| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 49.86666666666667, | |
| "eval_steps": 500, | |
| "global_step": 9350, | |
| "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 | |
| }, | |
| { | |
| "epoch": 18.13, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4436, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 18.67, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4331, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.598974358974359, | |
| "eval_loss": 2.062859058380127, | |
| "eval_runtime": 4.7948, | |
| "eval_samples_per_second": 104.279, | |
| "eval_steps_per_second": 13.139, | |
| "step": 3562 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_exact_match": 20.6, | |
| "eval_f1": 32.42380952380953, | |
| "step": 3562 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4338, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 19.73, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4341, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5983076923076923, | |
| "eval_loss": 2.0815415382385254, | |
| "eval_runtime": 4.4041, | |
| "eval_samples_per_second": 113.531, | |
| "eval_steps_per_second": 14.305, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_exact_match": 21.6, | |
| "eval_f1": 33.345238095238116, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 20.27, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4316, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 3e-05, | |
| "loss": 0.434, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.5968205128205128, | |
| "eval_loss": 2.1252710819244385, | |
| "eval_runtime": 4.3531, | |
| "eval_samples_per_second": 114.862, | |
| "eval_steps_per_second": 14.473, | |
| "step": 3937 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_exact_match": 21.4, | |
| "eval_f1": 32.32285714285715, | |
| "step": 3937 | |
| }, | |
| { | |
| "epoch": 21.33, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4255, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 21.87, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4335, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.5971282051282051, | |
| "eval_loss": 2.178868293762207, | |
| "eval_runtime": 4.1999, | |
| "eval_samples_per_second": 119.05, | |
| "eval_steps_per_second": 15.0, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_exact_match": 21.2, | |
| "eval_f1": 31.35571428571428, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4201, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 22.93, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4346, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.5952307692307692, | |
| "eval_loss": 2.145519971847534, | |
| "eval_runtime": 4.0955, | |
| "eval_samples_per_second": 122.084, | |
| "eval_steps_per_second": 15.383, | |
| "step": 4312 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_exact_match": 21.0, | |
| "eval_f1": 32.04714285714286, | |
| "step": 4312 | |
| }, | |
| { | |
| "epoch": 23.47, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4177, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4326, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5971282051282051, | |
| "eval_loss": 2.1990349292755127, | |
| "eval_runtime": 4.0925, | |
| "eval_samples_per_second": 122.176, | |
| "eval_steps_per_second": 15.394, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_exact_match": 19.6, | |
| "eval_f1": 30.60428571428572, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 24.53, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4139, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.5976410256410256, | |
| "eval_loss": 2.1889517307281494, | |
| "eval_runtime": 4.1957, | |
| "eval_samples_per_second": 119.17, | |
| "eval_steps_per_second": 15.015, | |
| "step": 4687 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_exact_match": 20.4, | |
| "eval_f1": 31.052554112554116, | |
| "step": 4687 | |
| }, | |
| { | |
| "epoch": 25.07, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4268, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 25.6, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4139, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.5967692307692307, | |
| "eval_loss": 2.1938648223876953, | |
| "eval_runtime": 4.417, | |
| "eval_samples_per_second": 113.198, | |
| "eval_steps_per_second": 14.263, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_exact_match": 20.4, | |
| "eval_f1": 32.143030303030315, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 26.13, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4222, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4162, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.5964615384615385, | |
| "eval_loss": 2.218991279602051, | |
| "eval_runtime": 4.2318, | |
| "eval_samples_per_second": 118.152, | |
| "eval_steps_per_second": 14.887, | |
| "step": 5062 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_exact_match": 21.2, | |
| "eval_f1": 31.969523809523817, | |
| "step": 5062 | |
| }, | |
| { | |
| "epoch": 27.2, | |
| "learning_rate": 3e-05, | |
| "loss": 0.419, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 27.73, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4177, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5954871794871794, | |
| "eval_loss": 2.2781052589416504, | |
| "eval_runtime": 4.1923, | |
| "eval_samples_per_second": 119.265, | |
| "eval_steps_per_second": 15.027, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_exact_match": 19.8, | |
| "eval_f1": 30.879220779220788, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 28.27, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4143, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4173, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.5975897435897436, | |
| "eval_loss": 2.2680609226226807, | |
| "eval_runtime": 4.401, | |
| "eval_samples_per_second": 113.61, | |
| "eval_steps_per_second": 14.315, | |
| "step": 5437 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_exact_match": 20.6, | |
| "eval_f1": 31.83809523809525, | |
| "step": 5437 | |
| }, | |
| { | |
| "epoch": 29.33, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4117, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 29.87, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4187, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.5958974358974359, | |
| "eval_loss": 2.2996103763580322, | |
| "eval_runtime": 4.199, | |
| "eval_samples_per_second": 119.077, | |
| "eval_steps_per_second": 15.004, | |
| "step": 5625 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_exact_match": 20.4, | |
| "eval_f1": 32.61430014430016, | |
| "step": 5625 | |
| }, | |
| { | |
| "epoch": 30.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.408, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 30.93, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4199, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.5981025641025641, | |
| "eval_loss": 2.239492177963257, | |
| "eval_runtime": 4.8298, | |
| "eval_samples_per_second": 103.523, | |
| "eval_steps_per_second": 13.044, | |
| "step": 5812 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_exact_match": 20.2, | |
| "eval_f1": 31.465887445887457, | |
| "step": 5812 | |
| }, | |
| { | |
| "epoch": 31.47, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4055, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4213, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5957435897435898, | |
| "eval_loss": 2.299053907394409, | |
| "eval_runtime": 4.0871, | |
| "eval_samples_per_second": 122.335, | |
| "eval_steps_per_second": 15.414, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_exact_match": 20.0, | |
| "eval_f1": 31.56428571428572, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 32.53, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4015, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.5952307692307692, | |
| "eval_loss": 2.3223156929016113, | |
| "eval_runtime": 4.1951, | |
| "eval_samples_per_second": 119.187, | |
| "eval_steps_per_second": 15.018, | |
| "step": 6187 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_exact_match": 21.2, | |
| "eval_f1": 32.325714285714284, | |
| "step": 6187 | |
| }, | |
| { | |
| "epoch": 33.07, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4161, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 33.6, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4058, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.5956923076923077, | |
| "eval_loss": 2.3266167640686035, | |
| "eval_runtime": 4.0888, | |
| "eval_samples_per_second": 122.286, | |
| "eval_steps_per_second": 15.408, | |
| "step": 6375 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_exact_match": 20.2, | |
| "eval_f1": 31.896666666666672, | |
| "step": 6375 | |
| }, | |
| { | |
| "epoch": 34.13, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4118, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 34.67, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4056, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.5946153846153847, | |
| "eval_loss": 2.3779330253601074, | |
| "eval_runtime": 4.1964, | |
| "eval_samples_per_second": 119.15, | |
| "eval_steps_per_second": 15.013, | |
| "step": 6562 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_exact_match": 20.6, | |
| "eval_f1": 32.23207792207793, | |
| "step": 6562 | |
| }, | |
| { | |
| "epoch": 35.2, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4103, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 35.73, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4078, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.5950769230769231, | |
| "eval_loss": 2.345280408859253, | |
| "eval_runtime": 5.1209, | |
| "eval_samples_per_second": 97.64, | |
| "eval_steps_per_second": 12.303, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_exact_match": 21.0, | |
| "eval_f1": 32.03683982683983, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 36.27, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4048, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 36.8, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4097, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.5965128205128205, | |
| "eval_loss": 2.3379180431365967, | |
| "eval_runtime": 4.1922, | |
| "eval_samples_per_second": 119.268, | |
| "eval_steps_per_second": 15.028, | |
| "step": 6937 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_exact_match": 20.6, | |
| "eval_f1": 32.22937950937952, | |
| "step": 6937 | |
| }, | |
| { | |
| "epoch": 37.33, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4059, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 37.87, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4105, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.5969230769230769, | |
| "eval_loss": 2.362399101257324, | |
| "eval_runtime": 4.1963, | |
| "eval_samples_per_second": 119.153, | |
| "eval_steps_per_second": 15.013, | |
| "step": 7125 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_exact_match": 20.6, | |
| "eval_f1": 31.836363636363643, | |
| "step": 7125 | |
| }, | |
| { | |
| "epoch": 38.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4013, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 38.93, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4116, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.5961538461538461, | |
| "eval_loss": 2.3846352100372314, | |
| "eval_runtime": 4.0932, | |
| "eval_samples_per_second": 122.154, | |
| "eval_steps_per_second": 15.391, | |
| "step": 7312 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_exact_match": 20.0, | |
| "eval_f1": 31.287619047619053, | |
| "step": 7312 | |
| }, | |
| { | |
| "epoch": 39.47, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3993, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4121, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.5944615384615385, | |
| "eval_loss": 2.3748481273651123, | |
| "eval_runtime": 4.9211, | |
| "eval_samples_per_second": 101.603, | |
| "eval_steps_per_second": 12.802, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_exact_match": 20.4, | |
| "eval_f1": 32.21826839826841, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 40.53, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3973, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.5955897435897436, | |
| "eval_loss": 2.3797402381896973, | |
| "eval_runtime": 4.0885, | |
| "eval_samples_per_second": 122.293, | |
| "eval_steps_per_second": 15.409, | |
| "step": 7687 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_exact_match": 21.8, | |
| "eval_f1": 33.78445887445889, | |
| "step": 7687 | |
| }, | |
| { | |
| "epoch": 41.07, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4092, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 41.6, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3985, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.5966666666666667, | |
| "eval_loss": 2.359941005706787, | |
| "eval_runtime": 4.319, | |
| "eval_samples_per_second": 115.769, | |
| "eval_steps_per_second": 14.587, | |
| "step": 7875 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_exact_match": 22.6, | |
| "eval_f1": 34.189696969696975, | |
| "step": 7875 | |
| }, | |
| { | |
| "epoch": 42.13, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4075, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 42.67, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4014, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.5970769230769231, | |
| "eval_loss": 2.347489356994629, | |
| "eval_runtime": 4.0917, | |
| "eval_samples_per_second": 122.197, | |
| "eval_steps_per_second": 15.397, | |
| "step": 8062 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_exact_match": 21.4, | |
| "eval_f1": 33.01255411255412, | |
| "step": 8062 | |
| }, | |
| { | |
| "epoch": 43.2, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4032, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 43.73, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4032, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.5987179487179487, | |
| "eval_loss": 2.3937387466430664, | |
| "eval_runtime": 4.3012, | |
| "eval_samples_per_second": 116.246, | |
| "eval_steps_per_second": 14.647, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_exact_match": 20.6, | |
| "eval_f1": 31.90666666666667, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 44.27, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 44.8, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4028, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.5967179487179487, | |
| "eval_loss": 2.386258363723755, | |
| "eval_runtime": 4.1018, | |
| "eval_samples_per_second": 121.897, | |
| "eval_steps_per_second": 15.359, | |
| "step": 8437 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_exact_match": 20.2, | |
| "eval_f1": 31.19238095238096, | |
| "step": 8437 | |
| }, | |
| { | |
| "epoch": 45.33, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3992, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 45.87, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4027, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.5956410256410256, | |
| "eval_loss": 2.4195497035980225, | |
| "eval_runtime": 4.0913, | |
| "eval_samples_per_second": 122.211, | |
| "eval_steps_per_second": 15.399, | |
| "step": 8625 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_exact_match": 21.4, | |
| "eval_f1": 32.3282683982684, | |
| "step": 8625 | |
| }, | |
| { | |
| "epoch": 46.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3976, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 46.93, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4046, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.597025641025641, | |
| "eval_loss": 2.3832449913024902, | |
| "eval_runtime": 4.1919, | |
| "eval_samples_per_second": 119.277, | |
| "eval_steps_per_second": 15.029, | |
| "step": 8812 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_exact_match": 20.8, | |
| "eval_f1": 32.23841269841271, | |
| "step": 8812 | |
| }, | |
| { | |
| "epoch": 47.47, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3948, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4067, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.5973333333333334, | |
| "eval_loss": 2.380531072616577, | |
| "eval_runtime": 4.1918, | |
| "eval_samples_per_second": 119.282, | |
| "eval_steps_per_second": 15.03, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_exact_match": 22.4, | |
| "eval_f1": 32.9035064935065, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 48.53, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3923, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.5956923076923077, | |
| "eval_loss": 2.4460361003875732, | |
| "eval_runtime": 4.8374, | |
| "eval_samples_per_second": 103.362, | |
| "eval_steps_per_second": 13.024, | |
| "step": 9187 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_exact_match": 21.6, | |
| "eval_f1": 32.555714285714295, | |
| "step": 9187 | |
| }, | |
| { | |
| "epoch": 49.07, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4045, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 49.6, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3949, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 49.87, | |
| "eval_accuracy": 0.5965641025641025, | |
| "eval_loss": 2.4443256855010986, | |
| "eval_runtime": 4.0946, | |
| "eval_samples_per_second": 122.113, | |
| "eval_steps_per_second": 15.386, | |
| "step": 9350 | |
| }, | |
| { | |
| "epoch": 49.87, | |
| "eval_exact_match": 21.8, | |
| "eval_f1": 32.88666666666667, | |
| "step": 9350 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 9350, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 3.489165488087368e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |