| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 26.0, | |
| "eval_steps": 500, | |
| "global_step": 6500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0005, | |
| "loss": 2.3712, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0005, | |
| "loss": 1.8732, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.593873417721519, | |
| "eval_loss": 2.011091947555542, | |
| "eval_runtime": 4.861, | |
| "eval_samples_per_second": 102.859, | |
| "eval_steps_per_second": 12.96, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.32835692091264185, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0005, | |
| "loss": 1.7329, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0005, | |
| "loss": 1.6233, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0005, | |
| "loss": 1.6142, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6051139240506329, | |
| "eval_loss": 1.844284176826477, | |
| "eval_runtime": 4.8205, | |
| "eval_samples_per_second": 103.724, | |
| "eval_steps_per_second": 13.069, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.7215873015873017, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0005, | |
| "loss": 1.1625, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0005, | |
| "loss": 1.206, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6007341772151898, | |
| "eval_loss": 1.981817603111267, | |
| "eval_runtime": 4.671, | |
| "eval_samples_per_second": 107.044, | |
| "eval_steps_per_second": 13.487, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.4163636363636364, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0005, | |
| "loss": 1.0064, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0005, | |
| "loss": 0.8141, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0005, | |
| "loss": 0.8693, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.594126582278481, | |
| "eval_loss": 2.2100281715393066, | |
| "eval_runtime": 5.2488, | |
| "eval_samples_per_second": 95.26, | |
| "eval_steps_per_second": 12.003, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_exact_match": 0.2, | |
| "eval_f1": 1.559097642686839, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 0.0005, | |
| "loss": 0.5477, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.0005, | |
| "loss": 0.6023, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5910126582278481, | |
| "eval_loss": 2.375615119934082, | |
| "eval_runtime": 4.4156, | |
| "eval_samples_per_second": 113.235, | |
| "eval_steps_per_second": 14.268, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_exact_match": 0.4, | |
| "eval_f1": 2.41971862073413, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 0.0005, | |
| "loss": 0.514, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.0005, | |
| "loss": 0.4479, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.0005, | |
| "loss": 0.4717, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5895696202531645, | |
| "eval_loss": 2.542104482650757, | |
| "eval_runtime": 5.1447, | |
| "eval_samples_per_second": 97.188, | |
| "eval_steps_per_second": 12.246, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_exact_match": 9.8, | |
| "eval_f1": 17.19683736589267, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3723, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3938, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5891139240506329, | |
| "eval_loss": 2.658656597137451, | |
| "eval_runtime": 4.7175, | |
| "eval_samples_per_second": 105.987, | |
| "eval_steps_per_second": 13.354, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_exact_match": 3.6, | |
| "eval_f1": 8.743843476498272, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3727, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3538, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3697, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5872911392405064, | |
| "eval_loss": 2.7531518936157227, | |
| "eval_runtime": 4.7213, | |
| "eval_samples_per_second": 105.902, | |
| "eval_steps_per_second": 13.344, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.3733369257948794, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3431, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3617, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5870126582278481, | |
| "eval_loss": 2.7664403915405273, | |
| "eval_runtime": 4.7203, | |
| "eval_samples_per_second": 105.925, | |
| "eval_steps_per_second": 13.347, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.18, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3424, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3418, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3607, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5866835443037974, | |
| "eval_loss": 2.8513779640197754, | |
| "eval_runtime": 4.5658, | |
| "eval_samples_per_second": 109.511, | |
| "eval_steps_per_second": 13.798, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_exact_match": 15.8, | |
| "eval_f1": 24.03880563880564, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 0.0005, | |
| "loss": 0.339, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3414, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5860759493670886, | |
| "eval_loss": 2.8931641578674316, | |
| "eval_runtime": 5.1473, | |
| "eval_samples_per_second": 97.139, | |
| "eval_steps_per_second": 12.24, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 1.8004313814840132, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3302, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 11.6, | |
| "learning_rate": 0.0005, | |
| "loss": 0.329, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3439, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5854683544303797, | |
| "eval_loss": 2.9544754028320312, | |
| "eval_runtime": 4.5673, | |
| "eval_samples_per_second": 109.473, | |
| "eval_steps_per_second": 13.794, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_exact_match": 0.8, | |
| "eval_f1": 3.4374976657329603, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 12.4, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3182, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.0005, | |
| "loss": 0.335, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.5842784810126582, | |
| "eval_loss": 2.899094343185425, | |
| "eval_runtime": 4.7188, | |
| "eval_samples_per_second": 105.96, | |
| "eval_steps_per_second": 13.351, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_exact_match": 14.4, | |
| "eval_f1": 23.30256743256743, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 13.2, | |
| "learning_rate": 0.0005, | |
| "loss": 0.327, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3223, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3391, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.5839746835443038, | |
| "eval_loss": 2.879316806793213, | |
| "eval_runtime": 4.415, | |
| "eval_samples_per_second": 113.251, | |
| "eval_steps_per_second": 14.27, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_exact_match": 17.2, | |
| "eval_f1": 25.861362134597435, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3128, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 14.8, | |
| "learning_rate": 0.0005, | |
| "loss": 0.328, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.5851392405063292, | |
| "eval_loss": 2.8953680992126465, | |
| "eval_runtime": 4.7507, | |
| "eval_samples_per_second": 105.249, | |
| "eval_steps_per_second": 13.261, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_exact_match": 15.8, | |
| "eval_f1": 23.802868797868808, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 15.2, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3233, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3216, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3351, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5838481012658228, | |
| "eval_loss": 2.913999080657959, | |
| "eval_runtime": 4.5211, | |
| "eval_samples_per_second": 110.593, | |
| "eval_steps_per_second": 13.935, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_exact_match": 15.8, | |
| "eval_f1": 23.843056551291856, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3087, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 16.8, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3252, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.5825063291139241, | |
| "eval_loss": 2.929701089859009, | |
| "eval_runtime": 5.1507, | |
| "eval_samples_per_second": 97.074, | |
| "eval_steps_per_second": 12.231, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_exact_match": 17.6, | |
| "eval_f1": 25.533542568542583, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 17.2, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3152, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3161, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 0.0005, | |
| "loss": 0.332, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.5833924050632912, | |
| "eval_loss": 2.9811997413635254, | |
| "eval_runtime": 4.5707, | |
| "eval_samples_per_second": 109.392, | |
| "eval_steps_per_second": 13.783, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_exact_match": 14.8, | |
| "eval_f1": 23.04023215346745, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3089, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 18.8, | |
| "learning_rate": 0.0005, | |
| "loss": 0.324, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.5807848101265822, | |
| "eval_loss": 2.982343912124634, | |
| "eval_runtime": 4.5156, | |
| "eval_samples_per_second": 110.728, | |
| "eval_steps_per_second": 13.952, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_exact_match": 17.2, | |
| "eval_f1": 24.75057253204313, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3169, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 19.6, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3116, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3329, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5813164556962025, | |
| "eval_loss": 2.9419918060302734, | |
| "eval_runtime": 5.1584, | |
| "eval_samples_per_second": 96.929, | |
| "eval_steps_per_second": 12.213, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_exact_match": 17.0, | |
| "eval_f1": 24.681692914928213, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 20.4, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3052, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3242, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.5806075949367089, | |
| "eval_loss": 3.0366506576538086, | |
| "eval_runtime": 4.4135, | |
| "eval_samples_per_second": 113.288, | |
| "eval_steps_per_second": 14.274, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_exact_match": 15.8, | |
| "eval_f1": 24.507380005615303, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 21.2, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3169, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 21.6, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3104, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3271, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.5807594936708861, | |
| "eval_loss": 3.0456035137176514, | |
| "eval_runtime": 4.6241, | |
| "eval_samples_per_second": 108.13, | |
| "eval_steps_per_second": 13.624, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_exact_match": 16.8, | |
| "eval_f1": 24.95916155739686, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "learning_rate": 0.0005, | |
| "loss": 0.4232, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 22.8, | |
| "learning_rate": 0.0005, | |
| "loss": 2.1778, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.3285822784810127, | |
| "eval_loss": 10.320929527282715, | |
| "eval_runtime": 4.5651, | |
| "eval_samples_per_second": 109.526, | |
| "eval_steps_per_second": 13.8, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.0, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 23.2, | |
| "learning_rate": 0.0005, | |
| "loss": 10.0061, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 23.6, | |
| "learning_rate": 0.0005, | |
| "loss": 8.9834, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 0.0005, | |
| "loss": 9.433, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.3188607594936709, | |
| "eval_loss": 11.153959274291992, | |
| "eval_runtime": 5.1388, | |
| "eval_samples_per_second": 97.298, | |
| "eval_steps_per_second": 12.26, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.0, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 24.4, | |
| "learning_rate": 0.0005, | |
| "loss": 8.4619, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 24.8, | |
| "learning_rate": 0.0005, | |
| "loss": 7.8728, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.32336708860759494, | |
| "eval_loss": 10.043231964111328, | |
| "eval_runtime": 4.5666, | |
| "eval_samples_per_second": 109.491, | |
| "eval_steps_per_second": 13.796, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.0, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 25.2, | |
| "learning_rate": 0.0005, | |
| "loss": 9.8186, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 25.6, | |
| "learning_rate": 0.0005, | |
| "loss": 7.7736, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 0.0005, | |
| "loss": 8.6557, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.329873417721519, | |
| "eval_loss": 10.657929420471191, | |
| "eval_runtime": 4.5712, | |
| "eval_samples_per_second": 109.38, | |
| "eval_steps_per_second": 13.782, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.0, | |
| "step": 6500 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 12500, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 4.155661298769592e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |