| { | |
| "best_metric": 0.656362771987915, | |
| "best_model_checkpoint": "/home2/s5431786/jb-internship/results/t5-small-CodeXGLUE-CONCODE-faster/checkpoint-24500", | |
| "epoch": 8.64, | |
| "global_step": 27000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 7.968e-05, | |
| "loss": 3.6878, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_bleu": 0.0323034964768643, | |
| "eval_exact_match": 0.0035, | |
| "eval_loss": 1.9745688438415527, | |
| "eval_rouge1": 0.3491207907520632, | |
| "eval_rouge2": 0.14656678100405174, | |
| "eval_rougeL": 0.3285981582102612, | |
| "eval_runtime": 36.246, | |
| "eval_samples_per_second": 55.178, | |
| "eval_steps_per_second": 1.738, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00015968, | |
| "loss": 1.7625, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_bleu": 0.038749091873206375, | |
| "eval_exact_match": 0.0145, | |
| "eval_loss": 1.4567848443984985, | |
| "eval_rouge1": 0.35498006740689036, | |
| "eval_rouge2": 0.13946738339077275, | |
| "eval_rougeL": 0.3327171698914647, | |
| "eval_runtime": 34.5773, | |
| "eval_samples_per_second": 57.841, | |
| "eval_steps_per_second": 1.822, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00023952, | |
| "loss": 1.3575, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_bleu": 0.04388514972808356, | |
| "eval_exact_match": 0.053, | |
| "eval_loss": 1.1938430070877075, | |
| "eval_rouge1": 0.4102062635539608, | |
| "eval_rouge2": 0.21783150768420984, | |
| "eval_rougeL": 0.3907330178570455, | |
| "eval_runtime": 34.9414, | |
| "eval_samples_per_second": 57.239, | |
| "eval_steps_per_second": 1.803, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00031952000000000006, | |
| "loss": 1.1187, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_bleu": 0.05024016948621801, | |
| "eval_exact_match": 0.0735, | |
| "eval_loss": 1.0715276002883911, | |
| "eval_rouge1": 0.44133744346838355, | |
| "eval_rouge2": 0.25777725613931946, | |
| "eval_rougeL": 0.4219959477894215, | |
| "eval_runtime": 35.3166, | |
| "eval_samples_per_second": 56.631, | |
| "eval_steps_per_second": 1.784, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00039952, | |
| "loss": 1.0062, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_bleu": 0.0525791651130599, | |
| "eval_exact_match": 0.0815, | |
| "eval_loss": 0.9908952116966248, | |
| "eval_rouge1": 0.46270763973625095, | |
| "eval_rouge2": 0.2836027818482203, | |
| "eval_rougeL": 0.4457626023912316, | |
| "eval_runtime": 35.2306, | |
| "eval_samples_per_second": 56.769, | |
| "eval_steps_per_second": 1.788, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00047952, | |
| "loss": 0.9314, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_bleu": 0.056618620209005335, | |
| "eval_exact_match": 0.0845, | |
| "eval_loss": 0.9375063180923462, | |
| "eval_rouge1": 0.4688436524948581, | |
| "eval_rouge2": 0.2932255323307211, | |
| "eval_rougeL": 0.45301193641551707, | |
| "eval_runtime": 37.1868, | |
| "eval_samples_per_second": 53.783, | |
| "eval_steps_per_second": 1.694, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.0004968673684210526, | |
| "loss": 0.8581, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_bleu": 0.05920053067894586, | |
| "eval_exact_match": 0.0855, | |
| "eval_loss": 0.8949318528175354, | |
| "eval_rouge1": 0.475736898014815, | |
| "eval_rouge2": 0.30127214832048377, | |
| "eval_rougeL": 0.45988834774964316, | |
| "eval_runtime": 35.9797, | |
| "eval_samples_per_second": 55.587, | |
| "eval_steps_per_second": 1.751, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.0004926568421052632, | |
| "loss": 0.8077, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_bleu": 0.060850014141887766, | |
| "eval_exact_match": 0.088, | |
| "eval_loss": 0.8577723503112793, | |
| "eval_rouge1": 0.4857568042531289, | |
| "eval_rouge2": 0.3123549789085296, | |
| "eval_rougeL": 0.47152130323710484, | |
| "eval_runtime": 35.0315, | |
| "eval_samples_per_second": 57.091, | |
| "eval_steps_per_second": 1.798, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.0004884463157894737, | |
| "loss": 0.7742, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_bleu": 0.060679509741740224, | |
| "eval_exact_match": 0.091, | |
| "eval_loss": 0.8396906852722168, | |
| "eval_rouge1": 0.48584317956376577, | |
| "eval_rouge2": 0.31707363805938765, | |
| "eval_rougeL": 0.47246293765798464, | |
| "eval_runtime": 35.6286, | |
| "eval_samples_per_second": 56.135, | |
| "eval_steps_per_second": 1.768, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00048423578947368425, | |
| "loss": 0.7556, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_bleu": 0.062181241480247514, | |
| "eval_exact_match": 0.094, | |
| "eval_loss": 0.8199761509895325, | |
| "eval_rouge1": 0.49335199051744105, | |
| "eval_rouge2": 0.3273247481525058, | |
| "eval_rougeL": 0.47992085377826155, | |
| "eval_runtime": 37.2571, | |
| "eval_samples_per_second": 53.681, | |
| "eval_steps_per_second": 1.691, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00048002526315789475, | |
| "loss": 0.7246, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_bleu": 0.06360201700239017, | |
| "eval_exact_match": 0.096, | |
| "eval_loss": 0.7979766726493835, | |
| "eval_rouge1": 0.49459857323152784, | |
| "eval_rouge2": 0.32742424566936434, | |
| "eval_rougeL": 0.4795144854855635, | |
| "eval_runtime": 38.3013, | |
| "eval_samples_per_second": 52.218, | |
| "eval_steps_per_second": 1.645, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.0004758147368421053, | |
| "loss": 0.7045, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_bleu": 0.06323253320022208, | |
| "eval_exact_match": 0.099, | |
| "eval_loss": 0.7866926193237305, | |
| "eval_rouge1": 0.49270311763410307, | |
| "eval_rouge2": 0.3310082026030387, | |
| "eval_rougeL": 0.4800440014525599, | |
| "eval_runtime": 35.9272, | |
| "eval_samples_per_second": 55.668, | |
| "eval_steps_per_second": 1.754, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.0004716042105263158, | |
| "loss": 0.6778, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_bleu": 0.06526774731233032, | |
| "eval_exact_match": 0.097, | |
| "eval_loss": 0.7727072834968567, | |
| "eval_rouge1": 0.5015386585306862, | |
| "eval_rouge2": 0.3327019592611423, | |
| "eval_rougeL": 0.4858921391142662, | |
| "eval_runtime": 36.8788, | |
| "eval_samples_per_second": 54.232, | |
| "eval_steps_per_second": 1.708, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.0004673936842105263, | |
| "loss": 0.652, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_bleu": 0.06603832460136996, | |
| "eval_exact_match": 0.0945, | |
| "eval_loss": 0.7700557708740234, | |
| "eval_rouge1": 0.5007155856518921, | |
| "eval_rouge2": 0.33660838962730144, | |
| "eval_rougeL": 0.48704095883836096, | |
| "eval_runtime": 36.0428, | |
| "eval_samples_per_second": 55.49, | |
| "eval_steps_per_second": 1.748, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0004631831578947369, | |
| "loss": 0.6477, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_bleu": 0.0637326815837752, | |
| "eval_exact_match": 0.1015, | |
| "eval_loss": 0.7626734375953674, | |
| "eval_rouge1": 0.5067175865244784, | |
| "eval_rouge2": 0.3418343311421686, | |
| "eval_rougeL": 0.4932440074450731, | |
| "eval_runtime": 36.8678, | |
| "eval_samples_per_second": 54.248, | |
| "eval_steps_per_second": 1.709, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.0004589726315789474, | |
| "loss": 0.6337, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_bleu": 0.06513339878553655, | |
| "eval_exact_match": 0.1015, | |
| "eval_loss": 0.7461714148521423, | |
| "eval_rouge1": 0.4982517321430179, | |
| "eval_rouge2": 0.33681326926227717, | |
| "eval_rougeL": 0.4849838232405096, | |
| "eval_runtime": 36.8056, | |
| "eval_samples_per_second": 54.34, | |
| "eval_steps_per_second": 1.712, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 0.00045476210526315794, | |
| "loss": 0.6278, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "eval_bleu": 0.06459301207154695, | |
| "eval_exact_match": 0.0995, | |
| "eval_loss": 0.7355420589447021, | |
| "eval_rouge1": 0.503290317947929, | |
| "eval_rouge2": 0.34354898330754957, | |
| "eval_rougeL": 0.49001424087111184, | |
| "eval_runtime": 36.3871, | |
| "eval_samples_per_second": 54.965, | |
| "eval_steps_per_second": 1.731, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.00045055157894736844, | |
| "loss": 0.6224, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_bleu": 0.06439506753752011, | |
| "eval_exact_match": 0.102, | |
| "eval_loss": 0.7340560555458069, | |
| "eval_rouge1": 0.5052438839482049, | |
| "eval_rouge2": 0.34484226789208994, | |
| "eval_rougeL": 0.4921510782769857, | |
| "eval_runtime": 37.3894, | |
| "eval_samples_per_second": 53.491, | |
| "eval_steps_per_second": 1.685, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 0.000446341052631579, | |
| "loss": 0.6008, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "eval_bleu": 0.06461231818263671, | |
| "eval_exact_match": 0.101, | |
| "eval_loss": 0.7321618795394897, | |
| "eval_rouge1": 0.5033575057087202, | |
| "eval_rouge2": 0.3421619427853616, | |
| "eval_rougeL": 0.4902835308639695, | |
| "eval_runtime": 36.6808, | |
| "eval_samples_per_second": 54.524, | |
| "eval_steps_per_second": 1.718, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0004421305263157895, | |
| "loss": 0.589, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_bleu": 0.06509490313536258, | |
| "eval_exact_match": 0.1005, | |
| "eval_loss": 0.7238326668739319, | |
| "eval_rouge1": 0.5044211940127038, | |
| "eval_rouge2": 0.34197306671723443, | |
| "eval_rougeL": 0.4906146460856744, | |
| "eval_runtime": 35.0879, | |
| "eval_samples_per_second": 57.0, | |
| "eval_steps_per_second": 1.795, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 0.00043792, | |
| "loss": 0.5708, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "eval_bleu": 0.06631228557498069, | |
| "eval_exact_match": 0.0995, | |
| "eval_loss": 0.715421736240387, | |
| "eval_rouge1": 0.5068028046779569, | |
| "eval_rouge2": 0.34473925417295426, | |
| "eval_rougeL": 0.49254925133943384, | |
| "eval_runtime": 35.4516, | |
| "eval_samples_per_second": 56.415, | |
| "eval_steps_per_second": 1.777, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 0.00043370947368421057, | |
| "loss": 0.5712, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "eval_bleu": 0.06578551089272082, | |
| "eval_exact_match": 0.102, | |
| "eval_loss": 0.7109146118164062, | |
| "eval_rouge1": 0.5061889557930622, | |
| "eval_rouge2": 0.34828839953790597, | |
| "eval_rougeL": 0.4934676685845075, | |
| "eval_runtime": 35.6977, | |
| "eval_samples_per_second": 56.026, | |
| "eval_steps_per_second": 1.765, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.0004295157894736842, | |
| "loss": 0.5655, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "eval_bleu": 0.06635467115197174, | |
| "eval_exact_match": 0.1015, | |
| "eval_loss": 0.7136749029159546, | |
| "eval_rouge1": 0.5101572079631966, | |
| "eval_rouge2": 0.3491129312343716, | |
| "eval_rougeL": 0.49647385739953515, | |
| "eval_runtime": 35.4035, | |
| "eval_samples_per_second": 56.492, | |
| "eval_steps_per_second": 1.779, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 0.00042530526315789473, | |
| "loss": 0.5593, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "eval_bleu": 0.06743929871449864, | |
| "eval_exact_match": 0.104, | |
| "eval_loss": 0.7079191207885742, | |
| "eval_rouge1": 0.5109229447218538, | |
| "eval_rouge2": 0.3498840589135052, | |
| "eval_rougeL": 0.49719053454192585, | |
| "eval_runtime": 35.9294, | |
| "eval_samples_per_second": 55.665, | |
| "eval_steps_per_second": 1.753, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.00042109473684210524, | |
| "loss": 0.5501, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 0.06587822883992077, | |
| "eval_exact_match": 0.0965, | |
| "eval_loss": 0.7067587971687317, | |
| "eval_rouge1": 0.5063077878070544, | |
| "eval_rouge2": 0.34684822407233706, | |
| "eval_rougeL": 0.49414991531075403, | |
| "eval_runtime": 26.7311, | |
| "eval_samples_per_second": 74.819, | |
| "eval_steps_per_second": 2.357, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 0.0004168842105263158, | |
| "loss": 0.5349, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "eval_bleu": 0.06621007420514197, | |
| "eval_exact_match": 0.1035, | |
| "eval_loss": 0.7037115097045898, | |
| "eval_rouge1": 0.5099136668737985, | |
| "eval_rouge2": 0.35113419695580583, | |
| "eval_rougeL": 0.4975205835514277, | |
| "eval_runtime": 25.9643, | |
| "eval_samples_per_second": 77.029, | |
| "eval_steps_per_second": 2.426, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 0.0004126736842105263, | |
| "loss": 0.5303, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "eval_bleu": 0.06674010593523277, | |
| "eval_exact_match": 0.1, | |
| "eval_loss": 0.700024425983429, | |
| "eval_rouge1": 0.5072383570011582, | |
| "eval_rouge2": 0.35252886838350916, | |
| "eval_rougeL": 0.49584290875564585, | |
| "eval_runtime": 25.9074, | |
| "eval_samples_per_second": 77.198, | |
| "eval_steps_per_second": 2.432, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 0.00040846315789473686, | |
| "loss": 0.5175, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "eval_bleu": 0.06588129973614662, | |
| "eval_exact_match": 0.096, | |
| "eval_loss": 0.6980301737785339, | |
| "eval_rouge1": 0.5084177911192925, | |
| "eval_rouge2": 0.34934886295349316, | |
| "eval_rougeL": 0.4949379734561957, | |
| "eval_runtime": 26.6858, | |
| "eval_samples_per_second": 74.946, | |
| "eval_steps_per_second": 2.361, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 0.00040425263157894736, | |
| "loss": 0.5155, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "eval_bleu": 0.0671386377827142, | |
| "eval_exact_match": 0.0995, | |
| "eval_loss": 0.6891024708747864, | |
| "eval_rouge1": 0.5106225287363708, | |
| "eval_rouge2": 0.35235285505383196, | |
| "eval_rougeL": 0.4974072087383229, | |
| "eval_runtime": 26.1481, | |
| "eval_samples_per_second": 76.487, | |
| "eval_steps_per_second": 2.409, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.0004000505263157895, | |
| "loss": 0.517, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_bleu": 0.06712927692087013, | |
| "eval_exact_match": 0.0985, | |
| "eval_loss": 0.6874147653579712, | |
| "eval_rouge1": 0.5086994664859764, | |
| "eval_rouge2": 0.3479701230394284, | |
| "eval_rougeL": 0.4953283279428679, | |
| "eval_runtime": 25.8913, | |
| "eval_samples_per_second": 77.246, | |
| "eval_steps_per_second": 2.433, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 0.00039584000000000006, | |
| "loss": 0.5163, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_bleu": 0.06670745674633105, | |
| "eval_exact_match": 0.102, | |
| "eval_loss": 0.6896522045135498, | |
| "eval_rouge1": 0.5102024342202081, | |
| "eval_rouge2": 0.3510306264329252, | |
| "eval_rougeL": 0.49684909710826397, | |
| "eval_runtime": 26.1813, | |
| "eval_samples_per_second": 76.39, | |
| "eval_steps_per_second": 2.406, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 0.0003916294736842105, | |
| "loss": 0.4952, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "eval_bleu": 0.06623474967340444, | |
| "eval_exact_match": 0.103, | |
| "eval_loss": 0.6886290311813354, | |
| "eval_rouge1": 0.5123666040180792, | |
| "eval_rouge2": 0.3526184224019636, | |
| "eval_rougeL": 0.4984851483091739, | |
| "eval_runtime": 26.2009, | |
| "eval_samples_per_second": 76.333, | |
| "eval_steps_per_second": 2.405, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 0.00038742736842105265, | |
| "loss": 0.4922, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "eval_bleu": 0.06698301903860669, | |
| "eval_exact_match": 0.1045, | |
| "eval_loss": 0.6785948276519775, | |
| "eval_rouge1": 0.5117392179713256, | |
| "eval_rouge2": 0.35491013960107753, | |
| "eval_rougeL": 0.49895018657411827, | |
| "eval_runtime": 25.8834, | |
| "eval_samples_per_second": 77.269, | |
| "eval_steps_per_second": 2.434, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 0.00038321684210526315, | |
| "loss": 0.4874, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "eval_bleu": 0.0690414017532501, | |
| "eval_exact_match": 0.103, | |
| "eval_loss": 0.6765531301498413, | |
| "eval_rouge1": 0.5144668417632419, | |
| "eval_rouge2": 0.35703037655654435, | |
| "eval_rougeL": 0.5011385205263255, | |
| "eval_runtime": 26.2313, | |
| "eval_samples_per_second": 76.245, | |
| "eval_steps_per_second": 2.402, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.0003790063157894737, | |
| "loss": 0.4864, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "eval_bleu": 0.06679150292418984, | |
| "eval_exact_match": 0.102, | |
| "eval_loss": 0.6803983449935913, | |
| "eval_rouge1": 0.5130781619566334, | |
| "eval_rouge2": 0.35606833458373227, | |
| "eval_rougeL": 0.4989077764101931, | |
| "eval_runtime": 26.015, | |
| "eval_samples_per_second": 76.879, | |
| "eval_steps_per_second": 2.422, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.0003747957894736842, | |
| "loss": 0.4867, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "eval_bleu": 0.06910514516097221, | |
| "eval_exact_match": 0.104, | |
| "eval_loss": 0.6761348247528076, | |
| "eval_rouge1": 0.5136060106685832, | |
| "eval_rouge2": 0.357449154537545, | |
| "eval_rougeL": 0.5012406684437172, | |
| "eval_runtime": 26.5616, | |
| "eval_samples_per_second": 75.297, | |
| "eval_steps_per_second": 2.372, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 0.00037058526315789477, | |
| "loss": 0.4865, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "eval_bleu": 0.06815194117913387, | |
| "eval_exact_match": 0.1055, | |
| "eval_loss": 0.6767784357070923, | |
| "eval_rouge1": 0.5103735531673305, | |
| "eval_rouge2": 0.3571731763455166, | |
| "eval_rougeL": 0.49830522378417486, | |
| "eval_runtime": 25.9799, | |
| "eval_samples_per_second": 76.983, | |
| "eval_steps_per_second": 2.425, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 0.0003663747368421053, | |
| "loss": 0.4645, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "eval_bleu": 0.06602598544323403, | |
| "eval_exact_match": 0.099, | |
| "eval_loss": 0.6806569695472717, | |
| "eval_rouge1": 0.5086638191909727, | |
| "eval_rouge2": 0.3503888330985566, | |
| "eval_rougeL": 0.49554377885548156, | |
| "eval_runtime": 25.9049, | |
| "eval_samples_per_second": 77.205, | |
| "eval_steps_per_second": 2.432, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 0.00036216421052631583, | |
| "loss": 0.4559, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "eval_bleu": 0.06761756880359872, | |
| "eval_exact_match": 0.098, | |
| "eval_loss": 0.6715950965881348, | |
| "eval_rouge1": 0.5091972864925758, | |
| "eval_rouge2": 0.35173327385615816, | |
| "eval_rougeL": 0.49666185033111254, | |
| "eval_runtime": 26.1251, | |
| "eval_samples_per_second": 76.555, | |
| "eval_steps_per_second": 2.411, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.00035795368421052634, | |
| "loss": 0.4584, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "eval_bleu": 0.06754458066884086, | |
| "eval_exact_match": 0.106, | |
| "eval_loss": 0.6722220182418823, | |
| "eval_rouge1": 0.5120041534722938, | |
| "eval_rouge2": 0.35712607384056955, | |
| "eval_rougeL": 0.4993047634971517, | |
| "eval_runtime": 25.9887, | |
| "eval_samples_per_second": 76.957, | |
| "eval_steps_per_second": 2.424, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.00035374315789473684, | |
| "loss": 0.4563, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "eval_bleu": 0.06740545701870117, | |
| "eval_exact_match": 0.1015, | |
| "eval_loss": 0.6715394258499146, | |
| "eval_rouge1": 0.5120845038766327, | |
| "eval_rouge2": 0.35703709540753703, | |
| "eval_rougeL": 0.49983280781141204, | |
| "eval_runtime": 26.1007, | |
| "eval_samples_per_second": 76.626, | |
| "eval_steps_per_second": 2.414, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 0.0003495326315789474, | |
| "loss": 0.458, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_bleu": 0.06803331295950608, | |
| "eval_exact_match": 0.102, | |
| "eval_loss": 0.6695247888565063, | |
| "eval_rouge1": 0.5165415978526902, | |
| "eval_rouge2": 0.36002931999238674, | |
| "eval_rougeL": 0.5027446077282225, | |
| "eval_runtime": 26.1301, | |
| "eval_samples_per_second": 76.54, | |
| "eval_steps_per_second": 2.411, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 0.0003453221052631579, | |
| "loss": 0.4588, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "eval_bleu": 0.06913111094634318, | |
| "eval_exact_match": 0.102, | |
| "eval_loss": 0.6690553426742554, | |
| "eval_rouge1": 0.5125284973649041, | |
| "eval_rouge2": 0.35722843628208867, | |
| "eval_rougeL": 0.5004858865893127, | |
| "eval_runtime": 27.2998, | |
| "eval_samples_per_second": 73.26, | |
| "eval_steps_per_second": 2.308, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 0.00034111157894736846, | |
| "loss": 0.45, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "eval_bleu": 0.06819285477641512, | |
| "eval_exact_match": 0.1055, | |
| "eval_loss": 0.674951434135437, | |
| "eval_rouge1": 0.5165834203044295, | |
| "eval_rouge2": 0.3615776158637767, | |
| "eval_rougeL": 0.503606069989712, | |
| "eval_runtime": 28.1296, | |
| "eval_samples_per_second": 71.099, | |
| "eval_steps_per_second": 2.24, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 0.00033690105263157896, | |
| "loss": 0.4311, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "eval_bleu": 0.06724531081942378, | |
| "eval_exact_match": 0.1035, | |
| "eval_loss": 0.665252685546875, | |
| "eval_rouge1": 0.510047207004009, | |
| "eval_rouge2": 0.3579587667006371, | |
| "eval_rougeL": 0.4979863063664335, | |
| "eval_runtime": 26.0203, | |
| "eval_samples_per_second": 76.863, | |
| "eval_steps_per_second": 2.421, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 0.00033269894736842105, | |
| "loss": 0.4374, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "eval_bleu": 0.06757017098690109, | |
| "eval_exact_match": 0.1, | |
| "eval_loss": 0.6658646464347839, | |
| "eval_rouge1": 0.512652953906156, | |
| "eval_rouge2": 0.3561654225240507, | |
| "eval_rougeL": 0.49934790688943353, | |
| "eval_runtime": 26.0622, | |
| "eval_samples_per_second": 76.74, | |
| "eval_steps_per_second": 2.417, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 0.0003284884210526316, | |
| "loss": 0.433, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "eval_bleu": 0.06809047383108359, | |
| "eval_exact_match": 0.1035, | |
| "eval_loss": 0.6632941961288452, | |
| "eval_rouge1": 0.511074186368692, | |
| "eval_rouge2": 0.35738572766675536, | |
| "eval_rougeL": 0.4988662331382937, | |
| "eval_runtime": 26.1254, | |
| "eval_samples_per_second": 76.554, | |
| "eval_steps_per_second": 2.411, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "learning_rate": 0.0003242778947368421, | |
| "loss": 0.4334, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "eval_bleu": 0.06857766001536686, | |
| "eval_exact_match": 0.105, | |
| "eval_loss": 0.6601866483688354, | |
| "eval_rouge1": 0.5140953364016511, | |
| "eval_rouge2": 0.35736194915396335, | |
| "eval_rougeL": 0.5012302345945796, | |
| "eval_runtime": 26.229, | |
| "eval_samples_per_second": 76.252, | |
| "eval_steps_per_second": 2.402, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 0.0003200757894736842, | |
| "loss": 0.4345, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "eval_bleu": 0.06814284730693101, | |
| "eval_exact_match": 0.1025, | |
| "eval_loss": 0.656362771987915, | |
| "eval_rouge1": 0.5143728841578871, | |
| "eval_rouge2": 0.35808146099426286, | |
| "eval_rougeL": 0.5020021673160826, | |
| "eval_runtime": 26.1074, | |
| "eval_samples_per_second": 76.607, | |
| "eval_steps_per_second": 2.413, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.00031586526315789475, | |
| "loss": 0.4321, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu": 0.06891070414721397, | |
| "eval_exact_match": 0.102, | |
| "eval_loss": 0.6621795892715454, | |
| "eval_rouge1": 0.5134088474234255, | |
| "eval_rouge2": 0.35716681673325246, | |
| "eval_rougeL": 0.5008715932059962, | |
| "eval_runtime": 27.5712, | |
| "eval_samples_per_second": 72.539, | |
| "eval_steps_per_second": 2.285, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "learning_rate": 0.00031165473684210526, | |
| "loss": 0.4123, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "eval_bleu": 0.06810253515300806, | |
| "eval_exact_match": 0.1015, | |
| "eval_loss": 0.672680675983429, | |
| "eval_rouge1": 0.517027248239716, | |
| "eval_rouge2": 0.3614838699580787, | |
| "eval_rougeL": 0.5045490816379123, | |
| "eval_runtime": 25.969, | |
| "eval_samples_per_second": 77.015, | |
| "eval_steps_per_second": 2.426, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 0.00030744421052631576, | |
| "loss": 0.4124, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "eval_bleu": 0.06826764323139825, | |
| "eval_exact_match": 0.1045, | |
| "eval_loss": 0.665459394454956, | |
| "eval_rouge1": 0.5122977931541812, | |
| "eval_rouge2": 0.35973042474946837, | |
| "eval_rougeL": 0.5005467940238904, | |
| "eval_runtime": 26.2994, | |
| "eval_samples_per_second": 76.047, | |
| "eval_steps_per_second": 2.395, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 0.0003032336842105263, | |
| "loss": 0.405, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "eval_bleu": 0.06948334775074264, | |
| "eval_exact_match": 0.1005, | |
| "eval_loss": 0.663256824016571, | |
| "eval_rouge1": 0.518724388856783, | |
| "eval_rouge2": 0.3607352300107304, | |
| "eval_rougeL": 0.5051343067349954, | |
| "eval_runtime": 26.328, | |
| "eval_samples_per_second": 75.965, | |
| "eval_steps_per_second": 2.393, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "learning_rate": 0.0002990231578947368, | |
| "loss": 0.4122, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "eval_bleu": 0.06882410125002518, | |
| "eval_exact_match": 0.1015, | |
| "eval_loss": 0.658465564250946, | |
| "eval_rouge1": 0.5144908155613723, | |
| "eval_rouge2": 0.3583930821930816, | |
| "eval_rougeL": 0.5013138652290803, | |
| "eval_runtime": 27.8274, | |
| "eval_samples_per_second": 71.872, | |
| "eval_steps_per_second": 2.264, | |
| "step": 27000 | |
| } | |
| ], | |
| "max_steps": 62500, | |
| "num_train_epochs": 20, | |
| "total_flos": 1.1286810924259738e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |