| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.29301745635910226, |
| "eval_steps": 500, |
| "global_step": 235, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.006234413965087282, |
| "grad_norm": 14.221988677978516, |
| "learning_rate": 2.162496240601504e-06, |
| "loss": 0.4075, |
| "mean_token_accuracy": 0.8846674263477325, |
| "num_tokens": 830507.0, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.012468827930174564, |
| "grad_norm": 2.5054450035095215, |
| "learning_rate": 4.865616541353383e-06, |
| "loss": 0.2149, |
| "mean_token_accuracy": 0.9441515356302261, |
| "num_tokens": 1652510.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.018703241895261846, |
| "grad_norm": 1.1095941066741943, |
| "learning_rate": 7.568736842105265e-06, |
| "loss": 0.1824, |
| "mean_token_accuracy": 0.9492746978998184, |
| "num_tokens": 2482550.0, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.02493765586034913, |
| "grad_norm": 1.1062263250350952, |
| "learning_rate": 1.0271857142857143e-05, |
| "loss": 0.1744, |
| "mean_token_accuracy": 0.9496640861034393, |
| "num_tokens": 3302052.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.03117206982543641, |
| "grad_norm": 0.7716283798217773, |
| "learning_rate": 1.2974977443609024e-05, |
| "loss": 0.147, |
| "mean_token_accuracy": 0.9584015130996704, |
| "num_tokens": 4132883.0, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.03740648379052369, |
| "grad_norm": 0.8777795433998108, |
| "learning_rate": 1.5678097744360906e-05, |
| "loss": 0.1324, |
| "mean_token_accuracy": 0.9609348356723786, |
| "num_tokens": 4963455.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.043640897755610975, |
| "grad_norm": 0.7445647120475769, |
| "learning_rate": 1.8381218045112784e-05, |
| "loss": 0.1244, |
| "mean_token_accuracy": 0.9621825248003006, |
| "num_tokens": 5792675.0, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.04987531172069826, |
| "grad_norm": 0.7758142352104187, |
| "learning_rate": 1.8911699600975878e-05, |
| "loss": 0.1304, |
| "mean_token_accuracy": 0.9598606675863266, |
| "num_tokens": 6624423.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.05610972568578554, |
| "grad_norm": 0.722838282585144, |
| "learning_rate": 1.8870545366153036e-05, |
| "loss": 0.1379, |
| "mean_token_accuracy": 0.9606135696172714, |
| "num_tokens": 7457619.0, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.06234413965087282, |
| "grad_norm": 0.7575320601463318, |
| "learning_rate": 1.8797929157448593e-05, |
| "loss": 0.1484, |
| "mean_token_accuracy": 0.9562382131814957, |
| "num_tokens": 8280329.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.0685785536159601, |
| "grad_norm": 0.8119825720787048, |
| "learning_rate": 1.8694175296478475e-05, |
| "loss": 0.144, |
| "mean_token_accuracy": 0.9597607880830765, |
| "num_tokens": 9094769.0, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.07481296758104738, |
| "grad_norm": 0.9321528673171997, |
| "learning_rate": 1.8559747173172413e-05, |
| "loss": 0.13, |
| "mean_token_accuracy": 0.9616361707448959, |
| "num_tokens": 9924881.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.08104738154613467, |
| "grad_norm": 0.8656273484230042, |
| "learning_rate": 1.8395245176162015e-05, |
| "loss": 0.1692, |
| "mean_token_accuracy": 0.9533779501914978, |
| "num_tokens": 10733869.0, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.08728179551122195, |
| "grad_norm": 0.4899306893348694, |
| "learning_rate": 1.820140401129939e-05, |
| "loss": 0.1366, |
| "mean_token_accuracy": 0.9600318193435669, |
| "num_tokens": 11542671.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.09351620947630923, |
| "grad_norm": 0.6704471707344055, |
| "learning_rate": 1.797908942028248e-05, |
| "loss": 0.1182, |
| "mean_token_accuracy": 0.964169979095459, |
| "num_tokens": 12368511.0, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.09975062344139651, |
| "grad_norm": 0.966526985168457, |
| "learning_rate": 1.7729294314042456e-05, |
| "loss": 0.1242, |
| "mean_token_accuracy": 0.9644079297780991, |
| "num_tokens": 13194259.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.1059850374064838, |
| "grad_norm": 0.6235336661338806, |
| "learning_rate": 1.745313433816246e-05, |
| "loss": 0.1129, |
| "mean_token_accuracy": 0.9659302473068238, |
| "num_tokens": 14015391.0, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.11221945137157108, |
| "grad_norm": 0.5269407629966736, |
| "learning_rate": 1.7151842890133638e-05, |
| "loss": 0.108, |
| "mean_token_accuracy": 0.9666280597448349, |
| "num_tokens": 14850900.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.11845386533665836, |
| "grad_norm": 0.7091829776763916, |
| "learning_rate": 1.6826765610702495e-05, |
| "loss": 0.1193, |
| "mean_token_accuracy": 0.9637973874807357, |
| "num_tokens": 15679071.0, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.12468827930174564, |
| "grad_norm": 0.8810948729515076, |
| "learning_rate": 1.6479354373912677e-05, |
| "loss": 0.1318, |
| "mean_token_accuracy": 0.9605223119258881, |
| "num_tokens": 16509130.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.13092269326683292, |
| "grad_norm": 0.47906363010406494, |
| "learning_rate": 1.611116080268305e-05, |
| "loss": 0.1075, |
| "mean_token_accuracy": 0.9674216449260712, |
| "num_tokens": 17330423.0, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.1371571072319202, |
| "grad_norm": 0.502947211265564, |
| "learning_rate": 1.572382933888308e-05, |
| "loss": 0.1241, |
| "mean_token_accuracy": 0.9633585959672928, |
| "num_tokens": 18152170.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.1433915211970075, |
| "grad_norm": 0.7405344843864441, |
| "learning_rate": 1.5319089898856233e-05, |
| "loss": 0.1045, |
| "mean_token_accuracy": 0.9681421965360641, |
| "num_tokens": 18955798.0, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.14962593516209477, |
| "grad_norm": 0.5441982746124268, |
| "learning_rate": 1.4898750147193666e-05, |
| "loss": 0.1096, |
| "mean_token_accuracy": 0.9671149224042892, |
| "num_tokens": 19779617.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.15586034912718205, |
| "grad_norm": 0.5530046224594116, |
| "learning_rate": 1.4464687423265335e-05, |
| "loss": 0.116, |
| "mean_token_accuracy": 0.9664247155189514, |
| "num_tokens": 20593859.0, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.16209476309226933, |
| "grad_norm": 0.7859145998954773, |
| "learning_rate": 1.4018840356566632e-05, |
| "loss": 0.1148, |
| "mean_token_accuracy": 0.9646897494792939, |
| "num_tokens": 21407432.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.16832917705735662, |
| "grad_norm": 0.5086043477058411, |
| "learning_rate": 1.3563200208328454e-05, |
| "loss": 0.1195, |
| "mean_token_accuracy": 0.9639829397201538, |
| "num_tokens": 22234975.0, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.1745635910224439, |
| "grad_norm": 0.5094566941261292, |
| "learning_rate": 1.309980197806116e-05, |
| "loss": 0.114, |
| "mean_token_accuracy": 0.9666442781686783, |
| "num_tokens": 23072659.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.18079800498753118, |
| "grad_norm": 0.7448077201843262, |
| "learning_rate": 1.2630715314752758e-05, |
| "loss": 0.115, |
| "mean_token_accuracy": 0.9651498794555664, |
| "num_tokens": 23890121.0, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.18703241895261846, |
| "grad_norm": 0.5883263349533081, |
| "learning_rate": 1.2158035273314107e-05, |
| "loss": 0.1276, |
| "mean_token_accuracy": 0.9617824375629425, |
| "num_tokens": 24726512.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.19326683291770574, |
| "grad_norm": 0.5835016369819641, |
| "learning_rate": 1.168387295755509e-05, |
| "loss": 0.1103, |
| "mean_token_accuracy": 0.966162720322609, |
| "num_tokens": 25555769.0, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.19950124688279303, |
| "grad_norm": 0.6837282180786133, |
| "learning_rate": 1.1210346091482546e-05, |
| "loss": 0.108, |
| "mean_token_accuracy": 0.9675496876239776, |
| "num_tokens": 26377083.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.2057356608478803, |
| "grad_norm": 0.6743711829185486, |
| "learning_rate": 1.073956956103075e-05, |
| "loss": 0.1375, |
| "mean_token_accuracy": 0.9600846111774445, |
| "num_tokens": 27204765.0, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.2119700748129676, |
| "grad_norm": 0.5510470867156982, |
| "learning_rate": 1.0273645968467439e-05, |
| "loss": 0.1148, |
| "mean_token_accuracy": 0.9651833415031433, |
| "num_tokens": 28012010.0, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.21820448877805487, |
| "grad_norm": 0.6454848647117615, |
| "learning_rate": 9.814656241661682e-06, |
| "loss": 0.1331, |
| "mean_token_accuracy": 0.961262908577919, |
| "num_tokens": 28823845.0, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.22443890274314215, |
| "grad_norm": 0.678521990776062, |
| "learning_rate": 9.364650340154834e-06, |
| "loss": 0.1309, |
| "mean_token_accuracy": 0.9619206607341766, |
| "num_tokens": 29651835.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.23067331670822944, |
| "grad_norm": 0.47411710023880005, |
| "learning_rate": 8.925638099543624e-06, |
| "loss": 0.1096, |
| "mean_token_accuracy": 0.9666814059019089, |
| "num_tokens": 30479280.0, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.23690773067331672, |
| "grad_norm": 0.5466766953468323, |
| "learning_rate": 8.499580255066467e-06, |
| "loss": 0.1109, |
| "mean_token_accuracy": 0.9657571285963058, |
| "num_tokens": 31310265.0, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.243142144638404, |
| "grad_norm": 0.6293110251426697, |
| "learning_rate": 8.088379684483909e-06, |
| "loss": 0.1197, |
| "mean_token_accuracy": 0.9659775942564011, |
| "num_tokens": 32130329.0, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.24937655860349128, |
| "grad_norm": 0.47574180364608765, |
| "learning_rate": 7.693872909364632e-06, |
| "loss": 0.1174, |
| "mean_token_accuracy": 0.965189179778099, |
| "num_tokens": 32948565.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.25561097256857856, |
| "grad_norm": 0.7879202365875244, |
| "learning_rate": 7.317821892734294e-06, |
| "loss": 0.112, |
| "mean_token_accuracy": 0.9663525938987731, |
| "num_tokens": 33778180.0, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.26184538653366585, |
| "grad_norm": 0.8077799081802368, |
| "learning_rate": 6.961906169720941e-06, |
| "loss": 0.1291, |
| "mean_token_accuracy": 0.9621292024850845, |
| "num_tokens": 34587229.0, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.26807980049875313, |
| "grad_norm": 0.6214185357093811, |
| "learning_rate": 6.627715346343407e-06, |
| "loss": 0.119, |
| "mean_token_accuracy": 0.9644369512796402, |
| "num_tokens": 35401895.0, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.2743142144638404, |
| "grad_norm": 1.0910162925720215, |
| "learning_rate": 6.316741999944951e-06, |
| "loss": 0.1248, |
| "mean_token_accuracy": 0.9634313136339188, |
| "num_tokens": 36217604.0, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.2805486284289277, |
| "grad_norm": 0.5792627930641174, |
| "learning_rate": 6.030375012980519e-06, |
| "loss": 0.1028, |
| "mean_token_accuracy": 0.9697955340147019, |
| "num_tokens": 37035942.0, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.286783042394015, |
| "grad_norm": 0.5335521101951599, |
| "learning_rate": 5.769893369930536e-06, |
| "loss": 0.1277, |
| "mean_token_accuracy": 0.9631022900342942, |
| "num_tokens": 37865680.0, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.29301745635910226, |
| "grad_norm": 0.4827023446559906, |
| "learning_rate": 5.5364604450457455e-06, |
| "loss": 0.1132, |
| "mean_token_accuracy": 0.9672250717878341, |
| "num_tokens": 38694201.0, |
| "step": 235 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 235, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.3229666486858547e+18, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
