| { | |
| "best_global_step": 26000, | |
| "best_metric": 74.62046528623556, | |
| "best_model_checkpoint": "./aramaic_diacritization_model_deep/checkpoint-26000", | |
| "epoch": 36.32760898282695, | |
| "eval_steps": 500, | |
| "global_step": 27500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13210039630118892, | |
| "grad_norm": 49.535423278808594, | |
| "learning_rate": 9.400000000000001e-07, | |
| "loss": 13.2291, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.26420079260237783, | |
| "grad_norm": 8.345534324645996, | |
| "learning_rate": 1.94e-06, | |
| "loss": 4.9005, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3963011889035667, | |
| "grad_norm": 6.039630889892578, | |
| "learning_rate": 2.9400000000000002e-06, | |
| "loss": 3.1719, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5284015852047557, | |
| "grad_norm": 5.1691412925720215, | |
| "learning_rate": 3.94e-06, | |
| "loss": 2.6108, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6605019815059445, | |
| "grad_norm": 4.929126262664795, | |
| "learning_rate": 4.94e-06, | |
| "loss": 2.2213, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6605019815059445, | |
| "eval_bleu": 0.19172411762066338, | |
| "eval_char_accuracy": 10.030946701760158, | |
| "eval_loss": 1.71331787109375, | |
| "eval_runtime": 308.7303, | |
| "eval_samples_per_second": 4.904, | |
| "eval_steps_per_second": 0.615, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7926023778071334, | |
| "grad_norm": 5.598935604095459, | |
| "learning_rate": 5.94e-06, | |
| "loss": 1.9171, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9247027741083224, | |
| "grad_norm": 5.103219032287598, | |
| "learning_rate": 6.9400000000000005e-06, | |
| "loss": 1.6624, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.0568031704095113, | |
| "grad_norm": 4.852541923522949, | |
| "learning_rate": 7.94e-06, | |
| "loss": 1.473, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.1889035667107002, | |
| "grad_norm": 4.4410624504089355, | |
| "learning_rate": 8.94e-06, | |
| "loss": 1.3081, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.321003963011889, | |
| "grad_norm": 3.9946470260620117, | |
| "learning_rate": 9.940000000000001e-06, | |
| "loss": 1.1709, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.321003963011889, | |
| "eval_bleu": 5.2707215480174545, | |
| "eval_char_accuracy": 18.391696825135714, | |
| "eval_loss": 0.8503363132476807, | |
| "eval_runtime": 326.5655, | |
| "eval_samples_per_second": 4.636, | |
| "eval_steps_per_second": 0.582, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.453104359313078, | |
| "grad_norm": 5.828360557556152, | |
| "learning_rate": 9.99996092907511e-06, | |
| "loss": 1.0774, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.5852047556142668, | |
| "grad_norm": 3.975123643875122, | |
| "learning_rate": 9.999833582267183e-06, | |
| "loss": 0.9803, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.7173051519154559, | |
| "grad_norm": 4.110162258148193, | |
| "learning_rate": 9.999617802644021e-06, | |
| "loss": 0.9023, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.8494055482166445, | |
| "grad_norm": 4.341949462890625, | |
| "learning_rate": 9.999313594022158e-06, | |
| "loss": 0.8494, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.9815059445178336, | |
| "grad_norm": 3.7582991123199463, | |
| "learning_rate": 9.99892096178217e-06, | |
| "loss": 0.7841, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.9815059445178336, | |
| "eval_bleu": 13.790865530076871, | |
| "eval_char_accuracy": 25.990602895213026, | |
| "eval_loss": 0.5515339374542236, | |
| "eval_runtime": 310.3546, | |
| "eval_samples_per_second": 4.878, | |
| "eval_steps_per_second": 0.612, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.1136063408190227, | |
| "grad_norm": 4.066399574279785, | |
| "learning_rate": 9.998439912868608e-06, | |
| "loss": 0.7379, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.2457067371202113, | |
| "grad_norm": 3.739553928375244, | |
| "learning_rate": 9.997870455789855e-06, | |
| "loss": 0.6859, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.3778071334214004, | |
| "grad_norm": 4.019917964935303, | |
| "learning_rate": 9.997212600617986e-06, | |
| "loss": 0.6547, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.509907529722589, | |
| "grad_norm": 3.1273276805877686, | |
| "learning_rate": 9.99646635898858e-06, | |
| "loss": 0.6313, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.642007926023778, | |
| "grad_norm": 3.0856070518493652, | |
| "learning_rate": 9.995631744100536e-06, | |
| "loss": 0.6058, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.642007926023778, | |
| "eval_bleu": 22.141127198903924, | |
| "eval_char_accuracy": 30.873190491857212, | |
| "eval_loss": 0.4170660674571991, | |
| "eval_runtime": 310.921, | |
| "eval_samples_per_second": 4.869, | |
| "eval_steps_per_second": 0.611, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.7741083223249667, | |
| "grad_norm": 3.660649299621582, | |
| "learning_rate": 9.994708770715807e-06, | |
| "loss": 0.5758, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.906208718626156, | |
| "grad_norm": 3.3534188270568848, | |
| "learning_rate": 9.993697455159165e-06, | |
| "loss": 0.5507, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.038309114927345, | |
| "grad_norm": 2.831392526626587, | |
| "learning_rate": 9.992597815317901e-06, | |
| "loss": 0.5334, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.1704095112285335, | |
| "grad_norm": 3.2069804668426514, | |
| "learning_rate": 9.991409870641512e-06, | |
| "loss": 0.508, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.3025099075297226, | |
| "grad_norm": 3.3302793502807617, | |
| "learning_rate": 9.990133642141359e-06, | |
| "loss": 0.4816, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.3025099075297226, | |
| "eval_bleu": 28.8217714543364, | |
| "eval_char_accuracy": 34.30457312057904, | |
| "eval_loss": 0.34086647629737854, | |
| "eval_runtime": 312.1625, | |
| "eval_samples_per_second": 4.85, | |
| "eval_steps_per_second": 0.609, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.4346103038309117, | |
| "grad_norm": 3.0527114868164062, | |
| "learning_rate": 9.988769152390284e-06, | |
| "loss": 0.4779, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.5667107001321003, | |
| "grad_norm": 2.557722568511963, | |
| "learning_rate": 9.987316425522226e-06, | |
| "loss": 0.4626, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.6988110964332894, | |
| "grad_norm": 2.993014097213745, | |
| "learning_rate": 9.985775487231788e-06, | |
| "loss": 0.4452, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.830911492734478, | |
| "grad_norm": 2.7321043014526367, | |
| "learning_rate": 9.984146364773777e-06, | |
| "loss": 0.4408, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.963011889035667, | |
| "grad_norm": 2.8790836334228516, | |
| "learning_rate": 9.982429086962729e-06, | |
| "loss": 0.4108, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.963011889035667, | |
| "eval_bleu": 33.52667574407562, | |
| "eval_char_accuracy": 37.42289027800625, | |
| "eval_loss": 0.29226553440093994, | |
| "eval_runtime": 316.4732, | |
| "eval_samples_per_second": 4.784, | |
| "eval_steps_per_second": 0.6, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.095112285336856, | |
| "grad_norm": 2.8862805366516113, | |
| "learning_rate": 9.980623684172396e-06, | |
| "loss": 0.4134, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.227212681638045, | |
| "grad_norm": 2.4621527194976807, | |
| "learning_rate": 9.978730188335215e-06, | |
| "loss": 0.3919, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.359313077939234, | |
| "grad_norm": 2.822957992553711, | |
| "learning_rate": 9.976748632941733e-06, | |
| "loss": 0.384, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.491413474240423, | |
| "grad_norm": 2.448110818862915, | |
| "learning_rate": 9.974679053040018e-06, | |
| "loss": 0.3735, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.623513870541611, | |
| "grad_norm": 2.2914109230041504, | |
| "learning_rate": 9.972521485235045e-06, | |
| "loss": 0.3604, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.623513870541611, | |
| "eval_bleu": 37.45556816331904, | |
| "eval_char_accuracy": 40.02354416844876, | |
| "eval_loss": 0.25823718309402466, | |
| "eval_runtime": 320.7507, | |
| "eval_samples_per_second": 4.72, | |
| "eval_steps_per_second": 0.592, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.755614266842801, | |
| "grad_norm": 2.5924477577209473, | |
| "learning_rate": 9.970275967688047e-06, | |
| "loss": 0.3624, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.887714663143989, | |
| "grad_norm": 2.49125075340271, | |
| "learning_rate": 9.967942540115829e-06, | |
| "loss": 0.3508, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 5.019815059445178, | |
| "grad_norm": 2.464569330215454, | |
| "learning_rate": 9.965521243790079e-06, | |
| "loss": 0.3355, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 5.1519154557463676, | |
| "grad_norm": 2.6026785373687744, | |
| "learning_rate": 9.963012121536635e-06, | |
| "loss": 0.3284, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 5.284015852047556, | |
| "grad_norm": 2.351313591003418, | |
| "learning_rate": 9.96041521773472e-06, | |
| "loss": 0.328, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.284015852047556, | |
| "eval_bleu": 40.737027833366824, | |
| "eval_char_accuracy": 42.49156933706202, | |
| "eval_loss": 0.23189863562583923, | |
| "eval_runtime": 320.7889, | |
| "eval_samples_per_second": 4.72, | |
| "eval_steps_per_second": 0.592, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.416116248348745, | |
| "grad_norm": 3.0100066661834717, | |
| "learning_rate": 9.95773057831617e-06, | |
| "loss": 0.311, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 5.5482166446499335, | |
| "grad_norm": 2.049722671508789, | |
| "learning_rate": 9.954958250764604e-06, | |
| "loss": 0.3136, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.680317040951123, | |
| "grad_norm": 2.1132702827453613, | |
| "learning_rate": 9.952098284114604e-06, | |
| "loss": 0.3, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 5.812417437252312, | |
| "grad_norm": 2.174574613571167, | |
| "learning_rate": 9.949150728950833e-06, | |
| "loss": 0.3093, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.9445178335535, | |
| "grad_norm": 2.8350446224212646, | |
| "learning_rate": 9.946115637407145e-06, | |
| "loss": 0.2988, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.9445178335535, | |
| "eval_bleu": 43.8559984358133, | |
| "eval_char_accuracy": 44.56273646981411, | |
| "eval_loss": 0.21132107079029083, | |
| "eval_runtime": 319.4845, | |
| "eval_samples_per_second": 4.739, | |
| "eval_steps_per_second": 0.595, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.07661822985469, | |
| "grad_norm": 2.289716958999634, | |
| "learning_rate": 9.94299306316567e-06, | |
| "loss": 0.2938, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 6.208718626155878, | |
| "grad_norm": 2.9307034015655518, | |
| "learning_rate": 9.939783061455845e-06, | |
| "loss": 0.2814, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 6.340819022457067, | |
| "grad_norm": 2.3431613445281982, | |
| "learning_rate": 9.936485689053462e-06, | |
| "loss": 0.2782, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 6.472919418758257, | |
| "grad_norm": 2.2339768409729004, | |
| "learning_rate": 9.933101004279647e-06, | |
| "loss": 0.2752, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 6.605019815059445, | |
| "grad_norm": 2.076145887374878, | |
| "learning_rate": 9.92962906699983e-06, | |
| "loss": 0.265, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.605019815059445, | |
| "eval_bleu": 47.09645003475602, | |
| "eval_char_accuracy": 46.29205050172725, | |
| "eval_loss": 0.1954895406961441, | |
| "eval_runtime": 322.9917, | |
| "eval_samples_per_second": 4.687, | |
| "eval_steps_per_second": 0.588, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.737120211360634, | |
| "grad_norm": 1.715720295906067, | |
| "learning_rate": 9.926069938622698e-06, | |
| "loss": 0.266, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 6.869220607661823, | |
| "grad_norm": 2.501234531402588, | |
| "learning_rate": 9.922423682099088e-06, | |
| "loss": 0.2633, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 7.001321003963012, | |
| "grad_norm": 1.929123044013977, | |
| "learning_rate": 9.918690361920898e-06, | |
| "loss": 0.2584, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 7.133421400264201, | |
| "grad_norm": 2.0370264053344727, | |
| "learning_rate": 9.914870044119924e-06, | |
| "loss": 0.2451, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 7.265521796565389, | |
| "grad_norm": 2.3562278747558594, | |
| "learning_rate": 9.91096279626671e-06, | |
| "loss": 0.2476, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.265521796565389, | |
| "eval_bleu": 48.90758910970442, | |
| "eval_char_accuracy": 46.93308932390195, | |
| "eval_loss": 0.18339309096336365, | |
| "eval_runtime": 328.302, | |
| "eval_samples_per_second": 4.612, | |
| "eval_steps_per_second": 0.579, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.397622192866579, | |
| "grad_norm": 2.410529613494873, | |
| "learning_rate": 9.90696868746934e-06, | |
| "loss": 0.2419, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 7.5297225891677675, | |
| "grad_norm": 1.685939908027649, | |
| "learning_rate": 9.902887788372223e-06, | |
| "loss": 0.2448, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 7.661822985468956, | |
| "grad_norm": 2.3180549144744873, | |
| "learning_rate": 9.89872017115484e-06, | |
| "loss": 0.2379, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 7.793923381770146, | |
| "grad_norm": 2.4159021377563477, | |
| "learning_rate": 9.894465909530471e-06, | |
| "loss": 0.2339, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 7.926023778071334, | |
| "grad_norm": 2.0757477283477783, | |
| "learning_rate": 9.890125078744884e-06, | |
| "loss": 0.2356, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.926023778071334, | |
| "eval_bleu": 51.273473767746786, | |
| "eval_char_accuracy": 49.088563086033886, | |
| "eval_loss": 0.1728673279285431, | |
| "eval_runtime": 317.183, | |
| "eval_samples_per_second": 4.773, | |
| "eval_steps_per_second": 0.599, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.058124174372523, | |
| "grad_norm": 1.9405860900878906, | |
| "learning_rate": 9.885697755575015e-06, | |
| "loss": 0.2251, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 8.190224570673712, | |
| "grad_norm": 1.7473342418670654, | |
| "learning_rate": 9.881184018327597e-06, | |
| "loss": 0.2195, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 8.3223249669749, | |
| "grad_norm": 1.7633724212646484, | |
| "learning_rate": 9.876583946837787e-06, | |
| "loss": 0.219, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 8.45442536327609, | |
| "grad_norm": 2.1117053031921387, | |
| "learning_rate": 9.871897622467748e-06, | |
| "loss": 0.2148, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 8.58652575957728, | |
| "grad_norm": 2.114854574203491, | |
| "learning_rate": 9.867125128105211e-06, | |
| "loss": 0.2222, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.58652575957728, | |
| "eval_bleu": 53.15738681385049, | |
| "eval_char_accuracy": 49.680251686132586, | |
| "eval_loss": 0.1632871925830841, | |
| "eval_runtime": 325.9103, | |
| "eval_samples_per_second": 4.645, | |
| "eval_steps_per_second": 0.583, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.718626155878468, | |
| "grad_norm": 2.6163322925567627, | |
| "learning_rate": 9.862266548162008e-06, | |
| "loss": 0.2141, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 8.850726552179657, | |
| "grad_norm": 2.1705501079559326, | |
| "learning_rate": 9.857321968572577e-06, | |
| "loss": 0.2126, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 8.982826948480845, | |
| "grad_norm": 2.4352259635925293, | |
| "learning_rate": 9.85229147679245e-06, | |
| "loss": 0.2124, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 9.114927344782034, | |
| "grad_norm": 1.912975788116455, | |
| "learning_rate": 9.847175161796696e-06, | |
| "loss": 0.2032, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 9.247027741083222, | |
| "grad_norm": 1.7575359344482422, | |
| "learning_rate": 9.841973114078358e-06, | |
| "loss": 0.2005, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.247027741083222, | |
| "eval_bleu": 54.81436388056976, | |
| "eval_char_accuracy": 50.20562592531667, | |
| "eval_loss": 0.1546466052532196, | |
| "eval_runtime": 319.5902, | |
| "eval_samples_per_second": 4.737, | |
| "eval_steps_per_second": 0.595, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.379128137384413, | |
| "grad_norm": 1.6994798183441162, | |
| "learning_rate": 9.836685425646842e-06, | |
| "loss": 0.1929, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 9.511228533685602, | |
| "grad_norm": 1.8375500440597534, | |
| "learning_rate": 9.831312190026295e-06, | |
| "loss": 0.1954, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 9.64332892998679, | |
| "grad_norm": 2.735320568084717, | |
| "learning_rate": 9.825853502253951e-06, | |
| "loss": 0.1949, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 9.775429326287979, | |
| "grad_norm": 1.9880143404006958, | |
| "learning_rate": 9.820309458878447e-06, | |
| "loss": 0.196, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 9.907529722589167, | |
| "grad_norm": 3.1160881519317627, | |
| "learning_rate": 9.814680157958122e-06, | |
| "loss": 0.1957, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.907529722589167, | |
| "eval_bleu": 56.684180257446236, | |
| "eval_char_accuracy": 51.796656522454356, | |
| "eval_loss": 0.14744216203689575, | |
| "eval_runtime": 318.7305, | |
| "eval_samples_per_second": 4.75, | |
| "eval_steps_per_second": 0.596, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 10.039630118890356, | |
| "grad_norm": 1.975994348526001, | |
| "learning_rate": 9.808965699059276e-06, | |
| "loss": 0.1964, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 10.171730515191545, | |
| "grad_norm": 1.6857510805130005, | |
| "learning_rate": 9.80316618325441e-06, | |
| "loss": 0.1832, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 10.303830911492735, | |
| "grad_norm": 1.6473827362060547, | |
| "learning_rate": 9.797281713120438e-06, | |
| "loss": 0.1846, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 10.435931307793924, | |
| "grad_norm": 2.1330363750457764, | |
| "learning_rate": 9.79131239273688e-06, | |
| "loss": 0.1783, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 10.568031704095112, | |
| "grad_norm": 2.0598771572113037, | |
| "learning_rate": 9.785258327684007e-06, | |
| "loss": 0.183, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.568031704095112, | |
| "eval_bleu": 57.92928420103779, | |
| "eval_char_accuracy": 51.580235236058556, | |
| "eval_loss": 0.1413801610469818, | |
| "eval_runtime": 318.3694, | |
| "eval_samples_per_second": 4.755, | |
| "eval_steps_per_second": 0.597, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.700132100396301, | |
| "grad_norm": 1.8669029474258423, | |
| "learning_rate": 9.779119625040988e-06, | |
| "loss": 0.1801, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 10.83223249669749, | |
| "grad_norm": 1.967623233795166, | |
| "learning_rate": 9.772896393383991e-06, | |
| "loss": 0.1772, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 10.964332892998678, | |
| "grad_norm": 2.0888118743896484, | |
| "learning_rate": 9.766588742784255e-06, | |
| "loss": 0.1741, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 11.096433289299869, | |
| "grad_norm": 1.8615264892578125, | |
| "learning_rate": 9.760196784806155e-06, | |
| "loss": 0.1733, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 11.228533685601057, | |
| "grad_norm": 1.786023736000061, | |
| "learning_rate": 9.753720632505219e-06, | |
| "loss": 0.171, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 11.228533685601057, | |
| "eval_bleu": 58.97794662773473, | |
| "eval_char_accuracy": 53.53111120250041, | |
| "eval_loss": 0.13721999526023865, | |
| "eval_runtime": 310.3442, | |
| "eval_samples_per_second": 4.878, | |
| "eval_steps_per_second": 0.612, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 11.360634081902246, | |
| "grad_norm": 2.22037935256958, | |
| "learning_rate": 9.74716040042614e-06, | |
| "loss": 0.169, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 11.492734478203435, | |
| "grad_norm": 2.2769389152526855, | |
| "learning_rate": 9.740516204600734e-06, | |
| "loss": 0.1631, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 11.624834874504623, | |
| "grad_norm": 2.1513566970825195, | |
| "learning_rate": 9.733788162545902e-06, | |
| "loss": 0.1669, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 11.756935270805812, | |
| "grad_norm": 1.640358328819275, | |
| "learning_rate": 9.726976393261547e-06, | |
| "loss": 0.1674, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 11.889035667107, | |
| "grad_norm": 1.6976934671401978, | |
| "learning_rate": 9.720081017228462e-06, | |
| "loss": 0.1646, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.889035667107, | |
| "eval_bleu": 60.38328284328657, | |
| "eval_char_accuracy": 54.70215084717881, | |
| "eval_loss": 0.13088105618953705, | |
| "eval_runtime": 316.5706, | |
| "eval_samples_per_second": 4.783, | |
| "eval_steps_per_second": 0.6, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.021136063408191, | |
| "grad_norm": 1.7420779466629028, | |
| "learning_rate": 9.713102156406213e-06, | |
| "loss": 0.1629, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 12.15323645970938, | |
| "grad_norm": 1.9723796844482422, | |
| "learning_rate": 9.706039934230967e-06, | |
| "loss": 0.1578, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 12.285336856010568, | |
| "grad_norm": 2.3517324924468994, | |
| "learning_rate": 9.698894475613323e-06, | |
| "loss": 0.1561, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 12.417437252311757, | |
| "grad_norm": 1.5132865905761719, | |
| "learning_rate": 9.691665906936088e-06, | |
| "loss": 0.157, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 12.549537648612946, | |
| "grad_norm": 2.435624599456787, | |
| "learning_rate": 9.684354356052055e-06, | |
| "loss": 0.1538, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 12.549537648612946, | |
| "eval_bleu": 61.08870144349316, | |
| "eval_char_accuracy": 55.10055107747986, | |
| "eval_loss": 0.12675440311431885, | |
| "eval_runtime": 314.3152, | |
| "eval_samples_per_second": 4.817, | |
| "eval_steps_per_second": 0.604, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 12.681638044914134, | |
| "grad_norm": 2.092256784439087, | |
| "learning_rate": 9.676959952281733e-06, | |
| "loss": 0.1518, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 12.813738441215325, | |
| "grad_norm": 1.7985179424285889, | |
| "learning_rate": 9.669482826411065e-06, | |
| "loss": 0.158, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 12.945838837516513, | |
| "grad_norm": 1.7571889162063599, | |
| "learning_rate": 9.66192311068911e-06, | |
| "loss": 0.152, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 13.077939233817702, | |
| "grad_norm": 1.3526843786239624, | |
| "learning_rate": 9.654280938825705e-06, | |
| "loss": 0.1426, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 13.21003963011889, | |
| "grad_norm": 1.6651784181594849, | |
| "learning_rate": 9.646556445989106e-06, | |
| "loss": 0.1476, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 13.21003963011889, | |
| "eval_bleu": 61.91608356246592, | |
| "eval_char_accuracy": 55.006477216647475, | |
| "eval_loss": 0.12389995902776718, | |
| "eval_runtime": 315.2745, | |
| "eval_samples_per_second": 4.802, | |
| "eval_steps_per_second": 0.603, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 13.34214002642008, | |
| "grad_norm": 1.5739635229110718, | |
| "learning_rate": 9.63874976880359e-06, | |
| "loss": 0.148, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 13.474240422721268, | |
| "grad_norm": 1.779477834701538, | |
| "learning_rate": 9.63086104534704e-06, | |
| "loss": 0.1469, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 13.606340819022456, | |
| "grad_norm": 1.5422449111938477, | |
| "learning_rate": 9.622890415148505e-06, | |
| "loss": 0.143, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 13.738441215323647, | |
| "grad_norm": 1.80446457862854, | |
| "learning_rate": 9.61483801918573e-06, | |
| "loss": 0.1424, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 13.870541611624835, | |
| "grad_norm": 1.7641818523406982, | |
| "learning_rate": 9.606703999882667e-06, | |
| "loss": 0.1406, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 13.870541611624835, | |
| "eval_bleu": 62.97532147351367, | |
| "eval_char_accuracy": 56.364636453364035, | |
| "eval_loss": 0.12048687040805817, | |
| "eval_runtime": 315.4459, | |
| "eval_samples_per_second": 4.8, | |
| "eval_steps_per_second": 0.602, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 14.002642007926024, | |
| "grad_norm": 1.5506337881088257, | |
| "learning_rate": 9.598488501106947e-06, | |
| "loss": 0.1436, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 14.134742404227213, | |
| "grad_norm": 1.338537335395813, | |
| "learning_rate": 9.590191668167343e-06, | |
| "loss": 0.1396, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 14.266842800528401, | |
| "grad_norm": 1.9432475566864014, | |
| "learning_rate": 9.581813647811199e-06, | |
| "loss": 0.1427, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 14.39894319682959, | |
| "grad_norm": 1.8222265243530273, | |
| "learning_rate": 9.573354588221833e-06, | |
| "loss": 0.1352, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 14.531043593130779, | |
| "grad_norm": 1.7141766548156738, | |
| "learning_rate": 9.564814639015915e-06, | |
| "loss": 0.1361, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 14.531043593130779, | |
| "eval_bleu": 63.45481969831487, | |
| "eval_char_accuracy": 56.61755634150354, | |
| "eval_loss": 0.11702162027359009, | |
| "eval_runtime": 318.0259, | |
| "eval_samples_per_second": 4.761, | |
| "eval_steps_per_second": 0.597, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 14.663143989431969, | |
| "grad_norm": 1.473866581916809, | |
| "learning_rate": 9.556193951240821e-06, | |
| "loss": 0.1302, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 14.795244385733158, | |
| "grad_norm": 1.3840774297714233, | |
| "learning_rate": 9.547492677371968e-06, | |
| "loss": 0.1355, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 14.927344782034346, | |
| "grad_norm": 1.8060030937194824, | |
| "learning_rate": 9.538710971310104e-06, | |
| "loss": 0.1332, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 15.059445178335535, | |
| "grad_norm": 1.8242031335830688, | |
| "learning_rate": 9.529848988378597e-06, | |
| "loss": 0.1247, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 15.191545574636724, | |
| "grad_norm": 1.3062950372695923, | |
| "learning_rate": 9.520906885320682e-06, | |
| "loss": 0.1295, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 15.191545574636724, | |
| "eval_bleu": 64.22910948855701, | |
| "eval_char_accuracy": 57.40099111696003, | |
| "eval_loss": 0.11363621801137924, | |
| "eval_runtime": 315.2122, | |
| "eval_samples_per_second": 4.803, | |
| "eval_steps_per_second": 0.603, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 15.323645970937912, | |
| "grad_norm": 1.5021114349365234, | |
| "learning_rate": 9.511884820296695e-06, | |
| "loss": 0.1292, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 15.455746367239101, | |
| "grad_norm": 1.8947999477386475, | |
| "learning_rate": 9.502782952881268e-06, | |
| "loss": 0.128, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 15.587846763540291, | |
| "grad_norm": 1.5116643905639648, | |
| "learning_rate": 9.493601444060514e-06, | |
| "loss": 0.1276, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 15.71994715984148, | |
| "grad_norm": 1.5134871006011963, | |
| "learning_rate": 9.48434045622917e-06, | |
| "loss": 0.132, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 15.852047556142669, | |
| "grad_norm": 1.7438267469406128, | |
| "learning_rate": 9.475000153187733e-06, | |
| "loss": 0.1243, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 15.852047556142669, | |
| "eval_bleu": 65.05289431113529, | |
| "eval_char_accuracy": 57.989081263365684, | |
| "eval_loss": 0.1115257665514946, | |
| "eval_runtime": 309.8437, | |
| "eval_samples_per_second": 4.886, | |
| "eval_steps_per_second": 0.613, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 15.984147952443857, | |
| "grad_norm": 1.913855791091919, | |
| "learning_rate": 9.46558070013956e-06, | |
| "loss": 0.1261, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 16.116248348745046, | |
| "grad_norm": 1.4257742166519165, | |
| "learning_rate": 9.456082263687946e-06, | |
| "loss": 0.117, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 16.248348745046236, | |
| "grad_norm": 1.49489164352417, | |
| "learning_rate": 9.44650501183318e-06, | |
| "loss": 0.1232, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 16.380449141347423, | |
| "grad_norm": 1.7977708578109741, | |
| "learning_rate": 9.436849113969567e-06, | |
| "loss": 0.1212, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 16.512549537648614, | |
| "grad_norm": 1.2978798151016235, | |
| "learning_rate": 9.427212472501483e-06, | |
| "loss": 0.122, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 16.512549537648614, | |
| "eval_bleu": 65.59844085659223, | |
| "eval_char_accuracy": 58.32784997532489, | |
| "eval_loss": 0.10947112739086151, | |
| "eval_runtime": 332.8226, | |
| "eval_samples_per_second": 4.549, | |
| "eval_steps_per_second": 0.571, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 16.6446499339498, | |
| "grad_norm": 1.5628472566604614, | |
| "learning_rate": 9.417400578537868e-06, | |
| "loss": 0.1219, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 16.77675033025099, | |
| "grad_norm": 1.5304769277572632, | |
| "learning_rate": 9.407510553339931e-06, | |
| "loss": 0.1192, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 16.90885072655218, | |
| "grad_norm": 1.6370787620544434, | |
| "learning_rate": 9.397542571834054e-06, | |
| "loss": 0.1181, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 17.040951122853368, | |
| "grad_norm": 1.5388678312301636, | |
| "learning_rate": 9.387496810325436e-06, | |
| "loss": 0.1137, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 17.17305151915456, | |
| "grad_norm": 1.8251888751983643, | |
| "learning_rate": 9.377373446494984e-06, | |
| "loss": 0.1122, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 17.17305151915456, | |
| "eval_bleu": 66.79852257149867, | |
| "eval_char_accuracy": 58.47692877117947, | |
| "eval_loss": 0.1069813147187233, | |
| "eval_runtime": 350.9901, | |
| "eval_samples_per_second": 4.314, | |
| "eval_steps_per_second": 0.541, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 17.305151915455745, | |
| "grad_norm": 1.6730031967163086, | |
| "learning_rate": 9.367172659396172e-06, | |
| "loss": 0.1123, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 17.437252311756936, | |
| "grad_norm": 1.2402830123901367, | |
| "learning_rate": 9.35689462945187e-06, | |
| "loss": 0.1125, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 17.569352708058123, | |
| "grad_norm": 1.9425466060638428, | |
| "learning_rate": 9.34653953845115e-06, | |
| "loss": 0.1109, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 17.701453104359313, | |
| "grad_norm": 3.105457067489624, | |
| "learning_rate": 9.33610756954608e-06, | |
| "loss": 0.1161, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 17.833553500660503, | |
| "grad_norm": 1.679442048072815, | |
| "learning_rate": 9.325598907248478e-06, | |
| "loss": 0.1131, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 17.833553500660503, | |
| "eval_bleu": 66.72469802657548, | |
| "eval_char_accuracy": 59.26910264846191, | |
| "eval_loss": 0.10426344722509384, | |
| "eval_runtime": 353.0152, | |
| "eval_samples_per_second": 4.289, | |
| "eval_steps_per_second": 0.538, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 17.96565389696169, | |
| "grad_norm": 1.1806349754333496, | |
| "learning_rate": 9.315013737426645e-06, | |
| "loss": 0.115, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 18.09775429326288, | |
| "grad_norm": 1.5437259674072266, | |
| "learning_rate": 9.304352247302091e-06, | |
| "loss": 0.1071, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 18.229854689564068, | |
| "grad_norm": 3.9584083557128906, | |
| "learning_rate": 9.293614625446205e-06, | |
| "loss": 0.11, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 18.361955085865258, | |
| "grad_norm": 1.4071292877197266, | |
| "learning_rate": 9.282801061776937e-06, | |
| "loss": 0.1093, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 18.494055482166445, | |
| "grad_norm": 1.599787950515747, | |
| "learning_rate": 9.271911747555425e-06, | |
| "loss": 0.1057, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 18.494055482166445, | |
| "eval_bleu": 67.26763967365643, | |
| "eval_char_accuracy": 58.966318473433134, | |
| "eval_loss": 0.10266197472810745, | |
| "eval_runtime": 318.6046, | |
| "eval_samples_per_second": 4.752, | |
| "eval_steps_per_second": 0.596, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 18.626155878467635, | |
| "grad_norm": 1.3243365287780762, | |
| "learning_rate": 9.260946875382624e-06, | |
| "loss": 0.1054, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 18.758256274768826, | |
| "grad_norm": 1.8453656435012817, | |
| "learning_rate": 9.249906639195894e-06, | |
| "loss": 0.1096, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 18.890356671070013, | |
| "grad_norm": 1.2308754920959473, | |
| "learning_rate": 9.238791234265565e-06, | |
| "loss": 0.1045, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 19.022457067371203, | |
| "grad_norm": 1.3015666007995605, | |
| "learning_rate": 9.22760085719149e-06, | |
| "loss": 0.1065, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 19.15455746367239, | |
| "grad_norm": 1.3568576574325562, | |
| "learning_rate": 9.21633570589957e-06, | |
| "loss": 0.1024, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 19.15455746367239, | |
| "eval_bleu": 67.65790078495372, | |
| "eval_char_accuracy": 59.99547622964303, | |
| "eval_loss": 0.10274580866098404, | |
| "eval_runtime": 317.6719, | |
| "eval_samples_per_second": 4.766, | |
| "eval_steps_per_second": 0.598, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 19.28665785997358, | |
| "grad_norm": 1.6260790824890137, | |
| "learning_rate": 9.204995979638241e-06, | |
| "loss": 0.1025, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 19.418758256274767, | |
| "grad_norm": 1.1852062940597534, | |
| "learning_rate": 9.193581878974964e-06, | |
| "loss": 0.101, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 19.550858652575958, | |
| "grad_norm": 2.5521767139434814, | |
| "learning_rate": 9.18209360579267e-06, | |
| "loss": 0.0999, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 19.682959048877148, | |
| "grad_norm": 1.0231155157089233, | |
| "learning_rate": 9.17053136328619e-06, | |
| "loss": 0.1055, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 19.815059445178335, | |
| "grad_norm": 1.3518396615982056, | |
| "learning_rate": 9.15889535595866e-06, | |
| "loss": 0.1014, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 19.815059445178335, | |
| "eval_bleu": 68.43528395628573, | |
| "eval_char_accuracy": 61.64870866918901, | |
| "eval_loss": 0.09970895200967789, | |
| "eval_runtime": 314.0763, | |
| "eval_samples_per_second": 4.82, | |
| "eval_steps_per_second": 0.605, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 19.947159841479525, | |
| "grad_norm": 1.2678903341293335, | |
| "learning_rate": 9.147185789617907e-06, | |
| "loss": 0.1005, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 20.079260237780712, | |
| "grad_norm": 1.2439242601394653, | |
| "learning_rate": 9.13540287137281e-06, | |
| "loss": 0.0989, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 20.211360634081903, | |
| "grad_norm": 1.3169798851013184, | |
| "learning_rate": 9.123546809629632e-06, | |
| "loss": 0.1006, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 20.34346103038309, | |
| "grad_norm": 1.3063526153564453, | |
| "learning_rate": 9.111617814088332e-06, | |
| "loss": 0.0966, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 20.47556142668428, | |
| "grad_norm": 1.8397212028503418, | |
| "learning_rate": 9.099616095738867e-06, | |
| "loss": 0.0965, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 20.47556142668428, | |
| "eval_bleu": 68.40618966106221, | |
| "eval_char_accuracy": 60.30751357131107, | |
| "eval_loss": 0.09965521842241287, | |
| "eval_runtime": 315.776, | |
| "eval_samples_per_second": 4.795, | |
| "eval_steps_per_second": 0.602, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 20.60766182298547, | |
| "grad_norm": 1.544118046760559, | |
| "learning_rate": 9.087541866857453e-06, | |
| "loss": 0.0954, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 20.739762219286657, | |
| "grad_norm": 1.5378237962722778, | |
| "learning_rate": 9.075395341002804e-06, | |
| "loss": 0.0975, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 20.871862615587848, | |
| "grad_norm": 1.0197510719299316, | |
| "learning_rate": 9.06317673301237e-06, | |
| "loss": 0.0964, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 21.003963011889034, | |
| "grad_norm": 0.9621543884277344, | |
| "learning_rate": 9.05088625899852e-06, | |
| "loss": 0.0925, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 21.136063408190225, | |
| "grad_norm": 1.356550931930542, | |
| "learning_rate": 9.038524136344736e-06, | |
| "loss": 0.0917, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 21.136063408190225, | |
| "eval_bleu": 69.05043210174456, | |
| "eval_char_accuracy": 60.879153643691396, | |
| "eval_loss": 0.0960288867354393, | |
| "eval_runtime": 317.6216, | |
| "eval_samples_per_second": 4.767, | |
| "eval_steps_per_second": 0.598, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 21.268163804491415, | |
| "grad_norm": 2.0014472007751465, | |
| "learning_rate": 9.026090583701755e-06, | |
| "loss": 0.0962, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 21.400264200792602, | |
| "grad_norm": 1.4762715101242065, | |
| "learning_rate": 9.013585820983713e-06, | |
| "loss": 0.0917, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 21.532364597093792, | |
| "grad_norm": 1.242245078086853, | |
| "learning_rate": 9.001010069364241e-06, | |
| "loss": 0.0907, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 21.66446499339498, | |
| "grad_norm": 1.9390721321105957, | |
| "learning_rate": 8.98836355127257e-06, | |
| "loss": 0.0918, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 21.79656538969617, | |
| "grad_norm": 1.070357322692871, | |
| "learning_rate": 8.975646490389581e-06, | |
| "loss": 0.0903, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 21.79656538969617, | |
| "eval_bleu": 69.68531857632678, | |
| "eval_char_accuracy": 62.48920463892087, | |
| "eval_loss": 0.09554192423820496, | |
| "eval_runtime": 335.0637, | |
| "eval_samples_per_second": 4.519, | |
| "eval_steps_per_second": 0.567, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 21.928665785997357, | |
| "grad_norm": 1.8990777730941772, | |
| "learning_rate": 8.962859111643862e-06, | |
| "loss": 0.0946, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 22.060766182298547, | |
| "grad_norm": 1.6244333982467651, | |
| "learning_rate": 8.950001641207719e-06, | |
| "loss": 0.0895, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 22.192866578599737, | |
| "grad_norm": 1.6511205434799194, | |
| "learning_rate": 8.937074306493187e-06, | |
| "loss": 0.0907, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 22.324966974900924, | |
| "grad_norm": 1.421342372894287, | |
| "learning_rate": 8.924077336147992e-06, | |
| "loss": 0.0864, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 22.457067371202115, | |
| "grad_norm": 1.572800874710083, | |
| "learning_rate": 8.911010960051522e-06, | |
| "loss": 0.088, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 22.457067371202115, | |
| "eval_bleu": 69.94746089002493, | |
| "eval_char_accuracy": 61.5572051324231, | |
| "eval_loss": 0.09443064033985138, | |
| "eval_runtime": 345.5738, | |
| "eval_samples_per_second": 4.381, | |
| "eval_steps_per_second": 0.55, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 22.5891677675033, | |
| "grad_norm": 1.2412383556365967, | |
| "learning_rate": 8.897875409310755e-06, | |
| "loss": 0.085, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 22.721268163804492, | |
| "grad_norm": 1.4429903030395508, | |
| "learning_rate": 8.884803301685314e-06, | |
| "loss": 0.0908, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 22.85336856010568, | |
| "grad_norm": 1.1933690309524536, | |
| "learning_rate": 8.871530785794356e-06, | |
| "loss": 0.092, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 22.98546895640687, | |
| "grad_norm": 1.399186611175537, | |
| "learning_rate": 8.85818979355093e-06, | |
| "loss": 0.0837, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 23.11756935270806, | |
| "grad_norm": 1.2939783334732056, | |
| "learning_rate": 8.844780560919194e-06, | |
| "loss": 0.0871, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 23.11756935270806, | |
| "eval_bleu": 70.43896205976631, | |
| "eval_char_accuracy": 61.825032900148045, | |
| "eval_loss": 0.0947960913181305, | |
| "eval_runtime": 324.1925, | |
| "eval_samples_per_second": 4.67, | |
| "eval_steps_per_second": 0.586, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 23.249669749009247, | |
| "grad_norm": 1.2098222970962524, | |
| "learning_rate": 8.831303325070279e-06, | |
| "loss": 0.0827, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 23.381770145310437, | |
| "grad_norm": 1.5045851469039917, | |
| "learning_rate": 8.8177583243781e-06, | |
| "loss": 0.0838, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 23.513870541611624, | |
| "grad_norm": 1.5295897722244263, | |
| "learning_rate": 8.80414579841514e-06, | |
| "loss": 0.0858, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 23.645970937912814, | |
| "grad_norm": 1.4860461950302124, | |
| "learning_rate": 8.790465987948212e-06, | |
| "loss": 0.0875, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 23.778071334214, | |
| "grad_norm": 1.4711731672286987, | |
| "learning_rate": 8.776719134934199e-06, | |
| "loss": 0.0828, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 23.778071334214, | |
| "eval_bleu": 70.5623767627479, | |
| "eval_char_accuracy": 62.505140648132915, | |
| "eval_loss": 0.0924154594540596, | |
| "eval_runtime": 342.2412, | |
| "eval_samples_per_second": 4.424, | |
| "eval_steps_per_second": 0.555, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 23.91017173051519, | |
| "grad_norm": 1.4447731971740723, | |
| "learning_rate": 8.762905482515775e-06, | |
| "loss": 0.0814, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 24.042272126816382, | |
| "grad_norm": 1.350907802581787, | |
| "learning_rate": 8.749025275017107e-06, | |
| "loss": 0.0806, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 24.17437252311757, | |
| "grad_norm": 1.7207551002502441, | |
| "learning_rate": 8.735078757939532e-06, | |
| "loss": 0.08, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 24.30647291941876, | |
| "grad_norm": 1.0851505994796753, | |
| "learning_rate": 8.721066177957213e-06, | |
| "loss": 0.0779, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 24.438573315719946, | |
| "grad_norm": 1.2182328701019287, | |
| "learning_rate": 8.70698778291278e-06, | |
| "loss": 0.0814, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 24.438573315719946, | |
| "eval_bleu": 70.99257164437572, | |
| "eval_char_accuracy": 62.22189093600922, | |
| "eval_loss": 0.09152651578187943, | |
| "eval_runtime": 323.7174, | |
| "eval_samples_per_second": 4.677, | |
| "eval_steps_per_second": 0.587, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 24.570673712021136, | |
| "grad_norm": 2.0363285541534424, | |
| "learning_rate": 8.69284382181294e-06, | |
| "loss": 0.0821, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 24.702774108322323, | |
| "grad_norm": 1.3864432573318481, | |
| "learning_rate": 8.67863454482408e-06, | |
| "loss": 0.0809, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 24.834874504623514, | |
| "grad_norm": 2.032351493835449, | |
| "learning_rate": 8.664360203267838e-06, | |
| "loss": 0.0819, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 24.966974900924704, | |
| "grad_norm": 1.2007182836532593, | |
| "learning_rate": 8.65002104961666e-06, | |
| "loss": 0.0819, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 25.09907529722589, | |
| "grad_norm": 1.167693853378296, | |
| "learning_rate": 8.635617337489331e-06, | |
| "loss": 0.0778, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 25.09907529722589, | |
| "eval_bleu": 70.99070971451881, | |
| "eval_char_accuracy": 63.15440450732028, | |
| "eval_loss": 0.09109245985746384, | |
| "eval_runtime": 331.1976, | |
| "eval_samples_per_second": 4.571, | |
| "eval_steps_per_second": 0.574, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 25.23117569352708, | |
| "grad_norm": 1.9939295053482056, | |
| "learning_rate": 8.621149321646495e-06, | |
| "loss": 0.076, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 25.36327608982827, | |
| "grad_norm": 1.148555874824524, | |
| "learning_rate": 8.60661725798614e-06, | |
| "loss": 0.078, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 25.49537648612946, | |
| "grad_norm": 1.159621238708496, | |
| "learning_rate": 8.592167677001219e-06, | |
| "loss": 0.0823, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 25.627476882430646, | |
| "grad_norm": 1.7136179208755493, | |
| "learning_rate": 8.57750892397125e-06, | |
| "loss": 0.0755, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 25.759577278731836, | |
| "grad_norm": 1.105714201927185, | |
| "learning_rate": 8.5627868949981e-06, | |
| "loss": 0.0756, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 25.759577278731836, | |
| "eval_bleu": 71.15813151741806, | |
| "eval_char_accuracy": 63.43405576575094, | |
| "eval_loss": 0.09030098468065262, | |
| "eval_runtime": 316.3492, | |
| "eval_samples_per_second": 4.786, | |
| "eval_steps_per_second": 0.601, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 25.891677675033026, | |
| "grad_norm": 1.6091099977493286, | |
| "learning_rate": 8.548001850472529e-06, | |
| "loss": 0.0778, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 26.023778071334213, | |
| "grad_norm": 1.1730881929397583, | |
| "learning_rate": 8.533154051899864e-06, | |
| "loss": 0.0787, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 26.155878467635404, | |
| "grad_norm": 1.2703460454940796, | |
| "learning_rate": 8.518243761895369e-06, | |
| "loss": 0.0711, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 26.28797886393659, | |
| "grad_norm": 1.3379662036895752, | |
| "learning_rate": 8.503271244179608e-06, | |
| "loss": 0.075, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 26.42007926023778, | |
| "grad_norm": 1.370871901512146, | |
| "learning_rate": 8.488236763573772e-06, | |
| "loss": 0.0717, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 26.42007926023778, | |
| "eval_bleu": 72.10017178272712, | |
| "eval_char_accuracy": 63.853018588583645, | |
| "eval_loss": 0.08871379494667053, | |
| "eval_runtime": 316.1038, | |
| "eval_samples_per_second": 4.79, | |
| "eval_steps_per_second": 0.601, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 26.552179656538968, | |
| "grad_norm": 1.3396387100219727, | |
| "learning_rate": 8.473140585995004e-06, | |
| "loss": 0.0726, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 26.68428005284016, | |
| "grad_norm": 1.1219794750213623, | |
| "learning_rate": 8.457982978451683e-06, | |
| "loss": 0.0754, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 26.81638044914135, | |
| "grad_norm": 1.0815324783325195, | |
| "learning_rate": 8.442764209038717e-06, | |
| "loss": 0.0745, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 26.948480845442536, | |
| "grad_norm": 1.4396206140518188, | |
| "learning_rate": 8.427484546932789e-06, | |
| "loss": 0.0749, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 27.080581241743726, | |
| "grad_norm": 1.2644987106323242, | |
| "learning_rate": 8.4121442623876e-06, | |
| "loss": 0.0731, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 27.080581241743726, | |
| "eval_bleu": 71.61514434619416, | |
| "eval_char_accuracy": 63.06290097055437, | |
| "eval_loss": 0.08903466165065765, | |
| "eval_runtime": 316.4287, | |
| "eval_samples_per_second": 4.785, | |
| "eval_steps_per_second": 0.6, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 27.212681638044913, | |
| "grad_norm": 1.3456913232803345, | |
| "learning_rate": 8.396743626729093e-06, | |
| "loss": 0.0728, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 27.344782034346103, | |
| "grad_norm": 1.1268248558044434, | |
| "learning_rate": 8.381282912350646e-06, | |
| "loss": 0.072, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 27.476882430647294, | |
| "grad_norm": 0.964856743812561, | |
| "learning_rate": 8.365762392708259e-06, | |
| "loss": 0.0711, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 27.60898282694848, | |
| "grad_norm": 1.2877197265625, | |
| "learning_rate": 8.350182342315719e-06, | |
| "loss": 0.0681, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 27.74108322324967, | |
| "grad_norm": 3.0796854496002197, | |
| "learning_rate": 8.334543036739743e-06, | |
| "loss": 0.0681, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 27.74108322324967, | |
| "eval_bleu": 72.44362762449254, | |
| "eval_char_accuracy": 64.56139990129955, | |
| "eval_loss": 0.08809462934732437, | |
| "eval_runtime": 319.538, | |
| "eval_samples_per_second": 4.738, | |
| "eval_steps_per_second": 0.595, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 27.873183619550858, | |
| "grad_norm": 1.445993185043335, | |
| "learning_rate": 8.3188447525951e-06, | |
| "loss": 0.0701, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 28.005284015852048, | |
| "grad_norm": 0.9414767622947693, | |
| "learning_rate": 8.303087767539723e-06, | |
| "loss": 0.0698, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 28.137384412153235, | |
| "grad_norm": 1.0644490718841553, | |
| "learning_rate": 8.28758923914531e-06, | |
| "loss": 0.0674, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 28.269484808454425, | |
| "grad_norm": 1.2708711624145508, | |
| "learning_rate": 8.27171684949204e-06, | |
| "loss": 0.0689, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 28.401585204755616, | |
| "grad_norm": 1.2177033424377441, | |
| "learning_rate": 8.25578659248641e-06, | |
| "loss": 0.0677, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 28.401585204755616, | |
| "eval_bleu": 72.7907604984095, | |
| "eval_char_accuracy": 64.78141964138838, | |
| "eval_loss": 0.0869230180978775, | |
| "eval_runtime": 315.3901, | |
| "eval_samples_per_second": 4.8, | |
| "eval_steps_per_second": 0.602, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 28.533685601056803, | |
| "grad_norm": 1.1179392337799072, | |
| "learning_rate": 8.239798749889293e-06, | |
| "loss": 0.0673, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 28.665785997357993, | |
| "grad_norm": 1.2472251653671265, | |
| "learning_rate": 8.223753604480086e-06, | |
| "loss": 0.0682, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 28.79788639365918, | |
| "grad_norm": 0.8336161375045776, | |
| "learning_rate": 8.207651440051714e-06, | |
| "loss": 0.0689, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 28.92998678996037, | |
| "grad_norm": 1.2652006149291992, | |
| "learning_rate": 8.1914925414056e-06, | |
| "loss": 0.0688, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 29.062087186261557, | |
| "grad_norm": 1.1424204111099243, | |
| "learning_rate": 8.175277194346636e-06, | |
| "loss": 0.0677, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 29.062087186261557, | |
| "eval_bleu": 72.84923297683062, | |
| "eval_char_accuracy": 64.63285491034709, | |
| "eval_loss": 0.08696427941322327, | |
| "eval_runtime": 316.7435, | |
| "eval_samples_per_second": 4.78, | |
| "eval_steps_per_second": 0.6, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 29.194187582562748, | |
| "grad_norm": 1.617885708808899, | |
| "learning_rate": 8.159005685678126e-06, | |
| "loss": 0.0638, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 29.326287978863938, | |
| "grad_norm": 1.2440978288650513, | |
| "learning_rate": 8.142678303196715e-06, | |
| "loss": 0.0606, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 29.458388375165125, | |
| "grad_norm": 1.1825751066207886, | |
| "learning_rate": 8.12629533568729e-06, | |
| "loss": 0.0661, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 29.590488771466315, | |
| "grad_norm": 1.5328364372253418, | |
| "learning_rate": 8.109857072917887e-06, | |
| "loss": 0.0647, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 29.722589167767502, | |
| "grad_norm": 1.1308438777923584, | |
| "learning_rate": 8.093363805634556e-06, | |
| "loss": 0.0666, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 29.722589167767502, | |
| "eval_bleu": 73.15661772633777, | |
| "eval_char_accuracy": 63.81086527389373, | |
| "eval_loss": 0.08648520708084106, | |
| "eval_runtime": 315.6708, | |
| "eval_samples_per_second": 4.796, | |
| "eval_steps_per_second": 0.602, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 29.854689564068693, | |
| "grad_norm": 1.0442135334014893, | |
| "learning_rate": 8.076815825556213e-06, | |
| "loss": 0.0648, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 29.98678996036988, | |
| "grad_norm": 1.363897681236267, | |
| "learning_rate": 8.060213425369492e-06, | |
| "loss": 0.0654, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 30.11889035667107, | |
| "grad_norm": 0.7751464247703552, | |
| "learning_rate": 8.043556898723568e-06, | |
| "loss": 0.0628, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 30.25099075297226, | |
| "grad_norm": 0.8685150742530823, | |
| "learning_rate": 8.026846540224956e-06, | |
| "loss": 0.0584, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 30.383091149273447, | |
| "grad_norm": 1.1484973430633545, | |
| "learning_rate": 8.0100826454323e-06, | |
| "loss": 0.0604, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 30.383091149273447, | |
| "eval_bleu": 73.25107285034876, | |
| "eval_char_accuracy": 65.01943164994243, | |
| "eval_loss": 0.08666232973337173, | |
| "eval_runtime": 323.7755, | |
| "eval_samples_per_second": 4.676, | |
| "eval_steps_per_second": 0.587, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 30.515191545574638, | |
| "grad_norm": 0.986289918422699, | |
| "learning_rate": 7.993265510851148e-06, | |
| "loss": 0.0688, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 30.647291941875825, | |
| "grad_norm": 1.0403423309326172, | |
| "learning_rate": 7.97639543392872e-06, | |
| "loss": 0.0638, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 30.779392338177015, | |
| "grad_norm": 1.517040729522705, | |
| "learning_rate": 7.959472713048617e-06, | |
| "loss": 0.0653, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 30.911492734478202, | |
| "grad_norm": 1.1347965002059937, | |
| "learning_rate": 7.942497647525576e-06, | |
| "loss": 0.0642, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 31.043593130779392, | |
| "grad_norm": 1.0789778232574463, | |
| "learning_rate": 7.925470537600155e-06, | |
| "loss": 0.0614, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 31.043593130779392, | |
| "eval_bleu": 73.23277401857816, | |
| "eval_char_accuracy": 65.0646693535121, | |
| "eval_loss": 0.08618722856044769, | |
| "eval_runtime": 313.4951, | |
| "eval_samples_per_second": 4.829, | |
| "eval_steps_per_second": 0.606, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 31.175693527080583, | |
| "grad_norm": 1.4853187799453735, | |
| "learning_rate": 7.908391684433432e-06, | |
| "loss": 0.0585, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 31.30779392338177, | |
| "grad_norm": 0.9656835794448853, | |
| "learning_rate": 7.891261390101675e-06, | |
| "loss": 0.0578, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 31.43989431968296, | |
| "grad_norm": 1.1521549224853516, | |
| "learning_rate": 7.874079957590997e-06, | |
| "loss": 0.0622, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 31.571994715984147, | |
| "grad_norm": 1.0636780261993408, | |
| "learning_rate": 7.856847690792002e-06, | |
| "loss": 0.0604, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 31.704095112285337, | |
| "grad_norm": 1.0833789110183716, | |
| "learning_rate": 7.839564894494409e-06, | |
| "loss": 0.0633, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 31.704095112285337, | |
| "eval_bleu": 73.62536575895216, | |
| "eval_char_accuracy": 65.15565882546471, | |
| "eval_loss": 0.08546082675457001, | |
| "eval_runtime": 318.0064, | |
| "eval_samples_per_second": 4.761, | |
| "eval_steps_per_second": 0.597, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 31.836195508586528, | |
| "grad_norm": 1.1620845794677734, | |
| "learning_rate": 7.822231874381658e-06, | |
| "loss": 0.0604, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 31.968295904887714, | |
| "grad_norm": 1.315012812614441, | |
| "learning_rate": 7.804848937025507e-06, | |
| "loss": 0.0593, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 32.1003963011889, | |
| "grad_norm": 0.8739562034606934, | |
| "learning_rate": 7.787416389880605e-06, | |
| "loss": 0.0608, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 32.23249669749009, | |
| "grad_norm": 0.9168538451194763, | |
| "learning_rate": 7.769934541279059e-06, | |
| "loss": 0.0577, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 32.36459709379128, | |
| "grad_norm": 0.9820032715797424, | |
| "learning_rate": 7.752403700424978e-06, | |
| "loss": 0.0569, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 32.36459709379128, | |
| "eval_bleu": 73.72149088930817, | |
| "eval_char_accuracy": 65.55457312057904, | |
| "eval_loss": 0.08627723157405853, | |
| "eval_runtime": 314.5801, | |
| "eval_samples_per_second": 4.813, | |
| "eval_steps_per_second": 0.604, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 32.49669749009247, | |
| "grad_norm": 1.0042686462402344, | |
| "learning_rate": 7.734824177389006e-06, | |
| "loss": 0.0582, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 32.628797886393656, | |
| "grad_norm": 1.251654863357544, | |
| "learning_rate": 7.71719628310283e-06, | |
| "loss": 0.0589, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 32.760898282694846, | |
| "grad_norm": 1.3150684833526611, | |
| "learning_rate": 7.699520329353694e-06, | |
| "loss": 0.0585, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 32.89299867899604, | |
| "grad_norm": 1.318556547164917, | |
| "learning_rate": 7.681796628778876e-06, | |
| "loss": 0.0588, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 33.02509907529723, | |
| "grad_norm": 1.2693874835968018, | |
| "learning_rate": 7.664025494860155e-06, | |
| "loss": 0.0605, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 33.02509907529723, | |
| "eval_bleu": 73.95631775899457, | |
| "eval_char_accuracy": 65.17056670505016, | |
| "eval_loss": 0.08447689563035965, | |
| "eval_runtime": 317.601, | |
| "eval_samples_per_second": 4.767, | |
| "eval_steps_per_second": 0.598, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 33.15719947159842, | |
| "grad_norm": 0.7866926193237305, | |
| "learning_rate": 7.646207241918272e-06, | |
| "loss": 0.055, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 33.2892998678996, | |
| "grad_norm": 1.0340533256530762, | |
| "learning_rate": 7.628342185107373e-06, | |
| "loss": 0.0563, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 33.42140026420079, | |
| "grad_norm": 1.6704190969467163, | |
| "learning_rate": 7.610430640409427e-06, | |
| "loss": 0.0568, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 33.55350066050198, | |
| "grad_norm": 1.4271676540374756, | |
| "learning_rate": 7.592472924628642e-06, | |
| "loss": 0.056, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 33.68560105680317, | |
| "grad_norm": 1.3886315822601318, | |
| "learning_rate": 7.574469355385865e-06, | |
| "loss": 0.0552, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 33.68560105680317, | |
| "eval_bleu": 73.67062952266826, | |
| "eval_char_accuracy": 65.04410676098043, | |
| "eval_loss": 0.08498267084360123, | |
| "eval_runtime": 314.7186, | |
| "eval_samples_per_second": 4.811, | |
| "eval_steps_per_second": 0.604, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 33.81770145310436, | |
| "grad_norm": 1.1037873029708862, | |
| "learning_rate": 7.556420251112956e-06, | |
| "loss": 0.0551, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 33.949801849405546, | |
| "grad_norm": 2.125624418258667, | |
| "learning_rate": 7.538325931047159e-06, | |
| "loss": 0.0591, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 34.081902245706736, | |
| "grad_norm": 1.674501895904541, | |
| "learning_rate": 7.52018671522546e-06, | |
| "loss": 0.0561, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 34.21400264200793, | |
| "grad_norm": 1.386206030845642, | |
| "learning_rate": 7.502002924478924e-06, | |
| "loss": 0.0509, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 34.34610303830912, | |
| "grad_norm": 1.0778214931488037, | |
| "learning_rate": 7.48377488042701e-06, | |
| "loss": 0.0544, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 34.34610303830912, | |
| "eval_bleu": 74.62046528623556, | |
| "eval_char_accuracy": 66.02442835992763, | |
| "eval_loss": 0.08464069664478302, | |
| "eval_runtime": 316.1374, | |
| "eval_samples_per_second": 4.789, | |
| "eval_steps_per_second": 0.601, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 34.4782034346103, | |
| "grad_norm": 0.8076276779174805, | |
| "learning_rate": 7.465502905471907e-06, | |
| "loss": 0.055, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 34.61030383091149, | |
| "grad_norm": 1.1508395671844482, | |
| "learning_rate": 7.447187322792806e-06, | |
| "loss": 0.057, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 34.74240422721268, | |
| "grad_norm": 1.2695698738098145, | |
| "learning_rate": 7.4288284563401945e-06, | |
| "loss": 0.055, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 34.87450462351387, | |
| "grad_norm": 1.166051983833313, | |
| "learning_rate": 7.410426630830131e-06, | |
| "loss": 0.0552, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 35.00660501981506, | |
| "grad_norm": 0.9517413973808289, | |
| "learning_rate": 7.391982171738496e-06, | |
| "loss": 0.0555, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 35.00660501981506, | |
| "eval_bleu": 74.18793581426324, | |
| "eval_char_accuracy": 65.63373910182597, | |
| "eval_loss": 0.08454510569572449, | |
| "eval_runtime": 313.1895, | |
| "eval_samples_per_second": 4.834, | |
| "eval_steps_per_second": 0.607, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 35.138705416116245, | |
| "grad_norm": 1.1265789270401, | |
| "learning_rate": 7.373495405295236e-06, | |
| "loss": 0.0529, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 35.270805812417436, | |
| "grad_norm": 1.0067466497421265, | |
| "learning_rate": 7.354966658478594e-06, | |
| "loss": 0.0502, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 35.402906208718626, | |
| "grad_norm": 0.9871610999107361, | |
| "learning_rate": 7.336396259009325e-06, | |
| "loss": 0.0508, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 35.53500660501982, | |
| "grad_norm": 1.4898390769958496, | |
| "learning_rate": 7.317784535344905e-06, | |
| "loss": 0.0544, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 35.66710700132101, | |
| "grad_norm": 1.1168763637542725, | |
| "learning_rate": 7.2991318166737126e-06, | |
| "loss": 0.0535, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 35.66710700132101, | |
| "eval_bleu": 74.4767569683976, | |
| "eval_char_accuracy": 65.44353512090805, | |
| "eval_loss": 0.08464961498975754, | |
| "eval_runtime": 313.4254, | |
| "eval_samples_per_second": 4.83, | |
| "eval_steps_per_second": 0.606, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 35.79920739762219, | |
| "grad_norm": 1.197704553604126, | |
| "learning_rate": 7.280625566954032e-06, | |
| "loss": 0.0547, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 35.93130779392338, | |
| "grad_norm": 1.6144860982894897, | |
| "learning_rate": 7.261892250434568e-06, | |
| "loss": 0.0516, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 36.06340819022457, | |
| "grad_norm": 1.1599304676055908, | |
| "learning_rate": 7.243118927483657e-06, | |
| "loss": 0.0502, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 36.19550858652576, | |
| "grad_norm": 1.043449878692627, | |
| "learning_rate": 7.22430593014791e-06, | |
| "loss": 0.0472, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 36.32760898282695, | |
| "grad_norm": 1.1489434242248535, | |
| "learning_rate": 7.205453591175666e-06, | |
| "loss": 0.0558, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 36.32760898282695, | |
| "eval_bleu": 74.20946250489693, | |
| "eval_char_accuracy": 65.89436996216483, | |
| "eval_loss": 0.08468983322381973, | |
| "eval_runtime": 314.7938, | |
| "eval_samples_per_second": 4.809, | |
| "eval_steps_per_second": 0.604, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 36.32760898282695, | |
| "step": 27500, | |
| "total_flos": 7035660725649408.0, | |
| "train_loss": 0.282735899699818, | |
| "train_runtime": 21727.9767, | |
| "train_samples_per_second": 55.735, | |
| "train_steps_per_second": 3.484 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 75700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 3 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7035660725649408.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |