| { | |
| "best_metric": 37.6244, | |
| "best_model_checkpoint": "./save/multi_news_longt5_base_weighted_ce/checkpoint-32000", | |
| "epoch": 3.0, | |
| "global_step": 33723, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.851733238442606e-05, | |
| "loss": 5.4403, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.703466476885213e-05, | |
| "loss": 4.2866, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 425.48674613058176, | |
| "eval_loss": 2.7076995372772217, | |
| "eval_rouge1": 23.4899, | |
| "eval_rouge2": 6.1612, | |
| "eval_rougeL": 15.6671, | |
| "eval_rougeLsum": 20.2981, | |
| "eval_runtime": 3496.9942, | |
| "eval_samples_per_second": 1.607, | |
| "eval_steps_per_second": 0.05, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.555199715327818e-05, | |
| "loss": 3.9482, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.406932953770424e-05, | |
| "loss": 3.7678, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 345.54794520547944, | |
| "eval_loss": 2.533926248550415, | |
| "eval_rouge1": 29.7909, | |
| "eval_rouge2": 9.3938, | |
| "eval_rougeL": 18.5202, | |
| "eval_rougeLsum": 26.3629, | |
| "eval_runtime": 3480.628, | |
| "eval_samples_per_second": 1.615, | |
| "eval_steps_per_second": 0.051, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.25866619221303e-05, | |
| "loss": 3.6022, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 9.110399430655636e-05, | |
| "loss": 3.5226, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 404.8505603985056, | |
| "eval_loss": 2.408491849899292, | |
| "eval_rouge1": 30.3236, | |
| "eval_rouge2": 9.683, | |
| "eval_rougeL": 18.7128, | |
| "eval_rougeLsum": 26.801, | |
| "eval_runtime": 3552.594, | |
| "eval_samples_per_second": 1.582, | |
| "eval_steps_per_second": 0.05, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 8.962132669098243e-05, | |
| "loss": 3.4401, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 8.813865907540848e-05, | |
| "loss": 3.3831, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 285.1569115815691, | |
| "eval_loss": 2.3446125984191895, | |
| "eval_rouge1": 33.4103, | |
| "eval_rouge2": 11.0056, | |
| "eval_rougeL": 20.1061, | |
| "eval_rougeLsum": 29.8896, | |
| "eval_runtime": 3417.5986, | |
| "eval_samples_per_second": 1.645, | |
| "eval_steps_per_second": 0.051, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 8.665599145983454e-05, | |
| "loss": 3.2864, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 8.51733238442606e-05, | |
| "loss": 3.2544, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 287.08663938800925, | |
| "eval_loss": 2.2861385345458984, | |
| "eval_rouge1": 34.4828, | |
| "eval_rouge2": 11.5674, | |
| "eval_rougeL": 20.789, | |
| "eval_rougeLsum": 31.0001, | |
| "eval_runtime": 3421.9145, | |
| "eval_samples_per_second": 1.643, | |
| "eval_steps_per_second": 0.051, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 8.369065622868666e-05, | |
| "loss": 3.2155, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 8.220798861311271e-05, | |
| "loss": 3.1714, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 294.1647393702188, | |
| "eval_loss": 2.2531518936157227, | |
| "eval_rouge1": 35.2859, | |
| "eval_rouge2": 12.0002, | |
| "eval_rougeL": 20.8964, | |
| "eval_rougeLsum": 31.7411, | |
| "eval_runtime": 3423.9736, | |
| "eval_samples_per_second": 1.642, | |
| "eval_steps_per_second": 0.051, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 8.072532099753878e-05, | |
| "loss": 3.1413, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 7.924265338196484e-05, | |
| "loss": 3.1154, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 310.6292474648639, | |
| "eval_loss": 2.2275736331939697, | |
| "eval_rouge1": 35.8542, | |
| "eval_rouge2": 12.3243, | |
| "eval_rougeL": 21.1165, | |
| "eval_rougeLsum": 32.2099, | |
| "eval_runtime": 3465.1719, | |
| "eval_samples_per_second": 1.622, | |
| "eval_steps_per_second": 0.051, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 7.775998576639089e-05, | |
| "loss": 3.0784, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 7.627731815081696e-05, | |
| "loss": 3.0757, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 203.26721223981497, | |
| "eval_loss": 2.194120407104492, | |
| "eval_rouge1": 36.9629, | |
| "eval_rouge2": 13.0355, | |
| "eval_rougeL": 21.6003, | |
| "eval_rougeLsum": 33.5146, | |
| "eval_runtime": 3112.0404, | |
| "eval_samples_per_second": 1.806, | |
| "eval_steps_per_second": 0.057, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 7.479465053524301e-05, | |
| "loss": 3.0279, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 7.331198291966908e-05, | |
| "loss": 3.0082, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 225.13271659847, | |
| "eval_loss": 2.182365655899048, | |
| "eval_rouge1": 36.8991, | |
| "eval_rouge2": 12.9026, | |
| "eval_rougeL": 21.6323, | |
| "eval_rougeLsum": 33.4198, | |
| "eval_runtime": 3290.7301, | |
| "eval_samples_per_second": 1.708, | |
| "eval_steps_per_second": 0.053, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.182931530409514e-05, | |
| "loss": 3.0088, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 7.034664768852119e-05, | |
| "loss": 2.9619, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 249.00177904287494, | |
| "eval_loss": 2.1777162551879883, | |
| "eval_rouge1": 38.0417, | |
| "eval_rouge2": 13.2345, | |
| "eval_rougeL": 21.668, | |
| "eval_rougeLsum": 34.3626, | |
| "eval_runtime": 3366.257, | |
| "eval_samples_per_second": 1.67, | |
| "eval_steps_per_second": 0.052, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 6.886398007294726e-05, | |
| "loss": 2.9599, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 6.738131245737331e-05, | |
| "loss": 2.9378, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 259.15815691158156, | |
| "eval_loss": 2.151700496673584, | |
| "eval_rouge1": 38.0194, | |
| "eval_rouge2": 13.365, | |
| "eval_rougeL": 21.9451, | |
| "eval_rougeLsum": 34.4388, | |
| "eval_runtime": 3366.4307, | |
| "eval_samples_per_second": 1.67, | |
| "eval_steps_per_second": 0.052, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 6.589864484179937e-05, | |
| "loss": 2.911, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 6.441597722622544e-05, | |
| "loss": 2.8819, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 238.5306884895926, | |
| "eval_loss": 2.1349971294403076, | |
| "eval_rouge1": 38.6339, | |
| "eval_rouge2": 13.5411, | |
| "eval_rougeL": 22.0768, | |
| "eval_rougeLsum": 35.0001, | |
| "eval_runtime": 3302.6091, | |
| "eval_samples_per_second": 1.702, | |
| "eval_steps_per_second": 0.053, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 6.293330961065149e-05, | |
| "loss": 2.876, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 6.145064199507754e-05, | |
| "loss": 2.8647, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 249.58797367016544, | |
| "eval_loss": 2.135904550552368, | |
| "eval_rouge1": 39.6584, | |
| "eval_rouge2": 14.1798, | |
| "eval_rougeL": 22.2953, | |
| "eval_rougeLsum": 35.9444, | |
| "eval_runtime": 3354.399, | |
| "eval_samples_per_second": 1.676, | |
| "eval_steps_per_second": 0.052, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 5.996797437950361e-05, | |
| "loss": 2.8483, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 5.848530676392967e-05, | |
| "loss": 2.8431, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 246.02045899306174, | |
| "eval_loss": 2.119384288787842, | |
| "eval_rouge1": 39.2036, | |
| "eval_rouge2": 13.9567, | |
| "eval_rougeL": 22.2848, | |
| "eval_rougeLsum": 35.6373, | |
| "eval_runtime": 3329.6362, | |
| "eval_samples_per_second": 1.688, | |
| "eval_steps_per_second": 0.053, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 5.700263914835573e-05, | |
| "loss": 2.834, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 5.551997153278178e-05, | |
| "loss": 2.8489, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 273.44209215442095, | |
| "eval_loss": 2.114163875579834, | |
| "eval_rouge1": 39.0102, | |
| "eval_rouge2": 13.665, | |
| "eval_rougeL": 22.2047, | |
| "eval_rougeLsum": 35.3681, | |
| "eval_runtime": 3369.9399, | |
| "eval_samples_per_second": 1.668, | |
| "eval_steps_per_second": 0.052, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 5.403730391720784e-05, | |
| "loss": 2.8122, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 5.25546363016339e-05, | |
| "loss": 2.7927, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 244.72140188578544, | |
| "eval_loss": 2.1085472106933594, | |
| "eval_rouge1": 39.9059, | |
| "eval_rouge2": 14.4122, | |
| "eval_rougeL": 22.5028, | |
| "eval_rougeLsum": 36.2593, | |
| "eval_runtime": 3298.5033, | |
| "eval_samples_per_second": 1.704, | |
| "eval_steps_per_second": 0.053, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 5.1071968686059954e-05, | |
| "loss": 2.8072, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 4.958930107048602e-05, | |
| "loss": 2.8051, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 230.1517523572318, | |
| "eval_loss": 2.1006970405578613, | |
| "eval_rouge1": 39.3406, | |
| "eval_rouge2": 13.9501, | |
| "eval_rougeL": 22.2845, | |
| "eval_rougeLsum": 35.8231, | |
| "eval_runtime": 3168.0344, | |
| "eval_samples_per_second": 1.774, | |
| "eval_steps_per_second": 0.056, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 4.810663345491208e-05, | |
| "loss": 2.8069, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4.662396583933814e-05, | |
| "loss": 2.7997, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 266.1441024728696, | |
| "eval_loss": 2.0949618816375732, | |
| "eval_rouge1": 39.4434, | |
| "eval_rouge2": 14.046, | |
| "eval_rougeL": 22.3371, | |
| "eval_rougeLsum": 35.8579, | |
| "eval_runtime": 3355.3929, | |
| "eval_samples_per_second": 1.675, | |
| "eval_steps_per_second": 0.052, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 4.51412982237642e-05, | |
| "loss": 2.766, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 4.365863060819026e-05, | |
| "loss": 2.7894, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 237.53833837395482, | |
| "eval_loss": 2.0886101722717285, | |
| "eval_rouge1": 40.6408, | |
| "eval_rouge2": 14.9391, | |
| "eval_rougeL": 22.8721, | |
| "eval_rougeLsum": 37.0402, | |
| "eval_runtime": 3398.8421, | |
| "eval_samples_per_second": 1.654, | |
| "eval_steps_per_second": 0.052, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 4.217596299261632e-05, | |
| "loss": 2.7678, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 4.0693295377042376e-05, | |
| "loss": 2.7881, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 238.2777085927771, | |
| "eval_loss": 2.082338809967041, | |
| "eval_rouge1": 40.1463, | |
| "eval_rouge2": 14.5452, | |
| "eval_rougeL": 22.6645, | |
| "eval_rougeLsum": 36.568, | |
| "eval_runtime": 3243.3567, | |
| "eval_samples_per_second": 1.733, | |
| "eval_steps_per_second": 0.054, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 3.921062776146844e-05, | |
| "loss": 2.7382, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.77279601458945e-05, | |
| "loss": 2.7694, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 242.4057996797723, | |
| "eval_loss": 2.0775482654571533, | |
| "eval_rouge1": 39.9889, | |
| "eval_rouge2": 14.4044, | |
| "eval_rougeL": 22.5574, | |
| "eval_rougeLsum": 36.3926, | |
| "eval_runtime": 3265.4881, | |
| "eval_samples_per_second": 1.721, | |
| "eval_steps_per_second": 0.054, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 3.624529253032056e-05, | |
| "loss": 2.7371, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 3.4762624914746615e-05, | |
| "loss": 2.7382, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 243.57089485856608, | |
| "eval_loss": 2.0729196071624756, | |
| "eval_rouge1": 40.4506, | |
| "eval_rouge2": 14.6977, | |
| "eval_rougeL": 22.7451, | |
| "eval_rougeLsum": 36.9094, | |
| "eval_runtime": 3321.1737, | |
| "eval_samples_per_second": 1.692, | |
| "eval_steps_per_second": 0.053, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.327995729917267e-05, | |
| "loss": 2.7799, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 3.179728968359873e-05, | |
| "loss": 2.7137, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 238.74114926169722, | |
| "eval_loss": 2.0669145584106445, | |
| "eval_rouge1": 40.7645, | |
| "eval_rouge2": 14.864, | |
| "eval_rougeL": 22.8929, | |
| "eval_rougeLsum": 37.2304, | |
| "eval_runtime": 4021.2465, | |
| "eval_samples_per_second": 1.398, | |
| "eval_steps_per_second": 0.044, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 3.031462206802479e-05, | |
| "loss": 2.7336, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 2.883195445245085e-05, | |
| "loss": 2.7185, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 238.28713752001423, | |
| "eval_loss": 2.07470965385437, | |
| "eval_rouge1": 40.5652, | |
| "eval_rouge2": 14.7185, | |
| "eval_rougeL": 22.7403, | |
| "eval_rougeLsum": 36.9759, | |
| "eval_runtime": 3243.9209, | |
| "eval_samples_per_second": 1.733, | |
| "eval_steps_per_second": 0.054, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 2.7349286836876908e-05, | |
| "loss": 2.715, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 2.586661922130297e-05, | |
| "loss": 2.7104, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 257.9969756271126, | |
| "eval_loss": 2.0688774585723877, | |
| "eval_rouge1": 41.0927, | |
| "eval_rouge2": 15.0971, | |
| "eval_rougeL": 22.9602, | |
| "eval_rougeLsum": 37.455, | |
| "eval_runtime": 3349.2799, | |
| "eval_samples_per_second": 1.678, | |
| "eval_steps_per_second": 0.053, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 2.438395160572903e-05, | |
| "loss": 2.7097, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 2.290128399015509e-05, | |
| "loss": 2.7038, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 245.09464508094646, | |
| "eval_loss": 2.0660908222198486, | |
| "eval_rouge1": 40.7524, | |
| "eval_rouge2": 14.7791, | |
| "eval_rougeL": 22.7677, | |
| "eval_rougeLsum": 37.1449, | |
| "eval_runtime": 3282.519, | |
| "eval_samples_per_second": 1.712, | |
| "eval_steps_per_second": 0.054, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 2.1418616374581146e-05, | |
| "loss": 2.7061, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.9935948759007204e-05, | |
| "loss": 2.6955, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 239.2723714641523, | |
| "eval_loss": 2.067803144454956, | |
| "eval_rouge1": 41.0961, | |
| "eval_rouge2": 15.0948, | |
| "eval_rougeL": 22.9771, | |
| "eval_rougeLsum": 37.4876, | |
| "eval_runtime": 3276.1425, | |
| "eval_samples_per_second": 1.716, | |
| "eval_steps_per_second": 0.054, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.8453281143433266e-05, | |
| "loss": 2.6783, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 1.6970613527859324e-05, | |
| "loss": 2.6906, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 249.8028820494574, | |
| "eval_loss": 2.0625176429748535, | |
| "eval_rouge1": 41.1383, | |
| "eval_rouge2": 14.9956, | |
| "eval_rougeL": 22.9293, | |
| "eval_rougeLsum": 37.5443, | |
| "eval_runtime": 3296.0562, | |
| "eval_samples_per_second": 1.705, | |
| "eval_steps_per_second": 0.053, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 1.5487945912285385e-05, | |
| "loss": 2.6802, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 1.4005278296711445e-05, | |
| "loss": 2.7057, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 238.6906244440491, | |
| "eval_loss": 2.060208559036255, | |
| "eval_rouge1": 41.1908, | |
| "eval_rouge2": 15.1675, | |
| "eval_rougeL": 23.0798, | |
| "eval_rougeLsum": 37.5906, | |
| "eval_runtime": 3211.8848, | |
| "eval_samples_per_second": 1.75, | |
| "eval_steps_per_second": 0.055, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 1.2522610681137503e-05, | |
| "loss": 2.6806, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 1.1039943065563562e-05, | |
| "loss": 2.6925, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 237.07045009784736, | |
| "eval_loss": 2.0573508739471436, | |
| "eval_rouge1": 40.9739, | |
| "eval_rouge2": 14.8989, | |
| "eval_rougeL": 22.8636, | |
| "eval_rougeLsum": 37.3584, | |
| "eval_runtime": 3212.5536, | |
| "eval_samples_per_second": 1.75, | |
| "eval_steps_per_second": 0.055, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 9.557275449989622e-06, | |
| "loss": 2.6881, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 8.074607834415681e-06, | |
| "loss": 2.6629, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 236.07898950364705, | |
| "eval_loss": 2.0589113235473633, | |
| "eval_rouge1": 41.2017, | |
| "eval_rouge2": 15.1175, | |
| "eval_rougeL": 22.9679, | |
| "eval_rougeLsum": 37.599, | |
| "eval_runtime": 3258.1795, | |
| "eval_samples_per_second": 1.725, | |
| "eval_steps_per_second": 0.054, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 6.59194021884174e-06, | |
| "loss": 2.65, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 5.1092726032678e-06, | |
| "loss": 2.6899, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 237.81409001956948, | |
| "eval_loss": 2.058128595352173, | |
| "eval_rouge1": 41.2137, | |
| "eval_rouge2": 15.15, | |
| "eval_rougeL": 23.0369, | |
| "eval_rougeLsum": 37.6244, | |
| "eval_runtime": 3226.4024, | |
| "eval_samples_per_second": 1.742, | |
| "eval_steps_per_second": 0.055, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 3.6266049876938586e-06, | |
| "loss": 2.6844, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 2.1439373721199182e-06, | |
| "loss": 2.6685, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "eval_block_avg": 0.0, | |
| "eval_gen_len": 239.2396370752535, | |
| "eval_loss": 2.0559887886047363, | |
| "eval_rouge1": 41.2047, | |
| "eval_rouge2": 15.0825, | |
| "eval_rougeL": 22.9882, | |
| "eval_rougeLsum": 37.5913, | |
| "eval_runtime": 3189.4588, | |
| "eval_samples_per_second": 1.762, | |
| "eval_steps_per_second": 0.055, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 6.612697565459775e-07, | |
| "loss": 2.6767, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 33723, | |
| "total_flos": 3.64032524422656e+17, | |
| "train_loss": 2.960794737135194, | |
| "train_runtime": 148191.3733, | |
| "train_samples_per_second": 0.91, | |
| "train_steps_per_second": 0.228 | |
| } | |
| ], | |
| "max_steps": 33723, | |
| "num_train_epochs": 3, | |
| "total_flos": 3.64032524422656e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |