| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 18.0, | |
| "eval_steps": 500, | |
| "global_step": 3375, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3e-05, | |
| "loss": 2.0369, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6128205128205129, | |
| "eval_loss": 1.2953312397003174, | |
| "eval_runtime": 4.1114, | |
| "eval_samples_per_second": 121.614, | |
| "eval_steps_per_second": 15.323, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 27.2, | |
| "eval_f1": 37.44218938149973, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3264, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2821, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6146153846153846, | |
| "eval_loss": 1.2741013765335083, | |
| "eval_runtime": 4.8302, | |
| "eval_samples_per_second": 103.516, | |
| "eval_steps_per_second": 13.043, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 25.6, | |
| "eval_f1": 38.026666666666664, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2348, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1987, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6161538461538462, | |
| "eval_loss": 1.2714948654174805, | |
| "eval_runtime": 4.868, | |
| "eval_samples_per_second": 102.712, | |
| "eval_steps_per_second": 12.942, | |
| "step": 562 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 28.0, | |
| "eval_f1": 40.6962091503268, | |
| "step": 562 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1537, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 3e-05, | |
| "loss": 1.066, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6151282051282051, | |
| "eval_loss": 1.3011157512664795, | |
| "eval_runtime": 4.407, | |
| "eval_samples_per_second": 113.455, | |
| "eval_steps_per_second": 14.295, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_exact_match": 26.2, | |
| "eval_f1": 38.19190476190477, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 3e-05, | |
| "loss": 1.001, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9381, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6125641025641025, | |
| "eval_loss": 1.3728467226028442, | |
| "eval_runtime": 5.153, | |
| "eval_samples_per_second": 97.03, | |
| "eval_steps_per_second": 12.226, | |
| "step": 937 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_exact_match": 24.6, | |
| "eval_f1": 37.08666666666668, | |
| "step": 937 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8667, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8238, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6090769230769231, | |
| "eval_loss": 1.4598820209503174, | |
| "eval_runtime": 4.0994, | |
| "eval_samples_per_second": 121.97, | |
| "eval_steps_per_second": 15.368, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_exact_match": 23.0, | |
| "eval_f1": 35.324444444444445, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7461, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7289, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6064102564102564, | |
| "eval_loss": 1.5455493927001953, | |
| "eval_runtime": 4.1683, | |
| "eval_samples_per_second": 119.952, | |
| "eval_steps_per_second": 15.114, | |
| "step": 1312 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_exact_match": 21.8, | |
| "eval_f1": 34.57222222222222, | |
| "step": 1312 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6464, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6559, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6025641025641025, | |
| "eval_loss": 1.6359158754348755, | |
| "eval_runtime": 4.3161, | |
| "eval_samples_per_second": 115.844, | |
| "eval_steps_per_second": 14.596, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_exact_match": 20.6, | |
| "eval_f1": 33.70507936507938, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5733, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6006153846153847, | |
| "eval_loss": 1.7149169445037842, | |
| "eval_runtime": 4.1084, | |
| "eval_samples_per_second": 121.702, | |
| "eval_steps_per_second": 15.334, | |
| "step": 1687 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_exact_match": 22.0, | |
| "eval_f1": 33.980476190476196, | |
| "step": 1687 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5859, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5336, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5989230769230769, | |
| "eval_loss": 1.8006359338760376, | |
| "eval_runtime": 4.4039, | |
| "eval_samples_per_second": 113.535, | |
| "eval_steps_per_second": 14.305, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_exact_match": 20.4, | |
| "eval_f1": 32.68793650793652, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5379, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 10.67, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5116, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5982051282051282, | |
| "eval_loss": 1.8851306438446045, | |
| "eval_runtime": 4.8059, | |
| "eval_samples_per_second": 104.038, | |
| "eval_steps_per_second": 13.109, | |
| "step": 2062 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_exact_match": 19.8, | |
| "eval_f1": 31.627460317460333, | |
| "step": 2062 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5025, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 11.73, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4934, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5982051282051282, | |
| "eval_loss": 1.9262274503707886, | |
| "eval_runtime": 4.2024, | |
| "eval_samples_per_second": 118.98, | |
| "eval_steps_per_second": 14.992, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_exact_match": 19.2, | |
| "eval_f1": 30.531746031746042, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 12.27, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4827, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4823, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.5974358974358974, | |
| "eval_loss": 1.9412546157836914, | |
| "eval_runtime": 4.4921, | |
| "eval_samples_per_second": 111.306, | |
| "eval_steps_per_second": 14.025, | |
| "step": 2437 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_exact_match": 18.8, | |
| "eval_f1": 30.277619047619062, | |
| "step": 2437 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 3e-05, | |
| "loss": 0.467, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.87, | |
| "learning_rate": 3e-05, | |
| "loss": 0.47, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.5967179487179487, | |
| "eval_loss": 2.01212739944458, | |
| "eval_runtime": 4.0908, | |
| "eval_samples_per_second": 122.226, | |
| "eval_steps_per_second": 15.4, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_exact_match": 18.8, | |
| "eval_f1": 30.478571428571446, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.454, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 14.93, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4661, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.5968205128205128, | |
| "eval_loss": 2.0249853134155273, | |
| "eval_runtime": 4.0875, | |
| "eval_samples_per_second": 122.323, | |
| "eval_steps_per_second": 15.413, | |
| "step": 2812 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_exact_match": 20.6, | |
| "eval_f1": 32.526031746031755, | |
| "step": 2812 | |
| }, | |
| { | |
| "epoch": 15.47, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4423, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.462, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.598974358974359, | |
| "eval_loss": 1.9804810285568237, | |
| "eval_runtime": 4.0893, | |
| "eval_samples_per_second": 122.269, | |
| "eval_steps_per_second": 15.406, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_exact_match": 19.8, | |
| "eval_f1": 31.533809523809534, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.53, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4357, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.5975897435897436, | |
| "eval_loss": 2.0656166076660156, | |
| "eval_runtime": 4.1949, | |
| "eval_samples_per_second": 119.193, | |
| "eval_steps_per_second": 15.018, | |
| "step": 3187 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_exact_match": 18.8, | |
| "eval_f1": 31.295238095238105, | |
| "step": 3187 | |
| }, | |
| { | |
| "epoch": 17.07, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4511, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4348, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.5978974358974359, | |
| "eval_loss": 2.030846118927002, | |
| "eval_runtime": 4.1974, | |
| "eval_samples_per_second": 119.122, | |
| "eval_steps_per_second": 15.009, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_exact_match": 19.6, | |
| "eval_f1": 31.137619047619065, | |
| "step": 3375 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 9350, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1.2593645472238797e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |