| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.024096385542169, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4e-05, | |
| "loss": 6.9946, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 8e-05, | |
| "loss": 5.455, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00012, | |
| "loss": 4.45, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00016, | |
| "loss": 3.8135, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0002, | |
| "loss": 3.4829, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00019555555555555556, | |
| "loss": 3.1942, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00019111111111111114, | |
| "loss": 2.9794, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.0001866666666666667, | |
| "loss": 2.8266, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00018222222222222224, | |
| "loss": 2.6337, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 2.5671, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00017333333333333334, | |
| "loss": 2.4814, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00016888888888888889, | |
| "loss": 2.4243, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.00016444444444444444, | |
| "loss": 2.3516, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00016, | |
| "loss": 2.3141, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.00015555555555555556, | |
| "loss": 2.2503, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.0001511111111111111, | |
| "loss": 2.1908, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.00014666666666666666, | |
| "loss": 2.1537, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 0.00014222222222222224, | |
| "loss": 2.0442, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 0.0001377777777777778, | |
| "loss": 2.0096, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 1.9876, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.00012888888888888892, | |
| "loss": 1.9789, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 0.00012444444444444444, | |
| "loss": 1.9375, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.00012, | |
| "loss": 1.9222, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.00011555555555555555, | |
| "loss": 1.8738, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.00011111111111111112, | |
| "loss": 1.8951, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "eval_bleu": 3.9246, | |
| "eval_gen_len": 18.6281, | |
| "eval_loss": 2.265472412109375, | |
| "eval_runtime": 47.3348, | |
| "eval_samples_per_second": 88.73, | |
| "eval_steps_per_second": 0.697, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 0.00010666666666666667, | |
| "loss": 1.7962, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 0.00010222222222222222, | |
| "loss": 1.7742, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 9.777777777777778e-05, | |
| "loss": 1.7561, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 1.7248, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 1.7343, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 8.444444444444444e-05, | |
| "loss": 1.7053, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 8e-05, | |
| "loss": 1.6866, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 7.555555555555556e-05, | |
| "loss": 1.665, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 7.111111111111112e-05, | |
| "loss": 1.6531, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 1.615, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 6.222222222222222e-05, | |
| "loss": 1.6093, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 5.7777777777777776e-05, | |
| "loss": 1.6087, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 1.586, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 4.888888888888889e-05, | |
| "loss": 1.5655, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 1.5485, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 4e-05, | |
| "loss": 1.582, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 1.5616, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 3.111111111111111e-05, | |
| "loss": 1.5579, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 1.5261, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 1.5194, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 1.777777777777778e-05, | |
| "loss": 1.5099, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 1.5227, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 8.88888888888889e-06, | |
| "loss": 1.4967, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "learning_rate": 4.444444444444445e-06, | |
| "loss": 1.5215, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 0.0, | |
| "loss": 1.5435, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "eval_bleu": 4.1139, | |
| "eval_gen_len": 18.6052, | |
| "eval_loss": 2.260429859161377, | |
| "eval_runtime": 18.0892, | |
| "eval_samples_per_second": 232.183, | |
| "eval_steps_per_second": 1.824, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "step": 500, | |
| "total_flos": 3.792343111041024e+16, | |
| "train_loss": 2.211637247085571, | |
| "train_runtime": 368.6992, | |
| "train_samples_per_second": 347.166, | |
| "train_steps_per_second": 1.356 | |
| } | |
| ], | |
| "max_steps": 500, | |
| "num_train_epochs": 7, | |
| "total_flos": 3.792343111041024e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |