| { | |
| "best_metric": 1.5397371053695679, | |
| "best_model_checkpoint": "./results/t5-bow-decoder-base\\checkpoint-13500", | |
| "epoch": 6.0, | |
| "global_step": 13500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 5e-05, | |
| "loss": 4.0305, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0001, | |
| "loss": 2.3805, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 9.943820224719102e-05, | |
| "loss": 2.1631, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 9.887640449438202e-05, | |
| "loss": 2.0932, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 9.831460674157303e-05, | |
| "loss": 1.9341, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 9.775280898876405e-05, | |
| "loss": 1.9336, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 9.719101123595506e-05, | |
| "loss": 1.8884, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 9.662921348314608e-05, | |
| "loss": 1.899, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.606741573033708e-05, | |
| "loss": 1.867, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.7203162908554077, | |
| "eval_runtime": 17.6762, | |
| "eval_samples_per_second": 56.573, | |
| "eval_steps_per_second": 3.564, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 9.550561797752809e-05, | |
| "loss": 1.8113, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 9.49438202247191e-05, | |
| "loss": 1.8082, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 9.438202247191012e-05, | |
| "loss": 1.8161, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 9.382022471910112e-05, | |
| "loss": 1.7548, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 9.325842696629214e-05, | |
| "loss": 1.7057, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 9.269662921348315e-05, | |
| "loss": 1.7636, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 9.213483146067416e-05, | |
| "loss": 1.7652, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 9.157303370786518e-05, | |
| "loss": 1.693, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.101123595505618e-05, | |
| "loss": 1.7073, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.624514102935791, | |
| "eval_runtime": 17.0363, | |
| "eval_samples_per_second": 58.698, | |
| "eval_steps_per_second": 3.698, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 9.04494382022472e-05, | |
| "loss": 1.668, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 8.988764044943821e-05, | |
| "loss": 1.6579, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 8.932584269662921e-05, | |
| "loss": 1.6282, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 8.876404494382022e-05, | |
| "loss": 1.6937, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 8.820224719101124e-05, | |
| "loss": 1.6562, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 8.764044943820225e-05, | |
| "loss": 1.61, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 8.707865168539327e-05, | |
| "loss": 1.6144, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 8.651685393258427e-05, | |
| "loss": 1.5554, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.595505617977528e-05, | |
| "loss": 1.6599, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.593631386756897, | |
| "eval_runtime": 19.1501, | |
| "eval_samples_per_second": 52.219, | |
| "eval_steps_per_second": 3.29, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 8.53932584269663e-05, | |
| "loss": 1.5611, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 8.483146067415731e-05, | |
| "loss": 1.5177, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 8.426966292134831e-05, | |
| "loss": 1.5332, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 8.370786516853934e-05, | |
| "loss": 1.6529, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 8.314606741573034e-05, | |
| "loss": 1.5575, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 8.258426966292135e-05, | |
| "loss": 1.5424, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 8.202247191011237e-05, | |
| "loss": 1.5496, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 8.146067415730337e-05, | |
| "loss": 1.5465, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.089887640449438e-05, | |
| "loss": 1.5903, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.5624333620071411, | |
| "eval_runtime": 19.0571, | |
| "eval_samples_per_second": 52.474, | |
| "eval_steps_per_second": 3.306, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 8.03370786516854e-05, | |
| "loss": 1.4976, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 7.97752808988764e-05, | |
| "loss": 1.4719, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 7.921348314606743e-05, | |
| "loss": 1.4781, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 7.865168539325843e-05, | |
| "loss": 1.4652, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 7.808988764044944e-05, | |
| "loss": 1.4997, | |
| "step": 10250 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 7.752808988764046e-05, | |
| "loss": 1.5537, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 7.696629213483147e-05, | |
| "loss": 1.5558, | |
| "step": 10750 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 7.640449438202247e-05, | |
| "loss": 1.4763, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 7.584269662921349e-05, | |
| "loss": 1.5011, | |
| "step": 11250 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.5619711875915527, | |
| "eval_runtime": 19.201, | |
| "eval_samples_per_second": 52.081, | |
| "eval_steps_per_second": 3.281, | |
| "step": 11250 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 7.52808988764045e-05, | |
| "loss": 1.4501, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 7.471910112359551e-05, | |
| "loss": 1.434, | |
| "step": 11750 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 7.415730337078653e-05, | |
| "loss": 1.3839, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 7.359550561797753e-05, | |
| "loss": 1.4426, | |
| "step": 12250 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 7.303370786516854e-05, | |
| "loss": 1.4481, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 7.247191011235956e-05, | |
| "loss": 1.452, | |
| "step": 12750 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 7.191011235955056e-05, | |
| "loss": 1.485, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 7.134831460674157e-05, | |
| "loss": 1.5027, | |
| "step": 13250 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7.078651685393259e-05, | |
| "loss": 1.4556, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.5397371053695679, | |
| "eval_runtime": 22.8085, | |
| "eval_samples_per_second": 43.843, | |
| "eval_steps_per_second": 2.762, | |
| "step": 13500 | |
| } | |
| ], | |
| "max_steps": 45000, | |
| "num_train_epochs": 20, | |
| "total_flos": 3.288372609024e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |