| { | |
| "best_metric": 0.675777792930603, | |
| "best_model_checkpoint": "trainer/checkpoint-372428", | |
| "epoch": 50.0, | |
| "eval_steps": 500, | |
| "global_step": 396200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2e-05, | |
| "loss": 2.2513, | |
| "step": 7924 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.2939835786819458, | |
| "eval_runtime": 37.6044, | |
| "eval_samples_per_second": 749.168, | |
| "eval_steps_per_second": 23.428, | |
| "step": 7924 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4e-05, | |
| "loss": 1.3752, | |
| "step": 15848 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1416432857513428, | |
| "eval_runtime": 37.8169, | |
| "eval_samples_per_second": 744.958, | |
| "eval_steps_per_second": 23.296, | |
| "step": 15848 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.9473684210526315e-05, | |
| "loss": 1.2436, | |
| "step": 23772 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.0676215887069702, | |
| "eval_runtime": 37.7184, | |
| "eval_samples_per_second": 746.904, | |
| "eval_steps_per_second": 23.357, | |
| "step": 23772 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.842105263157895e-05, | |
| "loss": 1.1458, | |
| "step": 31696 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.9978280067443848, | |
| "eval_runtime": 36.3705, | |
| "eval_samples_per_second": 774.584, | |
| "eval_steps_per_second": 24.223, | |
| "step": 31696 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.736842105263158e-05, | |
| "loss": 1.0841, | |
| "step": 39620 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.9508912563323975, | |
| "eval_runtime": 36.5102, | |
| "eval_samples_per_second": 771.62, | |
| "eval_steps_per_second": 24.13, | |
| "step": 39620 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.6315789473684214e-05, | |
| "loss": 1.0386, | |
| "step": 47544 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.9356514811515808, | |
| "eval_runtime": 36.323, | |
| "eval_samples_per_second": 775.598, | |
| "eval_steps_per_second": 24.255, | |
| "step": 47544 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.5263157894736846e-05, | |
| "loss": 0.9989, | |
| "step": 55468 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.9180737137794495, | |
| "eval_runtime": 36.4427, | |
| "eval_samples_per_second": 773.05, | |
| "eval_steps_per_second": 24.175, | |
| "step": 55468 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.421052631578947e-05, | |
| "loss": 0.9686, | |
| "step": 63392 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.8952454328536987, | |
| "eval_runtime": 36.1822, | |
| "eval_samples_per_second": 778.615, | |
| "eval_steps_per_second": 24.349, | |
| "step": 63392 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.3157894736842105e-05, | |
| "loss": 0.9426, | |
| "step": 71316 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.8876378536224365, | |
| "eval_runtime": 36.5182, | |
| "eval_samples_per_second": 771.451, | |
| "eval_steps_per_second": 24.125, | |
| "step": 71316 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4.210526315789474e-05, | |
| "loss": 0.9198, | |
| "step": 79240 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.8818822503089905, | |
| "eval_runtime": 36.3065, | |
| "eval_samples_per_second": 775.949, | |
| "eval_steps_per_second": 24.266, | |
| "step": 79240 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 4.105263157894737e-05, | |
| "loss": 0.9053, | |
| "step": 87164 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.8349147439002991, | |
| "eval_runtime": 36.3771, | |
| "eval_samples_per_second": 774.444, | |
| "eval_steps_per_second": 24.219, | |
| "step": 87164 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.8881, | |
| "step": 95088 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.8407663702964783, | |
| "eval_runtime": 36.4047, | |
| "eval_samples_per_second": 773.857, | |
| "eval_steps_per_second": 24.2, | |
| "step": 95088 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.894736842105263e-05, | |
| "loss": 0.8704, | |
| "step": 103012 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.8339666128158569, | |
| "eval_runtime": 36.314, | |
| "eval_samples_per_second": 775.79, | |
| "eval_steps_per_second": 24.261, | |
| "step": 103012 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.789473684210527e-05, | |
| "loss": 0.8533, | |
| "step": 110936 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.8264057636260986, | |
| "eval_runtime": 36.2648, | |
| "eval_samples_per_second": 776.841, | |
| "eval_steps_per_second": 24.294, | |
| "step": 110936 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 0.8418, | |
| "step": 118860 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.8100990653038025, | |
| "eval_runtime": 36.5073, | |
| "eval_samples_per_second": 771.682, | |
| "eval_steps_per_second": 24.132, | |
| "step": 118860 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 3.578947368421053e-05, | |
| "loss": 0.8307, | |
| "step": 126784 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.8106787800788879, | |
| "eval_runtime": 36.2767, | |
| "eval_samples_per_second": 776.586, | |
| "eval_steps_per_second": 24.286, | |
| "step": 126784 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 3.473684210526316e-05, | |
| "loss": 0.815, | |
| "step": 134708 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.7991083264350891, | |
| "eval_runtime": 36.2711, | |
| "eval_samples_per_second": 776.707, | |
| "eval_steps_per_second": 24.289, | |
| "step": 134708 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 3.368421052631579e-05, | |
| "loss": 0.8015, | |
| "step": 142632 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.7952774167060852, | |
| "eval_runtime": 37.5556, | |
| "eval_samples_per_second": 750.142, | |
| "eval_steps_per_second": 23.459, | |
| "step": 142632 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 3.2631578947368426e-05, | |
| "loss": 0.7894, | |
| "step": 150556 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.7720882296562195, | |
| "eval_runtime": 36.9729, | |
| "eval_samples_per_second": 761.964, | |
| "eval_steps_per_second": 23.828, | |
| "step": 150556 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 3.157894736842105e-05, | |
| "loss": 0.7789, | |
| "step": 158480 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.7802249789237976, | |
| "eval_runtime": 37.1298, | |
| "eval_samples_per_second": 758.744, | |
| "eval_steps_per_second": 23.728, | |
| "step": 158480 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 3.0526315789473684e-05, | |
| "loss": 0.7678, | |
| "step": 166404 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 0.7610885500907898, | |
| "eval_runtime": 36.2573, | |
| "eval_samples_per_second": 777.002, | |
| "eval_steps_per_second": 24.299, | |
| "step": 166404 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 2.9473684210526314e-05, | |
| "loss": 0.7534, | |
| "step": 174328 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.782088577747345, | |
| "eval_runtime": 37.4893, | |
| "eval_samples_per_second": 751.467, | |
| "eval_steps_per_second": 23.5, | |
| "step": 174328 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 2.842105263157895e-05, | |
| "loss": 0.7502, | |
| "step": 182252 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 0.7673630714416504, | |
| "eval_runtime": 36.6312, | |
| "eval_samples_per_second": 769.07, | |
| "eval_steps_per_second": 24.051, | |
| "step": 182252 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 2.7368421052631583e-05, | |
| "loss": 0.7345, | |
| "step": 190176 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.7627705335617065, | |
| "eval_runtime": 37.0149, | |
| "eval_samples_per_second": 761.099, | |
| "eval_steps_per_second": 23.801, | |
| "step": 190176 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.7264, | |
| "step": 198100 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.7561437487602234, | |
| "eval_runtime": 37.2677, | |
| "eval_samples_per_second": 755.937, | |
| "eval_steps_per_second": 23.64, | |
| "step": 198100 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 2.5263157894736845e-05, | |
| "loss": 0.7142, | |
| "step": 206024 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.7509896159172058, | |
| "eval_runtime": 37.3161, | |
| "eval_samples_per_second": 754.956, | |
| "eval_steps_per_second": 23.609, | |
| "step": 206024 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 2.4210526315789474e-05, | |
| "loss": 0.7018, | |
| "step": 213948 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 0.7464780807495117, | |
| "eval_runtime": 36.7538, | |
| "eval_samples_per_second": 766.505, | |
| "eval_steps_per_second": 23.97, | |
| "step": 213948 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 2.3157894736842107e-05, | |
| "loss": 0.6897, | |
| "step": 221872 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.7344001531600952, | |
| "eval_runtime": 37.0652, | |
| "eval_samples_per_second": 760.066, | |
| "eval_steps_per_second": 23.769, | |
| "step": 221872 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 2.2105263157894736e-05, | |
| "loss": 0.682, | |
| "step": 229796 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 0.7429642081260681, | |
| "eval_runtime": 37.2732, | |
| "eval_samples_per_second": 755.824, | |
| "eval_steps_per_second": 23.636, | |
| "step": 229796 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 2.105263157894737e-05, | |
| "loss": 0.6754, | |
| "step": 237720 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.7481978535652161, | |
| "eval_runtime": 37.0222, | |
| "eval_samples_per_second": 760.949, | |
| "eval_steps_per_second": 23.797, | |
| "step": 237720 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6679, | |
| "step": 245644 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_loss": 0.7224923968315125, | |
| "eval_runtime": 37.2439, | |
| "eval_samples_per_second": 756.419, | |
| "eval_steps_per_second": 23.655, | |
| "step": 245644 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 1.8947368421052634e-05, | |
| "loss": 0.6566, | |
| "step": 253568 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.715844452381134, | |
| "eval_runtime": 36.5457, | |
| "eval_samples_per_second": 770.871, | |
| "eval_steps_per_second": 24.107, | |
| "step": 253568 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "learning_rate": 1.7894736842105264e-05, | |
| "loss": 0.6492, | |
| "step": 261492 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_loss": 0.7234057188034058, | |
| "eval_runtime": 36.3822, | |
| "eval_samples_per_second": 774.335, | |
| "eval_steps_per_second": 24.215, | |
| "step": 261492 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 1.6842105263157896e-05, | |
| "loss": 0.642, | |
| "step": 269416 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 0.7132413983345032, | |
| "eval_runtime": 36.3793, | |
| "eval_samples_per_second": 774.396, | |
| "eval_steps_per_second": 24.217, | |
| "step": 269416 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "learning_rate": 1.5789473684210526e-05, | |
| "loss": 0.6342, | |
| "step": 277340 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_loss": 0.7007443904876709, | |
| "eval_runtime": 36.3032, | |
| "eval_samples_per_second": 776.02, | |
| "eval_steps_per_second": 24.268, | |
| "step": 277340 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 1.4736842105263157e-05, | |
| "loss": 0.6236, | |
| "step": 285264 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 0.69706791639328, | |
| "eval_runtime": 37.0755, | |
| "eval_samples_per_second": 759.854, | |
| "eval_steps_per_second": 23.762, | |
| "step": 285264 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "learning_rate": 1.3684210526315791e-05, | |
| "loss": 0.6146, | |
| "step": 293188 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_loss": 0.6900755167007446, | |
| "eval_runtime": 36.5007, | |
| "eval_samples_per_second": 771.822, | |
| "eval_steps_per_second": 24.137, | |
| "step": 293188 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "learning_rate": 1.2631578947368422e-05, | |
| "loss": 0.6087, | |
| "step": 301112 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 0.6962341666221619, | |
| "eval_runtime": 36.6631, | |
| "eval_samples_per_second": 768.402, | |
| "eval_steps_per_second": 24.03, | |
| "step": 301112 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "learning_rate": 1.1578947368421053e-05, | |
| "loss": 0.5989, | |
| "step": 309036 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_loss": 0.7045713067054749, | |
| "eval_runtime": 36.6758, | |
| "eval_samples_per_second": 768.136, | |
| "eval_steps_per_second": 24.021, | |
| "step": 309036 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.5924, | |
| "step": 316960 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 0.6984645128250122, | |
| "eval_runtime": 36.7394, | |
| "eval_samples_per_second": 766.807, | |
| "eval_steps_per_second": 23.98, | |
| "step": 316960 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "learning_rate": 9.473684210526317e-06, | |
| "loss": 0.5827, | |
| "step": 324884 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_loss": 0.6994604468345642, | |
| "eval_runtime": 36.8305, | |
| "eval_samples_per_second": 764.91, | |
| "eval_steps_per_second": 23.92, | |
| "step": 324884 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "learning_rate": 8.421052631578948e-06, | |
| "loss": 0.5731, | |
| "step": 332808 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 0.6827645301818848, | |
| "eval_runtime": 36.7809, | |
| "eval_samples_per_second": 765.941, | |
| "eval_steps_per_second": 23.953, | |
| "step": 332808 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "learning_rate": 7.3684210526315784e-06, | |
| "loss": 0.5718, | |
| "step": 340732 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_loss": 0.7020975947380066, | |
| "eval_runtime": 36.6627, | |
| "eval_samples_per_second": 768.411, | |
| "eval_steps_per_second": 24.03, | |
| "step": 340732 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "learning_rate": 6.315789473684211e-06, | |
| "loss": 0.5663, | |
| "step": 348656 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_loss": 0.6774910092353821, | |
| "eval_runtime": 36.8941, | |
| "eval_samples_per_second": 763.59, | |
| "eval_steps_per_second": 23.879, | |
| "step": 348656 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "learning_rate": 5.263157894736842e-06, | |
| "loss": 0.5575, | |
| "step": 356580 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_loss": 0.683965265750885, | |
| "eval_runtime": 36.8405, | |
| "eval_samples_per_second": 764.703, | |
| "eval_steps_per_second": 23.914, | |
| "step": 356580 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "learning_rate": 4.210526315789474e-06, | |
| "loss": 0.5524, | |
| "step": 364504 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_loss": 0.6812577247619629, | |
| "eval_runtime": 36.3196, | |
| "eval_samples_per_second": 775.669, | |
| "eval_steps_per_second": 24.257, | |
| "step": 364504 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "learning_rate": 3.1578947368421056e-06, | |
| "loss": 0.5499, | |
| "step": 372428 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_loss": 0.675777792930603, | |
| "eval_runtime": 36.282, | |
| "eval_samples_per_second": 776.473, | |
| "eval_steps_per_second": 24.282, | |
| "step": 372428 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "learning_rate": 2.105263157894737e-06, | |
| "loss": 0.541, | |
| "step": 380352 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 0.678913950920105, | |
| "eval_runtime": 36.4409, | |
| "eval_samples_per_second": 773.087, | |
| "eval_steps_per_second": 24.176, | |
| "step": 380352 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "learning_rate": 1.0526315789473685e-06, | |
| "loss": 0.5372, | |
| "step": 388276 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_loss": 0.6796761155128479, | |
| "eval_runtime": 36.5663, | |
| "eval_samples_per_second": 770.435, | |
| "eval_steps_per_second": 24.093, | |
| "step": 388276 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.535, | |
| "step": 396200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 0.6791965365409851, | |
| "eval_runtime": 36.555, | |
| "eval_samples_per_second": 770.674, | |
| "eval_steps_per_second": 24.101, | |
| "step": 396200 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 396200, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 7.079826371258392e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |