| { | |
| "best_metric": 0.8987509608268738, | |
| "best_model_checkpoint": "./mbartLarge_koja_37p_exp2/checkpoint-31250", | |
| "epoch": 3.1927074158886737, | |
| "eval_steps": 1250, | |
| "global_step": 36250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.995587196987527e-05, | |
| "loss": 3.3474, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.9808778536126146e-05, | |
| "loss": 2.0622, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_bleu": 1.2834, | |
| "eval_gen_len": 17.8009, | |
| "eval_loss": 1.6679450273513794, | |
| "eval_runtime": 1488.7514, | |
| "eval_samples_per_second": 15.252, | |
| "eval_steps_per_second": 0.954, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.966168510237703e-05, | |
| "loss": 1.7123, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.951459166862791e-05, | |
| "loss": 1.59, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.936749823487879e-05, | |
| "loss": 1.5139, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_bleu": 2.0427, | |
| "eval_gen_len": 17.8496, | |
| "eval_loss": 1.4377689361572266, | |
| "eval_runtime": 1494.0047, | |
| "eval_samples_per_second": 15.199, | |
| "eval_steps_per_second": 0.95, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.922040480112968e-05, | |
| "loss": 1.4565, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.9073311367380564e-05, | |
| "loss": 1.4121, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_bleu": 2.7599, | |
| "eval_gen_len": 17.7667, | |
| "eval_loss": 1.3115850687026978, | |
| "eval_runtime": 1502.2991, | |
| "eval_samples_per_second": 15.115, | |
| "eval_steps_per_second": 0.945, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.892621793363145e-05, | |
| "loss": 1.3511, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.877912449988233e-05, | |
| "loss": 1.3052, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.863203106613321e-05, | |
| "loss": 1.2879, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_bleu": 3.1444, | |
| "eval_gen_len": 17.8887, | |
| "eval_loss": 1.23811936378479, | |
| "eval_runtime": 1519.5015, | |
| "eval_samples_per_second": 14.944, | |
| "eval_steps_per_second": 0.935, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.8484937632384094e-05, | |
| "loss": 1.2634, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.8337844198634976e-05, | |
| "loss": 1.2344, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_bleu": 3.3835, | |
| "eval_gen_len": 17.8323, | |
| "eval_loss": 1.1769133806228638, | |
| "eval_runtime": 1506.2285, | |
| "eval_samples_per_second": 15.075, | |
| "eval_steps_per_second": 0.943, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.819075076488586e-05, | |
| "loss": 1.2213, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.804365733113674e-05, | |
| "loss": 1.1918, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.789656389738762e-05, | |
| "loss": 1.1778, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_bleu": 3.9511, | |
| "eval_gen_len": 17.4892, | |
| "eval_loss": 1.1381820440292358, | |
| "eval_runtime": 1485.4261, | |
| "eval_samples_per_second": 15.287, | |
| "eval_steps_per_second": 0.956, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.7749470463638505e-05, | |
| "loss": 1.1651, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.760237702988939e-05, | |
| "loss": 1.1461, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_bleu": 3.9402, | |
| "eval_gen_len": 18.0136, | |
| "eval_loss": 1.09383225440979, | |
| "eval_runtime": 1503.0162, | |
| "eval_samples_per_second": 15.108, | |
| "eval_steps_per_second": 0.945, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.745528359614027e-05, | |
| "loss": 1.1341, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.730819016239115e-05, | |
| "loss": 1.1294, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.7161096728642035e-05, | |
| "loss": 1.1151, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_bleu": 4.2134, | |
| "eval_gen_len": 18.0537, | |
| "eval_loss": 1.074916124343872, | |
| "eval_runtime": 1564.4468, | |
| "eval_samples_per_second": 14.514, | |
| "eval_steps_per_second": 0.908, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.701400329489292e-05, | |
| "loss": 1.1102, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 4.68669098611438e-05, | |
| "loss": 1.093, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_bleu": 3.9587, | |
| "eval_gen_len": 17.8715, | |
| "eval_loss": 1.0418165922164917, | |
| "eval_runtime": 1534.6245, | |
| "eval_samples_per_second": 14.796, | |
| "eval_steps_per_second": 0.925, | |
| "step": 11250 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 4.671981642739469e-05, | |
| "loss": 1.084, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 4.657272299364557e-05, | |
| "loss": 1.0641, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 4.6425629559896446e-05, | |
| "loss": 1.0626, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_bleu": 4.6251, | |
| "eval_gen_len": 17.9406, | |
| "eval_loss": 1.0315195322036743, | |
| "eval_runtime": 1499.6187, | |
| "eval_samples_per_second": 15.142, | |
| "eval_steps_per_second": 0.947, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 4.627853612614733e-05, | |
| "loss": 1.0332, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.613144269239821e-05, | |
| "loss": 1.0192, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "eval_bleu": 4.9573, | |
| "eval_gen_len": 18.1266, | |
| "eval_loss": 1.0131869316101074, | |
| "eval_runtime": 1514.9755, | |
| "eval_samples_per_second": 14.988, | |
| "eval_steps_per_second": 0.937, | |
| "step": 13750 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.598434925864909e-05, | |
| "loss": 1.0234, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.5837255824899976e-05, | |
| "loss": 1.0062, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 4.569016239115086e-05, | |
| "loss": 0.9957, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_bleu": 4.3068, | |
| "eval_gen_len": 18.0925, | |
| "eval_loss": 0.9988749623298645, | |
| "eval_runtime": 1565.2457, | |
| "eval_samples_per_second": 14.507, | |
| "eval_steps_per_second": 0.907, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 4.554306895740174e-05, | |
| "loss": 0.9843, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 4.539597552365263e-05, | |
| "loss": 0.9778, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_bleu": 5.0517, | |
| "eval_gen_len": 17.8783, | |
| "eval_loss": 0.984965980052948, | |
| "eval_runtime": 1508.5724, | |
| "eval_samples_per_second": 15.052, | |
| "eval_steps_per_second": 0.941, | |
| "step": 16250 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.524888208990351e-05, | |
| "loss": 0.961, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.5101788656154394e-05, | |
| "loss": 0.9735, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 4.4954695222405276e-05, | |
| "loss": 0.9446, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_bleu": 5.0194, | |
| "eval_gen_len": 17.9348, | |
| "eval_loss": 0.9747512936592102, | |
| "eval_runtime": 1574.1886, | |
| "eval_samples_per_second": 14.425, | |
| "eval_steps_per_second": 0.902, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 4.480760178865616e-05, | |
| "loss": 0.9352, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 4.466050835490704e-05, | |
| "loss": 0.9236, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_bleu": 4.6011, | |
| "eval_gen_len": 17.7926, | |
| "eval_loss": 0.9619229435920715, | |
| "eval_runtime": 1510.4722, | |
| "eval_samples_per_second": 15.033, | |
| "eval_steps_per_second": 0.94, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 4.4513414921157917e-05, | |
| "loss": 0.9289, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 4.43663214874088e-05, | |
| "loss": 0.9216, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 4.421922805365969e-05, | |
| "loss": 0.9091, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_bleu": 4.6035, | |
| "eval_gen_len": 17.9399, | |
| "eval_loss": 0.9564162492752075, | |
| "eval_runtime": 1530.1826, | |
| "eval_samples_per_second": 14.839, | |
| "eval_steps_per_second": 0.928, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 4.407213461991057e-05, | |
| "loss": 0.9333, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 4.392504118616145e-05, | |
| "loss": 0.9072, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_bleu": 4.8313, | |
| "eval_gen_len": 17.6221, | |
| "eval_loss": 0.9533364176750183, | |
| "eval_runtime": 1475.3262, | |
| "eval_samples_per_second": 15.391, | |
| "eval_steps_per_second": 0.962, | |
| "step": 21250 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 4.3777947752412335e-05, | |
| "loss": 0.8896, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 4.363085431866322e-05, | |
| "loss": 0.8945, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 4.34837608849141e-05, | |
| "loss": 0.8758, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_bleu": 5.2707, | |
| "eval_gen_len": 17.5851, | |
| "eval_loss": 0.9421375393867493, | |
| "eval_runtime": 1516.2122, | |
| "eval_samples_per_second": 14.976, | |
| "eval_steps_per_second": 0.937, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 4.333666745116498e-05, | |
| "loss": 0.8723, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 4.3189574017415864e-05, | |
| "loss": 0.8539, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "eval_bleu": 5.2661, | |
| "eval_gen_len": 17.821, | |
| "eval_loss": 0.9304332137107849, | |
| "eval_runtime": 1492.9401, | |
| "eval_samples_per_second": 15.21, | |
| "eval_steps_per_second": 0.951, | |
| "step": 23750 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 4.3042480583666747e-05, | |
| "loss": 0.8461, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 4.289538714991763e-05, | |
| "loss": 0.842, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 4.274829371616851e-05, | |
| "loss": 0.8575, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_bleu": 4.9143, | |
| "eval_gen_len": 17.8879, | |
| "eval_loss": 0.9329252243041992, | |
| "eval_runtime": 1560.9154, | |
| "eval_samples_per_second": 14.547, | |
| "eval_steps_per_second": 0.91, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 4.2601200282419394e-05, | |
| "loss": 0.8424, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 4.2454106848670276e-05, | |
| "loss": 0.8314, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "eval_bleu": 5.106, | |
| "eval_gen_len": 18.0037, | |
| "eval_loss": 0.9262479543685913, | |
| "eval_runtime": 1520.8802, | |
| "eval_samples_per_second": 14.93, | |
| "eval_steps_per_second": 0.934, | |
| "step": 26250 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 4.230701341492116e-05, | |
| "loss": 0.8172, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 4.215991998117204e-05, | |
| "loss": 0.8219, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 4.201282654742292e-05, | |
| "loss": 0.8248, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "eval_bleu": 5.3073, | |
| "eval_gen_len": 17.6632, | |
| "eval_loss": 0.9241103529930115, | |
| "eval_runtime": 1501.7783, | |
| "eval_samples_per_second": 15.12, | |
| "eval_steps_per_second": 0.946, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 4.1865733113673805e-05, | |
| "loss": 0.8046, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 4.1718639679924694e-05, | |
| "loss": 0.8151, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_bleu": 5.5675, | |
| "eval_gen_len": 17.7676, | |
| "eval_loss": 0.9301652312278748, | |
| "eval_runtime": 1487.7016, | |
| "eval_samples_per_second": 15.263, | |
| "eval_steps_per_second": 0.954, | |
| "step": 28750 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.157154624617558e-05, | |
| "loss": 0.8019, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.142445281242646e-05, | |
| "loss": 0.8423, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 4.127735937867734e-05, | |
| "loss": 0.8093, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "eval_bleu": 6.2644, | |
| "eval_gen_len": 17.8475, | |
| "eval_loss": 0.9149118065834045, | |
| "eval_runtime": 1510.1876, | |
| "eval_samples_per_second": 15.036, | |
| "eval_steps_per_second": 0.94, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 4.113026594492822e-05, | |
| "loss": 0.7833, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 4.09831725111791e-05, | |
| "loss": 0.7691, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_bleu": 6.6682, | |
| "eval_gen_len": 17.7685, | |
| "eval_loss": 0.8987509608268738, | |
| "eval_runtime": 1501.6272, | |
| "eval_samples_per_second": 15.122, | |
| "eval_steps_per_second": 0.946, | |
| "step": 31250 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 4.083607907742998e-05, | |
| "loss": 0.7673, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 4.0688985643680864e-05, | |
| "loss": 0.8064, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 4.054189220993175e-05, | |
| "loss": 0.771, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "eval_bleu": 5.7856, | |
| "eval_gen_len": 17.8678, | |
| "eval_loss": 0.9189176559448242, | |
| "eval_runtime": 1503.2313, | |
| "eval_samples_per_second": 15.105, | |
| "eval_steps_per_second": 0.945, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 4.0394798776182635e-05, | |
| "loss": 0.7652, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 4.024770534243352e-05, | |
| "loss": 0.7658, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_bleu": 6.2468, | |
| "eval_gen_len": 17.7313, | |
| "eval_loss": 0.9175418019294739, | |
| "eval_runtime": 1487.9561, | |
| "eval_samples_per_second": 15.261, | |
| "eval_steps_per_second": 0.954, | |
| "step": 33750 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 4.01006119086844e-05, | |
| "loss": 0.7568, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 3.995351847493528e-05, | |
| "loss": 0.7913, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 3.9806425041186165e-05, | |
| "loss": 0.7914, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "eval_bleu": 5.5525, | |
| "eval_gen_len": 17.7627, | |
| "eval_loss": 0.9019802808761597, | |
| "eval_runtime": 1501.9062, | |
| "eval_samples_per_second": 15.119, | |
| "eval_steps_per_second": 0.945, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 3.965933160743705e-05, | |
| "loss": 0.7333, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 3.951223817368793e-05, | |
| "loss": 0.7264, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "eval_bleu": 6.2055, | |
| "eval_gen_len": 17.7662, | |
| "eval_loss": 0.9046412110328674, | |
| "eval_runtime": 1496.9385, | |
| "eval_samples_per_second": 15.169, | |
| "eval_steps_per_second": 0.949, | |
| "step": 36250 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "step": 36250, | |
| "total_flos": 1.257073961708028e+18, | |
| "train_loss": 1.0393056295460668, | |
| "train_runtime": 72152.6791, | |
| "train_samples_per_second": 37.766, | |
| "train_steps_per_second": 2.36 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 170310, | |
| "num_train_epochs": 15, | |
| "save_steps": 1250, | |
| "total_flos": 1.257073961708028e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |