| { | |
| "best_metric": 0.7126582278481013, | |
| "best_model_checkpoint": "results/facebook/wav2vec2-base/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": 1.0702794790267944, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 4.416, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.505369551484523, | |
| "grad_norm": 1.4368451833724976, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 4.0736, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7580543272267846, | |
| "grad_norm": 2.1718220710754395, | |
| "learning_rate": 5e-05, | |
| "loss": 3.8685, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.010739102969046, | |
| "grad_norm": 3.2043306827545166, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 3.7358, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.010739102969046, | |
| "eval_accuracy": 0.10667433831990794, | |
| "eval_f1_macro": 0.010564416636306704, | |
| "eval_loss": 3.6359970569610596, | |
| "eval_runtime": 79.6332, | |
| "eval_samples_per_second": 109.125, | |
| "eval_steps_per_second": 3.416, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2634238787113077, | |
| "grad_norm": 7.089096546173096, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 3.5513, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5161086544535691, | |
| "grad_norm": 15.851802825927734, | |
| "learning_rate": 0.0001, | |
| "loss": 3.2697, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.7687934301958306, | |
| "grad_norm": 7.881278991699219, | |
| "learning_rate": 0.00011666666666666667, | |
| "loss": 2.9246, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.021478205938092, | |
| "grad_norm": 9.194243431091309, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 2.5677, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.021478205938092, | |
| "eval_accuracy": 0.4115074798619102, | |
| "eval_f1_macro": 0.17958228612174398, | |
| "eval_loss": 2.4761385917663574, | |
| "eval_runtime": 79.6601, | |
| "eval_samples_per_second": 109.088, | |
| "eval_steps_per_second": 3.415, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.274162981680354, | |
| "grad_norm": 11.996749877929688, | |
| "learning_rate": 0.00015, | |
| "loss": 2.2068, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.5268477574226154, | |
| "grad_norm": 15.755786895751953, | |
| "learning_rate": 0.00016666666666666666, | |
| "loss": 2.0173, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.779532533164877, | |
| "grad_norm": 15.083039283752441, | |
| "learning_rate": 0.00018333333333333334, | |
| "loss": 1.8694, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.0322173089071383, | |
| "grad_norm": 10.2857666015625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7515, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.0322173089071383, | |
| "eval_accuracy": 0.567433831990794, | |
| "eval_f1_macro": 0.36659150895414866, | |
| "eval_loss": 1.8158797025680542, | |
| "eval_runtime": 78.9641, | |
| "eval_samples_per_second": 110.05, | |
| "eval_steps_per_second": 3.445, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.2849020846493997, | |
| "grad_norm": 11.110665321350098, | |
| "learning_rate": 0.00021666666666666668, | |
| "loss": 1.5794, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.537586860391661, | |
| "grad_norm": 11.364448547363281, | |
| "learning_rate": 0.00023333333333333333, | |
| "loss": 1.5002, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.790271636133923, | |
| "grad_norm": 7.655250072479248, | |
| "learning_rate": 0.00025, | |
| "loss": 1.491, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.042956411876184, | |
| "grad_norm": 9.095359802246094, | |
| "learning_rate": 0.0002666666666666667, | |
| "loss": 1.4511, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.042956411876184, | |
| "eval_accuracy": 0.5951668584579977, | |
| "eval_f1_macro": 0.4026927353503452, | |
| "eval_loss": 1.6677496433258057, | |
| "eval_runtime": 79.0574, | |
| "eval_samples_per_second": 109.92, | |
| "eval_steps_per_second": 3.441, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.295641187618446, | |
| "grad_norm": 11.648326873779297, | |
| "learning_rate": 0.00028333333333333335, | |
| "loss": 1.3253, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.548325963360708, | |
| "grad_norm": 8.05283260345459, | |
| "learning_rate": 0.0003, | |
| "loss": 1.3594, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.801010739102969, | |
| "grad_norm": 8.383731842041016, | |
| "learning_rate": 0.00031666666666666665, | |
| "loss": 1.381, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 5.053695514845231, | |
| "grad_norm": 8.625487327575684, | |
| "learning_rate": 0.0003333333333333333, | |
| "loss": 1.3643, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.053695514845231, | |
| "eval_accuracy": 0.6102416570771001, | |
| "eval_f1_macro": 0.4299186371835549, | |
| "eval_loss": 1.6622495651245117, | |
| "eval_runtime": 79.0621, | |
| "eval_samples_per_second": 109.914, | |
| "eval_steps_per_second": 3.44, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.306380290587492, | |
| "grad_norm": 8.265228271484375, | |
| "learning_rate": 0.00035, | |
| "loss": 1.3271, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 5.559065066329754, | |
| "grad_norm": 15.364546775817871, | |
| "learning_rate": 0.00036666666666666667, | |
| "loss": 1.352, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 5.8117498420720155, | |
| "grad_norm": 8.970512390136719, | |
| "learning_rate": 0.00038333333333333334, | |
| "loss": 1.347, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 6.0644346178142765, | |
| "grad_norm": 11.065812110900879, | |
| "learning_rate": 0.0004, | |
| "loss": 1.3603, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.0644346178142765, | |
| "eval_accuracy": 0.5935558112773303, | |
| "eval_f1_macro": 0.43463154274064014, | |
| "eval_loss": 1.7598559856414795, | |
| "eval_runtime": 79.0019, | |
| "eval_samples_per_second": 109.997, | |
| "eval_steps_per_second": 3.443, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.317119393556538, | |
| "grad_norm": 7.496682643890381, | |
| "learning_rate": 0.0004166666666666667, | |
| "loss": 1.307, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.569804169298799, | |
| "grad_norm": 10.478510856628418, | |
| "learning_rate": 0.00043333333333333337, | |
| "loss": 1.3713, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 6.822488945041061, | |
| "grad_norm": 9.119362831115723, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 1.4236, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 7.075173720783323, | |
| "grad_norm": 6.770716667175293, | |
| "learning_rate": 0.00046666666666666666, | |
| "loss": 1.353, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.075173720783323, | |
| "eval_accuracy": 0.5903337169159953, | |
| "eval_f1_macro": 0.44580064776313444, | |
| "eval_loss": 1.7876626253128052, | |
| "eval_runtime": 79.2177, | |
| "eval_samples_per_second": 109.698, | |
| "eval_steps_per_second": 3.434, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.327858496525584, | |
| "grad_norm": 11.353608131408691, | |
| "learning_rate": 0.00048333333333333334, | |
| "loss": 1.3166, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 7.580543272267846, | |
| "grad_norm": 8.012442588806152, | |
| "learning_rate": 0.0005, | |
| "loss": 1.4728, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.833228048010107, | |
| "grad_norm": 12.510762214660645, | |
| "learning_rate": 0.0004981481481481482, | |
| "loss": 1.3979, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 8.085912823752368, | |
| "grad_norm": 7.419615268707275, | |
| "learning_rate": 0.0004962962962962963, | |
| "loss": 1.3836, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.085912823752368, | |
| "eval_accuracy": 0.6051783659378596, | |
| "eval_f1_macro": 0.46417323352244966, | |
| "eval_loss": 1.7455965280532837, | |
| "eval_runtime": 79.2025, | |
| "eval_samples_per_second": 109.719, | |
| "eval_steps_per_second": 3.434, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.33859759949463, | |
| "grad_norm": 7.640713214874268, | |
| "learning_rate": 0.0004944444444444445, | |
| "loss": 1.2883, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 8.591282375236892, | |
| "grad_norm": 8.823062896728516, | |
| "learning_rate": 0.0004925925925925925, | |
| "loss": 1.3007, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 8.843967150979154, | |
| "grad_norm": 7.705187797546387, | |
| "learning_rate": 0.0004907407407407408, | |
| "loss": 1.311, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.096651926721416, | |
| "grad_norm": 5.811334133148193, | |
| "learning_rate": 0.0004888888888888889, | |
| "loss": 1.2593, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.096651926721416, | |
| "eval_accuracy": 0.5981588032220944, | |
| "eval_f1_macro": 0.444749614864958, | |
| "eval_loss": 1.800374150276184, | |
| "eval_runtime": 74.8953, | |
| "eval_samples_per_second": 116.029, | |
| "eval_steps_per_second": 3.632, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.349336702463676, | |
| "grad_norm": 7.506765365600586, | |
| "learning_rate": 0.00048703703703703707, | |
| "loss": 1.2283, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 9.602021478205938, | |
| "grad_norm": 7.739789962768555, | |
| "learning_rate": 0.0004851851851851852, | |
| "loss": 1.2141, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 9.8547062539482, | |
| "grad_norm": 8.539153099060059, | |
| "learning_rate": 0.00048333333333333334, | |
| "loss": 1.2104, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 10.107391029690461, | |
| "grad_norm": 11.42418098449707, | |
| "learning_rate": 0.00048148148148148144, | |
| "loss": 1.1221, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.107391029690461, | |
| "eval_accuracy": 0.6199079401611047, | |
| "eval_f1_macro": 0.4625532357188315, | |
| "eval_loss": 1.7226791381835938, | |
| "eval_runtime": 79.2693, | |
| "eval_samples_per_second": 109.626, | |
| "eval_steps_per_second": 3.431, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.360075805432723, | |
| "grad_norm": 9.52509593963623, | |
| "learning_rate": 0.00047962962962962965, | |
| "loss": 1.1435, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 10.612760581174983, | |
| "grad_norm": 10.423213005065918, | |
| "learning_rate": 0.0004777777777777778, | |
| "loss": 1.1839, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 10.865445356917245, | |
| "grad_norm": 9.375807762145996, | |
| "learning_rate": 0.0004759259259259259, | |
| "loss": 1.237, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 11.118130132659507, | |
| "grad_norm": 11.8602294921875, | |
| "learning_rate": 0.0004740740740740741, | |
| "loss": 1.1433, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 11.118130132659507, | |
| "eval_accuracy": 0.607940161104718, | |
| "eval_f1_macro": 0.4688655321201303, | |
| "eval_loss": 1.7294870615005493, | |
| "eval_runtime": 80.4318, | |
| "eval_samples_per_second": 108.042, | |
| "eval_steps_per_second": 3.382, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 11.37081490840177, | |
| "grad_norm": 8.036584854125977, | |
| "learning_rate": 0.00047222222222222224, | |
| "loss": 1.1633, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.623499684144031, | |
| "grad_norm": 9.331839561462402, | |
| "learning_rate": 0.0004703703703703704, | |
| "loss": 1.1189, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 11.876184459886291, | |
| "grad_norm": 11.122476577758789, | |
| "learning_rate": 0.0004685185185185185, | |
| "loss": 1.2844, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 12.128869235628553, | |
| "grad_norm": 47.187034606933594, | |
| "learning_rate": 0.00046666666666666666, | |
| "loss": 1.413, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 12.128869235628553, | |
| "eval_accuracy": 0.5714614499424626, | |
| "eval_f1_macro": 0.4344627522389952, | |
| "eval_loss": 1.9414366483688354, | |
| "eval_runtime": 80.4563, | |
| "eval_samples_per_second": 108.009, | |
| "eval_steps_per_second": 3.381, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 12.381554011370815, | |
| "grad_norm": 10.95539379119873, | |
| "learning_rate": 0.0004648148148148148, | |
| "loss": 1.2141, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 12.634238787113077, | |
| "grad_norm": 9.24135684967041, | |
| "learning_rate": 0.000462962962962963, | |
| "loss": 1.1796, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 12.886923562855339, | |
| "grad_norm": 6.320292949676514, | |
| "learning_rate": 0.00046111111111111114, | |
| "loss": 1.0804, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 13.139608338597599, | |
| "grad_norm": 9.179173469543457, | |
| "learning_rate": 0.00045925925925925925, | |
| "loss": 1.0691, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 13.139608338597599, | |
| "eval_accuracy": 0.5985040276179516, | |
| "eval_f1_macro": 0.45639628530630205, | |
| "eval_loss": 1.7746257781982422, | |
| "eval_runtime": 81.5973, | |
| "eval_samples_per_second": 106.499, | |
| "eval_steps_per_second": 3.333, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 13.39229311433986, | |
| "grad_norm": 9.278477668762207, | |
| "learning_rate": 0.00045740740740740746, | |
| "loss": 1.0505, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 13.644977890082123, | |
| "grad_norm": 9.767425537109375, | |
| "learning_rate": 0.00045555555555555556, | |
| "loss": 1.0226, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 13.897662665824384, | |
| "grad_norm": 9.29463005065918, | |
| "learning_rate": 0.0004537037037037037, | |
| "loss": 1.0615, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.150347441566646, | |
| "grad_norm": 8.730624198913574, | |
| "learning_rate": 0.00045185185185185183, | |
| "loss": 0.9554, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 14.150347441566646, | |
| "eval_accuracy": 0.6289988492520138, | |
| "eval_f1_macro": 0.46507624040175854, | |
| "eval_loss": 1.723509669303894, | |
| "eval_runtime": 80.3276, | |
| "eval_samples_per_second": 108.182, | |
| "eval_steps_per_second": 3.386, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 14.403032217308906, | |
| "grad_norm": 5.3180718421936035, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 0.8988, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 14.655716993051168, | |
| "grad_norm": 7.094490051269531, | |
| "learning_rate": 0.00044814814814814815, | |
| "loss": 0.9003, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 14.90840176879343, | |
| "grad_norm": 9.134806632995605, | |
| "learning_rate": 0.0004462962962962963, | |
| "loss": 0.9316, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 15.161086544535692, | |
| "grad_norm": 6.943215847015381, | |
| "learning_rate": 0.0004444444444444444, | |
| "loss": 0.8504, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.161086544535692, | |
| "eval_accuracy": 0.6323360184119677, | |
| "eval_f1_macro": 0.4826347855547535, | |
| "eval_loss": 1.763763666152954, | |
| "eval_runtime": 81.3634, | |
| "eval_samples_per_second": 106.805, | |
| "eval_steps_per_second": 3.343, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.413771320277954, | |
| "grad_norm": 11.100159645080566, | |
| "learning_rate": 0.0004425925925925926, | |
| "loss": 0.8468, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 15.666456096020214, | |
| "grad_norm": 7.49467134475708, | |
| "learning_rate": 0.0004407407407407408, | |
| "loss": 0.8795, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 15.919140871762476, | |
| "grad_norm": 6.621813774108887, | |
| "learning_rate": 0.0004388888888888889, | |
| "loss": 0.8668, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 16.171825647504736, | |
| "grad_norm": 6.006112575531006, | |
| "learning_rate": 0.00043703703703703705, | |
| "loss": 0.803, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 16.171825647504736, | |
| "eval_accuracy": 0.625891829689298, | |
| "eval_f1_macro": 0.4487951994480194, | |
| "eval_loss": 1.7311913967132568, | |
| "eval_runtime": 37.9312, | |
| "eval_samples_per_second": 229.099, | |
| "eval_steps_per_second": 7.171, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 16.424510423247, | |
| "grad_norm": 6.613837718963623, | |
| "learning_rate": 0.0004351851851851852, | |
| "loss": 0.7517, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 16.67719519898926, | |
| "grad_norm": 9.915642738342285, | |
| "learning_rate": 0.00043333333333333337, | |
| "loss": 0.796, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 16.929879974731524, | |
| "grad_norm": 6.998866081237793, | |
| "learning_rate": 0.00043148148148148147, | |
| "loss": 0.7857, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 17.182564750473784, | |
| "grad_norm": 6.572287559509277, | |
| "learning_rate": 0.00042962962962962963, | |
| "loss": 0.7277, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 17.182564750473784, | |
| "eval_accuracy": 0.643843498273878, | |
| "eval_f1_macro": 0.48647759132797, | |
| "eval_loss": 1.7235004901885986, | |
| "eval_runtime": 37.9801, | |
| "eval_samples_per_second": 228.804, | |
| "eval_steps_per_second": 7.162, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 17.435249526216044, | |
| "grad_norm": 6.055329322814941, | |
| "learning_rate": 0.0004277777777777778, | |
| "loss": 0.7454, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 17.687934301958308, | |
| "grad_norm": 5.803674697875977, | |
| "learning_rate": 0.00042592592592592595, | |
| "loss": 0.7308, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 17.940619077700568, | |
| "grad_norm": 5.9112701416015625, | |
| "learning_rate": 0.00042407407407407406, | |
| "loss": 0.7351, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 18.19330385344283, | |
| "grad_norm": 7.775882720947266, | |
| "learning_rate": 0.0004222222222222222, | |
| "loss": 0.6472, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 18.19330385344283, | |
| "eval_accuracy": 0.6408515535097814, | |
| "eval_f1_macro": 0.48463566719085455, | |
| "eval_loss": 1.8127564191818237, | |
| "eval_runtime": 80.5215, | |
| "eval_samples_per_second": 107.921, | |
| "eval_steps_per_second": 3.378, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 18.44598862918509, | |
| "grad_norm": 6.309839248657227, | |
| "learning_rate": 0.00042037037037037043, | |
| "loss": 0.6687, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 18.69867340492735, | |
| "grad_norm": 7.941220760345459, | |
| "learning_rate": 0.00041851851851851853, | |
| "loss": 0.6467, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 18.951358180669615, | |
| "grad_norm": 6.115121364593506, | |
| "learning_rate": 0.0004166666666666667, | |
| "loss": 0.6331, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.204042956411875, | |
| "grad_norm": 6.009261131286621, | |
| "learning_rate": 0.0004148148148148148, | |
| "loss": 0.5843, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 19.204042956411875, | |
| "eval_accuracy": 0.6463751438434983, | |
| "eval_f1_macro": 0.4784169737107719, | |
| "eval_loss": 1.7239370346069336, | |
| "eval_runtime": 80.0088, | |
| "eval_samples_per_second": 108.613, | |
| "eval_steps_per_second": 3.4, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 19.45672773215414, | |
| "grad_norm": 7.258364200592041, | |
| "learning_rate": 0.00041296296296296296, | |
| "loss": 0.5705, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 19.7094125078964, | |
| "grad_norm": 4.382850170135498, | |
| "learning_rate": 0.0004111111111111111, | |
| "loss": 0.5821, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 19.96209728363866, | |
| "grad_norm": 5.137075424194336, | |
| "learning_rate": 0.0004092592592592593, | |
| "loss": 0.6053, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 20.214782059380923, | |
| "grad_norm": 7.7077412605285645, | |
| "learning_rate": 0.0004074074074074074, | |
| "loss": 0.54, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 20.214782059380923, | |
| "eval_accuracy": 0.643843498273878, | |
| "eval_f1_macro": 0.4918783795756223, | |
| "eval_loss": 1.8257795572280884, | |
| "eval_runtime": 79.2622, | |
| "eval_samples_per_second": 109.636, | |
| "eval_steps_per_second": 3.432, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 20.467466835123183, | |
| "grad_norm": 6.132132530212402, | |
| "learning_rate": 0.00040555555555555554, | |
| "loss": 0.5463, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 20.720151610865447, | |
| "grad_norm": 7.697502136230469, | |
| "learning_rate": 0.00040370370370370375, | |
| "loss": 0.5541, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 20.972836386607707, | |
| "grad_norm": 7.310309410095215, | |
| "learning_rate": 0.00040185185185185186, | |
| "loss": 0.5605, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 21.225521162349967, | |
| "grad_norm": 6.562226295471191, | |
| "learning_rate": 0.0004, | |
| "loss": 0.5041, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 21.225521162349967, | |
| "eval_accuracy": 0.6448791714614499, | |
| "eval_f1_macro": 0.4649201259680266, | |
| "eval_loss": 1.8011974096298218, | |
| "eval_runtime": 80.2121, | |
| "eval_samples_per_second": 108.338, | |
| "eval_steps_per_second": 3.391, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 21.47820593809223, | |
| "grad_norm": 9.611831665039062, | |
| "learning_rate": 0.0003981481481481481, | |
| "loss": 0.4949, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 21.73089071383449, | |
| "grad_norm": 7.182919025421143, | |
| "learning_rate": 0.00039629629629629634, | |
| "loss": 0.5248, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 21.983575489576754, | |
| "grad_norm": 6.601507186889648, | |
| "learning_rate": 0.00039444444444444444, | |
| "loss": 0.5251, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 22.236260265319014, | |
| "grad_norm": 6.742110729217529, | |
| "learning_rate": 0.0003925925925925926, | |
| "loss": 0.4582, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 22.236260265319014, | |
| "eval_accuracy": 0.6514384349827388, | |
| "eval_f1_macro": 0.49038024738617103, | |
| "eval_loss": 1.8499749898910522, | |
| "eval_runtime": 80.3812, | |
| "eval_samples_per_second": 108.11, | |
| "eval_steps_per_second": 3.384, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 22.488945041061275, | |
| "grad_norm": 6.984791278839111, | |
| "learning_rate": 0.0003907407407407407, | |
| "loss": 0.4905, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 22.74162981680354, | |
| "grad_norm": 10.236854553222656, | |
| "learning_rate": 0.0003888888888888889, | |
| "loss": 0.4511, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 22.9943145925458, | |
| "grad_norm": 6.726616859436035, | |
| "learning_rate": 0.000387037037037037, | |
| "loss": 0.4669, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 23.246999368288062, | |
| "grad_norm": 6.4510087966918945, | |
| "learning_rate": 0.0003851851851851852, | |
| "loss": 0.4323, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 23.246999368288062, | |
| "eval_accuracy": 0.6431530494821635, | |
| "eval_f1_macro": 0.47550601524271274, | |
| "eval_loss": 1.8550491333007812, | |
| "eval_runtime": 80.5385, | |
| "eval_samples_per_second": 107.899, | |
| "eval_steps_per_second": 3.377, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 23.499684144030322, | |
| "grad_norm": 4.697257995605469, | |
| "learning_rate": 0.00038333333333333334, | |
| "loss": 0.4415, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 23.752368919772582, | |
| "grad_norm": 5.567733287811279, | |
| "learning_rate": 0.0003814814814814815, | |
| "loss": 0.4638, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 24.005053695514846, | |
| "grad_norm": 5.837888240814209, | |
| "learning_rate": 0.00037962962962962966, | |
| "loss": 0.4391, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 24.257738471257106, | |
| "grad_norm": 5.660379886627197, | |
| "learning_rate": 0.00037777777777777777, | |
| "loss": 0.42, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 24.257738471257106, | |
| "eval_accuracy": 0.6355581127733027, | |
| "eval_f1_macro": 0.47185979811513606, | |
| "eval_loss": 2.105299949645996, | |
| "eval_runtime": 38.0724, | |
| "eval_samples_per_second": 228.249, | |
| "eval_steps_per_second": 7.144, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 24.51042324699937, | |
| "grad_norm": 8.816080093383789, | |
| "learning_rate": 0.00037592592592592593, | |
| "loss": 0.3997, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 24.76310802274163, | |
| "grad_norm": 6.397033214569092, | |
| "learning_rate": 0.0003740740740740741, | |
| "loss": 0.4096, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 25.01579279848389, | |
| "grad_norm": 3.873905897140503, | |
| "learning_rate": 0.00037222222222222225, | |
| "loss": 0.419, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 25.268477574226154, | |
| "grad_norm": 4.635372161865234, | |
| "learning_rate": 0.00037037037037037035, | |
| "loss": 0.3916, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 25.268477574226154, | |
| "eval_accuracy": 0.649367088607595, | |
| "eval_f1_macro": 0.46888469395951493, | |
| "eval_loss": 1.9272565841674805, | |
| "eval_runtime": 37.8797, | |
| "eval_samples_per_second": 229.41, | |
| "eval_steps_per_second": 7.181, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 25.521162349968414, | |
| "grad_norm": 10.858739852905273, | |
| "learning_rate": 0.0003685185185185185, | |
| "loss": 0.3817, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 25.773847125710677, | |
| "grad_norm": 6.400647163391113, | |
| "learning_rate": 0.00036666666666666667, | |
| "loss": 0.4088, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 26.026531901452937, | |
| "grad_norm": 6.833063125610352, | |
| "learning_rate": 0.00036481481481481483, | |
| "loss": 0.3963, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 26.279216677195198, | |
| "grad_norm": 4.5142340660095215, | |
| "learning_rate": 0.000362962962962963, | |
| "loss": 0.3393, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 26.279216677195198, | |
| "eval_accuracy": 0.6559263521288837, | |
| "eval_f1_macro": 0.4863981237740537, | |
| "eval_loss": 1.950628399848938, | |
| "eval_runtime": 37.9287, | |
| "eval_samples_per_second": 229.114, | |
| "eval_steps_per_second": 7.171, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 26.53190145293746, | |
| "grad_norm": 6.590590953826904, | |
| "learning_rate": 0.0003611111111111111, | |
| "loss": 0.3463, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 26.78458622867972, | |
| "grad_norm": 6.042336463928223, | |
| "learning_rate": 0.0003592592592592593, | |
| "loss": 0.3779, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 27.037271004421985, | |
| "grad_norm": 5.936365604400635, | |
| "learning_rate": 0.0003574074074074074, | |
| "loss": 0.3846, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 27.289955780164245, | |
| "grad_norm": 5.114433765411377, | |
| "learning_rate": 0.00035555555555555557, | |
| "loss": 0.3416, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 27.289955780164245, | |
| "eval_accuracy": 0.6508630609896433, | |
| "eval_f1_macro": 0.47374348013820844, | |
| "eval_loss": 1.941726565361023, | |
| "eval_runtime": 38.0186, | |
| "eval_samples_per_second": 228.572, | |
| "eval_steps_per_second": 7.154, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 27.542640555906505, | |
| "grad_norm": 5.3969902992248535, | |
| "learning_rate": 0.0003537037037037037, | |
| "loss": 0.3384, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 27.79532533164877, | |
| "grad_norm": 6.363078594207764, | |
| "learning_rate": 0.0003518518518518519, | |
| "loss": 0.3605, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 28.04801010739103, | |
| "grad_norm": 5.5530500411987305, | |
| "learning_rate": 0.00035, | |
| "loss": 0.3549, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 28.300694883133293, | |
| "grad_norm": 5.311258316040039, | |
| "learning_rate": 0.00034814814814814816, | |
| "loss": 0.3201, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 28.300694883133293, | |
| "eval_accuracy": 0.65512082853855, | |
| "eval_f1_macro": 0.48392342519236453, | |
| "eval_loss": 1.914045810699463, | |
| "eval_runtime": 37.9055, | |
| "eval_samples_per_second": 229.254, | |
| "eval_steps_per_second": 7.176, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 28.553379658875553, | |
| "grad_norm": 5.631585121154785, | |
| "learning_rate": 0.00034629629629629626, | |
| "loss": 0.3135, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 28.806064434617813, | |
| "grad_norm": 3.972892999649048, | |
| "learning_rate": 0.0003444444444444445, | |
| "loss": 0.3396, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 29.058749210360077, | |
| "grad_norm": 12.790987014770508, | |
| "learning_rate": 0.00034259259259259263, | |
| "loss": 0.3333, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 29.311433986102337, | |
| "grad_norm": 7.01127290725708, | |
| "learning_rate": 0.00034074074074074074, | |
| "loss": 0.2831, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 29.311433986102337, | |
| "eval_accuracy": 0.646029919447641, | |
| "eval_f1_macro": 0.4654633218896299, | |
| "eval_loss": 1.9762213230133057, | |
| "eval_runtime": 38.0033, | |
| "eval_samples_per_second": 228.664, | |
| "eval_steps_per_second": 7.157, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 29.5641187618446, | |
| "grad_norm": 9.544097900390625, | |
| "learning_rate": 0.0003388888888888889, | |
| "loss": 0.3253, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 29.81680353758686, | |
| "grad_norm": 4.85942268371582, | |
| "learning_rate": 0.00033703703703703706, | |
| "loss": 0.2931, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 30.06948831332912, | |
| "grad_norm": 5.682592868804932, | |
| "learning_rate": 0.0003351851851851852, | |
| "loss": 0.2976, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 30.322173089071384, | |
| "grad_norm": 4.502624034881592, | |
| "learning_rate": 0.0003333333333333333, | |
| "loss": 0.2701, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 30.322173089071384, | |
| "eval_accuracy": 0.6638665132336018, | |
| "eval_f1_macro": 0.48639052612692846, | |
| "eval_loss": 1.9732766151428223, | |
| "eval_runtime": 38.1849, | |
| "eval_samples_per_second": 227.577, | |
| "eval_steps_per_second": 7.123, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 30.574857864813644, | |
| "grad_norm": 3.922891139984131, | |
| "learning_rate": 0.0003314814814814815, | |
| "loss": 0.2949, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 30.827542640555908, | |
| "grad_norm": 5.285627365112305, | |
| "learning_rate": 0.0003296296296296296, | |
| "loss": 0.301, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 31.080227416298168, | |
| "grad_norm": 4.516997814178467, | |
| "learning_rate": 0.0003277777777777778, | |
| "loss": 0.2843, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 31.33291219204043, | |
| "grad_norm": 6.056753158569336, | |
| "learning_rate": 0.00032592592592592596, | |
| "loss": 0.2521, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 31.33291219204043, | |
| "eval_accuracy": 0.6558112773302647, | |
| "eval_f1_macro": 0.4871080465941763, | |
| "eval_loss": 2.021097421646118, | |
| "eval_runtime": 37.6773, | |
| "eval_samples_per_second": 230.643, | |
| "eval_steps_per_second": 7.219, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 31.585596967782692, | |
| "grad_norm": 5.199705600738525, | |
| "learning_rate": 0.00032407407407407406, | |
| "loss": 0.2742, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 31.838281743524952, | |
| "grad_norm": 4.554365634918213, | |
| "learning_rate": 0.0003222222222222222, | |
| "loss": 0.2721, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 32.09096651926721, | |
| "grad_norm": 5.793283462524414, | |
| "learning_rate": 0.0003203703703703704, | |
| "loss": 0.2662, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 32.34365129500947, | |
| "grad_norm": 4.629561424255371, | |
| "learning_rate": 0.00031851851851851854, | |
| "loss": 0.2498, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 32.34365129500947, | |
| "eval_accuracy": 0.6605293440736478, | |
| "eval_f1_macro": 0.47829496929439297, | |
| "eval_loss": 2.015612840652466, | |
| "eval_runtime": 37.6372, | |
| "eval_samples_per_second": 230.889, | |
| "eval_steps_per_second": 7.227, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 32.59633607075174, | |
| "grad_norm": 5.273705005645752, | |
| "learning_rate": 0.00031666666666666665, | |
| "loss": 0.2563, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 32.849020846494, | |
| "grad_norm": 6.036963939666748, | |
| "learning_rate": 0.0003148148148148148, | |
| "loss": 0.2452, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 33.10170562223626, | |
| "grad_norm": 6.134855270385742, | |
| "learning_rate": 0.00031296296296296297, | |
| "loss": 0.2465, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 33.35439039797852, | |
| "grad_norm": 5.270202159881592, | |
| "learning_rate": 0.0003111111111111111, | |
| "loss": 0.2292, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 33.35439039797852, | |
| "eval_accuracy": 0.6576524741081703, | |
| "eval_f1_macro": 0.4612500793066927, | |
| "eval_loss": 1.9660779237747192, | |
| "eval_runtime": 37.6766, | |
| "eval_samples_per_second": 230.647, | |
| "eval_steps_per_second": 7.219, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 33.60707517372078, | |
| "grad_norm": 5.208720684051514, | |
| "learning_rate": 0.00030925925925925923, | |
| "loss": 0.2426, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 33.85975994946305, | |
| "grad_norm": 6.54056453704834, | |
| "learning_rate": 0.0003074074074074074, | |
| "loss": 0.2382, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 34.11244472520531, | |
| "grad_norm": 5.9190473556518555, | |
| "learning_rate": 0.0003055555555555556, | |
| "loss": 0.2311, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 34.36512950094757, | |
| "grad_norm": 7.287013530731201, | |
| "learning_rate": 0.0003037037037037037, | |
| "loss": 0.2234, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 34.36512950094757, | |
| "eval_accuracy": 0.6574223245109321, | |
| "eval_f1_macro": 0.4842256883150983, | |
| "eval_loss": 2.1211564540863037, | |
| "eval_runtime": 37.6699, | |
| "eval_samples_per_second": 230.688, | |
| "eval_steps_per_second": 7.221, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 34.61781427668983, | |
| "grad_norm": 4.2805891036987305, | |
| "learning_rate": 0.00030185185185185187, | |
| "loss": 0.2349, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 34.87049905243209, | |
| "grad_norm": 5.430190563201904, | |
| "learning_rate": 0.0003, | |
| "loss": 0.22, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 35.123183828174355, | |
| "grad_norm": 3.191734552383423, | |
| "learning_rate": 0.0002981481481481482, | |
| "loss": 0.2062, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 35.375868603916615, | |
| "grad_norm": 13.889403343200684, | |
| "learning_rate": 0.0002962962962962963, | |
| "loss": 0.2236, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 35.375868603916615, | |
| "eval_accuracy": 0.6704257767548907, | |
| "eval_f1_macro": 0.47431074694572617, | |
| "eval_loss": 1.8960274457931519, | |
| "eval_runtime": 37.6349, | |
| "eval_samples_per_second": 230.903, | |
| "eval_steps_per_second": 7.227, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 35.628553379658875, | |
| "grad_norm": 4.789304256439209, | |
| "learning_rate": 0.00029444444444444445, | |
| "loss": 0.2185, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 35.881238155401135, | |
| "grad_norm": 6.029698848724365, | |
| "learning_rate": 0.00029259259259259256, | |
| "loss": 0.2044, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 36.133922931143395, | |
| "grad_norm": 5.329158306121826, | |
| "learning_rate": 0.00029074074074074077, | |
| "loss": 0.1985, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 36.38660770688566, | |
| "grad_norm": 7.056049823760986, | |
| "learning_rate": 0.0002888888888888889, | |
| "loss": 0.1917, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 36.38660770688566, | |
| "eval_accuracy": 0.6565017261219793, | |
| "eval_f1_macro": 0.465885996111531, | |
| "eval_loss": 2.1065523624420166, | |
| "eval_runtime": 37.6412, | |
| "eval_samples_per_second": 230.864, | |
| "eval_steps_per_second": 7.226, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 36.63929248262792, | |
| "grad_norm": 3.2443530559539795, | |
| "learning_rate": 0.00028703703703703703, | |
| "loss": 0.204, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 36.89197725837018, | |
| "grad_norm": 3.2316298484802246, | |
| "learning_rate": 0.0002851851851851852, | |
| "loss": 0.2049, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 37.14466203411244, | |
| "grad_norm": 4.233123779296875, | |
| "learning_rate": 0.00028333333333333335, | |
| "loss": 0.1971, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 37.3973468098547, | |
| "grad_norm": 5.144405364990234, | |
| "learning_rate": 0.0002814814814814815, | |
| "loss": 0.1903, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 37.3973468098547, | |
| "eval_accuracy": 0.6695051783659378, | |
| "eval_f1_macro": 0.49018342728532344, | |
| "eval_loss": 2.024235248565674, | |
| "eval_runtime": 37.6406, | |
| "eval_samples_per_second": 230.868, | |
| "eval_steps_per_second": 7.226, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 37.65003158559697, | |
| "grad_norm": 5.579497814178467, | |
| "learning_rate": 0.0002796296296296296, | |
| "loss": 0.1971, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 37.90271636133923, | |
| "grad_norm": 6.287644386291504, | |
| "learning_rate": 0.0002777777777777778, | |
| "loss": 0.1956, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 38.15540113708149, | |
| "grad_norm": 2.8196208477020264, | |
| "learning_rate": 0.00027592592592592594, | |
| "loss": 0.1735, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 38.40808591282375, | |
| "grad_norm": 6.207739353179932, | |
| "learning_rate": 0.0002740740740740741, | |
| "loss": 0.1832, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 38.40808591282375, | |
| "eval_accuracy": 0.6682393555811277, | |
| "eval_f1_macro": 0.489118101346677, | |
| "eval_loss": 2.0150773525238037, | |
| "eval_runtime": 37.612, | |
| "eval_samples_per_second": 231.043, | |
| "eval_steps_per_second": 7.232, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 38.66077068856601, | |
| "grad_norm": 3.1211068630218506, | |
| "learning_rate": 0.0002722222222222222, | |
| "loss": 0.1926, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 38.91345546430828, | |
| "grad_norm": 5.007220268249512, | |
| "learning_rate": 0.00027037037037037036, | |
| "loss": 0.2029, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 39.16614024005054, | |
| "grad_norm": 3.8774144649505615, | |
| "learning_rate": 0.0002685185185185186, | |
| "loss": 0.1706, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 39.4188250157928, | |
| "grad_norm": 5.49664831161499, | |
| "learning_rate": 0.0002666666666666667, | |
| "loss": 0.1696, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 39.4188250157928, | |
| "eval_accuracy": 0.6718066743383199, | |
| "eval_f1_macro": 0.4890009527209215, | |
| "eval_loss": 2.0123181343078613, | |
| "eval_runtime": 37.6065, | |
| "eval_samples_per_second": 231.077, | |
| "eval_steps_per_second": 7.233, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 39.67150979153506, | |
| "grad_norm": 1.5239861011505127, | |
| "learning_rate": 0.00026481481481481484, | |
| "loss": 0.1689, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 39.92419456727732, | |
| "grad_norm": 3.3123672008514404, | |
| "learning_rate": 0.00026296296296296294, | |
| "loss": 0.1716, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 40.176879343019586, | |
| "grad_norm": 4.820850372314453, | |
| "learning_rate": 0.00026111111111111116, | |
| "loss": 0.1548, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 40.429564118761846, | |
| "grad_norm": 13.379387855529785, | |
| "learning_rate": 0.00025925925925925926, | |
| "loss": 0.1622, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 40.429564118761846, | |
| "eval_accuracy": 0.6690448791714615, | |
| "eval_f1_macro": 0.5054021628227856, | |
| "eval_loss": 1.9521493911743164, | |
| "eval_runtime": 37.6399, | |
| "eval_samples_per_second": 230.872, | |
| "eval_steps_per_second": 7.226, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 40.682248894504106, | |
| "grad_norm": 5.211925983428955, | |
| "learning_rate": 0.0002574074074074074, | |
| "loss": 0.1565, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 40.934933670246366, | |
| "grad_norm": 4.36137580871582, | |
| "learning_rate": 0.00025555555555555553, | |
| "loss": 0.1653, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 41.187618445988626, | |
| "grad_norm": 4.851421356201172, | |
| "learning_rate": 0.00025370370370370374, | |
| "loss": 0.1553, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 41.44030322173089, | |
| "grad_norm": 3.541837692260742, | |
| "learning_rate": 0.00025185185185185185, | |
| "loss": 0.1589, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 41.44030322173089, | |
| "eval_accuracy": 0.6802071346375144, | |
| "eval_f1_macro": 0.4877877911094619, | |
| "eval_loss": 1.9544481039047241, | |
| "eval_runtime": 37.6399, | |
| "eval_samples_per_second": 230.872, | |
| "eval_steps_per_second": 7.226, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 41.69298799747315, | |
| "grad_norm": 3.3895256519317627, | |
| "learning_rate": 0.00025, | |
| "loss": 0.1515, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 41.94567277321541, | |
| "grad_norm": 4.256431579589844, | |
| "learning_rate": 0.00024814814814814816, | |
| "loss": 0.163, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 42.198357548957674, | |
| "grad_norm": 3.3639395236968994, | |
| "learning_rate": 0.00024629629629629627, | |
| "loss": 0.1497, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 42.451042324699934, | |
| "grad_norm": 6.572113990783691, | |
| "learning_rate": 0.00024444444444444443, | |
| "loss": 0.1479, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 42.451042324699934, | |
| "eval_accuracy": 0.6668584579976985, | |
| "eval_f1_macro": 0.49095063827235297, | |
| "eval_loss": 2.067105293273926, | |
| "eval_runtime": 37.6344, | |
| "eval_samples_per_second": 230.906, | |
| "eval_steps_per_second": 7.227, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 42.7037271004422, | |
| "grad_norm": 4.106721878051758, | |
| "learning_rate": 0.0002425925925925926, | |
| "loss": 0.1539, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 42.95641187618446, | |
| "grad_norm": 7.537932872772217, | |
| "learning_rate": 0.00024074074074074072, | |
| "loss": 0.1546, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 43.20909665192672, | |
| "grad_norm": 1.923229694366455, | |
| "learning_rate": 0.0002388888888888889, | |
| "loss": 0.1221, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 43.46178142766898, | |
| "grad_norm": 4.5005106925964355, | |
| "learning_rate": 0.00023703703703703704, | |
| "loss": 0.137, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 43.46178142766898, | |
| "eval_accuracy": 0.6779056386651323, | |
| "eval_f1_macro": 0.46761720782549016, | |
| "eval_loss": 2.0958869457244873, | |
| "eval_runtime": 37.5801, | |
| "eval_samples_per_second": 231.24, | |
| "eval_steps_per_second": 7.238, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 43.71446620341124, | |
| "grad_norm": 5.6225905418396, | |
| "learning_rate": 0.0002351851851851852, | |
| "loss": 0.1468, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 43.96715097915351, | |
| "grad_norm": 3.400129556655884, | |
| "learning_rate": 0.00023333333333333333, | |
| "loss": 0.1475, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 44.21983575489577, | |
| "grad_norm": 6.243558883666992, | |
| "learning_rate": 0.0002314814814814815, | |
| "loss": 0.1263, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 44.47252053063803, | |
| "grad_norm": 3.487706184387207, | |
| "learning_rate": 0.00022962962962962962, | |
| "loss": 0.1436, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 44.47252053063803, | |
| "eval_accuracy": 0.6789413118527042, | |
| "eval_f1_macro": 0.49075561338585527, | |
| "eval_loss": 2.021660566329956, | |
| "eval_runtime": 37.6019, | |
| "eval_samples_per_second": 231.105, | |
| "eval_steps_per_second": 7.234, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 44.72520530638029, | |
| "grad_norm": 3.1665256023406982, | |
| "learning_rate": 0.00022777777777777778, | |
| "loss": 0.122, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 44.97789008212255, | |
| "grad_norm": 6.8903937339782715, | |
| "learning_rate": 0.00022592592592592591, | |
| "loss": 0.1325, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 45.230574857864816, | |
| "grad_norm": 5.111687660217285, | |
| "learning_rate": 0.00022407407407407407, | |
| "loss": 0.1215, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 45.48325963360708, | |
| "grad_norm": 5.702262878417969, | |
| "learning_rate": 0.0002222222222222222, | |
| "loss": 0.1376, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 45.48325963360708, | |
| "eval_accuracy": 0.6742232451093211, | |
| "eval_f1_macro": 0.4685316288362612, | |
| "eval_loss": 2.1048314571380615, | |
| "eval_runtime": 37.6304, | |
| "eval_samples_per_second": 230.931, | |
| "eval_steps_per_second": 7.228, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 45.73594440934934, | |
| "grad_norm": 3.1774654388427734, | |
| "learning_rate": 0.0002203703703703704, | |
| "loss": 0.1168, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 45.9886291850916, | |
| "grad_norm": 2.6389782428741455, | |
| "learning_rate": 0.00021851851851851852, | |
| "loss": 0.1269, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 46.24131396083386, | |
| "grad_norm": 6.534961700439453, | |
| "learning_rate": 0.00021666666666666668, | |
| "loss": 0.1153, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 46.493998736576124, | |
| "grad_norm": 2.352351665496826, | |
| "learning_rate": 0.00021481481481481482, | |
| "loss": 0.1144, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 46.493998736576124, | |
| "eval_accuracy": 0.67307249712313, | |
| "eval_f1_macro": 0.48375331337278604, | |
| "eval_loss": 2.1653566360473633, | |
| "eval_runtime": 37.61, | |
| "eval_samples_per_second": 231.055, | |
| "eval_steps_per_second": 7.232, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 46.746683512318384, | |
| "grad_norm": 5.173340320587158, | |
| "learning_rate": 0.00021296296296296298, | |
| "loss": 0.1283, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 46.999368288060644, | |
| "grad_norm": 3.5517702102661133, | |
| "learning_rate": 0.0002111111111111111, | |
| "loss": 0.1251, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 47.252053063802904, | |
| "grad_norm": 3.369018793106079, | |
| "learning_rate": 0.00020925925925925927, | |
| "loss": 0.123, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 47.504737839545164, | |
| "grad_norm": 3.6757843494415283, | |
| "learning_rate": 0.0002074074074074074, | |
| "loss": 0.1074, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 47.504737839545164, | |
| "eval_accuracy": 0.6790563866513234, | |
| "eval_f1_macro": 0.5064960804642658, | |
| "eval_loss": 2.1274356842041016, | |
| "eval_runtime": 37.6079, | |
| "eval_samples_per_second": 231.069, | |
| "eval_steps_per_second": 7.233, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 47.75742261528743, | |
| "grad_norm": 4.2758355140686035, | |
| "learning_rate": 0.00020555555555555556, | |
| "loss": 0.105, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 48.01010739102969, | |
| "grad_norm": 3.747074604034424, | |
| "learning_rate": 0.0002037037037037037, | |
| "loss": 0.1192, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 48.26279216677195, | |
| "grad_norm": 2.5319418907165527, | |
| "learning_rate": 0.00020185185185185188, | |
| "loss": 0.1069, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 48.51547694251421, | |
| "grad_norm": 2.424844264984131, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1096, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 48.51547694251421, | |
| "eval_accuracy": 0.6822784810126582, | |
| "eval_f1_macro": 0.47328825429255844, | |
| "eval_loss": 2.1717238426208496, | |
| "eval_runtime": 37.5912, | |
| "eval_samples_per_second": 231.171, | |
| "eval_steps_per_second": 7.236, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 48.76816171825647, | |
| "grad_norm": 2.063728094100952, | |
| "learning_rate": 0.00019814814814814817, | |
| "loss": 0.1037, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 49.02084649399874, | |
| "grad_norm": 4.786061763763428, | |
| "learning_rate": 0.0001962962962962963, | |
| "loss": 0.115, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 49.273531269741, | |
| "grad_norm": 4.9217047691345215, | |
| "learning_rate": 0.00019444444444444446, | |
| "loss": 0.1019, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 49.52621604548326, | |
| "grad_norm": 1.9982796907424927, | |
| "learning_rate": 0.0001925925925925926, | |
| "loss": 0.0913, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 49.52621604548326, | |
| "eval_accuracy": 0.6803222094361335, | |
| "eval_f1_macro": 0.49140242513418214, | |
| "eval_loss": 2.189754009246826, | |
| "eval_runtime": 37.6051, | |
| "eval_samples_per_second": 231.086, | |
| "eval_steps_per_second": 7.233, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 49.77890082122552, | |
| "grad_norm": 6.895193576812744, | |
| "learning_rate": 0.00019074074074074075, | |
| "loss": 0.0921, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 50.03158559696778, | |
| "grad_norm": 4.949813365936279, | |
| "learning_rate": 0.00018888888888888888, | |
| "loss": 0.0972, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 50.28427037271005, | |
| "grad_norm": 3.9698879718780518, | |
| "learning_rate": 0.00018703703703703704, | |
| "loss": 0.0872, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 50.53695514845231, | |
| "grad_norm": 6.452556610107422, | |
| "learning_rate": 0.00018518518518518518, | |
| "loss": 0.0873, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 50.53695514845231, | |
| "eval_accuracy": 0.6830840046029919, | |
| "eval_f1_macro": 0.5074439371494469, | |
| "eval_loss": 2.1859612464904785, | |
| "eval_runtime": 37.4194, | |
| "eval_samples_per_second": 232.232, | |
| "eval_steps_per_second": 7.269, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 50.78963992419457, | |
| "grad_norm": 3.542201042175293, | |
| "learning_rate": 0.00018333333333333334, | |
| "loss": 0.1025, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 51.04232469993683, | |
| "grad_norm": 2.0884666442871094, | |
| "learning_rate": 0.0001814814814814815, | |
| "loss": 0.0918, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 51.29500947567909, | |
| "grad_norm": 2.5354607105255127, | |
| "learning_rate": 0.00017962962962962965, | |
| "loss": 0.0742, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 51.547694251421355, | |
| "grad_norm": 3.340730905532837, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 0.0914, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 51.547694251421355, | |
| "eval_accuracy": 0.6856156501726122, | |
| "eval_f1_macro": 0.5103333986880413, | |
| "eval_loss": 2.116797924041748, | |
| "eval_runtime": 37.578, | |
| "eval_samples_per_second": 231.252, | |
| "eval_steps_per_second": 7.238, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 51.800379027163615, | |
| "grad_norm": 4.722923278808594, | |
| "learning_rate": 0.00017592592592592595, | |
| "loss": 0.093, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 52.053063802905875, | |
| "grad_norm": 3.596904754638672, | |
| "learning_rate": 0.00017407407407407408, | |
| "loss": 0.0861, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 52.305748578648135, | |
| "grad_norm": 5.971360206604004, | |
| "learning_rate": 0.00017222222222222224, | |
| "loss": 0.0803, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 52.558433354390395, | |
| "grad_norm": 3.8663809299468994, | |
| "learning_rate": 0.00017037037037037037, | |
| "loss": 0.0911, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 52.558433354390395, | |
| "eval_accuracy": 0.6852704257767549, | |
| "eval_f1_macro": 0.49064573818288437, | |
| "eval_loss": 2.1430506706237793, | |
| "eval_runtime": 37.5953, | |
| "eval_samples_per_second": 231.146, | |
| "eval_steps_per_second": 7.235, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 52.81111813013266, | |
| "grad_norm": 3.418740749359131, | |
| "learning_rate": 0.00016851851851851853, | |
| "loss": 0.0783, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 53.06380290587492, | |
| "grad_norm": 1.8313372135162354, | |
| "learning_rate": 0.00016666666666666666, | |
| "loss": 0.0777, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 53.31648768161718, | |
| "grad_norm": 2.064988851547241, | |
| "learning_rate": 0.0001648148148148148, | |
| "loss": 0.0734, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 53.56917245735944, | |
| "grad_norm": 2.213332414627075, | |
| "learning_rate": 0.00016296296296296298, | |
| "loss": 0.0807, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 53.56917245735944, | |
| "eval_accuracy": 0.69102416570771, | |
| "eval_f1_macro": 0.49167230060507133, | |
| "eval_loss": 2.02927303314209, | |
| "eval_runtime": 37.5762, | |
| "eval_samples_per_second": 231.263, | |
| "eval_steps_per_second": 7.239, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 53.8218572331017, | |
| "grad_norm": 2.934459686279297, | |
| "learning_rate": 0.0001611111111111111, | |
| "loss": 0.0748, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 54.07454200884397, | |
| "grad_norm": 1.5093865394592285, | |
| "learning_rate": 0.00015925925925925927, | |
| "loss": 0.0804, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 54.32722678458623, | |
| "grad_norm": 4.368808746337891, | |
| "learning_rate": 0.0001574074074074074, | |
| "loss": 0.0744, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 54.57991156032849, | |
| "grad_norm": 3.199536085128784, | |
| "learning_rate": 0.00015555555555555556, | |
| "loss": 0.07, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 54.57991156032849, | |
| "eval_accuracy": 0.6880322209436134, | |
| "eval_f1_macro": 0.5191774594924041, | |
| "eval_loss": 2.117441177368164, | |
| "eval_runtime": 37.5744, | |
| "eval_samples_per_second": 231.274, | |
| "eval_steps_per_second": 7.239, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 54.83259633607075, | |
| "grad_norm": 5.144430637359619, | |
| "learning_rate": 0.0001537037037037037, | |
| "loss": 0.0755, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 55.08528111181301, | |
| "grad_norm": 1.9680262804031372, | |
| "learning_rate": 0.00015185185185185185, | |
| "loss": 0.0839, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 55.33796588755528, | |
| "grad_norm": 6.2530837059021, | |
| "learning_rate": 0.00015, | |
| "loss": 0.0692, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 55.59065066329754, | |
| "grad_norm": 3.312941551208496, | |
| "learning_rate": 0.00014814814814814815, | |
| "loss": 0.0723, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 55.59065066329754, | |
| "eval_accuracy": 0.6882623705408516, | |
| "eval_f1_macro": 0.4923514471900252, | |
| "eval_loss": 2.14955735206604, | |
| "eval_runtime": 37.5658, | |
| "eval_samples_per_second": 231.327, | |
| "eval_steps_per_second": 7.241, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 55.8433354390398, | |
| "grad_norm": 2.790160655975342, | |
| "learning_rate": 0.00014629629629629628, | |
| "loss": 0.0666, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 56.09602021478206, | |
| "grad_norm": 0.18920756876468658, | |
| "learning_rate": 0.00014444444444444444, | |
| "loss": 0.0673, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 56.34870499052432, | |
| "grad_norm": 3.6166439056396484, | |
| "learning_rate": 0.0001425925925925926, | |
| "loss": 0.0658, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 56.601389766266585, | |
| "grad_norm": 4.014189720153809, | |
| "learning_rate": 0.00014074074074074076, | |
| "loss": 0.0666, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 56.601389766266585, | |
| "eval_accuracy": 0.6874568469505178, | |
| "eval_f1_macro": 0.4789205528590129, | |
| "eval_loss": 2.2032995223999023, | |
| "eval_runtime": 37.5869, | |
| "eval_samples_per_second": 231.198, | |
| "eval_steps_per_second": 7.237, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 56.854074542008846, | |
| "grad_norm": 3.889650583267212, | |
| "learning_rate": 0.0001388888888888889, | |
| "loss": 0.0664, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 57.106759317751106, | |
| "grad_norm": 1.407404899597168, | |
| "learning_rate": 0.00013703703703703705, | |
| "loss": 0.063, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 57.359444093493366, | |
| "grad_norm": 2.4211344718933105, | |
| "learning_rate": 0.00013518518518518518, | |
| "loss": 0.0576, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 57.612128869235626, | |
| "grad_norm": 2.3689253330230713, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 0.0543, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 57.612128869235626, | |
| "eval_accuracy": 0.6943613348676639, | |
| "eval_f1_macro": 0.4842565035933064, | |
| "eval_loss": 2.2275025844573975, | |
| "eval_runtime": 37.6062, | |
| "eval_samples_per_second": 231.079, | |
| "eval_steps_per_second": 7.233, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 57.86481364497789, | |
| "grad_norm": 3.9035773277282715, | |
| "learning_rate": 0.00013148148148148147, | |
| "loss": 0.0616, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 58.11749842072015, | |
| "grad_norm": 1.0310261249542236, | |
| "learning_rate": 0.00012962962962962963, | |
| "loss": 0.0637, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 58.37018319646241, | |
| "grad_norm": 4.086586952209473, | |
| "learning_rate": 0.00012777777777777776, | |
| "loss": 0.0485, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 58.62286797220467, | |
| "grad_norm": 3.3585879802703857, | |
| "learning_rate": 0.00012592592592592592, | |
| "loss": 0.052, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 58.62286797220467, | |
| "eval_accuracy": 0.6880322209436134, | |
| "eval_f1_macro": 0.48385398427069204, | |
| "eval_loss": 2.2420876026153564, | |
| "eval_runtime": 37.594, | |
| "eval_samples_per_second": 231.154, | |
| "eval_steps_per_second": 7.235, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 58.875552747946934, | |
| "grad_norm": 1.1262246370315552, | |
| "learning_rate": 0.00012407407407407408, | |
| "loss": 0.0611, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 59.1282375236892, | |
| "grad_norm": 3.428152084350586, | |
| "learning_rate": 0.00012222222222222221, | |
| "loss": 0.0478, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 59.38092229943146, | |
| "grad_norm": 2.774853467941284, | |
| "learning_rate": 0.00012037037037037036, | |
| "loss": 0.0596, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 59.63360707517372, | |
| "grad_norm": 1.6087123155593872, | |
| "learning_rate": 0.00011851851851851852, | |
| "loss": 0.0485, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 59.63360707517372, | |
| "eval_accuracy": 0.6960874568469505, | |
| "eval_f1_macro": 0.49161132476399, | |
| "eval_loss": 2.205902099609375, | |
| "eval_runtime": 37.5949, | |
| "eval_samples_per_second": 231.149, | |
| "eval_steps_per_second": 7.235, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 59.88629185091598, | |
| "grad_norm": 5.234879970550537, | |
| "learning_rate": 0.00011666666666666667, | |
| "loss": 0.0525, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 60.13897662665824, | |
| "grad_norm": 1.3339439630508423, | |
| "learning_rate": 0.00011481481481481481, | |
| "loss": 0.0446, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 60.39166140240051, | |
| "grad_norm": 3.555692434310913, | |
| "learning_rate": 0.00011296296296296296, | |
| "loss": 0.0491, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 60.64434617814277, | |
| "grad_norm": 2.1915643215179443, | |
| "learning_rate": 0.0001111111111111111, | |
| "loss": 0.0466, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 60.64434617814277, | |
| "eval_accuracy": 0.6993095512082854, | |
| "eval_f1_macro": 0.4888962992436562, | |
| "eval_loss": 2.21913743019104, | |
| "eval_runtime": 37.5927, | |
| "eval_samples_per_second": 231.162, | |
| "eval_steps_per_second": 7.235, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 60.89703095388503, | |
| "grad_norm": 2.9716124534606934, | |
| "learning_rate": 0.00010925925925925926, | |
| "loss": 0.0439, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 61.14971572962729, | |
| "grad_norm": 4.863760471343994, | |
| "learning_rate": 0.00010740740740740741, | |
| "loss": 0.0439, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 61.40240050536955, | |
| "grad_norm": 1.8985756635665894, | |
| "learning_rate": 0.00010555555555555555, | |
| "loss": 0.0444, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 61.655085281111816, | |
| "grad_norm": 1.8271839618682861, | |
| "learning_rate": 0.0001037037037037037, | |
| "loss": 0.0492, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 61.655085281111816, | |
| "eval_accuracy": 0.6978135788262371, | |
| "eval_f1_macro": 0.5091662249913542, | |
| "eval_loss": 2.2356083393096924, | |
| "eval_runtime": 37.5823, | |
| "eval_samples_per_second": 231.226, | |
| "eval_steps_per_second": 7.237, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 61.907770056854076, | |
| "grad_norm": 4.40165376663208, | |
| "learning_rate": 0.00010185185185185185, | |
| "loss": 0.0474, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 62.160454832596336, | |
| "grad_norm": 5.742025852203369, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0415, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 62.4131396083386, | |
| "grad_norm": 2.5721802711486816, | |
| "learning_rate": 9.814814814814815e-05, | |
| "loss": 0.0377, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 62.66582438408086, | |
| "grad_norm": 0.10644868016242981, | |
| "learning_rate": 9.62962962962963e-05, | |
| "loss": 0.0467, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 62.66582438408086, | |
| "eval_accuracy": 0.6944764096662831, | |
| "eval_f1_macro": 0.4841950729208342, | |
| "eval_loss": 2.234914541244507, | |
| "eval_runtime": 38.1342, | |
| "eval_samples_per_second": 227.88, | |
| "eval_steps_per_second": 7.133, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 62.918509159823124, | |
| "grad_norm": 2.9092166423797607, | |
| "learning_rate": 9.444444444444444e-05, | |
| "loss": 0.0435, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 63.171193935565384, | |
| "grad_norm": 1.0841565132141113, | |
| "learning_rate": 9.259259259259259e-05, | |
| "loss": 0.0375, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 63.423878711307644, | |
| "grad_norm": 2.1147305965423584, | |
| "learning_rate": 9.074074074074075e-05, | |
| "loss": 0.0361, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 63.676563487049904, | |
| "grad_norm": 1.9272477626800537, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.0425, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 63.676563487049904, | |
| "eval_accuracy": 0.6973532796317606, | |
| "eval_f1_macro": 0.49266241462604743, | |
| "eval_loss": 2.2320919036865234, | |
| "eval_runtime": 37.8637, | |
| "eval_samples_per_second": 229.507, | |
| "eval_steps_per_second": 7.184, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 63.929248262792164, | |
| "grad_norm": 0.2638062834739685, | |
| "learning_rate": 8.703703703703704e-05, | |
| "loss": 0.0312, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 64.18193303853442, | |
| "grad_norm": 0.4505729079246521, | |
| "learning_rate": 8.518518518518518e-05, | |
| "loss": 0.0368, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 64.43461781427669, | |
| "grad_norm": 2.9775772094726562, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 0.0307, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 64.68730259001894, | |
| "grad_norm": 0.03926827386021614, | |
| "learning_rate": 8.148148148148149e-05, | |
| "loss": 0.0329, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 64.68730259001894, | |
| "eval_accuracy": 0.7001150747986191, | |
| "eval_f1_macro": 0.49594084798647425, | |
| "eval_loss": 2.309037208557129, | |
| "eval_runtime": 38.1104, | |
| "eval_samples_per_second": 228.022, | |
| "eval_steps_per_second": 7.137, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 64.93998736576121, | |
| "grad_norm": 3.131605625152588, | |
| "learning_rate": 7.962962962962964e-05, | |
| "loss": 0.0349, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 65.19267214150348, | |
| "grad_norm": 5.090763568878174, | |
| "learning_rate": 7.777777777777778e-05, | |
| "loss": 0.0347, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 65.44535691724573, | |
| "grad_norm": 2.136350393295288, | |
| "learning_rate": 7.592592592592593e-05, | |
| "loss": 0.0296, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 65.698041692988, | |
| "grad_norm": 1.2904014587402344, | |
| "learning_rate": 7.407407407407407e-05, | |
| "loss": 0.0344, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 65.698041692988, | |
| "eval_accuracy": 0.6997698504027619, | |
| "eval_f1_macro": 0.4852459931479808, | |
| "eval_loss": 2.3282463550567627, | |
| "eval_runtime": 37.8105, | |
| "eval_samples_per_second": 229.831, | |
| "eval_steps_per_second": 7.194, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 65.95072646873025, | |
| "grad_norm": 0.028127046301960945, | |
| "learning_rate": 7.222222222222222e-05, | |
| "loss": 0.0282, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 66.20341124447252, | |
| "grad_norm": 0.11164356768131256, | |
| "learning_rate": 7.037037037037038e-05, | |
| "loss": 0.0296, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 66.45609602021479, | |
| "grad_norm": 2.002295732498169, | |
| "learning_rate": 6.851851851851852e-05, | |
| "loss": 0.0324, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 66.70878079595704, | |
| "grad_norm": 2.676612615585327, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.0316, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 66.70878079595704, | |
| "eval_accuracy": 0.7020713463751438, | |
| "eval_f1_macro": 0.5033910282926815, | |
| "eval_loss": 2.2500393390655518, | |
| "eval_runtime": 37.7344, | |
| "eval_samples_per_second": 230.294, | |
| "eval_steps_per_second": 7.208, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 66.9614655716993, | |
| "grad_norm": 0.524612307548523, | |
| "learning_rate": 6.481481481481482e-05, | |
| "loss": 0.0272, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 67.21415034744156, | |
| "grad_norm": 5.645469665527344, | |
| "learning_rate": 6.296296296296296e-05, | |
| "loss": 0.0266, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 67.46683512318383, | |
| "grad_norm": 3.0155763626098633, | |
| "learning_rate": 6.111111111111111e-05, | |
| "loss": 0.0265, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 67.7195198989261, | |
| "grad_norm": 1.4998551607131958, | |
| "learning_rate": 5.925925925925926e-05, | |
| "loss": 0.0292, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 67.7195198989261, | |
| "eval_accuracy": 0.7002301495972382, | |
| "eval_f1_macro": 0.4923917551018341, | |
| "eval_loss": 2.336308240890503, | |
| "eval_runtime": 37.8308, | |
| "eval_samples_per_second": 229.707, | |
| "eval_steps_per_second": 7.19, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 67.97220467466835, | |
| "grad_norm": 3.1026313304901123, | |
| "learning_rate": 5.7407407407407406e-05, | |
| "loss": 0.0207, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 68.22488945041061, | |
| "grad_norm": 0.32973551750183105, | |
| "learning_rate": 5.555555555555555e-05, | |
| "loss": 0.0237, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 68.47757422615287, | |
| "grad_norm": 0.02183300070464611, | |
| "learning_rate": 5.3703703703703704e-05, | |
| "loss": 0.0316, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 68.73025900189513, | |
| "grad_norm": 0.04604117199778557, | |
| "learning_rate": 5.185185185185185e-05, | |
| "loss": 0.0276, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 68.73025900189513, | |
| "eval_accuracy": 0.7085155350978136, | |
| "eval_f1_macro": 0.5007118239174911, | |
| "eval_loss": 2.2208006381988525, | |
| "eval_runtime": 37.8426, | |
| "eval_samples_per_second": 229.636, | |
| "eval_steps_per_second": 7.188, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 68.9829437776374, | |
| "grad_norm": 1.813504934310913, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0236, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 69.23562855337966, | |
| "grad_norm": 1.0662651062011719, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.0208, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 69.48831332912192, | |
| "grad_norm": 0.9533895254135132, | |
| "learning_rate": 4.6296296296296294e-05, | |
| "loss": 0.0155, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 69.74099810486418, | |
| "grad_norm": 2.4781551361083984, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.0238, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 69.74099810486418, | |
| "eval_accuracy": 0.7037974683544304, | |
| "eval_f1_macro": 0.509261197731877, | |
| "eval_loss": 2.343860626220703, | |
| "eval_runtime": 38.0748, | |
| "eval_samples_per_second": 228.235, | |
| "eval_steps_per_second": 7.144, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 69.99368288060644, | |
| "grad_norm": 0.08948739618062973, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.0188, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 70.24636765634871, | |
| "grad_norm": 0.07096489518880844, | |
| "learning_rate": 4.0740740740740745e-05, | |
| "loss": 0.0181, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 70.49905243209096, | |
| "grad_norm": 0.2083720564842224, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.0185, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 70.75173720783323, | |
| "grad_norm": 2.5475051403045654, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.0216, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 70.75173720783323, | |
| "eval_accuracy": 0.7064441887226698, | |
| "eval_f1_macro": 0.484759251686816, | |
| "eval_loss": 2.3872413635253906, | |
| "eval_runtime": 37.8342, | |
| "eval_samples_per_second": 229.687, | |
| "eval_steps_per_second": 7.189, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 71.00442198357548, | |
| "grad_norm": 2.4494497776031494, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.018, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 71.25710675931775, | |
| "grad_norm": 8.410572052001953, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.016, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 71.50979153506002, | |
| "grad_norm": 1.5673861503601074, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.0208, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 71.76247631080227, | |
| "grad_norm": 0.37676000595092773, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.0217, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 71.76247631080227, | |
| "eval_accuracy": 0.7111622554660529, | |
| "eval_f1_macro": 0.5045859447220015, | |
| "eval_loss": 2.31369686126709, | |
| "eval_runtime": 38.0029, | |
| "eval_samples_per_second": 228.667, | |
| "eval_steps_per_second": 7.157, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 72.01516108654454, | |
| "grad_norm": 0.03591804578900337, | |
| "learning_rate": 2.7777777777777776e-05, | |
| "loss": 0.0211, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 72.26784586228679, | |
| "grad_norm": 3.384413957595825, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.0205, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 72.52053063802906, | |
| "grad_norm": 2.233063220977783, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.0132, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 72.77321541377133, | |
| "grad_norm": 0.5727182626724243, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0164, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 72.77321541377133, | |
| "eval_accuracy": 0.7100115074798619, | |
| "eval_f1_macro": 0.5025403728786204, | |
| "eval_loss": 2.311816930770874, | |
| "eval_runtime": 38.0077, | |
| "eval_samples_per_second": 228.638, | |
| "eval_steps_per_second": 7.156, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 73.02590018951358, | |
| "grad_norm": 0.052409425377845764, | |
| "learning_rate": 2.0370370370370372e-05, | |
| "loss": 0.0113, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 73.27858496525585, | |
| "grad_norm": 0.27832725644111633, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.0139, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 73.5312697409981, | |
| "grad_norm": 0.003998135682195425, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.016, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 73.78395451674037, | |
| "grad_norm": 0.08611119538545609, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.0143, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 73.78395451674037, | |
| "eval_accuracy": 0.7118527042577676, | |
| "eval_f1_macro": 0.4964154294350458, | |
| "eval_loss": 2.3175480365753174, | |
| "eval_runtime": 37.8473, | |
| "eval_samples_per_second": 229.607, | |
| "eval_steps_per_second": 7.187, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 74.03663929248263, | |
| "grad_norm": 1.426530361175537, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.0122, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 74.28932406822489, | |
| "grad_norm": 1.8630322217941284, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0132, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 74.54200884396715, | |
| "grad_norm": 0.13493819534778595, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.015, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 74.7946936197094, | |
| "grad_norm": 0.07671820372343063, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.015, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 74.7946936197094, | |
| "eval_accuracy": 0.711277330264672, | |
| "eval_f1_macro": 0.5024794350293241, | |
| "eval_loss": 2.3110148906707764, | |
| "eval_runtime": 38.1019, | |
| "eval_samples_per_second": 228.073, | |
| "eval_steps_per_second": 7.139, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 75.04737839545167, | |
| "grad_norm": 4.481829643249512, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0132, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 75.30006317119394, | |
| "grad_norm": 0.49408701062202454, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0109, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 75.5527479469362, | |
| "grad_norm": 0.8821873664855957, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.0144, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 75.80543272267846, | |
| "grad_norm": 1.098057746887207, | |
| "learning_rate": 0.0, | |
| "loss": 0.0149, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 75.80543272267846, | |
| "eval_accuracy": 0.7126582278481013, | |
| "eval_f1_macro": 0.502384644292337, | |
| "eval_loss": 2.3140313625335693, | |
| "eval_runtime": 37.7895, | |
| "eval_samples_per_second": 229.958, | |
| "eval_steps_per_second": 7.198, | |
| "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": 1.7426336812423224e+20, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |