| { | |
| "best_metric": 0.7934407364787112, | |
| "best_model_checkpoint": "results/facebook/hubert-large-ll60k/42/_retain/checkpoint-30000", | |
| "epoch": 75.80543272267846, | |
| "eval_steps": 400, | |
| "global_step": 30000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2526847757422615, | |
| "grad_norm": 7.531186580657959, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 4.4765, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.505369551484523, | |
| "grad_norm": 1.3038771152496338, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 4.1328, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7580543272267846, | |
| "grad_norm": 1.8434199094772339, | |
| "learning_rate": 5e-05, | |
| "loss": 3.8858, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.010739102969046, | |
| "grad_norm": 4.255322456359863, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 3.8141, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.010739102969046, | |
| "eval_accuracy": 0.08400460299194476, | |
| "eval_f1_macro": 0.008669601975968978, | |
| "eval_loss": 3.7715396881103516, | |
| "eval_runtime": 108.6457, | |
| "eval_samples_per_second": 79.985, | |
| "eval_steps_per_second": 2.504, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2634238787113077, | |
| "grad_norm": 5.987393379211426, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 3.7488, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5161086544535691, | |
| "grad_norm": 4.6245269775390625, | |
| "learning_rate": 0.0001, | |
| "loss": 3.6051, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.7687934301958306, | |
| "grad_norm": 7.047222137451172, | |
| "learning_rate": 0.00011666666666666667, | |
| "loss": 3.2059, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.021478205938092, | |
| "grad_norm": 11.504085540771484, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 2.7291, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.021478205938092, | |
| "eval_accuracy": 0.42681242807825087, | |
| "eval_f1_macro": 0.17578343384188652, | |
| "eval_loss": 2.323087215423584, | |
| "eval_runtime": 108.6246, | |
| "eval_samples_per_second": 80.0, | |
| "eval_steps_per_second": 2.504, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.274162981680354, | |
| "grad_norm": 9.302848815917969, | |
| "learning_rate": 0.00015, | |
| "loss": 2.2862, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.5268477574226154, | |
| "grad_norm": 16.832901000976562, | |
| "learning_rate": 0.00016666666666666666, | |
| "loss": 1.9558, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.779532533164877, | |
| "grad_norm": 18.594715118408203, | |
| "learning_rate": 0.00018333333333333334, | |
| "loss": 1.6921, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.0322173089071383, | |
| "grad_norm": 21.0137882232666, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4799, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.0322173089071383, | |
| "eval_accuracy": 0.68204833141542, | |
| "eval_f1_macro": 0.4656587521593648, | |
| "eval_loss": 1.2915173768997192, | |
| "eval_runtime": 108.6127, | |
| "eval_samples_per_second": 80.009, | |
| "eval_steps_per_second": 2.504, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.2849020846493997, | |
| "grad_norm": 10.087828636169434, | |
| "learning_rate": 0.00021666666666666668, | |
| "loss": 1.264, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.537586860391661, | |
| "grad_norm": 8.730626106262207, | |
| "learning_rate": 0.00023333333333333333, | |
| "loss": 1.1697, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.790271636133923, | |
| "grad_norm": 19.08763885498047, | |
| "learning_rate": 0.00025, | |
| "loss": 1.0774, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.042956411876184, | |
| "grad_norm": 12.934535026550293, | |
| "learning_rate": 0.0002666666666666667, | |
| "loss": 1.0037, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.042956411876184, | |
| "eval_accuracy": 0.7395857307249712, | |
| "eval_f1_macro": 0.5527706813137244, | |
| "eval_loss": 1.1226997375488281, | |
| "eval_runtime": 108.4906, | |
| "eval_samples_per_second": 80.099, | |
| "eval_steps_per_second": 2.507, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.295641187618446, | |
| "grad_norm": 11.546894073486328, | |
| "learning_rate": 0.00028333333333333335, | |
| "loss": 0.8737, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.548325963360708, | |
| "grad_norm": 10.242780685424805, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8296, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.801010739102969, | |
| "grad_norm": 9.696259498596191, | |
| "learning_rate": 0.00031666666666666665, | |
| "loss": 0.8139, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 5.053695514845231, | |
| "grad_norm": 6.417578220367432, | |
| "learning_rate": 0.0003333333333333333, | |
| "loss": 0.7768, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.053695514845231, | |
| "eval_accuracy": 0.7643268124280782, | |
| "eval_f1_macro": 0.5816116111187802, | |
| "eval_loss": 1.0890038013458252, | |
| "eval_runtime": 108.7404, | |
| "eval_samples_per_second": 79.915, | |
| "eval_steps_per_second": 2.501, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.306380290587492, | |
| "grad_norm": 9.01593017578125, | |
| "learning_rate": 0.00035, | |
| "loss": 0.7057, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 5.559065066329754, | |
| "grad_norm": 6.170841217041016, | |
| "learning_rate": 0.00036666666666666667, | |
| "loss": 0.6655, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 5.8117498420720155, | |
| "grad_norm": 9.565030097961426, | |
| "learning_rate": 0.00038333333333333334, | |
| "loss": 0.6767, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 6.0644346178142765, | |
| "grad_norm": 11.446185111999512, | |
| "learning_rate": 0.0004, | |
| "loss": 0.6207, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.0644346178142765, | |
| "eval_accuracy": 0.756156501726122, | |
| "eval_f1_macro": 0.5962013864268475, | |
| "eval_loss": 1.1590509414672852, | |
| "eval_runtime": 108.3732, | |
| "eval_samples_per_second": 80.186, | |
| "eval_steps_per_second": 2.51, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.317119393556538, | |
| "grad_norm": 6.382347583770752, | |
| "learning_rate": 0.0004166666666666667, | |
| "loss": 0.5489, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.569804169298799, | |
| "grad_norm": 7.796228408813477, | |
| "learning_rate": 0.00043333333333333337, | |
| "loss": 0.5615, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 6.822488945041061, | |
| "grad_norm": 10.157086372375488, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 0.61, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 7.075173720783323, | |
| "grad_norm": 9.361109733581543, | |
| "learning_rate": 0.00046666666666666666, | |
| "loss": 0.5457, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.075173720783323, | |
| "eval_accuracy": 0.75385500575374, | |
| "eval_f1_macro": 0.594201593201007, | |
| "eval_loss": 1.2054548263549805, | |
| "eval_runtime": 108.935, | |
| "eval_samples_per_second": 79.772, | |
| "eval_steps_per_second": 2.497, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.327858496525584, | |
| "grad_norm": 4.478394031524658, | |
| "learning_rate": 0.00048333333333333334, | |
| "loss": 0.5006, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 7.580543272267846, | |
| "grad_norm": 7.6330366134643555, | |
| "learning_rate": 0.0005, | |
| "loss": 0.5456, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.833228048010107, | |
| "grad_norm": 6.51686429977417, | |
| "learning_rate": 0.0004981481481481482, | |
| "loss": 0.5245, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 8.085912823752368, | |
| "grad_norm": 6.197965621948242, | |
| "learning_rate": 0.0004962962962962963, | |
| "loss": 0.4905, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.085912823752368, | |
| "eval_accuracy": 0.7558112773302647, | |
| "eval_f1_macro": 0.591751644264711, | |
| "eval_loss": 1.168681263923645, | |
| "eval_runtime": 108.6923, | |
| "eval_samples_per_second": 79.95, | |
| "eval_steps_per_second": 2.502, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.33859759949463, | |
| "grad_norm": 5.975191116333008, | |
| "learning_rate": 0.0004944444444444445, | |
| "loss": 0.4324, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 8.591282375236892, | |
| "grad_norm": 6.450153827667236, | |
| "learning_rate": 0.0004925925925925925, | |
| "loss": 0.4385, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 8.843967150979154, | |
| "grad_norm": 3.6089916229248047, | |
| "learning_rate": 0.0004907407407407408, | |
| "loss": 0.4312, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.096651926721416, | |
| "grad_norm": 5.325081825256348, | |
| "learning_rate": 0.0004888888888888889, | |
| "loss": 0.3977, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.096651926721416, | |
| "eval_accuracy": 0.7486766398158803, | |
| "eval_f1_macro": 0.5505823611427755, | |
| "eval_loss": 1.3269578218460083, | |
| "eval_runtime": 108.9968, | |
| "eval_samples_per_second": 79.727, | |
| "eval_steps_per_second": 2.495, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.349336702463676, | |
| "grad_norm": 6.121856212615967, | |
| "learning_rate": 0.00048703703703703707, | |
| "loss": 0.3766, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 9.602021478205938, | |
| "grad_norm": 3.0020909309387207, | |
| "learning_rate": 0.0004851851851851852, | |
| "loss": 0.3605, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 9.8547062539482, | |
| "grad_norm": 3.841557502746582, | |
| "learning_rate": 0.00048333333333333334, | |
| "loss": 0.3802, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 10.107391029690461, | |
| "grad_norm": 7.770158767700195, | |
| "learning_rate": 0.00048148148148148144, | |
| "loss": 0.3404, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.107391029690461, | |
| "eval_accuracy": 0.7620253164556962, | |
| "eval_f1_macro": 0.5823323463909736, | |
| "eval_loss": 1.2837411165237427, | |
| "eval_runtime": 108.7289, | |
| "eval_samples_per_second": 79.924, | |
| "eval_steps_per_second": 2.502, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.360075805432723, | |
| "grad_norm": 4.366064071655273, | |
| "learning_rate": 0.00047962962962962965, | |
| "loss": 0.314, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 10.612760581174983, | |
| "grad_norm": 9.094232559204102, | |
| "learning_rate": 0.0004777777777777778, | |
| "loss": 0.3174, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 10.865445356917245, | |
| "grad_norm": 4.532145023345947, | |
| "learning_rate": 0.0004759259259259259, | |
| "loss": 0.3083, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 11.118130132659507, | |
| "grad_norm": 2.8124845027923584, | |
| "learning_rate": 0.0004740740740740741, | |
| "loss": 0.28, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 11.118130132659507, | |
| "eval_accuracy": 0.7566168009205984, | |
| "eval_f1_macro": 0.5792237564159676, | |
| "eval_loss": 1.4293328523635864, | |
| "eval_runtime": 108.8172, | |
| "eval_samples_per_second": 79.859, | |
| "eval_steps_per_second": 2.5, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 11.37081490840177, | |
| "grad_norm": 3.8208773136138916, | |
| "learning_rate": 0.00047222222222222224, | |
| "loss": 0.2614, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.623499684144031, | |
| "grad_norm": 4.572866916656494, | |
| "learning_rate": 0.0004703703703703704, | |
| "loss": 0.2696, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 11.876184459886291, | |
| "grad_norm": 3.239572763442993, | |
| "learning_rate": 0.0004685185185185185, | |
| "loss": 0.2691, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 12.128869235628553, | |
| "grad_norm": 4.106109619140625, | |
| "learning_rate": 0.00046666666666666666, | |
| "loss": 0.2564, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 12.128869235628553, | |
| "eval_accuracy": 0.7440736478711162, | |
| "eval_f1_macro": 0.5907392176160607, | |
| "eval_loss": 1.45040762424469, | |
| "eval_runtime": 108.4161, | |
| "eval_samples_per_second": 80.154, | |
| "eval_steps_per_second": 2.509, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 12.381554011370815, | |
| "grad_norm": 2.1584527492523193, | |
| "learning_rate": 0.0004648148148148148, | |
| "loss": 0.2412, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 12.634238787113077, | |
| "grad_norm": 5.035179138183594, | |
| "learning_rate": 0.000462962962962963, | |
| "loss": 0.2345, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 12.886923562855339, | |
| "grad_norm": 4.4065752029418945, | |
| "learning_rate": 0.00046111111111111114, | |
| "loss": 0.2373, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 13.139608338597599, | |
| "grad_norm": 3.795588254928589, | |
| "learning_rate": 0.00045925925925925925, | |
| "loss": 0.2166, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 13.139608338597599, | |
| "eval_accuracy": 0.7540851553509781, | |
| "eval_f1_macro": 0.5753080086949651, | |
| "eval_loss": 1.4374622106552124, | |
| "eval_runtime": 108.3257, | |
| "eval_samples_per_second": 80.221, | |
| "eval_steps_per_second": 2.511, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 13.39229311433986, | |
| "grad_norm": 3.1705026626586914, | |
| "learning_rate": 0.00045740740740740746, | |
| "loss": 0.2114, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 13.644977890082123, | |
| "grad_norm": 5.109796047210693, | |
| "learning_rate": 0.00045555555555555556, | |
| "loss": 0.2207, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 13.897662665824384, | |
| "grad_norm": 5.884572982788086, | |
| "learning_rate": 0.0004537037037037037, | |
| "loss": 0.2135, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.150347441566646, | |
| "grad_norm": 4.0508246421813965, | |
| "learning_rate": 0.00045185185185185183, | |
| "loss": 0.2029, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 14.150347441566646, | |
| "eval_accuracy": 0.7579976985040277, | |
| "eval_f1_macro": 0.5636077065338464, | |
| "eval_loss": 1.3845158815383911, | |
| "eval_runtime": 108.3008, | |
| "eval_samples_per_second": 80.239, | |
| "eval_steps_per_second": 2.512, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 14.403032217308906, | |
| "grad_norm": 3.048769235610962, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 0.1842, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 14.655716993051168, | |
| "grad_norm": 5.531657695770264, | |
| "learning_rate": 0.00044814814814814815, | |
| "loss": 0.1969, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 14.90840176879343, | |
| "grad_norm": 4.874760150909424, | |
| "learning_rate": 0.0004462962962962963, | |
| "loss": 0.187, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 15.161086544535692, | |
| "grad_norm": 4.56043815612793, | |
| "learning_rate": 0.0004444444444444444, | |
| "loss": 0.1781, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.161086544535692, | |
| "eval_accuracy": 0.7592635212888378, | |
| "eval_f1_macro": 0.6140090678141539, | |
| "eval_loss": 1.4697922468185425, | |
| "eval_runtime": 108.1742, | |
| "eval_samples_per_second": 80.333, | |
| "eval_steps_per_second": 2.514, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.413771320277954, | |
| "grad_norm": 2.3174920082092285, | |
| "learning_rate": 0.0004425925925925926, | |
| "loss": 0.1646, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 15.666456096020214, | |
| "grad_norm": 6.532171726226807, | |
| "learning_rate": 0.0004407407407407408, | |
| "loss": 0.176, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 15.919140871762476, | |
| "grad_norm": 8.349685668945312, | |
| "learning_rate": 0.0004388888888888889, | |
| "loss": 0.1819, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 16.171825647504736, | |
| "grad_norm": 2.4073634147644043, | |
| "learning_rate": 0.00043703703703703705, | |
| "loss": 0.1559, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 16.171825647504736, | |
| "eval_accuracy": 0.7581127733026467, | |
| "eval_f1_macro": 0.6005969862926049, | |
| "eval_loss": 1.6600337028503418, | |
| "eval_runtime": 108.333, | |
| "eval_samples_per_second": 80.216, | |
| "eval_steps_per_second": 2.511, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 16.424510423247, | |
| "grad_norm": 5.727371692657471, | |
| "learning_rate": 0.0004351851851851852, | |
| "loss": 0.1574, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 16.67719519898926, | |
| "grad_norm": 5.15314245223999, | |
| "learning_rate": 0.00043333333333333337, | |
| "loss": 0.162, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 16.929879974731524, | |
| "grad_norm": 4.066269397735596, | |
| "learning_rate": 0.00043148148148148147, | |
| "loss": 0.1541, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 17.182564750473784, | |
| "grad_norm": 2.442532777786255, | |
| "learning_rate": 0.00042962962962962963, | |
| "loss": 0.1449, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 17.182564750473784, | |
| "eval_accuracy": 0.7632911392405063, | |
| "eval_f1_macro": 0.5902398414509258, | |
| "eval_loss": 1.5675625801086426, | |
| "eval_runtime": 108.2685, | |
| "eval_samples_per_second": 80.263, | |
| "eval_steps_per_second": 2.512, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 17.435249526216044, | |
| "grad_norm": 2.888688564300537, | |
| "learning_rate": 0.0004277777777777778, | |
| "loss": 0.1602, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 17.687934301958308, | |
| "grad_norm": 1.7030471563339233, | |
| "learning_rate": 0.00042592592592592595, | |
| "loss": 0.141, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 17.940619077700568, | |
| "grad_norm": 3.838587760925293, | |
| "learning_rate": 0.00042407407407407406, | |
| "loss": 0.14, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 18.19330385344283, | |
| "grad_norm": 2.5156667232513428, | |
| "learning_rate": 0.0004222222222222222, | |
| "loss": 0.144, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 18.19330385344283, | |
| "eval_accuracy": 0.7498273878020714, | |
| "eval_f1_macro": 0.5654132069159379, | |
| "eval_loss": 1.4789820909500122, | |
| "eval_runtime": 108.261, | |
| "eval_samples_per_second": 80.269, | |
| "eval_steps_per_second": 2.512, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 18.44598862918509, | |
| "grad_norm": 4.657927513122559, | |
| "learning_rate": 0.00042037037037037043, | |
| "loss": 0.1385, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 18.69867340492735, | |
| "grad_norm": 5.224093914031982, | |
| "learning_rate": 0.00041851851851851853, | |
| "loss": 0.1333, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 18.951358180669615, | |
| "grad_norm": 2.436795949935913, | |
| "learning_rate": 0.0004166666666666667, | |
| "loss": 0.1333, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.204042956411875, | |
| "grad_norm": 5.1690354347229, | |
| "learning_rate": 0.0004148148148148148, | |
| "loss": 0.1136, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 19.204042956411875, | |
| "eval_accuracy": 0.7498273878020714, | |
| "eval_f1_macro": 0.5531386036990584, | |
| "eval_loss": 1.6063206195831299, | |
| "eval_runtime": 108.3182, | |
| "eval_samples_per_second": 80.227, | |
| "eval_steps_per_second": 2.511, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 19.45672773215414, | |
| "grad_norm": 1.983140230178833, | |
| "learning_rate": 0.00041296296296296296, | |
| "loss": 0.1265, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 19.7094125078964, | |
| "grad_norm": 2.0450029373168945, | |
| "learning_rate": 0.0004111111111111111, | |
| "loss": 0.1221, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 19.96209728363866, | |
| "grad_norm": 6.7355804443359375, | |
| "learning_rate": 0.0004092592592592593, | |
| "loss": 0.1369, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 20.214782059380923, | |
| "grad_norm": 3.7778196334838867, | |
| "learning_rate": 0.0004074074074074074, | |
| "loss": 0.1082, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 20.214782059380923, | |
| "eval_accuracy": 0.7598388952819333, | |
| "eval_f1_macro": 0.5650987069075204, | |
| "eval_loss": 1.6282719373703003, | |
| "eval_runtime": 108.2582, | |
| "eval_samples_per_second": 80.271, | |
| "eval_steps_per_second": 2.513, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 20.467466835123183, | |
| "grad_norm": 3.0025575160980225, | |
| "learning_rate": 0.00040555555555555554, | |
| "loss": 0.1108, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 20.720151610865447, | |
| "grad_norm": 3.365360736846924, | |
| "learning_rate": 0.00040370370370370375, | |
| "loss": 0.1188, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 20.972836386607707, | |
| "grad_norm": 1.7243456840515137, | |
| "learning_rate": 0.00040185185185185186, | |
| "loss": 0.1152, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 21.225521162349967, | |
| "grad_norm": 2.3120861053466797, | |
| "learning_rate": 0.0004, | |
| "loss": 0.109, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 21.225521162349967, | |
| "eval_accuracy": 0.7635212888377445, | |
| "eval_f1_macro": 0.569418631056331, | |
| "eval_loss": 1.752591848373413, | |
| "eval_runtime": 108.3807, | |
| "eval_samples_per_second": 80.18, | |
| "eval_steps_per_second": 2.51, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 21.47820593809223, | |
| "grad_norm": 3.04634165763855, | |
| "learning_rate": 0.0003981481481481481, | |
| "loss": 0.1219, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 21.73089071383449, | |
| "grad_norm": 5.413991928100586, | |
| "learning_rate": 0.00039629629629629634, | |
| "loss": 0.1122, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 21.983575489576754, | |
| "grad_norm": 3.94024395942688, | |
| "learning_rate": 0.00039444444444444444, | |
| "loss": 0.1023, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 22.236260265319014, | |
| "grad_norm": 4.625360488891602, | |
| "learning_rate": 0.0003925925925925926, | |
| "loss": 0.1173, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 22.236260265319014, | |
| "eval_accuracy": 0.7597238204833141, | |
| "eval_f1_macro": 0.5837145389837423, | |
| "eval_loss": 1.5915985107421875, | |
| "eval_runtime": 108.3288, | |
| "eval_samples_per_second": 80.219, | |
| "eval_steps_per_second": 2.511, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 22.488945041061275, | |
| "grad_norm": 4.926342010498047, | |
| "learning_rate": 0.0003907407407407407, | |
| "loss": 0.1059, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 22.74162981680354, | |
| "grad_norm": 3.0098073482513428, | |
| "learning_rate": 0.0003888888888888889, | |
| "loss": 0.1021, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 22.9943145925458, | |
| "grad_norm": 1.6922105550765991, | |
| "learning_rate": 0.000387037037037037, | |
| "loss": 0.1046, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 23.246999368288062, | |
| "grad_norm": 4.048877239227295, | |
| "learning_rate": 0.0003851851851851852, | |
| "loss": 0.094, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 23.246999368288062, | |
| "eval_accuracy": 0.7614499424626007, | |
| "eval_f1_macro": 0.575035765001433, | |
| "eval_loss": 1.631859302520752, | |
| "eval_runtime": 108.7827, | |
| "eval_samples_per_second": 79.884, | |
| "eval_steps_per_second": 2.5, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 23.499684144030322, | |
| "grad_norm": 3.0817179679870605, | |
| "learning_rate": 0.00038333333333333334, | |
| "loss": 0.1134, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 23.752368919772582, | |
| "grad_norm": 0.7254201769828796, | |
| "learning_rate": 0.0003814814814814815, | |
| "loss": 0.0916, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 24.005053695514846, | |
| "grad_norm": 1.1015946865081787, | |
| "learning_rate": 0.00037962962962962966, | |
| "loss": 0.1014, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 24.257738471257106, | |
| "grad_norm": 1.6841648817062378, | |
| "learning_rate": 0.00037777777777777777, | |
| "loss": 0.0819, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 24.257738471257106, | |
| "eval_accuracy": 0.7601841196777905, | |
| "eval_f1_macro": 0.5741548033552556, | |
| "eval_loss": 1.7696462869644165, | |
| "eval_runtime": 108.7945, | |
| "eval_samples_per_second": 79.875, | |
| "eval_steps_per_second": 2.5, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 24.51042324699937, | |
| "grad_norm": 1.9033466577529907, | |
| "learning_rate": 0.00037592592592592593, | |
| "loss": 0.1013, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 24.76310802274163, | |
| "grad_norm": 3.788855791091919, | |
| "learning_rate": 0.0003740740740740741, | |
| "loss": 0.1057, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 25.01579279848389, | |
| "grad_norm": 1.6506637334823608, | |
| "learning_rate": 0.00037222222222222225, | |
| "loss": 0.0979, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 25.268477574226154, | |
| "grad_norm": 4.115843296051025, | |
| "learning_rate": 0.00037037037037037035, | |
| "loss": 0.0978, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 25.268477574226154, | |
| "eval_accuracy": 0.7659378596087457, | |
| "eval_f1_macro": 0.5667775496828193, | |
| "eval_loss": 1.633931040763855, | |
| "eval_runtime": 108.6433, | |
| "eval_samples_per_second": 79.987, | |
| "eval_steps_per_second": 2.504, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 25.521162349968414, | |
| "grad_norm": 3.1308891773223877, | |
| "learning_rate": 0.0003685185185185185, | |
| "loss": 0.0804, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 25.773847125710677, | |
| "grad_norm": 3.1427507400512695, | |
| "learning_rate": 0.00036666666666666667, | |
| "loss": 0.0861, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 26.026531901452937, | |
| "grad_norm": 2.7073075771331787, | |
| "learning_rate": 0.00036481481481481483, | |
| "loss": 0.0846, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 26.279216677195198, | |
| "grad_norm": 0.8588293194770813, | |
| "learning_rate": 0.000362962962962963, | |
| "loss": 0.0854, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 26.279216677195198, | |
| "eval_accuracy": 0.7668584579976985, | |
| "eval_f1_macro": 0.566287765137334, | |
| "eval_loss": 1.6085623502731323, | |
| "eval_runtime": 108.9617, | |
| "eval_samples_per_second": 79.753, | |
| "eval_steps_per_second": 2.496, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 26.53190145293746, | |
| "grad_norm": 2.541748285293579, | |
| "learning_rate": 0.0003611111111111111, | |
| "loss": 0.0828, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 26.78458622867972, | |
| "grad_norm": 4.645153999328613, | |
| "learning_rate": 0.0003592592592592593, | |
| "loss": 0.0943, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 27.037271004421985, | |
| "grad_norm": 2.1111841201782227, | |
| "learning_rate": 0.0003574074074074074, | |
| "loss": 0.0757, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 27.289955780164245, | |
| "grad_norm": 1.5534864664077759, | |
| "learning_rate": 0.00035555555555555557, | |
| "loss": 0.0818, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 27.289955780164245, | |
| "eval_accuracy": 0.768584579976985, | |
| "eval_f1_macro": 0.5967195650070578, | |
| "eval_loss": 1.6560544967651367, | |
| "eval_runtime": 108.8685, | |
| "eval_samples_per_second": 79.821, | |
| "eval_steps_per_second": 2.498, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 27.542640555906505, | |
| "grad_norm": 1.9038456678390503, | |
| "learning_rate": 0.0003537037037037037, | |
| "loss": 0.081, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 27.79532533164877, | |
| "grad_norm": 1.55673348903656, | |
| "learning_rate": 0.0003518518518518519, | |
| "loss": 0.0828, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 28.04801010739103, | |
| "grad_norm": 5.855372905731201, | |
| "learning_rate": 0.00035, | |
| "loss": 0.0862, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 28.300694883133293, | |
| "grad_norm": 3.111088991165161, | |
| "learning_rate": 0.00034814814814814816, | |
| "loss": 0.0725, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 28.300694883133293, | |
| "eval_accuracy": 0.7542002301495973, | |
| "eval_f1_macro": 0.5740863312977772, | |
| "eval_loss": 1.7682907581329346, | |
| "eval_runtime": 144.2676, | |
| "eval_samples_per_second": 60.235, | |
| "eval_steps_per_second": 1.885, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 28.553379658875553, | |
| "grad_norm": 1.5844389200210571, | |
| "learning_rate": 0.00034629629629629626, | |
| "loss": 0.0771, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 28.806064434617813, | |
| "grad_norm": 3.93422269821167, | |
| "learning_rate": 0.0003444444444444445, | |
| "loss": 0.0863, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 29.058749210360077, | |
| "grad_norm": 1.9465683698654175, | |
| "learning_rate": 0.00034259259259259263, | |
| "loss": 0.0777, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 29.311433986102337, | |
| "grad_norm": 1.951295018196106, | |
| "learning_rate": 0.00034074074074074074, | |
| "loss": 0.0672, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 29.311433986102337, | |
| "eval_accuracy": 0.7616800920598389, | |
| "eval_f1_macro": 0.5831099181645485, | |
| "eval_loss": 1.8726670742034912, | |
| "eval_runtime": 143.7286, | |
| "eval_samples_per_second": 60.461, | |
| "eval_steps_per_second": 1.892, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 29.5641187618446, | |
| "grad_norm": 2.5593600273132324, | |
| "learning_rate": 0.0003388888888888889, | |
| "loss": 0.0707, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 29.81680353758686, | |
| "grad_norm": 1.8622692823410034, | |
| "learning_rate": 0.00033703703703703706, | |
| "loss": 0.0709, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 30.06948831332912, | |
| "grad_norm": 2.1602845191955566, | |
| "learning_rate": 0.0003351851851851852, | |
| "loss": 0.0653, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 30.322173089071384, | |
| "grad_norm": 3.438899278640747, | |
| "learning_rate": 0.0003333333333333333, | |
| "loss": 0.076, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 30.322173089071384, | |
| "eval_accuracy": 0.7635212888377445, | |
| "eval_f1_macro": 0.608710633839459, | |
| "eval_loss": 1.6385372877120972, | |
| "eval_runtime": 143.5758, | |
| "eval_samples_per_second": 60.526, | |
| "eval_steps_per_second": 1.894, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 30.574857864813644, | |
| "grad_norm": 4.342224597930908, | |
| "learning_rate": 0.0003314814814814815, | |
| "loss": 0.0682, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 30.827542640555908, | |
| "grad_norm": 3.186768054962158, | |
| "learning_rate": 0.0003296296296296296, | |
| "loss": 0.0705, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 31.080227416298168, | |
| "grad_norm": 1.2158719301223755, | |
| "learning_rate": 0.0003277777777777778, | |
| "loss": 0.0665, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 31.33291219204043, | |
| "grad_norm": 1.0089023113250732, | |
| "learning_rate": 0.00032592592592592596, | |
| "loss": 0.0702, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 31.33291219204043, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_f1_macro": 0.5988474647551651, | |
| "eval_loss": 1.7596935033798218, | |
| "eval_runtime": 145.7714, | |
| "eval_samples_per_second": 59.614, | |
| "eval_steps_per_second": 1.866, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 31.585596967782692, | |
| "grad_norm": 1.9418379068374634, | |
| "learning_rate": 0.00032407407407407406, | |
| "loss": 0.0674, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 31.838281743524952, | |
| "grad_norm": 2.2405946254730225, | |
| "learning_rate": 0.0003222222222222222, | |
| "loss": 0.0645, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 32.09096651926721, | |
| "grad_norm": 1.296802043914795, | |
| "learning_rate": 0.0003203703703703704, | |
| "loss": 0.0654, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 32.34365129500947, | |
| "grad_norm": 5.712021827697754, | |
| "learning_rate": 0.00031851851851851854, | |
| "loss": 0.0644, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 32.34365129500947, | |
| "eval_accuracy": 0.7686996547756041, | |
| "eval_f1_macro": 0.5840964451710614, | |
| "eval_loss": 1.795427918434143, | |
| "eval_runtime": 137.5518, | |
| "eval_samples_per_second": 63.176, | |
| "eval_steps_per_second": 1.977, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 32.59633607075174, | |
| "grad_norm": 1.1909292936325073, | |
| "learning_rate": 0.00031666666666666665, | |
| "loss": 0.0638, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 32.849020846494, | |
| "grad_norm": 4.957464694976807, | |
| "learning_rate": 0.0003148148148148148, | |
| "loss": 0.0566, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 33.10170562223626, | |
| "grad_norm": 0.9219366908073425, | |
| "learning_rate": 0.00031296296296296297, | |
| "loss": 0.0622, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 33.35439039797852, | |
| "grad_norm": 0.8782810568809509, | |
| "learning_rate": 0.0003111111111111111, | |
| "loss": 0.0521, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 33.35439039797852, | |
| "eval_accuracy": 0.7619102416570771, | |
| "eval_f1_macro": 0.6192131512606447, | |
| "eval_loss": 1.901371955871582, | |
| "eval_runtime": 134.6686, | |
| "eval_samples_per_second": 64.529, | |
| "eval_steps_per_second": 2.02, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 33.60707517372078, | |
| "grad_norm": 2.130397319793701, | |
| "learning_rate": 0.00030925925925925923, | |
| "loss": 0.0686, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 33.85975994946305, | |
| "grad_norm": 1.2521294355392456, | |
| "learning_rate": 0.0003074074074074074, | |
| "loss": 0.058, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 34.11244472520531, | |
| "grad_norm": 2.1233527660369873, | |
| "learning_rate": 0.0003055555555555556, | |
| "loss": 0.0528, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 34.36512950094757, | |
| "grad_norm": 2.557140827178955, | |
| "learning_rate": 0.0003037037037037037, | |
| "loss": 0.0489, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 34.36512950094757, | |
| "eval_accuracy": 0.768584579976985, | |
| "eval_f1_macro": 0.6008906251293721, | |
| "eval_loss": 1.8431710004806519, | |
| "eval_runtime": 109.9242, | |
| "eval_samples_per_second": 79.054, | |
| "eval_steps_per_second": 2.474, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 34.61781427668983, | |
| "grad_norm": 1.0921064615249634, | |
| "learning_rate": 0.00030185185185185187, | |
| "loss": 0.0561, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 34.87049905243209, | |
| "grad_norm": 2.349817991256714, | |
| "learning_rate": 0.0003, | |
| "loss": 0.0565, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 35.123183828174355, | |
| "grad_norm": 1.187012791633606, | |
| "learning_rate": 0.0002981481481481482, | |
| "loss": 0.049, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 35.375868603916615, | |
| "grad_norm": 0.04951922222971916, | |
| "learning_rate": 0.0002962962962962963, | |
| "loss": 0.0447, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 35.375868603916615, | |
| "eval_accuracy": 0.769620253164557, | |
| "eval_f1_macro": 0.5941725933407251, | |
| "eval_loss": 1.8310672044754028, | |
| "eval_runtime": 109.8974, | |
| "eval_samples_per_second": 79.074, | |
| "eval_steps_per_second": 2.475, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 35.628553379658875, | |
| "grad_norm": 1.9226619005203247, | |
| "learning_rate": 0.00029444444444444445, | |
| "loss": 0.0575, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 35.881238155401135, | |
| "grad_norm": 0.9691677093505859, | |
| "learning_rate": 0.00029259259259259256, | |
| "loss": 0.0569, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 36.133922931143395, | |
| "grad_norm": 3.869662046432495, | |
| "learning_rate": 0.00029074074074074077, | |
| "loss": 0.0572, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 36.38660770688566, | |
| "grad_norm": 1.026167869567871, | |
| "learning_rate": 0.0002888888888888889, | |
| "loss": 0.0506, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 36.38660770688566, | |
| "eval_accuracy": 0.7692750287686997, | |
| "eval_f1_macro": 0.5692835911324542, | |
| "eval_loss": 1.7118504047393799, | |
| "eval_runtime": 110.0615, | |
| "eval_samples_per_second": 78.956, | |
| "eval_steps_per_second": 2.471, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 36.63929248262792, | |
| "grad_norm": 0.6115164756774902, | |
| "learning_rate": 0.00028703703703703703, | |
| "loss": 0.0472, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 36.89197725837018, | |
| "grad_norm": 2.26706862449646, | |
| "learning_rate": 0.0002851851851851852, | |
| "loss": 0.0507, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 37.14466203411244, | |
| "grad_norm": 1.0484291315078735, | |
| "learning_rate": 0.00028333333333333335, | |
| "loss": 0.0533, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 37.3973468098547, | |
| "grad_norm": 1.587339162826538, | |
| "learning_rate": 0.0002814814814814815, | |
| "loss": 0.0489, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 37.3973468098547, | |
| "eval_accuracy": 0.7665132336018412, | |
| "eval_f1_macro": 0.5648640266723668, | |
| "eval_loss": 1.8756378889083862, | |
| "eval_runtime": 109.745, | |
| "eval_samples_per_second": 79.184, | |
| "eval_steps_per_second": 2.478, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 37.65003158559697, | |
| "grad_norm": 0.5741543173789978, | |
| "learning_rate": 0.0002796296296296296, | |
| "loss": 0.0539, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 37.90271636133923, | |
| "grad_norm": 2.3613216876983643, | |
| "learning_rate": 0.0002777777777777778, | |
| "loss": 0.0601, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 38.15540113708149, | |
| "grad_norm": 2.9896764755249023, | |
| "learning_rate": 0.00027592592592592594, | |
| "loss": 0.0441, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 38.40808591282375, | |
| "grad_norm": 2.065026044845581, | |
| "learning_rate": 0.0002740740740740741, | |
| "loss": 0.0523, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 38.40808591282375, | |
| "eval_accuracy": 0.777445339470656, | |
| "eval_f1_macro": 0.5952390760705957, | |
| "eval_loss": 1.6472355127334595, | |
| "eval_runtime": 109.7433, | |
| "eval_samples_per_second": 79.185, | |
| "eval_steps_per_second": 2.479, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 38.66077068856601, | |
| "grad_norm": 0.533388078212738, | |
| "learning_rate": 0.0002722222222222222, | |
| "loss": 0.0465, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 38.91345546430828, | |
| "grad_norm": 2.1074326038360596, | |
| "learning_rate": 0.00027037037037037036, | |
| "loss": 0.0398, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 39.16614024005054, | |
| "grad_norm": 1.6379950046539307, | |
| "learning_rate": 0.0002685185185185186, | |
| "loss": 0.04, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 39.4188250157928, | |
| "grad_norm": 1.9422637224197388, | |
| "learning_rate": 0.0002666666666666667, | |
| "loss": 0.0384, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 39.4188250157928, | |
| "eval_accuracy": 0.7736478711162256, | |
| "eval_f1_macro": 0.6025598876629095, | |
| "eval_loss": 1.8879679441452026, | |
| "eval_runtime": 109.8175, | |
| "eval_samples_per_second": 79.131, | |
| "eval_steps_per_second": 2.477, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 39.67150979153506, | |
| "grad_norm": 1.7022106647491455, | |
| "learning_rate": 0.00026481481481481484, | |
| "loss": 0.036, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 39.92419456727732, | |
| "grad_norm": 1.5560113191604614, | |
| "learning_rate": 0.00026296296296296294, | |
| "loss": 0.0419, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 40.176879343019586, | |
| "grad_norm": 3.473234176635742, | |
| "learning_rate": 0.00026111111111111116, | |
| "loss": 0.0375, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 40.429564118761846, | |
| "grad_norm": 1.8860403299331665, | |
| "learning_rate": 0.00025925925925925926, | |
| "loss": 0.0423, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 40.429564118761846, | |
| "eval_accuracy": 0.775258918296893, | |
| "eval_f1_macro": 0.5970358418964591, | |
| "eval_loss": 1.8289481401443481, | |
| "eval_runtime": 109.6274, | |
| "eval_samples_per_second": 79.268, | |
| "eval_steps_per_second": 2.481, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 40.682248894504106, | |
| "grad_norm": 3.3260364532470703, | |
| "learning_rate": 0.0002574074074074074, | |
| "loss": 0.0384, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 40.934933670246366, | |
| "grad_norm": 0.306608110666275, | |
| "learning_rate": 0.00025555555555555553, | |
| "loss": 0.0485, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 41.187618445988626, | |
| "grad_norm": 1.0209686756134033, | |
| "learning_rate": 0.00025370370370370374, | |
| "loss": 0.0388, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 41.44030322173089, | |
| "grad_norm": 1.1874313354492188, | |
| "learning_rate": 0.00025185185185185185, | |
| "loss": 0.0421, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 41.44030322173089, | |
| "eval_accuracy": 0.7693901035673187, | |
| "eval_f1_macro": 0.6018061729163716, | |
| "eval_loss": 1.6877504587173462, | |
| "eval_runtime": 110.175, | |
| "eval_samples_per_second": 78.874, | |
| "eval_steps_per_second": 2.469, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 41.69298799747315, | |
| "grad_norm": 1.121667742729187, | |
| "learning_rate": 0.00025, | |
| "loss": 0.0382, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 41.94567277321541, | |
| "grad_norm": 1.1757502555847168, | |
| "learning_rate": 0.00024814814814814816, | |
| "loss": 0.0397, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 42.198357548957674, | |
| "grad_norm": 9.486854553222656, | |
| "learning_rate": 0.00024629629629629627, | |
| "loss": 0.0409, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 42.451042324699934, | |
| "grad_norm": 0.4830320179462433, | |
| "learning_rate": 0.00024444444444444443, | |
| "loss": 0.0404, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 42.451042324699934, | |
| "eval_accuracy": 0.7681242807825086, | |
| "eval_f1_macro": 0.5853419424167555, | |
| "eval_loss": 1.8393748998641968, | |
| "eval_runtime": 109.7881, | |
| "eval_samples_per_second": 79.152, | |
| "eval_steps_per_second": 2.478, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 42.7037271004422, | |
| "grad_norm": 1.6823731660842896, | |
| "learning_rate": 0.0002425925925925926, | |
| "loss": 0.0398, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 42.95641187618446, | |
| "grad_norm": 6.2725749015808105, | |
| "learning_rate": 0.00024074074074074072, | |
| "loss": 0.041, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 43.20909665192672, | |
| "grad_norm": 1.0358606576919556, | |
| "learning_rate": 0.0002388888888888889, | |
| "loss": 0.0359, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 43.46178142766898, | |
| "grad_norm": 0.5175994634628296, | |
| "learning_rate": 0.00023703703703703704, | |
| "loss": 0.0311, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 43.46178142766898, | |
| "eval_accuracy": 0.7772151898734178, | |
| "eval_f1_macro": 0.5963074285064315, | |
| "eval_loss": 1.7879575490951538, | |
| "eval_runtime": 109.7858, | |
| "eval_samples_per_second": 79.154, | |
| "eval_steps_per_second": 2.478, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 43.71446620341124, | |
| "grad_norm": 3.763942003250122, | |
| "learning_rate": 0.0002351851851851852, | |
| "loss": 0.0283, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 43.96715097915351, | |
| "grad_norm": 0.7010844349861145, | |
| "learning_rate": 0.00023333333333333333, | |
| "loss": 0.0392, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 44.21983575489577, | |
| "grad_norm": 1.6681573390960693, | |
| "learning_rate": 0.0002314814814814815, | |
| "loss": 0.0306, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 44.47252053063803, | |
| "grad_norm": 3.1810269355773926, | |
| "learning_rate": 0.00022962962962962962, | |
| "loss": 0.0421, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 44.47252053063803, | |
| "eval_accuracy": 0.7758342922899885, | |
| "eval_f1_macro": 0.601629063986742, | |
| "eval_loss": 1.70499587059021, | |
| "eval_runtime": 109.8026, | |
| "eval_samples_per_second": 79.142, | |
| "eval_steps_per_second": 2.477, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 44.72520530638029, | |
| "grad_norm": 1.0086809396743774, | |
| "learning_rate": 0.00022777777777777778, | |
| "loss": 0.0337, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 44.97789008212255, | |
| "grad_norm": 0.7673502564430237, | |
| "learning_rate": 0.00022592592592592591, | |
| "loss": 0.0381, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 45.230574857864816, | |
| "grad_norm": 0.9791069626808167, | |
| "learning_rate": 0.00022407407407407407, | |
| "loss": 0.0263, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 45.48325963360708, | |
| "grad_norm": 2.0957231521606445, | |
| "learning_rate": 0.0002222222222222222, | |
| "loss": 0.039, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 45.48325963360708, | |
| "eval_accuracy": 0.7800920598388953, | |
| "eval_f1_macro": 0.6025897002203272, | |
| "eval_loss": 1.7869104146957397, | |
| "eval_runtime": 109.5691, | |
| "eval_samples_per_second": 79.311, | |
| "eval_steps_per_second": 2.482, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 45.73594440934934, | |
| "grad_norm": 2.0811707973480225, | |
| "learning_rate": 0.0002203703703703704, | |
| "loss": 0.0284, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 45.9886291850916, | |
| "grad_norm": 4.134197235107422, | |
| "learning_rate": 0.00021851851851851852, | |
| "loss": 0.0347, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 46.24131396083386, | |
| "grad_norm": 0.5542005300521851, | |
| "learning_rate": 0.00021666666666666668, | |
| "loss": 0.0293, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 46.493998736576124, | |
| "grad_norm": 2.4530539512634277, | |
| "learning_rate": 0.00021481481481481482, | |
| "loss": 0.0268, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 46.493998736576124, | |
| "eval_accuracy": 0.7783659378596087, | |
| "eval_f1_macro": 0.5935536715539459, | |
| "eval_loss": 1.9122555255889893, | |
| "eval_runtime": 109.4826, | |
| "eval_samples_per_second": 79.373, | |
| "eval_steps_per_second": 2.484, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 46.746683512318384, | |
| "grad_norm": 0.5376617312431335, | |
| "learning_rate": 0.00021296296296296298, | |
| "loss": 0.0348, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 46.999368288060644, | |
| "grad_norm": 1.4315904378890991, | |
| "learning_rate": 0.0002111111111111111, | |
| "loss": 0.036, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 47.252053063802904, | |
| "grad_norm": 0.866862952709198, | |
| "learning_rate": 0.00020925925925925927, | |
| "loss": 0.0273, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 47.504737839545164, | |
| "grad_norm": 2.0620691776275635, | |
| "learning_rate": 0.0002074074074074074, | |
| "loss": 0.0305, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 47.504737839545164, | |
| "eval_accuracy": 0.7820483314154201, | |
| "eval_f1_macro": 0.5937346523730859, | |
| "eval_loss": 1.7703049182891846, | |
| "eval_runtime": 109.6053, | |
| "eval_samples_per_second": 79.284, | |
| "eval_steps_per_second": 2.482, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 47.75742261528743, | |
| "grad_norm": 1.7660882472991943, | |
| "learning_rate": 0.00020555555555555556, | |
| "loss": 0.0248, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 48.01010739102969, | |
| "grad_norm": 0.0962400734424591, | |
| "learning_rate": 0.0002037037037037037, | |
| "loss": 0.0316, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 48.26279216677195, | |
| "grad_norm": 1.133728265762329, | |
| "learning_rate": 0.00020185185185185188, | |
| "loss": 0.0308, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 48.51547694251421, | |
| "grad_norm": 0.4573793411254883, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0279, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 48.51547694251421, | |
| "eval_accuracy": 0.7795166858457998, | |
| "eval_f1_macro": 0.5972420776230559, | |
| "eval_loss": 1.6281923055648804, | |
| "eval_runtime": 109.5117, | |
| "eval_samples_per_second": 79.352, | |
| "eval_steps_per_second": 2.484, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 48.76816171825647, | |
| "grad_norm": 1.396851658821106, | |
| "learning_rate": 0.00019814814814814817, | |
| "loss": 0.0216, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 49.02084649399874, | |
| "grad_norm": 1.0762335062026978, | |
| "learning_rate": 0.0001962962962962963, | |
| "loss": 0.0219, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 49.273531269741, | |
| "grad_norm": 5.425212860107422, | |
| "learning_rate": 0.00019444444444444446, | |
| "loss": 0.0276, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 49.52621604548326, | |
| "grad_norm": 0.31777116656303406, | |
| "learning_rate": 0.0001925925925925926, | |
| "loss": 0.0238, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 49.52621604548326, | |
| "eval_accuracy": 0.768584579976985, | |
| "eval_f1_macro": 0.5921492404881852, | |
| "eval_loss": 1.8642910718917847, | |
| "eval_runtime": 109.7124, | |
| "eval_samples_per_second": 79.207, | |
| "eval_steps_per_second": 2.479, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 49.77890082122552, | |
| "grad_norm": 2.751400947570801, | |
| "learning_rate": 0.00019074074074074075, | |
| "loss": 0.0271, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 50.03158559696778, | |
| "grad_norm": 2.2741167545318604, | |
| "learning_rate": 0.00018888888888888888, | |
| "loss": 0.0238, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 50.28427037271005, | |
| "grad_norm": 0.2512575387954712, | |
| "learning_rate": 0.00018703703703703704, | |
| "loss": 0.0288, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 50.53695514845231, | |
| "grad_norm": 1.9498554468154907, | |
| "learning_rate": 0.00018518518518518518, | |
| "loss": 0.0243, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 50.53695514845231, | |
| "eval_accuracy": 0.770771001150748, | |
| "eval_f1_macro": 0.598576194898749, | |
| "eval_loss": 1.976339340209961, | |
| "eval_runtime": 109.5523, | |
| "eval_samples_per_second": 79.323, | |
| "eval_steps_per_second": 2.483, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 50.78963992419457, | |
| "grad_norm": 0.7483634948730469, | |
| "learning_rate": 0.00018333333333333334, | |
| "loss": 0.0253, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 51.04232469993683, | |
| "grad_norm": 1.7610820531845093, | |
| "learning_rate": 0.0001814814814814815, | |
| "loss": 0.025, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 51.29500947567909, | |
| "grad_norm": 2.2282676696777344, | |
| "learning_rate": 0.00017962962962962965, | |
| "loss": 0.0205, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 51.547694251421355, | |
| "grad_norm": 1.0908398628234863, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 0.0264, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 51.547694251421355, | |
| "eval_accuracy": 0.7785960874568469, | |
| "eval_f1_macro": 0.6159147237572236, | |
| "eval_loss": 1.9196436405181885, | |
| "eval_runtime": 109.4823, | |
| "eval_samples_per_second": 79.374, | |
| "eval_steps_per_second": 2.484, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 51.800379027163615, | |
| "grad_norm": 3.9686079025268555, | |
| "learning_rate": 0.00017592592592592595, | |
| "loss": 0.0215, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 52.053063802905875, | |
| "grad_norm": 1.9715676307678223, | |
| "learning_rate": 0.00017407407407407408, | |
| "loss": 0.0247, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 52.305748578648135, | |
| "grad_norm": 0.6495607495307922, | |
| "learning_rate": 0.00017222222222222224, | |
| "loss": 0.0249, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 52.558433354390395, | |
| "grad_norm": 1.0535669326782227, | |
| "learning_rate": 0.00017037037037037037, | |
| "loss": 0.0236, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 52.558433354390395, | |
| "eval_accuracy": 0.7753739930955121, | |
| "eval_f1_macro": 0.5985595705467979, | |
| "eval_loss": 1.9067500829696655, | |
| "eval_runtime": 109.3981, | |
| "eval_samples_per_second": 79.435, | |
| "eval_steps_per_second": 2.486, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 52.81111813013266, | |
| "grad_norm": 1.502255916595459, | |
| "learning_rate": 0.00016851851851851853, | |
| "loss": 0.018, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 53.06380290587492, | |
| "grad_norm": 0.7225372791290283, | |
| "learning_rate": 0.00016666666666666666, | |
| "loss": 0.0238, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 53.31648768161718, | |
| "grad_norm": 1.2900406122207642, | |
| "learning_rate": 0.0001648148148148148, | |
| "loss": 0.0203, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 53.56917245735944, | |
| "grad_norm": 1.0755679607391357, | |
| "learning_rate": 0.00016296296296296298, | |
| "loss": 0.0243, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 53.56917245735944, | |
| "eval_accuracy": 0.7788262370540852, | |
| "eval_f1_macro": 0.6106899497765403, | |
| "eval_loss": 1.8558076620101929, | |
| "eval_runtime": 109.5516, | |
| "eval_samples_per_second": 79.323, | |
| "eval_steps_per_second": 2.483, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 53.8218572331017, | |
| "grad_norm": 0.06415720283985138, | |
| "learning_rate": 0.0001611111111111111, | |
| "loss": 0.0205, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 54.07454200884397, | |
| "grad_norm": 0.032455410808324814, | |
| "learning_rate": 0.00015925925925925927, | |
| "loss": 0.0221, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 54.32722678458623, | |
| "grad_norm": 0.04551871493458748, | |
| "learning_rate": 0.0001574074074074074, | |
| "loss": 0.0173, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 54.57991156032849, | |
| "grad_norm": 1.3965240716934204, | |
| "learning_rate": 0.00015555555555555556, | |
| "loss": 0.0168, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 54.57991156032849, | |
| "eval_accuracy": 0.7788262370540852, | |
| "eval_f1_macro": 0.5821705462809209, | |
| "eval_loss": 1.8485394716262817, | |
| "eval_runtime": 110.0893, | |
| "eval_samples_per_second": 78.936, | |
| "eval_steps_per_second": 2.471, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 54.83259633607075, | |
| "grad_norm": 0.8048573136329651, | |
| "learning_rate": 0.0001537037037037037, | |
| "loss": 0.0212, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 55.08528111181301, | |
| "grad_norm": 3.6472041606903076, | |
| "learning_rate": 0.00015185185185185185, | |
| "loss": 0.0229, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 55.33796588755528, | |
| "grad_norm": 1.6580862998962402, | |
| "learning_rate": 0.00015, | |
| "loss": 0.0185, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 55.59065066329754, | |
| "grad_norm": 0.0511002242565155, | |
| "learning_rate": 0.00014814814814814815, | |
| "loss": 0.019, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 55.59065066329754, | |
| "eval_accuracy": 0.7772151898734178, | |
| "eval_f1_macro": 0.6042884739793327, | |
| "eval_loss": 1.8060083389282227, | |
| "eval_runtime": 109.6887, | |
| "eval_samples_per_second": 79.224, | |
| "eval_steps_per_second": 2.48, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 55.8433354390398, | |
| "grad_norm": 0.7249517440795898, | |
| "learning_rate": 0.00014629629629629628, | |
| "loss": 0.022, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 56.09602021478206, | |
| "grad_norm": 0.0043795122765004635, | |
| "learning_rate": 0.00014444444444444444, | |
| "loss": 0.0156, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 56.34870499052432, | |
| "grad_norm": 1.0969189405441284, | |
| "learning_rate": 0.0001425925925925926, | |
| "loss": 0.0186, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 56.601389766266585, | |
| "grad_norm": 0.018035294488072395, | |
| "learning_rate": 0.00014074074074074076, | |
| "loss": 0.0193, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 56.601389766266585, | |
| "eval_accuracy": 0.7785960874568469, | |
| "eval_f1_macro": 0.6042077712818668, | |
| "eval_loss": 1.7848457098007202, | |
| "eval_runtime": 109.4669, | |
| "eval_samples_per_second": 79.385, | |
| "eval_steps_per_second": 2.485, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 56.854074542008846, | |
| "grad_norm": 1.3541306257247925, | |
| "learning_rate": 0.0001388888888888889, | |
| "loss": 0.014, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 57.106759317751106, | |
| "grad_norm": 0.32108399271965027, | |
| "learning_rate": 0.00013703703703703705, | |
| "loss": 0.0147, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 57.359444093493366, | |
| "grad_norm": 1.5764517784118652, | |
| "learning_rate": 0.00013518518518518518, | |
| "loss": 0.0121, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 57.612128869235626, | |
| "grad_norm": 0.08151420950889587, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 0.0167, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 57.612128869235626, | |
| "eval_accuracy": 0.7812428078250863, | |
| "eval_f1_macro": 0.5890612998790621, | |
| "eval_loss": 1.9067610502243042, | |
| "eval_runtime": 109.8252, | |
| "eval_samples_per_second": 79.126, | |
| "eval_steps_per_second": 2.477, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 57.86481364497789, | |
| "grad_norm": 0.2571534514427185, | |
| "learning_rate": 0.00013148148148148147, | |
| "loss": 0.0175, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 58.11749842072015, | |
| "grad_norm": 1.1256097555160522, | |
| "learning_rate": 0.00012962962962962963, | |
| "loss": 0.0185, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 58.37018319646241, | |
| "grad_norm": 0.036026712507009506, | |
| "learning_rate": 0.00012777777777777776, | |
| "loss": 0.0149, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 58.62286797220467, | |
| "grad_norm": 0.8259644508361816, | |
| "learning_rate": 0.00012592592592592592, | |
| "loss": 0.0103, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 58.62286797220467, | |
| "eval_accuracy": 0.7819332566168009, | |
| "eval_f1_macro": 0.5807141259596321, | |
| "eval_loss": 1.945181965827942, | |
| "eval_runtime": 109.5353, | |
| "eval_samples_per_second": 79.335, | |
| "eval_steps_per_second": 2.483, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 58.875552747946934, | |
| "grad_norm": 0.3167104423046112, | |
| "learning_rate": 0.00012407407407407408, | |
| "loss": 0.0111, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 59.1282375236892, | |
| "grad_norm": 0.19305191934108734, | |
| "learning_rate": 0.00012222222222222221, | |
| "loss": 0.0198, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 59.38092229943146, | |
| "grad_norm": 0.9000421762466431, | |
| "learning_rate": 0.00012037037037037036, | |
| "loss": 0.0175, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 59.63360707517372, | |
| "grad_norm": 0.16270488500595093, | |
| "learning_rate": 0.00011851851851851852, | |
| "loss": 0.0134, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 59.63360707517372, | |
| "eval_accuracy": 0.7856156501726121, | |
| "eval_f1_macro": 0.6008830980384848, | |
| "eval_loss": 1.8185406923294067, | |
| "eval_runtime": 110.5509, | |
| "eval_samples_per_second": 78.606, | |
| "eval_steps_per_second": 2.46, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 59.88629185091598, | |
| "grad_norm": 0.15302228927612305, | |
| "learning_rate": 0.00011666666666666667, | |
| "loss": 0.0114, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 60.13897662665824, | |
| "grad_norm": 0.07428427040576935, | |
| "learning_rate": 0.00011481481481481481, | |
| "loss": 0.0115, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 60.39166140240051, | |
| "grad_norm": 0.025211408734321594, | |
| "learning_rate": 0.00011296296296296296, | |
| "loss": 0.0129, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 60.64434617814277, | |
| "grad_norm": 0.12629875540733337, | |
| "learning_rate": 0.0001111111111111111, | |
| "loss": 0.0105, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 60.64434617814277, | |
| "eval_accuracy": 0.7771001150747986, | |
| "eval_f1_macro": 0.5844044138241861, | |
| "eval_loss": 1.9464218616485596, | |
| "eval_runtime": 109.5176, | |
| "eval_samples_per_second": 79.348, | |
| "eval_steps_per_second": 2.484, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 60.89703095388503, | |
| "grad_norm": 0.08720462769269943, | |
| "learning_rate": 0.00010925925925925926, | |
| "loss": 0.0104, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 61.14971572962729, | |
| "grad_norm": 2.3958306312561035, | |
| "learning_rate": 0.00010740740740740741, | |
| "loss": 0.011, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 61.40240050536955, | |
| "grad_norm": 0.0953073650598526, | |
| "learning_rate": 0.00010555555555555555, | |
| "loss": 0.0115, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 61.655085281111816, | |
| "grad_norm": 0.013684218749403954, | |
| "learning_rate": 0.0001037037037037037, | |
| "loss": 0.0154, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 61.655085281111816, | |
| "eval_accuracy": 0.7811277330264672, | |
| "eval_f1_macro": 0.5880759609293783, | |
| "eval_loss": 1.9170596599578857, | |
| "eval_runtime": 109.5265, | |
| "eval_samples_per_second": 79.342, | |
| "eval_steps_per_second": 2.483, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 61.907770056854076, | |
| "grad_norm": 0.02879270352423191, | |
| "learning_rate": 0.00010185185185185185, | |
| "loss": 0.0133, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 62.160454832596336, | |
| "grad_norm": 2.672358512878418, | |
| "learning_rate": 0.0001, | |
| "loss": 0.014, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 62.4131396083386, | |
| "grad_norm": 0.22067667543888092, | |
| "learning_rate": 9.814814814814815e-05, | |
| "loss": 0.0152, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 62.66582438408086, | |
| "grad_norm": 1.4633599519729614, | |
| "learning_rate": 9.62962962962963e-05, | |
| "loss": 0.0132, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 62.66582438408086, | |
| "eval_accuracy": 0.7836593785960875, | |
| "eval_f1_macro": 0.5992199483923372, | |
| "eval_loss": 1.8896235227584839, | |
| "eval_runtime": 109.6805, | |
| "eval_samples_per_second": 79.23, | |
| "eval_steps_per_second": 2.48, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 62.918509159823124, | |
| "grad_norm": 1.5682035684585571, | |
| "learning_rate": 9.444444444444444e-05, | |
| "loss": 0.0103, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 63.171193935565384, | |
| "grad_norm": 0.010575806722044945, | |
| "learning_rate": 9.259259259259259e-05, | |
| "loss": 0.0099, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 63.423878711307644, | |
| "grad_norm": 1.3311010599136353, | |
| "learning_rate": 9.074074074074075e-05, | |
| "loss": 0.0082, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 63.676563487049904, | |
| "grad_norm": 0.13537296652793884, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.0059, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 63.676563487049904, | |
| "eval_accuracy": 0.785385500575374, | |
| "eval_f1_macro": 0.5910071350304901, | |
| "eval_loss": 1.93367600440979, | |
| "eval_runtime": 109.8909, | |
| "eval_samples_per_second": 79.078, | |
| "eval_steps_per_second": 2.475, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 63.929248262792164, | |
| "grad_norm": 0.13171835243701935, | |
| "learning_rate": 8.703703703703704e-05, | |
| "loss": 0.0099, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 64.18193303853442, | |
| "grad_norm": 0.03450850397348404, | |
| "learning_rate": 8.518518518518518e-05, | |
| "loss": 0.0125, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 64.43461781427669, | |
| "grad_norm": 2.559318780899048, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 0.0134, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 64.68730259001894, | |
| "grad_norm": 0.4207484722137451, | |
| "learning_rate": 8.148148148148149e-05, | |
| "loss": 0.0082, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 64.68730259001894, | |
| "eval_accuracy": 0.7844649021864212, | |
| "eval_f1_macro": 0.5981216790852985, | |
| "eval_loss": 1.9173579216003418, | |
| "eval_runtime": 109.5317, | |
| "eval_samples_per_second": 79.338, | |
| "eval_steps_per_second": 2.483, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 64.93998736576121, | |
| "grad_norm": 0.02727402001619339, | |
| "learning_rate": 7.962962962962964e-05, | |
| "loss": 0.0079, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 65.19267214150348, | |
| "grad_norm": 0.055505283176898956, | |
| "learning_rate": 7.777777777777778e-05, | |
| "loss": 0.0088, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 65.44535691724573, | |
| "grad_norm": 0.0938887745141983, | |
| "learning_rate": 7.592592592592593e-05, | |
| "loss": 0.0081, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 65.698041692988, | |
| "grad_norm": 0.06274864077568054, | |
| "learning_rate": 7.407407407407407e-05, | |
| "loss": 0.01, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 65.698041692988, | |
| "eval_accuracy": 0.7873417721518987, | |
| "eval_f1_macro": 0.5952116194605911, | |
| "eval_loss": 1.9076104164123535, | |
| "eval_runtime": 109.6326, | |
| "eval_samples_per_second": 79.265, | |
| "eval_steps_per_second": 2.481, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 65.95072646873025, | |
| "grad_norm": 0.395120769739151, | |
| "learning_rate": 7.222222222222222e-05, | |
| "loss": 0.013, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 66.20341124447252, | |
| "grad_norm": 0.0025830611120909452, | |
| "learning_rate": 7.037037037037038e-05, | |
| "loss": 0.0072, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 66.45609602021479, | |
| "grad_norm": 0.08498067408800125, | |
| "learning_rate": 6.851851851851852e-05, | |
| "loss": 0.0092, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 66.70878079595704, | |
| "grad_norm": 0.03530678525567055, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.0119, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 66.70878079595704, | |
| "eval_accuracy": 0.7857307249712313, | |
| "eval_f1_macro": 0.6071046848722826, | |
| "eval_loss": 1.8940664529800415, | |
| "eval_runtime": 109.9317, | |
| "eval_samples_per_second": 79.049, | |
| "eval_steps_per_second": 2.474, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 66.9614655716993, | |
| "grad_norm": 0.031329553574323654, | |
| "learning_rate": 6.481481481481482e-05, | |
| "loss": 0.0096, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 67.21415034744156, | |
| "grad_norm": 0.30799156427383423, | |
| "learning_rate": 6.296296296296296e-05, | |
| "loss": 0.0095, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 67.46683512318383, | |
| "grad_norm": 1.5470929145812988, | |
| "learning_rate": 6.111111111111111e-05, | |
| "loss": 0.0089, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 67.7195198989261, | |
| "grad_norm": 2.5448250770568848, | |
| "learning_rate": 5.925925925925926e-05, | |
| "loss": 0.0067, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 67.7195198989261, | |
| "eval_accuracy": 0.7878020713463751, | |
| "eval_f1_macro": 0.5999522395147027, | |
| "eval_loss": 1.930634617805481, | |
| "eval_runtime": 109.663, | |
| "eval_samples_per_second": 79.243, | |
| "eval_steps_per_second": 2.48, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 67.97220467466835, | |
| "grad_norm": 0.7930640578269958, | |
| "learning_rate": 5.7407407407407406e-05, | |
| "loss": 0.0069, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 68.22488945041061, | |
| "grad_norm": 0.26925787329673767, | |
| "learning_rate": 5.555555555555555e-05, | |
| "loss": 0.007, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 68.47757422615287, | |
| "grad_norm": 0.1341189444065094, | |
| "learning_rate": 5.3703703703703704e-05, | |
| "loss": 0.0087, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 68.73025900189513, | |
| "grad_norm": 0.02547445334494114, | |
| "learning_rate": 5.185185185185185e-05, | |
| "loss": 0.0081, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 68.73025900189513, | |
| "eval_accuracy": 0.792174913693901, | |
| "eval_f1_macro": 0.6008077464320003, | |
| "eval_loss": 1.9338207244873047, | |
| "eval_runtime": 109.813, | |
| "eval_samples_per_second": 79.135, | |
| "eval_steps_per_second": 2.477, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 68.9829437776374, | |
| "grad_norm": 0.03730254992842674, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0065, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 69.23562855337966, | |
| "grad_norm": 0.11829439550638199, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.0058, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 69.48831332912192, | |
| "grad_norm": 1.238792896270752, | |
| "learning_rate": 4.6296296296296294e-05, | |
| "loss": 0.0047, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 69.74099810486418, | |
| "grad_norm": 0.08776523917913437, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.0069, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 69.74099810486418, | |
| "eval_accuracy": 0.7914844649021864, | |
| "eval_f1_macro": 0.5974683158925355, | |
| "eval_loss": 1.9561039209365845, | |
| "eval_runtime": 109.6013, | |
| "eval_samples_per_second": 79.287, | |
| "eval_steps_per_second": 2.482, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 69.99368288060644, | |
| "grad_norm": 0.02279621548950672, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.0077, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 70.24636765634871, | |
| "grad_norm": 1.407116413116455, | |
| "learning_rate": 4.0740740740740745e-05, | |
| "loss": 0.0083, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 70.49905243209096, | |
| "grad_norm": 0.8993849754333496, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.0037, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 70.75173720783323, | |
| "grad_norm": 0.01359980646520853, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.0037, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 70.75173720783323, | |
| "eval_accuracy": 0.7919447640966628, | |
| "eval_f1_macro": 0.5933817664756982, | |
| "eval_loss": 1.9411331415176392, | |
| "eval_runtime": 109.976, | |
| "eval_samples_per_second": 79.017, | |
| "eval_steps_per_second": 2.473, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 71.00442198357548, | |
| "grad_norm": 0.00356766814365983, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.0053, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 71.25710675931775, | |
| "grad_norm": 0.0005093297804705799, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0046, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 71.50979153506002, | |
| "grad_norm": 0.007542046718299389, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.0063, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 71.76247631080227, | |
| "grad_norm": 0.008996374905109406, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.0055, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 71.76247631080227, | |
| "eval_accuracy": 0.7947065592635213, | |
| "eval_f1_macro": 0.606388230425645, | |
| "eval_loss": 1.9448614120483398, | |
| "eval_runtime": 109.7807, | |
| "eval_samples_per_second": 79.158, | |
| "eval_steps_per_second": 2.478, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 72.01516108654454, | |
| "grad_norm": 0.27878421545028687, | |
| "learning_rate": 2.7777777777777776e-05, | |
| "loss": 0.0046, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 72.26784586228679, | |
| "grad_norm": 0.011219154112040997, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.0067, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 72.52053063802906, | |
| "grad_norm": 0.0006956435390748084, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.0063, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 72.77321541377133, | |
| "grad_norm": 0.07210740447044373, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0046, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 72.77321541377133, | |
| "eval_accuracy": 0.7941311852704258, | |
| "eval_f1_macro": 0.6069777778254928, | |
| "eval_loss": 1.9499226808547974, | |
| "eval_runtime": 109.6721, | |
| "eval_samples_per_second": 79.236, | |
| "eval_steps_per_second": 2.48, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 73.02590018951358, | |
| "grad_norm": 0.04884390905499458, | |
| "learning_rate": 2.0370370370370372e-05, | |
| "loss": 0.006, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 73.27858496525585, | |
| "grad_norm": 0.06649865210056305, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.0028, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 73.5312697409981, | |
| "grad_norm": 0.0004434076545294374, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.009, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 73.78395451674037, | |
| "grad_norm": 0.15433602035045624, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.0063, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 73.78395451674037, | |
| "eval_accuracy": 0.7918296892980438, | |
| "eval_f1_macro": 0.5927699398802871, | |
| "eval_loss": 1.9630827903747559, | |
| "eval_runtime": 109.0221, | |
| "eval_samples_per_second": 79.709, | |
| "eval_steps_per_second": 2.495, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 74.03663929248263, | |
| "grad_norm": 0.0017094516661018133, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.0046, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 74.28932406822489, | |
| "grad_norm": 1.173220157623291, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0037, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 74.54200884396715, | |
| "grad_norm": 0.006894304417073727, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.0042, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 74.7946936197094, | |
| "grad_norm": 0.005111688282340765, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.0052, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 74.7946936197094, | |
| "eval_accuracy": 0.7937859608745684, | |
| "eval_f1_macro": 0.5940671707873302, | |
| "eval_loss": 1.951813817024231, | |
| "eval_runtime": 108.972, | |
| "eval_samples_per_second": 79.745, | |
| "eval_steps_per_second": 2.496, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 75.04737839545167, | |
| "grad_norm": 0.0013843182241544127, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0057, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 75.30006317119394, | |
| "grad_norm": 9.947911894414574e-05, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0052, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 75.5527479469362, | |
| "grad_norm": 0.3020813763141632, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.006, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 75.80543272267846, | |
| "grad_norm": 1.3610678911209106, | |
| "learning_rate": 0.0, | |
| "loss": 0.0058, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 75.80543272267846, | |
| "eval_accuracy": 0.7934407364787112, | |
| "eval_f1_macro": 0.6028809095714917, | |
| "eval_loss": 1.9492535591125488, | |
| "eval_runtime": 108.7371, | |
| "eval_samples_per_second": 79.918, | |
| "eval_steps_per_second": 2.501, | |
| "step": 30000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 30000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 76, | |
| "save_steps": 30000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.8164789316384843e+20, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |