| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.351610095735422, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.8262837249782422e-05, | |
| "loss": 2.0421, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.652219321148825e-05, | |
| "loss": 1.9451, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 18.9987, | |
| "eval_loss": 1.7494806051254272, | |
| "eval_rouge1": 0.2437, | |
| "eval_rouge2": 0.1156, | |
| "eval_rougeL": 0.2014, | |
| "eval_rougeLsum": 0.2016, | |
| "eval_runtime": 182.5028, | |
| "eval_samples_per_second": 12.592, | |
| "eval_steps_per_second": 1.578, | |
| "step": 1149 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.4781549173194083e-05, | |
| "loss": 1.9164, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 1.3044386422976502e-05, | |
| "loss": 1.9045, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 18.9909, | |
| "eval_loss": 1.7317167520523071, | |
| "eval_rouge1": 0.2439, | |
| "eval_rouge2": 0.1161, | |
| "eval_rougeL": 0.2017, | |
| "eval_rougeLsum": 0.2017, | |
| "eval_runtime": 183.2456, | |
| "eval_samples_per_second": 12.541, | |
| "eval_steps_per_second": 1.572, | |
| "step": 2298 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 1.1303742384682332e-05, | |
| "loss": 1.8751, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 9.563098346388164e-06, | |
| "loss": 1.8821, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 18.9909, | |
| "eval_loss": 1.7243174314498901, | |
| "eval_rouge1": 0.2441, | |
| "eval_rouge2": 0.1157, | |
| "eval_rougeL": 0.2014, | |
| "eval_rougeLsum": 0.2014, | |
| "eval_runtime": 180.4343, | |
| "eval_samples_per_second": 12.736, | |
| "eval_steps_per_second": 1.596, | |
| "step": 3447 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 7.822454308093995e-06, | |
| "loss": 1.874, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 6.081810269799826e-06, | |
| "loss": 1.8722, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 4.3411662315056575e-06, | |
| "loss": 1.8444, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 18.993, | |
| "eval_loss": 1.7211626768112183, | |
| "eval_rouge1": 0.2439, | |
| "eval_rouge2": 0.1154, | |
| "eval_rougeL": 0.2012, | |
| "eval_rougeLsum": 0.2011, | |
| "eval_runtime": 181.6857, | |
| "eval_samples_per_second": 12.648, | |
| "eval_steps_per_second": 1.585, | |
| "step": 4596 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 2.6005221932114883e-06, | |
| "loss": 1.8488, | |
| "step": 5000 | |
| } | |
| ], | |
| "max_steps": 5745, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.080444174729216e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |