| { | |
| "best_metric": 2.0858356952667236, | |
| "best_model_checkpoint": "./models/checkpoint-70000", | |
| "epoch": 0.9752155923041558, | |
| "global_step": 70000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.7678058113561536e-05, | |
| "loss": 2.5513, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.201458692550659, | |
| "eval_rouge1": 12.1725, | |
| "eval_rouge2": 2.4216, | |
| "eval_rougeL": 12.077, | |
| "eval_rougeLsum": 12.1411, | |
| "eval_runtime": 1560.686, | |
| "eval_samples_per_second": 8.565, | |
| "eval_steps_per_second": 2.141, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.535611622712307e-05, | |
| "loss": 2.4881, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1785953044891357, | |
| "eval_rouge1": 12.7586, | |
| "eval_rouge2": 2.6851, | |
| "eval_rougeL": 12.6518, | |
| "eval_rougeLsum": 12.733, | |
| "eval_runtime": 1548.5251, | |
| "eval_samples_per_second": 8.633, | |
| "eval_steps_per_second": 2.158, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.3034174340684604e-05, | |
| "loss": 2.459, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1513352394104004, | |
| "eval_rouge1": 12.7004, | |
| "eval_rouge2": 2.6318, | |
| "eval_rougeL": 12.5832, | |
| "eval_rougeLsum": 12.668, | |
| "eval_runtime": 1564.1475, | |
| "eval_samples_per_second": 8.547, | |
| "eval_steps_per_second": 2.137, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.071223245424614e-05, | |
| "loss": 2.4361, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.127916097640991, | |
| "eval_rouge1": 13.1946, | |
| "eval_rouge2": 2.8216, | |
| "eval_rougeL": 13.0738, | |
| "eval_rougeLsum": 13.1592, | |
| "eval_runtime": 1584.478, | |
| "eval_samples_per_second": 8.437, | |
| "eval_steps_per_second": 2.109, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 3.8390290567807665e-05, | |
| "loss": 2.4613, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.098001003265381, | |
| "eval_rouge1": 12.8818, | |
| "eval_rouge2": 2.8541, | |
| "eval_rougeL": 12.7484, | |
| "eval_rougeLsum": 12.8289, | |
| "eval_runtime": 1565.1942, | |
| "eval_samples_per_second": 8.541, | |
| "eval_steps_per_second": 2.135, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.60683486813692e-05, | |
| "loss": 2.462, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.0897421836853027, | |
| "eval_rouge1": 13.1141, | |
| "eval_rouge2": 2.8014, | |
| "eval_rougeL": 12.9971, | |
| "eval_rougeLsum": 13.0729, | |
| "eval_runtime": 1561.4037, | |
| "eval_samples_per_second": 8.562, | |
| "eval_steps_per_second": 2.14, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 3.374640679493074e-05, | |
| "loss": 2.4478, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.0858356952667236, | |
| "eval_rouge1": 13.4344, | |
| "eval_rouge2": 2.8499, | |
| "eval_rougeL": 13.3259, | |
| "eval_rougeLsum": 13.3837, | |
| "eval_runtime": 1559.9208, | |
| "eval_samples_per_second": 8.57, | |
| "eval_steps_per_second": 2.142, | |
| "step": 70000 | |
| } | |
| ], | |
| "max_steps": 215337, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.4804990931554304e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |