| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 1100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 4.545454545454545e-07, | |
| "loss": 2.5611, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 2.5692, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 2.644, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.304738759994507, | |
| "eval_runtime": 1.217, | |
| "eval_samples_per_second": 18.899, | |
| "eval_steps_per_second": 4.93, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 2.3827, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 2.0781, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 1.9548, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.9419394731521606, | |
| "eval_runtime": 1.2171, | |
| "eval_samples_per_second": 18.898, | |
| "eval_steps_per_second": 4.93, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 4.9987413559579636e-05, | |
| "loss": 1.8022, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.988679806432712e-05, | |
| "loss": 1.8295, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 4.968597221690986e-05, | |
| "loss": 1.788, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.9134601354599, | |
| "eval_runtime": 1.2132, | |
| "eval_samples_per_second": 18.958, | |
| "eval_steps_per_second": 4.946, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 4.938574467213518e-05, | |
| "loss": 1.6784, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 4.898732434036244e-05, | |
| "loss": 1.6528, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.849231551964771e-05, | |
| "loss": 1.6342, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.9498955011367798, | |
| "eval_runtime": 1.2502, | |
| "eval_samples_per_second": 18.397, | |
| "eval_steps_per_second": 4.799, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 4.790271143580174e-05, | |
| "loss": 1.3818, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 4.722088621637309e-05, | |
| "loss": 1.3781, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.1321451663970947, | |
| "eval_runtime": 1.2118, | |
| "eval_samples_per_second": 18.98, | |
| "eval_steps_per_second": 4.951, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 4.644958533087443e-05, | |
| "loss": 1.2512, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 4.559191453574582e-05, | |
| "loss": 1.0475, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "learning_rate": 4.465132736856969e-05, | |
| "loss": 1.0617, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.3518364429473877, | |
| "eval_runtime": 1.216, | |
| "eval_samples_per_second": 18.914, | |
| "eval_steps_per_second": 4.934, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 4.3631611241893874e-05, | |
| "loss": 0.9003, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 4.2536872192658036e-05, | |
| "loss": 0.7805, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 4.137151834863213e-05, | |
| "loss": 0.8104, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.609004259109497, | |
| "eval_runtime": 1.2167, | |
| "eval_samples_per_second": 18.904, | |
| "eval_steps_per_second": 4.932, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 4.014024217844167e-05, | |
| "loss": 0.6542, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 3.884800159665276e-05, | |
| "loss": 0.5753, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.5864, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.8889544010162354, | |
| "eval_runtime": 1.2434, | |
| "eval_samples_per_second": 18.498, | |
| "eval_steps_per_second": 4.825, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "learning_rate": 3.610166531514436e-05, | |
| "loss": 0.4181, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 8.73, | |
| "learning_rate": 3.465862814232822e-05, | |
| "loss": 0.4159, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 3.1356287002563477, | |
| "eval_runtime": 1.2192, | |
| "eval_samples_per_second": 18.865, | |
| "eval_steps_per_second": 4.921, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 3.3176699082935545e-05, | |
| "loss": 0.4188, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 3.166184534225087e-05, | |
| "loss": 0.3131, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "learning_rate": 3.012016670162977e-05, | |
| "loss": 0.3344, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 3.3189520835876465, | |
| "eval_runtime": 1.2131, | |
| "eval_samples_per_second": 18.96, | |
| "eval_steps_per_second": 4.946, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 10.18, | |
| "learning_rate": 2.8557870956832132e-05, | |
| "loss": 0.3005, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 2.698124892141971e-05, | |
| "loss": 0.2527, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "learning_rate": 2.5396649095870202e-05, | |
| "loss": 0.2446, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 3.5470495223999023, | |
| "eval_runtime": 1.216, | |
| "eval_samples_per_second": 18.914, | |
| "eval_steps_per_second": 4.934, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 11.27, | |
| "learning_rate": 2.3810452104406444e-05, | |
| "loss": 0.1855, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 11.64, | |
| "learning_rate": 2.222904500247473e-05, | |
| "loss": 0.1745, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.199, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 3.68398118019104, | |
| "eval_runtime": 1.2409, | |
| "eval_samples_per_second": 18.535, | |
| "eval_steps_per_second": 4.835, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 12.36, | |
| "learning_rate": 1.9106026612264316e-05, | |
| "loss": 0.1455, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "learning_rate": 1.7576990616793137e-05, | |
| "loss": 0.1245, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 3.765277147293091, | |
| "eval_runtime": 1.2162, | |
| "eval_samples_per_second": 18.911, | |
| "eval_steps_per_second": 4.933, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 13.09, | |
| "learning_rate": 1.6077844460203206e-05, | |
| "loss": 0.1351, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 1.4614624674952842e-05, | |
| "loss": 0.0967, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "learning_rate": 1.3193223130682936e-05, | |
| "loss": 0.1208, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 3.8721702098846436, | |
| "eval_runtime": 1.2167, | |
| "eval_samples_per_second": 18.903, | |
| "eval_steps_per_second": 4.931, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "learning_rate": 1.181936330973744e-05, | |
| "loss": 0.0853, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 1.049857726072005e-05, | |
| "loss": 0.0854, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "learning_rate": 9.236183322886945e-06, | |
| "loss": 0.1003, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 3.9574601650238037, | |
| "eval_runtime": 1.2205, | |
| "eval_samples_per_second": 18.844, | |
| "eval_steps_per_second": 4.916, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 15.27, | |
| "learning_rate": 8.0372647110717e-06, | |
| "loss": 0.0753, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 15.64, | |
| "learning_rate": 6.906649047373246e-06, | |
| "loss": 0.0928, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 5.848888922025553e-06, | |
| "loss": 0.0767, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 3.9671382904052734, | |
| "eval_runtime": 1.2561, | |
| "eval_samples_per_second": 18.311, | |
| "eval_steps_per_second": 4.777, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 16.36, | |
| "learning_rate": 4.868243561723535e-06, | |
| "loss": 0.0702, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 16.73, | |
| "learning_rate": 3.968661679220468e-06, | |
| "loss": 0.0913, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 3.9921278953552246, | |
| "eval_runtime": 1.2166, | |
| "eval_samples_per_second": 18.905, | |
| "eval_steps_per_second": 4.932, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 17.09, | |
| "learning_rate": 3.1537655732553768e-06, | |
| "loss": 0.0698, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "learning_rate": 2.4268365428344736e-06, | |
| "loss": 0.0661, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 17.82, | |
| "learning_rate": 1.790801674598186e-06, | |
| "loss": 0.0895, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 3.9939558506011963, | |
| "eval_runtime": 1.2161, | |
| "eval_samples_per_second": 18.913, | |
| "eval_steps_per_second": 4.934, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 1.248222056476367e-06, | |
| "loss": 0.0695, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 8.012824650910938e-07, | |
| "loss": 0.086, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 18.91, | |
| "learning_rate": 4.517825684323324e-07, | |
| "loss": 0.0671, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 3.9915316104888916, | |
| "eval_runtime": 1.2143, | |
| "eval_samples_per_second": 18.941, | |
| "eval_steps_per_second": 4.941, | |
| "step": 1045 | |
| }, | |
| { | |
| "epoch": 19.27, | |
| "learning_rate": 2.011296792301165e-07, | |
| "loss": 0.0681, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 19.64, | |
| "learning_rate": 5.033308820289184e-08, | |
| "loss": 0.09, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0671, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 3.990852117538452, | |
| "eval_runtime": 1.2506, | |
| "eval_samples_per_second": 18.391, | |
| "eval_steps_per_second": 4.798, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 1100, | |
| "total_flos": 3.807078373542298e+16, | |
| "train_loss": 0.6849299516461113, | |
| "train_runtime": 917.6848, | |
| "train_samples_per_second": 4.729, | |
| "train_steps_per_second": 1.199 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 1100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 20, | |
| "total_flos": 3.807078373542298e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |