| { | |
| "best_metric": 0.6006270051002502, | |
| "best_model_checkpoint": "/mnt/ml_drive/kcardenas/target_hold_hands/checkpoint-624", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 1040, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4807692307692308, | |
| "grad_norm": 772638.0, | |
| "learning_rate": 5e-06, | |
| "loss": 2.0721, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 57469.69140625, | |
| "learning_rate": 1e-05, | |
| "loss": 0.957, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_IoU": 0.0, | |
| "eval_loss": 0.8556730151176453, | |
| "eval_runtime": 8.489, | |
| "eval_samples_per_second": 68.324, | |
| "eval_steps_per_second": 1.178, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.4423076923076923, | |
| "grad_norm": 77164.875, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.8264, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 95215.3828125, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7261, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_IoU": 7.7851798261353e-09, | |
| "eval_loss": 0.6803418397903442, | |
| "eval_runtime": 8.5727, | |
| "eval_samples_per_second": 67.657, | |
| "eval_steps_per_second": 1.166, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.4038461538461537, | |
| "grad_norm": 78873.3359375, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6825, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.8846153846153846, | |
| "grad_norm": 65023.43359375, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6689, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_IoU": 3.1203175320106754e-07, | |
| "eval_loss": 0.6544784903526306, | |
| "eval_runtime": 8.4506, | |
| "eval_samples_per_second": 68.634, | |
| "eval_steps_per_second": 1.183, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 3.3653846153846154, | |
| "grad_norm": 187234.5, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.7322, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 207238.46875, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7126, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_IoU": 7.16540375536177e-08, | |
| "eval_loss": 0.702006459236145, | |
| "eval_runtime": 8.1834, | |
| "eval_samples_per_second": 70.875, | |
| "eval_steps_per_second": 1.222, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 4.326923076923077, | |
| "grad_norm": 70302.0625, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.7118, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 4.8076923076923075, | |
| "grad_norm": 106005.375, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6688, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_IoU": 5.327572640987815e-07, | |
| "eval_loss": 0.6712496876716614, | |
| "eval_runtime": 8.4417, | |
| "eval_samples_per_second": 68.707, | |
| "eval_steps_per_second": 1.185, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.288461538461538, | |
| "grad_norm": 181192.4375, | |
| "learning_rate": 4.8417721518987346e-05, | |
| "loss": 0.729, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.769230769230769, | |
| "grad_norm": 527063.125, | |
| "learning_rate": 4.683544303797468e-05, | |
| "loss": 0.7126, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_IoU": 8.267290868345697e-08, | |
| "eval_loss": 0.6633245944976807, | |
| "eval_runtime": 8.2611, | |
| "eval_samples_per_second": 70.208, | |
| "eval_steps_per_second": 1.21, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 36219.71875, | |
| "learning_rate": 4.525316455696203e-05, | |
| "loss": 0.6747, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 6.730769230769231, | |
| "grad_norm": 89007.9921875, | |
| "learning_rate": 4.367088607594937e-05, | |
| "loss": 0.6633, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_IoU": 1.3435934524923367e-07, | |
| "eval_loss": 0.6083313226699829, | |
| "eval_runtime": 8.3505, | |
| "eval_samples_per_second": 69.457, | |
| "eval_steps_per_second": 1.198, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 7.211538461538462, | |
| "grad_norm": 32232.62109375, | |
| "learning_rate": 4.208860759493671e-05, | |
| "loss": 0.611, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 54438.1484375, | |
| "learning_rate": 4.050632911392405e-05, | |
| "loss": 0.6113, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_IoU": 2.002574751254223e-07, | |
| "eval_loss": 0.6060946583747864, | |
| "eval_runtime": 8.5578, | |
| "eval_samples_per_second": 67.774, | |
| "eval_steps_per_second": 1.169, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 8.173076923076923, | |
| "grad_norm": 47105.93359375, | |
| "learning_rate": 3.89240506329114e-05, | |
| "loss": 0.6072, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 8.653846153846153, | |
| "grad_norm": 37276.02734375, | |
| "learning_rate": 3.7341772151898736e-05, | |
| "loss": 0.6101, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_IoU": 1.3619726913206721e-07, | |
| "eval_loss": 0.6026849150657654, | |
| "eval_runtime": 8.1149, | |
| "eval_samples_per_second": 71.473, | |
| "eval_steps_per_second": 1.232, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 9.134615384615385, | |
| "grad_norm": 27913.052734375, | |
| "learning_rate": 3.575949367088608e-05, | |
| "loss": 0.6014, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 9.615384615384615, | |
| "grad_norm": 44065.1796875, | |
| "learning_rate": 3.4177215189873416e-05, | |
| "loss": 0.6028, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_IoU": 0.0, | |
| "eval_loss": 0.6006531715393066, | |
| "eval_runtime": 8.1108, | |
| "eval_samples_per_second": 71.509, | |
| "eval_steps_per_second": 1.233, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 10.096153846153847, | |
| "grad_norm": 54902.17578125, | |
| "learning_rate": 3.2594936708860766e-05, | |
| "loss": 0.6069, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 10.576923076923077, | |
| "grad_norm": 47223.47265625, | |
| "learning_rate": 3.10126582278481e-05, | |
| "loss": 0.5983, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_IoU": 3.227043016392884e-09, | |
| "eval_loss": 0.6019152402877808, | |
| "eval_runtime": 8.2699, | |
| "eval_samples_per_second": 70.134, | |
| "eval_steps_per_second": 1.209, | |
| "step": 572 | |
| }, | |
| { | |
| "epoch": 11.057692307692308, | |
| "grad_norm": 44127.5, | |
| "learning_rate": 2.9430379746835446e-05, | |
| "loss": 0.5984, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 11.538461538461538, | |
| "grad_norm": 65407.2578125, | |
| "learning_rate": 2.7848101265822786e-05, | |
| "loss": 0.6014, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_IoU": 1.2153273194260533e-07, | |
| "eval_loss": 0.6006270051002502, | |
| "eval_runtime": 8.3368, | |
| "eval_samples_per_second": 69.571, | |
| "eval_steps_per_second": 1.199, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 12.01923076923077, | |
| "grad_norm": 72088.734375, | |
| "learning_rate": 2.626582278481013e-05, | |
| "loss": 0.5996, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 54947.15234375, | |
| "learning_rate": 2.468354430379747e-05, | |
| "loss": 0.5988, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 12.98076923076923, | |
| "grad_norm": 36718.6796875, | |
| "learning_rate": 2.3101265822784813e-05, | |
| "loss": 0.5968, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_IoU": 1.0654944237144431e-07, | |
| "eval_loss": 0.6013975739479065, | |
| "eval_runtime": 8.4006, | |
| "eval_samples_per_second": 69.043, | |
| "eval_steps_per_second": 1.19, | |
| "step": 676 | |
| }, | |
| { | |
| "epoch": 13.461538461538462, | |
| "grad_norm": 44921.89453125, | |
| "learning_rate": 2.1518987341772153e-05, | |
| "loss": 0.599, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 13.942307692307692, | |
| "grad_norm": 45908.14453125, | |
| "learning_rate": 1.9936708860759496e-05, | |
| "loss": 0.5932, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_IoU": 4.530420583131857e-07, | |
| "eval_loss": 0.6021161675453186, | |
| "eval_runtime": 7.9829, | |
| "eval_samples_per_second": 72.656, | |
| "eval_steps_per_second": 1.253, | |
| "step": 728 | |
| }, | |
| { | |
| "epoch": 14.423076923076923, | |
| "grad_norm": 51494.0, | |
| "learning_rate": 1.8354430379746836e-05, | |
| "loss": 0.5967, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 14.903846153846153, | |
| "grad_norm": 122982.484375, | |
| "learning_rate": 1.677215189873418e-05, | |
| "loss": 0.592, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_IoU": 0.0, | |
| "eval_loss": 0.6046732664108276, | |
| "eval_runtime": 8.5652, | |
| "eval_samples_per_second": 67.716, | |
| "eval_steps_per_second": 1.168, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 15.384615384615385, | |
| "grad_norm": 51938.21484375, | |
| "learning_rate": 1.5189873417721521e-05, | |
| "loss": 0.5907, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 15.865384615384615, | |
| "grad_norm": 316787.84375, | |
| "learning_rate": 1.3607594936708861e-05, | |
| "loss": 0.5935, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_IoU": 1.7956697673278814e-07, | |
| "eval_loss": 0.6020213961601257, | |
| "eval_runtime": 8.4008, | |
| "eval_samples_per_second": 69.041, | |
| "eval_steps_per_second": 1.19, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 16.346153846153847, | |
| "grad_norm": 31160.560546875, | |
| "learning_rate": 1.2025316455696203e-05, | |
| "loss": 0.59, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 16.826923076923077, | |
| "grad_norm": 61832.81640625, | |
| "learning_rate": 1.0443037974683544e-05, | |
| "loss": 0.5873, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_IoU": 2.442868828912477e-08, | |
| "eval_loss": 0.6026484966278076, | |
| "eval_runtime": 8.4706, | |
| "eval_samples_per_second": 68.472, | |
| "eval_steps_per_second": 1.181, | |
| "step": 884 | |
| }, | |
| { | |
| "epoch": 17.307692307692307, | |
| "grad_norm": 37131.21484375, | |
| "learning_rate": 8.860759493670886e-06, | |
| "loss": 0.591, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 17.78846153846154, | |
| "grad_norm": 42635.22265625, | |
| "learning_rate": 7.2784810126582285e-06, | |
| "loss": 0.5853, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_IoU": 6.648389312912634e-09, | |
| "eval_loss": 0.6114510893821716, | |
| "eval_runtime": 8.1171, | |
| "eval_samples_per_second": 71.454, | |
| "eval_steps_per_second": 1.232, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 18.26923076923077, | |
| "grad_norm": 40143.515625, | |
| "learning_rate": 5.69620253164557e-06, | |
| "loss": 0.5918, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "grad_norm": 38210.26953125, | |
| "learning_rate": 4.113924050632911e-06, | |
| "loss": 0.5835, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_IoU": 1.853719873045982e-07, | |
| "eval_loss": 0.6087284088134766, | |
| "eval_runtime": 8.3102, | |
| "eval_samples_per_second": 69.794, | |
| "eval_steps_per_second": 1.203, | |
| "step": 988 | |
| }, | |
| { | |
| "epoch": 19.23076923076923, | |
| "grad_norm": 54590.265625, | |
| "learning_rate": 2.531645569620253e-06, | |
| "loss": 0.5897, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 19.71153846153846, | |
| "grad_norm": 47306.3046875, | |
| "learning_rate": 9.493670886075951e-07, | |
| "loss": 0.5868, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_IoU": 1.5521619976116134e-07, | |
| "eval_loss": 0.6075738072395325, | |
| "eval_runtime": 8.5226, | |
| "eval_samples_per_second": 68.055, | |
| "eval_steps_per_second": 1.173, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 1040, | |
| "total_flos": 0.0, | |
| "train_loss": 0.6731257475339449, | |
| "train_runtime": 3556.481, | |
| "train_samples_per_second": 18.479, | |
| "train_steps_per_second": 0.292 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1040, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "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": 0.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |