| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.997867803837953, | |
| "eval_steps": 10, | |
| "global_step": 351, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 2.303, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_accuracy": 0.2334, | |
| "eval_loss": 2.1672284603118896, | |
| "eval_runtime": 61.2464, | |
| "eval_samples_per_second": 81.637, | |
| "eval_steps_per_second": 2.563, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 2.0158, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_accuracy": 0.657, | |
| "eval_loss": 1.6672015190124512, | |
| "eval_runtime": 57.6151, | |
| "eval_samples_per_second": 86.783, | |
| "eval_steps_per_second": 2.725, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.4855, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_accuracy": 0.8704, | |
| "eval_loss": 0.8292349576950073, | |
| "eval_runtime": 58.3323, | |
| "eval_samples_per_second": 85.716, | |
| "eval_steps_per_second": 2.691, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.936507936507937e-05, | |
| "loss": 0.7451, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_accuracy": 0.93, | |
| "eval_loss": 0.2578178346157074, | |
| "eval_runtime": 58.1914, | |
| "eval_samples_per_second": 85.923, | |
| "eval_steps_per_second": 2.698, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.7777777777777784e-05, | |
| "loss": 0.5618, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.962, | |
| "eval_loss": 0.14759203791618347, | |
| "eval_runtime": 58.5488, | |
| "eval_samples_per_second": 85.399, | |
| "eval_steps_per_second": 2.682, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.6190476190476194e-05, | |
| "loss": 0.4545, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.9642, | |
| "eval_loss": 0.12480442970991135, | |
| "eval_runtime": 61.2292, | |
| "eval_samples_per_second": 81.66, | |
| "eval_steps_per_second": 2.564, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.460317460317461e-05, | |
| "loss": 0.4587, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.9748, | |
| "eval_loss": 0.09407833218574524, | |
| "eval_runtime": 57.8655, | |
| "eval_samples_per_second": 86.407, | |
| "eval_steps_per_second": 2.713, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.301587301587302e-05, | |
| "loss": 0.3911, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_accuracy": 0.9712, | |
| "eval_loss": 0.09437316656112671, | |
| "eval_runtime": 57.5812, | |
| "eval_samples_per_second": 86.834, | |
| "eval_steps_per_second": 2.727, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.1428571428571437e-05, | |
| "loss": 0.3839, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_accuracy": 0.9756, | |
| "eval_loss": 0.0848420187830925, | |
| "eval_runtime": 58.1027, | |
| "eval_samples_per_second": 86.055, | |
| "eval_steps_per_second": 2.702, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3.984126984126984e-05, | |
| "loss": 0.3864, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_accuracy": 0.978, | |
| "eval_loss": 0.0744096040725708, | |
| "eval_runtime": 57.7886, | |
| "eval_samples_per_second": 86.522, | |
| "eval_steps_per_second": 2.717, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3.8253968253968256e-05, | |
| "loss": 0.3141, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.98, | |
| "eval_loss": 0.06731583178043365, | |
| "eval_runtime": 58.0242, | |
| "eval_samples_per_second": 86.171, | |
| "eval_steps_per_second": 2.706, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 0.3764, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_accuracy": 0.9764, | |
| "eval_loss": 0.07063312083482742, | |
| "eval_runtime": 58.0733, | |
| "eval_samples_per_second": 86.098, | |
| "eval_steps_per_second": 2.703, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.5079365079365075e-05, | |
| "loss": 0.3003, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.984, | |
| "eval_loss": 0.05995839089155197, | |
| "eval_runtime": 58.0208, | |
| "eval_samples_per_second": 86.176, | |
| "eval_steps_per_second": 2.706, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3.349206349206349e-05, | |
| "loss": 0.3566, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.9826, | |
| "eval_loss": 0.05617852881550789, | |
| "eval_runtime": 57.9727, | |
| "eval_samples_per_second": 86.248, | |
| "eval_steps_per_second": 2.708, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.19047619047619e-05, | |
| "loss": 0.2855, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.9816, | |
| "eval_loss": 0.056736305356025696, | |
| "eval_runtime": 57.7871, | |
| "eval_samples_per_second": 86.525, | |
| "eval_steps_per_second": 2.717, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.0317460317460318e-05, | |
| "loss": 0.3351, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_accuracy": 0.9828, | |
| "eval_loss": 0.0543464832007885, | |
| "eval_runtime": 58.1354, | |
| "eval_samples_per_second": 86.006, | |
| "eval_steps_per_second": 2.701, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 2.8730158730158728e-05, | |
| "loss": 0.2977, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.9798, | |
| "eval_loss": 0.05677906423807144, | |
| "eval_runtime": 57.6217, | |
| "eval_samples_per_second": 86.773, | |
| "eval_steps_per_second": 2.725, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 2.714285714285714e-05, | |
| "loss": 0.2924, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_accuracy": 0.9804, | |
| "eval_loss": 0.05774379149079323, | |
| "eval_runtime": 58.0352, | |
| "eval_samples_per_second": 86.155, | |
| "eval_steps_per_second": 2.705, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.5555555555555554e-05, | |
| "loss": 0.2884, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_accuracy": 0.983, | |
| "eval_loss": 0.05509978160262108, | |
| "eval_runtime": 57.5779, | |
| "eval_samples_per_second": 86.839, | |
| "eval_steps_per_second": 2.727, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.396825396825397e-05, | |
| "loss": 0.3067, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_accuracy": 0.983, | |
| "eval_loss": 0.04865285009145737, | |
| "eval_runtime": 57.833, | |
| "eval_samples_per_second": 86.456, | |
| "eval_steps_per_second": 2.715, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.2380952380952384e-05, | |
| "loss": 0.3159, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.984, | |
| "eval_loss": 0.05127624422311783, | |
| "eval_runtime": 57.9626, | |
| "eval_samples_per_second": 86.263, | |
| "eval_steps_per_second": 2.709, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 2.0793650793650797e-05, | |
| "loss": 0.2795, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_accuracy": 0.9846, | |
| "eval_loss": 0.046046771109104156, | |
| "eval_runtime": 57.8964, | |
| "eval_samples_per_second": 86.361, | |
| "eval_steps_per_second": 2.712, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.920634920634921e-05, | |
| "loss": 0.3113, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.9832, | |
| "eval_loss": 0.04946137219667435, | |
| "eval_runtime": 57.9735, | |
| "eval_samples_per_second": 86.246, | |
| "eval_steps_per_second": 2.708, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.761904761904762e-05, | |
| "loss": 0.2882, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.9838, | |
| "eval_loss": 0.047487206757068634, | |
| "eval_runtime": 57.8304, | |
| "eval_samples_per_second": 86.46, | |
| "eval_steps_per_second": 2.715, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.6031746031746033e-05, | |
| "loss": 0.263, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_accuracy": 0.9854, | |
| "eval_loss": 0.0448877215385437, | |
| "eval_runtime": 57.7851, | |
| "eval_samples_per_second": 86.528, | |
| "eval_steps_per_second": 2.717, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.4444444444444444e-05, | |
| "loss": 0.2686, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.9826, | |
| "eval_loss": 0.05095283314585686, | |
| "eval_runtime": 57.6101, | |
| "eval_samples_per_second": 86.79, | |
| "eval_steps_per_second": 2.725, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1.2857142857142857e-05, | |
| "loss": 0.2705, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_accuracy": 0.9846, | |
| "eval_loss": 0.04833937808871269, | |
| "eval_runtime": 57.637, | |
| "eval_samples_per_second": 86.75, | |
| "eval_steps_per_second": 2.724, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1.126984126984127e-05, | |
| "loss": 0.2807, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.9854, | |
| "eval_loss": 0.04302287474274635, | |
| "eval_runtime": 57.6918, | |
| "eval_samples_per_second": 86.667, | |
| "eval_steps_per_second": 2.721, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 9.682539682539683e-06, | |
| "loss": 0.2583, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.9858, | |
| "eval_loss": 0.04517200589179993, | |
| "eval_runtime": 57.63, | |
| "eval_samples_per_second": 86.76, | |
| "eval_steps_per_second": 2.724, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 8.095238095238097e-06, | |
| "loss": 0.2346, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_accuracy": 0.9858, | |
| "eval_loss": 0.04349066689610481, | |
| "eval_runtime": 57.5683, | |
| "eval_samples_per_second": 86.853, | |
| "eval_steps_per_second": 2.727, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 6.507936507936509e-06, | |
| "loss": 0.2294, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_accuracy": 0.986, | |
| "eval_loss": 0.04344068095088005, | |
| "eval_runtime": 57.6423, | |
| "eval_samples_per_second": 86.742, | |
| "eval_steps_per_second": 2.724, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.920634920634921e-06, | |
| "loss": 0.2608, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_accuracy": 0.986, | |
| "eval_loss": 0.04333638399839401, | |
| "eval_runtime": 57.6441, | |
| "eval_samples_per_second": 86.739, | |
| "eval_steps_per_second": 2.724, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.2642, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_accuracy": 0.9866, | |
| "eval_loss": 0.04251227155327797, | |
| "eval_runtime": 59.7952, | |
| "eval_samples_per_second": 83.619, | |
| "eval_steps_per_second": 2.626, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.7460317460317462e-06, | |
| "loss": 0.2781, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_accuracy": 0.986, | |
| "eval_loss": 0.041707251220941544, | |
| "eval_runtime": 57.5428, | |
| "eval_samples_per_second": 86.892, | |
| "eval_steps_per_second": 2.728, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.5873015873015874e-07, | |
| "loss": 0.247, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9858, | |
| "eval_loss": 0.041396625339984894, | |
| "eval_runtime": 57.5657, | |
| "eval_samples_per_second": 86.857, | |
| "eval_steps_per_second": 2.727, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 351, | |
| "total_flos": 3.5202418590178345e+18, | |
| "train_loss": 0.46742354324910035, | |
| "train_runtime": 3479.7951, | |
| "train_samples_per_second": 12.932, | |
| "train_steps_per_second": 0.101 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 351, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 3.5202418590178345e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |