| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.074688796680498, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 87.4023, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 52.564, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 6e-06, | |
| "loss": 41.0172, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 36.522, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1e-05, | |
| "loss": 33.7109, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.2e-05, | |
| "loss": 31.384, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 29.2667, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 27.6322, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 1.8e-05, | |
| "loss": 25.95, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 2e-05, | |
| "loss": 24.3416, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 22.5298, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.4e-05, | |
| "loss": 20.7681, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 19.4657, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 17.7138, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3e-05, | |
| "loss": 16.3022, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 14.6267, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 13.0298, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 3.6e-05, | |
| "loss": 11.5227, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3.8e-05, | |
| "loss": 9.8717, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4e-05, | |
| "loss": 8.0718, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_bleu": 5.8841, | |
| "eval_em": 0.0, | |
| "eval_gen_len": 126.98, | |
| "eval_loss": 6.968929290771484, | |
| "eval_rm": 0.0, | |
| "eval_runtime": 84.2898, | |
| "eval_samples_per_second": 0.593, | |
| "eval_steps_per_second": 0.083, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.2e-05, | |
| "loss": 7.3621, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 6.6663, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 6.4464, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 6.0977, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 5e-05, | |
| "loss": 5.8656, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 4.976851851851852e-05, | |
| "loss": 5.6319, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.9537037037037035e-05, | |
| "loss": 5.2821, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.930555555555556e-05, | |
| "loss": 4.9816, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.9074074074074075e-05, | |
| "loss": 5.0416, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 4.8842592592592595e-05, | |
| "loss": 4.9786, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 4.7885, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 4.837962962962963e-05, | |
| "loss": 4.6081, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 4.4985, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 4.791666666666667e-05, | |
| "loss": 4.3774, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.768518518518519e-05, | |
| "loss": 4.2805, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 4.745370370370371e-05, | |
| "loss": 3.9643, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 3.9951, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.699074074074074e-05, | |
| "loss": 3.9219, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 4.675925925925926e-05, | |
| "loss": 3.7154, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 4.652777777777778e-05, | |
| "loss": 3.6551, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_bleu": 20.8847, | |
| "eval_em": 0.0, | |
| "eval_gen_len": 30.46, | |
| "eval_loss": 3.524771213531494, | |
| "eval_rm": 0.0, | |
| "eval_runtime": 30.3441, | |
| "eval_samples_per_second": 1.648, | |
| "eval_steps_per_second": 0.231, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 3.6855, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 4.6064814814814814e-05, | |
| "loss": 3.5543, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 3.4561, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 4.5601851851851854e-05, | |
| "loss": 3.3651, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 4.5370370370370374e-05, | |
| "loss": 3.1537, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 3.3103, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 4.490740740740741e-05, | |
| "loss": 3.1797, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 4.467592592592593e-05, | |
| "loss": 3.1758, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 2.9908, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 4.4212962962962966e-05, | |
| "loss": 2.7989, | |
| "step": 5000 | |
| } | |
| ], | |
| "max_steps": 24100, | |
| "num_train_epochs": 10, | |
| "total_flos": 1401098107488768.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |