| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.99712, | |
| "eval_steps": 500, | |
| "global_step": 12496, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.8399487836107557e-05, | |
| "loss": 2.1501, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.9072, | |
| "eval_gen_len": 25.465454545454545, | |
| "eval_loss": 1.7061578035354614, | |
| "eval_precision": 0.9082, | |
| "eval_recall": 0.9065, | |
| "eval_rouge1": 0.4566, | |
| "eval_rouge2": 0.209, | |
| "eval_rougeL": 0.3745, | |
| "eval_rougeLsum": 0.3744, | |
| "eval_runtime": 882.5908, | |
| "eval_samples_per_second": 6.232, | |
| "eval_steps_per_second": 0.39, | |
| "step": 781 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.679897567221511e-05, | |
| "loss": 1.839, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.5198463508322665e-05, | |
| "loss": 1.7722, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.9097, | |
| "eval_gen_len": 25.42981818181818, | |
| "eval_loss": 1.6313680410385132, | |
| "eval_precision": 0.9107, | |
| "eval_recall": 0.909, | |
| "eval_rouge1": 0.4712, | |
| "eval_rouge2": 0.2226, | |
| "eval_rougeL": 0.3906, | |
| "eval_rougeLsum": 0.3904, | |
| "eval_runtime": 869.6253, | |
| "eval_samples_per_second": 6.325, | |
| "eval_steps_per_second": 0.396, | |
| "step": 1562 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 1.3597951344430219e-05, | |
| "loss": 1.7218, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.9106, | |
| "eval_gen_len": 25.656909090909092, | |
| "eval_loss": 1.594840168952942, | |
| "eval_precision": 0.9112, | |
| "eval_recall": 0.9103, | |
| "eval_rouge1": 0.4776, | |
| "eval_rouge2": 0.2284, | |
| "eval_rougeL": 0.3965, | |
| "eval_rougeLsum": 0.3963, | |
| "eval_runtime": 866.3975, | |
| "eval_samples_per_second": 6.348, | |
| "eval_steps_per_second": 0.397, | |
| "step": 2343 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 1.1997439180537773e-05, | |
| "loss": 1.6923, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 1.0396927016645328e-05, | |
| "loss": 1.6668, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.9112, | |
| "eval_gen_len": 25.34509090909091, | |
| "eval_loss": 1.5707985162734985, | |
| "eval_precision": 0.9122, | |
| "eval_recall": 0.9107, | |
| "eval_rouge1": 0.481, | |
| "eval_rouge2": 0.2316, | |
| "eval_rougeL": 0.4002, | |
| "eval_rougeLsum": 0.4, | |
| "eval_runtime": 864.0634, | |
| "eval_samples_per_second": 6.365, | |
| "eval_steps_per_second": 0.398, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 8.796414852752882e-06, | |
| "loss": 1.6437, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_f1": 0.9118, | |
| "eval_gen_len": 25.482, | |
| "eval_loss": 1.5564521551132202, | |
| "eval_precision": 0.9127, | |
| "eval_recall": 0.9113, | |
| "eval_rouge1": 0.4844, | |
| "eval_rouge2": 0.2346, | |
| "eval_rougeL": 0.4034, | |
| "eval_rougeLsum": 0.4031, | |
| "eval_runtime": 868.5536, | |
| "eval_samples_per_second": 6.332, | |
| "eval_steps_per_second": 0.396, | |
| "step": 3906 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 7.195902688860435e-06, | |
| "loss": 1.6338, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 5.595390524967991e-06, | |
| "loss": 1.6186, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_f1": 0.912, | |
| "eval_gen_len": 25.419090909090908, | |
| "eval_loss": 1.5476245880126953, | |
| "eval_precision": 0.9129, | |
| "eval_recall": 0.9115, | |
| "eval_rouge1": 0.4852, | |
| "eval_rouge2": 0.236, | |
| "eval_rougeL": 0.4047, | |
| "eval_rougeLsum": 0.4044, | |
| "eval_runtime": 867.5381, | |
| "eval_samples_per_second": 6.34, | |
| "eval_steps_per_second": 0.397, | |
| "step": 4687 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 3.9948783610755446e-06, | |
| "loss": 1.607, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_f1": 0.9122, | |
| "eval_gen_len": 25.49490909090909, | |
| "eval_loss": 1.5425922870635986, | |
| "eval_precision": 0.9129, | |
| "eval_recall": 0.9118, | |
| "eval_rouge1": 0.486, | |
| "eval_rouge2": 0.2367, | |
| "eval_rougeL": 0.4052, | |
| "eval_rougeLsum": 0.405, | |
| "eval_runtime": 866.1986, | |
| "eval_samples_per_second": 6.35, | |
| "eval_steps_per_second": 0.397, | |
| "step": 5468 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 2.3943661971830984e-06, | |
| "loss": 1.6029, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "learning_rate": 7.93854033290653e-07, | |
| "loss": 1.5972, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_f1": 0.9123, | |
| "eval_gen_len": 25.383636363636363, | |
| "eval_loss": 1.5380274057388306, | |
| "eval_precision": 0.9131, | |
| "eval_recall": 0.9118, | |
| "eval_rouge1": 0.4872, | |
| "eval_rouge2": 0.2387, | |
| "eval_rougeL": 0.407, | |
| "eval_rougeLsum": 0.4071, | |
| "eval_runtime": 863.1004, | |
| "eval_samples_per_second": 6.372, | |
| "eval_steps_per_second": 0.399, | |
| "step": 6248 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 9.596670934699104e-06, | |
| "loss": 1.5886, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 8.796414852752882e-06, | |
| "loss": 1.5836, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_f1": 0.9126, | |
| "eval_gen_len": 25.499454545454544, | |
| "eval_loss": 1.527321696281433, | |
| "eval_precision": 0.9133, | |
| "eval_recall": 0.9122, | |
| "eval_rouge1": 0.4891, | |
| "eval_rouge2": 0.2399, | |
| "eval_rougeL": 0.4088, | |
| "eval_rougeLsum": 0.4089, | |
| "eval_runtime": 859.8962, | |
| "eval_samples_per_second": 6.396, | |
| "eval_steps_per_second": 0.4, | |
| "step": 7029 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 7.996158770806658e-06, | |
| "loss": 1.5667, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_f1": 0.9127, | |
| "eval_gen_len": 25.386727272727274, | |
| "eval_loss": 1.5195879936218262, | |
| "eval_precision": 0.9135, | |
| "eval_recall": 0.9123, | |
| "eval_rouge1": 0.4906, | |
| "eval_rouge2": 0.2416, | |
| "eval_rougeL": 0.411, | |
| "eval_rougeLsum": 0.4112, | |
| "eval_runtime": 855.4017, | |
| "eval_samples_per_second": 6.43, | |
| "eval_steps_per_second": 0.402, | |
| "step": 7810 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "learning_rate": 7.195902688860435e-06, | |
| "loss": 1.5577, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "learning_rate": 6.395646606914213e-06, | |
| "loss": 1.5521, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_f1": 0.9127, | |
| "eval_gen_len": 25.21909090909091, | |
| "eval_loss": 1.5124093294143677, | |
| "eval_precision": 0.9137, | |
| "eval_recall": 0.912, | |
| "eval_rouge1": 0.4899, | |
| "eval_rouge2": 0.2406, | |
| "eval_rougeL": 0.4102, | |
| "eval_rougeLsum": 0.4103, | |
| "eval_runtime": 851.6688, | |
| "eval_samples_per_second": 6.458, | |
| "eval_steps_per_second": 0.404, | |
| "step": 8592 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 5.595390524967991e-06, | |
| "loss": 1.5413, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_f1": 0.9128, | |
| "eval_gen_len": 25.349090909090908, | |
| "eval_loss": 1.5083255767822266, | |
| "eval_precision": 0.9137, | |
| "eval_recall": 0.9123, | |
| "eval_rouge1": 0.4914, | |
| "eval_rouge2": 0.2416, | |
| "eval_rougeL": 0.4118, | |
| "eval_rougeLsum": 0.412, | |
| "eval_runtime": 856.2804, | |
| "eval_samples_per_second": 6.423, | |
| "eval_steps_per_second": 0.402, | |
| "step": 9373 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "learning_rate": 4.795134443021768e-06, | |
| "loss": 1.5354, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 3.9948783610755446e-06, | |
| "loss": 1.5291, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_f1": 0.913, | |
| "eval_gen_len": 25.208181818181817, | |
| "eval_loss": 1.5044068098068237, | |
| "eval_precision": 0.914, | |
| "eval_recall": 0.9123, | |
| "eval_rouge1": 0.4913, | |
| "eval_rouge2": 0.2419, | |
| "eval_rougeL": 0.4118, | |
| "eval_rougeLsum": 0.4119, | |
| "eval_runtime": 858.747, | |
| "eval_samples_per_second": 6.405, | |
| "eval_steps_per_second": 0.401, | |
| "step": 10154 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "learning_rate": 3.194622279129322e-06, | |
| "loss": 1.527, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_f1": 0.913, | |
| "eval_gen_len": 25.10690909090909, | |
| "eval_loss": 1.5025616884231567, | |
| "eval_precision": 0.9141, | |
| "eval_recall": 0.9123, | |
| "eval_rouge1": 0.4917, | |
| "eval_rouge2": 0.2426, | |
| "eval_rougeL": 0.4126, | |
| "eval_rougeLsum": 0.4128, | |
| "eval_runtime": 849.8377, | |
| "eval_samples_per_second": 6.472, | |
| "eval_steps_per_second": 0.405, | |
| "step": 10935 | |
| }, | |
| { | |
| "epoch": 14.08, | |
| "learning_rate": 2.3943661971830984e-06, | |
| "loss": 1.5206, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 14.72, | |
| "learning_rate": 1.594110115236876e-06, | |
| "loss": 1.5203, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_f1": 0.9131, | |
| "eval_gen_len": 25.106181818181817, | |
| "eval_loss": 1.5006238222122192, | |
| "eval_precision": 0.9143, | |
| "eval_recall": 0.9123, | |
| "eval_rouge1": 0.4921, | |
| "eval_rouge2": 0.243, | |
| "eval_rougeL": 0.4135, | |
| "eval_rougeLsum": 0.4136, | |
| "eval_runtime": 854.2433, | |
| "eval_samples_per_second": 6.438, | |
| "eval_steps_per_second": 0.403, | |
| "step": 11717 | |
| }, | |
| { | |
| "epoch": 15.36, | |
| "learning_rate": 7.93854033290653e-07, | |
| "loss": 1.5126, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_f1": 0.9132, | |
| "eval_gen_len": 25.133454545454544, | |
| "eval_loss": 1.5003753900527954, | |
| "eval_precision": 0.9143, | |
| "eval_recall": 0.9124, | |
| "eval_rouge1": 0.4923, | |
| "eval_rouge2": 0.2429, | |
| "eval_rougeL": 0.4134, | |
| "eval_rougeLsum": 0.4134, | |
| "eval_runtime": 851.4198, | |
| "eval_samples_per_second": 6.46, | |
| "eval_steps_per_second": 0.404, | |
| "step": 12496 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "step": 12496, | |
| "total_flos": 2.310831828476363e+18, | |
| "train_loss": 0.8017544373965294, | |
| "train_runtime": 29679.9237, | |
| "train_samples_per_second": 53.908, | |
| "train_steps_per_second": 0.421 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 12496, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 16, | |
| "save_steps": 500, | |
| "total_flos": 2.310831828476363e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |