| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.73913043478261, | |
| "global_step": 140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 9.285714285714286e-05, | |
| "loss": 3.382, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 2.3120291233062744, | |
| "eval_runtime": 3.0033, | |
| "eval_samples_per_second": 112.543, | |
| "eval_steps_per_second": 14.318, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 8.571428571428571e-05, | |
| "loss": 2.5354, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_loss": 2.1625144481658936, | |
| "eval_runtime": 2.8876, | |
| "eval_samples_per_second": 117.052, | |
| "eval_steps_per_second": 14.891, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 7.857142857142858e-05, | |
| "loss": 2.3745, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 7.142857142857143e-05, | |
| "loss": 2.2504, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_loss": 2.103144645690918, | |
| "eval_runtime": 2.866, | |
| "eval_samples_per_second": 117.932, | |
| "eval_steps_per_second": 15.003, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 6.428571428571429e-05, | |
| "loss": 2.1809, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_loss": 2.0733468532562256, | |
| "eval_runtime": 2.6867, | |
| "eval_samples_per_second": 125.803, | |
| "eval_steps_per_second": 16.005, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 5.714285714285714e-05, | |
| "loss": 2.1339, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 5e-05, | |
| "loss": 2.0889, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_loss": 2.0512640476226807, | |
| "eval_runtime": 2.8196, | |
| "eval_samples_per_second": 119.875, | |
| "eval_steps_per_second": 15.25, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 2.0586, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "eval_loss": 2.0358786582946777, | |
| "eval_runtime": 2.6297, | |
| "eval_samples_per_second": 128.531, | |
| "eval_steps_per_second": 16.352, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 2.0399, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 2.0117, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_loss": 2.032505750656128, | |
| "eval_runtime": 3.3953, | |
| "eval_samples_per_second": 99.549, | |
| "eval_steps_per_second": 12.665, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 1.9849, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.025442600250244, | |
| "eval_runtime": 3.0905, | |
| "eval_samples_per_second": 109.368, | |
| "eval_steps_per_second": 13.914, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 1.9803, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_loss": 2.0218234062194824, | |
| "eval_runtime": 3.3517, | |
| "eval_samples_per_second": 100.846, | |
| "eval_steps_per_second": 12.829, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 1.9694, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 0.0, | |
| "loss": 1.9584, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "eval_loss": 2.020766019821167, | |
| "eval_runtime": 2.8283, | |
| "eval_samples_per_second": 119.505, | |
| "eval_steps_per_second": 15.203, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "step": 140, | |
| "total_flos": 2.181454126645248e+16, | |
| "train_loss": 2.210656370435442, | |
| "train_runtime": 1086.7433, | |
| "train_samples_per_second": 33.798, | |
| "train_steps_per_second": 0.129 | |
| } | |
| ], | |
| "max_steps": 140, | |
| "num_train_epochs": 10, | |
| "total_flos": 2.181454126645248e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |