| { | |
| "best_metric": NaN, | |
| "best_model_checkpoint": "/mnt/bn/qingyi-bn-lq/llama/saved_bloomz-7b1-mt_TQA/checkpoint-200", | |
| "epoch": 0.6990715456034954, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 56.1213, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 46.3785, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 27.2008, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 58.3372, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003, | |
| "loss": 21.5097, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029208443271767806, | |
| "loss": 149.5568, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002841688654353562, | |
| "loss": 12.1125, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0002762532981530343, | |
| "loss": 64.6813, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00026833773087071237, | |
| "loss": 12.8244, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0002604221635883905, | |
| "loss": 13.6392, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": NaN, | |
| "eval_runtime": 39.2226, | |
| "eval_samples_per_second": 50.991, | |
| "eval_steps_per_second": 1.606, | |
| "step": 200 | |
| } | |
| ], | |
| "max_steps": 858, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.887154186856956e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |