| { | |
| "best_global_step": 1600, | |
| "best_metric": 62.760154724121094, | |
| "best_model_checkpoint": "rtdetr_voucher_model/checkpoint-1600", | |
| "epoch": 80.0, | |
| "eval_steps": 200, | |
| "global_step": 1680, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 23553.73828125, | |
| "learning_rate": 7.8e-07, | |
| "loss": 1390.8067, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 7427.158203125, | |
| "learning_rate": 1.5800000000000001e-06, | |
| "loss": 1273.2542, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 6636.22314453125, | |
| "learning_rate": 2.38e-06, | |
| "loss": 1059.9522, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 7.619047619047619, | |
| "grad_norm": 1760.4281005859375, | |
| "learning_rate": 3.1800000000000005e-06, | |
| "loss": 790.0425, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 9.523809523809524, | |
| "grad_norm": 1625.7744140625, | |
| "learning_rate": 3.980000000000001e-06, | |
| "loss": 517.1978, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.523809523809524, | |
| "eval_loss": 535.0709838867188, | |
| "eval_runtime": 8.3595, | |
| "eval_samples_per_second": 13.876, | |
| "eval_steps_per_second": 0.478, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 1354.9310302734375, | |
| "learning_rate": 4.78e-06, | |
| "loss": 353.8703, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 873.0985107421875, | |
| "learning_rate": 5.580000000000001e-06, | |
| "loss": 262.4233, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 15.238095238095237, | |
| "grad_norm": 1516.567626953125, | |
| "learning_rate": 6.380000000000001e-06, | |
| "loss": 205.628, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 5139.3779296875, | |
| "learning_rate": 7.180000000000001e-06, | |
| "loss": 174.0845, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 19.047619047619047, | |
| "grad_norm": 480.6575622558594, | |
| "learning_rate": 7.980000000000002e-06, | |
| "loss": 142.4243, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.047619047619047, | |
| "eval_loss": 214.48715209960938, | |
| "eval_runtime": 7.9717, | |
| "eval_samples_per_second": 14.551, | |
| "eval_steps_per_second": 0.502, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 20.952380952380953, | |
| "grad_norm": 366.5032958984375, | |
| "learning_rate": 8.78e-06, | |
| "loss": 115.4496, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 435.5931701660156, | |
| "learning_rate": 9.58e-06, | |
| "loss": 92.9752, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 24.761904761904763, | |
| "grad_norm": 232.87234497070312, | |
| "learning_rate": 9.993604266900276e-06, | |
| "loss": 78.7805, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 26.666666666666668, | |
| "grad_norm": 188.52627563476562, | |
| "learning_rate": 9.938441702975689e-06, | |
| "loss": 59.0797, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 177.4958038330078, | |
| "learning_rate": 9.827324621255206e-06, | |
| "loss": 46.9344, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_loss": 129.46548461914062, | |
| "eval_runtime": 7.2365, | |
| "eval_samples_per_second": 16.03, | |
| "eval_steps_per_second": 0.553, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 30.476190476190474, | |
| "grad_norm": 135.09072875976562, | |
| "learning_rate": 9.661512022674852e-06, | |
| "loss": 39.1039, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 32.38095238095238, | |
| "grad_norm": 128.56063842773438, | |
| "learning_rate": 9.4428826302319e-06, | |
| "loss": 32.5044, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 99.94125366210938, | |
| "learning_rate": 9.173913602302594e-06, | |
| "loss": 27.9798, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 36.19047619047619, | |
| "grad_norm": 128.35171508789062, | |
| "learning_rate": 8.857652465447947e-06, | |
| "loss": 24.9475, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 38.095238095238095, | |
| "grad_norm": 113.99468231201172, | |
| "learning_rate": 8.497682584720065e-06, | |
| "loss": 22.38, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 38.095238095238095, | |
| "eval_loss": 75.64146423339844, | |
| "eval_runtime": 7.294, | |
| "eval_samples_per_second": 15.904, | |
| "eval_steps_per_second": 0.548, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 190.8120574951172, | |
| "learning_rate": 8.098082562704253e-06, | |
| "loss": 19.868, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 41.904761904761905, | |
| "grad_norm": 58.219482421875, | |
| "learning_rate": 7.663380027322022e-06, | |
| "loss": 22.0343, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 43.80952380952381, | |
| "grad_norm": 89.1216049194336, | |
| "learning_rate": 7.1985003319975696e-06, | |
| "loss": 16.9501, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 147.2601776123047, | |
| "learning_rate": 6.708710749435427e-06, | |
| "loss": 15.9704, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 47.61904761904762, | |
| "grad_norm": 103.56001281738281, | |
| "learning_rate": 6.199560791316009e-06, | |
| "loss": 15.38, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 47.61904761904762, | |
| "eval_loss": 79.28771209716797, | |
| "eval_runtime": 7.3262, | |
| "eval_samples_per_second": 15.834, | |
| "eval_steps_per_second": 0.546, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 49.523809523809526, | |
| "grad_norm": 49.25249099731445, | |
| "learning_rate": 5.676819330110808e-06, | |
| "loss": 14.5406, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 116.98213195800781, | |
| "learning_rate": 5.14640923545217e-06, | |
| "loss": 14.0173, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 53.333333333333336, | |
| "grad_norm": 72.93340301513672, | |
| "learning_rate": 4.614340265653695e-06, | |
| "loss": 13.7052, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 55.23809523809524, | |
| "grad_norm": 105.87384796142578, | |
| "learning_rate": 4.086640974747176e-06, | |
| "loss": 15.4002, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 59.23909378051758, | |
| "learning_rate": 3.569290406556519e-06, | |
| "loss": 14.1171, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_loss": 64.76508331298828, | |
| "eval_runtime": 7.1521, | |
| "eval_samples_per_second": 16.219, | |
| "eval_steps_per_second": 0.559, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 59.04761904761905, | |
| "grad_norm": 98.89187622070312, | |
| "learning_rate": 3.068150349742125e-06, | |
| "loss": 12.3597, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 60.95238095238095, | |
| "grad_norm": 179.5811004638672, | |
| "learning_rate": 2.5888989213931047e-06, | |
| "loss": 12.125, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 62.857142857142854, | |
| "grad_norm": 41.74519729614258, | |
| "learning_rate": 2.1369662316918095e-06, | |
| "loss": 11.7567, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 64.76190476190476, | |
| "grad_norm": 60.97040557861328, | |
| "learning_rate": 1.717472858595694e-06, | |
| "loss": 12.3442, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 66.66666666666667, | |
| "grad_norm": 197.75689697265625, | |
| "learning_rate": 1.335171829642984e-06, | |
| "loss": 11.2475, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 66.66666666666667, | |
| "eval_loss": 67.15937805175781, | |
| "eval_runtime": 7.2137, | |
| "eval_samples_per_second": 16.08, | |
| "eval_steps_per_second": 0.554, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 68.57142857142857, | |
| "grad_norm": 44.10291290283203, | |
| "learning_rate": 9.943947682514538e-07, | |
| "loss": 11.0752, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 70.47619047619048, | |
| "grad_norm": 54.24775314331055, | |
| "learning_rate": 6.990028146942752e-07, | |
| "loss": 10.9762, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 72.38095238095238, | |
| "grad_norm": 41.610679626464844, | |
| "learning_rate": 4.523428778378852e-07, | |
| "loss": 10.9561, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 74.28571428571429, | |
| "grad_norm": 109.53748321533203, | |
| "learning_rate": 2.5720971332717783e-07, | |
| "loss": 10.7585, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 76.19047619047619, | |
| "grad_norm": 49.795623779296875, | |
| "learning_rate": 1.158142578873156e-07, | |
| "loss": 11.5847, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 76.19047619047619, | |
| "eval_loss": 62.760154724121094, | |
| "eval_runtime": 7.2145, | |
| "eval_samples_per_second": 16.079, | |
| "eval_steps_per_second": 0.554, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 78.0952380952381, | |
| "grad_norm": 58.66199493408203, | |
| "learning_rate": 2.9758578527189776e-08, | |
| "loss": 11.5135, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 53.418094635009766, | |
| "learning_rate": 1.7720480054994606e-11, | |
| "loss": 10.746, | |
| "step": 1680 | |
| } | |
| ], | |
| "logging_steps": 40, | |
| "max_steps": 1680, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 80, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.965158266419806e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |