{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.0033444256785003595, "eval_steps": 9, "global_step": 100, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 3.3444256785003593e-05, "grad_norm": 4.028215408325195, "learning_rate": 1e-05, "loss": 3.5489, "step": 1 }, { "epoch": 3.3444256785003593e-05, "eval_loss": 3.722480297088623, "eval_runtime": 2739.1214, "eval_samples_per_second": 9.193, "eval_steps_per_second": 1.149, "step": 1 }, { "epoch": 6.688851357000719e-05, "grad_norm": 4.290927886962891, "learning_rate": 2e-05, "loss": 3.6949, "step": 2 }, { "epoch": 0.00010033277035501078, "grad_norm": 4.205336570739746, "learning_rate": 3e-05, "loss": 3.5552, "step": 3 }, { "epoch": 0.00013377702714001437, "grad_norm": 4.705852508544922, "learning_rate": 4e-05, "loss": 3.9999, "step": 4 }, { "epoch": 0.00016722128392501797, "grad_norm": 4.193447589874268, "learning_rate": 5e-05, "loss": 3.3292, "step": 5 }, { "epoch": 0.00020066554071002156, "grad_norm": 4.9734721183776855, "learning_rate": 6e-05, "loss": 3.9245, "step": 6 }, { "epoch": 0.00023410979749502515, "grad_norm": 5.040078163146973, "learning_rate": 7e-05, "loss": 3.5262, "step": 7 }, { "epoch": 0.00026755405428002875, "grad_norm": 3.55013370513916, "learning_rate": 8e-05, "loss": 3.2288, "step": 8 }, { "epoch": 0.00030099831106503237, "grad_norm": 3.3910467624664307, "learning_rate": 9e-05, "loss": 3.0514, "step": 9 }, { "epoch": 0.00030099831106503237, "eval_loss": 2.8550124168395996, "eval_runtime": 2745.5024, "eval_samples_per_second": 9.171, "eval_steps_per_second": 1.147, "step": 9 }, { "epoch": 0.00033444256785003593, "grad_norm": 3.46765398979187, "learning_rate": 0.0001, "loss": 2.8576, "step": 10 }, { "epoch": 0.00036788682463503956, "grad_norm": 3.2942867279052734, "learning_rate": 9.99695413509548e-05, "loss": 2.7446, "step": 11 }, { "epoch": 0.0004013310814200431, "grad_norm": 3.2802114486694336, "learning_rate": 9.987820251299122e-05, "loss": 2.4887, "step": 12 }, { "epoch": 0.00043477533820504674, "grad_norm": 3.1480672359466553, "learning_rate": 9.972609476841367e-05, "loss": 2.4346, "step": 13 }, { "epoch": 0.0004682195949900503, "grad_norm": 3.0352845191955566, "learning_rate": 9.951340343707852e-05, "loss": 2.1358, "step": 14 }, { "epoch": 0.0005016638517750539, "grad_norm": 2.5337398052215576, "learning_rate": 9.924038765061042e-05, "loss": 1.797, "step": 15 }, { "epoch": 0.0005351081085600575, "grad_norm": 2.5257315635681152, "learning_rate": 9.890738003669029e-05, "loss": 1.66, "step": 16 }, { "epoch": 0.0005685523653450611, "grad_norm": 2.703648567199707, "learning_rate": 9.851478631379982e-05, "loss": 1.5367, "step": 17 }, { "epoch": 0.0006019966221300647, "grad_norm": 3.4314801692962646, "learning_rate": 9.806308479691595e-05, "loss": 1.6421, "step": 18 }, { "epoch": 0.0006019966221300647, "eval_loss": 1.629733681678772, "eval_runtime": 2742.8537, "eval_samples_per_second": 9.18, "eval_steps_per_second": 1.148, "step": 18 }, { "epoch": 0.0006354408789150684, "grad_norm": 3.213797092437744, "learning_rate": 9.755282581475769e-05, "loss": 1.591, "step": 19 }, { "epoch": 0.0006688851357000719, "grad_norm": 2.8766260147094727, "learning_rate": 9.698463103929542e-05, "loss": 1.5007, "step": 20 }, { "epoch": 0.0007023293924850755, "grad_norm": 3.298490285873413, "learning_rate": 9.635919272833938e-05, "loss": 1.6326, "step": 21 }, { "epoch": 0.0007357736492700791, "grad_norm": 3.046107530593872, "learning_rate": 9.567727288213005e-05, "loss": 1.6194, "step": 22 }, { "epoch": 0.0007692179060550827, "grad_norm": 3.1948232650756836, "learning_rate": 9.493970231495835e-05, "loss": 1.466, "step": 23 }, { "epoch": 0.0008026621628400862, "grad_norm": 3.6307601928710938, "learning_rate": 9.414737964294636e-05, "loss": 1.5915, "step": 24 }, { "epoch": 0.0008361064196250899, "grad_norm": 3.745431423187256, "learning_rate": 9.330127018922194e-05, "loss": 1.6644, "step": 25 }, { "epoch": 0.0008695506764100935, "grad_norm": 3.1015143394470215, "learning_rate": 9.24024048078213e-05, "loss": 1.4194, "step": 26 }, { "epoch": 0.0009029949331950971, "grad_norm": 2.8244731426239014, "learning_rate": 9.145187862775209e-05, "loss": 1.3943, "step": 27 }, { "epoch": 0.0009029949331950971, "eval_loss": 1.4489030838012695, "eval_runtime": 2740.1542, "eval_samples_per_second": 9.189, "eval_steps_per_second": 1.149, "step": 27 }, { "epoch": 0.0009364391899801006, "grad_norm": 2.156521797180176, "learning_rate": 9.045084971874738e-05, "loss": 1.21, "step": 28 }, { "epoch": 0.0009698834467651042, "grad_norm": 2.6699202060699463, "learning_rate": 8.940053768033609e-05, "loss": 1.4729, "step": 29 }, { "epoch": 0.0010033277035501078, "grad_norm": 2.460568904876709, "learning_rate": 8.83022221559489e-05, "loss": 1.5675, "step": 30 }, { "epoch": 0.0010367719603351115, "grad_norm": 2.557603120803833, "learning_rate": 8.715724127386972e-05, "loss": 1.5623, "step": 31 }, { "epoch": 0.001070216217120115, "grad_norm": 2.9280195236206055, "learning_rate": 8.596699001693255e-05, "loss": 1.4773, "step": 32 }, { "epoch": 0.0011036604739051187, "grad_norm": 2.2872281074523926, "learning_rate": 8.473291852294987e-05, "loss": 1.2972, "step": 33 }, { "epoch": 0.0011371047306901222, "grad_norm": 1.872541904449463, "learning_rate": 8.345653031794292e-05, "loss": 1.3728, "step": 34 }, { "epoch": 0.0011705489874751257, "grad_norm": 1.8109902143478394, "learning_rate": 8.213938048432697e-05, "loss": 1.2932, "step": 35 }, { "epoch": 0.0012039932442601295, "grad_norm": 2.0873613357543945, "learning_rate": 8.07830737662829e-05, "loss": 1.3242, "step": 36 }, { "epoch": 0.0012039932442601295, "eval_loss": 1.4143481254577637, "eval_runtime": 2741.3995, "eval_samples_per_second": 9.185, "eval_steps_per_second": 1.148, "step": 36 }, { "epoch": 0.001237437501045133, "grad_norm": 2.4181251525878906, "learning_rate": 7.938926261462366e-05, "loss": 1.4656, "step": 37 }, { "epoch": 0.0012708817578301367, "grad_norm": 2.4675228595733643, "learning_rate": 7.795964517353735e-05, "loss": 1.4617, "step": 38 }, { "epoch": 0.0013043260146151402, "grad_norm": 2.8824567794799805, "learning_rate": 7.649596321166024e-05, "loss": 1.5205, "step": 39 }, { "epoch": 0.0013377702714001437, "grad_norm": 2.1439270973205566, "learning_rate": 7.500000000000001e-05, "loss": 1.3624, "step": 40 }, { "epoch": 0.0013712145281851475, "grad_norm": 2.3218116760253906, "learning_rate": 7.347357813929454e-05, "loss": 1.4908, "step": 41 }, { "epoch": 0.001404658784970151, "grad_norm": 2.236236572265625, "learning_rate": 7.191855733945387e-05, "loss": 1.2628, "step": 42 }, { "epoch": 0.0014381030417551545, "grad_norm": 1.9307564496994019, "learning_rate": 7.033683215379002e-05, "loss": 1.3752, "step": 43 }, { "epoch": 0.0014715472985401582, "grad_norm": 1.8109971284866333, "learning_rate": 6.873032967079561e-05, "loss": 1.3783, "step": 44 }, { "epoch": 0.0015049915553251617, "grad_norm": 1.8418580293655396, "learning_rate": 6.710100716628344e-05, "loss": 1.3572, "step": 45 }, { "epoch": 0.0015049915553251617, "eval_loss": 1.3692147731781006, "eval_runtime": 2741.1491, "eval_samples_per_second": 9.186, "eval_steps_per_second": 1.148, "step": 45 }, { "epoch": 0.0015384358121101655, "grad_norm": 1.9104599952697754, "learning_rate": 6.545084971874738e-05, "loss": 1.2735, "step": 46 }, { "epoch": 0.001571880068895169, "grad_norm": 2.1751577854156494, "learning_rate": 6.378186779084995e-05, "loss": 1.3924, "step": 47 }, { "epoch": 0.0016053243256801725, "grad_norm": 1.5567160844802856, "learning_rate": 6.209609477998338e-05, "loss": 1.3338, "step": 48 }, { "epoch": 0.0016387685824651762, "grad_norm": 1.7269301414489746, "learning_rate": 6.0395584540887963e-05, "loss": 1.3239, "step": 49 }, { "epoch": 0.0016722128392501797, "grad_norm": 1.800885558128357, "learning_rate": 5.868240888334653e-05, "loss": 1.3558, "step": 50 }, { "epoch": 0.0017056570960351835, "grad_norm": 2.009683132171631, "learning_rate": 5.695865504800327e-05, "loss": 1.3781, "step": 51 }, { "epoch": 0.001739101352820187, "grad_norm": 1.7342225313186646, "learning_rate": 5.522642316338268e-05, "loss": 1.5557, "step": 52 }, { "epoch": 0.0017725456096051905, "grad_norm": 1.5290874242782593, "learning_rate": 5.348782368720626e-05, "loss": 1.3127, "step": 53 }, { "epoch": 0.0018059898663901942, "grad_norm": 1.653159499168396, "learning_rate": 5.174497483512506e-05, "loss": 1.1634, "step": 54 }, { "epoch": 0.0018059898663901942, "eval_loss": 1.3522837162017822, "eval_runtime": 2739.9035, "eval_samples_per_second": 9.19, "eval_steps_per_second": 1.149, "step": 54 }, { "epoch": 0.0018394341231751977, "grad_norm": 1.6415021419525146, "learning_rate": 5e-05, "loss": 1.4116, "step": 55 }, { "epoch": 0.0018728783799602012, "grad_norm": 1.749309778213501, "learning_rate": 4.825502516487497e-05, "loss": 1.4152, "step": 56 }, { "epoch": 0.001906322636745205, "grad_norm": 1.564772367477417, "learning_rate": 4.6512176312793736e-05, "loss": 1.417, "step": 57 }, { "epoch": 0.0019397668935302085, "grad_norm": 1.849936842918396, "learning_rate": 4.477357683661734e-05, "loss": 1.3394, "step": 58 }, { "epoch": 0.001973211150315212, "grad_norm": 1.7286560535430908, "learning_rate": 4.3041344951996746e-05, "loss": 1.5013, "step": 59 }, { "epoch": 0.0020066554071002155, "grad_norm": 1.9632574319839478, "learning_rate": 4.131759111665349e-05, "loss": 1.3443, "step": 60 }, { "epoch": 0.0020400996638852194, "grad_norm": 1.7394282817840576, "learning_rate": 3.960441545911204e-05, "loss": 1.3558, "step": 61 }, { "epoch": 0.002073543920670223, "grad_norm": 2.260873794555664, "learning_rate": 3.790390522001662e-05, "loss": 1.3278, "step": 62 }, { "epoch": 0.0021069881774552265, "grad_norm": 1.8529855012893677, "learning_rate": 3.6218132209150045e-05, "loss": 1.2319, "step": 63 }, { "epoch": 0.0021069881774552265, "eval_loss": 1.3400989770889282, "eval_runtime": 2741.0245, "eval_samples_per_second": 9.186, "eval_steps_per_second": 1.148, "step": 63 }, { "epoch": 0.00214043243424023, "grad_norm": 1.778730034828186, "learning_rate": 3.4549150281252636e-05, "loss": 1.3719, "step": 64 }, { "epoch": 0.0021738766910252335, "grad_norm": 1.7320828437805176, "learning_rate": 3.289899283371657e-05, "loss": 1.3139, "step": 65 }, { "epoch": 0.0022073209478102374, "grad_norm": 1.8604722023010254, "learning_rate": 3.12696703292044e-05, "loss": 1.1788, "step": 66 }, { "epoch": 0.002240765204595241, "grad_norm": 2.2654762268066406, "learning_rate": 2.9663167846209998e-05, "loss": 1.4146, "step": 67 }, { "epoch": 0.0022742094613802445, "grad_norm": 1.7686560153961182, "learning_rate": 2.8081442660546125e-05, "loss": 1.2696, "step": 68 }, { "epoch": 0.002307653718165248, "grad_norm": 1.5691628456115723, "learning_rate": 2.6526421860705473e-05, "loss": 1.2514, "step": 69 }, { "epoch": 0.0023410979749502515, "grad_norm": 1.543271541595459, "learning_rate": 2.500000000000001e-05, "loss": 1.4583, "step": 70 }, { "epoch": 0.0023745422317352554, "grad_norm": 1.897420048713684, "learning_rate": 2.350403678833976e-05, "loss": 1.3972, "step": 71 }, { "epoch": 0.002407986488520259, "grad_norm": 1.782452940940857, "learning_rate": 2.2040354826462668e-05, "loss": 1.4984, "step": 72 }, { "epoch": 0.002407986488520259, "eval_loss": 1.331598162651062, "eval_runtime": 2742.8962, "eval_samples_per_second": 9.18, "eval_steps_per_second": 1.148, "step": 72 }, { "epoch": 0.0024414307453052625, "grad_norm": 1.812255620956421, "learning_rate": 2.061073738537635e-05, "loss": 1.3558, "step": 73 }, { "epoch": 0.002474875002090266, "grad_norm": 1.7149916887283325, "learning_rate": 1.9216926233717085e-05, "loss": 1.2338, "step": 74 }, { "epoch": 0.0025083192588752695, "grad_norm": 1.5673575401306152, "learning_rate": 1.7860619515673033e-05, "loss": 1.5045, "step": 75 }, { "epoch": 0.0025417635156602734, "grad_norm": 1.7105762958526611, "learning_rate": 1.6543469682057106e-05, "loss": 1.3942, "step": 76 }, { "epoch": 0.002575207772445277, "grad_norm": 1.7119174003601074, "learning_rate": 1.526708147705013e-05, "loss": 1.2635, "step": 77 }, { "epoch": 0.0026086520292302805, "grad_norm": 1.553362488746643, "learning_rate": 1.4033009983067452e-05, "loss": 1.2464, "step": 78 }, { "epoch": 0.002642096286015284, "grad_norm": 1.757306456565857, "learning_rate": 1.2842758726130283e-05, "loss": 1.3831, "step": 79 }, { "epoch": 0.0026755405428002875, "grad_norm": 1.9788403511047363, "learning_rate": 1.1697777844051105e-05, "loss": 1.2947, "step": 80 }, { "epoch": 0.0027089847995852914, "grad_norm": 1.7730344533920288, "learning_rate": 1.0599462319663905e-05, "loss": 1.2519, "step": 81 }, { "epoch": 0.0027089847995852914, "eval_loss": 1.3251186609268188, "eval_runtime": 2741.7754, "eval_samples_per_second": 9.184, "eval_steps_per_second": 1.148, "step": 81 }, { "epoch": 0.002742429056370295, "grad_norm": 1.6482566595077515, "learning_rate": 9.549150281252633e-06, "loss": 1.2612, "step": 82 }, { "epoch": 0.0027758733131552984, "grad_norm": 1.8458375930786133, "learning_rate": 8.548121372247918e-06, "loss": 1.4599, "step": 83 }, { "epoch": 0.002809317569940302, "grad_norm": 1.6900582313537598, "learning_rate": 7.597595192178702e-06, "loss": 1.2102, "step": 84 }, { "epoch": 0.0028427618267253055, "grad_norm": 2.316399574279785, "learning_rate": 6.698729810778065e-06, "loss": 1.3149, "step": 85 }, { "epoch": 0.002876206083510309, "grad_norm": 1.6527814865112305, "learning_rate": 5.852620357053651e-06, "loss": 1.1052, "step": 86 }, { "epoch": 0.002909650340295313, "grad_norm": 1.7383328676223755, "learning_rate": 5.060297685041659e-06, "loss": 1.2462, "step": 87 }, { "epoch": 0.0029430945970803164, "grad_norm": 1.8759413957595825, "learning_rate": 4.322727117869951e-06, "loss": 1.5503, "step": 88 }, { "epoch": 0.00297653885386532, "grad_norm": 1.686247706413269, "learning_rate": 3.6408072716606346e-06, "loss": 1.4226, "step": 89 }, { "epoch": 0.0030099831106503235, "grad_norm": 1.983102798461914, "learning_rate": 3.0153689607045845e-06, "loss": 1.4, "step": 90 }, { "epoch": 0.0030099831106503235, "eval_loss": 1.3215969800949097, "eval_runtime": 2741.5437, "eval_samples_per_second": 9.185, "eval_steps_per_second": 1.148, "step": 90 }, { "epoch": 0.003043427367435327, "grad_norm": 1.666930079460144, "learning_rate": 2.4471741852423237e-06, "loss": 1.4906, "step": 91 }, { "epoch": 0.003076871624220331, "grad_norm": 1.5354820489883423, "learning_rate": 1.9369152030840556e-06, "loss": 1.203, "step": 92 }, { "epoch": 0.0031103158810053344, "grad_norm": 1.64034903049469, "learning_rate": 1.4852136862001764e-06, "loss": 1.3875, "step": 93 }, { "epoch": 0.003143760137790338, "grad_norm": 2.111876964569092, "learning_rate": 1.0926199633097157e-06, "loss": 1.4371, "step": 94 }, { "epoch": 0.0031772043945753415, "grad_norm": 1.6222939491271973, "learning_rate": 7.596123493895991e-07, "loss": 1.4024, "step": 95 }, { "epoch": 0.003210648651360345, "grad_norm": 1.8614100217819214, "learning_rate": 4.865965629214819e-07, "loss": 1.3613, "step": 96 }, { "epoch": 0.003244092908145349, "grad_norm": 1.6315127611160278, "learning_rate": 2.7390523158633554e-07, "loss": 1.3897, "step": 97 }, { "epoch": 0.0032775371649303524, "grad_norm": 1.4587796926498413, "learning_rate": 1.2179748700879012e-07, "loss": 1.1787, "step": 98 }, { "epoch": 0.003310981421715356, "grad_norm": 1.7069193124771118, "learning_rate": 3.04586490452119e-08, "loss": 1.3006, "step": 99 }, { "epoch": 0.003310981421715356, "eval_loss": 1.3205249309539795, "eval_runtime": 2741.3417, "eval_samples_per_second": 9.185, "eval_steps_per_second": 1.148, "step": 99 }, { "epoch": 0.0033444256785003595, "grad_norm": 1.4645100831985474, "learning_rate": 0.0, "loss": 1.3523, "step": 100 } ], "logging_steps": 1, "max_steps": 100, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 25, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 7.41887283560448e+16, "train_batch_size": 8, "trial_name": null, "trial_params": null }