| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 371720, | |
| "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": 6.8789, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.875e-05, | |
| "loss": 5.3718, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.8125e-05, | |
| "loss": 5.041, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.75e-05, | |
| "loss": 4.8245, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.6874999999999994e-05, | |
| "loss": 4.6461, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 5.625e-05, | |
| "loss": 4.5097, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.5625e-05, | |
| "loss": 4.3902, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 7.5e-05, | |
| "loss": 4.2887, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 8.437499999999999e-05, | |
| "loss": 4.2059, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 9.374999999999999e-05, | |
| "loss": 4.1373, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.000103115625, | |
| "loss": 4.0669, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.000112490625, | |
| "loss": 4.006, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00012185624999999998, | |
| "loss": 3.9433, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00013123125, | |
| "loss": 3.8913, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.000140596875, | |
| "loss": 3.8398, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0001499625, | |
| "loss": 3.8043, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.00015933749999999996, | |
| "loss": 3.7652, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.000168703125, | |
| "loss": 3.7399, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3474263441322527, | |
| "eval_loss": 3.9066405296325684, | |
| "eval_runtime": 152.4173, | |
| "eval_samples_per_second": 380.009, | |
| "eval_steps_per_second": 5.938, | |
| "step": 18586 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.000178078125, | |
| "loss": 3.698, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00018743437499999996, | |
| "loss": 3.6664, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.00019680937499999996, | |
| "loss": 3.6503, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00020618437499999995, | |
| "loss": 3.6308, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.00021555937499999998, | |
| "loss": 3.6101, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00022493437499999998, | |
| "loss": 3.5997, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00023429999999999998, | |
| "loss": 3.576, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00024367499999999997, | |
| "loss": 3.5679, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.000253040625, | |
| "loss": 3.5542, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.000262415625, | |
| "loss": 3.5366, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00027178124999999994, | |
| "loss": 3.5274, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.00028115624999999994, | |
| "loss": 3.5159, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.000290521875, | |
| "loss": 3.51, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.000299896875, | |
| "loss": 3.4984, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00029912663369833977, | |
| "loss": 3.4859, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.00029824443659484277, | |
| "loss": 3.47, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.00029736135641116214, | |
| "loss": 3.4615, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00029647827622748146, | |
| "loss": 3.4497, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00029559519604380077, | |
| "loss": 3.4368, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3750012764150741, | |
| "eval_loss": 3.6278717517852783, | |
| "eval_runtime": 153.9243, | |
| "eval_samples_per_second": 376.289, | |
| "eval_steps_per_second": 5.88, | |
| "step": 37172 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002947129989403037, | |
| "loss": 3.3908, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.00029382991875662304, | |
| "loss": 3.377, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.0002929477216531261, | |
| "loss": 3.3793, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.0002920646414694454, | |
| "loss": 3.3727, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002911824443659484, | |
| "loss": 3.364, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.0002903002472624514, | |
| "loss": 3.3582, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.0002894171670787707, | |
| "loss": 3.3503, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.00028853408689509004, | |
| "loss": 3.3459, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.00028765100671140936, | |
| "loss": 3.3414, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.00028676792652772867, | |
| "loss": 3.3321, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.000285884846344048, | |
| "loss": 3.3273, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.00028500264924055104, | |
| "loss": 3.323, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.00028411956905687036, | |
| "loss": 3.318, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.0002832364888731897, | |
| "loss": 3.3165, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0002823542917696927, | |
| "loss": 3.308, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.000281471211586012, | |
| "loss": 3.3037, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0002805881314023313, | |
| "loss": 3.2996, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.0002797059342988343, | |
| "loss": 3.294, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.38844945127587766, | |
| "eval_loss": 3.4854393005371094, | |
| "eval_runtime": 154.1208, | |
| "eval_samples_per_second": 375.809, | |
| "eval_steps_per_second": 5.872, | |
| "step": 55758 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.0002788228541151536, | |
| "loss": 3.2753, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.00027793977393147294, | |
| "loss": 3.2238, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.00027705757682797594, | |
| "loss": 3.2294, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 0.00027617449664429526, | |
| "loss": 3.2333, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.00027529229954079826, | |
| "loss": 3.2321, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 0.0002744092193571176, | |
| "loss": 3.2265, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.00027352702225362063, | |
| "loss": 3.2317, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.00027264394206993995, | |
| "loss": 3.2312, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.00027176174496644295, | |
| "loss": 3.2297, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 0.00027087866478276226, | |
| "loss": 3.2261, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.0002699955845990816, | |
| "loss": 3.224, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0002691125044154009, | |
| "loss": 3.224, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 0.0002682303073119039, | |
| "loss": 3.2252, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.0002673472271282232, | |
| "loss": 3.2151, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 0.00026646414694454253, | |
| "loss": 3.2164, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 0.00026558106676086184, | |
| "loss": 3.2184, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.0002646988696573649, | |
| "loss": 3.2157, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.0002638157894736842, | |
| "loss": 3.2151, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 0.00026293447545037085, | |
| "loss": 3.2094, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3968133961777415, | |
| "eval_loss": 3.417848587036133, | |
| "eval_runtime": 153.5385, | |
| "eval_samples_per_second": 377.234, | |
| "eval_steps_per_second": 5.894, | |
| "step": 74344 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 0.00026205139526669016, | |
| "loss": 3.1653, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.00026116831508300953, | |
| "loss": 3.147, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 0.00026028611797951253, | |
| "loss": 3.1487, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 0.00025940303779583185, | |
| "loss": 3.1529, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.00025852084069233485, | |
| "loss": 3.1547, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 0.00025763776050865417, | |
| "loss": 3.1556, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.00025675556340515717, | |
| "loss": 3.1526, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 0.0002558724832214765, | |
| "loss": 3.1545, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 0.0002549894030377958, | |
| "loss": 3.1529, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.0002541063228541151, | |
| "loss": 3.1541, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.00025322412575061817, | |
| "loss": 3.1537, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 0.0002523410455669375, | |
| "loss": 3.1532, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 0.00025145884846344043, | |
| "loss": 3.1499, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.00025057576827975975, | |
| "loss": 3.1522, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 0.0002496944542564465, | |
| "loss": 3.1509, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.0002488113740727658, | |
| "loss": 3.1508, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 0.0002479282938890851, | |
| "loss": 3.1537, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.00024704521370540444, | |
| "loss": 3.1515, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.40092110142528536, | |
| "eval_loss": 3.3861262798309326, | |
| "eval_runtime": 154.276, | |
| "eval_samples_per_second": 375.431, | |
| "eval_steps_per_second": 5.866, | |
| "step": 92930 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.00024616213352172375, | |
| "loss": 3.1449, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 0.00024527993641822675, | |
| "loss": 3.0802, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.00024439685623454607, | |
| "loss": 3.0841, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 0.00024351377605086539, | |
| "loss": 3.0862, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 0.0002426306958671847, | |
| "loss": 3.0941, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.00024174849876368773, | |
| "loss": 3.0972, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.00024086630166019073, | |
| "loss": 3.1007, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.00023998322147651005, | |
| "loss": 3.1025, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.00023910014129282936, | |
| "loss": 3.1013, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 0.00023821794418933236, | |
| "loss": 3.1017, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 0.00023733663016601905, | |
| "loss": 3.1005, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.00023645354998233837, | |
| "loss": 3.1006, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 0.00023557046979865768, | |
| "loss": 3.1027, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.000234687389614977, | |
| "loss": 3.1011, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.00023380430943129637, | |
| "loss": 3.1077, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.00023292122924761568, | |
| "loss": 3.1042, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 0.00023203814906393497, | |
| "loss": 3.1058, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 0.00023115595196043797, | |
| "loss": 3.1069, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 0.0002302728717767573, | |
| "loss": 3.1023, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.40412974020250664, | |
| "eval_loss": 3.3599889278411865, | |
| "eval_runtime": 153.6955, | |
| "eval_samples_per_second": 376.849, | |
| "eval_steps_per_second": 5.888, | |
| "step": 111516 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 0.00022938979159307663, | |
| "loss": 3.0668, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 0.00022850671140939595, | |
| "loss": 3.0364, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 0.00022762451430589895, | |
| "loss": 3.0447, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 0.00022674143412221827, | |
| "loss": 3.0456, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 0.0002258592370187213, | |
| "loss": 3.0496, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 0.0002249761568350406, | |
| "loss": 3.0544, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 0.0002240939597315436, | |
| "loss": 3.0547, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.00022321087954786293, | |
| "loss": 3.0562, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 0.00022232779936418224, | |
| "loss": 3.0571, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 0.00022144648534086893, | |
| "loss": 3.0577, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.00022056340515718824, | |
| "loss": 3.0587, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 0.00021968032497350756, | |
| "loss": 3.0642, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.00021879724478982688, | |
| "loss": 3.0635, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 0.0002179159307665136, | |
| "loss": 3.0656, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0002170328505828329, | |
| "loss": 3.0616, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 0.0002161506534793359, | |
| "loss": 3.0616, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 0.00021526757329565522, | |
| "loss": 3.0652, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.00021438449311197457, | |
| "loss": 3.0628, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 0.00021350229600847757, | |
| "loss": 3.0643, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.40467913612227785, | |
| "eval_loss": 3.3565268516540527, | |
| "eval_runtime": 153.5526, | |
| "eval_samples_per_second": 377.2, | |
| "eval_steps_per_second": 5.894, | |
| "step": 130102 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 0.00021261921582479688, | |
| "loss": 3.0021, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.0002117361356411162, | |
| "loss": 2.9991, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 0.00021085305545743551, | |
| "loss": 3.0031, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 0.00020997085835393854, | |
| "loss": 3.0117, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.00020908777817025786, | |
| "loss": 3.0148, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 0.00020820558106676083, | |
| "loss": 3.0193, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.00020732250088308015, | |
| "loss": 3.0201, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 0.00020643942069939946, | |
| "loss": 3.022, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 0.0002055572235959025, | |
| "loss": 3.024, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "learning_rate": 0.0002046741434122218, | |
| "loss": 3.0277, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 0.0002037919463087248, | |
| "loss": 3.0261, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 0.00020290886612504412, | |
| "loss": 3.0273, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 0.00020202755210173084, | |
| "loss": 3.028, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "learning_rate": 0.00020114447191805015, | |
| "loss": 3.0314, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 0.00020026227481455313, | |
| "loss": 3.0292, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 0.00019937919463087244, | |
| "loss": 3.0295, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 0.00019849611444719181, | |
| "loss": 3.0302, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 0.00019761303426351113, | |
| "loss": 3.0294, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.40586479136659714, | |
| "eval_loss": 3.357483386993408, | |
| "eval_runtime": 154.0037, | |
| "eval_samples_per_second": 376.095, | |
| "eval_steps_per_second": 5.876, | |
| "step": 148688 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 0.0001967308371600141, | |
| "loss": 3.0109, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.0001958486400565171, | |
| "loss": 2.9662, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 0.00019496555987283645, | |
| "loss": 2.9733, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 0.00019408247968915576, | |
| "loss": 2.9748, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 0.00019319939950547508, | |
| "loss": 2.9811, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 0.00019231720240197808, | |
| "loss": 2.981, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 0.00019143500529848108, | |
| "loss": 2.9822, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 0.00019055192511480042, | |
| "loss": 2.9871, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "learning_rate": 0.00018966884493111974, | |
| "loss": 2.9908, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 0.00018878576474743906, | |
| "loss": 2.9932, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 0.00018790356764394203, | |
| "loss": 2.9941, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 0.00018702048746026135, | |
| "loss": 2.9974, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 0.00018613829035676437, | |
| "loss": 2.9987, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "learning_rate": 0.0001852552101730837, | |
| "loss": 2.9974, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 0.0001843730130695867, | |
| "loss": 3.0005, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 0.00018349081596608972, | |
| "loss": 3.0045, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 0.00018260773578240904, | |
| "loss": 3.0006, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 0.00018172465559872835, | |
| "loss": 3.001, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "learning_rate": 0.00018084245849523135, | |
| "loss": 3.0007, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.40681383956403033, | |
| "eval_loss": 3.365987539291382, | |
| "eval_runtime": 153.9577, | |
| "eval_samples_per_second": 376.207, | |
| "eval_steps_per_second": 5.878, | |
| "step": 167274 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 0.00017995937831155067, | |
| "loss": 2.9554, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 0.00017907629812787, | |
| "loss": 2.9409, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 0.000178194101024373, | |
| "loss": 2.9439, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 0.00017731102084069233, | |
| "loss": 2.9518, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 0.0001764288237371953, | |
| "loss": 2.9546, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 0.00017554574355351462, | |
| "loss": 2.9557, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "learning_rate": 0.00017466354645001765, | |
| "loss": 2.9588, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 0.00017378046626633696, | |
| "loss": 2.9603, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 0.00017289738608265628, | |
| "loss": 2.965, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 0.0001720143058989756, | |
| "loss": 2.9669, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 0.0001711312257152949, | |
| "loss": 2.9681, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "learning_rate": 0.00017024991169198162, | |
| "loss": 2.9694, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 0.00016936683150830094, | |
| "loss": 2.9686, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 0.00016848375132462025, | |
| "loss": 2.9715, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 0.00016760067114093957, | |
| "loss": 2.9763, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 0.0001667184740374426, | |
| "loss": 2.9743, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 0.00016583539385376191, | |
| "loss": 2.973, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 0.00016495319675026492, | |
| "loss": 2.9771, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.40750740320742956, | |
| "eval_loss": 3.3512990474700928, | |
| "eval_runtime": 154.0319, | |
| "eval_samples_per_second": 376.026, | |
| "eval_steps_per_second": 5.875, | |
| "step": 185860 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 0.00016407011656658423, | |
| "loss": 2.9677, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 0.00016318703638290357, | |
| "loss": 2.9122, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 0.00016230483927940658, | |
| "loss": 2.9153, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 0.00016142175909572586, | |
| "loss": 2.9229, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 0.00016053867891204518, | |
| "loss": 2.9286, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 0.00015965648180854818, | |
| "loss": 2.9353, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 10.33, | |
| "learning_rate": 0.00015877340162486752, | |
| "loss": 2.9351, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 0.00015789032144118684, | |
| "loss": 2.9354, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 10.44, | |
| "learning_rate": 0.00015700812433768984, | |
| "loss": 2.9409, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 0.00015612504415400916, | |
| "loss": 2.9417, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 0.00015524196397032847, | |
| "loss": 2.9419, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 10.6, | |
| "learning_rate": 0.00015435888378664782, | |
| "loss": 2.9422, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "learning_rate": 0.00015347580360296713, | |
| "loss": 2.9414, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 0.00015259448957965382, | |
| "loss": 2.947, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 0.00015171140939597313, | |
| "loss": 2.9468, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 0.00015082832921229248, | |
| "loss": 2.9493, | |
| "step": 201000 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 0.00014994613210879548, | |
| "loss": 2.9447, | |
| "step": 202000 | |
| }, | |
| { | |
| "epoch": 10.92, | |
| "learning_rate": 0.0001490630519251148, | |
| "loss": 2.9464, | |
| "step": 203000 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "learning_rate": 0.0001481808548216178, | |
| "loss": 2.9526, | |
| "step": 204000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.4091822613238171, | |
| "eval_loss": 3.343301296234131, | |
| "eval_runtime": 157.5863, | |
| "eval_samples_per_second": 367.545, | |
| "eval_steps_per_second": 5.743, | |
| "step": 204446 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 0.0001472977746379371, | |
| "loss": 2.9178, | |
| "step": 205000 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 0.00014641469445425643, | |
| "loss": 2.8858, | |
| "step": 206000 | |
| }, | |
| { | |
| "epoch": 11.14, | |
| "learning_rate": 0.00014553161427057577, | |
| "loss": 2.8958, | |
| "step": 207000 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "learning_rate": 0.00014464853408689509, | |
| "loss": 2.8988, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 0.0001437663369833981, | |
| "loss": 2.9033, | |
| "step": 209000 | |
| }, | |
| { | |
| "epoch": 11.3, | |
| "learning_rate": 0.0001428832567997174, | |
| "loss": 2.9055, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 0.00014200017661603672, | |
| "loss": 2.9159, | |
| "step": 211000 | |
| }, | |
| { | |
| "epoch": 11.41, | |
| "learning_rate": 0.00014111709643235604, | |
| "loss": 2.9115, | |
| "step": 212000 | |
| }, | |
| { | |
| "epoch": 11.46, | |
| "learning_rate": 0.00014023578240904275, | |
| "loss": 2.9168, | |
| "step": 213000 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 0.00013935270222536206, | |
| "loss": 2.9187, | |
| "step": 214000 | |
| }, | |
| { | |
| "epoch": 11.57, | |
| "learning_rate": 0.00013846962204168135, | |
| "loss": 2.9202, | |
| "step": 215000 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 0.00013758742493818438, | |
| "loss": 2.9214, | |
| "step": 216000 | |
| }, | |
| { | |
| "epoch": 11.68, | |
| "learning_rate": 0.0001367043447545037, | |
| "loss": 2.922, | |
| "step": 217000 | |
| }, | |
| { | |
| "epoch": 11.73, | |
| "learning_rate": 0.000135821264570823, | |
| "loss": 2.9262, | |
| "step": 218000 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 0.00013493818438714233, | |
| "loss": 2.9226, | |
| "step": 219000 | |
| }, | |
| { | |
| "epoch": 11.84, | |
| "learning_rate": 0.00013405598728364536, | |
| "loss": 2.9286, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "learning_rate": 0.00013317290709996467, | |
| "loss": 2.9273, | |
| "step": 221000 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "learning_rate": 0.00013229070999646767, | |
| "loss": 2.9324, | |
| "step": 222000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 0.000131407629812787, | |
| "loss": 2.9307, | |
| "step": 223000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4093733876862222, | |
| "eval_loss": 3.354173421859741, | |
| "eval_runtime": 153.4542, | |
| "eval_samples_per_second": 377.442, | |
| "eval_steps_per_second": 5.898, | |
| "step": 223032 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 0.0001305245496291063, | |
| "loss": 2.8693, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 12.11, | |
| "learning_rate": 0.00012964146944542565, | |
| "loss": 2.8708, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "learning_rate": 0.00012875927234192862, | |
| "loss": 2.8783, | |
| "step": 226000 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 0.00012787619215824797, | |
| "loss": 2.8832, | |
| "step": 227000 | |
| }, | |
| { | |
| "epoch": 12.27, | |
| "learning_rate": 0.00012699311197456728, | |
| "loss": 2.8846, | |
| "step": 228000 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 0.0001261100317908866, | |
| "loss": 2.89, | |
| "step": 229000 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "learning_rate": 0.0001252278346873896, | |
| "loss": 2.8897, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 12.43, | |
| "learning_rate": 0.00012434475450370891, | |
| "loss": 2.8912, | |
| "step": 231000 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 0.00012346167432002826, | |
| "loss": 2.8953, | |
| "step": 232000 | |
| }, | |
| { | |
| "epoch": 12.54, | |
| "learning_rate": 0.00012257947721653126, | |
| "loss": 2.8985, | |
| "step": 233000 | |
| }, | |
| { | |
| "epoch": 12.59, | |
| "learning_rate": 0.00012169639703285057, | |
| "loss": 2.8963, | |
| "step": 234000 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "learning_rate": 0.00012081331684916989, | |
| "loss": 2.8986, | |
| "step": 235000 | |
| }, | |
| { | |
| "epoch": 12.7, | |
| "learning_rate": 0.0001199311197456729, | |
| "loss": 2.9023, | |
| "step": 236000 | |
| }, | |
| { | |
| "epoch": 12.75, | |
| "learning_rate": 0.00011904803956199222, | |
| "loss": 2.9006, | |
| "step": 237000 | |
| }, | |
| { | |
| "epoch": 12.81, | |
| "learning_rate": 0.00011816495937831154, | |
| "loss": 2.9026, | |
| "step": 238000 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "learning_rate": 0.00011728276227481455, | |
| "loss": 2.9053, | |
| "step": 239000 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "learning_rate": 0.00011639968209113387, | |
| "loss": 2.9046, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 12.97, | |
| "learning_rate": 0.0001155166019074532, | |
| "loss": 2.91, | |
| "step": 241000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.40952716211277596, | |
| "eval_loss": 3.344587564468384, | |
| "eval_runtime": 153.7111, | |
| "eval_samples_per_second": 376.811, | |
| "eval_steps_per_second": 5.888, | |
| "step": 241618 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 0.00011463352172377251, | |
| "loss": 2.889, | |
| "step": 242000 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "learning_rate": 0.0001137522077004592, | |
| "loss": 2.8496, | |
| "step": 243000 | |
| }, | |
| { | |
| "epoch": 13.13, | |
| "learning_rate": 0.00011286912751677851, | |
| "loss": 2.8514, | |
| "step": 244000 | |
| }, | |
| { | |
| "epoch": 13.18, | |
| "learning_rate": 0.00011198604733309783, | |
| "loss": 2.8588, | |
| "step": 245000 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 0.00011110385022960083, | |
| "loss": 2.8668, | |
| "step": 246000 | |
| }, | |
| { | |
| "epoch": 13.29, | |
| "learning_rate": 0.00011022077004592015, | |
| "loss": 2.8654, | |
| "step": 247000 | |
| }, | |
| { | |
| "epoch": 13.34, | |
| "learning_rate": 0.00010933768986223949, | |
| "loss": 2.8695, | |
| "step": 248000 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "learning_rate": 0.00010845549275874248, | |
| "loss": 2.8721, | |
| "step": 249000 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 0.00010757241257506181, | |
| "loss": 2.8733, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 0.00010668933239138112, | |
| "loss": 2.875, | |
| "step": 251000 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 0.00010580713528788414, | |
| "loss": 2.8777, | |
| "step": 252000 | |
| }, | |
| { | |
| "epoch": 13.61, | |
| "learning_rate": 0.00010492405510420345, | |
| "loss": 2.8804, | |
| "step": 253000 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "learning_rate": 0.00010404185800070647, | |
| "loss": 2.8783, | |
| "step": 254000 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "learning_rate": 0.00010315966089720946, | |
| "loss": 2.8809, | |
| "step": 255000 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 0.00010227658071352877, | |
| "loss": 2.8813, | |
| "step": 256000 | |
| }, | |
| { | |
| "epoch": 13.83, | |
| "learning_rate": 0.0001013935005298481, | |
| "loss": 2.8846, | |
| "step": 257000 | |
| }, | |
| { | |
| "epoch": 13.88, | |
| "learning_rate": 0.00010051042034616742, | |
| "loss": 2.8827, | |
| "step": 258000 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "learning_rate": 9.962822324267043e-05, | |
| "loss": 2.8856, | |
| "step": 259000 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 9.874514305898975e-05, | |
| "loss": 2.888, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.40998700743814087, | |
| "eval_loss": 3.3463170528411865, | |
| "eval_runtime": 154.1738, | |
| "eval_samples_per_second": 375.68, | |
| "eval_steps_per_second": 5.87, | |
| "step": 260204 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 9.786206287530908e-05, | |
| "loss": 2.8448, | |
| "step": 261000 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 9.697986577181206e-05, | |
| "loss": 2.8359, | |
| "step": 262000 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 9.609678558813138e-05, | |
| "loss": 2.837, | |
| "step": 263000 | |
| }, | |
| { | |
| "epoch": 14.2, | |
| "learning_rate": 9.52145884846344e-05, | |
| "loss": 2.8421, | |
| "step": 264000 | |
| }, | |
| { | |
| "epoch": 14.26, | |
| "learning_rate": 9.433150830095371e-05, | |
| "loss": 2.8459, | |
| "step": 265000 | |
| }, | |
| { | |
| "epoch": 14.31, | |
| "learning_rate": 9.344842811727304e-05, | |
| "loss": 2.8498, | |
| "step": 266000 | |
| }, | |
| { | |
| "epoch": 14.37, | |
| "learning_rate": 9.256623101377604e-05, | |
| "loss": 2.8529, | |
| "step": 267000 | |
| }, | |
| { | |
| "epoch": 14.42, | |
| "learning_rate": 9.168315083009537e-05, | |
| "loss": 2.8552, | |
| "step": 268000 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 9.080007064641469e-05, | |
| "loss": 2.8549, | |
| "step": 269000 | |
| }, | |
| { | |
| "epoch": 14.53, | |
| "learning_rate": 8.9916990462734e-05, | |
| "loss": 2.8557, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 14.58, | |
| "learning_rate": 8.903479335923702e-05, | |
| "loss": 2.8618, | |
| "step": 271000 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 8.815171317555632e-05, | |
| "loss": 2.8594, | |
| "step": 272000 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 8.726951607205933e-05, | |
| "loss": 2.8608, | |
| "step": 273000 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "learning_rate": 8.638643588837865e-05, | |
| "loss": 2.863, | |
| "step": 274000 | |
| }, | |
| { | |
| "epoch": 14.8, | |
| "learning_rate": 8.550423878488167e-05, | |
| "loss": 2.8642, | |
| "step": 275000 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "learning_rate": 8.462204168138465e-05, | |
| "loss": 2.8625, | |
| "step": 276000 | |
| }, | |
| { | |
| "epoch": 14.9, | |
| "learning_rate": 8.373896149770398e-05, | |
| "loss": 2.8607, | |
| "step": 277000 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "learning_rate": 8.28558813140233e-05, | |
| "loss": 2.862, | |
| "step": 278000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.4102595556463228, | |
| "eval_loss": 3.3529999256134033, | |
| "eval_runtime": 153.645, | |
| "eval_samples_per_second": 376.973, | |
| "eval_steps_per_second": 5.89, | |
| "step": 278790 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 8.197280113034264e-05, | |
| "loss": 2.8539, | |
| "step": 279000 | |
| }, | |
| { | |
| "epoch": 15.07, | |
| "learning_rate": 8.109060402684563e-05, | |
| "loss": 2.818, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 15.12, | |
| "learning_rate": 8.020840692334864e-05, | |
| "loss": 2.8218, | |
| "step": 281000 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 7.932532673966796e-05, | |
| "loss": 2.8211, | |
| "step": 282000 | |
| }, | |
| { | |
| "epoch": 15.23, | |
| "learning_rate": 7.844224655598727e-05, | |
| "loss": 2.8246, | |
| "step": 283000 | |
| }, | |
| { | |
| "epoch": 15.28, | |
| "learning_rate": 7.756004945249028e-05, | |
| "loss": 2.828, | |
| "step": 284000 | |
| }, | |
| { | |
| "epoch": 15.33, | |
| "learning_rate": 7.667696926880959e-05, | |
| "loss": 2.8318, | |
| "step": 285000 | |
| }, | |
| { | |
| "epoch": 15.39, | |
| "learning_rate": 7.57947721653126e-05, | |
| "loss": 2.832, | |
| "step": 286000 | |
| }, | |
| { | |
| "epoch": 15.44, | |
| "learning_rate": 7.491169198163192e-05, | |
| "loss": 2.8373, | |
| "step": 287000 | |
| }, | |
| { | |
| "epoch": 15.5, | |
| "learning_rate": 7.402861179795125e-05, | |
| "loss": 2.8389, | |
| "step": 288000 | |
| }, | |
| { | |
| "epoch": 15.55, | |
| "learning_rate": 7.314553161427057e-05, | |
| "loss": 2.8422, | |
| "step": 289000 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "learning_rate": 7.22624514305899e-05, | |
| "loss": 2.8461, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "learning_rate": 7.13802543270929e-05, | |
| "loss": 2.8391, | |
| "step": 291000 | |
| }, | |
| { | |
| "epoch": 15.71, | |
| "learning_rate": 7.049717414341221e-05, | |
| "loss": 2.8404, | |
| "step": 292000 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "learning_rate": 6.961409395973153e-05, | |
| "loss": 2.8426, | |
| "step": 293000 | |
| }, | |
| { | |
| "epoch": 15.82, | |
| "learning_rate": 6.873189685623453e-05, | |
| "loss": 2.844, | |
| "step": 294000 | |
| }, | |
| { | |
| "epoch": 15.87, | |
| "learning_rate": 6.784881667255386e-05, | |
| "loss": 2.8412, | |
| "step": 295000 | |
| }, | |
| { | |
| "epoch": 15.93, | |
| "learning_rate": 6.696573648887319e-05, | |
| "loss": 2.8461, | |
| "step": 296000 | |
| }, | |
| { | |
| "epoch": 15.98, | |
| "learning_rate": 6.608353938537619e-05, | |
| "loss": 2.8465, | |
| "step": 297000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.4098270524754391, | |
| "eval_loss": 3.3666415214538574, | |
| "eval_runtime": 153.3685, | |
| "eval_samples_per_second": 377.652, | |
| "eval_steps_per_second": 5.901, | |
| "step": 297376 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "learning_rate": 6.520045920169551e-05, | |
| "loss": 2.8188, | |
| "step": 298000 | |
| }, | |
| { | |
| "epoch": 16.09, | |
| "learning_rate": 6.431737901801484e-05, | |
| "loss": 2.8009, | |
| "step": 299000 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 6.343429883433415e-05, | |
| "loss": 2.8057, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "learning_rate": 6.255121865065347e-05, | |
| "loss": 2.8082, | |
| "step": 301000 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "learning_rate": 6.166902154715647e-05, | |
| "loss": 2.8104, | |
| "step": 302000 | |
| }, | |
| { | |
| "epoch": 16.3, | |
| "learning_rate": 6.07859413634758e-05, | |
| "loss": 2.813, | |
| "step": 303000 | |
| }, | |
| { | |
| "epoch": 16.36, | |
| "learning_rate": 5.99037442599788e-05, | |
| "loss": 2.8164, | |
| "step": 304000 | |
| }, | |
| { | |
| "epoch": 16.41, | |
| "learning_rate": 5.902066407629812e-05, | |
| "loss": 2.8178, | |
| "step": 305000 | |
| }, | |
| { | |
| "epoch": 16.46, | |
| "learning_rate": 5.8138466972801124e-05, | |
| "loss": 2.8203, | |
| "step": 306000 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "learning_rate": 5.7256269869304125e-05, | |
| "loss": 2.8181, | |
| "step": 307000 | |
| }, | |
| { | |
| "epoch": 16.57, | |
| "learning_rate": 5.6373189685623454e-05, | |
| "loss": 2.8204, | |
| "step": 308000 | |
| }, | |
| { | |
| "epoch": 16.63, | |
| "learning_rate": 5.549010950194278e-05, | |
| "loss": 2.8222, | |
| "step": 309000 | |
| }, | |
| { | |
| "epoch": 16.68, | |
| "learning_rate": 5.460791239844578e-05, | |
| "loss": 2.8243, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 16.73, | |
| "learning_rate": 5.3724832214765094e-05, | |
| "loss": 2.8238, | |
| "step": 311000 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "learning_rate": 5.284175203108442e-05, | |
| "loss": 2.8264, | |
| "step": 312000 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "learning_rate": 5.195867184740374e-05, | |
| "loss": 2.8258, | |
| "step": 313000 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "learning_rate": 5.107647474390674e-05, | |
| "loss": 2.8246, | |
| "step": 314000 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 5.019339456022606e-05, | |
| "loss": 2.8291, | |
| "step": 315000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.40987307059784595, | |
| "eval_loss": 3.3779566287994385, | |
| "eval_runtime": 153.7263, | |
| "eval_samples_per_second": 376.773, | |
| "eval_steps_per_second": 5.887, | |
| "step": 315962 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 4.931031437654539e-05, | |
| "loss": 2.8256, | |
| "step": 316000 | |
| }, | |
| { | |
| "epoch": 17.06, | |
| "learning_rate": 4.8428117273048394e-05, | |
| "loss": 2.7855, | |
| "step": 317000 | |
| }, | |
| { | |
| "epoch": 17.11, | |
| "learning_rate": 4.754503708936771e-05, | |
| "loss": 2.7903, | |
| "step": 318000 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "learning_rate": 4.666283998587071e-05, | |
| "loss": 2.7904, | |
| "step": 319000 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "learning_rate": 4.577975980219003e-05, | |
| "loss": 2.7981, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 17.27, | |
| "learning_rate": 4.4896679618509356e-05, | |
| "loss": 2.7943, | |
| "step": 321000 | |
| }, | |
| { | |
| "epoch": 17.32, | |
| "learning_rate": 4.401359943482868e-05, | |
| "loss": 2.8028, | |
| "step": 322000 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 4.313140233133168e-05, | |
| "loss": 2.7975, | |
| "step": 323000 | |
| }, | |
| { | |
| "epoch": 17.43, | |
| "learning_rate": 4.224832214765101e-05, | |
| "loss": 2.8009, | |
| "step": 324000 | |
| }, | |
| { | |
| "epoch": 17.49, | |
| "learning_rate": 4.1365241963970325e-05, | |
| "loss": 2.8034, | |
| "step": 325000 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "learning_rate": 4.0483044860473326e-05, | |
| "loss": 2.802, | |
| "step": 326000 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 3.959996467679265e-05, | |
| "loss": 2.8079, | |
| "step": 327000 | |
| }, | |
| { | |
| "epoch": 17.65, | |
| "learning_rate": 3.871688449311197e-05, | |
| "loss": 2.8056, | |
| "step": 328000 | |
| }, | |
| { | |
| "epoch": 17.7, | |
| "learning_rate": 3.7833804309431294e-05, | |
| "loss": 2.8024, | |
| "step": 329000 | |
| }, | |
| { | |
| "epoch": 17.76, | |
| "learning_rate": 3.695072412575062e-05, | |
| "loss": 2.8028, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 17.81, | |
| "learning_rate": 3.6069410102437296e-05, | |
| "loss": 2.8101, | |
| "step": 331000 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 3.518632991875662e-05, | |
| "loss": 2.809, | |
| "step": 332000 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "learning_rate": 3.430324973507594e-05, | |
| "loss": 2.8098, | |
| "step": 333000 | |
| }, | |
| { | |
| "epoch": 17.97, | |
| "learning_rate": 3.342105263157894e-05, | |
| "loss": 2.8072, | |
| "step": 334000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.40994703549240064, | |
| "eval_loss": 3.38580584526062, | |
| "eval_runtime": 153.5136, | |
| "eval_samples_per_second": 377.296, | |
| "eval_steps_per_second": 5.895, | |
| "step": 334548 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "learning_rate": 3.2538855528081944e-05, | |
| "loss": 2.7925, | |
| "step": 335000 | |
| }, | |
| { | |
| "epoch": 18.08, | |
| "learning_rate": 3.165577534440127e-05, | |
| "loss": 2.7758, | |
| "step": 336000 | |
| }, | |
| { | |
| "epoch": 18.13, | |
| "learning_rate": 3.077269516072059e-05, | |
| "loss": 2.7784, | |
| "step": 337000 | |
| }, | |
| { | |
| "epoch": 18.19, | |
| "learning_rate": 2.9889614977039912e-05, | |
| "loss": 2.7829, | |
| "step": 338000 | |
| }, | |
| { | |
| "epoch": 18.24, | |
| "learning_rate": 2.9006534793359235e-05, | |
| "loss": 2.784, | |
| "step": 339000 | |
| }, | |
| { | |
| "epoch": 18.29, | |
| "learning_rate": 2.8124337689862236e-05, | |
| "loss": 2.782, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 18.35, | |
| "learning_rate": 2.724125750618156e-05, | |
| "loss": 2.7808, | |
| "step": 341000 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 2.635906040268456e-05, | |
| "loss": 2.7866, | |
| "step": 342000 | |
| }, | |
| { | |
| "epoch": 18.45, | |
| "learning_rate": 2.5475980219003885e-05, | |
| "loss": 2.7903, | |
| "step": 343000 | |
| }, | |
| { | |
| "epoch": 18.51, | |
| "learning_rate": 2.4592900035323205e-05, | |
| "loss": 2.7834, | |
| "step": 344000 | |
| }, | |
| { | |
| "epoch": 18.56, | |
| "learning_rate": 2.3709819851642528e-05, | |
| "loss": 2.7924, | |
| "step": 345000 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 2.2826739667961847e-05, | |
| "loss": 2.7874, | |
| "step": 346000 | |
| }, | |
| { | |
| "epoch": 18.67, | |
| "learning_rate": 2.194454256446485e-05, | |
| "loss": 2.7869, | |
| "step": 347000 | |
| }, | |
| { | |
| "epoch": 18.72, | |
| "learning_rate": 2.1061462380784174e-05, | |
| "loss": 2.7917, | |
| "step": 348000 | |
| }, | |
| { | |
| "epoch": 18.78, | |
| "learning_rate": 2.0179265277287175e-05, | |
| "loss": 2.7869, | |
| "step": 349000 | |
| }, | |
| { | |
| "epoch": 18.83, | |
| "learning_rate": 1.9296185093606498e-05, | |
| "loss": 2.7932, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 18.89, | |
| "learning_rate": 1.8413987990109502e-05, | |
| "loss": 2.7917, | |
| "step": 351000 | |
| }, | |
| { | |
| "epoch": 18.94, | |
| "learning_rate": 1.7531790886612503e-05, | |
| "loss": 2.7874, | |
| "step": 352000 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 1.6648710702931826e-05, | |
| "loss": 2.786, | |
| "step": 353000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.4103528011264699, | |
| "eval_loss": 3.384690999984741, | |
| "eval_runtime": 153.4095, | |
| "eval_samples_per_second": 377.551, | |
| "eval_steps_per_second": 5.899, | |
| "step": 353134 | |
| }, | |
| { | |
| "epoch": 19.05, | |
| "learning_rate": 1.5765630519251145e-05, | |
| "loss": 2.7729, | |
| "step": 354000 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 1.4882550335570468e-05, | |
| "loss": 2.769, | |
| "step": 355000 | |
| }, | |
| { | |
| "epoch": 19.15, | |
| "learning_rate": 1.4000353232073472e-05, | |
| "loss": 2.7722, | |
| "step": 356000 | |
| }, | |
| { | |
| "epoch": 19.21, | |
| "learning_rate": 1.3117273048392793e-05, | |
| "loss": 2.7702, | |
| "step": 357000 | |
| }, | |
| { | |
| "epoch": 19.26, | |
| "learning_rate": 1.2234192864712116e-05, | |
| "loss": 2.7715, | |
| "step": 358000 | |
| }, | |
| { | |
| "epoch": 19.32, | |
| "learning_rate": 1.1351995761215117e-05, | |
| "loss": 2.7706, | |
| "step": 359000 | |
| }, | |
| { | |
| "epoch": 19.37, | |
| "learning_rate": 1.046891557753444e-05, | |
| "loss": 2.7727, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 19.42, | |
| "learning_rate": 9.58671847403744e-06, | |
| "loss": 2.7738, | |
| "step": 361000 | |
| }, | |
| { | |
| "epoch": 19.48, | |
| "learning_rate": 8.703638290356763e-06, | |
| "loss": 2.7756, | |
| "step": 362000 | |
| }, | |
| { | |
| "epoch": 19.53, | |
| "learning_rate": 7.821441186859766e-06, | |
| "loss": 2.7715, | |
| "step": 363000 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "learning_rate": 6.938361003179088e-06, | |
| "loss": 2.773, | |
| "step": 364000 | |
| }, | |
| { | |
| "epoch": 19.64, | |
| "learning_rate": 6.05528081949841e-06, | |
| "loss": 2.7653, | |
| "step": 365000 | |
| }, | |
| { | |
| "epoch": 19.69, | |
| "learning_rate": 5.173083716001413e-06, | |
| "loss": 2.7717, | |
| "step": 366000 | |
| }, | |
| { | |
| "epoch": 19.75, | |
| "learning_rate": 4.290003532320734e-06, | |
| "loss": 2.7718, | |
| "step": 367000 | |
| }, | |
| { | |
| "epoch": 19.8, | |
| "learning_rate": 3.406923348640056e-06, | |
| "loss": 2.7748, | |
| "step": 368000 | |
| }, | |
| { | |
| "epoch": 19.85, | |
| "learning_rate": 2.523843164959378e-06, | |
| "loss": 2.7719, | |
| "step": 369000 | |
| }, | |
| { | |
| "epoch": 19.91, | |
| "learning_rate": 1.6425291416460613e-06, | |
| "loss": 2.7704, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 19.96, | |
| "learning_rate": 7.594489579653833e-07, | |
| "loss": 2.773, | |
| "step": 371000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.4102484709891008, | |
| "eval_loss": 3.3909249305725098, | |
| "eval_runtime": 153.8319, | |
| "eval_samples_per_second": 376.515, | |
| "eval_steps_per_second": 5.883, | |
| "step": 371720 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 371720, | |
| "total_flos": 1.56614628708864e+18, | |
| "train_loss": 3.062808654778588, | |
| "train_runtime": 81067.646, | |
| "train_samples_per_second": 146.727, | |
| "train_steps_per_second": 4.585 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 371720, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 5000, | |
| "total_flos": 1.56614628708864e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |