| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 14403, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.479275151010207e-05, | |
| "loss": 0.8616, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.9585503020204125e-05, | |
| "loss": 0.7086, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3.437825453030619e-05, | |
| "loss": 0.6876, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 2.917100604040825e-05, | |
| "loss": 0.6658, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 2.396375755051031e-05, | |
| "loss": 0.6513, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.8756509060612374e-05, | |
| "loss": 0.6115, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.3549260570714433e-05, | |
| "loss": 0.5986, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 8.342012080816496e-06, | |
| "loss": 0.6065, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.134763590918559e-06, | |
| "loss": 0.5997, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 20.0, | |
| "eval_loss": 0.7166666984558105, | |
| "eval_rouge1": 15.1597, | |
| "eval_rouge2": 11.4975, | |
| "eval_rougeL": 14.0709, | |
| "eval_rougeLsum": 14.9737, | |
| "eval_runtime": 13.0755, | |
| "eval_samples_per_second": 6.501, | |
| "eval_steps_per_second": 3.289, | |
| "step": 4801 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 2.396375755051031e-05, | |
| "loss": 0.535, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 2.136013330556134e-05, | |
| "loss": 0.5328, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.8756509060612374e-05, | |
| "loss": 0.555, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 1.6152884815663406e-05, | |
| "loss": 0.5381, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 1.3549260570714433e-05, | |
| "loss": 0.5375, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 1.0945636325765466e-05, | |
| "loss": 0.5333, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 8.342012080816496e-06, | |
| "loss": 0.5291, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 5.738387835867528e-06, | |
| "loss": 0.5196, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.134763590918559e-06, | |
| "loss": 0.4995, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 5.311393459695897e-07, | |
| "loss": 0.5057, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 20.0, | |
| "eval_loss": 0.6984732747077942, | |
| "eval_rouge1": 15.2746, | |
| "eval_rouge2": 11.537, | |
| "eval_rougeL": 14.1481, | |
| "eval_rougeLsum": 15.1267, | |
| "eval_runtime": 13.3919, | |
| "eval_samples_per_second": 6.347, | |
| "eval_steps_per_second": 3.211, | |
| "step": 9602 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.528501006734708e-05, | |
| "loss": 0.4867, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.3549260570714433e-05, | |
| "loss": 0.4811, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 1.181351107408179e-05, | |
| "loss": 0.4842, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.0077761577449144e-05, | |
| "loss": 0.4909, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 8.342012080816496e-06, | |
| "loss": 0.4659, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 6.606262584183851e-06, | |
| "loss": 0.4747, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 4.870513087551205e-06, | |
| "loss": 0.473, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 3.134763590918559e-06, | |
| "loss": 0.4484, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.3990140942859125e-06, | |
| "loss": 0.4804, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 20.0, | |
| "eval_loss": 0.6930598020553589, | |
| "eval_rouge1": 15.2658, | |
| "eval_rouge2": 11.5104, | |
| "eval_rougeL": 14.117, | |
| "eval_rougeLsum": 15.1103, | |
| "eval_runtime": 13.632, | |
| "eval_samples_per_second": 6.235, | |
| "eval_steps_per_second": 3.154, | |
| "step": 14403 | |
| } | |
| ], | |
| "max_steps": 14403, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.113119515662336e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |