| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.797101449275362, | |
| "eval_steps": 20, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 1.4612, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 1.6819, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.493472933769226, | |
| "eval_runtime": 146.9486, | |
| "eval_samples_per_second": 6.669, | |
| "eval_steps_per_second": 0.837, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 1.3849, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 1.3102, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.0922572612762451, | |
| "eval_runtime": 146.6688, | |
| "eval_samples_per_second": 6.682, | |
| "eval_steps_per_second": 0.839, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00015, | |
| "loss": 0.9074, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.8606, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 0.7950681447982788, | |
| "eval_runtime": 147.0034, | |
| "eval_samples_per_second": 6.667, | |
| "eval_steps_per_second": 0.837, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.00020999999999999998, | |
| "loss": 0.7308, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.769, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_loss": 0.7387300729751587, | |
| "eval_runtime": 146.9341, | |
| "eval_samples_per_second": 6.67, | |
| "eval_steps_per_second": 0.837, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.00027, | |
| "loss": 0.7344, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6724, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_loss": 0.7141170501708984, | |
| "eval_runtime": 146.8124, | |
| "eval_samples_per_second": 6.675, | |
| "eval_steps_per_second": 0.838, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.00029, | |
| "loss": 0.7295, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00028, | |
| "loss": 0.6233, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_loss": 0.6994462013244629, | |
| "eval_runtime": 146.8753, | |
| "eval_samples_per_second": 6.672, | |
| "eval_steps_per_second": 0.837, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.00027, | |
| "loss": 0.7352, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 0.00026, | |
| "loss": 0.6463, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_loss": 0.6910027861595154, | |
| "eval_runtime": 147.0376, | |
| "eval_samples_per_second": 6.665, | |
| "eval_steps_per_second": 0.837, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 0.00025, | |
| "loss": 0.6765, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.683, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "eval_loss": 0.6793327331542969, | |
| "eval_runtime": 147.9244, | |
| "eval_samples_per_second": 6.625, | |
| "eval_steps_per_second": 0.832, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.00023, | |
| "loss": 0.6206, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 0.00021999999999999995, | |
| "loss": 0.7008, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_loss": 0.6654186248779297, | |
| "eval_runtime": 147.0825, | |
| "eval_samples_per_second": 6.663, | |
| "eval_steps_per_second": 0.836, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 0.00020999999999999998, | |
| "loss": 0.5981, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.00019999999999999998, | |
| "loss": 0.6754, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_loss": 0.6593416929244995, | |
| "eval_runtime": 146.8493, | |
| "eval_samples_per_second": 6.674, | |
| "eval_steps_per_second": 0.838, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 0.00018999999999999998, | |
| "loss": 0.6217, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.6349, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "eval_loss": 0.6553820967674255, | |
| "eval_runtime": 146.9525, | |
| "eval_samples_per_second": 6.669, | |
| "eval_steps_per_second": 0.837, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.00016999999999999999, | |
| "loss": 0.6592, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 0.00015999999999999999, | |
| "loss": 0.5813, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "eval_loss": 0.6516290307044983, | |
| "eval_runtime": 147.0111, | |
| "eval_samples_per_second": 6.666, | |
| "eval_steps_per_second": 0.837, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 0.00015, | |
| "loss": 0.6827, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 0.00014, | |
| "loss": 0.6018, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "eval_loss": 0.6494865417480469, | |
| "eval_runtime": 146.9051, | |
| "eval_samples_per_second": 6.671, | |
| "eval_steps_per_second": 0.837, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 0.00013, | |
| "loss": 0.6457, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 0.644, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "eval_loss": 0.6468493342399597, | |
| "eval_runtime": 146.9249, | |
| "eval_samples_per_second": 6.67, | |
| "eval_steps_per_second": 0.837, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 0.00010999999999999998, | |
| "loss": 0.5954, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 9.999999999999999e-05, | |
| "loss": 0.6548, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "eval_loss": 0.6439973711967468, | |
| "eval_runtime": 146.9023, | |
| "eval_samples_per_second": 6.671, | |
| "eval_steps_per_second": 0.837, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 0.5504, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 7.999999999999999e-05, | |
| "loss": 0.6869, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "eval_loss": 0.6418542861938477, | |
| "eval_runtime": 146.8382, | |
| "eval_samples_per_second": 6.674, | |
| "eval_steps_per_second": 0.838, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 7e-05, | |
| "loss": 0.6151, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 0.6089, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "eval_loss": 0.6405566334724426, | |
| "eval_runtime": 146.9207, | |
| "eval_samples_per_second": 6.67, | |
| "eval_steps_per_second": 0.837, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 4.9999999999999996e-05, | |
| "loss": 0.6438, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 3.9999999999999996e-05, | |
| "loss": 0.577, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "eval_loss": 0.6396059393882751, | |
| "eval_runtime": 146.9947, | |
| "eval_samples_per_second": 6.667, | |
| "eval_steps_per_second": 0.837, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 0.6599, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 1.9999999999999998e-05, | |
| "loss": 0.5724, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "eval_loss": 0.6387772560119629, | |
| "eval_runtime": 146.77, | |
| "eval_samples_per_second": 6.677, | |
| "eval_steps_per_second": 0.838, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 9.999999999999999e-06, | |
| "loss": 0.6487, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 0.0, | |
| "loss": 0.6236, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "eval_loss": 0.6381928324699402, | |
| "eval_runtime": 146.9718, | |
| "eval_samples_per_second": 6.668, | |
| "eval_steps_per_second": 0.837, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 400, | |
| "num_train_epochs": 6, | |
| "save_steps": 20, | |
| "total_flos": 7.971097386781901e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |