| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.039846590626089556, | |
| "eval_steps": 17, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00019923295313044778, | |
| "eval_loss": 0.9888173341751099, | |
| "eval_runtime": 244.5615, | |
| "eval_samples_per_second": 34.568, | |
| "eval_steps_per_second": 4.322, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005976988593913434, | |
| "grad_norm": 0.2933606505393982, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9647, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0011953977187826868, | |
| "grad_norm": 0.30875223875045776, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9835, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00179309657817403, | |
| "grad_norm": 0.27685415744781494, | |
| "learning_rate": 9e-05, | |
| "loss": 1.0392, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0023907954375653735, | |
| "grad_norm": 0.19165818393230438, | |
| "learning_rate": 9.997266286704631e-05, | |
| "loss": 0.8622, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0029884942969567166, | |
| "grad_norm": 0.21981056034564972, | |
| "learning_rate": 9.98292246503335e-05, | |
| "loss": 0.8396, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.003386960203217612, | |
| "eval_loss": 0.854656457901001, | |
| "eval_runtime": 246.6307, | |
| "eval_samples_per_second": 34.278, | |
| "eval_steps_per_second": 4.286, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.00358619315634806, | |
| "grad_norm": 0.18528936803340912, | |
| "learning_rate": 9.956320346634876e-05, | |
| "loss": 0.8472, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0041838920157394035, | |
| "grad_norm": 0.17711381614208221, | |
| "learning_rate": 9.917525374361912e-05, | |
| "loss": 0.8624, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.004781590875130747, | |
| "grad_norm": 0.18825452029705048, | |
| "learning_rate": 9.86663298624003e-05, | |
| "loss": 0.7784, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.00537928973452209, | |
| "grad_norm": 0.2445794641971588, | |
| "learning_rate": 9.803768380684242e-05, | |
| "loss": 0.8382, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.005976988593913433, | |
| "grad_norm": 0.25549814105033875, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 0.83, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.006574687453304777, | |
| "grad_norm": 0.2030678689479828, | |
| "learning_rate": 9.642770192448536e-05, | |
| "loss": 0.7497, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.006773920406435224, | |
| "eval_loss": 0.7824097871780396, | |
| "eval_runtime": 246.0525, | |
| "eval_samples_per_second": 34.359, | |
| "eval_steps_per_second": 4.296, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.00717238631269612, | |
| "grad_norm": 0.24549758434295654, | |
| "learning_rate": 9.545032675245813e-05, | |
| "loss": 0.863, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0077700851720874636, | |
| "grad_norm": 0.20169614255428314, | |
| "learning_rate": 9.43611409721806e-05, | |
| "loss": 0.7357, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.008367784031478807, | |
| "grad_norm": 0.1820153445005417, | |
| "learning_rate": 9.316282404787871e-05, | |
| "loss": 0.7357, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.00896548289087015, | |
| "grad_norm": 0.1945677399635315, | |
| "learning_rate": 9.185832391312644e-05, | |
| "loss": 0.764, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.009563181750261494, | |
| "grad_norm": 0.19976423680782318, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.7432, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.010160880609652836, | |
| "grad_norm": 0.2329975664615631, | |
| "learning_rate": 8.894386393810563e-05, | |
| "loss": 0.7484, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.010160880609652836, | |
| "eval_loss": 0.7552871704101562, | |
| "eval_runtime": 246.9974, | |
| "eval_samples_per_second": 34.227, | |
| "eval_steps_per_second": 4.279, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.01075857946904418, | |
| "grad_norm": 0.24253146350383759, | |
| "learning_rate": 8.73410738492077e-05, | |
| "loss": 0.7833, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.011356278328435523, | |
| "grad_norm": 0.2245894819498062, | |
| "learning_rate": 8.564642241456986e-05, | |
| "loss": 0.7807, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.011953977187826866, | |
| "grad_norm": 0.2301652878522873, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 0.7355, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.01255167604721821, | |
| "grad_norm": 0.21688085794448853, | |
| "learning_rate": 8.199842702516583e-05, | |
| "loss": 0.7521, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.013149374906609553, | |
| "grad_norm": 0.2115517556667328, | |
| "learning_rate": 8.005405736415126e-05, | |
| "loss": 0.7837, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.013547840812870448, | |
| "eval_loss": 0.7411483526229858, | |
| "eval_runtime": 246.9838, | |
| "eval_samples_per_second": 34.229, | |
| "eval_steps_per_second": 4.28, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.013747073766000897, | |
| "grad_norm": 0.2068549543619156, | |
| "learning_rate": 7.803575286758364e-05, | |
| "loss": 0.6639, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.01434477262539224, | |
| "grad_norm": 0.1979922354221344, | |
| "learning_rate": 7.594847868906076e-05, | |
| "loss": 0.7337, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.014942471484783584, | |
| "grad_norm": 0.20813895761966705, | |
| "learning_rate": 7.379736965185368e-05, | |
| "loss": 0.7659, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.015540170344174927, | |
| "grad_norm": 0.20922593772411346, | |
| "learning_rate": 7.158771761692464e-05, | |
| "loss": 0.7355, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.01613786920356627, | |
| "grad_norm": 0.1938929706811905, | |
| "learning_rate": 6.932495846462261e-05, | |
| "loss": 0.7499, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.016735568062957614, | |
| "grad_norm": 0.21271425485610962, | |
| "learning_rate": 6.701465872208216e-05, | |
| "loss": 0.6686, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.01693480101608806, | |
| "eval_loss": 0.7316516041755676, | |
| "eval_runtime": 246.9875, | |
| "eval_samples_per_second": 34.228, | |
| "eval_steps_per_second": 4.28, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.017333266922348958, | |
| "grad_norm": 0.2120652049779892, | |
| "learning_rate": 6.466250186922325e-05, | |
| "loss": 0.7224, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.0179309657817403, | |
| "grad_norm": 0.22158583998680115, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 0.7273, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.018528664641131645, | |
| "grad_norm": 0.21288037300109863, | |
| "learning_rate": 5.985585137257401e-05, | |
| "loss": 0.7102, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.019126363500522988, | |
| "grad_norm": 0.20418782532215118, | |
| "learning_rate": 5.74131823855921e-05, | |
| "loss": 0.7021, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.01972406235991433, | |
| "grad_norm": 0.22563454508781433, | |
| "learning_rate": 5.495227651252315e-05, | |
| "loss": 0.6756, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.02032176121930567, | |
| "grad_norm": 0.21469104290008545, | |
| "learning_rate": 5.247918773366112e-05, | |
| "loss": 0.6921, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.02032176121930567, | |
| "eval_loss": 0.7249154448509216, | |
| "eval_runtime": 246.7439, | |
| "eval_samples_per_second": 34.262, | |
| "eval_steps_per_second": 4.284, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.020919460078697015, | |
| "grad_norm": 0.2427963763475418, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7018, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.02151715893808836, | |
| "grad_norm": 0.2553952932357788, | |
| "learning_rate": 4.7520812266338885e-05, | |
| "loss": 0.7219, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.022114857797479702, | |
| "grad_norm": 0.23926398158073425, | |
| "learning_rate": 4.504772348747687e-05, | |
| "loss": 0.7707, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.022712556656871045, | |
| "grad_norm": 0.24447274208068848, | |
| "learning_rate": 4.2586817614407895e-05, | |
| "loss": 0.7402, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.02331025551626239, | |
| "grad_norm": 0.218000590801239, | |
| "learning_rate": 4.0144148627425993e-05, | |
| "loss": 0.7143, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.023708721422523286, | |
| "eval_loss": 0.7203857898712158, | |
| "eval_runtime": 246.9873, | |
| "eval_samples_per_second": 34.228, | |
| "eval_steps_per_second": 4.28, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.023907954375653732, | |
| "grad_norm": 0.21252404153347015, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 0.6713, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.024505653235045076, | |
| "grad_norm": 0.20347541570663452, | |
| "learning_rate": 3.533749813077677e-05, | |
| "loss": 0.7159, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.02510335209443642, | |
| "grad_norm": 0.2782200574874878, | |
| "learning_rate": 3.298534127791785e-05, | |
| "loss": 0.749, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.025701050953827763, | |
| "grad_norm": 0.21792350709438324, | |
| "learning_rate": 3.0675041535377405e-05, | |
| "loss": 0.6967, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.026298749813219106, | |
| "grad_norm": 0.19734562933444977, | |
| "learning_rate": 2.8412282383075363e-05, | |
| "loss": 0.6717, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.02689644867261045, | |
| "grad_norm": 0.25884565711021423, | |
| "learning_rate": 2.6202630348146324e-05, | |
| "loss": 0.7098, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.027095681625740897, | |
| "eval_loss": 0.7171828746795654, | |
| "eval_runtime": 247.1294, | |
| "eval_samples_per_second": 34.209, | |
| "eval_steps_per_second": 4.277, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.027494147532001793, | |
| "grad_norm": 0.2502513527870178, | |
| "learning_rate": 2.405152131093926e-05, | |
| "loss": 0.7678, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.028091846391393137, | |
| "grad_norm": 0.21426473557949066, | |
| "learning_rate": 2.196424713241637e-05, | |
| "loss": 0.7028, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.02868954525078448, | |
| "grad_norm": 0.2242855429649353, | |
| "learning_rate": 1.9945942635848748e-05, | |
| "loss": 0.7267, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.029287244110175824, | |
| "grad_norm": 0.2083127796649933, | |
| "learning_rate": 1.800157297483417e-05, | |
| "loss": 0.6915, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.029884942969567167, | |
| "grad_norm": 0.20724868774414062, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 0.7249, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.03048264182895851, | |
| "grad_norm": 0.23638051748275757, | |
| "learning_rate": 1.435357758543015e-05, | |
| "loss": 0.6968, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.03048264182895851, | |
| "eval_loss": 0.7152303457260132, | |
| "eval_runtime": 246.9292, | |
| "eval_samples_per_second": 34.237, | |
| "eval_steps_per_second": 4.281, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.031080340688349854, | |
| "grad_norm": 0.21780776977539062, | |
| "learning_rate": 1.2658926150792322e-05, | |
| "loss": 0.754, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.0316780395477412, | |
| "grad_norm": 0.23965197801589966, | |
| "learning_rate": 1.1056136061894384e-05, | |
| "loss": 0.6907, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.03227573840713254, | |
| "grad_norm": 0.21216127276420593, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.6891, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.032873437266523885, | |
| "grad_norm": 0.21560019254684448, | |
| "learning_rate": 8.141676086873572e-06, | |
| "loss": 0.7348, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.03347113612591523, | |
| "grad_norm": 0.24270105361938477, | |
| "learning_rate": 6.837175952121306e-06, | |
| "loss": 0.7507, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.03386960203217612, | |
| "eval_loss": 0.7143241167068481, | |
| "eval_runtime": 246.9514, | |
| "eval_samples_per_second": 34.233, | |
| "eval_steps_per_second": 4.28, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.03406883498530657, | |
| "grad_norm": 0.19754469394683838, | |
| "learning_rate": 5.6388590278194096e-06, | |
| "loss": 0.7157, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 0.034666533844697915, | |
| "grad_norm": 0.21476249396800995, | |
| "learning_rate": 4.549673247541875e-06, | |
| "loss": 0.7158, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.03526423270408926, | |
| "grad_norm": 0.2122727334499359, | |
| "learning_rate": 3.5722980755146517e-06, | |
| "loss": 0.6921, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 0.0358619315634806, | |
| "grad_norm": 0.22367681562900543, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 0.7906, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.036459630422871946, | |
| "grad_norm": 0.22603553533554077, | |
| "learning_rate": 1.962316193157593e-06, | |
| "loss": 0.6743, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 0.03705732928226329, | |
| "grad_norm": 0.23509639501571655, | |
| "learning_rate": 1.333670137599713e-06, | |
| "loss": 0.7062, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.03725656223539373, | |
| "eval_loss": 0.7139464020729065, | |
| "eval_runtime": 246.9924, | |
| "eval_samples_per_second": 34.228, | |
| "eval_steps_per_second": 4.279, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 0.03765502814165463, | |
| "grad_norm": 0.221495121717453, | |
| "learning_rate": 8.247462563808817e-07, | |
| "loss": 0.7116, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.038252727001045976, | |
| "grad_norm": 0.22564756870269775, | |
| "learning_rate": 4.367965336512403e-07, | |
| "loss": 0.7366, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.03885042586043732, | |
| "grad_norm": 0.23282335698604584, | |
| "learning_rate": 1.7077534966650766e-07, | |
| "loss": 0.7022, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.03944812471982866, | |
| "grad_norm": 0.23939348757266998, | |
| "learning_rate": 2.7337132953697554e-08, | |
| "loss": 0.7262, | |
| "step": 198 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 17, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.516788593636147e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |