| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 13.0, | |
| "eval_steps": 500, | |
| "global_step": 6877, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3e-05, | |
| "loss": 1.7194, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4369, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4257, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3947, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3948, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6132307692307692, | |
| "eval_loss": 1.3087366819381714, | |
| "eval_runtime": 4.3553, | |
| "eval_samples_per_second": 114.803, | |
| "eval_steps_per_second": 14.465, | |
| "step": 529 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 24.4, | |
| "eval_f1": 37.614420394420414, | |
| "step": 529 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3927, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3644, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3791, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3627, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3789, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6146153846153846, | |
| "eval_loss": 1.289727807044983, | |
| "eval_runtime": 4.4281, | |
| "eval_samples_per_second": 112.915, | |
| "eval_steps_per_second": 14.227, | |
| "step": 1058 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 23.2, | |
| "eval_f1": 35.56775335775336, | |
| "step": 1058 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3605, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3306, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3318, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3491, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3259, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6178974358974358, | |
| "eval_loss": 1.2849316596984863, | |
| "eval_runtime": 4.875, | |
| "eval_samples_per_second": 102.565, | |
| "eval_steps_per_second": 12.923, | |
| "step": 1587 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 25.2, | |
| "eval_f1": 37.32698412698414, | |
| "step": 1587 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3255, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 3e-05, | |
| "loss": 1.28, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2995, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2991, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3024, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2853, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.615948717948718, | |
| "eval_loss": 1.316924810409546, | |
| "eval_runtime": 5.3385, | |
| "eval_samples_per_second": 93.659, | |
| "eval_steps_per_second": 11.801, | |
| "step": 2116 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_exact_match": 23.8, | |
| "eval_f1": 36.284920634920645, | |
| "step": 2116 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2575, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2261, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2536, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2376, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2556, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6131794871794872, | |
| "eval_loss": 1.3531945943832397, | |
| "eval_runtime": 4.4382, | |
| "eval_samples_per_second": 112.659, | |
| "eval_steps_per_second": 14.195, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_exact_match": 25.8, | |
| "eval_f1": 37.67571428571429, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2195, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2003, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1864, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2231, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1972, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6125641025641025, | |
| "eval_loss": 1.4135066270828247, | |
| "eval_runtime": 5.0312, | |
| "eval_samples_per_second": 99.38, | |
| "eval_steps_per_second": 12.522, | |
| "step": 3174 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_exact_match": 24.0, | |
| "eval_f1": 36.140952380952385, | |
| "step": 3174 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1889, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1471, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1509, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1654, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1716, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1839, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6081025641025641, | |
| "eval_loss": 1.500714898109436, | |
| "eval_runtime": 4.4677, | |
| "eval_samples_per_second": 111.916, | |
| "eval_steps_per_second": 14.101, | |
| "step": 3703 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_exact_match": 22.8, | |
| "eval_f1": 33.44777777777779, | |
| "step": 3703 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1059, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1146, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1121, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1286, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1334, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6074358974358974, | |
| "eval_loss": 1.524243950843811, | |
| "eval_runtime": 4.2161, | |
| "eval_samples_per_second": 118.592, | |
| "eval_steps_per_second": 14.943, | |
| "step": 4232 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_exact_match": 23.2, | |
| "eval_f1": 34.507792207792214, | |
| "step": 4232 | |
| }, | |
| { | |
| "epoch": 8.13, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1072, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0626, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0842, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0812, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0966, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5802564102564103, | |
| "eval_loss": 1.6106736660003662, | |
| "eval_runtime": 4.2194, | |
| "eval_samples_per_second": 118.5, | |
| "eval_steps_per_second": 14.931, | |
| "step": 4761 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_exact_match": 22.8, | |
| "eval_f1": 34.45666666666666, | |
| "step": 4761 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0755, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0189, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0565, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0635, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 9.83, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0485, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6049230769230769, | |
| "eval_loss": 1.6749216318130493, | |
| "eval_runtime": 4.8521, | |
| "eval_samples_per_second": 103.048, | |
| "eval_steps_per_second": 12.984, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_exact_match": 22.8, | |
| "eval_f1": 34.3535064935065, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0585, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 10.21, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0098, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9935, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0178, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 10.78, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0217, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "learning_rate": 3e-05, | |
| "loss": 1.021, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6014871794871794, | |
| "eval_loss": 1.7323671579360962, | |
| "eval_runtime": 4.4139, | |
| "eval_samples_per_second": 113.28, | |
| "eval_steps_per_second": 14.273, | |
| "step": 5819 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_exact_match": 22.4, | |
| "eval_f1": 34.060490620490626, | |
| "step": 5819 | |
| }, | |
| { | |
| "epoch": 11.15, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9788, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 11.34, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9799, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 11.53, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9749, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 3e-05, | |
| "loss": 0.981, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 11.91, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9918, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6007179487179487, | |
| "eval_loss": 1.7632156610488892, | |
| "eval_runtime": 4.4251, | |
| "eval_samples_per_second": 112.992, | |
| "eval_steps_per_second": 14.237, | |
| "step": 6348 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_exact_match": 22.4, | |
| "eval_f1": 33.017142857142865, | |
| "step": 6348 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 3e-05, | |
| "loss": 0.967, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 12.29, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9341, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9656, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9519, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 3e-05, | |
| "loss": 0.947, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.6010769230769231, | |
| "eval_loss": 1.8302537202835083, | |
| "eval_runtime": 4.849, | |
| "eval_samples_per_second": 103.115, | |
| "eval_steps_per_second": 12.992, | |
| "step": 6877 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_exact_match": 21.6, | |
| "eval_f1": 32.89746031746034, | |
| "step": 6877 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 26450, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1.1617845809994793e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |