| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 14.682926829268293, | |
| "eval_steps": 500, | |
| "global_step": 217, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.9999451693655125e-05, | |
| "loss": 1.3914, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.9986295347545738e-05, | |
| "loss": 1.1489, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.9945218953682736e-05, | |
| "loss": 1.0541, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 1.0341095924377441, | |
| "eval_runtime": 379.2893, | |
| "eval_samples_per_second": 60.93, | |
| "eval_steps_per_second": 0.477, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.9876883405951378e-05, | |
| "loss": 1.0441, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.9781476007338058e-05, | |
| "loss": 0.9858, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.9659258262890683e-05, | |
| "loss": 0.9708, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_loss": 1.0142368078231812, | |
| "eval_runtime": 378.349, | |
| "eval_samples_per_second": 61.081, | |
| "eval_steps_per_second": 0.478, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.9510565162951538e-05, | |
| "loss": 0.9679, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 1.9335804264972018e-05, | |
| "loss": 0.9274, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 1.913545457642601e-05, | |
| "loss": 0.9142, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "eval_loss": 1.0111455917358398, | |
| "eval_runtime": 378.1977, | |
| "eval_samples_per_second": 61.106, | |
| "eval_steps_per_second": 0.479, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 1.891006524188368e-05, | |
| "loss": 0.8994, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 1.866025403784439e-05, | |
| "loss": 0.8775, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 1.8386705679454243e-05, | |
| "loss": 0.8637, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "eval_loss": 1.023858666419983, | |
| "eval_runtime": 378.3558, | |
| "eval_samples_per_second": 61.08, | |
| "eval_steps_per_second": 0.478, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 1.8090169943749477e-05, | |
| "loss": 0.8323, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 1.777145961456971e-05, | |
| "loss": 0.8209, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 1.7431448254773943e-05, | |
| "loss": 0.8091, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "eval_loss": 1.0362672805786133, | |
| "eval_runtime": 378.1433, | |
| "eval_samples_per_second": 61.114, | |
| "eval_steps_per_second": 0.479, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 1.7071067811865477e-05, | |
| "loss": 0.7588, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 1.6691306063588583e-05, | |
| "loss": 0.7629, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 1.6293203910498375e-05, | |
| "loss": 0.7516, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "eval_loss": 1.0779954195022583, | |
| "eval_runtime": 378.7856, | |
| "eval_samples_per_second": 61.011, | |
| "eval_steps_per_second": 0.478, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 1.5877852522924733e-05, | |
| "loss": 0.6866, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 1.5446390350150272e-05, | |
| "loss": 0.6993, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6884, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "eval_loss": 1.0987329483032227, | |
| "eval_runtime": 378.7519, | |
| "eval_samples_per_second": 61.016, | |
| "eval_steps_per_second": 0.478, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 1.4539904997395468e-05, | |
| "loss": 0.6135, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 1.4067366430758004e-05, | |
| "loss": 0.6301, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 1.3583679495453e-05, | |
| "loss": 0.6309, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 7.71, | |
| "eval_loss": 1.1394284963607788, | |
| "eval_runtime": 378.7271, | |
| "eval_samples_per_second": 61.02, | |
| "eval_steps_per_second": 0.478, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 1.3090169943749475e-05, | |
| "loss": 0.5431, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 1.2588190451025209e-05, | |
| "loss": 0.5686, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 8.68, | |
| "learning_rate": 1.2079116908177592e-05, | |
| "loss": 0.5696, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 8.68, | |
| "eval_loss": 1.1820148229599, | |
| "eval_runtime": 378.7544, | |
| "eval_samples_per_second": 61.016, | |
| "eval_steps_per_second": 0.478, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 1.156434465040231e-05, | |
| "loss": 0.489, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "learning_rate": 1.1045284632676535e-05, | |
| "loss": 0.5067, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.71, | |
| "learning_rate": 1.0523359562429441e-05, | |
| "loss": 0.4752, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 9.71, | |
| "eval_loss": 1.2695468664169312, | |
| "eval_runtime": 379.3356, | |
| "eval_samples_per_second": 60.922, | |
| "eval_steps_per_second": 0.477, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4579, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 9.476640437570562e-06, | |
| "loss": 0.448, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 10.68, | |
| "eval_loss": 1.3108839988708496, | |
| "eval_runtime": 379.098, | |
| "eval_samples_per_second": 60.96, | |
| "eval_steps_per_second": 0.477, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 8.954715367323468e-06, | |
| "loss": 0.4175, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 11.27, | |
| "learning_rate": 8.43565534959769e-06, | |
| "loss": 0.4065, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 7.92088309182241e-06, | |
| "loss": 0.3955, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 11.71, | |
| "eval_loss": 1.387709379196167, | |
| "eval_runtime": 379.1872, | |
| "eval_samples_per_second": 60.946, | |
| "eval_steps_per_second": 0.477, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 7.411809548974792e-06, | |
| "loss": 0.3736, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 12.29, | |
| "learning_rate": 6.909830056250527e-06, | |
| "loss": 0.3623, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 12.54, | |
| "learning_rate": 6.4163205045469975e-06, | |
| "loss": 0.3579, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 12.68, | |
| "eval_loss": 1.3922604322433472, | |
| "eval_runtime": 378.9155, | |
| "eval_samples_per_second": 60.99, | |
| "eval_steps_per_second": 0.478, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "learning_rate": 5.932633569242e-06, | |
| "loss": 0.3336, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 13.32, | |
| "learning_rate": 5.460095002604533e-06, | |
| "loss": 0.3216, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3228, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "eval_loss": 1.4064093828201294, | |
| "eval_runtime": 378.8946, | |
| "eval_samples_per_second": 60.993, | |
| "eval_steps_per_second": 0.478, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 4.5536096498497295e-06, | |
| "loss": 0.2974, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 14.34, | |
| "learning_rate": 4.12214747707527e-06, | |
| "loss": 0.29, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 14.59, | |
| "learning_rate": 3.7067960895016277e-06, | |
| "loss": 0.2914, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "eval_loss": 1.4377079010009766, | |
| "eval_runtime": 378.9608, | |
| "eval_samples_per_second": 60.983, | |
| "eval_steps_per_second": 0.478, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "step": 217, | |
| "total_flos": 353275824046080.0, | |
| "train_loss": 0.6393404012451523, | |
| "train_runtime": 14968.9879, | |
| "train_samples_per_second": 10.415, | |
| "train_steps_per_second": 0.02 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 300, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "total_flos": 353275824046080.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |