| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0832671336751643, | |
| "eval_steps": 500, | |
| "global_step": 7500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.9986295347545738e-05, | |
| "loss": 4.4247, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_gen_len": 6.267, | |
| "eval_loss": 3.5801377296447754, | |
| "eval_rouge1": 27.9882, | |
| "eval_rouge2": 8.9331, | |
| "eval_rougeL": 27.6011, | |
| "eval_rougeLsum": 27.5353, | |
| "eval_runtime": 1554.6657, | |
| "eval_samples_per_second": 1.93, | |
| "eval_steps_per_second": 0.482, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.9945218953682736e-05, | |
| "loss": 2.959, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_gen_len": 6.267333333333333, | |
| "eval_loss": 3.238922119140625, | |
| "eval_rouge1": 34.4913, | |
| "eval_rouge2": 14.1739, | |
| "eval_rougeL": 34.2505, | |
| "eval_rougeLsum": 34.1717, | |
| "eval_runtime": 1549.8046, | |
| "eval_samples_per_second": 1.936, | |
| "eval_steps_per_second": 0.484, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 1.9876883405951378e-05, | |
| "loss": 2.7367, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 2.983262300491333, | |
| "eval_rouge1": 40.5052, | |
| "eval_rouge2": 18.725, | |
| "eval_rougeL": 40.1333, | |
| "eval_rougeLsum": 40.0554, | |
| "eval_runtime": 1542.1597, | |
| "eval_samples_per_second": 1.945, | |
| "eval_steps_per_second": 0.486, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.9781476007338058e-05, | |
| "loss": 2.553, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_gen_len": 6.267333333333333, | |
| "eval_loss": 2.781419277191162, | |
| "eval_rouge1": 45.972, | |
| "eval_rouge2": 24.75, | |
| "eval_rougeL": 45.6989, | |
| "eval_rougeLsum": 45.6257, | |
| "eval_runtime": 1547.6212, | |
| "eval_samples_per_second": 1.938, | |
| "eval_steps_per_second": 0.485, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.9659258262890683e-05, | |
| "loss": 2.3988, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 2.6185145378112793, | |
| "eval_rouge1": 51.3237, | |
| "eval_rouge2": 30.8584, | |
| "eval_rougeL": 51.0697, | |
| "eval_rougeLsum": 50.9947, | |
| "eval_runtime": 1538.5542, | |
| "eval_samples_per_second": 1.95, | |
| "eval_steps_per_second": 0.487, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1.9510565162951538e-05, | |
| "loss": 2.2788, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 2.4821877479553223, | |
| "eval_rouge1": 55.054, | |
| "eval_rouge2": 35.9558, | |
| "eval_rougeL": 54.885, | |
| "eval_rougeLsum": 54.8263, | |
| "eval_runtime": 1535.9732, | |
| "eval_samples_per_second": 1.953, | |
| "eval_steps_per_second": 0.488, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 1.9335804264972018e-05, | |
| "loss": 2.185, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 2.372607469558716, | |
| "eval_rouge1": 58.4125, | |
| "eval_rouge2": 39.7017, | |
| "eval_rougeL": 58.2864, | |
| "eval_rougeLsum": 58.2323, | |
| "eval_runtime": 1555.498, | |
| "eval_samples_per_second": 1.929, | |
| "eval_steps_per_second": 0.482, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.913545457642601e-05, | |
| "loss": 2.1024, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 2.2966315746307373, | |
| "eval_rouge1": 60.2269, | |
| "eval_rouge2": 42.343, | |
| "eval_rougeL": 60.1064, | |
| "eval_rougeLsum": 60.015, | |
| "eval_runtime": 1539.5016, | |
| "eval_samples_per_second": 1.949, | |
| "eval_steps_per_second": 0.487, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.891006524188368e-05, | |
| "loss": 2.0395, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 2.2079367637634277, | |
| "eval_rouge1": 63.9442, | |
| "eval_rouge2": 47.1262, | |
| "eval_rougeL": 63.8226, | |
| "eval_rougeLsum": 63.783, | |
| "eval_runtime": 1534.6718, | |
| "eval_samples_per_second": 1.955, | |
| "eval_steps_per_second": 0.489, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.866025403784439e-05, | |
| "loss": 1.9929, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 2.142320156097412, | |
| "eval_rouge1": 66.0535, | |
| "eval_rouge2": 49.403, | |
| "eval_rougeL": 65.9837, | |
| "eval_rougeLsum": 65.9295, | |
| "eval_runtime": 1531.6986, | |
| "eval_samples_per_second": 1.959, | |
| "eval_steps_per_second": 0.49, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 1.8386705679454243e-05, | |
| "loss": 1.944, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 2.094369888305664, | |
| "eval_rouge1": 67.4662, | |
| "eval_rouge2": 50.9242, | |
| "eval_rougeL": 67.3906, | |
| "eval_rougeLsum": 67.3416, | |
| "eval_runtime": 1527.1607, | |
| "eval_samples_per_second": 1.964, | |
| "eval_steps_per_second": 0.491, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.8090169943749477e-05, | |
| "loss": 1.902, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 2.0382838249206543, | |
| "eval_rouge1": 69.3267, | |
| "eval_rouge2": 53.8395, | |
| "eval_rougeL": 69.2759, | |
| "eval_rougeLsum": 69.2151, | |
| "eval_runtime": 1532.3017, | |
| "eval_samples_per_second": 1.958, | |
| "eval_steps_per_second": 0.489, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.777145961456971e-05, | |
| "loss": 1.8712, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 2.010657548904419, | |
| "eval_rouge1": 70.2771, | |
| "eval_rouge2": 54.5208, | |
| "eval_rougeL": 70.2492, | |
| "eval_rougeLsum": 70.2095, | |
| "eval_runtime": 1563.3117, | |
| "eval_samples_per_second": 1.919, | |
| "eval_steps_per_second": 0.48, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 1.7431448254773943e-05, | |
| "loss": 1.8387, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 1.9761910438537598, | |
| "eval_rouge1": 71.145, | |
| "eval_rouge2": 56.4319, | |
| "eval_rougeL": 71.1008, | |
| "eval_rougeLsum": 71.0876, | |
| "eval_runtime": 1552.4923, | |
| "eval_samples_per_second": 1.932, | |
| "eval_steps_per_second": 0.483, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.7071067811865477e-05, | |
| "loss": 1.7558, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_gen_len": 6.267666666666667, | |
| "eval_loss": 1.9575979709625244, | |
| "eval_rouge1": 72.527, | |
| "eval_rouge2": 58.0461, | |
| "eval_rougeL": 72.4801, | |
| "eval_rougeLsum": 72.461, | |
| "eval_runtime": 1522.5751, | |
| "eval_samples_per_second": 1.97, | |
| "eval_steps_per_second": 0.493, | |
| "step": 7500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 30000, | |
| "num_train_epochs": 5, | |
| "save_steps": 2500, | |
| "total_flos": 3.0856111823601216e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |