| { | |
| "best_metric": 2.206695079803467, | |
| "best_model_checkpoint": "detr-r50-finetuned-mist1-gb-4ah-6l/checkpoint-5060", | |
| "epoch": 50.0, | |
| "eval_steps": 500, | |
| "global_step": 5750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.812173913043479e-06, | |
| "loss": 3.4044, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.004835844039917, | |
| "eval_runtime": 5.8385, | |
| "eval_samples_per_second": 6.851, | |
| "eval_steps_per_second": 0.856, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.612173913043479e-06, | |
| "loss": 3.1708, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.902806520462036, | |
| "eval_runtime": 5.7867, | |
| "eval_samples_per_second": 6.912, | |
| "eval_steps_per_second": 0.864, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.412173913043479e-06, | |
| "loss": 3.0756, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.853818416595459, | |
| "eval_runtime": 5.8759, | |
| "eval_samples_per_second": 6.808, | |
| "eval_steps_per_second": 0.851, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 9.21217391304348e-06, | |
| "loss": 2.9769, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.814894914627075, | |
| "eval_runtime": 5.8445, | |
| "eval_samples_per_second": 6.844, | |
| "eval_steps_per_second": 0.855, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 9.01217391304348e-06, | |
| "loss": 2.8999, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.7331762313842773, | |
| "eval_runtime": 5.9217, | |
| "eval_samples_per_second": 6.755, | |
| "eval_steps_per_second": 0.844, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 8.81217391304348e-06, | |
| "loss": 2.8609, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.7212414741516113, | |
| "eval_runtime": 5.879, | |
| "eval_samples_per_second": 6.804, | |
| "eval_steps_per_second": 0.85, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 8.613913043478262e-06, | |
| "loss": 2.8338, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.689382553100586, | |
| "eval_runtime": 5.8299, | |
| "eval_samples_per_second": 6.861, | |
| "eval_steps_per_second": 0.858, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 8.413913043478262e-06, | |
| "loss": 2.8103, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.704505681991577, | |
| "eval_runtime": 5.7873, | |
| "eval_samples_per_second": 6.912, | |
| "eval_steps_per_second": 0.864, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 8.21391304347826e-06, | |
| "loss": 2.8036, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.778614044189453, | |
| "eval_runtime": 5.7444, | |
| "eval_samples_per_second": 6.963, | |
| "eval_steps_per_second": 0.87, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 8.013913043478262e-06, | |
| "loss": 2.7486, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.688080310821533, | |
| "eval_runtime": 5.7616, | |
| "eval_samples_per_second": 6.943, | |
| "eval_steps_per_second": 0.868, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 7.813913043478263e-06, | |
| "loss": 2.7076, | |
| "step": 1265 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 2.605870485305786, | |
| "eval_runtime": 5.8415, | |
| "eval_samples_per_second": 6.848, | |
| "eval_steps_per_second": 0.856, | |
| "step": 1265 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 7.615652173913044e-06, | |
| "loss": 2.7156, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 2.648322343826294, | |
| "eval_runtime": 5.8314, | |
| "eval_samples_per_second": 6.859, | |
| "eval_steps_per_second": 0.857, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 7.415652173913044e-06, | |
| "loss": 2.6655, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 2.5438005924224854, | |
| "eval_runtime": 5.7667, | |
| "eval_samples_per_second": 6.936, | |
| "eval_steps_per_second": 0.867, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 7.215652173913043e-06, | |
| "loss": 2.6368, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 2.5342297554016113, | |
| "eval_runtime": 5.7673, | |
| "eval_samples_per_second": 6.936, | |
| "eval_steps_per_second": 0.867, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 7.015652173913044e-06, | |
| "loss": 2.5982, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 2.5287182331085205, | |
| "eval_runtime": 5.8321, | |
| "eval_samples_per_second": 6.859, | |
| "eval_steps_per_second": 0.857, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 6.815652173913045e-06, | |
| "loss": 2.6116, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 2.4446094036102295, | |
| "eval_runtime": 5.7529, | |
| "eval_samples_per_second": 6.953, | |
| "eval_steps_per_second": 0.869, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 6.615652173913044e-06, | |
| "loss": 2.5592, | |
| "step": 1955 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 2.4365103244781494, | |
| "eval_runtime": 5.7499, | |
| "eval_samples_per_second": 6.957, | |
| "eval_steps_per_second": 0.87, | |
| "step": 1955 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 6.415652173913044e-06, | |
| "loss": 2.5528, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 2.484395980834961, | |
| "eval_runtime": 5.7801, | |
| "eval_samples_per_second": 6.92, | |
| "eval_steps_per_second": 0.865, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 6.215652173913044e-06, | |
| "loss": 2.5248, | |
| "step": 2185 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 2.4194891452789307, | |
| "eval_runtime": 5.7656, | |
| "eval_samples_per_second": 6.938, | |
| "eval_steps_per_second": 0.867, | |
| "step": 2185 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 6.015652173913044e-06, | |
| "loss": 2.4853, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 2.453843832015991, | |
| "eval_runtime": 5.8034, | |
| "eval_samples_per_second": 6.893, | |
| "eval_steps_per_second": 0.862, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 5.815652173913045e-06, | |
| "loss": 2.5295, | |
| "step": 2415 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 2.569608211517334, | |
| "eval_runtime": 5.8097, | |
| "eval_samples_per_second": 6.885, | |
| "eval_steps_per_second": 0.861, | |
| "step": 2415 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 5.615652173913044e-06, | |
| "loss": 2.5069, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 2.4536538124084473, | |
| "eval_runtime": 5.7892, | |
| "eval_samples_per_second": 6.909, | |
| "eval_steps_per_second": 0.864, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 5.4156521739130445e-06, | |
| "loss": 2.4504, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 2.515150547027588, | |
| "eval_runtime": 5.8674, | |
| "eval_samples_per_second": 6.817, | |
| "eval_steps_per_second": 0.852, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 5.215652173913044e-06, | |
| "loss": 2.4447, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 2.4431958198547363, | |
| "eval_runtime": 5.9181, | |
| "eval_samples_per_second": 6.759, | |
| "eval_steps_per_second": 0.845, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 5.015652173913044e-06, | |
| "loss": 2.4303, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 2.4032533168792725, | |
| "eval_runtime": 5.9198, | |
| "eval_samples_per_second": 6.757, | |
| "eval_steps_per_second": 0.845, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 4.815652173913044e-06, | |
| "loss": 2.4137, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 2.3795552253723145, | |
| "eval_runtime": 5.8418, | |
| "eval_samples_per_second": 6.847, | |
| "eval_steps_per_second": 0.856, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 4.615652173913044e-06, | |
| "loss": 2.41, | |
| "step": 3105 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 2.3598742485046387, | |
| "eval_runtime": 5.8868, | |
| "eval_samples_per_second": 6.795, | |
| "eval_steps_per_second": 0.849, | |
| "step": 3105 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 4.415652173913044e-06, | |
| "loss": 2.3816, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 2.401759386062622, | |
| "eval_runtime": 5.8373, | |
| "eval_samples_per_second": 6.852, | |
| "eval_steps_per_second": 0.857, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 4.215652173913044e-06, | |
| "loss": 2.3752, | |
| "step": 3335 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 2.3115689754486084, | |
| "eval_runtime": 5.8962, | |
| "eval_samples_per_second": 6.784, | |
| "eval_steps_per_second": 0.848, | |
| "step": 3335 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 4.0156521739130435e-06, | |
| "loss": 2.3929, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 2.310469150543213, | |
| "eval_runtime": 5.8122, | |
| "eval_samples_per_second": 6.882, | |
| "eval_steps_per_second": 0.86, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "learning_rate": 3.815652173913044e-06, | |
| "loss": 2.3791, | |
| "step": 3565 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_loss": 2.367689371109009, | |
| "eval_runtime": 5.837, | |
| "eval_samples_per_second": 6.853, | |
| "eval_steps_per_second": 0.857, | |
| "step": 3565 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 3.615652173913044e-06, | |
| "loss": 2.3639, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 2.431199550628662, | |
| "eval_runtime": 5.8179, | |
| "eval_samples_per_second": 6.875, | |
| "eval_steps_per_second": 0.859, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "learning_rate": 3.4156521739130437e-06, | |
| "loss": 2.3475, | |
| "step": 3795 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_loss": 2.3051953315734863, | |
| "eval_runtime": 5.7888, | |
| "eval_samples_per_second": 6.91, | |
| "eval_steps_per_second": 0.864, | |
| "step": 3795 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 3.2156521739130435e-06, | |
| "loss": 2.3429, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 2.322197675704956, | |
| "eval_runtime": 5.7936, | |
| "eval_samples_per_second": 6.904, | |
| "eval_steps_per_second": 0.863, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "learning_rate": 3.015652173913044e-06, | |
| "loss": 2.3115, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_loss": 2.3126182556152344, | |
| "eval_runtime": 5.8416, | |
| "eval_samples_per_second": 6.847, | |
| "eval_steps_per_second": 0.856, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 2.815652173913044e-06, | |
| "loss": 2.3276, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 2.3154168128967285, | |
| "eval_runtime": 5.7988, | |
| "eval_samples_per_second": 6.898, | |
| "eval_steps_per_second": 0.862, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "learning_rate": 2.6156521739130438e-06, | |
| "loss": 2.3126, | |
| "step": 4255 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_loss": 2.353442430496216, | |
| "eval_runtime": 5.7487, | |
| "eval_samples_per_second": 6.958, | |
| "eval_steps_per_second": 0.87, | |
| "step": 4255 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "learning_rate": 2.417391304347826e-06, | |
| "loss": 2.2934, | |
| "step": 4370 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 2.2566468715667725, | |
| "eval_runtime": 5.902, | |
| "eval_samples_per_second": 6.777, | |
| "eval_steps_per_second": 0.847, | |
| "step": 4370 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "learning_rate": 2.2173913043478264e-06, | |
| "loss": 2.2901, | |
| "step": 4485 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_loss": 2.274752140045166, | |
| "eval_runtime": 5.8191, | |
| "eval_samples_per_second": 6.874, | |
| "eval_steps_per_second": 0.859, | |
| "step": 4485 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 2.017391304347826e-06, | |
| "loss": 2.2622, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 2.2620463371276855, | |
| "eval_runtime": 5.7456, | |
| "eval_samples_per_second": 6.962, | |
| "eval_steps_per_second": 0.87, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "learning_rate": 1.8173913043478262e-06, | |
| "loss": 2.2707, | |
| "step": 4715 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_loss": 2.2336184978485107, | |
| "eval_runtime": 5.8344, | |
| "eval_samples_per_second": 6.856, | |
| "eval_steps_per_second": 0.857, | |
| "step": 4715 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "learning_rate": 1.6173913043478262e-06, | |
| "loss": 2.2338, | |
| "step": 4830 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 2.224193811416626, | |
| "eval_runtime": 5.7896, | |
| "eval_samples_per_second": 6.909, | |
| "eval_steps_per_second": 0.864, | |
| "step": 4830 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "learning_rate": 1.4173913043478262e-06, | |
| "loss": 2.2457, | |
| "step": 4945 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_loss": 2.219238758087158, | |
| "eval_runtime": 5.8096, | |
| "eval_samples_per_second": 6.885, | |
| "eval_steps_per_second": 0.861, | |
| "step": 4945 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "learning_rate": 1.2173913043478262e-06, | |
| "loss": 2.227, | |
| "step": 5060 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_loss": 2.206695079803467, | |
| "eval_runtime": 5.7447, | |
| "eval_samples_per_second": 6.963, | |
| "eval_steps_per_second": 0.87, | |
| "step": 5060 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "learning_rate": 1.0173913043478262e-06, | |
| "loss": 2.2215, | |
| "step": 5175 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_loss": 2.2183449268341064, | |
| "eval_runtime": 5.7702, | |
| "eval_samples_per_second": 6.932, | |
| "eval_steps_per_second": 0.867, | |
| "step": 5175 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "learning_rate": 8.173913043478261e-07, | |
| "loss": 2.2075, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_loss": 2.2187769412994385, | |
| "eval_runtime": 5.7737, | |
| "eval_samples_per_second": 6.928, | |
| "eval_steps_per_second": 0.866, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "learning_rate": 6.173913043478262e-07, | |
| "loss": 2.2286, | |
| "step": 5405 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_loss": 2.2306272983551025, | |
| "eval_runtime": 5.7527, | |
| "eval_samples_per_second": 6.953, | |
| "eval_steps_per_second": 0.869, | |
| "step": 5405 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "learning_rate": 4.1739130434782616e-07, | |
| "loss": 2.2292, | |
| "step": 5520 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 2.2159781455993652, | |
| "eval_runtime": 5.8705, | |
| "eval_samples_per_second": 6.814, | |
| "eval_steps_per_second": 0.852, | |
| "step": 5520 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "learning_rate": 2.173913043478261e-07, | |
| "loss": 2.219, | |
| "step": 5635 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_loss": 2.2207822799682617, | |
| "eval_runtime": 5.7679, | |
| "eval_samples_per_second": 6.935, | |
| "eval_steps_per_second": 0.867, | |
| "step": 5635 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 1.739130434782609e-08, | |
| "loss": 2.2125, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 2.216578722000122, | |
| "eval_runtime": 5.8453, | |
| "eval_samples_per_second": 6.843, | |
| "eval_steps_per_second": 0.855, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 5750, | |
| "total_flos": 1.098949102848e+19, | |
| "train_loss": 2.5100708697775134, | |
| "train_runtime": 4619.9285, | |
| "train_samples_per_second": 4.978, | |
| "train_steps_per_second": 1.245 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5750, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1.098949102848e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |