| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.996592844974447, | |
| "global_step": 88000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0002985397907033341, | |
| "loss": 4.2075, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00029707958140666826, | |
| "loss": 3.7047, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0002956193721100024, | |
| "loss": 3.5253, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00029415916281333654, | |
| "loss": 3.4404, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0002926989535166707, | |
| "loss": 3.3437, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00029123874422000483, | |
| "loss": 3.3357, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.000289778534923339, | |
| "loss": 3.275, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00028831832562667317, | |
| "loss": 3.2801, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0002868581163300073, | |
| "loss": 3.2396, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00028539790703334145, | |
| "loss": 3.2389, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_gen_len": 15.0482, | |
| "eval_loss": 2.7936947345733643, | |
| "eval_rouge1": 26.3204, | |
| "eval_rouge2": 9.4077, | |
| "eval_rougeL": 23.6598, | |
| "eval_rougeLsum": 23.8655, | |
| "eval_runtime": 530.9697, | |
| "eval_samples_per_second": 18.837, | |
| "eval_steps_per_second": 2.356, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00028393769773667557, | |
| "loss": 3.2308, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00028247748844000974, | |
| "loss": 3.1836, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00028101727914334385, | |
| "loss": 3.2004, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00027955706984667797, | |
| "loss": 3.1561, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00027809686055001214, | |
| "loss": 3.1582, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00027663665125334625, | |
| "loss": 3.1249, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002751764419566804, | |
| "loss": 3.1236, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0002737162326600146, | |
| "loss": 3.0904, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0002722560233633487, | |
| "loss": 3.1179, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0002707958140666829, | |
| "loss": 3.0741, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_gen_len": 14.4006, | |
| "eval_loss": 2.6971848011016846, | |
| "eval_rouge1": 26.9258, | |
| "eval_rouge2": 9.7445, | |
| "eval_rougeL": 24.2045, | |
| "eval_rougeLsum": 24.4155, | |
| "eval_runtime": 528.5359, | |
| "eval_samples_per_second": 18.924, | |
| "eval_steps_per_second": 2.367, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00026933560477001705, | |
| "loss": 3.0814, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00026787539547335116, | |
| "loss": 3.1, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00026641518617668533, | |
| "loss": 3.0672, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00026495497688001945, | |
| "loss": 3.0728, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.0002634947675833536, | |
| "loss": 3.0516, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00026203455828668773, | |
| "loss": 3.0587, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00026057434899002185, | |
| "loss": 3.061, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.000259114139693356, | |
| "loss": 3.0404, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00025765393039669013, | |
| "loss": 3.0352, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002561937211000243, | |
| "loss": 2.9119, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_gen_len": 15.1769, | |
| "eval_loss": 2.641094923019409, | |
| "eval_rouge1": 27.6132, | |
| "eval_rouge2": 10.2843, | |
| "eval_rougeL": 24.6804, | |
| "eval_rougeLsum": 24.9483, | |
| "eval_runtime": 519.8172, | |
| "eval_samples_per_second": 19.241, | |
| "eval_steps_per_second": 2.407, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.00025473351180335847, | |
| "loss": 2.8945, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002532733025066926, | |
| "loss": 2.8809, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00025181309321002676, | |
| "loss": 2.8548, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.0002503528839133609, | |
| "loss": 2.8613, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.00024889267461669504, | |
| "loss": 2.8463, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0002474324653200292, | |
| "loss": 2.8409, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.0002459722560233633, | |
| "loss": 2.8519, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.0002445120467266975, | |
| "loss": 2.8457, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002430518374300316, | |
| "loss": 2.8603, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.00024159162813336578, | |
| "loss": 2.8272, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_gen_len": 15.1337, | |
| "eval_loss": 2.596390962600708, | |
| "eval_rouge1": 27.5723, | |
| "eval_rouge2": 10.131, | |
| "eval_rougeL": 24.6773, | |
| "eval_rougeLsum": 24.9496, | |
| "eval_runtime": 529.2072, | |
| "eval_samples_per_second": 18.9, | |
| "eval_steps_per_second": 2.364, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.0002401314188366999, | |
| "loss": 2.8576, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00023867120954003404, | |
| "loss": 2.8396, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0002372110002433682, | |
| "loss": 2.838, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00023575079094670232, | |
| "loss": 2.842, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002342905816500365, | |
| "loss": 2.8255, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.0002328303723533706, | |
| "loss": 2.8173, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00023137016305670478, | |
| "loss": 2.8115, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00022990995376003892, | |
| "loss": 2.8148, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00022844974446337306, | |
| "loss": 2.8155, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.0002269895351667072, | |
| "loss": 2.8135, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_gen_len": 15.4393, | |
| "eval_loss": 2.5782437324523926, | |
| "eval_rouge1": 28.7444, | |
| "eval_rouge2": 10.8958, | |
| "eval_rougeL": 25.6842, | |
| "eval_rougeLsum": 25.9691, | |
| "eval_runtime": 520.1126, | |
| "eval_samples_per_second": 19.23, | |
| "eval_steps_per_second": 2.405, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00022552932587004137, | |
| "loss": 2.8161, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.0002240691165733755, | |
| "loss": 2.8132, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.00022260890727670966, | |
| "loss": 2.8192, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.00022114869798004378, | |
| "loss": 2.8098, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.00021968848868337792, | |
| "loss": 2.776, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.0002182282793867121, | |
| "loss": 2.7931, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.0002167680700900462, | |
| "loss": 2.7675, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.00021530786079338037, | |
| "loss": 2.788, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.0002138476514967145, | |
| "loss": 2.7194, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.00021238744220004866, | |
| "loss": 2.6235, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_gen_len": 14.9003, | |
| "eval_loss": 2.551361322402954, | |
| "eval_rouge1": 29.1138, | |
| "eval_rouge2": 11.1582, | |
| "eval_rougeL": 25.9524, | |
| "eval_rougeLsum": 26.2659, | |
| "eval_runtime": 524.2558, | |
| "eval_samples_per_second": 19.078, | |
| "eval_steps_per_second": 2.386, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.0002109272329033828, | |
| "loss": 2.6189, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.00020946702360671694, | |
| "loss": 2.6318, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.00020800681431005108, | |
| "loss": 2.6356, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.00020654660501338525, | |
| "loss": 2.6379, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.00020508639571671937, | |
| "loss": 2.6018, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 0.00020362618642005354, | |
| "loss": 2.6223, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 0.00020216597712338765, | |
| "loss": 2.6421, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 0.0002007057678267218, | |
| "loss": 2.6397, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 0.00019924555853005597, | |
| "loss": 2.6401, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 0.00019778534923339008, | |
| "loss": 2.641, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_gen_len": 14.8827, | |
| "eval_loss": 2.526247978210449, | |
| "eval_rouge1": 29.1569, | |
| "eval_rouge2": 11.2473, | |
| "eval_rougeL": 26.0977, | |
| "eval_rougeLsum": 26.4149, | |
| "eval_runtime": 516.1021, | |
| "eval_samples_per_second": 19.38, | |
| "eval_steps_per_second": 2.424, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.00019632513993672425, | |
| "loss": 2.6228, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 0.0001948649306400584, | |
| "loss": 2.6487, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00019340472134339254, | |
| "loss": 2.6373, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 0.00019194451204672668, | |
| "loss": 2.6493, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.00019048430275006085, | |
| "loss": 2.6378, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 0.00018902409345339496, | |
| "loss": 2.6264, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.00018756388415672913, | |
| "loss": 2.6223, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 0.00018610367486006325, | |
| "loss": 2.6156, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.00018464346556339742, | |
| "loss": 2.6221, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 0.00018318325626673156, | |
| "loss": 2.6342, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "eval_gen_len": 14.8778, | |
| "eval_loss": 2.5141656398773193, | |
| "eval_rouge1": 29.2214, | |
| "eval_rouge2": 11.1812, | |
| "eval_rougeL": 26.1029, | |
| "eval_rougeLsum": 26.3802, | |
| "eval_runtime": 526.5352, | |
| "eval_samples_per_second": 18.996, | |
| "eval_steps_per_second": 2.376, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.00018172304697006568, | |
| "loss": 2.6198, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 0.00018026283767339985, | |
| "loss": 2.6206, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 0.00017880262837673396, | |
| "loss": 2.6106, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 0.00017734241908006813, | |
| "loss": 2.6162, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.00017588220978340227, | |
| "loss": 2.6218, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 0.00017442200048673642, | |
| "loss": 2.6045, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.00017296179119007056, | |
| "loss": 2.579, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00017150158189340473, | |
| "loss": 2.6103, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 0.00017004137259673884, | |
| "loss": 2.454, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.000168581163300073, | |
| "loss": 2.4637, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "eval_gen_len": 15.2305, | |
| "eval_loss": 2.511312246322632, | |
| "eval_rouge1": 29.6711, | |
| "eval_rouge2": 11.5, | |
| "eval_rougeL": 26.4325, | |
| "eval_rougeLsum": 26.7634, | |
| "eval_runtime": 524.3633, | |
| "eval_samples_per_second": 19.075, | |
| "eval_steps_per_second": 2.386, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 0.00016712095400340713, | |
| "loss": 2.4633, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 0.0001656607447067413, | |
| "loss": 2.4437, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 0.00016420053541007544, | |
| "loss": 2.4744, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.00016274032611340956, | |
| "loss": 2.4618, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 0.00016128011681674372, | |
| "loss": 2.4945, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 0.00015981990752007784, | |
| "loss": 2.4747, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 0.000158359698223412, | |
| "loss": 2.4632, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.00015689948892674615, | |
| "loss": 2.4892, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 0.0001554392796300803, | |
| "loss": 2.4389, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 0.00015397907033341444, | |
| "loss": 2.5028, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "eval_gen_len": 15.2348, | |
| "eval_loss": 2.494405746459961, | |
| "eval_rouge1": 29.8317, | |
| "eval_rouge2": 11.5675, | |
| "eval_rougeL": 26.4741, | |
| "eval_rougeLsum": 26.7943, | |
| "eval_runtime": 526.1631, | |
| "eval_samples_per_second": 19.009, | |
| "eval_steps_per_second": 2.378, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.0001525188610367486, | |
| "loss": 2.457, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 0.00015105865174008272, | |
| "loss": 2.4586, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 0.00014959844244341686, | |
| "loss": 2.4731, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 0.000148138233146751, | |
| "loss": 2.492, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 0.00014667802385008518, | |
| "loss": 2.485, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 0.00014521781455341932, | |
| "loss": 2.4679, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 0.00014375760525675346, | |
| "loss": 2.4675, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.0001422973959600876, | |
| "loss": 2.4659, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.00014083718666342175, | |
| "loss": 2.472, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 0.0001393769773667559, | |
| "loss": 2.4507, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "eval_gen_len": 14.9608, | |
| "eval_loss": 2.479571580886841, | |
| "eval_rouge1": 29.8158, | |
| "eval_rouge2": 11.6394, | |
| "eval_rougeL": 26.5371, | |
| "eval_rougeLsum": 26.8625, | |
| "eval_runtime": 518.7283, | |
| "eval_samples_per_second": 19.282, | |
| "eval_steps_per_second": 2.412, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.00013791676807009003, | |
| "loss": 2.489, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 0.00013645655877342417, | |
| "loss": 2.4626, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 0.00013499634947675834, | |
| "loss": 2.4534, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 0.00013353614018009246, | |
| "loss": 2.4664, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 0.0001320759308834266, | |
| "loss": 2.4927, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 0.00013061572158676074, | |
| "loss": 2.4654, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 0.00012915551229009489, | |
| "loss": 2.48, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 0.00012769530299342906, | |
| "loss": 2.3542, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 0.0001262350936967632, | |
| "loss": 2.3249, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.00012477488440009734, | |
| "loss": 2.3282, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "eval_gen_len": 14.9966, | |
| "eval_loss": 2.4929299354553223, | |
| "eval_rouge1": 29.8621, | |
| "eval_rouge2": 11.5076, | |
| "eval_rougeL": 26.5598, | |
| "eval_rougeLsum": 26.9032, | |
| "eval_runtime": 525.8633, | |
| "eval_samples_per_second": 19.02, | |
| "eval_steps_per_second": 2.379, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 0.00012331467510343148, | |
| "loss": 2.3419, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 0.00012185446580676563, | |
| "loss": 2.3247, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 0.00012039425651009977, | |
| "loss": 2.3349, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 0.00011893404721343392, | |
| "loss": 2.3203, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 0.00011747383791676807, | |
| "loss": 2.3494, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 0.00011601362862010221, | |
| "loss": 2.3575, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 0.00011455341932343635, | |
| "loss": 2.343, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.00011309321002677048, | |
| "loss": 2.3609, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 0.00011163300073010464, | |
| "loss": 2.3294, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 0.00011017279143343878, | |
| "loss": 2.3218, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "eval_gen_len": 15.2109, | |
| "eval_loss": 2.473184585571289, | |
| "eval_rouge1": 30.1018, | |
| "eval_rouge2": 11.8166, | |
| "eval_rougeL": 26.72, | |
| "eval_rougeLsum": 27.0393, | |
| "eval_runtime": 525.9361, | |
| "eval_samples_per_second": 19.018, | |
| "eval_steps_per_second": 2.379, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 0.00010871258213677292, | |
| "loss": 2.3574, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 0.00010725237284010706, | |
| "loss": 2.3396, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 0.00010579216354344122, | |
| "loss": 2.3742, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.00010433195424677536, | |
| "loss": 2.3382, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 0.0001028717449501095, | |
| "loss": 2.3518, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 0.00010141153565344366, | |
| "loss": 2.3659, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 9.99513263567778e-05, | |
| "loss": 2.3487, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 9.849111706011195e-05, | |
| "loss": 2.3337, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 9.703090776344609e-05, | |
| "loss": 2.3426, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 9.557069846678024e-05, | |
| "loss": 2.3321, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "eval_gen_len": 14.7861, | |
| "eval_loss": 2.462064266204834, | |
| "eval_rouge1": 30.3513, | |
| "eval_rouge2": 11.9542, | |
| "eval_rougeL": 27.074, | |
| "eval_rougeLsum": 27.39, | |
| "eval_runtime": 529.0668, | |
| "eval_samples_per_second": 18.905, | |
| "eval_steps_per_second": 2.365, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 9.411048917011437e-05, | |
| "loss": 2.3369, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 9.265027987344852e-05, | |
| "loss": 2.3536, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 9.119007057678266e-05, | |
| "loss": 2.3364, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 8.97298612801168e-05, | |
| "loss": 2.3412, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 8.826965198345096e-05, | |
| "loss": 2.3364, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 8.68094426867851e-05, | |
| "loss": 2.3605, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 8.534923339011924e-05, | |
| "loss": 2.3124, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 8.388902409345338e-05, | |
| "loss": 2.2233, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 8.242881479678754e-05, | |
| "loss": 2.2063, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 8.096860550012168e-05, | |
| "loss": 2.2197, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "eval_gen_len": 14.9783, | |
| "eval_loss": 2.4723336696624756, | |
| "eval_rouge1": 30.3337, | |
| "eval_rouge2": 11.9362, | |
| "eval_rougeL": 27.0161, | |
| "eval_rougeLsum": 27.365, | |
| "eval_runtime": 522.7349, | |
| "eval_samples_per_second": 19.134, | |
| "eval_steps_per_second": 2.393, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 7.950839620345582e-05, | |
| "loss": 2.2218, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 7.804818690678997e-05, | |
| "loss": 2.2415, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 7.658797761012412e-05, | |
| "loss": 2.2132, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "learning_rate": 7.512776831345825e-05, | |
| "loss": 2.2604, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 7.366755901679241e-05, | |
| "loss": 2.2573, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 7.220734972012655e-05, | |
| "loss": 2.2291, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 7.074714042346069e-05, | |
| "loss": 2.2489, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 6.928693112679484e-05, | |
| "loss": 2.225, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "learning_rate": 6.782672183012898e-05, | |
| "loss": 2.2077, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 6.636651253346312e-05, | |
| "loss": 2.2273, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "eval_gen_len": 14.8784, | |
| "eval_loss": 2.4681761264801025, | |
| "eval_rouge1": 30.4217, | |
| "eval_rouge2": 12.0307, | |
| "eval_rougeL": 27.11, | |
| "eval_rougeLsum": 27.4155, | |
| "eval_runtime": 524.8636, | |
| "eval_samples_per_second": 19.056, | |
| "eval_steps_per_second": 2.383, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 6.490630323679726e-05, | |
| "loss": 2.2231, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 6.344609394013142e-05, | |
| "loss": 2.2679, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "learning_rate": 6.198588464346556e-05, | |
| "loss": 2.2423, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 6.05256753467997e-05, | |
| "loss": 2.2195, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 5.9065466050133846e-05, | |
| "loss": 2.2439, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 5.760525675346799e-05, | |
| "loss": 2.2261, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 5.614504745680214e-05, | |
| "loss": 2.2562, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 5.468483816013628e-05, | |
| "loss": 2.2477, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 5.322462886347043e-05, | |
| "loss": 2.2257, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 5.176441956680457e-05, | |
| "loss": 2.2407, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "eval_gen_len": 14.8097, | |
| "eval_loss": 2.4504895210266113, | |
| "eval_rouge1": 30.4682, | |
| "eval_rouge2": 11.9828, | |
| "eval_rougeL": 27.0725, | |
| "eval_rougeLsum": 27.3849, | |
| "eval_runtime": 526.291, | |
| "eval_samples_per_second": 19.005, | |
| "eval_steps_per_second": 2.377, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 5.0304210270138715e-05, | |
| "loss": 2.2334, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 4.884400097347286e-05, | |
| "loss": 2.2468, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 4.7383791676807007e-05, | |
| "loss": 2.2107, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 4.592358238014115e-05, | |
| "loss": 2.2534, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 4.44633730834753e-05, | |
| "loss": 2.2311, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.300316378680944e-05, | |
| "loss": 2.2225, | |
| "step": 88000 | |
| } | |
| ], | |
| "max_steps": 102725, | |
| "num_train_epochs": 7, | |
| "total_flos": 2.7215384355038822e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |