| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 371760, | |
| "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": 6.2312, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 6.25e-05, | |
| "loss": 5.0188, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.375e-05, | |
| "loss": 4.6798, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.000125, | |
| "loss": 4.4554, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00015625, | |
| "loss": 4.2939, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0001875, | |
| "loss": 4.1794, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00021875, | |
| "loss": 4.0791, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00025, | |
| "loss": 3.9863, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00028121875, | |
| "loss": 3.9116, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00031246875000000003, | |
| "loss": 3.8526, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00034368749999999997, | |
| "loss": 3.8069, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0003749375, | |
| "loss": 3.765, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0004061875, | |
| "loss": 3.7259, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00043740625, | |
| "loss": 3.6976, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.000468625, | |
| "loss": 3.6673, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.000499875, | |
| "loss": 3.6486, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.000531125, | |
| "loss": 3.6223, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00056234375, | |
| "loss": 3.6055, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.35899377253205417, | |
| "eval_loss": 3.777111053466797, | |
| "eval_runtime": 152.4765, | |
| "eval_samples_per_second": 379.855, | |
| "eval_steps_per_second": 5.935, | |
| "step": 18588 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00059359375, | |
| "loss": 3.5729, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0006248125, | |
| "loss": 3.5466, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.0006560625, | |
| "loss": 3.5344, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00068728125, | |
| "loss": 3.5278, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.00071853125, | |
| "loss": 3.5148, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.0007497500000000001, | |
| "loss": 3.5048, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.000781, | |
| "loss": 3.4967, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00081221875, | |
| "loss": 3.4883, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.0008434375, | |
| "loss": 3.4759, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.0008746874999999999, | |
| "loss": 3.4723, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0009059375, | |
| "loss": 3.4626, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.0009371875, | |
| "loss": 3.4579, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.0009684062500000001, | |
| "loss": 3.4481, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.000999625, | |
| "loss": 3.4401, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.0009970920649870498, | |
| "loss": 3.433, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.0009941488109253592, | |
| "loss": 3.4158, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.0009912055568636685, | |
| "loss": 3.4076, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.0009882652460560396, | |
| "loss": 3.3956, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.0009853249352484107, | |
| "loss": 3.3861, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3803195432666162, | |
| "eval_loss": 3.5827529430389404, | |
| "eval_runtime": 154.5043, | |
| "eval_samples_per_second": 374.87, | |
| "eval_steps_per_second": 5.857, | |
| "step": 37176 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0009823816811867201, | |
| "loss": 3.3372, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.0009794384271250296, | |
| "loss": 3.3307, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.0009764951730633388, | |
| "loss": 3.3263, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.0009735548622557099, | |
| "loss": 3.3211, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0009706116081940193, | |
| "loss": 3.3218, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.0009676712973863904, | |
| "loss": 3.3117, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.0009647280433246997, | |
| "loss": 3.3057, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.0009617847892630091, | |
| "loss": 3.3084, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.0009588444784553803, | |
| "loss": 3.2993, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0009559012243936896, | |
| "loss": 3.2901, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.0009529609135860607, | |
| "loss": 3.2862, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.0009500176595243702, | |
| "loss": 3.2841, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.0009470802919708029, | |
| "loss": 3.2814, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.0009441370379091123, | |
| "loss": 3.2775, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0009411937838474218, | |
| "loss": 3.2718, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.0009382505297857311, | |
| "loss": 3.2708, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0009353102189781022, | |
| "loss": 3.263, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.0009323699081704733, | |
| "loss": 3.2583, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.3912598185726034, | |
| "eval_loss": 3.506493091583252, | |
| "eval_runtime": 154.3502, | |
| "eval_samples_per_second": 375.244, | |
| "eval_steps_per_second": 5.863, | |
| "step": 55764 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.0009294266541087827, | |
| "loss": 3.2401, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.0009264834000470921, | |
| "loss": 3.1911, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.0009235401459854014, | |
| "loss": 3.1913, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 0.0009205998351777726, | |
| "loss": 3.2004, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.000917656581116082, | |
| "loss": 3.1999, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 0.0009147133270543913, | |
| "loss": 3.1958, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.0009117730162467625, | |
| "loss": 3.2006, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.0009088297621850719, | |
| "loss": 3.1962, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.0009058865081233812, | |
| "loss": 3.2013, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 0.0009029461973157523, | |
| "loss": 3.1961, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.0009000029432540618, | |
| "loss": 3.1969, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0008970596891923711, | |
| "loss": 3.1927, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 0.0008941193783847422, | |
| "loss": 3.1953, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.0008911761243230517, | |
| "loss": 3.1906, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 0.0008882358135154227, | |
| "loss": 3.1874, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 0.0008852955027077938, | |
| "loss": 3.187, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.0008823522486461031, | |
| "loss": 3.177, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.0008794089945844126, | |
| "loss": 3.181, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 0.000876465740522722, | |
| "loss": 3.176, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3972729794453988, | |
| "eval_loss": 3.431795358657837, | |
| "eval_runtime": 154.2897, | |
| "eval_samples_per_second": 375.391, | |
| "eval_steps_per_second": 5.866, | |
| "step": 74352 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.000873525429715093, | |
| "loss": 3.136, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.0008705821756534025, | |
| "loss": 3.1093, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 0.0008676418648457735, | |
| "loss": 3.1209, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 0.0008646986107840828, | |
| "loss": 3.1257, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.0008617553567223923, | |
| "loss": 3.1243, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 0.0008588150459147634, | |
| "loss": 3.1306, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.0008558747351071344, | |
| "loss": 3.1262, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 0.0008529314810454438, | |
| "loss": 3.1282, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 0.0008499882269837532, | |
| "loss": 3.1309, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.0008470449729220626, | |
| "loss": 3.1302, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.000844101718860372, | |
| "loss": 3.129, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 0.0008411614080527431, | |
| "loss": 3.131, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 0.0008382210972451142, | |
| "loss": 3.1299, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.0008352778431834236, | |
| "loss": 3.1274, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 0.000832334589121733, | |
| "loss": 3.1264, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.0008293913350600424, | |
| "loss": 3.129, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 0.0008264480809983518, | |
| "loss": 3.129, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.0008235107134447845, | |
| "loss": 3.1227, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.40129751847861056, | |
| "eval_loss": 3.413170337677002, | |
| "eval_runtime": 154.2545, | |
| "eval_samples_per_second": 375.477, | |
| "eval_steps_per_second": 5.867, | |
| "step": 92940 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.000820567459383094, | |
| "loss": 3.1215, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 0.0008176242053214034, | |
| "loss": 3.0545, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.0008146809512597127, | |
| "loss": 3.0645, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.0008117376971980222, | |
| "loss": 3.067, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 0.0008087973863903932, | |
| "loss": 3.072, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.0008058541323287026, | |
| "loss": 3.0721, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.0008029138215210738, | |
| "loss": 3.0765, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.0007999735107134448, | |
| "loss": 3.0751, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.0007970302566517542, | |
| "loss": 3.0792, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 0.0007940870025900636, | |
| "loss": 3.0806, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 0.000791143748528373, | |
| "loss": 3.0825, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.0007882034377207441, | |
| "loss": 3.077, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 0.0007852631269131152, | |
| "loss": 3.0845, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.0007823198728514246, | |
| "loss": 3.0816, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.0007793795620437957, | |
| "loss": 3.0831, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.000776436307982105, | |
| "loss": 3.08, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 0.0007734930539204145, | |
| "loss": 3.0819, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 0.0007705497998587239, | |
| "loss": 3.0834, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 0.0007676065457970332, | |
| "loss": 3.0828, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.40357407765829956, | |
| "eval_loss": 3.3846945762634277, | |
| "eval_runtime": 154.4025, | |
| "eval_samples_per_second": 375.117, | |
| "eval_steps_per_second": 5.861, | |
| "step": 111528 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 0.0007646662349894043, | |
| "loss": 3.0473, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 0.0007617259241817754, | |
| "loss": 3.016, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 0.0007587826701200847, | |
| "loss": 3.0227, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 0.0007558394160583941, | |
| "loss": 3.0298, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 0.0007528961619967036, | |
| "loss": 3.0296, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 0.0007499558511890746, | |
| "loss": 3.0314, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 0.000747012597127384, | |
| "loss": 3.0328, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.0007440722863197552, | |
| "loss": 3.0409, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 0.0007411290322580645, | |
| "loss": 3.0419, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 0.0007381857781963739, | |
| "loss": 3.0391, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.0007352425241346832, | |
| "loss": 3.0439, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 0.0007323022133270544, | |
| "loss": 3.0436, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.0007293589592653638, | |
| "loss": 3.04, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 0.0007264186484577348, | |
| "loss": 3.0447, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0007234753943960443, | |
| "loss": 3.0434, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 0.0007205321403343537, | |
| "loss": 3.0461, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 0.0007175918295267247, | |
| "loss": 3.0458, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.0007146485754650342, | |
| "loss": 3.0441, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 0.0007117053214033436, | |
| "loss": 3.0461, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.40513838492949666, | |
| "eval_loss": 3.3778438568115234, | |
| "eval_runtime": 154.6119, | |
| "eval_samples_per_second": 374.609, | |
| "eval_steps_per_second": 5.853, | |
| "step": 130116 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 0.0007087650105957146, | |
| "loss": 2.988, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.000705821756534024, | |
| "loss": 2.9811, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 0.0007028785024723335, | |
| "loss": 2.9861, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 0.0006999381916647045, | |
| "loss": 2.9932, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.0006969949376030139, | |
| "loss": 2.9978, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 0.0006940516835413233, | |
| "loss": 3.0015, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.0006911113727336944, | |
| "loss": 3.0033, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 0.0006881681186720038, | |
| "loss": 3.0043, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 0.0006852278078643748, | |
| "loss": 3.006, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "learning_rate": 0.0006822845538026843, | |
| "loss": 3.0098, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 0.0006793442429950554, | |
| "loss": 3.0068, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 0.0006764009889333647, | |
| "loss": 3.0113, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 0.0006734606781257359, | |
| "loss": 3.0073, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "learning_rate": 0.0006705174240640453, | |
| "loss": 3.0093, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 0.0006675741700023546, | |
| "loss": 3.0162, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 0.0006646338591947258, | |
| "loss": 3.0135, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 0.0006616906051330351, | |
| "loss": 3.0119, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 0.0006587473510713445, | |
| "loss": 3.0138, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.40690873602292965, | |
| "eval_loss": 3.3611538410186768, | |
| "eval_runtime": 154.5822, | |
| "eval_samples_per_second": 374.681, | |
| "eval_steps_per_second": 5.854, | |
| "step": 148704 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 0.0006558070402637157, | |
| "loss": 2.992, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.000652863786202025, | |
| "loss": 2.9452, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 0.0006499205321403344, | |
| "loss": 2.9566, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 0.0006469802213327054, | |
| "loss": 2.9569, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 0.0006440369672710148, | |
| "loss": 2.967, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 0.0006410966564633859, | |
| "loss": 2.9682, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 0.0006381534024016953, | |
| "loss": 2.9707, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 0.0006352101483400047, | |
| "loss": 2.9738, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "learning_rate": 0.0006322698375323758, | |
| "loss": 2.9788, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 0.0006293265834706852, | |
| "loss": 2.9785, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 0.0006263862726630562, | |
| "loss": 2.9808, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 0.0006234430186013657, | |
| "loss": 2.9812, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 0.0006205027077937367, | |
| "loss": 2.9814, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "learning_rate": 0.0006175623969861078, | |
| "loss": 2.9819, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 0.0006146191429244173, | |
| "loss": 2.9849, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 0.0006116758888627266, | |
| "loss": 2.9844, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 0.0006087355780550977, | |
| "loss": 2.9832, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 0.0006057923239934072, | |
| "loss": 2.9871, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 0.0006028490699317165, | |
| "loss": 2.9878, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.4077603883002312, | |
| "eval_loss": 3.3628807067871094, | |
| "eval_runtime": 153.5098, | |
| "eval_samples_per_second": 377.298, | |
| "eval_steps_per_second": 5.895, | |
| "step": 167292 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 0.0005999058158700259, | |
| "loss": 2.9379, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 0.0005969684483164587, | |
| "loss": 2.9276, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 0.0005940251942547681, | |
| "loss": 2.9287, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 0.0005910819401930775, | |
| "loss": 2.9377, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 0.0005881386861313868, | |
| "loss": 2.9419, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 0.000585198375323758, | |
| "loss": 2.9415, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "learning_rate": 0.0005822551212620674, | |
| "loss": 2.9457, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 0.0005793118672003767, | |
| "loss": 2.9493, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 0.0005763715563927478, | |
| "loss": 2.9504, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 0.0005734283023310573, | |
| "loss": 2.9514, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 0.0005704879915234283, | |
| "loss": 2.9518, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "learning_rate": 0.0005675447374617377, | |
| "loss": 2.9501, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 0.0005646044266541089, | |
| "loss": 2.9538, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 0.0005616611725924182, | |
| "loss": 2.9542, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 0.0005587179185307276, | |
| "loss": 2.9587, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 0.0005557776077230988, | |
| "loss": 2.9585, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 0.0005528343536614081, | |
| "loss": 2.958, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 0.0005498910995997175, | |
| "loss": 2.9634, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.40929372517623736, | |
| "eval_loss": 3.348891496658325, | |
| "eval_runtime": 153.7083, | |
| "eval_samples_per_second": 376.811, | |
| "eval_steps_per_second": 5.888, | |
| "step": 185880 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 0.0005469507887920885, | |
| "loss": 2.9567, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 0.0005440104779844597, | |
| "loss": 2.8986, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 0.000541067223922769, | |
| "loss": 2.8997, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 0.0005381239698610783, | |
| "loss": 2.9067, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 0.0005351807157993878, | |
| "loss": 2.9087, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 0.0005322404049917589, | |
| "loss": 2.9194, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 10.33, | |
| "learning_rate": 0.0005292971509300682, | |
| "loss": 2.9176, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 0.0005263568401224393, | |
| "loss": 2.9252, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 10.44, | |
| "learning_rate": 0.0005234135860607488, | |
| "loss": 2.9248, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 0.0005204703319990581, | |
| "loss": 2.9242, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 10.54, | |
| "learning_rate": 0.0005175300211914292, | |
| "loss": 2.9289, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 10.6, | |
| "learning_rate": 0.0005145897103838003, | |
| "loss": 2.9293, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "learning_rate": 0.0005116493995761714, | |
| "loss": 2.9309, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 0.0005087061455144808, | |
| "loss": 2.9319, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 0.0005057628914527902, | |
| "loss": 2.9337, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 0.0005028196373910996, | |
| "loss": 2.9396, | |
| "step": 201000 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 0.0004998793265834707, | |
| "loss": 2.9383, | |
| "step": 202000 | |
| }, | |
| { | |
| "epoch": 10.92, | |
| "learning_rate": 0.00049693607252178, | |
| "loss": 2.9397, | |
| "step": 203000 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 0.0004939928184600895, | |
| "loss": 2.9347, | |
| "step": 204000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.40961330472053986, | |
| "eval_loss": 3.3615686893463135, | |
| "eval_runtime": 153.6771, | |
| "eval_samples_per_second": 376.888, | |
| "eval_steps_per_second": 5.889, | |
| "step": 204468 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 0.0004910495643983989, | |
| "loss": 2.9056, | |
| "step": 205000 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 0.0004881063103367083, | |
| "loss": 2.88, | |
| "step": 206000 | |
| }, | |
| { | |
| "epoch": 11.14, | |
| "learning_rate": 0.0004851630562750177, | |
| "loss": 2.8839, | |
| "step": 207000 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "learning_rate": 0.00048222274546738875, | |
| "loss": 2.892, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 0.00047927949140569813, | |
| "loss": 2.8946, | |
| "step": 209000 | |
| }, | |
| { | |
| "epoch": 11.3, | |
| "learning_rate": 0.00047633918059806926, | |
| "loss": 2.8973, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 0.00047339592653637864, | |
| "loss": 2.8936, | |
| "step": 211000 | |
| }, | |
| { | |
| "epoch": 11.41, | |
| "learning_rate": 0.000470452672474688, | |
| "loss": 2.8995, | |
| "step": 212000 | |
| }, | |
| { | |
| "epoch": 11.46, | |
| "learning_rate": 0.00046750941841299746, | |
| "loss": 2.9055, | |
| "step": 213000 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 0.00046456910760536853, | |
| "loss": 2.9058, | |
| "step": 214000 | |
| }, | |
| { | |
| "epoch": 11.57, | |
| "learning_rate": 0.0004616258535436779, | |
| "loss": 2.9054, | |
| "step": 215000 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 0.000458685542736049, | |
| "loss": 2.9091, | |
| "step": 216000 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 0.0004557422886743584, | |
| "loss": 2.9104, | |
| "step": 217000 | |
| }, | |
| { | |
| "epoch": 11.73, | |
| "learning_rate": 0.00045279903461266775, | |
| "loss": 2.9121, | |
| "step": 218000 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 0.00044985578055097713, | |
| "loss": 2.9101, | |
| "step": 219000 | |
| }, | |
| { | |
| "epoch": 11.84, | |
| "learning_rate": 0.00044691252648928657, | |
| "loss": 2.9087, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "learning_rate": 0.00044397515893571934, | |
| "loss": 2.9115, | |
| "step": 221000 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "learning_rate": 0.0004410319048740287, | |
| "loss": 2.9145, | |
| "step": 222000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 0.0004380915940663998, | |
| "loss": 2.9136, | |
| "step": 223000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4097346334270798, | |
| "eval_loss": 3.3726375102996826, | |
| "eval_runtime": 153.7904, | |
| "eval_samples_per_second": 376.61, | |
| "eval_steps_per_second": 5.885, | |
| "step": 223056 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 0.00043514834000470923, | |
| "loss": 2.8577, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 0.0004322050859430186, | |
| "loss": 2.8637, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "learning_rate": 0.0004292647751353897, | |
| "loss": 2.8622, | |
| "step": 226000 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 0.00042632152107369907, | |
| "loss": 2.866, | |
| "step": 227000 | |
| }, | |
| { | |
| "epoch": 12.27, | |
| "learning_rate": 0.0004233812102660702, | |
| "loss": 2.872, | |
| "step": 228000 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 0.0004204379562043796, | |
| "loss": 2.8781, | |
| "step": 229000 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "learning_rate": 0.00041749764539675065, | |
| "loss": 2.8778, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 12.43, | |
| "learning_rate": 0.0004145543913350601, | |
| "loss": 2.8812, | |
| "step": 231000 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 0.00041161408052743117, | |
| "loss": 2.8793, | |
| "step": 232000 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "learning_rate": 0.00040867082646574055, | |
| "loss": 2.8836, | |
| "step": 233000 | |
| }, | |
| { | |
| "epoch": 12.59, | |
| "learning_rate": 0.0004057305156581116, | |
| "loss": 2.8861, | |
| "step": 234000 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "learning_rate": 0.000402787261596421, | |
| "loss": 2.8888, | |
| "step": 235000 | |
| }, | |
| { | |
| "epoch": 12.7, | |
| "learning_rate": 0.0003998440075347304, | |
| "loss": 2.889, | |
| "step": 236000 | |
| }, | |
| { | |
| "epoch": 12.75, | |
| "learning_rate": 0.00039690369672710146, | |
| "loss": 2.8932, | |
| "step": 237000 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.0003939604426654109, | |
| "loss": 2.8912, | |
| "step": 238000 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "learning_rate": 0.00039102013185778197, | |
| "loss": 2.8937, | |
| "step": 239000 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "learning_rate": 0.00038807687779609135, | |
| "loss": 2.8919, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 12.97, | |
| "learning_rate": 0.0003851365669884624, | |
| "loss": 2.8947, | |
| "step": 241000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.4099391320886011, | |
| "eval_loss": 3.3682475090026855, | |
| "eval_runtime": 153.4886, | |
| "eval_samples_per_second": 377.35, | |
| "eval_steps_per_second": 5.896, | |
| "step": 241644 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 0.00038219331292677186, | |
| "loss": 2.872, | |
| "step": 242000 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "learning_rate": 0.00037925005886508124, | |
| "loss": 2.8318, | |
| "step": 243000 | |
| }, | |
| { | |
| "epoch": 13.13, | |
| "learning_rate": 0.0003763068048033906, | |
| "loss": 2.8474, | |
| "step": 244000 | |
| }, | |
| { | |
| "epoch": 13.18, | |
| "learning_rate": 0.0003733664939957617, | |
| "loss": 2.8451, | |
| "step": 245000 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "learning_rate": 0.00037042323993407113, | |
| "loss": 2.8491, | |
| "step": 246000 | |
| }, | |
| { | |
| "epoch": 13.29, | |
| "learning_rate": 0.0003674858723805039, | |
| "loss": 2.8551, | |
| "step": 247000 | |
| }, | |
| { | |
| "epoch": 13.34, | |
| "learning_rate": 0.0003645426183188133, | |
| "loss": 2.8536, | |
| "step": 248000 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "learning_rate": 0.0003615993642571227, | |
| "loss": 2.8543, | |
| "step": 249000 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 0.0003586561101954321, | |
| "loss": 2.8591, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 0.0003557157993878032, | |
| "loss": 2.8684, | |
| "step": 251000 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 0.00035277254532611256, | |
| "loss": 2.8659, | |
| "step": 252000 | |
| }, | |
| { | |
| "epoch": 13.61, | |
| "learning_rate": 0.00034983223451848363, | |
| "loss": 2.8646, | |
| "step": 253000 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 0.000346888980456793, | |
| "loss": 2.8666, | |
| "step": 254000 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "learning_rate": 0.0003439457263951024, | |
| "loss": 2.8756, | |
| "step": 255000 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 0.00034100247233341183, | |
| "loss": 2.8695, | |
| "step": 256000 | |
| }, | |
| { | |
| "epoch": 13.83, | |
| "learning_rate": 0.0003380621615257829, | |
| "loss": 2.8698, | |
| "step": 257000 | |
| }, | |
| { | |
| "epoch": 13.88, | |
| "learning_rate": 0.0003351189074640923, | |
| "loss": 2.8761, | |
| "step": 258000 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 0.00033217859665646336, | |
| "loss": 2.8731, | |
| "step": 259000 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 0.0003292353425947728, | |
| "loss": 2.8789, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.40986637517321756, | |
| "eval_loss": 3.381711483001709, | |
| "eval_runtime": 153.7177, | |
| "eval_samples_per_second": 376.788, | |
| "eval_steps_per_second": 5.887, | |
| "step": 260232 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 0.0003262920885330822, | |
| "loss": 2.8316, | |
| "step": 261000 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 0.00032335177772545325, | |
| "loss": 2.8181, | |
| "step": 262000 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 0.0003204114669178244, | |
| "loss": 2.8218, | |
| "step": 263000 | |
| }, | |
| { | |
| "epoch": 14.2, | |
| "learning_rate": 0.00031746821285613376, | |
| "loss": 2.833, | |
| "step": 264000 | |
| }, | |
| { | |
| "epoch": 14.26, | |
| "learning_rate": 0.00031452495879444314, | |
| "loss": 2.8368, | |
| "step": 265000 | |
| }, | |
| { | |
| "epoch": 14.31, | |
| "learning_rate": 0.0003115817047327525, | |
| "loss": 2.8377, | |
| "step": 266000 | |
| }, | |
| { | |
| "epoch": 14.36, | |
| "learning_rate": 0.00030864139392512366, | |
| "loss": 2.8404, | |
| "step": 267000 | |
| }, | |
| { | |
| "epoch": 14.42, | |
| "learning_rate": 0.00030569813986343304, | |
| "loss": 2.8433, | |
| "step": 268000 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 0.0003027578290558041, | |
| "loss": 2.8446, | |
| "step": 269000 | |
| }, | |
| { | |
| "epoch": 14.53, | |
| "learning_rate": 0.00029981457499411355, | |
| "loss": 2.8467, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 14.58, | |
| "learning_rate": 0.00029687132093242293, | |
| "loss": 2.8458, | |
| "step": 271000 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 0.00029393101012479395, | |
| "loss": 2.8488, | |
| "step": 272000 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 0.00029098775606310333, | |
| "loss": 2.8482, | |
| "step": 273000 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "learning_rate": 0.00028804744525547446, | |
| "loss": 2.8502, | |
| "step": 274000 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "learning_rate": 0.00028510419119378384, | |
| "loss": 2.8478, | |
| "step": 275000 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "learning_rate": 0.0002821609371320932, | |
| "loss": 2.8539, | |
| "step": 276000 | |
| }, | |
| { | |
| "epoch": 14.9, | |
| "learning_rate": 0.0002792176830704026, | |
| "loss": 2.8538, | |
| "step": 277000 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "learning_rate": 0.00027628031551683543, | |
| "loss": 2.8559, | |
| "step": 278000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.4098763851274116, | |
| "eval_loss": 3.384662628173828, | |
| "eval_runtime": 153.8197, | |
| "eval_samples_per_second": 376.538, | |
| "eval_steps_per_second": 5.884, | |
| "step": 278820 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 0.0002733370614551448, | |
| "loss": 2.8412, | |
| "step": 279000 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "learning_rate": 0.0002703938073934542, | |
| "loss": 2.8071, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 15.12, | |
| "learning_rate": 0.0002674534965858253, | |
| "loss": 2.8047, | |
| "step": 281000 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 0.0002645102425241347, | |
| "loss": 2.8115, | |
| "step": 282000 | |
| }, | |
| { | |
| "epoch": 15.22, | |
| "learning_rate": 0.0002615699317165058, | |
| "loss": 2.815, | |
| "step": 283000 | |
| }, | |
| { | |
| "epoch": 15.28, | |
| "learning_rate": 0.00025862667765481516, | |
| "loss": 2.8199, | |
| "step": 284000 | |
| }, | |
| { | |
| "epoch": 15.33, | |
| "learning_rate": 0.0002556834235931246, | |
| "loss": 2.8153, | |
| "step": 285000 | |
| }, | |
| { | |
| "epoch": 15.39, | |
| "learning_rate": 0.00025274016953143397, | |
| "loss": 2.8235, | |
| "step": 286000 | |
| }, | |
| { | |
| "epoch": 15.44, | |
| "learning_rate": 0.00024979691546974335, | |
| "loss": 2.8218, | |
| "step": 287000 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 0.00024685366140805273, | |
| "loss": 2.83, | |
| "step": 288000 | |
| }, | |
| { | |
| "epoch": 15.55, | |
| "learning_rate": 0.00024391629385448553, | |
| "loss": 2.8257, | |
| "step": 289000 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "learning_rate": 0.0002409730397927949, | |
| "loss": 2.8301, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "learning_rate": 0.00023802978573110432, | |
| "loss": 2.8286, | |
| "step": 291000 | |
| }, | |
| { | |
| "epoch": 15.71, | |
| "learning_rate": 0.0002350894749234754, | |
| "loss": 2.8299, | |
| "step": 292000 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "learning_rate": 0.0002321462208617848, | |
| "loss": 2.8329, | |
| "step": 293000 | |
| }, | |
| { | |
| "epoch": 15.82, | |
| "learning_rate": 0.00022920591005415588, | |
| "loss": 2.8295, | |
| "step": 294000 | |
| }, | |
| { | |
| "epoch": 15.87, | |
| "learning_rate": 0.00022626265599246526, | |
| "loss": 2.8303, | |
| "step": 295000 | |
| }, | |
| { | |
| "epoch": 15.92, | |
| "learning_rate": 0.00022332234518483636, | |
| "loss": 2.8323, | |
| "step": 296000 | |
| }, | |
| { | |
| "epoch": 15.98, | |
| "learning_rate": 0.00022037909112314574, | |
| "loss": 2.8374, | |
| "step": 297000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.4101721826328907, | |
| "eval_loss": 3.3835206031799316, | |
| "eval_runtime": 153.6226, | |
| "eval_samples_per_second": 377.021, | |
| "eval_steps_per_second": 5.891, | |
| "step": 297408 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "learning_rate": 0.00021743583706145515, | |
| "loss": 2.8042, | |
| "step": 298000 | |
| }, | |
| { | |
| "epoch": 16.09, | |
| "learning_rate": 0.00021449258299976453, | |
| "loss": 2.793, | |
| "step": 299000 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 0.00021155227219213564, | |
| "loss": 2.795, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "learning_rate": 0.00020860901813044502, | |
| "loss": 2.7934, | |
| "step": 301000 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "learning_rate": 0.00020566870732281612, | |
| "loss": 2.803, | |
| "step": 302000 | |
| }, | |
| { | |
| "epoch": 16.3, | |
| "learning_rate": 0.0002027283965151872, | |
| "loss": 2.8032, | |
| "step": 303000 | |
| }, | |
| { | |
| "epoch": 16.35, | |
| "learning_rate": 0.00019978514245349658, | |
| "loss": 2.8072, | |
| "step": 304000 | |
| }, | |
| { | |
| "epoch": 16.41, | |
| "learning_rate": 0.00019684188839180598, | |
| "loss": 2.7976, | |
| "step": 305000 | |
| }, | |
| { | |
| "epoch": 16.46, | |
| "learning_rate": 0.00019389863433011536, | |
| "loss": 2.8067, | |
| "step": 306000 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "learning_rate": 0.00019095538026842477, | |
| "loss": 2.8109, | |
| "step": 307000 | |
| }, | |
| { | |
| "epoch": 16.57, | |
| "learning_rate": 0.00018801212620673415, | |
| "loss": 2.8114, | |
| "step": 308000 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 0.00018507181539910526, | |
| "loss": 2.8102, | |
| "step": 309000 | |
| }, | |
| { | |
| "epoch": 16.68, | |
| "learning_rate": 0.00018213150459147633, | |
| "loss": 2.8082, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 16.73, | |
| "learning_rate": 0.00017918825052978574, | |
| "loss": 2.8124, | |
| "step": 311000 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "learning_rate": 0.00017624499646809515, | |
| "loss": 2.8151, | |
| "step": 312000 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "learning_rate": 0.00017330174240640453, | |
| "loss": 2.8114, | |
| "step": 313000 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "learning_rate": 0.00017036143159877563, | |
| "loss": 2.8139, | |
| "step": 314000 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 0.00016742112079114668, | |
| "loss": 2.8117, | |
| "step": 315000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.4099636531173315, | |
| "eval_loss": 3.3939976692199707, | |
| "eval_runtime": 153.623, | |
| "eval_samples_per_second": 377.02, | |
| "eval_steps_per_second": 5.891, | |
| "step": 315996 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 0.0001644778667294561, | |
| "loss": 2.8163, | |
| "step": 316000 | |
| }, | |
| { | |
| "epoch": 17.05, | |
| "learning_rate": 0.00016153461266776547, | |
| "loss": 2.774, | |
| "step": 317000 | |
| }, | |
| { | |
| "epoch": 17.11, | |
| "learning_rate": 0.00015859135860607488, | |
| "loss": 2.776, | |
| "step": 318000 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "learning_rate": 0.00015565104779844598, | |
| "loss": 2.7788, | |
| "step": 319000 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "learning_rate": 0.00015270779373675536, | |
| "loss": 2.7803, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 17.27, | |
| "learning_rate": 0.00014976748292912646, | |
| "loss": 2.7843, | |
| "step": 321000 | |
| }, | |
| { | |
| "epoch": 17.32, | |
| "learning_rate": 0.00014682422886743584, | |
| "loss": 2.7863, | |
| "step": 322000 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 0.00014388097480574525, | |
| "loss": 2.7861, | |
| "step": 323000 | |
| }, | |
| { | |
| "epoch": 17.43, | |
| "learning_rate": 0.00014093772074405463, | |
| "loss": 2.787, | |
| "step": 324000 | |
| }, | |
| { | |
| "epoch": 17.48, | |
| "learning_rate": 0.0001380003531904874, | |
| "loss": 2.7927, | |
| "step": 325000 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "learning_rate": 0.00013505709912879678, | |
| "loss": 2.7909, | |
| "step": 326000 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 0.0001321138450671062, | |
| "loss": 2.7929, | |
| "step": 327000 | |
| }, | |
| { | |
| "epoch": 17.65, | |
| "learning_rate": 0.0001291705910054156, | |
| "loss": 2.7935, | |
| "step": 328000 | |
| }, | |
| { | |
| "epoch": 17.7, | |
| "learning_rate": 0.00012623028019778667, | |
| "loss": 2.7976, | |
| "step": 329000 | |
| }, | |
| { | |
| "epoch": 17.75, | |
| "learning_rate": 0.00012328702613609608, | |
| "loss": 2.7943, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 17.81, | |
| "learning_rate": 0.00012034377207440548, | |
| "loss": 2.7978, | |
| "step": 331000 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 0.00011740346126677655, | |
| "loss": 2.7958, | |
| "step": 332000 | |
| }, | |
| { | |
| "epoch": 17.91, | |
| "learning_rate": 0.00011446315045914764, | |
| "loss": 2.7956, | |
| "step": 333000 | |
| }, | |
| { | |
| "epoch": 17.97, | |
| "learning_rate": 0.00011151989639745704, | |
| "loss": 2.7969, | |
| "step": 334000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.41020543717870317, | |
| "eval_loss": 3.4024271965026855, | |
| "eval_runtime": 153.4119, | |
| "eval_samples_per_second": 377.539, | |
| "eval_steps_per_second": 5.899, | |
| "step": 334584 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "learning_rate": 0.00010857664233576643, | |
| "loss": 2.7848, | |
| "step": 335000 | |
| }, | |
| { | |
| "epoch": 18.08, | |
| "learning_rate": 0.00010563338827407581, | |
| "loss": 2.7625, | |
| "step": 336000 | |
| }, | |
| { | |
| "epoch": 18.13, | |
| "learning_rate": 0.0001026901342123852, | |
| "loss": 2.7651, | |
| "step": 337000 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 9.975276665881799e-05, | |
| "loss": 2.7679, | |
| "step": 338000 | |
| }, | |
| { | |
| "epoch": 18.24, | |
| "learning_rate": 9.68095125971274e-05, | |
| "loss": 2.7718, | |
| "step": 339000 | |
| }, | |
| { | |
| "epoch": 18.29, | |
| "learning_rate": 9.386625853543678e-05, | |
| "loss": 2.7709, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 18.35, | |
| "learning_rate": 9.092300447374617e-05, | |
| "loss": 2.7699, | |
| "step": 341000 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 8.798269366611726e-05, | |
| "loss": 2.7741, | |
| "step": 342000 | |
| }, | |
| { | |
| "epoch": 18.45, | |
| "learning_rate": 8.503943960442666e-05, | |
| "loss": 2.773, | |
| "step": 343000 | |
| }, | |
| { | |
| "epoch": 18.51, | |
| "learning_rate": 8.209618554273605e-05, | |
| "loss": 2.7768, | |
| "step": 344000 | |
| }, | |
| { | |
| "epoch": 18.56, | |
| "learning_rate": 7.915587473510713e-05, | |
| "loss": 2.7741, | |
| "step": 345000 | |
| }, | |
| { | |
| "epoch": 18.61, | |
| "learning_rate": 7.621262067341653e-05, | |
| "loss": 2.7753, | |
| "step": 346000 | |
| }, | |
| { | |
| "epoch": 18.67, | |
| "learning_rate": 7.327230986578762e-05, | |
| "loss": 2.78, | |
| "step": 347000 | |
| }, | |
| { | |
| "epoch": 18.72, | |
| "learning_rate": 7.032905580409702e-05, | |
| "loss": 2.7727, | |
| "step": 348000 | |
| }, | |
| { | |
| "epoch": 18.78, | |
| "learning_rate": 6.738580174240641e-05, | |
| "loss": 2.7799, | |
| "step": 349000 | |
| }, | |
| { | |
| "epoch": 18.83, | |
| "learning_rate": 6.444549093477749e-05, | |
| "loss": 2.7722, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 18.88, | |
| "learning_rate": 6.150223687308688e-05, | |
| "loss": 2.7772, | |
| "step": 351000 | |
| }, | |
| { | |
| "epoch": 18.94, | |
| "learning_rate": 5.856192606545797e-05, | |
| "loss": 2.7803, | |
| "step": 352000 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 5.562455851189075e-05, | |
| "loss": 2.7772, | |
| "step": 353000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.41049068728278315, | |
| "eval_loss": 3.403170108795166, | |
| "eval_runtime": 153.4546, | |
| "eval_samples_per_second": 377.434, | |
| "eval_steps_per_second": 5.898, | |
| "step": 353172 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "learning_rate": 5.2681304450200143e-05, | |
| "loss": 2.7594, | |
| "step": 354000 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 4.973805038850954e-05, | |
| "loss": 2.7572, | |
| "step": 355000 | |
| }, | |
| { | |
| "epoch": 19.15, | |
| "learning_rate": 4.679479632681893e-05, | |
| "loss": 2.7582, | |
| "step": 356000 | |
| }, | |
| { | |
| "epoch": 19.21, | |
| "learning_rate": 4.385154226512833e-05, | |
| "loss": 2.7597, | |
| "step": 357000 | |
| }, | |
| { | |
| "epoch": 19.26, | |
| "learning_rate": 4.090828820343772e-05, | |
| "loss": 2.7565, | |
| "step": 358000 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "learning_rate": 3.7965034141747114e-05, | |
| "loss": 2.7595, | |
| "step": 359000 | |
| }, | |
| { | |
| "epoch": 19.37, | |
| "learning_rate": 3.50247233341182e-05, | |
| "loss": 2.7607, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 19.42, | |
| "learning_rate": 3.20814692724276e-05, | |
| "loss": 2.7571, | |
| "step": 361000 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "learning_rate": 2.9138215210736992e-05, | |
| "loss": 2.7573, | |
| "step": 362000 | |
| }, | |
| { | |
| "epoch": 19.53, | |
| "learning_rate": 2.6197904403108075e-05, | |
| "loss": 2.7611, | |
| "step": 363000 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "learning_rate": 2.3254650341417473e-05, | |
| "loss": 2.7623, | |
| "step": 364000 | |
| }, | |
| { | |
| "epoch": 19.64, | |
| "learning_rate": 2.0314339533788555e-05, | |
| "loss": 2.7598, | |
| "step": 365000 | |
| }, | |
| { | |
| "epoch": 19.69, | |
| "learning_rate": 1.737402872615964e-05, | |
| "loss": 2.756, | |
| "step": 366000 | |
| }, | |
| { | |
| "epoch": 19.74, | |
| "learning_rate": 1.4430774664469037e-05, | |
| "loss": 2.7586, | |
| "step": 367000 | |
| }, | |
| { | |
| "epoch": 19.8, | |
| "learning_rate": 1.1487520602778433e-05, | |
| "loss": 2.7585, | |
| "step": 368000 | |
| }, | |
| { | |
| "epoch": 19.85, | |
| "learning_rate": 8.544266541087827e-06, | |
| "loss": 2.7575, | |
| "step": 369000 | |
| }, | |
| { | |
| "epoch": 19.91, | |
| "learning_rate": 5.6010124793972215e-06, | |
| "loss": 2.7629, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 19.96, | |
| "learning_rate": 2.6607016717683068e-06, | |
| "loss": 2.76, | |
| "step": 371000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.41051635038682427, | |
| "eval_loss": 3.4108598232269287, | |
| "eval_runtime": 153.6005, | |
| "eval_samples_per_second": 377.076, | |
| "eval_steps_per_second": 5.892, | |
| "step": 371760 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 371760, | |
| "total_flos": 1.5663200871168e+18, | |
| "train_loss": 3.0294145256159046, | |
| "train_runtime": 80976.6072, | |
| "train_samples_per_second": 146.908, | |
| "train_steps_per_second": 4.591 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 371760, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 5000, | |
| "total_flos": 1.5663200871168e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |