| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 371940, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.375e-06, | |
| "loss": 6.8593, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.875e-05, | |
| "loss": 5.3643, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.8125e-05, | |
| "loss": 5.0364, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.75e-05, | |
| "loss": 4.8218, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.6874999999999994e-05, | |
| "loss": 4.6436, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 5.625e-05, | |
| "loss": 4.5074, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.5625e-05, | |
| "loss": 4.3958, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 7.5e-05, | |
| "loss": 4.2967, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 8.437499999999999e-05, | |
| "loss": 4.2068, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 9.374999999999999e-05, | |
| "loss": 4.1332, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.000103115625, | |
| "loss": 4.0676, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.000112490625, | |
| "loss": 4.0093, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.000121865625, | |
| "loss": 3.9526, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00013123125, | |
| "loss": 3.8934, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00014060625, | |
| "loss": 3.8436, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.000149971875, | |
| "loss": 3.8065, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.000159346875, | |
| "loss": 3.7709, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00016871249999999996, | |
| "loss": 3.7368, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.34518001634673895, | |
| "eval_loss": 3.9162497520446777, | |
| "eval_runtime": 152.9029, | |
| "eval_samples_per_second": 378.789, | |
| "eval_steps_per_second": 5.919, | |
| "step": 18597 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.000178078125, | |
| "loss": 3.6974, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00018745312499999998, | |
| "loss": 3.6666, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.00019682812499999998, | |
| "loss": 3.6461, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00020619374999999998, | |
| "loss": 3.6223, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.00021556874999999998, | |
| "loss": 3.6097, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00022494374999999998, | |
| "loss": 3.5886, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.00023430937499999997, | |
| "loss": 3.5771, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00024368437499999997, | |
| "loss": 3.5614, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00025305, | |
| "loss": 3.5469, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.000262425, | |
| "loss": 3.5355, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0002718, | |
| "loss": 3.5275, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.000281165625, | |
| "loss": 3.5145, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00029053124999999994, | |
| "loss": 3.5052, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.00029990624999999993, | |
| "loss": 3.4957, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.000299126316408778, | |
| "loss": 3.4811, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.000298243807730776, | |
| "loss": 3.4709, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.000297361299052774, | |
| "loss": 3.4514, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00029647967288344996, | |
| "loss": 3.4411, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.000295597164205448, | |
| "loss": 3.4348, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.37482168201575206, | |
| "eval_loss": 3.6327266693115234, | |
| "eval_runtime": 153.9354, | |
| "eval_samples_per_second": 376.249, | |
| "eval_steps_per_second": 5.879, | |
| "step": 37194 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.00029471553803612404, | |
| "loss": 3.3847, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.000293833029358122, | |
| "loss": 3.3766, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.000292951403188798, | |
| "loss": 3.3701, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 0.000292069777019474, | |
| "loss": 3.3642, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.000291187268341472, | |
| "loss": 3.3595, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.00029030475966347, | |
| "loss": 3.3568, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.000289422250985468, | |
| "loss": 3.3475, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.00028854062481614397, | |
| "loss": 3.3403, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.000287658116138142, | |
| "loss": 3.3384, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.00028677560746014, | |
| "loss": 3.3311, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.000285893981290816, | |
| "loss": 3.3236, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 0.00028501235512149196, | |
| "loss": 3.3185, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.00028412984644349, | |
| "loss": 3.3144, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.00028324822027416604, | |
| "loss": 3.3129, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.000282365711596164, | |
| "loss": 3.3052, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.000281483202918162, | |
| "loss": 3.2975, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.000280601576748838, | |
| "loss": 3.2951, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.000279719068070836, | |
| "loss": 3.2919, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.3899740583191344, | |
| "eval_loss": 3.508404493331909, | |
| "eval_runtime": 154.0998, | |
| "eval_samples_per_second": 375.847, | |
| "eval_steps_per_second": 5.873, | |
| "step": 55791 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.000278836559392834, | |
| "loss": 3.2736, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.000277954050714832, | |
| "loss": 3.2226, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.00027707154203683, | |
| "loss": 3.2278, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 0.000276189915867506, | |
| "loss": 3.2315, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.000275307407189504, | |
| "loss": 3.2244, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 0.00027442578102017997, | |
| "loss": 3.229, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.000273543272342178, | |
| "loss": 3.231, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.00027266076366417604, | |
| "loss": 3.224, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.000271779137494852, | |
| "loss": 3.2256, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 0.00027089662881685, | |
| "loss": 3.2241, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.000270015002647526, | |
| "loss": 3.2223, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.00026913249396952403, | |
| "loss": 3.2191, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 0.000268249985291522, | |
| "loss": 3.2179, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.000267368359122198, | |
| "loss": 3.2133, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 0.00026648585044419603, | |
| "loss": 3.2172, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 0.000265603341766194, | |
| "loss": 3.2109, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.000264720833088192, | |
| "loss": 3.2077, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.00026383832441018997, | |
| "loss": 3.2076, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 0.000262956698240866, | |
| "loss": 3.2086, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3956238680662571, | |
| "eval_loss": 3.4501850605010986, | |
| "eval_runtime": 154.25, | |
| "eval_samples_per_second": 375.481, | |
| "eval_steps_per_second": 5.867, | |
| "step": 74388 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.000262075072071542, | |
| "loss": 3.1628, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.00026119256339354003, | |
| "loss": 3.1437, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 0.000260310054715538, | |
| "loss": 3.1446, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 0.000259428428546214, | |
| "loss": 3.1476, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.00025854591986821204, | |
| "loss": 3.1475, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 0.00025766341119021, | |
| "loss": 3.1509, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.000256780902512208, | |
| "loss": 3.1536, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 0.00025589839383420603, | |
| "loss": 3.1514, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 0.000255015885156204, | |
| "loss": 3.149, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.00025413425898688, | |
| "loss": 3.1531, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.000253251750308878, | |
| "loss": 3.1487, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 0.000252370124139554, | |
| "loss": 3.1467, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 0.000251487615461552, | |
| "loss": 3.1496, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.00025060510678355004, | |
| "loss": 3.1494, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 0.000249723480614226, | |
| "loss": 3.148, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.000248840971936224, | |
| "loss": 3.1496, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.00024795846325822204, | |
| "loss": 3.1477, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.00024707595458022, | |
| "loss": 3.1474, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.3994938234828981, | |
| "eval_loss": 3.423523426055908, | |
| "eval_runtime": 154.6648, | |
| "eval_samples_per_second": 374.474, | |
| "eval_steps_per_second": 5.851, | |
| "step": 92985 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.000246194328410896, | |
| "loss": 3.1438, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 0.000245311819732894, | |
| "loss": 3.0764, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.000244429311054892, | |
| "loss": 3.0802, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.000243547684885568, | |
| "loss": 3.0875, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 0.00024266605871624405, | |
| "loss": 3.0881, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.00024178355003824203, | |
| "loss": 3.0951, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 0.00024090192386891804, | |
| "loss": 3.0977, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.00024001941519091602, | |
| "loss": 3.093, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.00023913778902159204, | |
| "loss": 3.097, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 0.00023825528034359002, | |
| "loss": 3.0942, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 0.00023737277166558803, | |
| "loss": 3.0994, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 0.000236491145496264, | |
| "loss": 3.1007, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 0.00023560863681826202, | |
| "loss": 3.1006, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.00023472612814026003, | |
| "loss": 3.1013, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 0.00023384450197093602, | |
| "loss": 3.103, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.00023296199329293402, | |
| "loss": 3.1001, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 0.00023208036712361004, | |
| "loss": 3.0997, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 0.00023119785844560804, | |
| "loss": 3.0997, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 0.00023031623227628403, | |
| "loss": 3.1012, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.4019542992689383, | |
| "eval_loss": 3.4030864238739014, | |
| "eval_runtime": 154.526, | |
| "eval_samples_per_second": 374.811, | |
| "eval_steps_per_second": 5.857, | |
| "step": 111582 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 0.00022943372359828204, | |
| "loss": 3.0671, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 0.00022855121492028005, | |
| "loss": 3.0316, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 0.00022766870624227803, | |
| "loss": 3.0381, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 0.00022678619756427604, | |
| "loss": 3.0465, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 0.00022590368888627402, | |
| "loss": 3.0474, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 0.00022502294522562804, | |
| "loss": 3.0522, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 0.00022414043654762602, | |
| "loss": 3.0489, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.00022325881037830203, | |
| "loss": 3.0538, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.0002223763017003, | |
| "loss": 3.0525, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 0.00022149379302229805, | |
| "loss": 3.0535, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.00022061128434429605, | |
| "loss": 3.0572, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 0.00021972877566629403, | |
| "loss": 3.06, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.00021884714949697005, | |
| "loss": 3.0585, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 0.00021796552332764603, | |
| "loss": 3.0623, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.00021708301464964404, | |
| "loss": 3.0628, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 0.00021620138848032003, | |
| "loss": 3.0624, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 0.00021531887980231804, | |
| "loss": 3.0592, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.00021443637112431604, | |
| "loss": 3.0606, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 0.00021355386244631402, | |
| "loss": 3.0638, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.4030296153303013, | |
| "eval_loss": 3.4128024578094482, | |
| "eval_runtime": 154.0025, | |
| "eval_samples_per_second": 376.085, | |
| "eval_steps_per_second": 5.877, | |
| "step": 130179 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 0.00021267311878566802, | |
| "loss": 3.0018, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.00021179061010766602, | |
| "loss": 2.9983, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "learning_rate": 0.00021090810142966406, | |
| "loss": 3.0046, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 0.00021002559275166204, | |
| "loss": 3.0095, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.00020914396658233805, | |
| "loss": 3.0113, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 0.00020826145790433606, | |
| "loss": 3.0139, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.00020737894922633404, | |
| "loss": 3.0141, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 0.00020649732305701006, | |
| "loss": 3.0165, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "learning_rate": 0.00020561481437900804, | |
| "loss": 3.0232, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "learning_rate": 0.00020473230570100604, | |
| "loss": 3.0236, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 0.00020384979702300403, | |
| "loss": 3.0217, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 0.00020296817085368004, | |
| "loss": 3.0193, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 0.00020208654468435603, | |
| "loss": 3.0279, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 0.00020120403600635403, | |
| "loss": 3.0243, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 0.00020032152732835207, | |
| "loss": 3.0241, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 0.00019943901865035005, | |
| "loss": 3.0299, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 7.9, | |
| "learning_rate": 0.00019855650997234806, | |
| "loss": 3.0313, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 0.00019767400129434606, | |
| "loss": 3.0262, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4046377523139853, | |
| "eval_loss": 3.3997907638549805, | |
| "eval_runtime": 154.087, | |
| "eval_samples_per_second": 375.879, | |
| "eval_steps_per_second": 5.873, | |
| "step": 148776 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 0.00019679237512502205, | |
| "loss": 3.0108, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.00019590986644702006, | |
| "loss": 2.9607, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 0.00019502735776901804, | |
| "loss": 2.9699, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "learning_rate": 0.00019414484909101605, | |
| "loss": 2.9736, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 0.00019326322292169203, | |
| "loss": 2.9743, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 0.00019238159675236805, | |
| "loss": 2.9832, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 0.00019149908807436603, | |
| "loss": 2.982, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 0.00019061657939636404, | |
| "loss": 2.9869, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 0.00018973495322704008, | |
| "loss": 2.9853, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 0.00018885244454903806, | |
| "loss": 2.9898, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 0.00018796993587103607, | |
| "loss": 2.9966, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "learning_rate": 0.00018708742719303405, | |
| "loss": 2.9916, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 0.00018620580102371006, | |
| "loss": 2.9973, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 0.00018532329234570804, | |
| "loss": 2.9947, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 0.00018444166617638405, | |
| "loss": 2.9926, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 0.00018356004000706004, | |
| "loss": 2.9954, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 0.00018267753132905805, | |
| "loss": 2.9968, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 0.00018179502265105606, | |
| "loss": 2.9997, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 0.00018091251397305404, | |
| "loss": 3.0016, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.40699604418590996, | |
| "eval_loss": 3.373080015182495, | |
| "eval_runtime": 153.8917, | |
| "eval_samples_per_second": 376.355, | |
| "eval_steps_per_second": 5.881, | |
| "step": 167373 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 0.00018003088780373005, | |
| "loss": 2.9567, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 0.00017914926163440606, | |
| "loss": 2.9362, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "learning_rate": 0.00017826675295640407, | |
| "loss": 2.9438, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 0.00017738424427840205, | |
| "loss": 2.9495, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 0.00017650261810907807, | |
| "loss": 2.9516, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 0.00017562010943107607, | |
| "loss": 2.9533, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "learning_rate": 0.00017473760075307405, | |
| "loss": 2.9566, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 0.00017385509207507206, | |
| "loss": 2.9633, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "learning_rate": 0.00017297258339707004, | |
| "loss": 2.9596, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 0.00017209095722774606, | |
| "loss": 2.9628, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 0.00017120844854974404, | |
| "loss": 2.9617, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "learning_rate": 0.00017032770488909806, | |
| "loss": 2.9647, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 0.00016944519621109604, | |
| "loss": 2.964, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "learning_rate": 0.00016856268753309407, | |
| "loss": 2.9676, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 0.00016768017885509208, | |
| "loss": 2.9731, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "learning_rate": 0.00016679767017709006, | |
| "loss": 2.9734, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 0.00016591604400776607, | |
| "loss": 2.9745, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 0.00016503353532976406, | |
| "loss": 2.9715, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.40619785412436715, | |
| "eval_loss": 3.4057776927948, | |
| "eval_runtime": 153.2918, | |
| "eval_samples_per_second": 377.828, | |
| "eval_steps_per_second": 5.904, | |
| "step": 185970 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.00016415190916044007, | |
| "loss": 2.97, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 0.00016326940048243805, | |
| "loss": 2.9101, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 0.00016238777431311406, | |
| "loss": 2.9121, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 10.16, | |
| "learning_rate": 0.00016150703065246806, | |
| "loss": 2.9209, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 0.00016062452197446606, | |
| "loss": 2.922, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "learning_rate": 0.00015974201329646404, | |
| "loss": 2.9243, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 0.00015885950461846205, | |
| "loss": 2.9304, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 0.0001579769959404601, | |
| "loss": 2.9305, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "learning_rate": 0.00015709448726245807, | |
| "loss": 2.9363, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 0.00015621286109313408, | |
| "loss": 2.9379, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 10.54, | |
| "learning_rate": 0.0001553303524151321, | |
| "loss": 2.941, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "learning_rate": 0.00015444872624580808, | |
| "loss": 2.9366, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "learning_rate": 0.00015356621756780608, | |
| "loss": 2.9413, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 10.7, | |
| "learning_rate": 0.00015268370888980406, | |
| "loss": 2.9473, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 10.75, | |
| "learning_rate": 0.00015180120021180207, | |
| "loss": 2.9472, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 0.00015091957404247806, | |
| "loss": 2.9504, | |
| "step": 201000 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "learning_rate": 0.00015003706536447607, | |
| "loss": 2.9458, | |
| "step": 202000 | |
| }, | |
| { | |
| "epoch": 10.92, | |
| "learning_rate": 0.00014915455668647407, | |
| "loss": 2.9471, | |
| "step": 203000 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 0.00014827293051715006, | |
| "loss": 2.9481, | |
| "step": 204000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.4068837157806495, | |
| "eval_loss": 3.38750958442688, | |
| "eval_runtime": 153.5658, | |
| "eval_samples_per_second": 377.154, | |
| "eval_steps_per_second": 5.893, | |
| "step": 204567 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 0.00014739042183914807, | |
| "loss": 2.9156, | |
| "step": 205000 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 0.00014650879566982408, | |
| "loss": 2.8891, | |
| "step": 206000 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 0.00014562628699182206, | |
| "loss": 2.8942, | |
| "step": 207000 | |
| }, | |
| { | |
| "epoch": 11.18, | |
| "learning_rate": 0.00014474554333117608, | |
| "loss": 2.9003, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 0.0001438630346531741, | |
| "loss": 2.8988, | |
| "step": 209000 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "learning_rate": 0.00014298052597517207, | |
| "loss": 2.9105, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 0.00014209889980584808, | |
| "loss": 2.9077, | |
| "step": 211000 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "learning_rate": 0.00014121639112784606, | |
| "loss": 2.9168, | |
| "step": 212000 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 0.00014033388244984407, | |
| "loss": 2.9148, | |
| "step": 213000 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 0.00013945137377184208, | |
| "loss": 2.9142, | |
| "step": 214000 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 0.0001385697476025181, | |
| "loss": 2.9159, | |
| "step": 215000 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 0.00013768723892451607, | |
| "loss": 2.9195, | |
| "step": 216000 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 0.00013680473024651408, | |
| "loss": 2.9189, | |
| "step": 217000 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 0.00013592398658586807, | |
| "loss": 2.9207, | |
| "step": 218000 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 0.00013504147790786608, | |
| "loss": 2.9228, | |
| "step": 219000 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "learning_rate": 0.0001341589692298641, | |
| "loss": 2.9244, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "learning_rate": 0.00013327734306054008, | |
| "loss": 2.9194, | |
| "step": 221000 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "learning_rate": 0.00013239483438253808, | |
| "loss": 2.9257, | |
| "step": 222000 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 0.0001315123257045361, | |
| "loss": 2.9243, | |
| "step": 223000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.407012167880445, | |
| "eval_loss": 3.4070396423339844, | |
| "eval_runtime": 153.2437, | |
| "eval_samples_per_second": 377.947, | |
| "eval_steps_per_second": 5.906, | |
| "step": 223164 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 0.00013063069953521208, | |
| "loss": 2.874, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 0.0001297481908572101, | |
| "loss": 2.8707, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 0.0001288656821792081, | |
| "loss": 2.8776, | |
| "step": 226000 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 0.00012798317350120608, | |
| "loss": 2.876, | |
| "step": 227000 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 0.0001271015473318821, | |
| "loss": 2.8805, | |
| "step": 228000 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 0.0001262190386538801, | |
| "loss": 2.8861, | |
| "step": 229000 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "learning_rate": 0.00012533741248455608, | |
| "loss": 2.8871, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 12.42, | |
| "learning_rate": 0.0001244549038065541, | |
| "loss": 2.8902, | |
| "step": 231000 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 0.0001235723951285521, | |
| "loss": 2.8965, | |
| "step": 232000 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "learning_rate": 0.00012269076895922809, | |
| "loss": 2.8889, | |
| "step": 233000 | |
| }, | |
| { | |
| "epoch": 12.58, | |
| "learning_rate": 0.0001218082602812261, | |
| "loss": 2.9003, | |
| "step": 234000 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "learning_rate": 0.00012092575160322409, | |
| "loss": 2.896, | |
| "step": 235000 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 0.00012004324292522208, | |
| "loss": 2.8995, | |
| "step": 236000 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "learning_rate": 0.00011916161675589808, | |
| "loss": 2.8987, | |
| "step": 237000 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.00011827910807789609, | |
| "loss": 2.8998, | |
| "step": 238000 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 0.0001173965993998941, | |
| "loss": 2.9073, | |
| "step": 239000 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "learning_rate": 0.0001165149732305701, | |
| "loss": 2.9013, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "learning_rate": 0.00011563246455256809, | |
| "loss": 2.9047, | |
| "step": 241000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.4079228207113693, | |
| "eval_loss": 3.4015119075775146, | |
| "eval_runtime": 153.5554, | |
| "eval_samples_per_second": 377.18, | |
| "eval_steps_per_second": 5.894, | |
| "step": 241761 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 0.00011475083838324409, | |
| "loss": 2.8895, | |
| "step": 242000 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "learning_rate": 0.00011386832970524209, | |
| "loss": 2.8458, | |
| "step": 243000 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 0.0001129867035359181, | |
| "loss": 2.8516, | |
| "step": 244000 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 0.0001121041948579161, | |
| "loss": 2.8585, | |
| "step": 245000 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "learning_rate": 0.0001112225686885921, | |
| "loss": 2.8576, | |
| "step": 246000 | |
| }, | |
| { | |
| "epoch": 13.28, | |
| "learning_rate": 0.00011034006001059009, | |
| "loss": 2.8645, | |
| "step": 247000 | |
| }, | |
| { | |
| "epoch": 13.34, | |
| "learning_rate": 0.00010945843384126609, | |
| "loss": 2.87, | |
| "step": 248000 | |
| }, | |
| { | |
| "epoch": 13.39, | |
| "learning_rate": 0.00010857592516326408, | |
| "loss": 2.8666, | |
| "step": 249000 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "learning_rate": 0.0001076934164852621, | |
| "loss": 2.872, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 0.0001068109078072601, | |
| "loss": 2.8725, | |
| "step": 251000 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "learning_rate": 0.0001059292816379361, | |
| "loss": 2.8739, | |
| "step": 252000 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 0.00010504677295993409, | |
| "loss": 2.878, | |
| "step": 253000 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 0.00010416514679061009, | |
| "loss": 2.8767, | |
| "step": 254000 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "learning_rate": 0.00010328263811260809, | |
| "loss": 2.877, | |
| "step": 255000 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 0.00010240012943460611, | |
| "loss": 2.882, | |
| "step": 256000 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "learning_rate": 0.00010151850326528211, | |
| "loss": 2.8804, | |
| "step": 257000 | |
| }, | |
| { | |
| "epoch": 13.87, | |
| "learning_rate": 0.0001006368770959581, | |
| "loss": 2.8858, | |
| "step": 258000 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 9.97543684179561e-05, | |
| "loss": 2.882, | |
| "step": 259000 | |
| }, | |
| { | |
| "epoch": 13.98, | |
| "learning_rate": 9.88718597399541e-05, | |
| "loss": 2.8797, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.40766477441674887, | |
| "eval_loss": 3.411367177963257, | |
| "eval_runtime": 153.8972, | |
| "eval_samples_per_second": 376.342, | |
| "eval_steps_per_second": 5.881, | |
| "step": 260358 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 9.79902335706301e-05, | |
| "loss": 2.8484, | |
| "step": 261000 | |
| }, | |
| { | |
| "epoch": 14.09, | |
| "learning_rate": 9.71077248926281e-05, | |
| "loss": 2.8351, | |
| "step": 262000 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "learning_rate": 9.62252162146261e-05, | |
| "loss": 2.8377, | |
| "step": 263000 | |
| }, | |
| { | |
| "epoch": 14.2, | |
| "learning_rate": 9.534270753662411e-05, | |
| "loss": 2.8434, | |
| "step": 264000 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 9.446108136730011e-05, | |
| "loss": 2.841, | |
| "step": 265000 | |
| }, | |
| { | |
| "epoch": 14.3, | |
| "learning_rate": 9.35785726892981e-05, | |
| "loss": 2.844, | |
| "step": 266000 | |
| }, | |
| { | |
| "epoch": 14.36, | |
| "learning_rate": 9.26960640112961e-05, | |
| "loss": 2.8465, | |
| "step": 267000 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 9.18135553332941e-05, | |
| "loss": 2.8546, | |
| "step": 268000 | |
| }, | |
| { | |
| "epoch": 14.46, | |
| "learning_rate": 9.09319291639701e-05, | |
| "loss": 2.8524, | |
| "step": 269000 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "learning_rate": 9.00503029946461e-05, | |
| "loss": 2.8515, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 14.57, | |
| "learning_rate": 8.91677943166441e-05, | |
| "loss": 2.8541, | |
| "step": 271000 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 8.82861681473201e-05, | |
| "loss": 2.8572, | |
| "step": 272000 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "learning_rate": 8.74036594693181e-05, | |
| "loss": 2.8568, | |
| "step": 273000 | |
| }, | |
| { | |
| "epoch": 14.73, | |
| "learning_rate": 8.652115079131611e-05, | |
| "loss": 2.8618, | |
| "step": 274000 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "learning_rate": 8.563952462199211e-05, | |
| "loss": 2.8609, | |
| "step": 275000 | |
| }, | |
| { | |
| "epoch": 14.84, | |
| "learning_rate": 8.475701594399011e-05, | |
| "loss": 2.8582, | |
| "step": 276000 | |
| }, | |
| { | |
| "epoch": 14.89, | |
| "learning_rate": 8.38745072659881e-05, | |
| "loss": 2.8602, | |
| "step": 277000 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "learning_rate": 8.29919985879861e-05, | |
| "loss": 2.8651, | |
| "step": 278000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.40826464303551124, | |
| "eval_loss": 3.4071929454803467, | |
| "eval_runtime": 153.9494, | |
| "eval_samples_per_second": 376.215, | |
| "eval_steps_per_second": 5.879, | |
| "step": 278955 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 8.21103724186621e-05, | |
| "loss": 2.8635, | |
| "step": 279000 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "learning_rate": 8.122786374066012e-05, | |
| "loss": 2.8126, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "learning_rate": 8.034535506265811e-05, | |
| "loss": 2.8181, | |
| "step": 281000 | |
| }, | |
| { | |
| "epoch": 15.16, | |
| "learning_rate": 7.946372889333411e-05, | |
| "loss": 2.82, | |
| "step": 282000 | |
| }, | |
| { | |
| "epoch": 15.22, | |
| "learning_rate": 7.85812202153321e-05, | |
| "loss": 2.826, | |
| "step": 283000 | |
| }, | |
| { | |
| "epoch": 15.27, | |
| "learning_rate": 7.76995940460081e-05, | |
| "loss": 2.8282, | |
| "step": 284000 | |
| }, | |
| { | |
| "epoch": 15.33, | |
| "learning_rate": 7.681796787668412e-05, | |
| "loss": 2.8297, | |
| "step": 285000 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 7.593545919868211e-05, | |
| "loss": 2.8314, | |
| "step": 286000 | |
| }, | |
| { | |
| "epoch": 15.43, | |
| "learning_rate": 7.505383302935811e-05, | |
| "loss": 2.8329, | |
| "step": 287000 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 7.417132435135611e-05, | |
| "loss": 2.8348, | |
| "step": 288000 | |
| }, | |
| { | |
| "epoch": 15.54, | |
| "learning_rate": 7.328881567335412e-05, | |
| "loss": 2.8374, | |
| "step": 289000 | |
| }, | |
| { | |
| "epoch": 15.59, | |
| "learning_rate": 7.240630699535211e-05, | |
| "loss": 2.8359, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "learning_rate": 7.152468082602811e-05, | |
| "loss": 2.8397, | |
| "step": 291000 | |
| }, | |
| { | |
| "epoch": 15.7, | |
| "learning_rate": 7.064217214802612e-05, | |
| "loss": 2.8381, | |
| "step": 292000 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "learning_rate": 6.975966347002411e-05, | |
| "loss": 2.8392, | |
| "step": 293000 | |
| }, | |
| { | |
| "epoch": 15.81, | |
| "learning_rate": 6.887803730070011e-05, | |
| "loss": 2.8425, | |
| "step": 294000 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 6.799552862269812e-05, | |
| "loss": 2.842, | |
| "step": 295000 | |
| }, | |
| { | |
| "epoch": 15.92, | |
| "learning_rate": 6.711301994469612e-05, | |
| "loss": 2.8431, | |
| "step": 296000 | |
| }, | |
| { | |
| "epoch": 15.97, | |
| "learning_rate": 6.623051126669412e-05, | |
| "loss": 2.8434, | |
| "step": 297000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.40750031273249193, | |
| "eval_loss": 3.424048900604248, | |
| "eval_runtime": 153.7747, | |
| "eval_samples_per_second": 376.642, | |
| "eval_steps_per_second": 5.885, | |
| "step": 297552 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "learning_rate": 6.534888509737011e-05, | |
| "loss": 2.8248, | |
| "step": 298000 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "learning_rate": 6.446637641936812e-05, | |
| "loss": 2.7992, | |
| "step": 299000 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 6.358386774136611e-05, | |
| "loss": 2.808, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "learning_rate": 6.270135906336412e-05, | |
| "loss": 2.807, | |
| "step": 301000 | |
| }, | |
| { | |
| "epoch": 16.24, | |
| "learning_rate": 6.181973289404012e-05, | |
| "loss": 2.8109, | |
| "step": 302000 | |
| }, | |
| { | |
| "epoch": 16.29, | |
| "learning_rate": 6.093722421603812e-05, | |
| "loss": 2.8098, | |
| "step": 303000 | |
| }, | |
| { | |
| "epoch": 16.35, | |
| "learning_rate": 6.005648055539212e-05, | |
| "loss": 2.811, | |
| "step": 304000 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "learning_rate": 5.917397187739013e-05, | |
| "loss": 2.8134, | |
| "step": 305000 | |
| }, | |
| { | |
| "epoch": 16.45, | |
| "learning_rate": 5.829146319938812e-05, | |
| "loss": 2.8165, | |
| "step": 306000 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "learning_rate": 5.740983703006412e-05, | |
| "loss": 2.8168, | |
| "step": 307000 | |
| }, | |
| { | |
| "epoch": 16.56, | |
| "learning_rate": 5.652732835206213e-05, | |
| "loss": 2.8195, | |
| "step": 308000 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 5.5644819674060124e-05, | |
| "loss": 2.8171, | |
| "step": 309000 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 5.4763193504736124e-05, | |
| "loss": 2.8199, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 16.72, | |
| "learning_rate": 5.3880684826734125e-05, | |
| "loss": 2.8202, | |
| "step": 311000 | |
| }, | |
| { | |
| "epoch": 16.78, | |
| "learning_rate": 5.2998176148732126e-05, | |
| "loss": 2.8227, | |
| "step": 312000 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 5.211566747073012e-05, | |
| "loss": 2.826, | |
| "step": 313000 | |
| }, | |
| { | |
| "epoch": 16.88, | |
| "learning_rate": 5.123404130140613e-05, | |
| "loss": 2.8246, | |
| "step": 314000 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "learning_rate": 5.035153262340413e-05, | |
| "loss": 2.824, | |
| "step": 315000 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "learning_rate": 4.946902394540212e-05, | |
| "loss": 2.8255, | |
| "step": 316000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.40828197600713634, | |
| "eval_loss": 3.417881727218628, | |
| "eval_runtime": 153.4343, | |
| "eval_samples_per_second": 377.478, | |
| "eval_steps_per_second": 5.898, | |
| "step": 316149 | |
| }, | |
| { | |
| "epoch": 17.05, | |
| "learning_rate": 4.858739777607813e-05, | |
| "loss": 2.7937, | |
| "step": 317000 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "learning_rate": 4.770488909807612e-05, | |
| "loss": 2.7869, | |
| "step": 318000 | |
| }, | |
| { | |
| "epoch": 17.15, | |
| "learning_rate": 4.682326292875212e-05, | |
| "loss": 2.792, | |
| "step": 319000 | |
| }, | |
| { | |
| "epoch": 17.21, | |
| "learning_rate": 4.594163675942813e-05, | |
| "loss": 2.7937, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 17.26, | |
| "learning_rate": 4.505912808142613e-05, | |
| "loss": 2.7897, | |
| "step": 321000 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 4.4176619403424126e-05, | |
| "loss": 2.7967, | |
| "step": 322000 | |
| }, | |
| { | |
| "epoch": 17.37, | |
| "learning_rate": 4.329411072542213e-05, | |
| "loss": 2.7942, | |
| "step": 323000 | |
| }, | |
| { | |
| "epoch": 17.42, | |
| "learning_rate": 4.241160204742013e-05, | |
| "loss": 2.8025, | |
| "step": 324000 | |
| }, | |
| { | |
| "epoch": 17.48, | |
| "learning_rate": 4.152909336941813e-05, | |
| "loss": 2.8014, | |
| "step": 325000 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "learning_rate": 4.0648349708772134e-05, | |
| "loss": 2.8022, | |
| "step": 326000 | |
| }, | |
| { | |
| "epoch": 17.58, | |
| "learning_rate": 3.976584103077013e-05, | |
| "loss": 2.8024, | |
| "step": 327000 | |
| }, | |
| { | |
| "epoch": 17.64, | |
| "learning_rate": 3.888333235276813e-05, | |
| "loss": 2.8028, | |
| "step": 328000 | |
| }, | |
| { | |
| "epoch": 17.69, | |
| "learning_rate": 3.800082367476614e-05, | |
| "loss": 2.8021, | |
| "step": 329000 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "learning_rate": 3.711831499676413e-05, | |
| "loss": 2.8048, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 17.8, | |
| "learning_rate": 3.623668882744013e-05, | |
| "loss": 2.8058, | |
| "step": 331000 | |
| }, | |
| { | |
| "epoch": 17.85, | |
| "learning_rate": 3.535506265811613e-05, | |
| "loss": 2.8078, | |
| "step": 332000 | |
| }, | |
| { | |
| "epoch": 17.91, | |
| "learning_rate": 3.447255398011413e-05, | |
| "loss": 2.8043, | |
| "step": 333000 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "learning_rate": 3.3590045302112134e-05, | |
| "loss": 2.8036, | |
| "step": 334000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.4081968563364037, | |
| "eval_loss": 3.4256491661071777, | |
| "eval_runtime": 153.5909, | |
| "eval_samples_per_second": 377.093, | |
| "eval_steps_per_second": 5.892, | |
| "step": 334746 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "learning_rate": 3.2708419132788134e-05, | |
| "loss": 2.7978, | |
| "step": 335000 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 3.1825910454786135e-05, | |
| "loss": 2.7744, | |
| "step": 336000 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "learning_rate": 3.0943401776784136e-05, | |
| "loss": 2.7766, | |
| "step": 337000 | |
| }, | |
| { | |
| "epoch": 18.17, | |
| "learning_rate": 3.0060893098782134e-05, | |
| "loss": 2.7851, | |
| "step": 338000 | |
| }, | |
| { | |
| "epoch": 18.23, | |
| "learning_rate": 2.9179266929458137e-05, | |
| "loss": 2.7799, | |
| "step": 339000 | |
| }, | |
| { | |
| "epoch": 18.28, | |
| "learning_rate": 2.829675825145614e-05, | |
| "loss": 2.7785, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "learning_rate": 2.7415132082132135e-05, | |
| "loss": 2.7824, | |
| "step": 341000 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "learning_rate": 2.6532623404130136e-05, | |
| "loss": 2.7789, | |
| "step": 342000 | |
| }, | |
| { | |
| "epoch": 18.44, | |
| "learning_rate": 2.565011472612814e-05, | |
| "loss": 2.7847, | |
| "step": 343000 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "learning_rate": 2.4767606048126135e-05, | |
| "loss": 2.7849, | |
| "step": 344000 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 2.3885979878802138e-05, | |
| "loss": 2.7844, | |
| "step": 345000 | |
| }, | |
| { | |
| "epoch": 18.61, | |
| "learning_rate": 2.300435370947814e-05, | |
| "loss": 2.7845, | |
| "step": 346000 | |
| }, | |
| { | |
| "epoch": 18.66, | |
| "learning_rate": 2.212184503147614e-05, | |
| "loss": 2.7858, | |
| "step": 347000 | |
| }, | |
| { | |
| "epoch": 18.71, | |
| "learning_rate": 2.123933635347414e-05, | |
| "loss": 2.788, | |
| "step": 348000 | |
| }, | |
| { | |
| "epoch": 18.77, | |
| "learning_rate": 2.035682767547214e-05, | |
| "loss": 2.7841, | |
| "step": 349000 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "learning_rate": 1.947431899747014e-05, | |
| "loss": 2.7931, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 18.87, | |
| "learning_rate": 1.859357533682414e-05, | |
| "loss": 2.7847, | |
| "step": 351000 | |
| }, | |
| { | |
| "epoch": 18.93, | |
| "learning_rate": 1.7711066658822143e-05, | |
| "loss": 2.7911, | |
| "step": 352000 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "learning_rate": 1.6828557980820144e-05, | |
| "loss": 2.7888, | |
| "step": 353000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.4083161044939021, | |
| "eval_loss": 3.4363324642181396, | |
| "eval_runtime": 153.9158, | |
| "eval_samples_per_second": 376.297, | |
| "eval_steps_per_second": 5.88, | |
| "step": 353343 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "learning_rate": 1.5946931811496144e-05, | |
| "loss": 2.7724, | |
| "step": 354000 | |
| }, | |
| { | |
| "epoch": 19.09, | |
| "learning_rate": 1.5064423133494145e-05, | |
| "loss": 2.7653, | |
| "step": 355000 | |
| }, | |
| { | |
| "epoch": 19.14, | |
| "learning_rate": 1.4181914455492144e-05, | |
| "loss": 2.7712, | |
| "step": 356000 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 1.3299405777490143e-05, | |
| "loss": 2.7696, | |
| "step": 357000 | |
| }, | |
| { | |
| "epoch": 19.25, | |
| "learning_rate": 1.2416897099488144e-05, | |
| "loss": 2.7658, | |
| "step": 358000 | |
| }, | |
| { | |
| "epoch": 19.3, | |
| "learning_rate": 1.1535270930164146e-05, | |
| "loss": 2.7637, | |
| "step": 359000 | |
| }, | |
| { | |
| "epoch": 19.36, | |
| "learning_rate": 1.0653644760840146e-05, | |
| "loss": 2.769, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 19.41, | |
| "learning_rate": 9.771136082838147e-06, | |
| "loss": 2.7734, | |
| "step": 361000 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "learning_rate": 8.888627404836147e-06, | |
| "loss": 2.7685, | |
| "step": 362000 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "learning_rate": 8.006118726834148e-06, | |
| "loss": 2.7733, | |
| "step": 363000 | |
| }, | |
| { | |
| "epoch": 19.57, | |
| "learning_rate": 7.125375066188151e-06, | |
| "loss": 2.7755, | |
| "step": 364000 | |
| }, | |
| { | |
| "epoch": 19.63, | |
| "learning_rate": 6.2428663881861495e-06, | |
| "loss": 2.7693, | |
| "step": 365000 | |
| }, | |
| { | |
| "epoch": 19.68, | |
| "learning_rate": 5.36035771018415e-06, | |
| "loss": 2.7692, | |
| "step": 366000 | |
| }, | |
| { | |
| "epoch": 19.73, | |
| "learning_rate": 4.47784903218215e-06, | |
| "loss": 2.7709, | |
| "step": 367000 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "learning_rate": 3.5953403541801493e-06, | |
| "loss": 2.7716, | |
| "step": 368000 | |
| }, | |
| { | |
| "epoch": 19.84, | |
| "learning_rate": 2.7128316761781486e-06, | |
| "loss": 2.7701, | |
| "step": 369000 | |
| }, | |
| { | |
| "epoch": 19.9, | |
| "learning_rate": 1.8312055068541506e-06, | |
| "loss": 2.7714, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 19.95, | |
| "learning_rate": 9.486968288521502e-07, | |
| "loss": 2.7701, | |
| "step": 371000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.40813531756892846, | |
| "eval_loss": 3.4419291019439697, | |
| "eval_runtime": 153.8158, | |
| "eval_samples_per_second": 376.541, | |
| "eval_steps_per_second": 5.884, | |
| "step": 371940 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 371940, | |
| "total_flos": 1.56702845389824e+18, | |
| "train_loss": 3.060115090571354, | |
| "train_runtime": 80942.7407, | |
| "train_samples_per_second": 147.036, | |
| "train_steps_per_second": 4.595 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 371940, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 5000, | |
| "total_flos": 1.56702845389824e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |