| { | |
| "best_metric": 0.38235806339254613, | |
| "best_model_checkpoint": "/groups/claytonm/enoriega/kw_pubmed/kw_pubmed_1000_0.000006/checkpoint-108", | |
| "epoch": 3.283101045296167, | |
| "global_step": 148, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "eval_accuracy": 0.36533437826541276, | |
| "eval_loss": 4.0918426513671875, | |
| "eval_runtime": 16.5878, | |
| "eval_samples_per_second": 602.854, | |
| "eval_steps_per_second": 18.869, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 5.92e-06, | |
| "loss": 4.4572, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.36964472309299895, | |
| "eval_loss": 4.006958961486816, | |
| "eval_runtime": 16.4206, | |
| "eval_samples_per_second": 608.992, | |
| "eval_steps_per_second": 19.061, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 5.786666666666667e-06, | |
| "loss": 4.0734, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_accuracy": 0.37338906304423547, | |
| "eval_loss": 3.9496026039123535, | |
| "eval_runtime": 16.513, | |
| "eval_samples_per_second": 605.585, | |
| "eval_steps_per_second": 18.955, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 5.653333333333333e-06, | |
| "loss": 3.8884, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_accuracy": 0.37656739811912227, | |
| "eval_loss": 3.918774127960205, | |
| "eval_runtime": 16.5065, | |
| "eval_samples_per_second": 605.823, | |
| "eval_steps_per_second": 18.962, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 5.5200000000000005e-06, | |
| "loss": 3.8037, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_accuracy": 0.37495646116335773, | |
| "eval_loss": 3.9296493530273438, | |
| "eval_runtime": 16.4773, | |
| "eval_samples_per_second": 606.897, | |
| "eval_steps_per_second": 18.996, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_accuracy": 0.37891849529780564, | |
| "eval_loss": 3.903676748275757, | |
| "eval_runtime": 16.4879, | |
| "eval_samples_per_second": 606.507, | |
| "eval_steps_per_second": 18.984, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 5.386666666666667e-06, | |
| "loss": 3.7326, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.3787878787878788, | |
| "eval_loss": 3.8809974193573, | |
| "eval_runtime": 16.5092, | |
| "eval_samples_per_second": 605.724, | |
| "eval_steps_per_second": 18.959, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 5.2533333333333336e-06, | |
| "loss": 3.6766, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_accuracy": 0.3770463253221874, | |
| "eval_loss": 3.8815793991088867, | |
| "eval_runtime": 16.4832, | |
| "eval_samples_per_second": 606.677, | |
| "eval_steps_per_second": 18.989, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 5.12e-06, | |
| "loss": 3.6152, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_accuracy": 0.3797892720306513, | |
| "eval_loss": 3.8686890602111816, | |
| "eval_runtime": 16.4735, | |
| "eval_samples_per_second": 607.035, | |
| "eval_steps_per_second": 19.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.986666666666667e-06, | |
| "loss": 3.6143, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.3770463253221874, | |
| "eval_loss": 3.874457597732544, | |
| "eval_runtime": 16.4674, | |
| "eval_samples_per_second": 607.26, | |
| "eval_steps_per_second": 19.007, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.3777429467084639, | |
| "eval_loss": 3.872670888900757, | |
| "eval_runtime": 16.447, | |
| "eval_samples_per_second": 608.012, | |
| "eval_steps_per_second": 19.031, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 4.853333333333333e-06, | |
| "loss": 3.54, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_accuracy": 0.3747387669801463, | |
| "eval_loss": 3.882456064224243, | |
| "eval_runtime": 16.5148, | |
| "eval_samples_per_second": 605.517, | |
| "eval_steps_per_second": 18.953, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.72e-06, | |
| "loss": 4.1293, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.38079066527342387, | |
| "eval_loss": 3.8683102130889893, | |
| "eval_runtime": 16.4717, | |
| "eval_samples_per_second": 607.102, | |
| "eval_steps_per_second": 19.002, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.586666666666667e-06, | |
| "loss": 3.4379, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_accuracy": 0.37983281086729365, | |
| "eval_loss": 3.8646554946899414, | |
| "eval_runtime": 16.4967, | |
| "eval_samples_per_second": 606.182, | |
| "eval_steps_per_second": 18.973, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 4.453333333333334e-06, | |
| "loss": 3.4215, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_accuracy": 0.3801375827237896, | |
| "eval_loss": 3.8663816452026367, | |
| "eval_runtime": 16.4695, | |
| "eval_samples_per_second": 607.185, | |
| "eval_steps_per_second": 19.005, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_accuracy": 0.3804858934169279, | |
| "eval_loss": 3.8687705993652344, | |
| "eval_runtime": 16.462, | |
| "eval_samples_per_second": 607.461, | |
| "eval_steps_per_second": 19.014, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 4.32e-06, | |
| "loss": 3.3448, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_accuracy": 0.38140020898641586, | |
| "eval_loss": 3.876781463623047, | |
| "eval_runtime": 16.4781, | |
| "eval_samples_per_second": 606.866, | |
| "eval_steps_per_second": 18.995, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 4.186666666666667e-06, | |
| "loss": 3.3324, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_accuracy": 0.37717694183211425, | |
| "eval_loss": 3.913390636444092, | |
| "eval_runtime": 16.4943, | |
| "eval_samples_per_second": 606.268, | |
| "eval_steps_per_second": 18.976, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 4.053333333333333e-06, | |
| "loss": 3.3088, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_accuracy": 0.3799634273772205, | |
| "eval_loss": 3.882460832595825, | |
| "eval_runtime": 16.4932, | |
| "eval_samples_per_second": 606.311, | |
| "eval_steps_per_second": 18.978, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 3.92e-06, | |
| "loss": 3.2765, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_accuracy": 0.37787356321839083, | |
| "eval_loss": 3.8959219455718994, | |
| "eval_runtime": 16.5043, | |
| "eval_samples_per_second": 605.902, | |
| "eval_steps_per_second": 18.965, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.37913618948101707, | |
| "eval_loss": 3.9021382331848145, | |
| "eval_runtime": 16.4868, | |
| "eval_samples_per_second": 606.546, | |
| "eval_steps_per_second": 18.985, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.7866666666666667e-06, | |
| "loss": 3.2439, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.38148728665970044, | |
| "eval_loss": 3.8731861114501953, | |
| "eval_runtime": 16.4672, | |
| "eval_samples_per_second": 607.269, | |
| "eval_steps_per_second": 19.008, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 3.6533333333333336e-06, | |
| "loss": 3.2546, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_accuracy": 0.38044235458028564, | |
| "eval_loss": 3.8909058570861816, | |
| "eval_runtime": 16.4399, | |
| "eval_samples_per_second": 608.277, | |
| "eval_steps_per_second": 19.039, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 3.52e-06, | |
| "loss": 3.818, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_accuracy": 0.3787443399512365, | |
| "eval_loss": 3.917863368988037, | |
| "eval_runtime": 16.4722, | |
| "eval_samples_per_second": 607.083, | |
| "eval_steps_per_second": 19.002, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 3.3866666666666667e-06, | |
| "loss": 3.1567, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_accuracy": 0.3803988157436433, | |
| "eval_loss": 3.906223773956299, | |
| "eval_runtime": 16.5109, | |
| "eval_samples_per_second": 605.66, | |
| "eval_steps_per_second": 18.957, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_accuracy": 0.3793974225008708, | |
| "eval_loss": 3.9190924167633057, | |
| "eval_runtime": 16.4677, | |
| "eval_samples_per_second": 607.249, | |
| "eval_steps_per_second": 19.007, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 3.2533333333333337e-06, | |
| "loss": 3.1563, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_accuracy": 0.38235806339254613, | |
| "eval_loss": 3.8962931632995605, | |
| "eval_runtime": 16.4504, | |
| "eval_samples_per_second": 607.888, | |
| "eval_steps_per_second": 19.027, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 3.12e-06, | |
| "loss": 3.1417, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_accuracy": 0.37948450017415536, | |
| "eval_loss": 3.924445152282715, | |
| "eval_runtime": 16.4336, | |
| "eval_samples_per_second": 608.508, | |
| "eval_steps_per_second": 19.046, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 2.9866666666666667e-06, | |
| "loss": 3.1247, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_accuracy": 0.3790055729710902, | |
| "eval_loss": 3.9178736209869385, | |
| "eval_runtime": 16.4751, | |
| "eval_samples_per_second": 606.976, | |
| "eval_steps_per_second": 18.998, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 2.8533333333333333e-06, | |
| "loss": 3.1307, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "eval_accuracy": 0.3778300243817485, | |
| "eval_loss": 3.939385175704956, | |
| "eval_runtime": 16.4922, | |
| "eval_samples_per_second": 606.346, | |
| "eval_steps_per_second": 18.979, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "eval_accuracy": 0.38179205851619646, | |
| "eval_loss": 3.9248249530792236, | |
| "eval_runtime": 16.503, | |
| "eval_samples_per_second": 605.949, | |
| "eval_steps_per_second": 18.966, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 2.72e-06, | |
| "loss": 3.0893, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_accuracy": 0.37752525252525254, | |
| "eval_loss": 3.94762921333313, | |
| "eval_runtime": 16.4815, | |
| "eval_samples_per_second": 606.74, | |
| "eval_steps_per_second": 18.991, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 2.5866666666666667e-06, | |
| "loss": 3.0906, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_accuracy": 0.3811825148032045, | |
| "eval_loss": 3.9266726970672607, | |
| "eval_runtime": 16.4398, | |
| "eval_samples_per_second": 608.279, | |
| "eval_steps_per_second": 19.039, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 2.4533333333333337e-06, | |
| "loss": 3.0575, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_accuracy": 0.37813479623824453, | |
| "eval_loss": 3.937913417816162, | |
| "eval_runtime": 16.4796, | |
| "eval_samples_per_second": 606.812, | |
| "eval_steps_per_second": 18.993, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 2.32e-06, | |
| "loss": 3.6148, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "eval_accuracy": 0.3803988157436433, | |
| "eval_loss": 3.9217216968536377, | |
| "eval_runtime": 16.4193, | |
| "eval_samples_per_second": 609.04, | |
| "eval_steps_per_second": 19.063, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_accuracy": 0.37861372344130967, | |
| "eval_loss": 3.934666156768799, | |
| "eval_runtime": 16.4991, | |
| "eval_samples_per_second": 606.093, | |
| "eval_steps_per_second": 18.971, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 2.1866666666666668e-06, | |
| "loss": 3.035, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "eval_accuracy": 0.37944096133751304, | |
| "eval_loss": 3.9348745346069336, | |
| "eval_runtime": 16.4542, | |
| "eval_samples_per_second": 607.749, | |
| "eval_steps_per_second": 19.023, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "step": 148, | |
| "total_flos": 9.09876408404562e+16, | |
| "train_loss": 3.4590183206506677, | |
| "train_runtime": 5524.5303, | |
| "train_samples_per_second": 332.471, | |
| "train_steps_per_second": 0.041 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "eval_accuracy": 0.38079066527342387, | |
| "eval_loss": 3.901303291320801, | |
| "eval_runtime": 16.4247, | |
| "eval_samples_per_second": 608.84, | |
| "eval_steps_per_second": 19.057, | |
| "step": 148 | |
| } | |
| ], | |
| "max_steps": 225, | |
| "num_train_epochs": 5, | |
| "total_flos": 9.09876408404562e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |