second": 23.069, |
|
"eval_steps_per_second": 2.902, |
|
"step": 7500 |
|
} |
|
], |
|
"logging_steps": 50, |
|
"max_steps": 8000, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 348, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 1.1135871544625309e+17, |
|
"train_batch_size": 64, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|