| { | |
| "best_metric": 9.025093793163277, | |
| "best_model_checkpoint": "./envi_checkpoints/checkpoint-31250", | |
| "epoch": 2.0, | |
| "global_step": 31250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3.1808000000000005e-06, | |
| "loss": 7.9834, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 6.380800000000001e-06, | |
| "loss": 5.5297, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.580800000000002e-06, | |
| "loss": 4.7262, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.2780800000000001e-05, | |
| "loss": 4.4162, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.59808e-05, | |
| "loss": 4.2072, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.91808e-05, | |
| "loss": 4.0103, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 2.23808e-05, | |
| "loss": 3.7644, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 2.55808e-05, | |
| "loss": 3.3987, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.87808e-05, | |
| "loss": 2.8234, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.19808e-05, | |
| "loss": 2.0638, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 3.51808e-05, | |
| "loss": 1.5503, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 3.83776e-05, | |
| "loss": 1.2126, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.15776e-05, | |
| "loss": 0.9281, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.47776e-05, | |
| "loss": 0.713, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.79776e-05, | |
| "loss": 0.6009, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 8.6159300527749, | |
| "eval_loss": 0.45428285002708435, | |
| "eval_runtime": 93.2143, | |
| "eval_samples_per_second": 26.82, | |
| "eval_steps_per_second": 0.848, | |
| "step": 15625 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 4.88224e-05, | |
| "loss": 0.54, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 4.56256e-05, | |
| "loss": 0.5017, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.24288e-05, | |
| "loss": 0.4777, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 3.9228800000000006e-05, | |
| "loss": 0.4661, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3.60288e-05, | |
| "loss": 0.4526, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 3.2828800000000007e-05, | |
| "loss": 0.4457, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 2.96288e-05, | |
| "loss": 0.4386, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 2.6428800000000004e-05, | |
| "loss": 0.4365, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 2.32288e-05, | |
| "loss": 0.4305, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.0032e-05, | |
| "loss": 0.4216, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 1.6832e-05, | |
| "loss": 0.4096, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 1.3632000000000001e-05, | |
| "loss": 0.4066, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1.0432e-05, | |
| "loss": 0.4071, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 7.2319999999999995e-06, | |
| "loss": 0.4048, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 4.0320000000000005e-06, | |
| "loss": 0.3997, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 8.352000000000001e-07, | |
| "loss": 0.4015, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 9.025093793163277, | |
| "eval_loss": 0.36966755986213684, | |
| "eval_runtime": 94.07, | |
| "eval_samples_per_second": 26.576, | |
| "eval_steps_per_second": 0.84, | |
| "step": 31250 | |
| } | |
| ], | |
| "max_steps": 31250, | |
| "num_train_epochs": 2, | |
| "total_flos": 1.505048664e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |