| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 750000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1e-05, | |
| "loss": 3.3081, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.83344888888889e-06, | |
| "loss": 0.6857, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_accuracy": 0.5564, | |
| "eval_loss": 0.6798537373542786, | |
| "eval_runtime": 9.7283, | |
| "eval_samples_per_second": 1027.927, | |
| "eval_steps_per_second": 16.138, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.666862222222223e-06, | |
| "loss": 0.6804, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.5607, | |
| "eval_loss": 0.677202045917511, | |
| "eval_runtime": 9.8699, | |
| "eval_samples_per_second": 1013.18, | |
| "eval_steps_per_second": 15.907, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.500275555555555e-06, | |
| "loss": 0.679, | |
| "step": 56250 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_accuracy": 0.5559, | |
| "eval_loss": 0.6789325475692749, | |
| "eval_runtime": 9.7877, | |
| "eval_samples_per_second": 1021.686, | |
| "eval_steps_per_second": 16.04, | |
| "step": 56250 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.33368888888889e-06, | |
| "loss": 0.6785, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.5639, | |
| "eval_loss": 0.6760360598564148, | |
| "eval_runtime": 9.8616, | |
| "eval_samples_per_second": 1014.039, | |
| "eval_steps_per_second": 15.92, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 9.167093333333335e-06, | |
| "loss": 0.6777, | |
| "step": 93750 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_accuracy": 0.5672, | |
| "eval_loss": 0.6752009987831116, | |
| "eval_runtime": 9.8899, | |
| "eval_samples_per_second": 1011.135, | |
| "eval_steps_per_second": 15.875, | |
| "step": 93750 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 9.000515555555556e-06, | |
| "loss": 0.6771, | |
| "step": 112500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_accuracy": 0.5647, | |
| "eval_loss": 0.6761014461517334, | |
| "eval_runtime": 9.9903, | |
| "eval_samples_per_second": 1000.97, | |
| "eval_steps_per_second": 15.715, | |
| "step": 112500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 8.833928888888889e-06, | |
| "loss": 0.6765, | |
| "step": 131250 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_accuracy": 0.5665, | |
| "eval_loss": 0.6758425235748291, | |
| "eval_runtime": 9.7731, | |
| "eval_samples_per_second": 1023.22, | |
| "eval_steps_per_second": 16.065, | |
| "step": 131250 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 8.667333333333334e-06, | |
| "loss": 0.6765, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.5669, | |
| "eval_loss": 0.6743916869163513, | |
| "eval_runtime": 9.7984, | |
| "eval_samples_per_second": 1020.577, | |
| "eval_steps_per_second": 16.023, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 8.50073777777778e-06, | |
| "loss": 0.6762, | |
| "step": 168750 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_accuracy": 0.5682, | |
| "eval_loss": 0.6749153137207031, | |
| "eval_runtime": 9.9597, | |
| "eval_samples_per_second": 1004.049, | |
| "eval_steps_per_second": 15.764, | |
| "step": 168750 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 8.334142222222223e-06, | |
| "loss": 0.676, | |
| "step": 187500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.5673, | |
| "eval_loss": 0.6741730570793152, | |
| "eval_runtime": 9.9295, | |
| "eval_samples_per_second": 1007.098, | |
| "eval_steps_per_second": 15.811, | |
| "step": 187500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 8.167546666666666e-06, | |
| "loss": 0.676, | |
| "step": 206250 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_accuracy": 0.5708, | |
| "eval_loss": 0.6741572618484497, | |
| "eval_runtime": 9.6873, | |
| "eval_samples_per_second": 1032.278, | |
| "eval_steps_per_second": 16.207, | |
| "step": 206250 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.000951111111112e-06, | |
| "loss": 0.6758, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.569, | |
| "eval_loss": 0.6750027537345886, | |
| "eval_runtime": 10.218, | |
| "eval_samples_per_second": 978.668, | |
| "eval_steps_per_second": 15.365, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 7.834364444444446e-06, | |
| "loss": 0.6755, | |
| "step": 243750 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.5688, | |
| "eval_loss": 0.6740711331367493, | |
| "eval_runtime": 10.1203, | |
| "eval_samples_per_second": 988.117, | |
| "eval_steps_per_second": 15.513, | |
| "step": 243750 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 7.66776e-06, | |
| "loss": 0.6758, | |
| "step": 262500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_accuracy": 0.5676, | |
| "eval_loss": 0.6746249198913574, | |
| "eval_runtime": 9.7352, | |
| "eval_samples_per_second": 1027.2, | |
| "eval_steps_per_second": 16.127, | |
| "step": 262500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 7.501173333333334e-06, | |
| "loss": 0.6754, | |
| "step": 281250 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_accuracy": 0.5728, | |
| "eval_loss": 0.673089861869812, | |
| "eval_runtime": 9.866, | |
| "eval_samples_per_second": 1013.582, | |
| "eval_steps_per_second": 15.913, | |
| "step": 281250 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 7.334586666666668e-06, | |
| "loss": 0.6752, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.5678, | |
| "eval_loss": 0.6737684607505798, | |
| "eval_runtime": 10.0756, | |
| "eval_samples_per_second": 992.497, | |
| "eval_steps_per_second": 15.582, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.167982222222223e-06, | |
| "loss": 0.6753, | |
| "step": 318750 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_accuracy": 0.5744, | |
| "eval_loss": 0.6733357310295105, | |
| "eval_runtime": 9.7112, | |
| "eval_samples_per_second": 1029.741, | |
| "eval_steps_per_second": 16.167, | |
| "step": 318750 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 7.0013777777777784e-06, | |
| "loss": 0.6752, | |
| "step": 337500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.5692, | |
| "eval_loss": 0.6732030510902405, | |
| "eval_runtime": 9.8389, | |
| "eval_samples_per_second": 1016.377, | |
| "eval_steps_per_second": 15.957, | |
| "step": 337500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 6.834764444444445e-06, | |
| "loss": 0.6753, | |
| "step": 356250 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.5714, | |
| "eval_loss": 0.6738024950027466, | |
| "eval_runtime": 9.6915, | |
| "eval_samples_per_second": 1031.834, | |
| "eval_steps_per_second": 16.2, | |
| "step": 356250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.66816888888889e-06, | |
| "loss": 0.6749, | |
| "step": 375000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5711, | |
| "eval_loss": 0.6741089820861816, | |
| "eval_runtime": 10.1172, | |
| "eval_samples_per_second": 988.413, | |
| "eval_steps_per_second": 15.518, | |
| "step": 375000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 6.501564444444445e-06, | |
| "loss": 0.6732, | |
| "step": 393750 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_accuracy": 0.572, | |
| "eval_loss": 0.6729293465614319, | |
| "eval_runtime": 9.7745, | |
| "eval_samples_per_second": 1023.073, | |
| "eval_steps_per_second": 16.062, | |
| "step": 393750 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 6.33496e-06, | |
| "loss": 0.6734, | |
| "step": 412500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.5715, | |
| "eval_loss": 0.6731555461883545, | |
| "eval_runtime": 9.9359, | |
| "eval_samples_per_second": 1006.454, | |
| "eval_steps_per_second": 15.801, | |
| "step": 412500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 6.168355555555556e-06, | |
| "loss": 0.6733, | |
| "step": 431250 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.5726, | |
| "eval_loss": 0.6726440191268921, | |
| "eval_runtime": 9.7596, | |
| "eval_samples_per_second": 1024.628, | |
| "eval_steps_per_second": 16.087, | |
| "step": 431250 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 6.001751111111111e-06, | |
| "loss": 0.6734, | |
| "step": 450000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 0.5725, | |
| "eval_loss": 0.6719476580619812, | |
| "eval_runtime": 10.1776, | |
| "eval_samples_per_second": 982.551, | |
| "eval_steps_per_second": 15.426, | |
| "step": 450000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 5.8351555555555565e-06, | |
| "loss": 0.6732, | |
| "step": 468750 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_accuracy": 0.5721, | |
| "eval_loss": 0.6720800995826721, | |
| "eval_runtime": 9.8195, | |
| "eval_samples_per_second": 1018.38, | |
| "eval_steps_per_second": 15.989, | |
| "step": 468750 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 5.668542222222223e-06, | |
| "loss": 0.6732, | |
| "step": 487500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.5719, | |
| "eval_loss": 0.6730443239212036, | |
| "eval_runtime": 9.8464, | |
| "eval_samples_per_second": 1015.602, | |
| "eval_steps_per_second": 15.945, | |
| "step": 487500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 5.501928888888889e-06, | |
| "loss": 0.6732, | |
| "step": 506250 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_accuracy": 0.5731, | |
| "eval_loss": 0.67209392786026, | |
| "eval_runtime": 9.9162, | |
| "eval_samples_per_second": 1008.449, | |
| "eval_steps_per_second": 15.833, | |
| "step": 506250 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 5.335342222222223e-06, | |
| "loss": 0.6733, | |
| "step": 525000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_accuracy": 0.5715, | |
| "eval_loss": 0.6724963188171387, | |
| "eval_runtime": 9.8765, | |
| "eval_samples_per_second": 1012.507, | |
| "eval_steps_per_second": 15.896, | |
| "step": 525000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 5.168737777777778e-06, | |
| "loss": 0.6733, | |
| "step": 543750 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_accuracy": 0.5757, | |
| "eval_loss": 0.6714832186698914, | |
| "eval_runtime": 10.1309, | |
| "eval_samples_per_second": 987.074, | |
| "eval_steps_per_second": 15.497, | |
| "step": 543750 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 5.002133333333333e-06, | |
| "loss": 0.673, | |
| "step": 562500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_accuracy": 0.5762, | |
| "eval_loss": 0.6717364192008972, | |
| "eval_runtime": 9.8941, | |
| "eval_samples_per_second": 1010.706, | |
| "eval_steps_per_second": 15.868, | |
| "step": 562500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 4.835546666666668e-06, | |
| "loss": 0.6732, | |
| "step": 581250 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_accuracy": 0.5735, | |
| "eval_loss": 0.672535240650177, | |
| "eval_runtime": 9.8695, | |
| "eval_samples_per_second": 1013.226, | |
| "eval_steps_per_second": 15.908, | |
| "step": 581250 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4.668942222222223e-06, | |
| "loss": 0.6733, | |
| "step": 600000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_accuracy": 0.5729, | |
| "eval_loss": 0.6723589897155762, | |
| "eval_runtime": 9.9575, | |
| "eval_samples_per_second": 1004.27, | |
| "eval_steps_per_second": 15.767, | |
| "step": 600000 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 4.502346666666667e-06, | |
| "loss": 0.6729, | |
| "step": 618750 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_accuracy": 0.5714, | |
| "eval_loss": 0.6716250777244568, | |
| "eval_runtime": 10.0075, | |
| "eval_samples_per_second": 999.255, | |
| "eval_steps_per_second": 15.688, | |
| "step": 618750 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 4.335742222222223e-06, | |
| "loss": 0.6729, | |
| "step": 637500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_accuracy": 0.576, | |
| "eval_loss": 0.6714919805526733, | |
| "eval_runtime": 9.9861, | |
| "eval_samples_per_second": 1001.39, | |
| "eval_steps_per_second": 15.722, | |
| "step": 637500 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 4.169155555555556e-06, | |
| "loss": 0.6726, | |
| "step": 656250 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_accuracy": 0.5785, | |
| "eval_loss": 0.6716631054878235, | |
| "eval_runtime": 9.8047, | |
| "eval_samples_per_second": 1019.923, | |
| "eval_steps_per_second": 16.013, | |
| "step": 656250 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 4.00256e-06, | |
| "loss": 0.6723, | |
| "step": 675000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_accuracy": 0.5791, | |
| "eval_loss": 0.6703997850418091, | |
| "eval_runtime": 9.8558, | |
| "eval_samples_per_second": 1014.636, | |
| "eval_steps_per_second": 15.93, | |
| "step": 675000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 3.835955555555556e-06, | |
| "loss": 0.6728, | |
| "step": 693750 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.5763, | |
| "eval_loss": 0.6714270710945129, | |
| "eval_runtime": 9.7423, | |
| "eval_samples_per_second": 1026.453, | |
| "eval_steps_per_second": 16.115, | |
| "step": 693750 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 3.6693511111111114e-06, | |
| "loss": 0.6724, | |
| "step": 712500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.5795, | |
| "eval_loss": 0.6705731749534607, | |
| "eval_runtime": 10.1221, | |
| "eval_samples_per_second": 987.933, | |
| "eval_steps_per_second": 15.511, | |
| "step": 712500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 3.502746666666667e-06, | |
| "loss": 0.6724, | |
| "step": 731250 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_accuracy": 0.5789, | |
| "eval_loss": 0.6716538071632385, | |
| "eval_runtime": 9.891, | |
| "eval_samples_per_second": 1011.024, | |
| "eval_steps_per_second": 15.873, | |
| "step": 731250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.3361422222222222e-06, | |
| "loss": 0.6727, | |
| "step": 750000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5789, | |
| "eval_loss": 0.6710340976715088, | |
| "eval_runtime": 9.6333, | |
| "eval_samples_per_second": 1038.067, | |
| "eval_steps_per_second": 16.298, | |
| "step": 750000 | |
| } | |
| ], | |
| "max_steps": 1125000, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.2542244576561201e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |