| { | |
| "best_metric": 0.3955613076686859, | |
| "best_model_checkpoint": "./TransparentBagClassifier/checkpoint-580", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 580, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12195121951219512, | |
| "grad_norm": 1.9288002252578735, | |
| "learning_rate": 1.9512195121951222e-05, | |
| "loss": 0.639, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.24390243902439024, | |
| "grad_norm": 1.795539379119873, | |
| "learning_rate": 1.902439024390244e-05, | |
| "loss": 0.6215, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.36585365853658536, | |
| "grad_norm": 3.186365842819214, | |
| "learning_rate": 1.8536585365853663e-05, | |
| "loss": 0.5834, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4878048780487805, | |
| "grad_norm": 2.8957085609436035, | |
| "learning_rate": 1.804878048780488e-05, | |
| "loss": 0.608, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6097560975609756, | |
| "grad_norm": 2.6288256645202637, | |
| "learning_rate": 1.75609756097561e-05, | |
| "loss": 0.6492, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7317073170731707, | |
| "grad_norm": 2.7670204639434814, | |
| "learning_rate": 1.7073170731707317e-05, | |
| "loss": 0.6062, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8536585365853658, | |
| "grad_norm": 2.380261182785034, | |
| "learning_rate": 1.6585365853658537e-05, | |
| "loss": 0.5372, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.975609756097561, | |
| "grad_norm": 2.4240152835845947, | |
| "learning_rate": 1.6097560975609757e-05, | |
| "loss": 0.448, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7304347826086957, | |
| "eval_loss": 0.5725326538085938, | |
| "eval_runtime": 13.2774, | |
| "eval_samples_per_second": 8.661, | |
| "eval_steps_per_second": 1.13, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.0975609756097562, | |
| "grad_norm": 1.8054224252700806, | |
| "learning_rate": 1.5609756097560978e-05, | |
| "loss": 0.5777, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "grad_norm": 2.1294147968292236, | |
| "learning_rate": 1.5121951219512196e-05, | |
| "loss": 0.6131, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.3414634146341464, | |
| "grad_norm": 2.426273822784424, | |
| "learning_rate": 1.4634146341463415e-05, | |
| "loss": 0.5493, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.4634146341463414, | |
| "grad_norm": 2.6345632076263428, | |
| "learning_rate": 1.4146341463414635e-05, | |
| "loss": 0.5506, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.5853658536585367, | |
| "grad_norm": 2.231985569000244, | |
| "learning_rate": 1.3658536585365855e-05, | |
| "loss": 0.4764, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.7073170731707317, | |
| "grad_norm": 2.486055850982666, | |
| "learning_rate": 1.3170731707317076e-05, | |
| "loss": 0.3848, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.8292682926829267, | |
| "grad_norm": 4.828785419464111, | |
| "learning_rate": 1.2682926829268294e-05, | |
| "loss": 0.5178, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.951219512195122, | |
| "grad_norm": 2.4976108074188232, | |
| "learning_rate": 1.2195121951219513e-05, | |
| "loss": 0.5097, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7652173913043478, | |
| "eval_loss": 0.4945763945579529, | |
| "eval_runtime": 13.2493, | |
| "eval_samples_per_second": 8.68, | |
| "eval_steps_per_second": 1.132, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 2.073170731707317, | |
| "grad_norm": 3.168379068374634, | |
| "learning_rate": 1.1707317073170731e-05, | |
| "loss": 0.4253, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.1951219512195124, | |
| "grad_norm": 4.656283378601074, | |
| "learning_rate": 1.1219512195121953e-05, | |
| "loss": 0.5243, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.317073170731707, | |
| "grad_norm": 2.9091641902923584, | |
| "learning_rate": 1.0731707317073172e-05, | |
| "loss": 0.4482, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "grad_norm": 3.3256161212921143, | |
| "learning_rate": 1.024390243902439e-05, | |
| "loss": 0.6001, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.5609756097560976, | |
| "grad_norm": 4.491713047027588, | |
| "learning_rate": 9.756097560975611e-06, | |
| "loss": 0.4818, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.682926829268293, | |
| "grad_norm": 2.668192148208618, | |
| "learning_rate": 9.268292682926831e-06, | |
| "loss": 0.4202, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.8048780487804876, | |
| "grad_norm": 3.90390944480896, | |
| "learning_rate": 8.78048780487805e-06, | |
| "loss": 0.447, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.926829268292683, | |
| "grad_norm": 4.300001621246338, | |
| "learning_rate": 8.292682926829268e-06, | |
| "loss": 0.452, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7565217391304347, | |
| "eval_loss": 0.48406168818473816, | |
| "eval_runtime": 12.8922, | |
| "eval_samples_per_second": 8.92, | |
| "eval_steps_per_second": 1.163, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 3.048780487804878, | |
| "grad_norm": 1.386968970298767, | |
| "learning_rate": 7.804878048780489e-06, | |
| "loss": 0.4124, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.1707317073170733, | |
| "grad_norm": 3.373535633087158, | |
| "learning_rate": 7.317073170731707e-06, | |
| "loss": 0.4285, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.292682926829268, | |
| "grad_norm": 4.440028190612793, | |
| "learning_rate": 6.829268292682928e-06, | |
| "loss": 0.5181, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.4146341463414633, | |
| "grad_norm": 3.297926425933838, | |
| "learning_rate": 6.341463414634147e-06, | |
| "loss": 0.452, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.5365853658536586, | |
| "grad_norm": 1.8083200454711914, | |
| "learning_rate": 5.853658536585366e-06, | |
| "loss": 0.4684, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.658536585365854, | |
| "grad_norm": 2.8701059818267822, | |
| "learning_rate": 5.365853658536586e-06, | |
| "loss": 0.3982, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.7804878048780486, | |
| "grad_norm": 5.826732635498047, | |
| "learning_rate": 4.8780487804878055e-06, | |
| "loss": 0.4161, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.902439024390244, | |
| "grad_norm": 1.7296645641326904, | |
| "learning_rate": 4.390243902439025e-06, | |
| "loss": 0.3885, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7565217391304347, | |
| "eval_loss": 0.48118799924850464, | |
| "eval_runtime": 13.093, | |
| "eval_samples_per_second": 8.783, | |
| "eval_steps_per_second": 1.146, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 4.024390243902439, | |
| "grad_norm": 1.9118701219558716, | |
| "learning_rate": 3.902439024390244e-06, | |
| "loss": 0.4334, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.146341463414634, | |
| "grad_norm": 2.3522226810455322, | |
| "learning_rate": 3.414634146341464e-06, | |
| "loss": 0.4392, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.2682926829268295, | |
| "grad_norm": 2.4978580474853516, | |
| "learning_rate": 2.926829268292683e-06, | |
| "loss": 0.3711, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.390243902439025, | |
| "grad_norm": 7.186872959136963, | |
| "learning_rate": 2.4390243902439027e-06, | |
| "loss": 0.524, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.512195121951219, | |
| "grad_norm": 2.313222885131836, | |
| "learning_rate": 1.951219512195122e-06, | |
| "loss": 0.3826, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.634146341463414, | |
| "grad_norm": 4.187412261962891, | |
| "learning_rate": 1.4634146341463414e-06, | |
| "loss": 0.3764, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.7560975609756095, | |
| "grad_norm": 3.5783486366271973, | |
| "learning_rate": 9.75609756097561e-07, | |
| "loss": 0.414, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.878048780487805, | |
| "grad_norm": 6.643193244934082, | |
| "learning_rate": 4.878048780487805e-07, | |
| "loss": 0.4708, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.836371898651123, | |
| "learning_rate": 0.0, | |
| "loss": 0.4743, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7739130434782608, | |
| "eval_loss": 0.46259966492652893, | |
| "eval_runtime": 13.4253, | |
| "eval_samples_per_second": 8.566, | |
| "eval_steps_per_second": 1.117, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.6206896551724137, | |
| "grad_norm": 5.8970112800598145, | |
| "learning_rate": 5.517241379310345e-06, | |
| "loss": 0.4971, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.706896551724138, | |
| "grad_norm": 3.7783031463623047, | |
| "learning_rate": 5.172413793103449e-06, | |
| "loss": 0.5238, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.793103448275862, | |
| "grad_norm": 4.057668209075928, | |
| "learning_rate": 4.8275862068965525e-06, | |
| "loss": 0.476, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.8793103448275863, | |
| "grad_norm": 4.21135950088501, | |
| "learning_rate": 4.482758620689656e-06, | |
| "loss": 0.487, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.9655172413793105, | |
| "grad_norm": 6.149752616882324, | |
| "learning_rate": 4.137931034482759e-06, | |
| "loss": 0.4749, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7987804878048781, | |
| "eval_loss": 0.45718589425086975, | |
| "eval_runtime": 18.5139, | |
| "eval_samples_per_second": 8.858, | |
| "eval_steps_per_second": 1.134, | |
| "step": 464 | |
| }, | |
| { | |
| "epoch": 4.051724137931035, | |
| "grad_norm": 5.206491470336914, | |
| "learning_rate": 3.793103448275862e-06, | |
| "loss": 0.5889, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.137931034482759, | |
| "grad_norm": 4.442436218261719, | |
| "learning_rate": 3.448275862068966e-06, | |
| "loss": 0.5011, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.224137931034483, | |
| "grad_norm": 3.270550012588501, | |
| "learning_rate": 3.103448275862069e-06, | |
| "loss": 0.4479, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.310344827586207, | |
| "grad_norm": 2.839463710784912, | |
| "learning_rate": 2.7586206896551725e-06, | |
| "loss": 0.4054, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.396551724137931, | |
| "grad_norm": 3.1818559169769287, | |
| "learning_rate": 2.4137931034482762e-06, | |
| "loss": 0.3631, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.482758620689655, | |
| "grad_norm": 7.189451217651367, | |
| "learning_rate": 2.0689655172413796e-06, | |
| "loss": 0.4301, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.568965517241379, | |
| "grad_norm": 3.522141695022583, | |
| "learning_rate": 1.724137931034483e-06, | |
| "loss": 0.4977, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.655172413793103, | |
| "grad_norm": 2.4986538887023926, | |
| "learning_rate": 1.3793103448275862e-06, | |
| "loss": 0.381, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.741379310344827, | |
| "grad_norm": 5.762561321258545, | |
| "learning_rate": 1.0344827586206898e-06, | |
| "loss": 0.4179, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.827586206896552, | |
| "grad_norm": 1.9201445579528809, | |
| "learning_rate": 6.896551724137931e-07, | |
| "loss": 0.4958, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.913793103448276, | |
| "grad_norm": 5.6399407386779785, | |
| "learning_rate": 3.4482758620689656e-07, | |
| "loss": 0.4647, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.986402988433838, | |
| "learning_rate": 0.0, | |
| "loss": 0.4319, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8597560975609756, | |
| "eval_loss": 0.3955613076686859, | |
| "eval_runtime": 18.7383, | |
| "eval_samples_per_second": 8.752, | |
| "eval_steps_per_second": 1.121, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 580, | |
| "total_flos": 3.571605801336545e+17, | |
| "train_loss": 0.13593495262080224, | |
| "train_runtime": 495.489, | |
| "train_samples_per_second": 9.334, | |
| "train_steps_per_second": 1.171 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 580, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.571605801336545e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |