| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.98463901689708, | |
| "global_step": 13000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.923195084485407e-05, | |
| "loss": 2.0051, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1.8463901689708145e-05, | |
| "loss": 1.6247, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.3888643980026245, | |
| "eval_runtime": 7.9545, | |
| "eval_samples_per_second": 653.846, | |
| "eval_steps_per_second": 81.841, | |
| "step": 1302 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.7695852534562215e-05, | |
| "loss": 1.4928, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 1.6927803379416285e-05, | |
| "loss": 1.3947, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.6159754224270355e-05, | |
| "loss": 1.3445, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.201552152633667, | |
| "eval_runtime": 8.1327, | |
| "eval_samples_per_second": 639.513, | |
| "eval_steps_per_second": 80.047, | |
| "step": 2604 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.5391705069124425e-05, | |
| "loss": 1.2667, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 1.4623655913978497e-05, | |
| "loss": 1.2231, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.0806881189346313, | |
| "eval_runtime": 7.9883, | |
| "eval_samples_per_second": 651.08, | |
| "eval_steps_per_second": 81.495, | |
| "step": 3906 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 1.3855606758832567e-05, | |
| "loss": 1.2226, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 1.3087557603686638e-05, | |
| "loss": 1.1666, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 1.2319508448540707e-05, | |
| "loss": 1.1263, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.024440050125122, | |
| "eval_runtime": 7.9462, | |
| "eval_samples_per_second": 654.524, | |
| "eval_steps_per_second": 81.926, | |
| "step": 5208 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 1.1551459293394778e-05, | |
| "loss": 1.1107, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 1.0783410138248848e-05, | |
| "loss": 1.0675, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 1.001536098310292e-05, | |
| "loss": 1.062, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.9445285797119141, | |
| "eval_runtime": 8.0272, | |
| "eval_samples_per_second": 647.921, | |
| "eval_steps_per_second": 81.099, | |
| "step": 6510 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 9.24731182795699e-06, | |
| "loss": 1.03, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 8.47926267281106e-06, | |
| "loss": 1.0055, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.9380741715431213, | |
| "eval_runtime": 8.0529, | |
| "eval_samples_per_second": 645.851, | |
| "eval_steps_per_second": 80.84, | |
| "step": 7812 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 7.711213517665132e-06, | |
| "loss": 0.9869, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 6.9431643625192015e-06, | |
| "loss": 1.0125, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 6.175115207373272e-06, | |
| "loss": 1.004, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.902702271938324, | |
| "eval_runtime": 8.0927, | |
| "eval_samples_per_second": 642.681, | |
| "eval_steps_per_second": 80.443, | |
| "step": 9114 | |
| }, | |
| { | |
| "epoch": 7.3, | |
| "learning_rate": 5.407066052227343e-06, | |
| "loss": 0.9893, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "learning_rate": 4.639016897081414e-06, | |
| "loss": 0.9525, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.8645159006118774, | |
| "eval_runtime": 8.0556, | |
| "eval_samples_per_second": 645.641, | |
| "eval_steps_per_second": 80.814, | |
| "step": 10416 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 3.870967741935484e-06, | |
| "loss": 0.9761, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "learning_rate": 3.1029185867895553e-06, | |
| "loss": 0.9469, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 2.3348694316436257e-06, | |
| "loss": 0.9166, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.8442177176475525, | |
| "eval_runtime": 8.0728, | |
| "eval_samples_per_second": 644.264, | |
| "eval_steps_per_second": 80.641, | |
| "step": 11718 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 1.5668202764976959e-06, | |
| "loss": 0.9201, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 7.987711213517666e-07, | |
| "loss": 0.9143, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "learning_rate": 3.0721966205837177e-08, | |
| "loss": 0.9314, | |
| "step": 13000 | |
| } | |
| ], | |
| "max_steps": 13020, | |
| "num_train_epochs": 10, | |
| "total_flos": 673236987330600.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |