|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 20.0, |
|
"eval_steps": 500, |
|
"global_step": 371900, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 3.125e-05, |
|
"loss": 6.2252, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 6.25e-05, |
|
"loss": 5.0003, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 9.375e-05, |
|
"loss": 4.6751, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 0.000125, |
|
"loss": 4.4573, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 0.00015625, |
|
"loss": 4.2943, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 0.0001875, |
|
"loss": 4.1687, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 0.00021875, |
|
"loss": 4.0718, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 0.00025, |
|
"loss": 3.9742, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 0.00028125000000000003, |
|
"loss": 3.9037, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 0.0003125, |
|
"loss": 3.8415, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"learning_rate": 0.00034375, |
|
"loss": 3.8015, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"learning_rate": 0.000375, |
|
"loss": 3.7642, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 0.00040625000000000004, |
|
"loss": 3.7283, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"learning_rate": 0.0004375, |
|
"loss": 3.7009, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"learning_rate": 0.00046875, |
|
"loss": 3.6715, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"learning_rate": 0.0005, |
|
"loss": 3.6474, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 0.00053125, |
|
"loss": 3.6209, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 0.0005625000000000001, |
|
"loss": 3.6026, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.3578007129931504, |
|
"eval_loss": 3.7870304584503174, |
|
"eval_runtime": 222.9054, |
|
"eval_samples_per_second": 259.846, |
|
"eval_steps_per_second": 2.032, |
|
"step": 18595 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 0.00059375, |
|
"loss": 3.5668, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 1.08, |
|
"learning_rate": 0.000625, |
|
"loss": 3.5444, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 1.13, |
|
"learning_rate": 0.00065625, |
|
"loss": 3.5358, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 1.18, |
|
"learning_rate": 0.0006875, |
|
"loss": 3.5237, |
|
"step": 22000 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"learning_rate": 0.00071875, |
|
"loss": 3.5107, |
|
"step": 23000 |
|
}, |
|
{ |
|
"epoch": 1.29, |
|
"learning_rate": 0.00075, |
|
"loss": 3.5037, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"learning_rate": 0.00078125, |
|
"loss": 3.4942, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"learning_rate": 0.0008125000000000001, |
|
"loss": 3.4814, |
|
"step": 26000 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"learning_rate": 0.00084375, |
|
"loss": 3.4718, |
|
"step": 27000 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"learning_rate": 0.000875, |
|
"loss": 3.4596, |
|
"step": 28000 |
|
}, |
|
{ |
|
"epoch": 1.56, |
|
"learning_rate": 0.00090625, |
|
"loss": 3.4533, |
|
"step": 29000 |
|
}, |
|
{ |
|
"epoch": 1.61, |
|
"learning_rate": 0.0009375, |
|
"loss": 3.4508, |
|
"step": 30000 |
|
}, |
|
{ |
|
"epoch": 1.67, |
|
"learning_rate": 0.00096875, |
|
"loss": 3.4414, |
|
"step": 31000 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"learning_rate": 0.001, |
|
"loss": 3.4282, |
|
"step": 32000 |
|
}, |
|
{ |
|
"epoch": 1.77, |
|
"learning_rate": 0.0009970579582230068, |
|
"loss": 3.427, |
|
"step": 33000 |
|
}, |
|
{ |
|
"epoch": 1.83, |
|
"learning_rate": 0.0009941159164460135, |
|
"loss": 3.4162, |
|
"step": 34000 |
|
}, |
|
{ |
|
"epoch": 1.88, |
|
"learning_rate": 0.0009911738746690203, |
|
"loss": 3.3995, |
|
"step": 35000 |
|
}, |
|
{ |
|
"epoch": 1.94, |
|
"learning_rate": 0.000988231832892027, |
|
"loss": 3.3924, |
|
"step": 36000 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"learning_rate": 0.0009852897911150338, |
|
"loss": 3.3834, |
|
"step": 37000 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.37945673622135395, |
|
"eval_loss": 3.5754754543304443, |
|
"eval_runtime": 223.4252, |
|
"eval_samples_per_second": 259.241, |
|
"eval_steps_per_second": 2.028, |
|
"step": 37190 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"learning_rate": 0.0009823477493380406, |
|
"loss": 3.3269, |
|
"step": 38000 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"learning_rate": 0.0009794057075610473, |
|
"loss": 3.3225, |
|
"step": 39000 |
|
}, |
|
{ |
|
"epoch": 2.15, |
|
"learning_rate": 0.0009764636657840541, |
|
"loss": 3.3158, |
|
"step": 40000 |
|
}, |
|
{ |
|
"epoch": 2.2, |
|
"learning_rate": 0.000973521624007061, |
|
"loss": 3.3188, |
|
"step": 41000 |
|
}, |
|
{ |
|
"epoch": 2.26, |
|
"learning_rate": 0.0009705795822300677, |
|
"loss": 3.3131, |
|
"step": 42000 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"learning_rate": 0.0009676375404530745, |
|
"loss": 3.3084, |
|
"step": 43000 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"learning_rate": 0.0009646954986760812, |
|
"loss": 3.304, |
|
"step": 44000 |
|
}, |
|
{ |
|
"epoch": 2.42, |
|
"learning_rate": 0.000961753456899088, |
|
"loss": 3.2992, |
|
"step": 45000 |
|
}, |
|
{ |
|
"epoch": 2.47, |
|
"learning_rate": 0.0009588114151220947, |
|
"loss": 3.2997, |
|
"step": 46000 |
|
}, |
|
{ |
|
"epoch": 2.53, |
|
"learning_rate": 0.0009558693733451015, |
|
"loss": 3.2882, |
|
"step": 47000 |
|
}, |
|
{ |
|
"epoch": 2.58, |
|
"learning_rate": 0.0009529273315681083, |
|
"loss": 3.2855, |
|
"step": 48000 |
|
}, |
|
{ |
|
"epoch": 2.64, |
|
"learning_rate": 0.0009499852897911151, |
|
"loss": 3.2749, |
|
"step": 49000 |
|
}, |
|
{ |
|
"epoch": 2.69, |
|
"learning_rate": 0.0009470432480141218, |
|
"loss": 3.2752, |
|
"step": 50000 |
|
}, |
|
{ |
|
"epoch": 2.74, |
|
"learning_rate": 0.0009441012062371285, |
|
"loss": 3.2708, |
|
"step": 51000 |
|
}, |
|
{ |
|
"epoch": 2.8, |
|
"learning_rate": 0.0009411591644601354, |
|
"loss": 3.2664, |
|
"step": 52000 |
|
}, |
|
{ |
|
"epoch": 2.85, |
|
"learning_rate": 0.000938217122683142, |
|
"loss": 3.2592, |
|
"step": 53000 |
|
}, |
|
{ |
|
"epoch": 2.9, |
|
"learning_rate": 0.0009352750809061489, |
|
"loss": 3.2619, |
|
"step": 54000 |
|
}, |
|
{ |
|
"epoch": 2.96, |
|
"learning_rate": 0.0009323330391291557, |
|
"loss": 3.2539, |
|
"step": 55000 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.3916382282939119, |
|
"eval_loss": 3.465026378631592, |
|
"eval_runtime": 224.5851, |
|
"eval_samples_per_second": 257.902, |
|
"eval_steps_per_second": 2.017, |
|
"step": 55785 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"learning_rate": 0.0009293909973521623, |
|
"loss": 3.237, |
|
"step": 56000 |
|
}, |
|
{ |
|
"epoch": 3.07, |
|
"learning_rate": 0.0009264489555751692, |
|
"loss": 3.1861, |
|
"step": 57000 |
|
}, |
|
{ |
|
"epoch": 3.12, |
|
"learning_rate": 0.000923506913798176, |
|
"loss": 3.1925, |
|
"step": 58000 |
|
}, |
|
{ |
|
"epoch": 3.17, |
|
"learning_rate": 0.0009205648720211828, |
|
"loss": 3.193, |
|
"step": 59000 |
|
}, |
|
{ |
|
"epoch": 3.23, |
|
"learning_rate": 0.0009176228302441895, |
|
"loss": 3.1928, |
|
"step": 60000 |
|
}, |
|
{ |
|
"epoch": 3.28, |
|
"learning_rate": 0.0009146807884671962, |
|
"loss": 3.1971, |
|
"step": 61000 |
|
}, |
|
{ |
|
"epoch": 3.33, |
|
"learning_rate": 0.0009117387466902031, |
|
"loss": 3.1926, |
|
"step": 62000 |
|
}, |
|
{ |
|
"epoch": 3.39, |
|
"learning_rate": 0.0009087967049132097, |
|
"loss": 3.1922, |
|
"step": 63000 |
|
}, |
|
{ |
|
"epoch": 3.44, |
|
"learning_rate": 0.0009058546631362166, |
|
"loss": 3.196, |
|
"step": 64000 |
|
}, |
|
{ |
|
"epoch": 3.5, |
|
"learning_rate": 0.0009029126213592234, |
|
"loss": 3.1898, |
|
"step": 65000 |
|
}, |
|
{ |
|
"epoch": 3.55, |
|
"learning_rate": 0.00089997057958223, |
|
"loss": 3.1874, |
|
"step": 66000 |
|
}, |
|
{ |
|
"epoch": 3.6, |
|
"learning_rate": 0.0008970285378052369, |
|
"loss": 3.1885, |
|
"step": 67000 |
|
}, |
|
{ |
|
"epoch": 3.66, |
|
"learning_rate": 0.0008940864960282436, |
|
"loss": 3.187, |
|
"step": 68000 |
|
}, |
|
{ |
|
"epoch": 3.71, |
|
"learning_rate": 0.0008911444542512503, |
|
"loss": 3.1892, |
|
"step": 69000 |
|
}, |
|
{ |
|
"epoch": 3.76, |
|
"learning_rate": 0.0008882024124742572, |
|
"loss": 3.1863, |
|
"step": 70000 |
|
}, |
|
{ |
|
"epoch": 3.82, |
|
"learning_rate": 0.0008852603706972639, |
|
"loss": 3.1827, |
|
"step": 71000 |
|
}, |
|
{ |
|
"epoch": 3.87, |
|
"learning_rate": 0.0008823183289202707, |
|
"loss": 3.1772, |
|
"step": 72000 |
|
}, |
|
{ |
|
"epoch": 3.93, |
|
"learning_rate": 0.0008793762871432774, |
|
"loss": 3.1803, |
|
"step": 73000 |
|
}, |
|
{ |
|
"epoch": 3.98, |
|
"learning_rate": 0.0008764342453662842, |
|
"loss": 3.1778, |
|
"step": 74000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.39928409129918474, |
|
"eval_loss": 3.4013407230377197, |
|
"eval_runtime": 224.8276, |
|
"eval_samples_per_second": 257.624, |
|
"eval_steps_per_second": 2.015, |
|
"step": 74380 |
|
}, |
|
{ |
|
"epoch": 4.03, |
|
"learning_rate": 0.000873492203589291, |
|
"loss": 3.1326, |
|
"step": 75000 |
|
}, |
|
{ |
|
"epoch": 4.09, |
|
"learning_rate": 0.0008705501618122977, |
|
"loss": 3.1121, |
|
"step": 76000 |
|
}, |
|
{ |
|
"epoch": 4.14, |
|
"learning_rate": 0.0008676081200353046, |
|
"loss": 3.1156, |
|
"step": 77000 |
|
}, |
|
{ |
|
"epoch": 4.19, |
|
"learning_rate": 0.0008646660782583113, |
|
"loss": 3.1184, |
|
"step": 78000 |
|
}, |
|
{ |
|
"epoch": 4.25, |
|
"learning_rate": 0.000861724036481318, |
|
"loss": 3.1201, |
|
"step": 79000 |
|
}, |
|
{ |
|
"epoch": 4.3, |
|
"learning_rate": 0.0008587819947043249, |
|
"loss": 3.1283, |
|
"step": 80000 |
|
}, |
|
{ |
|
"epoch": 4.36, |
|
"learning_rate": 0.0008558399529273316, |
|
"loss": 3.1222, |
|
"step": 81000 |
|
}, |
|
{ |
|
"epoch": 4.41, |
|
"learning_rate": 0.0008528979111503384, |
|
"loss": 3.128, |
|
"step": 82000 |
|
}, |
|
{ |
|
"epoch": 4.46, |
|
"learning_rate": 0.0008499558693733451, |
|
"loss": 3.1293, |
|
"step": 83000 |
|
}, |
|
{ |
|
"epoch": 4.52, |
|
"learning_rate": 0.0008470138275963519, |
|
"loss": 3.1233, |
|
"step": 84000 |
|
}, |
|
{ |
|
"epoch": 4.57, |
|
"learning_rate": 0.0008440717858193586, |
|
"loss": 3.1256, |
|
"step": 85000 |
|
}, |
|
{ |
|
"epoch": 4.62, |
|
"learning_rate": 0.0008411297440423654, |
|
"loss": 3.126, |
|
"step": 86000 |
|
}, |
|
{ |
|
"epoch": 4.68, |
|
"learning_rate": 0.0008381877022653723, |
|
"loss": 3.1279, |
|
"step": 87000 |
|
}, |
|
{ |
|
"epoch": 4.73, |
|
"learning_rate": 0.0008352456604883789, |
|
"loss": 3.1248, |
|
"step": 88000 |
|
}, |
|
{ |
|
"epoch": 4.79, |
|
"learning_rate": 0.0008323036187113857, |
|
"loss": 3.1262, |
|
"step": 89000 |
|
}, |
|
{ |
|
"epoch": 4.84, |
|
"learning_rate": 0.0008293615769343925, |
|
"loss": 3.1282, |
|
"step": 90000 |
|
}, |
|
{ |
|
"epoch": 4.89, |
|
"learning_rate": 0.0008264195351573992, |
|
"loss": 3.1215, |
|
"step": 91000 |
|
}, |
|
{ |
|
"epoch": 4.95, |
|
"learning_rate": 0.0008234774933804061, |
|
"loss": 3.1219, |
|
"step": 92000 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.4011568957772014, |
|
"eval_loss": 3.3871572017669678, |
|
"eval_runtime": 225.0611, |
|
"eval_samples_per_second": 257.357, |
|
"eval_steps_per_second": 2.013, |
|
"step": 92975 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 0.0008205354516034128, |
|
"loss": 3.116, |
|
"step": 93000 |
|
}, |
|
{ |
|
"epoch": 5.06, |
|
"learning_rate": 0.0008175934098264195, |
|
"loss": 3.0544, |
|
"step": 94000 |
|
}, |
|
{ |
|
"epoch": 5.11, |
|
"learning_rate": 0.0008146513680494263, |
|
"loss": 3.0598, |
|
"step": 95000 |
|
}, |
|
{ |
|
"epoch": 5.16, |
|
"learning_rate": 0.0008117093262724331, |
|
"loss": 3.0631, |
|
"step": 96000 |
|
}, |
|
{ |
|
"epoch": 5.22, |
|
"learning_rate": 0.0008087672844954399, |
|
"loss": 3.0724, |
|
"step": 97000 |
|
}, |
|
{ |
|
"epoch": 5.27, |
|
"learning_rate": 0.0008058252427184466, |
|
"loss": 3.0667, |
|
"step": 98000 |
|
}, |
|
{ |
|
"epoch": 5.32, |
|
"learning_rate": 0.0008028832009414534, |
|
"loss": 3.07, |
|
"step": 99000 |
|
}, |
|
{ |
|
"epoch": 5.38, |
|
"learning_rate": 0.0007999411591644602, |
|
"loss": 3.0746, |
|
"step": 100000 |
|
}, |
|
{ |
|
"epoch": 5.43, |
|
"learning_rate": 0.0007969991173874669, |
|
"loss": 3.0764, |
|
"step": 101000 |
|
}, |
|
{ |
|
"epoch": 5.49, |
|
"learning_rate": 0.0007940570756104736, |
|
"loss": 3.075, |
|
"step": 102000 |
|
}, |
|
{ |
|
"epoch": 5.54, |
|
"learning_rate": 0.0007911150338334805, |
|
"loss": 3.0812, |
|
"step": 103000 |
|
}, |
|
{ |
|
"epoch": 5.59, |
|
"learning_rate": 0.0007881729920564872, |
|
"loss": 3.0791, |
|
"step": 104000 |
|
}, |
|
{ |
|
"epoch": 5.65, |
|
"learning_rate": 0.000785230950279494, |
|
"loss": 3.0778, |
|
"step": 105000 |
|
}, |
|
{ |
|
"epoch": 5.7, |
|
"learning_rate": 0.0007822889085025008, |
|
"loss": 3.0805, |
|
"step": 106000 |
|
}, |
|
{ |
|
"epoch": 5.75, |
|
"learning_rate": 0.0007793468667255074, |
|
"loss": 3.0753, |
|
"step": 107000 |
|
}, |
|
{ |
|
"epoch": 5.81, |
|
"learning_rate": 0.0007764048249485143, |
|
"loss": 3.0851, |
|
"step": 108000 |
|
}, |
|
{ |
|
"epoch": 5.86, |
|
"learning_rate": 0.0007734627831715211, |
|
"loss": 3.0815, |
|
"step": 109000 |
|
}, |
|
{ |
|
"epoch": 5.92, |
|
"learning_rate": 0.0007705207413945278, |
|
"loss": 3.0767, |
|
"step": 110000 |
|
}, |
|
{ |
|
"epoch": 5.97, |
|
"learning_rate": 0.0007675786996175346, |
|
"loss": 3.0829, |
|
"step": 111000 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.40481396336362346, |
|
"eval_loss": 3.359344482421875, |
|
"eval_runtime": 226.3271, |
|
"eval_samples_per_second": 255.917, |
|
"eval_steps_per_second": 2.002, |
|
"step": 111570 |
|
}, |
|
{ |
|
"epoch": 6.02, |
|
"learning_rate": 0.0007646366578405413, |
|
"loss": 3.0485, |
|
"step": 112000 |
|
}, |
|
{ |
|
"epoch": 6.08, |
|
"learning_rate": 0.0007616946160635482, |
|
"loss": 3.0146, |
|
"step": 113000 |
|
}, |
|
{ |
|
"epoch": 6.13, |
|
"learning_rate": 0.0007587525742865549, |
|
"loss": 3.0181, |
|
"step": 114000 |
|
}, |
|
{ |
|
"epoch": 6.18, |
|
"learning_rate": 0.0007558105325095617, |
|
"loss": 3.0209, |
|
"step": 115000 |
|
}, |
|
{ |
|
"epoch": 6.24, |
|
"learning_rate": 0.0007528684907325685, |
|
"loss": 3.027, |
|
"step": 116000 |
|
}, |
|
{ |
|
"epoch": 6.29, |
|
"learning_rate": 0.0007499264489555751, |
|
"loss": 3.0328, |
|
"step": 117000 |
|
}, |
|
{ |
|
"epoch": 6.35, |
|
"learning_rate": 0.000746984407178582, |
|
"loss": 3.0339, |
|
"step": 118000 |
|
}, |
|
{ |
|
"epoch": 6.4, |
|
"learning_rate": 0.0007440423654015887, |
|
"loss": 3.0353, |
|
"step": 119000 |
|
}, |
|
{ |
|
"epoch": 6.45, |
|
"learning_rate": 0.0007411003236245954, |
|
"loss": 3.0357, |
|
"step": 120000 |
|
}, |
|
{ |
|
"epoch": 6.51, |
|
"learning_rate": 0.0007381582818476023, |
|
"loss": 3.0401, |
|
"step": 121000 |
|
}, |
|
{ |
|
"epoch": 6.56, |
|
"learning_rate": 0.000735216240070609, |
|
"loss": 3.04, |
|
"step": 122000 |
|
}, |
|
{ |
|
"epoch": 6.61, |
|
"learning_rate": 0.0007322741982936158, |
|
"loss": 3.0398, |
|
"step": 123000 |
|
}, |
|
{ |
|
"epoch": 6.67, |
|
"learning_rate": 0.0007293321565166225, |
|
"loss": 3.0423, |
|
"step": 124000 |
|
}, |
|
{ |
|
"epoch": 6.72, |
|
"learning_rate": 0.0007263901147396293, |
|
"loss": 3.0433, |
|
"step": 125000 |
|
}, |
|
{ |
|
"epoch": 6.78, |
|
"learning_rate": 0.0007234480729626361, |
|
"loss": 3.0416, |
|
"step": 126000 |
|
}, |
|
{ |
|
"epoch": 6.83, |
|
"learning_rate": 0.0007205060311856428, |
|
"loss": 3.0424, |
|
"step": 127000 |
|
}, |
|
{ |
|
"epoch": 6.88, |
|
"learning_rate": 0.0007175639894086497, |
|
"loss": 3.0473, |
|
"step": 128000 |
|
}, |
|
{ |
|
"epoch": 6.94, |
|
"learning_rate": 0.0007146219476316563, |
|
"loss": 3.0406, |
|
"step": 129000 |
|
}, |
|
{ |
|
"epoch": 6.99, |
|
"learning_rate": 0.0007116799058546631, |
|
"loss": 3.0448, |
|
"step": 130000 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.40765299737056315, |
|
"eval_loss": 3.347322940826416, |
|
"eval_runtime": 224.9608, |
|
"eval_samples_per_second": 257.472, |
|
"eval_steps_per_second": 2.014, |
|
"step": 130165 |
|
}, |
|
{ |
|
"epoch": 7.04, |
|
"learning_rate": 0.00070873786407767, |
|
"loss": 2.9818, |
|
"step": 131000 |
|
}, |
|
{ |
|
"epoch": 7.1, |
|
"learning_rate": 0.0007057958223006767, |
|
"loss": 2.9802, |
|
"step": 132000 |
|
}, |
|
{ |
|
"epoch": 7.15, |
|
"learning_rate": 0.0007028537805236835, |
|
"loss": 2.9845, |
|
"step": 133000 |
|
}, |
|
{ |
|
"epoch": 7.21, |
|
"learning_rate": 0.0006999117387466902, |
|
"loss": 2.9942, |
|
"step": 134000 |
|
}, |
|
{ |
|
"epoch": 7.26, |
|
"learning_rate": 0.000696969696969697, |
|
"loss": 2.9957, |
|
"step": 135000 |
|
}, |
|
{ |
|
"epoch": 7.31, |
|
"learning_rate": 0.0006940276551927037, |
|
"loss": 2.9952, |
|
"step": 136000 |
|
}, |
|
{ |
|
"epoch": 7.37, |
|
"learning_rate": 0.0006910856134157105, |
|
"loss": 2.9985, |
|
"step": 137000 |
|
}, |
|
{ |
|
"epoch": 7.42, |
|
"learning_rate": 0.0006881435716387174, |
|
"loss": 3.0067, |
|
"step": 138000 |
|
}, |
|
{ |
|
"epoch": 7.48, |
|
"learning_rate": 0.000685201529861724, |
|
"loss": 3.0054, |
|
"step": 139000 |
|
}, |
|
{ |
|
"epoch": 7.53, |
|
"learning_rate": 0.0006822594880847308, |
|
"loss": 3.0052, |
|
"step": 140000 |
|
}, |
|
{ |
|
"epoch": 7.58, |
|
"learning_rate": 0.0006793174463077376, |
|
"loss": 3.0097, |
|
"step": 141000 |
|
}, |
|
{ |
|
"epoch": 7.64, |
|
"learning_rate": 0.0006763754045307443, |
|
"loss": 3.0043, |
|
"step": 142000 |
|
}, |
|
{ |
|
"epoch": 7.69, |
|
"learning_rate": 0.0006734333627537511, |
|
"loss": 3.0131, |
|
"step": 143000 |
|
}, |
|
{ |
|
"epoch": 7.74, |
|
"learning_rate": 0.0006704913209767579, |
|
"loss": 3.0087, |
|
"step": 144000 |
|
}, |
|
{ |
|
"epoch": 7.8, |
|
"learning_rate": 0.0006675492791997646, |
|
"loss": 3.0118, |
|
"step": 145000 |
|
}, |
|
{ |
|
"epoch": 7.85, |
|
"learning_rate": 0.0006646072374227714, |
|
"loss": 3.0117, |
|
"step": 146000 |
|
}, |
|
{ |
|
"epoch": 7.91, |
|
"learning_rate": 0.0006616651956457782, |
|
"loss": 3.0117, |
|
"step": 147000 |
|
}, |
|
{ |
|
"epoch": 7.96, |
|
"learning_rate": 0.0006587231538687848, |
|
"loss": 3.0113, |
|
"step": 148000 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.40743070344640225, |
|
"eval_loss": 3.3619937896728516, |
|
"eval_runtime": 225.9207, |
|
"eval_samples_per_second": 256.378, |
|
"eval_steps_per_second": 2.005, |
|
"step": 148760 |
|
}, |
|
{ |
|
"epoch": 8.01, |
|
"learning_rate": 0.0006557811120917917, |
|
"loss": 2.9963, |
|
"step": 149000 |
|
}, |
|
{ |
|
"epoch": 8.07, |
|
"learning_rate": 0.0006528390703147985, |
|
"loss": 2.9487, |
|
"step": 150000 |
|
}, |
|
{ |
|
"epoch": 8.12, |
|
"learning_rate": 0.0006498970285378053, |
|
"loss": 2.9544, |
|
"step": 151000 |
|
}, |
|
{ |
|
"epoch": 8.17, |
|
"learning_rate": 0.000646954986760812, |
|
"loss": 2.9617, |
|
"step": 152000 |
|
}, |
|
{ |
|
"epoch": 8.23, |
|
"learning_rate": 0.0006440129449838187, |
|
"loss": 2.9614, |
|
"step": 153000 |
|
}, |
|
{ |
|
"epoch": 8.28, |
|
"learning_rate": 0.0006410709032068256, |
|
"loss": 2.9667, |
|
"step": 154000 |
|
}, |
|
{ |
|
"epoch": 8.34, |
|
"learning_rate": 0.0006381288614298323, |
|
"loss": 2.9703, |
|
"step": 155000 |
|
}, |
|
{ |
|
"epoch": 8.39, |
|
"learning_rate": 0.0006351868196528391, |
|
"loss": 2.9682, |
|
"step": 156000 |
|
}, |
|
{ |
|
"epoch": 8.44, |
|
"learning_rate": 0.0006322447778758459, |
|
"loss": 2.9722, |
|
"step": 157000 |
|
}, |
|
{ |
|
"epoch": 8.5, |
|
"learning_rate": 0.0006293027360988525, |
|
"loss": 2.9739, |
|
"step": 158000 |
|
}, |
|
{ |
|
"epoch": 8.55, |
|
"learning_rate": 0.0006263606943218594, |
|
"loss": 2.9805, |
|
"step": 159000 |
|
}, |
|
{ |
|
"epoch": 8.6, |
|
"learning_rate": 0.0006234186525448662, |
|
"loss": 2.9768, |
|
"step": 160000 |
|
}, |
|
{ |
|
"epoch": 8.66, |
|
"learning_rate": 0.0006204766107678729, |
|
"loss": 2.9796, |
|
"step": 161000 |
|
}, |
|
{ |
|
"epoch": 8.71, |
|
"learning_rate": 0.0006175345689908797, |
|
"loss": 2.9839, |
|
"step": 162000 |
|
}, |
|
{ |
|
"epoch": 8.77, |
|
"learning_rate": 0.0006145925272138864, |
|
"loss": 2.9793, |
|
"step": 163000 |
|
}, |
|
{ |
|
"epoch": 8.82, |
|
"learning_rate": 0.0006116504854368932, |
|
"loss": 2.9824, |
|
"step": 164000 |
|
}, |
|
{ |
|
"epoch": 8.87, |
|
"learning_rate": 0.0006087084436599, |
|
"loss": 2.983, |
|
"step": 165000 |
|
}, |
|
{ |
|
"epoch": 8.93, |
|
"learning_rate": 0.0006057664018829068, |
|
"loss": 2.9841, |
|
"step": 166000 |
|
}, |
|
{ |
|
"epoch": 8.98, |
|
"learning_rate": 0.0006028243601059136, |
|
"loss": 2.9876, |
|
"step": 167000 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.4083816834374635, |
|
"eval_loss": 3.3487634658813477, |
|
"eval_runtime": 224.2026, |
|
"eval_samples_per_second": 258.342, |
|
"eval_steps_per_second": 2.02, |
|
"step": 167355 |
|
}, |
|
{ |
|
"epoch": 9.03, |
|
"learning_rate": 0.0005998823183289202, |
|
"loss": 2.9442, |
|
"step": 168000 |
|
}, |
|
{ |
|
"epoch": 9.09, |
|
"learning_rate": 0.0005969402765519271, |
|
"loss": 2.9246, |
|
"step": 169000 |
|
}, |
|
{ |
|
"epoch": 9.14, |
|
"learning_rate": 0.0005939982347749339, |
|
"loss": 2.9274, |
|
"step": 170000 |
|
}, |
|
{ |
|
"epoch": 9.2, |
|
"learning_rate": 0.0005910561929979405, |
|
"loss": 2.9331, |
|
"step": 171000 |
|
}, |
|
{ |
|
"epoch": 9.25, |
|
"learning_rate": 0.0005881141512209474, |
|
"loss": 2.937, |
|
"step": 172000 |
|
}, |
|
{ |
|
"epoch": 9.3, |
|
"learning_rate": 0.0005851721094439541, |
|
"loss": 2.9395, |
|
"step": 173000 |
|
}, |
|
{ |
|
"epoch": 9.36, |
|
"learning_rate": 0.0005822300676669609, |
|
"loss": 2.9441, |
|
"step": 174000 |
|
}, |
|
{ |
|
"epoch": 9.41, |
|
"learning_rate": 0.0005792880258899676, |
|
"loss": 2.9432, |
|
"step": 175000 |
|
}, |
|
{ |
|
"epoch": 9.46, |
|
"learning_rate": 0.0005763459841129744, |
|
"loss": 2.9466, |
|
"step": 176000 |
|
}, |
|
{ |
|
"epoch": 9.52, |
|
"learning_rate": 0.0005734039423359812, |
|
"loss": 2.9496, |
|
"step": 177000 |
|
}, |
|
{ |
|
"epoch": 9.57, |
|
"learning_rate": 0.0005704619005589879, |
|
"loss": 2.9494, |
|
"step": 178000 |
|
}, |
|
{ |
|
"epoch": 9.63, |
|
"learning_rate": 0.0005675198587819948, |
|
"loss": 2.9545, |
|
"step": 179000 |
|
}, |
|
{ |
|
"epoch": 9.68, |
|
"learning_rate": 0.0005645778170050014, |
|
"loss": 2.9543, |
|
"step": 180000 |
|
}, |
|
{ |
|
"epoch": 9.73, |
|
"learning_rate": 0.0005616357752280082, |
|
"loss": 2.956, |
|
"step": 181000 |
|
}, |
|
{ |
|
"epoch": 9.79, |
|
"learning_rate": 0.0005586937334510151, |
|
"loss": 2.9582, |
|
"step": 182000 |
|
}, |
|
{ |
|
"epoch": 9.84, |
|
"learning_rate": 0.0005557516916740217, |
|
"loss": 2.9595, |
|
"step": 183000 |
|
}, |
|
{ |
|
"epoch": 9.9, |
|
"learning_rate": 0.0005528096498970286, |
|
"loss": 2.9613, |
|
"step": 184000 |
|
}, |
|
{ |
|
"epoch": 9.95, |
|
"learning_rate": 0.0005498676081200353, |
|
"loss": 2.961, |
|
"step": 185000 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.4087031329470162, |
|
"eval_loss": 3.3556947708129883, |
|
"eval_runtime": 224.7185, |
|
"eval_samples_per_second": 257.749, |
|
"eval_steps_per_second": 2.016, |
|
"step": 185950 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 0.0005469255663430421, |
|
"loss": 2.9585, |
|
"step": 186000 |
|
}, |
|
{ |
|
"epoch": 10.06, |
|
"learning_rate": 0.0005439835245660489, |
|
"loss": 2.8923, |
|
"step": 187000 |
|
}, |
|
{ |
|
"epoch": 10.11, |
|
"learning_rate": 0.0005410414827890556, |
|
"loss": 2.9022, |
|
"step": 188000 |
|
}, |
|
{ |
|
"epoch": 10.16, |
|
"learning_rate": 0.0005380994410120625, |
|
"loss": 2.9089, |
|
"step": 189000 |
|
}, |
|
{ |
|
"epoch": 10.22, |
|
"learning_rate": 0.0005351573992350691, |
|
"loss": 2.9148, |
|
"step": 190000 |
|
}, |
|
{ |
|
"epoch": 10.27, |
|
"learning_rate": 0.0005322153574580759, |
|
"loss": 2.9124, |
|
"step": 191000 |
|
}, |
|
{ |
|
"epoch": 10.33, |
|
"learning_rate": 0.0005292733156810828, |
|
"loss": 2.9155, |
|
"step": 192000 |
|
}, |
|
{ |
|
"epoch": 10.38, |
|
"learning_rate": 0.0005263312739040894, |
|
"loss": 2.9206, |
|
"step": 193000 |
|
}, |
|
{ |
|
"epoch": 10.43, |
|
"learning_rate": 0.0005233892321270962, |
|
"loss": 2.9236, |
|
"step": 194000 |
|
}, |
|
{ |
|
"epoch": 10.49, |
|
"learning_rate": 0.000520447190350103, |
|
"loss": 2.9245, |
|
"step": 195000 |
|
}, |
|
{ |
|
"epoch": 10.54, |
|
"learning_rate": 0.0005175051485731097, |
|
"loss": 2.9219, |
|
"step": 196000 |
|
}, |
|
{ |
|
"epoch": 10.59, |
|
"learning_rate": 0.0005145631067961165, |
|
"loss": 2.9295, |
|
"step": 197000 |
|
}, |
|
{ |
|
"epoch": 10.65, |
|
"learning_rate": 0.0005116210650191233, |
|
"loss": 2.9283, |
|
"step": 198000 |
|
}, |
|
{ |
|
"epoch": 10.7, |
|
"learning_rate": 0.00050867902324213, |
|
"loss": 2.9306, |
|
"step": 199000 |
|
}, |
|
{ |
|
"epoch": 10.76, |
|
"learning_rate": 0.0005057369814651368, |
|
"loss": 2.9344, |
|
"step": 200000 |
|
}, |
|
{ |
|
"epoch": 10.81, |
|
"learning_rate": 0.0005027949396881436, |
|
"loss": 2.9343, |
|
"step": 201000 |
|
}, |
|
{ |
|
"epoch": 10.86, |
|
"learning_rate": 0.0004998528979111503, |
|
"loss": 2.9381, |
|
"step": 202000 |
|
}, |
|
{ |
|
"epoch": 10.92, |
|
"learning_rate": 0.0004969108561341571, |
|
"loss": 2.9374, |
|
"step": 203000 |
|
}, |
|
{ |
|
"epoch": 10.97, |
|
"learning_rate": 0.0004939688143571639, |
|
"loss": 2.9378, |
|
"step": 204000 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"eval_accuracy": 0.40959009175049044, |
|
"eval_loss": 3.359853744506836, |
|
"eval_runtime": 225.3675, |
|
"eval_samples_per_second": 257.007, |
|
"eval_steps_per_second": 2.01, |
|
"step": 204545 |
|
}, |
|
{ |
|
"epoch": 11.02, |
|
"learning_rate": 0.0004910267725801706, |
|
"loss": 2.9112, |
|
"step": 205000 |
|
}, |
|
{ |
|
"epoch": 11.08, |
|
"learning_rate": 0.00048808473080317737, |
|
"loss": 2.8779, |
|
"step": 206000 |
|
}, |
|
{ |
|
"epoch": 11.13, |
|
"learning_rate": 0.0004851426890261842, |
|
"loss": 2.8836, |
|
"step": 207000 |
|
}, |
|
{ |
|
"epoch": 11.19, |
|
"learning_rate": 0.00048220064724919094, |
|
"loss": 2.8859, |
|
"step": 208000 |
|
}, |
|
{ |
|
"epoch": 11.24, |
|
"learning_rate": 0.00047925860547219775, |
|
"loss": 2.8924, |
|
"step": 209000 |
|
}, |
|
{ |
|
"epoch": 11.29, |
|
"learning_rate": 0.0004763165636952045, |
|
"loss": 2.8916, |
|
"step": 210000 |
|
}, |
|
{ |
|
"epoch": 11.35, |
|
"learning_rate": 0.0004733745219182112, |
|
"loss": 2.8993, |
|
"step": 211000 |
|
}, |
|
{ |
|
"epoch": 11.4, |
|
"learning_rate": 0.00047043248014121803, |
|
"loss": 2.8945, |
|
"step": 212000 |
|
}, |
|
{ |
|
"epoch": 11.45, |
|
"learning_rate": 0.0004674904383642248, |
|
"loss": 2.9001, |
|
"step": 213000 |
|
}, |
|
{ |
|
"epoch": 11.51, |
|
"learning_rate": 0.00046454839658723155, |
|
"loss": 2.9025, |
|
"step": 214000 |
|
}, |
|
{ |
|
"epoch": 11.56, |
|
"learning_rate": 0.0004616063548102383, |
|
"loss": 2.9072, |
|
"step": 215000 |
|
}, |
|
{ |
|
"epoch": 11.62, |
|
"learning_rate": 0.00045866431303324506, |
|
"loss": 2.908, |
|
"step": 216000 |
|
}, |
|
{ |
|
"epoch": 11.67, |
|
"learning_rate": 0.0004557222712562519, |
|
"loss": 2.9057, |
|
"step": 217000 |
|
}, |
|
{ |
|
"epoch": 11.72, |
|
"learning_rate": 0.00045278022947925863, |
|
"loss": 2.9115, |
|
"step": 218000 |
|
}, |
|
{ |
|
"epoch": 11.78, |
|
"learning_rate": 0.0004498381877022654, |
|
"loss": 2.9082, |
|
"step": 219000 |
|
}, |
|
{ |
|
"epoch": 11.83, |
|
"learning_rate": 0.00044689614592527215, |
|
"loss": 2.9129, |
|
"step": 220000 |
|
}, |
|
{ |
|
"epoch": 11.88, |
|
"learning_rate": 0.0004439541041482789, |
|
"loss": 2.9145, |
|
"step": 221000 |
|
}, |
|
{ |
|
"epoch": 11.94, |
|
"learning_rate": 0.00044101206237128567, |
|
"loss": 2.9153, |
|
"step": 222000 |
|
}, |
|
{ |
|
"epoch": 11.99, |
|
"learning_rate": 0.0004380700205942925, |
|
"loss": 2.9178, |
|
"step": 223000 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.41031938242461874, |
|
"eval_loss": 3.3431034088134766, |
|
"eval_runtime": 225.2573, |
|
"eval_samples_per_second": 257.133, |
|
"eval_steps_per_second": 2.011, |
|
"step": 223140 |
|
}, |
|
{ |
|
"epoch": 12.05, |
|
"learning_rate": 0.0004351279788172992, |
|
"loss": 2.8578, |
|
"step": 224000 |
|
}, |
|
{ |
|
"epoch": 12.1, |
|
"learning_rate": 0.00043218593704030594, |
|
"loss": 2.8647, |
|
"step": 225000 |
|
}, |
|
{ |
|
"epoch": 12.15, |
|
"learning_rate": 0.00042924389526331275, |
|
"loss": 2.8661, |
|
"step": 226000 |
|
}, |
|
{ |
|
"epoch": 12.21, |
|
"learning_rate": 0.0004263018534863195, |
|
"loss": 2.8714, |
|
"step": 227000 |
|
}, |
|
{ |
|
"epoch": 12.26, |
|
"learning_rate": 0.0004233598117093263, |
|
"loss": 2.8709, |
|
"step": 228000 |
|
}, |
|
{ |
|
"epoch": 12.32, |
|
"learning_rate": 0.00042041776993233303, |
|
"loss": 2.8761, |
|
"step": 229000 |
|
}, |
|
{ |
|
"epoch": 12.37, |
|
"learning_rate": 0.0004174757281553398, |
|
"loss": 2.873, |
|
"step": 230000 |
|
}, |
|
{ |
|
"epoch": 12.42, |
|
"learning_rate": 0.0004145336863783466, |
|
"loss": 2.8839, |
|
"step": 231000 |
|
}, |
|
{ |
|
"epoch": 12.48, |
|
"learning_rate": 0.00041159164460135336, |
|
"loss": 2.8799, |
|
"step": 232000 |
|
}, |
|
{ |
|
"epoch": 12.53, |
|
"learning_rate": 0.0004086496028243601, |
|
"loss": 2.8806, |
|
"step": 233000 |
|
}, |
|
{ |
|
"epoch": 12.58, |
|
"learning_rate": 0.0004057075610473669, |
|
"loss": 2.8864, |
|
"step": 234000 |
|
}, |
|
{ |
|
"epoch": 12.64, |
|
"learning_rate": 0.00040276551927037363, |
|
"loss": 2.8875, |
|
"step": 235000 |
|
}, |
|
{ |
|
"epoch": 12.69, |
|
"learning_rate": 0.00039982347749338044, |
|
"loss": 2.8879, |
|
"step": 236000 |
|
}, |
|
{ |
|
"epoch": 12.75, |
|
"learning_rate": 0.0003968814357163872, |
|
"loss": 2.8885, |
|
"step": 237000 |
|
}, |
|
{ |
|
"epoch": 12.8, |
|
"learning_rate": 0.0003939393939393939, |
|
"loss": 2.8938, |
|
"step": 238000 |
|
}, |
|
{ |
|
"epoch": 12.85, |
|
"learning_rate": 0.0003909973521624007, |
|
"loss": 2.8887, |
|
"step": 239000 |
|
}, |
|
{ |
|
"epoch": 12.91, |
|
"learning_rate": 0.0003880553103854075, |
|
"loss": 2.8892, |
|
"step": 240000 |
|
}, |
|
{ |
|
"epoch": 12.96, |
|
"learning_rate": 0.00038511326860841424, |
|
"loss": 2.8924, |
|
"step": 241000 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"eval_accuracy": 0.4108899972906878, |
|
"eval_loss": 3.3626227378845215, |
|
"eval_runtime": 224.5071, |
|
"eval_samples_per_second": 257.992, |
|
"eval_steps_per_second": 2.018, |
|
"step": 241735 |
|
}, |
|
{ |
|
"epoch": 13.01, |
|
"learning_rate": 0.00038217122683142105, |
|
"loss": 2.8765, |
|
"step": 242000 |
|
}, |
|
{ |
|
"epoch": 13.07, |
|
"learning_rate": 0.00037922918505442775, |
|
"loss": 2.8377, |
|
"step": 243000 |
|
}, |
|
{ |
|
"epoch": 13.12, |
|
"learning_rate": 0.00037628714327743456, |
|
"loss": 2.8411, |
|
"step": 244000 |
|
}, |
|
{ |
|
"epoch": 13.18, |
|
"learning_rate": 0.0003733451015004413, |
|
"loss": 2.8425, |
|
"step": 245000 |
|
}, |
|
{ |
|
"epoch": 13.23, |
|
"learning_rate": 0.0003704030597234481, |
|
"loss": 2.8521, |
|
"step": 246000 |
|
}, |
|
{ |
|
"epoch": 13.28, |
|
"learning_rate": 0.0003674610179464549, |
|
"loss": 2.8537, |
|
"step": 247000 |
|
}, |
|
{ |
|
"epoch": 13.34, |
|
"learning_rate": 0.0003645189761694616, |
|
"loss": 2.8564, |
|
"step": 248000 |
|
}, |
|
{ |
|
"epoch": 13.39, |
|
"learning_rate": 0.00036157693439246836, |
|
"loss": 2.8607, |
|
"step": 249000 |
|
}, |
|
{ |
|
"epoch": 13.44, |
|
"learning_rate": 0.00035863489261547517, |
|
"loss": 2.8609, |
|
"step": 250000 |
|
}, |
|
{ |
|
"epoch": 13.5, |
|
"learning_rate": 0.0003556928508384819, |
|
"loss": 2.8622, |
|
"step": 251000 |
|
}, |
|
{ |
|
"epoch": 13.55, |
|
"learning_rate": 0.00035275080906148863, |
|
"loss": 2.8638, |
|
"step": 252000 |
|
}, |
|
{ |
|
"epoch": 13.61, |
|
"learning_rate": 0.00034980876728449544, |
|
"loss": 2.8661, |
|
"step": 253000 |
|
}, |
|
{ |
|
"epoch": 13.66, |
|
"learning_rate": 0.0003468667255075022, |
|
"loss": 2.8676, |
|
"step": 254000 |
|
}, |
|
{ |
|
"epoch": 13.71, |
|
"learning_rate": 0.000343924683730509, |
|
"loss": 2.8689, |
|
"step": 255000 |
|
}, |
|
{ |
|
"epoch": 13.77, |
|
"learning_rate": 0.00034098264195351577, |
|
"loss": 2.8704, |
|
"step": 256000 |
|
}, |
|
{ |
|
"epoch": 13.82, |
|
"learning_rate": 0.0003380406001765225, |
|
"loss": 2.8734, |
|
"step": 257000 |
|
}, |
|
{ |
|
"epoch": 13.87, |
|
"learning_rate": 0.0003350985583995293, |
|
"loss": 2.8705, |
|
"step": 258000 |
|
}, |
|
{ |
|
"epoch": 13.93, |
|
"learning_rate": 0.00033215651662253605, |
|
"loss": 2.8739, |
|
"step": 259000 |
|
}, |
|
{ |
|
"epoch": 13.98, |
|
"learning_rate": 0.0003292144748455428, |
|
"loss": 2.8737, |
|
"step": 260000 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_accuracy": 0.4114068692920459, |
|
"eval_loss": 3.3568508625030518, |
|
"eval_runtime": 225.3346, |
|
"eval_samples_per_second": 257.044, |
|
"eval_steps_per_second": 2.01, |
|
"step": 260330 |
|
}, |
|
{ |
|
"epoch": 14.04, |
|
"learning_rate": 0.00032627243306854956, |
|
"loss": 2.8336, |
|
"step": 261000 |
|
}, |
|
{ |
|
"epoch": 14.09, |
|
"learning_rate": 0.0003233303912915563, |
|
"loss": 2.8234, |
|
"step": 262000 |
|
}, |
|
{ |
|
"epoch": 14.14, |
|
"learning_rate": 0.00032038834951456313, |
|
"loss": 2.8274, |
|
"step": 263000 |
|
}, |
|
{ |
|
"epoch": 14.2, |
|
"learning_rate": 0.0003174463077375699, |
|
"loss": 2.8291, |
|
"step": 264000 |
|
}, |
|
{ |
|
"epoch": 14.25, |
|
"learning_rate": 0.00031450426596057665, |
|
"loss": 2.836, |
|
"step": 265000 |
|
}, |
|
{ |
|
"epoch": 14.3, |
|
"learning_rate": 0.0003115622241835834, |
|
"loss": 2.8393, |
|
"step": 266000 |
|
}, |
|
{ |
|
"epoch": 14.36, |
|
"learning_rate": 0.00030862018240659017, |
|
"loss": 2.839, |
|
"step": 267000 |
|
}, |
|
{ |
|
"epoch": 14.41, |
|
"learning_rate": 0.0003056781406295969, |
|
"loss": 2.8416, |
|
"step": 268000 |
|
}, |
|
{ |
|
"epoch": 14.47, |
|
"learning_rate": 0.00030273609885260374, |
|
"loss": 2.8408, |
|
"step": 269000 |
|
}, |
|
{ |
|
"epoch": 14.52, |
|
"learning_rate": 0.0002997940570756105, |
|
"loss": 2.844, |
|
"step": 270000 |
|
}, |
|
{ |
|
"epoch": 14.57, |
|
"learning_rate": 0.00029685201529861725, |
|
"loss": 2.8446, |
|
"step": 271000 |
|
}, |
|
{ |
|
"epoch": 14.63, |
|
"learning_rate": 0.000293909973521624, |
|
"loss": 2.8471, |
|
"step": 272000 |
|
}, |
|
{ |
|
"epoch": 14.68, |
|
"learning_rate": 0.00029096793174463077, |
|
"loss": 2.8491, |
|
"step": 273000 |
|
}, |
|
{ |
|
"epoch": 14.74, |
|
"learning_rate": 0.0002880258899676376, |
|
"loss": 2.8497, |
|
"step": 274000 |
|
}, |
|
{ |
|
"epoch": 14.79, |
|
"learning_rate": 0.0002850838481906443, |
|
"loss": 2.8496, |
|
"step": 275000 |
|
}, |
|
{ |
|
"epoch": 14.84, |
|
"learning_rate": 0.00028214180641365105, |
|
"loss": 2.8512, |
|
"step": 276000 |
|
}, |
|
{ |
|
"epoch": 14.9, |
|
"learning_rate": 0.00027919976463665786, |
|
"loss": 2.8562, |
|
"step": 277000 |
|
}, |
|
{ |
|
"epoch": 14.95, |
|
"learning_rate": 0.0002762577228596646, |
|
"loss": 2.8492, |
|
"step": 278000 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"eval_accuracy": 0.4112024448704014, |
|
"eval_loss": 3.359900712966919, |
|
"eval_runtime": 226.1495, |
|
"eval_samples_per_second": 256.118, |
|
"eval_steps_per_second": 2.003, |
|
"step": 278925 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"learning_rate": 0.0002733156810826714, |
|
"loss": 2.8496, |
|
"step": 279000 |
|
}, |
|
{ |
|
"epoch": 15.06, |
|
"learning_rate": 0.00027037363930567813, |
|
"loss": 2.8052, |
|
"step": 280000 |
|
}, |
|
{ |
|
"epoch": 15.11, |
|
"learning_rate": 0.0002674315975286849, |
|
"loss": 2.8082, |
|
"step": 281000 |
|
}, |
|
{ |
|
"epoch": 15.17, |
|
"learning_rate": 0.0002644895557516917, |
|
"loss": 2.8147, |
|
"step": 282000 |
|
}, |
|
{ |
|
"epoch": 15.22, |
|
"learning_rate": 0.00026154751397469846, |
|
"loss": 2.8129, |
|
"step": 283000 |
|
}, |
|
{ |
|
"epoch": 15.27, |
|
"learning_rate": 0.00025860547219770517, |
|
"loss": 2.8164, |
|
"step": 284000 |
|
}, |
|
{ |
|
"epoch": 15.33, |
|
"learning_rate": 0.000255663430420712, |
|
"loss": 2.8224, |
|
"step": 285000 |
|
}, |
|
{ |
|
"epoch": 15.38, |
|
"learning_rate": 0.00025272138864371874, |
|
"loss": 2.818, |
|
"step": 286000 |
|
}, |
|
{ |
|
"epoch": 15.43, |
|
"learning_rate": 0.0002497793468667255, |
|
"loss": 2.8244, |
|
"step": 287000 |
|
}, |
|
{ |
|
"epoch": 15.49, |
|
"learning_rate": 0.0002468373050897323, |
|
"loss": 2.8188, |
|
"step": 288000 |
|
}, |
|
{ |
|
"epoch": 15.54, |
|
"learning_rate": 0.00024389526331273904, |
|
"loss": 2.8261, |
|
"step": 289000 |
|
}, |
|
{ |
|
"epoch": 15.6, |
|
"learning_rate": 0.0002409532215357458, |
|
"loss": 2.8273, |
|
"step": 290000 |
|
}, |
|
{ |
|
"epoch": 15.65, |
|
"learning_rate": 0.00023801117975875258, |
|
"loss": 2.8247, |
|
"step": 291000 |
|
}, |
|
{ |
|
"epoch": 15.7, |
|
"learning_rate": 0.00023506913798175937, |
|
"loss": 2.8368, |
|
"step": 292000 |
|
}, |
|
{ |
|
"epoch": 15.76, |
|
"learning_rate": 0.0002321270962047661, |
|
"loss": 2.8333, |
|
"step": 293000 |
|
}, |
|
{ |
|
"epoch": 15.81, |
|
"learning_rate": 0.00022918505442777289, |
|
"loss": 2.834, |
|
"step": 294000 |
|
}, |
|
{ |
|
"epoch": 15.86, |
|
"learning_rate": 0.00022624301265077964, |
|
"loss": 2.83, |
|
"step": 295000 |
|
}, |
|
{ |
|
"epoch": 15.92, |
|
"learning_rate": 0.00022330097087378643, |
|
"loss": 2.831, |
|
"step": 296000 |
|
}, |
|
{ |
|
"epoch": 15.97, |
|
"learning_rate": 0.00022035892909679316, |
|
"loss": 2.8356, |
|
"step": 297000 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_accuracy": 0.41057862456826844, |
|
"eval_loss": 3.3829264640808105, |
|
"eval_runtime": 225.1635, |
|
"eval_samples_per_second": 257.24, |
|
"eval_steps_per_second": 2.012, |
|
"step": 297520 |
|
}, |
|
{ |
|
"epoch": 16.03, |
|
"learning_rate": 0.00021741688731979995, |
|
"loss": 2.8149, |
|
"step": 298000 |
|
}, |
|
{ |
|
"epoch": 16.08, |
|
"learning_rate": 0.00021447484554280673, |
|
"loss": 2.7935, |
|
"step": 299000 |
|
}, |
|
{ |
|
"epoch": 16.13, |
|
"learning_rate": 0.0002115328037658135, |
|
"loss": 2.7946, |
|
"step": 300000 |
|
}, |
|
{ |
|
"epoch": 16.19, |
|
"learning_rate": 0.00020859076198882025, |
|
"loss": 2.7958, |
|
"step": 301000 |
|
}, |
|
{ |
|
"epoch": 16.24, |
|
"learning_rate": 0.000205648720211827, |
|
"loss": 2.798, |
|
"step": 302000 |
|
}, |
|
{ |
|
"epoch": 16.29, |
|
"learning_rate": 0.0002027066784348338, |
|
"loss": 2.802, |
|
"step": 303000 |
|
}, |
|
{ |
|
"epoch": 16.35, |
|
"learning_rate": 0.00019976463665784052, |
|
"loss": 2.8043, |
|
"step": 304000 |
|
}, |
|
{ |
|
"epoch": 16.4, |
|
"learning_rate": 0.0001968225948808473, |
|
"loss": 2.8092, |
|
"step": 305000 |
|
}, |
|
{ |
|
"epoch": 16.46, |
|
"learning_rate": 0.0001938805531038541, |
|
"loss": 2.806, |
|
"step": 306000 |
|
}, |
|
{ |
|
"epoch": 16.51, |
|
"learning_rate": 0.00019093851132686085, |
|
"loss": 2.8097, |
|
"step": 307000 |
|
}, |
|
{ |
|
"epoch": 16.56, |
|
"learning_rate": 0.0001879964695498676, |
|
"loss": 2.8087, |
|
"step": 308000 |
|
}, |
|
{ |
|
"epoch": 16.62, |
|
"learning_rate": 0.00018505442777287437, |
|
"loss": 2.8105, |
|
"step": 309000 |
|
}, |
|
{ |
|
"epoch": 16.67, |
|
"learning_rate": 0.00018211238599588115, |
|
"loss": 2.8141, |
|
"step": 310000 |
|
}, |
|
{ |
|
"epoch": 16.72, |
|
"learning_rate": 0.0001791703442188879, |
|
"loss": 2.8134, |
|
"step": 311000 |
|
}, |
|
{ |
|
"epoch": 16.78, |
|
"learning_rate": 0.00017622830244189467, |
|
"loss": 2.8076, |
|
"step": 312000 |
|
}, |
|
{ |
|
"epoch": 16.83, |
|
"learning_rate": 0.00017328626066490143, |
|
"loss": 2.8149, |
|
"step": 313000 |
|
}, |
|
{ |
|
"epoch": 16.89, |
|
"learning_rate": 0.0001703442188879082, |
|
"loss": 2.8099, |
|
"step": 314000 |
|
}, |
|
{ |
|
"epoch": 16.94, |
|
"learning_rate": 0.000167402177110915, |
|
"loss": 2.8156, |
|
"step": 315000 |
|
}, |
|
{ |
|
"epoch": 16.99, |
|
"learning_rate": 0.00016446013533392173, |
|
"loss": 2.8162, |
|
"step": 316000 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"eval_accuracy": 0.4110514274205635, |
|
"eval_loss": 3.38189697265625, |
|
"eval_runtime": 223.7075, |
|
"eval_samples_per_second": 258.914, |
|
"eval_steps_per_second": 2.025, |
|
"step": 316115 |
|
}, |
|
{ |
|
"epoch": 17.05, |
|
"learning_rate": 0.00016151809355692852, |
|
"loss": 2.7837, |
|
"step": 317000 |
|
}, |
|
{ |
|
"epoch": 17.1, |
|
"learning_rate": 0.00015857605177993527, |
|
"loss": 2.7806, |
|
"step": 318000 |
|
}, |
|
{ |
|
"epoch": 17.16, |
|
"learning_rate": 0.00015563401000294206, |
|
"loss": 2.7838, |
|
"step": 319000 |
|
}, |
|
{ |
|
"epoch": 17.21, |
|
"learning_rate": 0.0001526919682259488, |
|
"loss": 2.7806, |
|
"step": 320000 |
|
}, |
|
{ |
|
"epoch": 17.26, |
|
"learning_rate": 0.00014974992644895558, |
|
"loss": 2.7854, |
|
"step": 321000 |
|
}, |
|
{ |
|
"epoch": 17.32, |
|
"learning_rate": 0.00014680788467196236, |
|
"loss": 2.7837, |
|
"step": 322000 |
|
}, |
|
{ |
|
"epoch": 17.37, |
|
"learning_rate": 0.00014386584289496912, |
|
"loss": 2.7902, |
|
"step": 323000 |
|
}, |
|
{ |
|
"epoch": 17.42, |
|
"learning_rate": 0.00014092380111797588, |
|
"loss": 2.7881, |
|
"step": 324000 |
|
}, |
|
{ |
|
"epoch": 17.48, |
|
"learning_rate": 0.00013798175934098264, |
|
"loss": 2.791, |
|
"step": 325000 |
|
}, |
|
{ |
|
"epoch": 17.53, |
|
"learning_rate": 0.00013503971756398942, |
|
"loss": 2.7905, |
|
"step": 326000 |
|
}, |
|
{ |
|
"epoch": 17.59, |
|
"learning_rate": 0.00013209767578699618, |
|
"loss": 2.7923, |
|
"step": 327000 |
|
}, |
|
{ |
|
"epoch": 17.64, |
|
"learning_rate": 0.00012915563401000294, |
|
"loss": 2.7942, |
|
"step": 328000 |
|
}, |
|
{ |
|
"epoch": 17.69, |
|
"learning_rate": 0.00012621359223300972, |
|
"loss": 2.7939, |
|
"step": 329000 |
|
}, |
|
{ |
|
"epoch": 17.75, |
|
"learning_rate": 0.00012327155045601648, |
|
"loss": 2.7944, |
|
"step": 330000 |
|
}, |
|
{ |
|
"epoch": 17.8, |
|
"learning_rate": 0.00012032950867902324, |
|
"loss": 2.7961, |
|
"step": 331000 |
|
}, |
|
{ |
|
"epoch": 17.85, |
|
"learning_rate": 0.00011738746690203001, |
|
"loss": 2.7949, |
|
"step": 332000 |
|
}, |
|
{ |
|
"epoch": 17.91, |
|
"learning_rate": 0.00011444542512503678, |
|
"loss": 2.797, |
|
"step": 333000 |
|
}, |
|
{ |
|
"epoch": 17.96, |
|
"learning_rate": 0.00011150338334804354, |
|
"loss": 2.7966, |
|
"step": 334000 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_accuracy": 0.41103738709715776, |
|
"eval_loss": 3.392622709274292, |
|
"eval_runtime": 225.4581, |
|
"eval_samples_per_second": 256.904, |
|
"eval_steps_per_second": 2.009, |
|
"step": 334710 |
|
}, |
|
{ |
|
"epoch": 18.02, |
|
"learning_rate": 0.00010856134157105031, |
|
"loss": 2.7874, |
|
"step": 335000 |
|
}, |
|
{ |
|
"epoch": 18.07, |
|
"learning_rate": 0.00010561929979405707, |
|
"loss": 2.767, |
|
"step": 336000 |
|
}, |
|
{ |
|
"epoch": 18.12, |
|
"learning_rate": 0.00010267725801706384, |
|
"loss": 2.7668, |
|
"step": 337000 |
|
}, |
|
{ |
|
"epoch": 18.18, |
|
"learning_rate": 9.97352162400706e-05, |
|
"loss": 2.7713, |
|
"step": 338000 |
|
}, |
|
{ |
|
"epoch": 18.23, |
|
"learning_rate": 9.679317446307737e-05, |
|
"loss": 2.77, |
|
"step": 339000 |
|
}, |
|
{ |
|
"epoch": 18.28, |
|
"learning_rate": 9.385113268608415e-05, |
|
"loss": 2.7696, |
|
"step": 340000 |
|
}, |
|
{ |
|
"epoch": 18.34, |
|
"learning_rate": 9.090909090909092e-05, |
|
"loss": 2.7766, |
|
"step": 341000 |
|
}, |
|
{ |
|
"epoch": 18.39, |
|
"learning_rate": 8.796704913209768e-05, |
|
"loss": 2.7739, |
|
"step": 342000 |
|
}, |
|
{ |
|
"epoch": 18.45, |
|
"learning_rate": 8.502500735510445e-05, |
|
"loss": 2.7716, |
|
"step": 343000 |
|
}, |
|
{ |
|
"epoch": 18.5, |
|
"learning_rate": 8.20829655781112e-05, |
|
"loss": 2.7766, |
|
"step": 344000 |
|
}, |
|
{ |
|
"epoch": 18.55, |
|
"learning_rate": 7.914092380111798e-05, |
|
"loss": 2.7769, |
|
"step": 345000 |
|
}, |
|
{ |
|
"epoch": 18.61, |
|
"learning_rate": 7.619888202412474e-05, |
|
"loss": 2.7757, |
|
"step": 346000 |
|
}, |
|
{ |
|
"epoch": 18.66, |
|
"learning_rate": 7.325684024713152e-05, |
|
"loss": 2.7777, |
|
"step": 347000 |
|
}, |
|
{ |
|
"epoch": 18.71, |
|
"learning_rate": 7.031479847013828e-05, |
|
"loss": 2.7777, |
|
"step": 348000 |
|
}, |
|
{ |
|
"epoch": 18.77, |
|
"learning_rate": 6.737275669314505e-05, |
|
"loss": 2.7801, |
|
"step": 349000 |
|
}, |
|
{ |
|
"epoch": 18.82, |
|
"learning_rate": 6.443071491615181e-05, |
|
"loss": 2.7749, |
|
"step": 350000 |
|
}, |
|
{ |
|
"epoch": 18.88, |
|
"learning_rate": 6.148867313915858e-05, |
|
"loss": 2.7776, |
|
"step": 351000 |
|
}, |
|
{ |
|
"epoch": 18.93, |
|
"learning_rate": 5.854663136216534e-05, |
|
"loss": 2.7747, |
|
"step": 352000 |
|
}, |
|
{ |
|
"epoch": 18.98, |
|
"learning_rate": 5.560458958517211e-05, |
|
"loss": 2.7802, |
|
"step": 353000 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"eval_accuracy": 0.4114727714798978, |
|
"eval_loss": 3.397742986679077, |
|
"eval_runtime": 224.2743, |
|
"eval_samples_per_second": 258.26, |
|
"eval_steps_per_second": 2.02, |
|
"step": 353305 |
|
}, |
|
{ |
|
"epoch": 19.04, |
|
"learning_rate": 5.2662547808178876e-05, |
|
"loss": 2.7662, |
|
"step": 354000 |
|
}, |
|
{ |
|
"epoch": 19.09, |
|
"learning_rate": 4.972050603118564e-05, |
|
"loss": 2.7581, |
|
"step": 355000 |
|
}, |
|
{ |
|
"epoch": 19.14, |
|
"learning_rate": 4.677846425419241e-05, |
|
"loss": 2.7572, |
|
"step": 356000 |
|
}, |
|
{ |
|
"epoch": 19.2, |
|
"learning_rate": 4.383642247719918e-05, |
|
"loss": 2.7579, |
|
"step": 357000 |
|
}, |
|
{ |
|
"epoch": 19.25, |
|
"learning_rate": 4.0894380700205943e-05, |
|
"loss": 2.7615, |
|
"step": 358000 |
|
}, |
|
{ |
|
"epoch": 19.31, |
|
"learning_rate": 3.795233892321271e-05, |
|
"loss": 2.763, |
|
"step": 359000 |
|
}, |
|
{ |
|
"epoch": 19.36, |
|
"learning_rate": 3.501029714621948e-05, |
|
"loss": 2.7573, |
|
"step": 360000 |
|
}, |
|
{ |
|
"epoch": 19.41, |
|
"learning_rate": 3.2068255369226245e-05, |
|
"loss": 2.7605, |
|
"step": 361000 |
|
}, |
|
{ |
|
"epoch": 19.47, |
|
"learning_rate": 2.912621359223301e-05, |
|
"loss": 2.758, |
|
"step": 362000 |
|
}, |
|
{ |
|
"epoch": 19.52, |
|
"learning_rate": 2.618417181523978e-05, |
|
"loss": 2.7615, |
|
"step": 363000 |
|
}, |
|
{ |
|
"epoch": 19.58, |
|
"learning_rate": 2.3242130038246544e-05, |
|
"loss": 2.7584, |
|
"step": 364000 |
|
}, |
|
{ |
|
"epoch": 19.63, |
|
"learning_rate": 2.0300088261253312e-05, |
|
"loss": 2.7616, |
|
"step": 365000 |
|
}, |
|
{ |
|
"epoch": 19.68, |
|
"learning_rate": 1.7358046484260074e-05, |
|
"loss": 2.7628, |
|
"step": 366000 |
|
}, |
|
{ |
|
"epoch": 19.74, |
|
"learning_rate": 1.4416004707266844e-05, |
|
"loss": 2.7596, |
|
"step": 367000 |
|
}, |
|
{ |
|
"epoch": 19.79, |
|
"learning_rate": 1.1473962930273611e-05, |
|
"loss": 2.7624, |
|
"step": 368000 |
|
}, |
|
{ |
|
"epoch": 19.84, |
|
"learning_rate": 8.531921153280376e-06, |
|
"loss": 2.7636, |
|
"step": 369000 |
|
}, |
|
{ |
|
"epoch": 19.9, |
|
"learning_rate": 5.5898793762871435e-06, |
|
"loss": 2.758, |
|
"step": 370000 |
|
}, |
|
{ |
|
"epoch": 19.95, |
|
"learning_rate": 2.6478375992939103e-06, |
|
"loss": 2.7617, |
|
"step": 371000 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_accuracy": 0.41106143702911596, |
|
"eval_loss": 3.407801628112793, |
|
"eval_runtime": 225.2112, |
|
"eval_samples_per_second": 257.185, |
|
"eval_steps_per_second": 2.011, |
|
"step": 371900 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"step": 371900, |
|
"total_flos": 1.56692312054784e+18, |
|
"train_loss": 3.0278301616277896, |
|
"train_runtime": 112002.0514, |
|
"train_samples_per_second": 106.254, |
|
"train_steps_per_second": 3.32 |
|
} |
|
], |
|
"logging_steps": 1000, |
|
"max_steps": 371900, |
|
"num_train_epochs": 20, |
|
"save_steps": 5000, |
|
"total_flos": 1.56692312054784e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|