| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.8, |
| "eval_steps": 100, |
| "global_step": 1600, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.0125, |
| "grad_norm": 0.08136817067861557, |
| "learning_rate": 0.0004, |
| "loss": 1.158, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.025, |
| "grad_norm": 0.0653829574584961, |
| "learning_rate": 0.0004998852503731983, |
| "loss": 1.0957, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.0375, |
| "grad_norm": 0.11592712253332138, |
| "learning_rate": 0.0004993848168027977, |
| "loss": 0.9276, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.05, |
| "grad_norm": 0.08926476538181305, |
| "learning_rate": 0.0004984880506341147, |
| "loss": 1.0337, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.05, |
| "eval_loss": 0.9111642837524414, |
| "eval_runtime": 843.8058, |
| "eval_samples_per_second": 1.3, |
| "eval_steps_per_second": 0.021, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.0625, |
| "grad_norm": 0.08632034063339233, |
| "learning_rate": 0.0004971963770447935, |
| "loss": 1.0219, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.075, |
| "grad_norm": 0.12949417531490326, |
| "learning_rate": 0.0004955118488155782, |
| "loss": 0.784, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.0875, |
| "grad_norm": 0.09222520887851715, |
| "learning_rate": 0.0004934371430679492, |
| "loss": 1.0, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.1, |
| "grad_norm": 0.10075929015874863, |
| "learning_rate": 0.0004909755570095319, |
| "loss": 0.9617, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.1, |
| "eval_loss": 0.9380430579185486, |
| "eval_runtime": 856.156, |
| "eval_samples_per_second": 1.281, |
| "eval_steps_per_second": 0.021, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.1125, |
| "grad_norm": 0.09567111730575562, |
| "learning_rate": 0.0004881310026940389, |
| "loss": 0.7051, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.125, |
| "grad_norm": 0.09723825007677078, |
| "learning_rate": 0.0004849080008040734, |
| "loss": 0.9906, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.1375, |
| "grad_norm": 0.09430444985628128, |
| "learning_rate": 0.00048131167346667446, |
| "loss": 0.9113, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.15, |
| "grad_norm": 0.09439756721258163, |
| "learning_rate": 0.00047734773611302284, |
| "loss": 0.674, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.15, |
| "eval_loss": 0.9544369578361511, |
| "eval_runtime": 850.9461, |
| "eval_samples_per_second": 1.289, |
| "eval_steps_per_second": 0.021, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.1625, |
| "grad_norm": 0.10961435735225677, |
| "learning_rate": 0.0004730224883952422, |
| "loss": 0.9701, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.175, |
| "grad_norm": 0.09628895670175552, |
| "learning_rate": 0.0004683428041747334, |
| "loss": 0.8976, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.1875, |
| "grad_norm": 0.11085063964128494, |
| "learning_rate": 0.0004633161205979517, |
| "loss": 0.6683, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.2, |
| "grad_norm": 0.10140710324048996, |
| "learning_rate": 0.0004579504262769877, |
| "loss": 0.9373, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.2, |
| "eval_loss": 0.968166172504425, |
| "eval_runtime": 853.0258, |
| "eval_samples_per_second": 1.286, |
| "eval_steps_per_second": 0.021, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.2125, |
| "grad_norm": 0.13868793845176697, |
| "learning_rate": 0.0004522542485937369, |
| "loss": 0.8822, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.225, |
| "grad_norm": 0.11853040754795074, |
| "learning_rate": 0.00044623664014783386, |
| "loss": 0.6483, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.2375, |
| "grad_norm": 0.11668186634778976, |
| "learning_rate": 0.00043990716436988924, |
| "loss": 0.9374, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.25, |
| "grad_norm": 0.11911585181951523, |
| "learning_rate": 0.0004332758803228925, |
| "loss": 0.8434, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.25, |
| "eval_loss": 0.9753186702728271, |
| "eval_runtime": 851.2812, |
| "eval_samples_per_second": 1.289, |
| "eval_steps_per_second": 0.021, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.2625, |
| "grad_norm": 0.09807440638542175, |
| "learning_rate": 0.00042635332671593575, |
| "loss": 0.6661, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.275, |
| "grad_norm": 0.09675773978233337, |
| "learning_rate": 0.00041915050515566445, |
| "loss": 0.8999, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.2875, |
| "grad_norm": 0.0979737937450409, |
| "learning_rate": 0.00041167886266207167, |
| "loss": 0.8616, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.3, |
| "grad_norm": 0.08990875631570816, |
| "learning_rate": 0.0004039502734764241, |
| "loss": 0.8167, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.3, |
| "eval_loss": 0.9926204681396484, |
| "eval_runtime": 851.6484, |
| "eval_samples_per_second": 1.288, |
| "eval_steps_per_second": 0.021, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.3125, |
| "grad_norm": 0.09505137801170349, |
| "learning_rate": 0.0003959770201902294, |
| "loss": 0.8191, |
| "step": 625 |
| }, |
| { |
| "epoch": 0.325, |
| "grad_norm": 0.09105059504508972, |
| "learning_rate": 0.0003877717742252371, |
| "loss": 0.7203, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.3375, |
| "grad_norm": 0.09274734556674957, |
| "learning_rate": 0.00037934757569549495, |
| "loss": 0.8154, |
| "step": 675 |
| }, |
| { |
| "epoch": 0.35, |
| "grad_norm": 0.10026416182518005, |
| "learning_rate": 0.00037071781268346345, |
| "loss": 0.8336, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.35, |
| "eval_loss": 0.9959912896156311, |
| "eval_runtime": 849.7662, |
| "eval_samples_per_second": 1.291, |
| "eval_steps_per_second": 0.021, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.3625, |
| "grad_norm": 0.09517039358615875, |
| "learning_rate": 0.00036189619996312495, |
| "loss": 0.6845, |
| "step": 725 |
| }, |
| { |
| "epoch": 0.375, |
| "grad_norm": 0.11264201998710632, |
| "learning_rate": 0.00035289675720390174, |
| "loss": 0.8445, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.3875, |
| "grad_norm": 0.0979958102107048, |
| "learning_rate": 0.00034373378669002105, |
| "loss": 0.8269, |
| "step": 775 |
| }, |
| { |
| "epoch": 0.4, |
| "grad_norm": 0.10223093628883362, |
| "learning_rate": 0.00033442185059073706, |
| "loss": 0.6517, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_loss": 1.006990671157837, |
| "eval_runtime": 845.1184, |
| "eval_samples_per_second": 1.298, |
| "eval_steps_per_second": 0.021, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.4125, |
| "grad_norm": 0.09464213252067566, |
| "learning_rate": 0.00032497574781753367, |
| "loss": 0.8455, |
| "step": 825 |
| }, |
| { |
| "epoch": 0.425, |
| "grad_norm": 0.10187377035617828, |
| "learning_rate": 0.000315410490505086, |
| "loss": 0.8217, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.4375, |
| "grad_norm": 0.09793733805418015, |
| "learning_rate": 0.0003057412801533589, |
| "loss": 0.6218, |
| "step": 875 |
| }, |
| { |
| "epoch": 0.45, |
| "grad_norm": 0.09902466833591461, |
| "learning_rate": 0.0002959834834687587, |
| "loss": 0.869, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.45, |
| "eval_loss": 1.0097707509994507, |
| "eval_runtime": 847.3981, |
| "eval_samples_per_second": 1.295, |
| "eval_steps_per_second": 0.021, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.4625, |
| "grad_norm": 0.10546339303255081, |
| "learning_rate": 0.00028615260794273236, |
| "loss": 0.8204, |
| "step": 925 |
| }, |
| { |
| "epoch": 0.475, |
| "grad_norm": 0.10600671917200089, |
| "learning_rate": 0.00027626427720662416, |
| "loss": 0.5917, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.4875, |
| "grad_norm": 0.09627766162157059, |
| "learning_rate": 0.00026633420620195917, |
| "loss": 0.8667, |
| "step": 975 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 0.09961821138858795, |
| "learning_rate": 0.00025637817620561263, |
| "loss": 0.8215, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_loss": 1.023415446281433, |
| "eval_runtime": 844.006, |
| "eval_samples_per_second": 1.3, |
| "eval_steps_per_second": 0.021, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.5125, |
| "grad_norm": 0.09977111220359802, |
| "learning_rate": 0.0002464120097495559, |
| "loss": 0.5829, |
| "step": 1025 |
| }, |
| { |
| "epoch": 0.525, |
| "grad_norm": 0.10276953876018524, |
| "learning_rate": 0.00023645154547503855, |
| "loss": 0.8857, |
| "step": 1050 |
| }, |
| { |
| "epoch": 0.5375, |
| "grad_norm": 0.10077586770057678, |
| "learning_rate": 0.00022651261296116894, |
| "loss": 0.8018, |
| "step": 1075 |
| }, |
| { |
| "epoch": 0.55, |
| "grad_norm": 0.09652134776115417, |
| "learning_rate": 0.00021661100756789666, |
| "loss": 0.561, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.55, |
| "eval_loss": 1.0231441259384155, |
| "eval_runtime": 845.0138, |
| "eval_samples_per_second": 1.298, |
| "eval_steps_per_second": 0.021, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.5625, |
| "grad_norm": 0.0990302637219429, |
| "learning_rate": 0.00020676246533337764, |
| "loss": 0.9092, |
| "step": 1125 |
| }, |
| { |
| "epoch": 0.575, |
| "grad_norm": 0.11152709275484085, |
| "learning_rate": 0.00019698263796561526, |
| "loss": 0.8159, |
| "step": 1150 |
| }, |
| { |
| "epoch": 0.5875, |
| "grad_norm": 0.10092920064926147, |
| "learning_rate": 0.00018728706796812333, |
| "loss": 0.7329, |
| "step": 1175 |
| }, |
| { |
| "epoch": 0.6, |
| "grad_norm": 0.10138432681560516, |
| "learning_rate": 0.00017769116393914037, |
| "loss": 0.7873, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.6, |
| "eval_loss": 1.0180704593658447, |
| "eval_runtime": 843.0775, |
| "eval_samples_per_second": 1.301, |
| "eval_steps_per_second": 0.021, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.6125, |
| "grad_norm": 0.10279645770788193, |
| "learning_rate": 0.00016821017608365264, |
| "loss": 0.7186, |
| "step": 1225 |
| }, |
| { |
| "epoch": 0.625, |
| "grad_norm": 0.10766831040382385, |
| "learning_rate": 0.00015885917197714112, |
| "loss": 0.7201, |
| "step": 1250 |
| }, |
| { |
| "epoch": 0.6375, |
| "grad_norm": 0.10177863389253616, |
| "learning_rate": 0.00014965301261957238, |
| "loss": 0.8009, |
| "step": 1275 |
| }, |
| { |
| "epoch": 0.65, |
| "grad_norm": 0.0983508974313736, |
| "learning_rate": 0.00014060632881768558, |
| "loss": 0.7023, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.65, |
| "eval_loss": 1.0266674757003784, |
| "eval_runtime": 854.7884, |
| "eval_samples_per_second": 1.283, |
| "eval_steps_per_second": 0.021, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.6625, |
| "grad_norm": 0.10579918324947357, |
| "learning_rate": 0.00013173349793311424, |
| "loss": 0.7624, |
| "step": 1325 |
| }, |
| { |
| "epoch": 0.675, |
| "grad_norm": 0.10350169986486435, |
| "learning_rate": 0.0001230486210332916, |
| "loss": 0.7857, |
| "step": 1350 |
| }, |
| { |
| "epoch": 0.6875, |
| "grad_norm": 0.10701841115951538, |
| "learning_rate": 0.00011456550048145536, |
| "loss": 0.6771, |
| "step": 1375 |
| }, |
| { |
| "epoch": 0.7, |
| "grad_norm": 0.10333641618490219, |
| "learning_rate": 0.00010629761800136473, |
| "loss": 0.7669, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.7, |
| "eval_loss": 1.0311139822006226, |
| "eval_runtime": 850.5438, |
| "eval_samples_per_second": 1.29, |
| "eval_steps_per_second": 0.021, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.7125, |
| "grad_norm": 0.10531915724277496, |
| "learning_rate": 9.82581132515907e-05, |
| "loss": 0.7869, |
| "step": 1425 |
| }, |
| { |
| "epoch": 0.725, |
| "grad_norm": 0.08779594302177429, |
| "learning_rate": 9.045976294343145e-05, |
| "loss": 0.6651, |
| "step": 1450 |
| }, |
| { |
| "epoch": 0.7375, |
| "grad_norm": 0.09893123060464859, |
| "learning_rate": 8.291496053563699e-05, |
| "loss": 0.7938, |
| "step": 1475 |
| }, |
| { |
| "epoch": 0.75, |
| "grad_norm": 0.09708540141582489, |
| "learning_rate": 7.563569653821565e-05, |
| "loss": 0.7873, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.75, |
| "eval_loss": 1.0259246826171875, |
| "eval_runtime": 843.8488, |
| "eval_samples_per_second": 1.3, |
| "eval_steps_per_second": 0.021, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.7625, |
| "grad_norm": 0.1001739501953125, |
| "learning_rate": 6.863353945662288e-05, |
| "loss": 0.6202, |
| "step": 1525 |
| }, |
| { |
| "epoch": 0.775, |
| "grad_norm": 0.10724864155054092, |
| "learning_rate": 6.191961740661687e-05, |
| "loss": 0.8107, |
| "step": 1550 |
| }, |
| { |
| "epoch": 0.7875, |
| "grad_norm": 0.10225515067577362, |
| "learning_rate": 5.550460042899982e-05, |
| "loss": 0.8041, |
| "step": 1575 |
| }, |
| { |
| "epoch": 0.8, |
| "grad_norm": 0.09861259162425995, |
| "learning_rate": 4.9398683532350855e-05, |
| "loss": 0.5894, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_loss": 1.0317810773849487, |
| "eval_runtime": 852.0414, |
| "eval_samples_per_second": 1.287, |
| "eval_steps_per_second": 0.021, |
| "step": 1600 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 2000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9223372036854775807, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.608283832933417e+17, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
