| { | |
| "best_metric": 29.1144, | |
| "best_model_checkpoint": "./ko-en_mbartLarge_exp20p_linear_alpha/checkpoint-60000", | |
| "epoch": 8.81772827474185, | |
| "eval_steps": 4000, | |
| "global_step": 76000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.4920234365935726e-05, | |
| "loss": 2.3766, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 5.484046873187145e-05, | |
| "loss": 1.7994, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 5.476070309780717e-05, | |
| "loss": 1.6333, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 5.4680937463742896e-05, | |
| "loss": 1.554, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 5.460117182967862e-05, | |
| "loss": 1.5092, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 5.4521406195614344e-05, | |
| "loss": 1.4709, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5.444164056155007e-05, | |
| "loss": 1.4354, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 5.4361874927485784e-05, | |
| "loss": 1.404, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_bleu": 22.5375, | |
| "eval_gen_len": 18.6852, | |
| "eval_loss": 1.373841404914856, | |
| "eval_runtime": 1144.9051, | |
| "eval_samples_per_second": 15.055, | |
| "eval_steps_per_second": 0.942, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 5.4282109293421515e-05, | |
| "loss": 1.3753, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 5.420234365935724e-05, | |
| "loss": 1.3399, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 5.412257802529296e-05, | |
| "loss": 1.3272, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 5.404281239122868e-05, | |
| "loss": 1.3226, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 5.39630467571644e-05, | |
| "loss": 1.2978, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5.388328112310013e-05, | |
| "loss": 1.269, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 5.3803515489035856e-05, | |
| "loss": 1.2787, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 5.372374985497158e-05, | |
| "loss": 1.2629, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_bleu": 25.3741, | |
| "eval_gen_len": 18.7797, | |
| "eval_loss": 1.245756983757019, | |
| "eval_runtime": 1156.1784, | |
| "eval_samples_per_second": 14.909, | |
| "eval_steps_per_second": 0.932, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 5.36439842209073e-05, | |
| "loss": 1.2572, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 5.356421858684302e-05, | |
| "loss": 1.2035, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 5.348445295277875e-05, | |
| "loss": 1.1945, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 5.3404687318714474e-05, | |
| "loss": 1.179, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 5.332492168465019e-05, | |
| "loss": 1.2032, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 5.3245156050585915e-05, | |
| "loss": 1.192, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 5.316539041652164e-05, | |
| "loss": 1.1795, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 5.308562478245736e-05, | |
| "loss": 1.1951, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_bleu": 26.1281, | |
| "eval_gen_len": 18.6597, | |
| "eval_loss": 1.2066867351531982, | |
| "eval_runtime": 1158.2083, | |
| "eval_samples_per_second": 14.882, | |
| "eval_steps_per_second": 0.931, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 5.300585914839309e-05, | |
| "loss": 1.1704, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 5.292609351432881e-05, | |
| "loss": 1.1547, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 5.284632788026453e-05, | |
| "loss": 1.1419, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 5.276656224620026e-05, | |
| "loss": 1.1396, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 5.268679661213598e-05, | |
| "loss": 1.1549, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 5.2607030978071704e-05, | |
| "loss": 1.1255, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 5.252726534400743e-05, | |
| "loss": 1.1395, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 5.244749970994315e-05, | |
| "loss": 1.1317, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_bleu": 26.5384, | |
| "eval_gen_len": 19.2055, | |
| "eval_loss": 1.1767752170562744, | |
| "eval_runtime": 1178.3467, | |
| "eval_samples_per_second": 14.628, | |
| "eval_steps_per_second": 0.915, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 5.2367734075878875e-05, | |
| "loss": 1.1557, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 5.22879684418146e-05, | |
| "loss": 1.1478, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 5.2208202807750315e-05, | |
| "loss": 1.0771, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 5.2128437173686046e-05, | |
| "loss": 1.0135, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 5.204867153962177e-05, | |
| "loss": 1.0028, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 5.196890590555749e-05, | |
| "loss": 0.9894, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 5.1889140271493216e-05, | |
| "loss": 1.0178, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 5.180937463742893e-05, | |
| "loss": 0.9906, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "eval_bleu": 28.2459, | |
| "eval_gen_len": 18.7269, | |
| "eval_loss": 1.1362603902816772, | |
| "eval_runtime": 1157.7765, | |
| "eval_samples_per_second": 14.888, | |
| "eval_steps_per_second": 0.931, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 5.1729609003364664e-05, | |
| "loss": 1.0083, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 5.164984336930039e-05, | |
| "loss": 0.9965, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 5.157007773523611e-05, | |
| "loss": 0.9992, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 5.149031210117183e-05, | |
| "loss": 1.0042, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 5.141054646710755e-05, | |
| "loss": 0.9998, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 5.1330780833043275e-05, | |
| "loss": 1.0076, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 5.1251015198979005e-05, | |
| "loss": 1.0047, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 5.117124956491473e-05, | |
| "loss": 0.9894, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "eval_bleu": 28.5124, | |
| "eval_gen_len": 18.6882, | |
| "eval_loss": 1.1238752603530884, | |
| "eval_runtime": 1152.8912, | |
| "eval_samples_per_second": 14.951, | |
| "eval_steps_per_second": 0.935, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 5.1091483930850446e-05, | |
| "loss": 1.0115, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 5.101171829678617e-05, | |
| "loss": 0.9958, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 5.093195266272189e-05, | |
| "loss": 1.0044, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 5.0852187028657624e-05, | |
| "loss": 0.9668, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 5.077242139459334e-05, | |
| "loss": 0.8751, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 5.0692655760529064e-05, | |
| "loss": 0.8916, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 5.061289012646479e-05, | |
| "loss": 0.8861, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 5.053312449240051e-05, | |
| "loss": 0.8965, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_bleu": 28.5335, | |
| "eval_gen_len": 18.4917, | |
| "eval_loss": 1.1277672052383423, | |
| "eval_runtime": 1142.9887, | |
| "eval_samples_per_second": 15.081, | |
| "eval_steps_per_second": 0.943, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 5.045335885833624e-05, | |
| "loss": 0.8898, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 5.037359322427196e-05, | |
| "loss": 0.8982, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 5.029382759020768e-05, | |
| "loss": 0.896, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 5.0214061956143406e-05, | |
| "loss": 0.8889, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 5.013429632207913e-05, | |
| "loss": 0.9056, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 5.005453068801485e-05, | |
| "loss": 0.8867, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 4.997476505395058e-05, | |
| "loss": 0.911, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 4.98949994198863e-05, | |
| "loss": 0.9138, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "eval_bleu": 28.8189, | |
| "eval_gen_len": 18.7873, | |
| "eval_loss": 1.1216284036636353, | |
| "eval_runtime": 1156.9444, | |
| "eval_samples_per_second": 14.899, | |
| "eval_steps_per_second": 0.932, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 4.9815233785822024e-05, | |
| "loss": 0.9023, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 4.973546815175775e-05, | |
| "loss": 0.9127, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 4.9655702517693464e-05, | |
| "loss": 0.9168, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 4.9575936883629195e-05, | |
| "loss": 0.8978, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.949617124956492e-05, | |
| "loss": 0.9229, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 4.941640561550064e-05, | |
| "loss": 0.828, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 4.9336639981436366e-05, | |
| "loss": 0.8326, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 4.925687434737208e-05, | |
| "loss": 0.8272, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "eval_bleu": 28.332, | |
| "eval_gen_len": 18.6516, | |
| "eval_loss": 1.1468099355697632, | |
| "eval_runtime": 1152.7736, | |
| "eval_samples_per_second": 14.953, | |
| "eval_steps_per_second": 0.935, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 4.9177108713307806e-05, | |
| "loss": 0.8259, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 4.9097343079243536e-05, | |
| "loss": 0.8243, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 4.901757744517926e-05, | |
| "loss": 0.8664, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 4.893781181111498e-05, | |
| "loss": 0.8893, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 4.88580461770507e-05, | |
| "loss": 0.8958, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 4.8778280542986424e-05, | |
| "loss": 0.8924, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 4.8698514908922155e-05, | |
| "loss": 0.885, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 4.861874927485788e-05, | |
| "loss": 0.8753, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "eval_bleu": 28.2695, | |
| "eval_gen_len": 18.4919, | |
| "eval_loss": 1.1344704627990723, | |
| "eval_runtime": 1148.1227, | |
| "eval_samples_per_second": 15.013, | |
| "eval_steps_per_second": 0.939, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 4.8538983640793595e-05, | |
| "loss": 0.8749, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 4.845921800672932e-05, | |
| "loss": 0.8621, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 4.837945237266504e-05, | |
| "loss": 0.8616, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 4.8299686738600766e-05, | |
| "loss": 0.8551, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 4.821992110453649e-05, | |
| "loss": 0.8504, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 4.814015547047221e-05, | |
| "loss": 0.8459, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 4.806038983640794e-05, | |
| "loss": 0.7255, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 4.798062420234366e-05, | |
| "loss": 0.6855, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "eval_bleu": 28.7913, | |
| "eval_gen_len": 18.7596, | |
| "eval_loss": 1.154221773147583, | |
| "eval_runtime": 1168.8964, | |
| "eval_samples_per_second": 14.746, | |
| "eval_steps_per_second": 0.922, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.7900858568279384e-05, | |
| "loss": 0.7002, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 4.782109293421511e-05, | |
| "loss": 0.6982, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 4.774132730015083e-05, | |
| "loss": 0.6976, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 4.7661561666086555e-05, | |
| "loss": 0.7028, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 4.758179603202228e-05, | |
| "loss": 0.7138, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 4.7502030397958e-05, | |
| "loss": 0.7121, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 4.7422264763893726e-05, | |
| "loss": 0.7043, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 4.734249912982945e-05, | |
| "loss": 0.7088, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "eval_bleu": 29.0865, | |
| "eval_gen_len": 18.6626, | |
| "eval_loss": 1.153067708015442, | |
| "eval_runtime": 1149.2413, | |
| "eval_samples_per_second": 14.999, | |
| "eval_steps_per_second": 0.938, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 5.63, | |
| "learning_rate": 4.726273349576517e-05, | |
| "loss": 0.7116, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 4.71829678617009e-05, | |
| "loss": 0.7292, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 4.710320222763662e-05, | |
| "loss": 0.7289, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 4.702343659357234e-05, | |
| "loss": 0.7514, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 4.694367095950807e-05, | |
| "loss": 0.7545, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 4.686390532544379e-05, | |
| "loss": 0.7362, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 4.6784139691379515e-05, | |
| "loss": 0.7413, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 4.670437405731523e-05, | |
| "loss": 0.6738, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "eval_bleu": 28.0235, | |
| "eval_gen_len": 18.4243, | |
| "eval_loss": 1.190636396408081, | |
| "eval_runtime": 1139.2873, | |
| "eval_samples_per_second": 15.13, | |
| "eval_steps_per_second": 0.946, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 4.6624608423250955e-05, | |
| "loss": 0.6467, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 4.6544842789186686e-05, | |
| "loss": 0.6508, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 4.646507715512241e-05, | |
| "loss": 0.6742, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "learning_rate": 4.638531152105813e-05, | |
| "loss": 0.6609, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 4.630554588699385e-05, | |
| "loss": 0.6464, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 4.622578025292957e-05, | |
| "loss": 0.6362, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 4.61460146188653e-05, | |
| "loss": 0.6552, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 4.606624898480103e-05, | |
| "loss": 0.6763, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "eval_bleu": 28.1501, | |
| "eval_gen_len": 18.6932, | |
| "eval_loss": 1.1940782070159912, | |
| "eval_runtime": 1154.4139, | |
| "eval_samples_per_second": 14.931, | |
| "eval_steps_per_second": 0.934, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 4.5986483350736744e-05, | |
| "loss": 0.6775, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 4.590671771667247e-05, | |
| "loss": 0.6765, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 4.582695208260819e-05, | |
| "loss": 0.681, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 4.5747186448543915e-05, | |
| "loss": 0.6747, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 4.5667420814479645e-05, | |
| "loss": 0.6612, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 4.558765518041536e-05, | |
| "loss": 0.6662, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 4.5507889546351086e-05, | |
| "loss": 0.6653, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 4.542812391228681e-05, | |
| "loss": 0.6594, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_bleu": 29.1144, | |
| "eval_gen_len": 18.5459, | |
| "eval_loss": 1.1682050228118896, | |
| "eval_runtime": 1139.7821, | |
| "eval_samples_per_second": 15.123, | |
| "eval_steps_per_second": 0.946, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 4.534835827822253e-05, | |
| "loss": 0.6333, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 4.526859264415826e-05, | |
| "loss": 0.5491, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 4.518882701009398e-05, | |
| "loss": 0.5476, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 4.5109061376029704e-05, | |
| "loss": 0.5469, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 4.502929574196543e-05, | |
| "loss": 0.5708, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 4.494953010790115e-05, | |
| "loss": 0.5785, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 4.486976447383687e-05, | |
| "loss": 0.5803, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 4.47899988397726e-05, | |
| "loss": 0.5971, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "eval_bleu": 27.9464, | |
| "eval_gen_len": 18.4482, | |
| "eval_loss": 1.2448896169662476, | |
| "eval_runtime": 1136.4568, | |
| "eval_samples_per_second": 15.167, | |
| "eval_steps_per_second": 0.949, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 4.471023320570832e-05, | |
| "loss": 0.5985, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 4.4630467571644046e-05, | |
| "loss": 0.6032, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 4.455070193757977e-05, | |
| "loss": 0.6079, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 4.4470936303515486e-05, | |
| "loss": 0.6104, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 4.439117066945122e-05, | |
| "loss": 0.6063, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "learning_rate": 4.431140503538694e-05, | |
| "loss": 0.607, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 4.4231639401322664e-05, | |
| "loss": 0.5891, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 4.415187376725838e-05, | |
| "loss": 0.5935, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "eval_bleu": 28.6034, | |
| "eval_gen_len": 18.5967, | |
| "eval_loss": 1.2156028747558594, | |
| "eval_runtime": 1146.9398, | |
| "eval_samples_per_second": 15.029, | |
| "eval_steps_per_second": 0.94, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 4.4072108133194104e-05, | |
| "loss": 0.5981, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 4.399234249912983e-05, | |
| "loss": 0.5881, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 4.391257686506556e-05, | |
| "loss": 0.5013, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 4.383281123100128e-05, | |
| "loss": 0.5109, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 4.3753045596937e-05, | |
| "loss": 0.5187, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "learning_rate": 4.367327996287272e-05, | |
| "loss": 0.5198, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 8.3, | |
| "learning_rate": 4.3593514328808446e-05, | |
| "loss": 0.5175, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 4.3513748694744177e-05, | |
| "loss": 0.5383, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "eval_bleu": 27.891, | |
| "eval_gen_len": 18.6539, | |
| "eval_loss": 1.2927179336547852, | |
| "eval_runtime": 1152.5208, | |
| "eval_samples_per_second": 14.956, | |
| "eval_steps_per_second": 0.935, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 4.343398306067989e-05, | |
| "loss": 0.5285, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 4.335421742661562e-05, | |
| "loss": 0.5346, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "learning_rate": 4.327445179255134e-05, | |
| "loss": 0.5601, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 8.59, | |
| "learning_rate": 4.3194686158487064e-05, | |
| "loss": 0.5869, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "learning_rate": 4.311492052442279e-05, | |
| "loss": 0.5873, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 4.303515489035851e-05, | |
| "loss": 0.5924, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 4.2955389256294235e-05, | |
| "loss": 0.601, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 4.287562362222996e-05, | |
| "loss": 0.6022, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "eval_bleu": 27.7624, | |
| "eval_gen_len": 18.5558, | |
| "eval_loss": 1.283076286315918, | |
| "eval_runtime": 1144.2558, | |
| "eval_samples_per_second": 15.064, | |
| "eval_steps_per_second": 0.942, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "step": 76000, | |
| "total_flos": 2.6353338197706015e+18, | |
| "train_loss": 0.8847039678473222, | |
| "train_runtime": 77092.0129, | |
| "train_samples_per_second": 71.551, | |
| "train_steps_per_second": 4.472 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 344760, | |
| "num_train_epochs": 40, | |
| "save_steps": 4000, | |
| "total_flos": 2.6353338197706015e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |