| { | |
| "best_metric": 1.1539206504821777, | |
| "best_model_checkpoint": "./jako_mbartLarge_6p_run1/checkpoint-4000", | |
| "epoch": 1.9198464122870171, | |
| "eval_steps": 1000, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.951290793960059e-05, | |
| "loss": 1.8861, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.829517778860205e-05, | |
| "loss": 1.4641, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_bleu": 21.6162, | |
| "eval_gen_len": 19.4434, | |
| "eval_loss": 1.3276299238204956, | |
| "eval_runtime": 299.0357, | |
| "eval_samples_per_second": 13.931, | |
| "eval_steps_per_second": 0.873, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.707744763760351e-05, | |
| "loss": 1.3282, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.585971748660497e-05, | |
| "loss": 1.2615, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_bleu": 24.346, | |
| "eval_gen_len": 19.4734, | |
| "eval_loss": 1.186624526977539, | |
| "eval_runtime": 297.0522, | |
| "eval_samples_per_second": 14.024, | |
| "eval_steps_per_second": 0.879, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.4641987335606436e-05, | |
| "loss": 1.0805, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 4.342425718460789e-05, | |
| "loss": 0.9103, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_bleu": 25.4249, | |
| "eval_gen_len": 19.0086, | |
| "eval_loss": 1.1637648344039917, | |
| "eval_runtime": 293.4921, | |
| "eval_samples_per_second": 14.195, | |
| "eval_steps_per_second": 0.889, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 4.2206527033609356e-05, | |
| "loss": 0.8534, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 4.0988796882610817e-05, | |
| "loss": 0.8285, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_bleu": 26.2658, | |
| "eval_gen_len": 19.3961, | |
| "eval_loss": 1.1539206504821777, | |
| "eval_runtime": 298.2089, | |
| "eval_samples_per_second": 13.97, | |
| "eval_steps_per_second": 0.875, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 20830, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000, | |
| "total_flos": 1.3880032710832947e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |