| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "global_step": 2700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.8478, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.3793, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 16.2484, | |
| "eval_gen_len": 18.4167, | |
| "eval_loss": 0.3601939380168915, | |
| "eval_meteor": 0.4221, | |
| "eval_runtime": 13.5749, | |
| "eval_samples_per_second": 4.42, | |
| "eval_steps_per_second": 2.21, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.3099, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.274, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.2437, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 17.2012, | |
| "eval_gen_len": 18.45, | |
| "eval_loss": 0.32714077830314636, | |
| "eval_meteor": 0.4235, | |
| "eval_runtime": 11.4881, | |
| "eval_samples_per_second": 5.223, | |
| "eval_steps_per_second": 2.611, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2115, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.1758, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.1911, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 17.6526, | |
| "eval_gen_len": 18.5, | |
| "eval_loss": 0.3079971969127655, | |
| "eval_meteor": 0.4244, | |
| "eval_runtime": 11.4927, | |
| "eval_samples_per_second": 5.221, | |
| "eval_steps_per_second": 2.61, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.136, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.1508, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 17.6733, | |
| "eval_gen_len": 18.5, | |
| "eval_loss": 0.30199024081230164, | |
| "eval_meteor": 0.4227, | |
| "eval_runtime": 11.3901, | |
| "eval_samples_per_second": 5.268, | |
| "eval_steps_per_second": 2.634, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.1336, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0853, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.1042, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 17.1567, | |
| "eval_gen_len": 18.45, | |
| "eval_loss": 0.43005862832069397, | |
| "eval_meteor": 0.421, | |
| "eval_runtime": 11.4383, | |
| "eval_samples_per_second": 5.246, | |
| "eval_steps_per_second": 2.623, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.0795, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0776, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.0712, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu": 17.5338, | |
| "eval_gen_len": 18.5, | |
| "eval_loss": 0.39506202936172485, | |
| "eval_meteor": 0.4211, | |
| "eval_runtime": 11.4467, | |
| "eval_samples_per_second": 5.242, | |
| "eval_steps_per_second": 2.621, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.0548, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0516, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bleu": 17.1366, | |
| "eval_gen_len": 18.4667, | |
| "eval_loss": 0.44624677300453186, | |
| "eval_meteor": 0.4185, | |
| "eval_runtime": 11.4875, | |
| "eval_samples_per_second": 5.223, | |
| "eval_steps_per_second": 2.612, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.0557, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.0485, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.043, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu": 17.6692, | |
| "eval_gen_len": 18.5, | |
| "eval_loss": 0.4671081006526947, | |
| "eval_meteor": 0.4217, | |
| "eval_runtime": 11.4117, | |
| "eval_samples_per_second": 5.258, | |
| "eval_steps_per_second": 2.629, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.0412, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.0402, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0243, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_bleu": 17.5591, | |
| "eval_gen_len": 18.5, | |
| "eval_loss": 0.5049145221710205, | |
| "eval_meteor": 0.4214, | |
| "eval_runtime": 11.458, | |
| "eval_samples_per_second": 5.237, | |
| "eval_steps_per_second": 2.618, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0475, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.0324, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0313, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_bleu": 17.4599, | |
| "eval_gen_len": 18.5, | |
| "eval_loss": 0.5202592015266418, | |
| "eval_meteor": 0.4218, | |
| "eval_runtime": 11.5675, | |
| "eval_samples_per_second": 5.187, | |
| "eval_steps_per_second": 2.593, | |
| "step": 2700 | |
| } | |
| ], | |
| "max_steps": 2700, | |
| "num_train_epochs": 10, | |
| "total_flos": 163370071388160.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |