| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0034081710901887274, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.816342180377454e-05, | |
| "grad_norm": 10.726434707641602, | |
| "learning_rate": 1e-05, | |
| "loss": 6.7949, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 6.816342180377454e-05, | |
| "eval_loss": 7.433081150054932, | |
| "eval_runtime": 3610.8008, | |
| "eval_samples_per_second": 6.843, | |
| "eval_steps_per_second": 3.422, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001363268436075491, | |
| "grad_norm": 13.176578521728516, | |
| "learning_rate": 2e-05, | |
| "loss": 7.2107, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00020449026541132366, | |
| "grad_norm": 12.275219917297363, | |
| "learning_rate": 3e-05, | |
| "loss": 7.3524, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0002726536872150982, | |
| "grad_norm": 13.537530899047852, | |
| "learning_rate": 4e-05, | |
| "loss": 7.3109, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00034081710901887275, | |
| "grad_norm": 16.737390518188477, | |
| "learning_rate": 5e-05, | |
| "loss": 7.1299, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0004089805308226473, | |
| "grad_norm": 16.55790138244629, | |
| "learning_rate": 6e-05, | |
| "loss": 5.7001, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00047714395262642183, | |
| "grad_norm": 12.568345069885254, | |
| "learning_rate": 7e-05, | |
| "loss": 4.4402, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0005453073744301963, | |
| "grad_norm": 14.332297325134277, | |
| "learning_rate": 8e-05, | |
| "loss": 1.9874, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.000613470796233971, | |
| "grad_norm": 13.18618392944336, | |
| "learning_rate": 9e-05, | |
| "loss": 2.3065, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0006816342180377455, | |
| "grad_norm": 8.724907875061035, | |
| "learning_rate": 0.0001, | |
| "loss": 2.3028, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00074979763984152, | |
| "grad_norm": 17.249616622924805, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 1.5665, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0008179610616452946, | |
| "grad_norm": 12.082292556762695, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.1444, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0008861244834490692, | |
| "grad_norm": 6.85310697555542, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 1.3781, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0008861244834490692, | |
| "eval_loss": 1.087676763534546, | |
| "eval_runtime": 3614.8284, | |
| "eval_samples_per_second": 6.835, | |
| "eval_steps_per_second": 3.418, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0009542879052528437, | |
| "grad_norm": 10.216886520385742, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.108, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0010224513270566183, | |
| "grad_norm": 9.747845649719238, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.9771, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0010906147488603927, | |
| "grad_norm": 5.7691216468811035, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.9156, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0011587781706641673, | |
| "grad_norm": 5.967700958251953, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.803, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.001226941592467942, | |
| "grad_norm": 5.386447906494141, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.0213, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0012951050142717164, | |
| "grad_norm": 5.265084743499756, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 1.1579, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.001363268436075491, | |
| "grad_norm": 3.2680466175079346, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.7542, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0014314318578792656, | |
| "grad_norm": 5.0342020988464355, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 1.1486, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00149959527968304, | |
| "grad_norm": 3.5533130168914795, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.7507, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0015677587014868146, | |
| "grad_norm": 3.747938871383667, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.7606, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0016359221232905893, | |
| "grad_norm": 3.583609104156494, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.5768, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0017040855450943637, | |
| "grad_norm": 3.155369281768799, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.6934, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0017722489668981383, | |
| "grad_norm": 3.687058448791504, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.7966, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0017722489668981383, | |
| "eval_loss": 0.6649627685546875, | |
| "eval_runtime": 3621.1469, | |
| "eval_samples_per_second": 6.824, | |
| "eval_steps_per_second": 3.412, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.001840412388701913, | |
| "grad_norm": 3.756096839904785, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.6924, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0019085758105056873, | |
| "grad_norm": 2.65657901763916, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.4216, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.001976739232309462, | |
| "grad_norm": 3.051575183868408, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.6558, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0020449026541132366, | |
| "grad_norm": 3.750715494155884, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5206, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.002113066075917011, | |
| "grad_norm": 2.827176094055176, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.531, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0021812294977207854, | |
| "grad_norm": 3.170285940170288, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.7352, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.00224939291952456, | |
| "grad_norm": 4.301424026489258, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.9347, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0023175563413283347, | |
| "grad_norm": 3.8904523849487305, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.58, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0023857197631321093, | |
| "grad_norm": 3.3350067138671875, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.652, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.002453883184935884, | |
| "grad_norm": 3.35461163520813, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.8035, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0025220466067396585, | |
| "grad_norm": 3.971146583557129, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.6434, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0025902100285434327, | |
| "grad_norm": 2.744048833847046, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.468, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0026583734503472073, | |
| "grad_norm": 3.2330892086029053, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.5389, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0026583734503472073, | |
| "eval_loss": 0.612258791923523, | |
| "eval_runtime": 3622.0619, | |
| "eval_samples_per_second": 6.822, | |
| "eval_steps_per_second": 3.411, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.002726536872150982, | |
| "grad_norm": 3.731128454208374, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.7914, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0027947002939547566, | |
| "grad_norm": 2.8531653881073, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.7838, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0028628637157585312, | |
| "grad_norm": 2.8280680179595947, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.4223, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.002931027137562306, | |
| "grad_norm": 2.862281084060669, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.5895, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.00299919055936608, | |
| "grad_norm": 2.7972028255462646, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.6017, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0030673539811698547, | |
| "grad_norm": 4.903096675872803, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.5924, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0031355174029736293, | |
| "grad_norm": 5.058813571929932, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.473, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.003203680824777404, | |
| "grad_norm": 3.77838134765625, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.6534, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0032718442465811785, | |
| "grad_norm": 3.0591816902160645, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.6966, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0033400076683849527, | |
| "grad_norm": 2.853864908218384, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.5413, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0034081710901887274, | |
| "grad_norm": 2.967231035232544, | |
| "learning_rate": 0.0, | |
| "loss": 0.5969, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.967549134241792e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |