| { | |
| "best_metric": 0.4122372269630432, | |
| "best_model_checkpoint": "mikhail-panzo/zlm-fil-ceb_b64_le5_s8000/checkpoint-5000", | |
| "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": 0.7357823252677917, | |
| "learning_rate": 2.5000000000000004e-07, | |
| "loss": 0.4055, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.3478260869565215, | |
| "grad_norm": 0.7944665551185608, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 0.3988, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.521739130434782, | |
| "grad_norm": 0.7226569652557373, | |
| "learning_rate": 7.5e-07, | |
| "loss": 0.4072, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 0.7845238447189331, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.4028, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.869565217391305, | |
| "grad_norm": 0.9769566655158997, | |
| "learning_rate": 1.25e-06, | |
| "loss": 0.4066, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 13.043478260869565, | |
| "grad_norm": 0.8634074330329895, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.405, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 15.217391304347826, | |
| "grad_norm": 0.9029327034950256, | |
| "learning_rate": 1.75e-06, | |
| "loss": 0.4074, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "grad_norm": 0.7624074220657349, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.406, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.565217391304348, | |
| "grad_norm": 1.005196213722229, | |
| "learning_rate": 2.25e-06, | |
| "loss": 0.4052, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 21.73913043478261, | |
| "grad_norm": 1.0276380777359009, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.4019, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 21.73913043478261, | |
| "eval_loss": 0.4143177568912506, | |
| "eval_runtime": 6.4074, | |
| "eval_samples_per_second": 24.815, | |
| "eval_steps_per_second": 3.121, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 23.91304347826087, | |
| "grad_norm": 0.9067140817642212, | |
| "learning_rate": 2.7500000000000004e-06, | |
| "loss": 0.4021, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 26.08695652173913, | |
| "grad_norm": 0.8799042701721191, | |
| "learning_rate": 3e-06, | |
| "loss": 0.4071, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 28.26086956521739, | |
| "grad_norm": 1.0451533794403076, | |
| "learning_rate": 3.2500000000000002e-06, | |
| "loss": 0.4146, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 30.434782608695652, | |
| "grad_norm": 0.7998800277709961, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.4023, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 32.608695652173914, | |
| "grad_norm": 1.0301792621612549, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 0.4045, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 34.78260869565217, | |
| "grad_norm": 0.6388185620307922, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.4045, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 36.95652173913044, | |
| "grad_norm": 0.9676840901374817, | |
| "learning_rate": 4.25e-06, | |
| "loss": 0.4116, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 39.130434782608695, | |
| "grad_norm": 0.8355514407157898, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.4013, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 41.30434782608695, | |
| "grad_norm": 0.7534067630767822, | |
| "learning_rate": 4.75e-06, | |
| "loss": 0.4019, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 43.47826086956522, | |
| "grad_norm": 0.9366117119789124, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4064, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 43.47826086956522, | |
| "eval_loss": 0.41316837072372437, | |
| "eval_runtime": 6.5616, | |
| "eval_samples_per_second": 24.232, | |
| "eval_steps_per_second": 3.048, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 45.65217391304348, | |
| "grad_norm": 0.7025179266929626, | |
| "learning_rate": 5.2500000000000006e-06, | |
| "loss": 0.4061, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 47.82608695652174, | |
| "grad_norm": 0.9240646958351135, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.4061, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.9685391187667847, | |
| "learning_rate": 5.75e-06, | |
| "loss": 0.4063, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 52.17391304347826, | |
| "grad_norm": 0.8591094017028809, | |
| "learning_rate": 6e-06, | |
| "loss": 0.4013, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 54.34782608695652, | |
| "grad_norm": 0.9348228573799133, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.402, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 56.52173913043478, | |
| "grad_norm": 0.7908409237861633, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.4046, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 58.69565217391305, | |
| "grad_norm": 1.109017014503479, | |
| "learning_rate": 6.750000000000001e-06, | |
| "loss": 0.3995, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 60.869565217391305, | |
| "grad_norm": 0.8529478311538696, | |
| "learning_rate": 7e-06, | |
| "loss": 0.4018, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 63.04347826086956, | |
| "grad_norm": 0.9947476387023926, | |
| "learning_rate": 7.25e-06, | |
| "loss": 0.401, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 65.21739130434783, | |
| "grad_norm": 1.2476813793182373, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.4034, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 65.21739130434783, | |
| "eval_loss": 0.41223829984664917, | |
| "eval_runtime": 6.5846, | |
| "eval_samples_per_second": 24.147, | |
| "eval_steps_per_second": 3.037, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 67.3913043478261, | |
| "grad_norm": 0.8658013343811035, | |
| "learning_rate": 7.75e-06, | |
| "loss": 0.4007, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 69.56521739130434, | |
| "grad_norm": 0.7932788133621216, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.4054, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 71.73913043478261, | |
| "grad_norm": 2.33347225189209, | |
| "learning_rate": 8.25e-06, | |
| "loss": 0.4075, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 73.91304347826087, | |
| "grad_norm": 1.0095164775848389, | |
| "learning_rate": 8.5e-06, | |
| "loss": 0.3982, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 76.08695652173913, | |
| "grad_norm": 0.8895902633666992, | |
| "learning_rate": 8.750000000000001e-06, | |
| "loss": 0.4018, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 78.26086956521739, | |
| "grad_norm": 1.2252330780029297, | |
| "learning_rate": 9e-06, | |
| "loss": 0.4024, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 80.43478260869566, | |
| "grad_norm": 0.7731898427009583, | |
| "learning_rate": 9.250000000000001e-06, | |
| "loss": 0.3994, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 82.6086956521739, | |
| "grad_norm": 1.047607660293579, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.4013, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 84.78260869565217, | |
| "grad_norm": 0.9284490346908569, | |
| "learning_rate": 9.75e-06, | |
| "loss": 0.3986, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 86.95652173913044, | |
| "grad_norm": 4.725491046905518, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3985, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 86.95652173913044, | |
| "eval_loss": 0.4147759974002838, | |
| "eval_runtime": 6.5475, | |
| "eval_samples_per_second": 24.284, | |
| "eval_steps_per_second": 3.055, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 89.1304347826087, | |
| "grad_norm": 1.1678683757781982, | |
| "learning_rate": 9.916666666666668e-06, | |
| "loss": 0.4066, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 91.30434782608695, | |
| "grad_norm": 1.988890528678894, | |
| "learning_rate": 9.835000000000002e-06, | |
| "loss": 0.3997, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 93.47826086956522, | |
| "grad_norm": 1.0278427600860596, | |
| "learning_rate": 9.751666666666667e-06, | |
| "loss": 0.3948, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 95.65217391304348, | |
| "grad_norm": 0.9606735110282898, | |
| "learning_rate": 9.67e-06, | |
| "loss": 0.42, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 97.82608695652173, | |
| "grad_norm": 1.0861903429031372, | |
| "learning_rate": 9.586666666666667e-06, | |
| "loss": 0.398, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 1.4872643947601318, | |
| "learning_rate": 9.503333333333334e-06, | |
| "loss": 0.4059, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 102.17391304347827, | |
| "grad_norm": 1.0460981130599976, | |
| "learning_rate": 9.42e-06, | |
| "loss": 0.4014, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 104.34782608695652, | |
| "grad_norm": 1.906476378440857, | |
| "learning_rate": 9.336666666666666e-06, | |
| "loss": 0.4015, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 106.52173913043478, | |
| "grad_norm": 1.3965051174163818, | |
| "learning_rate": 9.253333333333333e-06, | |
| "loss": 0.4063, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 108.69565217391305, | |
| "grad_norm": 1.7410107851028442, | |
| "learning_rate": 9.17e-06, | |
| "loss": 0.396, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 108.69565217391305, | |
| "eval_loss": 0.41233959794044495, | |
| "eval_runtime": 6.6031, | |
| "eval_samples_per_second": 24.08, | |
| "eval_steps_per_second": 3.029, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 110.8695652173913, | |
| "grad_norm": 1.4562053680419922, | |
| "learning_rate": 9.088333333333334e-06, | |
| "loss": 0.4042, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 113.04347826086956, | |
| "grad_norm": 0.8741047978401184, | |
| "learning_rate": 9.005000000000001e-06, | |
| "loss": 0.3985, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 115.21739130434783, | |
| "grad_norm": 1.1817028522491455, | |
| "learning_rate": 8.921666666666668e-06, | |
| "loss": 0.4036, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 117.3913043478261, | |
| "grad_norm": 0.8515534996986389, | |
| "learning_rate": 8.838333333333335e-06, | |
| "loss": 0.3986, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 119.56521739130434, | |
| "grad_norm": 0.9545428156852722, | |
| "learning_rate": 8.755e-06, | |
| "loss": 0.3946, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 121.73913043478261, | |
| "grad_norm": 0.6653321385383606, | |
| "learning_rate": 8.671666666666667e-06, | |
| "loss": 0.3957, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 123.91304347826087, | |
| "grad_norm": 0.8100114464759827, | |
| "learning_rate": 8.588333333333334e-06, | |
| "loss": 0.3953, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 126.08695652173913, | |
| "grad_norm": 1.4174237251281738, | |
| "learning_rate": 8.505e-06, | |
| "loss": 0.3933, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 128.2608695652174, | |
| "grad_norm": 1.6101123094558716, | |
| "learning_rate": 8.421666666666668e-06, | |
| "loss": 0.4027, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 130.43478260869566, | |
| "grad_norm": 0.7453967332839966, | |
| "learning_rate": 8.338333333333335e-06, | |
| "loss": 0.3958, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 130.43478260869566, | |
| "eval_loss": 0.4138609766960144, | |
| "eval_runtime": 6.6093, | |
| "eval_samples_per_second": 24.057, | |
| "eval_steps_per_second": 3.026, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 132.6086956521739, | |
| "grad_norm": 1.061684012413025, | |
| "learning_rate": 8.255000000000001e-06, | |
| "loss": 0.3966, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 134.7826086956522, | |
| "grad_norm": 1.0820105075836182, | |
| "learning_rate": 8.171666666666668e-06, | |
| "loss": 0.3937, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 136.95652173913044, | |
| "grad_norm": 0.8147062063217163, | |
| "learning_rate": 8.088333333333334e-06, | |
| "loss": 0.3961, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 139.1304347826087, | |
| "grad_norm": 0.7450351715087891, | |
| "learning_rate": 8.005e-06, | |
| "loss": 0.3923, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 141.30434782608697, | |
| "grad_norm": 1.05226731300354, | |
| "learning_rate": 7.921666666666667e-06, | |
| "loss": 0.3951, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 143.47826086956522, | |
| "grad_norm": 1.5119163990020752, | |
| "learning_rate": 7.838333333333334e-06, | |
| "loss": 0.3957, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 145.65217391304347, | |
| "grad_norm": 1.0422383546829224, | |
| "learning_rate": 7.755000000000001e-06, | |
| "loss": 0.3942, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 147.82608695652175, | |
| "grad_norm": 1.2120935916900635, | |
| "learning_rate": 7.671666666666668e-06, | |
| "loss": 0.3946, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "grad_norm": 1.3127976655960083, | |
| "learning_rate": 7.588333333333334e-06, | |
| "loss": 0.3933, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 152.17391304347825, | |
| "grad_norm": 0.7704339027404785, | |
| "learning_rate": 7.505e-06, | |
| "loss": 0.3928, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 152.17391304347825, | |
| "eval_loss": 0.4132215976715088, | |
| "eval_runtime": 6.6162, | |
| "eval_samples_per_second": 24.032, | |
| "eval_steps_per_second": 3.023, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 154.34782608695653, | |
| "grad_norm": 1.4827122688293457, | |
| "learning_rate": 7.421666666666667e-06, | |
| "loss": 0.3908, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 156.52173913043478, | |
| "grad_norm": 1.0464049577713013, | |
| "learning_rate": 7.338333333333334e-06, | |
| "loss": 0.3993, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 158.69565217391303, | |
| "grad_norm": 0.8279913663864136, | |
| "learning_rate": 7.255000000000001e-06, | |
| "loss": 0.3967, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 160.8695652173913, | |
| "grad_norm": 0.9175034165382385, | |
| "learning_rate": 7.171666666666667e-06, | |
| "loss": 0.3911, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 163.04347826086956, | |
| "grad_norm": 1.0080233812332153, | |
| "learning_rate": 7.088333333333334e-06, | |
| "loss": 0.3898, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 165.2173913043478, | |
| "grad_norm": 0.9944823980331421, | |
| "learning_rate": 7.005000000000001e-06, | |
| "loss": 0.3901, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 167.3913043478261, | |
| "grad_norm": 1.8222392797470093, | |
| "learning_rate": 6.921666666666668e-06, | |
| "loss": 0.3901, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 169.56521739130434, | |
| "grad_norm": 0.9506675004959106, | |
| "learning_rate": 6.838333333333334e-06, | |
| "loss": 0.3945, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 171.7391304347826, | |
| "grad_norm": 4.512270450592041, | |
| "learning_rate": 6.7550000000000005e-06, | |
| "loss": 0.3913, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 173.91304347826087, | |
| "grad_norm": 0.74826580286026, | |
| "learning_rate": 6.6716666666666674e-06, | |
| "loss": 0.3914, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 173.91304347826087, | |
| "eval_loss": 0.4138501286506653, | |
| "eval_runtime": 6.5399, | |
| "eval_samples_per_second": 24.312, | |
| "eval_steps_per_second": 3.058, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 176.08695652173913, | |
| "grad_norm": 1.4560046195983887, | |
| "learning_rate": 6.588333333333334e-06, | |
| "loss": 0.3903, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 178.2608695652174, | |
| "grad_norm": 1.183271050453186, | |
| "learning_rate": 6.505e-06, | |
| "loss": 0.3901, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 180.43478260869566, | |
| "grad_norm": 0.6697027683258057, | |
| "learning_rate": 6.421666666666667e-06, | |
| "loss": 0.399, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 182.6086956521739, | |
| "grad_norm": 0.96327143907547, | |
| "learning_rate": 6.338333333333334e-06, | |
| "loss": 0.3895, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 184.7826086956522, | |
| "grad_norm": 0.9833754301071167, | |
| "learning_rate": 6.255e-06, | |
| "loss": 0.3903, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 186.95652173913044, | |
| "grad_norm": 0.857953667640686, | |
| "learning_rate": 6.171666666666667e-06, | |
| "loss": 0.39, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 189.1304347826087, | |
| "grad_norm": 0.7506987452507019, | |
| "learning_rate": 6.088333333333334e-06, | |
| "loss": 0.394, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 191.30434782608697, | |
| "grad_norm": 0.8283782601356506, | |
| "learning_rate": 6.005000000000001e-06, | |
| "loss": 0.3869, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 193.47826086956522, | |
| "grad_norm": 1.0252714157104492, | |
| "learning_rate": 5.921666666666667e-06, | |
| "loss": 0.3967, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 195.65217391304347, | |
| "grad_norm": 0.9400416612625122, | |
| "learning_rate": 5.838333333333334e-06, | |
| "loss": 0.3916, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 195.65217391304347, | |
| "eval_loss": 0.41232919692993164, | |
| "eval_runtime": 6.3912, | |
| "eval_samples_per_second": 24.878, | |
| "eval_steps_per_second": 3.129, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 197.82608695652175, | |
| "grad_norm": 0.9093385934829712, | |
| "learning_rate": 5.755000000000001e-06, | |
| "loss": 0.3894, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "grad_norm": 1.078837513923645, | |
| "learning_rate": 5.671666666666668e-06, | |
| "loss": 0.3899, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 202.17391304347825, | |
| "grad_norm": 0.7748771905899048, | |
| "learning_rate": 5.588333333333334e-06, | |
| "loss": 0.3943, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 204.34782608695653, | |
| "grad_norm": 1.1522585153579712, | |
| "learning_rate": 5.505000000000001e-06, | |
| "loss": 0.389, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 206.52173913043478, | |
| "grad_norm": 1.6818900108337402, | |
| "learning_rate": 5.4216666666666676e-06, | |
| "loss": 0.3961, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 208.69565217391303, | |
| "grad_norm": 0.9630489349365234, | |
| "learning_rate": 5.3383333333333345e-06, | |
| "loss": 0.3926, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 210.8695652173913, | |
| "grad_norm": 0.915107250213623, | |
| "learning_rate": 5.2550000000000005e-06, | |
| "loss": 0.3843, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 213.04347826086956, | |
| "grad_norm": 0.7301034331321716, | |
| "learning_rate": 5.171666666666667e-06, | |
| "loss": 0.3929, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 215.2173913043478, | |
| "grad_norm": 0.8523257970809937, | |
| "learning_rate": 5.088333333333334e-06, | |
| "loss": 0.3873, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 217.3913043478261, | |
| "grad_norm": 0.7538474798202515, | |
| "learning_rate": 5.0049999999999995e-06, | |
| "loss": 0.3919, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 217.3913043478261, | |
| "eval_loss": 0.4122372269630432, | |
| "eval_runtime": 6.5201, | |
| "eval_samples_per_second": 24.386, | |
| "eval_steps_per_second": 3.067, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 219.56521739130434, | |
| "grad_norm": 1.006597638130188, | |
| "learning_rate": 4.921666666666666e-06, | |
| "loss": 0.3844, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 221.7391304347826, | |
| "grad_norm": 0.6486532092094421, | |
| "learning_rate": 4.838333333333334e-06, | |
| "loss": 0.3853, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 223.91304347826087, | |
| "grad_norm": 0.7955175638198853, | |
| "learning_rate": 4.755e-06, | |
| "loss": 0.3853, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 226.08695652173913, | |
| "grad_norm": 0.8011429309844971, | |
| "learning_rate": 4.671666666666667e-06, | |
| "loss": 0.3888, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 228.2608695652174, | |
| "grad_norm": 0.7421734929084778, | |
| "learning_rate": 4.588333333333333e-06, | |
| "loss": 0.3945, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 230.43478260869566, | |
| "grad_norm": 1.5567547082901, | |
| "learning_rate": 4.505e-06, | |
| "loss": 0.3916, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 232.6086956521739, | |
| "grad_norm": 0.8348387479782104, | |
| "learning_rate": 4.421666666666667e-06, | |
| "loss": 0.3887, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 234.7826086956522, | |
| "grad_norm": 0.965840756893158, | |
| "learning_rate": 4.338333333333334e-06, | |
| "loss": 0.3907, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 236.95652173913044, | |
| "grad_norm": 1.0057460069656372, | |
| "learning_rate": 4.255e-06, | |
| "loss": 0.4004, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 239.1304347826087, | |
| "grad_norm": 1.2100751399993896, | |
| "learning_rate": 4.171666666666667e-06, | |
| "loss": 0.3844, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 239.1304347826087, | |
| "eval_loss": 0.41257867217063904, | |
| "eval_runtime": 6.47, | |
| "eval_samples_per_second": 24.575, | |
| "eval_steps_per_second": 3.091, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 241.30434782608697, | |
| "grad_norm": 0.7117742896080017, | |
| "learning_rate": 4.088333333333334e-06, | |
| "loss": 0.3845, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 243.47826086956522, | |
| "grad_norm": 0.6881281733512878, | |
| "learning_rate": 4.005000000000001e-06, | |
| "loss": 0.3858, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 245.65217391304347, | |
| "grad_norm": 1.3413081169128418, | |
| "learning_rate": 3.921666666666667e-06, | |
| "loss": 0.3842, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 247.82608695652175, | |
| "grad_norm": 0.7762352228164673, | |
| "learning_rate": 3.8383333333333336e-06, | |
| "loss": 0.3839, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 250.0, | |
| "grad_norm": 1.2861838340759277, | |
| "learning_rate": 3.7550000000000005e-06, | |
| "loss": 0.3836, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 252.17391304347825, | |
| "grad_norm": 1.12875497341156, | |
| "learning_rate": 3.6716666666666665e-06, | |
| "loss": 0.3828, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 254.34782608695653, | |
| "grad_norm": 0.9641019105911255, | |
| "learning_rate": 3.588333333333334e-06, | |
| "loss": 0.3815, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 256.5217391304348, | |
| "grad_norm": 0.7063067555427551, | |
| "learning_rate": 3.505e-06, | |
| "loss": 0.3838, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 258.69565217391306, | |
| "grad_norm": 1.0992182493209839, | |
| "learning_rate": 3.4216666666666672e-06, | |
| "loss": 0.3824, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 260.8695652173913, | |
| "grad_norm": 0.6153731942176819, | |
| "learning_rate": 3.3383333333333333e-06, | |
| "loss": 0.3908, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 260.8695652173913, | |
| "eval_loss": 0.4136714041233063, | |
| "eval_runtime": 6.4416, | |
| "eval_samples_per_second": 24.683, | |
| "eval_steps_per_second": 3.105, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 263.04347826086956, | |
| "grad_norm": 0.8322598934173584, | |
| "learning_rate": 3.255e-06, | |
| "loss": 0.3918, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 265.2173913043478, | |
| "grad_norm": 0.7336969971656799, | |
| "learning_rate": 3.1716666666666667e-06, | |
| "loss": 0.3908, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 267.39130434782606, | |
| "grad_norm": 0.7218737006187439, | |
| "learning_rate": 3.0883333333333336e-06, | |
| "loss": 0.3848, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 269.5652173913044, | |
| "grad_norm": 1.2083243131637573, | |
| "learning_rate": 3.005e-06, | |
| "loss": 0.3868, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 271.7391304347826, | |
| "grad_norm": 0.9182942509651184, | |
| "learning_rate": 2.921666666666667e-06, | |
| "loss": 0.3895, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 273.9130434782609, | |
| "grad_norm": 0.8747413158416748, | |
| "learning_rate": 2.8383333333333334e-06, | |
| "loss": 0.3842, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 276.0869565217391, | |
| "grad_norm": 0.8834450840950012, | |
| "learning_rate": 2.7550000000000003e-06, | |
| "loss": 0.3879, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 278.2608695652174, | |
| "grad_norm": 1.0217564105987549, | |
| "learning_rate": 2.6716666666666668e-06, | |
| "loss": 0.3892, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 280.4347826086956, | |
| "grad_norm": 0.6895825862884521, | |
| "learning_rate": 2.5883333333333337e-06, | |
| "loss": 0.389, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 282.60869565217394, | |
| "grad_norm": 0.6623283624649048, | |
| "learning_rate": 2.505e-06, | |
| "loss": 0.3854, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 282.60869565217394, | |
| "eval_loss": 0.413311630487442, | |
| "eval_runtime": 6.7085, | |
| "eval_samples_per_second": 23.701, | |
| "eval_steps_per_second": 2.981, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 284.7826086956522, | |
| "grad_norm": 0.7647244930267334, | |
| "learning_rate": 2.421666666666667e-06, | |
| "loss": 0.384, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 286.95652173913044, | |
| "grad_norm": 0.7481832504272461, | |
| "learning_rate": 2.3383333333333335e-06, | |
| "loss": 0.3897, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 289.1304347826087, | |
| "grad_norm": 0.9159369468688965, | |
| "learning_rate": 2.2550000000000004e-06, | |
| "loss": 0.3852, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 291.30434782608694, | |
| "grad_norm": 0.5772305727005005, | |
| "learning_rate": 2.171666666666667e-06, | |
| "loss": 0.3864, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 293.4782608695652, | |
| "grad_norm": 0.8706581592559814, | |
| "learning_rate": 2.088333333333334e-06, | |
| "loss": 0.3818, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 295.6521739130435, | |
| "grad_norm": 0.6359920501708984, | |
| "learning_rate": 2.0050000000000003e-06, | |
| "loss": 0.3866, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 297.82608695652175, | |
| "grad_norm": 0.6971527934074402, | |
| "learning_rate": 1.9216666666666668e-06, | |
| "loss": 0.3835, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 300.0, | |
| "grad_norm": 1.396098017692566, | |
| "learning_rate": 1.8383333333333334e-06, | |
| "loss": 0.3878, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 302.17391304347825, | |
| "grad_norm": 0.6345895528793335, | |
| "learning_rate": 1.7550000000000001e-06, | |
| "loss": 0.382, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 304.3478260869565, | |
| "grad_norm": 1.3194807767868042, | |
| "learning_rate": 1.6716666666666666e-06, | |
| "loss": 0.3885, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 304.3478260869565, | |
| "eval_loss": 0.4140428602695465, | |
| "eval_runtime": 6.4587, | |
| "eval_samples_per_second": 24.618, | |
| "eval_steps_per_second": 3.097, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 306.5217391304348, | |
| "grad_norm": 0.8248822689056396, | |
| "learning_rate": 1.5883333333333333e-06, | |
| "loss": 0.3847, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 308.69565217391306, | |
| "grad_norm": 0.635430097579956, | |
| "learning_rate": 1.505e-06, | |
| "loss": 0.3894, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 310.8695652173913, | |
| "grad_norm": 0.8129938244819641, | |
| "learning_rate": 1.4216666666666667e-06, | |
| "loss": 0.3862, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 313.04347826086956, | |
| "grad_norm": 0.8096001148223877, | |
| "learning_rate": 1.3383333333333334e-06, | |
| "loss": 0.3859, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 315.2173913043478, | |
| "grad_norm": 0.7998645901679993, | |
| "learning_rate": 1.255e-06, | |
| "loss": 0.3826, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 317.39130434782606, | |
| "grad_norm": 0.6718857288360596, | |
| "learning_rate": 1.1716666666666667e-06, | |
| "loss": 0.3896, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 319.5652173913044, | |
| "grad_norm": 0.5565139055252075, | |
| "learning_rate": 1.0883333333333334e-06, | |
| "loss": 0.3852, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 321.7391304347826, | |
| "grad_norm": 0.7727565765380859, | |
| "learning_rate": 1.0050000000000001e-06, | |
| "loss": 0.3866, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 323.9130434782609, | |
| "grad_norm": 0.7541866302490234, | |
| "learning_rate": 9.216666666666667e-07, | |
| "loss": 0.3818, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 326.0869565217391, | |
| "grad_norm": 0.7003363966941833, | |
| "learning_rate": 8.383333333333334e-07, | |
| "loss": 0.3849, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 326.0869565217391, | |
| "eval_loss": 0.41430386900901794, | |
| "eval_runtime": 6.7779, | |
| "eval_samples_per_second": 23.458, | |
| "eval_steps_per_second": 2.951, | |
| "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": 1.1098441388045722e+17, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |