| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 750000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1e-05, | |
| "loss": 3.5986, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.83344888888889e-06, | |
| "loss": 0.685, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_accuracy": 0.5626, | |
| "eval_loss": 0.676932692527771, | |
| "eval_runtime": 23.4325, | |
| "eval_samples_per_second": 426.757, | |
| "eval_steps_per_second": 6.7, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.666853333333333e-06, | |
| "loss": 0.6788, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.5642, | |
| "eval_loss": 0.6758846640586853, | |
| "eval_runtime": 23.4496, | |
| "eval_samples_per_second": 426.447, | |
| "eval_steps_per_second": 6.695, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.500266666666668e-06, | |
| "loss": 0.6774, | |
| "step": 56250 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_accuracy": 0.5681, | |
| "eval_loss": 0.6747780442237854, | |
| "eval_runtime": 23.6291, | |
| "eval_samples_per_second": 423.207, | |
| "eval_steps_per_second": 6.644, | |
| "step": 56250 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.333671111111113e-06, | |
| "loss": 0.6771, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.5748, | |
| "eval_loss": 0.6741061806678772, | |
| "eval_runtime": 23.3134, | |
| "eval_samples_per_second": 428.938, | |
| "eval_steps_per_second": 6.734, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 9.167075555555556e-06, | |
| "loss": 0.6762, | |
| "step": 93750 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_accuracy": 0.5738, | |
| "eval_loss": 0.6736128926277161, | |
| "eval_runtime": 23.6397, | |
| "eval_samples_per_second": 423.017, | |
| "eval_steps_per_second": 6.641, | |
| "step": 93750 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 9.00048e-06, | |
| "loss": 0.6755, | |
| "step": 112500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_accuracy": 0.5736, | |
| "eval_loss": 0.6737410426139832, | |
| "eval_runtime": 23.3251, | |
| "eval_samples_per_second": 428.723, | |
| "eval_steps_per_second": 6.731, | |
| "step": 112500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 8.833866666666667e-06, | |
| "loss": 0.6751, | |
| "step": 131250 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_accuracy": 0.5731, | |
| "eval_loss": 0.6734435558319092, | |
| "eval_runtime": 23.3345, | |
| "eval_samples_per_second": 428.549, | |
| "eval_steps_per_second": 6.728, | |
| "step": 131250 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 8.667271111111112e-06, | |
| "loss": 0.6749, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.5714, | |
| "eval_loss": 0.6720886826515198, | |
| "eval_runtime": 23.4003, | |
| "eval_samples_per_second": 427.345, | |
| "eval_steps_per_second": 6.709, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 8.500666666666666e-06, | |
| "loss": 0.6746, | |
| "step": 168750 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_accuracy": 0.5767, | |
| "eval_loss": 0.6722772717475891, | |
| "eval_runtime": 23.5061, | |
| "eval_samples_per_second": 425.422, | |
| "eval_steps_per_second": 6.679, | |
| "step": 168750 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 8.334071111111112e-06, | |
| "loss": 0.6745, | |
| "step": 187500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.5756, | |
| "eval_loss": 0.6716436147689819, | |
| "eval_runtime": 23.4294, | |
| "eval_samples_per_second": 426.814, | |
| "eval_steps_per_second": 6.701, | |
| "step": 187500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 8.167475555555557e-06, | |
| "loss": 0.6745, | |
| "step": 206250 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_accuracy": 0.5785, | |
| "eval_loss": 0.6716175675392151, | |
| "eval_runtime": 23.3298, | |
| "eval_samples_per_second": 428.636, | |
| "eval_steps_per_second": 6.73, | |
| "step": 206250 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.00088e-06, | |
| "loss": 0.6744, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.5721, | |
| "eval_loss": 0.6720548272132874, | |
| "eval_runtime": 23.2772, | |
| "eval_samples_per_second": 429.606, | |
| "eval_steps_per_second": 6.745, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 7.834284444444445e-06, | |
| "loss": 0.674, | |
| "step": 243750 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.5742, | |
| "eval_loss": 0.6714185476303101, | |
| "eval_runtime": 23.567, | |
| "eval_samples_per_second": 424.321, | |
| "eval_steps_per_second": 6.662, | |
| "step": 243750 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 7.66768888888889e-06, | |
| "loss": 0.6743, | |
| "step": 262500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_accuracy": 0.5699, | |
| "eval_loss": 0.6723877191543579, | |
| "eval_runtime": 23.1824, | |
| "eval_samples_per_second": 431.361, | |
| "eval_steps_per_second": 6.772, | |
| "step": 262500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 7.501084444444445e-06, | |
| "loss": 0.6737, | |
| "step": 281250 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_accuracy": 0.5775, | |
| "eval_loss": 0.6708235144615173, | |
| "eval_runtime": 23.3586, | |
| "eval_samples_per_second": 428.108, | |
| "eval_steps_per_second": 6.721, | |
| "step": 281250 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 7.33448888888889e-06, | |
| "loss": 0.6736, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.5749, | |
| "eval_loss": 0.6718671917915344, | |
| "eval_runtime": 23.412, | |
| "eval_samples_per_second": 427.132, | |
| "eval_steps_per_second": 6.706, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.167893333333334e-06, | |
| "loss": 0.6737, | |
| "step": 318750 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_accuracy": 0.5756, | |
| "eval_loss": 0.6709334254264832, | |
| "eval_runtime": 23.5575, | |
| "eval_samples_per_second": 424.493, | |
| "eval_steps_per_second": 6.665, | |
| "step": 318750 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 7.001297777777778e-06, | |
| "loss": 0.6736, | |
| "step": 337500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.5725, | |
| "eval_loss": 0.6705034375190735, | |
| "eval_runtime": 23.4245, | |
| "eval_samples_per_second": 426.903, | |
| "eval_steps_per_second": 6.702, | |
| "step": 337500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 6.834711111111112e-06, | |
| "loss": 0.6736, | |
| "step": 356250 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.5734, | |
| "eval_loss": 0.6707730293273926, | |
| "eval_runtime": 23.4532, | |
| "eval_samples_per_second": 426.381, | |
| "eval_steps_per_second": 6.694, | |
| "step": 356250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.6681155555555566e-06, | |
| "loss": 0.6731, | |
| "step": 375000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5765, | |
| "eval_loss": 0.6712862253189087, | |
| "eval_runtime": 23.3551, | |
| "eval_samples_per_second": 428.173, | |
| "eval_steps_per_second": 6.722, | |
| "step": 375000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 6.50152e-06, | |
| "loss": 0.6704, | |
| "step": 393750 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_accuracy": 0.5772, | |
| "eval_loss": 0.6697613000869751, | |
| "eval_runtime": 23.5543, | |
| "eval_samples_per_second": 424.551, | |
| "eval_steps_per_second": 6.665, | |
| "step": 393750 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 6.334924444444444e-06, | |
| "loss": 0.6703, | |
| "step": 412500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.5747, | |
| "eval_loss": 0.6700881719589233, | |
| "eval_runtime": 23.557, | |
| "eval_samples_per_second": 424.502, | |
| "eval_steps_per_second": 6.665, | |
| "step": 412500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 6.1683288888888896e-06, | |
| "loss": 0.6703, | |
| "step": 431250 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.578, | |
| "eval_loss": 0.669475257396698, | |
| "eval_runtime": 23.4497, | |
| "eval_samples_per_second": 426.445, | |
| "eval_steps_per_second": 6.695, | |
| "step": 431250 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 6.001733333333334e-06, | |
| "loss": 0.6705, | |
| "step": 450000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 0.577, | |
| "eval_loss": 0.6688240170478821, | |
| "eval_runtime": 23.5829, | |
| "eval_samples_per_second": 424.035, | |
| "eval_steps_per_second": 6.657, | |
| "step": 450000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 5.8351466666666665e-06, | |
| "loss": 0.6703, | |
| "step": 468750 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_accuracy": 0.5795, | |
| "eval_loss": 0.6692911982536316, | |
| "eval_runtime": 23.4227, | |
| "eval_samples_per_second": 426.936, | |
| "eval_steps_per_second": 6.703, | |
| "step": 468750 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 5.668551111111112e-06, | |
| "loss": 0.6704, | |
| "step": 487500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.5783, | |
| "eval_loss": 0.6689985394477844, | |
| "eval_runtime": 23.436, | |
| "eval_samples_per_second": 426.694, | |
| "eval_steps_per_second": 6.699, | |
| "step": 487500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 5.501955555555556e-06, | |
| "loss": 0.6703, | |
| "step": 506250 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_accuracy": 0.5783, | |
| "eval_loss": 0.6690422296524048, | |
| "eval_runtime": 23.5352, | |
| "eval_samples_per_second": 424.896, | |
| "eval_steps_per_second": 6.671, | |
| "step": 506250 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 5.33536e-06, | |
| "loss": 0.6705, | |
| "step": 525000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_accuracy": 0.5785, | |
| "eval_loss": 0.6693896055221558, | |
| "eval_runtime": 23.5799, | |
| "eval_samples_per_second": 424.09, | |
| "eval_steps_per_second": 6.658, | |
| "step": 525000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 5.168782222222223e-06, | |
| "loss": 0.6705, | |
| "step": 543750 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_accuracy": 0.5806, | |
| "eval_loss": 0.6683958768844604, | |
| "eval_runtime": 23.4942, | |
| "eval_samples_per_second": 425.637, | |
| "eval_steps_per_second": 6.683, | |
| "step": 543750 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 5.002186666666667e-06, | |
| "loss": 0.6704, | |
| "step": 562500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_accuracy": 0.5792, | |
| "eval_loss": 0.6692180633544922, | |
| "eval_runtime": 23.4522, | |
| "eval_samples_per_second": 426.4, | |
| "eval_steps_per_second": 6.694, | |
| "step": 562500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 4.8356e-06, | |
| "loss": 0.6706, | |
| "step": 581250 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_accuracy": 0.5787, | |
| "eval_loss": 0.6688229441642761, | |
| "eval_runtime": 23.3586, | |
| "eval_samples_per_second": 428.109, | |
| "eval_steps_per_second": 6.721, | |
| "step": 581250 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4.669004444444445e-06, | |
| "loss": 0.6706, | |
| "step": 600000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_accuracy": 0.5814, | |
| "eval_loss": 0.6699367165565491, | |
| "eval_runtime": 23.5906, | |
| "eval_samples_per_second": 423.898, | |
| "eval_steps_per_second": 6.655, | |
| "step": 600000 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 4.5024177777777786e-06, | |
| "loss": 0.6702, | |
| "step": 618750 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_accuracy": 0.5766, | |
| "eval_loss": 0.669127881526947, | |
| "eval_runtime": 23.4357, | |
| "eval_samples_per_second": 426.7, | |
| "eval_steps_per_second": 6.699, | |
| "step": 618750 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 4.335804444444445e-06, | |
| "loss": 0.6702, | |
| "step": 637500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_accuracy": 0.5782, | |
| "eval_loss": 0.6684728264808655, | |
| "eval_runtime": 23.4934, | |
| "eval_samples_per_second": 425.651, | |
| "eval_steps_per_second": 6.683, | |
| "step": 637500 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 4.169217777777778e-06, | |
| "loss": 0.67, | |
| "step": 656250 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_accuracy": 0.5817, | |
| "eval_loss": 0.6683481931686401, | |
| "eval_runtime": 23.6396, | |
| "eval_samples_per_second": 423.02, | |
| "eval_steps_per_second": 6.641, | |
| "step": 656250 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 4.002613333333334e-06, | |
| "loss": 0.6697, | |
| "step": 675000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_accuracy": 0.582, | |
| "eval_loss": 0.667320966720581, | |
| "eval_runtime": 23.7429, | |
| "eval_samples_per_second": 421.178, | |
| "eval_steps_per_second": 6.612, | |
| "step": 675000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 3.836026666666667e-06, | |
| "loss": 0.6703, | |
| "step": 693750 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.581, | |
| "eval_loss": 0.6678736209869385, | |
| "eval_runtime": 23.3426, | |
| "eval_samples_per_second": 428.401, | |
| "eval_steps_per_second": 6.726, | |
| "step": 693750 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 3.6694400000000002e-06, | |
| "loss": 0.6697, | |
| "step": 712500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.5854, | |
| "eval_loss": 0.6670705080032349, | |
| "eval_runtime": 23.4267, | |
| "eval_samples_per_second": 426.862, | |
| "eval_steps_per_second": 6.702, | |
| "step": 712500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 3.502835555555556e-06, | |
| "loss": 0.6697, | |
| "step": 731250 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_accuracy": 0.5837, | |
| "eval_loss": 0.6675453186035156, | |
| "eval_runtime": 23.4269, | |
| "eval_samples_per_second": 426.859, | |
| "eval_steps_per_second": 6.702, | |
| "step": 731250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.33624e-06, | |
| "loss": 0.6702, | |
| "step": 750000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5857, | |
| "eval_loss": 0.6678363084793091, | |
| "eval_runtime": 23.4622, | |
| "eval_samples_per_second": 426.218, | |
| "eval_steps_per_second": 6.692, | |
| "step": 750000 | |
| } | |
| ], | |
| "max_steps": 1125000, | |
| "num_train_epochs": 3, | |
| "total_flos": 4.4577934226915066e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |